the Office of Evaluation which is in the Department of Health, Education, and Welfare or in HUD or other departments, but gives you an independent source of information?

Now I certainly think that is an excellent idea because one of the principal conclusions of my theory of bureaucracy is that one way to increase the effectiveness of control at the top is redundancy. That means having multiple channels of information up from the bottom to the top, multiple ways of looking at the bottom.

Senator HARRIS. It seems to me that one of the reasons why you don't fix the responsibility very well is that decisions are being made all across the board by all sorts of people in various agencies. Now, what about something like this

Mr. Downs. May I say one more thing?

Senator HARRIS. Yes, please.

Senator RIBICOFF. If the Senator would yield, may I say we are at the present time drafting a bill for an Office of Legislative Evaluation.

PLACEMENT OF OFFICE OF LEGISLATIVE EVALUATION

Mr. Downs. But don't put it in the Government Accounting Office. Senator RIBICOFF. What is that?

Mr. Downs. Don't put it in the Government Accounting Office. Senator RIBICOFF. I don't know where we are going to put it yet. Mr. Downs. Well, don't put it in the Government Accounting Office because then you are committing the very error which you are trying to react against. You are attaching a small innovative agency to a large on-going bureaucracy which in my opinion has a very narrowminded view of its function. Its function is to keep people honest. That is an excellent objective and it has done a wonderful job in accomplishing that. But don't put the Office of Program Evaluation in there, or you are going to have exactly the same kind of dominance of that agency by people who don't really want to perform that function that you and I think is wrong in other agencies.

Senator RIBICOFF. I appreciate your point. I just want to say to you that much of the contribution being made, if the Senator will yieldSenator HARRIS. Yes.

Senator RIBICOFF. Many of the contributions that are being made by the witnesses are very, very valuable, and I have recognized for a long time that not all the brains in America are in Washington, D.C. I recognize that there is a large reservoir of intelligence and ideas throughout our Nation. These hearings have proved that.

Many of the bills that we introduce come from men like yourself who came here with some commonsense approaches to solve some of our programs. Mr. Moynihan proposed an Office of Legislative Evaluation. I think Mr. Moynihan made a very, very good suggestion. I also appreciate and welcome your comments. I apologize for the interruption, Senator Harris.

COORDINATION OF DOMESTIC PROGRAMS

Senator HARRIS. Thank you, Mr. Chairman. I wonder, though, if we are still being innovative enough. I think the central political science problem of our day is how to coordinate and get maximum re

sults out of these myriad and sometimes overlapping programs that we have put on the books in the last years, and the coordination between them and local and State effort in the same general fields.

For example, in the field of foreign relations, we had this same kind of fragmentation of authority in our representation in foreign countries, and a subcommittee of this same parent committee, headed by Senator Jackson, did a study called "The Secretary and the Ambassador." Out of that grew, with the inauguration of President Kennedy, the "country team" concept, where now everybody comes under the authority of the ambassador.

Now, I understand that that is difficult to do, but it was done in that smaller area. Now, if we can do it in a foreign country, if we can make all our programs go under some kind of central authority in a foreign country, why can't we do it in a place like New York City or Chicago or some of these more rural but regional areas, which have the same kind of general problems?

Mr. Downs. Well, there are several problems. In the first place, you could do it in a city like New York City or Chicago in terms of placing the authority and the power under the mayor. But if you create another official, a Federal official, who is the sort of Federal czar of programs in Chicago or New York it seems to me you are creating a direct rival of local government. I would be opposed to that because I think the problem is to improve the power of the local government, not to weaken it. Now the local government may be organized in such a way as it is in New York

Senator HARRIS. I am not sure whether you want to turn it all over. I am not sure that Congress would be responding to its duty in regard to the expenditure of this money, if you turned it over completely to the mayors of the country.

LOCAL DISTRIBUTION OF FEDERAL MONEY

Mr. Downs. No; I don't mean a block grant to mayors, but I think that you would have to do several things. If you could get the mayors

to exercise more centralized control over the coordination of Federal programs, and the Governors in their States, to do so, then that is the appropriate route to go, in my opinion, rather than trying to set up Federal coordinators in those areas.

The second thing you could do is merge a lot of grants, but not eliminate the idea of the special grant. As a mayor of a large city told me recently. "Don't give us block grants," he said, "because all the good ideas that have come into city programs in the last 20 years, have not come from local government, and they haven't come from the State government; they have come from the Federal Government. It is only when the money is attached to a certain set of things we have to do that we actually do anything new that is worth while, instead of spending the money for more streets or something else. Otherwise we would just fritter it away and we wouldn't innovate at all."

So I agree with you we shouldn't eliminate the congressional strings on what is done with the money. But instead of having 247 different packages, maybe we could have 20 different packages, with each one

containing a little bit broader mandate. This might greatly ease coordination problems, if used to supplement and work through the mayor, who is the equivalent of the ambassador in the countries that you are referring to.

Now, one of the difficulties is that the mayor has a much bigger empire. He has a lot bigger country than any ambassador does in terms of the number of people who are working for him. Mayor Lindsay in New York or Mayor Daley in Chicago have large organizations. They dwarf anything which any ambassador has under him except in Vietnam.

CENTRALIZED METROPOLITAN PLANNING RECOMMENDED

Senator HARRIS. Just take something like this. Almost any program that is applicable in a city requires a plan. You must have a plan, and the plan must be approved locally, and then it must meet certain Federal standards to be approved, but they are all separate plans. If you are going to come in and do something about open spaces, why that is one plan. If you are going to do something about air and water pollution, those are two separate plans. If you are going to do something in housing that is another plan, and urban renewal and so forth, that is another plan.

Well, it just seems to me that that necessarily diffuses the authority and confuses the people involved, that you just cannot afford it any more if you are going to have any kind of concentrated attack on some of these urban problems. Mr. Downs, you could argue that, if you want to accept the metropolitan area as the scale on which you want to coordinate these plans, the Federal Government could, if Congress would go along with it, use the power of financing these different programs and plans to evoke some kind of metropolitan decisionmaking. I won't call it metropolitan government, because it would not replace most of the activities of smaller communities. But it would be tralized metropolitan decisionmaking force, where these plans would have to be coordinated.

cen

Isn't that a problem with your mayor concept, that often you have a good many of the small municipalities involved?

GOVERNMENTAL AUTHORITY MUST SUPPORT METROPOLITAN PLAN

Mr. Downs. Yes, in an area where a program operates on a metropolitanwide scale, that makes more sense. But then Congress has to have the will to do that. Once a metropolitan coordinating authority or decisionmaking group gets into existence, it has to be supported by somebody. It can take its support from the Federal level, if the Congress attaches strings to funds saying that not only does there have to be a sewer and water plan but it has to be developed by a certain kind of governmental agency which has to have power over certain other kinds of plans too. Then we might move toward this coordination of planning for metropolitanwide functions.

Senator HARRIS. Thank you, Mr. Downs, for a very stimulating discussion. Thank you.

Senator RIBICOFF. Senator Baker?

SMALLER EXPERIMENTAL PROJECTS ARE A POSSIBILITY

Senator BAKER. Mr. Downs, I want to say at the outset that I am much taken with all of your presentation, but especially with one part which I believe forms the basis for the balance of your proposals. That is the proposition that in the course of the National Government's devotion to solving problems in urban areas, and eradicating the problems of poverty or the disadvantages, that apparently we have tended to find a program and adapt it to nationwide application instead of following its traditional time-tried and true scientific formula of devising a laboratory experiment, observing the reaction and modifying the next experiment.

In that connection let me ask you this to clarify my comprehension of what you have said. Would you propose that social experimentation along these lines, whether in urban or rural areas, would be undertaken on a statewide basis, a regional basis, a citywide basis, or might it in fact be undertaken on a neighborhood basis, like an area as small as 16 or 20 square blocks?

Mr. Downs. I think that depends greatly on the nature of the particular program being experimented with. In some cases the neighborhood would be very appropriate. In other cases a metropolitan area or a State would be more appropriate. It depends on the kind of program involved.

NATIONAL EXPERIMENTS CANNOT BE OVERLOOKED

For example, if you were interested in some kind of a crime prevention program, that part of crime prevention which involves a patrol in the streets could be conducted in a very small area. However, that part which involves improving the parole and prison system might have to be conducted in a wider area because of the way that particular system is structured. So I don't have any preconceived notion about the scale of the kinds of experimenting that should be done. It depends on the program.

Let me also say that I do not mean to imply that we should never tackle a problem head-on with immediate gearing up for operating programs because we want to wait around and see how the experiments come out. Where we have a strong national desire to do some thing right away, and to do at least those things which can be identi fied as possibly effective right away on a national scale, as we did with the Office of Economic Opportunity, I don't mean that we should stop doing everything until we have had a 10-year experiment to see whether it is really as productive as we think it should be, not at all.

VALIDITY OF LABORATORY CONCEPT OF EXPERIMENTATION

Senator BAKER. There is another basic consideration, however, and that is that no single solution, as a matter of fact no single formula or theory, is likely to solve all the problems in all of the areas of this Nation.

Mr. Downs. That is exactly right.

Senator BAKER. And isn't a part of the validity of the experimental or laboratory concept of this proposal, that you can measure not only

what will work and what won't work in this way, and you can also find out what works better where?

Mr. Downs. If you do it under enough of a variety of conditions, yes, you can do that.

RESULTS OF EXPERIMENTS CAN BE REGIONALLY ADAPTED

Senator BAKER. And on a small laboratory basis, we cannot only judge the probable validity and accuracy of the theory involved or the proposal involved, but we probably can adapt it on a regional basis to peculiar and unique concerns of that neighborhood or that city or that region.

Mr. Downs. Yes, and obviously this is closely related to the concept of the model city program. But the model city program, in my opinion, is handicapped in this regard by the rather stringent restrictions on how much experimentation of what type would be done. These restrictions are in the legislation, and therefore are expressed in the guidelines that have been sent to potential model cities.

REASONS FOR SEPARATE INNOVATIVE AGENCY

Senator BAKER. And this is a concern which leads you to the recommendation that we have a separate semi-independent innovative agency that is not directly related to present bureaucratic or administrative procedures?

Mr. Downs. That is one reason. Another reason is that the model cities program leaves the innovation to existing agencies. Take the schools, for example. It is the school board or the school operating agency in each city which is supposed to come up with the innovations on how to operate schools in the model cities program.

Now, many school boards may come up with some pretty good ideas. But in my opinion you are not going to get as wide a spectrum of possibilities tested if you rely on existing agencies who are operating programs as if you say "You, Mr. So-and-so, are in charge of this creative agency. It is your job to think of new ideas, ranging from obviously practical to wild and apparently impractical, and test them. That is your function."

With this approach, we are going to get a much bigger spectrum of things tested in my opinion, than if we rely on the operating head of an agency 99 percent of whose budget is in carrying out a certain program, and ask him with the other 1 percent of his attention to invent something new. I just don't think we will get the same range of things tested.

Senator BAKER. To try and pull all this together for my own comprehension and understanding, would it occur to you and would you agree with the proposition that one of the principal functions of the Federal Government might be to examine and understand the nature, variety, and complexity of the problems, both nationwide and from area to area?

Mr. Downs. Yes, I certainly would.

Senator BAKER. And to test these new proposals and innovations on a scientific basis, with some expectation of proposals that would apply in different areas with a different impact?