ever on top-level positions. I cite particularly the Atomic Energy Commission, Central Intelligence Agency, State Department Foreign Service, and the National Science Foundation, all of whom have tasks which are similar to one portion of the mission of the Department of Defense.

We, therefore, made a very strong recommendation, Mr. Chairman, that the present ceiling on the number of top-level positions in the Department of Defense either be removed or at least be lifted to a reasonable level. This would really have potent effect on the ability of the Defense Department to attract new talent and to keep new talent, because if there is no room to rise to the top, the good man just plain is not going to stay.

Now our third proposal was that an interim pay adjustment be made pending the time when the joint commission which we recommended should make its report. We are very well aware of the complexity of the compensation systems in Government. We started out expecting to make our study and to complete our report in a period of maybe 4 or 5 months. We ended up by taking 2 days under a year, and we were investigating only one segment of the picture. Any commission formed would inevitably take 2 years, perhaps longer, to complete its recommendations. In the meantime, you have all these people leaving. You have turnover increasing, you have a decrease in quality, you have low output both in quantity and quality. We feel that some interim measure is terribly important in order to stop this trend.

Our proposal does not cover the entire gamut of the civil-service scale. We do not do so deliberately because we found no problem at the lower end of the scale, and we knew there was a problem and a very serious one at the higher end of the scale.

First of all, we thought it was imperative to have competitive rates at the entrance level and so our suggestion was instead of a statutory requirement that college graduates be taken into the level of GS-5, that they be permitted to be brought in at the level of GS-7. This in effect legalizes a practice that is going on now because the GS-5 level for anybody who is worth his salt is nothing but a vestibule through which he passes very rapidly to the next higher level.

We felt that the top pay now with respect to administrators, $16,000, was completely inadequate and we suggested that this top level be raised to $19,000. We picked this figure quite deliberately. It is $1,000 below the annual salary of the lowest echelon of Presidential appointees in Defense who get $20,000 a year. It is also the same as the top salary which is payable to engineers and scientists under the authority of Public Law 313, which, as you know, Mr. Chairman, permits a certain number of people to be hired in top scientific positions at a salary which is based upon the man's individual worth.

Then what we tried to do between a competitive entrance level and this $19,000 top ceiling was to arrange a scale which would correspond to proper pay practices as reflected in industry, with material increases at the top end of the scale.

The precise proposal is set out in tabular form (table IV) on page 49 of the report and is shown in graphic form (chart 12) on page 48 of the report. If I may suggest, Mr. Chairman, I think page 48 could well be reproduced in your own record as I would like to refer to it. Senator NEUBERGER. It will be included.

Mr. PRATT. This chart shows a comparison of our proposed pay scale along with the midpoints of the present pay scale and the midpoints of the salary survey referred to earlier in my testimony, Mr. Chairman. You can see that we have restored the balance somewhat between industry and Government pay.

Now, there is no information on this chart as to what top industrial pay is at the equivalent of the civil service grades 16, 17, and 18, because you go way off the chart there and we, of course, were held down by this arbitrary $19,000 top ceiling.

Senator NEUBERGER. Yes.

Mr. PRATT. Which we felt was only realistic. We feel this is a reasonable interim measure. We do not think it is a final solution at all but we feel that something must be done until a long-range solution and complete overhaul of the Government's pay systems can be effected after a select blue ribbon commission makes its report.

One of the important things which our proposed salary scale does is to relieve the compression which presently exists in our salary scale. There has been a tendency to give an even number of dollar increases across the board with the result that the gap percentagewise between the lower people and the highest people has been narrowed and narrowed, so promotions mean less and less. Our interim proposal removes that compression factor as best it can with the $19,000 ceiling. Now, our other recommendations I will not go into. They relate to measures which can be taken to improve the prestige of our top level people in Government and the respect with which they are held by the community. They cover matters of personnel improvement opportunities in the Government service, comparable to those offered by industry. None of these require legislation with one exception, a minor change in the conflict-of-interest laws, which is set out in detail in the report.

We also made recommendations as to recruitment practices. Here. again, many of them can be adopted administratively and I believe will be adopted administratively with the exception that legislation is needed to authorize the reimbursement of travel expenses for people who are brought in for interview or who go to their first station. This. again, corresponds to industry practice.

Now, if I may take just a couple of more minutes to talk about cost. Total cost of the proposals made in the civilian personnel field by the Cordiner Committee amount to about $150 million for the Department of Defense. This, I might note, is four-tenths of 1 percent of the total Department of Defense budget and only three-quarters of 1 percent of the amount of money spent on materiel in the current year. It is interesting to note that the $150 million is about equivalent to the amount of increases expected this year to the blue collar employees who are paid under a different compensation system.

Now, what savings can you get from putting these proposals into effect? It is very hard to document. The Hoover Commission, I understand, estimated that if you could reduce turnover of your top civil servants by 200,000, that you would save

Mr. MAHARAY. Five billion.

Mr. PRATT. Five billion? That sounds pretty high.

Senator NEUBERGER. That is a lot of money.

Mr. PRATT. It is interesting to note that Industrial Relations Counselors prepared a report for industry on this turnover problem and

they figured that it would cost industry anywhere from five to twenty thousand dollars to replace an employee of the type we are talking about, just in terms of training and recruiting expenses and getting them into the system, entirely aside from the loss in experience that you suffer every time you lose an experienced man. That will give you some measure of the improvements that can be made by increasing the stability of the work force.

Another indication of what this means to the Government, to the taxpayer, is to take a look at some of the people who have left and see what they have done. We asked for a documentation of specific individual cases from installations all over the country. We selected more or less at random a dozen of those which are included in our report.

The salary of those 12 people, their aggregate annual salary, was $100,000. Those 12 individuals during the year preceding the time they left the Government saved in 1 year a total of $7,500,000. It is hard to believe, but that figure actually does represent the annual cash savings from measures that those people instituted.

Now, that is the kind of job that is being done by top civil servants who are leaving the Government, and that is the kind of fellow we want to keep in Government.

I think, myself, that the cost of our proposed interim pay raises can be absorbed. I think it a pretty small price to pay for maintaining the high level of management in professional and scientific fields that the Defense Department has been fortunate to have in the past 10 years. Thank you.

Senator NEUBERGER. Thank you very, very much, Mr. Pratt, for this very informative testimony which you have provided us with today. Senator Morton, do you have some questions you would like to ask? Senator MORTON. I think we are all grateful to Mr. Pratt for coming down today. The study of your committee was, as I understand it, confined to the Defense Department; is that not correct?

Mr. PRATT. That is correct.

Senator MORTON. And the problem that you were asked to look into was the turnover and the difficulty in recruitment and holding of specialists, scientists, and people in higher echelons of Government primarily?

Mr. PRATT. Yes, sir.

Senator MORTON. I just wanted to clear that for the record, Mr. Chairman, because we have been, in our hearings, on the Post Office mostly so far and we got into a different area today. I am happy that we did.

Mr. PRATT. Senator, the individuals to whom the report refers are in civil service levels GS-7 and above. Our studies indicated that 99 percent of the civil servants in those grade levels in the Department of Defense were either scientists, engineers, managerial personnel, professional people such as lawyers or accountants, or technicians supporting these professional and scientific people.

Senator NEUBERGER. The one thing along the line of Senator Morton's question, however, if I may just supplement it with this further question, you stressed the desirability of paying these skilled and trained and professional people in Government comparable salaries to what they could receive in private employment in industry. In your opinion, Mr. Pratt, would not the same policy be equally desir