Are Government Organizations Immortal?
Total Page:16
File Type:pdf, Size:1020Kb
Are Government HERBERT KAUFMAN THE BROOKINGS INSTITUTION ARE GOVERNMENT ORGANIZATIONS IMMORTAL? ARE GOVERNMENT ORGANIZATIONS IMMORTAL? Herbert Kaufman THE BROOKINGS INSTITUTION Washington, D.C. Copyright © igj6 by THE BROOKINGS INSTITUTION 7775 Massachusetts Avenue, N.W., Washington, D.C. 20036 Library of Congress Cataloging in Publication Data: Kaufman, Herbert, 1922- Are government organizations immortal? Includes bibliographical references. 1. United States—Executive departments. 2. United States—Politics and government—20th century. I. Title. JK2I.K37 353.04 75-43466 ISBN O-8157-4839-6 23456789 Board of Trustees the brookings institution is an inde- Ill Robert V. Roosa pendent organization devoted to nonpartisan Chairman ill research, education, and publication in eco- nomics, government, foreign policy, and the social Louis W. Cabot sciences generally. Its principal Chairman, purposes are to aid in Executive Committee the development of sound public policies and to pro- mote public understanding of issues of national Vincent M. Barnett, Jr. importance. Lucy Wilson Benson The Institution was founded on December 8, 1927, Biggs Barton M. to merge the activities of the Institute for Government Edward W. Carter Research, founded in 191 6, the Institute of Econom- George M. Elsey ics, founded in 1922, and the Robert Brookings Grad- John Fischer uate School of Economics and Government, founded Kermit Gordon in 1924. Huntington Harris The Board of Trustees is responsible for the general administration of the Institution, while the immediate Roger W. Heyns direction of the policies, program, and staff is vested Luther G. Holbrook in the President, assisted by an advisory committee of Robert S. McNamara the officers and staff. The bylaws of the Institution William McC. Martin, Jr. state: "It is the function of the Trustees to make Arjay Miller possible the conduct of scientific research, and pub- Barbara W. Newell lication, under the most favorable conditions, and to Herbert P. Patterson safeguard the independence of the research staff in the pursuit of their studies and in the publication of J. Woodward Redmond H. Chapman Rose the results of such studies. It is not a part of their function to determine, control, or influence the con- Warren M. Shapleigh duct of particular investigations or the conclusions Gerard C. Smith reached." Phyllis A. Wallace The President bears final responsibility for the de- Wilkinson, J. Harvie Jr. cision to publish a manuscript as a Brookings book. In accuracy, Honorary Trustees reaching his judgment on the competence, and objectivity of each study, the President is advised Arthur Stanton Adams by the director of the appropriate research program Eugene R. Black and weighs the views of a panel of expert outside Robert D. Calkins readers who report to him in confidence on the quality Colgate W. Darden, Jr. of the work. Publication of a work signifies that it is Douglas Dillon deemed a competent treatment worthy of public con- Marion B. Folsom sideration but does not imply endorsement of conclu- John E. Lockwood sions or recommendations. Institution maintains its position of neutrality Robert Brookings Smith The on issues of public policy in order to safeguard the Sydney Stein, Jr. intellectual freedom of the staff. Hence interpretations or conclusions in Brookings publications should be understood to be solely those of the authors and should not be attributed to the Institution, to its trustees, officers, or other staff members, or to the organiza- tions that support its research. Digitized by the Internet Archive in 2012 http://archive.org/details/aregovernnnentorgOOkauf FOREWORD In government and out, among sophisticated students of politics and casual observers of the political scene, in this country and abroad, belief in the virtual indestructibility of government organizations is entrenched. Many would say yes to the question asked in the title of this book. But little systematic evidence supports such an answer. Somewhat more is known about the life span of private organizations ; among them a short life is not unusual. Few of their governmental counter- parts are assured of indefinite existence when they are created, but many of them seem methuselan if not everlasting. Yet the ensuing political and managerial difficulties would be staggering if it were true that old ones seldom die and new ones appear continually. In this study, Herbert Kaufman, a Brookings senior fellow, brings together data bearing on the birth, longevity, and death of govern- ment organizations in an attempt to determine whether the common belief is valid. Treating his title as a question to be investigated empirically, rather than as a rhetorical flourish, he examined more than four hundred units of federal executive departments. His con- clusions support the widespread impression that agencies stay alive once they have been born, but they also make clear that organiza- tional death is not an unknown phenomenon. By showing that both births and deaths occur in spurts, this venture into unexplored terri- tory demonstrates that such research can be of benefit to managerial practice and organization theory. Many people in the federal bureaucracy cooperated with the author in this study; it is not possible to thank each of them individ- ually here. On behalf of the author and the Brookings Institution, however, their contributions to the project are gratefully acknowl- edged. The editor of the manuscript, Elizabeth H. Cross, improved it substantively as well as stylistically, as did suggestions and criti- cisms by Gilbert Y. Steiner, Paul Quirk, and other readers. The author is grateful to Christine de Fontenay for her skillful and imaginative computer analysis of his data. Radmila Reinhart un- Vlll FOREWORD complainingly typed and retyped it and kept track of the seemingly endless changes. The views expressed in this study are the author's and should not be ascribed to the trustees, officers, or other staff members of the Brookings Institution. KERMIT GORDON President February igj6 Washington, D.C. CONTENTS Maybe Yes, Maybe No / Factors Favoring Long Agency Life 3 Hazards to Agency Survival 12 A Logical Dilemma 22 The Quest for an Answer 23 The Data 23 Research Problems 24 Procedure 2g Findings and Evidence 34 The Findings Interpreted 64 Organizational Turnover 64 Organizational Birth 66 Organizational Death 68 Can the Number of Agencies Be Controlled? yo How Little Is Known yy Tables i. Organizational Births, Deaths, and Survivors, by Age Group, 1923-73 35 2. Changes in Organization Population and Median Age, by De- partment, 1923-73 36 3. Organizations by Department and Type, 1923-73 40 4. Organizations by Instrument of Creation and Department, i9 2 3-73 44 5. Deaths of Organizations, by Instruments of Termination and of Creation, 1924-73 46 6. Births and Deaths of Organizations, by Presidential Term, 1 789-1973 48 7. Deaths and Death Rates of Organizations Compared with Business Failure Rates, 1924-73 53 Figures 1. Births and Deaths of Organizations, by Presidential Term, 1 789^973 5° 2. Percentage Change in Organizational Population in Each Presi- dential Term, 1 789-1973 51 3. Age Distribution at Time of Death of Organizations That Died between 1923 and 1973 61 4. Cumulative Totals of Organizations That Survived to 1973, by Presidential Term, 1 789-1973 62 . MAYBE YES, MAYBE NO You frequently hear the opinion that government agencies hardly ever die—or fade away, either. 1 Once established, according to this view, they continue indefinitely, whether or not they are needed or useful or even wanted. They keep going because they enjoy some kind of natural immunity to the forces that kill off nongovernmental organizations. If you stop to think about it, it's a startling idea. Even with an ex- tremely low birth rate, a population of immortals would gradually attain immense proportions. Since it is known that the birth rate of government organizations is above zero, it follows that if they are im- mortal, public administration is headed for (or maybe even has ar- rived at) deep trouble. Galvanizing administrative machinery made up of everlasting or- ganizations would be a formidable task. Where time is unlimited, there is no sense of urgency ; members of an organization assured of indefinite existence would probably be inclined to move at a much more leisurely pace than the rest of us can afford or tolerate. In their long view, matters pressing to most of us might well seem deferrable. Where we hold speedy decisions more important than carefully re- see i . For examples of published allegations of public-agency immortality, James M. Beck, Our Wonderland of Bureaucracy (Macmillan, 1932), p. 67; Luther Gulick, "Notes on the Theory of Organization," in Luther Gulick and L. Ur- wick, eds., Papers on the Science of Administration (New York: Institute of Public Administration, ; and Peter Drucker, The Age of Discontinuity (Harper 1 937) , p. 43 and Row, 1968), p. 222. The opposite thesis—that government organizations survive only with great difficulty—is seldom advanced either orally or in print. A notable exception is Herbert A. Simon, Donald W. Smithburg, and Victor A. Thompson, Public Administration (Knopf, 1950), chaps. 18 and 19 and pp. 560-61 An even more extreme argument—that organizations in general tend to be short- lived—was put forth most emphatically by Chester I. Barnard in The Functions of the Executive (Harvard University Press, 1938), p. 5. But this seems to be a minor- ity position, at least as far as public agencies are concerned. For the most part, then, the mortality and longevity of public organizations are rarely studied and seldom discussed in the literature on public administra- tion, and even writers' assumptions about them are hardly ever made explicit. The subject is largely neglected. 2 MAYBE YES, MAYBE NO searched judgments, they would doubtless prefer painstaking study before taking any action. Choices that look critical to those of us whose mortality compels us to get on with things would appear triv- ial to an organization with an infinite time horizon.