If it is somewhat difficult to do this kind of work within a university setting then it might be asked: Why should policy research and analysis be connected with a university? What are the benefits to policy research and to the university from such a connection?

I think probably the most important thing is, as seems to be very clearly the case, there is this need for trained personnel and the logical place to train people is within the university. And it is difficult to train people unless you are actually doing this kind of work within the university.

The second thing is that major academic institutions and particularly State universities supported by public funds have a responsibility to contribute to more effective public policy.

A third thing that at least entered into our decisions in thinking about the policy research organization was that it seemed that a university setting might be able to minimize some of the problems that the nonprofit corporations experienced. If, from the point of view of policy research, the traditional university institute is tied too closely to the academic structure, then it could be said that, on the other hand, the nonprofit corporations have sometimes suffered from being tied too closely to governmental agencies. The nonprofits have had some problems in attaining and maintaining both independence and significance for this reason.

Government agencies are bound to exert some pressures on organizations that are doing research for them; these pressures will be greater if the research organization is wholly supported by one Government agency. Many of these pressures are quite legitimate; the agency needs some help on a problem and hopes to get that help from its research organization. Nevertheless, some of these pressures have to be resisted by a research organization if it is to maintain the quality and independence of its work. Certainly those pressures to come up with conclusions and recommendations supportive of the preexisting views or positions of the client agency fall within that category; so do some of the pressures to do short-term, rush projects to meet immediate needs.

While it is important for a policy research organization to work closely and cooperatively with Government agencies, one of the great values to a client of an outside research organization is its independence and its attention to long-term and fundamental problems. Client agencies confronted with crises from day to day often forget this, and a research organization too closely tied to a client may also be prone to forget it from time to time.

An additional liability that some of the nonprofit corporations have had is that of being relatively isolated from any larger intellectual community. It should be to the benefit of both policy research and basic research that they not become so insulated from each other that reciprocal influences cannot be exerted.

A fourth factor suggesting a university connection is that perhaps the best way of assuring the increased use and further development of systems analytic techniques and related approaches is to securely link this field to the institutions of higher education. Here, I think, it is not just a matter of training people but of increasing the acceptance and legitimacy of the field.

Now to the question you raised earlier about the structure. As I mentioned, it is a hybrid structure, in some ways it looks like a nonprofit corporation and in other ways like a university institute.

It looks like a nonprofit corporation in that it will have its own full-time research staff; in selecting research programs on the basis of their relevance and utility to the solution of current and future public problems; in doing interdisciplinary work involving the physical sciences, engineering, and the social sciences; in developing a research strategy that will pay attention to the rhythms and needs of government as well as to the rhythms of academia; and in being concerned about developing the in-house capabilities of client agencies it works with.

It will look more like a university institute in having a relatively small full-time research staff and in drawing flexibly on faculty resources, particularly the Irvine faculty, but also the other campuses of the University of California; in maintaining close contact with the academic departments; in contributing to formal educational and training programs; and in concentrating on fundamental problems and the development of the state of the art.

Does that answer your earlier question as to the structure?

Senator NELSON. Not that it is pertinent to these hearings, but, is it still an institution within the jurisdiction of the management of the university?

Miss ARCHIBALD. Very much so, yes.

We will pay some particular attention to problems at the State and local levels. California, with its size and growth rate, provides an ideal social laboratory. Also, our location in Orange County means that we are in the middle of a new and rapidly developing urban complex.

We are fortunate in California in having considerable interest in and experience with research and analysis in both the executive and legislative branches of the State government. The further development of this interest and expertise would seem to be important to all parties concerned.

One way of developing both the in-house capability of the policymaking agency and the policy sensitivity of the researchers and analysts is to work out a system whereby the analysts spend some time in government agencies and government personnel spend some time in the research organization. We hope to be able to work out such an exchange and we are currently exploring the potentialities here.

During your hearings in May, both Mr. Enthoven and Mr. Rowen mentioned that we do not yet have a clear idea how best to train people for systems analysis. Since Mr. Enthoven and Mr. Rowen are probably the two people in the country who have given most thought to such training and have had most experience with setting up programs, they have indeed pointed out a serious problem.

Public Policy Research Organization expects to start a training program on a modest scale within the next 2 years. We plan first to take a careful look at some of the training programs that are now in existence, and to learn from their experinece and, as well, to pay some attention to the training function that Rand, in fact, has performed so well for many years even though it has not been a formal training institution.

There seems to be, as of now, general agreement on two of the ingredients needed for training: (1) a strong and quantitatively oriented training, perhaps not solely in one discipline but in relevant aspects of several disciplines; (2) involvement in the actual conduct of systems analysis and related research.

A third possible ingredient would be some experience in working in a governmental agency. Even if these are the right ingredients, we still do not have a recipe that tells us how to mix them or in what amounts. When should a student start on such a program? At the beginning of his graduate career, in the midst of it, or perhaps at the postdoctoral level? How much time would be required? Can the time be shortened by use of such devices as case studies, possibly simulation, and other such techniques?

These are important questions since the successful use of systems analysis at the State and the local levels is dependent upon the availability of training both for people already within government and for younger people who will become those to be hired by State and local governments in future years.

I am pleased that the discussions on this bill are helping to shed some light on the problems in this area.

Senator NELSON. I take it one of the key functions of your institute will be to train systems analysts?

Miss ARCHIBALD. Right. We will start the research programs first; however, since that is our second function and it should be established before the training program. A little bit before, a year or so.

Senator NELSON. What have you projected as your objective in terms of numbers that you will train annually?

Miss ARCHIBALD. We have not projected that figure yet. I would think we would start relatively modestly. The University of California at Irvine is a new campus, we are only in our second year. I would think the growth rate in systems analysis would be rather similar to the general growth rate. I would think a beginning in this are would be about five or six students.

Senator NELSON. You will be training analysts in any and all disciplines in that area.

Miss ARCHIBALD. Right, and we also hope both to provide training for degree students who want to get-I don't know what the degree would be in—but who want to get a degree program, and also for people who have been out working in government agencies and want to come back for some training.

And another thing that has interested us as well: the interest in foreign countries in getting training in systems analysis is increasing and we would like to make our programs open for people from abroad as well as to American citizens.

Senator NELSON. Will your capacity to train analysts depend upon the number of contracts you have, or will you be running a separate academic program along with your contracting work?

Miss ARCHIBALD. Well, the academic program will be very much in conjunction with the academic departments, and the training function would not be supported out of contracts.

Senator NELSON. Will not?

Miss ARCHIBALD. Will not be supported out of contracts.

Although it will be important to have a strong ongoing research program before we get too much involved in training functions.

Senator NELSON. Do you have any views on what is the best way in which the Congress could draft a proposal that would help implement the utilization concept of systems engineering by local, State, and regional, and other governments around the country?

We have two bills before us, I do not think either one of them is the

answer.

The one I drafted simply aims at following roughly the policy of 701 grants to be made by the Congress for planning purposes- regional, city planning, and so forth. I am sure there are better ways. The other one establishes a national commission which would evaluate the problem.

The question is: What would be the best legislation, perhaps the best method to adopt when Congress is prepared to do so, to give grants and aid in the most effective fashion to induce the use of the systems engineering techniques at various levels of government?

Miss ARCHIBALD. I have only read the first of the bills you mentioned, the bill you drafted. It seems to me important, in terms of longrun purposes, to think about ways to get the most mileage out of the money, to use the first grants to prime the pump in order to get continuing programs and continuing interest at the State level.

In large part I would agree with Professor Geyer's comment that one of the key things will be to have a body of experts within the State or local government, monitoring work done outside the government and doing in-house work. This development of in-house capability at the State and local level seems very important.

I think I might disagree with Professor Geyer on his suggestion not to worry too much about institutional arrangements. This may come out of my discipline. I am a sociologist and I think institutions are important. It is not just a matter of individuals being interested; it is also a matter of institutional arrangements that channel, develop, and mobilize their interests.

I think the key one of those institutions is the one within the user agent having in-house capability.

Related to that would be not trying to scatter the money too thinly among the 50 States, which I think you do not plan to do. It might be possible to have some criteria concerning longrun impact built into the process used to select grantees. Organizations submitting proposals could be asked to pay explicit attention to the way their project would have longrun impact on systems analysis. This would turn the worry about this question over to those organizations who are actually doing systems analysis. That might be a possibility.

Senator NELSON. Professor Geyer thought it would be feasible to provide some training in academic institutions in a brief period of time, maybe 3 months, for example, to budget analysts and administrators in various levels of governments-city, State, and local-to give them some good idea of the concept of systems engineering and some idea of what tools are available so that they would then be in a position to do something about implementing such a system even if they were not analysts themselves.

Miss ARCHIBALD. I think that is an excellent idea. The reason we had thought about making training programs available to professional people already working within public agencies was because this was one of the very great needs. Such a training program would have to meet their requirements and might have to be fairly short.

Senator NELSON. Do I remember correctly from your testimony that you would intend to invite exchanges between your own personnel to work in Government or private areas and vice versa for the exchange of orders?

Miss ARCHIBALD. Yes. This is not an ongoing program. We just are exploring it now, but this one we think has a great deal of advantage to it. This would certainly be tied into a training program

as well.

Senator NELSON. I appreciate very much your coming here today and your very fine testimony.

Thank you.

Miss ARCHIBALD. Thank you very much, Senator.

Senator NELSON. Our next witness was to be Leonard Woodcock, vice president of the International Union of the United Auto Workers, and he is grounded in Detroit where they had some snow.

Mr. Frank Wallick is here from the United Auto Workers, on the legislative side.

Frank, would you like to present Mr. Woodcock's testimony?

STATEMENT OF FRANK WALLICK, WASHINGTON LEGISLATIVE REPRESENTATIVE, UNITED AUTOMOBILE, AEROSPACE, & AGRICULTURAL IMPLEMENT WORKERS OF AMERICA, AFL-CIO, ON BEHALF OF LEONARD WOODCOCK, VICE PRESIDENT, UNITED AUTOMOBILE, AEROSPACE, & AGRICULTURAL IMPLEMENT WORKERS OF AMERICA, AFL-CIO

Mr. WALLICK. Senator, I am sorry that systems engineering has not found a better way to get from Detroit to Washington in a snowstorm. Mr. Woodcock was anxious to be here and I have discussed the testimony with him a great deal and I know that he regrets that he cannot be here.

I am Frank Wallick, the Washington legislative representative of the United Automobile, Aerospace, & Agricultural Implement Workers of America, AFL-CIÓ, and I would like to present Mr. Woodcock's

statement.

On behalf of our organization, let me say first we do support the objectives of the Scientific Manpower Utilization Act of 1967, S. 430. One of Mr. Woodcock's responsibilities as an officer of the UAW is to represent our membership in the aerospace industries, some of which have expressed interest in and support for this program.

We in the UAW believe that systems analysis, systems engineering, and systems management can give the American people new insights, new ideas, new directions, and new hope in grappling with many of the perplexing problems of our time.

At the risk of being trite, if our technology can hurtle a man through space around the world in 90 minutes, we can and must find better and easier ways to get to and from work, design and build cheaper and better homes to live in, rebuild our cities, clean up our air and water, reduce crime and rehabilitate criminals, and improve the quality of life for all Americans.

Senator NELSON. We might even be able to make microphone systems work.

76-510-67-13