Read an Excerpt
Systematic Thinking for Social Action
By Alice M. Rivlin Brookings Institution Press
Copyright © 2015 The Brookings Institution
All rights reserved.
ISBN: 978-0-8157-2645-6
CHAPTER 1
INTRODUCTION
The Gaither Lectures, by an unbroken if brief tradition, address the question: How can government make decisions in a more rational way? To care about this question one has to have faith in the ability of nations to solve at least some of their problems by collective action. One also has to have sufficient faith in rationality to believe that analysis of a problem generally leads to a better decision. H. Rowan Gaither believed in governments, rationality, and the ability of people trained in systematic analysis to improve government decision making. Those who have delivered these lectures in his honor share that faith.
Charles Hitch focused the first series of Gaither Lectures on the problem of making better decisions about the national defense. He recounted the history of the fragmented United States defense establishment and showed the importance of central control for the effective use of defense resources. He described the reorganization and the tools through which control was to be achieved—the planning-programming-budgeting system (PPBS) introduced into the Department of Defense in 1961.
Charles Schultze, in the second series of Gaither Lectures, widened the discussion to include the whole domestic side of the federal government. He contrasted PPBS with an alternative decision-making model, which he called, after Lindblom, "muddling through," and, analyzing the politics of decision making, demonstrated how the system is compatible with a democratic political process.
Both Hitch and Schultze argued cogently that government should make decisions as systematically as possible—arraying alternative policies, assembling information on the advantages and disadvantages of each, and estimating the costs and benefits of public action. From their analyses of such tools of systematic decision making as program budgets, multiyear plans, and program memoranda, two major messages come through: (1) It is better to have some idea where you are going than to fly blind; and (2) it is better to be orderly than haphazard about decision making.
Neither Hitch nor Schultze, however, overstated the case for system or analysis. Both recognized the important role of judgment and values in the decision-making process. As Hitch put it, "Systems analysis is simply a method to get before the decisionmaker the relevant data, organized in a way most useful to him."
This third series of lectures will continue this discussion of how to make government decisions more rationally, but will change the focus in two ways. First, its emphasis will be on a particular set of what I call "social action" programs—education, health, manpower training, and income maintenance and various other efforts to alleviate poverty. These are human investment, or "people," programs, designed to help individuals function better, that at the federal level fall under the jurisdiction of the Department of Health, Education, and Welfare, the Office of Economic Opportunity, and parts of the Department of Labor.
Second, I will concentrate on the substance, not the process, of decision making. No purpose would be served by a reexamination of PPBS or reiteration of the case for systematic analysis of social action programs. That case has been ably made and widely accepted. By now the analysts have been on the scene for several years and it is time to evaluate their contribution. With what kinds of decisions have they proved helpful? Where have they been unable to help? What should be done to make systematic analysis more useful? These lectures will attempt to give a midterm report card on the contributions of analysis to decisions concerning social action.
PPBS AND COMMON SENSE
Despite its elaborate terminology, PPBS seems to me simply a commonsense approach to decision making. Anyone faced with the problem of running a government program, or, indeed, any large organization, would want to take these steps to assure a good job: (1) Define the objectives of the organization as clearly as possible; (2) find out what the money was being spent for and what was being accomplished; (3) define alternative policies for the future and collect as much information as possible about what each would cost and what it would do; (4) set up a systematic procedure for bringing the relevant information together at the time the decisions were to be made. PPBS was simply an attempt to institutionalize this common-sense approach in the government budgeting process. It was not the first such attempt and it will not be the last.
The tools and the terminology may change, but the approach to decision making implicit in PPBS has largely, I think, been accepted in Washington, in principle if not always in practice. It is regarded as a desirable way to make decisions—if the time and information are available. Hardly anyone explicitly favors a return to muddling through.
Indeed, PPBS is only one manifestation of a quiet revolution in the government in the last few years: The level of discussion of major decisions has gone up. The result is reflected in the questions decision makers ask about new programs, as well as those already in effect. What will it do? Why do we need it? What does it cost? They do not get very good answers yet, but they keep asking, and the standards of staff work are rising. The quiet revolution is also reflected in the acceptance of analysis as part of the decision process and of the analyst as a participant at the decision table. No one demands of the analyst, "Why are you here?" Now they ask, "What have you got to contribute?" Sometimes he has a lot to say and sometimes he is very quiet.
In the process, the mystique has gone out of planning and systems analysis. The practitioners themselves, in fact, never invoked as much mystique as the nonpractitioners alleged. Moreover, a touch of mystique was probably useful. A bit of bravado is necessary to overcome the inertia of government, to get attention, and to win a place at the decision table. Now, however, educators, doctors, and ordinary civil servants realize that systems analysts do not have pointy heads, that they can be helpful and sometimes even right. The analysts in turn have recognized that educators, doctors, civil servants—even generals—are knowledgeable, necessary, and not always wrong.
Moreover, the tools of the trade have become more familiar and thus less frightening. As recently as five or six years ago the average administrator viewed computers as somewhat menacing new instruments. In his Gaither Lectures, Hitch felt called upon to say, "Let it be said, here and now, that computers do not make decisions ..." That is certainly correct, but it already sounds dated. Decision makers no longer need to be reassured that they are not about to be put out of business by a computer.
The cockiness of systems analysts has disappeared with the mystique. If any analyst thought it was going to be easy to make social action programs work better or to make more rational choices among programs, he is by now a sadder and a wiser man. The choices are genuinely hard and the problems are extraordinarily complex and difficult. It is hard to design an income maintenance system that will both assure adequate incomes to the needy and encourage people to work, or a health financing system that will both assure proper care to the sick and encourage efficient use of health resources. It is hard to decide how the government should allocate its resources among different kinds of social action programs. So far the analysts have probably done more to reveal how difficult the problems and choices are than to make the decisions easier.
A WORD ABOUT PPBS IN HEW
These lectures will not evaluate PPBS as a set of procedures, but a word about the role of PPBS in government decision making is perhaps in order. In the Department of Health, Education, and Welfare, its most important effect was the creation of an analytical staff at the department level, which brought into the secretary's office a group of people who were trained to think analytically and whose job it was to improve the process of decision making. Although I am not an unbiased judge, I think the small planning and evaluation staff in HEW has accomplished an astonishing amount since it was created in late 1965. It has developed a program budget and information system that, while far from perfect, gives the secretary a better idea of how department funds are being spent, of what they are buying and for whom. It has started evaluation studies to try to measure the impact of departmental programs. The five-year plan for the department it has developed and periodically revised forces a lot of people to think harder about the objectives of programs and their future directions. It has produced analytical work that has had considerable impact on major policy choices. Finally, but perhaps most important, it has helped create a regular process for bringing analysis to bear on budgetary and legislative decisions, and it has established the voice of the analysts and planners in the decision process. (The voice is important; decision makers rarely have time to read!) Indeed, some of the procedural steps taken in the department under the impetus of PPBS seem so obviously useful that it is hard to remember that they are so new.
But now the process exists. Missionaries need no longer be sent to convert the heathen to the virtues of systematic analysis. If the analysts have something relevant and useful to say they will be listened to. Now that they have the floor, what do they have to say?
FOUR PROPOSITIONS
Anyone who makes decisions about social action programs—a congressman, the Secretary of HEW, the Governor of Nebraska, or a concerned voter—would want answers to such questions as these:
1. How do we define the problems, and how are they distributed? Who is poor or sick or inadequately educated?
2. Who would be helped by specific social action programs, and how much?
3. What would do the most good? How do the benefits of different kinds of programs compare?
4. How can particular kinds of social services be produced most effectively?
In the last few years, economists, statisticians, and other analysts have worked on all of these sets of problems, with results that are highly uneven. Four propositions, two positive and two negative, sum up the progress so far. Chapter 2 discusses the first two.
The first proposition is positive. Considerable progress has been made in identifying and measuring social problems in our society. Much more is known about who is poor or sick or badly educated, and this knowledge itself has helped clarify policy choices.
The second proposition is also positive. Systematic analysis has improved our knowledge of the distribution of the initial costs and benefits of social action programs. Much more is known about who wins and who loses.
The third proposition is negative. Little progress has been made in comparing the benefits of different social action programs. It is not possible, for example, to say whether it would do society more good to cure cancer or to teach poor children to read. I do not think this situation is temporary or that it matters much, for reasons explained in Chapter 3.
The fourth proposition, which is more important, is also negative. Little is known about how to produce more effective health, education, and other social services. Unfortunately, moreover, neither social service systems nor federal programs are organized to find out. Chapter 4 discusses this dismal state of affairs and the reasons for it.
If these propositions are correct, what can be done to improve the situation? Chapter 5 discusses the potentialities of social experimentation and makes the case for judicious use of this new technique to improve the effectiveness of social services.
But even if we knew how to produce more effective services, how would we insure that these methods were actually used? How can producers of social services be induced to do a more effective job? How can they be held accountable to the taxpayers and to the communities they serve? Chapter 6 discusses the concept of "accountability," what it might mean, and how it might work.
CHAPTER 2
WHO WINS AND WHO LOSES?
The first step in making public policy is to get a picture of what the problem is. Social action programs should be based on answers to questions like these: How many people are poor? Who are they? Where are they? Why are they poor? Who is in bad health? Who is not receiving treatment? How many people need more education or better job skills? How are these social problems related to each other?
Statisticians and analysts have worked hard on these questions in the last few years. For their efforts they get high marks: We know a lot more than we did about the distribution of social problems in our society. We have much more detailed and usable knowledge about who is poor or sick or inadequately educated. This knowledge has exploded some myths and thrown a good deal of light on policy choices.
WHAT ARE THE PROBLEMS?
The distribution of social problems has been illuminated by two important technical developments. The first is the improvement and wider use of sample survey techniques. The second is the astonishing increase in the data processing capacity of computers.
Surveys, of course, are not new. A census has been taken every decade since 1790 and social and economic characteristics of the population have been tabulated for some time. Until recently, however, census and other survey information was of limited usefulness to those interested in the interrelations of social characteristics. Data were usually displayed in tables in which only two or three characteristics could be considered at once. It was not possible to establish the simultaneous distribution of, or relation among, a large number of characteristics—say, age, race, sex, income, work experience, education, region, and number of children—either because the sample was too small or because the computation was too involved and expensive.
In late 1963, when the launching of the war on poverty was under consideration, those who made the decision had only the vaguest notions about how many people were poor and who they were. The Council of Economic Advisers picked $3,000 as the "poverty line" for family income even though everyone, including the council, was dissatisfied with the criterion. Quite apart from the question of its level, it did not allow for the facts that larger families need higher incomes, that it costs more to live in a city than on a farm, and that prices rise over time. At that time, however, a more sophisticated definition could not have been used even if it had been available. No technical capability existed for estimating quickly from the census, or any other source, the characteristics of the poor according to more complex definitions.
Now, less than eight years later, the situation has entirely changed. More sensitive and useful definitions of poverty, which take into account family size, place of residence, and other characteristics, have been developed. A special census survey, the Survey of Economic Opportunity (SEO), has been designed to yield more detailed and more accurate information about low-income groups. From this and other surveys, masses of information have been collected, detailing the characteristics and whereabouts of the poor. One can now literally punch a button and find out how many of the poor, by any appropriate definition, are black or aged or have six children or work full-time or live in cities. This does not solve the problem of poverty, but it certainly helps in establishing what and where the problem is.
Making the survey data useful has involved ingenuity and hard work, because in raw form they leave much to be desired. Certain kinds of people—especially the poor and men in their twenties—tend to be missed or fail to respond to surveys. Certain kinds of questions are often answered inaccurately or not at all. On the average, people underreport their incomes. There is evidence, for example, of considerable underreporting of transfer income (social security benefits, welfare payments, and the like) on the Survey of Economic Opportunity. Many manhours have gone into improving survey accuracy, estimating missing data, devising ways of merging different files of data on the same people, and otherwise improving the consistency and usefulness of survey information.
Better statistics on poverty have influenced the way people think about the problem and have dispelled some myths and false impressions. For example, the first analyses of poverty based on the Council of Economic Advisers' $3,000 definition created the impression that a very high proportion of the poor are aged. More refined criteria, such as that of the Social Security Administration, which allows variations by family size in the definition of poverty income, show the proportion of aged persons in the poor population to be considerably lower. It is lower still if an allowance is made for assets as well as income, since older persons are much more likely than the young to own their own homes or to have savings. While the plight of the aged poor should not be minimized, more refined definitions and statistics on poverty have rightly tended to dramatize the fact that a high proportion of the poor are children and relatively young heads of families.
The popular myth that most of the poor are black mothers with lots of children living in big cities also disappears quickly when the numbers are examined. Although many of the poor are, indeed, black mothers with several children, most are white and over half of all poor families have male heads. This distribution appears even in analyses using the Social Security Administration definition of poverty (about $3,700 for a family of four), and the proportions of white families and of those with male heads rise as the poverty line is set higher.
Another myth that will not stand up to the statistics is that poverty is largely a problem of people who cannot or will not work. About 55 percent of all poor families in 1969 had members who worked full- or part-time and almost a third were headed by males who worked full-time. Sheer numbers are persuasive. The statistics dramatized the plight of the working poor and focused attention on the inadequacy of an income-transfer system designed to aid only those who could not work.
(Continues...)
Excerpted from Systematic Thinking for Social Action by Alice M. Rivlin. Copyright © 2015 The Brookings Institution. Excerpted by permission of Brookings Institution Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
