policy would be regarding the purchase of Comsat services. It seemed, in the earlier instance, that the Government's policy formulated in a commonsense way was that bulk traffic would be purchased from commercial sources and certain unique and vital traffic would be performed in Government installations.

Then, in the latest formulation, as I recall, the Defense Department and the Government would utilize Government systems for unique and vital traffic which could not be supplied satisfactorily by commercial sources, something along that line.

Now, is it not a fact that you can contract for any service, even unique and vital services? These could be contracted to Comsat, and conceivably Comsat could design any kind of system the Government put on them; isn't that so?

Mr. O'CONNELL. Well, I think that is extending it further than I would be willing to go.

Mr. ROBACK. Well, suppose, just hypothetically, the Government put a contract obligation on Comsat to devise a tactical system. Why can't they do it? The Government writes any specification it wants, and Comsat goes to the same contractors. The only difference is it funnels through Comsat rather than through Government contracting officers.

Mr. O'CONNELL. Well, let us consider operation on naval vessels, as an example. You might not want to have contract personnel aboard a military vessel or in a theater of operations.

Mr. ROBACK. You may not want to have them, but many vessels have them; all instrumentation ships have them.

Mr. O'CONNELL. If you had no other means at your disposal and did not have or could not rapidly create a military system, you certainly might seek to get service that way.

Mr. ROBACK. What I am trying to find out for the committee is, when can you decide, on the basis of any definition that has been offered to date, whether the service should be supplied by the Government directly or under contract? You said in your testimony yesterday you really did not have any economy basis for comparison. So, what is the basis? Is it doctrinal, merely that we ought to support as much private enterprise in this field as we can?

Mr. O'CONNELL. We have an economic basis. We feel that it is a reasonable assumption that the commercial common carriers can supply the service, in general, more economically than the Government can create special organizations to handle its own communications. Mr. ROBACK. All right. That is an understandable

Mr. O'CONNELL. It is an assumption. It is an unproven assumption, but it is also equally unproven that the Government can do it cheaper.

Mr. ROBACK. I am not assuming either way.

Mr. O'CONNELL. Right.

Mr. ROBACK. I am trying to find out the state of the information. Mr. O'CONNELL. Well, the state of the information has been uncertain for many years. We are very anxious to pin it down and to get an accurate comparable methodology for handling this sort of thing.

Mr. ROBACK. What is the significance

Mr. O'CONNELL. I think it would be very wrong for us to try to pin down and specify precisely and exactly what the criteria were in

all respects, which the Department of Defense would use in seeking service through their own system or through a commercial system. I think that only the Department of Defense can make this ultimate determination. Certain broad policies and guidelines can be furnished, and I think this is as far as my office ought to go.

When it gets to a question of considering all the pros and cons, all the military requirements of a particular and specific situation, latitude, considerable latitude, must, in my opinion, be given to the Department of Defense which has responsibility for military operations.

REQUIREMENTS FOR ADCSP

Mr. ROBACK. The Department of Defense is very sensitive to what you might call national policy issues, and if it thinks there is a firm national policy issue that it buy from Comsat, then obviously that will make considerable and continuing difference as to what kind of system it designs.

For example, when a bidders' conference was held on the ADCSP, the advanced system, some inspired theater commander asked for some television channels, and there was a big flap in town. The idea that the military system might have some television channels created virtually a scandal and-well, I should not say that-I mean, a great

concern.

Mr. O'CONNELL. An awareness.

Mr. ROBACK. A great awareness.

Mr. HOLIFIELD. A deeply felt apprehension.

Mr. ROBACK. Comsat was very much concerned about it, too. So the bidders' conference was reinstructed, and ADCSP emerged as a skinny little fellow in terms of capacity.

Mr. O'CONNELL. Well, that is not my understanding at the present time.

Mr. ROBACK. What is your understanding? Can you enlighten us? Mr. O'CONNELL. Of ADCSP as a skinny fellow? It is a pretty substantial system.

Mr. ROBACK. It was a skinny fellow at the bidders' conference. Maybe it has put on a little flesh.

Mr. O'CONNELL. I do not get the feeling that the Department of Defense is seriously inhibited in putting adequate bandwidth in the ADCSP system.

Mr. ROBACK. And you do not espouse any policy of undue restriction. This is a matter of reason and balance?

Mr. O'CONNELL. That is correct.

Mr. ROBACK. You do not take a doctrinaire approach to the matter. Mr. O'CoNNELL. Absolutely not.

Mr. ROBACK. You do not put any impositions on them in saying, "Well, you cannot do this. It is against policy."

Mr. O'CONNELL. I do not say we could not, or would not, but I do not think we have done so. I do not want to abdicate from any authority to do this sort of thing. We have the next one here

Mr. ROBACK. We will come back to this. Go ahead.

DRAFT POLICY ON GOVERNMENT

Mr. O'CONNELL. You asked for the draft correspondence which we sent to the FCC and to the Department of Defense. I have that

here. Chairman Hyde said he has no objection to furnishing Mr. Strassburg's response to this.

I have also been in touch with the Department of Defense, and they are consulting with the various offices who have a share in considerations of this kind. The DOD will let us know at a later date on its release, but here they are.

Mr. ROBACK. Thank you.

Mr. O'CONNELL. Here, also, is the list of contracts of my office for fiscal year 1966 and, if you wish, a copy of my letter to Chairman Hyde, which does not have the "official use only" stamp on it.

(The documents above-referred to were inserted in the record at appropriate points.)

Mr. ROBACK. Thank you.

Mr. O'CONNELL. Let me see if I have anything else. I think that covers my extensions and compliances with your requests of yesterday.

ODTM-DOD RELATIONSHIPS

Mr. ROBACK. I gathered somewhere in your testimony that your relationship to the Defense Department, as far as policy issues go, left some areas that needed clarification. Is that a proper inference?

Mr. O'CONNELL. Well, it is always difficult to decide in any particular case whether any matter is a policy matter or an operational matter, and some of these matters are by no means clear. I think we are doing very well in reaching understandings with the Department of Defense. I believe our relationships are progressively getting better, and our understandings are improving in this respect.

But I would not want to give the impression that this is not a difficult area in which there can be

Mr. ROBACK. Well, Secretary McNamara acquired Governmentwide responsibility as Executive Agent for the NCS by a memorandum of President Kennedy in August of 1963; is that right? Is the date approximately right?

Mr. O'CONNELL. That is right.

Mr. ROBACK. Now, suppose you had been in office then-I believe your office then was vacant; was it not?

Mr. O'CONNELL. It was.

Mr. ROBACK. So that the President logically turned to a forceful personality, a topnotch administrator, and a major user of communications, and said, "This is your job.”

But you are familiar with the report that we issued not so long ago in which we said that the Defense Department is a major claimant on communications, and, also, in effect, is deciding what the other claimants could have.

Now, that seemed to us the kind of policy issue that ought to repose in your office. Are you familiar with that?

Mr. O'CONNELL. Yes.

Mr. ROBACK. Do you agree with it?

Mr. O'CONNELL. I think that policy does repose in the office. We can get into the act any time that we think that we should.

Mr. ROBACK. Just for example.

Mr. O'CONNELL. For example, at the present time, we seem not to have any disagreements between any of the agencies and the Executive Agent and Manager of the NCS on the long-range plan. We did have

one, which appeared in the 1965 long-range report. This was settled and compromised between GSA and the Department of Defense to the satisfaction of both parties.

Mr. ROBACK. You did not see that report until after it was submitted to the President?

Mr. O'CONNELL. We had discussed it prior to that time. We had discussed the problems involved, and had been in touch with both GSA and the Department of Defense in connection with the issues that were involved in it.

We now have a practice that the Executive Agent will furnish us with a draft copy of the long-range report before he finalizes the plan. So we have a chance to get a look at the plan in advance of its appearing in final form.

Mr. ROBACK. Assuming your office had the resources that it ought to have to carry out all these responsibilities which now are vested in you or assigned to you, should your office be the Executive Agent, logically?

Mr. O'CONNELL. No.

Mr. ROBACK. No. What is the difference? Is it a matter of resources or is it a matter of the nature of the function? Should one of the Government agencies be the Executive Agent?

Mr. O'CONNELL. I think it is quite proper and most effective to have the Department of Defense as the Executive Agent. I believe that my office should be the office of policy without, you might say, the biases or partisanships which could enter into the picture if we were an actual operator of a communications system.

Mr. ROBACK. But you are asking an Executive Agent with a welldefined and broad bias; namely, the Department of Defense, which is biased in favor of Defense interests, and properly so, to make a decision about NASA and everybody else, who shall do what in relation to Comsat. The Executive Agent really makes these decisions. The decision for NASA to go Comsat was as much a policy decision as any policy decision that comes to your office because, not only does it affect the way in which NASA does its business, but it has important consequences to other Government users-obviously, the Defense Department now buying in-and it has consequences for the industry because it affects the structure and the competitive relationship.

Mr. O'CONNELL. That is one which forms a good example of the inextricability of policy from operational considerations. Actually, from the standpoint of policy, we were very much involved in that particular situation.

Mr. ROBACK. If Secretary McNamara had turned down, refused approval to NASA, could you have approved it?

Mr. O'CONNELL. Would you repeat that?

Mr. ROBACK. Suppose the Executive Agent had turned thumbs down. on the NASA proposal. It is conceivable because they asked to review it, and the only purpose in reviewing something, apart from sheer curiosity, is to say yes, or no, or maybe.

Mr. O'CONNELL. In a case like that, if we felt that the policy implications were strong enough for us to act, and there was a technological question, we would call in outside independent experts and pay them to make an objective study for my office of the technical aspects and objections which might be involved. We would then make a decision.

Mr. ROBACK. I would think that General Starbird's office, with a big supporting organization, a contract organization, would be the place where you would want to get technical decisions."

Mr. O'CONNELL. We often go to him.

Mr. ROBACK. You should be making the policy decisions.

Mr. O'CONNELL. That is right. Sometimes these decisions get so mixed, though-particularly in the frequency management area—that you cannot separate them out because

Mr. ROBACK. That is understandable.

Mr. O'CONNELL. Because the decision on the policy is dependent on a technical issue, and you have to get your facts straight on the technical issue before you can make a satisfactory and sound policy.

POLICY UNDER COMSAT ACT

Mr. ROBACK. The NASA decision when it came to your office was not a technical matter, was it?

Mr. O'CONNELL, No.

Mr. ROBACK. It was a policy matter.

Mr. O'CONNELL. The Department of Defense analyzed it from a technical point of view. We accepted their analysis.

Mr. ROBACK. Well, do you think that this is an issue that needs clarification, that is where the policy begins and ends?

Mr. O'CONNELL. I think we are working it out satisfactorily.

Mr. ROBACK. You know, we said in our report that policy with regard to Government operation of systems as alternatives to dealing with Comsat or others, was a policy that the Government had formulated, so far as we could understand it, which was narrower than the law. Are you familiar with that statement?

Mr. O'CONNELL, No.

Mr. ROBACK. We stated that in our report. Do you think that the Government policy is narrower than the law or do you believe, as Defense Department representative Secretary Horwitz appeared to believe, that it was up to the President to decide how much of Comsat's services would be procured?

Excuse me, Mr. O'Connell, you know, I asked you a question before and your statement said that, "We needed to arrive at a common position on policy."

Mr. O'CONNELL. We are doing that now.

Mr. ROBACK. Implying that there is room for improvement.

Mr. O'CONNELL. In my opinion we are reaching an understanding on that right now in this discussion.

Mr. ROBACK. All right. If we are making any contribution, we are happy. [Laughter.]

Mr. O'CONNELL. Is there a question on the floor, here?

Mr. ROBACK. Well, to get a handle on this policy is like trying to put your finger in a goldfish bowl and to grab one of the goldfish.

Mr. O'CONNELL. This is largely a matter of commonsense and good relationships between the agencies involved. I think that as our capabilities improve our relationships are also improving. At the present time, I have nothing to recommend on any changes in specifications as to what and what is not policy.