Incrementalism and Policymaking in the USA: Adaptations for a Partisan Age 3031384849, 9783031384844

Citation preview

Incrementalism and Policymaking in the USA Adaptations for a Partisan Age

Michael T. Hayes

Incrementalism and Policymaking in the USA “Michael Hayes has done it again, delivering another important contribution to the literature about public policy in the United States. In Incrementalism and Policymaking in the USA, Hayes makes an important argument: that incrementalism is not just the way things usually work in American government, but the way things should work. With patient logic and clear prose, Hayes makes a compelling case for realism in our policy debates. Significantly, he shows how incrementalism even makes sense in a time of intense political polarization. This book should be required reading for every American politician before they take office.” —Greg Berman, Co-editor, Vital City “‘In Incrementalism and Policymaking in the USA’ Michael Hayes makes the classic case for gradualism in policymaking as well as anyone has done. He also illustrates it with welfare reform and other recent national issues. And he writes very well. This is a fine primer for how most legislation actually gets done in America.” —Professor Lawrence M. Mead, New York University (NYU) “Michael Hayes has already written several interesting works on incrementalism, but this book surpasses them. As well as being a discussion of policymaking, and the relevance of incremental solutions for difficult problems, this is a work of political theory. Anyone interested in public policy, and in governance more generally, whether a realist or an idealist, should read and profit from this book.” —Professor B. Guy Peters, University of Pittsburg

Michael T. Hayes

Incrementalism and Policymaking in the USA Adaptations for a Partisan Age

Michael T. Hayes Political Science Colgate University Hamilton, NY, USA

ISBN 978-3-031-38484-4    ISBN 978-3-031-38485-1 (eBook) https://doi.org/10.1007/978-3-031-38485-1 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

Preface

This book began as an attempt to improve and update The Limits of Policy Change, but it quickly evolved into a very different book. This is not another attempt at explanatory theory. Readers should not expect an introduction to theories of the policymaking process here or a comparison of incrementalism with other theories of the policy process. To the contrary, I take it as self-evident that policymaking is incremental most of the time and devote myself instead to arguing that this is a good thing—not something to be lamented. I do acknowledge that there are times when nonincremental policies are warranted, even though they inevitably go beyond the available knowledge base, and I identify those circumstances in Chap. 2. But most of the time, candidates calling for transformative change are pandering to voters’ unrealistic sense of what is possible, making promises they cannot really fulfill. Compromise is out of vogue these days, too often characterized as a violation of strongly held moral principles. But there are both moral and intellectual reasons why compromise is both necessary and desirable. Hopefully, this book will help readers understand why this is so. Hamilton, NY

Michael T. Hayes

v

Acknowledgments

I would like to thank Colgate University, particularly the Dean of Faculty, Tracey Hucks, for approving two nonconsecutive one-semester leaves in the spring of 2021 and the spring of 2022. Much of the book was written (and rewritten and rewritten) during those two leaves. I also want to thank my very good friend and longtime Colgate colleague, Michael Johnston, for reading every chapter of the book. I am also indebted to the three anonymous reviewers for many excellent suggestions that led to a major rewriting of the book. Bob Samuels, the assistant director of every Washington Study Group I have directed, has been a consistent source of encouragement, editorial advice, and friendship. Thanks also to my editor at Palgrave, Stewart Beale, for the many ways he has helped with the volume, as well as to my production editor, Tryphena, and all those involved in the production process. My wife Candace contributed to this book in two major ways. First, she graciously put up with my irritability and obsessiveness while working on the book. She also read over the many op-eds I have written in recent years, helping me to eliminate jargon, rearrange my ideas more efficiently, and reach a larger non-academic audience. In so doing, she has made me a much better writer, which has had a positive impact on the book as well. Above all, I want to thank the many students who went on five Washington Study Groups with me. They stimulated my thinking in a wide variety of ways and helped me see flaws in my previous book. I also

vii

viii 

ACKNOWLEDGMENTS

want to thank the students in my upper-level Politics of Poverty course, especially those in the fall of 2021, who read chapters from a previous draft of this book. I dedicate this book to all those Colgate students who have been a delight to teach and from whom I have learned so much. Finally, I want to thank God for all the blessings alluded to above and countless more besides. All errors, of course, are my own.

Contents

1 Incrementalism,  Realism, and Idealism 1 Realists vs. Idealists  3 Attitudes Toward Incrementalism  5 Pursuing Idealist Visions Through Realist Means  6 Realism as a Prescriptive Model  7 Outline of the Book  8 Policies as Experiments from Which We Can Learn 10 References 10 2 A  Realist Case for Incrementalism13 Obstacles to Rational Decision-Making 14 Four Decision Environments 15 Incrementalism as an Alternative to Rationality 17 Incrementalism Circumvents Obstacles to Rationality 18 What Makes a Policy Non-incremental? 19 When Is Non-incremental Policy Change Possible? 21 When Is Non-incremental Policy Change Desirable? 22 Conclusion 24 References 25 3 Partisan Incrementalism27 Partisan Incrementalism Defined 29 Welfare Reform as a Case Study of Partisan Incrementalism 31 The House Republicans Move First 31 ix

x 

Contents

Welfare Reform Moves to the Senate 34 Budget Considerations Intrude and Clinton Vetoes Two Bills 35 The Governors’ Proposal Revives Welfare Reform 36 Modification of the Law Through Reauthorization 37 Conclusion 40 References 41 4 Welfare  Reform as Normal Incrementalism43 Conflict over Problem Definition 44 Building on Past Policy 46 Negotiation and Tapering Down 49 Focus on Incremental Alternatives 50 The Final Outcome: Incremental or Non-incremental? 53 References 56 5 Incrementalism  and Political Inequality59 Biases to the Group Universe 60 Group Resources: Cumulative or Dispersed? 63 The Privileged Position of Business in Capitalist Societies 65 Political Inequality Plagues All Decision-Making Methods 67 The Special Problem of Nondecision-Making 69 Conclusion 72 References 73 6 Recapitulation and Implications77 Incrementalism Can Be Adapted to a Partisan Age 78 Eliminating the Senate Filibuster 79 What Does It Mean to Be a Realist? 81 Voters Must Also Become Realists 82 Partisan Incrementalism in Parliamentary Systems 84 References 85 Index87

About the Author

Michael T. Hayes  graduated from the University of Kansas in 1971. He received his PhD from Indiana University in 1977. His dissertation won the E.  E. Schattschneider Award from the American Political Science Association as the best dissertation in American Politics for the 1977–1978 academic year. His first article, “The Semi-Sovereign Pressure Groups: A Critique of Current Theory and an Alternative Typology,” Journal of Politics 40 (1978), 134–161, won the Jack L. Walker Outstanding Article Award from the American Political Science Association, Section on Interest Groups and Political Parties, 1991. Professor Hayes taught at Lawrence University from 1975 to 1977, at Rutgers University from 1977 to 1984, and at Colgate University from 1984 to 2022. He has written two previous books on incrementalism: Incrementalism and Public Policy (2006; originally published by Longman in 1992) and The Limits of Policy Change: Incrementalism, Worldview, and the Rule of Law (Georgetown University Press, 2001). He also wrote Lobbyists and Legislators: A Theory of Political Markets (Rutgers University Press, 1981) and co-edited Inside the House: Former Members Tell How Congress Really Works (University Press of America, 2001) with Lou Frey, Jr. In addition, he has written numerous journal articles and book chapters on domestic and foreign policymaking.

xi

CHAPTER 1

Incrementalism, Realism, and Idealism

Abstract  This chapter previews the argument that incrementalism, as a method of policymaking, is both inevitable and desirable. While idealists have an optimistic view of human nature and believe transformative policy changes can be achieved in a single step, realists recognize the inevitability of conflict and the limited knowledge base available to policymakers. Ideally, policymaking would be dominated by realists on both the left and the right. The chapter concludes by providing an outline of the book and identifying a recurring theme: policies should be understood as experiments from which we can learn. Keywords  Incrementalism • Realists • Idealists • Policies as experiments Can we really enact new laws that will yield transformative change? Or are incremental outcomes all we can realistically expect? These questions are central to the policy process, and they never lose their importance. Every new presidential administration, having promised voters major policy initiatives, must face these questions. Individual members of the House or Senate, when they act as policy entrepreneurs, drafting bills and trying to get them enacted, must also face them (Price, 2021, 95–131).

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. T. Hayes, Incrementalism and Policymaking in the USA, https://doi.org/10.1007/978-3-031-38485-1_1

1

2 

M. T. HAYES

Two recent examples from the Biden presidency show the continuing relevance of these questions. Example 1: In his first year in office, President Biden pushed for what he called the “Build Back Better” bill. Although Democrats nominally controlled both the House and Senate, they only had 50 seats in the Senate and were therefore only able to pass bills because of Vice President Harris’s ability to cast tie-breaking votes. Moreover, the need for 60 votes to invoke cloture on filibustered bills meant that Senate Democrats could only pass bills that could be considered part of the budgetary reconciliation process, which required a simple majority. Overly optimistic progressives nevertheless advanced a wish list of initiatives that would have cost more than $7 trillion. A more concrete version of the bill, which Biden characterized as the centerpiece of his legislative agenda, would have cost “only” $3.5 trillion. Although this figure was eventually reduced further through negotiations, the bill failed to reach the needed 50 votes that would have enabled Vice President Harris to cast the tie-breaking vote in the Senate. Negotiations broke down, and the bill was pronounced dead (Martin & Burns, 2022). During Biden’s second year in office, however, a compromise emerged between Majority Leader Chuck Schumer and Senator Joe Manchin (D-WV), who had been the chief obstacle to passage of the original bill. The updated version, severely limited in scope and much less expensive than the original bill (about $1.85 trillion), finally passed as the Inflation Reduction Act. While the provisions dealing with energy and the environment represented the biggest investment ever made in addressing climate change, most non-climate portions of the original bill were eliminated (Raju et al., 2022). Could a more skillful president have gotten a bigger bill through? Or was the new bill the most any president could realistically achieve, as the Biden administration claimed? Example 2: During the President’s second year in office, two horrific mass shootings occurred within a few days of one another, creating an overwhelming mass public demand for at least some action to address this long-neglected problem. In response to this mass public outrage, several Republican Senators agreed to engage in negotiations with Democrats, eventually providing the extra votes needed to overcome a party filibuster. While the result was the first gun control legislation passed by Congress in decades, the final product was viewed by most observers in the media as disappointingly weak (Cochrane & Kanno-Youngs, 2022). Was a bill so incremental in scope worth passing? Could a more skillful president have

1  INCREMENTALISM, REALISM, AND IDEALISM 

3

gotten more? Or was the bill a remarkable achievement, given the intense opposition from the NRA, gun owners, and many congressional Republicans? While incremental policies are by far the most common outcome of the legislative process, incremental policy changes are often dismissed as not worth pursuing. For example, in one 2020 Democratic presidential debate, Senator Elizabeth Warren famously questioned why anyone would run for president just to talk about “what we really can’t do and shouldn’t fight for” (Kaplan, 2019). But bigger is not always better. To the contrary, I will argue in this book that incrementalism is both inevitable and desirable. Disagreement over the merits of incrementalism arises because different underlying assumptions about human nature give rise to different expectations regarding what it is possible to achieve through politics. Scholars with a positive view of human nature have great confidence in our capacity both to cooperate with one another and to solve problems through individual reasoning. By contrast, those with a pessimistic view of human nature see man as fallible, fallen, and prone to conflict. Accordingly, these pessimists minimize our capacity to solve problems through analysis, instead expecting good policy outcomes to emerge, if at all, from a pluralistic process involving a multiplicity of participants (Hayek, 1948; Spicer, 1995; Sowell, 1987, 1995; Hayes, 2001, 28–35).

Realists vs. Idealists International relations theorist Hans J. Morgenthau (1962, 3) characterized the history of modern political thought as a contest between two competing schools of thought with fundamentally different conceptions of the nature of man. He called these two schools idealists and realists. Idealists view human nature as both good and malleable; it follows that a rational and moral political order is attainable. Confronted with our perpetual failure to establish such a moral order, idealists typically blame inadequate knowledge, obsolete social institutions, or the depravity of threatened individuals or groups. By contrast, realists believe that the world will forever be imperfect due to forces rooted in human nature. To improve the world even marginally, political leaders must work with those forces, not against them. Conflicts are inevitable in a world where nations have different interests and inevitably seek power to advance those interests. Within nation-states, domestic

4 

M. T. HAYES

policymaking is subject to these same forces. Because conflicts among opposing interests are inevitable, moral principles can only be approximated within heterogeneous societies, and checks-and-balances are indispensable. All political leaders can realistically hope to achieve are lesser evils rather than absolute goods (Morgenthau, 1962, 4). It is tempting to equate idealism with political liberalism and realism with political conservatism because liberals are optimistic about man’s potential while most forms of political conservatism start with a pessimistic view human nature (Contrast Burke, 1999 with Frankel, 1959; Spitz, 1964; Garry, 1992). For example, Thomas Sowell (1995) characterized liberals as utopian thinkers who continually overreach, and Kenneth Minogue (1996) drew a similar distinction between liberals and what he called conservative realists. Identifying liberalism with utopianism and conservatism with realism is misleading, however. Rather, we can identify four distinct worldviews, as there are idealists and realists on both the left and the right. First, we can identify two distinct types of liberals. Liberal idealists have great confidence in their ability to solve social problems through the power of human reason. For example, Elizabeth Warren asserted that “I have a plan for that” in response to every question about any current problem during the 2020 Democratic presidential debates. Michael Oakeshott had someone very much like Elizabeth Warren in mind when he said that rationalists (his term for liberal idealists) “cannot imagine a politics which do not consist in solving problems, or a political problem of which there is no ‘rational’ solution at all” (Oakeshott, 1991, 10). By contrast, liberal realists share the liberal idealists’ desire for social reform but view all knowledge as necessarily incomplete. As a result, they regard politics as a means by which to ameliorate public problems gradually. Because knowledge is dispersed throughout the system, policies must be made through an interactive social process in which all important interests are represented. President Biden is a liberal realist. His understanding of policymaking as a process of social interaction is manifested not only in his repeated calls for bipartisanship but also in his reputation for always being at the center of his party. Second, we can similarly identify two distinct varieties of conservatives. In an earlier work (Hayes, 2001, 29–33) I distinguished between adaptive and nostalgic conservatives. While adaptive conservatives want to preserve cherished institutions that have been eroded by modernity, nostalgic conservatives want to restore a previous way of life. I no longer see this as the

1  INCREMENTALISM, REALISM, AND IDEALISM 

5

central distinction, however. The real issue, I now believe, is not whether conservatives want to go forward (to adapt) or backward (to restore), but rather whether they are realist or idealist in their pursuit of their chosen goal. Conservative realists want to preserve cherished institutions or practices but recognize both their own fallibility and the right of other actors to hold different visions. By contrast, conservative idealists, following Morgenthau’s distinction, believe a rational, moral order is attainable. Many (although not all) contemporary conservative idealists believe this moral order was revealed by God through scripture (Weaver, 1984) or the teachings of the Catholic Church (Linker, 2006). Until recently, religious conservatives took a pessimistic view of human nature; for example, both Edmund Burke and Russell Kirk were explicitly anti-utopian (Kirk, 1993, 1997; Burke, 1999). With the rise of conservative evangelicals in the 1970s and 1980s (Kristol, 1996) and the emergence of the theoconservative movement in the 1980s and 1990s (Linker, 2006), a new variety of religious conservative emerged, quintessentially idealist in their desire to impose a rational moral order on American society. These religious conservative idealists are anti-science, anti-intellectual, and belligerently anti-pluralist (Linker, 2006, 176–207). They see the increasing secularization of American life as continually eroding the moral order and view political disagreement as evidence of complicity in an ongoing project to undermine moral norms. Conservative idealists are not all religious, however. Following Hayek (1994) many contemporary libertarians see a long drift away from free markets along a “road to serfdom.” Tea Party libertarians, and their Freedom Caucus successors, find their moral order in the market and champion draconian reductions in federal spending (Draper, 2012).

Attitudes Toward Incrementalism Realists are more likely than idealists to embrace incrementalism. Liberal idealists prefer once-and-for-all comprehensive solutions to problems, and they are necessarily impatient with incrementalism because they are in a hurry to get where they are going and resent any process that forces them to settle for one piece at a time. As noted earlier, Senator Warren questioned why anybody would run for President just to talk about what we cannot accomplish (Kaplan, 2019). Conservative idealists reject incrementalism for a different reason, equating it with a long drift away from fundamental truths (Hayek, 1994). For this reason, they view compromise as

6 

M. T. HAYES

corrupt, and an increasing number of them are willing to use violence to restore what they see as the lost moral order. Liberal realists are by far the most comfortable with incremental change. They regard inherited policies as the product of fallible, fallen people and therefore always subject to improvement. While sharing the liberal idealist’s belief that social and economic changes are potentially beneficial, liberal realists believe policy change is necessarily pursued through a pluralistic process involving a multiplicity of actors with different objectives and partial information. Where liberal idealists believe problems can be fully solved through the exercise of human reason, liberal realists are humbler, using politics to mitigate problems gradually through a series of iterations. Conservative realists also embrace incrementalism, although for a different reason. Because they see society and government as complicated beyond human power to grasp easily, they try to avoid unintended adverse consequences by proceeding piecemeal. While they typically resist change, when change is forced upon them conservative realists prefer that it proceed gradually. Historically, the typical response of the conservative realist to liberal proposals for change has been initial skepticism followed by grudging acceptance and a desire that changes be both smaller and slower.

Pursuing Idealist Visions Through Realist Means While all four worldviews have virtues as well as vices (Hayes, 2001, 154–161), both the liberal and conservative idealists fail to recognize the inevitability of conflict among actors with different interests as well as the need for some process of bargaining and compromise to accommodate differences. However, these two schools nevertheless make a potential contribution by identifying positive visions toward which more realistic policymakers can move incrementally. For example, as Yuval Levin has observed (2012, 6), although capitalism produces great prosperity, it also creates great insecurity for many people. For this reason, liberals believe capitalism must be constrained by government regulations and buttressed by a system of social insurance that provides a safety net. However, this liberal vision of a social-democratic welfare state has necessarily been pursued incrementally. Both liberals and progressives view this succession of small steps as constituting progress toward goals they both share (Levin, 2012, 8). By contrast, Levin emphasizes the potential of a free economy to facilitate social mobility and elevate the poor (Levin, 2012, 16), and the real

1  INCREMENTALISM, REALISM, AND IDEALISM 

7

problem with capitalism is that too many people are denied access to its benefits. Similarly, family, church, and civil society are the primary means by which human beings find fulfillment—not outmoded institutions that need to be circumvented by government regulators. As such, they should be encouraged rather than undermined (Levin, 2012, 16; see also Glazer, 1990). Significantly, these contemporary conservative realists explicitly recognize the need to pursue their alternative vision through incremental steps and acknowledge that their ideal vision will never be fully realized. Even if all their proposed reforms were enacted, Levin concedes, there would still be an extensive federal government and no shortage of bureaucratic inefficiency (Levin, 2012, 18–19). More important than any one of the specific policies they propose is the development of a reform-oriented conservatism (Levin, 2012, 15). In short, in the same way that liberal realists pursue an expansionary agenda through incremental steps, conservative realists may pursue a contractionary agenda through decremental steps (Dommel, 1974; see also Hayes, 2001, 159–162).

Realism as a Prescriptive Model While Morgenthau advanced realism as the basis for a descriptive theory of international politics rooted in objective laws that have their roots in unchanging human nature, he repeatedly lamented that real-world foreign policies were often driven instead by idealistic motives. As a result, his classic work may be better understood as advancing realism as a superior normative model for how foreign policies ought to be made. In Morgenthau’s view, foreign policies would be better, and the world would be a safer place, if all political leaders were realists rather than idealists. I am making a similar normative argument in this volume. Incrementalism is, in my view, a superior way to make public policies. In my ideal world—a world in which millions of people read this book and come to appreciate the virtues of incrementalism—liberal and conservative realists would dominate policy debates. In such a world, all the participants would have a proper humility, recognizing both their own moral and intellectual limitations and the consequent need for policymaking through incrementalism. They would also recognize that different actors and groups necessarily have different political interests and a legitimate right to advance them. Viewed through the realist lens, those opposed to our vision are not evil, but rather have as much right to advance their interests as we do. It

8 

M. T. HAYES

follows that compromise is a normal part of politics, not a violation of principles understood as sacrosanct. In this ideal world, antipluralist conservatives would be supplanted by conservative realists who pursue decremental reforms with an explicit vision of the end toward which they are working. Realistic conservatives understand that the breakdown of the rational-comprehensive ideal applies to conservatives as well as progressives, and that incrementalism is the only viable way to make public policies, whether the goal is to expand the scope of government or contract it. Similarly, liberal idealists would be both outnumbered and outargued by liberal realists who recognize both the limitations on human rationality and the inevitability of policymaking through bargaining, compromise, and future modifications. The belief, common among liberal idealists, that a policy change is not worth pursuing unless it is transformative would be supplanted by a vision of policies as necessarily experimental.

Outline of the Book Chapter 2 begins by exploring both why incrementalism is so prevalent and why it is also a superior method of policymaking, Subsequent sections identify the conditions under which major policy changes are sometimes possible as well as the conditions under which major policy departures are—and are not—desirable. Chapter 3 shows how incrementalism can still function effectively under our current regime of extreme party polarization. There can be no doubt that polarized parties superimposed on our Madisonian system often yield gridlock, and incremental policymaking was certainly easier in the past, when the Democratic and the Republican parties both featured liberal and conservative wings. Nevertheless, I contend that the development of ideologically cohesive, polarized parties creates the potential for an alternative form of incrementalism which I will call “partisan incrementalism.” The second half of the chapter presents a case study illustrating what partisan incrementalism looks like in practice. The 1996 welfare reform converted the Aid to Families with Dependent Children program from a categorical grant aimed at income maintenance to a block grant program promoting work. The conservative realist drafters of the original bill sought a distinctively Republican approach to a chronic public problem that might help them regain control of the House. Along the way, their

1  INCREMENTALISM, REALISM, AND IDEALISM 

9

welfare reform bill was incorporated into Newt Gingrich’s “Contract with America,” a ten-point program that House Republican candidates pledged to enact within the first one hundred days of the next Congress, a quintessential instance of responsible party government. Chapter 4 reexamines this same case study with a different theoretical focus. Five core elements of incrementalism, identified in Chap. 2, will be applied to the welfare reform case. While the primary drafter of the legislation characterized the new law as “revolutionary” (Haskins, 2006), Chap. 4 will show that the welfare reform of 1996 conformed to the incremental model in every respect. Political necessities forced proponents of work-­ oriented welfare reform to build on past policies, negotiate with a variety of actors in both parties, and eventually accept a final product that fell far short of their most ambitious goals. Chapter 5 asks whether political inequality—a central feature of our political system—precludes the proper functioning of incrementalism. There are predictable biases to the interest group universe, as many interests fail to mobilize successfully, and political resources are distributed very unevenly across groups. As a result, incrementalism necessarily works imperfectly. However, political inequality constitutes a problem for all policy-making methods—not just incrementalism—and the effects of political inequality on incremental policymaking are typically exaggerated. Chapter 5 also examines the problem of nondecision-making: a process through which powerful interests can prevent issues threatening them from arising at all (Bachrach & Baratz, 1970). The concept of nondecision-­ making will be applied to our dual system of criminal justice—a system in which blacks and other minorities are systematically treated differently than whites (Moore, 2015). Attempts to reform this system were “organized out of politics” (Schattschneider, 1960) between 1988 and 2012, as the fear of crime displaced concerns over racial inequalities in the criminal justice system. I will argue that this potential for nondecision-making provides the greatest obstacle to the effective operation of incrementalism. Often, the persistence of pressing public problems results less from the inevitability of incremental policy outcomes than from a failure to take any action at all. Put another way, we should worry less about the small size of individual steps when the real problem is our failure to start moving in the first place. Chapter 6 will recapitulate the argument of the book and tease out implications. In sum, this book will advance three arguments. First, incrementalism is both inevitable and desirable as a method of policymaking.

10 

M. T. HAYES

Second, incrementalism can be adapted to function effectively in our highly partisan age through a process I term partisan incrementalism. And third, realists are more likely to see the merits of incrementalism than idealists, and in my ideal world policymaking would be dominated by realists on both the left and right. My goal is to persuade both policymakers and voters to be more realistic about the limited potential for transformative one-step solutions to public problems.

Policies as Experiments from Which We Can Learn In this regard, a recurring theme of my argument throughout is that limitations on our understanding of most policy problems preclude the design of non-incremental policy changes that would work effectively. All policies yield unanticipated consequences that will need to be remedied in subsequent policy cycles. For this reason, public policies should be viewed as experiments from which we can learn. It follows that when presidential candidates promise to solve major problems through transformational legislation, they are promising more than anyone can really deliver. Because there are upper limits on what we can achieve through any one program or bill, our two parties should compete by offering voters platforms that propose incremental (or decremental) solutions to public problems. When policies fail to solve problems fully (as all policies do), they can then be modified through subsequent legislation. Particularly promising policies can be built upon and expanded. Voters as well as elected officials need to recognize that there are both political and intellectual limits on policy change. As the late economist Frank H. Knight observed, limitations on our ability to predict outcomes open the way for wishful thinkers to underestimate both the intellectual and political obstacles to the success of proposals they endorse. Political science must give both citizens and political leaders more realistic ideas as to what it is possible to achieve through political action (Knight, 1982, 395–396).

References Bachrach, P. M., & Baratz, M. (1970). Power and Poverty: Theory and Practice. Oxford University Press. Burke, E. (1999). Reflections on the Revolution in France. In I. Kramnick (Ed.), The Portable Edmund Burke. Penguin Books.

1  INCREMENTALISM, REALISM, AND IDEALISM 

11

Cochrane, E., & Kanno-Youngs, Z. (2022, June 25). Biden Signs Gun Bill into Law, Ending Years of Stalemate. New York Times. Biden Signs Bipartisan Gun Bill Into Law - The New York Times (nytimes.com) Dommel, P. R. (1974). The Politics of Revenue Sharing. Indiana University Press. Draper, R. (2012). When the Tea Party Came to Town. Simon & Schuster. Frankel, C. (1959). The Case for Modern Man. Beacon Press. Garry, P. M. (1992). Liberalism and American Identity. Kent State University Press. Glazer, N. (1990). The Limits of Social Policy. Harvard University Press. Haskins, R. (2006). Work Over Welfare: The Inside Story of the 1996 Welfare Reform Law. Brookings Institution. Hayek, F.  A. (1948). Individualism: True and False. In F.  A. Hayek (Ed.), Individualism and Economic Order (pp. 1–32). University of Chicago Press. Hayek, F. A. (1994). The Road to Serfdom (Fiftieth Anniversary ed.). University of Chicago Press. Hayes, M.  T. (2001). The Limits of Policy Change: Incrementalism, Worldviews, and the Rule of Law. Georgetown University Press. Kaplan, T. (2019, July 30). Elizabeth Warren’s Slam on John Delaney Was Called the Line of the Night. Here’s What She Said. New York Times. Elizabeth Warren’s Slam on John Delaney Was Called the Line of the Night. Here’s What She Said. - The New York Times (nytimes.com). Kirk, R. (1993). The Politics of Prudence. Intercollegiate Institute. Kirk, R. (1997). Edmund Burke: A Genius Reconsidered. Intercollegiate Studies Institute. Knight, F. H. (1982). Economics, Political Science, and Education. In F. H. Knight (Ed.), Freedom and Reform: Essays in Economics and Social Philosophy. Liberty Fund. Originally published in 1947 by Harper & Brothers. Kristol, I. (1996). America’s “Exceptional Conservatism”. In K. Minogue (Ed.), Conservative Realism. Harper Collins Publishers in Association with the Centre for Policy Studies. Levin, Y. (2012). Beyond the Welfare State. In Y. Levin & M. Clyne (Eds.), A Time for Governing: Policy Solutions from the Pages of National Affairs. Encounter Books. Linker, D. (2006). The Theocons: Secular America Under Siege. Doubleday. Martin, J., & Burns, A. (2022). This Will Not Pass: Trump, Biden, and the Battle for America’s Future. Simon & Schuster. Minogue, K. (1996). Introduction: On Conservative Realism. In K.  Minogue (Ed.), Conservative Realism: New Essays in Conservatism. HarperCollins Publishers in Association with the Centre for Policy Studies. Moore, N.  M. (2015). The Political Roots of Racial Tracking in American Criminal Justice. Cambridge University Press. Morgenthau, H.  J. (1962). Politics Among Nations: The Struggle for Power and Peace (3rd ed.). Alfred A. Knopf.

12 

M. T. HAYES

Oakeshott, M. (1991). Rationalism in Politics and Other Essays (New and Expanded ed.). Liberty Fund. Price, D.  E. (2021). The Congressional Experience: An Institution Transformed (4th ed.). Routledge. Raju, M., Nilsen, E., & Luhby, T. (2022, July 28). In Major Boost to Democrats, Manchin and Schumer Announce Deal for Energy and Health Care Bill. cnnpolitics. Joe Manchin and Chuck Schumer announce deal for energy and health care bill | CNN Politics. Schattschneider, E.  E. (1960). The Semi-Sovereign People: A Realist’s View of Democracy in America. Holt, Rinehart, and Winston. Sowell, T. (1987). A Conflict of Visions: Ideological Origins of Political Struggles. William Morrow, Quill Books. Sowell, T. (1995). The Vision of the Anointed: Self-Congratulation as a Basis for Social Policy. Basic Books. Spicer, M. W. (1995). The Founders, the Constitution, and Public Administration: A Conflict in Worldviews. Georgetown University Press. Spitz, D. (1964). The Liberal Idea of Freedom. University of Arizona Press. Weaver, R. (1984). Ideas Have Consequences. University of Chicago Press.

CHAPTER 2

A Realist Case for Incrementalism

Abstract  Rational decision-making is typically precluded by disagreement over values and/or a limited knowledge base among participants. Incrementalism circumvents these problems, permitting the formulation of policies even where rationality breaks down. While single-step non-­ incremental policies are possible under exceptional circumstances, these major policy changes necessarily go beyond the available knowledge base. By contrast, non-incremental policy changes are most likely to be enduring and effective when they constitute rationalizing breakthroughs that occur late in the life cycle of an issue. Keywords  Breakthrough vs. rationalizing policies • Calculated risks • Life cycle of policies • Rationalizing breakthroughs • Tapering down from the optimal to the acceptable Experience teaches us that the realists are right: policymaking is typically incremental. This chapter will argue that the prevalence of incrementalism is a good thing. I will begin by reviewing the main obstacles to rational decision-making, identify the key elements of incrementalism as an alternative to rationality, and then show how incrementalism circumvents these obstacles. Next, I will ask what characteristics differentiate incremental

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. T. Hayes, Incrementalism and Policymaking in the USA, https://doi.org/10.1007/978-3-031-38485-1_2

13

14 

M. T. HAYES

polices from those that qualify as non-incremental. I will conclude by exploring when non-incremental policy change is possible and whether such change is desirable.

Obstacles to Rational Decision-Making If public policies were made through a process of rational comprehensive decision-making, decision-makers would be free to formulate the “best” policy without regard to whether it was incremental or non-incremental. Starting with agreed-upon goals, decision-makers would examine all available alternatives, accurately identify the consequences of each, and choose the one that maximized their values. Unfortunately, rational decision-making requires both a consensus on objectives that is almost never attained and an understanding of the problem at hand sufficient to permit accurate prediction of the consequences of available alternatives. The all-too-common combination of value conflicts and an inadequate knowledge base precludes rational decision-­ making (Braybrooke & Lindblom, 1970, 81–110). In the current polarized political era, where Democrats and Republicans disagree across the board on most issues—education, the environment, immigration, health care reform, and tax policy—the prevalence of conflict over objectives should be self-evident. On issues where they might seem to agree, like the need for a strong national defense, the two parties disagree over how much spending is enough. Trade-offs are another major source of conflict. Trade-offs occur where two or more goals are in conflict: we cannot simultaneously maximize both goals, so to gain more of one, we must have less of the other. For example, after 9/11 we had to face the trade-off between homeland security and civil liberties. To gain more security meant sacrificing some liberties, and vice versa. Different people made this trade-off in different ways, however, with some valuing security more while others emphasized civil liberties. Rational decision-making also requires a knowledge base sufficient to predict consequences of available alternatives accurately. The disagreement over how to trade-off shared values reviewed above is made worse by imperfect knowledge: no one can say for sure how much security we are gaining by adopting policies that reduce our civil liberties. Much the same thing could be said regarding the more recent trade-off between public health and individual freedoms posed by the COVID epidemic. Not only

2  A REALIST CASE FOR INCREMENTALISM 

15

do different people make the trade-off between public health and individual liberty in diverse ways but also the empirical relationship between the two (how much health we gain for each unit of liberty given up) is hotly contested. Policymakers almost never have sufficient knowledge to estimate policy consequences accurately. If we create a health insurance exchange under the Affordable Care Act, how many people will sign up? What will the premiums be, and how will they vary in different places? How will they change over time? Similarly, if we cut taxes on the wealthy, will investment increase, increasing jobs and prosperity so that the tax cuts pay for themselves? If we raise the minimum wage significantly, what will be the effects on employment? Will the wage increase help more people than it hurts? Will this vary from one industry to another, and if so, how much? Answers to these questions, and others like them, are hotly contested. There are often long running disputes over what consequences flowed from policies that were enacted in the past.

Four Decision Environments Lindblom believed there would be both a conflict over objectives and an inadequate knowledge base on most issues, rendering rational decision-­ making impossible and making incrementalism inevitable. Sometimes policymakers do agree on basic objectives, however, and sometimes they share a mutual understanding of how a particular policy area works and agree on basic facts. Where policymakers achieve agreement on what variables are important and how they are causally related to one another, we can term this consensual knowledge (Rothstein, 1984, 736). If we treat the degree of agreement over objectives and policy consequences as variables rather than as inherent limitations on decision-­making, we can identify four distinct decision environments (Hayes, 2006, 136–143, 2022), (1) Where there is no agreement on objectives and where the knowledge base is insufficient to permit accurate estimation of consequences—Lindblom’s normal case—incrementalism is inevitable. (2) By contrast, where policymakers agree on both policy objectives and the consequences associated with different alternatives, decision-making through an analytic process is possible. (3) Where the knowledge base is uncontested, but policymakers disagree over objectives, the result is a pure problem of value conflict. The abortion issue would fall here, as participants

16 

M. T. HAYES

disagree sharply over whose interests should be protected (pregnant women or fetuses) while the available scientific knowledge base on fetal development is extensive. (4) By contrast, where policymakers agree on objectives but lack the knowledge of how to achieve them, the result is a pure problem of knowledge base. Macroeconomic policy is a pure problem of knowledge base since all policymakers want to achieve full employment without inflation but disagree sharply over how to achieve these goals, or on what the empirical trade-off is between them. While rational decision-making may yield non-incremental policy outcomes whenever a major policy change is judged to be value-maximizing, incremental outcomes are inevitable in the other three decision environments. However, the policy process giving rise to these incremental outcomes will differ from one decision environment to another. Normal incrementalism will occur whenever conflict over objectives is combined with an inadequate knowledge base, just as Lindblom predicted. For pure problems of value disagreement, however, conflicts will be especially intense. Where an inadequate knowledge base tempers conflict by making participants unsure of the consequences of the policies they are advancing, a fully adequate knowledge base allows for no such uncertainties. For pure problems of knowledge base, policy changes will be viewed as experiments from which policymakers can draw lessons. The smaller the policy change, the easier it is for policymakers to identify the causes of undesirable consequences. True rational decision-making is extremely rare, if not nonexistent. When policymakers agree on goals, it may be because the circle of decision-­ makers excludes dissenters. When debating whether to drop an atomic bomb on Japan at the close of World War II, for example, all those involved in the decision process agreed on the goal of unconditional surrender by Japan, but dissenters who rejected this goal existed outside the decision-­ making circle (Stimson & Bundy, 1948). Similarly, the fact that policymakers exhibit consensual knowledge does not mean that their shared knowledge base is objectively correct, only that they agree upon it at the time the decision is made. What we can say with confidence is that where policymakers achieve consensus on both objectives and knowledge base, policies will result from a calculated decision rather than a process of bargaining and compromise.

2  A REALIST CASE FOR INCREMENTALISM 

17

Incrementalism as an Alternative to Rationality The ideal of rationality holds great appeal for students of the policy process, including many who would not characterize themselves as utopian dreamers, because it involves specification of objectives, a rigorous analysis of options, and value-maximizing decisions. This is what most people believe decision-making should look like. Unfortunately, however, Lindblom is correct in asserting that comprehensive rationality is simply out of the question for most policy problems. As an alternative to the rational ideal, disjointed incrementalism has five major elements (Lindblom, 1959; Braybrooke & Lindblom, 1970; Hayes, 2006). (Two additional elements need not concern us here: adjustment of objectives to policies and margin-dependent choice. For definitions of these two elements see Hayes, 2006, 18–19.) First, policymakers do not try to maximize abstract positive values but rather move away from problems, expecting only to mitigate or ameliorate these problems without ever fully solving them. In so doing, policymakers are not typically faced with a “given” problem (Lindblom, 1980, 24). Rather, problems must be perceived and defined, a process that is anything but automatic. Not only do different actors advance different problems for consideration, but problems will usually be defined differently by different actors. Second, policymakers maximize available information by building on past policies. They have gained experience and learned from inevitable mistakes through implementation of existing policies. Building on past policies takes full advantage of this experience. By contrast, designing innovative, transformative policies is a leap into the unknown. As Edmund Burke understood, the spirit of innovation is hostile to reform (Burke, 1999; see also Burke’s “Letter to William Elliot,” Kramnick, 1999, 569–573). Third, policymakers typically confine their attention to proposals that differ only incrementally from existing policies. Incremental policy changes tend to be more politically viable than proposals for non-incremental change, so focusing on incremental alternatives minimizes the costs of search and evaluation by excluding from the outset policy alternatives that are almost certain to be rejected. The consequences of incremental policy changes are also easier to predict. Fourth, policies emerge from a process of social interaction that necessarily involves negotiation and compromise. To secure the votes necessary

18 

M. T. HAYES

for passage, policy proponents must “taper down proposals from the optimal to the acceptable” (Jones, 1974). Finally, under incrementalism policymakers converge on good policies by modifying programs through successive policy cycles. This reduces the need for an adequate knowledge base by permitting policymakers to learn from experience. They do not need to formulate comprehensive solutions that solve problems in a single step. Rather, unanticipated consequences can be addressed through subsequent legislation. Lindblom used several alternative terms to describe this process of gradual improvement of policies over time: seriality, successive approximations, iteration, and convergence on a solution.

Incrementalism Circumvents Obstacles to Rationality Incrementalism circumvents the problems plaguing the rational ideal in at least five ways. First, as noted above, policymakers need not make a single, comprehensive decision that will solve a problem fully in a single step. To the contrary, policies can always be modified as necessary in subsequent iterations of the policy cycle. Of course, policy evaluation in subsequent policy cycles is inevitably a political process characterized by fragmentation, conflict, and imperfect knowledge. Many of the same actors who disagreed over values, trade-offs, and even problem definition will be central participants in the process of policy evaluation and modification, precluding rational analysis at this stage as well. Problems remain “unsolved” as long as some publics continue to express dissatisfaction with existing policies. By contrast, a problem may be regarded as “solved” when it is eventually crowded off the agenda by other problems that are now considered more pressing (Wildavsky, 1979, 60). Second, incrementalism reduces the costs of analysis by providing a defensible basis for confining attention to some alternatives over others. The cost of analysis, in time and money, must be considered in any attempt to make value-maximizing decisions. Analyzing all conceivable options is neither possible nor desirable. Setting an aspiration level and then randomly examining alternatives until one is found that satisfies the aspiration level, a method Herbert Simon (1976) called “satisficing” offers no guidance on which alternatives should be examined first. By contrast, incrementalism restricts attention to alternatives that are politically viable and for which policy analysis is most likely to be accurate.

2  A REALIST CASE FOR INCREMENTALISM 

19

Third, incrementalism facilitates majority building by minimizing disruption to established practices. As Karl Popper observed, any political reform will mobilize opposing forces in rough proportion to the scope of the reform. Therefore, minimizing the scope of the reform should minimize opposition (Popper, 1994, 62). Fourth, incrementalism facilitates learning from mistakes. Popper (a forerunner of Lindblom) distinguished between what he called holistic and piecemeal social engineering. Holistic social engineering—a form of rationalism—aims at “remodeling the ‘whole of society’ in accordance with a definite plan or blueprint.” By contrast, piecemeal social engineers focus instead on “small adjustments and readjustments which can be continually improved upon” (Popper, 1994, 67). Popper stressed the incompatibility of holistic social engineering with the scientific method. Attempts to remodel the whole of society change many things at one time; when a policy works imperfectly, as most policies do (Hall, 2014), it is impossible to determine which of the many changes in the holistic reform are responsible for the undesirable consequences. By contrast, piecemeal reforms make changes at the margins of existing policy, facilitating efforts to trace unanticipated consequences to their true causes (Popper, 1994, 66–67). Finally, the failure of any given step to solve a particular problem often makes the best case for taking the next step. In their study of air pollution policymaking in California, Krier and Ursin (1977) emphasized the critical role of crises, particularly smog disasters, in the policy process at both the state and federal levels. Although policy change is not always a response to crisis, occasional crises are inevitable in a system that by its very nature produces marginal responses to steadily worsening problems. Fortunately, dramatic performance failures make a powerful case for moving on to the next step, however unpalatable that step may have seemed previously. (For a recent defense of incrementalism identifying advantages different from those advanced here, see Berman & Fox, 2023.)

What Makes a Policy Non-incremental? Before examining the conditions under which non-incremental policy changes are possible and desirable, we must first clarify what quality or qualities make a policy outcome non-incremental. Lindblom answered

20 

M. T. HAYES

this question in two quite different ways. Early in his career, in a book written with David Braybrooke, he treated non-incremental change as a matter of degree rather than kind. He and Braybrooke advanced a four-­ quadrant model (reviewed later in this chapter) with two interacting dimensions. One was a continuum extending from incremental policy change at one end to non-incremental policy change at the other. The further one moves from left to right along the continuum, the less incremental a policy becomes (Braybrooke & Lindblom, 1970, 66–79). For example, raising the minimum wage by only a dollar would clearly be incremental. Similarly, a minimum wage of $15/hour would be less incremental than a minimum wage of $13/hour. However, for Braybrooke and Lindblom, there was no point along the continuum where a proposal to increase the minimum wage would definitively become non-incremental; whether a policy qualified as incremental or non-incremental was a matter of degree. Later in his career, Lindblom defined non-incremental change very differently. To qualify as non-incremental, a policy proposal would have to transform the economic or political system—for example, moving our economic system from communism to capitalism or moving from our system of checks-and-balances to a parliamentary system. Policy changes that fell short of systemic transformation would be merely “large increments” in Lindblom’s (1977) terms. Of course, defining major policy changes that fall short of systemic transformation as merely large increments helped Lindblom make his case that policymaking is typically incremental. Policy changes everyone would characterize as major that nevertheless fall short of systemic transformation can be readily identified: for example, the Civil Rights Act of 1964, the Voting Rights Act of 1965, the Clean Air Act of 1970, the 1986 tax reform act, and the Affordable Care Act, among others. But these large increments are the exception rather than the rule, and political scientists find it well worth investigating why our political system sometimes produces large increments rather than small ones. These two quite different definitions of non-incremental change can be reconciled by reformulating the Braybrooke and Lindblom continuum to range from incremental change at one end to systemic transformation at the other. As we move from left to right along this continuum, policies progress from small increments to larger ones. However, systemic transformations will be extremely rare.

2  A REALIST CASE FOR INCREMENTALISM 

21

When Is Non-incremental Policy Change Possible? While Lindblom clearly believed disjointed incrementalism was the normal case, he saw non-incremental policy changes as possible under exceptional circumstances. His most extensive treatment of this issue involved the four-quadrant model he developed with David Braybrook that I alluded to in the previous section. The four quadrants were derived from the interaction of two continuous dimensions: (1) the degree of change involved in a policy proposal, which ranged from incremental at one end to non-incremental at the other and (2) the degree to which policymakers understood the policy problem, which ranged from low understanding at one end to high understanding at the other (Braybrooke & Lindblom, 1970, 78–79). In quadrant 3 (incremental change/low understanding), incrementalism will be the norm. In quadrant 2 (incremental change/high understanding), rational decision-making is possible but narrowly confined to small, technical issues. Braybrooke and Lindblom predict that quadrant 1 (non-incremental change/high understanding) will be empty. However, systemic transformations may result in quadrant 4 (non-incremental change/low understanding), where economic and/or political institutions undergo cataclysmic change (what they call “wars, revolutions, and grand opportunities”). These transformative policies will necessarily go beyond the existing knowledge base. However, large policy increments that fall short of systemic transformation may be adopted under less cataclysmic circumstances. For example, in an earlier work with Robert Dahl (Dahl & Lindblom, 1963, 85), Lindblom suggested that policymakers may decide to take a “calculated risk” where an existing policy is too broken to fix, opting for non-incremental change despite a limited knowledge base. For example, the various “large increment” policy proposals advanced by President Biden in his first year in office would all qualify here. These calculated risks would also fit within this fourth quadrant (non-incremental change/low understanding). Large increments of this sort may also occur within quadrant 4  in response to an aroused public opinion. In a study of air pollution policymaking over a 30-year period, Charles O. Jones characterized the 1970 Clean Air Act as non-incremental. In his study, policymakers opted for what he called “speculative augmentation” in response to a “pre-formed majority.” A sharp increase in public concern over air pollution between 1967 and 1970 led to a bandwagon effect and policy escalation; clean air

22 

M. T. HAYES

legislation got progressively stronger at every stage of the legislative process, resulting in a significant augmentation in federal authority over the issue. This augmentation was speculative because it went beyond the available knowledge to enact technology-forcing legislation. While this suggests that non-incremental policy changes may be achieved in response to an aroused public opinion, Jones (1974, 438–464; see also Jones, 1975) cautioned that the 1970 act “legislated beyond capability.”

When Is Non-incremental Policy Change Desirable? The authors just cited all viewed non-incremental policy changes as incompatible with an adequate knowledge base, calling into serious question whether such major policy changes are even desirable. However, non-­ incremental policy changes can result from an increase in the available knowledge base after long years of experience implementing and modifying a policy. In two earlier works, I identified a life cycle of policies that consists of three distinct stages (Hayes, 2006, 153–158; see also Hayes, 2006, 125–127). This life cycle modified Lawrence D.  Brown’s (1983) distinction between breakthrough and rationalizing policies. Breakthrough policies occur when the federal government takes on a new responsibility. For example, the Social Security Act of 1935 involved the federal government in pensions for retirees for the first time. The passage of Medicare similarly created a program of health insurance for retirees. Examples abound since every federal program was a brand-new policy initially. By contrast, rationalizing policies are subsequent policies designed to fix problems that arise in implementing breakthrough policies. Rationalizing policies are typically incremental changes as advocated by Lindblom. It is not necessary for breakthrough policies to solve problems once-and-for-all. To the contrary, all policies fall short in some way: they may need more money, or to be redesigned, or to be extended to more people. Because Brown’s rationalizing policies are examples of what Lindblom terms seriality or successive approximations, I call them incremental rationalizing policies. In stage 1 of the life cycle, Congress enacts a federal role breakthrough. In stage 2, which can last for decades, incremental rationalizing policies are enacted in subsequent policy cycles. For some policies—but not necessarily all—a third stage follows. The available knowledge base should grow over time as legislators and

2  A REALIST CASE FOR INCREMENTALISM 

23

administrators gain experience with incremental rationalizing policies. The conditions for rational decision-making may eventually be met, as policymakers arrive at a consensus on objectives, and the knowledge base becomes consensual rather than conflictual. When this happens, the result is something I call a rationalizing breakthrough. A rationalizing breakthrough is a policy innovation that permits policymakers to address a longstanding problem in a new way that is both effective and efficient. Put another way, rationalizing breakthroughs are major policy changes that are driven by an increase in the available knowledge base. They do not involve legislating beyond capability; to the contrary, they legislate in response to increased capability. Legislation creating a market for emissions rights provides one example. Before 1991 clean air policies set air quality standards for new coal-­ fired plants and attempted to enforce them through various penalties. Polluters had an incentive to repair old plants (which did not have to meet the standards) rather than building new plants that would have to comply, so progress in reducing emissions was slow. Setting up a market where emissions rights could be bought and sold made it legal for plants to pollute, but polluters would now have to pay for the right to do so. The government could determine the total allowable amount of a particular pollutant by setting an overall limit on the number of emissions rights that would be sold in the market. Firms that had just built a new plant that met the new source performance standards could sell their emissions rights to other companies whose older plants failed to meet the standards. This made it impossible for companies to pollute the air for free, as they had before, creating instead an incentive for them to reduce emissions, which now cost them money. When the owners of a company decided a plant had reached the end of its useful life, they could build a new one that met the new source standards—at which point they could stop buying costly emissions rights and start selling them instead. With the rare exception of policies that are so unpopular with the public that they are terminated almost immediately—like President Reagan’s short-lived proposal for catastrophic health insurance, for example—most policies go through the first two stages of the life cycle, starting with a federal role breakthrough which is then followed by incremental rationalizing policies. There is no guarantee that any given policy area will ever make it to stage 3, but, where rationalizing breakthroughs do occur, they both work effectively and endure over time.

24 

M. T. HAYES

Conclusion To return to the vocabulary advanced in Chap. 1, idealists are naïve in asserting that public problems can be solved, once-and-for-all, through reason and rational decision-making. Rational decision-making requires both a consensus on objectives and a knowledge base sufficient to permit accurate estimation of policy consequences. When neither of these conditions are met, which is true most of the time, normal incrementalism is inevitable. When policymakers trapped in this cell attempt to pass a large-increment policy, they will be unsuccessful. Something resembling rational decision-making can occur where consensual knowledge is combined with consensual objectives, but most of the time the consensual objectives are a function of a restricted circle of decision-makers, and consensual knowledge reflects an agreement on basic facts that may later prove to have been inaccurate. By contrast, realists are correct in viewing incrementalism as inevitable under normal circumstances. Lindblom’s model of disjointed incrementalism describes how policymakers manage to produce good outcomes even though rational decision-making is precluded. The five elements identified earlier (no “given” problem, building on past policy, tapering down, a focus on incremental options, and policy modification via incremental rationalizing policies) are aids to calculation in a complex environment where the conditions for rational decision-making are unmet. In other words, there is a “science” to muddling through (Lindblom, 1959). Does all this mean that large increments (or even systemic transformations) are impossible? Lindblom answered this question differently in different works. Writing with David Braybrooke, he treated non-incremental policy changes as confined to a single quadrant: the realm of wars, revolutions, and grand opportunities. While exceptional circumstances of this sort are surely necessary for systemic transformations to occur, large increments may result from calculated risks where policymakers believe existing policies are too broken to fix. According to Charles O.  Jones, an aroused mass public can create a “pre-formed majority” demanding non-incremental change. As a result, policymakers will engage in a bandwagon effect and policy escalation, resulting in a significant expansion in federal capacity that goes well beyond the available knowledge base. Jones explicitly placed this policy change via a large increment in Braybrooke and Lindblom’s fourth quadrant, which combined non-incremental change with an inadequate knowledge base.

2  A REALIST CASE FOR INCREMENTALISM 

25

Policymaking by large increments does not have to go beyond the available knowledge base, however. Rationalizing breakthroughs can occasionally occur at the end of the life cycle of a policy. These rationalizing breakthroughs are the product of a gradual increase in the knowledge base over time combined with an emerging consensus on objectives. In short, the conditions for rational decision-making may be met after many years of experience with a policy, permitting enactment of an innovative new policy that addresses an old problem in a new and more effective way. Rationalizing breakthroughs would fall within the quadrant in the Braybrooke and Lindblom scheme combining major policy change with high understanding. These more effective and enduring forms of major policy change are the result of incrementalism, however, not an alternative to it.

References Berman, G., & Fox, A. (2023). Gradual: The Case for Incremental Change in a Radical Age. Oxford University Press. Braybrooke, D., & Lindblom, C. E. (1970). A Strategy of Decision. Free Press. Brown, L. D. (1983). New Policies, New Politics: Government’s Response to Government’s Growth. Brookings Institution. Burke, E. (1999). Reflections on the Revolution in France. In I. Kramnick (Ed.), The Portable Edmund Burke (pp. 416–474). Penguin Books. Dahl, R. A., & Lindblom, C. E. (1963). Politics, Economics and Welfare: Planning and Politico-Economic Systems Resolved into Basic Social Processes. Harper and Row, Harper Torch books. Hall, T.  E. (2014). Aftermath: The Unintended Consequences of Public Policies. Cato Institute. Hayes, M.  T. (2006). Incrementalism and Public Policy. University Press of America. Hayes, M.  T. (2022, March 23). Incrementalism and Public Policy-Making. Oxford Research of Politics. Ed. Oxford University Press. Incrementalism and Public Policy-Making | Oxford Research Encyclopedia of Politics. Jones, C. O. (1974). Speculative Augmentation in Federal Air Pollution Policy-­ Making. Journal of Politics, 36(May), 438–464. Jones, C.  O. (1975). Clean Air: The Policies and Politics of Pollution Control. University of Pittsburgh Press. Krier, J. E., & Ursin, E. (1977). Pollution & Policy: A Case Essay on California and Federal Experience with Motor Vehicle Air Pollution, 1940–1975. University of California Press.

26 

M. T. HAYES

Lindblom, C.  E. (1959). The Science of “Muddling Through”. Public Administration Review, 29(Spring), 79–88. Lindblom, C. E. (1977). Politics and Markets: The World’s Political-Economic Systems. Basic Books. Lindblom, C. E. (1980). The Policy-Making Process, 2nd ed. Prentice-Hall. Popper, K. (1994). The Poverty of Historicism. Routledge and Kegan Paul. Rothstein, R. L. (1984). Consensual Knowledge and International Collaboration: Some Lessons from the Commodity Negotiations. International Organization, 38(Autumn), 733–762. Simon, H. A. (1976). Administrative Behavior: A Study of Decision-Making Process in Administrative Organization. Free Press. Stimson, H. L., & Bundy, M. (1948). On Active Service in Peace and War. Harper. Wildavsky, A. (1979). Speaking Truth to Power: The Art and Craft of Policy Analysis. Little, Brown.

CHAPTER 3

Partisan Incrementalism

Abstract  Incrementalism can be adapted to our contemporary system of polarized parties through partisan incrementalism. Seriality works best for pure problems of knowledge base. Fortunately, where parties are homogeneous internally and far apart ideologically, the range of disagreement over goals within each party should be narrow, transforming conflictual issues into pure problems of knowledge base within one or both parties. Parties could compete for votes by offering alternative programs of incremental (or decremental) policy initiatives. This chapter illustrates how partisan incrementalism can work by applying the concept to the struggle over the 1996 welfare reform. Keywords  Aid to Families with Dependent Children • Contract with America • Partisan incrementalism • Pure problems of knowledge base • Seriality • Temporary Assistance for Needy Families Americans currently live in an age of polarized parties. Critics of our Madisonian system of checks-and-balances argued for issue-oriented parties that would offer alternative policies to voters as early as the 1940s (Schattschneider, 1942; American Political Science Association, 1950). In response, party activists in both parties sought to make the parties

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. T. Hayes, Incrementalism and Policymaking in the USA, https://doi.org/10.1007/978-3-031-38485-1_3

27

28 

M. T. HAYES

ideologically distinct, and they succeeded (Rosenfeld, 2018). For a variety of reasons, the parties have realigned along issue lines and their voter bases have followed along. Fortunately, incrementalism can be adapted to our partisan age through a process I call partisan incrementalism. This chapter will show what partisan incrementalism looks like in practice through application to a case study of the welfare reform of 1996, which abolished the Aid to Families with Dependent Children program and replaced it with a new system of block grants aimed at encouraging welfare recipients to work (Temporary Assistance for Needy Families, or TANF). Why look at a case from so long ago? First, poverty policy is intrinsically important as it determines how we will treat the nation’s poor, and the 1996 reform remains the most recent major legislative reform of welfare policy at the national level. Second, this case provides an excellent example of policy entrepreneurship (Price, 2021, 95–131; Walker, 1977). as the legislation originated as an effort by a moderate faction within the Republican Party in the House to strengthen the weak work requirements enacted in the most recent previous welfare reform. Third, it also provides an unusually clear example of responsible party government, as the moderates’ carefully negotiated bill was incorporated into the House Republicans’ Contract with America, which offered voters a clear-cut set of Republican policies that contrasted with Democratic policy approaches. Finally, the welfare reform case shows that partisan incrementalism can operate effectively even under conditions of divided government. Contrary to widespread belief, the 1996 welfare reform was driven primarily by a small group of House Republicans, who overcame factional disagreements over priorities to pass a strong bill in the House and then compromised with Senate Republicans to pass legislation that President Clinton would eventually sign. These House Republicans were so determined to achieve work-oriented welfare that they passed three different versions of the legislation, overcoming two Clinton vetoes, to achieve their legislative goals. The first section of this chapter will develop the concept of partisan incrementalism. The second section will explore at length how partisan incrementalism operated in the welfare reform case. The third section will focus on how the new law was modified when reauthorized in 2005. The concluding section will draw together the arguments offered here and briefly preview Chaps. 4 and 5.

3  PARTISAN INCREMENTALISM 

29

Partisan Incrementalism Defined Incrementalism was easier when the two congressional parties were both internally divided and ideologically overlapping, and in our current climate polarized parties yield gridlock far more often than normal incrementalism (Davis et  al., 2014). However, parties that offer contrasting positions on issues of the day will remain a feature of our politics for the foreseeable future. Is there a way to reconcile our Madisonian system of checks-and-balances with the development of responsible parties? Fortunately, there is. While responsible parties operating within our Madisonian system can produce gridlock, they also create at least the potential for a different kind of incremental policy-making that was impossible before—one that combines bargaining, compromise, and incremental outcomes with the increased electoral accountability that issue-oriented parties can give (Hayes, 2022). Incrementalism as a system of iterations or successive approximations only works well where there are no obstacles to what Lindblom called “seriality,” the gradual improvement of policies through subsequent policy cycles (Lindblom, 1979). As we saw in Chap. 2, seriality works best for “pure problems of knowledge base,” where policymakers agree on objectives but lack the knowledge base necessary to achieve them. The failure of a law to achieve a desired goal is much more likely trigger efforts to fix emergent problems through subsequent legislation where most participants agree on that goal. Where the two parties are internally homogeneous but far apart ideologically, as they are now, the range of disagreement over objectives within each party will often be narrow. This narrow range of disagreement on goals has the potential to convert problems combining conflictual objectives with conflictual knowledge into pure problems of knowledge base, at least internally. While bargaining and compromise will typically be necessary to secure complete agreement on policies within each party, compromises should be more easily reached within the parties than between the two parties. Where members of the majority can reach agreement on objectives, thus creating pure problems of knowledge base, policies that fail to work as intended should stimulate efforts to modify them through rationalizing policies in subsequent policy cycles. Moreover, as policymakers gain more experience with policies, the knowledge base may eventually improve sufficiently to permit development of rationalizing breakthroughs. As we saw

30 

M. T. HAYES

in Chap. 2, these rationalizing breakthroughs would result from expansions of the available knowledge base rather than venturing beyond it, and they would be products of incrementalism rather than alternatives to it. Under such a system, the party in control of government could develop incremental (or decremental) policies across a wide range of issues. Moreover, a majority party that can enact its legislative program could establish a positive record that appeals to voters, creating at least the potential to remain in power for an extended period. A sustained period in control of government would facilitate successive approximations that gradually converge on workable policies. For any of this to happen, however, the Senate filibuster would clearly have to be abolished or reformed; unified control of government means nothing if the minority party can block any legislation through a party filibuster. (The case for eliminating the Senate filibuster will be made in Chap. 6.) Partisan incrementalism is also possible under divided government, although it necessarily functions differently. In the case study that follows, President Clinton had campaigned on a promise to reform welfare by shifting its purpose from income maintenance to the promotion of work. While the President’s party controlled both Houses of Congress for his first two years, the Democrats were badly split on the issue; ironically, the Republicans in the House and Senate were more receptive to the President’s ideas than his own party was. Recognizing the opportunity to enact a welfare reform to their liking, a core group of House Republican leaders and staffers attempted to develop a bill of their own that would unite their party while attracting support in Congress from moderate Democrats. When the Republicans gained control over both the House and Senate in the 1994 midterm elections, creating divided government, this core group of policy entrepreneurs no longer needed Democratic support in Congress for their initiative. The only Democrat whose approval they still needed was President Clinton—if the Republicans could maintain party unity. Fortunately, the range of views on welfare reform was sufficiently narrow within their party to convert the issue into a pure problem of knowledge base for the Republicans. The remainder of this chapter will review the operation of partisan incrementalism in this case. Chapter 4 will reexamine this same case to show how all the elements of normal incrementalism were also manifest throughout. Achieving and maintaining party unity would permit House Republicans to pass bills that were stronger than the President wanted but

3  PARTISAN INCREMENTALISM 

31

building a majority within their own party proved far from automatic. Republican reformers were forced to taper down proposals from what they considered optimal to what would attract enough votes to pass bills in both houses, and it took passage of three different bills to finally secure the President’s signature.

Welfare Reform as a Case Study of Partisan Incrementalism In February 1991, a group of moderate Republicans in the House of Representatives chose welfare reform as an issue on which they might offer a distinctively Republican alternative to Democratic policies. The Wednesday Group, an informal caucus of moderate Republicans, asked Ron Haskins, chief staffer to Clay Shaw (R-FL), the chair of the Ways and Means subcommittee on human resources, to draft a report on existing welfare programs that might lead to legislation (Haskins, 2006, 24–26). The following year, a bill promoting work-oriented welfare reform was introduced in response to Haskins’ report (Haskins, 2006, 30–31). Prospects for enacting some version of welfare reform improved in 1992 when Bill Clinton won the presidency after campaigning, in part, on the need to “end welfare as we know it” (Haskins, 2006, 14). Clinton’s victory created at least the possibility that congressional Republicans could draft a welfare reform bill that a Democratic president might sign. Because Republicans were in the minority in the House, and had been for four decades, these House Republican moderates knew they would need to draw enough support from moderate Democrats to pass both the House and Senate. The House Republicans Move First In February 1993, President Clinton promised to appoint a task force to develop a welfare proposal. However, he decided to move on health care reform before taking up welfare, and no welfare task force members were appointed for several months (Haskins, 2006, 37). In May, an administration spokesperson asked Congress for a two-year delay in implementing the 1988 JOBS program, which was part of the 1988 Family Support Act. A bipartisan compromise, the 1988 act expanded welfare to previously excluded male heads of households while offering weak incentives to

32 

M. T. HAYES

encourage welfare recipients to seek jobs. Republicans already doubted congressional Democrats’ commitment to promote work as an alternative to welfare, and the administration’s request for a delay reinforced these suspicions. A small. core group of House Republican members and staffers decided to move ahead without waiting for the President’s bill to emerge from a task force he seemed in no hurry to appoint. Meanwhile, welfare rolls had recently begun rising sharply, suggesting strongly that the JOBS program, in its current form, was too weak to be effective (Haskins, 2006, 40). Months later, a group of conservative Republicans joined the legislative struggle, dismissing the importance of promoting work and calling instead for a primary focus on the need to reduce illegitimacy. This strong challenge forced the moderates to negotiate a revised version of the bill incorporating policies aimed at illegitimacy. But any such accommodations would necessarily make the bill less appealing to moderate Democrats and the president, and it seemed for a time as if the effort would stall (Haskins, 2006, 65–77). However, welfare reform was reinvigorated when Newt Gingrich incorporated it into his ten-item Contract with America going into the 1994 midterm elections. When the Republicans captured both the Senate and the House that year, they no longer needed to attract moderate Democrats. Rather, they could now pass a bill to their liking, send it to the President, and put him in the uncomfortable position of having to sign the bill or break an important campaign promise to “end welfare as we know it” (Haskins, 2006, 70–84). For House Republicans, welfare reform was a pure problem of knowledge base. Although they differed over priorities and strategies, they all agreed on the need to reform welfare. Both factions agreed on the desirability of moving recipients into work as well as on the need to reduce illegitimacy even though they disagreed fiercely over which aspect of the problem to emphasize. Partisan incrementalism was possible because the two Republican factions could negotiate their differences, reach a compromise, and produce a bill that addressed both problems simultaneously. By contrast, Democrats were badly divided on the issue. As noted earlier, most congressional Democrats rejected Clinton’s proposals to put a time limit on welfare and make work pay, preferring instead to preserve welfare as a cash entitlement. President Clinton never succeeded in uniting his party on the issue; as a result, he would be forced to decide whether to sign or veto bills advanced by congressional Republicans.

3  PARTISAN INCREMENTALISM 

33

The big question for House Republicans initially was how to achieve these goals while producing a bill the President would sign. While a majority of House and Senate Democrats wanted to preserve welfare as a cash entitlement program, there were moderate Democrats in both houses who might be receptive to work-oriented welfare reform. However, after the 1994 midterm elections, the only Democrat whose opinion mattered was President Clinton. As the majority party in the House and Senate, Republicans no longer needed the votes of moderate Democrats to pass legislation. To the contrary, if they maintained party unity, Republicans could now control the content of their bills. Block grants were now possible where they were not before. While Republicans embrace block grants because they give states more freedom, Democrats disagree, fearing that states (especially in the South) will use that freedom to adopt policies that would disadvantage minority groups. When the Democrats controlled both ends of Pennsylvania Avenue, block grants were a nonstarter. Immediately after the election, however, Haskins asked the Congressional Research Service and the General Accounting Office for a review of eight domains of social policy (cash welfare, nutrition, employment and training, childcare, foster care and adoption, social services, housing, and medical). The CRS and GAO identified 336 programs that together cost $375 billion in 1994. Haskins and other House Republicans proposed to consolidate these 336 programs into eight large block grants. This ambitious attempt to convert domestic social policies into block grants was a major departure from normal incrementalism. Certainly, devolution of responsibility for social programs to the states was viewed as revolutionary by its proponents: a first step toward a “true federalist revolution” that would get the federal government out of the social policy business entirely (Haskins, 2006, 84–85). Republican proposals to convert both Food Stamps and Medicaid into block grants were particularly controversial. Early in 1995 the House Ways and Means Committee held a series of hearings at both the full committee and subcommittee levels; these hearings were carefully designed to convince both the public and rank-and-file Republicans that the AFDC program was badly broken, and that work-­ oriented welfare reform was the way to fix it (Haskins, 2006, 106–109). (Hearings on the proposals to block grant various other social programs were held in different House committees.) Markups followed in late February and early March at both the subcommittee and full committee

34 

M. T. HAYES

levels. Republicans prevailed in a series of votes by maintaining almost complete party unity (Haskins, 2006, 135–164). After being reported out of committee, the bill went to the Rules Committee where Republicans (again prevailing on a series of party line votes) chose to permit a number of amendments, including two Democratic alternatives: one that would preserve welfare as a cash entitlement and the other a more moderate bill that differed only slightly from the House Republican version (Haskins, 2006, 165–171). After prevailing on every amendment, as well as the motion to recommit, House Republicans passed the bill in late March, with all votes closely following party lines (Haskins, 2006, 189–193). Welfare Reform Moves to the Senate Although the Senate was also controlled by the Republicans, the path to enactment there would be more challenging. Because they represent entire states, which are typically much more heterogeneous than House districts, Senators tend to be more moderate than their House counterparts. Although a number of very conservative new members had come over from the House in recent years, the Senate Republican caucus remained much more moderate than its House counterpart, and a number of the more extreme House proposals (big spending cuts and at least some of the block grant proposals, for example) could not pass the Senate because several key Senate committees were chaired by Republican moderates. Particularly problematic were the Senate Agriculture Committee, chaired by Richard Lugar (R-IN), which had authority over both Food Stamps and child nutrition programs, and the Senate Finance Committee, where moderate Republican John Chafee (R-RI) was a powerful member. Finally, the Senate Majority Leader, Robert Dole (R-KS) was a notorious moderate, and his chief of staff, Sheila Burke, was widely perceived as a liberal by conservative House Republicans. An experienced and extremely skillful legislator, Dole was committed to pursuing legislation with broad bipartisan majorities and had an excellent working relationship with the Senate Democratic leader, Tom Daschle (Haskins, 2006, 194–226). Welfare reform originated in the Senate Finance Committee, but Finance Committee Chair Robert Packwood produced a bill that was far too accommodating to Democrats and Republican moderates to be acceptable to House or Senate conservatives. Dole took control of the bill

3  PARTISAN INCREMENTALISM 

35

in August and, through an impressive series of negotiations with Senate Democrats, conservative Republicans, and Republican moderates, produced a bill that had strong bipartisan support. On September 19th, the Senate passed its version of welfare reform by a vote of 87-12 (Haskins, 2006, 205–226). A conference committee was created to reconcile differences between the House and Senate versions. Republican conferees from both houses realized they could control the outcome of the conference committee by outvoting the Democratic members if they were successful in maintaining party unity. For this reason, the House and Senate Republican conferees began meeting together in advance of the conference, striving to reach agreement on the numerous differences between the House and Senate versions (Haskins, 2006, 227–230). During this stage, the more extreme provisions of the House bill were jettisoned; at the same time, House Republican conferees communicated to their more moderate Senate counterparts which of the Senate provisions would doom the bill in the House. Budget Considerations Intrude and Clinton Vetoes Two Bills Before the conferees could reconcile all the differences between the two bills, welfare reform was overtaken by the budget process. As part of the Contract for America, House Republicans had promised to balance the federal budget while, at the same time, enacting tax cuts for married couples and the middle class that would make the deficit worse. By December, budget reconciliation was in full swing, and because welfare reform (in whatever form it emerged) would have budgetary implications, it was inevitably included in the reconciliation process. The budget negotiations were made more complicated by the need to raise the debt ceiling, a problem that also arose late in the year. Clinton vetoed two attempts at a debt ceiling extension as well as the first version of the reconciliation bill, which now included welfare reform. Fortunately for Clinton, the public blamed Republicans rather than the Democrats for the two resulting government shutdowns (Haskins, 2006, 227–253). After the veto of the reconciliation bill (which necessarily included only an abridged version of welfare reform confined to provisions with a significant fiscal impact), House and Senate Republican conferees resumed their negotiations. In late December, the conferees finally reached agreement

36 

M. T. HAYES

and the full House and Senate each passed the conference report, sending a second version of welfare reform to the President. However, on January 9th President Clinton announced his veto of this second welfare reform bill as well (Haskins, 2006, 254–267). By all rights, two presidential vetoes should have brought an end to welfare reform, at least for that Congress, but the core proponents of the bill in the House (Clay Shaw, Haskins, and other staffers) were unusually determined. The most promising path to success would have been for the House to hold its nose and pass the Senate version of the bill, which President Clinton had indicated during negotiations that he could sign. That unpalatable prospect was rendered unnecessary, however, when the nation’s governors, at the annual meeting of the National Governors Association, held a press conference announcing that they had reached bipartisan agreement on the main elements of a welfare reform bill (Haskins 268–275). The Governors’ Proposal Revives Welfare Reform The governors’ press conference resurrected welfare reform. Although there were elements in the governors’ plan that would never be acceptable to House Republicans, and any modifications made to that plan to make welfare reform acceptable to that group would forfeit the support of the Democratic governors, Haskins, Shaw, and others started all over again. Using the vetoed version of welfare reform as their baseline, House Republicans modified the bill in negotiations that included House conservatives as well as moderates, Senate Republicans, members of the administration, and the Republican governors. Among the GOP governors, John Engler’s (R-MI) approval was especially vital as he served as chair of the Republican Governors’ Association. Along with Tommy Thompson of Wisconsin, Engler was also one of the two most prominent Republican governors who had pioneered work-oriented welfare reforms at the state level (Haskins, 2006, 268–287). Over time, extreme provisions in the House version were modified or dropped. For example, the Senate blocked cuts in the Earned Income Tax Credit that had passed the House and inserted a maintenance of effort provision (MOE) that prevented states from reducing what they were currently spending on welfare. Over the summer, House Republicans reluctantly acknowledged that converting Medicaid to a block grant would

3  PARTISAN INCREMENTALISM 

37

doom the entire bill, so that provision was finally jettisoned. On July 18th, the House passed their version of welfare reform; the Senate followed on July 23rd. At this point, the process paused while Democrats in both houses waited for the President to say whether he would sign the new bill before deciding how to cast their votes. On August 1st, after a long internal debate among his advisers, President Clinton announced that he would sign the third version of welfare reform. The House passed the final version on August 21st, and the Senate followed on August 22nd. With President Clinton finally on record, the bill received substantial Democratic support in both Houses. The President signed the bill into law later that same day (Haskins, 2006, 314–331). After two unsuccessful attempts, congressional Republicans had finally produced a version of welfare reform acceptable to the President. House Republican moderates succeeded in converting AFDC into a block grant and establishing strong work requirements for the first time. However, President Clinton would get the lion’s share of the credit for what most journalists and scholars now routinely call “President Clinton’s Welfare Reform,” and President Clinton got to keep his campaign promise to “end welfare as we know it,” removing from the 1996 campaign an issue that might have helped the Republican presidential candidate, Bob Dole.

Modification of the Law Through Reauthorization A major advantage of partisan incrementalism, as I argued above, is that a political party committed to a particular program of reform would be more committed to fixing problems that arise during policy implementation. Unfortunately, however, the incentives facing intellectuals who participate in the welfare policy community are quite distinct from those facing elected officeholders. (For a discussion of the relationship between academic experts and elected officials see Hayes, 2006, 134–136.) Implementation of TANF was subjected to a wide variety of policy evaluations by individual academics as well as professional research organizations and think tanks. A study of welfare policy changes in the George W.  Bush administration identified four potential issues to be addressed when the 1996 law was reauthorized: (1) Barriers to employment (lack of skills, mental health disorders, and so on) were greater for those still on welfare, as previous caseload reduction reflected movement off welfare by those who were most employable. (2) Support services were not uniformly

38 

M. T. HAYES

available across communities and neighborhoods. (3) TANF was not countercyclical; time limits, work requirements, and federal work participation benchmarks made it difficult for states to expand caseloads during economic downturns. And, finally, (4) programs aimed at reducing nonmarital birthrates and promoting two-parent families were unsuccessful, just as Haskins expected (Allard, 2007). These emergent problems were not addressed by the reauthorization, however, as the President and Congress had concerns quite different from academic welfare policy specialists. Although the 1996 law was up for reauthorization in 2002, Congress and the President could not reach agreement until 2005, when the law was finally reauthorized as part of the budget reconciliation process. The Bush administration succeeded in defining what counted as work more narrowly, eliminating loopholes in work requirements, and increasing the percentage of welfare recipients that must be working. Funding was added for programs to support the formation of two-parent families, and President Bush switched the emphasis toward counseling programs designed to prepare couples for marriage and parenthood However, the administration showed little interest in improving access to support services or making TANF countercyclical. Rather, their emphasis was on tightening work requirements, and the states lost a great deal of flexibility (Allard, 2007; see also Riccucci, 2005, 122–124). In sum, the Bush administration’s indifference to the consequences of policies for the target population at the implementation stage strongly suggests that the administration’s primary goal in implementing work-­oriented welfare reform was to save money rather than to improve the lives of the poor. Conflict over these two quite different objectives occurred throughout the process. As states experimented with innovative approaches to welfare reform in the 1980s and 1990s, some governors favored work-­oriented welfare reform primarily to save money by moving welfare recipients into work and diverting new applicants from welfare altogether. At the same time, however, others (like Tommy Thompson of Wisconsin) genuinely wanted to help the poor by facilitating their economic independence. A president with this second vision of welfare could have used reauthorization to address these unanticipated consequences of the original legislation. The idea of incremental policies as experiments from which we can learn should not be rejected, however. As the following chapter will demonstrate, this learning process can take time, and issues typically move through a life cycle. The Social Security Act of 1935 (a federal role breakthrough) was followed by six decades of incremental rationalizing policies

3  PARTISAN INCREMENTALISM 

39

that eventually culminated in a consensus within the welfare policy community on the desirability of promoting work. While work-oriented welfare may not prove to be the definitive answer to the problem of poverty, welfare policies have been modified repeatedly over several decades as policymakers experimented with a variety of incremental rationalizing policies. In this regard, the negative income tax, advocated by Milton Friedman (1962), can be categorized as a rationalizing breakthrough. Within the welfare policy community, a consensus had emerged in the late 1960s on the criteria by which welfare reforms were to be judged: the welfare economics principles of efficiency, horizontal equity, and vertical equity. (Policies that take the economy closer to its full productive potential provide efficiency benefits. Horizontal equity requires identical cases to be treated alike, while vertical equity requires aid to be given in proportion to need.) In theory, at least, the negative income tax would have increased incentives for welfare recipients to work, thus taking the economy closer to its maximum potential. Even more impressive, a negative income tax featured textbook horizontal and vertical equity. Families of identical size and income would receive the same level of benefits regardless of where they lived, and the net federal contribution to recipients’ disposable income under the plan would be directly proportional to need. Unfortunately, the negative income tax fell victim to two serious objections. First, the criteria by which any policy will be evaluated includes overall cost. Setting the income floor high enough to hold harmless welfare recipients in the states with the highest AFDC benefits would make the program prohibitively expensive, while setting the floor at a more affordable level (half the poverty line in Friedman’s version) would significantly reduce benefits for that portion of the welfare population that was successfully mobilized politically. Second, Friedman saw the negative income tax as an alternative to existing welfare programs, not as a supplement to them. President Nixon was unwilling to spend the political capital necessary to abolish the existing welfare system (Moynihan, 1973; Burke & Burke, 1974). While President Carter was willing to do so, he could not avoid the problem of overall cost identified above. As he quickly discovered, eliminating the entire welfare system would not yield enough money to fund a negative income tax that would hold harmless beneficiaries from high-benefit states (Lynn & Whitman, 1981). Despite the defeat of negative income tax proposals in two administrations, and its abandonment by subsequent administrations, the proposal still qualifies as a rationalizing breakthrough. If policymakers could go

40 

M. T. HAYES

through a time machine to 1935 and replace the welfare programs established in the Social Security Act with a negative income tax, it is at least possible that the subsequent problems plaguing welfare could have been avoided.

Conclusion In recent years, the development of ideological, polarized parties has subjected our political system to severe stress, calling into question whether responsible parties are in fact compatible with our Madisonian system. Certainly, incrementalism as originally conceived by Lindblom was easier to implement where both political parties featured liberal and conservative wings, creating more opportunities for bipartisanship. Can incrementalism be adapted to operate effectively in an era of highly polarized parties? I have argued here that it can. Where the parties are each ideologically homogeneous, the narrow range of views within each party creates the potential to convert contentious issues into pure problems of knowledge base. This is important because seriality operates more effectively for pure problems of knowledge than within any of the other decision environments. This makes possible an innovative approach to incrementalism, which I have here called partisan incrementalism, in which the two parties compete by offering alternative incremental proposals for solving pressing public problems. We can now draw four conclusions from the case study. First, the welfare reform of 1996 provides an example of partisan incrementalism, as President Clinton advanced a plan to “end welfare as we know it” by incrementally tightening the JOBS program, while the Republicans countered with a larger increment that would do more to make work requirements effective. In this case, however, the Republicans were able to overcome their narrow differences and maintain party discipline while the Democrats were badly divided. President Clinton forfeited the initiative on the welfare reform issue because he was unable to unite his party around the kind of work-oriented reforms that he favored. As a result, the Republicans effectively controlled the policy formulation process, producing a series of alternative bills until the President finally found one acceptable. But his ability to veto bills forced the Republicans to taper down their bills. Second, the 1996 welfare reform also shows that partisan incrementalism can work under conditions of divided government. In this case, a

3  PARTISAN INCREMENTALISM 

41

bipartisan welfare result was possible because President Clinton and the House Republicans shared an interest in reforming welfare along similar lines. The only real issue was how strong the final bill would be. President Clinton advocated reform that was a small increment; Republicans—especially House Republicans—wanted a much larger increment. As we will see in the following chapter, the ambitious House Republican bill had to be “tapered down from the optimal to the acceptable” throughout. The bill was perceived as “revolutionary” by its proponents primarily because it finally enacted strong work requirements rather than incrementally tightening the weak incentives to work included in the JOBs program. Third, the welfare reform case also illustrates what normal incrementalism looks like in practice, as reform proponents not only built on past policies but were forced to engage in bargaining and compromise throughout. Chapter 4 will review the core elements of incrementalism and show how each of them was manifested in the welfare reform case. Finally, while partisan incrementalism should facilitate learning from mistakes, the major shortcomings of the new law, as identified by subsequent policy evaluations, were disregarded by President Bush and the Republican Congress when TANF was reauthorized in 2005. The reasons for this gap between policy learning and legislative fixes will be discussed in the opening sections of Chap. 5, which explores how political inequalities across groups affect the operation of incrementalism. As noted in Chap. 2, many of the same interests that were involved in the policy formulation process will be represented in the policy evaluation process as well, guaranteeing that policy evaluation will be a political process rather than a rational one.

References Allard, S.  W. (2007). The Changing Face of Welfare During the Bush Administration. Publius, 37(3), 304–332. American Political Science Association, Committee on Political Parties. (1950). Toward a Responsible Party System. Holt, Rinehart, and Winston. Burke, V. J., & Burke, V. (1974). Nixon’s Good Deed: Welfare Reform. Columbia University Press. Davis, T., Frost, M., & Cohen, R. (2014). The Partisan Divide: Congress in Crisis. Premiere. Friedman, M. (1962). Capitalism and Freedom. University of Chicago Press. Haskins, R. (2006). Work Over Welfare: The Inside Story of the 1996 Welfare Reform Law. Brookings Institution.

42 

M. T. HAYES

Hayes, M.  T. (2006). Incrementalism and Public Policy. University Press of America. Hayes, M. T. (2022). Incrementalism and Public Policy-Making. Oxford Research Encyclopedia of Politics. Ed. Oxford University. Incrementalism and Public Policy-Making | Oxford Research Encyclopedia of Politics. Lindblom, C. E. (1979). Still Muddling, Not Yet Through. Public Administration Review, 39(November/December), 517–526. Lynn, L. E., Jr., & Whitman, D. D. (1981). The President as Policy-Maker: Jimmy Carter and Welfare Reform. Temple University Press. Moynihan, D.  P. (1973). The Politics of a Guaranteed Income: The Nixon Administration and the Family Assistance Plan. Random House, Vintage Books. Price, D.  E. (2021). The Congressional Experience: An Institution Transformed (4th ed.). Routledge. Riccucci, N. M. (2005). How Management Matters: Street-Level Bureaucrats and Welfare Reform. Georgetown University Press. Rosenfeld, S. (2018). The Polarizers: Postwar Architects of Our Partisan Era. University of Chicago Press. Schattschneider, E. E. (1942). Party Government. Rinehart. Walker, J. L. (1977). Setting the Agenda in the U.S. Senate: A Theory of Problem Selection. British Journal or Political Science, 7(October), 390–406.

CHAPTER 4

Welfare Reform as Normal Incrementalism

Abstract  While the 1996 welfare reform was viewed as revolutionary by its proponents, the legislative struggle over passage of the law nevertheless illustrated all the central elements of normal incrementalism. This chapter will show that the welfare reform case exhibited the following characteristics: (1) policy-makers engaged in remediality but were not presented with a given problem, (2) to maximize available information, they built on past policies, (3) they confined their attention primarily to incremental alternatives—and where they considered non-incremental reforms they were unsuccessful, (4) they were forced to bargain and compromise with multiple stakeholders throughout the process, and (5) the outcome was at best a large increment. Keywords  Build on past policies • Focus on incremental alternatives • No given problem • Policy succession • Policy termination • Taper down proposals from the optimal to the acceptable This chapter will reexamine the 1996 welfare reform case, focusing this time on the elements of normal incrementalism. As outlined in Chap. 2, disjointed incrementalism has five major elements: First, policymakers do not try to maximize abstract positive values but rather seek to mitigate or ameliorate problems without ever fully solving

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. T. Hayes, Incrementalism and Policymaking in the USA, https://doi.org/10.1007/978-3-031-38485-1_4

43

44 

M. T. HAYES

them. In so doing policymakers are not typically faced with a “given” problem (Lindblom, 1980, 24). Lindblom used the term remediality to refer to this process. Second, policymakers take full advantage of available information by building on past policies. They do not engage in holistic social engineering that ignores what has been done in the past. In Burke’s (1999) terms they engage in reform rather than innovation. Third, policymakers typically confine their attention to proposals differing only incrementally from existing policies. Where non-incremental proposals are advanced, they will typically be discarded or weakened along the way. Fourth, policies necessarily emerge from a process of bargaining and compromise. To secure the votes necessary for passage, policy proponents must “taper down” proposals from the optimal to the acceptable (Jones, 1974). Finally, under incrementalism policymakers converge on good policies by modifying programs through successive policy cycles. Lindblom termed this process of convergence through successive approximations seriality. The first four sections of this chapter will demonstrate that there was conflict over problem definition on the welfare reform issue, that policymakers built on previous policy, that the focus was primarily on incremental alternatives, and that tapering down was required throughout. Because seriality was addressed in the previous chapter, that discussion will not be repeated here. Instead, the fifth section of this chapter will focus on whether the final legislative outcome was incremental or non-incremental.

Conflict over Problem Definition According to Lindblom, policymakers will strive to mitigate problems rather than pursue abstract positive values. While most scholars and policymakers agree that poverty is a fundamental problem that should be reduced, they disagree strongly over how best to do it. Conflict arises over how to define the problem as well as which potential causes should be prioritized. In short, as Lindblom predicted, there is no “given” problem. While census data on poverty point to many distinct factors contributing to poverty including place of residence, age, race, marital status, and work experience during the year, two factors stand out as particularly important: failure to work full-time, and failure to marry or stay married

4  WELFARE REFORM AS NORMAL INCREMENTALISM 

45

(Mead & Bane, 2003, 58–64). According to Haskins, social science research strongly establishes both the importance of work and the adverse effects of welfare on work. However, research on whether welfare contributes to illegitimacy is less compelling. Although a substantial literature on the topic exists, the results are mixed, permitting Democrats to argue that welfare does not cause illegitimacy while Republicans argue that it does (Haskins, 2006, 6–7). Moreover, in Haskins’ view (2006, 8), no one has any workable ideas on how to promote marriage; by contrast, a great deal of research exists on what approaches are most effective in encouraging recipients to work. For these reasons, Haskins and the core group of House Republicans working on welfare reform emphasized work rather than illegitimacy in their initial proposals. In 1992 the Weber-Shaw bill was introduced, which included various elements drawn from Haskins’ report to the Wednesday group: a mandatory work requirement, specification of a minimum number of hours of work, and mandatory sanctions for those who refused to work (Haskins, 2006, 30). (Vin Weber, a prominent conservative member from Minnesota, was co-chair of the Wednesday Group; E. Clay Shaw was Haskins’ boss as chair of the Ways and Means Committee’s Subcommittee on Human Resources.) As we saw in the previous chapter, however, in January 1992, conservatives both inside and outside of Congress mounted an attack on the Weber-Shaw approach, minimizing the importance of work and emphasizing instead the alleged centrality of illegitimacy as the primary cause of poverty and welfare dependence. Prominent figures from conservative think tanks (Bill Bennett and Jack Kemp from Empower America, Robert Rector from the Heritage Foundation) were joined by conservative intellectuals like Charles Murray, George Will, and Charles Krauthammer, among others (Haskins, 2006, 41). An alternative bill stressing the need to address illegitimacy was introduced in the House by Jim Talent (R-MO) and Lauch Faircloth (R-NC). Conservative proponents dubbed this the “Real Welfare Reform Bill” and characterized moderates as being opposed to real change because true reform, in their view, must be radical, not incremental (Haskins, 2006, 69). There was a parallel conflict here over substantive vs. political priorities: while the moderates favoring work-oriented reform wanted to produce a bill the president would sign, improving current law by establishing a strong work requirement, conservatives emphasizing illegitimacy were less interested in writing a bill that would pass. To the contrary, they wanted

46 

M. T. HAYES

to sharpen the differences between the parties going into the next election (Haskins, 2006, 43–44). The moderates would have to accommodate the conservatives to maintain Republican unity on welfare reform, but in so doing they would have to move the bill further to the right, making it less attractive to moderate Democrats and the President.

Building on Past Policy The American welfare state began with the Social Security Act of 1935; there were no national-level income maintenance programs prior to that act. Written during the Great Depression, when unemployment levels were historically unprecedented, the Social Security Act created programs to aid four categories of people who were judged to be in need through no fault of their own: (1) Aid to Dependent Children (ADC—later to be renamed Aid to Families with Dependent Children), which was targeted at widows; (2) Aid to the disabled; (3) Aid to the blind; and (4) Aid to the aged poor. The states were free to cover any additional categories of people through a fifth program called General Assistance, but they would do so at their own expense. While the federal government would provide more than half the funds for each of the first four programs, states had to pay 100% of the costs of any groups covered under General Assistance. Over the years, this welfare state expanded to include a variety of additional programs, including Food Stamps and Medicaid. By 1991, when the Wednesday Group first asked Ron Haskins to develop a distinctly Republican approach to welfare reform, the welfare state could be divided into three tiers: (1) social insurance programs providing universal entitlements, including Social Security, Medicare, Disability Insurance, and Unemployment Compensation; (2) the largest entitlement programs targeted at the poor—AFDC, Medicaid, Food Stamps, and SSI (Supplemental Security Income, a program enacted in 1970 that combined aid to the disabled, aid to the blind, and aid to the aged poor into a single program with a uniform benefit level); and (3) a few hundred smaller programs for poor and low-income families and individuals that competed with one another for limited congressional appropriations (Haskins, 2006, 4–5). Medicaid was created along with Medicare in 1965 (Marmor, 1973). Food Stamps was created as a pilot program in 1964 and nationalized under President Nixon in 1971 (Kotz, 1969). Because spending on these two programs remained low initially, from 1935 through the 1970s the term “welfare” was reserved exclusively for AFDC, and the “welfare crisis”

4  WELFARE REFORM AS NORMAL INCREMENTALISM 

47

that policymakers perceived in the 1960s resulted from unanticipated increases in AFDC rolls over time (Steiner, 1966; Derthick, 1970). In 1962 the Kennedy administration attempted to mitigate this steadily worsening problem by significantly expanding the number of social workers (Steiner, 1966). In 1969, with welfare rolls again rising sharply, Richard Nixon introduced the Family Assistance Plan—a form of negative income tax that would guarantee half the poverty level to all recipients while allowing them to keep fifty percent of whatever they earned. Although FAP passed the House twice, it was defeated in the Senate both times (Moynihan, 1973; Burke & Burke, 1974). Jimmy Carter also proposed a negative income tax in 1977, but his version (the Program for Better Jobs and Income) never even made it to the floor of the House (Lynn & Whitman, 1981). Following these defeats of the negative income tax, a consensus gradually developed, among both liberals and conservatives within the welfare policy community, on the importance of promoting work as a strategy for escaping poverty. Work-oriented reformers at the national level were able to learn from a series of policy experiments conducted at the state level via the waiver process (Handler, 1995; Mead, 2004). As a result of these experiments, policymakers possessed a good deal of information on the potential effectiveness of various strategies for moving people from welfare to work. Where the negative income tax tried to motivate welfare recipients to work through incentives—under the NIT formula recipients would always be better off working than not—a consensus gradually emerged within the scholarly community that no real progress would be made without vigorously-­enforced work requirements. At one extreme was Charles Murray (1984), who called for the complete abolition of welfare; in his view, welfare encouraged both illegitimacy and long-term dependency because it had become more remunerative than low-wage work. For its part, the Reagan administration rejected the idea that government owned the problem of motivating people to work; instead, the administration made major cuts in welfare programs, placing the primary responsibility for family support on individual breadwinners, where it had traditionally fallen. As Nathan Glazer described it (1988, 44), welfare would once again become a form of charity, rather than an entitlement, with benefits that would be both limited and strictly administered. Ideally, people who could work would prefer work to welfare, even if work did not provide more income than welfare.

48 

M. T. HAYES

Another voice on the right, Lawrence Mead, wrote a series of volumes calling for welfare to shift away from its focus on entitlement to a new emphasis on work. Welfare recipients, according to Mead, share the same aspirations as other Americans but are too discouraged to achieve them. Government’s role should be to help the poor move from welfare to work—and thus into the economic and political mainstream—through a paternalistic approach, “helping and hassling” the poor as needed to help them succeed (Mead, 1986, 1992, 1997, 2004). Finally, in 1986 Michael Novak and Douglas Besharov of the American Enterprise Institute organized a seminar (1987) involving twenty scholars specializing in poverty policy. This working group, which included experts from both the center-­ right and the center-left, eventually issued a report that emphasized welfare dependency as a genuine problem to be addressed and called for government to help people find work and supplement their income. This emphasis on the need to encourage work was not limited to scholars on the right. David Ellwood (1989) called for time limits on welfare and a commitment by the federal government to guarantee a job for recipients who reached the time limit. Christopher Jencks of Harvard called on liberals to transform AFDC so that it reinforced, rather than undermining, American ideals about work and marriage (Haskins, 2006, 15). Finally, Mickey Kaus (1992) called for welfare to be replaced by a guaranteed jobs program; people who refused work would receive no aid. Given this evolving consensus on the importance of work as a solution to poverty and welfare dependency, it is not surprising that the next major welfare reform emphasized work. The 1988 Family Support Act established the Job Opportunities and Basic Skills program (JOBS). The JOBS program was a small increment at best. Single parents (about 95% of the caseload) were exempted from work requirements. While states were given mandatory participation standards, participation was defined extremely broadly to include education, training, and even job search. Although the legislation provided $1 billion/year to help adult recipients leave welfare, six years after the enactment of the bill less than 1% of welfare recipients were either working or enrolled in job search programs. A compromise between Republicans and Democrats, the law also expanded welfare by requiring states to provide AFDC to two-parent households in which a parent was unemployed (Haskins, 2006, 12–13). Because of the law’s weak work requirements, Republicans inferred that Democrats were not really serious about enforcing work, and the Clinton administration’s May 1993 request for a two-year delay in

4  WELFARE REFORM AS NORMAL INCREMENTALISM 

49

implementing work requirements under the JOBS program convinced House Republicans to start working seriously on a bill to strengthen the work requirements rather than waiting on President Clinton to formulate a bill of his own (Haskins, 2006, 40). These efforts, which originated within the Ways and Means subcommittee on Human Resources, initiated a protracted and pluralistic process of social interaction that would stretch over the next three years.

Negotiation and Tapering Down As Charles O. Jones (1974) has observed, policy entrepreneurs must taper down proposals from what they consider optimal to what enough of their colleagues find acceptable. House Republicans would have to compromise repeatedly to produce a bill that would be acceptable to all the major elements in their coalition. First, to become law, any welfare reform would also have to be passed by the Senate, and the Senate Republican caucus was much more moderate than its House counterpart for reasons discussed in the previous chapter. Moreover, the inevitable differences between the Senate bill and the much more conservative House bill would have to be resolved in a conference committee. Predictably, the final conference report proved much less ambitious—and thus more incremental— than the original House bill. Equally important, House Republicans would also have to negotiate with the nation’s Republican governors. Most of the work-oriented innovations in welfare at the state level were initiated by Republican governors; the two leading figures here were Tommy Thompson of Wisconsin and John Engler of Michigan. Because the governors would be responsible for implementing whatever bill Congress might pass, they qualified as major stakeholders. Moreover, they had more experience, and thus more expertise, with work-oriented welfare reform than any of the Republicans in Congress. For all these reasons, Haskins, and other members of the core group of House Republicans driving the process in Washington, made it a point to keep the governors updated and always involved. The governors’ interests were different from their Washington counterparts. They wanted to maintain the flow of federal welfare money to the states, and they wanted to retain as much freedom as possible in spending that money. They also objected to proposals that would impose new federal mandates on them or force them to raise money at the state level to pay for categories of the poor that congressional Republicans might want

50 

M. T. HAYES

to cutoff (for example, undocumented immigrants). Haskins also made it a point to update Robert Rector of the Heritage Foundation on proposed changes in the bill. Rector was so well respected among conservative Republicans, especially in the House, that an objection from Rector would doom a proposed change. Finally, the core group of House Republicans would have to negotiate with President Clinton. The President had first-hand experience with welfare as well, as the former governor of Arkansas, and his experiences as governor would lead him to veto two Republican welfare reform bills before finally signing the third.

Focus on Incremental Alternatives Lindblom argues that participants will confine themselves to proposals that make incremental changes from the status quo. Moreover, policymakers will be forced to bargain and compromise, virtually assuring incremental outcomes. The very inevitability of this “tapering down” is one of the primary motivations for limiting attention to incremental alternatives that are politically viable. While the core group of moderates (Haskins, Shaw, Ways & Means chief staffer Phil Mosely, and others) wanted to achieve the strongest possible work requirements, they were forced to operate incrementally since the Democrats were the majority party in the House and Senate. Any Republican welfare reform would have to be acceptable to moderate Democrats in Congress as well as President Clinton. House conservatives’ insistence on addressing illegitimacy would necessarily make the bill more extreme, and thus less palatable to those Democrats who might otherwise be receptive to work-oriented welfare reform. On the Democratic side, support for time limiting welfare and imposing strict work requirements originated within the Democratic Leadership Council, an organization of conservative and moderate Democrats founded in 1980 with the goal of rebranding the Democratic Party to broaden its appeal to middle-class voters (Marshall & Kamarck, 1993; Baer, 2000; From, 2013). These self-proclaimed “New Democrats” favored welfare reform, more money for police, and free trade agreements, among other things; in Bill Clinton, they found an articulate advocate for their moderate alternative to an increasingly unpopular Democratic liberalism. Although Clinton (and his fellow DLC member, Al Gore) won the

4  WELFARE REFORM AS NORMAL INCREMENTALISM 

51

White House in 1992, the DLC remained an insurgent faction within a liberal Democratic Party. Because of Clinton’s relative conservatism, and his campaign pledge to reform welfare, House Republicans saw an opportunity to pass a welfare reform that a Democratic president might sign even though they were the minority party in both houses of Congress— but only if they could exercise the self-control required to produce an incremental bill. Unfortunately, House conservatives favored two provisions on illegitimacy that would surely be unacceptable to moderate Democrats and the president. The first was Jan Meyers’ (R-KS) bill to convert AFDC to a block grant and cutoff welfare to teenage mothers. While Republicans favor block grants, Democrats do not, and block grants were a nonstarter whenever Democrats controlled both houses of Congress. (Block grants would become possible, however, when Republicans won control of both houses in the 1994 midterms.) Cutting off all welfare to teenage mothers was perceived as too harsh, as well as certain to trigger media characterizations of Republicans as mean spirited (Haskins, 2006, 53). The Meyer proposal was radical; it was dubbed “Murray Lite” only because it merely cut off welfare to a fraction of the welfare population, rather than abolishing welfare entirely as advocated by Charles Murray. However, Murray Lite could be made more incremental by giving states the option to enact the provision rather than mandating it in the law. In fact, proposals can be arrayed along a continuum ranging from complete abolition of welfare (the full Murray) at one end, Meyers’ Murray Lite proposal as a mandate for all states, Murray Lite as included in the bill but modified to permit states to opt out, and finally Murray Lite as a policy that states could opt into although not required to by the bill. Opt-outs and opt-ins require state action. Specifically, a state legislature would have to pass a bill opting in. Just as with Congress, this would involve passage of a bill to opt in by each house (except of course, in Nebraska, which had a unicameral legislature), a probable conference committee, passage of the conference report in each house, and the governor’s signature. The same thing would have to happen if the law mandated Murray Lite but allowed states to opt out. As bargaining took place at various stages of the legislative process, as well as in the three different versions of the bill that were passed and sent to Clinton, compromises would be reached by making the Murray Lite proposal incrementally stronger or weaker as needed. In addition to the Meyers bill, some conservatives also wanted to impose a family

52 

M. T. HAYES

cap on welfare, cutting off aid after the first two children. This proposal could be moderated in the same way by allowing states to opt out or, alternatively, by leaving the family cap out of the bill and allowing states to opt in. The last version of the bill, signed into law by the President, did not include either Murray Light or a family cap. Because of the flexibility offered by the conversion of AFDC into a block grant, states were free to opt in to either provision. Ten years after passage, however, no state had adopted any version of Murray Light, although about half had adopted a family cap (Haskins, 2006, 366). As politically unrealistic as these two proposals were, conservative House Republicans clung to them a long time before finally admitting defeat. Finding themselves in the majority after the 1994 midterm elections, House Republicans opted to pursue one genuinely non-incremental reform: converting 336 social programs from categorical grants into eight block grants. As Lindblom would predict, however, most of these block grant proposals were not politically viable. Child nutrition block grants and child protection block grants were dropped in the Senate due to resistance from Republican moderates. Food stamp block grants were dropped in response to opposition from Republicans on the House Agriculture Committee. The proposal to block grant Medicaid was finally jettisoned in June of 1996, as Republicans struggled to produce a third version of welfare reform after two Clinton vetoes. By this point, Republicans had begun to realize that they had overreached, trying to do too many things at once—pass all the items in the Contract, balance the federal budget, block grant 336 social policies, ameliorate illegitimacy, and strengthen work requirements. When it became clear that combining welfare reform with a Medicaid block grant would kill the bill, Republicans reluctantly agreed to treat the Medicaid block grant as a separate, albeit doomed, bill. (For a summary of the provisions of the bill, as well as the provisions that were dropped separately but eventually enacted separately, see Haskins, 2006, 364–376.) Only two of the proposed block grants survived. Five separate categorical programs for childcare were combined into a single childcare block grant. This was an incremental change, however, as the purpose of the program was unchanged and the new block grant was funded at a higher level than the combination of previous categorical programs. By far the most important block grant included in the new law was TANF. By

4  WELFARE REFORM AS NORMAL INCREMENTALISM 

53

abolishing AFDC and replacing it with a block grant, the purpose of welfare was converted from sustenance (monthly cash payments) to promoting and supporting work.

The Final Outcome: Incremental or Non-incremental? While Ron Haskins characterized the 1996 welfare reform as a revolution in social policy, the final legislative outcome does not qualify as a non-­ incremental policy change in Lindblom’s terms. Recall from Chap. 2 that Braybrooke and Lindblom treated the distinction between incremental and non-incremental policy changes as a matter of degree. Policies were evaluated by placing them along a continuum ranging from weak or marginal proposals at one end to systemic transformations at the other. The JOBS program would clearly fall toward the incremental end of the continuum, with its reliance on extremely weak work incentives, low participation standards, and an extremely broad definition of work-related activities that included going to school, engaging in job training, or merely searching for a job (Haskins, 2006, 12). House Republicans wanted to move the needle significantly here, requiring participants to work, setting high participation targets, and defining work-related activities as getting a job and keeping it. By contrast, successfully converting 336 social welfare programs into eight block grants would just as clearly have fallen toward the non-­ incremental side of the continuum. The conservative long-term vision of shifting responsibility for income maintenance and social welfare programs entirely to the states, along with the responsibility to pay for them, might qualify as a systemic transformation since it would achieve a return to the kind of federalism envisioned by the framers. If the House Republicans had achieved all, or even most, of their block grant proposals, the result would qualify as an exceptionally large increment indeed. But they did not. Several potentially large increments were dropped from the final bill. Medicaid block grants were jettisoned late in the process to avoid dooming the third version of the reform. Child nutrition block grants were killed by Lugar’s opposition in the Senate. Food Stamp block grants were removed after encountering stiff opposition from Republicans on the House Agriculture committee. Child protection block grants were killed in the Senate. The illegitimacy provisions (Murray Lite

54 

M. T. HAYES

and the family cap) were stripped from the bill while allowing states to opt in. However, some provisions that were dropped from the final bill were subsequently passed as separate laws, including the elimination of SSI payments for alcoholics and drug addicts and limitations on welfare benefits for noncitizens. Most provisions that were retained in the final bill are best understood as incremental. For example, as noted earlier, five separate categorical programs were consolidated into a single childcare block grant, but the purpose of the law was left unchanged, and the block grant was funded at a higher level than the combined funding for the five categorical grants. Child support enforcement was tightened, establishing strict requirements for determining paternity, developing computerized systems for identifying employers to facilitate withholding of wages for child support, applying sanctions to deadbeat dads, and altering the distribution formulas to deliver more child support money to mothers and less to state-level bureaucrats. Amendments to the SSI program established a much stricter definition of child disability and ended an assessment method that Republicans felt was much too lenient. In the end, the core of the bill was exactly what it had been at the outset of the legislative process in 1991, work-oriented welfare reform. Haskins and his allies were proven right: the illegitimacy provisions would be unacceptable to President Clinton, whose signature on the bill was necessary for it to become law, and measures to strengthen work requirements and time limit welfare represented the common ground on which a workable compromise could be built. AFDC was abolished and replaced by TANF. The goal of welfare was shifted from providing monthly cash benefits to the promotion of work. Replacing AFDC with TANF also marked a devolution in power from the federal government to the states by replacing AFDC categorical grants, with all their federal requirements, with new block grants oriented around the promotion of work. In short, both the purpose of welfare and the delivery system had been changed; welfare would now emphasize work over income maintenance, and federal assistance would arrive through flexible block grants rather than categorical grants. States would now have an extraordinary degree of freedom to experiment with alternative approaches to encouraging work. Two political scientists, Brian Hogwood and Guy Peters (1983) have identified four distinct forms of policy change: (1) policy innovation, (2) policy maintenance, (3) policy termination, and (4) policy succession. Policy innovation occurs when new policies are created through a federal

4  WELFARE REFORM AS NORMAL INCREMENTALISM 

55

role breakthrough. Policy maintenance occurs when policies are tweaked or adapted without being significantly changed through incremental rationalizing policies. Policy termination occurs when a government program is abolished, whether because it no longer works or there is no longer any need for it. Finally, policy succession occurs when a new policy replaces a previous policy. In Hogwood and Peters’ terms, the 1996 welfare reform qualifies as policy succession, terminating one program (AFDC) and replacing it with another (TANF). The Hogwood and Peters terminology is more precise than the ambiguous distinction between incremental and non-incremental advanced by Braybrooke and Lindblom. Characterizing the 1996 welfare reform as a case of policy succession tells us more than any attempt to define the outcome as incremental or non-incremental. Finally, the new welfare reform law was subsequently eroded by the effects of federalism on policy implementation. A study of how the new policy was implemented in four states found considerable erosion of the mandate to promote work at the intake point, or “street level,” (Riccucci, 2005; see also Lipsky, 1980). The problem was not an ambiguous legislative mandate. To the contrary, state-level administrators clearly understood the intent of the new law and communicated it just as clearly to administrators at lower levels. However, managers continued to emphasize established goals as well: providing income to applicants so that parents could remain with their children, collecting information on applicants and determining eligibility accurately, and maintaining low error rates. As a result, street level intake workers were effectively granted vast discretion to choose among multiple goals: not just promoting work and/or diverting applicants from applying but also accurately determining eligibility and providing cash benefits where needed (Riccucci, 2005, 91–114). While the 1996 welfare reforms produced a dramatic reduction in caseloads and a significant increase in work by welfare recipients, the reforms had negligible impact on the day-to-day activities of street level bureaucrats. Rather, most of the increase in work can be attributed to “front door” institutional changes: (1) strong messages that welfare was now hard to get, and that work was expected and (2) expanded paperwork requirements that made for a much longer application process that discouraged applicants. The strong state of the economy also contributed significantly to the sharp drop in welfare rolls (Allard, 2007). Finally, work requirements were implemented more vigorously by some states than others. Lawrence Mead (2004, 214–244) attributed this uneven implementation to political culture at the state level, as defined by

56 

M. T. HAYES

Daniel Elazar (1984). In Mead’s view, work-oriented welfare reforms were implemented most effectively in “good government” states that Elazar classified as moralistic because those states view legislation as problem solving and exhibit a tradition of neutral and competent public administration. Moreover, in the 1980s and early 1990s, many states were already reforming welfare—some in anticipation of reforms they expected to emerge at the national level—and the federal requirements arising from the 1996 act would be adapted idiosyncratically to what states were doing already (Winston, 2002). This considerable variation in implementation of the law across states can be viewed in a more positive light, however. Under the U.S. Constitution, checks-and-balances at the national level are reinforced by a further division of power between the new national government and the states as a “double security to the right of the people” (Rossiter, 1999, 291). Within this “compound republic,” laws passed by Congress are notoriously modified or eroded at state and/or local levels during the implementation process, to the chagrin of their original sponsors. As Alexander Hamilton observed in Federalist 73, however, subjecting a legislative proposal to examination by a variety of individuals from diverse institutional backgrounds reduces the potential for errors stemming from the lack of due deliberation, as well as the dangers flowing from transient majorities aroused by “the contagion of some common passion or interest,” (Rossiter, 1999, 411). While policymakers at the national level understandably strive for absolute compliance with legislative mandates, implementation of laws by state or local administrators brings into play an additional set of actors from extremely diverse situations, further reducing the dangers that flow from inadequate deliberation. It follows that modification of policies by state and local administrators—what Berman and Fox (2023, 37–51) call a “practitioner veto”—may be properly understood as an additional advantage of incrementalism.

References Allard, S.  W. (2007). The Changing Face of Welfare During the Bush Administration. Publius, 37(3), 304–332. Baer, K. S. (2000). Reinventing Liberalism: The Politics of Liberalism from Reagan to Clinton. University Press of Kansas. Berman, G., & Fox, A. (2023). Gradual: The Case for Incremental Change in a Radical Age. Oxford University Press.

4  WELFARE REFORM AS NORMAL INCREMENTALISM 

57

Besharov, D., & Novak, M. (Eds.). (1987). A Community of Self-Reliance: The New Consensus on Family and Welfare. American Enterprise Institute. Burke, E. (1999). Reflections on the Revolution in France. In I. Kramnick (Ed.), The Portable Edmund Burke. Penguin Books. Burke, V., & Burke, V. (1974). Nixon’s Good Deed: Welfare Reform. Columbia University Press. Derthick, M. (1970). The Influence of Federal Grants: Public Assistance in Massachusetts. Harvard University Press. Elazar, D. J. (1984). American Federalism: A View from the States. HarperCollins College Division. Ellwood, D. T. (1989). Poor Support: Poverty in the American Family. Basic Books. From, A. (2013). The New Democrats and the Return to Power. Palgrave Macmillan. Glazer, N. (1988). The Limits of Social Policy. Harvard University Press. Handler, J.  F. (1995). The Return to the States: Changing Social Behavior. In J. F. Handler (Ed.), The Poverty of Welfare Reform. Yale University Press. Haskins, R. (2006). Work Over Welfare: The Inside Story of the 1996 Welfare Reform Law. Brookings Institution. Hogwood, B., & Peters, B. G. (1983). Policy Dynamics. St. Martin’s. Jones, C.  O. (1974). Speculative Augmentation in Federal Air Pollution Policymaking. Journal of Politics, 36(May), 438–464. Kaus, M. (1992). The End of Inequality. Basic Books. Kotz, N. (1969). Let Them Eat Promises: The Politics of Hunger in America. Prentice-Hall. Lindblom, C. E. (1980). The Policy-Making Process (2nd ed.). Prentice-Hall. Lipsky, M. (1980). Street Level Bureaucrats: Dilemmas of the Individual in Public Service. Russell Sage. Lynn, L. E., Jr., & Whitman, D. D. (1981). The President as Policy-Maker: Jimmy Carter and Welfare Reform. Temple University Press. Marmor, T. R. (1973). The Politics of Medicare. Aldine. Marshall, W., & Kamarck, E.  C. (1993). Replacing Welfare with Work. In W.  Marshall & M.  Schram (Eds.), Mandate for Change (pp.  217–236). Berkley Books. Mead, L.  M. (1986). Beyond Entitlement: The Social Obligations of Citizenship. Free Press. Mead, L. M. (1992). The New Politics of Poverty: The Nonworking Poor in America. Basic Books. Mead, L.  M. (1997). The New Paternalism: Supervisory Approaches to Poverty. Brookings Institution. Mead, L. M. (2004). Government Matters: Welfare Reform in Wisconsin. Princeton University Press.

58 

M. T. HAYES

Mead, L. M., & Bane, M. J. (2003). Lifting Up the Poor: A Dialogue on Religion, Poverty & Welfare Reform. Brookings Institution Press. Moynihan, D.  P. (1973). The Politics of a Guaranteed Income: The Nixon Administration and the Family Assistance Plan. Random House. Murray, C. (1984). Losing Ground: American Social Policy, 1950–1980. Basic Books. Riccucci, N. M. (2005). How Management Matters: Street-Level Bureaucrats and Welfare Reform. Georgetown University Press. Rossiter, C. (1999). The Federalist Papers. New American Library, Mentor Books. Steiner, G. (1966). Social Insecurity: The Politics of Welfare. Rand McNally. Winston, P. (2002). Welfare Policymaking in the States: The Devil in Devolution. Georgetown University Press.

CHAPTER 5

Incrementalism and Political Inequality

Abstract  For incrementalism to work properly, all affected interests must be represented, and political resources must be distributed evenly across interests. Unfortunately, many interests fail to mobilize, and political resources are concentrated narrowly rather than distributed evenly across groups, creating a privileged position for business with capitalist societies. However, political inequality is a problem that plagues all decision-making methods—not just incrementalism—and political inequality within the United States may be exaggerated. The most serious objection to incrementalism is not the small size of incremental steps but rather nondecision-­ making—the ability of powerful interests to take any action at all on many issues. Keywords  Cumulative vs. dispersed resources • Free rider problem • Nondecision-making • Privileged position of business • Racial tracking For incrementalism to work in a way that is fair to all participants, two preconditions must be met. First, all affected interests succeed in mobilizing, or, as Lindblom expressed it (1959, 85), every social interest has its watchdog. Second, no single actor or coalition exercises disproportionate power in the competition among social interests.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. T. Hayes, Incrementalism and Policymaking in the USA, https://doi.org/10.1007/978-3-031-38485-1_5

59

60 

M. T. HAYES

Neither of these preconditions is met in practice. In the first section, I will identify three biases in the interest group universe. In the second section, I will identify the primary resources available to interest groups as they seek to influence public policy and show that they are concentrated narrowly rather than broadly dispersed (Dahl, 2005, 85–86). In the third section, I will argue that this cumulative inequality in resources favors business interests, who tend to have more of these various resources than any other groups. This advantage is so great that Lindblom (1977) can say corporations occupy a “privileged position” within capitalist societies. However, the fourth section will argue that political inequality is a problem that confronts all decision-making methods—not just incrementalism—and suggest that political inequality in the United States may be overstated. The concluding section of the chapter will shift focus to the special problems facing groups at the other end of the spectrum—groups that not only fail to mobilize but also lack access to all or almost all political resources. More specifically, I will explore how groups with disproportionate power can prevent or deflect challenges by disadvantaged groups to their privileged status through a process neo-elite theorists term nondecision-­making (Lukes, 1974; Bachrach & Baratz, 1970). I will apply the theory of nondecision-making to the case of unequal treatment of African Americans by every facet of our criminal justice system and ask whether elite control over the political agenda constitutes a fatal flaw in incrementalism.

Biases to the Group Universe Ideally, all potential groups with a stake in each issue would succeed in mobilizing. This seldom happens, however, because the universe of organized groups exhibits three distinct biases: Groups with only a few members are more likely to organize than groups with many members, groups seeking to mobilize middle- and upper-class interests will have an easier time than groups seeking to mobilize the poor, and institutions will be much more likely to survive over the long haul than membership groups. All potential interest groups must first overcome the free rider problem if they are to form organizations. Because the primary benefits of interest group activity are collective, necessarily available to all potential members whether they join the group or not, self-interested individuals will find it rational not to contribute, hoping instead to benefit from the efforts of

5  INCREMENTALISM AND POLITICAL INEQUALITY 

61

others. This free rider problem is especially severe for large diffuse interests like consumers, taxpayers, or environmentalists because the likelihood that any one member’s contribution will make a decisive difference to the group’s success diminishes as the number of potential members increases. By contrast, in groups with a handful of members, the importance of everyone’s contribution to the group’s success is so apparent—and the individual’s personal stake in the outcome so much larger—that collective action is much more likely (Olson, 1970). This free rider problem is one obstacle working against efforts to organize the poor, and it helps to explain why reauthorization of TANF failed to address their interests. To make matters worse, interest groups are harder to form during economic downturns, when they are most needed. Group membership is an economic transaction between potential members and entrepreneurs who offer a mix of tangible and intangible benefits in exchange for dues (Salisbury, 1969). To survive, a group must eventually reach a point at which income from member dues exceeds the costs of servicing the membership. Because group membership is often more of a luxury than a necessity, and the collective benefits pursued by most groups are vulnerable to the free rider problem, interest groups are most likely to form in prosperous times when potential members can easily afford the dues. In unfavorable economic times, when an organized interest group might make a greater difference to members’ lives, potential members’ incomes fall, and groups are more likely to lose members or fail altogether. By extension, this analysis implies a socioeconomic bias to interest group formation. Entrepreneurs who seek to mobilize the poor will face insurmountable difficulties, as potential members will have little income to spend on group memberships that yield collective benefits in the distant future, if at all. As E.  E. Schattschneider observed (1960, 35), in the struggle among organized interests, “the heavenly chorus sings with a strong upper-class accent.” Finally, institutions survive longer than membership groups. The free rider problem never goes away for membership groups; once a group has gotten off the ground, it still needs to motivate members to remain in the group in subsequent years. By contrast, corporations do not face a free rider problem when lobbying as individual organizations; they are already organized for the purpose of producing and selling products or services in the private marketplace. Where membership groups are vulnerable to economic cycles, corporations are much more likely to have surplus resources to devote to lobbying regardless of the circumstances.

62 

M. T. HAYES

In this regard, two political scientists compared the universe of active interest groups in 1960 with the group universe in 1981 (Schlozman & Tierney, 1986). Most corporations active in the earlier period were still on the scene two decades later. By contrast, most membership groups in the 1960 sample no longer existed in 1981. To the contrary, most membership groups active in the 1981 sample were founded in the preceding 10–15 years, suggesting that the new groups merely replaced earlier groups that had failed to survive. While far more groups were active in 1981, the relative balance of corporations and membership groups was unchanged. Governmental institutions do not face a free rider problem, and they often have interests of their own. This why Earl Latham (1952, 390) called them “official groups” and included them along with private interest groups as part of the larger group universe. The official group with the greatest continuing interest in welfare reform would be the nation’s governors. We saw the key role played by Republican governors in the enactment of the 1996 welfare reform in Chap. 4; for a bill to be enacted, they had to be included in negotiations throughout. While some governors wanted to reform welfare to make recipients better off, finding them jobs and bringing them into the economic mainstream, others favored the reforms because they promised to move many current recipients off the rolls while deterring others from applying at all. TANF was remarkably successful at reducing welfare rolls and, as we saw in Chap. 4, the deterrent effect on the front end was the most significant contributor to this success. The actors pushing hardest for welfare reform are usually governors who need to get welfare spending under control since most states operate under a constitutional requirement to balance their budgets. Accordingly, whenever welfare rolls rise significantly, the welfare portion of state budgets pose a direct threat to funding for other state programs—education, highways, and so on (Weissert, 2000). Governors initiate reforms at the state level to get control of welfare costs, and they approach Washington when they find they cannot do this by themselves. Once welfare reform is placed on the congressional agenda, liberal groups favoring broader coverage and higher benefits become active, making welfare policy intensely conflictual. But the issue typically originates in a desire to get control over welfare costs. There are always institutions with no free rider problem that have an ongoing interest in reducing welfare benefits, whether by

5  INCREMENTALISM AND POLITICAL INEQUALITY 

63

removing recipients from the rolls or moving them into jobs. This is why seriality did not function as well as predicted in the welfare reform case. However, partisan incrementalism should function more efficiently on other issues where all the major stakeholders are effectively organized.

Group Resources: Cumulative or Dispersed? Beyond this, the relative balance of the groups that do mobilize will depend on the resources available to the various participants and the extent to which each one can translate these resources into effective influence. The resources that groups may employ in their efforts to influence policy may be divided into two categories: tangible and intangible (Truman, 1951, 506–507). Tangible resources would include money and the size of the group’s membership. Money can be used to pay for staff, lobbyists, or advertisements on television or digital platforms, among other things. A century ago, vote mobilization and communication were labor-intensive processes that operated through patronage-oriented machine parties. Today’s campaigns, with their emphasis on television and social media platforms, are capital intensive. Supreme Court rulings equating money with speech have made money more important than ever, giving a huge advantage to corporations and wealthy individuals. The size of a group’s membership matters in a variety of ways. A large membership contributes to a group’s legitimacy by making it seem representative of a broad, significant interest. (The concept of legitimacy will be defined more fully below.) Endorsements by group leaders carry more weight if the group has a large membership. Finally, groups with many members can contribute campaign workers; for many years, labor unions were dominant actors in  local Democratic organizations, and today the same thing is true for the Christian right on the Republican side. Intangible resources may be even more important to a group’s success. For example, a group’s influence will depend in part on its reputation as a reliable source of information or expertise. Interest groups often function as “service bureaus” for legislators who are already sympathetic to their cause, providing information on alternative policies and educating their members on the need to accept compromises emerging from the bargaining process (Bauer et al., 1972; Wright, 1996). A reputation for honesty and accuracy is essential lest groups forfeit access to policymakers.

64 

M. T. HAYES

A second intangible resource is the group’s perceived legitimacy. A group is legitimate if policymakers view its members as worthy or meritorious. Doctors, church leaders, and heads of major corporations will be perceived as more legitimate than welfare recipients, for example. Similarly, a group’s demands are legitimate if policymakers perceive them as fair and reasonable. Senior citizens’ groups possess extraordinary legitimacy because the large medical bills they face in their retirement years are viewed as not of their own making, and because all of us must eventually confront the problems associated with aging. While Medicare and Social Security provide benefits to one part of the population at the expense of the rest, these programs are better understood as universal entitlements that eventually become available to all of us. By contrast, welfare recipients are easily portrayed as failures who live off the toil of others: men too unskilled or irresponsible to hold a steady job, women abandoned by their husbands or having children out of wedlock, “welfare queens” taking advantage of the system by collecting multiple benefits. The contrast is striking, and it helps explain why seniors are insulated from major program cuts while welfare programs are frequently cut back. Finally, some groups gain leverage from their control over goods or services vital to the economy. Public employee unions whose workers pick up garbage or operate mass transit systems are one example here. As will be discussed more fully in the following section, corporations occupy a privileged position within capitalist societies to the extent that they can deliver or withhold prosperity, which everyone needs. For incrementalism to operate properly, these various resources must be dispersed across groups rather than concentrated in the hands of a small number of groups (Dahl, 2005, 85–86). Where political resources are dispersed, there is more than one way to win. A group high in legitimacy might prevail over groups with lots of money, at least some of the time. Groups might offset small numbers or limited budgets with expertise. By contrast, where political resources are concentrated in a small number of groups, some groups will score high across the board while others rank near the bottom in every category. The following section will show that, within capitalistic societies, resources are cumulative rather than dispersed. Corporations rank at or near the top in every resource category, enabling them to block unwanted legislation and receive benefits from policymakers without even having to mobilize.

5  INCREMENTALISM AND POLITICAL INEQUALITY 

65

The Privileged Position of Business in Capitalist Societies There can be little doubt that business as a class exercises disproportionate power within capitalist societies. Surprisingly, the originator of the theory of incrementalism, Charles Lindblom, is the most influential theorist advancing this argument. In Lindblom’s view, the free market operates as a kind of prison, compelling policymakers to reject out of hand policy changes detrimental to business. Attempts to tax or regulate business automatically trigger “punishment” in the form of high unemployment or a sluggish economy (Lindblom, 1982, 324–325). Ralph Miliband advanced a more nuanced variation on this theme. Where Lindblom believed business could block any reforms that threatened their class interests, Miliband recognized that business groups are sometimes unable to prevent enactment of adverse regulatory or tax policies. While the market is not a prison in his view, business groups do prevail more often than any other groups in what he saw as an “imperfect competition” of social interests (Miliband, 1969, 164–165). According to Miliband, the advantage to business in this competition is rooted in the extraordinary degree of commitment to capitalism as a form of economic organization. Disagreements among the viable contenders for elective office are typically confined to secondary issues, leaving the fundamental question of the form of economic organization unaddressed. In short, debate is confined to the proper degree of intervention in an economy that all agree will remain capitalistic. To reformulate this argument in the language of Chap. 2, the privileged position of business in capitalistic societies precludes systemic transformation from capitalism to socialism. However, this privileged position does not enable business groups to keep all issues threatening to them off the agenda, nor does it preclude enactment of policies adverse to business interests, although these victories over business groups will typically take the form of small increments rather than large ones. Both Lindblom and Miliband emphasize the importance of intangible resources as the primary sources of business power. The most important intangible resource, as they see it, is the leverage business derives from its ability to provide or withhold prosperity. Corporate leaders serve as a second set of public officials within capitalistic economies insofar as a wide range of activities that affect the entire public have been delegated into their hands: decisions regarding what is to be produced, how labor and

66 

M. T. HAYES

other factors of production will be allocated to different products, what technologies will be employed, where plants will be located and when they may be closed down, and so on. Although all these activities are defined as private by most economists, they have momentous consequences for the average citizen, affecting the overall level of economic growth and the prospects for employment in various locations and lines of work (Lindblom, 1977, 171–175; see also Miliband, 1969, 147–155). To Lindblom, the central problem with the delegation of public functions into private hands is the corresponding forfeiture of command as a means of control. This delegation makes the performance of the economy dependent on the response of corporate leaders to changes in the business climate. Corporations cannot be commanded but rather must be induced to perform the functions on which society depends. This is where the “market as prison” analogy comes into play; in making public policies, government officials must always consider the effects of their proposals on business confidence. This delegation of private functions into public hands gives business interests an extraordinary level of legitimacy in the imperfect competition of social interests. Recall that groups are viewed as legitimate by policymakers to the extent that group members are seen as meritorious or worthy. Because corporate executives function as a second set of public officials within capitalist societies, they are naturally accorded high status. On another level, legitimacy derives from the extent to which a group’s demands are compatible with the values held by policymakers and the larger society—the extent to which their demands are viewed as fair and reasonable. Because capitalism is so entrenched within Western democracies, policymakers naturally equate the narrow class interests of business with the broader national interest. By comparison, all other groups, particularly organized labor, are perceived as narrow sectoral interests (Miliband, 1969, 75). The dependence of the state on business for the maintenance of prosperity creates a third intangible advantage. Government officials need reliable information when they formulate public policies—particularly when the subject matter is highly technical, as it typically is with business regulation. Corporations often possess a near monopoly on expertise relating to production processes. For example, while environmental groups exploited their high legitimacy with the public to place the regulation of surface mining of coal on the agenda in the 1970s, the coal industry’s advantage in expertise enabled them to favorably influence the language of the

5  INCREMENTALISM AND POLITICAL INEQUALITY 

67

resulting legislation, prevailing in Congress on the issues that mattered most to them—avoiding federal preemption of implementation and placement of the new agency within the industry-dominated Bureau of Mines (Vietor, 1980). This information advantage was even more important at the implementation stage, as environmental groups could not match the industry’s experts or its ability to monitor implementation continuously (Hayes, 2006, 97–117). While both Lindblom and Miliband minimized the importance of money as a source of business power, the financial advantage of business over other actors is exceptionally large and growing steadily larger. With the legalization of political action committees (PACs) in the mid-1970s, the total amount of business contributions to candidates grew dramatically, much faster than the contributions of any other groups. By the early 1980s, corporate and trade association PACs had amassed a better than two-to-one dollar advantage over organized labor, business’s richest competitor. Even more impressive, the total for all business groups (corporate plus trade association PACs) exceeded the combined resources of all the other organized interests, including labor. This remarkable monetary advantage to business interests enables corporations and trade associations to gain access to a broader range of legislators than any other group (Schlozman & Tierney, 1986, 249, Table 10-5). In sum, in our capitalistic society, political resources are concentrated rather than dispersed. Business groups combine intangible advantages (strategic position within the economy, unparalleled legitimacy, and unsurpassed expertise) with a truly staggering advantage in money. The result is a form of skewed pluralism in which business prevails more often than any other actor in the system.

Political Inequality Plagues All Decision-Making Methods Incrementalism should not be rejected as the preferred method of policymaking because it operates imperfectly. Winston Churchill famously said that democracy is the worst form of government except for all the others. The same thing might be said about incrementalism, since political inequality is a problem that affects all decision-making methods. Recall that incrementalism was never a utopian vision for Lindblom. At best, incrementalism moves slowly and iteratively toward a solution, and solving

68 

M. T. HAYES

a problem only means mitigating it sufficiently that other problems begin to seem more pressing. While we must constantly strive to mobilize the unorganized (Hayes, 2001, 161–166), we can never mobilize all groups no matter what we do. Fortunately, many interests are effectively represented by advocacy groups that do not rely exclusively on member dues for funding. Such groups may support themselves through consulting work or large donations from wealthy contributors attracted by favorable tax provisions. Others rely partially on member dues but supplement those dues with money obtained from outside sources (Walker, 1983). A study of groups active on the 1996 welfare reform issues (Winston, 2002) found that advocacy groups speaking for the poor were more numerous than those in any other category. This was true not only in Washington, D.C. but also in two of the three states singled out for comparative analysis (Texas and Maryland). Moreover, most members of Congress make an effort to identify the interests of potential groups that fail to organize successfully. A signature characteristic of congressional Democrats is a concern for the poor, while Republicans look out for the interests of unorganized taxpayers (Price, 2004, 284–285). This suggests that the larger problem may not be uneven representation but rather the maldistribution of resources for effective political influence. For example, the privileged position of business can only be mitigated, not eliminated, within our capitalistic society. Similarly, legitimacy is not a resource groups can possess, like money or expertise. Rather, legitimacy is granted to groups—or withheld from them—by policymakers and the public. Reformers, no matter how well-intentioned, cannot give a group legitimacy in the same way that they might give it money. But here too the problem may be exaggerated. The privileged position business interests occupy within capitalist societies primarily enables them to prevent a systemic transformation from capitalism to a different form of economic organization. However, Lindblom saw all such systemic transformations as extremely rare and confined to what he and Braybrooke termed the realm of “wars, revolutions, and grand opportunities.” Within capitalist societies incrementalism may be a system of imperfect competition among social interests, but business interests do not always prevail, and overall business power has varied significantly in different historical periods. For example, the regulatory state grew steadily between 1933 and 1980, and the number of regulations in the Federal Register grows whenever Democrats gain control over the White House and Congress.

5  INCREMENTALISM AND POLITICAL INEQUALITY 

69

The Special Problem of Nondecision-Making Incrementalism is a pluralist theory of policymaking. In the 1950s, a controversy raged between pluralists and power elite theorists. Power elite theorists argued that elites dictate outcomes behind the scenes, outside normal political institutions, although the exact identity of the hypothesized elite varied from one theorist to another. Pluralists countered that the political system is open and responsive to a wide variety of interests even if some fail to mobilize. The pluralists eventually prevailed with the publication of two important works by Robert Dahl: an article specifying the appropriate methodology for testing elite theories (Dahl, 1958) and a book examining political decision-making in New Haven, Connecticut from the founding of the city through the early 1960s (when Dahl wrote) that failed to find convincing evidence of a power elite (Dahl, 2005). Critics of pluralism refused to concede defeat, however. While policymaking on issues that succeed in reaching the agenda may appear pluralistic, they argued, political elites nevertheless exercise real power by restricting the agenda to issues that do not threaten their interests. Instances where elites prevent a latent issue from arising were termed nondecisions (Bachrach & Baratz, 1962, 1963, 1970, 1975; Lukes, 1974). While the empirical analysis of nondecision-making has focused primarily on the study of community power (see Gaventa (1980) for a particularly convincing example), it can also be applied to national politics. Nondecisions can take two distinct forms. The first occurs when elites succeed in preventing an issue from arising at all and there is no overt, observable challenge to prevailing institutions or values. The second occurs when an overt challenge arises but is deflected. Peter Bachrach and Morton Baratz called nondecisions within the first category barrier I nondecisions. Barrier I nondecisions typically occur because elites have succeeded in manipulating mass preferences so that grievances fail to even form. Steven Lukes (1974, 24) saw this ability to subtly persuade people or institutions to accept their subordinate status as natural, inevitable, or beneficial as the supreme form of power. By contrast, barrier II nondecisions occur when an overt challenge is mounted against prevailing institutions or power relations, but elites successfully deflect the threat to their privileged position. In short, while barrier I nondecisions stifle challenges before they are even raised, barrier II nondecisions are responses to overt challenges. Barrier II nondecisions thwart such challenges through a variety of means

70 

M. T. HAYES

including intimidation, cooptation, manipulation of precedents or procedures, and redirecting threatening demands into harmless and symbolic channels (Bachrach & Baratz, 1970, 44–46). While barrier II nondecisions are observable—challenges to the status quo have emerged and been defeated through the methods just listed—barrier I nondecisions are not. For a barrier I nondecision hypothesis to be falsifiable, the failure of issues to arise at all due to elite suppression must be clearly differentiated from normal issues that fail merely because they lack sufficient political support. In this regard, Bachrach and Baratz distinguished between conventional issues and what they called “key” issues. By their definition, a key issue is one that mounts “a genuine challenge to the resources, power, and authority of those who currently dominate the process by which policy outputs in the system are determined” (Bachrach & Baratz, 1970, 48). I have suggested a less ambiguous definition elsewhere: a key issue is one that all observers would agree should have reached the agenda but did not (Hayes, 2018). Failure to guarantee equal treatment under law to African Americans should qualify as one such key issue. Professor Nina Moore (2015) has convincingly demonstrated that African Americans are treated differently than whites at every stage of the criminal justice process by police, prosecutors, and judges. African Americans are not only subject to racial profiling; they are also more likely to be pulled over for traffic stops, more likely to die in custody, more likely to be denied access to positive alternatives to imprisonment and subjected to harsher sentences once convicted. There are clearly two criminal justice systems in the United States—one for whites and another for African Americans and other minorities. However, equal treatment under law is a fundamental element of the rule of law and thus an ideal to which we all claim to be committed. For this reason, evidence of unequal treatment—of chronic, systemic racial tracking within the criminal justice system—should trigger a response. It is an issue that should reach the agenda but does not. Failure to address this disparity denies African Americans full status as citizens (Thurman, 1976). This unequal treatment did not end with the abolition of slavery or the end of Jim Crow. Nor did it begin with George Floyd; it has been going on throughout our history as a nation. For long periods of time, as Moore shows, the racial tracking issue fails to arise at all, remaining within the realm of barrier I nondecisions. However, overt challenges have occasionally been mounted against this chronic pattern of unequal treatment. Focusing on the period between

5  INCREMENTALISM AND POLITICAL INEQUALITY 

71

1988 and 2012, Moore identified a variety of bills designed to remedy racial tracking at every stage of the criminal justice process, from police investigation and prosecution to the court and imprisonment stages. Almost all of these initiatives were deflected, with only a handful even making it out of congressional committees to the floor. Only two bills were enacted into law, and both were incredibly weak. One, the Fair Sentencing Act of 2010, ignored almost all aspects of racial tracking and weakened recommendations advanced by the U.S.  Sentencing Commission. The other was a minor provision appended to an infrastructure bill with vague requirements and inadequate funding (Moore, 2015, 100–135). Surprisingly, these disappointing results do not seem to have been the result of elite suppression. According to Moore (2015, 123–133), the racial justice agenda items failed in spite of a number favorable conditions: a wealth of thorough studies documenting the existence of racial tracking, advocacy from across all regions, sponsors from both parties, support from committee chairs, policy entrepreneurs with solid parliamentary skills, support from black lawmakers and a wide variety of outside groups, and strong support from presidents Clinton, Bush, and Obama. Rather, the racial justice agenda seems instead to have been the victim of what E. E. Schattschneider (1960, 171) called a “mobilization of bias”: all forms of political organization facilitate the raising of some issues and the suppression of others. As a result, some issues are “organized into politics while others are organized out.” As Moore convincingly shows, bills designed to address racial tracking in the criminal justice system were organized out of politics while crime bills were organized in. Although crime rates were not actually rising during this period, and most people regarded their own neighborhoods as safe, the public was more worried by the threat of a potential rise in crime rates than by the dual justice system adversely affecting African Americans (Moore, 2015, 166–218). The two parties began competing to offer tougher crime bills, and the conflict was redefined to focus on the threat from potentially rising crime rates, effectively displacing the original concern over racial tracking. Significantly, this concern over crime was not limited to white Americans. Although African Americans strongly supported racial justice policies, they also placed a premium on anti-crime policies (Moore, 2015, 257). This displacement of conflict was possible, in large part, because African Americans possessed less legitimacy than other population categories. While many white Americans recognized racial biases in policing, they also

72 

M. T. HAYES

saw police officers as heroes who risked their lives daily. The resulting cognitive dissonance was reduced by viewing Black Americans as primarily responsible for how the police treat them (Moore, 2015, 225). Partly because of the way news programs elevate criminal acts, often highlighting African Americans as suspects, and partly because of the way the entertainment industry reinforces these images, many whites perceive African Americans as lazy, irresponsible, and prone to violence (Moore, 2015, 220–232). In addition, the shared American ideology of individualism and personal responsibility suggests strongly that success in life is open to all if you are willing to work hard and play by the rules (Moore, 2015, 243). In terms of the group resources identified in the second section of this chapter, African Americans lack legitimacy on both dimensions: as a population, they are deemed responsible for their own plight, and thus unworthy, and their demands for racial justice are deemed unreasonable. In this regard, surveys suggest white Americans now see themselves as victims of affirmative action and other preferential programs. Surprisingly, Black Americans are as likely as white Americans to say they oppose reverse discrimination (Moore, 2015, 241), a finding consistent with Lukes’ argument that success in convincing a group to view its subordinate status as reasonable or warranted is the supreme form of power.

Conclusion The policy process in the United States fails to satisfy the preconditions that must be met for incrementalism to operate efficiently, but the failure of incrementalism to work optimally does not mean we should reject the method altogether. The system is open and responsive to interests that do mobilize, and interests that fail to mobilize may be represented by advocacy groups that avoid the free rider problem by securing funding from sources outside their memberships. As I put it earlier, incrementalism may be the worst method of policymaking except for all the others. However, Moore’s analysis strongly supports Schattschneider’s thesis that all systems have a mobilization of bias that organizes some issues into politics and others out of politics. In addition to the mobilization of bias in favor of corporate interests, there is also a mobilization of bias operating powerfully to organize African American issues out of politics. The prevalence of nondecision-making poses a fundamental problem for incrementalism—just as it does for all other policy-making methods.

5  INCREMENTALISM AND POLITICAL INEQUALITY 

73

It follows that the problem is not so much the incremental nature of individual policy steps but rather the failure of policymakers to take any action at all on critical issues. The Senate filibuster notoriously permitted segregationists to block or weaken civil rights bills for years before finally being overcome in 1964. It similarly prevented any action on police reform in both 2020 and 2021 despite a mass public demand for action in the wake of the killing of George Floyd, blocked for decades any legislation at all in response to mass shootings, and prevented passage of a stronger gun control bill in 2021. Even global warming could have been addressed through incremental steps if political opponents had been unable to exploit procedural advantages to block any action at all for years through barrier II nondecision-making. According to Naomi Klein, if the recommendations of the 1988 World Conference on the Changing Atmosphere, held in Toronto, had been adopted, the world would only have needed to reduce carbon emissions by two percent a year between 1988 and 2005. (Klein, 2014, 55–56). The same thing could be said for deficit reduction and the need to get control over the steadily increasing national debt. Clearly, it is not enough to facilitate the mobilization of more interests or to search for ways to make the distribution of political resources more equitable. We must also remove veto points that permit advantaged groups to block action. If the filibuster were eliminated, as advocated in the following chapter, partisan incrementalism would be able to function more effectively, not only by making it easier to pass initial legislative initiatives but also by removing the biggest single obstacle to the proper operation of seriality. None of these actions—mobilizing more interests, equalizing the distribution of political resources, eliminating veto points—will make incrementalism operate perfectly; we can only mitigate these problems, not eliminate them. But taken together they would make incrementalism work better, and that is certainly worth doing.

References Bachrach, P., & Baratz, M. (1962). Two Faces of Power. American Political Science Review, 56(December), 947–952. Bachrach, P., & Baratz, M. (1963). Decisions and Nondecisions: An Analytical Framework. American Political Science Review, 57(September), 632–642. Bachrach, P., & Baratz, M. (1970). Power and Poverty: Theory and Practice. Oxford University Press.

74 

M. T. HAYES

Bachrach, P., & Baratz, M. (1975). Power and Its Two Faces Revisited: A Reply to Geoffrey Debnam. American Political Science Review, 69(September), 900–904. Bauer, R.  A., Pool, I., & Dexter, L.  A. (1972). American Business and Public Policy: The Politics of Foreign Trade (2nd ed.). Aldine-Atherton. Dahl, R.  A. (1958). A Critique of the Ruling Elite Model. American Political Science Review, 52(June), 463–469. Dahl, R. A. (2005). Who Governs? Democracy and Power in an American City. Yale University Press. Gaventa, J. (1980). Power and Powerlessness: Quiescence and Rebellion in an Appalachian Valley. University of Illinois Press. Hayes, M. T. (2001). The Limits of Policy Change: Incrementalism, Worldview, and the Rule of Law. Georgetown University Press. Hayes, M.  T. (2006). Incrementalism and Public Policy. University Press of America. Hayes, M. T. (2018). Congress and War Powers: Symbolism and Nondecisions in the Struggle for Influence. Congress & the Presidency, 45(2), 185–207. Klein, N. (2014). This Changes Everything: Capitalism vs. the Climate. Simon & Schuster. Latham, E. (1952). The Group Basis of Politics: Notes for a Theory. American Political Science Review, 46(June), 376–397. Lindblom, C.  E. (1959). The Science of “Muddling Through”. Public Administration Review, 19(Spring), 79–88. Lindblom, C.  E. (1977). Politics and Markets: The World’s Political-Economic Systems. Basic Books. Lindblom, C.  E. (1982). The Market as Prison. Journal of Politics, 44(May), 324–336. Lukes, S. (1974). Power: A Radical View. The Macmillan Press LTD. Miliband, R. (1969). The State in Capitalist Society. Basic Books. Moore, N.  M. (2015). The Political Roots of Racial Tracking in American Criminal Justice. Cambridge University Press. Olson, M., Jr. (1970). The Logic of Collective Action: Public Goods and the Theory of Groups. Schocken Books. Price, D. E. (2004). The Congressional Experience (3rd ed.). Westview Press. Salisbury, R. H. (1969). An Exchange Theory of Interest Groups. Midwest Journal of Political Science, 8(February), 1–32. Schattschneider, E.  E. (1960). The Semi-Sovereign People: A Realist’s View of Democracy in America. Holt, Rinehart, and Winston. Schlozman, K.  L., & Tierney, J.  T. (1986). Organized Interests and American Democracy. Harper and Row. Thurman, H. (1976). Jesus and the Disinherited. Beacon Press. Truman, D. (1951). The Governmental Process: Political Interests and Public Opinion. Knopf.

5  INCREMENTALISM AND POLITICAL INEQUALITY 

75

Vietor, R.  H. K. (1980). Environmental Politics and the Coal Coalition. Texas A&M University Press. Walker, J. L. (1983). The Origins and Maintenance of Interest Groups in America. American Political Science Review, 77(June), 390–406. Weissert, C. (Ed.). (2000). Learning from Leaders: Welfare Reform Politics in Five Midwestern States. Rockefeller Institute Press. Winston, P. (2002). Welfare Policymaking in the States: The Devil in Devolution. Georgetown University Press. Wright, J. R. (1996). Interest Groups and Congress: Lobbying, Contributions, and Influence. Allyn and Bacon.

CHAPTER 6

Recapitulation and Implications

Abstract  To recapitulate, incrementalism is both inevitable and desirable under normal circumstances; fortunately, it can be adapted to function within our current regime of responsible parties. For partisan incrementalism to work effectively, however, the majority party must be able to enact its program; this requires abolition of the Senate filibuster, a reform consistent with the vision of the framers. Equally important, both policymakers and voters must become realists, rejecting not only idealism but also the emphasis on symbols over substance that has emerged in our social media age. Finally, while this volume has focused on the American context, partisan incrementalism should function especially well within first-­ past-­the-post parliamentary systems featuring two strong parties. Keywords  Agitators • Parliamentary systems • Rationalizing breakthroughs • Senate filibuster At the outset of this book I distinguished between realists, who embrace incrementalism, and idealists, who reject it. I have argued throughout that the realists are correct: the policy process necessarily involves conflict and compromise on most issues. Disagreements over values combined with inadequate understanding of most policy problems preclude rational decision-­making. Instead, policies result from the interplay of a variety of

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. T. Hayes, Incrementalism and Policymaking in the USA, https://doi.org/10.1007/978-3-031-38485-1_6

77

78 

M. T. HAYES

actors with different self-interests, different institutional interests, and different perspectives on the public interest. Negotiation and compromise are necessary if any progress is to be made, making incremental outcomes inevitable. I have also argued here that incrementalism is desirable under normal circumstances. There are distinct limits to what can be achieved through any individual piece of legislation, and we should expect the policies we enact to be imperfect and in need of improvement. Solving public problems is a gradual process involving trial and error and subsequent modification. Policies necessarily go through a life cycle, as federal role breakthroughs are followed by incremental rationalizing policies. Rationalizing breakthroughs may occur if experience gained through policy implementation yields both consensual objectives and consensual knowledge among policymakers. Where that happens, a single-step large increment can occur, one that is driven by an increase in the available knowledge base rather than by an ill-advised attempt to go beyond it. Rationalizing breakthroughs are rare, but where they do emerge, they are the most likely non-incremental changes to work effectively and endure.

Incrementalism Can Be Adapted to a Partisan Age Since Lindblom first advanced the theory of incrementalism in 1959, our old bimodal parties have been replaced by parties that are ideologically coherent and polarized. By choosing one party over the other, voters can now dictate the direction of policy in advance through elections. Information costs have been significantly reduced as well, as most of the information a voter really needs is now communicated by the party label. As our politics becomes more majoritarian, and less pluralistic, the old incrementalism—in which policy outcomes could be understood as an equilibrium of contending interests—might seem to be obsolete. Fortunately, as I argued in Chap. 3, incrementalism can be reconceptualized to function successfully within a system of responsible parties. Under partisan incrementalism, the two parties would offer voters competing platforms consisting of incremental policy proposals that are both realistic and viable politically. By converting contentious problems into pure problems of knowledge base, partisan incrementalism increases the likelihood that inevitable policy imperfections will lead to rationalizing policies in subsequent policy cycles, enhancing the prospects for effective seriality. Moreover, party platforms confined to incremental alternatives

6  RECAPITULATION AND IMPLICATIONS 

79

should, in theory, lower the stakes in presidential elections by reducing the level of threat voters associate with the other party’s agenda. However, effective partisan incrementalism is impossible where the majority party cannot enact the bulk of its program. Right now, the greatest obstacle to partisan incrementalism is the Senate filibuster.

Eliminating the Senate Filibuster Eliminating the filibuster is not as radical as it sounds, for at least two reasons. First, it is consistent with the vision of the framers. Second, changing just one feature of our political system and then monitoring the result is properly understood as an incremental change. The U.S. Constitution was designed to be a system requiring multiple, sequential majorities, not a system in which a minority of one house of Congress can prevent a majority from acting at all. The Senate filibuster is not part of the Constitution, and the authors of The Federalist were united in favoring majority rule as the normal method for reaching decisions in both the House and Senate. In Federalist 10, Madison counted on majority rule (which he called “the republican principle”) to check minority factions (Rossiter, 1999, 48). In Federalist 22, Hamilton similarly objected that requiring an extraordinary majority to reach decisions would necessarily permit small coalitions to triumph over larger ones, violating the republican principle. In the worst case, a cohesive minority can prevent the government from acting, sapping its energy, and making it look weak and ineffectual when forceful action is required (Rossiter, 1999, 116). This is a remarkably accurate description of how partisan filibusters operate to prevent the majority from acting in the contemporary Senate. (For a good illustration see Shapiro, 2022.) Defenders of the filibuster counter that requiring sixty votes to pass bills encourages the pursuit of bipartisan coalitions. This is wishful thinking, however. Right now, our two congressional parties are so far apart ideologically that the most liberal Republican is far to the right of the most conservative Democrat. To attract votes from the minority party, whose leaders will punish members who cross the aisle to work with the opposition, the majority party will be forced to acquiesce in legislation acceptable to the minority. This allows the minority to dictate the content of the legislation, effectively reversing the outcome of the election (Binder & Smith, 1996). This is precisely what happened when President Biden attempted to win minority support for his Covid relief plan during his first

80 

M. T. HAYES

days in office: the ten Republicans with whom the president met offered an alternative bill that provided only a small fraction of the funds the president sought. Not surprisingly, the President opted instead to use reconciliation procedures to enact his own plan along party lines. Resistance to eliminating the filibuster is primarily rooted in negative partisanship. An NBC News poll released just prior to the 2020 midterm elections found that 81% of Democratic voters believe the Republican Party’s agenda poses a threat that, if not stopped, will destroy America. The corresponding result for Republicans was similar: 79% believe the Democrats’ agenda poses such a threat (Murray, 2022). Abolishing the filibuster would enable the opposition to enact their agenda when they gain the majority, and this cannot be allowed to happen. But holding on tightly to the filibuster to preserve the ability to block unwanted legislation is a prescription for deadlock and an inability to govern, which is precisely what we have now. Neither party has a monopoly on wisdom or corruption, and both parties are needed for our democracy to work properly. Politics in a healthy two-party system is characterized by sequential attention to goals (Cyert & March, 1963). Republicans and Democrats pursue very different agendas, and in so doing each neglects some problems while focusing on others with tunnel vision. Over time, the incumbent party’s agenda reaches a point of diminishing returns, and the problems they have neglected become increasingly serious. Party alternation in power operates to give a variety of problems attention over the long-term. A procedural change of this magnitude can be expected to have unanticipated consequences. However, the development of the 60-vote Senate has also had unanticipated consequences, and they are clearly unacceptable. You do not have to be a responsible party theorist to believe that when voters give unified control over government to one party in a national election, they have a right to expect policy consequences to follow, even within our Madisonian system. Finally, eliminating the Senate filibuster could be viewed as an incremental change: why not modify just this one aspect of our political system and then see whether more changes are needed? Fears that majorities would ride roughshod over minorities are overblown; within the American system, Representatives, Senators, and the President represent different constituencies, inevitably creating different institutional interests. If the Senate filibuster were abolished, bills would still have to pass both houses in identical form and be acceptable to the President. The 1996 welfare

6  RECAPITULATION AND IMPLICATIONS 

81

reform case, discussed in depth in Chaps. 3 and 4, shows just how difficult it is for legislative entrepreneurs to build and maintain a majority in support of a legislative initiative, even within a single party.

What Does It Mean to Be a Realist? In my ideal world, as I noted in Chap. 1, vast numbers of people would read this book and be persuaded; as a result, liberal and conservative realists would dominate policymaking. Idealists on the left and right would contribute primarily by identifying visions of where we should go as a nation that other, more realistic, policymakers could pursue incrementally. Given the fallible and self-interested nature of man, our social, political, and economic systems will always need reform, creating continual opportunities for liberal realists. At the same time, all human attempts at reform will necessarily be both imperfect and incomplete, warranting challenges by conservative realists (Niebuhr, 1940, 1949). But what does it mean to be a realist, exactly? Am I being an idealist when I argue for incrementalism as a prescriptive model of how policies should be made? Am I guilty of making optimistic assumptions about human nature when I call on policymakers to change their behavior in the ways I suggest? Isn’t partisan incrementalism just another idealistic attempt to approximate rational decision-making in a world constrained by bounded rationality (Simon, 1976)? I would respond as follows. Focusing primarily on incremental policy initiatives is only one of five core elements of incrementalism, all of which are realistic rather than idealistic. For example, it is realistic to expect conflicts over problem definition. It is realistic to utilize available information by building on past policies whenever possible. It is realistic to focus primarily on incremental alternatives, both because they are more viable politically and because they are easier to evaluate when they fail to work perfectly. It is realistic to expect that proposals will need to be tapered down to secure enough votes to pass a bill. Finally, it is realistic to expect policies to fail initially and to improve over time through a succession of incremental rationalizing policies. We saw all these elements come into play in Chap. 4. By contrast, it is idealistic to expect public problems to emerge well defined in a form that will be readily accepted by all participants. It is foolish to throw away the information that could be gained by building on past policies—unless, of course, the existing policy is too broken to fix. It

82 

M. T. HAYES

is naïve to expect holistic approaches to social engineering to be politically viable and incredibly naïve to believe we can easily figure out how to fix them once they fail. It flies in the face of all experience to expect consensus on objectives and policy consequences for most issues. Rejecting compromise—insisting that others accept your proposals without modification— ignores the fact that other participants bring information to the table that you do not have and refuses to recognize that they have as much right to pursue their interests as you do. Finally, it is extraordinarily naïve to believe you can solve chronic public problems once-and-for-all in a single step. In short, it is idealistic to believe there are no limitations on our ability to design effective policies and secure their enactment in pristine form— or, in Morgenthau’s terms, to believe that a rational, moral order is attainable and its establishment merely a matter of political will. By contrast, it is realistic to recognize how the policy process really works and pursue one’s objectives within its constraints.

Voters Must Also Become Realists In my ideal world, voters would also become more realistic, preferring incremental proposals over non-incremental ones. There is evidence that this may already be happening. Greg Berman and Aubrey Fox recently conducted a survey of more than 1000 registered voters asking them to choose between three alternatives: (1) I would generally prefer that government make big, bold changes quickly, (2) I would generally prefer that government make small, gradual changes over time, or (3) I would generally prefer that government not change things (Berman & Fox, 2023, 62–64). Only 34% preferred bold change, 45% preferred incremental change, and 21% preferred no change at all. If voters really are starting to prefer incremental change, elected officials should respond by offering voters the kinds of policies they find attractive. In theory, at least, realistic voters should give rise to realistic leaders eventually. More realistic voters would also emphasize substance over symbolism. The rise of social media in recent decades has significantly increased the potential for policymakers to dispense symbolic reassurances without seriously addressing the substance of political issues (Edelman, 1964). Edelman saw organized interests pursuing economic advantages at the expense of the public. When a dramatic event makes the unorganized mass public aware of an issue, creating widespread anxiety, the conflict is socialized (Schattschneider, 1960) and the pursuit of private wealth by political

6  RECAPITULATION AND IMPLICATIONS 

83

means is potentially exposed. Consequently, policymakers respond by enacting public policies that seem to address the current problem but are merely symbolic reassurances designed to restore voters’ normal inattention. Cosmetic policy responses allow politicians to pose as reformers when they are really preserving a system that advantages interests that are organized and politically aware. Similarly, David Mayhew (1974) recognized that reelection-minded legislators may position themselves on the “right” side of an issue without making a genuine attempt to solve the problem through legislation. On divisive issues, the only thing that contributes to securing reelection is making sure constituents know the legislators share their values and are fighting for them. Winning on the issue does not matter. Today, it is far too easy for individual members of Congress to pretend they are addressing a serious public problem by the simple act of sending out a tweet. There is no need to introduce a bill or build a coalition sufficient to pass legislation. This performative politics is entirely cosmetic; there is no attempt to develop a serious solution, just a successful effort to get on the right side of the issue. To make matters worse, former President Trump took symbolic politics to an even higher level. Where Edelman saw elected officials as dispensing symbols only when forced to by a focusing event that aroused public concern, President Trump deliberately created anxieties to arouse his base and then dispensed symbolic reassurances to them. His goal was not to restore political quiescence but rather to reinforce his image as a tough leader who fought for them. Whether it was an invasion of immigrants, Chinese trade policies, the “deep state,” or liberals bent on socialism, Trump continually stoked fears to keep his base anxious. (He is still doing this, as of this writing, even though he is out of office.) From Trump’s perspective, the great advantage to this strategy was that he never had to deliver any real policy successes. To the contrary, it was enough to identify a series of enemies and then do battle with them. In this upside-down world, Trump’s policy failures were accepted as evidence of the strength of the evil forces against which he fought, making him appear indispensable. Trump and his imitators on both the left and the right are neither realists nor idealists. Rather, they fall into a third category we might call agitators. For all their differences, realists and idealists share a commitment to making good public policy. By contrast, agitators throw things into disorder, advancing themselves by undermining the legitimacy of the broader

84 

M. T. HAYES

political system. Although these agitators typically claim to be acting on principles, they show no interest in working to solve pressing public problems. Incendiary rhetoric enhances their personal power. When voters settle for symbols over substance, they absolve policymakers from responsibility for addressing public problems. Far too much of our politics today consists of posturing designed to dupe credulous voters. We must demand more from our elected officials and stop rewarding them for posing as our saviors on twitter. As former Speaker of the House John Boehner observed (2021, xiv), too many voters “…reward amateurs who know and care nothing about governing instead of having faith in seasoned officials with experience and patience, who are trying to make some changes to a democracy where change is meant to be hard.”

Partisan Incrementalism in Parliamentary Systems My focus throughout this volume has been on partisan incrementalism as a method of policymaking within the United States. Can this analysis be generalized to other western democracies? Certainly, the factors contributing to the breakdown of rational decision-­making would be operative within any political system. The factors that make incrementalism prevalent—conflict over objectives and disagreement about policy consequences—should be especially pronounced within those parliamentary systems in which proportional representation systems give rise to multiple parties. The need to negotiate with rival parties to form coalitions should make achieving agreement on objectives and policy consequences even harder. By contrast, partisan incrementalism should work particularly well within first-past-the-post parliamentary systems. Where the Parliament is sovereign—and the House of Commons is unchecked by the upper house—effective power is centralized, facilitating coherent action through majority rule. Centralizing power in the House of Commons also prevents divided government, thus making it easier for voters to hold policymakers accountable for performance in office (Forsey, 1991; Norton, 2010). The American system has different strengths of its own. It enhances deliberation by making the legislature independent of the executive, enabling it to defeat presidential initiatives without bringing down the government. It also facilitates policy entrepreneurship by individual

6  RECAPITULATION AND IMPLICATIONS 

85

members, whose legislative initiatives do not have to be adopted as part of the Cabinet’s agenda (in the UK) or the president’s agenda (in the United States). A preference for one system over the other will depend on which values matter more to the individual observer. Whatever the form of the democratic political system, incrementalism remains the best way to make successful public policies because it recognizes both the inevitability of conflict over values and limitations on our available knowledge base. By approaching policy proposals as experiments from which we can learn, it optimizes available information by building on past policies and improving them gradually through seriality. Major changes are almost never achieved in a single step, and where they are they are unlikely to work well. Learning from mistakes is essential, and the steady accumulation of small steps facilitates this. Non-incremental policy changes are more likely to be both successful and enduring where rationalizing breakthroughs emerge after years of experience with policies, and such breakthroughs are more likely to result from partisan incrementalism. Incrementalism is so prevalent, and works so well, precisely because it acknowledges man’s intellectual and moral limitations.

References Berman, G., & Fox, A. (2023). Gradual: The Case for Incremental Change in a Radical Age. Oxford University Press. Binder, S.  A., & Smith, S.  S. (1996). Politics or Principle: Filibustering in the United States Senate. Brookings Institution. Boehner, J. (2021). On the House: A Washington Memoir. St. Martin’s Press. Cyert, R. M., & March, J. G. (1963). A Behavioral Theory of the Firm. Prentice-Hall. Edelman, M. (1964). The Symbolic Uses of Politics. University of Illinois Press. Forsey, E. (1991). How Canadians Govern Themselves (3rd ed.). Public Information Office, House of Commons, Canada. Mayhew, D. R. (1974). Congress: The Electoral Connection. Yale University Press. Murray, M. (2022, October 23). “Anger on Their Minds”: NBC Poll Finds Sky High Interest and Polarization Ahead of Midterms. NBC News poll finds sky-­ high interest and polarization ahead of midterms (cnbc.com). Niebuhr, R. (1940). Christianity and Power Politics. Charles Scribner’s Sons. Niebuhr, R. (1949). Moral Man and Immoral Society: A Study in Ethics and Politics. Scribner. Norton, P. (2010). The British Polity (5th ed.). Routledge.

86 

M. T. HAYES

Rossiter, C. (Ed.). (1999). The Federalist Papers. New American Library, Mentor Books. Schattschneider, E.  E. (1960). The Semi-Sovereign People: A Realist’s Guide to Democracy in America. Holt, Rinehart, and Winston. Shapiro, I. (2022). The Betrayal: How Mitch McConnell and the Senate Republicans Abandoned America. Rowman and Littlefield. Simon, H. A. (1976). Administrative Behavior: A Study of Decision-Making Process in Administrative Organization (3rd ed.). Free Press.

Index

A Aid to Families with Dependent Children (AFDC, formerly ADC, Aid to Dependent Children), 8, 28, 33, 37, 39, 46–48, 51–55 B Bennett, Bill, 45 Berman, Greg, 19, 56, 82 Besharov, Douglas, 48 Biases to the interest group universe advantage of institutions over membership groups, 60, 61 advantage of small groups over large groups, 60 socioeconomic bias, 61 Biden, Joseph R., 2, 4, 21, 79 Block grants, 8, 28, 33, 34, 36, 37, 51–54 Breakthrough vs. rationalizing policies, 22 Build Back Better bill, 2

Building on past policy, 17, 24, 44, 46–49, 81, 85 Burke, Edmund, 5, 17 Burke, Sheila, 34 C Carter, Jimmy, 39, 47 Chafee, John, 34 Childcare block grant, 52, 54 Child nutrition block grant, 52, 53 Child protection block grant, 52, 53 Churchill, Winston, 67 Clinton, Bill, 28, 30–33, 35–37, 40, 41, 48–52, 54, 71 Conflict over objectives, 14–16, 84 Congressional committees conference committee, 35, 49, 51 Finance committee (Senate), 34 Rules Committee, 34 Ways and Means Committee (House), 33

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Hayes, Incrementalism and Policymaking in the USA, https://doi.org/10.1007/978-3-031-38485-1

87

88 

INDEX

Consensual vs. conflictual knowledge, 23 Contract with America, 9, 28, 32 Cumulative vs. dispersed political resources, 63–64 D Daschle, Tom, 34 Democratic Leadership Council, 50 Disability Insurance, 46 Displacement of conflict, 71 Dole, Bob, 37 E Earned Income Tax Credit (EITC), 36 Elazar, Daniel, 56 Ellwood, David, 48 Engler, John, 36, 49 Equality of treatment under the rule of law, 70 Equity horizontal, 39 vertical, 39 F Faircloth, Lauch, 45 Fair Sentencing Act of 2010, 71 Family Assistance Plan, 47 Family Support Act, 31, 48 Federal role breakthroughs, 22, 23, 38, 55, 78 Food Stamps, 33, 34, 46, 52, 53 Fox, Aubrey, 19, 56, 82 Free rider problem, 60–62, 72 Friedman, Milton, 39 Front door institutional changes, 55

G General Assistance, 46 Gun control legislation, 2 H Hamilton, Alexander, 56, 79 Haskins, Ron, 9, 31–38, 45, 46, 48–54 Hogwood, Brian, 54, 55 Holistic vs. piecemeal engineering, 19 I Idealism conservative idealists, 5, 6 conservative realists, 4–8, 81 defined, 4 Incrementalism advantages over rational decisionmaking, 18–19 as an alternative to rationality, 17–18 conditions for fair operation, 59–60 main elements, 17–18, 43–44 Incremental rationalizing policies, 22–24, 38, 39, 55, 78, 81 J Jencks, Christopher, 48 JOBS program, 31, 32, 40, 41, 48, 49, 53 K Kaus, Mickey, 48 Kemp, Jack, 45 Kirk, Russell, 5 Knight, Frank H., 10 Krauthammer, Charles, 45

 INDEX 

L Legitimacy as interest group resource class interest of business equated with the public interest, 66 defined, 64 disadvantages facing African Americans, 71–72 Levin, Yuval, 6, 7 Life cycle of policies, 22, 25 Lugar, Richard, 34, 53 M Madison, James, 79 Maintenance of effort (MOE), 36 Manchin, Joe, 2 Market as prison, 66 Mead, Lawrence, 45, 47, 48, 55, 56 Medicaid, 33, 36, 46, 52, 53 Medicare, 22, 46, 64 Meyers, Jan, 51 Minogue, Kenneth, 4 Mobilization of bias, 71, 72 Morgenthau, Hans J., 3–5, 7, 82 Mosely, Phil, 50 Murray, Charles, 45, 47, 51 Murray “lite,” 51, 53 N National Governors Association, 36 Negative income tax, 39, 40, 47 Negative partisanship, 80 New Democrats, 50 Nixon, Richard, 39, 46, 47 No “given” problem, 24, 44 Nondecisions barrier I nondecisions, 69, 70 barrier II nondecisions, 69, 70, 73 Non-incremental change as “calculated risk,” 21 defined, 20

89

preconditions for, 21–22 public attitudes towards, 82 Republican proposals to block grant Medicaid and Food Stamps, 52 Republican proposal to block grant 336 social programs, 52 as response to aroused public opinion, 21, 22 response to “wars, revolutions, and grand opportunities,” 21 Novak, Michael, 48 O Oakeshott, Michael, 4 Official groups, 62 Opt outs vs. opt ins, 51 Organizing issues out of politics, 71 P Packwood, Robert, 34 Partisan incrementalism, 8, 10, 27–41, 63, 73, 78, 79, 81, 84–85 Performance failures as best case for taking the next step, 19 Peters, Guy, 54 Policies as experiments, 10, 38 Policy entrepreneurs, 1, 30, 49, 71 Political resources available to groups, 63–64 Practitioner veto, 56 Pre-formed majorities, 21, 24 Privileged position of business in capitalist societies, 65–67 Program for Better Jobs and Income (PBJI), 47 Pure problem of knowledge base, 16, 30, 32 Pure problem of value conflict, 15

90 

INDEX

R Rationalizing policies, 22–24, 29, 38, 39, 55, 78, 81 Realism conservative realists, 4–8, 81 defined, 7 liberal realists, 4, 6–8, 81 as normative model, 7 “Real Welfare Reform” bill, 45 Reauthorization of TANF, 61 Remediality, 44 Republican principle, 79 S Satisficing, 18 Schattschneider, E. E., 9, 27, 61, 71, 72, 82 Schumer, Chuck, 2 Senate filibuster dangers flowing from minority veto, 79–81 elimination of, as incremental change, 80 Sequential attention to goals, 80 Seriality (convergence through successive approximations), 18, 22, 29, 40, 44, 63, 73, 78, 85 Shaw, E. Clay, 31, 36, 45, 50

Social Security Act of 1935, 22, 38, 40, 46 Sowell, Thomas, 3, 4 Speculative augmentation, 21 Street level bureaucrats, 55 Supplemental Security Income (SSI), 46, 54 T Talent, Jim, 45 Taper down from the optimal to the acceptable, 17, 43, 49–50 Temporary Assistance for Needy Families (TANF), 28, 37, 38, 41, 52, 54, 55, 61, 62 Thompson, Tommy, 36, 38, 49 Trade-offs, 14–16, 18 U Unemployment Compensation, 46 W Warren, Elizabeth, 3–5 Weber, Vin, 45 Weber-Shaw bill, 45 Wednesday Group, 31, 45, 46 Will, George, 45