Understanding Public Policy: Theories and Issues [2 ed.] 1137545194, 9781137545190

Table of contents :
Title Page
Copyright
Contents
List of Figures
List of Tables
List of Boxes
List of Abbreviations
Preface
1. Introduction to Policy and Policymaking
Introduction: Why Should We Study Public Policy?
The General Approach of This Book
Simple Models Help Us Understand How Policy Is Not Made
Policymaker Psychology
Policymaking Complexity
The Power of ‘the Centre’ Is Limited
Policy Networks and Subsystems are Pervasive
Complex Policymaking Environments Limit Policymaker Control
Ideas Matter
How to Analyse Policy and Policymaking
The Structure of the Book
Conclusion
2. What Is Policy and Policymaking?
Introduction: The Need to Define Policy and Policymaking
What Is Public Policy?
Measuring Public Policy
Narratives of Public Policy
Frameworks, Theories, Models, and Heuristics
What Is the Policy Cycle?
Agenda Setting and Policy Formulation
Implementation
Top-down and Bottom-up Implementation
Evaluation
Policy Maintenance, Succession, and Termination
Beyond the Policy Cycle
Conclusion
3. Power and Public Policy
Introduction: The Centrality of Power to Public Policy Studies
Definitions of Power
Three Dimensions of Power: Winning Key Decisions, Agenda Setting, and Thought Control
The First Dimension: A Debate on Elitism and Pluralism
The ‘Second Face’ of Power
The Third Dimension of Power
Observing the Unobservable
All Assessments of Power are Empirical and Normative
Beyond the Third Dimension: Foucault and Habermas
Power and Critical Theory: The Emancipatory Role for Research
Are Such Forms of Power ‘Structural’?
Where Does the Role of Power Stop and Ideas Begin?
Conclusion
4. Bounded Rationality and the Psychology of Policymaking
Introduction: The Profound Importance of Bounded Rationality
Comprehensive and Bounded Rationality
Bounded Rationality Is More Important Than Ever
Incrementalism
Is Incrementalism ‘Universal’ and Inevitable?
The Narrative Policy Framework
Social Construction and Policy Design
Is Social Construction and Policy Design ‘Universal’ and Inevitable?
Conclusion
5. Institutions and New Institutionalism
Introduction: Institutions Matter, but What are Institutions?
Identifying Formal and Informal Institutions
What Exactly Is an Institution? What Is Institutionalism?
Key Variants of New Institutionalism
Historical Institutionalism
Rational Choice Institutionalism
Normative and Sociological Institutionalism
Discursive and Constructivist Institutionalism
Feminist Institutionalism
Empirical vs. Network Institutionalism? Diverging and Converging Policy Styles
Conclusion
6. Structures, Environments, and Complex Systems
Introduction: Structure and Agency in the Policy Process
Do Structural Factors Determine Policy and Policymaking?
The Economic Context: Marxism and Globalization
Inheritance before Choice, and Policy Succession
The Evolutionary Metaphor: Context as a Policymaking Environment
The Policy Process as a Complex System
Conclusion
7. Collective Action Problems in Public Policy
Collective Action Problems in Rational Choice and Game Theory
The Prisoner’s Dilemma, Logic of Collective Action, and Tragedy of the Commons
Government as One Institutional Solution to Collective Action Problems
Institutional Analysis and Development (IAD) Framework
Key Approaches in the IAD’s Extended Family
Managing Common Pool Resources (CPRs) and Avoiding Tragedies
Social-Ecological Systems (SES) Framework
Polycentric Governance and Institutional Collective Action
Institutional Collective Action (ICA)
Institutional Complexity and the Ecology of Games
Conclusion
8. Multi-level Governance and Multi-centric Policymaking
Introduction: From One to Many ‘Centres’ of Policymaking
What Is Governance?
Global Governance
Governance as Systems Over Which the Central Government Has Limited Control
Governance as the Proposed or Actual Reform of the State
Governance as a Problem: The Westminster Model
Multi-level Governance (MLG) and the European Union (EU)
Empirical and Normative Visions: What MLG Is and Should Be
MLG and International Comparisons
Comparing Political Systems
Comparing Approaches to the Study of Political Systems
Comparing MLG and Policy Theories
Conclusion
9. Punctuated Equilibrium Theory
Introduction: The Profound Importance of Policymaker Attention
Why ‘Punctuated Equilibrium’ Theory (PET)?
PET’s Original Focus: Agendas and Instability
Policy Subsystems, Monopolies, and Subsystems
Issue Networks and Subsystems
Agenda Setting and Attention
Problem Definition
Problem Definition, Policy Monopolies, and Venue Shopping
Case Studies of Punctuated Equilibrium: ‘Some Issues Catch Fire’
From Case Studies to the ‘General Punctuation Hypothesis’
Government Budgets: Hyper-incremental and Dramatic Policy Change
The Comparative Policy Agendas Project
Conclusion
10. The Advocacy Coalition Framework
Introduction: Coalitions, Policy-oriented Learning, and Policy Change
A Picture of the Advocacy Coalition Framework (ACF)
The Role of Beliefs to Address Bounded Rationality
The Role of Beliefs to Boost Cooperation and Help Actors Exercise Power
The Dynamics of ‘Policy-Oriented Learning’
Policymaking Stability and Instability, Policy Continuity and Change
The ACF Goes International: New Empirical Applications
Conceptual Revisions and New Directions
Conclusion
11. Ideas and Multiple Streams Analysis
Introduction: The Role of Ideas in Policymaking
Defining Ideas
Ideas as the Primary Source of Explanation: Viruses and Norms
Hall’s Policy Paradigms and Third-Order Change
Multiple Streams Analysis (MSA)
A Garbage Can Model of Organizational Choice
The ‘Window of Opportunity’
Why Are the Three Streams Separate and How Do They Come Together?
The Impressive Generalizability of MSA
Conclusion
12. Policy Learning and Transfer
Introduction: The Politics of Policy Learning
Policy Learning: What Does It Mean? What are the Main Types?
Approaches to the Study of Policy Transfer
Lesson-drawing
Policy Diffusion
Policy Convergence
Policy Transfer: Who Does It?
Why Transfer? Is It Voluntary?
Coercive Transfer: How Is It Manifest and Demonstrated?
What Is Transferred? What Makes Policy Transfer Distinctive?
What Does ‘Successful’ Policy Transfer Mean?
How to Encourage ‘Evidence-based’ Policy Learning and Transfer
Conclusion
13. Conclusion: Policy Theory as Accumulated Wisdom
Introduction: Combining Theoretical Insights Is Useful but Tricky
Defining Policy and Telling a Story of Policy Change
A Theory-based Story of Policymaking
The Politics of Evidence-based Policymaking
Using Multiple Theories: Three Cautionary Tales
Policy Theory beyond the ‘West’ or Global North
Conclusion
Bibliography
Index

Citation preview

‘Paul Cairney has done the near impossible and made an outstanding textbook even better. The public policy landscape is always evolving and this second edition captures the new features, colours and dynamics in rich and vivid detail. It should be the go-to work for experienced scholars and students alike.’ – Allan McConnell, University of Sydney, Australia ‘This volume stands as the essential guide for students who seek to understand the forces that shape political decisions and the policy process. Clearly written and full of theoretical and empirical insights from contemporary issues, I have no doubt this book will have a prominent place in policy scholars’ reading lists.’ – Claire A. Dunlop, University of Exeter, UK ‘Rarely does a policy textbook title match its content better. Understanding Public Policy provides an efficient introduction to complex policy theory that is accessible to new learners by demystifying (necessary) policy jargon through plain language and current examples. Cairney’s work incorporates the views of top scholars across the policy process literature and around the world. New learners and experienced readers will find the book a useful reference for understanding and researching public policy.’ – Chris Koski, Reed College, USA ‘Paul Cairney’s book is the essential work for any public policy course. It clearly maps out all the key theories, and is packed with definitions and examples based on the latest research. It also offers a crucial insight into how policy works, and what happens when theory meets reality. It is usable, lucid and very well written.’ – Ben Worthy, Birkbeck, University of London, UK ‘Understanding Public Policy does what is says on the tin. It allows students, social scientists and those involved in public policy to access, understand and apply the most important findings of contemporary theories and empirical studies of the policy process. Cairney’s volume delivers with crystal clear language, powerful narratives, and a rich repertoire of examples.’ – Claudio Radaelli, University College London (UCL), UK ‘The value of Understanding Public Policy comes from the fact it is the only introductory text about public policy that properly locates the subject within the larger theories and ideas of political science. This means that policy and politics can be taught and understood simultaneously through a single high-quality source.

Understanding Public Policy remains an essential text for any social science student seeking to understand the relationship between policy, politics and society.’ – Alastair Stark, University of Queensland, Australia ‘In the first edition of this text Paul Cairney showed us that the untamed study of public policy could be conveyed in a way that is both accessible and useful, and yet also comprehensive and sophisticated. In this second edition he extends these qualities. At this point it is safe to say that Cairney’s aerial view of the study of public policy is itself becoming a standard in the field.’ – Michael D. Jones, Editor-in-Chief, Policy Studies Journal ‘Cairney’s second edition synthesizes the rich body of scholarship on public policy and policymaking to distil lessons on the multifaceted factors that structure, influence and produce public policy. This book serves as an essential resource for diagnosing the nature of policymaking while honouring its complexity and the diversity of lenses through which we can explore it.’ – Tanya Heikkila, University of Colorado Denver, USA ‘Paul Cairney’s book is exceptional in capturing the theoretical and conceptual complexity of the approaches to understanding the policy process. Cairney’s approach is unique in laying the foundation for understanding these approaches. The book will be useful to both students of the policy process as well as practitioners hoping to gain a lay of the land in policy studies.’ – Sam Workman, University of Oklahoma, USA

Textbooks in Policy Studies Series Editor: Paul Cairney, University of Stirling, UK Editorial Advisory Group: Jennifer Curtin, University of Auckland, New Zealand; Jordi Diez, University of Guelph, Canada; Claire Dunlop, University of Exeter, UK; Tanya Heikkila, University of Colorado Denver, USA; Karin Ingold, University of Bern, Switzerland; Allan McConnell, University of Sydney, Australia; Daniel Nohrstedt, Uppsala University, Sweden; Jale Tosun, Heidelberg University, Germany; Chris Weible, University of Colorado Denver, USA. There has never been a more important time in which to understand the dynamics of public policy. Key events across the globe have prompted debate about political crisis, from ‘post truth’ and emotionally-driven politics to the sense that international institutions or agreements are not well-equipped to solve pressing global crises such as public health epidemics, poverty and climate change. Yet, a key feature of policy studies is the identification and explanation of policymaking stability and policy continuity, with salient issues gathering most of our attention while most issues are processed out of the public spotlight. A major problem for democracy is that few people know about the choices governments make in their name. We need as many perspectives as possible on how such policy dynamics operate, and what effect they have on developments in different regions and policy areas. Such developments will not be covered fully by research monographs, with a limited focus and audience. We also need textbooks which combine academic rigour with a broad analytical focus and an appeal to a wide audience. This series helps develop this focus on research and learning, serving the global market of undergraduate, postgraduate and professional education in public policy and related disciplines such as political science and social policy. PUBLISHED Anneliese Dodds Comparative Public Policy (second edition) Paul Cairney Understanding Public Policy (second edition) FORTHCOMING Christoph Knill and Jale Tosun Public Policy: A New Introduction (second edition)

Madeleine Pill Urban Policy and Politics Alison Ritter Drug Policy PLANNED Researching Public Policy Global Public Policy … and more

UNDERSTANDING PUBLIC POLICY Theories and Issues 2nd edition Paul Cairney

© Paul Cairney, under exclusive licence to Springer Nature Limited 2011, 2020 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6–10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. This edition published 2020 by RED GLOBE PRESS Previous edition published under the imprint PALGRAVE Red Globe Press in the UK is an imprint of Springer Nature Limited, registered in England, company number 785998, of 4 Crinan Street, London N1 9XW. Red Globe Press® is a registered trademark in the United States, the United Kingdom, Europe and other countries. ISBN 978–1–137–54519–0 hardback ISBN 978–1–137–54518–3 paperback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. A catalog record for this book is available from the Library of Congress.

To be honest, when I wrote the dedication for the first edition, I did not anticipate so many path dependent issues with the second edition. Still, here goes. For my still-lovely partner Linda, our stillbeautiful-but-not-really-children-anymore children, Evie, Alfie and Frankie, their partners (if applicable), our grandchildren Charlie and Teddy, and our still-smelly-but-handsome dogs Coco and Mabel.

Contents List of Figures List of Tables List of Boxes List of Abbreviations Preface 1

Introduction to Policy and Policymaking Introduction: Why Should We Study Public Policy? The General Approach of This Book Simple Models Help Us Understand How Policy Is Not Made Policymaker Psychology Policymaking Complexity The Power of ‘the Centre’ Is Limited Policy Networks and Subsystems are Pervasive Complex Policymaking Environments Limit Policymaker Control Ideas Matter How to Analyse Policy and Policymaking The Structure of the Book Conclusion

2

What Is Policy and Policymaking? Introduction: The Need to Define Policy and Policymaking What Is Public Policy? Measuring Public Policy Narratives of Public Policy Frameworks, Theories, Models, and Heuristics What Is the Policy Cycle? Agenda Setting and Policy Formulation Implementation Top-down and Bottom-up Implementation Evaluation Policy Maintenance, Succession, and Termination Beyond the Policy Cycle

Conclusion 3

Power and Public Policy Introduction: The Centrality of Power to Public Policy Studies Definitions of Power Three Dimensions of Power: Winning Key Decisions, Agenda Setting, and Thought Control The First Dimension: A Debate on Elitism and Pluralism The ‘Second Face’ of Power The Third Dimension of Power Observing the Unobservable All Assessments of Power are Empirical and Normative Beyond the Third Dimension: Foucault and Habermas Power and Critical Theory: The Emancipatory Role for Research Are Such Forms of Power ‘Structural’? Where Does the Role of Power Stop and Ideas Begin? Conclusion

4

Bounded Rationality and the Psychology of Policymaking Introduction: The Profound Importance of Bounded Rationality Comprehensive and Bounded Rationality Bounded Rationality Is More Important Than Ever Incrementalism Is Incrementalism ‘Universal’ and Inevitable? The Narrative Policy Framework Social Construction and Policy Design Is Social Construction and Policy Design ‘Universal’ and Inevitable? Conclusion

5

Institutions and New Institutionalism Introduction: Institutions Matter, but What are Institutions? Identifying Formal and Informal Institutions What Exactly Is an Institution? What Is Institutionalism? Key Variants of New Institutionalism Historical Institutionalism Rational Choice Institutionalism

Normative and Sociological Institutionalism Discursive and Constructivist Institutionalism Feminist Institutionalism Empirical vs. Network Institutionalism? Diverging and Converging Policy Styles Conclusion 6

Structures, Environments, and Complex Systems Introduction: Structure and Agency in the Policy Process Do Structural Factors Determine Policy and Policymaking? The Economic Context: Marxism and Globalization Inheritance before Choice, and Policy Succession The Evolutionary Metaphor: Context as a Policymaking Environment The Policy Process as a Complex System Conclusion

7

Collective Action Problems in Public Policy Collective Action Problems in Rational Choice and Game Theory The Prisoner’s Dilemma, Logic of Collective Action, and Tragedy of the Commons Government as One Institutional Solution to Collective Action Problems Institutional Analysis and Development (IAD) Framework Key Approaches in the IAD’s Extended Family Managing Common Pool Resources (CPRs) and Avoiding Tragedies Social-Ecological Systems (SES) Framework Polycentric Governance and Institutional Collective Action Institutional Collective Action (ICA) Institutional Complexity and the Ecology of Games Conclusion

8

Multi-level Governance and Multi-centric Policymaking Introduction: From One to Many ‘Centres’ of Policymaking What Is Governance?

Global Governance Governance as Systems Over Which the Central Government Has Limited Control Governance as the Proposed or Actual Reform of the State Governance as a Problem: The Westminster Model Multi-level Governance (MLG) and the European Union (EU) Empirical and Normative Visions: What MLG Is and Should Be MLG and International Comparisons Comparing Political Systems Comparing Approaches to the Study of Political Systems Comparing MLG and Policy Theories Conclusion 9

Punctuated Equilibrium Theory Introduction: The Profound Importance of Policymaker Attention Why ‘Punctuated Equilibrium’ Theory (PET)? PET’s Original Focus: Agendas and Instability Policy Subsystems, Monopolies, and Subsystems Issue Networks and Subsystems Agenda Setting and Attention Problem Definition Problem Definition, Policy Monopolies, and Venue Shopping Case Studies of Punctuated Equilibrium: ‘Some Issues Catch Fire’ From Case Studies to the ‘General Punctuation Hypothesis’ Government Budgets: Hyper-incremental and Dramatic Policy Change The Comparative Policy Agendas Project Conclusion

10 The Advocacy Coalition Framework Introduction: Coalitions, Policy-oriented Learning, and Policy Change A Picture of the Advocacy Coalition Framework (ACF) The Role of Beliefs to Address Bounded Rationality The Role of Beliefs to Boost Cooperation and Help Actors Exercise Power

The Dynamics of ‘Policy-Oriented Learning’ Policymaking Stability and Instability, Policy Continuity and Change The ACF Goes International: New Empirical Applications Conceptual Revisions and New Directions Conclusion 11 Ideas and Multiple Streams Analysis Introduction: The Role of Ideas in Policymaking Defining Ideas Ideas as the Primary Source of Explanation: Viruses and Norms Hall’s Policy Paradigms and Third-Order Change Multiple Streams Analysis (MSA) A Garbage Can Model of Organizational Choice The ‘Window of Opportunity’ Why Are the Three Streams Separate and How Do They Come Together? The Impressive Generalizability of MSA Conclusion 12 Policy Learning and Transfer Introduction: The Politics of Policy Learning Policy Learning: What Does It Mean? What are the Main Types? Approaches to the Study of Policy Transfer Lesson-drawing Policy Diffusion Policy Convergence Policy Transfer: Who Does It? Why Transfer? Is It Voluntary? Coercive Transfer: How Is It Manifest and Demonstrated? What Is Transferred? What Makes Policy Transfer Distinctive? What Does ‘Successful’ Policy Transfer Mean? How to Encourage ‘Evidence-based’ Policy Learning and Transfer Conclusion 13 Conclusion: Policy Theory as Accumulated Wisdom

Introduction: Combining Theoretical Insights Is Useful but Tricky Defining Policy and Telling a Story of Policy Change A Theory-based Story of Policymaking The Politics of Evidence-based Policymaking Using Multiple Theories: Three Cautionary Tales Policy Theory beyond the ‘West’ or Global North Conclusion Bibliography Index

List of Figures 2.1 2.2 9.1 10.1 12.1 13.1

The generic policy cycle Key elements of policy choice in a complex policymaking environment A broad picture of the general punctuation hypothesis in action The advocacy coalition framework flow diagram The Dolowitz and Marsh policy transfer continuum An image of the policy process

List of Tables 4.1 5.1 7.1 7.2 8.1 8.2 9.1

Social construction and power Lijphart’s majoritarian–consensus dichotomy The prisoner’s dilemma Factors essential to self-organization in social-ecological systems Westminster model versus multi-level governance Types of multi-level governance Typology of policy networks

List of Boxes 2.1 2.2 2.3 2.4 3.1 3.2 3.3 4.1 4.2 4.3 4.4 4.5 5.1 5.2 5.3 6.1 6.2 6.3 6.4 7.1 7.2 7.3 8.1 8.2 8.3 9.1 9.2 9.3

Definitions and descriptions of public policy ‘Evidence-based’ policy and policymaking Policy tools: their use and potential abuse Types of policy problems Discussions of power The semi-sovereign people: A realist’s view of democracy Power and systematic luck Cognitive biases and fast and frugal heuristics The meaning of incrementalism Heresthetic and the manipulation of choice Narrative versus frames NPF positivism vs. interpretivist postpositivism Descriptions of institutions and new institutionalism Key questions for new institutionalism Path dependence and immigration Structural factors and policy conditions Japan’s nuclear policy The policy environment and ‘evidence-based policymaking’ Are complex systems theories deterministic? Nudge theory and behavioural public policy Collective action during repeated games Key Institutional Analysis and Development (IAD) terms Empirical and conceptual descriptions of multi-level governance Can Westminster take back control after Brexit? The normative debate: Should power be concentrated in the ‘centre’? The changing world of policy networks and group– government relations Participation in American politics: The dynamics of agenda building Up and down with ecology: The ‘issue attention cycle’

9.4 10.1 10.2 11.1 11.2 11.3 11.4 12.1 12.2 12.3 12.4 12.5 13.1 13.2 13.3 13.4

Key terms for statistical analysis Policy Conflict Framework (PCF) The case of hydraulic fracturing in ten countries Key quotations on ideas and interests Ideas and the case study of tobacco Policy entrepreneurs and ‘evidence-based policymaking’ The comparative application of multiple streams analysis (MSA) Individual and collective learning: Lessons from the advocacy coalition framework (ACF) The mechanisms, products, and facilitators of policy learning Examples of widespread policy convergence, diffusion, or transfer The policy transfer window Richard Rose’s ‘ten steps in lesson-drawing’ What is policy, how much has it changed, and why? Theory-informed policy analysis and the ‘practical lessons’ agenda Ways to compare theories of the policy process Theories and applications in China and India: Possibilities and problems

List of Abbreviations ACF CPR DSH EBP EBPM EG EU GDP IAD ICA IGR IMF MLG MSA NPF NPM OECD PCF PET PFT RCT SCPD SES UK USA WM

Advocacy Coalition Framework Common Pool Resource Dye–Sharkansky–Hofferbert (approach) Evidence-based Policy Evidence-based Policymaking Ecology of Games European Union Gross Domestic Product Institutional Analysis and Development (framework) Institutional Collective Action Intergovernmental Relations International Monetary Fund Multi-level Governance Multiple Streams Analysis Narrative Policy Framework New Public Management Organisation for Economic Co-operation and Development Policy Conflict Framework Punctuated Equilibrium Theory Policy Feedback Theory Randomized Control Trial (note: the first edition used RCT to denote Rational Choice Theory) Social Construction and Policy Design Social-Ecological Systems United Kingdom United States of America Westminster model

Preface In the first edition of this book, I started by moaning that most of my colleagues do not really appreciate the study of public policy. Well, things have changed since then. I have noticed a major increase in attention to policy studies from scholars in many disciplines and practitioners in many organizations. For me, this interest has come from countries like the USA, Canada, Australia, New Zealand, and the UK, as well as several EU member states and organizations like the European Commission, in which academics and practitioners are interested in the ‘impact’ of research on policy. When they want to take some insights from policy theories, where do they look? The field of policy studies, as a whole, is huge and unwieldy. However, if we can synthesize and explain their insights concisely, they offer profoundly important ways to understand the dynamics and effects of policymaking. I’ve tried to help out by writing about policy theories in a collection of complementary ways, such as by combining this book with a series of blog posts (and podcasts) which explain key concepts in the original 1000 words series (https://paulcairney.wordpress.com/1000-words/) and the newer 500 words series (https://paulcairney.wordpress.com/500-words/). Each post gives you the basics and helps you navigate a chapter or section in this book, and then the book introduces you to a much larger literature. You can then choose how far to go to find and use insights from policy theories (although I recommend strongly that you read the source material to gain a full understanding). I have also tried to help by generating an overall sense of what the literature adds up to, and how it might inform people who do not have the time to study it all the time. For example, the second edition now has a greater focus on the highly popular but problematic idea of ‘evidence-based policymaking’ (EBPM). I use policy theory to show why ‘the evidence’ will always be one of many influences on policy. So, if you do not want to jump into the study of many theories,

you can begin with the study of an overarching topic like EBPM and learn how policy theory fits in (there is also a blog page devoted to this subject https://paulcairney.wordpress.com/ebpm/). However, there is still a long way to go before I’m fully satisfied with Understanding Public Policy. In particular, the first edition relied too much on white male (usually US and UK) authors, reinforcing a general tendency of political science scholars to undervalue women and send ‘a very clear message to students about who (white male elites) and what (institutions) are important in political science’ (Amy Atchison, 2017). There are useful discussions taking place about how to modify the ways in which we redesign scholarship for teaching and learning. Some focus on syllabus/module design: Raul Pacheco-Vega (2015) on including ‘under-represented scholars’, Manjeet Ramgotra (2015) on how to include women and non-white men in curricula, Shannon Morreira and Kathy Luckett (2018) on ‘Questions academics can ask to decolonise their classrooms’, and Alice Evans (2018) on a tendency to ‘venerate men as knowledgeable authorities’. A special issue of Political Studies Review also includes reflections on ‘decolonising the curriculum’ by Neema Begum and Rima Saini (2019) and Akwugo Emejulu (2019). I think it will take many editions, over my entire career, before I address these issues really well, and – in this opening discussion - I am making a commitment to do so. In the meantime, I have addressed them quite well in the second edition, by: •

• •

Giving more respect to key literatures that should have appeared more in the first edition, including Social Construction and Policy Design, led by Anne Schneider and Helen Ingram, and the Institutional Analysis and Development framework, led by Elinor Ostrom. Giving more respect to individual pieces of work with high impact in the field, such as by Carol Bacchi and, now that I focus more on the use of evidence and evaluation, Carol Weiss. Expanding on key approaches that only got a little nod in the first edition, including feminist institutionalism, to which my introduction was via Fiona Mackay and Meryl Kenny, and policy learning which has been taken forward in important ways by

•

•

•

•

•

Claire Dunlop, usually as part of a dream team – with Claudio Radaelli – that has done a phenomenal amount of work in this field. Relying slightly less on a notional divide between the study of public administration and public policy, and connecting to similar debates on feminist research in the former (albeit as part of a ‘new conversation’; Carey et al., 2018). Relying less on the excuse that most of the ‘big’ US policy theories were founded and taken forward by men, such as multiple streams analysis initiated by John Kingdon and advanced by Nikolaos Zahariadis. By being a little less lazy – and by noting a rapid rise of interest in this theory – I have incorporated key insights from more people, such as Åsa Knaggård on the ‘problem broker’ and Nicole Herweg and colleagues on comparative applications. Thinking more about the links between comparable literatures that are or are not necessarily ‘badged’ as public policy. For example, in Chapter 3 on power, I discuss work by scholars including Akwugo Emejulu (who signed my copy of one of her books!), Sarah Ahmed, Sarah Childs, and Elizabeth Evans who engage with concepts that you will find discussed a lot in this book, including path dependence, norms, ideas, and power. Further, Chapter 3 has an expanded section on the relationship between decolonization, power, and knowledge claims, which connects issues of power and ideas to the expanded focus on ‘evidence-based policymaking’. Reflecting throughout on the simple but profound maxim: the way we understand the world has a major effect on the ways in which we engage, and whose interests we seek to serve (Guerrina et al., 2018: 253). Redesigning my MPP module guides/syllabus to provide a less narrow reading list (see the MPP page on my blog).

Overall, here are the changes from the first edition: more on policy formulation and tools (Chapter 2), power, knowledge, and inequality (Chapter 3), the psychology of policymaking, and the NPF and SCPD (Chapter 4), feminist institutionalism (Chapter 5), complexity

theory (Chapter 6), the IAD, polycentric governance, and ecology of games (Chapter 7), the links between MLG and polycentric governance (Chapter 8), new developments in PET, the ACF, and MSA (Chapters 9–11), and policy learning and ‘evidence-based’ policy transfer (Chapter 12). You will need to read the book to understand the abbreviations! I also spend more time in the conclusion, and a section in most chapters, on three key themes: what is the story of each theory (or multiple theories), what does policy theory tell us about issues like ‘evidence-based policymaking’ and how ‘universal’ are policy theories designed in the Global North? I then return to the Global North–Global South question in the conclusion. To make space for these new discussions, I’ve spent a bit less time on incrementalism (the old Chapter 5) and combining multiple theories (the old Chapter 13) but you can read more on the latter in Cairney (2013a) and Heikkila and Cairney (2018). In the first edition, I thanked many people for helping with key points and commenting on drafts, with Allan McConnell winning first prize. Allan holds on to that prize for the second edition. He gave me some valuable comments on the penultimate draft, which made it a joy to complete the final draft. Allan also showed me around Sydney in 2018, and he and Iris Kirkpatrick took me out to a memorable dinner. It was called a ‘dégustation’, which means loads of small courses (taster versions of full meals) with drinks to match (the veggie/teetotal version, with herb drinks, was a lot nicer than you might think). In fact, let’s say it was 13 courses so that I can suggest you treat this book like a huge and tasty collection of small meals. I asked for less advice for the second edition as a whole, but still owe a debt to the many people who helped with individual chapters. My new co-author buddies Chris Weible and Tanya Heikkila have influenced (a lot) the way I think, but I can’t really blame them for anything specific. The following people kept me right on key chapters or concepts: Tanya Heikkila (Chapter 7, IAD), Mike Jones (my story of the NPF), Chris Weible (Chapter 10, ACF), Claire Dunlop (Chapter 12, policy learning), Mark Lubell (EG), Andrew Hindmoor (rational choice), Serena (Seo Young) Kim and William Swann (ICA), Andrea Gerlak (the IAD and learning), and Jon Pierce (SCPD). Fiona Mackay gave me some excellent, detailed advice on how to write

about feminist institutionalism in Chapter 5, including giving me some text to use and identifying several gaps in my reading. I rewrote this text to take responsibility for it, but Fiona’s influence on this description remains. Chris Koski helped with Chapter 9 (PET), but I feel like I already paid him back by featuring as a Scottishsounding spinning head on his PowerPoint presentation (the Cairneyval, which works well if you pronounce my name incorrectly, like Barney, instead of to sound like hair). I attended lectures by Sam Workman, Chris Weible, Jennifer Dodge, William Swann, and Serena Kim at the IPPA Spring School in Denver (May 2019), and you will find bits and pieces of those insights in the book. Sam really clarified some key points just before I was about to revise Chapter 9 for the last time (including that PET – unlike most approaches in this book – tends to equate ‘institution’ with organization rather than the rules of organizations; see Chapter 5). Chris did the same for the ACF, allowing me to avoid repeating an embarrassing mistake in the way that I described internal/external ‘shocks’. My co-author and office-buddy Emily St Denny kept me right on new institutionalism, and hers is the only PhD thesis that I have read solely out of interest (and to inform Box 5.2). There is a lot of discussion of evidence and policy in the second edition, based heavily on co-authored work with Kathryn Oliver and Adam Wellstead. I first heard Woodward’s phrase ‘velvet triangle’ from my co-author Kirstein Rummery. I included a quotation from Keynes after hearing it in David Mair’s presentations, and I have learned more generally from David, Lene Topp, and Laura Smillie (at the European Commission Joint Research Centre) during our collaboration on ‘knowledge management for policy’ (Topp et al., 2018). Working with Ruth Mayne and her colleagues at Oxfam helped me think in different ways about policy studies (Mayne et al., 2018). I have drawn – in Chapter 10 – on my contribution to an ECPR paper on the ACF that I co-authored with Hannes Stephan and Irina Timonina, and from the work I did with my main ACF coauthor buddies Karin Ingold and Manuel Fischer. I have also drawn – in Chapter 12 – on European Research Council Horizon 2020 funded work (as part of the IMAJINE project) on policy learning (with Michael Keating and Emily St Denny). My update to the EU section

in chapter 8 is informed heavily by a paper co-authored with Andrew Glencross. In energy policy, I learned (again) that people use the term ‘system’ very loosely (with Fiona Munro) and that it is very difficult in practice to separate formal power and informal influence (with Nicola McEwen, Aileen McHargh, and Karen Turner). Bethany McKechnie helped me produce a very decent bibliography (while completing her MPP studies), and my hope is that she thrives in the academy. The same goes for many of my former MPP students, but I don’t want to jinx their chances. Harriet Simons made sure that the index was a proper index. I paid Bethany and Harriet £10 per hour, which is OK money if you don’t try to spend it outside the UK. Of course, the danger with a long list is that I’ve likely not described someone really important (and I was tempted to just list the name of every co-author ever), but it’s traditional to remember one or two people just as the book goes past the point of no return. I’d like to dedicate the book to my friend and mentor, Grant Jordan, who died in 2017. He wouldn’t have read this book, but it would have been nice to keep up a tradition: he would keep my signed books for a while before handing them back to me during a clear-out! More importantly, his ideas – generally as part of a dream team with Jeremy Richardson – have been central to the way in which I understand the policy process. You can find a longer tribute to Grant (by many people cited in this book) on my blog. Grant once told me a story of trying to buy a parcel of land from his neighbour. The price? Whatever it would cost his neighbour to fix his roof! He was focusing on what he needed from the process, not what the process could reasonably provide. People often do the same thing with politics: they focus on what they want from a policymaking system (or express a ‘functionalist’ imperative) rather than how it works. In contrast, this book focuses on how it works, and invites you to wonder what to do with that knowledge. I would also like to note the passing of Barry Hindess. I did not know him personally, but, if you have read his work, you will see that some of his insights underpin a lot of my discussion. Philosophy and Methodology in the Social Sciences is a classic, and I read it (and many more of his books) after browsing the library shelves (please contact me if you would like more tales of the pre-Internet age). In

Chapter 3, I also rely on his work to describe the work of Jürgen Habermas. Finally, I would like to thank Lloyd Langman at Red Globe Press for encouraging me to write the second edition and not complaining too much when it was (*coughs*) a few years behind schedule. However, Lloyd did not have to convince me of the value of these kinds of textbooks. I know I’m biased (as a contributor to, and editor of, Red Globe Press’s Textbooks in Policy Studies series), but I hope you will still indulge my plea to praise the textbook. Many colleagues describe their reluctance to write a textbook because it allegedly takes too much effort for too little reward; our status depends too highly on the journal article and monograph. I challenge this conclusion for four reasons. First, if you do it right, you can write as a package: the immense reading for the textbook informs the literature review that underpins any good piece of original research (and good textbooks incorporate new material). Second, almost all of my mostcited publications have been reviews of the literature (including ‘Standing on the shoulders of giants’, based on the first edition’s conclusion). Third, while a very small number of people (think they) know their entire field, most people benefit from these attempts at theoretical synthesis. This benefit is particularly high when you are trying to relay key insights to people in other disciplines, and practitioners, who do not have the time to read the same material. We rely on each other to make this stuff widely accessible, or accept that we are speaking to a very small group of people. Fourth, people actually read textbooks. Many students and practitioners get in touch with me to describe the usefulness of the first edition (and blog), and each moment of feedback makes the writing worthwhile. I’d have to rack my brain to remember anything as remotely positive in relation to any article or grant application I’ve sent for peer review. Of course, my story of public policy is potentially one of many. White male professors in the Global North – like me – seem the most likely to tell these stories of policymaking. Therefore, my second commitment is to find ways to encourage and enable more voices in this Textbooks in Policy Studies series. Paul Cairney

1 Introduction to Policy and Policymaking

Key themes of this chapter: • •

•

•

•

•

•

Why do we study public policy? To compare simple stories of how it should be made with complex stories of how it is actually made. Politics and power are central features of this story. Politics is about many actors, with different beliefs and preferences, exercising power to get what they want. We misunderstand the policy process if we ignore politics and focus only on vague ambitions such as ‘evidence-based policymaking’. Definitions matter. Policy can be defined as the sum total of government action, from signals of intent to final outcomes. It can refer to specific ‘tools’ or ‘instruments’. ‘Policymakers’ can be elected or unelected people or organizations. Simple models help us understand how policy is not made . Policy studies compare ‘ideal-types’ with the real world: there is no ‘comprehensive rationality’, in which governments process all information relevant to policy, or ‘policy cycle’, in which they make policy simply via a well-ordered series of ‘stages’. Policymakers have to use cognitive shortcuts. They deal with ‘bounded rationality’ in two ways: ‘rational’, to seek reliable sources of information and analysis, and ‘irrational’, using gut instinct, habits, emotions, and beliefs. The power of ‘the centre’ is limited. Political studies often focus on the concentration of power in central government. However, elected policymakers share power with other actors, by choice (to produce checks and balances in political systems) and necessity (because their controlling capacity is limited). Policy networks and subsystems are pervasive. Instead of one ‘centre’ in charge of all policy, we see many ‘communities’ of

•

•

•

actors spread across government. Complex policymaking environments limit policymaker control. Policy studies focus on the environments in which policymakers operate but do not control. An environment contains many policymakers and influencers operating across many levels and types of government, each with their own institutions, networks, ideas, socioeconomic conditions, and key events. Ideas matter. Policy theories describe ‘ideas’ as (1) general ways in which to understand the world and policy problems and (2) specific policy ‘solutions’. In some cases, governments transfer policy solutions from one system to another. How to analyse policy and policymaking. At the heart of the study of public policy are definitions and measurements (how do we gauge policy change?), power (who or what is responsible for policy change?), and unpredictability (policymaking is often stable then suddenly unstable).

INTRODUCTION: WHY SHOULD WE STUDY PUBLIC POLICY? Public policy is so important because it influences all aspects of our lives. It is difficult to think of any aspect of social life that has no connection to policy. So, it is important to get policy right, or at least explain what goes wrong and what anyone can be expected to do about it. In other words, we compare simple stories of how we think policy should be made with more complex stories of how it is actually made. For example, we can focus on why particular decisions are made. Why did so many governments decide to ‘bail out’ banks, rather than let them fold, after economic crisis? Why did many governments ‘privatize’ their industries and introduce private sector ideas to the public sector? Why have so many governments introduced major tobacco control policies while others have opposed further controls? Why did the UK government introduce the ‘poll tax’ or the Australian government reform gun laws (McConnell, 2010:

149–53)? Why do some groups seem to ‘win’ and others ‘lose’ when governments make key choices? We study concepts and theories of public policy because we recognize that there are many different answers to these questions. These answers come from different sources, and we can learn a lot by comparing many perspectives carefully (Allison, 1969, 1971). We can compare insights from different theories, by asking ourselves, for example, if they can explain different parts of a complex question or if we can combine them to produce more general explanation. Although we cannot simply mash theories together (Cairney, 2013a), it is possible to identify a small set of core concepts that all theories seem to use to explain policymaking. Put most simply, they try to capture the role of individual choice within a complex policymaking environment which contains many actors, rules, and networks spread across many levels and types of government. Things get complicated when we try to define each of these concepts, usually with reference to bounded rationality and the relationships between so many of the constituent parts of policy environments. For example, consider the many ways in which we can begin to study policymaking. We can focus on individual policymakers, examining how they analyse and understand policy problems. We can consider their beliefs and how receptive they are to particular ideas and approaches to the problem. We can focus on institutions, as the rules that influence policymaker choice. We can identify the powerful groups that influence how policymakers think and act. We can focus on the socioeconomic context and consider the pressures that governments face when making policy. Or, we can try to combine this analysis to present an overall picture of policymaking. Most contemporary accounts try to explain policy decisions by focusing on one factor or by combining an understanding of these factors into a theory. There will never be a single theory to unify all of these approaches to public policy, but we can at least generate a language to help us understand and compare many theories, and communicate our findings (Cairney and Weible, 2017). Most of this book is devoted to abstract concepts and theories, to help the reader understand how they provide a range of explanations of policymaking in the real world.

This conceptual understanding of policymaking informs discussions of how policy should be made. For example, there is currently a widespread interest in the idea of ‘evidence-based policymaking’ (EBPM) (Cairney, 2016a). For our purposes, EBPM is interesting because it sums up a classic debate in policy studies about the potential trade-off between a reliance on (1) expertise and evidence to pursue good policy outcomes and (2) the need to encourage the participation of as many people as possible to ensure good policy process. Of course we may want both things, but politics is about the choice between many preferences whenever they collide, and a process driven solely by expertise will produce very different outcomes to a process driven solely by participatory democracy. In practice, neither extreme is likely. Instead, the tensions between expertise and participation are reflected in the ways in which, for example, elected policymakers try to combine many relevant factors – research evidence, public opinion, their own beliefs, and their assessment of what policies are politically feasible – to produce what they consider to be good decisions under challenging circumstances. Policy theory is crucial to the critical analysis, and full understanding, of what we might call ‘good policymaking’, of which EBPM is only one part. Without the insights of policy theories, it is too tempting to throw up one’s hands and declare that politics and policy-making are dysfunctional, and that policy is made with too much reference to ideology and too little to evidence. Yet ‘there is a big difference between a policy process that does not work and a process we do not understand’ (Wellstead et al., 2018: 7). Without enough knowledge of the policy process, we will struggle to produce a thoughtful and practical understanding of good versus bad policymaking. A focus on power provides the most important way to think about the relationship between how policy is and should be made (Chapter 3). The most visible illustration of power is when elected policymakers are, in principle, given the authority to govern by the electorate. Our aim in this context is to examine the extent to which that power is used in good faith. For example, we may examine the

responsiveness of public policy to public opinion or the electoral and legal mechanisms that keep policymakers in check. Yet policy studies also analyse the least visible elements of power, often reproduced by unelected actors. For example, powerful groups often maintain their status and get what they want by minimizing attention to certain issues. Policy change requires attention from policymakers and other interested participants, but such attention is a rare commodity: a policymaker can only consider so many issues and most members of the public will only pay fleeting attention to political issues. So, actors exercise power to make sure that important issues do not arrive at the top of the policy agenda, because greater attention could undermine the status quo. This outcome may be achieved in several ways. First, by portraying issues as not worthy of attention. They can be portrayed as issues that have been solved, with only the technical issues of implementation to address. Or, they can be described as private issues in which the government should not interfere. Second, issues can be ‘crowded out’ of the policy agenda by other issues that command more attention. Third, the rules and procedures of government can be manipulated to make sure that proponents of certain issues find it difficult to command policymaker attention. This potential for a small number of unelected actors to be so powerful can be assessed empirically, in theory-driven accounts of policymaking, and normatively, in relation to a range of principles of good policymaking. Only then can we describe a realistic and pragmatic response to such findings. For example, instead of expecting evidence to speak for itself as part of EBPM, advocates of research may have to get their hands dirty by exercising power and seeking to persuade, by using the techniques of their opponents (Cairney, 2018a). In doing so, they will encounter many other actors pursuing different ways – combining knowledge, values, deliberation, and governance principles – to ensure good policymaking (Cairney, 2016a, 2018b; Jasanoff 1986: 5; Weale, 2001).

THE GENERAL APPROACH OF THIS BOOK

Public policy is difficult to study but worth the effort. The policy process is complex, messy, and often appears to be unpredictable. The idea of a single process is often a useful simplification. However, when we scratch beneath the surface we find that there are multiple policy processes: the behaviour of policymakers, the problems they face, the actors they meet, and the results of their decisions often vary remarkably. They often vary by region, political system, over time, and from policy issue to issue. Indeed, we might start to wonder how we can make convincing generalizations about all public policy. That is why much of the literature employs the case study method to make sense of very specific events in different political systems. There are well-established ways to make sense of the process as a whole. The first step is to ensure that our language is clear enough to make sure that we are discussing the same thing. For example, we define public policy and compare definitions of policy to help identify our key focus (Chapter 2). We identify types of policy choice, such as the ‘tools’ or ‘instruments’ of policy, to help us measure policy change. We also juggle definitions of ‘policymaker’, to recognize that they can be elected or unelected, identifiable as individuals or representatives of organizations, and part of wider networks in which it is often difficult to separate the effect of choices made by policymakers and influencers. The second step is to decide how to make the study of policymaking manageable (Chapter 2). For example, we can use ideal-types or artificial models to present a starting point to analysis. One classic approach is to describe the policy process as a cycle and break down its key features into a series of stages such as agenda setting, formulation, legitimation, implementation, and evaluation. Another is to describe ‘comprehensive rationality’, in which a policymaker has the perfect ability to produce, research, and introduce their policy preferences. We can then examine how useful are these concepts as a way to organize policy studies, and consider how best to supplement or replace them with policy theories. The main aim of this book is to identify those theories, explain how they work, and assess their current ability to explain key parts of the policy process.

I describe the overall plan of each chapter in this paragraph, and then expand on key themes in the remainder of this introduction. First, each chapter sets out a key theory or concern of public policy. In some cases, such as punctuated equilibrium and the advocacy coalition framework (ACF), this task is straightforward because the theory is linked to a small number of authors with a coherent research plan. In others, the chapter describes a disparate literature with many authors and approaches (new institutionalism, multi-level governance, policy transfer). Or, the chapter describes an important issue (defining policy, structural factors, bounded rationality, the role of ideas and power) and outlines key concepts or theories within that context. Second, it identifies the questions that each theory seeks to answer. It is important to assess each theory according to its stated aims. Too much of the literature complains about the things that a theory does not explain, often to introduce the self-serving agenda of the author, rather than simply using theories when they are useful. Third, it considers the value of each theory in different circumstances. For example, many theories have developed from a study of the US political system characterized by a focus on democratic elections and a separation of powers. Therefore, some thought is required when we apply their lessons to many other political systems (see also Chapter 5 on institutions). Fourth, it explores how the key themes and issues raised by each theory relate to concerns raised in other chapters. We explore the language of policy studies and the extent to which theories address the same issues, and often come to similar conclusions, in different ways. Finally, the concluding chapter tells an overall story of policy, policy change, and the policy process. It returns to the theme of EBPM as a case study, to show how we can combine the insights of many public policy theories to inform current debates on good policymaking. It also tells some cautionary tales about the limits to our knowledge of policy processes and global public policy.

SIMPLE MODELS HELP US UNDERSTAND HOW POLICY IS NOT MADE

The ideal-type ‘comprehensive rationality’ suggests that elected policymakers translate their values into policy in a straightforward manner (Chapter 4). They have a coherent and rank-ordered set of policy preferences which organizations carry out in a ‘logical, reasoned and neutral way’ (John, 1998: 33). There are clear-cut and ordered stages to the process (to identify aims, describe options to achieve those aims, and select the best option) and analysis of the policymaking context is comprehensive. This capability and process allows policymakers to maximize the benefits of policy to society in much the same way that an individual maximizes their own utility (as described in Chapter 7). This simplified account could be described in aspirational terms, as part of prescriptive or normative policy analysis. Indeed, modern references to EBPM seem to invoke these kinds of aims (Cairney, 2016a: 19). However, most of the policy theories described in this book treat it primarily as a useful comparison to how policymaking really works, as part of descriptive policy analysis. The key phrase is ‘bounded rationality’ which Simon (1957, 1976) coined to describe people or organizations adapting to their realworld limitations, using decision-making shortcuts rather than comprehensive analysis, and seeking satisfactory rather than ‘optimal’ solutions to policy problems. Lindblom’s (1959) theory of incrementalism also suggests that policymakers generally do not look everywhere for policy solutions. Rather, their values limit their search and their decisions often reflect decisions that were taken in the past and seem difficult to change (because, for example, they were the result of extensive negotiation) (Chapter 4).

Policymaker Psychology Although policy theories still describe ‘bounded rationality’, they are informed increasingly by modern studies of human and organizational psychology (Cairney and Kwiatkowski, 2017). There is still a focus on ‘rational’ ways to deal with bounded rationality, such as by setting goals and seeking high-quality information by relying on expertise and scientific evidence. However, there is also a growing focus on the ‘irrational’ ways in which actors, including policymakers,

limit their attention to information, to help them make decisions quickly. They may draw on their deeply held beliefs about how the world works and should work – expressed as ideology or moral values – as well as their emotions, gut instinct, habits, and lived experience. In other words, so-called ‘irrational’ thinking can often relate to well-thought-out, efficient, and useful shortcuts. Key policy theories use this simple insight, about human psychology, to show how actors exploit cognitive shortcuts to pursue their policy aims. The Narrative Policy Framework (NPF) identifies the role of narratives (Jones et al., 2014). Narratives are stylized accounts of the origins, aims, and likely impacts of policies. They are used strategically to reinforce or oppose policy measures. Narratives have a setting, characters, plot, and moral. They can be compared to marketing, as persuasion based more on appealing to an audience’s beliefs than on the ‘facts’. Actors, such as policymakers, will pay attention to certain narratives because they are seeking shortcuts to gather sufficient information, and are prone to accept simple stories that confirm their biases, exploit their emotions, and/or come from a source they trust. Social Construction and Policy Design (SCPD) shows that politicians describe moral judgements about who should be rewarded or punished by government (Schneider and Ingram, 1993). They link policy agendas to stereotypes of ‘target populations’, by (1) exploiting the ways in which many people already think about groups or (2) making emotional and superficial judgements, backed up with selective use of facts. These judgements are reproduced in policy and practice. Such ‘policy designs’ can endure for years or decades, since the distribution of rewards and sanctions is cumulative and difficult to overcome. The impact on citizens can be profound if they participate in politics according to how they are characterized by government and, therefore, how rewarding they think their participation will be.

Policymaking Complexity The idea of a policy cycle can also be used as a starting point for prescriptive and descriptive analysis (Althaus et al., 2013; Wu et al.,

2017). We might start by saying that policymakers should divide the process into a well-ordered series of stages to ensure policy success: identify policymaker aims, identify policies to achieve those aims, select the best policy measure, ensure that the selection is legitimized by the public or its legislature, identify the necessary resources, implement, and then evaluate the policy to decide if it should continue (Chapter 2). There are also modern texts dedicated to the empirical study of key stages such as agenda setting, formulation, and implementation, or built on the enduring point that policymaking seems cyclical and represents ‘its own cause’ (Wildavsky, 1980: 62; Hogwood and Gunn, 1984: 245). Yet few texts suggest that the policy process resembles an orderly cycle. Indeed, some refer to the ‘stages heuristic’ to highlight its descriptive weaknesses and to recommend new ways to study the policy process (Sabatier, 2007a: 7; John, 2012). In this context, the aim of the book is to focus on theories of public policy rather than devote a chapter to each stage of the policy cycle (compare with Knill and Tosun, 2019 who treat each ‘stage’ as a key function of government worthy of analysis).

THE POWER OF ‘THE CENTRE’ IS LIMITED One reason not to rely too much on the policy cycle approach is that it conjures up an image of a single centralized government: an elite group of people at the ‘centre’ decides what it wants and uses a series of stages to get it. Indeed, if we treat comprehensive rationality and policy cycles as ideals, they suggest that power should be held centrally, perhaps because central policymakers are elected and should have control over policy implementation (Hogwood and Gunn, 1984: 198). Yet implementing officials may also be elected and have a legitimate and practical role in the making of policy. Further, a concentration of power in one place without checks and balances may be the best formula for its abuse. This assumption of centralization would also be a big mistake empirically, because formal sources of power are concentrated but the actual power to make and influence policy is dispersed widely across political systems (Chapter 8).

Studies of federalism suggest that this diffusion of power results partly from choice, such as when written constitutions guarantee a separation and balance of powers between many levels of government (from local to supranational) and types of organization (including legislature, executive, and judiciary). In some cases, such as the USA and Switzerland, power is devolved to many elected organizations, and federalist studies examine how they negotiate with each other (intergovernmental relations). Policy theories also focus on the necessity of power diffusion caused by bounded rationality and the sheer scale of the policymaking environment. For example, studies of ‘multi-centric policymaking’ (Cairney et al., 2019a) – such as ‘multi-level governance’ (MLG), complexity theory, and polycentric governance – identify a further dispersal of power to bodies such as quasi-nongovernmental actors (‘quangos’) and the blurry boundaries between formal and informal sources of authority. MLG suggests that the process is messy: many actors are involved at many levels of government and their relationships vary across time and policy issue. Policy outcomes are difficult to predict because the formal responsibility for policy issues changes over time, different actors become involved, and there is scope to influence and change policy at different points.

POLICY NETWORKS AND SUBSYSTEMS ARE PERVASIVE The policy process is messy, but the logic of power diffusion is often simple: the scale of government responsibilities is so huge, and the environment so complex, that many decisions are effectively out of the control of individual policymakers. Bounded rationality dictates that they need to find ways to ignore almost all information and issues so that they can pay enough attention to some. If they focus on one issue they have to ignore most others. This type of decision becomes so routine that it is ‘institutionalized’ within government. Public policy is broken down into a series of sectors and subsectors and responsibility delegated largely to unelected

policymakers. Policymaking tends to be specialized, with the responsibilities of government processed in venues that are often called policy networks, communities, or subsystems (Chapter 9). The ‘policy communities’ literature suggests that policymakers devolve responsibility to civil servants who, in turn, rely on interest groups and other ‘pressure participants’ for information and advice (Richardson and Jordan, 1979; Jordan et al., 2004). These arrangements exist because there is a logic to devolving decisions and consulting with certain affected interests. Policymakers rely on their officials for information and advice. For specialist issues, officials rely on specialist organizations. Those organizations trade information (and other resources, such as the ability to implement government policy) for access to, and influence within, government. Consequently, much public policy is conducted primarily via policy networks which process ‘technical’ issues at a level of government not particularly visible to the public, often with minimal policymaker or senior civil service involvement. Participants are specialists and relationships develop between those who deliver policy and those who seek to influence it. The wider literature contains a mix of theories which try to capture this sense that there is no single ‘centre’ at the heart of policymaking, and that subsystems are pervasive, but without necessarily concluding that policy issues are insulated from wider attention. MLG (Chapter 8) often links power diffusion to the unintended consequences of decisions made by governments in the past. For example, many governments reformed their public sectors from the late 1970s in the spirit of ‘new public management’, which describes the application of private sector business methods to the public sector. Many sold off previously nationalized industries and contracted the delivery of services to non-governmental organizations. The reforms have reduced the ability of central governments to deliver public services directly. They are now more likely to ‘steer’ rather than ‘row’, negotiating and making shared decisions with actors outside the public sector. In countries such as the UK, the era of Europeanization, combined with the devolution of power to sub-national authorities, has further complicated the

national central government role. Responsibility is now shared across multiple levels and types of government. Consequently, the government is unable to centralize and insulate decision-making. Rather, the policy process contains a much larger number of actors that central government must negotiate with to pursue its policy aims. Further, the group-government world has become much more complicated, with many groups lobbying and seeking to influence policy in multiple venues. Punctuated equilibrium theory (Chapter 9) explains why political systems can be characterized as both stable and dynamic. Most policies stay the same for long periods while some change very quickly and dramatically. Punctuated equilibrium explains this dynamism with reference to bounded rationality and agenda setting. Since policymakers cannot consider all issues at all times, they ignore most and promote relatively few to the top of their agenda. This lack of attention to most issues helps explain why groups and officials can maintain policy communities and why most policies do not change. ‘Policy monopolies’ can be fostered by ‘framing’ an issue to limit the number of participants who can claim a legitimate role in the process. Actors argue that a policy problem has been solved, with only the technical and unimportant issues of implementation to address. If successful, the ‘technical’ description reduces public interest and the ‘specialist’ description excludes those groups considered to have no expertise. Yet sometimes policymakers do focus on these issues. In some cases, their levels of attention are disproportionate and their response is ‘hypersensitive’. Change comes from a successful challenge to the way that an issue is framed, by finding other influential audiences with an interest in new ways of thinking. In many cases, this shift can be explained by an increasingly crowded and multi-level policymaking process. When groups are excluded at one level, they ‘venue shop’, or seek influential audiences in other venues such as legislative committees, the courts, or other levels of government. If they catch the attention of another venue, newly involved policymakers increase their demand for new information and new ways to think about and solve old policy problems. In a process characterized by interdependence between groups and

government, and overlapping jurisdictional boundaries (in which many institutions can be influential in the same policy areas), these innovations can be infectious. The actions of one often catch the attention of others, producing a ‘bandwagon effect’ of attention and policy change. For the ACF (Chapter 10), the policy process is driven by actors attempting to translate their beliefs into public policy. Common beliefs bring people together within advocacy coalitions. In turn, different coalitions with different beliefs compete with each other within subsystems. The ACF describes a complex and crowded political system and focuses on policy change over ‘a decade or more’ to reflect the importance of policy actors spread across many levels and types of government. The ACF also represents an attempt to show how actors mediate socioeconomic factors within subsystems. The idea is that, within a subsystem, these advocacy coalitions are not only jostling for position but also learning from past policy and revising their strategic positions based on new evidence and the need to react to external events. While many of these external factors – such as global recession, environmental crises, and demographic changes – may be universally recognized as important, coalitions influence how policymakers understand, interpret, and respond to them. They adapt to events by drawing on their beliefs and promote their understanding of events within the subsystem.

COMPLEX POLICYMAKING ENVIRONMENTS LIMIT POLICYMAKER CONTROL When we seek to understand policy change, do we assign responsibility to the individuals that make policy decisions or the environment in which they operate (which includes the policy conditions and socioeconomic pressures that they face)? Of course, the answer is both, but that answer does not take us far. In analytical terms, we can separate and describe the separate effects of structure and agency (Chapter 6; Hay, 2002: 89). We may focus primarily on agency and attribute the exercise of power only to

individuals because socioeconomic and institutional structures do not act and therefore cannot exercise power. This appears to make things simpler, but it just raises another question: how else can we describe convincingly the importance of institutions and structures? In practice, structure and agency are difficult to separate: policymakers act in accordance with rules, and their action depends partly on the nature of the problem they face. Yet these concepts still have an analytical value. They help us simplify and make sense of complex processes, and therefore help us draw generalizations and develop theories to explain other events and processes. Further, although we may agree that policy change results from a wide range of factors, most policy theories try to relate policymaker choice to an environment which can be summed up by a small number of statements: 1. There are many policymakers spread across many policymaking venues. 2. Each venue contains its own ‘institutions’, or formal and informal rules. 3. Each venue can produce its own networks of policymakers and influencers. 4. Participants in each venue make reference to a dominant set of ‘ideas’ or beliefs about the nature of policy problems and the acceptable range of solutions. 5. Social and economic conditions, and events, help set the policy agenda and limit policymakers’ ability to address and solve policy problems. Different chapters describe these dynamics in different, but generally complementary, ways. The rationality approach, explored in Chapters 4 and 7, focuses on the characteristics of actors such as policymakers. Some studies outline a set of assumptions about how individuals think and behave, to gauge how much this approach explains. Others explore what happens if these assumptions prove unrealistic. Broadly speaking, rational choice theory employs ‘methodological individualism’ or a commitment to explain sociopolitical outcomes as the aggregation of the decisions of

individuals. The basic aim is to establish how many, or what proportion of, political outcomes one can explain with reference to the choices of individuals under particular conditions. Perhaps the biggest contrast is with complexity theory, which states that a policymaking system is greater than the sum of its parts, and that we cannot simply reduce outcomes to the choices of individuals (Chapter 6). Rationality approaches also compare with accounts that consider the nature of the policymaking environment and how it impacts on the policy process. ‘New institutionalism’ (Chapter 5) treats institutions as sets of rules, norms, established practices, and relationships that produce regular patterns of policymaking behaviour. Rules can be formal, such as when set out in the constitutions of political systems. Or, they can be informal, such as the norms, understandings and expectations that people share when they interact regularly with other people. In studies of ‘institutional rational choice’, we combine this focus on choice and institutions. For example, the Institutional Analysis and Development Framework (IAD) goes beyond an abstract discussion of choice to consider the most effective ways in which people can produce rules to cooperate with each other (Chapter 7). The premise of more structural accounts is that powerful external forces determine, or influence, the way that individuals or governments make policy decisions. As Chapter 6 suggests, there is a long list of arguments from which to choose, including • • •

•

the socioeconomic background of countries or regions may affect the size and scope of their welfare policies; there may be a strong imperative for governments to support the capitalist system and therefore the interests of the classes that benefit most from that system; ‘globalization’ describes an imperative for governments to compete with each other to protect their economy and secure foreign direct investment, by reducing regulations, corporation taxes, and public spending; and policymakers are part of a large complex system over which they have limited control.

Chapter 6 also explores the idea that previous policies create new policy problems or constrain choice. For example, governments inherit massive commitments that constrain their ability to change policy beyond the margins, and it is relatively difficult to introduce new policies and terminate others. A combination of individualist and structural approaches suggests that we need to recognize the importance of structures and rules without saying that they determine behaviour. Contemporary theories of public policy address this issue by examining how policymakers and pressure participants adapt to their policy environment by, for example, interpreting socioeconomic shifts in a specific way to set the policy agenda, and ‘learning’ and adapting their beliefs to reflect those shifts.

IDEAS MATTER Almost all policy theories refer to the role of ‘ideas’ – or shared beliefs – to sum up this relationship between the bounded rationality of policy actors and the constraints posed by complex policymaking environments. They define ideas and describe this dynamic in different ways, focusing on the potential solution to a policy problem (‘I have an idea’), the narrative designed to persuade people to select it, or a fundamental set of beliefs or understandings which seem to provide structure to policy debate (Cairney and Weible, 2015). For example, ideas may be the shared beliefs that give people a common aim (Chapter 10). They can represent the accumulation of knowledge within a political system. Ideas may be embedded in the institutions guiding action (Chapter 5), the paradigms taken for granted and acted on with little further thought (Hall, 1993), or the ‘monopolies’ on the way that we understand policy issues (Chapter 9). Ideas represent standards of behaviour that are considered to be normal, or ideologies used as the basis for policy action. In each case, people exercise power in a context in which only certain beliefs or norms of behaviour are acceptable, or only certain forms of knowledge count (see Chapter 3 on power). Chapter 11 shows how different discussions try to describe this dynamic. Ideas can be compared to ‘viruses’ to identify their ability to

spread, subject to the crucial role of the ‘host’, or ability of policymaker audiences, with ingrained beliefs, to resist ‘infection’. Norms of behaviour have a strong influence but only when people enforce them. Paradigms represent established beliefs about the nature of the policy problem and how to solve it; but crises also prompt policymakers to revisit, and often reject, those beliefs. Chapter 11 then describes multiple streams analysis (MSA) which suggests that the adoption of a new policy solution requires the coming together of three factors at the same time: a problem is high on the agenda and ‘framed’ in a particular way, a technically and politically feasible solution already exists, and policymakers have the motive and opportunity to adopt that solution and translate it into policy. Chapter 12 focuses on policy learning and the potential transfer of ideas. The literature on learning describes it as a political strategy. People learn by listening to experts, engaging in wider dialogue, using trial and error to get what they want, and identifying the limits to ‘top-down’ policymaking (Dunlop and Radaelli, 2018). The literature on policy transfer refers to the evidence for, and causes of, similarities in policy across political systems. It can relate to the transfer of ideologies and wholesale programmes, broad ideas, minor administrative changes, and even negative lessons (when one country learns not to follow another country). Our discussion of ideas suggests that the scale and likelihood of transfer depends partly on the beliefs already held within the importing country. To take an extreme example, a capitalist and a socialist country may not find much to learn from each other about how to balance the public and private sectors or pursue economic growth. In less extreme examples, we may focus on more technical reasons for the likelihood of transfer: if the borrowing country has an incentive to learn from another (e.g. if they share a common problem and have a similar political system); if the policy is simple and easy to adopt; if the values of borrower and lender coincide; and if their administrative arrangements are similar. Some countries may also become pressured to follow the lead of another. Chapter 12 explores the role of coercion, from the direct pressure of another country or supranational institution to the feeling among an importing country

that it should keep up with international trends. Overall, the study of transfer gives us the ability to explain why policy changes and the extent to which change relates to the widespread sharing of ideas.

HOW TO ANALYSE POLICY AND POLICYMAKING We focus so much on definitions and measurement because policy theories exist primarily to explain continuity and change in public policy. We need to know what policy is to be able to measure it. We need to measure it – and identify minor or major change – to know what we seek to explain. We need to explain the policymaking process to understand the role of key elements, such as evidence or participation, and to make an informed analysis of how good or bad policymaking tends to be. In particular, a central aim of policy theories is to explain why policymaking seems to involve stable relationships and policy continuity at one point but instability and policy change at another. There is a discussion of policy continuity and change in almost every chapter. Policy cycle models identify decision points at which policy will change, but also that decisions made in the past constrain present options. Theories of power highlight the ways in which powerful actors create or reinforce barriers to change, but also that they can be overcome. New institutionalism may focus on the constraints on policy change, often caused by events and decisions in the past, but also the extent to which rules and norms are challenged and reformed. Bounded rationality can produce both incremental and radical change. Structural factors are sources of constraint, but developments such as demographic change can put immense pressure on policymakers to change policy. Rational choice theory aims to identify points of equilibrium, in which there is no incentive for any individual to make a different choice, but much analysis seeks to explain its frequent absence. MLG identifies new sources of ‘veto points’ within political systems, but also new sources of policy innovation and diffusion. Punctuated equilibrium uses theories of bounded rationality and agenda setting to explain long periods of stasis but bursts of innovation. The ACF identifies the core beliefs of advocacy coalitions, and fundamental social structures, as

sources of continuity, but policy learning and ‘shocks’ to the subsystem as drivers for change. Ideas can represent paradigms that represent stability and continuity, but also new policy proposals that can be used to encourage change. We may also consider if policymakers seek lessons from elsewhere to pursue innovation or legitimize existing decisions. In each case, the theories only tell us that policymaking can be stable or unstable; that policy can change or stay the same. We must also establish what happens, using theories to guide empirical studies. In other words, it is up to us to identify what policy is and the extent to which it has changed. It would be a mistake to skip the boring definition and measurement part (Chapter 2) to get to the exciting discussions of policy dynamics, because the latter make no sense without the former.

THE STRUCTURE OF THE BOOK The book is designed to introduce theories and concepts in a particular order. Chapters 2–4 introduce key concepts and background to the study of public policy: •

•

•

Chapter 2. We consider the meaning of public policy and describe the main ways to identify policy tools and instruments, to help us measure policy change. We then examine the role of models, theories, and concepts of policymaking and examine the classic reference point: the policy cycle. Chapter 3. We explore power as a key concept that underpins discussion in all subsequent chapters. We derive insights from the ‘community power debate’ to consider how to define power and consider the methods to identify power within policymaking systems. Although ideas enjoys a separate discussion, Chapter 3 demonstrates the inextricable relationship between the exercise of power and shared beliefs. Chapter 4. This focus on power and ideas allows us to examine fully the second classic ideal-type: comprehensive rationality. We focus on policymakers at the heart of government and consider what conditions need to be met to ensure that they have the

ability to research and articulate a series of consistent policy aims and then make sure that they are carried out. Our identification of ‘bounded rationality’ allows us to consider what happens when these conditions are not met. Approaches such as SCPD and the NPF show how policy actors exploit the ways in which policy actors address bounded rationality. Chapters 5–7 explore the role of structure and agency in complex policymaking environments: •

•

•

Chapter 5. We consider how to define institutions and identify the study of ‘new institutionalism’. We explore the role of institutions as the sets of laws and rules that govern the operation of policymaking systems. These rules and laws vary by political system, but key concepts cut across them. For example, feminist institutionalism analyses the rules used to address gender and the ways in which they produce imbalances of power in all political systems. Chapter 6. We identify the role of factors that may be beyond the control of policymakers. Too many people assume that policy analysis should begin with a discussion of ‘rational’ policymakers rather than the environment in which they operate. We explore the key ‘structural’ causes of policy outcomes, the extent to which policymakers are influenced by economics, the idea that policymakers inherit policy commitments, and the idea that they form one part of a large complex policymaking system. Chapter 7. We identify the ways in which actors might, or might not, engage in collective action. Key examples of ‘game theory’ – including the prisoner’s dilemma and ‘tragedy of the commons’ – identify ways in which individuals appear to undermine long-term collective action by acting according to their own short-term rewards. However, the IAD helps identify the conditions under which people will decide to act collectively to produce and enforce rules to avert such tragedies.

Chapters 8–10 explore the relationships between policymakers and policy participants within policy networks or subsystems:

•

•

•

Chapter 8. We identify multi-level governance, or the dispersal of power from national central government to other levels of government and non-governmental actors. We consider how to define governance and MLG, identify its origins in studies of the UK and EU, consider its applicability to other political systems, and identify the links between MLG and other theories studying subsystems. Chapter 9. We consider punctuated equilibrium theory, whose aim is to explain long periods of policy stability and continuity and short but intense periods of instability and change. Punctuated equilibrium theory is based on the study of bounded rationality, complex systems, subsystems, and agenda setting. We explore the ability of interest groups to ‘venue shop’, or seek sympathetic audiences elsewhere, when they are dissatisfied with the way that policymakers understand and seek to solve policy problems. Chapter 10. We consider the ACF. It focuses on the role of coalitions driven by their common beliefs. Coalitions compete with each other within subsystems. Some coalitions can dominate the way that policy problems are understood and solved for long periods, but are also subject to challenge when ‘shocks’ force them to reconsider the adequacy of their beliefs or provide new opportunities for their competitors.

Chapters 11 and 12 consider the role and transfer of ideas: •

•

Chapter 11. Our basic premise is that policymaking is not just about power. Instead, we must identify how power and ideas combine to explain policy processes and outcomes. We consider how to define and identify ideas, explore the ways in which theories of public policy conceptualize the relationship between power and ideas, and outline the importance of MSA to capture the explanatory power of ideas and ways in which ideas are promoted and accepted within government. Chapter 12. We consider policy learning and the transfer of ideas from one policymaking system to another. In some cases, transfer follows research, when the knowledge of one system is used to inform policy in another. However, the transfer of ideas or

policy programmes is also about power and the ability of some countries to oblige or encourage others to follow their lead. We consider what policy transfer is, who does it, and why. However, the chapters do not need to be read in strict chronological order. For example, Chapters 3 and 11 can be usefully combined to consider the role of power and ideas. Chapters 3 and 9 consider power and agenda setting. Chapters 4 and 6 have a common focus on structure and agency. Chapters 6 and 9 discuss complex systems. Chapters 6–8 describe multi-centric policymaking. The role of Chapter 13 is to consider these multiple links in more detail and to consider how best to combine theories and concepts to address the big questions in public policy.

CONCLUSION The most basic question we ask ourselves when studying public policy is: what do we want to know? The answer is that we want to know what policy is; what policy measures exist; what measures have been taken; and how to best make sense of what has happened. Indeed, this book is mostly about policymaking rather than specific examples of policy in specific countries. In particular, we want to know what the most useful theories of public policy are, what each of them tells us, and what the cumulative effect of their knowledge is. We study theories of public policy because we want to ask why particular decisions are made, but recognize that there are many different answers to that question. The use of multiple theories allows us to examine the policy process and build up a detailed narrative of events and decisions, using multiple perspectives. We focus on policymakers and seek to identify how they understand policy problems, how high the problem is on their agenda, and to which arguments and solutions they are most receptive. We identify the political, social, and economic pressures that they face when making decisions. We identify the main governing institutions and examine the rules they follow when they process issues. We identify policy networks and the relationships between interest groups and

government. We also consider the extent to which policy is within the control of elected policymakers, and how much is devolved to other actors such as civil servants and groups, other types of government, and non-governmental organizations. Each chapter sets out a key theory or concern of public policy, identifies its value, and explores the questions that each theory seeks to answer. It then considers, where appropriate, how each theory or concept has been applied empirically and how much it tells us about different political systems, policy areas, and times. The role of Chapter 13 is to consider how to best combine the insights of these public policy theories. It helps us tell a story based on the insights of policy studies, use it to answer real-world questions – such as, can policymaking be ‘evidence-based?’ – and reflect on the tendency for that story to come from the Global North. Public policy – the sum total of government action, from signals of intent to the final outcomes (but see Chapter 2). Theory – a set of analytical principles designed to structure our observation and explanation of the world. Actors – the organizations or people who take action. Bounded rationality – the limits to human cognition described in Chapter 4. Empirical or descriptive – relating to how things are. Normative or prescriptive – relating to how things should be. Ideal-type – an abstract idea used to highlight hypothetical features of an organization or action. Sectors – broad policy areas, such as economic, foreign, agriculture, health, and education. Sub-sectors – specific policy issues or niches within sectors, such as seed potato regulations or growth hormones in dairy production. Policy conditions – the nature or structure of the policy environment and hence the specific problems that policy-makers face. Contextual factors include a political system’s size, demographic structure, economy, and mass behaviour (Chapter 6).

Structure – a set of parts put together in a particular way to form a whole. The implications are (1) that a structure is relatively fixed and difficult but not impossible to break down; and (2) that it influences the decisions that actors make. Agency – refers to the ability of an actor to deliberate and act to realize its goals. The implication is that this is intentional action based on an actor’s thought process and ability to choose (rather than determined by structure or environment).

2 What Is Policy and Policymaking?

Key themes of this chapter: • •

• • •

•

We need to define public policy to measure and explain policy change. However, there is no definition able to describe fully the richness of policy. One solution is to identify policy ‘tools’ and ‘instruments’ to measure policy change. This approach highlights the politics of policymaking, in which some policy changes are easier to sell or deliver than others. Another solution is to construct and compare narratives of policy change by interpreting the relative importance of key measures. We then need to define and identify the key elements of policymaking. A classic starting point is to identify stages – including policy formulation and implementation – in a ‘policy cycle’. The cycle is best understood as an ideal-type to compare with real-world policymaking. New concepts and theories help explain how policy is made in a more complex policymaking environment.

INTRODUCTION: THE NEED TO DEFINE POLICY AND POLICYMAKING ‘Public policy’ is important because the scope of the state extends to almost all aspects of our lives. However, it is one of many terms in political science – like democracy, equality, and power – that are well known but difficult to define. Our definition of policy affects how we analyse and understand real policy issues. Different definitions,

drawing on different aspects of the policy process, give us multiple perspectives and set the agenda for the study of policymaking. Therefore, the first aim of this chapter is to explore such definitions. It does not produce a comprehensive description. Rather, it identifies the key points that arise from the public policy literature when we analyse any definition in more detail. For example, policy can refer to a statement of intent, an aim, a decision, or an outcome. It may refer to issues that policymakers do not to address. Further, it is made and influenced by many actors who may or may not have formal authority. The second aim is to think about how to identify and measure policy change. This task is easier said than done, but a useful starting point is to draw on established ways to categorize policy measures, with reference to the broad policy ‘tools’ (such as regulations or taxes) or specific ‘instruments’ (such as a piece of legislation) used by governments to produce policy change. Such typologies show us that different kinds of policy measures are associated with different styles of politics because they are more or less easy to sell and deliver (John, 1998: 7–8; 2011; Lowi, 1964). A key tenet of the public policy literature is that we need to identify and account for this variation. Policy outcomes and processes vary according to the political system and territory, time, policy issue, and solutions that we study. This variation makes it difficult to characterize policy change. Although we can identify the types of policy instruments available to policymakers, the list of measures is long and it is difficult to know how significant each decision will be. Indeed, the more measures or decisions there are, the greater potential the researcher has to construct different stories of policy change. There is always more than one way to interpret the significance and direction of policy from the evidence available. This point may prompt us to consider different narratives of policymaking depending on which policy decisions we describe as the most important (Cairney, 2013a). This problem of defining and measuring policy is magnified when we account for the complex policymaking environment in which policy is made. Frameworks, theories, concepts, and models help us make sense of policy processes. For example, the policy cycle

model provides a classic starting point, in which policy is made in stages including formulation, implementation, and evaluation. The model helps us identify a difference between a policymaker’s intention and the policy outcome, and has prompted a considerable literature devoted to the ‘implementation gap’. The cycle metaphor also suggests that the evaluation stage of policy one represents the first stage of policy two, as lessons learned in the past inform choices in the future. Policymaking is a continuous process rather than a single event; different actors are influential at different stages and previous decisions set the agenda for future decisions. Yet, although a useful starting point, the cycle is best understood as an ideal-type to compare with real-world policymaking. Concepts and theories help explain how policy is actually made in a more complex policymaking environment. This chapter considers the limits to a focus on stages to help explain the benefit of policy theories.

WHAT IS PUBLIC POLICY? Hogwood and Gunn (1984: 13–19) identify different ways to understand ‘policy’: as a label for a field of activity (e.g. health policy); an expression of intent (e.g. ‘we will improve healthcare’); specific proposals (e.g. a manifesto or white paper); decisions of government and the formal authorization of decisions (e.g. legislation); a programme, or package of legislation, staffing, and funding; intermediate and ultimate outputs (e.g. more doctors, better medical care); outcomes, or what is actually achieved (better societal health); and a process and series of decisions, not an event and single decision. Box 2.1 provides further descriptions of public policy, but the literature generally questions our ability to define policy (Colebatch, 1998: 72–73; Smith and Larimer, 2009: 3–4, 25, 239). Box 2.1 Definitions and descriptions of public policy Whatever governments choose to do or not to do. (Dye in Birkland, 2005: 18)

The actions of government and the intentions that determine those actions. (Cochran et al. in Birkland, 2005: 18) Decisions by governments to retain the status quo are just as much policy as are decisions to alter it. (Howlett and Ramesh, 2003: 5) Diverse activities by different bodies are drawn together into stable and predictable patterns of action which (as often as not) come to be labelled ‘policy’. (Colebatch, 1998: x) Policy designs are observable phenomena found in statutes, administrative guidelines, court decrees, programs, and even the practices and procedures of street level bureaucrats. (Schneider and Ingram, 1997: 2) There is an underlying assumption that policy is a good thing, that it fixes things up. Policy makers are the ones who do the fixing . (Bacchi, 2009: ix on the perils of making too many assumptions in policy analysis) The next best thing is to explore the issues that arise when we start with a simple definition. Defining public policy as the ‘sum total of government action’ (Chapter 1) raises four key questions: 1. Does government action include what policymakers say they will do as well as what they actually do? Policymaking is often symbolic and policymakers often make statements to look like they are doing something (Althaus et al., 2007: 5; Dearlove in Colebatch, 1998: 9). Indeed, the election of US President Donald Trump in 2016 raised the possibility that a powerful policymaker could be announcing policy via Twitter or making heat-of-the-moment remarks with an unclear meaning and no expectation of follow-through. Further, political parties produce manifestos and elected officials produce speeches describing their plans, but with no guarantees that they will happen. If so, should we wait until those plans are carried out

until we call them ‘policy’? It is difficult to know when to say that an aim has been pursued enough to constitute policy. Few would say that legislation is not policy until the budget is agreed to carry it out, but what about a consultation document officially setting out the government’s plans? The identification of substantive policy measures is often more art than science. 2. Does policy include the effects of a decision as well as the decision itself? We need to (1) focus on policy outcomes as well as decisions; and (2) recognize that those outcomes are influenced by factors beyond the control of policymakers. For example, they can ensure that money is put aside to pay for an output such as doctors or teachers. However, they can do less to ensure an outcome, such as an improvement in the health or education of the population, because it also relies on the behaviour of that population (Colebatch, 1998: 114). 3. What is ‘the government’ and does it include elected and unelected policymakers? It is worth considering how actors gain membership within the policy process (Colebatch, 1998: 16; Hale, 1988). They may have a recognized source of authority, such as presidents and elected members of the legislature. Or, they may have a recognized form of expertise that gives them policy influence, or be required to make sure that policy decisions are carried out (Colebatch, 1998: 18–22, 31–32). Policy is made by ‘policy collectivities’ (1998: 23; Colebatch, 2006: 1). It is the ‘joint product of their interaction’ (Rose, 1987: 267–68). 4. Does public policy include what policymakers do not do? Policymakers may ignore an issue, or attempt to look like they are addressing it, because it is low on their list of priorities; they do not have the means to solve the problem; and/or, they know that they will face opposition if they try. They may not consider an issue to be a problem. Or, they may decide that they cannot, or should not try, to solve it. Many of these reasons raise crucial

issues of power and agenda setting discussed in Chapters 3 and 9. In many cases, issues are low on the agenda because actors exercise power to reinforce social attitudes – for example, about the causes of poverty (Bachrach and Baratz, 1970) – or manipulate policymaking procedures. In some cases, this absence of attention allows policy to be processed without a formal statement of intent by elected policymakers (Colebatch, 1998: 99), raising important questions about who is accountable for the outcomes. Overall, it is difficult to define policy, identify policymakers, and therefore provide certainty on the meaning of policymaking. This uncertainty makes it especially difficult to hope for ‘evidence-based’ policy (Box 2.2). Box 2.2 ‘Evidence-based’ policy and policymaking If it is hard to define policy and policymaking, it is even harder to define ‘evidence-based’ policy and policymaking (Cairney, 2016a: 2). There is no single accepted definition of policy. There are blurry lines between the actors making and influencing policy. Although we can define evidence simply ‘as an argument or assertion backed by information’, there is much dispute about what counts as good evidence (2016a: 3). There is even disagreement about the metaphor to employ, with ‘based’ often seeming too absolute and ‘informed’ more realistic (Nutley et al., 2007). Consequently, many discussions of ‘evidence-based policy’ are confusing ‘because people begin by complaining that they don’t have it without really saying what it is’ (Cairney, 2016a: 4). The same problem arises in definitions of policymaking. Key policymakers and practitioners describe a far messier policy process than the image portrayed by the policy cycle, which makes it difficult to imagine what an evidence-based policy process would even look like (Gluckman, 2016; Cairney, 2017a; Topp et al., 2018).

MEASURING PUBLIC POLICY However, we can rely on well-established techniques to make the task of policy studies more manageable. The first solution is to identify different types of public policy, including broadly defined policy ‘tools’ (Box 2.3) and more specific ‘instruments’. When we begin to identify a wide range of instruments, it soon looks like an overwhelming list of measures (Cairney et al., 2012; Birkland, 2017; Wu et al., 2017): 1. Public expenditure. Deciding how to tax, how much money to raise, on which policy areas (crime, health, education) to spend, and the balance between current (e.g. the wages of doctors) and capital (building a new hospital) spending. 2. Economic disincentives, such as taxation on the sale of certain products, or charges to use services. 3. Economic incentives, such as subsidies to farmers or tax expenditure (‘tax breaks’) on certain spending such as giving to charity or buying services such as health insurance. 4. Linking government-controlled benefits to behaviour (e.g. unemployment benefits only for people seeking work) or a means test (e.g. their income or savings). 5. The use of formal regulations or legislation to control behaviour. 6. Voluntary regulations, such as agreements between governments and other actors such as unions and business. 7. Linking the provision of public services to behaviour (e.g. restricting the ability of smokers to foster children). 8. Legal penalties, such as when the courts approve restrictions on, or economic sanctions against, organizations. 9. Public education and advertising to highlight the risks to certain behaviours, or ‘nudge’ techniques to influence individual choices (Box 2.3). 10. Providing services and resources to help change behaviour. 11. Providing resources to tackle illegal behaviour. 12. Funding organizations to influence public, media, and government attitudes. 13. Funding scientific research or advisory committee work.

14. Organizational change, such as the establishment of a new unit within a government department or a reform of local government structures. 15. Providing services directly, or via non-governmental organizations, or via networks of public and private actors. 16. Providing a single service or setting up quasi-markets. 17. Providing state services for free, charging, or expecting the market to deliver. Therefore, there is great value in providing a smaller and more manageable list of types of policy instrument, provided by the literature on policy tools (Box 2.3). This focus on tools highlights a strong relationship between the type of measure and its political costs (as well as the perceived nature of the policy problem – Box 2.4). For example, Lowi (1964; 1972: 299; Smith, 2002) argues that ‘policies determine politics’: the nature of the policy measure, and the level of coercion required to implement it, determines how policy is made. For example, in the case of redistribution, the difficulty relates to the sense that some populations clearly benefit at the expense of others (such as taxing A to provide benefits to B). Or, it highlights fundamental debates about how far the state should be allowed to go to tax and regulate the lives of its citizens. Distribution produces a reduced sense that one social group visibly benefits from public funding at the expense of another. Distributive policies, providing a benefit to a social group, also differ from regulatory policies if the latter incur costs (to comply or be fined for non-compliance) on individuals and private companies (Lowi, 1964: 690–91). In each case, the likely effect of policy on individuals will influence their levels of opposition, the ways in which they mobilize to influence policy, and the amount of government coercion required to ensure success. In other words, ‘every category of policy has its own political dynamic’ (Lowi, 1988: 726; McCool, 1995: 246–48). Box 2.3 Policy tools: their use and potential abuse

Lowi (1964; 1972) distinguishes between regulatory, distributive, redistributive, and constituent policies to argue that each measure is more or less difficult to sell and deliver. Subsequent typologies reflect some changes in the nature of measures available to, or preferred by, policymakers. Hood and Margetts (2007: 5–6) identify four types: Nodality describes being at the centre of the information network that underpins policy development; Authority refers to the power of policymakers to legislate and regulate; Treasure describes the money or resources available to support their policy decisions; Organization describes the resources and coordinative capacity – such as staff, buildings, and technology – at their disposal. Combined, these tools highlight a range of potential central government roles in ‘public administration’, from directing public service delivery at the centre to delegating authority and providing guidance with no obligation to comply (Hughes, 2017). John (2011) draws on an expanded list of tools, including a focus on persuasion and the rise of psychological techniques to influence social behaviour, including the now famous ‘nudge’ method based on exploiting the ways in which cognitive biases inform choices (Thaler and Sunstein, 2008; see also Box 7.1). Pykett et al. (2017) often describe nudge negatively as one type of’ ‘psychological governance’ to shape the way that citizens think or behave (see also Chapter 3 on Foucault). It contrasts with a more positive relationship with citizens during the ‘co-production’ of policy with service users (Durose and Richardson, 2015; Durose et al., 2017), as well as the many ‘tools of policy formulation’ used to generate good information before making choices (Jordan and Turnpenny, 2015). However, bear in mind that, if governments are aware of how their policy tools look, they may pay lip service to positive tools while actually using the full range. Or, they simply add new tools and instruments to a pile of existing measures without knowing what the overall effect will be (Peters et al., 2018: 8). There is a range of possible measures within broad categories, such as when ‘regulation’ can describe energetic ‘enforcement’ using

harsh penalties or to ‘standard setting’ and encouragement (Lodge and Weigrich, 2012). Further, the boundaries between each type of tool are blurry and many policy measures can include elements of each (Steinberger, 1980; John, 1998: 7; Smith and Larimer, 2009: 41). Policy measures also mean different things to different people, making it difficult for governments to predict and anticipate how individuals will mobilize in reaction to them (2009: 42). The idea that ‘policies determine politics’ should be treated with caution; Lowi’s point was that too many people assumed that policy analysis should begin with a discussion of policymaker aims rather than their environment, including the audience they face. Policymakers may act differently when they face different policy problems, or be more or less motivated to incur political costs. Different policy areas often have different characteristics, present different problems to solve, have different participants, and are associated with different styles of policymaking. They also have different starting points based on the legacies of past decisions (John, 1998: 7–8; Smith and Larimer, 2009: 234; Hill, 2009: ch. 7). Box 2.4 Types of policy problems The selection of policy tools relates strongly to the perceived nature of the problem and the politics involved. This calculation may vary strongly by policy ‘sector’, and it is common to discuss policy problems according to broad categories like health, education, environmental, social and criminal justice, and so on. It also relates to the problem definition (because there is no objective or apolitical definition of policy problems – Bacchi, 1999), and the extent to which policymaking is processed routinely in policy communities or breaks out into wider discussion (see Chapter 9). For example, McConnell (2010: 154–59) distinguishes between issues of high/low politicization or urgency, issues that affect a government’s reputation or image of governing competence, the pressure for consensus, and the level of party opposition. Newman (2019) compares issues of high/low certainty regarding risk (such as the health effects of e-cigarettes). The term ‘wicked’ often describes (quite loosely) the problems that seem highly ambiguous and intractable, and relate to

socioeconomic inequalities (Rittell and Webber, 1973; Newman and Head, 2017; Turnbull and Hoppe, 2018; McConnell, 2018; Cairney and St Denny, 2020; compare with ‘super wicked’ – Levin et al., 2012). ‘Morality’ policy describes issues debated on ‘first principles’ and subject to high ideological disagreement (Mooney, 1999; Knill, 2013). ‘Constitutional’ policies may be similarly unconducive to ‘normal’ policymaking (Richardson, 1982b).

NARRATIVES OF PUBLIC POLICY A focus on such a wide range of policy instruments and scenarios suggests that we may come to different conclusions on the nature of policy when we examine different indicators (or interpret the same indicators differently). Our ability to analyse all of the measures that constitute ‘policy’ will always be limited and we may have different ideas about which measures are the most important (Lavis et al., 2002). Similarly, governments have far more aims than they could possibly deliver, each aim competes for attention and resources, and some of them contradict each other. This problem often prompts them to identify a small list of priorities from a huge list of stated aims (Barber, 2012). For example, we may identify a high-profile policy measure that involves the use of legislation but does not have the money or resources to ensure that the policy aim is fulfilled. Or, we may find more successful examples of low-key, informal, and unfunded agreements, between the government and other actors, to pursue a common aim (although see Baggott, 1986). In practice, most areas will contain a complex mix of high- and low-profile, formal and informal, funded and unfunded, successful and unsuccessful decisions, and the aim of the researcher is to identify the most important policies from a long list of actions. The scope for different perspectives is demonstrated well by interview research. For example, I have interviewed interest groups favouring tobacco control who viewed UK legislation to ban tobacco advertising and smoking in public places as a watershed in public policy because previous regulatory measures were based on badly

enforced voluntary agreements between the government and the tobacco and leisure industries. I have also interviewed government actors who argue that such legislation represents incrementalism because other policy instruments demonstrate a clear direction of travel towards tobacco control (Cairney, 2007a). Such competing accounts cannot be rejected simply because they are biased. All accounts are biased in the sense that they are based on a limited number of indicators, the decision to give some indicators more weight, and an interpretation of the motivation behind, or effect of, policy measures. The potential for different interpretations comes from many sources, including •

•

•

•

The time frame. We may conclude that nothing much has changed in the short term but that a lot has changed in the long term, or vice versa. ‘Vignette’ studies focus in detail on one key event, while historical studies track policy change over decades. The level or type of government we study. Policy can often be made by a wide range of organizations (executives, legislatures, courts, devolved authorities, agencies) and influenced by many non-governmental actors. If we ‘zoom in’ to analyse key organizations, we might miss the dynamic of the wider political system. If we ‘zoom out’, we might miss important organizational details. Our expectations. When we identify policy change we link it, at least implicitly, to a yardstick based on how much we expect it to change (e.g. based on who we think is powerful) and how much we think policy should change under the circumstances (e.g. given the size of the problem or the level of public attention). Crenson’s (1971) study examines why some policymakers did not do enough to control pollution when its effect became well known (Chapter 3), while Cairney et al. (2012) explore the same argument on limited tobacco control. Motivation. Economic sanctions can be introduced to change behaviour or raise money. Policymakers can introduce measures to satisfy a particular interest or constituency and ensure a boost to their popularity; do something that is relatively easy to achieve in the short term; or focus on long-term outcomes (perhaps to

•

•

•

fulfil a commitment based on fundamental beliefs). The distinction is crucial if the political weight behind a measure affects its success (McConnell, 2010). Statistical comparisons. We may describe spending on an issue as a proportion of gross domestic product (GDP), the government budget, or the policy area’s budget, or in terms of change from last year vs. over many years. In some cases, the amount of money spent by government could be compared with that spent by industry (e.g. Cairney, 2007a compares health education with tobacco companies’ advertising budgets). Or, the level of fines or taxes could be compared to the incomes of individuals or organizations (e.g. a $1 million fine could bankrupt a small business but not affect Amazon). Contradictory and inconsistent policies. Things become complicated when governments appear to pursue contradictory policies in one area (e.g. subsidizing tobacco production while discouraging tobacco sales), or policies in one field that have unintended consequences for another (e.g. when school expulsion policies affect youth crime). Researcher bias. Richardson (1982a: 199) argues that most policy is processed in less visible and contentious arenas despite an understandable tendency for researchers to focus on ‘spectacular’ policy activity. Bacchi (1999: 200) argues that many policy scholars see themselves (incorrectly) as separate from the policy issues and processes they study, rather than people with personal beliefs about the nature of policy problems and who should solve them.

One possible response is to construct and share different narratives of policy development. It is possible to produce competing and equally plausible accounts of the type, speed, and nature of change for the same policy issue or series of events. This recommendation does not look particularly striking, but our choice of narrative has a crucial effect on the value of each theory. For example, theories of public policy that seek, primarily, to explain policy continuity are reinforced by narratives highlighting incremental change but not by narratives highlighting radical policy change. In other words, theories

are only valuable if the evidence supports them, but the evidence is rarely incontrovertible (Head, 2008). Since it is difficult to talk of the evidence with certainty, we should not pretend otherwise just to bolster our pet theories and make our published studies seem important.

FRAMEWORKS, THEORIES, MODELS, AND HEURISTICS The final way to make sense of policy complexity is to use frameworks and apply models and theories. The initial aim of a model is to provide a simplified representation of a specific object or process. It need not be a completely accurate representation. For example, think of a model car (McCool, 1995: x) or the London Underground map (Parsons, 1995: 59–61). A model can be used as a guide to description – to simplify and manage analysis by identifying the key features of a more complex entity. It serves as a heuristic device to help describe the real thing. A theory is more difficult to define, and there is some disagreement on how to define it, but let’s work with a set of analytical principles or statements designed to structure our observation and explanation of the world. For me, a theory sets out the object of study’s essential features and the relationship between those features (compare with frameworks in Chapter 7). Theories tell us what to look for and help explain what we find. Our initial aim when we use theories is to identify policy decisions and their main causes. Our ultimate aim is to examine the extent to which we can identify the same causes in other situations and therefore generalize from specific to numerous instances. Their usefulness is based on the idea that, although the world is immensely complex, we can still identify a manageable number of factors that explain what happens within it (John, 1998: 8; Sabatier, 2007a: 5). We might also hope that the theory is parsimonious; it explains to us what is happening in an efficient and not too complicated way. The theory-building process can often be divided into two stages. First, we move from the detailed identification of key events and

decisions in particular cases, towards the identification of abstract ideas that apply across all cases. Second, we use those abstract ideas to inform our study of subsequent case studies and, in doing so, attempt to confirm or deny the value of the theory (a problematic task, to say the least – see Chalmers, 1999; Cairney, 2013a; see also Chapter 13). For example, Baumgartner and Jones’ (1993) work on ‘punctuated equilibrium’ began as a way to explain why certain groups were able to exert power by monopolizing not only access to policymakers but also the way that people understood and addressed policy problems. They used rich case studies on tobacco, nuclear power, the environment, and pesticides in the USA. Then, Jones and Baumgartner (2005) developed the ‘general punctuation hypothesis’, a more abstract theory based on the ability of policymakers to process information and pay attention to policy issues, with the potential to be applicable to all policy areas and beyond the USA (Chapter 9). However, no one has managed to produce a theory applicable to public policy as a whole (Smith and Larimer, 2009: 15–17; see Chapter 13). The world is too complex, and there are too many causes of outcomes (some of which seem exclusive to individual cases) to allow for a truly parsimonious explanation. Parsimonious theories are valuable but only to explain part of the process. Consequently, much of the public policy literature represents ‘thick description’, or a process of modelling or mapping out complex terrain (Hill, 1997: 2; John, 1998: 8–9). Overall, the choice may be between using models to map particular outcomes and theories to explain many outcomes, but the boundaries between the two are not clear: both processes are based on the identification of the essential features of the world and how they relate to each other; and the same model may be applicable to many cases, while some theories do not explain much beyond their initial case studies (despite the best efforts of their authors). Or, if many theories are useful, perhaps we can combine their insights (Chapter 13).

WHAT IS THE POLICY CYCLE?

The policy cycle is the best known way to organize the study of policymaking (and the best way to introduce a discussion of policy theories). It is tempting to view the policy cycle as a theory because it is used to represent the policy process in multiple political systems. The same can be said for many of its stages – terms such as ‘agenda setting’ and ‘implementation’ are widely applicable. Yet it is generally viewed as a simple model of complex processes which does not explain, for example, why issues arise on the policy agenda or why a particular decision is made. It is a model in two related ways: 1. Prescription – a model for how policymakers should operate, to make sure that their decisions are made in a systematic way. 2. Description – a model to describe how they do operate, to simplify the study of how they make decisions (Hogwood and Gunn, 1984: 42–43). It divides the policy process into a series of stages, from a notional starting point at which policymakers begin to think about a policy problem to a notional end point at which a policy has been implemented and policymakers think about how successful it has been before deciding what to do next. The image is of a continuous process rather than a single event. There is some variation in the literature regarding the number of stages in a policy cycle, but most describe the identification of policymaker aims, the formulation of policies to achieve those aims, the selection and legitimation of policy measures, implementation, and evaluation. Lasswell (1956) set the ball rolling with a short text highlighting essential policymaking functions (or requirements) – intelligence, recommendation, prescription, invocation, application, appraisal, and termination – rather than stages in a cycle. In Jones’ (1970) book, each stage commands a chapter: defining the problem, setting the government’s agenda, formulating proposals, having a programme or coherent set of proposals legitimated by the legislature, assigning a budget, implementing, and evaluating policy (see also Anderson, 1975; and Brewer and deLeon, 1983: 18–20; Parsons, 1995: 78–79; Smith and Larimer, 2009: 30).

Most texts were based on the US political system, which enjoys a strong legislature and a budget process that must be agreed between the executive and legislative branches. This explains why the legitimation and budgeting stages receive more attention than they might in a study of centralized parliamentary systems. However, in a more abstract sense, the model portrays how any government attempts to translate ‘inputs’ such as public demands into ‘outputs’ such as public policies (Hill and Varone, 2016). Further, the idea that this is how the cycle should operate in a democratic system is widespread: the public expresses its will in elections, elected policymakers fulfil their promises and legitimize their decisions at separate stages, and then civil servants and other bodies carry them out (Jann and Wegrich, 2007: 44). Comparable models for the UK and USA (Jenkins, 1978; Hogwood and Peters, 1983: 8; Hogwood and Gunn, 1984: 7–11) and for Australia (Bridgman and Davis, 2003: 100; Howard, 2005: 10–11; Althaus et al., 2013), produced a very similar list of stages (set out in Figure 2.1): • • •

Agenda setting. Identifying problems that require government attention, deciding which issues deserve the most attention, and defining the nature of the problem. Policy formulation. Setting objectives, identifying the cost and estimating the effect of solutions, choosing from a list of solutions, and selecting policy instruments. Legitimation. Ensuring that the chosen policy instruments have support. Measures include legislative approval, executive approval, seeking consent through consultation with interest groups, and referenda.

Figure 2.1 The generic policy cycle

•

• •

Implementation. Establishing or employing an organization to take responsibility for implementation, and for ensuring that the organization has the resources (staffing, money, legal authority) and that decisions are carried out as planned. Evaluation. Assessing the extent to which the policy was successful: if the policy was appropriate, implemented correctly, and had the desired effect. Policy maintenance, succession, or termination. Considering if the policy should be continued, modified, or discontinued.

While the use of the cycle to explain policymaking has diminished, many of the stages or functions of policymaking still command their own literature, and the model provides simple descriptive lessons

(the study of public policy does not end when a decision has been made) and prescriptive advice (think about implementation when you engage in formulation – Malbon et al., 2019).

AGENDA SETTING AND POLICY FORMULATION The study of agenda setting is really the study of power and ideas (Chapter 3), and discussed at length as part of our discussion of punctuated equilibrium theory (Chapter 9). Similarly, policy formulation (and legitimation) involves the production and selection of technically and politically feasible ‘policy solutions’, which forms part of a wider discussion of multiple streams analysis (MSA; Chapter 11). However, since this book focuses somewhat on ‘evidence-based policymaking’ (Box 2.2), we should note two key points about these stages. First, it is tempting to see each stage as part of a dry and technical process. However, agenda setting involves a fundamentally important competition to exercise power to define issues as policy problems and establish their severity and cause. Bacchi (1999, 2009) warns against assuming that policy problems exist naturally. Rather, we need to ‘think about policy in anthropological terms’, to understand how any dominant way to think of an issue came to be: why is the government involved in the first place, how does it describe the nature of the problem, who does it blame, who should benefit from any solution, and how much is it willing to help (2009: ix)? This process can involve evidence, but politics is about exercising power to establish how we understand the world, which goes well beyond the routine search for information. Other practices, including telling stories about policy issues (i.e. Narrative Policy Framework, NPF) and characterizing target populations (i.e. Social Construction and Policy Design, SCPD), are a key feature of Chapter 4. Second, the ‘tools of policy formulation’ come in several forms, designed to ensure good policymaking procedures (e.g. to maximize public participation), policy implementation (e.g. regulations), and effective policy analysis (Howlett, 2000, 2011; Radin, 2000, 2013; Turnpenny et al., 2015). The latter involves using evidence to

generate a number of feasible solutions to meet a policy aim. This field has expanded in a big way, partly to reflect the proliferation of methods – from decision trees to cost-benefit analyses and complex systems modelling – to gather and analyse information (Turnpenny et al., 2015: 4). However, the availability of analytical tools does not match their use: there are waves of high and low government interest in them (2015: 4) and, even when they invest heavily in sophisticated tools, there is no guarantee that policymakers will actually use them (Nilsson et al., 2008: 335).

IMPLEMENTATION The study of implementation is based on the simple point that decisions made by policy-makers may not be carried out successfully. Instead, we can identify an implementation ‘gap’ which represents the difference between the stated expectations of policymakers and the actual policy outcomes (deLeon, 1999: 314– 15; Hill and Hupe, 2009: 11). Hogwood and Gunn (1984: 197) attribute implementation failure to three main factors: bad execution, when it is not carried out as intended; bad policy, when it is carried out but fails to have the desired effect; and bad luck, when it is carried out and should work but it is undermined by factors beyond the control of policymakers. Cairney (2009a: 354) also suggests that expectations gaps are wide because participants have unrealistic expectations. These explanations can be standardized in much the same way as the stages of a policy cycle. The common aim is to highlight the conditions that have to be met to ensure ‘perfect’ implementation success (Hood, 1976: 6; Hogwood and Gunn, 1984: 198; Sabatier, 1986: 23–24; Cairney, 2009a: 357). The understanding is that, in fact, they explain why policies fail or achieve partial success: 1. The policy’s objectives are clear, consistent, and well communicated and understood. A clear policy provides legal weight and acts as a ‘standard of evaluation’ (Sabatier, 1986: 23– 24). A vague policy is subject to multiple interpretations and the potential for bad execution even by implementers with the best of

2. 3.

4.

5.

6. 7.

intentions. For Hogwood and Gunn (1984: 204–206), perfect implementation requires that policymakers agree on a common understanding of objectives; fully specify those objectives in the correct sequence; and coordinate the implementation process with no breakdown in communication. The policy will work as intended when implemented. The policy must be a good solution to the problem, based on a ‘valid theory of cause and effect’ (if we do X, the result will be Y) (1984: 201). The required resources are committed to the programme. Commitment can refer to money, staffing, and physical materials (1984: 200), or giving the task to an organization that ‘would be supportive and give it a high priority’, providing the right legal and economic sanctions and incentives to ‘overcome resistance’ (Sabatier, 1986: 23), and investing attention and political will to an issue over the longer term. Policy is implemented by skilful and compliant officials. The term ‘perfect obedience’ (Hogwood and Gunn, 1984: 206) reminds us that we are discussing an ideal-type; in fact, the discretion held by implementing officials with specialized jobs is ‘unavoidable’ (Lipsky, 1980: 13–16; Sabatier, 1986: 23; Hill and Hupe, 2009: 26). Dependency relationships are minimal. There are few ‘veto points’ or links in a ‘delivery chain’; the implementing agency does not rely on the cooperation of others (Hogwood and Gunn, 1984: 202; Sabatier, 1986: 23; Barber, 2012: 85–87). Support from influential groups is maintained. The implementation process is long, and continuous support from key actors is crucial. Conditions beyond the control of policymakers do not significantly undermine the process. Socioeconomic conditions can often be unpredictable (Sabatier, 1986: 25) and affect the costs of, or support for, implementation (e.g. consider the effects of a ‘baby boom’ on education demand, or an ageing population on pensions or personal care policies).

Pressman and Wildavsky (1973; 1979) draw on such factors to explain why the provision of US federal funds, for public works

programmes to solve the problem of ‘high unemployment and racial unrest’, did not have the desired effect in Oakland, California. Many projects were not completed because the building materials were not secured; there were too many ‘decision points’; the implementing agencies relied on cooperation with many other authorities and private companies; the costs in some projects were higher than expected; the legal framework was inadequate; there was disagreement about how spending money on training programmes would boost employment; the business loans programme contained incompatible aims; local groups opposed aspects such as low-cost public housing; and the Economic Development Administration could not identify the influential groups from which it needed support (compare with Compton and ‘t Hart, 2019 on policy success). We may suppose that some of these problems, such as interdependence, are specific to the US model of government in which power is shared by many organizations. However, examples from the ‘majoritarian’ UK demonstrate comparable experiences (King and Crewe, 2014). Marsh and Rhodes (1992a) employ a similar framework to identify policy failures of the Thatcher government (1979–90): its privatization programme suffered because political objectives clashed with economic/ideological objectives; monetarist economic policy suffered because the causal link between inflation and the money supply was suspect; the control of local authority budgets was undermined by local authorities; industrial relations policies lacked an adequate legislative framework; a policy to reduce social security spending was undermined by rising unemployment; and demographic effects (an ageing population) undermined the policy of healthcare financial stringency. In particular, they identify problems of interdependence exacerbated by the government’s unwillingness to consult with interest groups before making decisions (1992a: 185). Those groups and agencies, which were affected by policy but not consulted, ‘failed to co-operate, or comply, with the administration of policy’ (1992a: 181). As Jordan and Richardson (1987: 242) show, there are four main advantages to consultation. First, it secures wider participation in the political system, and hence support for that

system. Second, it creates a sense of involvement, and thus ensures greater commitment to the success of a policy. Third, it allows the government to benefit from the practical experience of those consulted. Finally, it allows some portion of responsibility for that success to be transferred to other participants. The argument that consultation allows a more informed government with fewer problems of agreement and compliance is almost universally applicable (albeit not applied universally).

Top-down and Bottom-up Implementation This approach is often labelled ‘top-down’ because it is built on two related assumptions: 1. Descriptive – decisions are made at the ‘top’ (by central government or the legislature) and carried out at the ‘bottom’ (by implementing organizations). 2. Prescriptive – decisions should be made at the top and carried out at the bottom. Indeed, Barber (2012: 293–94) describes his approach as ‘modelled on how the chief executive of a business would drive key priorities’ to help strengthen the power of policymakers at the centre of government. In contrast, the ‘bottom-up’ literature questions the assumption that central government is – or should be – the main influence on policy outcomes. Lipsky’s (1980) classic argument is that policy is, to a large extent, made by the ‘street-level bureaucrats’ (including teachers, doctors, police officers, judges, and welfare officers) who deliver it. Bureaucrats are subject to an immense range of, often unclear, requirements laid down by regulations at the top, but are powerless to implement them all successfully (1980: 14). This is not necessarily an argument based on ‘disobedience’. Rather, even the most committed workers do not have the resources to fulfil all of their job requirements (1980: xii). Instead, they use their discretion to establish routines to satisfy a proportion of central government objectives while preserving a sense of professional autonomy

necessary to maintain morale. Indeed, there is now a large literature describing the ‘coping’ mechanisms of ‘front line workers … interacting with clients’ to ‘master, tolerate, or reduce external and internal demands and conflicts they face on an everyday basis’ (Tummers et al., 2015: 1101–12). The irony is that the added pressure from central government policy obliges implementers to manage their own day-to-day activities (see Cook, 2018). Therefore, policy change at the top will not necessarily translate to change at the bottom. Hjern (1982: 213–16) argues that the hope or assumption that policy is controlled by a single central actor (possessing clear and consistent aims) exacerbates policy failure. Inattention to the complexity of implementation causes difficulties in the administration of policy, producing feelings of powerlessness when no one seems to be in charge. Instead, we should recognize intra-departmental conflict, when central government departments pursue programmes with competing aims, and interdependence, when policies are implemented by multiple organizations. Programmes are implemented through ‘implementation structures’ where ‘parts of many public and private organizations cooperate in the implementation of a programme’. It is difficult to force decisions on actors who are employed by other organizations, so it is unrealistic to think that a sole central actor could secure its own aims and objectives irrespective of the actions of the others involved. Although national governments create the overall framework of regulations and resources, and there are ‘administrative imperatives’ behind the legislation authorizing a programme, the main shaping of policy takes place at local levels by implementation structures in which national considerations may play a small part (Hjern and Porter, 1981: 213). Although published decades ago, this narrative of limited central control is just as relevant today, and described with reference to complex, multi-level, or polycentric governance (Chapters 6–8; Cairney et al., 2019a). The ‘bottom-up’ alternative is summed up by Barrett and Fudge’s (1981: 4) description of decisions made at the top as part of a bargaining process, with policy modified continuously as each actor

involved in implementation, ‘attempts to negotiate to maximize its own interests and priorities’. Central government may prove to be the most influential actor, but we should not assume so, particularly when it appears not to have a strong interest in the subject. We should also not assume that everyone involved attaches the same ‘meaning’ to policy measures (Yanow, 1996; Schofield, 2004). Instead, we could shift our focus to the ‘bottom’ by identifying the implementing agencies involved and exploring how they operate. Policy from the top may represent only one of many factors (including the lower level ‘environment’ in which local demands and needs arise) relevant to the deliberations and actions of those agencies (Barrett and Fudge, 1981: 25; Colebatch, 1998: 12, 28– 29). Can we call the outcomes of practices so removed from the top ‘government policy’ (Colebatch, 1998: 30)? These differences in approach led initially to the ‘top-down versus bottom-up’ debate. Their differences centred on empirical issues (e.g. the extent to which implementing actors can thwart the intentions of the centre), research design (should we focus on decisions made at the top or the implementation structure?), and prescription (who should make policy? How clear should the policy aim be? Who should implementers be accountable to?) (Hudson, 1993: 392–95; deLeon and deLeon, 2002: 474; Hupe and Hill, 2007; Hill and Hupe, 2009: 15–16). The debate was followed by ‘thirdgeneration’ studies that sought in vain to distil the vast range of variables, or causes of policy outcome variation, into a manageable and testable general theory (Goggin et al., 1990; Parsons, 1995: 480–81; Howlett and Ramesh, 2003: 90; Smith and Larimer, 2009: 170–72). To some extent, top/bottom debates reflect biases in research design. A focus at the bottom highlights a multiplicity of influences and distance from central government, but does not tell us how many organizations meet targets set at the top. A focus at the top highlights the central control of a small number of issues, ignoring the bulk of government responsibilities which are delivered routinely and with minimal attention from elected policymakers (Cairney, 2009a: 360). Therefore, if you are a fan of easy compromise, you can conclude that different approaches are valuable in different

circumstances (Matland, 1995: 165–67; Lundin, 2007; compare with Hill, 2009: 140). There have also been attempts to combine the merits of top-down and bottom-up approaches (Hill and Hupe, 2009), set new directions in implementation research (Carey et al., 2019), and incorporate implementation issues into the study of policy learning (Chapter 12).

EVALUATION Evaluation involves assessing the extent to which a policy was successful. A top-down approach might identify success if the implementation gap is small or if there is evidence of compliance (Matland, 1995: 154). It might link success to common arguments about representative democracy, in which policymakers at the top enjoy the most legitimacy when they are elected and accountable, and policy success happens when they deliver on their promises to the public (Hill and Hupe, 2002: 71; Linder and Peters, 2006: 31; Cairney, 2009a: 358–59). Indeed, in some cases, policy is implemented using top-down evaluation techniques: published targets provide an expectation of implementer behaviour (but often produce unintended consequences – Hoque et al., 2004; Barrett, 2004; Hood, 2007; Carey and Harris, 2015) and dedicated units are created to monitor and evaluate levels of compliance (Lindquist, 2004; Duffy, 2017). Yet the ‘top’ often produces contradictory or vague policy aims. There may also be competing forms of legitimacy. Many local officials are elected. Even if not, it is sensible to build some form of discretion into policy design to allow locally knowledgeable experts to adapt policies to particular circumstances (Hogwood and Gunn, 1984: 207–8, 198; Colebatch, 1998: 68; deLeon, 1999: 320; deLeon and deLeon, 2002: 478). This normative dimension is a key feature of complexity theory, MLG, and and polycentric governance (Chapters 6–8). It has been debated for some time in the US literature on intergovernmental relations, studies of member state ‘compliance’ in the EU, and in relation to UK devolution (Duina, 1997; Falkner et al., 2004; Mastenbroek, 2005; Pülzl and Treib, 2007: 97–99; Cairney, 2009a).

Evaluation is also complicated because success is in the eye of the beholder and difficult to measure. Relatively technical forms of measurement involve the selection of a large mixture of indicators that are hard to compare, or a small number of measures that are handpicked and represent no more than crude proxies for success (Cairney, 2009a: 369; see also Andrews and Martin, 2007). Relatively political measurement involves a convoluted process that Marsh and McConnell (2010: 571, 580) distil heroically into seven factors. They include practical issues, such as how long we should wait before evaluating a policy, how much information there is, how we can separate the effects of this policy from others, and what the benchmark should be (e.g. the government’s intentions, past outcomes, the success of other countries). They also include the more fundamental questions: • •

Whose success are we measuring: the government, its stakeholders or its target group? What successes do (and should) policymakers care about: the impact of policy on their popularity, how easy policy change is to achieve, or the long-term impact on the population?

Such discussions demonstrate that evaluation measures are not ‘objective’ (Marsh and McConnell, 2010: 575) and that common evaluation terms such as ‘effectiveness, efficiency and equity’ give the misleading impression that they are technical (Brewer and deLeon, 1983: ix). Rather, any measure of success has to consider which measures we should select, who benefits the most from policy outcomes, and what should be done following evaluation. Evaluation is ‘an inherently political activity’ (Taylor and Balloch, 2005: 1; Hogwood and Gunn, 1984: 221; Bovens, ‘t Hart, and Kuipers, 2006). It should not be considered as an afterthought in our policy analysis. Indeed, the likelihood of success in terms of popularity may be the key reason that policymakers seek positive long-term outcomes (McConnell, 2010). Therefore, although there is now a massive technical literature devoted to evaluating policy (see, for example, the online Campbell Library https://campbellcollaboration.org/library.html), how this

evidence is used remains unclear, and almost no account highlights a clear and direct impact on policy (Oliver et al., 2014a, 2014b). For example, it may: have an ‘enlightenment’ function, to help reshape thinking over the long term (Weiss, 1976, 1977a, 1977b, 1979); be just another form of evidence to be used to set the policy agenda (Henry and Mark, 2003); and/or be used largely to justify existing policies (Boswell, 2009). For example, the UK New Labour government described evaluation as part of its evidence-based policy agenda – in which it projected the sense that evidence would trump ideology, empower citizens, and help join up government – but, like all governments operating in political contexts, actually used evaluation evidence politically (Sullivan, 2011: 501).

POLICY MAINTENANCE, SUCCESSION, AND TERMINATION The notional final stage is when policymakers decide if the policy should be continued. This stage reinforces the idea that ‘policy determines politics’ because new policies are often pursued largely to address the problems caused by the old (Wildavsky, 1980: 62). As Jones (1970: 135, 11) puts it, ‘the end is the beginning’, because government action results from the ‘continuing application and evaluation of ongoing policies’ (perhaps with scope for larger departures during periods of crisis – Geva-May, 2004). A decision made in the past influences how the next decision will be made (Colebatch, 1998: 10–11). For example, it may be easier to amend a policy, and present it as new, than create a new one. Hogwood and Peters (1983: 26–27) suggest that ‘policy succession’ is more likely than ‘innovation’ because most of the hard work has already been done: the issue is recognized as a legitimate problem for the government to solve; a service delivery organization exists; the policy has resources devoted to it; and it has an established clientele. More significant innovations would require a process to establish these factors and perhaps the termination of another policy. The latter has immediate financial costs, may produce the perception of policy failure, and may be opposed by

interest groups, clients, and the organizations that depend on the policy to survive (deLeon, 1978; Hogwood and Peters, 1983: 16–17; 1982; Hogwood, 1987; Parsons, 1995: 574–75; Geva-May, 2004). The policy cycle therefore tends to ‘end’ with succession (Hogwood and Gunn, 1984: 61–84).

BEYOND THE POLICY CYCLE The policy cycle has three good points: 1. Prescriptive. The idea that elected people make policy and unelected officials carry it out has normative weight (Colebatch, 1998: 77). 2. Practical. Policymakers might use a sequence of stages to describe or plan their work (see Althaus et al., 2013; Bridgman and Davis, 2003 on Australia; Marchildon, 2001 on Canada; and the European Commission, 2011). It often represents the ‘public face of public policies’ (McConnell, 2010: 222). 3. Descriptive. It is a simple model understood by most students (Wu et al., 2017). However, we can question its value with reference to all three points: 1. Prescriptive. There are many alternative models of good policymaking, many of which stress the benefits of multi-level, ‘polycentric’, or pragmatic governance to address (1) multiple sources of governing legitimacy and (2) the need to accept the limited coordinative capacity at the ‘top’ (Cairney et al., 2019a). Indeed, central policymakers would often rather delegate responsibility to other actors and find pragmatic ways to legitimize actions by unelected bodies (Jordan and Richardson, 1987: 233; John, 1998: 29). 2. Practical. Fewer policymaking organizations use the cycle to describe what they actually do (Box 2.2). Everett (2003: 65; see also Colebatch, 1998: 102) also questions its practical value: it is sold as the basis for ‘rigorous scientific method’ to produce good,

equitable, policies, but really exists to ‘smooth’ policymaking after difficult decisions have been made (Everett, 2003: 66–68). 3. Descriptive. It does not provide an accurate description or good scientific explanation of how policy is made. It is simplistic as well as simple: a better theory would describe ‘multiple interacting cycles involving numerous policy proposals and statutes at multiple levels of government’ (Sabatier, 2007a: 7; Topp et al., 2018). For such explanation, academic accounts draw primarily on policy theories. The cycle was useful when it emerged but has now ‘outlived its usefulness’ (Sabatier, 2007a: 7; compare with Hogwood and Gunn, 1984: 6–11; Parsons, 1995: 80–81; Jann and Wegrich, 2007: 56; Smith and Larimer, 2009: 235). Radin (2000: 15, 34; see also Howard, 2005: 4) associates the cycle and comprehensive rationality models with the early post-war period of radical change in countries like the USA. Policy scientists were able to use those models because the overall direction of policy was driven by a small number of policy-makers at the centre who relied on an elite group of analysts to produce ‘objective’ ways to gather facts and analyse and solve policy problems. This post-war image contrasts with modernday policymaking characterized by four main factors. The first is a diminished sense of optimism regarding the government’s ability to solve problems through objective scientific analysis. The status of policy analysis has diminished (John, 1998: 32–33) and policy scientists have to compete with many other actors for policymaker attention. In that context, the modern reference to ‘evidence-based policymaking’ is to bemoan its non-existence rather than identify its central role (Cairney, 2016a). The second is what Heclo (1978: 94) describes as an end to the ‘clubby days of Washington politics’ and Jordan (1981: 96–100) links, in other countries, to a shift from corporatism, in which group–government relations are centralized and exclusive, towards a more fragmented system with many more policy participants. The rise in government responsibilities mobilized more groups and stretched the government’s resources, producing its increased reliance on outside advice. This rise in activity from multiple sources often caused issues

which were once ‘quietly managed by a small group of insiders’ to become ‘controversial and politicized’ (Heclo, 1978: 105). The third is the focus on not one but multiple centres of authority – the dispersal of power from a single central actor towards many organizations and sources of authority and influence. The latter can range from highly visible sub-central organizations led by elected policymakers to almost invisible or specialized networks of unelected policymakers and influencers. Overall, we have a rather complex and shifting picture. On the one hand, policymaking is unstable: populated by a fragmented government and many participants, potentially with different values, perceptions, and preferences (Sabatier, 2007a: 3–4). On the other, influencers often share close and stable relationships with government officials who operate out of the public spotlight. Contemporary theories seek to capture and explain this picture of policy choice within a complex policymaking environment (Cairney and Weible, 2017; Heikkila and Cairney, 2018). As Chapters 3–12 demonstrate, there is a wide range of concepts and theories on which to draw, and each theory provides its own explanation. However, most of the theories in this book try to capture the elements summarized in Figure 2.2. In this image, we retain simple description but dispense with the idea of an orderly process of stages. It does not describe any substantial interaction between the concepts to show cause and effect. Rather, it provides a checklist of the concepts that we need to include to understand each theory and provide a full account of the policy process. Therefore, Figure 2.2 provides a more useful reference point for each chapter in the book: how does each theory account for policy choice, and how does it conceptualize an environment containing many actors, institutions, networks, and ideas, as well as influenced by events and changing contexts?

Figure 2.2 Key elements of policy choice in a complex policymaking environment Source: Cairney (2017a). Compare with the simpler Figure 13.1 on p.232.

CONCLUSION The study of public policy is challenging because the subject matter is so complex, with too many causes of variation in policy outcomes to make it as predictable as we would like. We can address this problem by producing a checklist of ways to simplify and manage our task: Define, categorize, and measure. Seek ways to define policy, and then measure and describe the nature of policy change by considering basic distinctions (such as between aims, decisions, and

outcomes) and identifying policy instruments and types. We use these indicators to describe the nature of policy decisions and measure the extent to which policy has changed in each area. Categories and typologies help us move from the general identification of complexity towards a specific understanding of key processes. Consider the factors that affect our narratives of policy development. All narratives of public policy are based on our selective interpretations of a limited number of measures. We need to be selfaware when constructing narratives and be open to different accounts of empirical results. One way to encourage reflexivity is to produce more than one narrative and compare their outcomes. It should make us more sensitive to some of the assumptions in our favoured version of events (rather than take them for granted). Particular narratives may support particular theories of public policy. Raise questions about power. We can explore issues of power (Chapter 3) by considering who has formal authority and the reputation for being powerful, who makes and influences the most important decisions, and why some key decisions seem not to be made (some problems and solutions receive attention while others remain off the agenda). Map the policymaking environment. We can appreciate the context within which decisions are made by considering the nature of the policy problem, the pressure that policymakers are under to address it, and the resources they can use to solve it. The idea that policy is influenced heavily by the policymaking environment is explored further in Chapter 6. Consider the policy cycle. The image of a cycle helps us consider the idea of ‘policy as its own cause’, the stages through which policies appear to go before an authoritative decision is made, and the difference between a policy decision and its implementation. Use theories of public policy. A general aim within science is to examine the proportion of outcomes that can be linked to the same

causes. Our aim is to apply this broad approach to public policy: to explain why certain policies exist, how they were made, and what happens when they are made. To do so, we identify the most important factors in the policy process, such as how and why choices were made; the role of institutions; the relationship between governments and interest groups; the socioeconomic context; and the role of ideas. We then examine how these factors relate to each other by learning from the most established theories of public policy. Our aim is to examine how each theory of public policy gives us a better understanding of policy processes. To this end, most chapters outline a particular theory and describe what it tells us about the policy process; what factors we should identify; and how they relate to each other. In most cases, we consider the application of theories to a single policy process (usually the political system in which the theory developed) before considering how applicable they are in other countries. Chapter 13 then considers what these theories tell us when considered together. Can we use these theories to accumulate knowledge of policy processes? Policy tool or instrument – a device used to help turn a broad policy aim into a specific action. The terms are often used synonymously, but I use ‘instrument’ to describe the wide range of measures and ‘tool’ to describe key categories. Narrative – an account of a series of events. A non-fictional story. An attempt to ‘render various series of events into an intelligible whole’ (Kay, 2006: 23). Compare with Box 4.4. Heuristic device – a tool used to guide investigation. Framework – for Ostrom (1999: 39–40), frameworks identify concepts and organize analysis, theories make general assumptions about the causal relationships between concepts, and models make specific assumptions about objects of enquiry (but see Cairney, 2013a).

3 Power and Public Policy Key themes of this chapter: • •

• • •

•

We need to define power to demonstrate its importance to public policy. The ‘three dimensions of power’ debate focused on our ability to measure power and its outcomes. The most visible forms of power are measurable, but the most worrying and least visible are often not. Actors exercise power to treat matters as public problems (worthy of our attention and government intervention) or private issues (none of our business). We can identify the right to exercise power, such as when popular consent underpins a state’s authority. Power also relates strongly to the role of ideas: the use of persuasion to set the agenda; and the beliefs, norms, and rules that are so taken for granted, or unshifting, that we often describe them as structures. A key aspect of ‘evidence-based’ policymaking is the power to decide what knowledge – and therefore who – counts.

INTRODUCTION: THE CENTRALITY OF POWER TO PUBLIC POLICY STUDIES Power is one of the most important but least clear concepts in political science. We need to define it to explain its role in public policy research, and our definition can have a profound effect on what we study. It says much about which aspects of political life we think are most worthy of research, in much the same way that governments declare the problems most worthy of their attention. It also highlights the importance of research methods, since

uncertainty about the meaning of power may lead us to wonder how to gather knowledge of it. In public policy, the question of power arises in its most basic form when we ask: who is in charge or who are the policymakers? Who is responsible for policy change? Who is thought to be in charge and who is actually in charge? We may also use discussions of power to explain why policies change or remain stable. For example, can policymakers exercise power to force or resist change in the face of opposition? Can actors use power to ‘set the agenda’ and encourage policy change in some areas at the expense of others? Or can actors use more hidden forms of power, such as the manipulation of knowledge and beliefs, to restrict debate and minimize attention to the need for change? To answer these questions, we may need to broaden our discussions from the exercise of power by individuals to the role of institutions, or from formal sources of authority to informal sources of influence, and to consider the routine and everyday ways in which inequalities of power produce unequal outcomes in society. Our starting point is the ‘community power’ debate, which prompted wider debate on three ‘dimensions’ of power. It began with the modern study of elitism and pluralism, examining the extent to which power is concentrated or diffuse within government and society. While Hunter (1953, 1980) and Mills (1956) identified the reputations of people in powerful positions to identify oligarchy, Dahl (1958) questioned the importance of reputations, focusing instead on the exercise of power and the effects of one actor’s power over another during key decisions. The debate highlighted different understandings of how to observe power, linking the problem of definition to methodology. It continued when Bachrach and Baratz (1962) questioned Dahl’s focus on directly observable behaviour (often linked to behaviouralism). Actors exercise power behind the scenes by ‘setting the agenda’ and limiting public debate. Or people may feel powerless because they do not have the opportunity to contribute to key decisions; they cannot find the arena to express their views, or feel unable to express them. The ‘second face’ of power describes ‘non-decisionmaking’, including visible attempts to argue that the government

should not be involved in private matters, or less visible actions to ensure that some individuals and groups do not engage. The ‘second face’ raised a research methods problem because power is not easy to observe. The ‘third dimensional’ view magnifies this problem. Lukes (1974) identifies a process in which some people benefit when others do not act according to their own ‘real’ interests. Others manipulate their preferences through the control of information and socialization. Power may be furthered by a ‘structure’ or force, independent of the actions of individuals, such as a dominant ideology or set of rules within government that block certain types of action. In broader terms, the exercise of power within institutions is rule-bound and not reducible to the sum total of the actions of individuals. Thus, power can be observed but also theorized from a broader examination of social, economic, and political relations. This understanding raises new problems about the language we use to describe the ‘actions’ of structures ‘exercising’ power and the methods to observe allegedly unobservable forms. During such debates, we examine proof and measurement: how do we demonstrate that some are powerful and others are powerless? What is the evidence and where can we find it? Does the right to possess power – such as by an elected government – correspond to the actual exercise of power? The simplest measure of power comes from certain people benefiting regularly from policy outcomes, but note that they can: exercise power directly to get what they want; manipulate others into thinking that the outcome benefits both parties; or benefit from an outcome without being responsible for it. However, a measurement-driven approach is not exhaustive. It runs a high risk of ignoring aspects of power that we can theorize convincingly without direct observation of action. It may account for the role of ideas in relation to persuasion to set the policy agenda, but not the beliefs, norms, and rules that seem so strong that we describe them as structures. Most importantly, it may prompt us to ignore fundamental issues of unequal power – in relation to factors such as gender, race, class, and sexuality – which are better served by critical theory than behaviouralist method.

DEFINITIONS OF POWER The concept of power encompasses a vast range of behaviours, including the ability to get what you want despite the resistance of others, the possession of authority based on consent, and the inability to exercise autonomy when subject to ‘structural’ power. Power may be exercised visibly or hidden from view; used for collective ends or at the expense of others; concentrated or diffuse; and used for legitimate or insidious purposes. It may be associated with visible reputations that affect the actions of others, or more subtle inequalities with less obvious effects. In some discussions, power may encompass or be treated as synonymous with terms such as influence, authority, and force; in others, it may be differentiated from them (Box 3.1). Box 3.1 Discussions of power ‘Power’ can refer to many concepts and arguments, including: • • • • • • • • • • •

The ability to get what you want despite the resistance of others The ability to influence the choices of others The ability to influence an actor’s decision-making environment Power as a resource or capacity, and the exercise of power Power based on popular support, used legitimately or illegitimately The power to change or obstruct Material sources of power – economic, military, governmental, cultural Colonial power, used for the ‘control of other people’s land and goods’, and built on slavery and indentured labour (Loomba, 2007: 8) Power as knowledge or embedded in language Reputational power Achieving compliance by using overt or tacit threats (‘power’), without using threats (‘influence’), restricting someone’s choices (‘manipulation’), and overcoming their non-compliance

• • • • • •

(‘force’); or when one’s position is respected (‘authority’) (Bachrach and Baratz, 1970: 17–38; Arendt, 1986: 64–65) The ability of a social class to realize its interests Decision or non-decision-making (two ‘faces’ of power) Three ‘dimensions’ of power Power diffusion or centralization Power versus systematic luck Inequalities of power related to factors such as gender, race and ethnicity, class, sexuality, disability, age

A key normative theme is that the use of power should not get in the way of ‘democracy’ (McGarvey and Cairney, 2008: 220). For example, we know that there are inequalities within politics and society, but hope that there are safeguards on governmental and non-governmental actors to limit their effects (Dahl, 1961). The study of power focuses on the extent to which inequalities in the possession of power translate into political outcomes. The key empirical themes are as follows. First, power can be the capacity and potential to act; as a resource to be used. However, second, we may not know how much power an actor has until they exercise it and we analyse the outcomes. Third, it is relatively difficult to identify the exercise of power by actors who are not individuals: can institutions and ‘structures’ act to exercise power? Finally, how do we study power? Should we use interviews to establish the reputations of elites or gauge the power of participants to influence key decisions? Should we deduce and theorize power relations from language, and observed and unobserved behaviour? Many answers to these questions feature in the ‘community power’ debate – which led to comparisons of three dimensions of power – and are explored in critical policy studies.

THREE DIMENSIONS OF POWER: WINNING KEY DECISIONS, AGENDA SETTING, AND THOUGHT CONTROL

The First Dimension: A Debate on Elitism and Pluralism Early twentieth-century studies of democracy addressed elitism based on the assumption that it is inevitable within society. Schumpeter (1942) developed an economic theory of democracy to show that the electorate could still be involved by choosing between (and influencing the policies of) elites. However, Lasswell (1936) identified a wider power structure less constrained by popular control. Power can be identified in a range of groups – the military, police, state bureaucracy, business, and professions that control the communication of knowledge – and, if combined, could contribute to authoritarian rule (Parsons, 1995: 248–50). Two books summed up elitism studies in the post-war period. Hunter’s (1953) study of power in Atlanta, USA – based on interviews asking respondents to list its most powerful people (who runs this community?) – identified a small group of elites. Mills’ (1956: 4) more theoretical study identified a US-wide male elite based on control of the ‘big corporations’, the ‘machinery of the state’, and the ‘military establishment’ (US President Eisenhower coined it the ‘military-industrial complex’). This identification of a centralized order, contrasting with the formal US system of checks and balances, confirms Lasswell’s greatest fears (Mills, 1956: 4–5). Elites at the ‘top’ have the means to control key events, while the fragmentation of ‘mass-like society’ at the ‘bottom’, combined with the ineffectiveness of ‘middling units of power’ (such as political parties), undermines the competition to influence policy outcomes (1956: 28–29, 361). The key critique of this conception of power regards its method: how do we demonstrate its accuracy empirically? Dahl (1958; see also Kaufman and Jones, 1954: 207; Polsby, 1960: 483; Wolfinger, 1960) criticizes the ‘ruling elite model’ because it does not demonstrate the exercise of power. Rather, it posits an unobservable process of covert control that is ‘virtually impossible to disprove’ (Dahl, 1958: 463; 1961: 185; Polsby, 1960: 476). The statement ‘A has more power than B’ has no meaning unless they have different

preferences and the ruling minority’s preferences are met at the expense of other groups (Dahl, 1958: 464). We must demonstrate that elites control large and powerful organizations and that inequalities in society translate into systematic advantages across the political system (1958: 465–66). Dahl’s classic statement is that, ‘A has power over B to the extent that he can [or does] get B to do something that B would not otherwise do’. To demonstrate power requires the identification of A’s resources, A’s means to exploit those resources, A’s willingness to engage in political action, the amount of power exerted (or threatened) by A, and the scope of that power, defined as effect of the action on B (Dahl, 1957: 202–203, 206; Polsby, 1960: 480). Overall, Dahl (1957: 214) points to the problems we face when we operationalize power. Dahl’s approach is to identify ‘key political choices’, involving a significant conflict of preferences and the outcomes of ‘concrete decisions’ (see also Polsby, 1960: 483–84). Using this test as part of a wider study of New Haven, USA, Dahl (1961) identifies three main processes. First, there has been a shift since the eighteenth century ‘from oligarchy to pluralism’. A ruling class based on the monopoly of elected positions by elites with high social status, education, and wealth (the ‘Notables’) has given way to elections of the ‘middle classes’, ‘ex-plebes’, and formerly ‘ethnic immigrant’ populations whose status, wealth, and education have risen (1961: 11, 32, 44). Although there are inequalities, they are ‘dispersed’ rather than ‘cumulative’: superior status and wealth no longer translates to a superior ability to control elected and unelected office (1961: 84). Although the power associated with money and status is significant, it now competes with the independent power of elected office and the vast range of political resources available to other actors, including time, esteem, support of the law, and control of information. Second, there is a democratic link – albeit imperfect – between policymakers and the electorate. Political participation, and entry into the ‘political stratum’, is highest among populations with the highest incomes, education, and social and professional standing (Dahl, 1961: 282–83). The ‘political stratum’ is small and much more active and influential in politics than the ‘apolitical strata’. Yet it is not a closed group based on class interests; it is penetrable by anyone

with the resources and motivation. Further, its members seek to build coalitions with the apolitical strata for straightforward practical reasons – to aid re-election (1961: 91–92) or avoid its wrath and potential to mobilize (1961: 310) – and cultural reasons: ‘democracy’ is a powerful idea supported by the political stratum and the apolitical strata (1961: 316–67). The political stratum acts as if apolitical strata are involved by anticipating their reactions. Further, since the political stratum is heterogeneous, most groups in the apolitical strata can find a powerful advocate (1961: 93). Third, although there are significant inequalities in politics, there is no overall control of the policy process. Although the Notables control policymaking positions in some sectors, they do not control others. Public policy is specialized. The sheer size, and fragmentation, of political systems ensures that the reach of one actor does not extend across all areas: ‘the individuals who spend time, energy and money in an attempt to influence policies in one issue-area are rather different to those who do so in another’ (1961: 273–74, 126, 169, 180; Polsby, 1960: 482; Moran, 2005: 16). Therefore, for pluralists, it is tempting to change the question from ‘Who runs this community’ to ‘Does anyone at all run this community?’ (Polsby, 1960: 476). However, we should not exaggerate their position, which is not to identify the total absence of power imbalances. In modern studies of policy networks and subsystems (Chapters 8–10), the literature supports Dahl’s argument that public policy is specialized; the size and fragmentation of political systems ensure that the reach of one actor does not extend across all areas. However, many actors are disproportionately powerful within subsystems, suggesting that the dividing lines between elitism and pluralism are blurry. Pluralism is elitism’s ‘close cousin’ (Moran; 2005: 16; see also Barry, 1980b: 350).

The ‘Second Face’ of Power Subsequent debates focused as much on a critique of pluralist methods as definitions of power. Bachrach and Baratz argue (1962: 948; against Polsby, 1960: 477) that, since there is no objective way

for pluralists to identify key decisions, we cannot demonstrate their representativeness. Therefore, the identification of pluralism in some case studies does not demonstrate pluralism overall. Indeed, the modern study of agenda setting (Chapter 9) suggests that power is about the issues we do not identify. A powerful actor may successfully focus our attention on one issue at the expense of attention to others without the need to engage or discourage action in those other areas. Since policymaking attention is limited, the agenda setting success of one group in one area may cause the failure of many groups in many others (Crenson, 1971: 25). Therefore, the pluralist focus on observable decision-making events does not take into account the second ‘face’ of power associated with the terms ‘non-decision-making’, ‘mobilization of bias’, and ‘unpolitics’. For Bachrach and Baratz (1970: 94), key decisions are not gauged by the size of the policy area or the degree of conflict, but by the extent to which a decision challenges the ‘authority of those who regularly enjoy a dominant position in the determination of policy outputs’. Actors may exercise power to protect that dominant position by restricting the attention of other actors to ‘safe’ issues: actor A devotes her ‘energies to creating or reinforcing social and political values and institutional practices that limit the scope of the political process to public consideration of only those issues which are comparatively innocuous to A’ (Bachrach and Baratz, 1962: 948). When A is successful, B does not engage in decisions that threaten A’s preferences. Therefore, issues displaying differences between A and B are not necessarily important; B may ‘win’ decisions that are innocuous to A. The less powerful face two major barriers to engagement. First, they may be disadvantaged by the dominant view within society that favours some ideologies over others. For example, most may feel that an issue is not a legitimate problem for governments to solve – such as when issues of poverty are defined as a matter of individual or family responsibility – or that the solution proposed is not worth considering (such as a ‘socialist’ solution in a capitalist society). Actors may exercise power to reinforce this view of an issue, to discourage people from thinking that it is a policy problem (Crenson,

1971: 180–81). Second, policymaking organizations and institutions may keep their grievances off the agenda. For example, a government may fill unelected posts with people committed to the status quo, or the most powerful may discourage the formal discussion of certain issues (Bachrach and Baratz, 1970: 54–59, 70; Hay, 2002: 175). Actor A may contribute to a social or institutional climate in which there is a high chance of failure and/or a fear of sanctions for challenging the status quo, contributing to B’s inability or reluctance to engage (Hay, 2002: 175 draws parallels to speaking out against the ‘local godfather’). Overall, Bachrach and Baratz (1970: 105–106) argue that the ‘dominant group’ manipulates the values of society and the procedures of government to ensure that the grievances of ‘subordinate groups’ are not aired. Political systems reinforce a ‘set of values, beliefs, rituals and procedures’ which cause an unequal distribution of ‘benefits and privileges’. Such undemocratic rule by elites meets minimal opposition because elites manipulate the decision-making process (Hindess, 1996: 5). The main problem with the ‘second face’ of power (from the perspective of pluralist method) is that some aspects may be impossible to demonstrate empirically. Bachrach and Baratz’s (1970) solution is to draw on Schattschneider’s ‘mobilization of bias’ to show that non-decision-making can be observed in, for example, agenda setting to draw attention to some issues at the expense of all others (Box 3.2). Box 3.2 The semi-sovereign people: A realist’s view of democracy Imagine two ways to consider democracy. First, politics will remain ‘broken’ until we rediscover popular democracy. Second, almost all decisions are made, necessarily, by a small number of people out of the public spotlight. How might we reconcile these arguments? Schattschneider (1960: 136) argues that political systems can run well if: (a) the government makes most decisions on behalf of the people, with minimal public involvement; and (b) there is a maximal public involvement in a small number of key decisions. In this scenario, ‘the people’ matter when they pay

attention and become mobilized. However, there are far more potential conflicts than anyone can pay attention to. The people must ignore most issues and are therefore ‘semi-sovereign’, or able to exercise their power in few areas. Further, political systems do not ensure that key issues receive the most public attention. Rather, actors exercise power to make sure that people pay attention to innocuous issues at the expense of the more important. Schattschneider (1960: 2–5) creates a thought experiment to demonstrate this argument. Think of two fighters surrounded by a massive crowd: its composition, bias towards each fighter, and willingness to engage are crucial. The outcome of conflict is determined by the extent to which the audience becomes involved. Since the audience is biased, and only a small part will become engaged, the mobilization of one part changes the balance of power. This possibility affects the strategies of participants: the ‘loser’ has the incentive to expand the scope of the conflict by encouraging a part of the audience to become involved; the ‘winner’ would prefer to isolate its opponent. Most political behaviour involves this competition to ‘socialize’ or ‘privatize’ conflict, often using widely held values such as ‘equality’ and ‘social protection’ versus ‘individualism’, ‘small government’ (1960: 7–8), or ‘this is a private matter’. The ability to privatize or socialize matters is distributed unevenly. Politics is largely the preserve of the well-educated, upper class, and business class seeking to minimize attention to their activities (1960: 30–37). Therefore, Schattschneider (1960: 12, 119) highlights the need for government to intervene: ‘Democratic government is the greatest single instrument for the socialization of conflict … big business has to be matched by … big democracy.’ The government becomes the audience to conflicts. However, there are more potential conflicts than any government can manage. Actors exercise power to determine the issues most worthy of government attention. The structures of government, such as legislative procedures controlling debate, reinforce this process by determining which conflicts receive

attention: ‘All forms of political organization have a bias in favour of the exploitation of some kinds of conflict and the suppression of others because organization is the mobilization of bias. Some issues are organized into politics while others are organized out’ (1960: 69). This ability to keep issues off the agenda means that we do not witness the exercise of power by all participants in ‘key decisions’. Crenson (1971) extended this argument in his study of air pollution policy in US cities. His three key points are that (1) post-war levels of public attention to air pollution were low, compared to the problem; (2) attention varied in different cities; and (3) while some cities passed legislation to regulate air pollution during manufacturing, others did not. In the case study of Gary (Indiana, USA), he finds evidence of non-decision-making when the Mayor delayed the study of air pollution, the study’s authors (funded by manufacturing industries) underestimated the contribution of manufacturing to air pollution and anticipated the response of US Steel while recommending weak enforcement policies, and the City Council delayed its ruling (1971: 64–65). Overall, important issues are kept off the political agenda by powerful interests who reinforce social attitudes and manipulate decision-making procedures, while the powerless pay minimal attention to an issue or feel unable to engage. Crucially, although Bachrach and Baratz (1970: 49–50) take the analysis beyond highly visible events, they share with pluralists their view on the limits to empirical analysis: one or more party has to recognize that a power struggle exists. If the researcher finds no grievance, ‘the presumption must be that there is consensus on the prevailing allocation of values, in which case non-decision making is impossible’.

The Third Dimension of Power Their statement represents the line between the second face and ‘third dimension’, which theorizes unequal power relationships

despite the appearance of consensus (Lukes, 1974, 2005). The third dimension suggests that everyone may seem to agree because B does not recognize her ‘real’ interests and A benefits from the relationship at her expense. The counterfactual is that, if given the chance, or made aware of a way to pursue her real interests, B would act differently. The difference in analysis features to some extent in Crenson’s argument that US Steel did not have to engage in non-decisionmaking. Rather, the population regulated itself, and its government promoted weak regulations, in anticipation of an unfavourable reaction to any different action (Crenson, 1971: 67–70, 122–23). Gary was a town built and kept alive by US Steel, which commanded considerable loyalty. US Steel’s strategy was to say very little in public to make sure that it did not contribute to the issue’s salience (1971: 72). This strategy, combined with the effect of US Steel’s reputation on the population’s behaviour, assured its dominance. US Steel’s inaction was emulated by Gary (1971: 78). The issue was not raised even though unregulated air pollution represents a source of profit for business and ill health for the population. By accepting the pollution unwittingly, the population was not acting in its real interests (assuming pollution control would not cause unemployment; Lukes, 2005: 48). If made aware fully of the facts, the population would act differently to promote environmental regulations. Instead, an unfavourable ‘political climate’ undermined its ability to articulate different preferences or support a new understanding of the situation (Crenson, 1971: 23). The population would not pay attention to the issue until US Steel was supportive. The third dimension of power is easier to theorize than observe. In this case, the ‘mere reputation for power, unsupported by acts of power, can be sufficient to restrict the scope of local decision making’ (1971: 125, 177). Therefore, Lukes (2005: 12, 1) warns against equating power with its exercise: ‘Power is a capacity not the exercise of that capacity (it may never be, and never need to be, exercised)’ and ‘power is at its most effective when least observable’. The behaviour of all concerned may suggest consensus rather than dominance, with no easy way to demonstrate one rather than the other. Indeed, it may be that B supports her own exploitation

while A does not believe that she is acting against B’s real interests (although Lukes’ argument seems to describe intended consequences). Yet, in some cases, the third dimension may still be visible if we can demonstrate that A manipulates B’s beliefs. This is the power to: Prevent people, to whatever degree, from having grievances by shaping their perceptions, cognitions and preferences in such a way that they accept their role in the existing order of things … the supreme exercise of power to get another or others to have the desires you want them to have – that is, to secure their compliance by controlling their thoughts and desires. (2005: 11– 12, 27, 14) The classic example of ‘false consciousness’ comes from Marxist descriptions of the exploitation of the working classes within a capitalist system: if only they knew the full facts, that capitalism worked against their real interests, they would rise up and overthrow it. They do not object because they are manipulated into thinking that capitalism is their best chance of increasing their standard of living. In effect, Lukes describes ‘hegemony’ in which the most powerful dominate state institutions and the intellectual and moral world, to decide which actions are most worthy of attention and which are right or wrong (Gramsci, 1971). Capitalist dominance is ‘based on a combination of coercion and consent’ (Hindess, 1996: 5). The ‘most effective and insidious use of power’ is to prevent conflict through the control of information and media, as well as a process of socialization (Lukes, 2005: 27).

Observing the Unobservable How do we identify the third dimension if it is so difficult to observe? For example, does Lukes satisfy the pluralist focus on demonstrable power? From a pluralist stand-point, the answer is ‘no’. Dahl (1958: 469) argues that if a ‘consensus is perpetual and unbreakable … there is no conceivable way of determining who is ruler and who is ruled’. Instead, concepts such as ‘real interests’ and ‘false class

consciousness’ represent the imposition of a theorist’s values on the research (Polsby, 1960: 479). Similarly, Polsby (1980: 97) argues that Crenson’s argument regarding ‘un-politics’ is undermined by the potentially infinite number of non-issues, which requires a choice about which non-issues are the most important; it would be inappropriate for an outsider to call an issue important when the population does not. Lukes (2005: 27) addresses the former problem by identifying third-dimensional processes in Dahl’s work, such as when the government shapes public preferences by restricting the flow of information or indoctrinates the population to ensure a widespread respect for the legitimacy of government and democracy. Crenson’s (1971: 26–28) response to the latter is to compare levels of attention to the same problem by similar populations in different cities, asking why air pollution is important in one but not the other. It predicts what a city’s population would do by establishing the actions of an equivalent city’s population in the absence of un-politics (1971: 33). This use of the comparative method is necessary to establish Dahl’s focus on ‘something that B would not otherwise do’. In other words, the counterfactual is that the population of Gary would do more about air pollution if US Steel were not so powerful. Crenson (1971: 80, 108, 182) also extends Dahl’s argument that the political stratum acts in anticipation of the reactions of other actors. If we accept that policymakers anticipate the reactions of the public, we should accept that they anticipate the reactions of ‘big business’. This selfgovernance may not be apparent in ‘key decisions’. Indeed, the pluralist focus on air pollution policy in other cities would exaggerate non-business power because it would focus only on the examples in which the issue had become politicized (1971: 131). Therefore, the focus on observable decisions may be as misleading and biased as the decision to deduce power from less visible relationships. Thus, non-pluralistic processes can be identified by extending pluralist methods, but has the burden of proof been met? The answer from Polsby is ‘no’. While Crenson’s comparative method is commendable, Polsby (1980: 214–17) argues that no data provided by Crenson demonstrates that US Steel was powerfully inactive. Rather, given the importance of US Steel to Gary’s economy, the

population decided to trade off clean air for employment. Lukes’ (2005: 48) assumption that pollution control would not cause unemployment had no basis in fact (suggesting that it may not have been in their real interests to challenge the status quo), while Jones’ (1975 in Polsby, 1980: 217) study of air pollution in Pittsburgh suggests that populations do not trade off their health for employment unwittingly. Although Lukes (2005: 148) argues that Gary could act in its ‘real interests’ by pursuing US-wide regulations, this is based on the assumption that there would be no adverse consequences.

ALL ASSESSMENTS OF POWER ARE EMPIRICAL AND NORMATIVE Empirical discussions of power are also normative. In the absence of infinite knowledge and perspective, ideological norms fill the gaps in policy studies. This argument is most clear in criticisms that the pluralist literature serves partly to legitimate the pluralist ideal by distracting us from the undemocratic role of elites (Hindess, 1996: 5; Hay, 2002: 175; but note that Dahl (1961: 3–5, 86, 330–36) goes to great lengths to identify and measure inequalities). However, we can make similar comments on bias within second face and thirddimensional discussions. For example, Crenson (1971: 180) is really arguing that power gets in the way of democracy when elites do not compete over what he thinks are relatively important issues. Or Marxists may bemoan the lack of working-class awareness of their real interests because they believe that the capitalist system exploits them. They may be right or wrong but the ‘facts’ will not adjudicate for us. Although we may be able to witness manipulation or the shaping of preferences, we cannot state with certainty if it is done to advance or thwart someone’s ‘real’ interests. In this context, most attempts to separate completely the empirical and analytical from the normative seem doomed (compare with Hay, 2002: 187). The same goes for discussions of state ‘paternalism’ when constraining the freedoms of its citizens ostensibly for their own good. Further, the government-

driven socialization of its subjects, in the promotion of ‘citizenship’, may be for legitimate reasons (to encourage participation in politics) or insidious (to foster passive consent) (Hindess, 1996: 72). The normative dimension suggests that power is about more than the ability to act. It is also about the right to act. For example, how much power is exercised with the ‘consent of those over whom it is exercised’ (Hindess, 1996: 1, 11; see also Arendt, 1986: 62)? The identification of consent, giving the capacity to exercise power over others, may be most clear with democratically elected governments. In this sense, government action notionally combines the power of all who consent (1996: 15). As Hindess (1996: 13) suggests, at the heart of such relationships is the notion of a contract in which those vested with the right to exercise power are under certain obligations not to abuse that right, in part by upholding the values of those who consent. On the other hand, since one function of government is to regulate the attitudes and behaviour of its citizens for the collective good, it produces a circular effect: consent for government action is based on government-influenced attitudes (1996: 43). Governments may also weigh up the potential trade-offs between the welfare and liberty of their citizens. In this sense, government control and legitimacy may be assumed until citizens have the ability to give ‘rational consent’ (1996: 74, 118), but the government may also determine the criteria used to gauge someone’s ability to reason. Therefore, much political theory regards the appropriateness and effectiveness of such contracts and our ability to demonstrate that consent has been given in a meaningful way, usually via representative democracy, to an organization with a recognizable unity of purpose and a clearly accountable ‘centre’. Similarly, discussions of democracy explore the extent to which other forms – such as participative, deliberative, and pluralist democracy – provide more legitimacy and a clearer link between consent and the right to exercise power.

BEYOND THE THIRD DIMENSION: FOUCAULT AND HABERMAS

Hay (2002: 170–71, 187–93; see also Hindess, 1996: 38) suggests that a debate on the extent to which power is measurable is a peculiarly Anglo-US activity. The main alternative examines whether power is so embedded in our language and practices that it is impossible to be liberated from it. Foucault (1977) presents two ways in which liberation may be impossible by drawing on the idea of society modelled on a prison. First, the power of the state to monitor and punish may reach the point at which its subjects assume they are always visible. This ‘perfection of power’ – associated with Jeremy Bentham’s ‘Panopticon’, in which it is possible for a guard to observe all prisoners from one position – renders the direct exercise of power unnecessary (1977: 201; Hay, 2002: 191). Rather, individuals accept that discipline is a fact of life and anticipate the consequences of their actions – from cheating at school and slacking off at work, to colluding in crimes and forming organizations that challenge the state – and regulate their own behaviour. Second, a broader form of control is so embedded in our psyches, knowledge, and language that it is ‘normalized’ and often rendered invisible. Consider, for example, mental health in which the ‘knowledge’ and identification of severe mental illness inevitably produce the perceived need for medical treatment. Or we ‘know’ which forms of behaviour are deviant and therefore should be regulated or punished. Therefore, power is exercised by not only the state but also communities and individuals; they reproduce and reinforce power by controlling their own behaviour and the behaviour of others. It is a ubiquitous form of social control which undermines notions of collective power or government authority granted freely by the consent of individuals (Hindess, 1996: 145). Foucault’s work contrasts with the idea that individuals are capable of giving consent based on their ability to reason and identify their own preferences (Chapter 7). Rather, individual bodies and minds represent the ultimate hub of repressive power because individuals regulate their own behaviour and suppress many expressions of their own preferences (Lukes, 2005: 91–93). As such, rationality is an ideal-type that we can use to theorize power relationships. For example, Habermas (1984) identifies an ‘ideal-

speech situation’ which ‘refers to a condition of uncoerced discussion between free and equal individuals in which … communication will be organized around the attempt to reach rationally motivated agreement’ (Hindess, 1996: 92). In that context, the use of language for communication – to act collectively to help understand the world and each other – is undermined by the use of language to dominate or coerce. It is built on the use of words to refer to traditions, norms, and rules that define the context in which any conversation takes place. Communication results ‘at best in agreement, or rather the appearance of agreement, arising from fear, deference, insecurity and other such non-rational motivations’ (1996: 93). In other words, significant power is exercised via social forces that are difficult for the researcher to detect far less measure because often the most threatening communications are made using innocuous-looking statements or commonplace language (1996: 93; compare with Fairclough, 2015). In this context, we theorize the profoundly imbalanced power relations from everyday interaction, in which some actors control others routinely, rather than in high-profile key events. Such discussions may have been precluded by the pluralist focus on method because it is not clear how you could recognize this form of power far less measure it. Still it highlights the limits to our reliance on research methods focusing entirely on observable behaviour. When individuals regulate themselves in the ‘private’ sphere, power is difficult to observe in public policy (Hay, 2002: 169). Therefore, a focus on the most observable forms of power should remind us of the story of the drunkard searching for her keys under a lamppost, not because they are there but because there is more light (Hogwood, 1992a). The most interesting and worrying forms of power – and the practices we may most want to challenge – are the most difficult to research.

POWER AND CRITICAL THEORY: THE EMANCIPATORY ROLE FOR RESEARCH

Wouldn’t it be odd if researchers became so removed from the real world that they had (or thought they had, or pretended to have) no normative positions? Donning a white lab coat and asserting a hierarchy of research methods makes us political actors, not objective scientists (Douglas, 2009; Fischer, 1998). A contrasting role for research, often described as ‘critical’ scholarship, is to: produce social change that will empower, enlighten, and emancipate … Critical theories need to offer their audience (those who are oppressed) an alternative conception of who they are, providing them with a new and radically different picture of their political, economic, and social order. Critical social science also aims to empower its audience to take action. (Schneider and Ingram, 1997: 51) A need to overthrow the status quo raises the stakes so high that you can understand why the problem of power measurement would not stop a critical scholar in her tracks. Instead, the urgent need to act provides a new perspective: issues of power measurement may seem technical from a narrowly scientific perspective, but highly political when we seek to relate them to ‘normative criteria such as social justice, democracy and empowerment’ (Fischer et al., 2015: 1). To explore these issues, let us describe examples of power in relation to gender, race, and other causes or indicators of inequality, notionally from most to least measurable. A classic observable case is the control of elected office and its impact on politics and policy. The Inter-Parliamentary Union (2018) identifies more male than female representation in 190 of 193 national parliaments. Men account for more than 70% of seats in more than 70% of cases. Such imbalances can be self-reinforcing. Male incumbency limits new opportunities for participation. It signals to women that their efforts will be relatively unrewarded, in the context of electoral campaigns in which women face misogyny routinely, including verbal assault, intimidation, and/or physical threat and assault (these imbalances also exist in positions of unelected power, such as the civil service, unions, and academia – Woodward, 2004; Curtin, 2018;

Carey et al., 2018). There is also some debate within feminist movements about the value of elected office. It can be a way to ensure the substantive representation of feminist issues, which could include violence, pornography, reproductive rights, poverty, equal pay, rights for women of colour, and LGBT rights. Or it can be a distraction from more radical ways to challenge patriarchal political systems (Evans, 2014: 148; see also Boyle, 2005). When describing non-decision-making and the privatization of issues, feminist studies identify power relationships that are reinforced in “‘personal” relationships such as child-rearing, housework and marriage and in all kinds of sexual practices including rape, prostitution, pornography, sexual harassment and sexual intercourse’ (Abbott et al., 2005: 35). These relationships are often exacerbated by other sources of inequalities – including class, race and ethnicity, sexuality, gender identity, and age – which reinforce social positions and undermine their ability to mobilize (e.g. some women may contribute to the subordination of others). Further, since power relations are multifaceted, debates on unequal power can be framed differently, from (for example) the power to be treated equally to the right to be different without fear of the consequences (Bock and James, 1992). A key link to the community power debate regards the ability to ‘socialize’ or ‘privatize’ such issues to place them on, or keep them off, the policy agenda. The phrase ‘this is a private matter’ often carries weight in the politics of families. Or socializing phrases can be repurposed cynically. For example, ‘community development’ can encourage a focus on social justice, using the state as a tool for redistribution, but also provide cover for social welfare cuts, using a focus on community autonomy to justify forms of state retrenchment, particularly during periods of state-led ‘austerity’ (Emejulu, 2016; MacLeod and Emejulu, 2014; compare with Beresford and Russo, 2016 on mental health self-management and recovery). Governments may frame such retrenchment as responsible action designed to be equally painful, but it has an unequal impact, particularly when many sources of inequality – ‘gender, race, class, sexual orientation, and national origin’ – intersect (Bassel and Emejulu, 2017: 9).

To describe third-dimensional power, Heyward (2007: 53) uses the example of a father denying his daughter education to illuminate the difference between dominance (power over someone that undermines their real interests) and potestas (power over someone with a benign effect). We must first take a normative position (does this action undermine her real interests because education should be available to all?) before deciding which form of power has been exercised. Also note the role of context, in which we are not sure if the dominant actor is reproducing traditions and norms or making strategic choices. Indeed, many forms of power relationship seem invisible because they are taken for granted or treated as natural. We use gender routinely as a way to understand our relationships with other people, and people draw routinely on norms of male and female behaviour from birth, such as when they link adjectives (strong/pretty), colours (blue/pink), and ambitions (to be an astronaut or nurse) to babies as if aspects of their lives are predetermined (Ridgeway, 2011; Ahmed, 2017). Key choices, such as to deny education, may be described as abhorrent in one context but natural in another. Or some of the most worrying forms of power may result from a selective approach to knowledge and sources of knowledge. Terms like ‘epistemic violence’ seem alien if we (a) focus on observable forms of power and (b) do not question a narrow scientific definition of knowledge. In this context, the request to prove a power relationship – ‘where is the evidence?’ – may seem reasonable. However, if we focus on dominance through language and everyday practices, epistemic violence seems like a logical extension of a profoundly important ‘critical’ argument. Examples include: •

•

privileging scientific knowledge ‘in Europe by white male scientists’ at the expense of ‘other epistemologies and other ways of representing knowledge’, for example by asserting that ‘personal and/or grounded experiences are unscientific’ (Hall and Tandon, 2017: 7; Fonow and Cook, 2005: 2213) marginalizing feminist research in mainstream social science (Lovenduski, 1998; Guerrina et al., 2018)

• •

•

erasing the voices of Women of Colour from the history of women’s activism (Emejulu, 2018) and intellectual history (Cooper, 2017) dismissing knowledge claims, such as about the impact of Western economic growth on climate change, by ‘indigenous peoples and poorer communities in a number of developing countries’ (Munshi and Kurian, 2005: 516) explaining patterns of social inequality in relation to the flaws of the unsuccessful individuals (and ‘merit’ of the successful) rather than the systemic rules and norms described by people who are excluded routinely from positions of power (Ahmed, 2017).

In this context, the request for evidence seems more like a political position designed to protect the status quo by undermining those who challenge it. To demand explanation is to demand high levels of intellectual labour from others without recognizing the cost.

ARE SUCH FORMS OF POWER ‘STRUCTURAL’? As Chapters 1 and 6 suggest, ‘structure and agency’ are at the heart of such discussions of power. The third dimension of power, and its example of the working classes unaware of their real interests, highlights the language used to describe how political structures influence behaviour. Poulantzas (1986: 146 and in Lukes, 2005: 54– 56) describes perhaps the most extreme version of this argument, defining power as ‘the capacity of a class to realize specific objective interests’. In this context, the state is an ‘objective system of regular connections’ to further the interests of a class. It is the ultimate expression of structural power if exercised on behalf of classes by individuals with no autonomy to make choices (Miliband in Lukes, 2005: 56). Although some accounts treat rules and norms as akin to physical structures (Ahmed, 2017: 30), few take it to the extreme of denying agency, while some reject this language completely. For Dowding (1991: 9–10; 2003: 306) and Lukes (2005: 57), the term ‘structural power’ makes no sense because the ‘exercise’ of power requires the exerciser to have an ability to choose how to act, and structures

cannot act. Only actors can act and be held responsible for their actions. Dowding suggests that ‘structural power’ is used to explain outcomes in the interests of certain actors (such as capitalists) when those actors do not exercise power themselves. This point is central to Crenson’s (1971: 125) argument that those with powerful reputations often enjoy favourable policy outcomes without exercising power. For Dowding (1991; compare with Barry, 1980a, 1980b, 2002), it may be better described as systematic ‘luck’ (Box 3.3). People are ‘lucky’ when they benefit from policy outcomes – without exercising power – because their interests coincide with those of someone else exercising power. It can explain why some groups appear to get more of what they want than their ‘powers’ would suggest (Dowding, 2003: 316). Box 3.3 Power and systematic luck The term ‘luck’ conjures an image of randomness and serendipity (Smith, 2009: 39; Lukes and Haglund, 2005: 49). ‘Systematic luck’ suggests that we are talking about someone who is randomly lucky a lot! Yet Barry (1980a: 184) and Dowding (1996: 71; 2009) are not describing randomness. Luck refers to someone who enjoys favourable political outcomes as the by-product of the behaviour of someone else. Systematic luck occurs ‘because of the way society is structured … Actors denoted by their social location have powers based upon their social resources, and they also have luck based upon their social location’ (1996: 71–72). For example, ‘capitalists’ benefit disproportionately from the decision by almost everyone, for their own reasons, to maintain capitalism and support economic growth rather than seek socialism. Socialist parties may know that socialism is impossible in the short term and will wreck their re-election chances, while the working classes may not want to endure decades of pain before securing long-term benefits (1996: 73). The result is more beneficial to capitalists than other actors even though they did not determine the outcome.

The identification of ‘luck’ does not preclude power. Capitalists can be lucky (enjoying outcomes caused by the actions of someone else) and powerful (exploiting their position and resources to influence outcomes). Similarly, groups can be powerless and unlucky. They are powerless when they struggle to mobilize effectively and have no effective leadership or powerful sponsor (1996: 38–40; McLean, 1987: 66–67; compare with Lukes and Haglund, 2005: 50–52). They are unlucky if they lose out when other people make decisions. However, we still need to conceptualize the socioeconomic pressures that actors (including policymakers) face; the feeling that they often seem powerless or act in an environment that is often beyond their control (Chapter 6). In this context, structural power describes situations that appear to make ‘certain acts unthinkable or physically impossible’ or ‘so costly that actors are structurally constrained from carrying them out’ (Ward, 1987: 602). This is not a million miles from Dowding’s (1996: 44; against Hay, 2004a: 51) suggestion that (a) ‘we have no choice’ really means ‘the best course of action seems obvious’ (see also Hindess, 1988: 97) and (b) Dowding’s (1991: 9) statement, ‘the power of individuals is in part determined (or rather structurally suggested) by their positions in the social structure’. There is widespread disagreement about how to describe this relationship between structure and agency, but more agreement that individuals do not act unconditionally and that some structures are more difficult to overcome than others.

WHERE DOES THE ROLE OF POWER STOP AND IDEAS BEGIN? Our discussion of power is really a discussion of power and ideas (Kettell and Cairney, 2010; Béland, 2010). Indeed, most chapters discuss their interaction. Agenda setting involves the ability of groups to frame issues to limit the number of participants in policy networks (Chapter 9). The advocacy coalition framework (ACF) highlights the shared beliefs of advocacy coalitions and their ability to establish a

dominant way to interpret policy-relevant evidence (Chapter 10). Hall (1993: 287) identifies policy paradigms, or ways of thinking about policy problems that are institutionalized or so ingrained in the psyche that they are taken for granted (Chapters 5 and 11). The literature also explores what it would take to change those arrangements; how power could be exercised to challenge existing beliefs and change the way policymakers think and act. For example, when actors exercise power, some invoke key aspects of widely shared beliefs to limit attention and participation, while others frame issues in new ways to challenge such barriers to policymaking engagement. Bachrach and Baratz’s (1970: 54–59, 98) first barrier to engagement is the dominant set of beliefs held within society, which can endure but be overcome. Their study of the politics of poverty in Baltimore suggests that these barriers were overcome over the longer term, as the previously excluded Black population became increasingly powerful, buoyed by anti-poverty groups and more able, over time, to identify, become aware of, and influence ‘arenas of conflict’. Further, air pollution now receives much more attention. Indeed, Crenson (1971: 69–71, 79) suggests that Gary shifted its policy first by learning lessons and transferring policy from Allegheny Pennsylvania, aided by federal government regulations that reduced the ability of large companies to play cities off against each other. However, many discussions of power and ideas seem more difficult to resolve. Luke’s third dimension focuses on what people believe to be their real interests and the extent to which those perceptions can be manipulated. Foucault’s social control is based on common knowledge of normality and deviant behaviour. The exercise of power through routine practice involves the correspondence between everyday language and firmly held beliefs with a profound impact. When processing policy issues, the identification of extreme failures or ‘shocks’ to the political system (such as a crisis or change in government) may prompt policymakers to engage in a fundamental rethink of their beliefs and seek out participants with new ideas. Or the ideas and policies adopted by other governments may prompt policymakers to learn lessons and transfer policy (Chapter 12). However, it is more difficult to identify such an

immediate and profound shift in attitudes to certain people, particularly when policymakers believe in, or exploit, social stereotypes to maintain inequalities (Chapter 4).

CONCLUSION Power has a wide variety of meanings, including the capacity for action: the ability to get what you want, affect the behaviour of others, and alter the decision-making environment. In turn, this ability can be related to inequalities, or the relative powers of individuals, social groupings, and institutions. It also refers to the exercise of power that, in some cases, is visible and measurable. This focus on measurement underpins key debates in the literature, including the community power debate. While the modern elitist position suggests that power can be inferred from reputations and the possession of powerful positions in society, the pluralist critique is that power must be demonstrated and observed. Pluralist empirical studies of ‘key issues’ suggest that, although power is dispersed unequally throughout society, it does not translate into overall control of the policy process. Rather, the control of elected and unelected office is diffuse and the sheer size and fragmentation of the state ensure that the reach of one powerful actor does not extend across all policy areas. Critics of the pluralist position argue that it ignores the role of nondecision-making, in which the powerless are prevented from engaging in policymaking, and the mobilization of bias, in which some issues are ‘organized out’ of the policymaking process. Therefore, a focus on observable decision-making, in which one actor wins at the expense of another, ignores the extent to which power is exercised, less visibly, to keep issues off the political agenda. This argument is extended in the ‘third-dimensional’ account of power in which potential conflicts are minimized following the manipulation of people’s beliefs. Although policy areas may appear to be consensual (suggesting that we cannot observe winners and losers in ‘key decisions’), this is because some actors do not recognize their ‘real interests’ and act accordingly.

Other accounts go further, to argue that power relations are even less observable because they are manifest in self-regulation, everyday language, and assessments of knowledge claims. We can use critical theory to (a) identify and explain the use of language to dominate, dismiss, and maintain fear and (b) seek ways to challenge such power relations to address major systematic inequalities in politics and policy. In that context, critical scholars may suggest that to insist on the measurement of such forms of power is to dismiss the argument and downplay the necessarily normative role of research in politics. Such debates suggest that the most important and interesting issues of power may also be the most difficult to research and demonstrate. Although we ‘know’ such power exists we do not agree on how to identify and theorize it. Therefore, what starts as a dry and technical issue of methodology becomes an issue that goes to the heart of normative debate on who possesses, and should possess, the means to exercise power. They also highlight the role of structure and agency. The exercise or effect of power relates to the resources of individuals, but they do not operate in a vacuum. They are surrounded by an audience more important than themselves. They engage with institutions and systems – representing the rules, beliefs, and actions of many others – that amplify or undermine the impact of their actions. Indeed, ideas often resemble structures, as the dominant ideologies that restrict debate and the rules of policymaking that limit popular participation and expression of meaningful consent. As the remaining chapters suggest, theories of public policy address these issues of method and evidence well. However, in analysing individual approaches, it is easy to lose sight of the issues raised when we focus directly on themes such as power and inequality. Studies of key actors – such as ‘policy entrepreneurs’ – may connect their success too much to individual skill rather than the nature of the policymaking environment that gives more opportunities to some people. Or studies of systems may focus more on complexity, unpredictability, and minimal central control than the endurance of major inequalities within complex systems. Therefore, the more critical reader should note the specific language of each

policy theory and the power-based context in which we can understand them as a whole. Elitism – power is concentrated in the hands of a small number of people or organizations that control policy outcomes. Pluralism – power is more diffuse; policy results from the competition between actors. Oligarchy – the control of government or society by elites. Methodology – the study of methods used to gather knowledge (based on epistemology, or a theory of what knowledge is and how we create it). Behaviouralism has two broad claims: ‘(a) observable behaviour … should be the focus of analysis; and (b) any explanation of that behaviour should be susceptible to empirical testing’ (Sanders, 2010: 23; 2017). Operationalize – to turn abstract concepts into observable and measurable units. Counterfactual – a statement used to explore what would have happened under different circumstances. Ideal-type – an analytical construct that serves as a comparator and point of departure for ‘real world’ descriptions of events and behaviour (compare with the definition in Chapter 1). Epistemic violence – the act of dismissing an individual, social group, or population by undermining the value of their knowledge or claim to knowledge. Spivak (1988) relates it primarily to the acts of the colonial West to subjugate colonized populations, with reference to the ‘subaltern’ (someone of low social status, oppressed or excluded from society) (see also Rutazibwa and Shilliam, 2018). Actors – entities such as individuals, groups, and governments with the means to deliberate and make choices. Ideas – shared beliefs or ways of thinking (Chapter 11).

4 Bounded Rationality and the Psychology of Policymaking Key themes of this chapter: • • • • • • •

‘Bounded rationality’ describes the limits to gathering and processing policy-relevant information. It suggests that ‘evidence-based policymaking’ is a misleading ideal. Policymakers need to find ways to ignore most evidence. They use cognitive shortcuts to gather enough evidence and make what they believe to be ‘good enough’ choices. Classic debates examined if bounded rationality would, and should, produce incrementalism in political systems. Modern theories focus more on ‘fast thinking’ and its potential to produce radical policy change. The Narrative Policy Framework (NPF) describes actors telling stories to set the agenda and influence policy change. Social Construction and Policy Design (SCPD) shows how policymakers use emotion and social stereotypes to reward some populations and punish others.

INTRODUCTION: THE PROFOUND IMPORTANCE OF BOUNDED RATIONALITY Chapter 3 suggests that some actors are powerful because they can exploit the ways in which other people think. Therefore, studies of power link strongly to studies in psychology about the cognitive ‘biases’ on which all people rely. In a complex world beyond our understanding, we all have to rely on efficient ways of thinking, ignoring almost all information available to us, to allow us to make choices decisively. Cognitive shortcuts can be described provocatively as ‘rational’, when we seek systematic methods to

identify the best information and ‘irrational’ when we rely on gut instinct and emotion to act almost instantly (Cairney and Kwiatkowski, 2017). However, the use of emotion and morality to take care of one’s loved ones without weighing up the costs and benefits is perhaps the most reasonable action of all (Gigerenzer, 2001)! A more relevant worry is that cognitive shortcuts leave us vulnerable to error and manipulation. Policymakers are no exception to this problem. To understand policymaking, policy studies compare the classic ideal-type, comprehensive rationality, with bounded rationality in the real world. Comprehensive rationality is an unrealistic simplification of reality that we can use to explore what really happens. For example, it is not realistic to expect policymakers to (1) possess the organizational and cognitive capacity to gather and process all information relevant to their decisions and then (2) make clear, consistent, and well-ranked choices, in (3) a policymaking system over which they have full knowledge and control. Rather, they face bounded rationality, in which their possession and grasp of evidence, and their ability to make and implement consistent policy choices, and know and control their effects is limited. They have to rely on cognitive and organizational shortcuts. Even though it is an impossibility, comprehensive rationality also seems to be an ideal to which many people aspire. Many advocates of ‘evidence-based policymaking’ seem to criticize the need for policymakers to ignore most information when they make decisions. Or, they display a passionate objection to any ‘irrational’ means to do so (Cairney, 2016a)! More broadly, comprehensive rationality sums up a common-sense view of how democracies should operate: elected policymakers employ reason to identify problems and clarify their aims, produce and process evidence comprehensively to assess the societal costs and benefits of solutions, use this evidence to make and legitimize their choices, and evaluate policy success after a comprehensive search for evidence (Cairney, 2019a). This double-ideal is what makes policymaker psychology and the debate on ‘incrementalism’ (Box 4.2) so interesting. It prompts us to think about the effect of bounded rationality on system-wide policymaking, and what we should do about it. Classic accounts

describe different ways to adjust to bounded rationality. Simon focuses on the need for policymaking organizations to ‘satisfice’, combining systematic searches for information with efficient procedural shortcuts to reach an acceptable proximity to the comprehensive ideal. However, when discussing the nature of policymaking systems, Lindblom appears to promote the practical and normative value of departing from the ideal of comprehensive rationality. Incrementalism suggests that a good strategy for boundedly rational policymakers with limited policymaking resources is to make a succession of incremental changes to public policy based on the lessons of past decisions. In a pluralist political system, it makes sense to limit information searches only to politically feasible options, learn from trial and error, and not depart radically from decisions based on a previous consensus (compare with Chapter 3 on pluralism). Modern policy theories add two profound insights. First, many theories link major policy change to bounded rationality (or they identify inertia that is not caused by pluralism). For example, punctuated equilibrium theory identifies incremental change and rapid and profound shifts of policy (Chapter 9). Second, theories identify new normative issues that arise from the use and abuse of cognitive shortcuts. In this chapter, the Narrative Policy Framework (NPF) identifies the ability of actors to tell stories to influence their audience. Further, Social Construction and Policy Design (SCPD) shows how policymakers draw on their own biases, and exploit social stereotypes, to make quick choices that reward some populations and punish others. Such responses to bounded rationality produce manipulative politics and ‘degenerative’ political systems rather than consensus-driven incrementalism. Overall, modern theories suggest that advances in science and information technology have not solved – and cannot solve – the problem of bounded rationality. Instead, politics and power will always matter.

COMPREHENSIVE AND BOUNDED RATIONALITY A focus on comprehensive rationality prompts us to question our assumptions about the power of the ‘centre’ to cause policy change:

• •

Do individual actors at the ‘top’ have the ability to research and articulate a series of consistent policy aims, and then make sure that they are carried out? Can an organization act in the same manner as a ‘rational’ individual?

Comprehensive rationality suggests that elected policymakers seek to translate their values into policy, aided by organizations that operate in a ‘logical, reasoned and neutral way’ (John, 1998: 33). The model includes a series of assumptions: 1. Organizations can separate values (required by policymakers to identify their aims) from facts (required by organizations to assess the best way to achieve those aims). 2. Organizations and policymakers can produce consistent policy preferences, and rank them, to help maximize societal gain (in the same way that an individual ranks preferences to help ‘maximize utility’). 3. Policy is made in a linear fashion, as in a policy cycle. First, policy aims are identified in terms of the values of the policymaker. Second, all means to achieve those aims are identified. Finally, the best means are selected. There are clearcut stages to the process – such as between agenda setting (identifying aims), formulation (identifying choices and making decisions), and implementation (carrying them out). 4. Analysis of the decision-making context is comprehensive. All relevant factors and possibilities have been explored, and all theories regarding how the policy process works have been considered (Lindblom, 1959; Simon, 1976; Jordan and Richardson, 1987: 9–10; John, 1998: 33; Hill, 2005: 146). Since this is an idealized version of the policy process, the assumptions are deliberately unrealistic. Most approaches begin their analysis by identifying the limitations to comprehensive rationality and the presence of bounded rationality. First, it is impossible to separate facts from values in such an artificial way. Simon (1983: 8) argued that the best way to demonstrate this point

is to read the alleged ‘facts’ in Hitler’s Mein Kampf (see also Brinkmann, 2009; Douglas, 2009; Etzioni, 1967: 386). More generally, the cause of policy problems is always subject to interpretation and debate, and there is no objective way to evaluate the success of a policy solution (McConnell, 2010). Second, policymakers have multiple, and often unclear, objectives which are difficult to rank in any meaningful way. Therefore, they tend to pursue a small number of those aims that command their attention at any one time. Policy aims can also be contradictory: choosing a policy to address aim A may mean undermining a policy to address aim B, producing clear winners and losers from the policy process. These problems are multiplied when extended to organizations. Governments contain a mass of organizations pursuing policy aims relatively independently of each other, with little regard to the idea of centralized and ranked preferences or the trade-offs in policy choices (there is no ‘holistic’ or ‘joined-up government’). Third, the policy process is not linear and it is difficult to separate the policy cycle into discrete stages. The ‘garbage can’ model and ‘multiple streams’ analysis (Chapter 11) present the most significant departure from an assumption of linearity, suggesting that three processes – problem definition, solution, choice – appear to act almost independently, with the potential for a completely different order. Policymaking often begins with solutions that ‘chase problems’. Finally, policymakers are faced with incomplete knowledge of the policy environment and the likely consequences of their solutions, and cognitive and time constraints limit their ability to consider and understand every possible solution. Organizations do not have the capacity to consider every fact and solution; the cost of research forces them to set priorities. The search for theories to explain policy problems is limited by the values and emotions of policymakers (which predispose them to consider only some solutions) and organizational rules. Consequently, Simon (1976: xxviii) presents a more realistic view of policymakers and policymaking organizations based on bounded rationality:

1. Individuals and organizations cannot ‘maximize’ their utility; instead, they ‘satisfice’, or seek ‘a course of action that is satisfactory or “good enough”’. 2. They have neither the ability nor the inclination to consider all facts. Instead, they combine intense searches for information with efficient approximations to focus on the factors considered most relevant and important. Simon expressed the aim to, as far as possible, use ‘administrative science’ (or, in Lasswell’s language, the ‘policy sciences’) to reduce the problem of bounded rationality within organizations by: training officials in policy analysis to open their mind to new possibilities; fostering the development of specialization and expertise in information processing; teaching officials the most appropriate informational shortcuts to make organizations more effective; and, supporting the long-term goals of an organization by providing the right incentives (1976: 242–43; 1960). His term ‘satisfice’ sums up the need to be pragmatic when facing limits to the cognitive power of individuals and coordinative capacity of organizations (Hill, 2005: 146–47; Parsons, 1995: 273–84).

Bounded Rationality Is More Important Than Ever We might be tempted – wrongly – to think that bounded rationality is less of a problem now than when first identified in the policy sciences (Cairney and Weible, 2017). Our ‘evidence-based age’, involving major advances in scientific research methods and information technology, has given ‘a new breath of life’ to hopes for comprehensive rationality and evidence-based policymaking (Botterill and Hindmoor, 2012: 367). Yet, when we revisit the four conditions for comprehensive rationality, we see that the accumulation of new evidence does not come close to solving the problem of bounded rationality (Cairney, 2016a: 14). It addresses the problem of gathering policy-relevant knowledge somewhat (point 4 above). However, the most pressing policy issues are characterized by ‘radically uncertainty’; their complexity makes them difficult to define and manage, partly

because so many aspects are interdependent, producing too many possible outcomes to predict (Tuckett and Nicolic, 2017). Further, more evidence does not help us adjudicate between unclear preferences (point 2) or simplify the policy process in which they are considered (point 3). Indeed, Chapter 11 suggests that the garbage can model is as relevant as ever. Modern science remains value-laden (point 1) even when so many people employ so many systematic methods to increase the replicability of research and reduce the reliance of evidence on individual scientists. The role of values is fundamental. Anyone engaging in research uses professional and personal values and beliefs to decide which research methods are the best; generate research questions, concepts and measures; evaluate the impact and policy relevance of the results; decide which issues are important problems; and assess the relative weight of ‘the evidence’ on policy effectiveness (Cairney, 2016a: 16; 2019a). We cannot simply focus on ‘what works’ to solve a problem without considering how we used our values to identify a problem in the first place (Botterill and Hindmoor, 2012). It is also impossible in practice to separate two choices: (1) how to gather the best evidence and (2) whether to centralize or localize policymaking (see Cairney, 2017b; Cairney and Oliver, 2017). Most importantly, as Chapter 3 suggests, the assertion that ‘my knowledge claim is superior to yours’ symbolizes one of the most worrying exercises of power. We may decide to favour some forms of evidence over others, but the choice is value-laden and political rather than objective and innocuous (compare Douglas, 2009; Pielke, 2007; Jasanoff, 2008). Further, our growing knowledge of psychology has reinforced the importance of bounded rationality and challenged a lot of the midtwentieth-century optimism about the prospect of comprehensive searches for information: ‘People are “cognitive misers” (Kam, 2005), using informational shortcuts and heuristics to gather just enough information to make decisions’ (Cairney and Kwiatkowski, 2017: 2; Cairney and Weible, 2017; Vis, 2019). Modern studies of psychology suggest that:

People use short cuts to gather enough information to make decisions quickly: the ‘rational’, by pursuing clear goals and prioritising certain kinds of information, and the ‘irrational’, by drawing on emotions, gut feelings, values, beliefs, habits, and the familiar, to make decisions quickly. (Cairney and Kwiatkowski, 2017: 2) All people – including policymakers – address bounded rationality by thinking ‘fast and slow’: ‘System 1 operates automatically and quickly, with little or no effort and no sense of voluntary control. System 2 allocates attention to the effortful mental activities that demand it, including complex computations’ (Kahneman, 2012: 20). Haidt (2001: 818; 2007, 2012) distinguishes between the ‘intuitive system’ and ‘reasoning system’ to explain ‘moral reasoning’ as a retrospective process. In theory, moral reasoning could represent an archetype of ‘rational’ judgement to determine what is right or wrong in a society. However, in practice, people often engage with policy issues using gut instinct and ‘when faced with a social demand for a verbal justification, one becomes a lawyer trying to build a case rather than a judge searching for the truth’ (2001: 814). Box 4.1 identifies how people limit their exposure to information before making choices. Box 4.1 Cognitive biases and fast and frugal heuristics Kahneman (2012) describes a series of cognitive shortcuts, many of which feature in policy theories (Lewis, 2013: 7; Cairney and Kwiatkowski, 2017): •

Cognitive biases are at the heart of agenda setting, attention, and the competition to resolve policy ambiguity: ‘framing effects’, based on emotional and moral judgements; the ‘representativeness heuristic’, when people overestimate the probability of events because of their vivid nature; and the ‘availability heuristic’ (or ‘processing fluency’), when people relate the size, frequency, or probability of a problem to how easy it is to remember or imagine (Chapters 9 and 11).

•

• • •

• •

‘Prospect theory’ describes people valuing the losses they fear more than the equivalent gains they might receive. It helps explain why advocacy coalitions think that their competitors are more powerful than they are (Chapter 10). A ‘need for coherence’, to identify patterns and causal relationships, helps explain the power of narratives with a simple hero and moral (Jones et al., 2014). Policymakers use exemplars of social groups to represent general experience and describe why they reward and punish populations (Schneider et al., 2014). Status quo bias, the ‘sunk costs fallacy’, and ‘optimism bias’, in which unrealistic expectations about our plans working out well when we commit to them, all help explain inertia and the endurance of institutions (Chapter 5). ‘Groupthink’ and other aspects of organizational psychology place crucial limits on comprehensive searches for policyrelevant information. Some issues may heighten public and policymaker anxiety, contributing to a sense of urgency or panic, which limits information searches dramatically (Pepin-Neff, 2019).

However, we are still scratching the surface of the links between psychology and policymaking (and indeed physiology and policymaking). For example, we know that ‘emotion and cognition are part of the same internal mental process’ but we have not considered how organizations might use this insight to modify policy processes (Cairney and Kwiatkowski, 2017: 3). Nor do we agree on the positive or negative aspects of psychology. While it is customary to describe cognitive ‘biases’ negatively, with reference to the comprehensive ideal, Gigerenzer (2001: 37–38) provides a more positive description of ‘how actual humans … make decisions, as opposed to heavenly beings being equipped with practically unlimited time, knowledge, memory, and other unlimited resources’. ‘Fast and frugal heuristics’ are the ‘computationally cheap’ methods people use to make choices. An ‘adaptive toolbox’ helps people ‘to experiment using trial and error, use emotions to limit needless searches for new choices (such as

considering the costs/benefits of keeping one’s children), and make choices based on a small number of simple rules rather than trying in vain to weigh all costs and benefits’ (Cairney and Kwiatkowski, 2017: 3; see also Gigerenzer, 2015; Tuckett et al., 2015; Durnová, 2018). One source of rules may be a core set of moral values in a political system to which policymakers can refer. Such well-thought-out values are more difficult to describe as ‘irrational’ compared to action driven by fire and fury. Or, we may find that emotions are crucial to political mobilization, such as to maintain campaigns to challenge injustice (Reger, 2004). Further, the application of studies of psychology and cognition to policymaking systems suggests that governments struggle to process policy-relevant information proportionately (Koski and Workman, 2018). Policymakers set goals but ‘they are not generally effective in judging the connections between [their] goals and the complex reality they face’ (Jones and Thomas, 2017: 49). Individuals communicate their narrow expertise within a system of which they have almost no knowledge (Sloman and Fernbach, 2017). In any situation, ‘most members of the system are not paying attention to most issues most of the time’ (Baumgartner, 2017: 72). This scarcity of attention helps explain, for example, why policymakers ignore most issues in the absence of a ‘focusing event’ (Shaffer, 2017), organizational searches for information miss key elements routinely (Workman et al., 2017), organizations fail to respond to events or changing circumstances proportionately (Epp, 2017), and the ‘centre’ will always lack coordinative capacity and the ability to make policy from the top-down (Cairney et al., 2019a). This description of policymaking must inform prescription: seeking comprehensive rationality is like trying to fly unaided rather than designing and using a plane (Lindblom, 1964).

INCREMENTALISM We want to know how policymaking systems operate – and should operate – when responding to bounded rationality. In that context,

incrementalism is a key attempt to consolidate empirical and normative aspects (Box 4.2). It describes the need to respond to bounded rationality pragmatically, by (1) gathering information efficiently (note the opportunity cost of research) and (2) situating this activity within a real policymaking environment. When we focus on comprehensive rationality and evidence-based policymaking we should not forget that choice is inherently political; it is about winners and losers. A policymaker has to consider their values, the balance of power within the legislature, and the reaction to policy change by interest groups. Policymaking is costly: it takes time and political will to persuade political parties, vested interests, and the public that major policy change is appropriate (and to ensure that policy is implemented). It is also unpredictable: policymakers often react to events or solve problems caused by previous policies rather than devote their time to major new initiatives. Consequently, boundedly rational policymakers are much more likely to introduce incremental policy changes – based on learning from experience and addressing the unintended consequences of previous decisions – than introducing radically new initiatives. Box 4.2 The meaning of incrementalism The meaning of incrementalism is often unclear in three main ways. First, ‘incremental’ is often equated with small. However, Lindblom (1979: 517) argues that increments can be large or small. The crucial distinction is between radical and non-radical change: does it follow logically from existing policy or mark a significant departure? This argument solves one problem but raises another: can we distinguish between radical and nonradical change? Second, incrementalism can be a description of how policy is made or a prescription of strategies to pursue (or a confusing mix of both). Third, incrementalism can refer to analytical strategies, in which organizations try to overcome problems related to comprehensive rationality:

•

• •

Simple incremental analysis. Analysis limited to a small number of policy choices that diverge incrementally from the status quo; it is better to analyse those issues comprehensively than seek comprehensive coverage of all issues. Disjointed incrementalism. The simplifying strategies used by organizations, including simple incremental analysis, trial and error, and ‘parallel processing’. Strategic analysis. Realistic policymaking strategies (including disjointed incrementalism and ‘mixed scanning’) used as an alternative to the ‘futile attempt at superhuman comprehensiveness’ (Lindblom, 1959: 88).

Or, it can relate to strategies to reach agreement. ‘Partisan mutual adjustment’ is a process in which actors pursue their interests and respond to other actors doing the same (by researching their positions, using persuasion, and seeking allies) (Lindblom, 1959: 85). Policymakers do not begin by articulating their values, translating them into policy aims in rank order, and seeking the best means to achieve them. Their willingness to trade-off one aim for another only becomes clear when they make decisions. The analysis is not comprehensive, considering all of the empirical and theoretical implications (Lindblom, 1965). Organizations analyse the effects of incremental change and ignore many important possible outcomes, alternative policies, theories, and values. Indeed, it is efficient for government organizations to spend most of their time focusing on the effects of incremental departures from current policies. Taking the comprehensive ideal seriously would ‘paralyse’ an organization by frustrating its staff and exhausting its resources on many radical options that are rarely acceptable to major political parties (Lindblom, 1964: 157; Etzioni, 1967: 386). It is also sensible for policy change to take place through a series of steps, to reduce the chance of making ‘serious lasting mistakes’. The effects of non-incremental decisions are unpredictable and difficult to solve (1959: 86).

Further, any test of ‘good’ policy shifts from the ability to satisfy a comprehensively rational policymaker’s objectives to whether or not it commands agreement within the political system (Lindblom, 1959: 81). An incremental strategy works in that context: if a previous policy commanded widespread respect, then policymakers recognize the analytical and political costs of a radical departure from it. To change policy radically may be undemocratic. In a pluralistic system, existing policy is based on a long-term process of negotiation, bargaining, and adjustment. No single actor commands the policy process as a whole (Lindblom’s comparator to the USA was Soviet command-and-control). Rather, multiple actors and agencies pursue their interests and command their own powerful ‘watch-dogs’ to anticipate or redress ‘damages done by other agencies’ (1959: 85). It encourages ‘partisan mutual adjustment’ (Box 4.2). These suggestions prompted one of the most significant post-war public policy debates in the USA (see Cairney, 2012a: 100–4). Lindblom’s critics suggested that incrementalism represents a more realistic model than comprehensive rationality, but has less to offer as a policymaking ideal. First, it relies on pluralism or the assumption of an equal balance of power to produce widely supported policy. In unequal societies, non-radical change preserves policies produced by elites (Dror, 1964; Etzioni, 1967: 387). Lindblom (1977; 1979) accepts this problem, but argues that ‘partisan mutual adjustment’ may be a preferable response: ‘strong central authority can be – and historically is, in case after case – an instrument for protecting historically inherited inequalities’ (1979: 523). Indeed, actors may use the appearance of a comprehensively rational process to minimize public, parliamentary, and pressure group attention to inequalities. Putting power in the hands of the few does not guarantee that they will use it wisely and in the spirit of benevolent neutrality. Second, debates focused on how to manage more radical change for the long term in ‘fundamental’ policy areas, by encouraging policymakers to plan their activities more efficiently, clarify their values, and identify alternative policies and issues worthy of comprehensive analysis (Dror, 1964: 156; Lindblom, 1979). Lindblom’s proposal – ‘one can make changes in the social structure

as rapidly through a series of incremental steps as through drastic – hence less frequent – alterations’ (1964: 157) – helped distinguish incrementalism from inertia (Lindblom, 1979: 520). It also signalled a greater role for central planning to ensure a consistent series of incremental steps towards a clear goal (Etzioni, 1967: 387–88; Cairney, 2007a). However, it did not help us make a clear distinction between incremental and radical change. Since the nature of policy problems is open to interpretation and debate, there is no objective way to identify ‘fundamental’ decisions (Smith and May, 1993: 203; Hill, 2005: 151). The incremental/radical policy change distinction seems obvious in theory, but where is the dividing line in practice?

Is Incrementalism ‘Universal’ and Inevitable? Incrementalism was based initially on a study of US politics, so how relevant is it to policy-making in other countries? When making such comparisons, it is important to avoid confusing the proposed advantages of incrementalism (the management of policy by consensus and the minimization of unintended consequences) with forms of inertia made more likely by veto points in particular political systems (Chapter 6). Hayes (2001) suggests that incrementalist strategies are used in many political systems characterized by bargaining and compromise between actors who have different information, interests, and views; and the need to build on past policies (Hayes, 2001: 3). Indeed, the identification of incrementalism as a routine ‘policy style’ – in majoritarian and consensus democracies – was a key theme of Richardson’s (1982b) Policy Styles in Western Europe. Incrementalism transcends formal political structures because it results from routine responses to bounded rationality (see Chapter 5 on ‘network institutionalism’ and Chapter 9 on ‘policy communities’). Consider, for example, how we might explain an incrementalist style in a system, such as the UK, which enjoys a reputation for ‘majoritarian’ and ‘top-down’ government (Cairney, 2019b). Most policy decisions are beyond the reach of elected policymakers such as UK ministers. The sheer size of government necessitates breaking policy down into more manageable issues involving a

smaller number of interested and knowledgeable participants. Therefore, most public policy is conducted primarily through specialist ‘policy communities’ which process ‘technical’ issues at a level of government not particularly visible to the public or Parliament, and with minimal ministerial involvement. These arrangements exist because there is a logic to devolving decisions and consulting with certain affected interests. Ministers rely on their officials for information and advice. For specialist issues, those officials rely on specialist organizations. Organizations trade information, advice, and other resources (such as the ability to implement or ‘deliver’ a large group membership) for access to, and influence within, government. Further, the logic of this relationship holds regardless of the party of government, and we are unlikely to witness radical policy change simply because there is a change of UK government (Richardson and Jordan, 1979; Jordan and Richardson, 1982; Jordan and Maloney, 1997; Jordan, 2005; Cairney, 2008; Jordan and Cairney, 2013). When comparing political systems, commentators are often expressing unrealistic expectations about how quickly policy can change, then complaining about unusually high inertia compared to countries like the UK (Burns, 1963; Krauss and Pekkanen, 2004; Fabbrini and Gilbert, 2000: 28; Kitschelt and Streeck, 2003: 2). The implication is that a majoritarian political system puts power in the hands of the few and gives them more opportunity to pursue the comprehensive rationality ideal. Yet, as Chapters 5 and 6 describe, inertia is not caused solely by veto. Rather, the effect of decades of cumulative policies is that newly elected policymakers inherit a huge government with massive commitments (Rose, 1990; 1986; Rose and Davies, 1994). The size and scope of the state is such that any ‘new’ policy is likely to be a minor revision of an old one, partly because policy builds up its own clientele resistant to change (Hogwood and Peters, 1982; 1983; Geva-May, 2004). Or, when a commitment to a policy has been established and resources devoted to it, it becomes increasingly costly to choose a different path (Pierson, 2000a). In other words, the abstract theme with which incrementalism engages – the lack of comprehensive rationality and its impact on

political systems – is universal (it applies to all times and places). Further, incrementalism as a policy style can be found in many systems, and the reasons for policy inertia are not restricted to systems with many veto points. In general, all policymakers face bounded rationality and a policymaking environment over which they have limited control. The differences associated with individual political systems relate more to the specific ways in which actors make sense of this problem. In some cases, policymakers appreciate the benefits of incrementalism. In others, they seek quicker, more radical change. However, modern theories suggest that incrementalism as a policy style is not inevitable. Bounded rationality is universal, but it often helps to explain two forms of activity that we should not confuse with incrementalism. 1. Policy obstruction driven by fast thinking Policy inertia often relates to the exercise of power to reinforce ‘fast’ responses to bounded rationality. Actors rely on ‘quick gut, instinct, emotional, and moral choices’ and become ‘biased’ or ‘motivated’ reasoners devoted to the status quo despite evidence that might support policy change (Cairney and Kwiatkowski, 2017: 3). For example, Chapter 10 describes the ‘advocacy coalitions’ who romanticize their own cause, demonize their opponents, and often try to obstruct policy change (Sabatier et al., 1987). 2. Incrementalism co-existing with sudden and profound policy change Many theories highlight the links between bounded rationality and non-incremental change. Punctuated equilibrium theory shows that political systems produce ‘hyper-incremental’ and radical change. Boundedly rational policymakers have limited resources (including time, knowledge, and attention) and cannot deal with all policy problems. So they ignore most and promote a few to the top of their agenda. This lack of attention to issues helps explain why most policies do not change dramatically, while intense periods of attention to some issues may prompt new demands for

change. By ‘reframing’ issues, policy actors can draw the attention of policymakers to new ways of seeing and solving problems. Successful persuasion prompts ‘positive feedback’, in which policymakers pay disproportionate attention to the issue, and a ‘bandwagon effect’, in which multiple actors in the policy process all suddenly pay attention to, and seek to influence, the same issue (Chapter 9; Jones, 1999, 2003). Multiple streams analysis (MSA) suggests that radical policy change happens when there is a ‘window of opportunity’ for three processes: a new or reframed problem gains attention, a feasible solution gains currency within policy networks, and policymakers have the motive and opportunity to translate the idea into policy. MSA predicts policy continuity and profound change (Chapter 11; Kingdon, 1984). Studies of policy diffusion demonstrate that non-incremental change can happen when one region emulates another on the assumption that it was successful. The policy transfer literature suggests that some governments feel pressure to engage in nonincremental policy change, often driven by anxiety that they are not keeping up with their neighbours (Chapter 12; Berry and Berry, 2007; Dolowitz and Marsh, 2000). In such cases of policy change, we can detect the role of cognitive misers driven by fast thinking rather than comprehensive analysis. People can understand policy problems in many different ways, but focus only on one, with the potential for their attention to lurch dramatically following a disproportionate response to new information. A key strategy is to encourage people to pay attention to policy problems and potential solutions in a particular order (Box 4.3). Box 4.3 Heresthetic and the manipulation of choice William Riker (1982; 1986) uses the term ‘heresthetic’ to describe the importance of a particular kind of manipulation: ‘one can help produce a particular choice if one can determine the context of, or order in which people make, choices’. In other words, we may conclude from studies of psychology that we cannot change

people’s minds when they engage in fast thinking to entrench their positions. However, we can exploit the fact that they possess many different – and often contradictory – preferences. If so, the aim is to draw their full attention to one issue or preference at the expense of another. Additional strategies may involve exploiting their cognitive miserliness to make sure that they do not engage in a full enough analysis to compare their preferences systematically. For example, one can present information in ways that increase the cognitive load and dissuade people from analysis, such as by presenting abstract concepts or excessive detail (Drew, 2019). Put simply, if you want to make something happen, it may be better to influence the rules that people use to make decisions – such as the order in which we consider each choice, or the ways in which we present limited choices – than change their minds about their preferences. This strategy includes the restriction of options during ‘the selection of … policies to which there are majority preferred alternatives’ (Riker, 1982; Hindmoor, 2006a: 80–87; Ward, 2002: 66; McLean, 2002; Ward and Weale, 2010). Many policy preferences are ‘intransitive’ or inconsistent: if A is preferred to B and B to C, A is not necessarily preferred to C (Chapter 7, p.117). Therefore, the order of the vote is crucial. This approach seems to contrast markedly with Aristotle’s logos, ethos, pathos approach: there is minimal appeal to pathos to persuade, or ethos to follow rules on policy debate. Instead, heresthetic is manipulation ‘concerned with the strategy value of sentences’ (Riker, 1986: x). Similarly, it may contrast with modern trends in ‘evidence-based policy’ to communicate evidence as simply as possible. Herestheticians may exacerbate the cognitive burden of choice to get what they want.

THE NARRATIVE POLICY FRAMEWORK There is far more to bounded rationality than a technical response in which governments seek ways to gather and process more evidence more effectively (Jones and Crow, 2017). Rather, all people – including policymakers – ‘develop quick and easy emotional

renderings of the world that fit with who they think they are and what they know’ (Crow and Jones, 2018: 217). If so, the shortcuts they use to limit their attention – and the small number of issues to which they pay attention – really matter. Our focus on agenda setting and ‘framing’ (Chapters 3 and 9) would be irrelevant if everyone had the willingness and ability to understand and compare all possible frames of reference! Instead, their attention to one way of seeing the world, at the expense of another, can prompt them to support radically different policies than if they had full information. However, we can draw two very different lessons from this starting point: 1. People may be highly vulnerable to manipulation if they rely on other actors to help them understand the world. 2. People are resistant to a lot of manipulation because they do not pay attention to or trust many sources of information. Put more simply, the stories that we tell people can reinforce or compete with the stories they tell to themselves (Tuckett and Nicolic, 2017). The same story can motivate some audiences, if it chimes with their beliefs or pulls their heartstrings, but backfire with others, if it grates with their view of the world (or if they feel they are being manipulated with stories). In that context, the NPF identifies the strategies of actors to exploit other actors’ cognitive biases (described in Box 4.1). Then, it tries to measure their success empirically (Box 4.4). Box 4.4 Narrative versus frames Framing and narrative are not the same. This point is confusing because its authors did not always make the distinction (Shanahan et al., 2018a: 204), and both terms are ambiguous and overlap greatly (Weible and Schlager, 2014: 242; see also Lakoff, 2014). In policy studies, framing is the definition of a policy’s image; how issues are portrayed and categorized (Chapter 9). Yet ‘framing’ is also a looser metaphor to describe the ways in which we understand and use language selectively. Many disciplines describe the metaphor differently, but in two key ways. First, akin

to a photo. The photographer decides what you see (where the lens points, how much to zoom or Photoshop). Framing describes our reliance on someone else to describe reality. Second, akin to a building’s timber frame, providing essential structure. It describes less visible assumptions (Chapter 3). Combined, framing sounds like storytelling. We limit information to influence how people understand the world, the issues to which they pay attention, and how they understand policy problems. The difference comes from the way in which the NPF (1) describes narrative to (2) conduct – generally quantitative – empirical research to test the effects of strategies (Shanahan et al., 2013; 2018b). To count as narrative , it must have a structure akin to a literary narrative: setting, character, plot, moral. In contrast, framing analysis could quantify the frequency of keywords rather than whole phrases (although narratives can contain few words (Merry, 2016) and NPF studies can use framing analysis (Lawlor and Crow, 2018)!). People use narratives to manipulate or persuade their audience to see a policy problem and its solution in a particular way. Such strategies, and the context in which they appear, can vary markedly in ‘rich and unique policy contexts’, but a narrative has the same basic structure containing four key elements (Shanahan et al., 2018a: 176–77): 1. Setting. It relates to a specific policymaking context consisting of ‘policy phenomena such as legal and constitutional parameters, geography, evidence, economic conditions, norms’. The nature of some settings may be taken for granted, while others are highly contested. 2. Characters. It contains at least one actor, such as a hero, villain, beneficiary, or victim. 3. Plot. A plot establishes a timeline or story arc. Common arcs include stories of heroes going on a journey or facing and overcoming adversity, often relating to villains causing trouble and sympathetic victims suffering tragedy.

4. Moral. A story’s simple take-home point describes the cause of, and hence solution to, the policy problem. For example, we must protect the lives of victims of mass shootings or the rights of gun owners to defend themselves (Merry, 2018). From this starting point, we can identify different strategies and factors that cause the relative success of a narrative (Shanahan et al., 2018a: 177–78; 2014: 70). Strategies to foster support from groups or coalitions include describing policy issues as private or in need of government intervention (see Box 3.2 (p 43) on Schattschneider); establishing the cause of a problem by (for example) assigning blame to a character; or romanticizing the work of your own coalition while demonizing others (see Chapter 10). In each case, such strategies are constrained but not determined by the firmly held beliefs of the narrator’s audience (Chapters 10 and 11). NPF empirical studies suggest that combinations of the following techniques or factors seem to improve narrative success: • • • • •

using an audience’s fundamental belief (e.g. in the value of the market) to influence a more malleable belief (e.g. business leaders will solve climate change) (Box 4.3); generating admiration for a hero in favour of policy change or empathy for a victim of policy failure/inaction; helping an audience imagine a problem and relate it to their own lives; exploiting trust in the narrator; and making sure that ‘micro level’ strategies (aimed at individuals) and ‘meso level’ strategies (aimed at coalitions) relate to ‘macro level’ or ‘grand narratives’ which capture fundamental societal beliefs, such as in the relative merits of capitalism/ socialism or freedom/security ( Shanahan et al. 2018a: 188–95; Jones, 2014).

The latter strategy is crucial. It suggests that you need to know your audiences to know how to influence it, and that this influence is bounded by context. In some cases, there are only brief ‘windows of opportunity’ in which a narrative will have the desired impact (Chapter 11). This window can relate to (1) the emotional state of the

audience or (2) a series of events in political systems (Cairney and Kwiatkowski, 2017: 5). In other cases, narrative success relates to the attitudes of left- and right-wing populations. Some audiences are relatively open to government action to solve problems, while others only accept arguments that emphasize individual responsibility. For example, the Frameworks Institute experimented with a personal story of a woman to highlight the difficulties of eating healthily as a family when poor, only to find that one audience engaged emotionally and sympathized, while another judged the victim harshly and assumed they could do better (Cairney, 2016b). Jones and Song (2014) also find that, in climate change debates, the stories with the biggest short-term impact – ‘egalitarian’ (it is caused by overconsumption) and ‘individualist’ (it is exaggerated by idealists and the self-interested) – affected the audiences predisposed to accept them (compare with Fadlallah et al.’s (2019) evaluation of narrative in health policy). The potential impact of NPF insights is not undermined by the fact that some findings seem obvious. There is a big difference between assuming that some stories matter and measuring their effect over the short and long term (Box 4.5). Further, it may not seem important that stories have most impact when telling people what they already think, but it could make the difference between thought and action, such as when people turn out to vote or pay attention to one problem at the expense of the rest. We may struggle to persuade people to change their minds, but we can encourage them to act by prioritizing their attention to one belief over another (Box 4.3; Jones and Crow, 2017: 3). Box 4.5 NPF positivism vs. interpretivist postpositivism To count as an NPF study, research should be ‘clear enough to be wrong’ (Jones and McBeth, 2010; Jones et al., 2014: 3–4). Its ‘positivist’ approach to science is driven by empiricism, hypothesis testing, generalizing findings, and replicating research (Sabatier, 2007b: 326–31; Eller and Krutz, 2009: 1; Cairney, 2013a: 10). It has provoked debate with ‘postpositivist’ scholars of narrative (Jones and Radaelli, 2015, 2016; Lejano, 2015; compare with

‘critical’ scholarship in Chapter 3, p.50). NPF authors hope that many people can conduct their empirical studies systematically to accumulate knowledge (Pierce et al., 2014), unlike some other theories which lack coherence and coordination (Cairney and Jones, 2016: 50). It requires replicability: two different people could use the same method and get the same results (Chapter 13). Its critics suggest that the policymaking context is too complex to boil down into such a basic description of narrative, and that researchers are subject to the same cognitive biases that the NPF ascribes to policy actors. Therefore, ‘researchers cannot escape the world they inhabit to make sense from some external position’; they tell one of many possible truths in each study rather than accumulate knowledge of the truth over time (Dodge, 2015: 362; compare with Jones, 2018). Yet NPF studies may not always provoke such intense debate or require such a strict approach. Gray and Jones’ (2016: 199) call for more qualitative NPF studies takes us from ‘clear enough to be wrong’ to consistent enough to seem credible to a critical reader, partly by reflecting continuously and openly about the application of NPF concepts (Jones, in correspondence, 2018). Some NPF articles encourage scientists to focus more on their narrative and less on the alleged superiority of their knowledge (Jones and Crow, 2017; Crow and Jones, 2018; Jones and Peterson, 2017). Further, McBeth’s (2014) story is that the NPF’s architects began as postmodernists ground down by the US positivist academic system. Its expansion outside of the USA (McBeth et al., 2018a: 198) changes the audience and maybe the story.

SOCIAL CONSTRUCTION AND POLICY DESIGN SCPD identifies the profoundly ‘degenerative’ effect of narratives and other ways to exploit people’s beliefs to make policy. Policymakers describe and use their value judgements to make fundamental choices about which social groups should be treated positively or negatively by government. When addressing highly politicized issues, they seek to reward ‘good’ groups with

government support and punish ‘bad’ groups with sanctions (Schneider et al., 2014). In psychology, such ‘moral reasoning’ (Haidt, 2001) is ‘motivated reasoning’, in which policymakers make quick, biased, emotional judgements, then seek to back up their actions with selective facts: Likes and dislikes are not the result of individual or collective reason and deliberation but mainly the product of emotion and heuristics … judgments begin with emotional reactions … and reason is used mainly to justify initial emotion responses. (Schneider et al., 2014: 106) Yet social constructions can also be based on conscious bias and the strategic exploitation of social stereotypes and other people’s emotions for political gain. Policies reflect the goal-driven use of constructions ‘strategically manipulated for political gain … to create political opportunities and avoid political risks’ or, at least, an anxiety by politicians ‘not to be caught in opposition to prevailing values’ if it affects their performance in elections (Schneider and Ingram, 1997: 6; 192). They seek support from the populations they describe as ‘deserving’, as well as a wider public satisfied with describing others as ‘undeserving’ (1997: 6). These judgements have an enduring ‘feed-forward’ effect (Ingram et al., 2007: 112; feed-forward means feedback – Schneider and Ingram, 2019: 230). Choices based on values are reproduced in ‘policy designs’: Policy designs are observable phenomena found in statutes, administrative guidelines, court decrees, programs, and even the practices and procedures of street level bureaucrats … [they] contain specific observable elements such as target populations (the recipients of policy benefits or burdens), goals or problems to be solved (the values to be distributed), rules (that guide or constrain action), rationales (that explain or legitimate the policy), and assumptions (logical connections that tie the other elements together). (Schneider and Ingram, 1997: 2)

Examples of positive feed-forward effects include policy designs that signal that ‘elderly citizens are worthy of respect and deserving of the funds they receive’, and prompt ‘a level of political participation rivaled by no other group’ (Schneider and Sidney, 2009: 110). Negative actions include signalling to welfare recipients that they have themselves to blame and deserve minimal support; and restricting voting rights directly or introducing convoluted rules to diminish participation (2009: 111). Policy designs based on moral choices often become routine and questioned rarely in government because they are ‘automatic rather than thought through’ (Schneider et al., 2014: 124). Emotional assignments of ‘deservingness’ act as important ‘decision heuristics’ because this process is ‘easy to use and recall and hard to change’ (2014: 124). In this respect, SCPD builds on classic discussions of actors exercising power to reinforce or challenge attitudes (Chapter 3; Pierce et al., 2014). Policy designs are difficult to overcome because a sequence of previous policies, based on a particular framing of target populations, helps produce ‘hegemony’: the public, media, and/or policymakers take this set of values for granted, as normal or natural, and rarely question them when engaging in politics (Pierce et al., 2014). For example, if most people assume that people in poverty deserve little government help, because they are largely responsible for their own fate, policymakers have little incentive to intervene. In such cases, powerlessness relates to the inability of disadvantaged groups to persuade the public, media, and/or government that there is a reason to make policy or a problem to be solved. Or, people may take for granted that criminals should be punished because they engage in deviant behaviour. Since policy design sends a powerful signal to the recipients of benefits or punishments, target populations participate more or less according to how they are characterized by government (Schneider and Ingram, 1993: 334). To challenge policy designs, groups often have to challenge fundamental public assumptions, reinforced by government policy, regarding what constitutes normal and deviant behaviour. Yet only some have the resources to mobilize to influence the way they are described by policymakers (Schneider and Ingram, 1997: 21–24; 2005b: 444; Pierce et al., 2014), or to persuade the

public, media, and/or government that there is a reason to act on their behalf. Social constructions are – in theory – ‘inherently unstable’ but few groups are categorized differently over time, at least in the absence of long-term change in social attitudes (Ingram and Schneider, 2005a: 10). For example, it can follow a major external event such as an economic crisis or game-changing election, exploited by ‘moral entrepreneurs’ to influence the way that policymakers view particular groups (Ingram and Schneider, 2005: 10–11; Nicholson-Crotty and Meier, 2005). Or, it can be prompted by policy designs which suit powerful populations but produce spillover effects for the relatively powerless. For example, a shift to drug treatment as an alternative to incarceration, designed for white people, benefits people of colour unintentionally (Schneider and Ingram, 2005a: 639). Ingram et al. (2007: 102) depict this dynamic with a table in which there are two spectrums (Table 4.1). One describes the positive or negative ways in which groups are portrayed by policymakers. The other describes the resources available to groups to challenge or reinforce that image. There are four categories of target population: the advantaged are treated positively and are powerful, so they receive benefits publicly. Deviants are their polar opposite. Dependents are treated positively in public but do not have the power to exploit their image, while contenders are treated negatively in public but powerful enough to secure benefits behind the scenes. Most issues are not politicized in this way because people can only pay attention to a small number of issues (Chapter 9). Yet low salience can exacerbate problems of citizen exclusion. Policies dominated by bureaucratic interests often alienate citizens receiving services (Schneider and Ingram, 1997: 79). Or, the pursuit of ‘evidence-based policymaking’ (Box 2.2) helps experts dominate policies, and many government agencies, when there is high scientific agreement and wide acceptance that the ‘public interest’ is served largely through the production and use of evidence. The process does not include ordinary citizens routinely. Rather, ‘experts with scientific credentials aid and abet the disappearance of the public sphere’ (1997: 153). Issues ‘with important social value implications’ transform into ‘a matter of elite scientific and

professional concern’, such as when official calculations of economic activity override personal experiences (1997: 153, 167). Overall, SCPD describes a political system with major potential to diminish democracy. Politicians politicize issues to reward or punish populations or depoliticize issues with reference to science and objectivity. Policy designs are not informed routinely by citizen participation. Schneider and Ingram (1997: 3) argue that, although the US political system may ‘meet some standard of fairness or openness’, the policies they produce may not be ‘conducive to democracy’. They describe an increasingly individualistic US system with declining rates of collective political participation (in elections), a tendency for actors to seek benefits for their own populations, and ‘degenerative’ policy which produces major inequalities along sex, race, and ethnicity lines (Ingram and Schneider, 2005: 22–26). For example, Soss (2005: 294–95) describes a ‘two-tier’ social security system to help some and humiliate others (compare with Herd and Moynihan, 2018). Public policies have failed to solve major problems – including inequality, poverty, crime, racism, sexism, and unequal access to effective healthcare and education – and policy failure contributes to the sense that the political process serves special interests at the expense of the general public (Schneider and Ingram, 1997: 4–7). Policy designs ‘are strongly implicated in the current crisis of democracy’ (1997: 5–6), because they have failed, and they discourage many target populations (the ‘undeserving’, ‘deviant’, or ‘demons’) from public participation: Table 4.1 Social construction and power

High power Low power

Described positively

Described negatively

advantaged, treated positively in public and receiving benefits publicly

contenders, treated negatively in public but negotiating benefits privately

dependents, treated positively in public but unable to

deviants, treated negatively in public

mobilize to negotiate and punished by benefits policy Source: adapted from Schneider and Ingram (1997).

These designs send messages, teach lessons, and allocate values that exacerbate injustice, trivialize citizenship, fail to solve problems, and undermine institutional cultures that might be more supportive of democratic designs. (1997: 192)

Is Social Construction and Policy Design ‘Universal’ and Inevitable? SCPD began as a study of US politics, and there are few direct international comparisons. Still, we can see that US-style ‘degenerative’ politics is more or less apparent in other political systems. The UK contains comparable processes. Hunter and Nixon’s (1999: 166) study of housing policy identifies the same ‘advantaged’ status for homeowners as in the USA. Media, court judges, and politicians express sympathy for owner-occupiers affected by a system over which they had minimal control, but describe disadvantaged social housing tenants as feckless individuals. In contrast, in their study of poverty policies in two provinces in Canada, Mondou and Montpetit (2010) argue that the worst excesses of degenerative politics are associated with adversarial contexts. Consensual policy styles, associated with proportional electoral systems and multi-party politics, produce less incentive to reward advantaged and demonize deviant groups. Such applications help us reflect on the ways in which abstract concepts have ‘universal’ relevance but very different results in different contexts. In the abstract, key SCPD processes are likely to be found anywhere: policymakers combine cognition and emotion to define highly salient problems and populations, their choices are reflected in policy design, and policy design sends signals to target populations, and only some populations can mobilize to exploit or challenge their image and seek government rewards. Further, the

tendency to process most other policies out of the public spotlight is a key feature of political systems (Chapter 9). However, different political systems and policymaking environments constrain and facilitate such behaviour, producing different results. Compared to the NPF and most other theories in this book, we may be relatively likely to find SCPD-style applications, in studies of other countries, which do not cite the foundational authors (Schneider and Ingram). For example, a large literature on the social construction of ‘problem’ families in the UK has direct parallels with SCPD and its depiction of degenerative US politics. Politicians have long blamed parents or an ‘underclass’ for family breakdown or dysfunction and its relationship to social problems such as school truancy, anti-social behaviour, crime, and low employability (Crossley, 2017; Cairney and St Denny, 2020: 188–94). This overlap of study, but not citation, partly reflects SCPD’s intellectual connection to ‘critical’ scholarship (Chapter 3) which is relatively diverse and does not require the same reference to a foundational text (compare with Chapter 11 on Kingdon).

CONCLUSION Comprehensive rationality is an ideal-type. The task is to identify its assumptions and consider the implications when these conditions are not met. Policy theories build on the concept of bounded rationality, which highlights the inability of organizations to separate facts from values; unclear and conflicting political objectives; a nonlinear decision-making process; and an incomplete search for knowledge combined with limited resources, time, and cognitive abilities. From this starting point, we can explore what happens (description) and what should happen (prescription). For example, policymakers ‘satisfice’ and use informational and cognitive shortcuts to make their task more manageable. While Simon uses this discussion to seek improvements to policymaking, Lindblom extends the analysis to a political context in which policymakers bargain with other actors and treat high levels of agreement in the political system as an indicator of good policy. Lindblom advocates this approach because it reduces the chances

of governments making big mistakes that are relatively difficult to reverse, and ensures that they do not use resources – including political will – unwisely. The argument that incrementalism does and should happen produced a US-specific debate focused on unequal power and inertia. Yet key themes are universal. The use of incrementalist strategies to address the limits to comprehensive rationality is common to many political systems. Most political systems face some need for bargaining and compromise, and to build on past policies. More generally, a focus on comprehensive rationality helps us focus on the big questions of political science: if comprehensive rationality cannot be achieved, what happens and what should happen? Should we spread analytical resources in an attempt to cover most areas, or focus those resources on the issues most likely to receive policymaker attention? How do we identify the issues to which everyone should pay most attention versus those we leave to the experts? Modern theories still focus on bounded rationality, but have taken forward two profoundly important insights. First, when policymakers respond to bounded rationality, it produces a combination of incremental and radical policy change. Second, all people – including policymakers and influencers – combine cognition and emotion to make decisions. Both points undermine the idea that we can get closer to the comprehensive ideal summed up in mythical discussions of ‘evidence-based policymaking’, or that policymaking is simply incremental. Incrementalism may seem appropriate to actors who want to be able to reverse mistakes more easily, establish ‘stable expectations in a complex and uncertain environment’, and address the spirit of compromise, particularly in political systems which have a formal separation of powers and ‘overlapping, conflicting and interacting institutions’ (Jones and Baumgartner, 2005: 119). Non-radical policy change may also be the norm if all policy actors must ignore almost all policy issues almost all of the time. Yet the desire by some for consensus-driven policy does not ensure that governments will act that way. Emotionally or ideologically driven governments may still make radical policy

choices even if their ability to produce a consistent and wellresearched plan is out of reach (Hill, 2005: 152). Radical change may follow a window of opportunity, disproportionate attention to one issue at the expense of others, or international emulation and coercion to transfer policy, all of which suggest that bounded rationality does not necessarily cause policymaking stability, policy continuity, or incremental change. Further, the NPF suggests that, although people are relatively susceptible to stories that reinforce their beliefs, an effective narrative can mobilize policymakers to act to protect or challenge the status quo. More worryingly, the SCPD equates policy inertia with a combination of emotional policymaking and profound inequalities of power. Lindblom’s image of pluralistic and consensus-driven incrementalism compares with Schneider and Ingram’s elitist and degenerative politics. If so, we can no longer fool ourselves that policymakers routinely produce good policymaking when dealing with bounded rationality. Comprehensive rationality – an ideal-type of policymaking in which policymakers translate their values and aims into consistent policy choices following a comprehensive study of all choices and their effects. Bounded rationality – a more realistic model which identifies the factors – such as uncertain aims and limited information – that undermine comprehensive rationality. Opportunity cost – the value of option B, which is foregone when we pursue option A. For example, the cost of a decision to study many policy options is the foregone chance to study fewer, more realistic, options in more depth.

5 Institutions and New Institutionalism Key themes of this chapter: • • • • • •

Institutions are formal and informal rules that influence political behaviour. Formal rules are usually written and visible, such as the ‘structures of government’ that vary across policymaking systems. Informal rules are unwritten, less visible, and profoundly important but difficult to detect. ‘New institutionalism’ is the study of ‘institutions’, but both terms are difficult to define. We know that institutions matter but often struggle to define institutions. Instead, we focus on the different messages provided by key variants of institutional studies: historical, rational choice, normative, constructivist, feminist, empirical, network. Almost all approaches associate institutions with rules or practices, not specific organizations (a key exception is punctuated equilibrium theory, or PET, in Chapter 9).

INTRODUCTION: INSTITUTIONS MATTER, BUT WHAT ARE INSTITUTIONS? ‘New institutionalism’ identifies the rules, norms, practices, and relationships that influence patterns of behaviour in politics and policymaking. ‘Institution’ once described policymaking organizations such as legislatures, courts, and executives (Judge, 2005: 2). Now, it describes the formal and informal rules that guide action. Institutions are not the buildings or arenas within which people make policy. They are the rules of behaviour that influence how they make it. Formal rules of political systems include the US federal model and its constitution which sets out the rules governing the separation of

powers between the executive, legislative, and judicial branches and guarantees a degree of sub-national policymaking autonomy. We may compare these rules with those of other political systems, such as the ‘Westminster model’ which has an executive, within the legislature, and less automatic sub-national autonomy. Or, a focus on particular formal rules may allow comparisons of policymaking ‘styles’. For example, Lijphart (1999) differentiates between the styles in ‘majoritarian’ systems with first-past-the-post electoral rules and ‘consensus’ democracies which use proportional representation. Informal rules are more difficult to define, identify, and study. Broadly speaking, they are the norms of behaviour that we find in many different settings, from societal norms on gender roles to the expectations about policy goals that are taken for granted in specific organizations. Although we write and speak about them, rules are often unwritten, unspoken, and difficult to unearth. They ‘exist in the mind’ of people who may know what they are but struggle to describe their meaning and effect (Ostrom, 2007: 23). As in the discussion of power in Chapter 3, the impact of some rules is measurable, while others rely on theory and interpretation. Our aim is to identify patterns of behaviour, then relate them to the relationships and shared understandings that policy actors develop. Institutionalism studies are tricky because no one is entirely sure what an institution is and what new institutionalism means. Lowndes (2010: 65; 2018) lists nine approaches, which suggests that it represents an umbrella term. Studies may be held together by the belief that institutions are the ‘central component of political life’ and that ‘they matter more than anything else that could be used to explain political decisions’ (Peters, 2005: 164). However, we may have to work hard to make sure that we are talking about institutions in comparable ways. There are two main ways to do so. First, describe the key messages associated with each variant. Second, identify the ‘causal mechanisms’ of institutions: how exactly do they cause patterns of behaviour – such as via socialization or enforcement – and why do institutions change? Key messages include (1) many institutions are akin to structures which endure for many decades (historical); (2) they provide incentives or constraints on action (rational choice); and

(3) they are communicated via socialization (normative). Some approaches seek to go beyond these three variants to add new insights: rules and norms are communicated via the spread of ‘ideas’ (Chapter 11), which are more flexible than structures (discursive/constructivist); and, institutions generally make it easier for men to engage in politics and policymaking (feminist). Other approaches are better understood in terms of specific debates, such as when empirical institutionalism identifies formal rules as the cause of different policy styles, while network institutionalism emphasizes common informal rules and policy styles across many political systems. Overall, we can agree quite easily that institutions matter but have to work harder to say how and why.

IDENTIFYING FORMAL AND INFORMAL INSTITUTIONS Newton and van Deth (2010: 9) describe formal institutions as ‘structures of government’. For example, constitutions represent a ‘set of fundamental laws that determines the central institutions and offices, and powers and duties of the state’ (2010: 71). In this context, the list of rules and associated organizations is long, and the potential for different policy outcomes in different political systems appears to be considerable: •

•

Confederal and federal. Confederal arrangements exist when countries give some powers to supranational bodies to work in their common interests but retain significant autonomy. Federal suggests a greater integration of territories, with common institutions more likely to have enforcement powers (2010: 107– 9). Federal and unitary. In federal systems, there is a balance of power between the executive, legislature, and judiciary, and between central and territorial government. The latter have powers and rights guaranteed in a written constitution. In unitary systems, the central government controls, and can reform or abolish, territorial government (2010: 109). These distinctions are

•

•

•

•

blurred by a trend towards centralization in federal states and the ‘quasi-federal’ status of some unitary systems that guarantee some autonomy to regions (2010: 114–15). Presidential, parliamentary, and semi-presidential. In presidential systems, an elected president is head of government and state. Many systems are based on the US model in which the president heads the executive and can initiate but not control the progress of legislation. In parliamentary systems, the head of government is a prime minister or chancellor. For example, in the ‘Westminster model’ the executive forms part of, and is accountable to, the legislature. Most legislation is initiated and controlled by the governing party which tends to have a majority under a plurality voting system. In semi-presidential systems, there is a fusion of executive and legislature but the prime minister is chosen by an elected president. Presidential and semi-presidential systems are the most likely to distribute power and are often associated with inertia and ‘deadlock’, with parliamentary systems associated with strength or effectiveness (2010: 92–98; Weaver and Rockman, 1993: 2; compare with Chapter 4). Unicameral and bicameral legislatures. While many federal states (Australia, Belgium, Germany, Switzerland, the USA) have a ‘strong’ second chamber, most are ‘weak’, with limited powers to initiate and amend legislation or control finance (2010: 79; compare with Galligan, 2006: 272–73). The role of the judiciary. Most states have judicial review, but some also have constitutional courts (e.g. the EU, South Africa) and others have particularly active judiciaries (the USA, Germany) (2010: 79–81). Direct democracy. Countries such as the USA and Switzerland have procedures for holding referendums regularly, prompted by popular local initiatives (Budge, 2006).

There is also a series of rules embedded within such political systems, including:

•

•

•

•

Electoral systems. In first-past-the-post systems, the candidate with a plurality (not necessarily majority) of votes in a constituency is elected. In proportional systems there is often more than one vote cast (or candidates are ranked in order of preference) to ensure a closer link between the share of votes and seats that political parties gain (Newton and van Deth, 2010: 248–49; Karp and Banducci, 2008). Party systems. The number of political parties is influenced by the number of ‘cleavages’ within society, such as between those representing labour or business, and ethnic and religious populations. Party systems are influenced heavily by electoral systems, particularly under plurality which exaggerates the results for the two largest parties at the expense of small parties (which are better represented under proportional representation). Rules of government formation and executive–legislative relations. The need for minority or coalition government is more prevalent under multi-party proportional representation systems (Newton and van Deth, 2010: 272–77). It combines with different rules in different countries regarding, for example, how long parties are given to negotiate government formation, how susceptible they are to motions of no confidence, and the extent to which parties can have policy influence in opposition (Strøm, 1990; Shugart, 2006). Group–government relations. ‘Corporatism’ refers to a close relationship between government and groups representing labour and business, in which ‘umbrella’ organizations are integrated within government policymaking structures. This is joint decisionmaking, with the results often implemented by groups (e.g. by trade unions enforcing wage agreements on members). ‘Tripartism’ is a less formal arrangement. ‘Pluralism’ describes groups competing with each other to influence government. In this case, informal ‘rules of the game’ are more important than formalized arrangements. Groups are ‘para-governmental’ when they receive money from governments to deliver public services (Newton and van Deth, 2010: 213–17; see also Martin and Swank, 2004; Martin and Thelen, 2007).

•

Structures of public bureaucracies. Bureaucracies are subject to rules to ensure that they are powerful enough to be efficient but also accountable and not too powerful or autonomous. This balancing act inspires continuous civil service reform, including ‘new public management’ based on applying private business methods to government (Kettl, 2006).

These formal rules are crucial but should be qualified in three ways. First, they largely describe variations across economically developed and Western liberal democracies, which tend to be the focus of most policy studies. Political systems which do not conduct regular freeand-fair elections, and the policymaking systems of low- and middleincome countries, have only recently become subject to significant empirical study using the theories described in this book (Chapter 13). Second, semi-formal rules have resulted from policy decisions ‘institutionalized’ in some way. Economic institutions refers to the system of state country-level rules governing the operation of economic organizations. For example, there may be a US (adversarial) or UK (self-regulation and oversight) model for government agencies to regulate large corporations. There may also be Anglo-American, Rhineland, and East Asian models of capitalism, each of which involve different relationships between the state and the market (Moran, 2006). Regulatory state refers to the collection of rules governing economic organizations and interest groups, agencies, and other organizations involved in the delivery of public services (Braithwaite, 2006). Welfare states represent a complex system of rules governing the provision of social security and public services. Third, our analysis is complete only when we consider informal rules, or shared understandings and standard operating procedures. For example, intergovernmental relations within federal and devolved systems involve formal dispute resolution mechanisms (intergovernmental committees, the courts) and informal contacts between political parties, civil servants, and individuals (Horgan, 2004; Watts, 2007). In some systems, including the UK, the norm is to avoid formal dispute resolution (Cairney, 2012c). While federal

systems are ‘cooperative’ (the centre and states share power and must coordinate policy efforts) or ‘dual’ (powers are separated), in practice most are administratively and financially interdependent and cooperate in similar ways (Newton and Van Deth, 2010: 114; Galligan, 2006: 274). Box 5.1 Descriptions of institutions and new institutionalism A major confusion exists between scholars who use the term to refer to an organizational entity such as the US Congress, a business firm, a political party, or a family, and scholars who use the term to refer to the rules, norms and strategies adopted by individuals operating within or across organizations. (Ostrom, 2007: 23) The formal rules, compliance procedures, and standard operating procedures that structure conflict. (Hall in Thelen and Steinmo, 1992: 2) Most of the major actors in modern economic and political systems are formal organizations, and the institutions of law and bureaucracy occupy a dominant role in contemporary life. (March and Olsen, 1984: 734) Building blocks of social order … organizing behaviour into predictable and reliable patterns. (Streek and Thelen, 2005: 9) Humanly devised constraints that shape human interaction. (North in Sanders, 2006: 42) Institutions … can be interpreted as reflecting habits and norms, more likely to be evolved than to be created. But institutions also may be seen as architecture and as rules that determine opportunities and incentives for behaviour. (Rhodes et al., 2006: xiii) Discursive institutionalism is an umbrella concept for the vast range of works in political science that take account of the substantive content of ideas and the interactive processes by

which ideas are conveyed and exchanged through discourse. (Schmidt, 2010: 3) The majority of new institutionalist research is gender blind, failing to consider issues of gender and rarely drawing on feminist political science, despite a wide-ranging body of feminist work that is primarily concerned with political institutions. (Kenny, 2007: 95)

WHAT EXACTLY IS AN INSTITUTION? WHAT IS INSTITUTIONALISM? This need to focus on formal, semi-formal, and informal rules really complicates the definition of institution. Rhodes et al. (2006) identify many things which count as institutions: the state, civil society, economic institutions, constitutions, federal and territorial institutions, executives, legislatures, courts, bicameral structures, public bureaucracies, the welfare state, regulations, local government, political parties, electoral systems, direct democracy, and international and non-governmental institutions. It is more difficult to move from this ‘I know it when I see it’ approach to listing institutions to define an institution and decide how to study it (Thelen and Steinmo, 1992: 29, n. 9; Peters, 2005: 74). Rhodes et al. (2006: xiii) suggest that there is no ‘singular definition of an institution on which students of politics can find wide agreement’. Dowding and King (1995: 10) also warn against ‘trivially true’ definitions that make it difficult to find something that is not an institution. If ‘institution’ means everything, then it means nothing (Rothstein in Lowndes, 2010: 72). As Box 5.1 suggests, in most definitions the focus is on regular patterns of behaviour and the rules and norms that influence relationships and behaviour (with the key exception of PET in Chapter 9). Yet we face problems when we move from studying regular patterns of behaviour towards defining rules, identifying which rules are the most important, and assessing the extent to which rules are followed and enforced, or if most processes are characterized by shared understandings and agreements about how

people should act. We need a language to describe institutions as influences on behaviour rather than forms of behaviour in their own right. One solution is to follow Hall’s (1986; 1993) idea of ‘standard operating procedures’ which are recognized by participants and ‘can be described and explained to the researcher’ (Lowndes, 2010: 73; Ostrom, 2007: 23). Another is to use Ostrom’s (2007: 23) distinction between rules, norms, and strategies as part of the Institutional Analysis and Development Framework (see Chapter 7 and Box 7.3). Most discussions tread a fine line between saying that rules influence or determine the behaviour of individuals (Thelen and Steinmo, 1992: 3), but there is no agreement on how to balance the two. For example, March and Olsen (2006a: 3) suggest that an institution can be treated effectively as a structure because rules often endure in the same basic form regardless of the individuals involved. Alternatively, Rhodes (2006a: 103) argues that ‘Even when an institution maintains similar routines while personnel change, it does so mainly because the successive personnel pass on similar beliefs and preferences’. In this case, a rule becomes akin to a language: it survives if there are enough individuals committed to its survival. Or, it is a ‘tradition’ – a ‘set of understandings someone receives during socialization’ (Rhodes, 2006a: 91; compare with Moyson et al., 2018). This difference in language highlights the profound philosophical debates that often divide the discipline (Box 5.2). For example, Bevir (2013: 5) argues that the use of the term ‘institution’ rather than ‘practice’ misleads because it suggests a real, fixed entity or set of rules that we can all point to and understand in the same way; and exaggerates the extent to which we can say that rules determine behaviour. Yet it is not inevitable that institutions endure over time, and we may question the extent to which they represent shared ideas. Instead, they may be reproduced in different ways by individuals who understand rules differently and act differently (Bevir, 2009). This makes the identification of institutions really tricky! They ‘exist in the minds of the participants and sometimes are shared as implicit knowledge rather than in an explicit and written form’ (Ostrom, 2007: 23). Further, the rules followed implicitly within organizations may contradict the rules described explicitly in written

statements (2007: 23). Therefore, while we can perhaps agree that institutions represent sets of rules and norms, this may be where the agreement ends: ‘the range of theoretical approaches underlying the contemporary study of institutions is remarkably diverse, let alone the range of empirical and methodological orientations’ (Rhodes et al., 2006: xiii). At best, we can ask questions (Box 5.2) and identify an answer in each variant. Box 5.2 Key questions for new institutionalism Lowndes (2010: 65; 2017) and Schmidt (2006: 115) provide a list of new institutionalisms and explore the differences between them. However, there are no hard-and-fast distinctions between each version and it would be misleading to provide a table of definitions with strong dividing lines. There is too much disagreement on which texts fit into which camps (Peters, 2005: 108), too much variation within them, and general problems of measurement and conceptualization are common to all of them (Peters, 2016: 134–36). Instead, particular texts provide different answers to six questions: 1. What is an institution? They are the rules, norms, and ‘standard operating procedures’ that influence behaviour, but some accounts only identify formal rules, or they treat rules as fixed structures more than others. For example, Streek and Thelen (2005: 10–11) describe institutions as ‘formalized rules that may be enforced by calling upon a third party’, not ‘a more or less voluntarily agreed social convention’. Others include formal rules (as ‘exogenous constraints’) and the informal rules enforced by ‘players’ (Shepsle, 2006: 24–25). Hay and Wincott (1998: 952) also warn against treating institutions as real, fixed structures. 2. How does an institution influence individual behaviour? Hall and Taylor (1996: 939–40) distinguish between the ‘calculus’ and the ‘cultural approach’ (compare with cognitive shortcuts in Chapter 4). ‘Calculus’ suggests that individuals make strategic decisions based on their preferences. Institutions provide the

3. 4.

5.

6.

payoffs or negative consequences of their actions (e.g. the ‘penalties for defection’). ‘Cultural’ suggests that individuals follow ‘established routines or familiar patterns of behaviour’ based on their ‘worldview’ or ‘interpretation of a situation’. Institutions provide ‘moral or cognitive templates’ to inform that interpretation. How does an institution become established? Approaches present different ideas on ‘intentional design, accident or evolution’ (Goodin in Lowndes, 2010: 75–76). How does an institution change? While most accounts suggest that institutions can be challenged or modified by agents, some treat institutions as structures that are relatively resistant to change; others focus on less-fixed rules agreed informally between participants, or the ‘institutionalization’ of ideas. This difference helps produce a distinction between quick versus gradual changes to institutions. Four key terms help explain changes to institutions without the crisis and profound challenge that we associate with ‘punctuated equilibrium’: drift (when the rules no longer address the policy problem), layering (new rules are added to old), conversion (redirecting existing institutions to solve new problems), and displacement (switching to another institution to solve the problem) (St Denny, 2016: 83; Waylen, 2014: 217; Thelen, 2004; Mahoney and Thelen, 2010). How does institutionalism inform comparative public policy? While ‘empirical institutionalism’ uses institutions to explain country-level differences, many ‘network institutionalism’ accounts identify similar ‘policy styles’ or ‘rules of the game’ in many countries. Why does this ambiguity matter, and what can we do about it? A lack of clarity on the mechanisms of institutions exacerbates the difficulty of challenging the most damaging rules, such as in relation to the reproduction of institutional racism (Williams, 1985; Bourne; 2001; Murji, 2017; Bhopal, 2018; compare with Michener, 2019). Clarity helps us study and seek to change unfair practices.

KEY VARIANTS OF NEW INSTITUTIONALISM Historical Institutionalism Historical institutionalism treats institutions as ‘the formal rules, compliance procedures, and standard operating procedures’ that ‘structure conflict’ (Hall in Thelen and Steinmo, 1992: 2) or ‘structure and shape behaviour and outcomes’ (Steinmo, 2008: 188). Its key terms are ‘historical contingency’, ‘path dependence’ (Thelen and Steinmo, 1992: 2), and ‘critical juncture’. Historical contingency refers to the extent to which events and decisions made in the past contributed to the formation of institutions that influence current practices. Path dependence suggests that when a commitment to an institution has been established and resources devoted to it, over time it produces ‘increasing returns’ and it becomes increasingly costly to choose a different path (Pierson, 2000a; Peters, 2005: 74; Greener, 2005; Kay, 2006). Therefore, institutions, and the practices they encourage, may remain stable for long periods of time. A ‘critical juncture’ is the point at which certain events and decisions led to the introduction of – or major change to – an institution. The timing of these decisions is crucial because it may be the order of events that sets an institution on its path. Note the phrase ‘sensitivity to initial conditions’ (also a feature of complexity theory in Chapter 6 , pp.105–6): •

The analogy is a thought experiment with two balls – one black and one red – in a container. What happens if, each time we take a ball from the container, we add another black or red ball to the container? If a black ball is chosen we add another black ball to the container and therefore increase the probability of a black ball being chosen next time (from 1 in 2 to 2 in 3). The more we choose, the greater the cumulative effect of this first event or choice. Choosing a black ball at this stage becomes selfreinforcing; it increases the chance of selecting a black ball next time. The immediate lessons from the experiment are (1) the

•

initial event involved a degree of chance and (2) its cumulative effect was significant. The usual example is the QWERTY keyboard. Better systems now exist, but the decision to adopt this keyboard in the past helps maintain its position now.

Path dependence suggests initial unpredictability but subsequent inertia, as relatively small events and actions have large and enduring effects on institutions (Pierson, 2000a; Sanders, 2006: 39; Box 5.3). It may suggest a ‘reproduction mechanism’ as some decisions close off other options and established organizations serve to uphold their own rules (Capoccia and Kelemen, 2007: 341). This aspect of historical institutionalism contributes to a distinct literature called ‘policy feedback theory’ (PFT) (Pierson, 1993; Mettler and SoRelle, 2018; Béland and Schlager, 2019; compare with ‘feedforward’, p.70). The wider lesson is that significant institutional differences are likely to develop in each country, because a different set of initial conditions produces a different set of actions and events which have a cumulative effect and set institutional development on a different path. Therefore, to help give a full explanation for current institutional and policy differences between countries we should look to the past to establish how and why they developed (Peters, 2005: 76; Thelen and Steinmo, 1992: 10). A wide variety of publications may qualify as historical institutionalist (see Sanders, 2006: 44; Schmidt, 2006: 104; Steinmo, 2008: 124; Peters, 2005: 75; Hall and Taylor, 1996: 938). Some seek to explain why institutions developed (Sanders, 2006: 51) or explain why similar levels of public support for public services translate into different levels of taxation and sizes of welfare states (Steinmo, 2008: 124–25; Rosamond, 2000: 116–17). Others explore the interaction between forces for change and relatively durable institutions. For example, why do country-level differences in welfare state institutions endure in the face of strong socioeconomic pressures? Pierson (2000b: 2–4; 2000c: 80–81) argues that economic ‘globalization’ (Chapter 6) may undermine the ability of countries to

control their economic and monetary policies. Further, the rising costs of high unemployment, pensions, and care for older people, have put an immense strain on many countries’ ability to maintain their welfare states. However, these pressures do not produce an inevitable ‘race to the bottom’, such as when countries radically reduce welfare provision and/or reduce corporation taxes and employment regulations to enable them to compete with other countries for foreign investment. Rather, different countries react differently according to (1) the scope and popularity of existing welfare state institutions and (2) political institutions which structure the interaction between political parties, interest groups, and government. Put more simply, even if a government would not design a welfare state as it exists just now, it would not necessarily seek to remove it. Box 5.3 Path dependence and immigration Bale (2017) combines a discussion of (1) the common pressures on immigration policy in Europe with (2) the identification of different policies that have endured for decades. Sources of common pressure include the new freedom of movement of European Union (EU) nationals within the EU; the legacy of decolonization (many EU countries had colonized African and South American countries); a rise in asylum applications, particularly from people in African and Middle East countries; attention to levels of unemployment and benefit-seeking among people born/not born in that country; rising ‘Islamophobia’ and new stereotypes of ‘migrants and minorities as terrorists’; the rising popularity of radical right-wing parties seeking strong immigration controls; and shifting ideas, from multiculturalism (the promotion of diverse cultures within states) to integration (the promotion of a common culture) (2017: 328–62). We can identify similar policies in many countries, such as the lack of attention to, or rejection of, the status of Roma travellers (2017: 359–60). However, we can also identify different policies based on their historical origins. For example, Germany’s post-war policy of

liberal immigration (initially to encourage Germans to return to Germany, followed by attempts to ‘demonstrate its liberal credentials’) but differential status (granting citizenship based on blood links, not birth or residence) has produced a population which includes millions of people described as ‘foreigners’. Its broad policy response is often described as ‘multiculturalism’. Chancellor Angela Merkel had described it as a failure, but she then oversaw major immigration from Syria in 2017, suggesting that this language does not mark a major policy shift (2017: 357). Bale (2017: 357) contrasts Germany with France, which has a history of granting citizenship more easily to so-called ‘illegal’ immigrants, linked to issues such as decolonization in North Africa and previous fears about underpopulation. The policy context is different, and the agenda is more about being ‘tough’ on illegal immigration and promoting national integration. The biggest contrast may soon be with the UK, following a critical juncture in 2016, in which 52% of voters chose to leave the EU, and many expressed support for greater restrictions on immigration from Europe (2017: 337).

Rational Choice Institutionalism Rational choice theory (Chapter 7) employs ‘methodological individualism’, or a commitment to explain sociopolitical outcomes as the aggregation of the decisions of individuals. Its aim is to establish what proportion of political outcomes one can explain with reference to the choices of individuals pursuing their preferences under particular conditions (the ‘calculus’ approach). Their preferences provide the ‘motivation of individual action’ and institutions provide the ‘context’ within which they operate (Dowding and King, 1995: 1). Individuals know that actions have different consequences in different contexts and it affects how they pursue their preferences. They ask themselves, ‘which action produces payoffs closest to my preferences?’ and institutions affect behaviour by providing the payoffs or information regarding the consequences of their actions. For example, institutions may provide the ‘enforcement mechanisms

for agreements’ and ‘penalties for defection’ (Hall and Taylor, 1996: 939–40). Sets of rules, influencing choices, often produce regular patterns of behaviour. ‘Equilibrium’ describes a stable point at which there is no incentive to divert from these patterns of behaviour. Institutions may be treated as solutions to at least three public policy problems identified by rational choice. First, to solve a ‘collective action problem’: the potential for choices made by individuals to have an adverse societal effect when there is an absence of trust, obligation, or other incentives to cooperate with each other (see Chapter 7 on the Institutional Analysis and Development (IAD) framework). The institution gives individuals that incentive if the rule dictates that they will be punished (e.g. if they flout the rules of the road), or provides a positive inducement (e.g. individuals are rewarded financially if they recycle) (Peters, 2005: 49). Second, to reduce ‘transactions costs’. In economics, institutions represent a set of – often formal and enforced – rules governing relationships between individuals or organizations. The aim is to reduce the likelihood that agreements between actors are broken. If so, they help reduce uncertainty and the costs associated with wasted effort on joint tasks that are not fulfilled. Studies have explored the relationship between legislative committees and government agencies, the rules of coalition government formation, and the development of EU institutions governing the interactions between member states (Hall and Taylor, 1996: 943). There is also a large literature identifying the incentives for governing bodies to cooperate with each other to produce more effective governance (see the Institutional Collective Action (ICA) framework in Chapter 7). Third, to address the instability of choice. As Chapter 7 suggests, there is no perfect rule to aggregate individual preferences into social preferences. Instead, there is always the potential for ‘intransitive’ policy decisions because different choices can win according to the order in which they are presented (Riker, 1982). Imagine the policymaking equivalent of a two-player game, in which scissors beats paper, paper beats stone, but stone beats scissors (please do not contact me to insist erroneously that stone should be rock)! If so, we might expect ‘cycling’ from one decision to another:

as soon as one choice is made, there is a more-preferred alternative ready to be selected as its replacement. It is fine for a fun game, but not in policymaking if each choice involves major upheavals in organizations and resources. In this case the role of an institution is to help manage how, and how often, people make choices (Dowding and King, 1995: 3). For example, a rule to ensure that elections take place once every four years, producing a governing majority with a relatively fixed policy programme, limits cycling.

Normative and Sociological Institutionalism Peters (2005: 26) describes March and Olsen’s work as ‘normative’ because they identify the ability of ‘norms and values within organizations’ to influence behaviour. March and Olsen (1984: 739) reject the idea that we should explain political behaviour primarily in terms of strategic choices based on the preferences of individuals. Instead, they refer to the role of political structures, defined as collections of ‘institutions, rules of behavior, norms, roles, physical arrangements, buildings, and archives that are relatively invariant in the face of turnover of individuals and relatively resilient to the idiosyncratic preferences and expectations of individuals’. Structures act as a key reference point for political actors driven by duty: ‘Members of an institution are expected to obey, and be the guardians of, its constitutive principles and standards’ (March and Olsen, 2006a: 7). The ‘rules of appropriateness’ associated with each institution are ‘transmitted through socialization’ and ‘followed because they are seen as natural, rightful, expected and legitimate’ (March and Olsen, 1984: 739; 2006a: 7). There are three main complications to such analysis. First, informal rules are often ambiguous and mean different things to different people. Second, different policy participants may be more or less willing or able to follow rules. Third, they may face a series of potentially contradictory rules associated with their role within multiple institutions. These issues prompt March and Olsen (2006a: 9) to stress the role of an individual’s identity and the link between forming a relatively strong attachment to one institution and adhering to its rules. This approach seems chicken-and-egg (or ‘two sides of

the same coin’ – see Chapter 6 on Giddens). Although institutions are structures that influence individual behaviour, ‘individuals must pick and choose among influences and interpret the meaning of their institutional commitments’ (Peters, 2005: 26). While the institution, as a structure, provides the ‘rules of appropriateness’, the individual is in a position to judge which ‘rules of appropriateness’ are appropriate to follow. Peters (2005: 108) suggests that this focus on institutional norms and cultures originated in sociological institutionalism. Normative and sociological institutionalism share a rejection of ‘functionalist’ accounts of institutions. In sociology, functionalism suggests that ‘societies have certain requisite functions that must be performed if they are to survive’; institutions represent the tools used by societies to produce the goods and services that they require (2005: 109). Similarly, in political science, institutions are ‘functions of political life’ (John, 1998: 39). Societies require that different institutions carry out different tasks to maintain an effective political system: ‘legislatures make laws, executives take decisions, bureaucracies implement them and courts resolve disputes’ (1998: 39). In each case, institutions help societies translate values into political outputs. Crucially, normative institutionalism rejects this idea that institutions have evolved to represent the most efficient way to solve the problems that societies face (March and Olsen, 1984; 1996). Institutional forms are not caused solely by the nature of their external surroundings and do not change inevitably to fit their new surroundings. Rather, institutions are somewhat autonomous and ‘the paths they follow seem determined in part by internal dynamics only loosely connected to changes in their environment’ (March and Olsen, 1996: 256). Organizations often live on after their aims have been met or they no longer serve a useful purpose (Box 5.2). They maintain their standard operating procedures in the face of changing environments and the decisions of higher-level policymakers (Peters, 2005: 110). Therefore, to understand how they operate, we must study their ‘culture’ or the manner in which people within them understand their world (Hall and Taylor, 1996: 947; Peters, 2005: 111–15).

One classic concept is ‘institutional isomorphism’. DiMaggio and Powell (1983) revisit Weber’s argument that competition among firms in a capitalist system, and between states, would cause bureaucracies to converge on the most efficient model. Hierarchical and unified bureaucracies, with strict rules to enforce formal roles, develop because they are ‘technically’ superior (Weber, 1978: 973). Rather, DiMaggio and Powell (1983: 147) argue that organizations converge without becoming efficient. They emulate the successful organizations that they compete with for resources, power, and legitimacy within a common legal and political framework. Or, they become driven by the rules and norms of public sector professions, developed through education and professional networks and encouraged by selective hiring practices (1983: 152). Indeed, if the professional network is homogeneous, these rules and practices may be taken for granted, akin to Bourdieu’s concept of ‘habitus’ or everyday, normal, embodied action based on actors’ similar perceptions of the world (Wang, 2016: 351). They become part of institutional ‘culture’ (Alasuutari, 2015). Such practices are not challenged routinely, particularly when the aims of an organization are unclear and oversight is performed by people who are not particularly aware of the organization’s activities. Although we might expect inefficient organizations to be ‘weeded out’ in a process of ‘natural selection’, economic efficiency is difficult to judge and organizations are often judged on their political performance (DiMaggio and Powell, 1983: 157). Or, they are valued for their symbolic or social value than their efficiency (Hall and Taylor, 1996: 949).

Discursive and Constructivist Institutionalism Schmidt (2010: 2) uses the label ‘discursive institutionalism’ to describe studies which ‘take ideas and discourse seriously’. It includes ‘constructivist’ institutionalism, which draws on an approach to science that prompts us to consider the status of our knowledge: ‘whether things are simply given and correctly perceived by our senses (empiricism) or whether the things we perceive are rather the

product of our conceptualizations (constructivism)’ (Kratochwil, 2008: 80; compare with Box 4.5). In this context, Schmidt (2008: 313) suggests that constructivism presents an ontological challenge: ‘about what institutions are and how they are created, maintained, and changed’. It distinguishes between institutions that are real structures that influence the interests of agents and institutions as ideas created by agents and used to inform their perceptions of their interests. People’s actions are not simply a ‘reflection of their material interests but, rather, a reflection of particular perceptions of their material interests’ (Hay, 2006a: 68). Or, if enough people believe a particular idea or argument, it becomes important as a means to explain behaviour, regardless of its relation to the ‘real’ world (Blyth 2002: vii–viii). Shared beliefs give people a common aim or reason to believe they have shared interests. Ideas are often institutionalized; they are taken for granted and rarely questioned, or at least accepted as the starting point when we consider policy problems (see Chapter 11 on paradigms). Therefore, discursive or constructivist accounts may identify path dependence, but relate it as much to ideas as governmental structures: ‘it is not just institutions, but the very ideas on which they are predicated and which inform their design and development, that exert constraints on political autonomy’ (Hay, 2006a: 65). As such, institutionalized ideas are not as rigid as structures. They are ‘constantly in flux, being reconsidered and redefined as actors communicate and debate with one another’ (Béland and Cox, 2010: 4; see also Béland, 2019). Institutions-as-ideas influence action but ‘exist’ in terms of the way that actors understand them. Hay (2006a: 65) notes that constructivists examine how actors ‘interpret environmental signals’ (compare with punctuated equilibrium theory, or PET, Chapter 9). Institutions represent ‘established ideas’ or ‘paradigms’ which act as ‘cognitive filters’ or the primary means through which people understand their environment. The agenda of constructivist institutionalism is to understand how such ideas are ‘contested, challenged and replaced’ (2006a: 65). Discursive institutionalists describe historical, rational choice, and sociological/normative institutionalism as ‘much better at explaining

continuity than change. In all three … institutions serve primarily as constraints’ (Schmidt, 2010: 2). They treat institutions as stable or relatively fixed, only to be changed radically – during a period of ‘punctuated equilibrium’ – before another stable institution forms (Blyth, 2002: 7; Hay, 2006a). Or, they are ‘overly structuralist’ and do ‘not grant purposeful actors a proper role’ (Olsen, 2009: 3; Hindmoor, 2010). The solution is to relate institutional stability to the maintenance of dominant ideas; change results from successful challenges to those ideas. This approach helps explore how institutions influence agents and how actors challenge ideas and beliefs. For example, Schmidt (2006: 2010: 3; Schmidt and Radaelli, 2004: 193) argues that stability and change relates to ‘coordinative’ or ‘communicative’ discourse. ‘Coordinative’ refers to the role of individuals, groups, or networks ‘who generate the ideas that form the bases for collective action and identity’ and compete with each other to have their ideas accepted. ‘Communicative’ refers to the use of those ideas in the wider public sphere, from elites using policy platforms to persuade voters, to debate among social movements or local voters. Institutions represent the most established ideas that are used to frame discussions and dominate debates, while coordinative and communicative processes suggest that these ideas can be challenged when debated. However, the role of ideas is not absent in other accounts (Peters, 2005: 75; Hay, 2006a: 66; Hall and Taylor, 1996: 942; Sanders, 2006: 42; Blyth, 1997: 229). Rather, Schmidt (2006: 113–14) suggests that discursive may be viewed as a supplement to other variants. Further, a large part of policy studies seeks ways to account for major policy and policymaking changes in the absence of crisis or punctuation (Box 5.2). Descriptions include ‘punctuated evolution’ (Hay, 2002: 163), ‘gradual change with transformative results’ (Streeck and Thelen, 2005: 9), ‘gradual but profound’ change (Palier, 2005: 129), and ‘phased transition towards paradigm change’ (Studlar and Cairney, 2014: 514).

Feminist Institutionalism

Chapter 3 shows how feminist research helps us understand the exercise of power and effects of power asymmetries. The ‘guiding principles of feminist methodology’ include a continuous focus on ‘gender and gender asymmetry as a basic feature of all social life, including the conduct of research’ and ‘the empowerment of women and transformation of patriarchal social institutions through research and research results’ (Fonow and Cook, 2005: 2213; see also Ackerly and True, 2020). Research often focuses on the ‘codes of masculinity and femininity’ which underpin institutions and political action (Lovenduski, 1998: 337; Chappell and Waylen, 2013: 600–1). The term ‘gendered’ denotes the ways in which actors exercise power by, for example, categorizing masculine and feminine roles or norms and assigning higher value or more rewards to the former (Scott, 1986: 1067; Cohn, 2013: 3). As Chapter 3 (pp. 49–50) discusses, visible examples arise most frequently during elections and the gendered norms of participation in public life, while less visible examples range from the treatment of feminist issues as private rather than public policy issues, ‘epistemic violence’ such as when Women of Colour are excluded from political debate and intellectual history, and everyday rules which seem innocuous but can symbolize immense inequalities of power (Kenny and Mackay, 2009: 277). Feminist institutionalism takes forward this approach, to show that institutions are gendered and have gendering effects (Mackay, in correspondence, 2019). It helps explain the ways in which formal rules, and informal norms and practices, reflect and reproduce power asymmetries. It shows how actors create, maintain, reproduce, or take for granted – but often challenge – the ‘gendered power dynamics’ (Lowndes, 2018: 60) or ‘gendered patterns of power’ (Kenny, 2007: 91) in political life. Examples range from openly expressed stereotypes of historically male roles (the aggressive soldier or police officer, the calculating bureaucrat) and female roles (‘the caring nurse, the compliant secretary’) to the informal rules which represent the ‘hidden life of institutions’ (Chappell and Waylen, 2013: 608). Feminist institutionalism provides a corrective to the lack of attention to gender in political science and public policy (Mackay,

2004: 99; Kenny and Mackay, 2018: 93–96; Bacchi, 1999: 200; Guerrina et al., 2018; Lovenduski, 2011). Further, like ‘critical’ scholarship (Chapter 3), it helps ‘produce social change that will empower, enlighten, and emancipate’ (Schneider and Ingram, 1997: 51; Cairney and Rummery, 2018: 2). As such, a focus on gendering can improve the insights of other variants of new institutionalism, albeit subject to the same limitations regarding the ambiguity of key terms (Box 5.2; Waylen, 2014: 212; Lowndes, 2014; Chappell and Mackay, 2017): •

•

Historical. Path dependence may be a routine feature of elections, such as in the incumbency effect: men hold more than 70% of seats in over 70% of national parliaments (InterParliamentary Union, 2018). This dominance began with formal rules limiting the ability of women to stand for election, and is reinforced by informal rules including the belief of too many party selectors that women are less electable, and men that they are entitled to harass women candidates (Norris and Lovenduski, 1995; Evans, 2014: 148; Kenny and Mackay, 2018). In that context, responses to such rules can be formal, such as via All Women Shortlists, and informal, to challenge deep-seated beliefs about the archetype of an elected politician or the rules of acceptable conduct during debate (Lamprinakou et al., 2016; Durose et al., 2012). Potential critical junctures include the introduction of new parliaments, presenting the possibility of ‘locking in’ elements that promote gender equality and gender justice at the stage of institutional design’ (Mackay, 2014: 549), or a tipping point when there is a sufficient ‘critical mass’ of women in an elected assembly to influence its culture (Mackay, 2004: 101–102). For example, in legislative committees, women are ‘tokens’ in the USA, segregated in Germany, but validated in Sweden which has the highest proportion of women in elected positions (Bolzendahl, 2014: 869–70; compare with the less clear relationship between gender and corruption, Stensöta et al., 2015). Rational choice. Men and women may adopt the same ‘calculus’ approach, but face very different rewards and punishments for

•

•

•

action (Driscoll and Krook, 2009). Normative. The transmission, via socialization, of gendered norms relating to participation in public policy can be traced across many generations. The ‘gendered logic of appropriateness’ (Chappell, 2006a) has been used to keep women in subordinate roles in public life and undermine challenges to this practice (Chappell and Waylen, 2013: 602– 603). Discursive. The use of discourse to reinforce ‘racial or gendered stereotypes’ may help maintain social inequalities (Kenny, 2007: 96; compare with Freidenvall and Krook, 2011; Erikson, 2017; and Chapter 4 on Social Construction and Policy Design, or SCPD). Network. The ‘velvet triangle’ describes the policy networks of ‘feminist bureaucrats, trusted academics, and organized voices in the women’s movement’ that develop partly because women are excluded routinely from the positions of power (Woodward, 2004: 78; Chappell and Waylen, 2013: 599).

The feminist institutionalism literature now contains many relevant case study examples. Discussing the UK, Annesley and Gains (2010: 909) argue that the ‘core executive’ (the ‘centre’ of power described in p.136) is ‘a gendered institution in terms of recruitment, resource allocation, relationships and rules’. For instance, although there developed a formal commitment to recruiting more women – in, say, their fifties – to the most senior positions in the UK civil service, it was undermined by an informal requirement to work intensively as a ‘private secretary’ in their twenties or thirties, which generally favoured men with fewer caring responsibilities (underpinned by a wider societal ‘logic of appropriateness’) (Chappell and Waylen, 2013: 611). Further, formal measures by EU institutions to foster greater gender equality were generally overshadowed by its unstated preference to focus on economic development (MacRae and Weiner, 2017). New formal measures for gender equality in reformed or newly formed institutions often do not ‘stick’ because they are ‘nested’ in a wider set of gendered rules and practices which

undermine them (Mackay, 2014: 550–54). Further, the distinction between the new ‘rule-makers’ versus the established ‘ruleenforcers’ who ‘can slip back into old ways’ (especially when the new rules are ambiguous) resembles the incomplete implementation and ‘old way of doing things’ described in Chapter 2 (Waylen, 2014: 219; Carey et al., 2019; Mamudu et al., 2015).

Empirical vs. Network Institutionalism? Diverging and Converging Policy Styles A focus on policy style allows us to compare the effects of formal and informal institutions on policymaking. Empirical institutionalism identifies country-level differences in styles based on formal institutions within political systems. Network institutionalism highlights similar styles in different countries based on their adherence to the same informal ‘rules of the game’. The former is exemplified by Lijphart’s (1999) Patterns of Democracy (Table 5.1). Table 5.1 Lijphart’s majoritarian–consensus dichotomy

Institutional divisions

Majoritarian democracy

Consensus democracy

Executive power

Concentrated in single-party majority cabinet

Shared in broad multi-party coalition

Executive–legislative Executive is relationship dominant

Balance of power between executive and legislature

Party system

Two-party system

Multi-party system

Electoral system

Majoritarian and disproportional (based on a plurality of votes)

Proportional

Interest group system

Pluralist free-for-all and competition among groups

Coordinated and corporatist, exhibiting compromise and concertation

Federal–unitary

Unitary and centralized

Federal and decentralized

Legislative power

Concentrated in unicameral legislature

Divided between two equally strong houses

Constitutions

Rigid constitutions Flexible constitutions that can be changed that can be amended only by large by simple majorities majorities

Constitutionality of laws

Decided by legislatures

Subject to judicial review

Central banks

Dependent on the executive

Independent

Illustrative examples

New Zealand, the UK

Switzerland, Belgium, the EU

Source: adapted from Lijphart (1999: 3–4). Note that the examples are illustrative. No system lives up to an ideal-type.

Lijphart (1999: 2) argues that there are two basic models of political system design: those that concentrate power in the hands of the few (majoritarian) and those that ‘share, disperse, and limit power’ (consensus). In a majoritarian democracy, the first-past-thepost voting system exaggerates governing majorities by granting most seats in the — legislature to a party that commands only a plurality of the vote. This result, combined with an imbalance of power towards the governing party’s leadership, a weak second chamber, and a unitary government, produces a concentration of power at the centre. Lijphart (1999: 2–3) associates majoritarian

democracies with an ‘exclusive, competitive and adversarial’ mentality in which parties compete within parliament, interest groups are more likely to compete than cooperate, and governments are more likely to impose policy from the top down than seek consensus. In a consensus democracy, the proportional electoral system produces no overall majority and power is dispersed across parties, encouraging the formation of coalitions based on common aims. This spirit of ‘inclusiveness, bargaining and compromise’ extends to group–government relations, with groups more likely to cooperate with each other and governments more willing to form corporatist alliances. In contrast, Richardson (1982b) suggests that different political systems produce similar policy styles. Although their political structures and electoral systems vary, they share a ‘standard operating procedure’ based on two factors: an incremental approach to policy and an attempt to reach consensus with interest groups, not impose decisions. Ministers and senior civil servants devolve most decision-making to less senior officials who consult with groups and exchange access for resources such as expertise. This exchange is based on the ‘logic of consultation’ with the most affected interests; it encourages group ‘ownership’ of policy and maximizes governmental knowledge of possible problems (Richardson and Jordan, 1979; Richardson et al., 1982: 2; Jordan and Maloney, 1997; Jordan, 1990a). ‘Policy community’ describes this close relationship between civil servants and groups (but see Box 9.1 (p.152) on different meanings of ‘community’). Membership is based partly on the willingness of actors to follow and enforce the ‘rules of the game’. They recognize the benefits – such as policymaking stability and policy continuity – of attempting to insulate their decisions from the wider political process. In such cases, we explain policy styles in terms of informal rules and norms of behaviour that transcend formal structures. Evidence from European comparisons suggests that countries do not live up to the reputations associated with their formal institutions (Jordan and Cairney, 2013). Kriesi et al.’s (2006: 345) study of seven Western European countries suggests that the British policy style is relatively consensual despite its majoritarian political system

(compare with Richardson, 2018). In contrast, ‘the Italian style of policy-making appears to be more unilateral’ even though it ‘has institutions which are rather of the more consensus-democratic type’, while the EU is ‘less co-operative than it appears at first sight’. Cairney et al. (2018) find that the majoritarian/consensus distinction between the UK and Switzerland has only a subtle effect on policy community dynamics. Cairney (2008, 2009b, 2019b) produces a similar assessment of majoritarian-UK versus consensus-devolved government (in Scotland, Wales, and Northern Ireland), while Cairney and Widfeldt (2015) find many similarities between the UK and Sweden (again, compare with Richardson, 2018; see also Larsen, Taylor-Gooby and Kananen, 2006; Atkinson and Coleman, 1989; Bovens et al., 2001; John, 1998: 42–44; Freeman, 1985). Barzelay and Gallego (2010: 298) also criticize historical institutionalist accounts that focus too much on national character traits (in France, Spain, and Italy) at the expense of knowledge of their subsystems. Of course, a global comparison would test this argument to its limits. Howlett and Tosun’s (2019) edited volume identifies potentially major differences in, for example, countries with or without elections, and the extent to which governments are more or less supportive of interest groups and non-governmental organizations (NGOs). Still, the evidence we have so far suggests that we need to demonstrate such differences systematically rather than assume that major differences in policymaking will be caused necessarily by formal institutional differences.

CONCLUSION Studies of new institutionalism argue that institutions are not the bricks-and-mortar arenas within which decisions are made. Rather, we study the rules, norms, and conventions that influence individual behaviour and social practices. While this research agenda has raised the status of institutional studies, it has also produced much soul-searching about what an institution is and how we should study it. We may agree that institutions matter and that they evolve and change, but explaining how and why is another matter. We can focus

on the formal rules that are enforced, such as the written constitutions that set out a political system’s separation of powers. Or, we can identify norms of behaviour and informal rules. We can regard institutions as sets of incentives used by individuals pursuing their preferences, or structures that influence those preferences. We can emphasize stability or change. Institutions can be relatively stable and durable structures that live longer than individuals, or unstable sets of ideas that are taken for granted on one day only to be challenged the next. We can treat them as structures that exist in the real world or as constructs that exist in the minds of participants. We may identify historical differences to explain why public policy is different in different political systems, or current practices which seem to be very similar. Or, as in the case of PET (Chapter 9), we can maintain the classic description: institutions are the organizations making the rules (Workman, in correspondence, 2019). These theoretical problems and terminological debates have the potential to distract us from the value of institutional analysis. Alternatively, we can identify a basic common understanding that institutions are important, and a set of questions to guide public policy research: how are institutions formed by agents? Why, and under what circumstances, do agents accept or follow rules? What patterns of policymaking behaviour can we attribute to those rules? (Peters, 2005: 156). The solution for ‘problem-oriented scholars’ is not to ignore problems of ambiguity. Rather, it is to get enough of a sense of the debate to allow us to conduct well-informed and theorydriven public policy research. A key theme of this book regards the extent to which we can use the same theories to explain policy developments in countries with different institutions. For example, the policy process in a federal USA may contrast with a devolved but unitary UK. While Germany has a federal structure, its size and relationship with the EU undermine direct comparisons with the USA. Australia’s system is federal but also parliamentary, not presidential. The checks and balances in the US system ensure a key policy role for the courts and a second legislative chamber; in the UK, there have been few avenues for judicial policy influence (at least until EU legal

procedures developed) and the second chamber is comparatively weak. The US reliance on plurality voting (producing a two-party system) is shared by the UK (Westminster elections) and Canada but not Germany or Japan. Its ‘weak’ party structure, in which the main national parties do not control their state or local counterparts (Atkinson and Coleman, 1989), is shared by Canada but not Germany which has integrated parties and formal links to coordinate policy (Horgan, 2004). Using these formal differences as a basis of comparison, we may produce questions relating to the application of theories of public policy to different countries. For example, are political systems with formal checks and balances the most conducive to pluralism (Chapter 3)? Are political systems with diffused powers and extensive ‘veto points’ less conducive to radical rather than incremental change (Chapter 4)? Is ‘venue shopping’ a feature of federal systems modelled on the USA but not unitary systems modelled on the UK (Chapter 9)? On the other hand, as the policy styles debate shows, countries with different formal institutions may produce similar ‘standard operating procedures’ when the behaviour of participants is driven by informal rules and norms. Further, much of the US literature combines a discussion of specific US institutions with universal themes. For example, Baumgartner and Jones’ (1993) study is based largely on the bounded rationality of policy-makers (Chapter 9). The advocacy coalition framework (ACF) relates primarily to the belief systems that bind policy participants together (Chapter 10). Kingdon’s (1995) multiple streams analysis (MSA) explores policymaker receptivity to new ideas (Chapter 11). The policy transfer literature combines a discussion of (1) the widespread adoption of similar policy ideas with (2) the institutional differences that ensure that they are formulated and implemented in different ways (Chapter 12). In other words, the remaining chapters recognize the difference that institutions make but also highlight the policy processes that transcend formal institutional boundaries, and the informal rules – such as on gender roles in public life – found in all systems. In Chapter 13, we note that such analyses are largely of countries in the Global North. Policy theories have only begun to

compare their institutions and experiences with countries in the Global South. Ontology – a theory of reality (is there a real world, does it exist independently of our knowledge of it, and how does it relate to the concepts we use to describe it?). Related terms are epistemology (a theory of knowledge and how we can gather it) and methodology (the analysis of methods used to gather knowledge). Discourse – the verbal or written exchange or communication of ideas. In some cases, it has a more ambitious meaning, linking discourse to the power to decide which forms of knowledge are legitimate or acceptable. Policy style – the way that governments make and implement policy.

6 Structures, Environments, and Complex Systems Key themes of this chapter: • •

• • • •

A focus on structures, policy conditions, environments, or complex systems is a useful corrective to individualist and ‘topdown’ accounts of policymaking. Policy and policymaking is not controlled by a small group of policymakers at the ‘centre’ of government. They do not control their policymaking context. Rather, it influences their attention to problems and the feasibility of solutions. Policy conditions include the demographic, social, technological, natural, and economic changes that often seem to drive the policy agenda. The wider policymaking environment includes the institutions, networks, and ideas that new governments inherit before they make new choices. Many policy theories use many different evolutionary metaphors to describe policy actors adapting to their environments. In complex policymaking systems, policy often seems to ‘emerge’ locally in the absence of central control.

INTRODUCTION: STRUCTURE AND AGENCY IN THE POLICY PROCESS When we seek to understand policy change, we attribute the exercise of power to the individuals that make policy but also recognize the context in which they operate and the pressures that they face. This context prompts us to consider a different perspective to ‘comprehensively rational’ policymaking and the idea that the policy process begins with the decision by a policymaker to identify a problem to solve (Chapters 2 and 4). Policy analysis begins with the

consideration of context, not a blank sheet of paper (Bardach and Patashnik, 2015; Cairney and Weible, 2017). Context includes geographical and socio-economic conditions, the government infrastructure and policies that are already in place, and the events that often seem to be out of the control of policymakers and prompt them to act. There are many ways to characterize this process, with different approaches striking a different balance between the role of structure and agency (Chapter 1). Some accounts seem to focus on ‘structural power’ and profound limits to agency (Chapter 3, pp.51–2), but most are really trying to ‘zoom out’ from elite individual action to describe a wider perspective on policymaking. The term ‘structure’ refers vaguely to a set of parts put together in a particular way to form a whole. We tend to attribute two key properties to structures: they are relatively fixed and difficult but not impossible to break down; and they influence the decisions that actors (‘agents’) make. Examples include the structure of the – domestic and global – economy and the rules and policies that new governments inherit when they enter office. Policy studies may not always use the language of structure and agency directly, but they supplement a focus on policymaker action by highlighting context. We know that the decisions made by policymakers can have a profound effect on society, the economy, and government. However, we also know that the choices they make are often based on policy problems over which they have limited control, such as an increasingly global economy; demographic change; and events that propel issues to the top of their agenda (Chapter 9). There are limits on their ability to anticipate the effects of policy and carry it out successfully (Chapter 2). Policymakers operate within the context of rules, norms and cultures that influence their behaviour (Chapter 5). Further, such is the scope of the state that individual policymakers could not hope to have a full understanding of it, and their decisions while in office will likely represent a small part of overall government activity. We miss a lot if we only focus on their action, not the big picture. This chapter identifies the main ways in which we can describe this policymaking context. Policy conditions describe what

policymakers take into account when identifying problems and deciding how to address them. They may be particularly aware of a political system’s size, demographic structure, economy and mass behaviour. These conditions represent a source of pressure on how policymakers operate. For example, an ageing population puts pressure on governments to address pensions policy and their provision of social care, while an economic crisis generally rises to the top of a government’s agenda (Box 6.1). Policy environment is an evolutionary metaphor. It brings together the key concepts of this book: the spread of actors and responsibilities across many levels and types of government (Chapter 8); the institutions associated with each venue (Chapter 5); the networks between policymakers and influencers (Chapter 9); the ideas on which policy actors draw (Chapter 11); and structural or socioeconomic factors (and events) (Box 6.1). Combined, these factors produce the sense that the main influences on policymaking often seem to be overwhelming or out of the control of individual policymakers. Further, each policy theory brings together these factors to provide their own story about how policy actors engage with their environment. Finally, in this chapter, we focus on the ways in which complexity theory tells that story. The language of complexity seems to downplay the role of central governments and the power of individual policymakers. Complexity theory suggests that we shift our analysis from individual parts of a political system to the system as a whole – as a network of elements and systemic behaviour that cannot be broken down easily into its constituent parts. Complex system describes the interdependence between a huge number of actors, albeit often in pockets of local activity from which outcomes ‘emerge’. Policy results from their interaction, the system seems to dampen or amplify the effects of action, and policy outcomes often appear to emerge locally despite attempts at central control. This approach serves as a corrective to accounts that focus too much on the ability of individual policymakers to change policy singlehandedly. It is useful as long as we do not go too far towards the idea that structures determine behaviour and remove the role of agency, or the ability of individuals to deliberate and make choices.

Box 6.1 Structural factors and policy conditions •

• •

•

•

•

Historic-geographic. Factors such as the climate and stock of natural resources influenced how countries and regions developed. The population size of political systems may inform how institutions are set up. Policies for dense, urban areas may not be popular in, or appropriate for, rural areas (Hofferbert, 1974: 228–29). Demographic. An ageing population puts pressure on the costs of social security and personal care. A ‘baby boom’ produces greater demand for schools. Economic. Global and domestic economic activity influences the size of government tax income, its ability to attract foreign investment, and the effect of interest rate changes on inflation and investment. Social and behavioural. Social attitudes and behaviour influence the popularity and effect of policy measures. Unhealthy behaviour in the population affects demand for healthcare services. Technological. The mass production of cars produces new demands on transport policy. Medical advances produce demands for new medicines and equipment (Hogwood, 1992a: 191–208). Information technology changes the way we engage in politics and design and use public services (Dunleavy et al., 2006). Institutional. Formal institutions describe the structure of government. Informal institutions and norms influence behaviour within it (Chapter 5).

DO STRUCTURAL FACTORS DETERMINE POLICY AND POLICYMAKING? A focus on ‘structures’ helps raise a fundamental question: how much do policymakers shape, and how much of their behaviour is shaped by, their surroundings (Hay, 2002: 89)? Box 6.1 outlines the

main ‘structural’ factors that help shape policy, while John (1998: 92) highlights how strongly we could associate them with policy outcomes: The simple idea is that the policy process, far from being a rational weighing up of alternatives, is driven by powerful socioeconomic forces that set the agenda, structure decision makers choices, constrain implementation and ensure that the interests of the most powerful (or of the system as a whole) determine the outputs and outcomes of the political system. Historical approaches suggest that structural factors are something to ignore at our peril. For example, Hogwood (1992a: 191–208) compares demography and changing parties in government and concludes that the former often had a larger effect on trends in public expenditure, employment, and regulation. Further, many empirical studies inspired by systems theory (Easton, 1953, 1965; Deutsch, 1970) test the statistical association between the socioeconomic factors that produce demands feeding into the political system and policy outputs (Blomquist, 2007: 261). Dawson and Robinson (1963), Dye (1966), and Sharkansky (1972: 21) argue that these factors explain variations in, for example, the demand for and provision of public services, albeit while mediated by political parties, legislatures, and executive administration. Further, Hofferbert’s (1974: 228–30; see also Wilder, 2017) ‘funnel of causality’ describes how: • • • •

historic-geographic conditions contribute to the socioeconomic composition of a region (climate and natural resources influence population density, employment, and levels of prosperity); the concentration of social groups in regions may be traced back to historical events; the socioeconomic composition of a region contributes to ‘mass political behaviour’ such as voter turnout, which influences the structure and fortunes of parties; and these factors combine with ‘government institutions’ to influence elite behaviour.

These descriptions are not deterministic. For example, Dye (1966: 299–301) rejects a sole focus on individual and group action, then explores how socioeconomic factors can be taken into account in the ‘little black box labelled political system’. Hofferbert (1974: 231–33) argues that, while elites cannot ignore their policy environment, there is still scope for ‘leadership’ and ‘vigorous action against strong historical and economic forces’, and structural impact varies by policy issue over time (1974: 237–41). In other words, they portray an iterative process: socioeconomic factors influence outcomes, but the impact of policy feeds back to the policy environment. Still, there was a strong academic reaction to the perception that the ‘Dye– Sharkansky–Hofferbert’ (DSH) approach down-played the idea that ‘politics matters’ (Blomquist, 2007: 270–71). Studies sought to ‘bring politics back in’ by comparing countries rather than US states and finding a much clearer link between, for example, levels of social welfare spending and the role of political parties (John, 1998: 105–6; Blomquist, 2007: 264–65). More generally, it is difficult to conclude that structural factors determine politics, for two main reasons. First, the structure and agency-inspired policy literature presents a more dynamic process in which structures influence agents and agents often make or reconstitute the structural constraints within which they operate (Marsh and Smith, 2000: 5; Adler and Haas, 1992: 371; Jacobsen, 1995: 300). Different texts describe this relationship differently (Hay, 2002: 118–21), but few present a completely structural account (compare with Chapter 3). Second, we cannot define structures objectively and there are few conditions that, like a meteor about to fall on your head, cannot be ignored. Rather, they are subject to interpretation; policymakers attach different meanings to what appear to be the same factors (Blomquist, 2007: 274–75). They mediate the effects of structural factors and external events by defining their importance and acting on them in a particular way (Chapter 10). They may pay attention to some and ignore others for long periods. Therefore, a focus on structures resembles the study of agenda setting (Chapter 9) in which actors try – with variable success – to convince policymakers that certain factors are causing a crisis that should not be ignored.

Yet we should not go too far in highlighting the role of discretion and attention. Indeed, the fact that policymakers in most countries could choose to focus on any issue but actually focus on a small number of issues – such as the economy – reinforces the importance of socioeconomic factors. A sensible way to address this problem analytically is to think in terms of the structure–agency mix, with some issues providing more or less constraint on choice than others. For example, the ageing population may give governments little choice but to plan for the consequences, but they can do so in a variety of ways. Technology-driven healthcare is resistible, and public health policies are often more cost-effective, but many governments still seem to favour medicines and acute hospitals strongly. Adapting to our physical environment also involves a complex mix of structure and perception (Box 6.2). For example, coastal conditions may appear to force us to build protective barriers, but policymakers can ignore the issue for some time (Werritty, 2006). Many policymakers also seem determined to deny or ignore climate change (Tangney, 2018). Box 6.2 Japan’s nuclear policy Japan’s choice of nuclear power as a key source of electricity demonstrates the strong, but not determining, nature of external factors on decisions (see Park and Sovacool, 2018). In this case, we are literally describing the policy environment because energy policy depends on two potentially competing factors. First, Japan has few indigenous sources of energy such as coal or gas, requiring it to import natural sources or rely on nuclear power. Second, it is vulnerable to natural disasters such as earthquakes that undermine reliance on nuclear power (even if considerable measures are taken to make it secure). The resultant policy, including nuclear as a key source, follows policymakers weighing up these sorts of constraints but without policy being determined by them. This case demonstrates how difficult it is to describe structural constraints. In the aftermath of the earthquakes in March 2011, one scientific commentator (on the rolling BBC news coverage) still maintained that Japanese governments effectively

had to rely on nuclear power; that they had no choice. This type of example may be what Ward (1987: 602) has in mind when he describes situations which appear to make ‘certain acts unthinkable or physically impossible’ or ‘so costly that actors are structurally constrained from carrying them out’. This constraint is far less apparent elsewhere, and the Japanese crisis has prompted many governments in other countries to reassess their energy portfolio (Pidd and Goldenberg, 2011; Wittneben, 2012; Huang et al., 2018; Chapter 9).

THE ECONOMIC CONTEXT: MARXISM AND GLOBALIZATION Chapter 3 discusses the continued relevance of Marxism to studies of power, while Kiely (2017) describes its importance to studies of globalization. Marxist accounts (broadly defined) help us construct a simple story of policymaking in which elites, and the structure of government, seem devoted to protecting the capitalist system of economic production, giving economic interests a privileged position within the policy process. In this scenario, our assumption is that the capitalist system benefits one class of people (those who own or control capital) at the expense of another (the working class, who do not). Elites in powerful positions in government share relationships with business elites. For example, the ruling and capitalist classes share a common, privileged, background that predisposes them towards working with each other. Or, the world of business provides an incentive for policymakers to cooperate, by providing campaign funding or a source of employment in the future (Cairney and Yamazaki, 2018: 259). Consequently, government elites forge close networks with representatives of business, ensuring that the latter dominate access to policymaking at the expense of other interests (John, 1998: 94). It is in the interests of most governments or policymakers to ensure that the capitalist system runs smoothly, since this provides employment for its voters and government income through taxes. Overall, we have a strong explanation for the

role of business within the policy process (compare with Lindblom, 1977). It is possible to take this argument to its extreme, to treat capitalist protection as a functional imperative, inevitable and immune from change, with agents merely functioning according to particular roles, with no real ability to make choices (see Hay, 2002: 116). In this scenario, education and healthcare systems exist to socialize and maintain a functioning workforce. However, the problem with such arguments is that it is possible to interpret all policy developments in this way without evidence or demonstrating how the process works (John, 1998: 96). When we ‘unpack’ the process, we find two problems. First, many policies in many countries appear to reflect the beliefs of governments or the power of unions, such as the welfare state, the living wage, or measures to protect union rights. Second, we can identify several types of capitalist class that may not share the same interests. For example, the issue of interest rates may divide the manufacturer and the banker. The same can be said for ‘globalization’, an important but vague term to describe the international spread of processes that used to be confined to a small number of countries. Economic globalization describes global or deregulated financial markets, technology which allows greater interactions – such as trade – across the globe, the role of ‘global corporations’, migration, and/or the unequal relationship between the Global North and South (Hill, 2005: 45; John, 1998: 103; Gangopadhyay, 2017; Hopper et al., 2017). It also describes the social and cultural integration that tends to follow greater interaction between countries and their populations (Parsons, 1995: 242–43). Globalization could describe what happens when nation-states experience a diminished ability to control their own economic and monetary policies and the actions of large multinational corporations (MNCs). Governments appear to be forced to compete with each other, react to shifts and crises in international financial conditions, and change to attract business from MNCs. The focus is on economics determining politics, with moves from the 1980s towards ‘deregulation, privatization and cutting back welfare’ representing a

worldwide response to globalization (John, 1998: 103; Friedman, 2000; Quiggin, 2006: 536). Such arguments are incomplete unless they identify a clear causal process and some evidence (Hay, 2002: 114). For example, we may identify an inevitable ‘race to the bottom’ as governments – acting on the belief that MNCs will not invest in countries with high taxes, a public sector-dominated economy, and restrictive labour and environmental regulations – reduce the size of the public sector and corporation taxes and deregulate to attract foreign direct investment. Yet this race has not run in a uniform way and, for example, the overall size of the public sector has not diminished (Hay, 2006b: 591). Rather, different policymakers in different countries have reacted to their environment in different ways (Hoberg, 2001: 128– 30; Hay, 2002: 253; Quiggin, 2006: 537). Policymakers are constrained by the need to maintain a reasonable balance between revenue and spending (the budget) and imports and exports (the balance of payments) (Quiggin, 2006: 530–35). However, there are many ways to attempt to do so, including running a deficit in one term of office, on the assumption that a future government will be obliged to produce a surplus (Quiggin, 2006: 540). Public expenditure as a percentage of GDP also varies, quite markedly, over time (Hogwood, 1992a: 43; Hindmoor, 2003: 208). More significantly, it fluctuates by policy area, as a combination of inertia and political choices produces minimal change in most areas but profound change in others (John and Margetts, 2003: 421; Chapter 9). Therefore, empirical policy studies suggest that ‘structural forces’ are important but they do not determine political behaviour; there is always some degree of choice. We may agree that one choice seems inevitable because it seems much better than the rest. Or, the appearance of constraint and inevitability may be convenient for policymakers attempting to introduce unpopular policies or avoid responsibility for poor results (Hay, 2002: 259). However, there is choice nonetheless and ‘the constraints of globalization are as much as anything else, what political actors make of them’ (Hay, 2006b: 587). Further, a key reason for country-level policies to stay distinctive is that they are influenced by other structural factors,

‘including distinctive national values, different political institutions, and the legacy of past policies’ (Hoberg, 2001: 127; Sinclair, 2004). Yet we should strike some note of caution: most of these studies are of the ‘West’ or Global North. As Chapter 12 suggests, policymakers in the South often face more political and economic pressures to reform to become more like the international norm (see also Kiely, 2017: 123).

INHERITANCE BEFORE CHOICE, AND POLICY SUCCESSION The ‘legacy of past policies’ is captured best by Rose (1990), who describes the cumulative effect of decades of policies: newly elected policymakers inherit a huge government with massive commitments. Most new decisions are based on legislation that already exists, and the bulk of public expenditure is spent on government activities that continue by routine, carried out by public employees recruited in the past. Current policymakers choose to ‘uphold the laws of the land’ before making new ones, and the effect of their new policy choices is small compared to the sum of government activity (1990: 263). Much current government policy results from the choices made by former policymakers, and policy-makers ‘choose’ to address the ‘consequences of inherited programmes that would not have been chosen by the current incumbents’ (Rose, 1990: 264, 229). Note the difference between inheritance and incrementalism (Chapter 4). Rose and Davies (1994: 31) suggest that incrementalist analyses tend to be relatively short term; longer-term analyses show that marginal steps made in the past have produced ‘massive commitments’ such as the modern welfare state and tax systems (or the breakdown of such systems – Richardson, 2018). These policies are maintained and reproduced by government organizations that operate without requiring further authorization. The logical implication of bounded rationality is that when policymakers pay attention to one issue they must ignore 99 others. When they invest a significant amount of effort and attention to policy change on one issue, they accept that they cannot do the same for

most others (Rose, 1986; compare with Chapter 9). Therefore, most of the day-to-day delivery of policy takes place without significant policymaker involvement. Most public sector organizations continue to implement policies that were legislated and budgeted for in the past (Rose, 1990: 264). ‘Change without choice’ describes external effects, such as demographic changes prompting new demands for services, and almost automatic changes in budgets (1990: 285). This argument supplements Rose’s (1984) suggestion that a new party in government may not make a major difference. They inherit legislation and a lot of draft legislation that tends to progress regardless of the party in government (Rose and Davies, 1994: 133). This theme of inertia is reinforced by ‘policy succession’, as ‘the replacement of an existing policy, program or organization by another’ (Hogwood and Peters, 1983: 1). While it may look new, and is different from ‘policy maintenance’, it is ‘directed at the same programme and/or clientele’ (1983: 18). Succession is likely for three main reasons. First, the size and scope of the state is large. Second, existing policy is often ‘its own cause’ (Wildavsky, 1980), since the delivery of policy often throws up problems that command the time of policymakers (compare with policy feedback theory – Mettler and SoRelle, 2018; Béland and Schlager, 2019). Third, the level of existing commitments is high and there is little scope to increase tax income (through growth or higher taxes) to fund new programmes (Hogwood and Peters, 1983: 2–5). Succession is generally more likely than innovation because the conditions for the introduction of policy are already in place: the issue already has legitimacy, the resources are in place, and policy has an established clientele (Chapter 2). These arguments are not about no choice. For example, Rose (1990: 288; Rose and Davies, 1994: 5) argues that inherited commitments can be moulded to suit current problems, some programmes can be terminated or introduced, and the policy innovations made during the administration may have a long-term, cumulative effect. Further, many of the policy successions described by Hogwood and Peters (1983: 30), such as the introduction of the National Health Service in the UK, were significant (in fact, they describe this as a hybrid of succession and innovation). Similarly, the

significance of a decision to reform welfare policies by spending the same amount but changing the balance between recipients may not be captured well by the term ‘consolidation’ (1983: 66). Rather, it shows us that any such reform may take a ‘huge political effort’ to produce ‘marginal improvements’ (1983: 129). It will not involve new money and will be carried out by the organizations that policymakers inherited (compare with Box 5.2 on ‘drift, layering, conversion, and displacement’, and Chapter 11 (pp.193–4) on first- and second-order change).

THE EVOLUTIONARY METAPHOR: CONTEXT AS A POLICYMAKING ENVIRONMENT Many theories draw on evolutionary metaphors to describe the interaction between actors and their environment. This approach can be useful in the abstract, to identify the key dynamics that might constitute that environment, and prompt us to consider why a narrow focus on elites will not tell us much about policymaking. For example, John (2003: 495) uses ‘evolutionary theory’ to capture the dynamics of the ‘five core causal processes’, described by Heikkila and Cairney (2018: 303–4) as follows: 1. Actors making choices. There are many policymakers and influencers spread across many types of policymaking venues (Chapter 8). If we only focus on a small group of elites, at the ‘centre’ of government, we miss the spread of policymaking responsibility across a political system. 2. Institutions. Each venue contains its own formal and informal rules governing behaviour (Chapter 5). Formal rules place limits on the roles of policymakers, and our explanation of their behaviour is incomplete if we do not consider the often-unspoken rules that they follow. 3. Networks. Each venue can produce its own networks of policymakers and influencers (Chapter 9). Even in highly centralized systems, there is a strong logic to the formation of policy ‘communities’. Policymakers at the ‘top’ must ignore and

delegate responsibility for most of the issues for which they are responsible. Bureaucrats tend to rely on specialists to help them process policy, and they often work closely together, building up trust and shared ideas about policy problems. If so, the lines between formal responsibility and informal influence are blurry (Chapter 8). 4. Ideas. Participants in each venue draw on a dominant set of beliefs about the nature of policy problems and the acceptable range of solutions (Chapter 11). Classic examples range from a broad adherence to capitalist principles in many countries, to the sense in some countries that some ideas seem off limits (e.g., when UK policymakers take for granted, but US policymakers reject, a ‘National Health Service’). 5. Social and economic factors, and events, help set the policy agenda and limit policymakers’ abilities to address and solve policy problems. Events can be routine (such as elections) or non-routine (such as social or natural crises). Combined, these factors produce the broad sense that environmental influences on policymaking often seem to be overwhelming or out of the control of individual policymakers. Their environments constrain or facilitate individual action. Since this process relates to an evolutionary metaphor, it is worth being clear on how evolutionary language translates into policymaking dynamics. For example, Lewis and Steinmo (2008: 15– 16) suggest that ‘all living things – including humans – want to pass on their genes’ and they ‘inherit powerful instincts to follow social rules’ to help them do so. In other words, they combine their desires to pursue self-interest and cooperate with others to protect themselves and their kith and kin (Cairney, 2013b: 289). In politics, the equivalent of passing on genes is to ensure the survival of memes (Dawkins, 1976; John, 2003: 493; Kerr, 2002), to help communicate and protect our beliefs across generations. The process involves three key evolutionary mechanisms: ‘Variation’ refers to the different rules adopted by different social groups to foster the collective action required to survive.

‘Selection’ describes the interaction between people and their environments; particular environments may provide an advantage to some groups over others and encourage certain behaviours (or, at least, some groups may respond by adapting their behaviour to their environment). ‘Retention’ describes the ways in which people pass on their genes (memes) to ensure the reproduction of their established rules. (Cairney, 2013b: 289) In other words, we are trying to identify the selection mechanism that causes some ideas to be successful while most fail. Evolutionary theory requires the identification of competing memes; selection and deselection processes; changes to policy memes by political actors seeking to adapt their aims to current circumstances; and the ‘random mutation of ideas and their fortuitous success in particular environments’ (John, 1999: 47). In other words, there are many possible ways to understand the world and produce many different rules on cooperation. However, policymaking environments advantage some groups and rules over others, and the maintenance of rules over generations is imperfect. As Chapter 5 describes, it is unclear how people produce, maintain, challenge, or pass down rules across generations. The use of evolutionary language in policy scholarship is also imperfect. Its meaning is contested strongly (Dowding, 2000), and too many scholars use the language too loosely, treating evolution as a vague synonym for change, or describing very different processes, including: 1. Gradual change in policy or institutions, or improvement over a long period. 2. ‘Punctuated equilibrium’, or a major disruption in policy or the institutions used to process policy. 3. The ‘coming together of multiple factors to create the conditions for major policy change’, such as a ‘perfect storm’ or ‘window of opportunity’. 4. The ‘trial-and-error strategies adopted by actors, such as policy entrepreneurs, when adapting to their environment’ (2013b: 279).

This lack of clarity will not go away, partly because it reflects wider debates on the meaning and mechanisms of evolution in nature (Cairney, 2012a: 266–69). Therefore, the onus is on us to think critically about the meaning of evolution in each case. For example, multiple streams analysis (MSA) (Chapter 11) describes the slow and gradual improvement of policy solutions towards political feasibility. In contrast, punctuated equilibrium theory (PET) (Chapter 9) describes long periods of political stability and policy continuity followed (sometimes) by instability and rapid and profound change. While MSA describes the need to ‘soften’ solutions to make them acceptable to their audience in one venue, PET describes the potential for venue shopping and shifting responsibility to a more sympathetic audience (Cairney, 2013b: 282). On the crucial role of ‘policy entrepreneurs’, Kingdon (1984: 173) describes ‘surfers waiting for the big wave’ rather than controllers of the sea. In other words, entrepreneurs possess the skills to exploit decisions made by other actors, not to control them directly (Lustick, 2011: 204). Drawing on rational choice theory (Chapter 7), John (1998: 184) suggests that their strategies change as actors learn to cooperate with each other and adapt to their environments. They learn, and their preferences change, when they formulate and adapt ideas. This process does not end when policies are selected at the formulation stage. Rather, policy changes further as actors – more powerful at the point of delivery – influence the progress of policy. John (1998: 186) describes the process as ‘structured evolution’ since the institutions, networks, and socioeconomic constraints that affect behaviour are more static than strategies and ideas. Box 6.3 The policy environment and ‘evidence-based policymaking’ This focus on policy environments helps us understand why the prospect of ‘evidence-based policymaking’ (Box 2.2, p.19) is so unrealistic. Chapter 4 already questions its possibility by describing why individual policymakers must disregard most evidence. Further, each of the five environmental factors provides a sense of the immense scale of the task. Consider the advice

that someone would give you if you sought the greater use of your evidence in policy (Cairney and Oliver, 2019; Oliver and Cairney, 2019). First, if there are so many potential authoritative venues, you will need to devote considerable energy to finding where the ‘action’ is. Second, even if you find the right venue, you will not know the unwritten rules unless you study them intensely. Third, some networks are close-knit and difficult to access because bureaucracies have operating procedures that favour some sources of evidence. You could be a privileged insider in one and be excluded in others. Fourth, if your evidence challenges an existing paradigm, you need a persuasion strategy good enough to prompt a shift of attention to a policy problem and a willingness to understand that problem in a new way. Finally, you can try to find the right time to use evidence to exploit a crisis leading to major policy change, but the spark for these opportunities is out of your control. Then consider how you could operate effectively in a complex system, in which the same action can be amplified or ignored, and outcomes seem to emerge locally, often with no central direction. In that context, policy studies recommend being entrepreneurial and investing your time to build up alliances, and trust in the messenger, gain knowledge of the system, and spot ‘windows of opportunity’ (Stoker, 2010; Weible et al., 2012; Cairney, 2016a: 124; Cairney and Oliver, 2017). However, they offer no assurances that this investment will pay off.

THE POLICY PROCESS AS A COMPLEX SYSTEM Scholars use complexity theory to identify complex systems across the natural and social sciences. Broadly speaking, the approach is based on a rejection of reductionist approaches to science, and there is a strong emphasis on the absence of linear or orderly processes. It seeks to explain why complex or system-wide behaviour emerges from the interaction between ‘large collections of simpler components’ (Blackman, 2001; Kernick, 2006; Mitchell, 2009: x; Geyer and Rihani, 2010). Cairney (2012b: 348) identifies five common assumptions regarding complex systems:

1. The whole is greater than the sum of its parts. A complex system cannot be explained merely by breaking it down into its component parts. Instead, we must shift our analysis to the system as a whole, as networks of elements that interact, share information, adapt, and combine to produce systemic behaviour. 2. Non-linear dynamics. Complex systems provide ‘feedback’ to actions: dampening some (negative feedback) and amplifying others (positive feedback). As a result, small actions can have large effects and large actions can have small effects. 3. ‘Strange attractors’. Although unstable, systems often display regularities of behaviour. ‘Punctuated equilibrium’ sums up the potential for long periods of stability disrupted by short bursts of change. 4. Sensitivity to initial conditions. Initial events or decisions often produce a long-term momentum. The ‘butterfly effect’ captures the idea of small factors producing large consequences and a profound cumulative impact. 5. ‘Emergence’ in the absence of central control. Behaviour and outcomes result from the interaction between elements at a local level rather than central direction. For example, swarming behaviour in bees cannot be explained merely by the actions of individual insects. Rather, we must study their interaction, the rules they follow, how they communicate those rules, and the extent to which a minor change in rules causes a large change in outcomes. The brain is also a complex system in which emergent processes, such as thoughts and feelings, are difficult to break down into the performance of individual cells and neurons. The brain also makes it difficult to treat political and social systems in the same way as natural systems. The human ‘capacity to reflect and to make deliberative choices and decisions among alternative paths of action’ makes the social world a different object of study (Mitleton-Kelly, 2003: 25–26). For example, social science complexity studies are less deterministic than physical models tracing outcomes back to a specific cause like the ‘butterfly effect’ (Box 6.4).

However, complexity theory’s application to public policy is not obvious, and ‘complexity’ – like ‘evolution’ – is often used very loosely as a metaphor or to mean complicated (2003: 26; Kernick, 2006: 389; Bovaird, 2008: 321). To make sense of ‘complexity’, we need to describe the key dynamics of a complex policymaking system. The whole is greater than its parts. Individual policy actors are one small part of a much larger system over which they have limited control. Jervis (1998: 5–6) suggests that: ‘We are dealing with a system when (a) a set of units or elements is interconnected so that changes in some elements or their relations produce changes in other parts of the system, and (b) the entire system exhibits properties and behaviours that are different from those of the parts.’ The metaphor of a telescope is useful to describe ‘zooming out’ from individuals to consider the system as a whole (Cairney and Geyer, 2015: 2). Non-linear dynamics and strange attractors. PET uses this language to explain policymaking stability and instability (Chapter 9). The ‘general punctuation hypothesis’ describes the importance of ‘feedback’ when policymakers process information. Jones and Baumgartner (2005: 7) define information processing as the ‘collecting, assembling, interpreting and prioritizing [of] signals from the environment’. Policymakers are surrounded by an almost infinite number of ‘signals’ – or information that could be relevant to their decisions – from their environment. Since policymakers are boundedly rational (Chapter 4), they do not have the ability to process all signals. Instead, they must simplify their environment by ignoring most information and issues (negative feedback) and promoting a few issues to the top of their agenda (positive feedback) (Baumgartner, 2017). Negative feedback may produce prolonged periods of equilibrium; existing policy relationships and responsibilities are more likely to remain stable, and policy is less likely to change, when the issue receives minimal attention from policymakers. Positive feedback may produce policy ‘punctuations’ because, when policymakers pay a disproportionate amount of attention to an issue, it is more likely that policy will change dramatically.

In a complex system with so many actors, non-linear dynamics may be found in many places, such as when actors in different venues pay maximal or minimal attention to an issue, or implementers pay unusually high attention to some directives and ignore others. Yet, systems also display regularities of behaviour over extended periods (Bovaird, 2008: 320). We might call such regularities ‘strange attractors’ and perhaps relate their cause to regularities in ways of thinking – summed up by the idea of organizational or political ‘culture’ – or the relatively stable role of powerful interests (Givel, 2015: 71–74). However, this term seems relatively vague and less used in policy studies (Tenbensel, 2015: 372). Sensitivity to initial conditions resembles the ‘path dependence’ described by historical institutionalism (Chapter 5). A ‘critical juncture’ is the point at which certain events and decisions led to the development of an institution. The timing of these decisions is crucial, because it may be the order of events that sets institutional development on its path. When a commitment to a policy has been established and resources devoted to it, over time it produces ‘increasing returns’ and it becomes increasingly costly to choose a different path (Pierson, 2000a). The link between complexity theory and historical institutionalism is strong: both identify the same sense of inertia and unpredictability, as relatively small events or actions can have a huge and enduring effect on policy change, and both use the black-and-red ball analogy and examples such as the QWERTY keyboard, outlined in page 82 (Pierson, 2000a: 253; Room, 2011: 16; 2016). Complexity theory may add a new dimension to the study of institutions, exploring the potential for unexpected instability and dramatic policy change even when the rules seem to change in a minor way (like a sudden shift of direction of a swarm of bees, or an attention shift in the brain). Emergence requires the most careful translation since there is no direct political equivalent to the complete absence of central control. For example, in biology, we may identify molecular interaction with no equivalent of a ‘centre’. In politics, there is at least one centre – central government – and so we need to describe (1) the emergence of policy despite a centre, or despite attempts by central

governments to control outcomes, and (2) the role of many possible ‘centres’ or authoritative sources of policymaking (Cairney and Geyer, 2015: 3). For example, complexity theory is one of three approaches exploring the idea of ‘multi-centric’ policymaking (Cairney et al., 2019a), alongside the Institutional Analysis and Development (IAD) framework (Chapter 7) and multi-level governance (MLG; Chapter 8). Further, emergence in the absence of central government control is a key feature of the implementation and governance literatures. Central governments face problems when they do not recognize the extent to which policy will change as it is implemented (Chapter 2). The level of interdependence between governments and implementing organizations has prompted the identification of ‘selforganizing networks’ (Rhodes, 1997: 50) and images of ‘bottom-up’ implementation through self-selecting clusters of organizations in which a variety of public and private organizations cooperate. Such arrangements have often prompted governments to embrace ‘new public management’, or the application of private business management methods to the public sector, to try to impose order through hierarchy and targets (Geyer, 2012; Chapters 8 and 12). However, implementation structures may not be amenable to such direct control, even if the organizations, responsible for delivery, are completely on board with central policy (Teisman and Klijn, 2008: 294; Bovaird, 2008: 339). Constant reforms of service delivery functions may be futile unless we recognize the nonlinear nature of policymaking. This recommendation is one of many possible practical lessons for policymakers, including (Teisman and Klijn, 2008: 288): 1. A policy that was successful in one context may not have the same effect in another. General behaviour is difficult to identify because the policy process is ‘guided by a variety of forces’. X will only influence Y under conditions that are difficult to specify. We need to know why it was successful in that instance, but there is so much interaction between so many variables that we cannot account for them all (or their separate effects).

2. Complex systems are difficult to control. They have ‘selforganizing capacities’, and the effect of the same central government intervention may be large or small. Policymakers should not be surprised if their interventions do not have the desired effect. 3. Find ways to adapt to a changing landscape. The metaphor of the ‘fitness landscape’ – or ‘surroundings in which living beings exist and behave’ – provides the context for the choices of agents. It is unstable and often rapidly changing. Therefore, actors must adapt quickly and not rely on a single policy strategy (2008: 289; Mitleton-Kelly, 2003: 35–36). Complexity scholars tend to recommend that central government policymakers let go of the idea of control, to experiment more, and give more local actors the discretion and flexibility to act. Sanderson (2009: 706) suggests that we do not know exactly how any policy measure will make a difference. Therefore, policymakers should be careful when making interventions, making greater use of ‘trial and error’ and learning from pilot projects (2009: 707), or giving implementing organizations the chance to learn from their experience and adapt to their environment (Bardach, 2006: 353; Haynes, 2008: 326). Such advice takes us back to the discussion in Chapter 4 on Lindblom (1959: 86), who rejects the idea of top-down radical change because policymaking is ‘a very rough process’ and no one knows ‘enough about the social world to avoid repeated error in predicting the consequences of policy moves’. Therefore, a ‘wise policymaker’ anticipates error and unintended consequences and avoids too-radical moves. See also Dunlop and Radaelli (2015: 140), who propose ‘regulatory humility’ to avoid accepting an ‘illusion of control’. Yet it is difficult to expect policymakers to accept such advice. For example, in Westminster-style systems, national elections revolve around the idea that we can hold elected politicians to account because we know they are in charge of policymaking. In that context, they need to link their image of governing competence to the idea that they are in control (Cairney, 2015a). Other forms of accountability – built on the idea of multi-centric policymaking, ‘co-

produced’ policy, and shared accountability – may exist, but they remain difficult to defend in the world of highly competitive and adversarial politics (Cairney et al., 2019a). Box 6.4 Are complex systems theories deterministic? A common complaint regarding systems theories is that they describe systemic logics that ‘operate in some sense independent of – and over the heads of – the actors themselves’ (Hay, 2002: 102). The danger is that, if the complex system is predominantly the causal factor, then we lose sight of the role that policymakers play; there may be a tendency to treat the system as a rule-bound structure, which leaves minimal room for the role of agency. What we need is an understanding of how agents perceive their decision-making environments; how they reproduce, accept, or challenge the structural, institutional, and wider systemic constraints that they face when making decisions. Policy studies seek to explain why different policymakers make different decisions under the same circumstances. If we are being sympathetic to complexity theory, we can treat its approach to structure and agency like Gidden’s ‘two sides of the same coin … If we look at social practices in one way, we can see actors and actions; if we look at them another way we can see structures’ (Craib, 1992: 3 on Giddens, 1984). On one side, we have ‘self-organizing landscapes’ (Teisman and Klijn, 2008: 289) or complex systems that adapt and change behaviour; behaviour is ‘emergent’ from the processes within systems and is not readily broken down to the agents within it. Further, much of the explanation for outcomes comes not from individuals but from the level of connectivity between them (Mitleton-Kelly, 2003: 28). On the other, we have the ‘self-referential behaviour’ of agents, reacting to ‘external forces and changes’ but also ‘creating their own perception of what they want and how to behave in the landscape they are in’ (Teisman and Klijn, 2008: 289). Some actors are more powerful than others because they are more able to adapt to the ‘fitness landscape’ (Room, 2016). If we are less sympathetic, each side appears to contradict the other: a focus on

separable, independent actions by agents appears to contradict the idea that a complex system cannot be broken down into its component parts (Cairney, 2012b). Perhaps this is the meaning of ‘two sides of the same coin’: two arguments that represent each other’s opposite. This problem of conceptualizing structure and agency is not unique to systems theory. Indeed, a common theme throughout the book is how to attribute explanatory power to entities that do not act but appear to influence the actions of agents.

CONCLUSION The strongest structural accounts suggest that powerful external forces constrain the ability of individuals or governments to make decisions, or that socioeconomic change determines policy change. Yet few contemporary theories take this approach. Rather, we can identify a range of approaches that try to strike the right balance between structure and agency. The DSH approach measures the association between socioeconomic factors and policy outputs. Marxist accounts explore the strong imperative for governments to support the capitalist system and therefore the interests of the classes that benefit most from that system. Globalization suggests that governments compete to protect their economy and secure foreign direct investment. Inheritance before choice suggests that governments inherit massive policy commitments and tend to change policy only at the margins, while the pervasiveness of policy succession reflects the constraints on policy innovation and termination. Complexity theory suggests that policymakers are part of a large complex system that seems to defy control. A focus on structural factors is really about the wider context in which policymaking takes place. It allows us to ‘zoom out’ from a focus on individual policymakers and their centrality to policymaking, towards thinking of the policymaking environment or system in which they operate. Broadly speaking, the environment is a metaphor to bring together key concepts – actors, institutions, networks, ideas,

socioeconomic factors and events – and then consider how specific theories use them to provide an overall explanation of policymaking. The advantage of such approaches is that they highlight the context within which policy is made. We know that policymakers make choices but recognize that policymaking does not begin with a blank slate or operate in a vacuum. Structural factors may influence what they pay attention to and how they act. There may be events and policy conditions outside of their control. There are limits on their ability to anticipate the effects of policy and the extent to which it is carried out successfully. They inherit a ‘ship of state’ that behaves more like a super-tanker (notoriously difficult to turn around) than a rowboat. In short, this discussion is a useful corrective to the idea that we could ever explain policymaking well by focusing on a small group of policymakers at the centre. The disadvantage is that such accounts often appear to favour structural explanations at the expense of a focus on agency. Some Marxist accounts highlight inevitability rather than choice. Some theories of globalization do not explain fully how policymakers interpret and react to their environment. Some accounts overemphasize the vague notion of evolution at the expense of careful analysis, or suggest that ‘self-organizing’ complex systems almost have a life of their own. It is difficult to present a convincing account of policymaking purely in terms of context and the constraints that policymakers face. We must also explain how and why they act. In some cases, policymakers have many choices even when the decision-making context appears to provide only one realistic option. We may often feel that policymakers have no choice, but a more accurate description is that ‘the best course of action seems obvious’ (Dowding, 1996: 44). While it may seem like a fine distinction, it helps separate social science contingency from the deterministic physical and natural sciences. Most modern theories try to conceptualize this dynamic process in which structures influence agents and agents mediate or reconstitute the structural constraints within which they operate. They identify how socioeconomic factors constrain behaviour and policymakers mediate these factors by interpreting or weighting their significance in different ways. PET shows how policymakers process information

in complex systems (Chapter 9). MSA shows how policy entrepreneurs navigate policymaking environments (Chapter 11). The advocacy coalition framework seeks to incorporate the DSH approach into a wider analysis. Its flow diagram (Figure 10.1) includes a discussion of ‘relatively stable parameters’, such as the ‘basic distribution of natural resources’, and ‘external events’ such as ‘changes in socioeconomic conditions’, but it also seeks to explain the ‘black box’ by identifying the ability of advocacy coalitions to interpret external effects and compete to define the policy problem within subsystems. System – ‘integrated group of interacting elements designed jointly to perform a given function’ (such as a political system that functions to ‘allocate values’ – Dawson and Robinson, 1963: 267). Meme – the ideational equivalent of a gene: ‘an information pattern, held in an individual’s memory, which is capable of being copied to another’ (John, 2003: 493). It is used by Dawkins (1976) to explain why types of information – common sense, cultural references, phrases, technology, traditions, theories – ‘survive’ and are reproduced across generations. Reductionist – explaining complex phenomena in relation to their separate components.

7 Collective Action Problems in Public Policy

Key themes of this chapter: • • • • • •

•

Political science models and policy theories help us understand collective action problems. Rational choice theory and the Institutional Analysis and Development (IAD) framework help us identify why people may, or may not, cooperate with each other. Simple examples of game theory describe individual action undermining collective action, such as when ‘free-riding’ or contributing to the ‘tragedy of the commons’. The IAD helps compare the effectiveness of institutions designed to foster collective action. Most famously, it identifies the conditions under which ‘common pool resource’ tragedies can be averted by communal rather than market or state action. A key theme is institutional complexity and the overlaps between many different rules. ‘Polycentric’ governance describes many decision-making centres, responsible for their own action, and cooperating or competing with others. These approaches provide important ways of thinking about policy problems rather than universal solutions or explanations.

Political systems often represent the main venue for collective action. For example, government often involves coercion, through policy tools such as regulation and taxation (Chapter 2). Or, governments use ‘behavioural public policy’ which incorporates psychological insights to ‘nudge’ people into changing their behaviour (Box 7.1). In this chapter, we take a step back from such specific measures, to consider more generally why people act individually and collectively. Why do people agree, or refuse, to cooperate with each other to produce rules to solve policy problems?

What are the consequences for policymaking? Should market, state, or communal practices drive collective action? In that context, rational choice theory and game theory provide important ways of thinking. Through simple models and thought experiments, they help identify collective action problems: the potential for choices made by individuals to have an adverse societal effect when there is an absence of trust, obligation, or other incentives to cooperate. People may have collective aims or interests that require cooperation but individual incentives to defect. While the action of one individual makes little difference, the sum total of individual actions may be catastrophic. It may cause (1) ‘market failure’ when there is scope to free-ride and no incentive to purchase or contribute to public goods (such as constructing national defences or maintaining clean air) and (2) the ‘tragedy of the commons’ when individual actions combine to ruin common pool resources (CPRs) such as land, forests, and fish stocks (Hardin, 1968; Heikkila and Carter, 2017). Government intervention is one solution to collective action problems: if the effect of non-cooperation is that everyone is worse off, let’s make people cooperate. Yet such solutions are also problematic for several reasons. First, policymakers may act in their own interests rather than the collective interest described by governments. Second, no government is able to make decisions that suit everyone. Rather, governments coerce everyone (through taxes) to contribute to policies that benefit only some, and regulate one group to benefit another (Wilson, 1980: 419–22). Third, government solutions may be expensive and produce unintended consequences. Fourth, policymaking systems contain many venues with policymaking authority – ‘polycentric’ or ‘multi-centric’ governance – which also need to find ways to cooperate with each other (Cairney et al., 2019a). Overall, if we need rules to encourage individual citizens to cooperate, we also need rules to encourage policymakers. These issues are taken forward by modern policy theories that describe a combination of important factors: (1) bounded rationality (Chapter 4); (2) trust and norms; (3) institutional solutions to collective action problems; (4) many layers of rules to encourage

cooperation among many actors and ‘centres’ of government; and (5) the complexity and unintended consequences of such activities. Most importantly, the Institutional Analysis and Development (IAD) framework provides a language to study these dynamics. It focuses on institutions (Chapter 5) as vehicles for problem-solving and collective action. Actors develop and modify rules to aid collective action. These rules can be primarily state, market, or communal but are often a complicated mix of all three. Indeed, one aim is to compare different types of institutions and evaluate their relative success, using criteria such as economic efficiency and equity. Elinor Ostrom won the Nobel Prize in economics for her application of the IAD to the study of CPRs, in which she often found that communal arrangements were more successful than purely state or market solutions. However, these conclusions come with a massive dose of caution against drawing too many simple conclusions from many different and complex contexts. A huge number of studies of CPR management suggest that certain ‘design principles’ can help produce effective communal collective action but do not provide a blueprint. The IAD also contains a ‘family of theories’ that can be applied to related topics, such as the sustainability of Social-Ecological Systems (SES Framework), how to deal with ‘polycentric governance’ (Institutional Collective Action), and how to study and manage institutional complexity (Ecology of Games). Box 7.1 Nudge theory and behavioural public policy ‘Nudge’ sums up the idea that we can take insights from psychology (Chapter 4) to influence human behaviour. Rather than simply coercing people to change their own behaviour, or cooperate with others, we can shape the environment in which they make choices and reduce the cognitive burden of the choice we would like people to make (Thaler and Sunstein, 2008; compare with Box 4.3). It has – very quickly – become a ‘mainstream’ idea in economics and an essential ‘policy tool’ (Chapter 2) for many governments (John, 2018: 49, 6–7). The superficial attraction is clear: many politicians like the idea of

policy change without direct imposition (coercion can be electorally damaging), and citizens often ‘resist such commands and find ways to avoid regulations’ or ‘resist messages that come from authority’ (2018: 8, 29; compare with Jones et al., 2013). Consequently, many governments established their own ‘nudge units’ and – despite some concerns about the ethics or effectiveness of many interventions – there has been a major increase in the use of ‘behavioural public policy’ (John, 2018: 82). Still, there remains uncertainty about how to use it (2018: 145–6; compare with Strassheim, 2019). For some, it provides subtle manipulations by reducing the cognitive load of certain options. For others, it encourages the conditions under which people can process information efficiently, and deliberate together, to allow them to think about and co-produce choices. Overall, these approaches prompt the big normative questions in public policy: is there a common good, and how should people cooperate to achieve it? Domestically, we might ask: should governments coerce citizens to pay for public goods, do governments perform better than markets, and do we need big or small, centralized or polycentric, government? However, the solutions to many problems – such as climate change and global public health – are not in the gift of single countries. They take the possibility of central coercion out of the equation, prompting us to wonder if citizens and governments across the globe can solve collective action problems together or are doomed to catastrophic failure. These approaches also highlight an intellectual trade-off between (1) the search for parsimonious explanations of many cases and (2) the acceptance that the world is complicated and each case is unique. Can we find a happy medium between too-simple and toocomplex analysis of collective action problems?

COLLECTIVE ACTION PROBLEMS IN RATIONAL CHOICE AND GAME THEORY

Rational choice theory describes models and approaches which combine insights from mathematics, economics, and political science (Hindmoor and Taylor, 2015). ‘Rational’ describes the ability of individuals to apply reason while processing information. They identify and prioritize their preferences, recognize the constraints of the environment in which they operate, and make choices based on their beliefs about the best way to satisfy their preferences. Many models are deliberately abstract and simplified to the point of being unrealistic. Some assume ‘perfect’ rationality to predict the effect of different rules or environments on collective action. Therefore, they differ from our discussion of ‘comprehensive rationality’ (Chapter 4) in which we automatically treat these assumptions as false and examine the consequences of bounded rationality (Cairney, 2012a: 133–35). Classic rational choice theory is based on three principles: 1. Deductive reasoning. Create models based on a small number of propositions (‘axioms’ or ‘assumptions’) and examine their logical implications. 2. Methodological individualism. Explain political outcomes as the aggregate of the intentional actions of individual actors. 3. Instrumental rationality. Individuals pursue the best means to fulfil their preferences (which can include direct self-interest but also protecting their families or giving to charity) (Shepsle and Bonchek, 1997: 16). There are two main variants (Dowding and King, 1995: 1; Hindmoor, 2010; Hay, 2004a: 50). The first is an abstract exercise to predict behaviour when individuals act ‘optimally’ in different environments. For example, to maximize their utility, actors would need to process all relevant information and be able to rank-order their preferences consistently (Hindmoor, 2006a: 182; Griggs, 2007: 174; Dowding, 2010). The second involves more detailed assumptions on the preferences of individuals and how they relate to institutions. For example, much of ‘institutional rational choice’ (Chapter 5) is built on the premise that (1) each decision-making situation provides particular incentives for individuals to act and (2) institutions are the formal and informal rules that can be used to change those

incentives. Note the trade-offs between simplified models that produce potential generalizations about many instances and more accurate and detailed explanations that are more difficult to apply generally (Dowding and King, 1995: 16; Dowding, 1995a: 49; Dowding, 1995c: 82).

The Prisoner’s Dilemma, Logic of Collective Action, and Tragedy of the Commons Game theory examines the choices that actors make when situated within a strategic decision-making environment, faced with a particular set of ‘pay-offs’ and anticipating the choices of other actors. In the simplest games, the assumptions are that all ‘players’ are instrumentally rational (focused on the means to a desired end), each player knows that the others are instrumentally rational, everyone understands the rules of the game and the pay-off from each choice, and that everyone would make the same choice in the same circumstances (Hindmoor, 2006a: 106–7). Other games introduce greater levels of uncertainty or different assumptions about the motivations of actors (Harsanyi, 1986: 90). A key aim is to identify points of equilibrium when actors make a choice and stick to it, such as the ‘Nash equilibrium’ when players have made their best choice and there is no incentive to change (Chwaszcza, 2008: 145). Three famous examples follow, which you should treat as thought experiments rather than actual predictions of behaviour. Table 7.1 The prisoner’s dilemma

First, the ‘prisoner’s dilemma’ (Laver, 1997: 45–46). Two people are caught red-handed and arrested for a minor crime, placed in separate rooms, and invited to confess to a major crime (they both did it and the police know it but cannot prove it). The pay-offs (Table 7.1) are: if Paul confesses and Linda does not, then Paul walks free and Linda receives a ten-year jail sentence (or vice versa); if both confess they receive a much higher sentence (eight years) than if neither confesses (one year). Also assume that they derive no utility from the shorter sentence of the other person (a non-cooperative game). It demonstrates a collective action problem: although the best outcome for the group requires that neither confess (both would go to jail for one year), the actual outcome is that both confess (and spend eight years each in jail). This point represents the Nash equilibrium since neither would be better off by changing their strategy unilaterally (Linda assumes that if she stays silent, and Paul confesses, Linda gets suckered into ten years). The effect of Paul and Linda acting rationally as individuals is that they are worse off than if they had cooperated. Both ‘defect’ when they should ‘cooperate’. The structure, pay-offs, and outcomes differ in other games. In the ‘chicken game’, in which we imagine two people driving towards each other, there are greater costs when both defect (death) and more likelihood that at least one will cooperate (by getting out of the other’s way). In many ‘assurance games’, there is a greater benefit to unilateral cooperation and/or a greater expectation by each player that the other will also cooperate (McLean, 1987: 127; Ostrom, 1990: 42; Dowding and King, 1995: 8; John, 1998: 120–21; Hindmoor, 2006a: 109–11). In other words, individuals do not simply cooperate; the situation must provide an incentive and pay-off. Second, the ‘logic of collective action’. Olson (1965; 1971) argues that, as the membership of an interest group rises, so does (1) the belief among individuals that their contribution to the group would make little difference and (2) their ability to free-ride. I may applaud the actions of a group but can enjoy the outcomes without leaving my sofa, paying them, or worrying that they will fail without me or punish me for not getting involved (Chwaszcza, 2008: 156–57). Therefore, ‘unless the number of individuals … is quite small, or

unless there is coercion or some other special device to make individuals act in their common interest, rational, self-interested individuals will not act to achieve their common or group interests’ (Olson, 1971: 2; Hindmoor and Taylor, 2015: ch. 6). Third, the ‘tragedy of the commons’ (Hardin, 1968). The scenario is that a group of farmers share a piece of land which can only support so many cattle before deteriorating and becoming useless. Although each farmer recognizes the collective benefit to an overall maximum number of cattle, each calculates that the marginal benefit they derive from one extra cow for themselves exceeds the marginal cost of overgrazing to the group. Individuals place more value on the resources they can extract for themselves now than the additional rewards they could all extract in the future. The tragedy is that, if all farmers act on the same calculation, then the common resource will be destroyed (1968: 1244). As described, the group is too large to track individual behaviour, individuals have ‘high discount rates’ – placing more value on current over future consumption – and low mutual trust, with minimal motive and opportunity to produce and enforce binding agreements (Ostrom, 1990: 183). This ‘tragedy’ appears to reflect current anxieties about one of the big problems of our time: global CPRs are scarce and the world’s consumption levels are rising; there is no magic solution; and collective action is necessary but not guaranteed. We may value sustainable water, air, energy, forests, crops, and fishing stocks but find it difficult to imagine how our small contribution to consumption will make much difference. As a group, we may fear climate change and seek to change our ways but, as individuals, contribute to the problem (Sandler, 1997; 2004; compare with profound critiques of Hardin – Aklin and Mildenberge, 2018; Hale, 2018; Brinkley, 2019). Overall, these simple scenarios suggest that instrumentally rational individuals have weak incentives to cooperate even if it is in their interests and they agree to do so. Hardin (1968: 1247) argued that collective action is unlikely without ‘mutual coercion, mutually agreed upon’, amplifying a major and ongoing debate about the state’s responsibility for that coercion (Ostrom, 1990: 2; Boyd et al., 2018; Frischmann, 2018). These problems appear to undermine the ‘idea that cooperation naturally follows from mutual interests’ (John,

1998: 122), although they are exaggerated by certain assumptions, such as that the games are not repeated (Box 7.2). Box 7.2 Collective action during repeated games The nature of collective action problems changes if we modify game theory assumptions. Games like the prisoner’s dilemma are one-off and self-contained, but in many games the players know that there are wider, longer-term consequences to defection (Ostrom, 2011: 14). Describing ‘nested games’, Tsebelis (1990) posits that the behaviour of individuals often seems suboptimal in one game until we recognize their involvement in a series of others. It may be optimal to act ‘irrationally’ in the short term to support a longer-term strategy. These games play out differently if they are connected to other games with different rules and payoffs, and/or the participants know they will be repeated. Long (1958: 251–53) describes an ‘ecology of games’, in which: • • • •

many games take place at roughly the same time in local communities, including buying and selling, banking, producing media, and ‘civic organization’ (1958: 253) ‘players in each game’ know and follow their distinct rules, but also ‘make use of players in the others for their particular purpose’ (1958: 253) these games overlap and interact to produce ‘unintended but systemically functional results for the ecology’ (1958: 251) key players foster ‘a vague set of commonly shared values that promotes co-operation in the system’ (1958: 251).

Axelrod (1984; 1986: 1087) uses evolutionary game theory to explore how behaviour changes over multiple games to reflect factors such as (1) feedback and learning from trial and error and (2) norms and norm enforcement (see Chapter 11 and Box 7.3). For example, player 2 may pursue a tit-for-tat strategy. She cooperates at first, then mimics the other player’s previous choice: defecting, to punish the other player’s defection, or cooperating if the other player cooperated. Knowledge of this strategy could

provide player 1 with the incentive to cooperate. Axelrod’s simulations suggest that norms develop when players enforce and expect sanctions for non-cooperation, and through socialization to discourage norm violation, while some norms become laws. Ostrom (2011: 15) extends the analysis to real-world settings, in which repeated face-to-face meetings affect how people make individual calculations.

GOVERNMENT AS ONE INSTITUTIONAL SOLUTION TO COLLECTIVE ACTION PROBLEMS Collective action problems underpin a convincing argument for the role of ‘institutions’. However, institutions come in many forms and encourage many types of behaviour, from the informal enforcement of norms to the governments with formal authority to tax, redistribute income, and spend on public services (Dowding and King, 1995: 10). Government can be a key solution to many collective action problems but is also problematic for four reasons. First, similar collective action problems relate to policymakers, who may act as individuals pursuing their own preferences, often at the expense of wider collective action (Buchanan, 1988: 13; Ostrom, V. and Ostrom, E., 1971). Classic examples include: • • •

Bureaucrats more interested in themselves than policy outcomes (Niskanen, 1971; Dunleavy, 1985, 2019). ‘Rent-seeking’, which relates to the ‘investment of resources by firms and pressure groups in the expectation of securing economic privileges’ from politicians (Hindmoor, 2006b: 87). ‘Political business cycles’, or the (not necessarily strong) incentive of incumbent governments to manipulate the economic cycle to make it boom before an election (Schultz, 1995: 81; Hindmoor, 2006a: 46–47; Dubois, 2016).

Second, governments may coerce everyone to contribute to government resources (through taxation) but cannot make choices that suit everyone. There is no such thing as a ‘public will’ or

‘national interest’ that we can determine by merely adding up the opinions or preferences of the population. There are always winners and losers when policymakers make choices. One way to understand this point is via the idea that no voting system can aggregate individual preferences to satisfy the collective will unproblematically (McLean, 1987: 10, 26). Instead, there is continuous potential for instability related to intransitive electoral results and policy decisions, or ‘the selection of candidates or policies to which there are majority preferred alternatives’ (Riker, 1982; Hindmoor, 2006a: 80–87; Ward, 2002: 66; Ward and Weale, 2010). The famous ‘Arrow problem’ (or ‘impossibility theorem’) describes the impossibility of action to make everyone better off. Arrow (1963) seeks to establish a rule, to aggregate individual preferences into social preferences, which can satisfy certain criteria – achieving Pareto efficiency without Kaldor–Hicks compensation, with choice determined by all individual preferences rather than dictatorship – and finds that no such rule exists (Hindmoor and Taylor, 2015: ch. 5; Dewan et al., 2009: xxi). Third, government (or state intervention) may be expensive and produce unintended consequences, particularly when policymakers lack enough information to make good choices (Ostrom, 1990: 11). Consequently, we should keep an open mind about the relative benefits of many options to solve collective action problems. Solutions may be achieved by ‘communal property arrangements’ as well as state intervention (Feeny et al., 1990: 13). Fourth, policymaking tends to be ‘polycentric’, producing the need for many policymaking organizations to solve collective action problems across government (Cairney et al., 2019a). Overall, if we need rules to encourage individual citizens to cooperate, we also need rules to encourage policy-makers and governmental bodies .

INSTITUTIONAL ANALYSIS AND DEVELOPMENT (IAD) FRAMEWORK The IAD provides a comprehensive guide to addressing collective action problems. To begin with, it has a simple aim – articulated and

developed by Elinor and Vincent Ostrom – to understand ‘the strengths and limitations of diverse forms of institutional arrangements in different settings’ (Ostrom et al., 2014: 268; note that the Ostrom references in this chapter are to Elinor). It focuses heavily on ‘problem-solving’, in which actors produce, use, analyse, and refine institutions to aid effective collective action and produce their preferred outcomes (2014: 269). However, the IAD requires a major investment of time to understand its many components and relate them to specific issues such as diagnosing collective action problems in relation to CPRs (Heikkila and Andersson, 2018: 309). To begin, consider key terms and insights – described in Box 7.3 – as part of a wider framework. It promotes a common language, including a ‘universal set of variables’ and collection of rules, to help ask the right questions during theory development and empirical analysis, on the basis that there are too many variables and too much complexity to contain within a single theory or model (Ostrom, 1990: 214; Ostrom et al., 2014: 267–71; Schlager and Cox, 2018: 216; McGinnis, 2011). Rather, a ‘family of theories’ can be used to specify the relative importance of some elements to answer a research question, and many models can provide many precise accounts of the interaction between key variables (Ostrom, 2011: 9). As described, this approach seems to contrast with our initial focus on very simple rational choice games, but think of these games as some of many ways to use the concepts described by the IAD (2011: 8; Ostrom et al., 2014: 269). Box 7.3 Key Institutional Analysis and Development (IAD) terms Action arena and situation. The action arena contains action situations and actors. Action situations are ‘social spaces where individuals interact, exchange goods and services, solve problems, dominate one another, or fight’ (Ostrom et al., 2014: 271; McGinnis, 2011: 172). Key questions include: who are the actors, what are their positions, what action is allowed, how much information do they have, what are the known costs and benefits of each action, and how does this action relate to the wider

‘biophysical’ and ‘community’ context? More philosophically, we may ask: under what conditions do actors favour social solidarity over selfish action (Ostrom et al., 2014: 272–73; Ostrom, 2011: 11–2; Heikkila and Andersson, 2018: 313)? Institutions. The ‘rules, norms, and shared strategies that structure human behaviour and choices’ (2018: 310). They can be (1) formal, written, and widely understood or (2) informal, unwritten, and difficult to detect (Chapter 5). We can be describing the rules that are ‘collectively created, adapted, monitored, and enforced’ (Heikkila and Andersson, 2018: 310) or understandings that ‘exist in the minds of the participants and sometimes are shared as implicit knowledge rather than in an explicit and written form’ (Ostrom, 2007: 23). Rules. ‘Shared prescriptions (must, must not, or may) that are mutually understood and predictably enforced in particular situations by agents responsible for monitoring conduct and for imposing sanctions’ (Ostrom, 2007: 23). ‘Working rules’ are common knowledge – participants use them ‘to explain and justify their actions to fellow participants’ – and actors can enforce them on the assumption that all know them (Ostrom, 1990: 51; 2011: 18). ‘Rules in use’ are ‘followed and respected by people’, while ‘rules in form’ are ‘formalized and written down’ but not necessarily respected (Heikkila and Andersson, 2018: 314). Norms. ‘Shared prescriptions that tend to be enforced by the participants themselves through internally and externally imposed costs and inducements’ (2007: 23). Internal costs include feelings of guilt. External costs include ‘social displeasure’ (Ostrom, 1990: 206). Strategies. ‘The regularised plans that individuals make within the structure of incentives produced by rules, norms and expectations of the likely behaviour of others’ (Ostrom, 2007: 23). Common pool resources (CPRs). Generally refers to finite natural resources with two properties: (1) it is difficult to exclude actors

from enjoying their benefits and (2) their use by one diminishes their value to another. Examples are groundwater basins, fish stocks, forests, and land to grow food (Heikkila and Carter, 2017). Evaluating outcomes. It is difficult to define success, and all measures come with trade-offs. Possible criteria include that outcomes were (1) economically efficient, (2) equitable (according to contribution/reward, or ability to pay), (3) in line with actors’ values on how to reward cooperation and punish defection, or conducive to (4) high-trust/low-conflict relationships, (5) effective accountability mechanisms, or (6) CPR sustainability (Ostrom, 2011: 16; Cox et al., 2010: 40; Baggio, 2016: 421). Self-governance. People often have the ability to cooperate, to design institutions, and to solve collective action problems without state intervention. The IAD analyses the conditions under which ‘bottom-up’ action is more effective than ‘command-and-control’ from central government, and how both approaches interact in practice (Heikkila and Andersson, 2018: 311–19; Schlager, 2004: 159; Schlager and Cox, 2018: 216). Polycentricity describes (1) ‘many decision centers’ with their own separate authority, (2) ‘operating under an overarching set of rules’ but with (3) a sense of ‘spontaneous order’ in which no single centre controls the rules or outcomes (Aligica and Tarko, 2012: 237, 254; Heikkila et al., 2011: 123). Polycentric governance describes ‘policymaking centres with overlapping authority; they often work together to make decisions, but may also engage in competition or conflict’ (Cairney et al., 2019a: 2). Multiple centres can be found across many organizations or within one. To understand a problem, who seeks to solve it, and how, we need to examine ‘multiple action arenas’ containing different actors (Ostrom et al., 2014: 271). We can use the IAD to combine several elements of this book. Institutional rational choice describes the rules that provide

incentives and sanctions to guide individual and collective practices (Chapter 5). Policymaking complexity describes the many actors, institutions, networks, ideas, and socioeconomic factors and events that combine to produce policy outcomes (Chapter 6). Multi-level governance describes the absence of one single government in control (Chapter 8) and the – necessary and inevitable – presence of ‘polycentric’ governance (Box 7.3). Altogether, we can say that institutions might foster collective action, but there are many available types, understood in different ways by many actors, and they coexist in a complex policymaking environment; multiple layers and types of rules are difficult to reconcile and use collectively. If so, how can people come together to generate the sense of a common aim, and cooperate to produce the rules to shape their collective practices?

KEY APPROACHES IN THE IAD’S EXTENDED FAMILY Since the IAD contains a ‘family of theories’ with a problem-solving bent, let’s try to understand it through the lens of important problems, including how to manage CPRs (Ostrom’s most famous work, providing a valuable lens through which to understand the IAD), polycentric governance, and institutional complexity. In the first two examples, the research has been led by the IAD’s matriarch, Elinor Ostrom. In others, such as Institutional Collective Action and the Ecology of Games, there is a clear family connection but more than one lineage.

Managing Common Pool Resources (CPRs) and Avoiding Tragedies We can understand the IAD through the lens of Governing the Commons (Ostrom, 1990) which explains how to challenge the ‘tragedy of the commons’ narrative and encourage more effective CPR management (Heikkila and Carter, 2017). The success of this

approach contributed to Elinor Ostrom winning the Nobel Prize in Economic Sciences in 2009: It was long unanimously held among economists that natural resources that were collectively used by their users would be overexploited and destroyed in the long-term. Elinor Ostrom disproved this idea by conducting field studies on how people in small, local communities manage shared natural resources, such as pastures, fishing waters, and forests. She showed that when natural resources are jointly used by their users, in time, rules are established for how these are to be cared for and used in a way that is both economically and ecologically sustainable. (Nobel Media AB, 2009) Ostrom (1990: 8–12) criticizes the uncritical use of rational choice games to conclude – too quickly – that disastrous collective action problems are inevitable and that we must secure major government intervention or ‘privatize’ commons and assign property rights (Ostrom, 1990: 12). These recommendations seem to rely on toosimple assumptions (Box 7.2) which prompt people to make sweeping statements about the need for one solution (Ostrom, 1990: 8–14). They also assume a highly informed, efficient, fair, and reliable government or market, and underestimate the unintended consequences of government intervention or costs of privatizing resources (Schlager, 2004: 148–49). Yet empirical studies highlight many examples of nationalized or privatized CPRs with disastrous effects (Ostrom, 1990: 23). Alternatively, what if there is much evidence that people often work collectively and effectively without extreme coercion? We can explore the relative benefits of more voluntarily cooperative arrangements, with people seeking agreements with each other that could be enforced – if necessary – by CPR users or a private rather than state authority. The commons would remain common, users would monitor each other’s behaviour, and, in some cases, they would sanction defections (1990: 17). On that basis, Ostrom (1990: 20–22) describes interlinked aims:

• •

Theoretical. To identify key variables and the conditions that help some groups or communities organize themselves to solve a collective action problem without state coercion or completely private property rights. Empirical. To code and compare many examples to identify their common elements (see also National Research Council, 1986).

Think of an iterative process of using the IAD to analyse case studies and using that analysis to build insights to enhance the IAD. It helps produce a long list of variables to explain success and failure in CPR management (Ostrom, 1990: 183). This approach requires intense and careful study to (1) gather enough information about vital factors, such as the nature of the CPR and ‘culturally acceptable rules’, (2) understand how specific rules work in specific contexts, and (3) account for inevitable variations in their impact on free-riding and cooperation. Ostrom (1990: xv) describes analysing a subset of 5,000 ‘small-scale’ cases if they reported four factors: the ‘structure of the resource system’, the ‘attributes’ and behaviour of the users (‘appropriators’) of CPRs, the rules they were using, and the outcomes resulting from their behaviour. In that context, the IAD serves initially as a framework, combining a language, way of thinking, and collection of methods and insights to apply to empirical cases (Ostrom, 2011). Then, we can develop specific theories or models to apply to particular fields such as CPR management. Ostrom (1990: 90; 140) describes eight (often mutually reinforcing) ‘design principles’ common to ‘long-enduring CPR institutions’, and further studies help clarify the principles (Cox et al., 2010: 15; Baggio et al., 2016: 431): 1. CPRs have ‘clearly defined boundaries’. Users (1) know what resource they are managing and (2) can distinguish ‘legitimate users and nonusers’ to exclude ‘outsiders’. 2. The rules are tailored to local conditions. Users know what they (1) are expected to contribute to management and (2) receive from CPRs. They enjoy high certainty about the likely costs and benefits of use.

3. The actors affected by the rules can also help shape them (preferably at low cost). 4. CPR monitors are users or accountable to users. They monitor (1) the conduct of users and (2) the state of the CPR. The costs of mutual monitoring tend to be low, and their consequences are felt relatively quickly, but such factors are contingent on context. For example, it is easier to monitor resource extraction than pollution (Stern et al., 2002: 462). New institutions often begin small and develop incrementally to minimize early set-up costs. 5. The penalties for rule-breaking are low if the choice is a one-off and understandable under the circumstances (to avoid alienating the user to the extent that they no longer believe in the rules). The penalties are high if the choice is part of a pattern which makes other users feel they are being treated like ‘suckers’ for following the rules, or if any rule-breaking would be catastrophic to the CPR. 6. Conflict resolution is frequent, rapid, and low-cost (to address the inevitable ambiguity of any rule which, if left unresolved, allows free-riding). 7. Users have the right to self-organize without external government intervention (or too much interference from hubristic public managers – Schlager, 2004: 167). 8. These rules and practices form part of ‘multiple layers of nested enterprises’. Many projects are connected geographically and at different scales – local, regional, national – hopefully in ways that do not undermine the function of individual projects (Heikkila et al., 2011). These design principles help explain why some communities manage CPRs successfully. They allow users to ‘adopt the same commitment’ and expect the long-term benefits to be relatively worthwhile (Ostrom, 1990: 186; Schlager, 2004: 145–46). However, there are three key reasons to not expect hard and fast conclusions to emerge. First, design principles contribute to – but do not determine – more fundamental requirements for collective action. For example, it requires the CPR itself to be conducive to effective design: good

management would improve its value to users, and users know enough about the CPR to predict its contribution to their livelihood and the effect of their action (Schlager, 2004: 151–52). Further, it requires the sense of a common goal to which many actors will commit, driven by: tangible factors such as local knowledge of likely rewards; less tangible factors such as a sense of community and local autonomy; and experience and leadership to promote common beliefs and interests, often when differences – such as economic inequality – might undermine solidarity (2004: 152, 168; Heikkila and Andersson, 2018: 317). Crucially, it requires high levels of trust to encourage norms of reciprocity (Poteete et al., 2010: 226). Trust, and factors such as predictability and access to information, is crucial to minimizing the costs of compliance monitoring and enforcement (Ostrom, 1990: 25– 26). Trust may develop when participants communicate regularly, share an understanding of their common interests, reciprocate each other’s cooperation, and have proven reliable in the past (Ostrom, 1990: 183; Keohane and Ostrom, 1995: 6; Ostrom, 1998; see also Putnam 2001 on ‘social capital’). Many additional factors are key to developing trust and solidarity, such as the ‘evolution’ of behaviour under conditions of uncertainty and complexity (Box 7.2). Actors have often learned about rule efficacy – to encourage cooperation and punish opportunism – through trial and error over a long period, beginning with simple, lowcost operational rules producing quick wins (1990: 14, 34, 140–42). They have also been motivated to change rules over time, when the costs are low and the perceived benefits of improved cooperation are high (1990: 211). The latter helps explain how actors solve ‘firstorder’ collective action problems, in which there are unclear incentives to invest resources to set up institutions in the first place (1990: 182). Indeed, Ostrom’s (1990: 59) discussion of actors discovering rules together resembles (somewhat) complexity theory’s discussion of locally ‘emergent’ rules and outcomes which are not explained simply by cost-benefit calculus or design factors such as the number of face-to-face meetings (Ostrom et al., 2014: 275).

Second, the IAD is complicated because it analyses many types of rule-influenced activity. The IAD classifies rules according to multiple purposes, including how many actors are part of an action situation and the role they might play (‘positional’); what they must/ must not do (‘choice’); who is eligible to participate (‘boundary’); who can move from one position to another (‘succession’); who controls membership and how (‘entry and exit’); how many participants have to be involved in a choice and what will happen if there is no agreement (‘aggregation’); how to manage and communicate information (‘information’); the rewards or sanctions (‘pay-off’); and the range of acceptable actions or outcomes from action (‘scope’) (Ostrom and Crawford, 2005: 191–209; Ostrom et al., 2014: 278–79; Schlager and Cox, 2018: 219–20). Good explanation requires ‘multiple levels of analysis’ to account for – and help design – the rules which influence the context of other rules, and are more or less costly to change: • • • •

‘operational-choice’ rules on the action ‘on the ground’ (including the pay-offs/ sanctions for behaviour); ‘collective-choice’ rules on how actors make and monitor operational-choice rules; ‘constitutional-choice’ rules on who performs legitimate roles, such as monitoring and enforcement (and who can modify collective-choice rules); and ‘metaconstitutional’ analysis of how to define and modify constitutions with reference to the wider political and social context (Ostrom, 1990: 51; Ostrom et al., 2014: 285).

Third, the IAD is complicated because each case study of physical conditions and social practices is different in some way (Stern et al., 2002: 446–47; Dietz et al., 2003: 1907). We can use the IAD to compare experiences but accept that a profoundly successful scheme in one context may fail miserably in another (Ostrom, 2008: 16–17).

The complexity and global spread of CPR studies

Compared to most theories in this book, the international spread and scale of Ostrom’s (1990) study is immense. Ostrom (1990: xv) describes trying to make sense of 5,000 ‘small-scale’ cases (containing 50 to 15,000 actors in a single country). The most analysed cases were in countries other than the USA, initially including Switzerland, Japan, Spain, the Philippines, Sri Lanka, and Turkey. They were followed by new fieldwork on multiple/larger-scale projects, including: irrigation projects in Nepal; a 17-country, fourcontinent study of forest management; and land basin management in Southern California (Ostrom et al., 2014: 288; Blomquist, 1992). Crucially, the newer analysis suggested that IAD insights held even when the projects were larger, with more actors, and even when some users had advantageous user rights that could foster resentment (Heikkila and Carter, 2017). CPR scholarship has grown significantly to compare many different – multi-level, small- and large-scale, simple and complex – settings (see Robson et al., 2014; Pacheco-Vega, 2014 on countries including Mexico, Eritrea, and Belgium) and use multiple methods including interview-rich fieldwork and laboratory experiments to explore new game theory scenarios (Ostrom et al., 2014: 286–87; Poteete et al., 2010). It has branched into new fields such as how to share research knowledge, manage tourism, maintain housing cooperatives, distribute international aid, avoid overusing antimicrobial drugs, and consider the global commons (Heikkila and Carter, 2017; Boyd et al., 2018: 1240; Ostrom et al., 2014: 289; Schlager and Cox, 2018: 242; Ostrom, 2008: 10). It also informs studies of the factors that influence collective action success, including the total number of actors, the number required to make cooperation viable, the ‘discount rate’, ‘similarities of interests’, and leadership. To this list, the IAD adds the value of learning (Box 12.2), the role of external actors influencing collective-choice rules, and how people gather information in real-world settings (Ostrom, 1990: 188–92; Schlager, 2004: 163; Heikkila and Gerlak, 2018). With a research programme covering so many diverse experiences, using a rather complicated approach, mixing methods, and identifying a huge number of potentially relevant variables, it is inevitable to find variation in the way that people think about,

research, and produce conclusions on CPR management (Ostrom, 2009a). Different scholars also relate insights to very different political contexts. We can use the IAD to analyse the relative value of government, market, and communal action in each context, but know that attitudes to (1) state versus market action, (2) local autonomy versus external ‘command-and-control’, and (3) how to evaluate success vary widely across the globe (Schlager, 2004: 159, 169; Dietz et al., 2003: 1910; Stern et al., 2002: 457). This is true even when a key aim is to produce commonly agreed rules and methodological standards (Ostrom, 2006: 5; Poteete et al., 2010).

Social-Ecological Systems (SES) Framework The SES framework sums up this scientific dilemma. On the one hand, Ostrom (2009b: 419) describes it as a way to address the lack of a common interdisciplinary framework to study ‘complex socialecological systems (SESs)’ and it is a ‘direct descendent’ of CPR studies (Ostrom et al., 2014: 291). On the other, it is a yet more complicated framework – with a high cognitive barrier to entry – to reflect Ostrom’s collaboration with natural scientists (2014: 289–91) and connections to the wider literature on complexity theory (Preiser et al., 2018). Put simply, ‘the ecological and social sciences have developed independently and do not combine easily’ (Ostrom, 2009b: 419). Table 7.2 Factors essential to self-organization in social-ecological systems

‘Second-level’ variables

Summary of their indicative effect*

Size of resource system

If too large, the costs of establishing boundaries are too high. If too small, there are insufficient benefits to prompt collective action.

Productivity of system

If too abundant, users will see little reason to worry. If too depleted, it will seem too late to

manage for the long term. Predictability of system dynamics; Resource unit mobility

Users need a clear sense of the effect of their rules, which is easier in static forests than fisheries, water, and mobile wildlife.

Number of users

Large numbers of actors are difficult to manage but may be needed to manage large resource systems.

Leadership

Self-organization is more likely when guided by actors with entrepreneurial skills and local respect.

Norms/social capital

Shared ethics and norms of reciprocity reduce the costs of agreement and monitoring compliance

Knowledge of the SES

Users will perceive the costs of selforganizing to be lower if they understand the resource system and the likely effect of their actions.

Importance of resource to users

Users are more likely to self-organize if they depend on the resource units for their livelihood.

Collective-choice rules

Users are more likely to self-organize if they have high autonomy to create and monitor their own rules.

Source: adapted from Ostrom (2009b: 420–21). * They are indicative since, ‘As in most complex systems, the variables interact in a nonlinear fashion … Simple blueprint policies do not work’ (2009b: 421).

The result is a framework that resembles CPR studies in key respects, with visual emphasis on the interactions between ‘firstlevel’ concepts including users, their governance system, resource system (such as a protected park), and resource units (such as its

trees) (2009b: 420). It also raises similar questions – such as ‘When will the users of a resource invest time and energy to avert “a tragedy of the commons”?’ – and answers them with reference to Ostrom’s ‘second-level’ concepts (or ‘variables’) describing factors that encourage users to (1) value long-term sustainability and (2) self-organize to secure this outcome (Table 7.2). While the SES has inspired many studies (particularly on fishing), the project is less organized than CPR research, partly because new aspects – and terms such as ‘governance system’ – are less well defined than terms in the IAD (Ostrom et al., 2014: 294–96). It remains a promising way to ‘diagnose’ the sustainability of socialecological systems (Partelow, 2018) and address problems with selforganization to manage resources, but largely by people who are already immersed in the IAD, or scholars using the factors in Table 7.2 as a checklist to structure discussion (Schlager and Cox, 2018: 237–44).

Polycentric Governance and Institutional Collective Action For our purposes, the potential benefit of polycentricity in policymaking (Box 7.3) relates to the best of both worlds: (1) the freedom of many actors to act according to their expertise and judgement but (2) the sense of a ‘common ideal’, combined with some agreement to contribute to a common aim (Aligica and Tarko, 2012: 238). As such, it could be relatively effective if it avoids costs associated with more impositional mono-centric arrangements, such as reduced freedom, high costs of finding information about the system, and a tendency not to consult all actors when making decisions (2012: 240). For example, a polycentric system may have a ‘built-in mechanism of self-correction’ as many actors innovate and adjust through trial and error, during an ‘evolutionary competition between the complementary ideas and methods of those different decision centers’ (2012: 240; 246; compare with ‘adaptive capacity’ in Pahl-Wostl and Knieper, 2014, and ‘The evolutionary metaphor’ in Chapter 6, p.101).

Polycentric governance studies pre-date the IAD’s CPR research and describe governmental and non-governmental cooperation more widely. Heikkila (in Cairney et al., 2019a: 16–18) links its early study from the 1960s ‘to the work of Elinor and Vincent Ostrom and the “Bloomington School” of institutional analysis’ in relation to US metropolitan governance, and describes the elements that became central to the IAD framework: •

•

Theoretical discussions emphasize (1) the ability of actors to selforganize to produce institutions to help them solve collective action problems and (2) the fact that public goods need not be delivered by a single central government. Many governments can deliver many different types of goods and service at different levels or scale, while working with each other and nongovernmental actors. They operate under a ‘shared system of laws’ (such as provided by a constitution) and produce ‘interorganizational agreements’ to aid cooperation. The empirical aim is to identify the conditions under which polycentric governance arrangements, held together by a collection of nested rules, provide more efficient, equitable, democratic, or sustainable arrangements during the provision of public goods.

This work challenged the idea that metropolitan regions would necessarily be inefficient simply because they contained many semiautonomous and overlapping jurisdictions (Aligica and Tarko, 2012: 241–44). Greater efficiency is not a given – since multi-centric governance could produce damaging practices associated with competition, duplication, and unclear accountability – but has been documented in an impressive number of cases (Aligica and Tarko, 2012: 243).

Institutional Collective Action (ICA) Polycentric governance may help address ‘institutional collective action’ (ICA) dilemmas when policy problems transcend the responsibility of one government (Feiock, 2013). It is difficult to

match the perceived ‘scale and nature of policy problems’ (including how to tax, deliver services, and allocate land use) to the ‘scale and coerciveness of policy intervention’ (2013: 397). Citizens demand services, jurisdictions exist, and each service has its own economies of scale. However, it is common for all three to diverge – when governments are too small, or government responsibilities overlap too much – when there is insufficient collaboration. Many policymakers, often at different levels of government, seek to coordinate their response, produce an agreement on how to act collectively, ensure that each actor trusts and maintains their agreement, and therefore minimize the transactions costs of ICA. Otherwise, their own action will produce negative externalities for others (2013: 398; Marks and Hooghe, 2000). Swann and Kim (2018: 273) summarize the conditions under which such ‘semiautonomous’ governing bodies ‘overcome barriers to collective action and reduce the risk and uncertainty of collaborative arrangements’. The most effective solutions appear to be: • •

• • •

participating heavily in informal social networks and speaking directly with people, to help build ‘social capital’ and share information effectively (2018: 281–82) tailoring strategies to the nature of the task, such as to maintain closed networks between a small number of organizations when you need to ‘defend against opportunism’, or flexible arrangements when you need to act quickly and reduce the burden of agreement (2018: 283) working with organizations who already share – or collaborating with them to encourage similar – values, ideals, policy preferences, and goals (2018: 284) keeping the costs and other barriers to participation low, to maintain high levels of ‘transparency, accountability and equity’ (2018: 285) producing ‘small wins’ then working incrementally to reduce the burden of cooperation on ill-resourced organizations (2018: 286; compare with Ansell and Gash, 2008).

However, as in the IAD more generally, the overall message is that no single arrangement provides a blueprint or panacea for collective action (Cairney et al., 2019a: 18).

Institutional Complexity and the Ecology of Games The ‘Ecology of Games Framework’ (EG) combines insights from many approaches to analyse ‘institutional complexity’ or ‘complex institutional systems’ (Lubell, 2013: 538). The focus is on actors learning how to secure ‘mutually beneficial outcomes’, cooperating to produce and deliver agreed solutions, and bargaining ‘over the distribution of the costs and benefits’ within a system over which no actor has control (Lubell, 2015: 42). First, EG connects Long’s (1958) ‘ecology of games’ (Box 7.2), the IAD, and ‘a complex adaptive system perspective’ to reinforce the idea that institutional arrangements are not simple and orderly. In simple games, we need only analyse the interaction between a small number of actors with reference to one set of self-contained rules providing clear sanctions or pay-offs. However, ‘governance involves multiple policy games operating simultaneously’ and over time (Lubell, 2013: 538). For example, Lubell and colleagues identified ‘over 100’ different venues ‘operating simultaneously’ in a single case study of San Francisco Bay water policy (Lubell, 2013: 537; 2015: 42). In that context, policy games may be contained within a geographical area, but there are no self-contained collective action problems. For example, ‘biodiversity policy’ is a collection of interdependent policies relating to issues like local planning, protected species, and water management (2013: 541; see also Lubell et al., 2010: 292–94). Many games associated with different policy issues – such as ‘water supply, water pollution, air pollution, traffic congestion, or loss of biodiversity’ – interact with each other, such as when the same actor participates in multiple games subject to different rules, or each game produces ‘externalities’ which affect the ‘pay-offs’ from others (2013: 540–41). A focus on complex adaptive systems suggests that central governments do not have the resources to control – or even understand fully – interaction at this

frequency and scale. Rather, drivers to change behaviour and outcomes are: • •

endogenous (internal), when actors (1) follow and shape the rules of each institution and (2) learn when they ‘try out different strategies’; and exogenous (external), when physical resources (such as water basins) change, or other levels of government change the resources of local actors (2013: 542).

Second, EG takes insights from the wider literature to analyse specific aspects of this process, such as actors forming networks with their allies to get what they want from different games (see ‘advocacy coalitions’ in Chapter 10) and engaging in ‘policy learning’ (Chapter 12) to update their knowledge (2013: 543). Actors deal with bounded rationality primarily with reference to their ‘social tribal instincts’, with EG emphasizing: ‘the importance of emotions, instinctual “fast thinking,” cooperation, in-group biases, and social learning … [producing] some individuals who are altruistic, others who conditionally cooperate, and a minority who pursue their selfinterest regardless of the welfare of others’ (2013: 544–45). Actors form alliances based on strongly held beliefs, process new information to refine their beliefs and strategies, but ‘never have complete knowledge’ of the environment in which they operate (2012: 544). Consequently, we have come a long way from the ‘utility maximizing’ approach outlined at the start of this chapter. EG also emphasizes the relative ‘fragmentation’ of wider policymaking arrangements, negative effects of political competition, and disproportionate power of some actors even in a ‘self-organizing system’ (2013: 546). For example, ‘pay-off externalities’ occur when organizations with overlapping jurisdictions produce policies that undermine each other, and ‘strategy externalities’ describe an actor developing behaviour that is useful in one game but damaging in another (2013: 546). Yet, as with the IAD, the theoretical emphasis is on solutions that come from initiatives consistent with self-organizing systems – such as ‘collaborative governance’ (Ansell and Gash, 2008) – rather than

‘consolidating all decision-making authority into a single institution … [because] it is naïve to believe that an optimal system could be exogenously designed’ (2013: 547). In that context, the ongoing empirical agenda is to identify how ‘collaborative partnerships promote cooperation, learning, and bargaining’, such as when actors recognize the benefit of ‘participating in a portfolio of forums’ to reduce strategy externalities (Lubell, 2015: 45; Mewhirter et al., 2018: 304). This approach must be flexible, because there are too many different – and often-changing – venues, institutions, and actors (with different motives) to produce a one-size-fits-all solution.

CONCLUSION It may seem strange to relate rational choice theory to public policy because it is difficult to apply parsimonious explanations based on a small number of variables to a field with ‘inherently messy circumstances’ (John, 1998: 117; Hay, 2004a: 46–50; Hindess, 1988: 115 with Laver, 1997: 11–12). Yet the issues raised by rational choice and game theory inform many empirical studies, prompting questions such as: how can people cooperate effectively, and what should be the role of government? There will always be winners and losers when policymakers make choices, and it is unrealistic to seek a rational-synoptic or technical solution to conflicts of preferences within society. Further, government solutions may be expensive and produce unintended consequences. We need institutions to produce more convergence between individual incentives and collective action, but their appropriate nature is an open question. You may have detected different assumptions about human behaviour and levels of expectations about these dilemmas from the stories people tell. For example, the ‘prisoner’s dilemma’ and ‘tragedy of the commons’ thought experiments seem to present a bleak view of human behaviour, in which it would be surprising to find much evidence of cooperation in the absence of coercion, even when the fate of the world is in their hands. In the bleakest scenarios, people act independently and make calculations with insufficient thought for other people or the future. Further, many

oppose government intervention because they argue that politicians are just as bad and would only make things worse. In contrast, IAD empirical scholarship describes a long history of cooperation to maintain CPRs. In that context, the IAD presents a more positive account of non-market and non-governmental solutions to collective action problems based on rules, norms, trust, and the monitoring and enforcement of collective agreements. In positive scenarios, people are social beings who make ‘informed judgments about uncertain benefits and costs’ while communicating regularly with others (Ostrom, 1990: 208). They share information, build trust by becoming known as reliable and predictable, and come together to produce, monitor, and enforce rules for the benefit of the group and the planet. They can be trusted to cooperate to produce rules to solve their own problems rather than relying completely on experts or governments to tell them what to do (Schlager, 2004: 168). These approaches help put problems in context and identify the conditions that encourage effective collective action. This context seems remarkably complex, which makes the IAD difficult to understand at first. While game theory laboratory experiments – built on simple rules and relatively small numbers of parameters – produce relatively parsimonious analysis, field studies embrace life’s complexities to try to understand the key dimensions of each case study’s context. There are too many concepts, variables, global applications, and variations-by-context to contain in a simple theory or model. Rather, the IAD offers a common language to help organize research and encourage a family of theories to explain different parts of the whole picture. As with any language, we learn it as we speak with people more fluent than us. Unfortunately, it is also a highly technical language, and knowledge of its rules is often as tacit, ill-communicated, and difficult to grasp as the working rules described by the IAD (Heikkila and Andersson, 2018: 310). So if you see an IAD scholar with a pained expression, saying ‘it depends’, ‘it’s complex’, and ‘there is no magic answer’, you know they really mean it. You should also give yourself a break if it takes a while to make sense. Like Chapters 6 and 8, this chapter explores the consequences when policy problems go beyond the boundaries of jurisdictions or

political systems and become vulnerable to complex decisionmaking arrangements that may undermine central control and simple institutional responses. The IAD makes frequent reference to ‘complexity’ and ‘evolution’. It highlights the more ‘spontaneous’ and ‘emergent’ outcomes that arise when many actors and institutions interact (Aligica and Tarko, 2012: 254; Cairney, 2013). Further, EG emphasizes fragmentation when the overlap between jurisdictions, combined with uncertainty and different ideas about how to solve problems, produces duplication or contradictory policies. In that context, compare with chapters that explore the role of norm enforcement (Chapter 11, p.192), policy learning, and coercive policy transfer (Chapter 12) to address such problems. Game theory – the use of simple mathematical models to compare how individuals interact with each other when faced with different incentives in different settings. Free ride – to enjoy the benefits of a collective resource without paying; to benefit from collective behaviour without engaging in that behaviour. Public goods – collective resources of which (1) no one can be excluded from enjoying their benefits (‘non-excludable’) and (2) their use by one actor does not diminish their value to another (‘non-rival’). Common pool resources (CPRs) – collective resources of which (1) it is difficult to exclude actors from enjoying their benefits and (2) their use by one does (usually) diminish their value to another. Utility – the satisfaction gained from fulfilling one’s preferences (e.g. consuming a good or service). Intransitive – when preferences are not consistent. If A is preferred to B and B to C, A is not necessarily preferred to C. Pareto efficiency/optimality – a point at which no one is better off without someone being worse off. Kaldor–Hicks compensation – A gains more than B loses, and compensates B, so that the outcome is Pareto efficient. Economies of scale – when the marginal cost of producing a good or service drops as you increase production (reducing the average cost).

Transactions costs – the costs of ‘doing business’ with others, including costs of searching for information on others, professional fees, and the setting up and enforcing of contracts. Externality – the (often unintended) ‘spillover’ effects of the actions of one actor on another.

8 Multi-level Governance and Multi-centric Policymaking Key themes of this chapter: • • • •

• •

Multi-level governance (MLG) and multi-centric policymaking describe the dispersion of power from national central governments to other ‘centres’. Multi-level usually describes the sharing of policymaking responsibility between supranational, national, regional, and local government. Governance often describes the sharing of power between those with formal responsibility and informal influence. MLG can result from choice and necessity. Choice describes constitutional decisions to share power with other levels and types of government. Necessity describes the consequence of bounded rationality and other causes of low coordinative capacity. MLG contrasts with the Westminster model which describes power centralization. MLG often describes Europe Union (EU) policymaking, but there are many parallels with federalism and other forms of multi-centric policymaking.

INTRODUCTION: FROM ONE TO MANY ‘CENTRES’ OF POLICYMAKING Multi-level governance (MLG) describes the sharing of power between national central government and other levels of government (hence multi-level) and non-governmental actors (hence governance rather than government). It identifies blurry boundaries between formal and informal sources of authority, which make it difficult to

identify clear-cut decisions or power relations. In the international arena, it is difficult to identify sovereignty within national governments. They are tied increasingly to the policies agreed between states and implemented by international organizations. In the domestic arena, the interdependence between multiple levels of governments and public and private actors suggests that policymaking does not result solely from the exercise of formal powers. Instead, central governments must try to influence, rather than control, policy outcomes. From this general starting point, we can identify different elements of an overall MLG story. MLG often describes changing relationships. We are witnessing a shift towards a new balance of policymaking responsibilities: from national governing institutions to supranational and sub-national governing institutions; and from central government to the different levels of government and nongovernmental organizations that interact with them. MLG results from a mixture of choice and necessity (Cairney et al., 2019a). Choice refers to constitutional arrangements with power sharing built in, such as the USA’s well-established balance of powers, the more recent and changing dynamic in the European Union (EU), and the UK’s devolution of powers to sub-central governments. Necessity describes the centre’s lack of coordinative capacity. Policymakers can only pay attention to a small proportion of their responsibilities, and must delegate or share power with many other actors. Therefore, MLG also describes a governance problem or dilemma, in which central governments must balance pragmatism, to recognize the limits to their powers, with an electoral imperative to emphasize being in control. We need to tell these stories in different ways in different political systems. First, MLG helps us describe policymaking in unitary and centralized systems such as the UK. The ‘Westminster model’ image of a strong, centralized state which acts unilaterally contrasts with MLG’s image of a segmented, disaggregated state which has to share power and negotiate with other political actors (Bache and Flinders, 2004a: 38). It describes the choice to become a more Europeanized, devolved, and ‘quasi-federal’ political system. It also describes a dilemma about how to deal with the necessity of multi-

centric policy-making. The term ‘governance problem’ largely describes the intended and unintended consequences of many central government initiatives to try to centralize politics and policymaking. This distinction between choice and necessity helps us make sense of the UK’s exit from the EU. Some commentators equate MLG with a loss of control, describing ‘Brexit’ as a way to take it back. However, the UK’s centre does not actually control policy outcomes in the way they suggest, and policy concepts are crucial to help us understand what terms like ‘Brexit’ mean in practice. Second, MLG developed as a way to describe and explain major changes in the EU policy process. The role of the EU – constituting the rules and organizations established by member states – has become increasingly important, producing the need for new ways to conceptualize its policy process. For example, Hooghe and Marks (2003) describe two ways to understand MLG. ‘Type 1’ refers to a relatively clear separation of powers by territory – local, regional, national, supranational – while ‘type 2’ refers to the more complex diffusion of power according to policy issue, involving a wide range of organizations across various levels of government and the public and private sectors (Table 8.2). Third, the key themes raised by MLG – power sharing, choice and necessity, and how to project accountability – are applicable to many other political systems. MLG studies help us compare policy processes, such as the emerging EU and the US federal system that has long enjoyed a separation of powers between the executive, legislative, and judicial branches, and devolved power to subnational governments and policy specific jurisdictions. MLG studies also extend to countries such as Canada and Australia. MLG helps us identify abstract and universally applicable themes to guide empirical study. For example, it draws on the policy networks literature to describe the blurry boundaries between formal power and informal influence. In doing so, it shares a focus with punctuated equilibrium theory (PET; Chapter 9) and the advocacy coalition framework (ACF; Chapter 10), which both began as studies of the USA before being applied to systems such as the EU and its member states.

This focus also helps us consider how central governments should respond: should there be MLG, or should governments centralize to address an electoral imperative? For example, we may welcome MLG in systems such as the USA and Switzerland that contain a plethora of well-established elected bodies. However, we may be less certain about systems such as the UK and EU, in which power often seems to be shared with bodies with less electoral accountability following pragmatic choices and the unintended consequences of decisions made in the past (Box 8.1). Box 8.1 Empirical and conceptual descriptions of multi-level governance Many MLG studies describe governance empirically, as part of a real and measurable trend: the centre is losing power or sharing it with many others. Others describe it conceptually, as part of a wider focus on how we should analyse and describe our world. ‘Decentred theory’ represents a philosophical argument against treating the ‘centre’ as a real entity: terms like ‘governance’ and ‘institution’ describe the ways in which actors interpret, narrate, and respond to their policymaking context. A system of continuous negotiation among nested governments at several territorial tiers – supranational, national, regional and local. (Marks, 1993: 392) We no longer have a mono-centric or unitary government; there is not one but many centres linking many levels of government – local, regional, national and supranational. (Rhodes, 1997: 1) Governing issues generally are not just public or private, they are frequently shared, and governing activity at all levels (from local to supra-national) is becoming diffused over various societal actors whose relationships with each other are constantly changing. (Kooiman, 2003: 3)

A key tenet of multi-level governance is the dispersal of authority and decision-making to a wide range of bodies through a process of negotiation. The net effect is that policy-making has been transformed from being state-centred and state-driven activity to become a complex mix of hierarchies, networks and markets. (Richards and Smith, 2004) While there is a view that states are losing control in the context of governance, the alternative view focuses on new state strategies for coping with the challenge of governance. (Bache and Flinders, 2004a: 36). Decentered theory thus explains types of governance as the contingent results of situated agents acting on beliefs that they reach by drawing on inherited traditions to respond to dilemmas. (Bevir, 2013: 5)

WHAT IS GOVERNANCE? The term ‘governance’ has many meanings and is used to describe a wide range of practices, relationships, and trends (Kjaer, 2004). However, three related characteristics are most relevant to our purposes: 1. There is interdependence between public and private organizations. Power is shared between actors formally responsible for making decisions and those with whom they consult and negotiate. Governments form policy networks with a range of interest groups and implementing organizations willing to trade their expertise and support for influence. 2. Governments often have the authority but not the capacity to make and carry out their own policies. They can influence but not control the networks of organizations that implement policy. 3. Commentators may describe MLG more or less positively according to the extent to which they are accustomed to

centralized authority or shared decision-making (Kooiman, 1993; Peters and Pierre, 2004: 78; Kjaer, 2004: 35).

Global Governance Many of the world’s major problems, such as climate change, economic crises, and pandemics are ‘transnational’ and require some degree of cooperation across governments (Kjaer, 2004; Weiss, 2009: 254). ‘Global governance’ describes the rise in importance of international organizations and agreements as the ‘collective efforts to identify, understand, or address worldwide problems that go beyond the capacities of individual states to solve; it reflects the capacity of the international system … to provide government-like services’ (2009: 257; see also Finkelstein, 1995; Slaughter, 2004). Studies describe new patterns of global politics in which: • • • • •

power is increasingly diffused to supranational organizations; there is more international cooperation and interdependence; there is a rise of nongovernmental and ‘civil society’ involvement in international policy (on issues such as land mines, torture, and human rights); there is a greater adherence to international norms; and policing efforts are increasingly transnational (Hooghe and Marks, 2003: 234; Welch and Kennedy-Pipe, 2004).

‘International regime complexity’ captures the idea that states create multiple agreements and international organizations, producing a messy and confusing process of international cooperation (Alter and Meunier, 2009). Further, the globalization of policymaking has prompted new forms of organization, such as ‘transnational networking among women’s movements’ (True, 2003).

Governance as Systems Over Which the Central Government Has Limited Control

Studies of governance connect to our focus on policymaking environments and complex systems (Chapter 6). Central government is one part of a wider system of governance which includes other governmental organizations (such as local and health authorities), non-governmental organizations (such as the private and ‘third’ or ‘non-profit’ sectors), and quangos (Macleavy and Gay, 2005). Governance studies draw heavily on studies of policy networks to highlight the interdependence between governmental and non-governmental actors (Chapter 9). Policymakers in government come together with other actors to negotiate and make shared decisions. These systems or networks are often described as ‘self-organizing’ to reflect the government’s limited capacity to control them (Rhodes, 1997: 46–53; Rhodes, 2006b). The policy networks literature – and the UK ‘policy communities’ literature in particular – has long identified shared, not unilateral, decision-making as a key tenet of public policy. Jordan and Richardson (1982: 84) draw on Habermas’ work to highlight a ‘rationality deficit’ within modern government: ‘Authorities with little informational and planning capacity … are dependent on the flow of information from their clients … thus unable to preserve the distance from them necessary for independent decisions’. The policymaking world is specialized, with the responsibilities of government divided into sectors and subsectors. As the scope of government has expanded, its departments have become more specialized, there has been a rise of ‘particularistic’ groups, and civil servants have taken on a ‘larger and larger part of the policy making load’ (1982: 86). Policymakers devolve the responsibility for policy management to civil servants. Given civil servants’ lack of political legitimacy, they are ‘ill placed to impose and conflict avoidance is likely to result’. Further, given civil servants’ lack of specialized knowledge, they are often dependent upon groups for information and advice. The result is policymaking relationships between those in formal positions of responsibility and those who seek to influence them. Specialist accommodation leads to ‘clientelism’, in which civil servants form a bond with, and promote the interests of, certain groups within government. A bargaining relationship develops, based on an exchange of information for influence. Therefore,

policymaking is interdependent and it is difficult to attribute responsibility for policy to specific individuals. Rather, public policy ‘is the joint product of their interaction’ (Rose, 1987: 267–68).

Governance as the Proposed or Actual Reform of the State Key aspects of governance often connect to the corporate world. For example, ‘good governance’ refers to decision-making principles and practices, to achieve full disclosure, integrity, and accountability. In business, they may be linked primarily to the roles of managers, while in government these aims are combined with a diffusion of power to the judiciary and external auditors, a free press, and respect for human rights. However, ‘new public management’ (NPM) is more relevant to our focus on coordinative capacity. NPM describes the application of private sector ideas to the public sector. NPM reforms includes setting standards of performance and gauging success by results, accountability to the ‘customer’, contracting out services and introducing quasi-markets in the public sector. From the late 1970s, public sector reforms were particularly extensive in the UK (Hogwood, 1997: 715; O’Toole and Jordan, 1995: 9–10; Cairney, 2009a; Flinders, 2009; DBERR, 2008: ii, 13) and more radical than in most European countries (Kjaer, 2004: 35). However, NPM became a popular way for many countries to try to reduce the size of the state. For example, the privatization of 8,500 state-owned enterprises took place in over 80 countries between 1980 and 1992 (including Communist, post-Communist and developing countries) and the broad principles of NPM are just as widespread (Box 12.1). Hooghe and Marks (2004: 15) report that a survey of 75 developing countries suggests 63 underwent a process of decentralization (compared to half of the EU countries since 1980). These developments prompted a huge number of academic publications on NPM’s spread and effect. More recent scholarship has focused on, for example, ‘public value’ (O’Flynn, 2007) and ‘new public governance’ (Osborne, 2006;

2010). Both incorporate the study of less hierarchical and more ‘coproduced’ policy between policymakers, citizens, and stakeholders, and the rejection of NPM by some policymakers (Alford and O’Flynn, 2009; Dickinson and Sullivan, 2014; Newig and Koontz, 2014; Housden, 2014; Durose and Richardson, 2015; Sicilia et al., 2016).

GOVERNANCE AS A PROBLEM: THE WESTMINSTER MODEL Central control is a key element of the ideal-type ‘Westminster model’ (WM), whose main characteristics include the: • • • •

reliance on representative, not participatory, democracy; first-past-the-post electoral system, which exaggerates the majority of the governing party and allows it to control Parliament; power of the prime minister to control cabinet and ministers; and politically neutral civil service which acts according to ministerial wishes (Bevir and Rhodes, 1999; Marsh et al., 2001; Richards and Smith, 2002: 3–4; Cairney and McGarvey, 2013: 20).

The WM suggests that power is centralized and government is topdown or ‘one way traffic from those governing (the Government) to those being governed (society)’ (Richards and Smith, 2002: 3). Therefore, power and responsibility go hand in hand since, if you know who is in charge, you know who to reward or punish in the next election. In that context, the characteristics of governance – including interdependence and the low coordinative capacity of the centre – are particularly important for Westminster systems. The low or diminished ability of central governments to direct policy is a problem for ministers who need to project an image of governing competence based on control (Cairney et al., 2019a; Duggett, 2009). Bache and Flinders (2004a: 38) contrast the UK’s version of the WM with MLG. While the WM suggests a centralized, unitary state in which there are clear lines of accountability and hierarchical control, MLG suggests a disaggregated, quasi-federal state in which influence replaces control within a political system with multiple lines

of accountability and uncertainty about where the ‘top’ (or centre) is in different situations. The international arena extends this loss of control, with ‘absolute sovereignty’ for one’s foreign affairs replaced by interdependence and negotiation with other countries and international organizations. MLG therefore provides us with a new set of dictums when exploring policymaking. The diffusion of power vertically (across multiple levels of government) and horizontally (to non-state actors) suggests: negotiation not compulsion; fragmentation not centralization; and multi-level but not hierarchical relationships (Table 8.1). Table 8.1 Westminster model versus multi-level governance

Westminster model

Multi-level governance

Centralized state

Disaggregated and fragmented state

Hierarchy and control

Power diffusion and negotiation

Clear lines of accountability

Blurry lines of accountability

Absolute sovereignty

Interdependence, unclear sovereignty

Unitary state

Quasi-federal state

Strong core executive

Segmented executive

Direct government

Delegated governance, ‘steering’ not ‘rowing’

Source: adapted from Bache and Flinders (2004a: 38).

In the UK context, MLG studies often focus on (1) the extent to which the centre is ‘hollowing out’, (2) how ministers have responded, and (3) the extent to which their reforms helped solve or exacerbate their governance problems. From the late 1970s, UK governments have addressed their governance problem by shifting ‘the balance between government and society away from the public

sector and more towards the private sector’ (Kooiman, 1993: 1). The process is associated with NPM reforms (Gray, 2000: 283–84), including: •

•

•

•

Privatization. The sale of public assets, break up of state monopolies, injection of competition, introduction of public– private partnerships for major capital projects, and charging for government services (Rhodes, 1994: 139; Goldsmith and Page, 1997: 150). The introduction of ‘quasi-markets’. In the absence of privatization, internal markets may be introduced in which one part of the public sector competes with another for the ‘business’ of commissioning agencies (Wistow, 1992; Day and Klein, 2000; Cairney, 2002; Greer, 2004). The reform of the civil service. This includes attempts to make civil servants more accountable by giving them more responsibility to manage their own budgets, and separating the policymaking and delivery functions in government departments by hiving off the latter and creating executive agencies (Greer, 1994: 6; O’Toole and Jordan, 1995: 3–5; Massey, 2001: 21; OPSR, 2002). The increased use of quangos (Greenwood et al., 2001: 153–57; Stoker, 2004: 32).

Rhodes (1994, 1997) argues that the overall effect of NPM reforms – combined with a process of devolution to subcentral governments and the Europeanization of policymaking – is a decline in the capacity of central government to control public policy. Policy delivery shifted from a unified civil service and accountable local government to a ‘patchwork quilt’ of quangos and NGOs, producing service fragmentation. It diminished accountability to Parliament via ministers, with much responsibility devolved to executive agencies, quangos, and the private sector or lost to European institutions (see also Hogwood et al., 2001: 97; Bevir and Rhodes, 2003: 6). However, the extent of the lack of central control is not clear. Many question the ‘hollowing out’ thesis by pointing to the growing range of powers that the core executive enjoys, and exploring the extent to

which changes at the centre have reinforced its power (Hogwood, 1997; Holliday, 2000; Marinetto, 2003). The government was never effective at controlling peripheral functions of the state such as the nationalized industries. Governance changes, such as privatization and civil service reforms, mark a return to core competencies, with the centre making strategic decisions and creating accountability mechanisms for delivery bodies. In other words, we may have witnessed a shift from ‘interventionist state’ to ‘regulatory state’ (Hood et al., 1999; Cairney and McGarvey, 2013: 149; Majone, 1994). Marsh (2008: 255; 2011; Marsh et al., 2003: 308) argues that ‘strong government, although increasingly challenged’ is a more realistic description than the ‘hollowed-out state’. Although there is competition for power between multiple organizations, some are more powerful than others. While the government is dependent on other organizations to deliver policy, and is one of many organizations involved, it is also the most powerful organization. Similarly, studies of ‘metagovernance’ suggest that governments possess key tools to manage the ‘rules of the game’ in policy networks, including the initial design and resourcing of the network, definition of the problem to solve, and access to policymakers (Sørensen and Torfing, 2009; Dommett and Flinders, 2015; Bailey and Wood, 2017). Until recently, some of this academic debate converged around the idea that there is more than one form of UK governance. If it intervenes and pays disproportionate attention to an issue, it can make a major difference in the short term, perhaps with a deep and enduring effect. If not, it makes little difference. Indeed, UK governments demonstrate a tendency to delegate then intervene in ad hoc manner (Gains and Stoker, 2009), or rely on indirect control through performance management. This approach is summed up by Matthews’ (2016) phrase ‘letting go and holding on’, Hay’s (2009) exploration of the stories that governments tell about their governing competence, and Hood’s (2010) identification of the tools that governments can use to avoid blame (see also Bache et al., 2015). However, in 2016, the debate took on a profoundly new dimension, as 52% of referendum voters chose for the UK to leave the EU, to ‘take back control’ of policy and policymaking (Box 8.2).

Box 8.2 Can Westminster take back control after Brexit? The ‘Brexit’ referendum was dominated by a narrative of taking back control of policy and policymaking. Control of policy would allow the UK government to make profound changes to immigration and spending. Control of policymaking would allow Parliament and the public to hold the UK government directly to account, in contrast to a more complex and distant EU policy process less subject to direct British scrutiny. Such high-level political debate relies on the false image of a small number of elected policymakers – and the prime minister in particular – responsible for the outcomes of the policy process. There is a strange disconnect between the ways in which elected politicians and elected policymakers describe UK policymaking. Ministers have mostly given up the language of control; modern manifestos no longer make claims – such as to secure ‘full employment’ or eradicate health inequalities – that suggest they control the economy or can solve problems by providing public services. Yet much Brexit rhetoric suggests that a vote to leave the EU will put control back in the hands of ministers to solve major problems. The main problem with the latter way of thinking is that it is rejected continuously in the modern literature on policymaking. Policymaking is multi-centric: responsibility for outcomes spreads across many levels and types of government, to the extent that it is not possible to know who is in charge and to blame. Some MLG relates to the choice to share power with EU, devolved, and local policymaking organizations. However, most MLG is necessary because ministers do not have the cognitive or coordinative capacity to control policy outcomes. They can only pay attention to a tiny proportion of their responsibilities, and have to delegate the rest. Most decisions are taken in their name but without their intervention. They occur within a policymaking environment over which ministers have limited knowledge and control (Chapter 6). The problem with using Brexit as a lens through which to understand British politics is that it emphasizes the choice to spread power across a political system no longer, without acknowledging the necessity of doing so. Our understanding of

the future of UK policy and policymaking is incomplete without a focus on the concepts and evidence that help us understand why UK ministers must accept their limitations and act accordingly. Yet clearly the Westminster model archetype remains important even if it does not exist (Duggett, 2009). Policy studies have challenged successfully its image of central control, but the model’s importance resides in its rhetorical power in wider politics when people maintain a simple argument during general election and referendum debates: we know who is – or should be – in charge. This perspective has a profound effect on the ways in which policymakers defend their actions, and political actors compete for votes, even when it is ridiculously misleading (Rhodes, 2013; Bevir, 2013).

MULTI-LEVEL GOVERNANCE (MLG) AND THE EUROPEAN UNION (EU) MLG is a vague but important and widely used phrase in EU studies (see Stephenson, 2013, for a review of its origins and coverage). To some extent, the woolly nature of MLG reflects the main object of its study: the rapidly changing EU. A relatively fluid EU policy process, with levels of power diffusion varying across time and policy issues, requires a flexible conceptual framework to accommodate empirical studies. The early literature on EU integration contrasted two grand theories of international relations (Jachtenfuchs, 2001: 246; compare with Rosamond, 2000: 145). Neofunctionalism describes a political project in which incremental change is preferred to a federal approach setting out a formal constitution and separation of EU powers (Rosamond, 2000: 51). It begins with greater economic integration and trade barrier removal. Cooperation increases and countries become economically interdependent. Economic interdependence produces political integration incrementally, beginning with ‘low politics’, or the least controversial policy areas. The logic of further cooperation in some policy areas contributes to a snowball effect in others. As this integration increases, so does the

role for supranational institutions to further the EU project. As the EU grows in power, groups representing ‘social interests’ shift their activities from states to European institutions (2000: 52). Overall, the creation of a political community in Europe is not only likely but also a social good, reducing political conflict and the prospect of war (Haas, 1958). In contrast, intergovernmentalism suggests that member states remain the most important actors, as ‘gatekeepers’ controlling EU development. This follows a ‘two-level game’ in which domestic preferences within member states are taken to the EU bargaining table (Moravscik, 1993; Rosamond, 2000: 136–38). Supranational institutions have little independent influence and get their power from states (Pollack, 2001: 225). Both approaches proved to be flawed (see Glencross, 2014 for a full account). The fundamental flaw of neofunctionalism is that its predictions did not come true. In the short term, key member states blocked the ‘inevitable’ process of political integration (Dinan, 2004; Rosamond, 2000: 75; Haas, 1975). In the longer term, while political integration has progressed significantly – including ‘high politics’ issues such as the Euro currency and monetary union – it is not uniform and has contributed to political crises (Moravcsik, 2012; Vilpišauskas, 2013; Bickerton et al., 2014; Crespy and Schmidt, 2014; Mény, 2014; Sandbu, 2015; Woodruff, 2016). EU measures to address ‘austerity’ have also exposed tensions which relate strongly to imbalances in member state powers (Bulmer and Paterson, 2013; White, 2015). Intergovernmentalism also proved less useful over time, as integration affected domestic preference and ‘locked-in’ each member state, while states are now one of many organizations within a complex set of multi-level governance arrangements (Pollack, 2001: 226–27). The EU is now a multi-level, not two-level, game (Rosamond, 2000: 147) and supranational institutions – such as the European Parliament, the European Court of Justice, and the European Commission – increasingly demonstrate decision-making that is not dominated by state-centred institutions such as the Council of Ministers and the European Council (Asare et al., 2009; Fabbrini, 2013; Torres, 2013). Overall, states are in a web of interdependence that allows supranational organizations and organized interests

(including sub-national authorities) to influence policy and integration. Policy outcomes are uncertain and dependant on levels of Europeanization, the entrenchment of member state laws, and strength of the policy agenda in each jurisdiction (Marks, 1993). Therefore, MLG’s focus on the variable diffusion of power – by policy issue and over time – captures the EU policy process better. Or, we may argue that it is more important to explain what the EU is and does now, rather than what it may be moving towards (Rosamond, 2000: 106–109). In that context, MLG developed as a way to address a false boundary between the study of domestic and international politics in the EU, which: Neither resembles domestic politics nor international organizations, and therefore defies explanation from approaches applied either to politics within states or politics between states … multi-level governance has been seen to capture the shifting and uncertain patterns of governance within which the EU is just one actor upon a contested stage. (Bache and Flinders, 2004b: 1–2) The study of MLG then developed to incorporate empirical developments in the EU, such as when key changes established the importance of multiple levels of government in their own right. For example, the Single European Act in 1987 extended the use of qualified majority voting and therefore a reduction in the ability of member states to reject new policies unanimously. The development of powerful EU administrative and legal institutions further established the role of the EU as an actor as well as an arena for intergovernmental relations (Bache and Flinders, 2004b: 2). At the other end of the governing spectrum, the reform of EU structural funding policy in 1988 advanced the ability of sub-national authorities to maintain direct links with the European Commission as well as their central governments (2004b: 3). Further, European cities and regions have established themselves as key actors in fields such as migration policy (Zapata-Barrero et al., 2017). Such developments highlight the potential for supranational and regional influence not conditional on national governmental support. Further, MLG informed by networks analysis highlights informal

relationships and the blurring of boundaries between public/ private action and levels of governmental sovereignty at multiple levels. The EU became an important hub for group–government relations. As its institutions grew in importance, many powerful interest groups shifted resources away from national governments towards the EU, and other (often less well resourced) groups lobbied the EU via member states or regions (Mazey and Richardson, 2006: 248–52; Keating et al., 2009).

EMPIRICAL AND NORMATIVE VISIONS: WHAT MLG IS AND SHOULD BE MLG empirical studies are generally descriptive, suggesting that decision-making authority is dispersed and policy outcomes are determined by a complex series of negotiations between various levels of government and interest groups. Our empirical focus shifts from formal powers and the capacity to make and enforce decisions to the messy systems in which the distinction between formal and informal sources of authority becomes less meaningful. With decision-making responsibility shared across multiple levels of government and with non-governmental actors, formal responsibility may be less important than a willingness to engage in policymaking and negotiate with other jurisdictions. MLG also prompts normative inquiry: this is how things are, but is it how they should be (Box 8.3)? Stories of multi-centric policymaking offer the following potential advantages: (1) more opportunities to elect policymakers in many venues; (2) a more open and efficient way to encourage many actors from many policymaking venues to cooperate (Chapter 7); (3) policy tailored to local context; and (4) the potential for policy innovation and learning by a large number of authorities (Hooghe and Marks, 2004: 16; Cairney et al., 2019a). Such normative potential relies on empirical examples, such as countries like Switzerland and the USA in which functionally specific bodies are elected locally and accountable to the populations they serve. In contrast, MLG in the UK is often via ‘distributed public governance’ (Bache and Flinders, 2004a: 45) in which unelected

quangos operate with minimal public and ministerial scrutiny, while executive agencies find themselves ‘sitting on top of complex publicprivate relationships they may only vaguely understand’ (Kettl, 1993, cited in Rhodes, 1997: 54; Flinders and Skelcher, 2012). Therefore, our empirical findings inform our normative conclusions. The key questions become: is MLG chosen or an unintended consequence of government reform? Does type 2 MLG signal the devolution of power to elected local bodies or the loss of central control and accountability? Box 8.3 The normative debate: Should power be concentrated in the ‘centre’? MLG’s empirical focus on the dispersal of power relates to a classic normative question in policy studies: should power be centralized? The ideal of comprehensive rationality includes an assumption that power is held centrally by policymakers; their decisions are carried out by neutral bureaucrats or other organizations via a series of well-ordered stages (Chapter 4). Studies of implementation and complexity theory demonstrate that this assumption is unrealistic. Many studies also suggest that complete top-down control would be inappropriate. We need to balance the authority of the top with the local knowledge at the bottom to connect (1) the delegation of policymaking to people who know how to do it with (2) the maintenance of a meaningful degree of accountability for the outcomes (Chapter 6; Cairney et al., 2019a). Yet this position challenges the WM emphasis on a clear and accountable ‘centre’ that seems necessary to make sense of the government’s contract with those who consent to its power. In the absence of such an arrangement, how do we hold policymakers to account? Hooghe and Marks (2003: 233) provide one of the strongest defences of MLG: ‘Centralized authority – command and control – has few advocates … the dispersion of governance across multiple jurisdictions is both more efficient than, and normatively superior to, central state monopoly.’ Since MLG is often vague, they describe

two empirical forms and ‘contrasting visions’ (2003: 236; see also Hueglin and Fenna, 2006: 316) based on their questions described in Table 8.2. With type 1, the dispersion of formal authority is limited to a small number of jurisdictions, with each enjoying an associated executive, legislature, and court system. Devolution tends to be on a territorial basis with a small number of discrete units of government and no overlap of membership. Policies are ‘bundled in a small number of packages’ at each level and the relationships are durable (Hooghe and Marks, 2003: 237). Type 1 perhaps minimizes the coordination problems associated with a proliferation of bodies (2003: 39), albeit often via an asymmetry of power during intergovernmental relations (Wilson, 2003; Cairney and McGarvey, 2013: 202–207; Curry, 2015; McEwen and Petersohn, 2015; Gormley-Heenan and Birrell, 2016). With type 2, we have a ‘complex, fluid patchwork of innumerable, overlapping jurisdictions’ (Bache and Flinders, 2004a: 39). Devolution takes place on a policy issue basis, with much larger numbers of authorities, memberships that span complementary policy areas, and less stable relationships (Hooghe and Marks, 2003: 237). Type 2 exhibits the most flexible and dynamic arrangements, in which many organizations can come together to cooperate to solve specific policy problems (compare with polycentric governance in Chapter 7). Types 1 and 2 are ideal-types to help us explore visions for governance and relate them to specific countries. For example, type 1 might describe the UK, while type 2 is often associated with Switzerland and the USA (Box 8.3). However, for example, the UK really contains elements of both. On the one hand, MLG is a modest departure from state-centred politics in which regional and supranational governments have more powers, but in a relatively ordered system in which each level’s responsibilities are relatively clear. For example, the UK’s devolution of powers to Scotland produced an extensive list of its responsibilities and a set of concordats to ensure that the governments coordinated their activities. On the other hand, in practice, these responsibilities overlap continuously when governments produce different policy instruments to deal with the same problem (Cairney and McGarvey,

2013; Bache and Flinders, 2004a: 40). For example, measures to increase tobacco control are spread across EU, UK, and devolved governments (Asare et al., 2009; Cairney et al., 2012) and energy powers are increasingly shared (Cairney et al., 2019b). Further, Rhodes (1997) argues that a ‘patchwork quilt’ of organizations delivering services has replaced central order. The UK may be following a type 1 ‘vision’ but actually delivering type 2 in practice (Marks and Hooghe, 2000; Hooghe and Marks, 2003: 234; 2001; Bache and Flinders, 2004b: 7; Jessop, 2004; Peters and Pierre, 2004). Table 8.2 Types of multi-level governance

Key questions

Type 1

Type 2

Should jurisdictions be designed around communities/levels of government or policy problems/tasks?

General-purpose jurisdictions

Task-specific jurisdictions

Should jurisdictions bundle competencies or be functionally specific?

Separate functions, ensure nonintersecting memberships

Bundle competencies, encourage intersecting memberships

Should jurisdictions be limited in number or proliferate?

Limit the number of jurisdictions

Do not limit jurisdictions

Should jurisdictions be designed to last or be fluid?

Produce a systemwide architecture

Maintain a flexible design

Source: adapted from Hooghe and Marks (2003: 236).

MLG AND INTERNATIONAL COMPARISONS In some countries, we can find the direct application of MLG concepts and studies. For example, Daniell and Kay’s (2017: 4–6) edited volume on Australian policymaking uses ‘the European view of MLG processes’ to examine the effects of the blurry boundaries between formal power and informal influence, and the ‘dispersion of authority upwards, downwards and sideways’ (see also Horak and Young, 2012 on Canada). In others, we can use MLG and related concepts to compare political systems and the ways in which we study them. The benefit of a flexible MLG framework is that it helps brings together (1) the comparisons of unitary, federal, and quasifederal systems (Chapter 5) and (2) studies of federalism, drawn primarily from US experiences, and MLG drawn primarily from studies of the EU. We therefore need to be very clear on what we are comparing: the normative evaluations of forms of governance, discussed in Box 8.3; or, in the following sections, (1) the different types of political systems (confederal, federal, quasi-federal), (2) the literatures to describe their formal institutions and relationships, and (3) the more abstract theories describing their policymaking dynamics.

Comparing Political Systems The EU is unusual. It represents a modern-day confederacy, or a quasi-federal system, in which EU institutions resemble the executive, judicial, and legislative branches (the European Commission, Court of Justice and Parliament) but member states retain some sovereignty on key issues of economic, social, and foreign policy (Burgess, 2006: 226; Hueglin and Fenna, 2006: 13– 16; Newton and Van Deth, 2010: 109; 2016). Yet we should not treat the EU as a simple departure from a model or norm of federalism since, in practice, there are many different types of federalist arrangements. Federalist states combine national government with a constitutional arrangement that guarantees a degree of sovereignty

to sub-national governments. This sovereignty may be justified in two main ways: to allow local decision-making to supplement ‘a distant centre of government’ in large federations such as the USA; and/or to address (often geographically specific) political divisions ‘based on language, ethnicity, religion, culture or history’ in systems such as Canada and Belgium (Newton and Van Deth, 2010: 111–13). Federations vary according to the number of sub-national governments, the balance of powers between national and subnational governments, the extent to which their respective responsibilities are clear, and the role of legislatures. They also have a different balance between national and sub-national shares of tax receipts and public employees (2010: 110–11; McKay, 2000: 31). There are three main consequences. First, the division between federal and other systems is often unclear. Many unitary states seem quasi-federal because they retain power at the centre but devolve significant policy responsibilities to certain populations (Newton and Van Deth, 2010: 114). Indeed, our rejection of the Westminster model, and the idea of one single type of federalism, suggests that political systems are more similar than they first appear. MLG relates as much to the unitary UK as it does, say, the federal USA. Second, different federations may provide different analytical links to, and policy lessons for, the EU. For example, Scharpf (1988) explores the potential for the EU to share with Germany a rather convoluted system of checks and balances that may stifle policy innovation (compare with Kitschelt and Streeck, 2003). Holland et al. (1996) draw similar links between inertia, federalism, and environmental policymaking in Australia, Canada, and the USA. Church and Dardanelli (2005: 181) draw lessons for the EU on the move from confederalism to federalism in Switzerland (the ‘near perfect embodiment of the federal idea’). McKay (2000) draws lessons from Australia, Canada, Germany, the USA, and Switzerland to determine if the EU can operate successfully as a system with almost no responsibility for raising taxes. Scharpf (1997) examines the scope for EU economic integration to mirror the development of ‘regulatory competition’ in the USA. Overall, the lessons from ‘federalism’ are there, but the sources are eclectic because MLG and

federalism are very broad concepts to describe a huge variety of governance arrangements (Stein and Turkewitsch, 2008: 5).

Comparing Approaches to the Study of Political Systems The MLG and federalism literatures identify the absence of a single centre of government. Rhodes (1997: 1), describing governance, suggests ‘there is not one but many centres linking many levels of government’ while Elazar (1971: 6), describing federalism, identifies ‘the diffusion of power among multiple centers which must cooperate in order to govern’. This overlap of concepts and concerns has prompted some attempts to use federalist theory to compare the USA and EU (Schain and Menon, 2007). For example, Keleman (2004) extends Majone’s (1993; 1994; 1996) argument that the EU is a ‘regulatory state’ to make two main points. First, the vertical relationship (or the ‘politics of competence’) is similar in federal systems: most policymaking takes place at the federal (or central) level, while the responsibility for most implementation rests with the states (or subcentral authorities). Second, the extent to which the centre allows sub-central authorities the freedom to implement federal policy (the ‘politics of discretion’) depends on the levels of horizontal fragmentation within the central government. A highly fragmented system ‘encourages an adversarial, litigious approach to regulation’ (Keleman, 2004: 2). The competition between institutions makes them more protective of their authority and more likely to write detailed laws for the provinces to follow (what Keleman calls ‘regulatory federalism’). In contrast, a concentration of power at the centre encourages ‘discretionary federalism’ or a ‘less judicialized’ approach (2004: 2). Written laws are broad, allowing for more flexibility in application by lower level jurisdictions. For example, the fragmentation of power in the EU is high, so Keleman (2004: 2) predicts a form of regulatory federalism in which ‘inflexible rulemaking and litigious enforcement’ characterizes EU member state relations. Yet the evidence is mixed: Kelemen confirms his

hypothesis in environmental policy, but Asare et al. (2009) do not in tobacco policy. To some extent, there is also an overlap in the study of intergovernmental relations (IGR). IGR explores the entanglement of policy issues when power is vested in more than one actor. As Keating (2005: 18) discusses, multi-level government issues are common to most states with the dual aim of devolving decisions but maintaining central/broad control when appropriate. Although many political systems outline in detail the policy domains of each level or type of government, the boundaries between policy areas become blurry in practice. IGR explores these issues by highlighting countrylevel differences according to institutional structures and the recourse to authority and formal resolution. It asks: is it a unitary, union, federal or quasi-federal state? What is the strength of the ‘centre’ and what is the frequency of formal dispute resolution? What is the role of parties, executives and courts in IGR (Watts, 2007; Horgan, 2004; Hueglin and Fenna, 2006: 215; McEwen et al., 2012; Bolleyer et al., 2014)? These IGR-led studies of federalism often diverge from MLG because the latter focuses more on informal relationships and the blurry boundaries between actions taken by policymakers and influential groups. It shifts our focus from formal institutions towards less formal policy networks, based on the finding that most policymaking takes place in specialized subsystems in which direct accountability for policy outcomes is more difficult to identify (Stein and Turkewitsch, 2008: 10).

COMPARING MLG AND POLICY THEORIES MLG has more in common with policy theories that developed from the exploration of bounded rationality and policymaking complexity: No single actor has all knowledge and information required to solve complex, dynamic and diversified problems, no actor has sufficient overview to make the application of needed instruments effective, no single actor has sufficient action potential to dominate unilaterally in a particular governing model. (Kooiman, 1993: 4)

As described, MLG bears close resemblance to studies of subsystems. For example, PET (Chapter 9) identifies a similar starting point in bounded rationality: there is not one, sole, comprehensively rational decision maker at the heart of government; policymakers cannot consider all issues at all times, and they ignore most and promote relatively few to the top of their agenda. Powerful interest groups are able to operate in relative anonymity within specialist policy subsystems for long periods. Yet, when policymakers focus on these issues, their levels of attention are disproportionate and their response is ‘hypersensitive’. In many cases, this shift can be explained by an increasingly crowded policymaking process and the types of MLG now identified in the EU. When interest groups experience little success at one level, they ‘venue-shop’, or seek influential audiences in other arenas (Mahoney, 2008). If they catch the attention of another venue, newly involved policymakers increase their demand for information and new ways to think about and solve old policy problems. In a process of government characterized by interdependence and overlapping jurisdictional boundaries – in which many institutions can be influential in the same policy areas – these innovations can be infectious. The actions of one often catch the attention of others, causing a ‘bandwagon effect’ of attention and major policy change. Both PET and multiple streams analysis (MSA; Chapter 11) describe a process in which policy actors compete to set the agenda within multiple venues (Baumgartner and Jones, 1993: 32). A large number of MSA studies of the EU describe the importance of problem definition (Ackrill et al., 2011: 874–77). Actors exercise power to frame policy problems to ensure that they are addressed by a particular decision-making venue. Since most policies can be influenced by more than one level of government, the relative involvement of each jurisdiction depends as much on how the policy issue is understood and portrayed as the powers and responsibilities of each jurisdiction (Ackrill and Kay, 2011; Sarmiento-Mirwaldt, 2013; Bache and Reardon, 2013, 2016). The analysis shifts our focus from formal institutions and intergovernmental procedures towards policy networks and the idea that most policymaking takes place in specialized subsystems. The identification of visible power relations

is qualified by less visible competition to define policy problems and perhaps remove them from formal intergovernmental agendas. MLG also relates to a family of concepts describing multi-centric policymaking, including polycentric governance (Chapter 7) and complexity theory (Chapter 6). The dispersal of power throughout a political system makes it difficult to track the effects of policy measures taken at one level of government when it travels through a series of other ‘centres’. While central government is powerful, its reach is limited in proportion to the size of the state. Its action in one area precludes the same degree of action in most others. Taken as a whole, these theories describe multiple sources of policy continuity and change. A degree of continuity can be found in the study of bounded rationality and policy networks. The lack of policymaker attention to most issues, combined with the tendency of governments to process policy in specialist policy subsectors containing civil servants and interest groups, suggests that we will not witness radical policy change in many instances (Richardson and Jordan, 1979). Further, when viewed from the top down, the problem of governance is an inability of central government to carry out its own policies directly (Pierre and Stoker, 2000: 43). Rather, the proliferation of governance organizations may produce a ‘coordination dilemma’ (Hooghe and Marks, 2003: 239) and an increase in the number of ‘veto players’ may ‘make significant policy changes difficult or impossible’ (Tsebelis, 2002: 7). Yet continuity is not inevitable, since there are many sources of major policy change in the EU context. The EU may impose new policies on member states that they would otherwise not entertain. The diffusion of power and responsibility may encourage policy innovation in one territory, followed by emulation in the others (Hooghe and Marks, 2003: 236); and there is continuous potential for venue shifts to spark new interest in old problems.

CONCLUSION The concept of multi-level governance is simple: governance refers to shared decision-making and the blurry boundaries between formal and informal sources of authority, while multi-level suggests that

these processes take place above and below the nation-state. Yet, beyond this basic understanding, MLG is often difficult to pin down and apply systematically or meaningfully to policy studies. There are two main ways to do so. First, use MLG to help describe specific empirical developments. Second, think of MLG more conceptually, as an abstract way to describe key policymaking developments. In practice, we need to do both. For example, we can tell the UK story in relation to the classic ideal-type Westminster model, used generally to explain how policymaking power could be centralized and held by a small number of people in the core executive. MLG describes a shift of power from the centre to different levels (EU and sub-national) and types of governing actor (including executive agencies and quangos). MLG partly represents the choice to share power with other elected policymakers and the necessity of power diffusion in relation to its low coordinative capacity. The UK government appeared to exacerbate its lack of central control by privatizing the public sector, reforming the civil service, introducing quasi-markets, and relying more on quangos. It became obliged to cooperate with many organizations charged with the delivery of public services. In this scenario, perhaps central government ministers once thought that they could make policy from the centre, and now they realize that they need to negotiate policy delivery with a wide range of other actors. Or, perhaps they remain caught in a vicious cycle, in which they try to reassert central control to maintain an image of governing competence, only to find that their efforts are counterproductive. On the other hand, these measures helped produce a ‘lean’ state and the potential for the central government to regulate, performance manage, and intervene in an ad hoc way. Perhaps ministers are content largely to oversee public policy and avoid blame for poor outcomes as much as they can. In any case, we must conclude that the centre is not in control in the way that the WM suggests. MLG also helps us understand the relatively new and quickly changing EU policy process. Much confusion stems from the object of study: a relatively complex policy process that is still developing and difficult to explain. In this sense, Hooghe and Marks’ (2003)

types 1 and 2 can be seen as different lenses through which to study the same political process. We can study the EU as a set of welldefined competencies held by different governments, or a more complicated process of problem solving by many overlapping jurisdictions. On that basis, the MLG literature allows us to make meaningful comparisons between such new quasi-federal systems – the UK and EU – and more established federal systems such as the USA. New studies of the EU connect to older studies of federalism that explore decentralized government, and questions regarding IGR apply to the new EU context. However, most importantly, MLG compares to several other policy theories. We have seen in Chapters 6–8 that MLG is part of a collection of concepts to identify multi-centric policymaking empirically and explore its practical and normative implications. Now, in Chapters 9–11, we explore theories that began life in the federal USA. All of these theories share with MLG a focus on bounded rationality, interest groups, subsystems, and policy-making in multiple venues (although they describe them in different ways). They help us make sense of the cognitive and coordinative limits of policymakers in a policymaking context out of their control. Quangos – quasi-non-governmental organizations. They are sponsored, but not controlled directly, by government departments. Core executive – the ‘centre’ of government, including senior ministers and the administrative arrangements that support them.

9 Punctuated Equilibrium Theory Key themes of this chapter: • • • • •

• •

Policymakers can only pay attention to a tiny proportion of their responsibilities. They prioritize some and pay minimal attention to the rest. Changes in levels of attention are not proportionate to changes in the size of policy problems or information about them. Governments process most issues despite minimal high-level political attention. This simple insight helps explain long periods of policymaking stability and policy continuity and instability and major policy change. The early punctuated equilibrium (PET) literature used decades-long case studies in the USA to show how the creation, longevity, and destruction of ‘policy monopolies’ explained policymaking stability and instability. Modern PET focuses on the ways in which institutions process information disproportionately to produce huge amounts of small, and small amounts of huge, policy changes. PET began as an explanation of US policymaking. The Comparative Policy Agendas project now finds similar dynamics and distributions of policy change in many political systems.

INTRODUCTION: THE PROFOUND IMPORTANCE OF POLICYMAKER ATTENTION Punctuated equilibrium theory (PET) tells a story of complex systems that are stable and dynamic. Most policymaking exhibits long periods of stability but with the ever-present potential for sudden instability. Most policies stay the same for long periods while some change very

quickly and dramatically. Policy change in a particular area may be incremental for decades, followed by profound change that sets an entirely new policy direction. We can begin to explain this dynamic with reference to the limits of comprehensive rationality: since policymakers cannot consider all issues at all times, they ignore most and promote relatively few to the top of their agenda (Chapter 4). This lack of attention to most issues helps explain why most policies may not change, while intense periods of attention to some issues may prompt new ways to frame and solve policy problems. Some explanation comes from the power of participants, either to minimize attention and maintain an established framing or to expand attention in the hope of attracting new audiences more sympathetic to new ways of thinking (True et al., 2007: 157). Further explanation comes from policymaking complexity, in which the scale of conflict is too large to understand, let alone control (Chapter 6). The original story of PET applies two approaches to the study of public policy – policy communities and agenda setting – to longterm case studies of US policymaking (Baumgartner and Jones, 1993; 2009). A key focus is the stable and enduring relationships between interest groups and policymakers. They endure because participants have built up trust and agreement – about the nature of a policy problem and how to address it – and few other actors are interested in the issue (Jordan and Maloney, 1997; Jordan and Cairney, 2013). In many instances, those most involved are able to protect a policy monopoly by framing the issue in a way that limits wide interest and excludes most actors: the policy problem has largely been solved, with only the implementation to be addressed. Or, it is very dull and technical, requiring a certain level of expertise or stoicism. Such group–government relations take place beyond the public spotlight, since the issues are presented as too technical and unimportant to invite attention, and most actors do not have the resources to engage. Policymaking is stable and policy is based on previous agreements between participants. However, the study of agenda setting also highlights sudden lurches of attention, policymaking instability, and non-incremental change (Baumgartner and Jones, 1993: 10). Some issues attract

high and fluctuating levels of attention. A rise in attention can follow a ‘focusing event’ (Birkland, 1997; 2016). Or, it can be caused by interest groups drawing attention to ‘their’ issues. Baumgartner and Jones (1993: 32–37) connect shifts of attention to venue shift, in which an issue can become the jurisdiction of more than one audience at the same time. Groups excluded from policy monopolies will try to shift the debate from within by, for example, appealing to public officials and questioning the existing approach. If unsuccessful, they have an incentive to venue shop, or seek influential audiences elsewhere. It may involve an appeal to a different level or type of governing organization (such as a legislative committee or court) capable of making decisions on the same policy issue. Or, groups make direct appeals to the public. When an issue reaches the ‘top’ of this wider political agenda, it is processed in a different way: more participants become involved, and they generate more ways to look at (and seek to solve) the policy problem. A combination of approaches explains continuity and change. Policies stay the same within policy communities because there is limited external interest and ability to engage. Policies change when there is sufficient external interest to cause the collapse of previously insulated communities. External attention rises and the issues are considered in a broader political environment where power is more evenly spread and new actors can set the agenda. In both cases, the key focus is the competition to define a policy image (Baumgartner and Jones, 1993: 31). The successful definition of a policy problem as technical or humdrum ensures that issues are monopolized and considered quietly in one venue. The reframing of that issue as crucial to other institutions, or the big political issues of the day, ensures that it will be considered by many audiences and processed in more than one venue (see Box 3.2 on Schattschneider). The modern PET story is about complex systems and attention (Chapter 6). Its analysis of bounded rationality (Chapter 4) remains crucial, since PET measures the consequences of the limited attention of individuals and organizations. However, note the much greater quantification of policy change across entire political systems and the Comparative Agendas Project (CAP) which compares

change in many countries (Baumgartner et al., 2018: 55). Still, the conciseness of explanation, and consistency of empirical findings, is remarkable. PET shows how institutions contribute to ‘disproportionate information processing’, in which attention to information fluctuates out of proportion to (1) the size of policy problems and (2) the information on problems available to policymakers. It also shows that the same basic distribution of policy change – ‘hyper-incremental’ in most cases but huge in some – is present in every political system studied by the CAP.

WHY ‘PUNCTUATED EQUILIBRIUM’ THEORY (PET)? The term ‘punctuated equilibrium’ describes a combination of minor and dramatic changes in policymaking and policy (Baumgartner and Jones, 1993: 19; compare with Hall, 1993; Hay, 2002: 161 on punctuated equilibrium in new institutionalism). In public policy, equilibrium (balance, or stability) can be the result of (1) dominance within government, based on a widely accepted framing of a policy problem and/or the enforcement of the status quo, and (2) balance between competing political forces. Punctuation refers to a policy change associated with (1) framing a policy problem in a different way to mobilize previously uninvolved actors and (2) new imbalances between competing political forces. The overall appearance of equilibrium masks the ever-present potential for instability. Although PET began with reference to an evolutionary biology metaphor, it is now better understood with reference to complex systems (Chapter 6): Complexity in political systems implies that destabilizing events, the accumulation of unaddressed grievances, or other political system processes can change the ‘normal’ process of equilibrium and status quo on the basis of negative feedback (which dampens down activities) into those rare periods when positive feedback (which reinforces activities) leads to explosive change for a short while and the establishment of a new policy equilibrium.

(Baumgartner et al., 2018: 57; see also Baumgartner and Jones, 2015: 203–7) This language of complexity contributes to two different but complementary ways in which to use a telescope metaphor to imagine the policy process. First, by zooming in to case studies, we can observe the ways in which actors promote stability or instability. Some actors try to foster equilibrium by creating or maintaining institutions to support a policy monopoly and defending it by mobilizing against challenges by other groups. Policymaking may be stable when those actors produce an arrangement on how they will govern, such as to agree to minimize visible conflict. Punctuated equilibrium occurs if this strategy is unsuccessful and the policy monopoly is destroyed. A new approach to defining and solving a policy problem legitimizes the involvement of previously excluded groups and encourages previously uninvolved actors (often in different venues) to become involved. The result may be long periods of policymaking stability disrupted by sudden instability. Some policy monopolies are created and maintained while others are destroyed. Second, by zooming out to the whole system, we can observe the aggregate effects of such activity. In most cases, institutions process information about issues routinely in the absence of any internal (‘endogenous’) or external (‘exogenous’) forces for change. In some cases, new ways of thinking can prompt a long-term ‘explosive process’ and we may not return to ‘equilibrium for a very long time’ (Baumgartner and Jones, 2009: 280–81). Relatively few policy punctuations are associated with ‘big bang’-style external events (Studlar and Cairney, 2014). In complex systems, internal changes to rules or procedures can contribute to large changes in outcomes (Chapter 6).

PET’S ORIGINAL FOCUS: AGENDAS AND INSTABILITY

The history of PET matters. If you are reading PET studies for the first time, and look at state-of-the-art PET studies now, it is difficult to relate them to PET’s original story (in Baumgartner and Jones, 1993). We could ask new researchers to stop reading the original, since it provides a misleading account of current research. However, this approach is problematic to say the least, and it would be difficult to set a dividing line for old and new approaches. Instead, it seems more sensible to analyse PET’s origin story, to show how this field began and developed, and relate it to current research. This story begins with a focus on policy monopolies, to describe (1) the oftenclose and often-stable relationships between policymakers and influencers in ‘subsystems’ (Box 9.1) and (2) the role of agenda setting in focusing the attention of policymakers. To be honest, these classic concepts are too important to ignore, and I need to explain them somewhere, so we’re doing it here.

POLICY SUBSYSTEMS, MONOPOLIES, AND SUBSYSTEMS In the early 1990s, PET studies in the USA resembled closely the study of policy networks in countries like the UK. Marsh and Rhodes (1992a) identify a continuum of group– government relationships based on factors such as the number of participants and frequency and nature of contact. As Table 9.1 suggests, they situate Richardson and Jordan’s (1979) concept of policy community at one end of the continuum, while Heclo’s (1978) issue network represents its polar opposite. For them, policy community suggests a close, stable, and often consensual relationship between a small number of groups and government (but see Box 9.1 on the Europe/US usage). In contrast, ‘issue network’ suggests a wide variety of links between the government and many groups, in which there is less agreement and less stability. Table 9.1 Typology of policy networks

Policy community

Issue network

Numbers of participants

Small – many are excluded

Large – the barriers to entry are low

Nature of consultation

Frequent, high quality

Variable frequency and quality

Nature of interaction

Stable and close

Less stable. Access fluctuates significantly

Levels of consensus

Participants share the same basic understanding of the policy problem and how to solve it. Members accept and support the outcomes

A measure of agreement may be reached but conflict and opposition is more likely

Sources: Adapted from Marsh and Rhodes (1992b: 251); see also Jordan (1981: 98).

This spectrum of group–government relations sums up PET’s initial focus. Baumgartner and Jones (1993) seek to explain the success or failure of attempts by certain groups to establish a policy monopoly. Policy monopoly refers partly to the ‘monopoly on political understandings’, or the ability of certain groups to maintain a dominant image of the policy problem (Baumgartner and Jones, 1993: 6). The maintenance of this monopoly requires a common adherence to the same agenda – on how to govern, and which policy image to favour – and an ability to exclude groups who do not sign up to it. Therefore, monopoly also describes the ‘institutional arrangement that reinforces that understanding’ (1993: 6). When bureaucrats and certain interest groups form relationships, they recognize the benefits – such as policymaking stability – of attempting to insulate their decisions from the wider political process (Richardson and Jordan, 1979). In some accounts, this stability hinges on socialization. Inclusion within the policy community

depends on gaining personal trust, through the awareness of, following, and reproduction of ‘rules of the game’. The learning process involves immersion within a ‘common culture’ in which there exists a great deal of agreement on the nature and solutions to policy problems (Wilks and Wright, 1987: 302–303; McPherson and Raab, 1988: 55). What explains the ‘insulation’ of these communities from the wider political process? First, policies are broken down – to such a low level of government and high level of niche specialization that few actors have the time or resources to become involved. Second, one ‘rule of the game’ is that participants resolve issues within the community rather than seeking change elsewhere. Participants know that, while they may not agree with all decisions taken, it is counterproductive to highlight these grievances in other arenas where more involvement may dilute their influence. Third, actors weaken political interest further by defining issues as humdrum and technical, limiting attention and the motive of other groups to participate. This is particularly true when the policy problem appears to have been solved, with only technical aspects of implementation to deliver. If successful, they maintain a policy community which is characterized by: • •

•

A limited group membership, based on the use of a certain policy image to exclude most participants and reduce the visibility of decisions; Good-quality relations between groups and government, based on shared values, a shared understanding of the policy problem, and the trust that comes from regular and rewarding exchange (Chapters 7 and 10); Policy and policy community stability, based on a lack of external attention and the fruitful exchange of resources between groups and public officials (Jordan and Maloney, 1997). Box 9.1 The changing world of policy networks and group– government relations

The study of group–government relations produced many approaches and names to describe the nature and frequency of contact. They include competitive pluralism, state corporatism, sub-government, iron triangles, subsystems, policy communities, and issue networks (Jordan, 1981; Baumgartner and Jones, 1993: 7; Jordan and Schubert, 1992). While ‘issue network’ describes relatively loose relationships between public officials and many interest groups, ‘policy community’ began in the UK as a wider description of networks (Richardson and Jordan, 1979; Jordan, 2005) before Marsh and Rhodes (1992b) assigned it to a more specific meaning akin to a monopoly. In the USA, ‘policy community’ is often understood as an open network (see Kingdon in Chapter 11). There, the ‘iron triangle’ once described ‘strong and exclusive relationships between congressional committees, federal agencies, and interest groups’ (‘cozy triangle’ and ‘policy monopoly’ describe far less extreme insulation), in contrast to ‘velvet triangles’ describing networks of actors who are ‘female in a predominantly male environment’ (Cairney and Rummery, 2018: 545; Woodward, 2004: 84). ‘Policy network’ became the generic umbrella term for group– government relationships in the UK, EU, Australia, and Canada. Many use this term metaphorically (Dowding, 1995b). Some try to measure and visualize key aspects of network activity, such as how actors share information (Ingold et al., 2016) and how many (close or distant) connections each actor has with others (Oliver and Faul, 2018). ‘Subsystem’ is the more frequently used general term in the USA, but it often describes a venue in which actors interact rather than the relationship between actors (see Chapter 10). In Europe, early studies compared pluralism (bargaining with government and competition between large numbers of participants) and corporatism (formal collaboration between the government and a restricted number of large groups representing business and labour). In the USA, early studies identified relatively insulated and uncompetitive group–government relationships within a political system assumed to be pluralistic (iron triangles). Now, there is a greater focus on the need to explain instability and

policy change. In the USA, it may be a response to the identification, since the 1970s, of a more crowded political system – containing a much larger number of groups, experts, and other policy participants – which makes it much more difficult for policy issues to be insulated from attention and for groups to restrict debate. In the UK, it may follow the experience of the Thatcher government (from 1979) and the imposition of policy change in the face of widespread opposition, rather than through negotiation and compromise within communities (Marsh and Rhodes, 1992a). In the EU, it may follow the identification of a policy process which is ‘more fluid and unpredictable – and less controllable – than seems to be implied by enthusiasts of the network approach’ (Richardson, 2000: 1008). In each case, the group–government world appears to be changing – and becoming more ‘multi-centric’ (Cairney et al., 2019) – and our aim is to explain the impact of this change. If the attempt is unsuccessful, it suggests that the participants cannot insulate the process from a wider audience and there is effective competition to define the policy’s image (partly as a problem that is urgent, important, and has not been solved). More groups become involved, and there is greater competition for access to government and instability caused by group conflict. In the USA, this shift is captured by the study of ‘issue networks’.

ISSUE NETWORKS AND SUBSYSTEMS Heclo’s (1978) study of the US executive challenges the ‘received opinion’ that explains most public policymaking with reference to iron triangles linking ‘executive bureaus, congressional committees and interest groups with a stake in particular programs’ and excluding other political actors: the iron triangle concept is not so much wrong as it is disastrously incomplete … Looking for the closed triangles of control, we tend

to miss the fairly open networks of people that increasingly impinge upon government. (1978: 88) Heclo argues that the huge post-war rise in federal responsibilities contributed to an administration becoming increasingly stretched, as more groups mobilized in response to the growth of government. The simple ‘clubby days of Washington politics’ gave way to ‘complex relationships’ among a huge, politically active population (1978: 94, 97; Baumgartner and Jones, 1993: 177–78; see Jordan, 1981: 98 for similar trends in the UK). Issue networks comprise many participants with ‘variable degrees of mutual commitment’, and participants ‘move in and out of the networks constantly’ (1978: 102). The boundaries of networks are fluid. The barriers to entry are low, and based more on the ability to contribute to debate than material power or a common adherence to a policy image. The network is relatively unstable and resolutions to debate are conducted ‘rarely in any controlled, well-organized way’ (1978: 104). Issues which were once ‘quietly managed by a small group of insiders’ have now become ‘controversial and politicized’ as the role of policy activists increase and previously insulated policy communities collide (1978: 105). Baumgartner and Jones (1993: 45) provide a range of examples demonstrating this process. Policy communities ‘relating to tobacco, pesticides, air and water pollution, airlines, trucking, telecommunications, and nuclear power were all destroyed or radically altered’. However, we should not exaggerate Heclo’s position and assume that group– government relations are pluralistic. This wider political process may mask ‘the real locus of decision’ (Richardson, Gustafsson and Jordan, 1982: 2) and it is possible for a policy community to exist within an issue network (Read, 1996). Heclo’s (1978: 105) qualification is crucial: issue networks ‘complicate calculations’ and ‘decrease predictability’, but it would be ‘foolish’ to argue that they replace ‘the more familiar politics of subgovernments’ (compare with True et al., 2007: 158). The ‘politics of subgovernments’ is still compelling. The sheer size of government means that most policy decisions are effectively beyond the reach (or interest) of government ministers or presidents (Heclo, 1978: 88;

Richardson and Jordan, 1979). When policymakers focus on one issue, they have to ignore 99 others. The policy process is broken down into more manageable issues involving a smaller number of interested and knowledgeable participants. Using the language we encountered with complexity theory (Chapter 6; and Simon, 1976: 242–43), PET describes a combination of parallel and serial processing (Jones, 1994). The government processes a huge number of issues simultaneously. Most public policy is conducted primarily through small and specialist policy communities which process ‘technical’ issues at a level of government not particularly visible to the public, and with minimal involvement from senior decision makers. Therefore, overall, we are unlikely to witness the types of radical policy shift associated with a change of government, president, or party control in Congress. In some cases, these relationships break down and policies change. Serial processing at the ‘macropolitical’ (highest, or most senior) level replaces parallel processing at a low level of government (True et al., 2007: 158–59). Thus, any characterization of group– government relationships is a ‘snapshot of a dynamic process’ in which stable relationships can endure or be destroyed (Baumgartner and Jones, 1993: 6; Baumgartner et al., 2018: 58). But what causes this shift? Early PET studies suggest that policy change follows a mutually reinforcing process of increased attention, venue shift, and shifting problem definition (True et al., 2007: 160). This dynamic is a key concern of agenda setting studies.

AGENDA SETTING AND ATTENTION The agenda setting literature focuses on (1) levels of public, media, and government attention to particular issues and (2) what causes attention to rise or fall (see Zahariadis, 2016). It relates to three main factors. First, the pre-existing prejudices of the audience. Different audiences will be receptive to different policy issues. Or, different issues may occupy the top of the agenda in different arenas. This effect is more marked if we divide the policy agenda into smaller parts and measure issue attention in local, regional, national, and supranational governments, the legislature, or the courts. Yet, in

some cases, a range of audiences may be receptive to similar issues simultaneously. Media represent the ‘means of communication’ between multiple venues which are often ‘tightly linked’, with ‘shifts in attention in one … quickly followed by shifts in others’ (Baumgartner and Jones, 1993: 107). Such attention does not necessarily prompt rapid and major policy change. The US legislative progress ‘slows down as media attention increases’ (Wolfe, 2012: 109) and issue expansion can make powerless groups even worse off (Baumgartner et al., 2018: 81). Second, the significance and immediacy of issues. Some policy problems are more important than others, while some require quick decisions. For example, economic issues (such as unemployment, interest rates, or tax) always remain high on the political agenda, while natural and human disasters often focus attention (Birkland, 2016). Yet, since the attention of audiences is limited, and the number of potential issues is almost infinite, the significance of each issue and event is subject to interpretation, debate, and competition. People combine cognition and emotion to process information, and what may seem like an objectively important and urgent problem may be perceived subjectively as too distant or abstract to command concern (Chapter 4). Third, the ability of actors to exercise power to draw attention to one issue at the expense of another. Agenda setting describes ‘an ongoing competition among issue proponents to gain the attention of media professionals, the public, and policy elites’ (Dearing and Rogers, 1996: 1; compare with McConnell, 2018 on ‘hidden agendas’). As Chapters 4 and 11 suggest, this competition is to draw selective attention to particular: facts to reduce uncertainty, and interpretations to reduce ambiguity (Zahariadis, 2007: 66; Cairney et al., 2016: 399; Cairney, 2019c). Baumgartner and Jones (1993: 11– 12, drawing on Kingdon, 1984: 3–4) focus on the effect this attention has on the governmental agenda (the problems that governments ‘are paying some serious attention at any given time’) and the decision agenda (the problems ‘that are up for an active decision’). In this context, agenda setting can be summed up with two key statements:

There is an almost unlimited amount of policy problems that 1. could reach the top of the policy agenda. Yet very few issues do while most others do not. 2. There is huge number of solutions that could be proposed. Yet few policy solutions will be considered while most others will not (Box 9.2). Box 9.2 Participation in American politics: The dynamics of agenda building Cobb and Elder (1972: 105–10) argue that the best chance for an issue to reach the top of the agenda is to maximize the size of its audience. They present the image of four different audiences as circles within larger circles. The smallest circle is the ‘identification group’ which is already sympathetic to an issue proponent’s aims. If an issue expands, it gains the attention of ‘attentive groups’ (who are interested and easily mobilized only on certain issues), the ‘attentive public’ (high income, high educated people with general political interests) and finally the ‘general public’ which is ‘less active, less interested and less informed’ and therefore only becomes aware of an issue if it has a striking symbolic value (1972: 107–108). In many cases, a rise in attention is linked to ‘triggering devices’ or ‘unforeseen events’ such as natural disasters, riots, and protests, while novel issues or issues amenable to piggybacking onto existing debates are the most likely to attract wider attention (1972: 84, 112–18). General public attention makes it likely that an issue will be prominent on the government agenda (1972: 157; see also King et al., 2017: 776). Baumgartner and Jones (1993: 36–37) argue that gaining public attention is one of two strategies open to issue proponents. The second is venue shopping, or seeking more sympathetic audiences in institutions such as congressional committees, state governments and the courts. Therefore, while ‘issue expansion’ (the engagement of a wider audience) requires the reframing of policy problems in terms likely to attract new attention, it need not be in the broadest or least technical terms required by the general public.

PROBLEM DEFINITION Problem definition is central to agenda setting. It refers to ‘what we choose to identify as public issues and how we think and talk about these concerns’ (Rochefort and Cobb, 1994: vii). Policy problems do not simply receive the most attention because they are the most important or immediate. There are no objective indicators to determine which ‘real-world conditions’ are the most deserving of our attention. Further, the ability of policymakers to receive and act on ‘signals’ or information about the severity of policy problems is imperfect (Jones and Baumgartner, 2005: 8; Chapter 4). Attention is linked more strongly to the ability of actors to convince enough people that ‘their’ issues are the most worthy of discussion (Dearing and Rodgers, 1996). This may not involve a battle over the overall accuracy of ‘facts’, but to direct attention to other facts which support a rival policy image (Baumgartner and Jones, 1993: 107–18, 113). Attention-grabbing strategies depend on the following factors: Framing. Framing involves the definition of a policy’s image, to portray and categorize issues in specific ways (Box 4.4). People combine cognition and emotion to understand the world, and framing is ‘a mixture of empirical information and emotive appeals’ (True et al., 2007: 161; Cairney and Kwiatkowski, 2017). Actors can frame policy problems to make them appear technical and relevant only to experts, or link them to wider social values to heighten participation (Rochefort and Cobb, 1994: 5). To attract attention, one may link ideas to popular values such as (in the USA): ‘progress, participation, patriotism, independence from foreign domination, fairness, economic growth’ (Baumgartner and Jones, 1993: 7). Or, issue expansion requires reframing from a focus on self-interest to a problem that the public (or powerful sections within it) can relate to (Hogwood, 1987: 30; Box 9.2). Most policy issues are multifaceted and can command a wide range of images. For example, many actors have framed smoking in terms of health, public health, health and safety, health inequalities, nuisance, employment, the economy, trade, customs and excise, the

role of multinational corporations, civil liberties, and human rights (Cairney et al., 2012). Yet, while there are many ways to frame the same problem, there is limited attention to devote to issues. So, highly complex issues are simplified, with attention to very few aspects coming at the expense of the rest. This selective focus has policy consequences. For example, a focus on smoking as an urgent public health issue prompts governments to restrict tobacco use, while an economic image prompts governments to support the tobacco industry (Mamudu et al., 2015). Baumgartner and Jones (1993: 109) suggest that, while certain groups may exploit selective attention to create and protect a policy monopoly based on one image for long periods, the agenda setting process is too dynamic to expect it to go on forever. Indeed, the basis for creating monopolies – limited time, attention, cognitive abilities – also contributes to radical shifts of attention to different policy images. Policy change can result from a major shift of attention rather than a major change of beliefs or preferences (Baumgartner et al., 2018: 65; Koski and Workman, 2018: 296). Causal stories, responsibility, and the availability of a solution. For Kingdon (1984: 115), policy issues only become problems when there is a solution and ‘we come to believe that we should do something about them’ (Chapter 11). This belief may follow new theories on the determinants of problems, who is to blame and who is responsible for solving the problem. For example, poverty became a US public policy issue in the 1960s because of a shift in the issue’s image ‘from that of a private misfortune to a public problem amenable to government solutions’ (Baumgartner and Jones, 1993: 28; Rochefort and Cobb, 1994: 15). Stone (1989: 282–83; 2002: 191) suggests that assigning responsibility is a strategy of organizations seeking to prompt or justify policy intervention. They use ‘causal stories’ which highlight the cause of the problem and who is to blame (compare with studies of ‘blame game’ management – Boin et al., 2010). For example, several studies shifted long-held beliefs about blame in health and safety (from the worker to the employer), environmental damage (from natural phenomenon to human pollution), and road safety (from careless drivers to car manufacturers). In each case, this shift of attention prompted

equivalent shifts in public policy (compare with the Narrative Policy Framework in Chapter 4). In most cases, there are likely to be many competing stories of blame. Rochefort and Cobb’s (1994: 1–4) discussion of the LA riots in 1992 highlights multiple causes: a justified reaction to police racism; poor immigration controls; a breakdown in law and order; the failure of Lyndon Johnson’s Great Society measures in inner cities; the Republican neglect of race relations; and a ‘poverty of values’ (compare with the UK Government’s ‘troubled families’ explanation of London riots in 2011 – Cairney, 2019d). The same can be said for valence issues. While most agree that child, drug, and alcohol abuses are problems, there is much less agreement on assigning responsibility and producing solutions (Baumgartner and Jones, 1993: 150). Studies of the ‘social construction of target populations’ (Schneider and Ingram, 1993) highlight a competition to assign praise or blame social groups to help decide if they should receive government benefits or sanctions (Chapter 4). Crises and triggering or focusing events. In many cases, crises such as environmental disasters act as a ‘triggering event’ or ‘focusing event’, focusing public, media, and government attention to an issue previously lower down on the agenda (Birkland, 2016; Dearing and Rogers, 1996: 37–39). They can act as ‘dramatic symbols of problems that are already rising to national attention’ (Baumgartner and Jones, 1993: 130). Events have high potential to focus attention if they appear to arise suddenly, are unusual or uncommon, seem immediately or potentially harmful, can be connected to a specific geographical area or social group, and their effects are known to the public and policymakers at the same time (Birkland, 1997: 22; 2016: 392). Examples include terrorist attacks, major oil spills causing damage to wildlife, and environmental disasters with human consequences, particularly when they are of an unusual scale (2016: 394). However, the event is only one part of the story: ‘Focusing events are advocacy opportunities that can be used by policy entrepreneurs to advance their preferred problem frames and solutions on the agenda’ (2016: 394; Chapter 11). This propensity to give disproportionate attention to disasters (rather than more ‘mundane’

events which cause more deaths) also allows issue proponents to highlight the potential for dramatic crises, such as a nuclear power plant disaster (Baumgartner and Jones, 1993: 118–21) or the global spread of ‘non-communicable diseases’ (NCDs) caused by factors such as smoking or obesity (Studlar and Cairney, 2019). In other cases, the appearance of crisis depends on a widespread belief that something has gone wrong, is causing catastrophic effects, and needs urgent action to prevent further catastrophe. For example, the global financial crisis from 2007 focused attention on solutions such as financial regulation (Gava, 2016: 433). Or, problems are simply labelled as crises ‘to elevate a concern when facing an environment overloaded with competing claims’ (Rochefort and Cobb, 1994: 21). Measurement. Although the measurement of a problem may seem like a straightforward and technical matter, it can be subject to as much interpretation and debate as issue framing. Policy problems may be so complex or ambiguous that there is room to interpret the measures we should use. For example, the way we understand and measure inequality can relate to factors such as income and wealth, gender, race and ethnicity, sexuality, disability, and mental ill-health (Cairney and St Denny, 2020: 9). The measurement of poverty can relate to a household or individual; mean or median incomes; absolute or relative poverty; income; wealth; and inequalities in public service access. It may also have a different definition in domestic and international settings, and change according to technical factors relating to measurement. For example, the UK government defines the ‘absolute poverty line’ in relation to an individual ‘if their household income is less than 60% of real median income’: 60% of approximately £64 (£38, or $60) per day in 2011 (Office for National Statistics, 2013: 5). The World Bank used $2 per person per day to define absolute poverty and $1 to define extreme absolute poverty, before raising the latter to $1.90 in 2014 (Hood and Waters, 2017: 27; Ferreira, 2015; World Bank, 2008; Jones and Baumgartner, 2005: 31). Measurement debates have policy consequences. For example, the treatment of civil rights issues in the USA changed significantly when the measure of racism shifted, from the need to prove a person’s intent to discriminate to statistical evidence proving that an

overall selection process could not have happened by chance (Stone, 1989: 291). Governments can also change the scope of their measurements. For example, in the UK, unemployment means the number of people claiming unemployment benefit, so the government reduced unemployment on paper by restricting the number of people who qualified for this benefit. It contributed to a high number of economically inactive people who claimed sickness rather than unemployment benefits (Webster, 2002; Machin, 2004) and a government response to restrict disability-related benefits (Carter and Whitworth, 2015). Box 9.3 Up and down with ecology: The ‘issue attention cycle’ Do peaks in attention cause policy change? Downs (1972) argues that public attention is fleeting, even when it involves ‘a continuing problem of crucial importance to society’. A rise in interest does not mean a worsening of the problem, while falling interest does not suggest that the problem has been solved. This point is a key tenet of the literature (Baumgartner and Jones, 1993: 47). The issue attention cycle has five stages: 1. Pre-problem – a problem alarms experts but does not yet capture public attention. 2. Alarmed discovery and euphoric enthusiasm – concentrated public attention is accompanied by a widespread hope that the problem can be solved quickly. 3. Realizing the cost of significant progress – when the public realizes that the solution involves major costs or a significant change in behaviour. 4. Gradual decline of intense public interest – people feel discouraged at the prospect of change and shift their attention to the next issue at stage two. 5. Post-problem – public attention is minimal or spasmodic and the problem has been replaced by another issue.

Downs posits a weak link between public attention and policy change, since the public is rarely engaged long enough to see matters through. However, in many cases, there is a policy response which creates new institutions that operate long after public interest has waned (Downs, 1972). Peters and Hogwood (1985: 251; see also Hogwood, 1992b) find a positive relationship between public attention and government reorganizations. As Baumgartner and Jones (1993: 87) argue, peak periods of organization change ‘generally coincided with Gallup Poll data showing public concern with the same problems’. Therefore, ‘the public is seriously involved in the agenda setting process, not an ignored bystander’ (Jones and Baumgartner, 2005: 269). The policy responses from public attention may have long-term effects. A realization of the costs of policy change may occur only after legislation has passed and policy is being implemented (Peters and Hogwood, 1985: 239). New government organizations are created but ‘do not simply “fade away” like public interest or media attention’ (Baumgartner and Jones, 1993: 84–7; 191; Baumgartner and Mahoney, 2005).

PROBLEM DEFINITION, POLICY MONOPOLIES, AND VENUE SHOPPING Venue shopping relates to Schattschneider’s (1960) suggestion that power involves the competition to draw more or less attention to conflict: the winners try to ‘privatize’ and the losers ‘socialize’ conflict (Box 3.2). Baumgartner and Jones (1993) highlight the role of government as a source for socialization and privatization (compare with Genieys and Smyrl, 2008a, 2008b; Jobert and Muller, 1987; Cairney, 2012a: 231). For example, if actors organize issues out of politics by creating a policy monopoly, their opponents ‘will have the incentive to look for allies elsewhere’ (1993: 35–7). Policy monopolies are common, but so too is the ability to challenge monopolies and make authoritative decisions in other venues. The main options for issue proponents are two-fold. First, they can challenge the dominance of a monopoly within this venue.

The attitudes of policymakers are susceptible to change: ‘decisionmakers value or weight preferences differently depending on the context in which they are evoked’ (Jones, 1994: 5). They have many (often-contradictory) objectives, most issues are multifaceted and there are many ways to solve policy problems. Therefore, it may be possible to shift the opinions of decision makers by ‘shifting the focus of their attention from one set of implications to another’ (Baumgartner and Jones, 1993: 30). As Box 4.4 on ‘heresthetic’ suggests, a group of people with the same beliefs can draw different conclusions according to the context and order in which they consider issues and make choices. Second, if unsuccessful, actors can venue shop to seek more sympathetic audiences (Baumgartner and Jones, 2009: 276). Each venue may be relatively receptive to a different policy image. The ability to venue shop has increased since the post-1945 period. An increasingly crowded political system makes it more difficult for policy issues to be insulated from the wider political process (Baumgartner and Jones, 1993: 43). In most cases, there is no natural jurisdiction for policy problems and no ‘ironclad logic’ for an issue to be considered at, say, the national rather than international or local level. It is also common for policymaking responsibility to shift over time (1993: 32–33; Cairney et al., 2019), or for many cross-cutting issues (such as climate change) to be ‘bundled’ together (Koski and Workman, 2018). This premise allows us to explain why policy change follows a mutually reinforcing process of increased attention, venue shift, and shifting policy images. The key strategy is to involve the previously uninvolved. If issues ‘break out’ of these policy communities and are considered in one or more alternative venues, then the scope for new ways to examine and solve policy problems increases. As people come to understand the nature of a policy problem in a different way (and this new understanding of the problem is more closely linked to their priorities), more people become interested and involved (1993: 8). The more ‘outside’ involvement there is, then the greater likelihood of a shift of the policy image, as new actors propose new ways to define and solve a policy problem.

CASE STUDIES OF PUNCTUATED EQUILIBRIUM: ‘SOME ISSUES CATCH FIRE’ Baumgartner and Jones (1993) compare case studies over several decades to demonstrate the creation and destruction of policy monopolies. Environmental policy demonstrates the use of venue shopping to change a policy’s image, with the new involvement of one venue producing a snowball effect. Environmental groups, unhappy at losing out on regulatory decisions made by the federal government, appealed to members of Congress. Congress became more sympathetic to a new image of environmental policy and passed legislation (the National Environmental Protection Act, NEPA, 1969) to regulate business, allow groups greater access to the courts, and pave the way for a federal response (the creation of the Environmental Protection Agency) in line with the new problem definition (1993: 38). In nuclear power, Baumgartner and Jones (1993: 59–82), describe the effect of a ‘Downsian mobilization’ (Box 9.3). The governmentinspired enthusiasm for nuclear produced a positive policy image – stressing the safe use of nuclear materials, reduced energy bills, independence from imported oil, employment, reduced air pollution, and the economic benefits of new technology – which supported the formation of a post-war policy monopoly. As the process moved from policy formulation to implementation, public, media, and government attention fell and the details of policy were left to the private sector experts, federal agencies (e.g. Atomic Energy Commission, AEC), and certain congressional committees. From the late 1960s, there was increased opposition from environmentalists, local activists, and nuclear scientists expressing concerns about safety. It suggests a ‘Schattschneider mobilization’ (Box 3.2), or an expansion of the scope of conflict. The safety agenda was pursued through public hearings on nuclear licensing and, by the 1970s, public and media attention rose to reflect new concerns (the number of negative news articles began to outnumber the positive). Many venues became involved. The focus of Congress (and an increasing number of committees) became increasingly

negative and NEPA reinforced more stringent regulation. It was followed by a series of decisions in the courts, including the retrospective application of NEPA to previous AEC licensing decisions, and supplementary action by state and local governments (including planning laws to delay the building of new plants). The new policy image was cemented in the public’s mind by the accident at the Three Mile Island nuclear plant. The private sector lost confidence in the industry. Overall, the combination of interest-group opposition and venue shopping shifted the policy image of nuclear power from highly positive to overwhelmingly negative. The policy monopoly endured for over 20 years before being destroyed (although the existing nuclear plants remained profitable, and nuclear still accounts for 20% of US electricity supply – Abdulla, 2018). Increased regulation replaced a post-1945 policy of power plant expansion. Baumgartner and Jones (2009: 260–64) suggest that a new policy image has emerged recently in the context of climate change, prompting some national governments to reinforce support for nuclear as a source of low carbon electricity. However, disasters such as Fukushima in 2011 also keep safety on the agenda (prompting plans to end nuclear capacity in Germany, and reform regulation in India – Fuchs, 2015: 208; Rabbi and Sabharwal, 2018) and, without continuous long-term US government support, nuclear power has struggled to compete economically and politically with other sources of electrical energy (Abdulla, 2018). In early twentieth-century tobacco policy, actors built a policy monopoly on a positive policy image (economic benefits) and deferral to the experts in agriculture. Tobacco consumption was high, media coverage was low and generally positive (smoking was normal, and had a glamorous image), and tobacco companies had a reputation for patriotism after ensuring a supply of cigarettes during the Second World War (Cairney et al., 2012: 48). From the 1950s, there was significant issue expansion, with anti-tobacco groups increasingly setting the media agenda with a negative policy image. Tobacco consumption fell dramatically from the 1950s and in 1964 the US Surgeon General’s report cemented a new image based on ill-health. Congressional attention followed in the mid-1970s, with

health committee hearings outnumbering agriculture and public health groups increasingly involved and able to reframe the debate (Baumgartner and Jones, 1993: 114; 210). By the 1980s, this new image was pursued through a series of high-profile court cases and by states such as New York and California providing best practice and the scope for policy diffusion (Studlar, 2002). By the 2000s, tobacco control became restrictive and the old pro-tobacco monopoly was ‘destroyed’ (Baumgartner and Jones, 2009: 280). In the post-war period, actors built a pesticides policy monopoly on a positive policy image: eradicating harmful insects, boosting agricultural exports, and ending world hunger by boosting food production. There was public and media ‘enthusiasm for progress through chemistry’ and deferral to experts in agriculture and the chemical industry, who enjoyed a close relationship with the Department of Agriculture, USDA, and the congressional agriculture committee (Baumgartner and Jones, 1993: 95). Environmental concerns were marginalized, consumer interests ignored, and the Food and Drug Administration (FDA; within the USDA) had no jurisdiction. This ‘golden age’ was punctured in 1957 following the devastating failure of two large insect eradication campaigns and in 1959 following the FDA’s decision to ban the sale of a crop tainted with pesticide residue. It shifted media and congressional attention from the economic benefits of pesticides to their ineffectiveness, adverse health effects, and environmental damage. By the late 1960s, environmental groups had found support within multiple venues (Congress, the courts, executive and state agencies) and the regulation of pesticides rocketed.

FROM CASE STUDIES TO THE ‘GENERAL PUNCTUATION HYPOTHESIS’ Jones and Baumgartner (2005: 278) show that punctuated equilibrium is a system-wide effect, rather than specific to these case studies (see also Baumgartner et al., 2018 for case studies by other authors). Their focus shifted to general explanations of agenda setting and policy change, based on non-linear dynamics in

complex systems (Chapter 6, p.104). Systems, and the actors within them, provide ‘feedback’ to actions: dampening some (negative feedback) and amplifying others (positive feedback). While the former acts as a counterbalance to political forces, the latter reinforces those forces to produce radical change (Baumgartner and Jones, 2002: 8–16). As a result, small actions can have large effects and large actions can have small effects. A key dynamic is ‘disproportionate information processing’: policymaker responses are not in proportion with the ‘signals’ that they receive from the outside world (Jones et al., 2009). Instead, they pay attention to some and ignore the rest: ‘most members of the system are not paying attention to most issues most of the time’ (Baumgartner, 2017: 72; Shaffer, 2017). Baumgartner and Jones’ (1993) case studies already describe key aspects of disproportionate information processing: when policy monopolies are maintained, policymakers appear unreceptive to new information. During periods of punctuation, they appear to become hypersensitive to new information (Jones and Baumgartner, 2005: 8; 18). The ‘general punctuation hypothesis’ goes further to identify the system-wide effect of policymaking organisations receiving ‘feedback’ on their policies (e.g., from interest groups, the media, or public opinion) without responding proportionately. The ‘selective attention’ of policymakers and organisations explains why issues can be relatively high on certain agendas, but not acted upon, and why powerful signals are often ignored: •

•

Policymakers are unwilling to focus on certain issues, either because ideology precludes action in some areas, because there is an established view within government about how to address the issue, or because the process of acting ‘rationally’ (making explicit trade-offs between choices) is often unpopular. They are unable to give issues significant attention. Boundedly rational policymakers set simple goals but struggle to generate enough information of complex systems to understand how to achieve them (Jones and Thomas, 2017: 49; Chapter 4). Further, their focus on one issue means ignoring 99 others.

Policymaking organizations are more able to gather information routinely, but are subject to the same maxim on a larger scale: environmental ‘signals’ on available information are almost infinite, while information processing capabilities are always finite (and often low). Low information capacity helps explain why, even when governments pay high attention to some issues, their searches for information miss key elements and they fail to respond to changing circumstances proportionately (Workman et al., 2017; Koski and Workman, 2018; Epp, 2017). The idea of ‘institutional friction’ or ‘stickiness’ is that rules in political systems make it more or less difficult to turn new information and other ‘inputs’ into policy ‘outputs’ (Baumgartner et al., 2018). For example: • •

Policymaking organizations establish rules and standard operating procedures to gather information. These rules skew information searches, missing or rejecting some information, and ruling out certain ways of thinking or acting.

A major change in policymaking and policy involves costs that policymakers are often unwilling or unable to bear: the ‘cognitive costs’ to understand a problem, and ‘institutional costs’ to change well-established rules (Baumgartner et al., 2018: 69). It may require a critical mass of attention and pressure to overcome the conservatism of policymakers – and institutional friction – and shift their attention to new problems or ways of interpreting them (Jones and Baumgartner, 2005: 19–20; 48–51). Partisan ideology may be one obstacle, but dramatic party changes are only one of many sources of ‘policy disruptions’ (Baumgartner and Jones, 2009: 287). Broadly speaking, if levels of external pressure reach a tipping point, they cause major and infrequent punctuations rather than smaller and more regular policy changes. A burst in attention and communication becomes self-reinforcing: new approaches are considered; different ‘weights’ are applied to the same categories of information; policy is driven ideologically by new actors; and/or the ‘new’ issue sparks off new conflicts (2005: 52; 69). The effect is like a dam that contains the flow of water, until there is so much pressure

that it breaks down and a huge amount of water flows quickly (indeed, Koski and Workman, 2018: 305 describe the ‘flow of information’). Information processing is characterized by ‘stasis interrupted by bursts of innovation’ and policy responses are unpredictable and episodic rather than continuous (2005: 20). Crucially, the more ‘friction’ we find, the more we expect higher stability followed by more dramatic change. The system ‘often works to reinforce conservatism, but it sometimes works to wash away existing policy subsystems’ (Baumgartner et al., 2018: 58). PET scholars do not really try to identify and measure precisely a just-right level of attention or reaction to a problem (akin to the story of Goldilocks). However, a reference to ‘disproportionate’ suggests too little or too much attention and the potential for major policy consequences. Further, the concept of ‘policy bubbles’ describes ‘underreaction and overreaction’ to policy problems, including a ‘severe’ and long-term ‘overinvestment’ of government resources in ‘a single policy instrument beyond its instrumental value in achieving a policy goal’ (Jones et al., 2014: 146–49; compare with Maor, 2014). For example, governments may over-invest in an instrument because it has ‘symbolic or ideological’ value, ‘solving’ an ill-defined problem politically rather than substantively (Jones et al., 2014: 149; compare with ‘good politics, bad policy’ – McConnell, 2010: 78). Actors cannot simply change this way of thinking with reference to ‘evidence-based’ or ‘rational’ arguments, particularly if (1) policy promises escalate following a ‘bidding war’ between political parties, or (2) the policy supports actors’ fundamental beliefs, such as their faith in a government or market solution (2014: 150–52; compare with Chapter 10, and Baekgaard et al., 2017; Cairney and St Denny, 2020: 204). Or, a solution appears to work at first, and achieves ‘lock in’ before its limited benefits are known (compare with path dependence, inertia, or incrementalism in Chapters 4 and 5). Similarly, most PET studies do not focus on improving system design, but the practical effect of diagnosing a potentially overwhelming amount of information, inefficient information searches, ‘sticky’ institutions, and the misallocation of budgets could be profound (Koski and Workman, 2018; Baumgartner et al., 2018: 68).

GOVERNMENT BUDGETS: HYPERINCREMENTAL AND DRAMATIC POLICY CHANGE The Policy Agendas Project extends the analysis from particular case studies to the dynamics of policy change, government budgets, elections, and policymaker attention. Most notably, it maintains a long-term and comprehensive analysis of post-1945 US public expenditure (Jones and Baumgartner, 2005: 111; True et al., 2007) and longer-term budgets (Jones et al., 2009). Public expenditure changes are key indicators in public policy because (1) they allow us to measure policy changes in a systematic way and (2) spending commitments are more concrete than more vague policy intentions (Chapter 2). The results are displayed in Figure 9.1 which highlights an aggregate of the annual percentage change of US budgets in each policy area (such as health, social security, and education). But what does the figure tell us and how should the results be interpreted? First, there is not a normal distribution. This is a statistical term which refers to (1) the mean and (2) the standard deviation from the mean. A normal distribution suggests that most (68%) of all the values are slightly different from the mean, while almost all (95%) are no more than moderately different. In other words, while there will be a small handful of instances in which the values differ markedly from the mean, most values are bunched closely together, while some are further away from the mean, but not remarkably so (Box 9.4). Yet Figure 9.1 demonstrates that the distribution of values is leptokurtic. It has two main features. First, a higher central peak (note how much higher the bar reaches above the peak of the normal curve) and lower level of dispersion (note the space between the bars and the dotted line within the normal curve) than we would expect. In the vast majority of cases, the real increase or decrease in annual spending is very low and the size of the increment is very small. Jones and Baumgartner (2005: 112) call this ‘hyperincrementalism’. Second, there are many more outliers than we would expect under a normal distribution. For example, in their post-

war data set with approximately 3,300 values, we may expect approximately 15–20 outliers. Yet the number of cases of annual change greater than +160% is 75 (2005: 110). The analysis suggests that, in a small but very important number of cases, ‘programs received huge boosts, propelling them to double or triple their original sizes or more’ (2005: 112). Overall, budget change is characterized – much like the distribution of natural seismic events like earthquakes – by a huge number of small changes in each policy category, combined with a small number of huge changes (True et al., 2007: 166; Baumgartner et al., 2018: 71).

Figure 9.1 A broad picture of the general punctuation hypothesis in action Source: adapted (normal curve added) from Jones et al. (2009: 861). The figure shows the percentage changes to US budgets from 1800 to 2000. See also the more detailed graphs in True et al. (2007: 170) and Jones and Baumgartner (2005: 111) which show an even higher central peak and longer tails when we analyse only the post-war budget changes. Note that the latter figures track the amount dedicated (not the amount spent) and variation from the mode (a

4% real annual rise is the most common) rather than the mean (Jones and Baumgartner, 2005: 110–11).

But why would we expect a normal distribution anyway? The answer to this question is interesting. At first, Jones and Baumgartner (2005) related it to classic US debates in which many scholars predicted budgeting incrementalism (e.g. Davis et al., 1966). They argued that, if the overall nature of budgeting changes was incremental, we would expect no more than a normal distribution because incrementalism suggests a common type of change in most policy areas: a non-radical change from previous years in most cases, combined with moderate change overall (Jones and Baumgartner, 2005: 120–23). Now, their explanation relates more strongly to an ideal-type of optimal information processing: ‘a perfect pattern of adjustment, to a complex multifaceted environment in which multiple informational input flows are processed by a political system will yield a normal distribution of output changes’ (Baumgartner et al., 2018: 66). In other words, if we assume that (in a democracy) the ‘signals’ to government exhibit a normal distribution, then an optimal adjustment would produce a normal distribution of policy change. Instead, policymaking systems exhibit ‘disproportionate information processing’: there is no shortage of information, but most issues are ignored or receive little attention, while some receive an intense level of attention which produces major policy consequences (Jones and Baumgartner, 2005: 112). Box 9.4 Key terms for statistical analysis Mean – the average, calculated by adding all values together and dividing by the number of values. Mode – the average, calculated by identifying the most frequent value. Standard deviation – a measure of statistical dispersion which generally denotes deviation from the mean (Figure 9.1 uses the mode). If all values are the same, then the SD is zero (e.g. the

numbers 100, 100, and 100). The greater the dispersion (e.g. 0, 100, 200), then the greater the SD. Normal distribution – denotes the level of SD from the mean. It suggests that 68% of all values fall within a range of plus or minus one standard deviation from the mean, while 95% fall within a range of plus or minus two standard deviations. Leptokurtic – a distribution which is not normal because it has a higher central peak (representing many small changes), few moderate changes, and more outliers (large changes). More than 68% fall within ±1 standard deviation but less than 95% fall within a range of ±2. Outlier – a value which is further away from the mean than the normal distribution suggests. Real increase – an increase in spending which takes inflation into account. This data provided initial confirmation of the general punctuation hypothesis: ‘400,000 observations collected as part of the Policy Agendas Project’ demonstrate that the hypothesis ‘is a fundamental characteristic of the American political system’ (2005: 278). Further, the absence of a normal distribution is a feature demonstrated by data on events such as US elections and congressional legislation and hearings (Baumgartner et al., 2009: 611).

THE COMPARATIVE POLICY AGENDAS PROJECT The Comparative Policy Agendas Project (CAP) took PET to another level. Initially, Baumgartner and Jones (1993) described good reasons to think that PET applies mostly to the USA. They used it to explain why the US political system ‘conservatively designed to resist many efforts at change’ also helped produce ‘bursts of change’ (True et al., 2007: 157). Punctuated equilibrium suggests that the key features that explain policymaking stability also help explain

major punctuations. They include the separation of powers (executive, legislative, and judicial), overlapping jurisdictions (between institutions or between federal, state, and local government), and the pluralistic interaction between groups (in which the ‘mobilization of one group will lead to the countermobilization of another’ – Baumgartner and Jones, 1993: 4–5). These checks and balances combine with the ability of organized interests to ‘countermobilize’ to block radical policy change. However, in a small number of cases, mobilizations are accompanied by renewed interest among one or more venues. In such cases, ‘the newcomers are proponents of changes … and they often overwhelm the previously controlling powers’ (True et al., 2007: 157). Therefore, the diffusion of power across US government increases the scope for venue shift and helps groups form new alliances with policymakers capable of challenging monopolies. Yet the causal factors identified by punctuated equilibrium apply to a wide range of political systems (Baumgartner et al., 2018: 55). First, the separation of powers and/or existence of overlapping jurisdictions is not limited to the USA. Indeed, the entanglement of policy issues (when decision-making power is vested in more than level of government) is common to unitary, quasi-federal, and federal systems (Keating, 2005: 18; Cairney, 2006). Second, the multi-level governance (MLG) literature demonstrates the increasing significance of multiple decision-making venues (and issue network rather than policy community relationships – Richardson, 2008: 25) in the EU and countries such as the UK (Chapter 8; Cairney et al., 2019a). Third, the component parts of punctuated equilibrium – bounded rationality, information processing, complex systems, agenda setting, and group– government relations – are central to the policy literature as a whole. Processes such as ‘disproportionate information processing’ are universal, since what we are really talking about is the effect of bounded rationality and friction in complex policymaking environments. The difference in each country relates to the specific ways in which people combine cognition and emotion to make choices, and the specific institutions that make routine and efficient policy change more or less possible (Baumgartner et al., 2018: 66).

By 2009, PET had been applied in Canada, the EU, and many European countries (True et al., 2007: 175; John, 2006; Baumgartner and Jones, 2009: 255; Baumgartner et al., 2006). Jones et al. (2009: 855) identify the same basic distribution of budget changes in the USA, Canada, the UK, France, Germany, Spain, Belgium, Denmark, and South Africa: ‘budgets are highly incremental, yet occasionally are punctuated by large changes’. Further, Baumgartner et al. (2009) suggest that variations in the data may relate as much to internal as well as comparative sources of variation. Much depends on the levels of friction, producing the transactions costs related to policy coordination, which seem to be higher at the beginning of policy development and lower during policy delivery. For example, it is easier for legislative committees to come together to focus on new issues than for large governments to shift their budgets to reflect new priorities. Consequently, the former will display lower levels of kurtosis (change is more ‘normal’) than the latter (which displays relatively high levels of both minimal and dramatic changes) (2009: 609). Koski and Workman (2018: 301– 303) also identify trade-offs between different ways to try to improve information processing: centralization may produce ‘bottlenecks’ and ignore too much relevant information, while the maintenance of multiple venues makes it difficult to coordinate information searches. By 2016, Baumgartner et al. (2018: 82–85) report ‘fifteen country projects’ as part of the Comparative Agendas Project, and 393 PET publications, including major growth and a shift in country focus: US studies accounted for 65% of applications before 2006 and 36% after. Almost all non-US studies are of democratic systems, particularly in Western Europe (e.g., Engeli et al., 2012; GreenPederson and Walgrave, 2014). The latter have prompted some shift of focus towards the role of political parties, but usually to find that the timing of shifting agendas ‘is not closely related to elections or shifts of governments’ (2018: 87). In theory, PET’s abstract logic should hold in authoritarian systems. Authoritarianism is the ultimate source of institutional friction, which we would expect to contribute to a more extreme combination of policymaking stability and relatively rare

‘system-destabilizing policy punctuations’ (Baumgartner et al., 2018: 68). While there are few studies on which to draw, they seem to support this expectation, albeit in relation to very different dynamics. In Hong Kong, Lam and Chan (2015: 549) find greater punctuations during centralized rather than democratized eras. In China, Chan and Zhao (2016: 137–38) associate ‘extreme stability’ and ‘drastic adjustment’ with highly centralized information processing, including a tendency for (1) ‘lower level administrators to manipulate information as it travels up the bureaucratic hierarchy’ and produce policymaker reliance on poor quality information and (2) highly infrequent and dramatic adjustments to policy only when policy problems ‘become serious enough to threaten regime survival’. Further, this effect is less pronounced in regions in which policymakers have greater exposure to policy relevant information (2016: 145). In Hungary, Sebők and Berki (2018: 604) confirm PET’s ‘general empirical law of public budgets’ and find ‘higher levels of punctuation … in autocratic or semi-autocratic regimes’. These studies suggest that PET continues to grow in importance as a truly comparative policy theory, even if it describes remarkably different ways of thinking, institutional design, and levels of punctuation.

CONCLUSION PET explains long periods of policymaking stability and intense periods of instability. A combination of bounded rationality and agenda setting explains how policy monopolies can be created and destroyed. While there are many ways to interpret policy problems, there is only so much time and energy to devote to issues. So, highly complex issues are simplified, with very few aspects considered at any one time. Problem definition is crucial because the allocation of resources follows the image of the policy problem. Policymaking stability depends on the ability of actors to maintain a policy monopoly. A policy monopoly involves the successful definition of a policy problem in a certain way, to limit the number of participants who can claim a legitimate role in the process. As the examples of pesticides and nuclear power suggest, it often follows an initial burst of public and governmental enthusiasm for policy change. Such

enthusiasm produces a supportive policy image, based on the idea that economic progress and technological advance have solved the policy problem, allowing policy monopolies to operate for long periods with very little external attention. After the main policy decision is made, the details are left to policy experts and specialists in government. This allows the participants to frame the process as ‘technical’ to reduce public interest or ‘specialist’ to exclude those groups considered to have no expertise. The lack of attention or external involvement allows communities to build up a policy-delivery infrastructure that is difficult to dismantle, even during periods of negative attention. Policy change is explained by a successful challenge to policy monopolies. Those excluded from monopolies have an interest in challenging or reshaping the dominant way of defining policy problems. This may come from within, by ensuring that new ideas or evidence prompt a shift in government attention to a new policy image. Or, if this new image is stifled, groups venue shop to expand the scope of conflict and promote it to a more sympathetic audience. As the examples of pesticides and nuclear power suggest, the new policy image based on safety concerns produced a bandwagon effect. It began with rising dissent among experts, followed by increased media coverage critical of the status quo. Previously excluded interest groups exploited this shift of focus to attract the attention of actors in other venues. The adoption of the safety agenda in one venue had a knock-on effect, providing more legitimacy for the new image and creating an incentive for decision makers in other venues to become involved. The result was profound policy change following a burst of new regulations by Congress, the courts and multiple levels of government. Policy change therefore follows a mutually reinforcing process of increased attention, venue shift, and shifting problem definition. As people come to understand the nature of a policy problem in a different way, then more people become interested and involved. The more ‘outside’ involvement there is, then the greater likelihood of a further shift in problem definition, as new ideas are discussed and new participants propose new solutions.

The ‘general punctuation hypothesis’ suggests that this effect is not limited to policy monopolies. Rather, the case studies help describe a wider process of ‘disproportionate information processing’. Most government responses are not in proportion with the ‘signals’ that they receive from the outside world. They are insensitive or hypersensitive to policy relevant information. As a result, most policies stay the same for long periods because policymakers are unwilling or unable to pay them sufficient attention. However, in a small number of cases, policy changes radically as policymakers respond to a critical mass of external attention. The burst of governmental attention is accompanied by a sense of policy malaise and a need to play ‘catch up’. New ideas are considered, different ‘weights’ are applied to the same kinds of information, and the ‘new’ issue sparks off new conflicts between political actors. These bursts of attention produce short bursts of radical policy change. The best demonstration of this picture of stability and change can be found in budgeting. Budget change is characterized by a huge number of small changes combined with a small number of huge changes. Overall, the picture is more dynamic than incrementalism suggests, even if most policymaking appears to be incremental (Chapter 4). Although most policy issues display continuity, and there are many sources of ‘institutional friction’, dominant rules and ways of thinking are constantly being created and destroyed. Therefore, any snapshot of the political system will be misleading since it shows an overall picture of stability, but not the process of profound change in many cases, or over a longer period. Instead, think of PET as a way to explain complex system dynamics (Chapter 6). A complex system can appear to produce stability, but with the ever-present potential for sudden instability and change, as it amplifies some signals and dampens others, and policies seem to ‘emerge’ in the absence of anyone’s control. Frame – to define a policy’s image or define the policy problem (Box 4.4). Policy communities – close relationships between interest groups and public officials, based on the exchange of information for influence (see Box 9.1).

Agenda setting – the study of public, media, and government attention to policy issues. Policy monopoly– describes institutional arrangements (as in policy community) and a dominant way to understand a policy issue (Chapter 11): ‘Policy monopolies have two important characteristics. First, a definable institutional structure is responsible for policymaking, and that structure limits its access to the policy process. Second, a powerful supporting idea is associated with the institution’ (Baumgartner and Jones, 1993: 7). Venues – arenas in which to make authoritative decisions, such as government departments or bureaucracies, US congressional committees, and the courts. Some discussions seem to equate a ‘venue’ with an ‘institution’, so compare with Chapter 5. Venue shop – the attempt to seek favourable audiences in other venues. Policy image – how an issue is understood and discussed as a policy problem. It is now more common to describe ‘problem definition’. Parallel processing – when many issues are considered at one time by component parts of a larger organization. Serial processing – when issues are considered one (or a few) at a time. Policy issue – a focus of discussion, debate, or conflict in politics. Top of the policy agenda – treated as the most important (or most immediate) policy problem to be addressed. Media – forms of mass communication which foster collective attention (such as TV, radio, print, and online news, and social media such as Facebook and Twitter). Policy problem – a policy issue to be solved. Valence issue – an issue in which there is only one legitimate position and public opinion is uniform. Non-linear dynamics – the same amount of energy can have a minimal or maximal effect when amplified or dampened by a system. Negative feedback – a mechanism to dampen a signal. For example, actors pay no attention to information, or resist forces for change (‘like a thermostat maintains constant temperature’ - Baumgartner et al, 2018: 61).

Positive feedback – a mechanism to amplify a signal. For example, actors pay high attention to information, or accentuate forces for change (‘feeding frenzy, cascade, tipping point, momentum, or bandwagon effect’ – 2018: 61).

10 The Advocacy Coalition Framework Key themes of this chapter: • • • • • •

The advocacy coalition framework (ACF) suggests that people engage in politics to turn their beliefs into policy. They form coalitions with people who share their beliefs, and compete with coalitions of people with different beliefs. Beliefs provide the glue to hold actors together, cooperate, and learn how to respond to new information and events (and often to compete with other coalitions). The action takes place in subsystems, surrounded by a wider political system. Minor policy change can come from routine ‘policy-oriented learning’, while major change often results from the ‘shocks’ undermining some coalitions and exploited by others. The output of the ACF has changed over time to reflect learning by its core team and the widening of ACF scholarship to new policy areas, political systems, and methods.

INTRODUCTION: COALITIONS, POLICY-ORIENTED LEARNING, AND POLICY CHANGE The advocacy coalition framework (ACF) tells a simple story of policy action within a complex policymaking system. People engage in politics to turn their beliefs into policy. As policy actors, they form advocacy coalitions with actors who share their beliefs, and often compete with other coalitions. The action takes place within a subsystem devoted to a specific policy issue, and a wider policymaking environment that can influence the subsystem dynamic and provide constraints and opportunities to policy actors. Things get more complicated when we consider the scale of policy activity contained within such analysis. The policy process contains multiple

actors and levels of government. It displays a mixture of intensely politicized conflicts coupled with technical and routine policy-oriented learning. There is much uncertainty and debate about the nature and severity of policy problems. The full effects of policy may be unclear for at least a decade (Sabatier, 2007a: 3–4). According to the ACF, the best way to understand this complex world is to focus on how policy actors themselves simplify it enough to act individually and as part of coalitions. They identify and promote their beliefs: ‘policy participants hold strong beliefs and are motivated to transfer those beliefs into actual policy … before their opponents can do the same’ (Sabatier and Weible, 2007: 192, 196). Beliefs about how to interpret the cause and solution of policy problems, and the role of government in solving them, act as a glue to bind actors together within coalitions. In turn, different coalitions compete with each other to secure policy outcomes consistent with their beliefs. If the policy issue is rather technical or humdrum, there may be room for routine cooperation aided by policy brokers or experts (Weible et al., 2010; Ingold and Gschwend, 2014). If the issue is salient, and the politics is highly charged, people in each coalition may romanticize their own cause and demonize their opponents (Sabatier et al., 1987). The outcome can be long-term policymaking stability and policy continuity because the core beliefs of coalitions are unlikely to shift, and one coalition may dominate the subsystem for long periods. There are several possible sources of changes to policy and policymaking. First, coalitions engage in policy-oriented learning to remain competitive and adapt continuously to new information about policy and their policymaking environment. This process often produces minor change because coalitions learn on their own terms. Learning can describe a change in thought or action after taking on board new scientific or technical information. However, it is also a political activity, driven by actors’ beliefs. People learn how to retain their coalition’s strategic advantage and select or interpret the information they hold to be most relevant (Chapter 12). Second, ‘shocks’ affect the positions of coalitions within subsystems. Shocks are the combination of events and the response by coalitions to events. An external shock relates to events external to the subsystem, such as the election of a new government with new

ideas, or the effect of socio-economic change. An internal shock relates to the failure of current policy to solve the policy problem. Events may prompt major change, such as when members of a dominant coalition question their beliefs in the light of new evidence. Or, another coalition may adapt more readily to its new policy environment and exploit events to improve its position within the subsystem. While these external factors – such as global recession or environmental crises – are clearly important, coalitions influence how key actors understand, interpret, and respond to them. Third, there is often some room for negotiated agreements between coalitions. Further, each process may involve working with a subsystem’s broker (such as a sovereign). The ACF provides distinctive analysis of several aspects of policy processes. It recognizes the importance of potential dominance within subsystems (Chapter 9), but challenges the idea that a small number of groups and government actors control entire policy processes. It identifies a wider variety of players, spread across levels and types of government. It suggests that we analyse change over a full ‘policy cycle’, but rejects the artificial distinctions between stages or clearly separated phases of policy development (Chapter 2; Sabatier, 1986; Weible et al., 2009). It highlights the importance of socioeconomic factors and events, but rejects the ‘black box’ approach (Chapter 6, p.97) in favour of a focus on how actors mediate change. It notes that people exercise power to get what they want, but beliefs and persuasion also drive policy debate (Sabatier, 1988: 158, drawing on Majone, 1980; compare with Radaelli, 1999: 664). Perhaps the most distinctive aspect of the ACF is the commitment to conceptual revision and empirical application (Wellstead, 2017). Many ACF publications have been cited more than 1,000 times each, and two reviews identified more than 240 applications since 1987 (Weible et al., 2009; Pierce et al., 2017). The output of the ACF has changed over time to reflect learning by its core team – including a process of clarification by Sabatier and Weible (2007) – and the widening of ACF scholarship to new policy areas, national and subnational contexts, political systems, and methods.

A PICTURE OF THE ADVOCACY COALITION FRAMEWORK (ACF) As Figure 10.1 suggests, the main focus of the ACF is on a policy subsystem within a wider political system. It shares with the policy networks literature a focus on the policy-making relationships that develop when the reach of government has extended to most aspects of society (Chapter 9). Governments deal with the potential for ‘overload’ by breaking policy down into more manageable sectors and subsectors. The system effectively produces ‘enormous pressures for specialization’ and it is unusual for groups to have enough relevant knowledge or resources to compete in more than one or two sectors (Sabatier, 1993: 23; Sabatier, 1988: 138; Sabatier and Weible, 2007: 192; compare with Dahl in Chapter 3). Much of the early post-war literature suggests that this pressure produced specialized and exclusive ‘policy communities’ such as ‘iron triangles’ (Jordan, 1981; Richardson and Jordan, 1979; Weible and Sabatier, 2005) (see Box 9.1 on terminology). However, the ACF follows Heclo’s (1978: 94–97) departure from a focus on the simple ‘clubby days of Washington politics’ towards ‘complex relationships’ among a huge, politically active population. Issues which were once ‘quietly managed by a small group of insiders’ have now become ‘controversial and politicized’ (1978: 105; McCool, 1998: 554–55). In the ACF, the wider politicization of issues is more common than Chapter 9 suggests, partly because advocacy coalitions often try to activate ‘latent constituencies’ rather than seek to create policy monopolies (Sabatier, 1993: 24). The ACF highlights a role for many actors: Our conception of policy subsystems should be broadened from traditional notions of iron triangles limited to administrative agencies, legislative committees, and interest groups at a single level of government to include actors at various levels of government, as well as journalists, researchers and policy analysts who play important roles in the generation, dissemination, and evaluation of policy ideas. (Jenkins-Smith and Sabatier, 1993a:

179; see also Sabatier, 1993: 24–25; Sabatier and Weible, 2007: 192)

Figure 10.1 The advocacy coalition framework flow diagram Source: Weible et al. (2016: 6). Note: The flow diagram is a simple guide, not a comprehensive model of actual political systems. This version no longer includes policy brokers (they are not always present) and it is now clear that ‘external events’ are external to the subsystem (Jenkins-Smith et al., 2018: 143–44, 161). There may be more or less than two coalitions in a subsystem, and the insulation of subsystems from each other is a simplification (Sabatier and Weible, 2007: 193; Weible et al., 2009: 134).

A key focus of the ACF is how coalitions emerge, the form they take, and the roles of members. For example, this sense of relatively wide participation does not suggest that actors necessarily form trivial connections or ‘coalitions of convenience’. Instead, they may cooperate to pursue their beliefs and ward off the threat from other coalitions (Weible and Ingold, 2018). ‘Principal coalition actors’

represent a constant presence, ‘auxiliary’ actors share their beliefs but are involved sporadically during each ‘episode of conflict’, and ‘entrepreneurs’ champion ‘a particular policy idea’ (2018: 332). These roles contribute to variations in activity: in coalitions with ‘strong coordination’, actors share resources and produce a common plan; while ‘weak coordination’ involves a less codified common goal and actors making sure their behaviour complements the actions of their allies (2018: 334). Other actors supplement such activity: the brokers, ‘whose primary goal is consensus and the mitigation of conflict, and general citizens, who are ‘interested in and/or affected by subsystem affairs’ without participating regularly (2018: 332). As such, advocacy coalitions are more stable and enduring than social movements, and more focused on single policy issues than political parties (2018: 328). They are closer to interest groups, but with a less defined organizational and leadership function, to the extent that phrases like ‘coalition behaviour’ are ‘used metaphorically in reference to the individuals comprising the coalition’ (Jenkins-Smith et al., 2018: 140). Policy subsystems exist within a wider political system that sets the context for action and often provides coalitions with constraints or opportunities (Jenkins-Smith et al., 2018: 144): •

•

•

•

Relatively stable parameters are the factors that we would least expect to fluctuate over a decade or so, such as the ‘social values’, ‘constitutional structure’, and ‘distribution of natural resources’. These stable parameters influence the system’s long-term coalition opportunity structures, which relate to the ‘societal cleavages’, ‘openness of the political system’ and ‘degree of consensus needed for major policy change’ in each system. Dynamic external events are more likely to fluctuate and have an impact on subsystems, including socioeconomic or technological change, a change in government, or the impact from decisions made in other subsystems. These events and opportunity structures help facilitate the ‘shortterm opportunities’ that coalitions may exploit to reinforce their position or promote policy change.

ACF analysis often takes place over ‘a decade or more’ for two main reasons. First, it helps capture the ‘enlightenment function’ that actors contribute through policy research. A key aim is to ‘integrate the role of such technical analysis into the overall process of policy change’ (Jenkins-Smith and Sabatier, 1993b: 41). Second, it highlights the role – within subsystems – of actors at all levels of government. For example, the inclusion of actors from multiple levels of US government reflects the frequency of their interactions and strength of their relationships. State and local governments implement federal programmes, and ‘intergovernmental transfers constitute a significant percentage of most state and local government budgets’. State and local government representatives have a strong presence within the policy networks surrounding the legislature and executive agencies, and interest groups lobby ‘a variety of agencies at different levels of government … A coalition doing poorly in Washington is not helpless but instead can focus its efforts at sub-national levels where it is more powerful’ (Sabatier and Jenkins-Smith, 1993: 215–16; 230; Sabatier, 1993: 36; compare with venue shopping in Chapter 9). One consequence of this approach is that the boundaries between coalition members and actors such as policy brokers and sovereigns are fluid (Sabatier and Jenkins-Smith, 1993: 227). ‘Governmental authorities’ may be formally responsible for policy decisions, but policymakers and influencers interact closely and may be members of advocacy coalitions (compare with the blurry boundaries between formal authority and informal influence in Chapter 8). Similarly, brokers may have a ‘policy bent’, while some members of coalitions seek compromise and subsystem stability. The ‘distinction between “advocate” and “broker” … rests on a continuum’ (Sabatier, 1993: 27) and may vary in different case studies.

THE ROLE OF BELIEFS TO ADDRESS BOUNDED RATIONALITY Chapter 4 shows that people address bounded rationality by combining cognition and emotion to limit their attention and make choices. This idea is a key feature of ACF analysis. Individuals use beliefs to decide to which information to pay attention, and how to

interpret its meaning and significance: ‘actors always perceive the world through a lens consisting of their pre-existing beliefs’ (Sabatier, 1998: 109). The ACF assumes that actors are ‘instrumentally rational – that is they seek to use information and other resources to achieve their goals’ but rejects simple rational choice assumptions of perfect cognition and information that allow individuals to rank their preferences (Chapter 7). Instead, preferences ‘should be ascertained empirically’ because people use different ‘cognitive biases and constraints’ to determine how they perceive their world and process information (1998: 109). The ACF suggests that actors act according to their enduring beliefs rather than ‘short-term self-interest’ (Sabatier, 1993: 27). Beliefs help explain, in a parsimonious way, why a wide range of different actors exhibit ‘coordinated activity over time’ (Sabatier, 1993: 25). They form coalitions and stick together because they have similar ‘belief systems’, which represent a complex mix of theories about how the world works, how it should work and what we should do to bring the former closer to the latter (compare with world views and ideologies in Chapter 11). Beliefs also relate specifically to policy action, influencing which problems coalition members feel, ‘should receive the highest priority, the causal factors that need to be examined most closely, and the governmental institutions most likely to be favourably disposed to the coalition’s point of view’ (JenkinsSmith and Sabatier, 1993b: 41). Broadly speaking, there are three main types: •

•

Deep core beliefs regard an actor’s ‘underlying personal philosophy’ about nature and human nature (Sabatier, 1993: 30). Examples include whether people are evil or socially redeemable; how we should rank values such as freedom and security; whose welfare should count the most; and the left-wing/right-wing (or, in the USA, liberal/ conservative) divide (Sabatier, 1998: 103). Policy core beliefs regard ‘fundamental policy positions’ that relate to a specific subsystem issue. They are some combination of ‘empirical’ (what is happening?) and ‘normative’ (what should happen?) (Jenkins-Smith et al., 2018: 140). Examples include what causes this policy problem; whether or not society can, and should attempt to, solve it; how a government should solve it

•

(‘policy core policy preferences’); the proper balance between the government and market when solving this problem; and the proper distribution of power across levels of government (Sabatier, 1993: 31; 1998: 110; 2018: 141). Secondary aspects relate to the funding, delivery, and implementation of policy goals and the information gathered to support the process (Sabatier, 1993: 31).

As described, there is a hierarchy of beliefs according to their scope, how strongly people adhere to them, how they influence learning, and how susceptible they are to change in the light of new experiences and events. Core beliefs span almost all policy issues and are the least susceptible to change. A short-term shift would be ‘akin to a religious conversion’ because they form part of a person’s identity, as ‘largely normative issues inculturated in childhood and largely impervious to empirical evidence’ (Sabatier, 1993: 31, 36). Policy core beliefs relate to particular subsystems. They are more susceptible to change, but short-term change is also unlikely. The ‘enlightenment function’ may take place over decades, because many beliefs are ‘primarily normative – and thus largely beyond direct empirical challenge’, and ‘powerful ego-defense, peer-group, and organizational forces create considerable resistance to change, even in the face of countervailing empirical evidence or internal inconsistencies’ (Sabatier, 1993: 44, 33; compare with Weiss, 1979). In some cases, coalition members may revisit their views on the ‘overall seriousness of the problem’ when compared to other issues, or regarding the most appropriate role of government (Sabatier and Jenkins-Smith, 1993: 221). For example, environmental coalitions may shift their beliefs about the best way to achieve their policy goals, from a top-down command-and-control approach towards education or the provision of economic incentives. However, in most cases, change refers to ‘secondary aspects’, in which actors modify their beliefs to refine the routine delivery of specific policies in line with new information (Sabatier and Jenkins-Smith, 1993: 31, 221). This process varies somewhat, to reflect the motivation of actors to engage in politics. For example, more self-interested ‘material groups’ representing business or trade associations may be more likely to adjust their beliefs than ‘purposive’ groups because they have a

weaker emotional connection to specific causes (Sabatier and Jenkins-Smith, 1993: 224–25; Cairney, 2007a: 58).

THE ROLE OF BELIEFS TO BOOST COOPERATION AND HELP ACTORS EXERCISE POWER A common set of beliefs does not necessarily produce collective action. As Chapter 7 suggests, individuals with the same beliefs, aims or interests may not cooperate well if they feel that they can benefit from the outcome of collective action without making a direct contribution, or if they favour their own short-term return over the long-term returns to a group. Indeed, ACF case studies demonstrate that coalition members ‘express similar policy beliefs’ without necessarily engaging ‘in collective action to realize policy goals’ (Schlager, 1995: 248). There are ‘transactions costs’ involved in collective action. Members of different organizations, with different rules and incentives, may find it difficult to trust each other enough to work together even if they share the same beliefs, particularly if they are competing with each other for income (such as donations, grants, or profits) or to claim credit for policy successes (Sabatier, 1998: 115– 16). Some actors might be all-in, while others merely share information when required (Jenkins-Smith et al., 2018: 150). Therefore, what makes the difference between shared beliefs and actual coordination to do something about them? Partly following the Institutional Analysis and Development framework (Chapter 7), the ACF identifies trust development as a key source of collective action (Schlager, 1995: 248; Zafonte and Sabatier, 2004: 78; Leach and Sabatier, 2005). The study of ‘nascent’ subsystems suggests that trust can develop in three main ways (Ingold et al., 2017; see Cairney and Wellstead, 2019 on the often-unclear meaning of trust). First, shared beliefs help reduce some transaction costs when people begin to work together. There are low costs to ‘weak participation’, which primarily involves being aware of the strategies of one’s allies and pursuing ‘complementary strategies’ rather than something akin to monitoring and enforcing a contract (Sabatier,

1998: 110; Princen, 2007: 21; Sabatier and Weible, 2007: 197; Weible et al., 2009: 122). Actors could use a reference to beliefs as a short cut to decide how much they can trust them, to begin to share information, or experiment with collective action (Ingold et al., 2017: 447). Or, their beliefs help them decide who not to trust. When issues are politically salient or actors have a strong emotional attachment to a potential outcome, they romanticise their own cause and demonize their opponents. The ‘devil shift’ describes coalition action built on distrust, anxiety, and fear of the power of their competitors, which may heighten conflict (Box 10.1). If so, people exaggerate the benefits of participation or do not feel that free-riding is possible when the stakes are so high (Sabatier et al., 1987). Second, the increased frequency of fruitful coalition interaction could foster greater trust. If so, it could reduce the costs they bear when sharing information and seeking agreements that each group perceives to be fair (Schlager; 1995: 262; Sabatier, 1998: 116; Sabatier and Weible, 2007: 195; Ingold et al., 2017: 449). Actors encourage further cooperation by co-creating and enforcing rules. Sabatier and Weible (2007: 194) note the normative institutionalist suggestion that ‘Rules are followed because they are seen as natural, rightful, expected and legitimate’ (March and Olsen, 2006b: 689). Such rules may emerge as ‘the product of strategies by advocacy coalitions over time’ (Sabatier, 1993: 37). Yet, as Chapter 5 (p.85) suggests, there are many such ‘rules of appropriateness’, and ‘individuals must pick and choose among influences and interpret the meaning of their institutional commitments’ (Peters, 2005: 26; March and Olsen, 2006a: 9). Therefore, a combination of ACF and Institutional Analysis and Development (IAD) insights may explain the most widely understood and applied rules that are particularly important: actors may use shared beliefs as the initial basis of weak cooperation, and experience the feedback – such as quick wins – necessary to see the value of further investment in specific rules. Third, when people know what they want, but are uncertain what to do, they often follow people with authority, such as leaders in formal positions of power or experts with scientific or policymaking experience (Weible et al., 2010; Ingold and Gschwend, 2014; Ingold

et al., 2017: 448). While the ACF’s emphasis on beliefs and policy learning gives prominence to the role of ideas, power and authority clearly matter. The ability of advocacy coalitions to push policy in their favoured direction is dependent on the resources of their members (Sabatier, 1993: 29). Coalition power depends on factors such as their: • • • • • • •

status, for example when ‘in positions of legal authority’ ability to ‘garner public support’ grasp of ‘information regarding the problem severity and causes and the costs and benefits of policy alternatives’ ability to spin their arguments convincingly (without losing respect from policymakers) ‘mobilizable troops’, or members willing to initiate public demonstrations and fundraisers money to fund research, ‘bankroll sympathetic candidates’ or otherwise gain policy access and disseminate their message persuasive ability, to present an ‘attractive vision for a coalition’ and exploit the opportunities for policy change prompted by external events (Sabatier and Weible, 2007: 201–3).

Thus, beliefs and resources reinforce each other. Members of an advocacy coalition seek as many allies as possible. Beliefs act as the glue to bind participants, while the coming together of a range of allies creates ‘pressures for common positions, which tend to harden over time’ (Sabatier, 1993: 26). Further, actors use legal authority and institutions to establish or reinforce the centrality of particular beliefs within subsystems (alongside a wider range of strategies – Sabatier and Jenkins-Smith, 1993: 227). The establishment of authority and rules often produces stability and tends to reinforce the balance (or asymmetry) of power between coalitions that is mediated by policy brokers. In turn, stability often leads to policy continuity, with a battle of ideas replaced by a focus on technical issues and secondary aspects. Box 10.1 Policy Conflict Framework (PCF)

The Policy Conflict Framework (PCF) draws on the ACF and the IAD (Chapter 7) to study conflict (Weible and Heikkila, 2017). It identifies a spectrum of conflict, from low to high intensity, in relation to three main factors: •

•

•

Policy setting. It describes the level of action (conflict takes place in the political system, a subsystem, or a specific ‘policy action situation’), policy actors (including their beliefs and networks), policy issue (such as its complexity and relationship to core beliefs), and the nature of events prompting action (such as how urgent and solvable they seem to be) (2017: 25– 27). Episodes of policy conflict. It describes the conflict’s ‘cognitive characteristics’, or how much: their policy positions diverge, they perceive threats from their competitors, and they are willing to compromise. It also describes ‘behavioural characteristics’, or the ‘political strategies’ they use, from direct lobbying to forming coalitions and telling stories (2017: 28). ‘Policy positions’ resemble the ‘policy core policy preferences’ described in the ACF, while a focus on threat connects to the idea of ‘devil shift’ (2017: 28–29). Feedback effects. They include changes to policy and institutions, and the effects of change on the conflict setting.

In other words, the PCF examines particular policy settings to understand the degree to which actors: agree on policy, believe that the policy positions of others will harm their cause, are willing and able to engage in politics to influence policy, and are willing to change their policy position to help produce consensus. It investigates in detail why actors might compromise without changing their policy core beliefs (2019: 29). For example, they might compromise more if they are part of a group with diverse views and experiences, are regularly in touch with many policy actors (they are not in their own bubble), they do not feel threatened by their competitors, or they see the risk of bad outcomes from losing a policy debate (or not getting it all their own way) as low.

Heikkila and Weible’s (2017) case study of oil and gas policy conflict in Colorado found that these factors were generally key to compromise, but also found some unexpected effects – including a high perception of risk reducing unwillingness to compromise – that seem to relate to this policy setting.

THE DYNAMICS OF ‘POLICY-ORIENTED LEARNING’ Chapter 12 suggests that ‘policy learning’ is a very useful but rather vague term to describe acquiring new knowledge and skills. As such, it could describe the use of new scientific information to reduce uncertainty, or new political information to gain advantage. Crucially, it is driven by the ways in which people combine cognition and emotion to address bounded rationality (Chapter 4). The ACF provides key insights to this process by describing individual and collective learning processes (see Box 12.1, p.201). Sabatier (1998: 104) begins by defining policy-oriented learning as: ‘relatively enduring alterations of thought or behavioral intentions which result from experience and/or new information and which are concerned with the attainment or revision of policy objectives’. Then, the ACF shows how coalition actors learn by interpreting such new information through the lens of their beliefs. In particular, relatively fixed policy core beliefs contribute to ‘selective perception and partisan analysis’ (Sabatier, 1998: 104) to influence the focus of learning and information use. In comparison, secondary aspects are less restrictive and modified more easily in light of experience. Learning within coalitions is based on three key processes: (1) individuals within coalitions use new information to inform and adjust their beliefs; (2) members of coalitions interact and influence each other’s views; and (3) coalitions experience a turnover of members (Jenkins-Smith and Sabatier, 1993b: 41). A typical process of change, over many years, involves the dissemination of new ideas within coalitions by new members or existing members using new evidence. The diffusion of, or resistance to, new ideas depends on factors such as the ‘rate of turnover, the compatibility of the

information with existing beliefs, the persuasiveness of the evidence, and the political pressures for change’. In most cases, learning follows the routine monitoring of policy implementation, as members consider how policy contributes to positive or unintended outcomes and whether their beliefs regarding the best way to solve the policy problem are challenged or supported by the evidence. A less typical experience regards a form of learning so profound that it prompts internal division or some actors to change their position enough to warrant leaving the coalition (see the discussion of ‘shocks’ below). Learning across coalitions resembles a form of adaptation by the members of coalition A to the beliefs of coalition B, particularly when B’s views become ‘too important to ignore’ (1993b: 43). This is a political process, to make sure that policy reflects the coalition’s beliefs and interpretation of facts, ‘not a disinterested search for “truth”’ (1993b: 45; Sabatier, 1988: 151). Such learning is not straightforward for three main reasons. First, the evidence is unclear and contested. For example, information on the success of policy is difficult to obtain and subject to framing by each coalition. In some cases, there are commonly accepted ways to measure policy performance. In others, it is a battle of ideas. Technical information is often used ‘primarily in an “advocacy” fashion’ when the exercise of power is not enough to secure a coalition’s policy. For example, a manufacturer accused of excessive pollution may ‘challenge the validity of the data concerning the seriousness of the problem’, or produce data on the costs of regulation, rather than merely lobby to minimize regulation (1993b: 45; Sabatier, 1988: 152). Second, competing coalitions may refuse to engage on each other’s terms, particularly when the ‘level of conflict’ is high and each coalition’s claims threaten the core beliefs of the other. Data may be used as ‘ammunition’ rather than something to exchange and learn from (unless the debate can be restricted to professional scientific arenas – 1993b: 49, 54). Third, dominant coalitions are unlikely to simply accept the arguments of other coalitions and adopt their preferred policies. Rather, they seek to incorporate those points in a way that acknowledges the flaws of their own approach but maintains, as far as possible, the link between their core beliefs and policy measures. For example, a coalition may favour amending a

piece of legislation instead of replacing it. Therefore, non-incremental change may require a ‘shock’ that undermines the position of that coalition within the subsystem (1993b: 43–44). Otherwise, a coalition can successfully challenge the data and the technical basis for policy change and delay the process of policy innovation for years (Sabatier, 1998: 104). Jenkins-Smith et al. (2018: 151) suggest that, although learning is central to ACF analysis, its effect is difficult to identify and measure, and its causes remain unclear. Key facilitators for cross-coalition learning could include (1) a supportive and prestigious forum, in which discussions are open and actors share the same training or commitment; (2) ‘intermediate levels of conflict’ in which people feel threatened enough to engage but not so much that they are closed off to cooperation; (3) a policy problem that does not seem intractable or subject to too much debate on its cause; and (4) policy brokers, and coalition actors with ‘moderate beliefs’, willing to listen and encourage compromise (2018: 152; Ingold and Varone, 2012; Ingold and Gschwend, 2014; Heikkila et al., 2014; Moyson, 2017).

POLICYMAKING STABILITY AND INSTABILITY, POLICY CONTINUITY AND CHANGE These subsystem dynamics help explain periods of policymaking stability and policy continuity. Well-established subsystems contain a small number of coalitions exhibiting relatively fixed beliefs and fundamental policy positions over one cycle of a ‘decade or so’ (Sabatier, 1993: 26; Sabatier and Weible, 2007: 193). Stability within subsystems is reinforced externally by ‘relatively stable parameters’ that are unlikely to change over one cycle. These parameters include the rules and constitutional structures of a political system, the fundamental social structures and values of the wider polity, and the ‘basic attributes of a problem area’, such as the properties of natural resources relevant to many environmental policies (Sabatier, 2007b: 193; Weible et al., 2009: 135). It is reinforced internally by coalition actors. Although they engage in learning, it tends to affect the secondary aspects of policy rather than policy core beliefs, combined with a tendency towards brokered compromise between established

advocacy coalitions. The main effect is minor policy change and routine learning, either by a dominant coalition adapting to policy evaluation and new information, or debates between competing coalitions that alter secondary aspects rather than shifting the balance of power. Sources of instability come largely from ‘shocks’, which are the combination of events and the response by coalitions to events. An external shock relates to events external to the subsystem. Examples include the election of a new government with different policy ideas, a sudden change in public opinion (Jones and Jenkins-Smith, 2009) that affects the positions of competing coalitions (such as a surge of environmental concern), a sudden change in socioeconomic conditions (such as a global recession), and the effect of policies made in other subsystems. An internal shock relates to the failure of current policy to solve the policy problem. These events and perceptions of failure can help some coalitions and constrain others, and less powerful coalitions are unlikely to challenge a dominant coalition without them (Sabatier, 1993: 35). They ‘can foster change in a subsystem by shifting and augmenting resources, tipping the power of coalitions, and changing beliefs’ (Weible et al., 2009: 124). However, the ACF emphasis is on mediation by the actors and coalitions with the resources and strategies to respond relatively well to events. An event on its own is not a ‘shock’. The ACF highlights two types of process. One is akin to a crisis of confidence in which actors reconsider their policy core beliefs or question their motive to remain in a coalition. An event or new evidence prompts members to revisit their beliefs, perhaps concluding that existing policies have failed profoundly. They may then leave the coalition or depart to another, which could undermine the coalition’s position in a subsystem. This departure is more likely among the ‘auxiliary coalition actors’, who engage sporadically, than the ‘principal coalition actors’ who engage more routinely and have a more central role (Weible and Ingold, 2018). A second process refers to the ability of another coalition to use an event or new resources (such as new evidence on the problem) to reinforce its position within the subsystem. The shock relates partly to one coalition’s ability to knock a formerly dominant coalition off its perch, for example by showing

that its approach is best equipped to interpret and solve the policy problem (compare with Hall, 1993 in Chapter 11). Policy change may be major following such shocks, because they have an impact on the policy core aspects of coalition beliefs: one coalition changes, or another coalition takes its place and pushes for major change (Sabatier, 1998: 119). Other sources of potential change – or factors which prompt coalitions to redouble their efforts to resist it – include enduring and ‘chronic threats’ from a coalition’s rivals, short-term ‘spillovers’ from other subsystems which force coalitions to respond, and longer term ‘changes in social values’ (Weible and Ingold, 2018: Table 5). In each case, coalitions compete to improve their positions within subsystems using resources such as the ability to gather and interpret information, mobilize public support, secure funding for campaigns, and show skilful leadership (Sabatier, 1993: 29; Sabatier and Weible, 2007: 201–203; Cairney, 2015b).

THE ACF GOES INTERNATIONAL: NEW EMPIRICAL APPLICATIONS The ACF was conceived in the USA and focused primarily on US natural and environmental issues. However, it is ‘intended to apply to all policy areas’ in most developed countries (Jenkins-Smith and Sabatier, 1993: 225), and newer studies address: One of the most frequent criticisms of the ACF … that it is too much of a product of its empirical origins in American pluralism. It makes largely tacit assumptions about well-organized interest groups, mission-oriented agencies, weak political parties, multiple decision making venues, and the need for supermajorities to enact and implement major policy change. (Sabatier and Weible, 2007: 199) On these terms, the ACF expansion has been successful. It is a useful framework for the comparative analysis of policymaking across many types of policy issue, national and sub-national governments, and countries, and by using many different methods, and there have been over 240 published applications of the ACF since 1987.

Weible et al. (2009) found 80 ACF studies conducted from 1987 to 2006. Many are best described as ACF-inspired case studies because 55% ‘do not explicitly test any of the hypotheses’ (2009: 128). Further, 41% have unspecified methods and only 6% use the questionnaire approach accentuated in early studies (Jenkins-Smith and Sabatier, 1993c; Sabatier and Weible, 2007: 196). Most of these ACF case studies ‘remain within environmental and energy policies’, with relatively few studies exploring areas that primarily affect people rather than natural resources (Weible et al., 2009: 125). This bias towards environmental policy may, for example, exaggerate the role of the EU or international policy cooperation (Litfin, 2000). Pierce et al. (2017) find a rise in activity and diversity from 2007 to 16, albeit with the same sense that most applications do not apply the ACF systematically. They find 161 applications, including 138 authors and 122 different research organizations in 25 countries, spread over 98 journals (2017: S20). Environment/energy is the most studied area (43%), followed by fields including public health, education, ‘social welfare’, science, defence, foreign policy, economics, and urban policy (2012: S23). Studies cover 201 countries, with the USA the most studied country but Europe (the EU and single countries) the most studied continent, and evidence of rising cross-country comparisons (2017: S25). Most of the stated methods are mixed method and qualitative, and the proportion using interviews (63%) has doubled since the first review (2017: S26). The majority of studies identify the number of subsystem coalitions and draw on ACF concepts to explain policy change. However, Pierce et al. (2017: S35) find a lot of ACF-inspired studies which, for example, do not apply key insights (e.g. the need to study policymaking beyond the national government level), explain their methods, or examine key dynamics such as policy-oriented learning. This limitation has prompted efforts to coordinate comparative studies (Box 10.2). Box 10.2 The case of hydraulic fracturing in ten countries The ACF has inspired over 240 applications but, in many cases, it is difficult to know how they compare with each other. Key projects address this issue by coordinating analysis in special issues of

journals (Weible et al., 2011) or edited books. For example, Weible et al. (2016) describe an initial seven country – the USA, Canada, the UK, France, Germany, Sweden, Switzerland – comparison of policy regarding the hydraulic fracturing (‘fracking’) of shale rock to extract oil and gas. To some extent, the case selection reflects (1) a tendency of ACF scholars to specialize in policy studies in North America and Europe (and interact with each other frequently) and (2) a focus on places in which there are sufficient energy reserves to make it a realistic prospect that governments will allow fracking to take place. Comparative analysis is challenging, even with this level of coordination. They reflect country-level differences in: the general structures and dynamics of each political system, the level/type of government addressing this policy issue, the framing of policy debate, and the balance between perceived economic benefits and environmental costs (2016: 2–3). The ACF provides a common language and focus for such analysis – to describe subsystems, identify a number of coalitions, and measure policy change – but it is rare for projects of this scale to adopt one uniform research method. Consequently, the book’s conclusions are a mix of positive lessons and tentative conclusions (Ingold et al., 2016). For example, the study confirms that each country contains a subsystem with two advocacy coalitions, However, there is much variation according to levels of policy change (from policy to allow high activity to a ban on all activity), reflecting the positions of each coalition (some ‘pro’ coalitions are in favour of fracking, while others want to explore feasibility), general political system differences, such as the level of government at which the action takes place, and issue-specific factors relating to: • • •

institutions, such as legal ownership of mineral rights (from private in the USA to government ownership in the UK) basic attributes, such as the economic feasibility of energy extraction (game-changing in the USA, but low in Sweden and Switzerland) subsystem autonomy, such as when fracking emerges from oil/gas subsystems (Germany, France, Switzerland) or is a new

issue in relation to factors such as new technology or companies (USA and Canada) (2016: 240–47). The greater challenge – and opportunity – is to produce more global systematic comparisons (Box 13.4). Weible et al.’s (2018) addition of India, China, and Argentina produces a ten-country ACF comparison. This scale of analysis unearths new findings, and prompts scholars to seek explanations for phenomenon that may not occur to them in single case studies. For example, two subsystems do not contain two coalitions: China has one supportive coalition, and India has a ‘dominant supportive coalition and limited opposition’. This finding prompts us not to assume that we can predict the highly salient issues that produce subsystems with binary coalitions. Instead, we should identify and research the conditions under which this dynamic arise (such as when there are sufficient oppositional political resources, low constraints on mobilization imposed by the political system, and a significantly high perceived threat to prompt actors to mobilize one way or the other) (2018: 15).

CONCEPTUAL REVISIONS AND NEW DIRECTIONS The ACF can apply to many policymaking contexts. This expansion has accentuated an already strong focus on conceptual revision following empirical application. The ACF’s original authors, Paul Sabatier and Hank Jenkins-Smith, inspired a research programme that resembles one of its coalitions, containing scholars that use and modify ACF insights via individual studies, edited volumes and journal special issues (see Jenkins-Smith et al., 2018; Weible et al., 2016). This approach is no accident. It partly reflects a ‘Lakatosian’ approach to science (Chalmers, 1999; Majone, 1980), in which ACF scholars engage in learning to refine the ACF’s secondary aspects while generally protecting its core argument and clarifying its analysis of coalitions, beliefs, learning, and policy change (Cairney, 2015b:

489–94; Wellstead, 2017: 550). Examples of key points of clarification by Sabatier and Jenkins-Smith (1993) include: 1. Administrative agencies are often coalition members, but they take ‘more centrist positions than their interest-group allies’ and could ‘switch sides’ after ‘major exogenous events’ such as elections (1993: 213–14). 2. Technical information can have ‘important impacts on policy’ by influencing policy brokers even when it does not change the views of the dominant coalition (1993: 219). 3. There is a difference, within policy core beliefs, between ‘normative precepts’ that are unlikely to change, and ‘precepts with a substantial empirical content’ that are more likely to change in the light of new evidence (1993: 220–21). Sabatier and Weible (2007) and Weible et al. (2009) took this agenda forward by encouraging revisions in light of new applications. First, for example, many studies suggest that external events are necessary but insufficient conditions for major policy change (Sabatier and Weible, 2007: 204–207). Events may prompt members to strengthen their resolve rather than drift apart. Or, a competing coalition may not have the resources to exploit a new opportunity. Therefore, potential shocks lead to major policy change in some areas but not others (Weible et al., 2009: 128). One rare exception may be when a ‘hierarchically superior jurisdiction’ overrides subsystem policy despite opposition from the dominant coalition (Sabatier and JenkinsSmith, 1993: 217), but even this is not the end of the story if there is substantial discretion to change policy during implementation (Sabatier, 1998: 119). Second, we need to distinguish between ‘nascent’ and ‘mature’ subsystems. In the latter, there are ‘specialized subunits within agencies at all relevant levels of government to deal with the topic’, while participants ‘regard themselves as a semi-autonomous community who share a domain of expertise’ and identify the issue as a ‘major policy topic’ (Sabatier, 1998: 110–14; Sabatier and Weible, 2007: 192). Only mature subsystems exhibit stability. Relationships are more fluid in nascent subsystems when issues are still emerging,

the problem is still being defined, and the costs of action and inaction are unclear (Radaelli, 1999: 665, 676; Ingold et al., 2017). Third, coalition membership ‘can be stable over time’ when there are ‘major controversies’ and ‘policy core beliefs are in dispute’, but ‘defection is also common’ following external events or when there is clear internal division on key aspects of policy core beliefs (Weible et al., 2009: 128–30, 135). Fourth, policy learning across coalitions is most likely to focus on secondary aspects (lower stakes issues with a strong empirical element), but some studies find learning at the policy core level. Finally, Sabatier and Weible (2007: 205–7) had highlighted the potential for major policy change to result from negotiated agreements between ‘previously warring coalitions’. Supportive factors include: a previous ‘stalemate’ that suits no-one; a negotiation process that builds trust, involves frequent meetings and minimal turnover of staff, gives ‘all relevant groups of stakeholders’ the power of veto and is chaired by a ‘respected “neutral”’; and a realization that there is no better alternative. The issue also has to be amenable to resolution though empirical research, rather than issues such as abortion that are too mired in normative beliefs to allow participants to engage with each other in a meaningful way. However, empirical applications suggest that professional fora do not guarantee success because the surrounding ‘high political conflict’ often limits their effectiveness (Weible et al., 2009: 130; compare with Koebele, 2019a, 2019b on the ACF and collaborative governance). The major expansion of ACF studies in Europe accelerated the desire of its authors to revise the ACF in light of empirical investigation. In particular, the ACF flow diagram’s reference to the political system’s long-term coalition opportunity structures is largely the response to insights from comparative studies. A focus on the ‘degree of consensus needed for major policy change’ reflects applications to Europe that highlighted the important of proportional electoral systems (Sabatier, 1998: 121). Drawing on Lijphart (1999), Sabatier and Weible (2007: 200) argue that the ‘norms of consensus and compromise’ vary markedly across political systems. In Westminster systems, a political majority can be commanded relatively easily and there is less incentive to seek broad support for

policy measures. This contrasts with consensus democracies with strong incentives to form coalitions. In turn, the incentives for coalitions to cooperate with each other and share information may be lower in Westminster systems and higher in consensus democracies (2007: 200). However, empirical applications suggest that these differences are often subtle (Cairney et al., 2018; Chapter 5). A focus on the ‘openness of the political system’ can relate to two developments. First, for Sabatier and Weible (2007: 200), openness refers to, ‘(1) the number of decision making venues that any major policy proposal must go through and (2) the accessibility of each venue’ (2007: 200). For example, the US tradition of multiple, open venues may contrast with ‘corporatist’ systems characterized by centralized decision making, restricted to a small number of leaders of business groups and unions. In the latter, we may find that coalitions have fewer active members and that more actors may take on the role of policy broker. Second, a focus on openness raises a more profound difference in applications to countries without free and fair elections, and/or systems that discourage or prohibit people from coming together as coalitions (although ACF studies of relevant countries have only just begun – Henry et al., 2014; Jenkins-Smith, 2018: 158–59). A range of further revisions connect strongly to new thinking about comparative analysis. First, for example, coalitions seeking stability in ‘separation-of-power systems’ may pursue legislative change because laws, once enacted, are difficult to change; in ‘Westminsterstyle’ systems, where laws are easier to change, coalitions may ‘rely upon a variety of more informal, and longer-lasting, arrangements’ (Sabatier, 1998: 103). Second, policy-oriented learning may be ‘more difficult in Britain than in many other countries because of the norms of secrecy so ingrained in the civil service’ (1998: 103). Third, a change in the ‘systemic governing coalition’ (Figure 10.1) may be difficult to define. In separation-of-powers systems, it refers to ‘the replacement of one coalition by another in both houses of the legislature and in the chief executive’ (1998: 120). In a two-party parliamentary system it may merely refer to the replacement of one party by another. In a multi-party system there may be changes only at the margins, perhaps with the largest party remaining in

government and forming a coalition with another party. Sabatier (1998: 120) suggests that a change may only be deemed to occur if there is a 60–70% turnover in the coalition. Not all non-US applications produce the need for such revision. For example, the argument that administrative agencies are often part of coalitions is based on the idea that civil servants are not neutral; they have a policy bent. This has academic support ‘even in European countries with a strong tradition of elitist – and supposedly neutral – civil servants’ (Sabatier and Jenkins-Smith, 1993: 214; Richardson and Jordan, 1983: 258; Cairney, 1997: 891). Further, the ACF already accommodates political system differences such as the intergovernmental relations that vary markedly according to federal, quasi-federal, and unitary systems (Chapter 5). However, more generally speaking, the ACF ethos is to seek new case studies and comparisons continuously, so as to revise the ACF in light of new information (such as new comparative analysis, Box 10.2). In some cases, these revisions are already apparent to a large extent, but perhaps accompanied by a sense of unfinished business (see Jenkins-Smith, 2018: 154–59 and Pierce et al., 2019 on new directions). For example, new work outlines in more detail how members of advocacy coalitions address collective action problems (Weible et al., 2011: 349), but there are few studies of the transition from nascent to mature subsystems containing well-established coalitions (Jenkins-Smith et al., 2018: 157). There is also some variation in the ways in which scholars seek to analyse entire subsystems, or key parts, or one territorial area or level-ofgovernment-specific subsystem ‘nested’ in a wider network (2018: 155; Cairney et al., 2018). The key question is whether to analyse one subsystem containing actors from all levels of government, or to identify a separate subsystem at each major territorial or governmental level. In practice, it is often treated as an empirical question supplemented by a degree of judgement regarding factors such as ‘the degree of (a) legal autonomy of each level and (b) actor integration among levels’ (Sabatier, 1998: 115; Sabatier and Weible, 2007: 193; Jones and Jenkins-Smith, 2009: 52).

CONCLUSION

When it first appeared, the ACF presented a novel discussion of at least four key aspects of the policy process. First, it went beyond the ‘iron triangle’ image of policy subsystems in which a small number of groups and government actors control the process by excluding most participants. It identifies a wider variety of players, including journalists and academics, to highlight their long-term ‘enlightenment function’ and the importance of technical information and knowledge, and actors across various levels of government, to reflect the need to study policy change over a full cycle. In this sense, the focus is less on competition and exclusion within insulated policy communities (as in Chapter 9) and more towards competition between different coalitions sharing different beliefs within a wider subsystem. Second, it rejected the analytical distinctions between stages in the policy cycle and sought to go beyond the implementation literature that became bogged down in an intractable debate between ‘top-down’ and ‘bottom-up’ approaches. Instead, it focuses on policy change ‘over a decade or more’ within subsystems that include actors involved at all stages. Third, it built on models that identify the importance of ‘external’ or ‘structural’ factors, by exploring how they affect the beliefs and strategies of members of advocacy coalitions. For example, the values of the population are framed and exploited better by some coalitions than others, while the nature of the constitution and the basic rules of political systems provide opportunities for coalitions to protect existing policy positions or seek audiences sympathetic to change. Further, while external events provide the impetus for major policy changes within subsystems, they do not do so inevitably. Rather, this also requires a major rethink by a dominant coalition suffering a crisis of confidence, or the skilful manipulation of the situation by another coalition. Last, but not least, it outlines the importance of beliefs. Much of the ACF’s value comes from its ability to theorize convincingly the relationship between power and ideas (we study this relationship in Chapter 11). The policy process is populated by a huge number of actors who try to influence events by engaging in both the technical and the highly politicized aspects of policymaking. Advocacy coalitions exercise power to compete for dominance, or the chance to translate

their beliefs into policy positions adopted by government authorities. There may be long periods of policy continuity, when one coalition dominates the subsystem, and less frequent periods of major change as its position is threatened by external events such as changes in socioeconomic conditions or government. We can link this process to at least three types of policy learning. First, the dominant coalition engages in learning that affects secondary aspects of its beliefs. In turn, it accepts or encourages minor policy changes that do not challenge its more fundamental beliefs. Second, a shock prompts it to engage in a more radical examination of its policy core beliefs (in other words, individual members revisit their own beliefs and encourage others to do the same) which may lead to its acceptance of major policy changes. Third, a shock provides a minority coalition with a window of opportunity to adapt to the new situation and persuade governmental authorities to revisit the nature, cause of, and solution to the policy problem. If successful, this leads to major policy change. Overall, we have a framework that helps explain both continuity and change, and subsystem stability and instability, across a full policy cycle of a ‘decade or so’. This approach has clearly found a prominent place in policy scholarship, producing at least 240 applications over 30 years. The empirical applications highlighted the need for conceptual revision to accommodate very different experiences in case studies, particularly when they do not focus on the USA or environmental, natural resource, or energy issues. However, they also showed that the basic insights from the ACF hold in many contexts: people go into politics to turn their beliefs into policy, they form coalitions with their allies, and they learn from new information about the policy problem and their experiences in competing with other coalitions. This process tends to produce many periods of subsystem stability and policy continuity and some periods of instability and change. As such, like many theories in this book, the ACF discusses elements that it would treat as (1) universally applicable, such as the use of beliefs to address bounded rationality, and (2) context-specific, such as the motive and opportunity of specific people to organize

collectively to translate their beliefs into policy (a theme to which we return in Chapter 13). Advocacy coalitions – ‘people from a variety of positions (elected and agency officials, interest group leaders, researchers) who share a particular belief system – i.e. a set of basic values, causal assumptions, and problem perceptions – and who show a nontrivial degree of coordinated activity over time’ (Sabatier, 1988: 139; compare with Hajer, 1995; Dodge, 2017 on ‘discourse coalitions’ which do not necessarily exhibit coordination). Policy subsystem – a ‘set of actors’ including advocacy coalitions, ‘who are involved in dealing with a policy problem’ (Sabatier, 1988: 138; McCool, 1998). The ACF’s description of subsystems suggest that they contain more actors and venues (in multiple levels of government) than in policy monopolies (Chapter 9). Policy brokers – the actors present within many (not all) subsystems that seek to minimize conflict and produce workable compromises between advocacy coalitions. Sovereign – the governmental authority or ultimate decision-maker in a subsystem.

11 Ideas and Multiple Streams Analysis Key themes of this chapter: • • • •

•

•

To explain power, we need to explain the role of ideas. Ideas are shared beliefs or ways of thinking, including knowledge, norms, world views, and ideology. Some discussions treat ideas as the main source of explanation for policy change while others treat them as resources used by actors during agenda setting. Policy theories focus on three elements: policy solutions (‘I have an idea’), the use of persuasion to connect solutions to the beliefs of audiences (agenda setting), and deeply held beliefs which represent the context for debate (‘paradigms’). Kingdon’s multiple streams analysis (MSA) combines these elements to identify a ‘window of opportunity’ for policy change: attention rises to a problem, a feasible solution is available, and policymakers have the motive to select it. ‘Policy entrepreneurs’ are the resourceful and well-connected actors adapting to their policymaking environment to try to exploit such windows of opportunity.

INTRODUCTION: THE ROLE OF IDEAS IN POLICYMAKING In Chapter 3, we found that no discussion of power is complete without a focus on the role of ideas. The exercise of power can involve framing issues to limit attention, manipulating someone’s way of thinking to change their behaviour, dismissing someone’s beliefs as based on inferior knowledge, and forming coalitions with actors whose beliefs you share. The quotations in Box 11.1 suggest that policy theories focus primarily on argumentation and persuasion, to tell stories that influence agendas. However, such stories may only

have impact when they tap into the deeply held beliefs of their audience and the stories they tell to themselves. For example, participants in many countries take certain reference points for granted when seeking to persuade: democracy, free speech, and capitalism are values expressed routinely in Western political systems; many scientists share the same methodological standards when producing knowledge; and members of the same political party may take for granted their ideology on the role of the state. In other cases, these ideas are a bone of contention. For example, there may be competition to establish ‘how the world works’ and what policy solutions are appropriate in these circumstances, or shared beliefs may change over time as people challenge them. Our aim is to conceptualize this relationship between power and ideas: where does one concept end and another begin? How can we use the analytically distinct concept of ideas to explain behaviour? Different approaches treat ideas as more or less important to overall explanation. For some, the role of ideas is the independent variable, or the source of explanation, which makes ‘history a contest of ideas rather than interests’ (John, 2003: 487; 1998: 149). The ‘power of ideas’ suggests that ‘ideas have a force of their own’ and that ‘the power of the idea itself explains its acceptance’ (Jacobsen, 1995: 285). Ideas are often associated with metaphors used to describe an irresistible force, as viruses infecting political systems or high tides sweeping obstacles aside. Or, we describe the norms and world views that restrict the ways in which individuals, groups, or institutions operate. They may even be treated as structures which constrain agents, particularly if we are talking about the big ideas – such as capitalism, socialism, or religions – which may be impervious to change for decades (Parsons, 1938: 659). Further, we can identify ‘paradigms’ as the ideas so fundamental that they are taken for granted (Hall, 1993). For others, the role of ideas is the dependent variable, or the phenomenon to be explained. Ideas are produced and promoted by many actors and accepted or rejected by policymakers. The focus is on the mediation of, and receptivity to, ideas. If there is an almost infinite number of ideas, then how and why do policymakers choose one at the expense of others (John, 1999; Cairney, 2009c)? If we

extend the metaphor of the virus, the immunity of the host may be the most important factor. Kingdon’s (1984; 1995) multiple streams analysis (MSA) (sometimes described as the multiple streams approach or framework) sums up this interaction between the supply of, and receptivity to, ideas designed to solve policy problems. The phrase ‘an idea whose time has come’ highlights its role as ‘an irresistible movement that sweeps over our politics and our society, pushing aside everything that might stand in its path’ (Kingdon, 1984: 1). Kingdon suggests that such explanations are incomplete since they ignore the conditions that have to be satisfied before a policy will change. A ‘window of opportunity’ must open when three separate ‘streams’ come together at the same time: attention rises to a policy problem that is framed in a certain way, a feasible solution to that problem is available, and the political conditions are conducive to action (Kingdon, 1984: 174). In other words, the promotion of a new idea (‘policy solution’) will not be successful unless policymakers have the motive and opportunity to adopt it. This window is fleeting, suggesting that ideas are powerful only when acceptable to their audience under certain conditions. Key actors – ‘policy entrepreneurs’ – can influence this process, but as ‘surfers waiting for the big wave’, not controllers of the sea (1984: 173). Box 11.1 Key quotations on ideas and interests To ‘take ideas seriously’ is to recognise the symbiotic relationship between power and the role of ideas, rather than explain policy primarily in terms of influence and material interest. (Kettell and Cairney, 2010: 301) We miss a great deal if we try to understand policy-making solely in terms of power, influence and bargaining, to the exclusion of debate and argument. Argumentation is the key process through which citizens and policymakers arrive at moral judgements and policy choices. (Majone, 1989: 2)

Lobbyists marshal their arguments as well as their number … The content of the ideas themselves, far from being mere smokescreens or rationalizations, are integral parts of decision making in and around government … in our preoccupation with power and influence, political scientists sometimes neglect the importance of content. (Kingdon, 1984: 133; 131) Decision makers need ideas, or a consistent political message that will command popular support and motivate those who carry out policy. (Hall, 1993: 291–92) Raw political power may carry the day against superior evidence, but the costs to one’s credibility in a democratic society can be considerable. Moreover, resources expended – particularly in the form of favours called in – are not available for future use. Thus those who can most effectively marshal persuasive evidence, thereby conserving their political resources, are more likely to win in the long run than those who ignore technical arguments. (Jenkins-Smith and Sabatier, 1993b: 44–45)

Defining Ideas Ideas can be defined broadly as shared beliefs, thoughts, or ways of thinking. When we try to go beyond this broad definition, towards something more concrete, we find a range of approaches which treat ideas as normative beliefs, regarding what is right and wrong, or empirical beliefs regarding what factors cause others to change. They also describe a strong or weak role for ideas in explanation (Parsons, 1938: 653; Jacobsen, 1995: 291; Campbell, 2002; John, 2003: 487; 1998: 144). The literature is large (Reich, 1988: 4; Majone, 1989: 25; Haas, 1992; Fischer and Forester, 1993; Radaelli, 1995; Richardson, 2000; Blyth, 2002; Campbell, 2002; Goodin and Tilly, 2006; Hay, 2006a: 65; Rhodes, 2006a: 91; Cairney, 2009c; Béland, 2010; Béland and Cox, 2010), and it suggests that ideas can be:

• • • • • • • • • • • • • • •

visible and discussed routinely, or almost taken for granted ‘viruses’ which ‘mutate’, take on a life of their own, and infect political systems world views, ideologies, or paradigms which place limits on policy debate norms which determine broad attitudes to the regulation of behaviour akin to institutions, or rules governing thought and action new thoughts used to reframe policy problems scientific knowledge or expertise which influences policy beliefs shared by interest groups and civil servants within ‘policy communities’ beliefs shared in ‘epistemic communities’ (international networks of experts) core beliefs shared by advocacy coalitions that shape policy learning and action traditions, or sets of understandings received through socialization understandings of the world used to understand and articulate our interests forms of language used to dominate the policy process visions employed by political leaders to encourage people to act in the public interest rather than self-interest policy solutions which survive by adapting and evolving to political circumstances

From this long list, we can identify three main ways in which actors draw on ideas (Cairney and Weible, 2015: 86–87). First, an idea is a proposed solution to a policy problem (‘I have an idea’). Second, actors draw on persuasion and argumentation skills – framing, heresthetic, and storytelling – to persuade their audience that a policy issue is a serious problem that can be solved by their action (Chapters 4 and 9). Third, ideas are the understandings – described variously as core beliefs (Chapter 10), monopolies of understanding (Chapter 9), hegemons (Chapter 3), and paradigms (Hall, 1993) – so deeply accepted that they can be taken for granted, as part of everyday language, despite having a profound effect on the terms of

policy debate. When we combine all three categories, we can see the intersection between many dimensions of power and ideas: actors engage in visible forms of agenda setting to sell solutions, but in the context of wider and less visible competition to set the terms of debate and ‘institutionalize’ ideas (Chapters 4 and 5; Jobert and Muller, 1987; Genieys and Smyrl, 2008b).

IDEAS AS THE PRIMARY SOURCE OF EXPLANATION: VIRUSES AND NORMS Some authors employ metaphors to describe ideas as the main source of explanation, but without privileging ideas completely at the expense of power. For example, Richardson (2000: 1019) studies the breakdown of closed policy communities in Western Europe from the 1980s (Chapters 8 and 9). He uses the analogy of viruses to describe the devastating effect that ideas can have on the policy process: ‘Our resort to the virus analogy is meant to convey the importance of exogenously generated ideas as a shock to both existing institutional arrangements and the actors that benefit from them.’ The preferences of actors, such as the public, are changed by ideas and knowledge which reside in the political community in much the same way that ‘viruses present in the atmosphere we breathe’. Similarly, ‘exogenous changes in policy fashion, ideas or policy frames’ cause major changes to previously stable policy communities. If this new agenda conflicts with the old, then policy communities become unbalanced and break down. Extending the analogy to viruses such as SARS, policy viruses often spread through international travel, with stakeholders and policymakers increasingly able to meet their international counterparts and ‘bring new ideas and policy frames back home’ (Richardson, 2000: 1018). The effect of new knowledge can be profound, particularly when it spreads across a range of countries to produce major policy transfers (2000: 1020; compare with Chapter 12). However, as Box 11.2 suggests, any virus analogy must consider the role of the host, as more or less immune to infection. Actors within policy networks survive by ‘mutating’ the virus or themselves

(changing their behaviour to embrace or deal with a new frame of reference). For example, new ideas may originate within existing policy networks (2000: 1018). If not, they are still only a ‘potential threat’ and this potential is only realized if such new ways of thinking cannot be accommodated within existing frames of reference. Baumgartner and Jones (1993: 237) present a similar account of ‘powerful forces of change that sweep through the entire system’, linked to the actions of interest groups, policymakers, and the ‘national mood’. They portray the image of an almost uncontrollable process that is independent of any one actor. Since political leaders cannot ‘reverse the tides’, their most advantageous course of action is to harness the energy that these ideas create and use them to promote sympathetic policy proposals. Yet these cases are unusual. The consequence of intense levels of attention to one idea is that most ideas with the potential to destabilize existing policy arrangements are ignored (Chapter 9; Cairney, 2009c). A stronger treatment of ideas as independent variables can be found in discussions of norms as determinants of behaviour. Axelrod (1986: 1095) suggests that when we see many people or nations displaying ‘a great deal of coordinated behaviour that serves to regulate conflict’, we tend to attribute this behaviour to norms. Further, ‘an established norm can have tremendous power’ as a ‘mechanism for regulating conflict in groups’ (1986: 1095). As we saw in Chapters 3 and 5, norms may also be used to explain why people regulate each other’s behaviour or reproduce the rules of institutions. Yet there are two main qualifications to Axelrod’s argument. First, he is really describing several forms of observable behaviour: political actors choosing to adhere to norms rather than challenge them, for fear of sanctions; the use of sanctions to punish a departure from a norm; and ‘metanorms’, or the punishment of someone who doesn’t punish the departure from a norm (historical examples include White men punished for not lynching Black men, the communist culture of calling on society to denounce others, and US sanctions on foreign companies unwilling to boycott the Soviet Union – 1986: 1097). Second, norms are challenged frequently and, when successful, ‘the standing of a norm can change in a

surprisingly short time’ (1986: 1096). In other words, norms can be taken more or less seriously according to the levels of collective adherence to them, the likely scale of sanctions, and the motivation of individuals (how ‘bold’ or ‘vengeful’ they are). Axelrod (1986: 1087) takes an evolutionary game theory approach to norms, modelling likely behaviour according to trial-and-error strategies (see Box 7.2). His simulations suggest that some fairly traditional factors explain why people adhere to norms, including the exercise of power to enforce sanctions; the use of disproportionate sanctions to head off future challenges; socialization, to ensure that people are uncomfortable with the idea of violating norms; and the formalization of norms into laws. The common theme is dominance: ‘it is easier to get a norm started if it serves the interests of the powerful few’ (1986: 1108). Otherwise, it would be difficult to explain why certain norms take hold and why policymakers pay more or less attention to them over time, particularly since ‘world culture’ is ‘rife’ with contradictions and competing norms (Campbell, 2002: 25–26). Therefore, the general theme of this literature is that, although many accounts appear to cite ideas as the main source of explanation, they are supplemented by references to the exercise of power. The appearance of an empirically independent effect of ideas is largely the result of the limitations to the language we can use to express social and political relationships. If there is a symbiotic relationship between ideas and interests, or if the concepts of power and ideas are inextricably linked, then a degree of confusion will always arise when we attempt to separate them analytically to explore their effects independently of each other (Jacobsen, 1995: 309; Weible et al., 2009: 122; Hoberg, 1996: 137; Sabatier, 1988: 161 note 11). Box 11.2 Ideas and the case study of tobacco Tobacco control policy is a good candidate for the phrase ‘an idea whose time has come’. Post-war policy in most countries was controlled by a strong policy community of tobacco companies and policymakers in finance, trade, and industry departments. A policy monopoly was maintained by establishing a dominant

image of tobacco related to its economic benefits, which helped exclude doctors and public health groups with stronger links to health departments (Cairney et al., 2012; Cairney, 2007a; 2019e). From the 1990s, the virus-like spread of new ideas infected policy communities and contributed to their collapse, with the balance of power shifting to public health actors. New ideas related to the scientific evidence that established increasingly strong links between smoking and ill health. The policy image of tobacco as an economic and civil liberties issue was replaced by a focus on public health and the need to intervene (Cairney, 2009c). Smoking prevalence has fallen, smoking behaviour has become ‘denormalised’ (Studlar, 2007a: 1), and many countries have used similar policy instruments to control tobacco (Studlar, 2004). The driver for convergence is the increased acceptance of the scientific evidence on smoking and passive smoking (Feldman and Bayer, 2004: 1). The post-war debate on the ‘facts’ was replaced by government acceptance of the scientific evidence, and the debate has shifted to the question: ‘what form of tobacco control works best’ (Cairney, 2009c)? Tobacco control based on public health knowledge is an idea ‘whose time has come’ (at least in many countries) and the spread of the ‘virus’ explains significant policy changes. However, there are two limitations to this explanation. First, the response to this idea has varied markedly according to the ‘vested economic interests, cultural practices, and political factors’ of each country (Studlar, 2004). Second, there have been significant time lags (20–30 years) between the proposal and acceptance of scientific knowledge and then the introduction of concrete policies to address the problem (Studlar, 2007b; Studlar and Cairney, 2014; Cairney and Mamudu, 2014; Mamudu et al., 2015; Cairney and Yamazaki, 2018). The spread of the virus depends on levels of resistance by the host. When we assess the strength of ideas, we must consider how successfully they are promoted and how receptive decision makers are to them.

HALL’S POLICY PARADIGMS AND THIRD-ORDER CHANGE Some studies portray ideas as ‘paradigms’, or overarching views of reality which limit the terms of discussion. Most notably, Hall’s (1993) study of economic policymaking gauges the link between ‘learning’ (Chapter 12) and first-, second-, or third-order changes in policy (compare with Campbell, 2002: 23). First-order change is incremental, with current policy based on the lessons of past policy decisions. It refers to a change in policy instrument settings while maintaining the instruments themselves (such as the minimum lending rate) and the government’s overall goals. Learning is generally internalized, with civil servants and experts receiving minimal attention from ministers and interest groups. Second-order change is also based on adapting to experience while maintaining overall goals, but with more wholesale changes to policy instruments (such as new systems of monetary and public spending control). While outside interests are more involved, their views are used by officials to support changes sanctioned from within. Third-order change refers to a radical shift in policy involving changes to the ‘hierarchy of goals behind policy’ (such as the replacement of unemployment by inflation as the key concern). Third-order change is rare and can be compared to Kuhn’s ‘paradigm shift’. Policy instruments are underpinned by policy goals that, in turn, are underpinned by a ‘policy paradigm’, or a wider understanding of the world: Policymakers customarily work within a framework of ideas and standards that specifies not only the goals of policy and the kind of instruments that can be used to attain them, but also the very nature of the problems they are meant to be addressing … this framework is embedded in the very terminology through which policymakers communicate about their work, and it is influential precisely because so much of it is taken for granted and unamenable to scrutiny. (Hall, 1993: 279)

First- and second-order changes represent ‘normal policymaking’, or the adjustments made within an existing paradigm. Third-order change is more radical and requires a major departure from the way that policymakers would normally think and act. It follows significant policy failures which command the attention of the wider political world, call into question current thinking, and undermine its advocates. It produces a shift of power within government, with new governments taking over and introducing radically different policies, or existing policymakers rejecting the advice of one set of experts in favour of another. In other words, those closest to the existing paradigm are only likely to make incremental changes, adapting and ‘stretching’ their theories to accommodate new information. At the brink of third-order change, their concepts are so stretched that they cannot meaningfully explain and address policy problems. This prompts a ‘first principles’ battle of ideas (in which current assumptions are challenged and much less is taken for granted) which ends ‘when the supporters of a new paradigm secure positions of authority over policymaking and are able to rearrange the organization and standard operating procedures of the policy process so as to institutionalize the new paradigm’ (Hall, 1993: 281). From this approach, we can ascribe two main roles to ideas. First, paradigms undermine policy change. Policymakers establish a language and set of policy assumptions that excludes most policy participants. For Hall (1993), first- or second-order changes are characterized by the insulation of civil servants and experts from ideational challenge. Third-order change is rare and only occurs when policy failures are so significant that they produce shifts in power that displace existing policymakers and cause current experts to fall out of favour. Second, when new ideas are adopted, they cause a complete shift in the way that policy is understood and made within government (Hall, 1993: 287). Such change is rare or exceptional, although Hall suggests that it happens more often than it used to; the ‘outside marketplace’ for ideas has expanded and opportunities are more likely to ‘spring up’ to ‘provide outsiders with influence over a formerly closed process’ (1993: 289; see also Baumgartner and Jones, 1993: 42–43).

Politics is therefore not just about policymakers exerting power, but also outside interests acquiring power ‘by trying to influence the political discourses of the day’ (Hall, 1993: 290). This ability to influence policy increases as policy is seen to fail, and the current policy paradigm breaks down if it cannot adapt to the problems it faces. It prompts political actors to provide a new way to understand and solve policy problems. The policy process is therefore characterized by ‘the presence of a policy paradigm [generating] long periods of continuity punctuated occasionally by the disjunctive experience of a paradigm shift’ (Hall, 1993: 291). In effect, the privileged status of a paradigmatic set of ideas ensures that most ideas are ignored or rejected, revisited only in rare cases of thirdorder change following policy failure. Hall’s work is a key reference point in policy studies, but note two competing frames of reference. First, Hall often appears to be describing policymaking stability/instability and policy continuity/change across an entire policy system, compared to Chapter 9 in which monopolies form in subsystems (Cairney and Weible, 2015: 88; Baumgartner, 2013: 251; 2014: 476). Second, a large part of the ‘new institutionalist’ literature now seeks ways to account for major policy and policymaking changes in the absence of crisis or punctuation (Chapter 5). In other words, Hall’s article remains influential but primarily as a well described account rather than an argument confirmed empirically over time and in many places.

MULTIPLE STREAMS ANALYSIS (MSA) Kingdon’s (1984; 1995) MSA is one of the most-cited and most-used approaches to policy studies. It provides an effective way to describe the role of ideas and agenda setting within a broader understanding of policymaking. As Chapter 9 discusses, there is an almost unlimited amount of policy problems and solutions that could be considered, but few reach the top of the policy agenda. So, what determines which problems and solutions are discussed, and if an idea is accepted? Kingdon (1984) suggests that new ideas are not accepted, and policy does not change significantly, unless three

separate streams – problems, policies, politics – come together at the same time during a ‘window of opportunity’. His argument is inspired by the ‘garbage can model’ (Cohen et al., 1972) which questions the assumption of comprehensive rationality in decisionmaking organizations (Chapter 4) and helps us understand why we should treat key ‘stages’ of a policy cycle as separate processes (Chapter 2).

A Garbage Can Model of Organizational Choice Cohen et al.’s (1972) concept of ‘organized anarchy’ may be comprehensive rationality’s polar opposite. Organizations do not make choices based on clearly defined and shared aims: •

•

•

There is no clear process in which the values of the policymaker determine policy aims, the organization identifies systematically all the means to achieve those aims, and then the policymaker selects the best solution. Rather, policymakers and organizations juggle a set of contradictory preferences which are difficult to define and rank. Indeed, people often fudge their objectives to avoid confronting their inconsistencies. Analysis of the decision-making context is never comprehensive, and organizations can never consider all relevant factors/possibilities. Rather, time is limited, the attention of participants to each possibility is uneven, the identification of preferences is based on ‘trial and error’, and decisions to act are based on ‘learning from the accidents of past experience’ (1972: 1). ‘Rationality’ can mean more than one thing in politics. While comprehensive rationality suggests that organizations use reason to identify problems and then find solutions, the conception of rationality as self-interest suggests that organizations seek ways to protect or enhance their position by justifying their own existence.

Such is the departure from comprehensively rational policymaking that solutions appear to be formulated before problems are identified

or articulated clearly. Organizations represent ‘collections of choices looking for problems, issues and feelings looking for decision situations in which they might be aired, solutions looking for issues to which they might be an answer, and decision makers looking for work’ (Cohen et al., 1972: 1). The garbage can is a receptacle into which a volatile mix of problems and solutions are dumped. Comprehensive rationality requires a policy cycle with a linear process involving the identification of a problem, the production of solutions, and the choice of a solution which solves the problem. In contrast, organized anarchy suggests that the organization consists of a number of ‘relatively independent streams’ which enjoy a much more complicated and unpredictable relationship. While problems are identified, solutions proposed, and choices made, it is by no means inevitable that the process is chronological or that these streams will come together in any meaningful way (1972: 3). Problems are poorly understood, goals are ambiguous, conflict is commonplace, and decision makers ‘may have other things on their mind’ (1972: 16). Therefore, outcomes often depend not on ‘organizational goals’ but on which participants within the organization become most involved. Cohen et al.’s (1972) exemplar of garbage can policymaking is a university. Kingdon (1984: 82–83) extends these themes to the policy process in the US federal government, highlighting the elusiveness of comprehensive rationality when people and organizations have limited cognitive abilities; people move in and out of the policy process; coalitions are built of people with different goals; and people appear to be creating problems to justify their policy desires. As a result, the policy process is as much accidental as rational. As the size of an organization or system increases, so does fragmentation, variations in participation/interest, the inability to define aims precisely, and the unpredictability of agenda setting and policy choice.

The ‘Window of Opportunity’ Consequently, we should think of three requirements – to define a policy problem, propose a solution, and select it – as three separate

streams rather than three linear stages. Identifying a problem does not set in motion the inevitable production and selection of a solution. Rather, three streams must come together at the right time. 1. Problem stream: attention lurches to a policy problem. Problems are policy issues that are deemed to require attention. There are no neutral indicators to determine which problems are the most deserving of our attention (Chapter 9; Majone, 1989: 24). This is not an ‘evidence-based’ process to provide an objective account of the size of the problem. It is a political process to exercise power to ensure that one subjective account wins at the expense of all others (Cairney, 2016a: 42). Problems receive attention based on how they are defined by policy participants. Agenda setting involves a process of issue framing, assigning causal responsibility, and competition to choose one indicator of policy problems over others (Kingdon, 1984: 98–99). Note the key distinction (introduced in Chapter 4) between • •

uncertainty, in which there is imprecision or a lack of evidence, and actors can inform their audience; and ambiguity, in which perceptions of problems can change and actors can exploit their audience’s ambivalence or ability to interpret the same problem in different ways (Zahariadis, 2007: 66).

In a political environment where the evidence is rarely conclusive enough to remove uncertainty, persuasion and argument (not ‘facts’) are the tools used by policy participants to address ambiguity (Majone, 1989: 8; 21; Cairney et al., 2016; Wellstead et al., 2018). Further, persuasiveness relies on the existing beliefs of the audience. In some cases, problems receive attention because of a crisis or focusing event (Birkland, 1997). In others, a change in the scale of the problem will draw attention. However, a focusing event on its own is insufficient. It may need to reinforce a problem ‘already in the back of people’s minds’ (Kingdon, 1984: 103). Most problems do not receive such attention by decision makers. Policymakers ‘could attend to a long list of problems’ but

they ‘pay serious attention to only a fraction of them’ (1984: 95, 120). The policy agenda is congested, policymakers often focus on solving the problems of previous policies (1984: 106, 196), and the perceived intractability or excessive costs of new solutions may undermine serious consideration. Therefore, raising an issue to the top of the policy agenda, and getting people to see new problems, or think about established problems in a new way, is a major accomplishment (1984: 121). It must be acted upon quickly before attention shifts elsewhere, partly by demonstrating that a feasible solution already exists. 2. Policy stream: a feasible solution to that problem is available. Policies are ideas/solutions proposed by actors. They should be considered as ‘relatively independent’ of the problem stream for two main reasons. First, ‘solutions’ may primarily be strategies to address some other aim, such as to protect an organization or get elected (Kingdon, 1984: 129–30). Second, problems rise and fall on the agenda relatively quickly, while solutions take much more time to develop and refine. Kingdon employs an evolutionary metaphor to describe the time and effort it takes for feasible policy solutions to develop (see Chapter 6 and Cairney, 2013b on the many, confusing, evolutionary metaphors in policy studies). They whirl around in the ‘policy primeval soup’, evolving and mutating as they are proposed by one actor but then reconsidered and modified by a large number of policy participants. The development of new ideas generally takes place within networks of ‘researchers, congressional staffers, people in planning and evaluation offices and in budget offices, academics, interest group analysts’ (Kingdon, 1984: 18). There are three key elements to this process. First, ideas ‘come from anywhere’ and tracing the original source is futile since ‘nobody really controls the information system’ to the extent that they can maintain the same basic solution from start to finish (Kingdon, 1984: 78; 81; compare with Smith, 2013). Second, the process of proposing new ideas and having them accepted usually takes a long time. Policy specialists with established

views have to be ‘softened up’ to new ideas (1984: 134–46; compare with 1984: 18, Durant and Diehl, 1989 and Zahariadis, 2007: 72 on the rapid adoption of some ideas). Third, while there are many ideas floating around in the ‘policy primeval soup’ – that is far more than are considered by policymakers at decision time – only some ‘survive and prosper’ by adapting to meet certain criteria (1984: 131; 123). Solutions with a realistic chance of success are characterized by their ‘feasibility’: • •

Technical feasibility. A solution will work as intended if implemented. Political feasibility. A solution will receive enough support, based on factors such as: ‘value acceptability within the policy community’; tolerable anticipated costs; public acceptability; and, a ‘reasonable chance for receptivity among elected decision makers’ (1984: 138–46).

On the basis of these criteria, actors in policy networks produce ‘a short list of ideas’ and accentuate some policy problems over others, to maximize the chance that an idea will be accepted (Kingdon, 1984: 146). In other words, to deal with this disconnect between lurches of attention and the time it takes to produce solutions, communities of policy specialists develop proposals in anticipation of future problems: ‘They try out their ideas on others by going to lunch, circulating papers, publishing articles, holding hearings, presenting testimony, and drafting and pushing legislative proposals’ (1984: 122–24). Then, proponents of those solutions either chase or help define policy problems. There are numerous instances of solutions chasing problems in politics. For example, the solution of mass public transport addressed the problems of congestion in the 1950s, pollution in the 1960s, and oil shortages in the 1970s (1984: 181), while it may now be linked to excessive fuel prices or climate change. Health maintenance organizations (HMOs) were initially framed in the USA as a means to extend healthcare delivery to the poor, then advanced to reduce healthcare costs in general (Stone, 1989: 298). In the UK, ‘harm reduction’ to make drug use safer

was adopted following the identification of HIV/AIDS, and its description shifts continuously to address new issues (Cairney, 2002; Zampini, 2018). The use of private healthcare was framed by the UK Conservative government as ‘rolling back the state’, while under Labour it was a means to reduce surgical waiting lists. In many Western countries, major public health reforms became attached to the agenda set by 9/11 and the threat of bioterrorism (Avery, 2004) and an ‘austerity’ agenda to reduce the cost of public services (Cairney et al., 2018; Cairney and St Denny, 2020). 3. Politics stream: policymakers have the motive to turn that solution into policy. The politics stream regards how receptive policymakers are to certain solutions at particular times. Changes in the political system often cause rising attention, and receptivity, to a particular problem and its solution. They include ‘Swings of national mood, vagaries of public opinion, election results, changes of administration … turnover in Congress … and interest group pressure campaigns’ (Kingdon, 1984: 19), although Howlett (1998) finds that elections provide the most frequent impetus. Each change may affect the receptivity of policymakers to ideas depending, for example, on the need to ‘build electoral coalitions’ or become re-elected (Kingdon, 1984: 19). Major political events, such as the election of a new president, make ‘some things possible that were impossible before’ and ‘other things out of the question’, and create ‘a receptivity to some ideas but not to others’ (1984: 152). Similarly, policymakers estimate how receptive the public is to government policy (e.g. when the ‘national mood’ appears to be against ‘big government’ and higher taxes – 1984: 154). They also weigh up the balance of interest group opinion and assess the political costs of going against the tide (1984: 157–58). On the one hand, this process is an important source of inertia: existing government programmes have an established clientele with much to lose from policy change, while many equally worthy policy alternatives may suffer from a lack of organized advocacy

(1984: 159). On the other hand, swings in national mood and changes in government can be enough to overcome these obstacles, while some policymakers thrive on a challenge to vested interests (1984: 160). Or, if there are a number of venues that could introduce policy changes, the competition between them could either constrain or accelerate the promotion of issues to the top of the agenda, depending on the perceived popularity of the issue and the political benefits to be had from jumping on the bandwagon (1984: 165).

Why Are the Three Streams Separate and How Do They Come Together? The ‘garbage can’ model has given us a general reason to separate the three streams and recognize that policymaking does not follow a linear policy cycle. Kingdon’s specific analysis of the USA also highlights the role of scale, the crowded policymaking environment, and the tendency to break the policy process down into more manageable sectors and processes. Different actors tend to play different roles. For example, the president may set the agenda but is ‘unable to determine the alternatives that are seriously considered, and is unable to determine the final outcome’ (Kingdon, 1984: 26– 27). Interest groups may get enough people to pay attention to an issue but cannot determine what solutions are considered or chosen within policy networks (1984: 53). While public and media attention may cause policymakers to focus on certain issues, it may not last long enough to influence the lengthy consideration of solutions (1984: 62). While elected politicians and senior officials may have a strong effect on the definition of problems, and the final decisions, civil servants and other policy specialists play a larger part in the coordination, selection, and presentation of solutions. Since policymakers do not have the time to devote to detailed policy work, they rely on civil servants who consult with interest groups, think tanks, and other policy specialists to consider ideas and produce policy solutions.

In turn, policymakers decide which problems they deem most worthy of their attention. They do not ‘discover issues’; they ‘elevate issues’, particularly when responding to relatively fleeting public and media opinion (1984: 31; 40). Therefore, the strategy of interest groups or policy proponents is either to create enough interest in policy problems to lobby for a solution or to wait for the right moment in politics that gives them the best chance to promote their solution. This emphasis on ‘solutions chasing problems’, often with the help of ‘policy entrepreneurs’ who frame issues and promote their ‘pet’ solutions to policymakers, is central to the acceptance of new ideas (Box 11.3). Crucially, if the motive and opportunity of policymakers to translate ideas into action are temporary, then this limits the time to find policy solutions when a new policy problem has been identified: ‘When the time for action arrives, when the policy window … opens, it is too late to develop a new proposal from scratch. It must have already gone through this process of consideration, floating up, discussion, revision and trying out again’ (1984: 149). Therefore, ‘advocates lie in wait in and around government with their solutions at hand, waiting for problems to float by to which they can attach their solutions, waiting for a development in the political stream they can use to their advantage’ (Kingdon, 1984: 165–66). This window of opportunity for major policy changes opens when: Box 11.3 Policy entrepreneurs and ‘evidence-based policymaking’ There are many, many ways to describe policy entrepreneurs (Cairney, 2012a: 272; Aviram et al., 2019; Beeson and Stone, 2013; Bakir, 2009; Bakir and Jarvis, 2017; Chappell, 2006b: 230; Christopoulos and Ingold, 2015; Faling et al., 2019; John, 1999: 45; Jones, 1994: 196; Laffan, 1997; Roberts and King, 1991; Smith, 2018). In business, entrepreneurs are innovative actors who take a chance to market and provide a good or service that otherwise may not be provided; in return for their services they take a profit. The amount of profit depends partly on the level of competition by different entrepreneurs (McLean, 1987: 28). In

domestic politics, entrepreneurs also engage in marketing, often via trial-and-error strategies to frame problems and seek support for solutions (Jones, 1994: 196; John, 1999: 45). They profit through other means, as leaders of a group or coalition receiving financial and political support; as politicians receiving votes; or as supporters of a cause securing policy change (Mintrom and Vergari, 1996: 431; Mintrom and Norman, 2009). In international politics, entrepreneurs can be consultants, NGOs, or think tanks, backed by countries or supranational institutions seeking policy change in other countries (Chapter 12). However, Kingdon (1984: 21; 104) describes entrepreneurs as actors who use their knowledge of the policy process to further their own ends. They may be elected politicians, leaders of interest groups, or unofficial spokespeople for particular causes. They are actors with the knowledge, power, tenacity, and luck to be able to exploit windows of opportunity (Kingdon, 1984: 165– 66). This focus on MSA helps us understand how proponents of ‘evidence-based policymaking’ (Box 2.2) can learn from the ‘three habits of successful policy entrepreneurs’ (Cairney, 2018a). First, they know that evidence does not speak for itself. Instead, entrepreneurs tell a story, to frame their evidence in relation to the beliefs of their audience (Knaggård, 2015; Blum, 2018). Second, they know that, when attention lurches to a policy problem, it is too late to begin to produce a new solution. Instead, they have a technically and politically feasible solution worked out and ready to go. Third, they know that they are part of a wider environment to which they must adapt. In crowded systems, like the US federal level, they are like surfers waiting for the big wave. In sub-national or niche areas, they can have a greater influence (Cairney and Jones, 2016: 46). Separate streams come together at critical times. A problem is recognized, a solution is developed and available in the policy community, a political change makes it the right time for policy change, and potential constraints are not severe … these policy windows, the opportunities for action on given initiatives, present

themselves and stay open for only short periods. (Kingdon, 1984: 174) The policy window can be opened by any one stream. For example, a policy problem may suddenly appear to be a crisis requiring immediate resolution, a new policy solution could make a problem solvable and therefore more worthy of attention, or a change of government could open up avenues of influence for policy entrepreneurs (1984: 21; 204). Then, if the three streams come together, each factor acts as an impetus to policy change: a policy problem is reframed, a new solution is identified, and the conditions are right for policy change. However, if they do not, then each factor can act as a constraint to further action (1984: 19). For example, a problem as currently defined, or the ideological stance of governing parties, may preclude certain solutions. Or, the problem may appear intractable, with solutions deemed ineffective or too expensive. The overall point is that the use of particular ideas to define and solve policy problems is far from a straightforward process. Therefore, to explain the adoption of ideas we must explore the particular circumstances in which policy change takes place. This description can involve a clumsy combination of metaphors. The policy window opens to allow the three streams to couple, and each stream is dumped into a garbage can, with policy outcomes depending on ‘the mix of elements present and how the various elements are coupled’ (Kingdon, 1984: 174) and the extent to which a solution has emerged from the policy primeval soup! In my ever-so-humble opinion, the clearest metaphor is of a timely space launch that requires key conditions to be met at the same time: ‘policymakers will abort the mission unless every relevant factor is just right’ and those conditions are out of the control of any single actor (Cairney, 2018a: 202). In other words, even when major policy change is likely to take place, the final outcome is rather unpredictable. It depends on factors such as the ability of the public to maintain their attention to a problem, the ideas available to solve it, and the spirit of compromise in the political stream (Kingdon, 1984: 186).

The image of ‘organized anarchy’ is furthered by the unpredictability of many windows of opportunity and the short period of time in which the window stays open: focusing events (Chapter 9) do not command attention for long, new governments only enjoy short honeymoons, and/or no feasible policy solutions appear to exist (1984: 177–78). The infrequency of policy windows also suggests that, when new legislation looks likely to be adopted, there is a deluge of interest and a range of participants keen to jump on an idea’s bandwagon: ‘the submission of a legislative proposal becomes a garbage can into which modifications, amendments, wholly new directions, and even extraneous items can be dumped as the bill wends its way through the legislative process’ (1984: 186). Overall, the image of messy politics portrayed by multiple streams reflects the ambiguity inher-ent in the policy process and a ‘partially comprehensible’ world. People may not know what they want and governments may make promises to solve problems they do not fully understand: ‘Yet, choices are made, problems are defined and solutions are implemented’ (Zahariadis, 2003: 1). MSA therefore combines two main views of ideas. An ‘idea whose time has come’ suggests inevitability and that the idea is the main source of explanation. The policy window suggests contingency; receptivity to an idea is more important than the idea itself (Kingdon, 1984: 76; 101–3). A combination of these views suggests that: An idea’s time arrives not simply because the idea is compelling on its own terms, but because opportune political circumstances favour it. At those moments when a political idea finds persuasive expression among actors whose institutional position gives them both the motive and the opportunity to translate it into policy – then, and only then, can we say that an idea has found a time. (Lieberman, 2002: 709) Further, since a policy window does not stay open very long, an ‘idea’s time comes, but it also passes’, particularly if the reasons for a particular level of attention to the policy problem fade before a coalition behind policy change can be mobilized (Kingdon, 1995: 169). There is an almost infinite number of ideas which could rise to

the top of the political agenda, producing immense competition to dedicate political time to one idea at the expense of the rest.

THE IMPRESSIVE GENERALIZABILITY OF MSA How relevant is a theory, based on the US political system in the 1980s, to modern-day policymaking in the USA and other countries? For example, the focus on a messy political process – linked strongly to the USA’s separation of powers – seems increasingly relevant to a European Union (EU) process characterized by power diffusion between EU organizations, member states, and sub-national authorities (Richardson, 2008; Cairney, 2009c). However, some aspects of the US system may seem alien to students of relatively centralized parliamentary systems. For example, Kingdon’s (1984: 186) picture of a legislative proposal as a garbage can ‘into which modifications, amendments, wholly new directions, and even extraneous items can be dumped’ only holds when multiple participants can amend legislation. This is unlikely to occur in a parliamentary system in which one party of government dominates proceedings (Page, 2006: 208). More generally, if we apply MSA to many other countries, or sub-national politics within those countries, we may find that the policy environment is less crowded, access to decision makers is more controlled, the turnover of politically appointed civil servants is less significant, and multiparty competition is more important (Zahariadis, 2003: 16–17; 2007: 77; Cairney, 2018a: 207; Herweg et al., 2015, 2018). In other words, context matters, and the direct relevance or implications of MSA may be more or less apparent in different systems. Yet Kingdon’s book has been cited (in Google Scholar) more than 22000 times and MSA studies have taken off internationally. Jones et al.’s (2016) review of the MSA literature from 2000 to 2014 finds 311 published applications in 165 journals (in English) covering 22 different policy areas in 65 countries, including 132 case studies of the USA, 205 European or EU case studies, and 140 outside both (many publications have multiple case studies, so the total is greater than 311). Further, there are more than 100 studies of sub-national policymaking.

A big part of MSA’s success comes from the attractiveness of its metaphor and the intuitive feel of its description of policymaking (Cairney and Jones, 2016: 37). Its abstract nature also allows highly flexible applications. The concept of ‘organized anarchy’ was originally developed to apply to decision-making within organizations such as universities, the model’s focus on bounded rationality applies to all policymaking, and the three streams metaphor can be applied somehow to most (if not all) political systems (Cairney, 2009c). In the abstract, the common aim of MSA studies is to identify the successful adoption of one particular idea at the expense of many others within a political process characterized by: • • • • •

Ambiguity. There are many ways to frame any policy problem. Competition for attention. Few problems or definitions of those problems reach the top of the policy agenda. An imperfect selection process. New information is difficult to gather and subject to manipulation and interpretation. Limited time. Policymakers have to make choices before their preferences are entirely clear. A departure from comprehensive rationality and a linear policymaking process in which problem definition precedes solution production and selection (Zahariadis, 2003: 2–15).

Consequently, we can identify three types of application. The first type demonstrates that MSA travels well to many other countries and systems, although comparative empirical studies tend to highlight very different windows of opportunity driven by major differences in political systems and policy histories. An exemplar is Zahariadis’ (2003) analysis of privatization policy in Western Europe (Box 11.4). Box 11.4 The comparative application of multiple streams analysis (MSA) Zahariadis (2003) demonstrates the explanatory power of MSA in political systems with different cultures and institutions. Using privatization policy as a key example of policy change, he shows the relatively high receptivity to this broad idea in the UK –

following the election of Margaret Thatcher’s Conservative Government in 1979 – but the need for specific windows of opportunity for specific sales of nationalized industries (2003: 59). For example, the privatization of British Telecom (BT) required its initial separation from the Post Office (which made the sale more feasible technically) and the refusal of the government to consider any other politically feasible option to fund its expansion activities (2003: 32–33). This move in 1984 created a strong precedent for privatization, since it provided political rewards to a government keen to reduce public sector spending, encourage popular capitalism through shareholding, and challenge public sector unions (2003: 34). However, further privatizations were not inevitable. It took a decade to sell British Rail, partly because it was far less technically feasible. Its finances were poorer than BT’s, it is hard to break up a rail network effectively, and the government would still need to oversee rail maintenance and safety. It took a Conservative Party election victory in 1992 (headed by John Major) to help open a new window (2003: 84). Zahariadis (2003) shows how streams couple in different ways in France. For example, in the case of telecoms, the problem was different from the UK: there were no equivalent budget problems or a split from the post office, and no political drive to challenge the industrial opposition to telecoms privatization. While a window for privatization was opened up by the politics stream, following the appointment of Jacques Chirac as prime minister in 1986, telecommunications did not join the list of 65 businesses sold. The French government was receptive to different ideas on the benefits of privatization, to ensure the state is not giving excessive subsidies for failing businesses or to encourage competition, rather than simply rolling back the state (2003: 38). In Germany, the problem and political streams were systematically different. The legacy of Nazi abuses of the state produced a lower propensity to nationalize industries, with German governments preferring to maintain a minority stake in companies. Further, previous pushes to promote popular capitalism in the 1960s were unsuccessful, while attempts in the 1980s (following the election of Helmut Kohl’s Christian

Democrats) to follow international privatization trends were undermined by more dominant problem definitions, including the value of certain industries to regional populations and the centrality of government-influenced industries to Germany’s international competitiveness (2003: 62). As a result, major privatizations only took place in the 1990s following the window opened by German reunification and the perceived need to dispose of much of East Germany’s industrial ownership (2003: 62). The second type highlights new issues that arise from empirical studies in other political systems (Cairney and Jones, 2016: 45; Rawat and Morris, 2016). For example, studies of EU policymaking demonstrate the close link between defining a policy problem and deciding which organisation is responsible for solving it (Ackrill and Kay, 2011; Sarmiento-Mirwaldt, 2013). There is therefore great potential for actors to engage in problem definition and ‘venue shopping’ to seek sympathetic audiences for their way of thinking (Chapter 9, p.149). Studies of policy transfer suggest that some policy solutions can be pre-softened elsewhere – at least to determine their technical feasibility – and reduce the time it takes to couple problems and policies (Bache and Reardon, 2013; Cairney, 2009c). Sub-national studies highlight that policy entrepreneurs can have a far more influential role in more than one stream, particularly when the environment is less crowded, such as when an issue is relatively niche or attached to a small budget (Cairney, 2018a: 209). Studies of countries like China highlight major political system differences, in which assumptions like the need for technical feasibility are misleading (Zhu, 2008). They prompt us to remember the assumptions that we make about MSA when we have US-style democracy in mind (Box 13.4). Consequently, the third type is theoretical refinement, in which scholars clarify aspects of MSA and update assumptions or hypotheses according to these new empirical results. Herweg et al. (2015; 2017) and Cairney and Zahariadis (2016) try to identify a core set of MSA hypotheses that would hold in comparative applications, and add more nuance to applications of MSA beyond the USA. For

example, Herweg et al. (2015: 437) remind us that political party competition has a major effect on all three streams (not just the political stream), particularly when there is high multi-party competition. Yet we should not overstate the effect of such refinement. MSA is characterized by a high level of ‘rudderless’ activity in which most scholars are inspired by Kingdon’s metaphor without engaging comprehensively with the state-of-the-art literature (Jones et al., 2016: 30).

CONCLUSION The role of ideas is important and its discussion within political science undoubtedly improves our ability to explain complex policy events and outcomes. However, the literature has the potential to confuse, for four main reasons. First, the term ‘idea’ can relate to a wide spectrum of concepts, including broad world views at one end and specific proposals on the other. Second, descriptions of the role of ideas do not make sense unless they are accompanied by descriptions of the ways in which political actors accept, reproduce, or make use of them. Therefore, even when the literature describes ideas as viruses, tides, norms, and paradigms shaping or limiting behaviour, such discussions generally complement theories which focus on power. Third, in our discussion of paradigms, we have encountered two different emphases. On the one hand, ideas command considerable explanatory power: a paradigm which becomes institutionalized can constrain and facilitate political behaviour for decades. On the other, the political system often seems resistant to new ideas. Indeed, the consequence of the dominance of one idea is that the vast majority of ideas are rejected or receive no attention. This gives us a dual picture of ideas: as important, sweeping everything aside and then setting the policy direction for decades, or as unimportant, since a very small number act in this way. Fourth, our discussion may highlight different links to power. For example, a focus on ideas signals a shift of focus from elitism to pluralism; from the power of a small band of elites towards a much wider range of experts and commentators engaged in the production

of, and argumentation about, ideas (John, 1998: 156). Public policy decisions are underpinned by societal values and moral judgements determined through a wider process of argumentation and reciprocal ‘persuasion’ within society rather than merely the exercise of interests (Majone, 1989: 2). Or, ideas – in the form of language or paradigms – limit wider debate and provide fertile conditions for the maintenance of power by elites (Chapter 3). Elected policymakers may not monopolize power when they are influenced by wider ideas which are outside of their control. However, this does not necessarily mean that power is dispersed. The use of ideas to establish policy paradigms suggests that other, unelected, elites operate under a cloak of anonymity for long periods. In this context, multiple streams analysis emphasizes the messy and often unpredictable nature of agenda setting and decisionmaking. The phrase ‘an idea whose time has come’ describes the notion that certain policy solutions and ways of thinking have profound effects on public policy. Yet major policy changes may not take place until these ideas ‘couple’ with newly framed policy problems and opportune political circumstances during a ‘window of opportunity’. In other words, an idea’s time only arrives when policymakers have the motive and opportunity to translate ideas into policy. Policy entrepreneurs can help ‘couple’ these streams, but as actors adapting to, not controlling, their environments. This messy image is something to bear in mind when, in Chapter 12, we extend our analysis to the transfer of ideas across political systems and subsystems. While the concept of policy transfer suggests that the same ideas may prompt similar actions across a range of countries, the ‘window of opportunity’ suggests that policy change often results from idiosyncratic elements linked to the particular circumstances of individual political systems. Ideology – a set of political beliefs and values held by an individual or social group. Independent variable – the object of study that one expects will cause a change in another object.

Norm – a rule or standard of behaviour considered to be normal and therefore acceptable. World view – a fundamental set of beliefs about how the world works and how we should engage with it. Dependent variable – the object whose change is caused by another. A change in the independent variable will cause change in the dependent variable. Kuhn’s (1970) – model of scientific progress suggests that it has not been incremental or based on the linear accumulation of knowledge. Rather, science is conducted and evaluated by communities who share a common paradigm or view of the world, and therefore do not question their own assumptions. When communities of experts fail to provide further advance, or cannot explain why the world does not work in the way they predict, they are replaced by communities with different ideas.

12 Policy Learning and Transfer Key themes of this chapter: • • • •

•

Policy learning describes acquiring new policy-relevant knowledge and skills. It sounds like a ‘rational’, ‘evidence-based’, and technical process, but it is just as political as other ways to exercise power to address bounded rationality. Policy transfer describes the import and export of policy. It can range from taking broad inspiration to importing a policy programme as a full package. Learning and transfer can be driven by (1) routine searches for evidence, dialogue, consensus-seeking, and bargaining but also (2) coercion, compliance, and a perceived need to keep up with international norms. The potential for ‘evidence-based’ transfer seems limited, so how can we produce more feasible models of learning?

INTRODUCTION: THE POLITICS OF POLICY LEARNING Policy learning is a vague but useful term to describe acquiring new knowledge to inform policy and policymaking. Knowledge can be based on information regarding a current policy problem, lessons from the past, or the experience of others. Policy transfer is also a broad term to describe the sharing of policy ideas from one place to another (Dolowitz and Marsh, 1996). Further, transfer is not the only term to describe the exchange of policies and ideas. Lesson-drawing brings together the study of learning from the past as well as other countries (Rose, 1993), policy diffusion describes the spread of solutions among, for example, US states (Walker, 1969), and policy

convergence refers to the factors causing similarities in policy across countries (Bennett, 1991a). Learning and transfer are political, not technical or objective, processes (Dunlop and Radaelli, 2018; Witting, 2017). In that context, the literature presents a fascinating dilemma. First, taken at face value, innovation, learning, and transfer seem crucial to evidence-informed global policy change (Page, 2018: viii). Why wouldn’t we want to learn from experience and, when appropriate, learn lessons from other actors? Why would policymakers keep making the same mistakes instead of learning from them? Why would a government try to reinvent the wheel when another government has shown how? Second, however, since the study of policy learning and transfer is about the power to produce and use particular forms of knowledge, we need to ask questions such as who is in charge, from whom do they learn, what rules do they follow, and is this process voluntary? Policy actors learn under the condition of bounded rather than comprehensive rationality, which means exercising power to decide how to understand a complex world in simple ways (Chapters 3 and 4). They have to act despite uncertainty (Nair and Howlett, 2017). They learn through the lens of their beliefs and compete with other actors to tell the most effective stories (Chapter 10). This competition to ‘learn’ can cause and exacerbate a sense of policy failure (Newman and Bird, 2017) or make it difficult to know if policy change results from the use of new data (Moyson et al., 2017: 165). Similarly, policy transfer involves agenda setting, to narrate the success of one government and pressure others to follow their lead. Transfer might follow a thoughtful ‘process of learning’ or be ‘the result of mimicry or good timing’ (Heikkila and Gerlak, 2013: 484). Overall, a focus on policy learning and transfer provides a lens through which to understand the policy process, without necessarily pinpointing the things that policymakers learn or the policies they import (Page, 2018; Stone, 2004: 549; James and Lodge, 2003; Evans and Davies, 1999). Third, we might still want policymakers to learn from each other and transfer policy success! We can recognize that the process is political and messy but still seek ways to do it effectively. Therefore,

we need to know exactly what policy learning and transfer are, and what studies of the process tell us, to know how to respond. Political science insights are crucial to the identification of sophisticated learning techniques that go beyond too-simple calls for ‘evidencebased’ learning and transfer (see Chapter 13). We need to understand the political choices associated with learning to address the dilemmas and trade-offs associated with ‘evidence-informed’ policy. Therefore, this chapter first describes the ways in which policy studies have made sense of the concept of policy learning. Second, it highlights the key questions associated with transfer. Does policy convergence follow the deliberate transfer of policy from one country to another? Is the transfer of policy voluntary? Which actors are involved? How much policy is transferred? How do we explain variations in levels of transfer? Third, it explores potential responses to the politics of learning and transfer: how can governments use knowledge, learn, and transfer policy in a way that would satisfy enthusiasts of ‘evidence-based policymaking’ (EBPM)?

POLICY LEARNING: WHAT DOES IT MEAN? WHAT ARE THE MAIN TYPES? Policy actors use the research process for many different purposes, from short-term problem-solving and long-term enlightenment to putting off decisions or using evidence cynically to support an existing policy (see Weiss, 1979 and Chapter 2, pp. 32–3 on evaluation). Similarly, policy learning is a highly political process akin to the ways in which actors exercise power to address bounded rationality (Chapters 3 and 4). They combine cognition and emotion (to produce heuristics) to understand, for example, their environment, how to exercise power to secure their political and policy aims, the nature of policy problems, and how to import policy lessons (Meseguer, 2005; Weyland, 2005; Witting, 2017). We should therefore reject the temptation to describe policy learning simplistically, with reference to a process that we might associate incorrectly with teachers transmitting facts to children. Nor

should we assume that adults simply change their beliefs when faced with new evidence, akin to the quotation often attributed to John Maynard Keynes: ‘when the facts change, I change my mind’. Rather, contrast this quotation with Hall’s (1993) account of Keynesian economics (Chapter 11, pp. 193–5) in which people mostly oversee limited policy change by interpreting new facts through the lens of their existing knowledge and beliefs. Consequently, profound change results more from experts being ignored or replaced; changing people, not changing their minds. Policy learning describes many types of political activity. Most people use the phrase loosely and instrumentally: we learn how to change our thought and behaviour to solve a problem or achieve an aim. It can describe many forms of instrumental learning, including how to adapt to a complex policymaking system or environment, use knowledge for policy, reduce uncertainty, encourage policy transfer, and learn about the most effective ways to learn (Dunlop et al., 2018: 1; Dunlop and Radaelli, 2013). Consequently, Bennett and Howlett’s (1992) review of classic studies (updated by Goyal and Howlett, 2018) reflects this wide range of possibilities, describing the ways in which governments or coalitions: • • • • • •

react to social and environmental pressures (Heclo, 1974) develop better policymaking functions (Etheredge, 1981) draw lessons from the experience of other governments (Rose, 1991) adapt to the success or failure of policy (Hall, 1993) use trial-and-error learning to make incremental changes (May, 1992: 333) use knowledge to update their beliefs (Box 12.1; Dunlop and Radaelli, 2013; 2018).

Therefore, when we explore policy learning, we need to seek clarity on our object of study, asking questions such as who learns, what do they learn, how do they learn, and what is the impact of learning on policy change (Moyson et al., 2017: 166; Moyson and Scholten, 2018)? For example, learning takes place at three levels:

1. Micro, as the ways in which individuals deliberate and act; 2. Meso, as the organizational rules to process information; and 3. Macro, as the learning that takes place in political systems, such as when governments draw lessons from each other (Moyson et al., 2017: 163–64). These levels are analytically separable but connected in practice. Individuals combine cognition and emotion to process information. However, they do it collectively, in (1) organizations, with norms of behaviour and modes of socialization that influence their motive and ability to learn, and (depending on what you read) either (2) systems, in which many groups of actors cooperate and compete to establish the rules of evidence gathering and analysis, or (3) environments that constrain or facilitate their action (Dunlop and Radaelli, 2017). In other words, policy actors do not simply learn by receiving clear information; they generate learning by engaging – individually and collectively – with many sources and forms of information (Freeman, 2006: 379). Boxes 12.1 and 12.2 show how to explore this process when informed by the advocacy coalition framework (ACF; Chapter 10) and the Institutional Analysis and Development (IAD) framework (Chapter 7). Learning may also have a cooperative or competitive purpose, from acquiring new knowledge and skills to solve a policy problem, with the help of an ‘informed, vigilant democratic public’, to enhancing knowledge on how to defeat our opponents in debate and dominate the policy process (Dunlop et al., 2018: 6; Dunlop and Radaelli, 2018: 256). We can update beliefs to ‘make policy more efficient, legitimate, democratic’, or ‘win consensus, to promote one’s strategy, to humiliate the opposition – without necessarily improving on efficiency or effectiveness’ (2018: 256). In that context, Dunlop and Radaelli (2013; 2018) draw on their major review of policy learning to describe four different categories, which relate partly to (1) general levels of knowledge and uncertainty about a policy problem or its solution and (2) the power and status of specific actors such as policymakers and experts:

Box 12.1 Individual and collective learning: Lessons from the advocacy coalition framework (ACF) The ACF describes the combination of individual and collective learning processes (Moyson, 2017: 321–22). It first describes two ways in which actors process new information (Sabatier, 1987). Sabatier (1998: 104) describes policy-oriented learning as ‘relatively enduring alterations of thought or behavioral intentions which result from experience and/or new information’. This type of learning often resembles Weiss’ (1979) classic description of the ‘enlightenment’ function of evidence that takes place over the long term. However, actors also learn through the lens of their beliefs and engage in ‘selective perception and partisan analysis’ (Sabatier, 1998: 104). They ‘learn’ more from the information they can add to their existing knowledge and beliefs, while challenging information can take far longer to make an impact. Second, the ACF describes learning within coalitions of actors who have some reason to trust each other enough to cooperate. For example, they share the same beliefs, have cooperated in the past, or follow authoritative actors (Ingold et al., 2017). Learning has three components: 1. Individuals use new information to inform and adjust their beliefs. 2. Members of coalitions interact and either reinforce or challenge each other’s views. 3. In some cases, internal conflict prompts a turnover of coalition members, which alters the knowledge and beliefs of the coalition (Jenkins-Smith and Sabatier, 1993b: 41). In most cases, learning follows the routine monitoring of implementation, in which new information does not challenge each coalition’s position. In rare cases, new information prompts ‘shocks’, such as when evidence emerges that policy is failing (prompting some coalitions to reconsider their beliefs). Finally, the ACF describes learning across coalitions. In low salience issues, coalition A may adapt to the beliefs of coalition B

when B’s views become ‘too important to ignore’ (1993b: 43). Or, scientists can become respected brokers between relatively moderate positions (Ingold and Gschwend, 2014). In high salience issues, coalitions romanticize their own cause and demonize their opponents (Sabatier et al., 1987). Learning is a political process, ‘not a disinterested search for “truth”’ (1993b: 45; Sabatier, 1988: 151). In some cases, there are commonly accepted ways to measure policy performance. In others, it is a battle of ideas, to frame issues, interpret experience, and challenge the status of an opponent’s knowledge. 1. Learning epistemically. In this scenario, there is high uncertainty, and there exists an ‘authoritative body of knowledge and experts who are willing and able to interact with policy-makers and take a proper role in the policy process’. This learning is the closest to teaching and ‘evidence-based policy-making’. It requires the scientific and ‘soft’ skills of researchers and the willingness and ability of policymakers to acquire new knowledge (2018: 259). 2. Learning from reflection. Uncertainty is high, but we cannot rely on expert authority, or ‘there is a predisposition to listen to what the others have to say and to re-consider one’s preferences’. Learning is closer to open dialogue in which people use deliberative techniques to (1) combine their diverse forms of knowledge and (2) encourage cooperation by agreeing on the social norms that guide their dialogue (2018: 260). 3. Learning as by-product of bargaining. There is low uncertainty because policy actors have ‘a repertoire of solutions, algorithms, or ways of doing things’. Many ‘interdependent’ actors – including ‘decision-makers, interest groups and civil society organizations’ – focus on how to bargain effectively (2018: 261). They learn (1) each other’s preferences, (2) which strategies work best, and (3) the cost of disagreement. In its ideal form, when power is not concentrated in a small number of actors, this form of bargaining is like Lindblom’s ‘partisan mutual adjustment’ (Box 4.2, p.62). It represents a highly defendable alternative to simply learning from experts (2018: 261).

4. Learning in hierarchies. Uncertainty is low and the authority of some actors is high. Subordinate actors learn that they are subject to the ‘shadow of hierarchy’, in which rules and norms appear to limit their options. Powerful actors learn about the levels of compliance they can achieve, or likely adherence to international norms (2018: 263). Since there are many different categories, there are also many factors than can facilitate or hinder learning. Facilitators include the routine consultation of experts (epistemic); the conditions for open, non-hierarchical, and transparent dialogue in which individuals are willing to change their minds (reflexive); mutual adjustment, by making trade-offs among a large bundle of issues, agreeing on the venues and procedures for negotiation, and engaging in many negotiations (bargaining); and trust in a clearly defined authority (hierarchy). Hindrances include unproductive debates among experts (epistemic); the ability to stifle dissent and unwillingness to compromise (reflexive); one-shot negotiations in which there is limited trust (bargaining); and too many ‘veto players’ or ‘jointdecision traps’ (2018: 263–66; Radaelli and Dunlop, 2013; Tsebelis, 2002; Scharpf, 1988). In other words, although learning sounds positive, it can be ‘dysfunctional’: actors may be dogmatic and coalitions subject to ‘groupthink’; reflection may be unproductive; and bargaining or hierarchy may focus on learning how to win arguments more than solve problems (Dunlop and Radaelli, 2018: 265–66; Dunlop, 2017a; 2017b). Further, Dunlop and Radaelli’s (2018) four categories show us what can happen when people have very different ideas about the meaning and purpose of policy learning. The classic example relates to our theme of ‘the politics of evidence-based policymaking’ (Cairney, 2016a): scientific experts will soon become disheartened if they assume that the process is epistemic and it is actually the byproduct of bargaining (Dunlop and Radaelli, 2016). Box 12.2 The mechanisms, products, and facilitators of policy learning

Heikkila and Gerlak (2013) describe the mechanisms of individual and collective learning: 1. Acquisition. Actors acquire information from a range of sources including direct experience, observation, expert networks, and deliberation forums. 2. Translation. They interpret new information to create policyrelevant knowledge. 3. Dissemination. Individuals share knowledge within groups, such as by telling stories of what they learned and how it affects the aims of group (2013: 488–91). Second, they identify products of policy learning: 1. Cognitive change, in which actors in a group exhibit ‘new or strengthened ideas, beliefs, or values’ about how to define and try to solve a policy problem. 2. Behavioural change, including new rules or strategies. As in studies of new institutionalism (Chapter 5), these rules are formal and written down or informal and communicated in different ways (2013: 491–92). Third, they identify key factors that influence these mechanisms and products: 1. The rules of collective action. Organizations can facilitate a relatively decentralized and open dialogue to encourage information to come from many sources, or centralized and rule-bound interaction to reduce the costs of coordination and dissemination. They can create actors as sources of authority, or ‘knowledge brokers’ to facilitate information exchange among actors with different ways of thinking. 2. Social dynamics, such as levels of trust or conflict among actors sharing information, a common language, and the shared rules or norms to which they refer during cooperation. For example, groups with high trust and a shared commitment to respectful criticism may facilitate effective learning. 3. Technology, to gather and store information.

4. External factors, including the sources of competition, political pressure, or crisis that prompt actors to learn more or less urgently (2013: 496–500). In each mechanism, all actors need to address bounded rationality. They use cognitive or institutional shortcuts to acquire information selectively, draw on heuristics and mental frames to interpret new information, and need to describe what they learned in terms of priorities and beliefs of the wider group (2013: 488– 91). Consequently, we should not assume that learning produces new and improved policy or policymaking (2013: 492). Nor should we assume that cognitive change is primarily responsible for policy change, compared to factors ‘including the resources of policy actors, the rules governing policy processes, and the political and economic climate’ (2013: 493–94). Dunlop (2017c) uses a case study – European Union (EU) policy on the supply of growth hormones for cattle – to describe this potential ‘irony of epistemic learning’ (see also Dunlop, 2014). First, a period of epistemic learning allowed scientists to teach policymakers the key facts on a newly emerging policy issue. The scientists, trusted to assess risk, engaged in the usual processes associated with scientific work: gathering evidence to reduce uncertainty, but always expressing the need to produce continuous research to address inevitable uncertainty. The ‘Lamming’ committee of experts commissioned and analysed scientific evidence systematically before reporting (1) that the use of ‘naturally occurring’ hormones in livestock was low risk for human consumers if administered according to regulations and guidance, but (2) it wanted more time to analyse the carcinogenic effects of two ‘synthetic compounds’ (2017c: 224). Second, a period of bargaining changed the context. EU officials in DG Agriculture responded to European Parliament concerns, fuelled by campaigning from consumer groups, which focused on uncertainty and worst-case scenarios. Officials suspended the committee’s deliberations before it was due to report, and banned the use of growth hormones. The

irony is two-fold. First, more ‘learning’ seemed to reduce effectiveness: 1. Scientists, operating in epistemic mode, described low risk but some uncertainty (in a scientific context where expressing uncertainty is routine good practice). 2. Policymakers, operating in bargaining mode, used this sense of uncertainty to reject scientific advice (in a political context where uncertainty can be damaging or used against you). The second irony is that scientists were there to help policymakers learn about the evidence, but were themselves unable to learn about how to communicate and form wider networks within a political system characterized by periods of bargaining. The moral is that we should not assume that ‘policy learning’ describes an apolitical process in which the most knowledgeable researchers are the most powerful participants. Rather, actors are learning how to make policy and get what they want, and new knowledge is one of many key factors relevant to that process.

APPROACHES TO THE STUDY OF POLICY TRANSFER Policy transfer could describe (1) the processes of policy learning that contribute to the sharing of policy from one government to another and/or (2) the products of that process, such as the import and export of policy. For example, Dolowitz and Marsh (2000: 5; 1996: 344) use transfer as an umbrella term to describe: the process by which knowledge about policies, administrative arrangements, institutions and ideas in one political system (past or present) is used in the development of policies, administrative arrangements, institutions and ideas in another political system. Their definition’s lack of parsimony reflects a diverse literature containing a family of concepts that overlap but ask different questions (Dussauge-Laguna, 2012a: 317; Heichel et al., 2005;

Benson and Jordan, 2011). Approaches include lesson-drawing, diffusion, and convergence studies.

Lesson-drawing Rose (1991: 4; 1993; 2005) asks: ‘Under what circumstances and to what extent would a programme now in effect elsewhere also work here’? He describes policymakers seeking to (1) learn from their own experiences and (2) take lessons from successful countries before calculating what it would require to import that success. Rose produces a series of questions about the nature and extent of lesson-drawing, such as from where are countries most likely to learn; how much of a policy programme is imported; and how easy will it be to adapt the programme, given the importing country’s social, economic, and political conditions? Programmes ‘are the stuff of public policy; they are concrete measures for doing such things as delivering hospital care’ (2005: 17). Further, a ‘lesson’ is ‘much more than a politician’s prescription of goals’. It also denotes ‘the laws, appropriations and personnel and organizational requirements needed for a programme to be put into effect’ (2005: 22).

Policy Diffusion The diffusion literature began as a study of policy innovation and emulation within the USA, before spreading to international studies (Busch and Jörgens, 2005). Walker (1969: 881–87) defines innovation as ‘policy which is new to the states adopting it’ and suggests it is most likely within: • • • •

The ‘larger, wealthier, more industrialized states’. States with high competition between candidates (producing a demand for new ideas) and turnover (most innovation takes place at the beginning of a new administration). States with the most extensive policy and research staffs. Urban states with more ‘cosmopolitan’ populations, more tolerant of change, or experiencing more planning and infrastructure-

based problems that require innovation. Berry and Berry (2018: 269) also identify ‘problem severity’ as a key impetus.

•

Walker offers several reasons for others to follow their lead: •

The policy may be viewed as successful (and/or popular among the public or a political party – Berry and Berry, 2018: 269–70). The federal government may encourage (or states may feel obliged to follow) policy uniformity and national norms. Interest groups promote best practice from the innovating states. States have close policy ties because they are close in another way; they have something in common, like a shared geography, ideology, or understanding of the policy problem. Boundedly rational policymakers use learning from other states as a shortcut. The innovation in one state may have a knock-on effect (‘spillover’) obliging other states to follow or lose out, because they compete with each other economically (see also Grey, 1973: 1182; Berry and Berry, 2007: 225–26; compare with Chapter 6, p.97 on the ‘Dye–Sharkansky–Hofferbert’ approach).

• • • • •

In some cases, policies are ‘virtually copied’ (Walker, 1969: 881; 898). Further, diffusion increases as the ability to communicate increases (Berry and Berry, 2007: 232). However, differences in policy conditions across different states, combined with varying degrees of willingness to follow the leader, create diffusion of different magnitudes and speed (Eyestone, 1977: 441). Further, problems with implementation can lead to ‘disinnovations’ (1977: 445). Box 12.3 Examples of widespread policy convergence, diffusion, or transfer •

The privatization of 8,500 state-owned enterprises took place in more than 80 countries from 1980 to 1992 (Kikeri, Nellis and Shirley, 1992).

•

•

• • • •

• •

The broad principles of ‘new public management’ (NPM) – the application of private sector business methods to the public sector – were just as widespread (Hood, 1995: 94; OECD, 1995; Ormond and Löffler, 1998), albeit with several variants (Common, 1998: 442; James, 2001). Now, scholars wonder if modern reforms are ‘post NPM’ (Ramia and Carney, 2010; Lodge and Gill, 2011) or in keeping with ‘public value’ or ‘new public governance’ (O’Flynn, 2007; Osborne, 2006; Goldfinch and Wallis, 2010). Policies to address money laundering were adopted by 170 countries in 20 years (Sharman, 2008). Most countries now have a ‘national environmental strategy’ (Busch and Jörgens, 2005: 868). Most countries have central banks, and have introduced reforms to make them legally independent (Marcussen, 2005: 903). The World Health Organization’s Framework Convention on Tobacco Control (FCTC) has been signed by 181 countries (90% of the world’s population), ‘which makes it the most widely embraced treaty in UN history’ (WHO, 2018). There have been many attempts to emulate the success of California’s ‘Silicon Valley’ (Giest, 2017). Governments can also learn what not to do (Stone, 2017). For example, many learned from the UK’s mistakes when tackling BSE in cows (Stone, 1999: 52).

Policy Convergence Policy convergence describes a movement towards policy similarities among different countries. Bennett’s (1991a: 216) review suggests that historical studies of ‘societal convergence’ sought to explain broad similarities with reference to worldwide industrialization, akin to the study of ‘globalization’ (Chapter 6) which produces common policy problems and some pressures to converge. In this scenario, countries with similar policy problems could address them in similar ways without knowledge of each other

and without transfer (Holzinger and Knill, 2005: 786; Hoberg, 2001: 127). However, Bennett (1991a: 217) argues that we should examine the political processes involved rather than imagine a ‘black box’ between global forces at one end and outcomes at the other. In other words, identify the decisions of policymakers when they weigh up external constraints to converge and domestic pressures to diverge or stay different (Hoberg, 2001: 127). Bennett (1991a: 231) identifies four main causes of convergence: • • •

•

Emulation, in which the model adopted by one country ‘serves as a blueprint that pushes a general idea on to the political agenda’ of another (1991a: 221–22); Exchange, of ideas between international policy networks (1991a: 224); Interdependence and the transnational nature of many policies. This prompts governments to search for ways to cooperate and ‘mitigate the unintended external consequences of domestic policy’. Such harmonization may be facilitated by supranational organizations such as the EU (1991a: 226); Penetration, in which the actions of one or more countries (or organizations) puts pressure on others to follow (see also Holzinger and Knill, 2005; Heichel et al., 2005).

POLICY TRANSFER: WHO DOES IT? Most actors involved in transfer are the regular participants in domestic policy processes, including, ‘elected officials, political parties, bureaucrats/civil servants [and] pressure groups’ (Dolowitz and Marsh, 1996: 345; compare with McCann and Ward, 2012: 326– 27). Further actors include: 1. Supranational organizations. They include the EU, the Organisation for Economic Cooperation and Development (OECD), World Bank, and United Nations (UN), and specific actors such as the European Commission and the World Health Organization (Benson and Jordan, 2011: 369).

2. Entrepreneurs. ‘Policy entrepreneur’ describes actors who sell policies (Box 11.3, p.200). NGOs (non-governmental organizations) such as think tanks or philanthropic organizations promote best practice internationally (Stone, 2000; 2010). Consultants use their experience in one country to sell that policy to another (O’Neill, 2000: 63; Common, 1998: 440; Dolowitz and Marsh, 1996: 345), often backed by an exporting country or supranational institution (Dolowitz and Marsh, 2000: 10). 3. International policy networks. An ‘epistemic community’ is an international ‘network of professionals with recognized expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge within that domain or issue-area’ (Haas, 1992: 3; see also Rose, 1991: 6; Campbell, 2002: 25; Holzinger and Knill, 2005: 784). Epistemic communities contain actors with shared beliefs and a ‘common policy enterprise’. They operate like domestic policy networks connected through international professional organizations (Bennett, 1991a: 224), with members of networks promoting shared ideas within their own countries (such as the ‘carbon capture’ community – Stephens et al., 2011). Other examples of networks include the ‘Windsor conference’ that brings together labour and social policy civil servants from ‘Australia, Canada, Ireland, New Zealand, the UK and the USA’ (Legrand, 2012a: 523). 4. Multinational or transnational corporations. Multinational corporations (MNCs) ‘hold a trump card’ because they can ‘threaten to take their business elsewhere’ if governments do not minimize their regulations on companies (Dolowitz and Marsh, 1996: 348). However, the effect of their demands on individual countries – or cities or regions within them – varies considerably (see Chapter 6, p.99 on the ‘race to the bottom’; Bennett, 1991a: 228–29). 5. Other countries.

Policy transfer often follows pressure from other countries, directly (through encouragement or coercion) or indirectly (by establishing new norms, inspiring others to follow, or causing unintended consequences for another country). Holzinger and Knill (2005: 785) suggest that the innovating country benefits from emulation because it reduces transactions costs when dealing with others (e.g. if they have the same standards or even adopt a common currency). In some cases, there are systematic leaders. The USA provides ideas and the main source of influence behind coercive forms of policy transfer (see the section ‘Why Transfer? Is It Voluntary?’). Sweden has long been a beacon for the social democratic state, and ‘Nordic welfare states’ provide wider lessons (Blum, 2014: 358). Germany has a history of effective inflation control, and Japan has become known for innovation in a range of areas (Rose, 1993: 107–108). Further, the decision to transfer once can establish a longer-term borrowing and lending relationship. For example, Japan borrowed Germany’s police system because it had already borrowed related aspects in law and local government (Page, 2000: 6; Westney, 1987). However, while countries like Canada and the UK tend to borrow from the USA, they also lend some policies in return (Dolowitz et al., 1999; Hoberg, 1991; Dolowitz and Marsh, 2000: 10). Relationships between other countries also demonstrate phases or issue-specific patterns. While Germany exported unemployment policies to Sweden and the UK in the early twentieth century, Sweden exported data protection laws to many countries from the 1970s (Dolowitz and Marsh, 1996: 352) and the UK was one of the frontrunners in privatization from the 1980s. While New Zealand initially borrowed aspects of NPM, it then became a beacon for lending (Armstrong, 1998; Common, 1998). To some extent, this variation reflects new possibilities to share experience beyond a country’s immediate border. Rose (1993: 98–102) links it to ‘propinquity’ (closeness) which could refer to one or more similarities in geography, policy conditions, ideology,

or problem definition (1993: 99–102; see Blum, 2014: 373 on Germany and Austria). Physical constraints to transfer have diminished, while the rapid growth of technology, communication, mobility, and external pressure prompt more transfer ‘across diverse countries’ (Dolowitz and Marsh, 2000: 7). 6. Sub-national governments. Multi-level governance (MLG; Chapter 8) and ‘multi-disciplinary’ studies suggest that the international export and import of policy can take place at local and regional government levels (McCann and Ward, 2012: 327; Benson and Jordan, 2011: 369). MLG may encourage more innovation as regions (or cities) compete with each other and have more scope to try out new ideas (Hooghe and Marks, 2003). It may produce more emulation as regions follow best practice, or the success of policies adopted by devolved governments puts pressure on the national government to follow (or vice versa). MLG may increase the demand for transfer since it introduces new audiences for ideas and new ways for epistemic communities or entrepreneurs to venue-shop. However, interdependence combined with uncooperative intergovernmental relations, may also undermine innovation, particularly when devolved government policy would undermine a national position (Keating and Cairney, 2012). Or, sub-national to sub-national transfer can be more ‘accidental’ and haphazard than based on ‘rational’ systematic learning from best practice (Dolowitz et al., 2012).

WHY TRANSFER? IS IT VOLUNTARY? Who decides if an importing country should transfer policy: the country itself or external agents? In many cases the answer is ‘both’ and the process contains a mix of voluntary and coercive elements. Dolowitz and Marsh (2000: 13) produce a policy transfer continuum (Figure 12.1) to describe such possibilities. At one end of the spectrum is voluntary transfer, which Dolowitz and Marsh (1996: 346) associate with lesson-drawing. Transfer may occur when policymakers express dissatisfaction with domestic

policy (1996: 347; Rose, 1993: 5) and ‘look abroad’ to see how others have addressed similar problems (Bennett, 1991a: 200). Or, they look abroad for evidence to legitimize their existing policies. In each case, there is no external pressure to examine policy change and minimal temptation to mimic another government to deal with bounded rationality (Dolowitz and Marsh, 2000: 22; Berry and Berry, 2007: 225). At the other end of the spectrum is direct coercive transfer. It refers to the influence exerted by an organization, or national or supranational government, to make another government adopt a policy. For Dolowitz and Marsh (1996: 348; 2000: 10–11), direct imposition is rare. Indeed, extreme examples such as regime change seem to be off the scale entirely (compare with Holzinger and Knill, 2005: 781). Rather, in most cases, there is some degree of (limited) choice within the importing country. Coercion may describe setting the agenda for transfer and the adverse consequences if the importing country does not import policy. The most significant example is conditionality: when ‘developing’ or low-income countries seek emergency loans from the International Monetary Fund (IMF) or World Bank, they are almost obliged to accept certain conditions, such as to make ‘good governance’ reforms (Chapter 8, p.134). NPM-style reforms and a reduction of the role of the state in the economy are also ‘encouraged’ (Biersteker, 1990; Hopkins, Powell, Roy and Gilbert, 1997). In turn, countries like the USA influence the ‘deregulation, liberalization and privatization’ agenda of the IMF or World Bank (Stone, 2017: 59; Wade, 2002). While ‘obligated transfer’ may seem more coercive than conditionality, the original decision to become subject to treaties is often voluntary, while many countries have the ability to influence international policies. Supranational organizations, created by governments to deal with interdependence, prompt governments to cooperate (Bennett, 1991a: 226). Similarly, while Dolowitz and Marsh (2000: 15) point out that the Court of Justice can ‘force member states to comply with European policy’, they ask: ‘since individual nations voluntarily joined the Union, can any act of the EU be considered coercive in terms of policy transfer’? Of course, the willingness and ability of member states to resist coercion can vary

dramatically, to include the UK’s decision to leave the EU (Brexit) and Greece’s relative powerlessness during its economic crisis (Blyth, 2013).

Figure 12.1 The Dolowitz and Marsh policy transfer continuum Source: modified from Dolowitz and Marsh (2000: 13).

The third main category – indirect coercive transfer – refers to voluntary transfer in which the importing country perceives the need to change. The policy agenda is driven elsewhere and there are consequences to non-emulation, but the push for change comes from the importing country. For Dolowitz and Marsh (1996: 348–49), it refers mostly to externalities, or the (often unintended) effects of the actions of one agent on another (Chapter 7). For example, countries react to a range of external pressures including norms (when countries adopt the same behaviour and expect others to follow suit), embarrassment (a feeling of being left behind), and economics (when countries react to the decisions made in competitor countries). Small states, sharing a border with larger states, are affected disproportionately (Hoberg, 1991). For example, Canada responds continuously to economic and regulatory decisions made in the USA (as its main market for many goods and services), while its attention to the US media and reliance on US expertise act as further sources of emulation (1991: 110). These effects can be seen in sub-national authorities, such as when UK government decisions prompt Scottish and Welsh government responses

(Cairney, 2008; Keating and Cairney, 2012; Keating et al., 2012; Cairney et al., 2019b).

Coercive Transfer: How Is It Manifest and Demonstrated? There is considerable debate regarding the means and strength of coercive transfer (Benson and Jordan, 2011: 370). For example, the use of conditionality to coerce developing countries can be an example of material or ideational power (Box 3.1, p.39). Cammack (2004) argues that the World Bank’s putative aims to reduce world poverty mask its main goal of capitalist hegemony (linked strongly to US attempts to dominate world politics – Wade, 2006). Wade (2002) argues that, while the USA may seek to influence directly the economic policies of other countries through the World Bank and IMF, its use of free-market ideas as part of a wider hegemonic project may be as important (but there are many ideas out there, competing for attention – Dassauge-Laguna, 2012a: 318). Stone (2004: 554) presents a similar discussion of knowledge and the dissemination of ‘best practice’ by the World Bank (see also Hwang and Song, 2019). Hopkins et al. (1997: 513) also question the assumption of direct and strong coercion, suggesting instead that the importing country would not otherwise have reformed without being bribed, induced, or rewarded (see also Meseguer, 2003: 24). Holzinger and Knill (2005: 781) caution against the assumption of high levels of coercion; there is a difference between a government being resistant to or uninterested in a policy transfer. Stone (2004: 554) suggests that since economic sanctions or inducements are often ineffective, it is difficult to identify successful coercion, particularly during implementation. While some countries must make certain choices, enforcement is less clear. There is some evidence of the ‘dismal’ implementation of agreements linked to conditionality (Meseguer, 2003: 24), particularly when there is no sense of ‘ownership’ of policy in the importing country (Hopkins et al., 1997: 512) or there are problems with the ‘transmission’ of key ideas (Stone, 2017: 60).

Similarly, studies debate the role of the EU, since member state approval is key to many EU obligations, making them ‘semi-coercive’ (Bulmer et al., 2007: 15 in Benson and Jordan, 2011: 370). The European Commission also has weak powers to implement in many policy areas. It may oblige the adoption of very broad laws that leave a lot of discretion in implementation (Duina, 1997; Mastenbroek, 2005; Falkner et al., 2004) or of administrative reforms that do not endure (Stone, 2017: 66).

What Is Transferred? What Makes Policy Transfer Distinctive? Page’s (2018: vii) reflection on the literature suggests that ‘what either looked like, or was claimed to be, a policy “borrowed” from another was, in fact, more significantly shaped by a range of other political and organizational constraints such that it was hard to identify what, precisely, was borrowed over and above terminology or the germ of an idea’. Policy transfer can range from the long-term decision to import completely the substantive aims and institutions associated with a major policy change, to the quick decision to pursue a vague idea (Dolowitz and Marsh, 1996: 350). Consequently, we need evidence of one government studying another and a way to show that it translated into concrete results. In other words, to measure the levels of transfer in much the same way that we measure policy change (Chapter 2, p.20). For example, Dolowitz and Marsh (1996: 349–50) identify seven measures: ‘policy goals, structure and content; policy instruments or administrative techniques; institutions; ideology; ideas; attitudes and concepts; and negative lessons’ (compare with Rose, 2005: 22; Bennett, 1991a). We can use such measures to identify variations in borrowing. Rose (1993: 30–32) provides five main categories along a continuum. At one end is complete duplication, possible within political systems with ‘close similarity of institutions and laws’, and possible to detect in the identical wording in legislation. More likely strategies are adaptation, when an importing country tailors the imported policy to domestic circumstances; making a hybrid policy

from the exporting and importing region; or creating synthesis by extracting aspects of one or more exporting programmes (such as when new democracies borrow elements of electoral systems and legislative models – Norris, 1997). Or, policymakers use lessons from others as inspiration, as the ‘transfer of policy knowledge but not a transfer of policy practice’ (Stone, 2004: 549). Other categories may be relevant, such as the use of information to package decisions already made (Dolowitz, 2003: 103). Or, the same programme may be repackaged by a government with a different ideological position (such as ‘Patient Choice’ in England and Wales or the adoption of ‘Reaganite’ ideas by a left-wing New Zealand government – Clancy, 1996; Cairney, 2009a). Such variations reflect the fact that policy transfer is a political process embedded in domestic policymaking. For example, Page (2000: 4) draws on agenda setting concepts to describe the supply of ideas: ‘the perception of how the policy operated in the exporter jurisdiction might be crucially shaped by the observer from outside who seeks to import it’. The demand for ideas is also dependant on the power dynamics described in several policy theories (Box 12.4). Punctuated equilibrium theory (PET; Chapter 9) identifies positive and negative feedback, in which there is disproportionately low or high attention to new ideas. Multiple streams analysis (MSA; Chapter 11) suggests that major policy transfer will not take place unless during a window of opportunity. The advocacy coalition framework (ACF; Chapter 10) suggests that the pressure to change policy comes from competition between domestic coalitions rather than coercion from another country. Yet most of these theories of framing and idea-adoption developed from studies of the USA, which tends to innovate rather than emulate and is less subject to pressure than most other countries (although pressure is apparent at sub-national levels) (Stone, 2017: 57). What happens if we shift our attention to countries more subject to external influence? Direct or indirect coercion is a stronger influence on the supply of and demand for new ideas. Levels of ‘positive feedback’, ‘receptivity’, and coalition competition may be less important than the pressure exerted from outside the system. External activity is a key factor in the supply and demand for new ideas. This external

dynamic makes the study of policy transfer more than simply the study of domestic policy processes (compare with James and Lodge, 2003; Evans and Davies, 1999; Holzinger and Knill, 2005: 775). Box 12.4 The policy transfer window Chapter 11 describes (1) the profound effect of ideas but (2) their conditional acceptance within political communities. We may treat ideas as ‘viruses’ but also consider the role of the ‘host’ and the ability of policymakers to resist ‘infection’. Our focus moves from the power of ideas themselves to their successful promotion and receptivity to them. Major policy change may only occur when established ideas – as paradigms, belief systems, monopolies, or institutions – are challenged. In some cases, change follows an event. Policy failure may prompt a shift in power when existing policymakers or experts fall out of favour. ‘Shocks’ may prompt dominant advocacy coalitions to revisit their beliefs or allow other coalitions to exploit the opportunity to gain more favour within subsystems. An established policy monopoly may break down completely when pressure participants, dissatisfied with current policy ideas in one venue, attract external attention from powerful actors more willing to consider radically new solutions. Or, institutional change may be gradual, as policymakers reinterpret policy problems and challenge current solutions. These discussions suggest that quick and radical policy transfers based on new ideas are unusual. The ‘policy transfer window’ represents one way to combine a focus on policy transfer and the adoption of ideas within political systems at a particular time (Cairney, 2012a: 269–71; see also Dussauge-Laguna, 2012b: 573; 581). It is based on Kingdon’s (1984) identification of the role of problems (how is the policy problem defined?), policies (which feasible solutions are available?), and politics (how receptive are policymakers to these solutions?) In Kingdon’s model, although multiple streams come together to produce change, policymaker receptivity seems to be the most important factor. In the policy transfer window, we consider the power of an external actor to oblige importing

governments to consider particular solutions. The window of opportunity now depends on the ability of external actors to set the agenda, the applicability of their ideas, and a combination of (1) the level of domestic actor receptivity to those ideas and (2) their ability to resist transfer. The window may still open rarely and with unpredictable results, but the process is not influenced solely by domestic actors (see, for example, Cairney et al., 2012).

WHAT DOES ‘SUCCESSFUL’ POLICY TRANSFER MEAN? Rose (1993: 132) describes factors that enhance the successful completion of transfer: 1. The policy is not unique or dependent on inimitable organizations. 2. There are few resource constraints to implement policy. 3. The policy is simple with a clear cause and effect. 4. There is adequate information available about what the policy is and how it works. 5. The new policy does not mark a radical departure from the importer’s original policy. This list relates strongly to the factors central to policy implementation (Chapter 2, pp.28–9). Two of the top-down conditions for successful implementation are that there is an understanding of clear and consistent policy objectives and that the policy will work as intended when implemented. Additionally, in the case of transfer, the importer must ask if the original policy worked as intended because of the wider political system, values, or administrative arrangements in the exporting country (Dolowitz, 2003: 106; Page, 2000: 10; Massey, 2010). We therefore have two separate but related elements: transfer success and implementation success. For Dolowitz and Marsh (2000: 17), transfer success refers to the extent to which a policy was adapted properly. In turn, failure can relate to one, or a combination, of three factors:

1. Uninformed transfer – when the borrowing country has incomplete information about the crucial elements that made the policy a success in the lending country. 2. Incomplete transfer – when those elements are not transferred. 3. Inappropriate transfer – when not enough attention is paid to adapting to the differences in policy conditions, structures, and aims of the lending country. To demonstrate, let us compare an example of unsuccessful and successful transfer. Dolowitz’s (2003) case study of the Child Support Agency in the UK identifies adaptation problems caused by different aims. In the UK, the driver was to raise revenue (focusing on a small number of absent parents) but policy in the USA was geared towards wider coverage regardless of ability to pay (see also Sharman, 2010). It compares badly to Dolowitz and Marsh’s (2000: 16) claim that the transfer of the US Earned Income Tax Credit to the UK (Working Families Tax Credit) was successful because the UK government had the time and resources to study the issue well. Yet there are three key qualifications. First, much of our earlier discussion suggests that it is difficult to think of transfer in terms of a successful exporter or motivated importer of a specific programme (Stone, 2017: 55; compare with Evans and Barakat, 2012). Rather, governments often take broad inspiration or translate policy into something different. This process prompts us to focus less on transfer as the export of a well-packaged product and more on the ways in which importing policymakers make sense of policy in local contexts (Stone, 2017: 64–65). Our understanding shifts from ‘transferred’ to ‘translated’ (Freeman, 2009; Stone, 2012; Park et al., 2014), ‘transformed’, or ‘mutated’ (McCann and Ward, 2012: 326–27; 2013: 10). Second, transfer success does not mean that the policy will be an implementation success. As Chapter 2 suggests, it depends on wider issues of implementation, such as compliance, support from interest groups, and the effects of wider socioeconomic factors. Third, wider evaluations of policy success also incorporate political assessments, including was the process legitimate and straightforward, and did the policy transfer improve the popularity of government (McConnell, 2010; Compton and ‘t Hart, 2019)?

HOW TO ENCOURAGE ‘EVIDENCE-BASED’ POLICY LEARNING AND TRANSFER Cairney (2018c; 2019a) explains why policymakers might not pay attention to scientific evidence in ways that scientists might expect. Although many scientists express a fear that they have entered an age of ‘post-truth politics’ in which people do not trust evidence or experts (Jasanoff and Simmet, 2017), it is more useful to identify the ever-present features of policymaking systems that influence evidence use. In turn, these factors help us understand the – often limited – role of academic research in policy learning and transfer. 1. Policymakers have different ideas about what counts as good evidence Few policymakers use the criteria developed by some scientists to describe a hierarchy of scientific evidence. For some scientists, at the top of this hierarchy is the systematic review of randomized control trials (RCTs), because RCTs often provide a way to isolate and measure the effect of a policy intervention (Petticrew and Roberts, 2006; Dunlop, 2014: 213). Other forms of knowledge – derived from scientific expertise, practitioner experience, and service user feedback – are nearer the bottom (Oliver et al., 2014a; 2014b; Oliver and Pearce, 2017; Parkhurst, 2017; Cairney, 2016c). Most policymakers, and many academics, prefer a wider range of information sources (Nutley et al., 2013; Boaz et al., 2019). They combine their own experience with information that may include peer reviewed scientific evidence, the ‘grey’ literature (not-peerreviewed by academics), public opinion data, and feedback from consultation (Nilsson et al., 2008; Lomas and Brown, 2009; Cairney and Oliver, 2017; Davidson, 2017; Kislov et al., 2019; Craig, 2019). Further, there are many policymakers spread across many levels and types of government, causing the same evidence to have different impacts in different parts of government (Cairney, 2016a).

2. Policymakers have to ignore almost all evidence Many scientists describe their hopes for EBPM in a way that is reminiscent of ‘comprehensive’ rationality (Cairney and Oliver, 2017). Yet comprehensive is an ideal-type, used to describe what cannot happen (Chapter 4). Instead, policymakers must use heuristics to prioritize evidence and free up time to make choices (Cairney and Kwiatkowski, 2017). In that context, a call for EBPM is naïve (Botterill and Hindmoor, 2012) or a way to prioritize scientific evidence and curtail debate (Boswell, 2018), rather than a realistic description of policymaking. 3. Policymakers do not control the policy process (in the way that a policy cycle suggests) Many scientists like the idea of the policy cycle because it offers a simple and appealing model, giving clear advice on how to engage at each stage (Oliver et al., 2014b). However, the cycle provides misleading practical advice: researchers will generally not find an orderly process with a clearly defined debate on problem definition, a single moment of authoritative choice, and a clear chance to use scientific evidence to evaluate policy (Chapter 2). Rather, they will engage with a complex policymaking system of which policymakers have limited knowledge and even less control (Chapter 6). All three reasons help us understand policy learning and transfer (Cairney, 2018d). Policymakers will draw on an eclectic mix of sources to generate evidence of another government’s policy success. This generation of lessons is political: actors exercise power to define policy problems and therefore determine the policyrelevance of evidence to produce and evaluate solutions. Bounded rationality suggests that policymakers may import policy solutions without knowing if, and why, they were successful. Further, policy transfer is not separate from the policy process; evidence of international experience competes with many other sources of ideas. Therefore, it is difficult to imagine how to understand policy transfer as a ‘rationalist’ project (Marsh and Evans, 2012: 589) even if some governments describe it as such (Legrand, 2012b). Instead,

the literature reminds us that policy learning and transfer are not ‘evidence-based’ and may not even be voluntary. For example, complete policy duplication can follow: 1. key scientific principles, such as the need for ‘fidelity’ to specific models whose evidence of effectiveness comes from RCTs (Cairney, 2017b), or 2. coercion, hierarchy, or unreflective mimicry (Dunlop and Radaelli, 2018). Epistemic forms of learning may often seem desirable, but prove to be less realistic than bargaining and no more defendable than reflexive (Dunlop and Radaelli, 2018: 261). Drawing evidence from the experience of others can be useful, but only as part of a wider process in which they learn from their own experience in their own context. Therefore, if we look for advice on lesson-drawing – such as in Rose’s ten steps in Box 12.5 – we need to ask more questions to inform ‘how to learn and transfer’ guides. They relate strongly to the three reasons not to expect EBPM: (1) the evidence for success, (2) the story policymakers tell, and (3) how conducive to comparison are the policymaking systems of each government. Box 12.5 Richard Rose’s ‘ten steps in lesson-drawing’ 1. Learn Rose’s language of lesson-drawing, including what ‘programme’ and ‘lesson’ mean 2. ‘Catch the attention of policymakers’ 3. Do some preliminary work to identify ‘where to look for lessons’ 4. ‘Learn by going abroad’ 5. Produce a model to describe how and why a programme works 6. ‘Turn the model into a lesson fitting your own national context’ 7. ‘Decide whether the lesson should be adopted’ 8. ‘Decide whether the lesson can be applied’ 9. Simplify the lesson and its application to increase its chance of success

10. Evaluate the success of your lesson-drawing process. Source: adapted from Rose (2005).

What is the evidence for success, and from where does it come? In general, policy actors debate what counts as good evidence, draw on an eclectic mix of evidence sources, and use evidence to support their beliefs about the value of a policy choice. The transfer process may exacerbate these issues in several ways. First, the exporting government and supporters of its policy have an incentive to declare that there is evidence of their success (Cairney, 2019d). Therefore, importing governments should rely on a combination of sources, including government-commissioned but relatively independent evaluations, and independent scientific reviews subject to peer review. These evaluations should use direct or counterfactual comparisons to ask if another solution would have been more successful. Second, it is not clear how long to wait to declare success. We may be identifying ‘Good practice’ based on positive experience, ‘Promising approaches’ but unsystematic findings, ‘Research-based’ or ‘sound theory informed by a growing body of empirical research’, or ‘Evidence-based’, when ‘the programme or practice has been rigorously evaluated and has consistently been shown to work’ (Perkins, 2010, cited in Nutley et al., 2013: 9) Third, the research methods we use to generate evidence relate strongly to our beliefs about the best way to deliberate, govern, and ‘scale up’ policy success (see Cairney, 2017b; 2018d; Cairney and Oliver, 2017; Cairney et al., 2017). Consider two very different approaches. First, we could identify the same basic model of success and transfer it uniformly across the national population. Centralization and hierarchy would be conducive to the rollout of uniform policy interventions, driven by evidence from RCTs. It might also help avoid a ‘postcode lottery’ of service provision (in which where you live seems to determine how good an ostensibly uniform service is). Second, we could tell a richly descriptive story about a model and invite local policymakers and stakeholders to deliberate and adapt policy to their circumstances. The routine delegation of

policy to local communities, service users, and practitioners would be conducive to sharing evidence personally via storytelling. It would not satisfy advocates of a ‘hierarchy’ of evidence. However, it would allow central governments to recognize a local democratic mandate, and admit that successful policy change in one context does not mean that the same solution will be successful in another. What story do policymakers tell about the problem they are trying to solve, the solutions they produced, and why? We need to know more about the agenda setting process (Chapter 9) that led policymakers to narrow their range of choices, and the ways in which new policy instruments related to existing policies. This step is crucial to comparability, since the exporting/importing government may have very different ways to (1) understand issues as problems, (2) identify feasible solutions, and (3) determine what policy success looks like. As Box 12.4 suggests, a window of opportunity for policy transfer opens only for technically and politically feasible solutions. Was the policy introduced in a comparable policymaking system? Issues of comparability often relate to factors such as the size and type of political systems. For example, it may be relatively easy for two US states to share lessons since their overall federal context is similar, but difficult for Scotland and a US state to share lessons since they are part of different national and supra-national arrangements. However, the focus on this book is the need to conceptualize and compare policymaking systems. For example, we can use the constituent parts of policymaking environments to think systematically about elements of comparability: • Actors. Compare the spread of policymakers and influencers across many venues: which levels or types of government are responsible for this policy? • Institutions. Compare the formal and informal rules of those venues: which rules matter? • Networks. Compare the relationships between policymakers and influencers: what role do these networks play in making and delivering this policy?

•

•

Ideas. Compare the paradigms within which new solutions are considered: do policymakers in the exporting and importing governments have the same fundamental beliefs about the nature of a problem and its feasible solution? Socioeconomic context and events. Compare the drivers for policy change: are both governments responding to crisis or routine events? Do they share similar reference points, including economic performance or social attitudes and behaviour?

The overall sense that we get from these questions is that evidence is an important element in the learning process but one of many factors to consider. Policy transfer may be evidence-informed but it is difficult to see how evidence alone could be the deciding factor in policy transfer (indeed, we are using these insights to encourage realistic policy learning across the EU – see Cairney et al., 2018c).

CONCLUSION Policy learning sounds like a ‘rational’ and technical process in which actors acquire new policy-relevant knowledge and skills. However, it describes a collection of political processes in which actors exercise power to learn how to get what they want. The process is akin to the ways in which actors address bounded rationality (Chapter 4). They use cognitive shortcuts to filter information and exercise power, to (1) reduce uncertainty by translating evidence and (2) reduce ambiguity by influencing how other actors use this knowledge to interpret policy problems. Learning is an individual and collective process: cognitive shortcuts matter but so do the rules and systems in which actors interact (Heikkila and Gerlak, 2013). Consequently, we can think of learning in terms of very different categories, as driven by expertise, dialogue, bargaining, and/or hierarchy (Dunlop and Radaelli, 2013; 2018). This wide-ranging definition of policy learning informs the study of policy transfer. The search for new knowledge, from the experience of other governments, is part of a highly political learning process. Policy transfer describes the international exchange of ideas, which can prompt governments to import a policy intervention or

programme entirely or, more likely, take broad inspiration or adapt lessons to local contexts. Key drivers can include routine searches for evidence, coercion, or a perceived need to keep up with the norms set by other governments and international organizations. We could focus on supply, as the ways in which actors try to ensure the widespread diffusion of policy solutions, or demand, as the ways in which actors become more or less receptive to new ideas and transform them into programmes specific to their political systems. We can add insights from learning and transfer to existing theories of policymaking. Implementation failure can derive from misunderstandings of international policy experiences. The external pressure to act may further exacerbate bounded rationality, and emulation may produce non-incremental policy change. A focus on coercion may inform formerly US-centric models of policy change – including PET, MSA, and the ACF – which initially ignored the kinds of political constraints that are more likely to be experienced outside the USA. These insights also help warn us against the assumption that policymaking, learning, and transfer can be ‘evidence-based’. Rather, scientific evidence is one of many sources of information for policymakers, and information is one of many factors crucial to explanation. To know how evidence-informed policy and policymaking are, we need to know more about the shortcuts to action that policymakers use, and the complex policymaking environment in which they operate. Only in that context can we produce a realistic ‘how to do it’ guide which encourages us to be pragmatic about the highly political ways in which governments ‘learn’ from one another. Hegemonic project – the pursuit of dominance in terms of the way we think about and address policy problems.

13 Conclusion: Policy Theory as Accumulated Wisdom Key themes of this chapter: • • •

• • • •

•

We can combine many insights to tell an overall story of policy and policymaking, but should proceed with caution. Policymakers combine cognition and emotion to make choices, in a complex policymaking environment of which they have limited knowledge and control. The result is frequent policymaking stability and infrequent instability. Most policy change is minor. Major policy change is unusual. It takes place over decades or, in rare cases, rapid change follows a major ‘shock’, punctuation, or window of opportunity. Highly abstract concepts apply universally. All actors use heuristics to address bounded rationality. All policymaking environments are complex. We can apply this accumulated wisdom to key issues, such as the vague meaning and impossibility of ‘evidence-based policymaking’. Yet people tell policy stories in profoundly different ways, applying universal concepts to very different contexts, often using a different conceptual language. Some provide complementary stories, akin to pieces of the same jigsaw. Others tell stories that contradict each other, prompting us to choose what to accept and question the extent to which we are accumulating knowledge. These stories come from theories and studies of the Global North (or the ‘West’), with the Global South often ignored or interpreted through an external lens. We should challenge this narrow perspective of policy processes.

INTRODUCTION: COMBINING THEORETICAL INSIGHTS IS USEFUL BUT TRICKY Wouldn’t it be nice if we could combine the insights from the theories and concepts in this book to accumulate knowledge of policymaking and make the whole greater than the sum of its parts? We have good reasons to be optimistic. Although there is a proliferation of approaches and concepts, they often display an impressive degree of unity and a common research agenda. Most theories of public policy identify the same key empirical themes and conceptual issues. They ask comparable questions and identify similar causal processes. At a conceptual level, most identify bounded rationality and the interplay between individuals and their policymaking environment (Box 13.3). Most consider the interplay between the exercise of power and the role of ideas, and seek to theorize the relationship between sources of stability or instability and policy continuity or change. At the empirical level, many theories consider the pervasiveness of specialization and policy subsystems, and seek to account for a shift from the centralization of power towards a more competitive group–government process and multi-centric policymaking (Cairney et al., 2019a). My aim in this chapter is to identify ways in which we can bring together these insights. First, I explore a profoundly important relationship between the analysis of policy change and policymaking. It seems impossible to assess the explanatory power of policy process theories without first defining and measuring policy change. However, it also seems impossible to define policy – and therefore measure policy change – without reference to policy process theories. Our response should be a combination of art and science: tell stories of policy change and compare the ways in which theories help identify and explain their causes (Box 13.1). Second, I tell a story of the policy process built on common insights about the ways in which actors address bounded rationality within a complex policymaking system. At this general level of theoretical analysis, we can focus on the richness of accounts, in which many different theories provide ways to understand the

implications of policymaker psychology and aggregate activity in policymaking systems. Box 13.2 also shows how policy process insights can inform practical policy analysis. Third, I apply this story to the myth of ‘evidence-based policymaking’. This theme is particularly important as a way to describe the interdisciplinary value of policy studies. There is a lot of confusion about the relationship between evidence and policy, and policy studies can help, but not everyone has the time to train as a policy theorist. Rather, our aim is often to synthesize disciplinary insights in a concise way, and show how they inform wider debates. In this case, policy studies helps enthusiasts of research evidence understand and navigate policy processes. Fourth, I warn against combining theories uncritically. Combining the merits of theories represents ‘standard advice’ in the literature (Sabatier, 2007a: 330). Yet it is less clear about how. We can produce a single theory that combines key insights from a range of others. We can maintain a separation between theories to provide multiple, complementary, perspectives. Or, we can pit competing theories against each other to see which provides better explanations. Each approach is problematic for different reasons, but they all provide cautionary tales about simple combinations of theories. Fifth, I explore the tendency of policy theories and studies to derive from, and apply to, a small number of Global North or ‘Western’ countries (usually the USA and Western Europe). Highly abstract concepts such as bounded rationality and institutions apply universally. However, people make sense of the world in very different ways and the rules of their political systems vary too much to expect a theory built on one part of the world to apply to it all. There is a promising increase in international applications, and some studies show that combining case studies is as fruitful as combining theories to generate a wider perspective. However, the Global South is still relatively ignored, or interpreted through a lens created in the North. This limitation reminds us that the power we identify in political systems also exists among the people who describe them (Chapter 3, p.50). Some voices are heard, and many are ignored. Policy theories endure because they are remarkably valuable, but it

is worth reflecting on who values them and why, how truly universal any policy theory can be, and the steps we would need to take to amplify more voices in the study of public policy.

DEFINING POLICY AND TELLING A STORY OF POLICY CHANGE We can combine the theories, concepts, and insights, described in this book, to tell an overall story of policy and policymaking. It begins with definitions of policy and measures of policy change (Chapter 2). When we try to define policy – as, for example, the sum total of government action, from signals of intent to the final outcomes – we quickly move into a discussion of policymaking processes. ‘Government action’ could include what policymakers say they will do, what they actually do, or the outcomes. This expansive definition may prompt us to focus on policy change over a whole ‘cycle’ before we come to firm conclusions. Or, if the policy cycle does not describe choices and events well, we need another way to connect choices to long-term outcomes. A focus on ‘the government’ leads us to the role of elected and unelected policymakers and influencers, prompting us to consider the extent to which policymaking takes place in political systems or subsystems, and a single central government or multicentric arrangements. Famously, public policy also includes what policymakers do not do, reminding us that (1) policymaking is about power, often exercised to keep important issues off the policy agenda, and (2) it is often more difficult to measure than conceptualize policy change. If we describe policy as the tangible product of authoritative choice, we need to describe the process in which choice takes place, and the extent to which a policymaking system helps translate stated intent into actual outcomes. An uncritical analysis of the policy cycle might give the impression of policy formulation followed inevitably by implementation and evaluation. However, most policy studies invite us to pause to consider (1) if a single choice at one ‘stage’ of the process constitutes ‘policy’ and (2) if that choice leads to the intended policy change.

These conceptual issues inform the ways in which we measure and evaluate policy change. A focus on policy tools and instruments shows us that some policies may begin with a broad statement of intent but they continue with more detailed strategies and outputs that contribute to more or less successful outcomes: 1. Governments already combine a large number of instruments to make policy, including legislation, expenditure, economic incentives and penalties, education, and various forms of service delivery. Each instrument adds to a large policy pile, and it is rare to introduce an entirely new policy measure. Further, many instruments provide incentives rather than obligations, which means that policymakers accept that policy outcomes are not entirely in their gift. 2. Key categories of policy tools highlight the relative political costs of action. To try to persuade, or simply share information, may be less costly than direct imposition via regulation. To redistribute (visibly) some resources from one group to another seems more difficult than to distribute resources with less of a clear sense of who wins at the others’ expense. The choice of tool is also a policy choice, and patterns of policy tool use help us build a picture of ‘policy styles’ (Howlett and Tosun, 2019; Greer and Jarman, 2008). To some extent, there is a technical or scientific dimension to measuring policy change, as well as quite high consensus about its overall nature. In other words, a key theme of the book is that policy studies describe most policy change as minor, coupled with infrequent major policy changes taking place over decades, and very unusual cases of rapid change following a major ‘shock’, punctuation, or window of opportunity. However, it would be a mistake to portray measurement as objective. It is necessarily selective (we cannot study everything) and evaluative (we make a judgement about the meaning of the results). In politics, actors such as political parties pay attention to evidence selectively, and compete to evaluate policy styles and declare policy success or failure. In research, we might be less partisan, but we still

face important choices about which measures to select, and how we tell a story of policy change. Key factors include the time frame we choose, the levels of government on which we focus, how much policy change we expect given the nature of the problem, the motive we ascribe to policymakers, and perhaps a tendency of researchers to focus on exciting but unrepresentative case studies of major change (Box 13.1). Box 13.1 What is policy, how much has it changed, and why? A common format for writing academic studies is theory, method, results. In other words, we might establish a theory to apply, select a method to do so, and explain clearly the new knowledge that arose from this process. However, a theory’s explanatory value ‘may depend on the interpretation of events, or the narrative of policy change, that we select, rather than just the nature of the theory’ (Cairney, 2013a: 13). In other words, our initial account of policy change matters, particularly (1) if there is room for debate on the size (minor or major), speed (over days or decades), and source (such one or more venues) of policy change and (2) because some theories are more useful in explaining long periods of continuity rather than rapid change. Adapting a theory–method– results approach is as much art as science (Cairney, 2017c): 1. Set a boundary on your chosen policy. For example, a focus on health policy is very broad and difficult to limit, particularly if the Health in All Policies agenda suggests that major influences on outcomes do not come from health services (de Leeuw, 2017; Shankardass et al., 2018). Policy to reduce population-wide smoking is more specific, but there are many relevant policy instruments in areas such as taxation, trade, and education (Mamudu et al., 2015: 860). 2. Identify the choices, instruments, and outcomes that are key to your measurement of change. A classic mistake is to equate policy with a strategy document that does little more than set the agenda or satisfy calls for action, often with no expectation of follow-through. On the other hand, it is not possible to

measure the entirety of policy outcomes or contribution of policy outputs to them. Most studies are pragmatic: they establish a small number of key actions and their effects, and situate this analysis within a wider context. For example, we could zoom in to provide a forensic study of changes to one tobacco policy instrument in one US state (Weible and Carter, 2015) or zoom out to study the cumulative effect of a global commitment to ‘comprehensive’ policy change (Cairney et al., 2012). 3. Connect your story of policy change to your policy theory story. The ‘empirical evidence is wide open to interpretation, and we can often produce competing, and equally convincing, empirical narratives based on solid evidence’ (Cairney, 2013a: 12). One tentative solution is to compare many narratives and theories, but this approach may raise as many problems as it solves (2013a: 15–18).

A THEORY-BASED STORY OF POLICYMAKING The first part of our universal story is that policymakers, like all humans, combine cognition and emotion to make choices. They use two cognitive processes to deal with information: ‘System 1 operates automatically and quickly, with little or no effort and no sense of voluntary control. System 2 allocates attention to the effortful mental activities that demand it, including complex computations’ (Kahneman, 2012: 20). This statement is based on a truism associated with bounded rationality: no individual has the ability to pay attention to, and understand, all information relevant to their choices. Rather, they must use heuristics or implicit and explicit cognitive rules to prioritize some information at the expense of the rest (Chapter 4). All relevant policy theories begin with some version of this story, but they tell it in different ways (see Box 4.1, p.60). For example, they provide a positive or negative spin . The language of ‘cognitive biases’ describes the ways in which humans make suboptimal decisions (some of which might be remedied through training or

organizational responses). For example, individuals are vulnerable to misjudge the size of a problem when they base their judgement on a vivid but unrepresentative event. They misjudge the motives of a social group based on a person that they take to represent many people. They may pay more attention to information to which they can easily relate to their current knowledge. They engage with a story designed to manipulate their emotions. People also tend towards ‘groupthink’ and other biases that limit their searches for information. A more positive spin is to describe ‘fast and frugal heuristics’ and marvel about the human ability to make almost instant, efficient, but still useful choices based on emotion, values, and simple strategies such as trial and error (Gigerenzer, 2001: 37–38). Theories also describe different ways in which responses to bounded rationality affect policy-making behaviour: •

• • •

•

•

Policymakers can only pay attention to a tiny proportion of their responsibilities, and policymaking organizations struggle to process all policy-relevant information. They prioritize some issues and information and ignore the rest (Chapter 9). Some ways of understanding and describing the world dominate policy debate, helping some actors and marginalizing others (Chapters 3 and 11). Policy actors see the world through the lens of their beliefs. Beliefs allow them to select and interpret policy-relevant information and decide who to trust (Chapter 10). Actors engage in ‘trial-and-error strategies’ or use their ‘social tribal instincts’ to rely on ‘different decision heuristics to deal with uncertain and dynamic environments’ (Cairney, 2013b: 279; Lubell, 2013: 544; Chapters 6, 7, and 11). Policy audiences are vulnerable to manipulation when they rely on other actors to help them understand the world. Actors tell simple stories to persuade their audience to see a policy problem and its solution in a particular way (Chapter 4). Policymakers draw on quick emotional judgements, and social stereotypes, to propose benefits to some target populations and punishments for others (Chapter 4).

•

•

Institutions include formal rules but also the informal understandings that ‘exist in the minds of the participants and sometimes are shared as implicit knowledge rather than in an explicit and written form’ (Ostrom, 2007: 23; Chapter 5). Policy learning is a political process in which actors engage selectively with information, not a rational search for truth (Chapter 12).

Figure 13.1 An image of the policy process Source: Cairney (2018b). Compare with the more complicated Figure 2.2 on p.35.

The second part of our universal story is that people respond to bounded rationality within complex policymaking environments. We can describe this environment with reference to five or six constituent parts (John, 2003: 495; Heikkila and Cairney, 2018). First, there are many actors – including policymakers and influencers – spread across many types of policymaking venues. Second, each venue contains its own ‘institutions’, or formal and informal rules governing behaviour. Third, each venue can produce its own networks of policymakers and influencers, and the lines between

formal responsibility and informal influence are blurry. Fourth, actors in each venue draw on a dominant set of ideas or beliefs about the nature of policy problems and the acceptable range of solutions. Fifth, natural, social, and economic factors limit policymakers’ abilities to address and solve policy problems. Finally, routine and non-routine events help set the policy agenda and influence the resources available to actors. Combined, these factors produce the broad sense that policymaking environments – or, in some accounts, ‘context’ or ‘systems’ – constrain and facilitate action and are out of the control (or even understanding) of individual actors. Figure 13.1 provides the simplest way to visualize these concepts, partly to compete with the visual simplicity of the policy cycle while maintaining the assumption of complexity. The classic way to combine the study of bounded rationality and policymaking systems is incrementalism, which describes pragmatic ways to (1) gather and use information, (2) engage in strategic analysis, and (3) negotiate political settlements that do not depart radically from the status quo (Chapter 4). It suggests that policy change is incremental for good reason: people see the benefits of only studying in depth, or experimenting with, the changes that would be technically feasible, in relation to available resources, and politically feasible, in relation to current policy and the balance of power. However, incrementalism does not capture well the tendency of policymaking systems to produce stability and instability, or policy continuity and major and often intense periods of change. Instead, a suite of modern policy theories helps explain this dual dynamic: •

•

Punctuated equilibrium theory (PET) provides the most convincing quantitative analysis of the combination of ‘hyperincrementalism’ and punctuated policy change. Initial case studies explained policy continuity and change in relation to the creation, maintenance, and destruction of policy monopolies. Modern studies highlight the role of ‘disproportionate information processing’ and institutional ‘friction’, producing minimal attention to most issues but maximal attention to some (Chapter 9). New institutionalism traditionally used the language of punctuated equilibrium to describe the long-term endurance of rules and

•

•

• •

•

ways of thinking, disrupted by critical junctures prompting new approaches. Studies now identify the conditions under which less dramatic and more gradual long-term changes have a transformative effect (Chapters 5 and 11). Multiple streams analysis (MSA) describes the general absence of major policy change unless key actors exploit successfully a ‘window of opportunity’ in which attention rises to a problem, a feasible solution already exists, and policymaker have the motive and opportunity to select it (Chapter 11). The advocacy coalition framework (ACF) associates long periods of subsystem continuity with the stability of factors external to the subsystem, as well as dominance by one coalition or stable agreement between coalitions. Instability and major change comes from ‘shocks’, in which new information (e.g. about policy failure) or external events prompt (1) actors in one coalition to question their beliefs or membership or (2) actors in another coalition to exploit new opportunities (Chapter 10). Policy narratives can help reinforce the inequalities that obstruct policy change or provide a powerful new motivation for people to act on their beliefs to challenge the status quo (Chapter 4). In complex systems, policy outcomes appear to emerge from the interactions between many actors using local rules, in the general absence of central government control (Chapter 6). Or, multicentric policymaking systems provide a combination of many ways to obstruct or propose new policies (Chapters 7 and 8; Cairney et al., 2019a). Non-incremental change can happen when one government emulates another, such as when it assumes that it was successful, or feels internal or external pressure to keep up with international norms (Chapter 12).

These insights can be used in multiple ways to generate more specific stories of policy processes. First, we can try to combine insights from many theories in one story (see page xv to understand the initials, and please tweet me @Cairneypaul to tell me which 4letter abbreviation is missing):

actors form coalitions to cooperate with each other and compete with their opponents (ACF); they exploit cultural stereotypes and cognitive biases to tell stories with heroes and a policy moral (NPF); the policy system dampens the effect of most stories and amplifies some (PET); the small number of amplified issues prompt policy change during a window of opportunity (MSF); and subsequent policies create feedback, or the rules that constrain and facilitate future coalition activity (PFT). (Heikkila and Cairney, 2018: 319–20) Second, we can use a specific lens, such as feminist theory and research, to inform narratives of inequality in policymaking. For example, we should not assume that certain policy problems receive government or scholarly attention in proportion to their ‘objective’ importance (Chapters 2 and 9). Rather, the low proportion of women in positions of elected and unelected influence contributes to the often low attention given to feminist issues and feminist research (Chapter 3, p.50). Institutions contribute to such marginalization, from the path-dependent effects of male incumbency to the norms and social stereotypes influencing the incentives (the higher costs and lower benefits) to run for office or seek leadership positions (Chapter 5, pp. 88–9). Further, the ‘velvet triangle’ describes networks of feminist actors that develop partly because women are excluded routinely from positions of power (Woodward, 2004; Chapter 9, p.152). Third, we can use these insights to inform policy analysis (Box 13.2). Box 13.2 Theory-informed policy analysis and the ‘practical lessons’ agenda The ‘new policy sciences’ describes an update to Harold Lasswell’s promotion of theory-informed policy analysis (Cairney and Weible, 2017: 620; Cairney, 2017c). In other words, let’s take what we know – about policymaker psychology and complex policy-making environments – to supplement the ways in which we analyse real-world policy problems (compare with Bardach and Patashnik, 2015). The new policy sciences would:

1. Expand our focus to more actors. Classic analyses see the world through the eyes of a small number of policymakers at the ‘centre’ (Chapter 2). New analyses identify many policymakers and influencers spread across multiple ‘centres’ in a system, to establish where the ‘action’ is really taking place. 2. Incorporate a focus on policymaker psychology. Classic analyses identify the costs and benefits of action in relation to ‘rational’ policymaking (Chapter 4). New studies recognize the profound importance of ‘disproportionate information processing’ (Chapter 9; Baumgartner et al., 2018). They frame issues and choices in relation to the ways in which (1) policymakers combine cognition and emotion and (2) policymaking organizations produce standard operating procedures to limit their search for information (Cairney and Kwiatkowski, 2017). 3. Incorporate the effect of a ‘large and messy policy environment’. Classic analysis involves identifying problems and generating solutions at the start of a policy cycle. New studies anticipate a more complex context for policymakers. This includes ‘the formal and informal institutions that guide their actions at each level; the networks they form; the ideas or beliefs that dominate the ways in which they think about problems and solutions; the biophysical and socio-economic conditions to which they respond; and the events that offer opportunities and shocks’ (Cairney and Weible, 2017: 624). This agenda is part of a larger effort to produce ‘practical lessons’ from a policy theory field that has become highly specialized and disconnected from policy analysis: ‘the field is vast and uncoordinated, and too many scholars hide behind a veil of jargon and obfuscation’ (Weible and Cairney, 2018). Simplification is not about ‘dumbing down’. On the contrary, clear and concise explanation has scientific and practical benefits. It allows us to be ‘clear enough to invite critical analysis’ and check ‘to see if the conceptual logic still holds and the empirical analysis is rewarding’ (2018: 184; compare with Crowley et al., 2019). If so, we can give

theory-informed practical advice to policy analysts, trying to solve real-world problems, with clarity and confidence. Examples include how people can: •

• •

Act as an entrepreneur, tell persuasive stories, become part of an advocacy coalition, engage in policy learning, and ‘navigate’ complex policymaking systems (Cairney, 2018a; Weible and Ingold, 2018; Dunlop and Radaelli, 2018; Simmons, 2018). Encourage more coordinated and effective institutional and system design (Heikkila and Andersson, 2018; Koski and Workman, 2018; Swann and Kim, 2018). Understand, and operate effectively and hold people to account in multi-centric policymaking systems (Cairney et al., 2019a).

THE POLITICS OF EVIDENCE-BASED POLICYMAKING Insights from policy studies and theories provide three main ways to understand the relationship between evidence and policy. First, they help clarify some rather vague ambitions. If it is hard to define policy and policymaking, and measure policy change, it is even harder to define ‘evidence-based’ policymaking (EBPM). Many discussions of EBPM are confusing ‘because people begin by complaining that they don’t have it without really saying what it is’ (Cairney, 2016a: 4). In contrast, careful theory-informed empirical analysis allows us to identify policy, measure change, explain it, and situate the role of evidence use within that wider context (Cairney, 2019d). In such cases, it is unrealistic to expect to come to an overall conclusion about the use of research in policy, but this approach is preferable to the tendency to cherry-pick examples of low evidence use to make a vague political point. Second, they help reject too-simple expectations about how policymakers will or should use research evidence. Our initial focus on ‘comprehensive rationality’ helps us understand why policymakers have to ignore almost all information, including a lot of

potentially policy-relevant evidence. They have to prioritize information by relying on a small amount of ways to understand and engage with the world, including their: beliefs, feedback from experience, emotional judgements, familiarity with evidence, and institutions. They have to ‘learn’ how to win battles and get what they want, as well as from experts and other actors. Therefore, it is unhelpful to hold them the standard of an all-knowing actor. Crucially, Chapter 4 rejects the idea that this problem is diminishing as our scientific methods become more sophisticated and our evidence base grows. On the contrary, there is a growing evidence base about the ways in which all humans (1) combine their beliefs, values, and other cognitive shortcuts to (2) find ways to ignore almost all information about the world, and (3) present evidence selectively in battles of ideas (see Mair et al., 2019 on how the European Commission describes this dynamic). Third, they help reject too-simple ways to model EBPM. There is no policy cycle containing clearly defined stages in which to present and use evidence. We should not expect to see a distinct evidencebased debate on problem definition, a single moment of evidencebased authoritative choice, or the use of evidence-based evaluations to drive choice. Instead, agenda setting is primarily about the exercise of power to define policy problems and influence the selective demand for evidence, there is low reliance on sophisticated tools to produce policy solutions, and evaluation is a highly political, not technical, way to measure success. We should look for more useful images and theories of the policy process to help us present a more realistic way to understand evidence use. The cycle image leads to inflated expectations followed by despair. More realistic images help produce pragmatic strategies to promote evidence within a system that no one fully understands or controls (Box 6.3, pp.103–4; Box 11.3, p.200; Chapter 12, pp.223–6). More generally, a focus on policy processes provides us with a different perspective on evidence use. Too many studies begin with an artificial construct – a focus on ‘the evidence’ – then bemoan its lack of progress or offer tips on how to make a greater impact (Cairney and Oliver, 2019; Oliver and Cairney, 2019). They ask questions such as ‘why is policy not evidence-based?’. In contrast,

the first edition of this book (Cairney, 2012a) used the word ‘evidence’ rarely, to reflect the need to understand a wide range of policymaking influences and dynamics (or a tendency to use broader terms like information or rationality). Policy theories can be used to ask similar questions, and answer them well, but we could ask 99 other, equally important, questions. For example, why do governments seem to privilege knowledge produced disproportionately by white, male, Western scientists at the expense of knowledge produced via experience and deliberation (p.50), or through the lens of the social groups most likely to be affected by policy? In other words, if we began by reading Chapter 3 on power, it might seem churlish and unreasonable to conclude that the big issue of our day is that scientific evidence is often ignored by policymakers. Instead, if our focus is on trying to diagnose and improve policy processes, we have many options from which to choose (see for example Box 13.2). Some might be about improvements to information processing, such as by governments ‘bundling’ issues to avoid policymaking silos (Koski and Workman, 2018), or to the ‘governance of evidence’ (Parkhurst, 2017). Others might be about ways to deliberate, produce rules, and foster sufficient trust to engage in meaningful collective action in many policymaking and non-governmental venues (Chapter 7). Further, such a wide focus should help clarify the salience of principles such as EBPM. For example, a lack of scientific evidence often reflects a greater focus on efforts to sustain democracy via elections or other forms of wide participation to encompass many views and preferences. Such processes reduce the tendency to rely exclusively on experts. Should we prioritize EBPM even if it means downplaying the importance of wider political participation? Or, should we accept that wide participation requires us to deliberate and share beliefs rather than simply seek to educate the public enough to change theirs?

USING MULTIPLE THEORIES: THREE CAUTIONARY TALES

So far, our assumption has been that we can, and should, combine multiple theoretical insights. Indeed, Sabatier (2007b: 330) highlights several advantages to the use of multiple theories: • • •

it provides ‘some guarantee against assuming that a particular theory is the valid one’ it shows us that ‘different theories may have comparative advantages in different settings’ the knowledge of other theories ‘should make one much more sensitive to some of the implicit assumptions in one’s favoured theory’.

Combining the merits of theories represents ‘standard advice’ in the literature (2007b: 330). Yet this initial appearance of consensus is illusory, reflecting a lack of clarity in the literature about what theories are and how we should combine them: ‘Few would disagree with the idea, largely because the sentiment is rather vague’ (Cairney, 2013a: 1). This lack of clarity about the definition of theory has several elements: there is no common definition of ‘theory’; only some scholars define theory in comparison with terms such as framework and model (p.25); and, statements about theories often act as – heavily debated – philosophical statements about our ability to explain and predict the social world (2013: 1–2). Indeed, Heikkila and Cairney (2018: 302) play it safe, using ‘the term theory in a generic sense for ease of description’ to allow them to focus largely on comparison (Box 13.3). Using this generic definition of theories, Cairney (2012a: 266–87; 2013a) identifies major problems in three main approaches to comparing and combining their insights. 1. Synthetic and hybrid theories One option is to assume that theories or concepts are broadly complementary and that we can combine their insights in a relatively straightforward manner. Indeed, John (2003: 488; 1998: 173) suggests that the most popular theories of public policy – such as MSA, PET, and the ACF – already adopt this approach to core concepts. They are ‘synthetic’ because they combine a discussion of many concepts to theorize ‘the relationship

between the five core causal processes … institutions, networks, socioeconomic process, choices, and ideas’. Although each theory has its limitations, public policy research still follows their agenda (2003: 495). John’s (1998: 183; 2012) proposed ‘evolutionary’ theory is also synthetic in that it describes how: institutions constrain actors but actors can challenge or reconstitute rules; ‘economic power structures’ give actors resources to pursue their interests through networks; and, policies are ‘ideas concretized by political action (i.e. interests)’. Similarly, we can treat complexity theory (in policy studies) as synthetic, since it brings together multiple insights from approaches such as PET, new institutionalism, and implementation and governance studies (Chapter 6, pp.105–6). Yet this approach is problematic because key insights cannot be put together so easily. A combination of theories, or the insights from those theories, requires that we are talking about the same thing. Yet our search for a common language is hindered by the use of the same term to mean different things in different theories. Thinking back, don’t you get the impression that (for example) sometimes people describe an institution when they mean an idea, or vice versa? Or, that PET equates an institution with an organization, while the Institutional Analysis and Development (IAD) framework equates it with rules? The chapters in this book show that terms like institution, network, idea, and policy entrepreneurs (Box 11.3) are contested, with many possible meanings. Each theory makes sense of these concepts in different ways, which makes it unwise to piece together their insights uncritically. Chapter 6 also shows that many theories make loose use of metaphors, such as the language of evolution (and related terms like punctuated equilibrium), which makes them hard to compare. Further problems arise when we are not sure what assumptions each theory contains about issues such as the relationship between structure and agency or, in many cases, the methods people use to produce new knowledge. An uncritical combination of theories masks these contradictions.

Or, as John (1998: 187; 1999: 48–49; 2012) often seems to suggest, synthesis may be about piecing together insights to produce new and better theories to replace the old, rather than try to accumulate insights and knowledge from them all. If so, this approach seems more likely to add to, rather than clear up, the conceptual sludge (Cairney et al., 2019a: 2). 2. Complementary theories A second option is to maintain a separation between theories to provide multiple, complementary, perspectives. As in our story of policymaking, we can identify bounded rationality and policymaking environments as our starting point, then use different theories to explore the consequences. Allison (1969; 1971) uses this general approach well. His aim is to explain why the US and Soviet Union made decisions that led to the Cuban missile crisis. To do so, he compares multiple sources of explanation for the same event: • • •

the ‘neorealist’ assumption of state rationality in which national governments are treated as ‘centrally coordinated, purposive individuals’; comprehensive and bounded rationality to examine the ‘standard operating procedures’ of the organizations that make up governments; and game theory to discern an overall pattern of behaviour from the rational choices of key individuals (1971: 3–7; 257).

There are many other examples of scholars combining multiple perspectives in the Allison tradition (Parker et al., 2009; Greenaway et al., 1992; Dunleavy, 1990; Gamble, 1990; Greener, 2002; Rosamond, 2000; Mintrom and Vergari, 1996), as well as debates that perform a similar function (John, 1999; McConnell, 2000). There is also a thriving field of studies drawing on multiple policy theories to explain one or more case studies (e.g. Bandelow et al., 2019). With this approach, the need to reconcile the meaning of each concept and theory is less pressing, because we are deliberately

seeking multiple perspectives and can accommodate their differences (Cairney, 2013a: 8–9). Instead, it presents a much more practical problem, in which it is difficult to create a research design that does justice to multiple separate theories. Imagine combining the quantitative methods of PET to identify budget punctuations, and conduct surveys and media analysis to identify advocacy coalitions, and compare in-depth IAD case studies, and conduct semi-structured interviews to piece together events during an MSA-style window of opportunity. This approach would be beyond most, if not all, PhD projects and few research projects are funded to make such comparisons. Instead, scholars tend to adopt a proxy version of this approach by producing a case study of policy change and providing quite general analyses of the ways in which we might use multiple theories to explain the same events (Cairney, 2007a; Cairney et al., 2012). Or, some scholars engage in meta-analysis to identify a small number of causal processes and extract from each study a finding that can be compared with findings from other studies (Exadaktylos and Radaelli, 2009; see also Newig and Fritsch, 2009: Kay, 2006; compare with van der Heijden et al., 2019). If so, the task of doing justice to each theory tends to be left to scholars specializing in each approach. 3. Contradictory theories A third option is to suggest that theories provide competing explanations and that one is better than another (or, at least, it should receive more of our attention). This approach is summed up in the (tongue-in-cheek) ‘policy shootout’ that promotes the most promising theories based on the (not tongue-in-cheek) ‘basic tenets of science’: 1. A theory’s methods should be explained so that they can be replicated by others. 2. Its concepts should be clearly defined, logically consistent, and give rise to empirically falsifiable hypotheses. 3. Its propositions should be as general as possible. 4. It should set out clearly what the causal processes are.

It should be subject to empirical testing and revision (Eller and 5. Krutz, 2009: 1; Sabatier, 2007a: 5; 2007b: 326–31). Several factors make this process seem complicated or unwise. First, only a small proportion of research for each theory in this book lives up to these principles consistently, and there is little evidence to suggest that the most prominent theories thrive on this basis (Meier, 2009). Second, shootout-style special issues or edited books do not really place this level of discipline on each scholar. Third, the principles give a misleading impression of social scientific research or present an artificial standard. Researchers may explain their methods and build ways to verify the validity and reliability of codes into the research design. Yet, in most studies, the results are accepted on trust; they are not replicated directly. For example, a theory’s main method may be the (well-respected) elite interview, but it would be absurd to expect one researcher’s interview data to be replicated by another (especially when interviews are anonymized). Fourth, very little policy research involves making a horoscope-style prediction. Rather, we tend to examine the evidence and assess the extent to which it is consistent with the logic of a theory after the fact. It would be disingenuous to suggest otherwise. Fifth, the idea of subjecting ‘empirically falsifiable’ hypotheses to testing is misleading. Policy change is open to interpretation and we can produce competing, and equally convincing, empirical narratives. Narratives of policy change are as open to debate as theories; if we cannot resolve what the evidence is, then we may struggle to choose decisively between theories. Indeed, the most valuable theory in each case may depend on the narrative of policy change that we select rather than just the nature of the theory. We can highlight instances in which theories provide competing assumptions or predictions, but be less certain about how to choose between them. Finally, these principles do not represent a universal understanding of science, or how we should decide between theories (or indeed what makes a theory useful – Box 13.3). Indeed, theories may be ‘incommensurable’ (Hindess, 1988: 73–75). The same words might be used in each theory, but they may mean

something else according to the paradigm from which they were derived. Researchers may look at the same object but view and interpret it differently. They may ‘not share a common set of perceptions which would allow scientists to choose between one paradigm and the other … there will be disputes between them that cannot all be settled by an appeal to the facts’ (1988: 74). Overall, these principles do not provide a convincing way to choose one theory and reject others. Instead, we have a form of methodological pluralism in which different studies take different approaches. While there are practical limits to the variety of acceptable methods (based on professional views expressed through the peer review process of journal publication and grant distribution), there is enough tolerance of variety and enough defences of methodological pluralism (Della Porta and Keating, 2008a; 2008b) to ensure that we do not have to select ‘one best way’. Indeed, if you speak to many exponents of particular theories, they will identify 101 problems with their own approach, which would make it a but rich to simply reject others. Box 13.3 Ways to compare theories of the policy process Cairney and Heikkila (2014) and Heikkila and Cairney (2018) describe the systematic comparison of policy theories. Chapter 7 shows how difficult it is to compare many cases, concepts, and theories even when there is a common framework. In the field as a whole, ‘major theories and frameworks have generally been produced independently of each other and were not designed with these comparisons in mind’ (2018: 302). Theories may provide ‘different frames of reference, foci, and concepts’, ‘attach different meanings to the same concepts’, try to explain different things, such as a small part of many cases or a large part of a single case, and define ‘theory’ differently (and, many theories have changed over time). Each theory takes a major investment to understand enough to apply empirically (the chapters in this book only scratch the surface), and we do not have the resources to give them all our full attention. Systematic comparisons help us understand the ways in which people might choose to apply some

more than others. Key ways to compare many theories systematically include (2018: 302–3): 1. Does it cover the ‘basic elements of a theory’? Whether each approach provides clarity on the scope and level of analysis, fosters a shared vocabulary and well-defined concepts, and presents clear assumptions about key dynamics, such as how people address bounded rationality (the ‘model of the individual’). This comparison allows us to specify interesting trade-offs, such as encouraging a general language to foster pluralism or very detailed agenda to foster close collaboration. 2. Is there an ‘active research’ programme? Indicators include the extent to which scholars are part of a thriving academic agenda based on many published empirical applications – in many political systems, policy areas, and over time, using many methods – and conceptual refinement in light of new knowledge. Most theories are thriving, with some evidence of a more international agenda, but a tendency for many applications to use the theories quite loosely (which makes it difficult to know if the research programme for each theory helps accumulate knowledge). 3. Does it ‘explain a large part of the policy process’? How does a theory conceptualize the relationship between key parts of policymaking environments: actors, institutions, networks, ideas, context, and events? Heikkila and Cairney’s (2018) chosen theories are MSA (Chapter 11 in this book), PET (Chapter 9), policy feedback theory (PFT, akin to historical institutionalism, Chapter 5), ACF (Chapter 10), NPF (Chapter 4), IAD (Chapter 7), and innovation and diffusion models (Chapter 12). These approaches perform well, partly because they are in the fourth edition of Theories of the Policy Process, following a preliminary assessment by its editors who use similar criteria. Earlier editions of the book (Sabatier, 1999; 2007a) contained chapters on the policy cycle and SCPD (Chapter 4 in this book), and Heikkila and Cairney (2018: 322) suggest that more expansive criteria – such as, do they connect

empirical and normative issues explicitly? – would have ensured SCPD inclusion (see also Simmons, 2018 on ‘cultural theory’). The most notable omission is ‘critical policy theory’ (Chapter 3) as a more general and thriving research agenda with dedicated journal venues (e.g. Critical Policy Theory) and conference meetings. This omission sums up the major difference that our criteria make to an assessment of the ‘best’ theories, such as if they (1) meet specific scientific criteria (Sabatier, 2007a: 5; 2007b: 326–31) or (2) help challenge social inequality (which often includes challenging these scientific criteria). This choice is an exercise of power that is often taken for granted rather than made explicitly in scientific debate (Douglas, 2009).

POLICY THEORY BEYOND THE ‘WEST’ OR GLOBAL NORTH In policy studies, we often describe the policy process when there are many policy processes. The singular description is analytically useful, to make our studies manageable, but it masks a key tenet in policy studies: policy and policymaking varies markedly from issue to issue, over time, and across space. Our aim is to find the right language to allow us to compare those processes. Without that common language, we would be unable to communicate our findings to each other in a systematic way (Box 13.3). Our case studies would describe parts of the policymaking world without an ability to explain how they relate to others, and without the sense that we can accumulate knowledge. Yet people make sense of the world in very different ways, and the rules of their political systems vary too much to expect a theory built on one part of the world to apply to it all. Indeed, we should not assume that the same concepts can describe all issues in all times and places. This is a particular issue when we treat concepts based on studies of ‘Western’ countries or those in the Global North as ‘universal’. Most of the policy theories in this book developed from studies of the USA before being applied to countries such as Canada and Australia and West European countries such as the UK,

Germany, and France, and other countries or regions within the European Union. There are too few applications across the globe to confirm the global applicability of these theories (at least in their current form). We can explore three different ways to address this problem (compare with Box 13.4). First, reflect continuously on a dualism between universal concepts and their specific application. Some concepts may be abstract enough to apply generally, such as when we describe a story of policymaking psychology within complex policymaking environments consisting of actors, institutions, networks, ideas, and socioeconomic factors. However, we need to explore in some depth the applications and limits of those terms. In each case, these concepts may provide us with a language to compare studies, but they come with a basic trade-off: if the language is abstract enough to be ‘universal’, it may be too abstract to suggest that, for example, country-level studies have substantial commonalities. The world is full of very different political systems, so our abstract language should not suggest that they are essentially the same. Context always matters. Therefore, with each concept, we need to balance a common reference point with the expectation of important differences: •

•

•

The truism that people are boundedly rational, engaging in fast and slow thinking, is an important starting point to analysis. However, we should not assume that policymakers use the same shortcuts or that their ways of thinking and acting have the same consequences. All political systems may be ‘complex’, with a multiplicity of actors involved in many levels or ‘centres’ of government. However, some systems are organized to recognize the benefit and necessity of multi-centric governance, while others operate as if power is centralized in the hands of a few key actors (Cairney et al., 2019a). All systems have institutions consisting of formal and informal rules. However, their constitutions and informal rules of institutions vary considerably (Chapter 5). In some cases, studies identify the effect of path dependence to help explain, for

•

•

•

example, why countries often have very different welfare states despite facing common pressures in a globalized world (Pierson, 2000b). However, if we take seriously the idea that rules ‘exist in the minds’ of participants (Ostrom, 2007: 23), then we accept that we have only scratched the surface of global comparisons. The general logic to explain the pervasiveness policy networks can be similar in each system. However, there can also be major differences in the extent to which interest groups are willing and able to mobilize to propose policy change, affecting the balance of power within networks (Cairney and Yamazaki, 2018). The paradigms, hegemonic ideas, or core beliefs in political systems can differ fundamentally and visibly (such as when some systems operate under conditions of capitalism or socialism) or be difficult to piece together through empirical study in subsystems. A focus on context and events generally suggests that policy choices in each country will result from specific events that are often unique to them. In some cases, countries have the same reference points – such as global attention to the Fukushima nuclear crisis in Japan – but react very differently to reflect previous events and decisions in their own countries (Box 6.2, p.98).

These differences become starker when we combine all of the above concepts to produce and apply theories of the policy process. For example, many of the specific ‘target populations’ subject to social stereotypes in the USA may have no equivalents in many other countries (Chapter 4), and the application of US theories to countries like China prompt us to revisit key assumptions (Box 13.4). Therefore, second, apply policy theories to more countries and take stock of the results. Several theories exhibit stocktaking exercises in which review authors: 1. Identify the numbers and proportions of applications in each country (Pierce et al., 2014; 2017; Jones et al., 2016). 2. Analyse key studies in depth to identify the difference that new applications make to theoretical conclusions (Henry et al., 2014;

Cairney and Jones, 2016). 3. Identify what it takes to coordinate the meaningful international application of a common framework (Poteete et al., 2010; Weible et al., 2018; Baumgartner et al., 2018). These studies tend to highlight a small number of, but a promising increase in, international applications. Some studies also show that combining case studies is as fruitful as combining theories to generate a wider perspective on policy processes, albeit with a tendency to treat newer perspectives as unusual (Box 13.4). Third, encourage dialogue between approaches and perspectives. This method would be relatively inductive and deliberative, in which we identify many different insights about policymaking, expect different and often implicit reference points, and move away from the sense that non-Western experiences are not the norm. US- or UKdriven approaches may take certain assumptions for granted, and they will affect evaluations of other systems. Put very simply, UK analyses often make reference to a relatively top-down and majoritarian culture, often in contrast to consensus democracies (Cairney et al., 2018). In comparison, we may associate a US system – post British colonial rule – with a story of breaking free from external interference, and creating a political system on principles such as power diffusion (rather than executive control), the freedom to mobilize against government policy or other actors, and elected representation in multiple sub-national jurisdictions. Further, many of the theories in this book began by using US-specific political system features to explain policymaking dynamics (Henry et al., 2014: 301; True et al., 2007: 157; Cairney, 2018a: 207). In countries less studied by Western theories, there may be new reference points that can be anticipated and operationalized (such as democratic/authoritarian state comparisons – Box 13.4). However, there are also concepts without a direct Western equivalent. For example, consider ideas in many African states about ‘traditional leadership’ and patriarchal or matriarchal (‘Queen Mother’) forms of informal or local governance. They may be dismissed unwisely as archaic because they were discouraged or ignored during colonial rule and have an unclear role in postcolonial

liberal democratic development. However, we would only provide partial analysis if we ignored their role in ‘informal governance’ or as part of a wider discussion of democracy beyond the ballot box (Mulaudzi, 2017: 45; Owusu-Mensah, 2015; Stoeltje, 1997; Ayres, 2017). Further, if our expectation is that such empirical and conceptual development takes place within Global South countries, without an expectation of using concepts developed in the North, we would also assume that meaningful comparisons could only be made in conversation across literatures rather than as part of the coordinated programmes described in Box 13.3. This process would not solve research dilemmas or the need for political choice in research, but it would at least make sure that we do not ignore them. Box 13.4 Theories and applications in China and India: Possibilities and problems China and India each account for approximately 18% of the world’s 7.6 billion population, compared to Europe as a whole at 10%, the USA at 4%, and the UK below 1% (United Nations Population Division, 2017: xxii). The development and application of policy theories does not come close to reflecting this proportion. There are three key ways in which to respond, each with their own advantages and limitations. 1. Applying the lens of existing theory to more countries. Applications to China identify distinctive issues in each theory. Chan and Zhao (2016: 137–38) draw on PET to identify ‘extreme stability’ and ‘drastic adjustment’ to policy, and link it to highly centralized and problematic information processing. Weible et al.’s (2018) ten-country study finds that China has one pro-fracking coalition compared to eight countries with two pro-/anti-coalitions (reflecting differences in coalition resources and governmental constraints on mobilization). Zhu (2008) flips one MSA assumption on its head by identifying the importance of policy solution infeasibility to major policy change. China’s authoritarian system produces an extreme example of the

same basic dynamic identified by PET (Chapter 9), an outlier for the ACF (Chapter 10), and a challenge to assumptions in MSA (Chapter 11). Although useful, this approach tells a story in a very particular way that would likely change if we began with the perspective of scholars in source countries. To tell a story from a US or UK perspective establishes what is normal enough practice to be described as routine, which can produce a tendency to describe the practices of ‘other’ countries as irregular or extreme. 2. Making comparative public policy analysis more global. Global comparisons of policy outputs and outcomes often identify major country-level differences. For example, Mamudu et al. (2015) and Cairney et al. (2012) identify the uneven implementation of a global commitment to the WHO Framework Convention on Tobacco Control (FCTC). Countries like the UK have policymaking environments conducive to this comprehensive policy change, while in China (which has a state monopoly in tobacco manufacturing) and India there is less evidence of FCTC implementation to ensure, for example, a meaningful ban on smoking in public places (although India has innovated with health information-prefaced movies which involve smoking). Although useful, this approach often encourages a highly normative comparison in which one type of country is advanced and another left behind (‘leaders’ and ‘laggards’ are common terms in policy transfer and diffusion studies). It reflects a long history of terminological debate, in which no description is neutral, and there are examples of category-use which seem to make an explicit normative judgement (such as democratic/authoritarian), represent an uneasy compromise (developed/developing countries), or try to project the lack of a value judgement (high-, low-, and middle-income countries). 3. New country specific studies.

Chakrabarti and Sanyal (2017) use a case study approach to apply policy theory insights to processes in India. Their spread of case studies is impressive. They include issues that are often part of international trends with country-level aspects (including criminal law, competition law, freedom of information, and corruption law reforms), combine domestic issues with global inequalities (food security), and have wider international implications even if many countries have had comparable laws for a longer time (the right to education, the regulation of child labour). They also engage specifically with ‘Western’ policy-making concepts to ask if they apply to ‘nonWestern’ countries (2017: 45–6; 301–2). However, even with a book-length treatment, they do not have the space in which to develop new theories systematically enough to provide an answer (except to conclude with reference to the value of complexity theory). Problems of comparability with other countries are magnified when we do not have comparable concepts on which to draw.

CONCLUSION From our review of the available literature, we can build a clear picture of the modern policy process, the key issues in public policy, and the main ways to study them. There is no single, authoritative decision maker. The policy process is marked by a diffusion of power from the centre towards other levels of government and, in many cases, quasi- and non-governmental actors. The lines are blurry between power exercised by executives, based on their formal roles and responsibilities, and influence pursued by other actors, based on their resources and experience of the process. Access to policymakers is not controlled by a small elite. The group– government world is not fully insulated. Rather, there are multiple channels of access to policymaking, and policy participants have access to a range of policymaking venues. At the same time, the policy world is clearly specialized. The logic of government is to divide the policy process into a large number of

smaller units and to devolve policymaking responsibility from the ‘top’ towards subsystems (and, in many cases, local or policy specific arrangements at the ‘bottom’). Consequently, the logical step for pressure participants is to specialize in a small number of policy areas and to participate within a small number of subsystems or policy networks. Policymakers devolve responsibility to officials who, in turn, seek information and advice from pressure participants. The currency of government may be power, in which people marshal their resources to represent groups and influence policy delivery, but it is also information and expertise which can be used to build reputations and trust. The modern policy process therefore contains sources of stability and policy continuity. Stable arrangements are a common feature of subsystems. The same small group of participants may be involved for long periods, at the expense of other actors, because they have the ability to exclude those actors. They have resources, based on their socioeconomic position. For example, businesses are central to the functioning of capitalist economies and the ability of governments to raise tax revenue, while doctors take centre stage in the treatment of illness. They also have resources based on their knowledge and expertise, which allows them to develop reputations valued within government and society. Such resources allow them to pursue privileged access to policymakers and positions within subsystems. They may also have the power to protect their positions by helping create policy monopolies: defining issues in particular ways to ensure that only certain groups are interested or deemed to have the necessary expertise, and presenting an image in which the policy problem has been solved in principle, with only the implementation to be discussed. Or, coalitions exercise power to further their beliefs and ensure that they have the dominant position within subsystems necessary to translate beliefs into policy action. These arrangements are often institutionalized: the frequent contact between officials and certain groups becomes routine, while the dominant ways to consider and address policy problems become taken for granted and rarely questioned. Further, policymakers are often unable or unwilling to challenge these arrangements. Paying attention to one

issue means ignoring most others, while policy innovation and termination require more political energy than policy succession. Yet there are also many sources of instability. Policy participants, dissatisfied with existing policy and institutional arrangements, can seek influence by identifying sympathetic audiences in other venues, and by lobbying policymakers who may be more willing to consider policy problems in a different way. Policy action taken in other jurisdictions can change policy overall (in systems where policymaking responsibilities overlap and multiple venues are involved) or act as a source of pressure for change. Events, such as major economic or environmental crises or changes of government, and significant policy failures, can prompt advocacy coalitions to reconsider their beliefs or other coalitions to exploit their position and increase their influence. Institutional barriers may be strong, but they can also be overcome, as new ideas are used to reframe policy problems and events prompt policymakers to reconsider their assumptions, process information in a different way, and seek alternative sources of information and advice. Finally, and perhaps most importantly, individuals make a difference. Some policymakers simply do not accept the existing arrangements and seek to challenge institutions and change policy. In short, policy processes are complex and often unpredictable. We can use theories and concepts to guide study and explanation but must also recognize their limitations. The identification of the conditions required for comprehensively rational policymaking serves primarily to show what actually happens when those conditions are not met. The division of policymaking into stages helps us analyse the process, not show us that it works this way. The identification of individuals making choices, institutional constraints, and socioeconomic pressures shows us that models focusing on only one of these aspects will not tell us the whole story. Theories based on the exercise of power are incomplete without theories that explain how ideas are promoted and accepted within government. Theories based on experiences in a small number of countries do not explain all experiences. We can combine the insights of these theories and account for most outcomes with reference to actors making choices in relation to institutions, networks, ideas, and socioeconomic

factors. However, we must also allow for an element of serendipity, as unpredictable events change how policy participants see the world and behave within it and windows of opportunity open to allow policy change. We should not become too dispirited by these constraints to our knowledge and understanding of the world. They are shared by all of the sciences, and should not be taken to suggest that public policy analysis is uniquely problematic. Rather, the public policy literature contains a rich and fascinating source of information on policy change and a variety of theories that help us explain how and why it changed. We can make tentative claims for knowledge accumulation and interdisciplinary application while seeking more global perspectives. We should also not be too dispirited by the conclusions about bounded rationality and complexity. It may be tempting to go too far, to wonder if there is any order and predictability in political systems. If so, we may give up on the idea that we can hold elected policymakers to account if they do not control policymaking and policy outcomes. Instead, I encourage you to use these insights to reflect continuously on the ways in which we expect policymakers to act. In other words, we can compare real-world policymaking with ideal-types (of order and central control) to help manage our expectations, rather than give up on the ability of actors to make a difference in the policy process. Well, you have reached the end of the book. If you would like to read more, please see my blog at https://paulcairney.wordpress.com/. Or, get in touch with me directly – via email – if you have any questions. I would be happy to help. Space – a geographical term to describe (1) a defined material area (such as a country or region), (2) the relationship between key elements within that area (such as humans interacting with their physical and social environment), and/or (3) the ways in which humans make sense of those material and social worlds (see Jones, 2009). We often compare across – and within – the jurisdictions of political systems but recognize that these boundaries are physically and socially constructed. For example, reference to the Global North/South seems more about relative levels of power and material resources than the equator.

Bibliography Abbott, P., Wallace, C. & Tyler, M. (2005) Introduction to Sociology: Feminist Perspectives (London: Routledge). Abdulla, A. (2018) ‘The Demise of US Nuclear Power in 4 Charts’, The Conversation http://theconversation.com/the-demise-of-us-nuclear-(power-in-4charts-98817) Ackerly, B. & True, J. (2020) Doing Feminist Research in Political and Social Science 2nd edn London: Red Globe Press. Ackrill, R. & Kay, A. (2011) ‘Multiple Streams in EU Policy-Making: The Case of the 2005 Sugar Reform’, Journal of European Public Policy, 18: 72–89. Adler, E. & Haas, P. (1992) ‘Conclusion: Epistemic Communities, World Order, and the Creation of a Reflective Research Program’, International Organization, 46, 1: 367–390. Ahmed, S. (2017) Living a Feminist Life (London: Duke University Press). Aklin, M. & Mildenberger, M. (2018) ‘Prisoners of the Wrong Dilemma: Why Distributive Conflict, Not Collective Action, Character-izes the Politics of Climate Change’, Working Paper at https://papers.ssrn.com/sol3/papers.cfm? abstract_id=3281045 Alasuutari, P. (2015) ‘The Discursive Side of New Institutionalism’, Cultural Sociology, 9, 2: 162–184. Alford, J. & O’Flynn, J. (2009) ‘Making Sense of Public Value: Concepts, Critiques and Emergent Meanings’, International Journal of Public Administration, 32, 3–4: 171–191. Aligica, P. & Tarko, V. (2012) ‘Polycentricity: From Polanyi to Ostrom, and Beyond’, Governance, 25, 2: 237–262. Allison, G. (1969) ‘Conceptual Models and the Cuban Missile Crisis’, American Political Science Review, 63, 3: 689–718. Allison, G. (1971) Essence of Decision: Explaining the Cuban Missile Crisis (Boston: Little, Brown). Alter, K. & Meunier, S. (2009) ‘The Politics of International Regime Complexity’, Perspectives on Politics, 7, 1: 13–24. Althaus, C., Bridgman, P. & Davis, G. (2007) The Australian Policy Handbook 4th edn (Sydney: Allen & Unwin). Althaus, C., Bridgman, P. & Davis, G. (2013) The Australian Policy Handbook 5th edn (Sydney: Allen & Unwin). Anderson, J. A. (1975) Public Policy Making (London: Thomas Nelson & Sons). Andrews, R. & Martin, S. (2007) ‘Has Devolution Improved Public Services?’ Public Money and Management, 27, 2: 149–156.

Annesley, C. & Gains, F. (2010) ‘The Core Executive: Gender, Power and Change’, Political Studies, 58, 5: 909–929 Ansell, C. & Gash, A. (2008) ‘Collaborative Governance in Theory and Practice’, Journal of Public Administration Research and Theory 18, 4: 543–571. Arendt, H. (1986) ‘Communicative Power’ in S. Lukes (ed.) Power (New York: New York University Press). Armstrong, A. (1998) ‘A Comparative Analysis: New Public Management – The Way Ahead?’, Australian Journal of Public Administration, 57, 2: 12–24. Arrow, K. (1963) Social Choice and Individual Values 2nd edn (New York: John Wiley & Sons) http://cowles.econ.yale.edu/P/cm/m12-2/index.htm Asare, B., Cairney, P. & Studlar, D. (2009) ‘Federalism and Multilevel Governance in Tobacco Control Policy: The European Union, United Kingdom, and Devolved Institutions’, Journal of Public Policy, 29, 1: 79–102. Atchison, A. (2017) ‘Women: Mostly Missing from Introductory Political Science Texts’, PSA Women and Politics Specialist Group 31 March https://psawomenpolitics.com/2017/03/31/women-mostly-missing-fromintroductory-political-science-texts/ Atkinson, M. & Coleman, W. (1989) ‘Strong States and Weak States: Sectoral Policy Networks in Advanced Capitalist Economies’, British Journal of Political Science, 19, 1: 47–67. Avery, G. (2004) ‘Bioterrorism, Fear, and Public Health Reform: Matching a Policy Solution to the Wrong Window’, Public Administration Review, 64, 3: 275–288. Aviram, N., Cohen, N. & Beeri, I. (2019) ‘Wind(ow) of Change: A Systematic Review of Policy Entrepreneurship Characteristics and Strategies’, Policy Studies Journal, Early View https://doi.org/10.1111/psj.12339 Axelrod, R. (1984) The Evolution of Cooperation (New York: Basic Books). Axelrod, R. (1986) ‘An Evolutionary Approach to Norms’, American Political Science Review, 80, 4: 1095–1111. Ayres, S. (2017) ‘Assessing the Impact of Informal Governance on Political Innovation’, Public Management Review, 19, 1: 90–107. Bacchi, C. (1999) Women, Policy and Politics (London: Sage). Bacchi, C. (2009) Analysing Policy: What’s the Problem Represented to be? (New South Wales: Pearson Australia). Bache, I. (1999) ‘The Extended Gatekeeper: Central Government and the Implementation of EC Regional Policy in the UK’, Journal of European Public Policy, 6, 1: 28–45. Bache, I., Bartle, A., Flinders, M. & Marsden, G. (2015) ‘Blame Games and Climate Change: Accountability, Multi-level Governance and Carbon Management’, British Journal of Politics and International Politics, 17, 1: 64–88. Bache, I. & Flinders, M. (2004a) ‘Multi-level Governance and the Study of the British State’, Public Policy and Administration, 19, 1: 31–51. Bache, I. & Flinders, M. (2004b) ‘Themes and Issues in Multi-level Governance’ in I. Bache & M. Flinders (eds.) Multi-level Governance (Oxford: Oxford University

Press). Bache, I. & Reardon, L. (2013) ‘An Idea Whose Time has Come? Explaining the Rise of Wellbeing in British Politics’, Political Studies, 61, 4: 898–914. Bache, I. & Reardon, L. (2016) The Politics and Policy of Wellbeing: Understanding the Rise and Significance of a New Agenda (Cheltenham: Edward Elgar). Bachrach, P. & Baratz, M. (1962) ‘Two Faces of Power’, American Political Science Review, 56, 4: 947–952. Bachrach, P. & Baratz, M. (1963) ‘Decisions and Nondecisions: An Analytical Framework’, The American Political Science Review, 57, 3: 632–642. Bachrach, P. & Baratz, M. (1970) Power and Poverty (New York: Oxford University Press). Baekgaard, M., Christensen, J., Dahlmann, C., Mathiasen, A. & Petersen, N. (2017) ‘The Role of Evidence in Politics: Motivated Reasoning and Persuasion Among Politicians’, British Journal of Political Science, https://doi.org/10.1017/S0007123417000084: 1–24 Baggio, J. A., Barnett, A. J., Perez-Ibara, I., Brady, U., Ratajczyk, E., Rollins, N., Rubiños, C., Shin, H. C., Yu, D. J., Aggarwal, R. & Anderies, J. M. (2016) ‘Explaining Success and Failure in the Commons: The Configural Nature of Ostrom’s Institutional Design Principles. International Journal of the Commons, 10, 2: 417–439. Baggott, R. (1986) ‘By Voluntary Agreement: The Politics of Instrument Selection’. Public Administration, 64, 1: 51–67. Bailey, D. & Wood, M. (2017) ‘The Metagovernance of English Devolution’, Local Government Studies, 43, 6: 966–991. Bakir, C. (2009) ‘Policy Entrepreneurship and Institutional Change: Multilevel Governance of Central Banking Reform’, Governance, 22, 4: 571–598. Bakir, C. & Jarvis, D. (2017) ‘Contextualising the Context in Policy Entrepreneurship and Institutional Change, Policy and Society, 36, 4: 465–478. Bale, T. (2008) European Politics 2nd edn (London: Red Globe Press). Bale, T. (2017) European Politics 4th edn (London: Red Globe Press). Bandelow, N., Vogeler,C., Hornung, J., Kuhlmann, J. & Heidrich, S. (2019) ‘Learning as a Necessary but Not Sufficient Condition for Major Health Policy Change: A Qualitative Comparative Analysis Combining ACF and MSF’, Journal of Comparative Policy Analysis: Research and Practice, 21, 2: 167–182. Barber, M. (2012) Instruction to Deliver (Croydon: Methuen). Bardach, E. (2006) ‘Policy Dynamics’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press), 336–66. Bardach, E. (2009) A Practical Guide for Policy Analysis: The Eightfold Path to More Effective Problem Solving (Washington, DC: CQ Press). Bardach, E. & Patashnik, E. M. (2015) A Practical Guide for Policy Analysis: The Eightfold Path to More Effective Problem Solving (Washington, DC: CQ Press).

Barrett, S. (2004) ‘Implementation Studies: Time for a Revival?’ Public Administration, 82, 2: 249–262. Barrett, S. & Fudge, C. (eds.) (1981) Policy and Action (London: Methuen). Barry, B. (1980a) ‘Is it Better to be Powerful or Lucky: Part 1’, Political Studies, 28, 2: 183–194. Barry, B. (1980b) ‘Is it Better to be Powerful or Lucky: Part 2’, Political Studies, 28, 3: 338–352. Barry, B. (2002) ‘Capitalists Rule OK? Some Puzzles About Power’, Politics Philosophy Economics, 1: 155–184. Barzelay, M. & Gallego, R. (2010) ‘The Comparative Historical Analysis of Public Management Policy Cycles in France, Italy, and Spain: Symposium Conclusion’, Governance, 23, 2: 297–307. Bassel, L. & Emejulu, A. (2017) Minority Women and Austerity: Survival and Resistance in France and Britain (Bristol: Policy Press). Baumgartner, F. (2013) ‘Ideas and Policy Change’, Governance, 26, 2: 239–258. Baumgartner, F. (2014) ‘Ideas, Paradigms and Confusions’, Journal of European Public Policy, 21, 3: 475–480. Baumgartner, F. (2017) ‘Endogenous Disjoint Change’, Cognitive Systems Research, 44: 69–73. http://dx.doi.org/10.1016/j.cogsys.2017.04.001 Baumgartner, F., Breunig, C., Green-Pedersen, C., Jones, B., Mortensen, P., Nuytemans, M. & Walgrave, S. (2009) ‘Punctuated Equilibrium in Comparative Perspective’, American Journal of Political Science, 53, 3: 603–620. Baumgartner, F., Green-Pedersen, C. & Jones, B. (eds.) (2006) ‘Comparative Studies of Policy Agendas’, special issue of the Journal of European Public Policy, 13: 7. Baumgartner, F. & Jones, B. (1993) Agendas and Instability in American Politics (Chicago: University of Chicago Press). Baumgartner, F. & Jones, B. (2002) Policy Dynamics (Chicago: University of Chicago Press). Baumgartner, F. & Jones, B. (2009) Agendas and Instability in American Politics 2nd edn (Chicago: University of Chicago Press). Baumgartner, F. & Jones, B. (2015) The Politics of Information (Chicago: University of Chicago Press). Baumgartner, F., Jones, B. & Mortensen, P. (2018) ‘Punctuated Equilibrium Theory’ in C. Weible & P. Sabatier (eds.) Theories of the Policy Process 4th edn (Chicago: Westview Press). Baumgartner, F. & Mahoney, C. (2005) ‘Social Movements, the Rise of New Issues, and the Public Agenda’ in D. Meyer, V. Jenness & H. Ingram (eds.) Routing the Opposition: Social Movements, Public Policy, and Democracy (Minneapolis: University of Minnesota Press). BBC News (2008) ‘G7 Passes Plan to Ease Credit Woe’, 11 April, http://news.bbc.co.uk/1/hi/business/7342419.stm

Beeson, M. and Stone, D. (2013) ‘The changing fortunes of a policy entrepreneur: The case of Ross Garnaut’, Australian Journal of Political Science 48, 1: 1–14. Béland, D. (2010) ‘The Idea of Power and the Role of Ideas’, Political Studies Review, 8, 2: 145–154. Béland, D. (2019) How Ideas and Institutions Shape the Politics of Public Policy (Cambridge: Cambridge University Press). Béland, D. & Cox, R. (2010) ‘Introduction: Ideas and Politics’ in D. Béland & R. Cox (eds.) Ideas and Politics in Social Science Research (Oxford: Oxford University Press). Béland, D. & Schlager, E. (2019) ‘Varieties of Policy Feedback Research: Looking Backward, Moving Forward’, Policy Studies Journal, 47, 2: 184–205. Bennett, C. (1991a) ‘What is Policy Convergence and What Causes It?’, British Journal of Political Science, 21, 2: 215–233. Bennett, C. (1991b) ‘How States Utilize Foreign Evidence’, Journal of Public Policy, 11, 1: 31–54. Bennett, C. & Howlett, M. (1992) ‘The Lessons of Learning: Reconciling Theories of Policy Learning and Policy Change’, Policy Sciences, 25: 275–294. Benson, D. & Jordan, A. (2011) ‘What Have We Learned from Policy Transfer Research? Dolowitz and Marsh Revisited’, Political Studies Review, 9, 3: 366– 378. Beresford, P. & Russo, J. (2016) ‘Supporting the Sustainability of Mad Studies and Preventing its Co-option’, Disability & Society, 31, 2: 270–274. Berry, F. & Berry, W. (2007) ‘Innovation and Diffusion Models in Policy Research’ in P. Sabatier (ed.) Theories of the Policy Process (Cambridge, MA: Westview Press). Berry, F. & Berry, W. (2018) ‘Innovation and Diffusion Models in Policy Research’ in C. Weible & P. Sabatier (eds.) Theories of the Policy Process 4th edn (Chicago: Westview Press). Bevir, M. (2009) ‘Anti-Foundationalism’ in M. Flinders, A. Gamble, C. Hay & M. Kenny (eds.) The Oxford Handbook of British Politics (Oxford: Oxford University Press). Bevir, M. (2013) A Theory of Governance (Berkeley: Gaia Books). Bevir, M. & Rhodes, R. A. W. (1999) ‘Studying British Government’, British Journal of Politics and International Relations, 1, 2: 215–239. Bevir, M. & Rhodes, R. A. W. (2003) Interpreting British Governance (London: Routledge). Bhopal, K. (2018) White Privilege: The Myth of a Post-Racial Society (Bristol: Policy Press). Bickerton, C. J., Hodson, D. & Puetter, U. (2014) ‘The New Intergovernmentalism: European Integration in the Post-Maastricht era’, Journal of Common Market Studies 53, 4: 703–722. Biersteker, T. (1990) ‘Reducing the Role of the State in the Economy: A Conceptual Exploration of IMF and World Bank Prescriptions’, International

Studies Quarterly, 34: 477–492. Birkland, T. (1997) After Disaster: Agenda Setting, Public Policy and Focusing Events (Washington, DC: Georgetown University Press). Birkland, T. (2005) An Introduction to the Policy Process: Theories, Concepts and Models of Public Policy Making 2nd edn (London: M. E. Sharpe). Birkland, T. (2016) ‘Attention and Natural Disasters’ in N. Zahariadis (ed.) Handbook of Public Policy Agenda-Setting (Cheltenham: Edward Elgar). Birkland, T. (2017) An Introduction to the Policy Process 4th edn (London: Routledge). Blackman, T. (2001) ‘Complexity Theory and the New Public Management’, Social Issues, 2: 1–11. Blomquist, W. (1992) Dividing the Waters: Governing Groundwater in Southern California (San Francisco: ICS). Blomquist, W. (2007) ‘The Policy Process and Large-N Comparative Studies’ in P. Sabatier (ed.) Theories of the Policy Process, 2nd edn (Cambridge, MA: Westview Press). Blum, S. (2014) ‘No Need to Reinvent the Wheel: Family Policy Transfers in Germany and Austria’, Policy Studies, 35, 4: 357–376. Blum, S. (2018) ‘The Multiple-Streams Framework and Knowledge Utilization: Argumentative Couplings of Problem, Policy, and Politics Issues’, European Policy Analysis, 4, 1: 94–117. Blyth, M. (1997) ‘Any More Bright Ideas: The Ideational Turn in Comparative Political Economy’, Comparative Politics, 29, 2: 229–250. Blyth, M. (2002) Great Transformations: Economic Ideas and Institutional Change in the Twentieth Century (Cambridge: Cambridge University Press). Blyth, M. (2013) Austerity: The History of a Dangerous Idea (Oxford: Oxford University Press). Boaz, A., Davies, H., Fraser, A. & Nutley, S. (eds.) (2019) What Works Now? (Bristol: PolicyPress). Bock, G. & James, S. (1992) Beyond Equality and Difference: Citizenship, Feminist Politics, and Female Subjectivity (London: Routledge). Boin, A., ‘t Hart, P., McConnell, A. & Preston, T. (2010) ‘Leadership Style, Crisis Response and Blame Management: The Case of Hurricane Katrina’, Public Administration, 88, 3: 706–720. Bolleyer, N., Swenden, W. & McEwen, N. (2014) ‘Constitutional Dynamics and Partisan Conflict: A Comparative Assessment of Multi-level Systems in Europe’, Comparative European Politics, 12, 4–5: 531–555. Bolzendahl, C. (2014) ‘Opportunities and Expectations: The Gendered Organization of Legislative Committees in Germany, Sweden, and the United States’, Gender and Society, 28, 6: 847–876. Boswell, C. (2009) The Political Uses of Expert Knowledge (Cambridge: Cambridge University Press).

Boswell, J. (2018) ‘What Makes Evidence‐based Policy Making Such a Useful Myth? The Case of NICE Guidance on Bariatric Surgery in the United Kingdom’, Governance, 31, 2: 199–214. Botterill, L. & Hindmoor, A. (2012) ‘Turtles All the Way Down: Bounded Rationality in an Evidence-based Age’, Policy Studies, 33, 5: 367–379. Bourne, J. (2001) ‘The Life and Times of Institutional Racism’, Race & Class, 43, 2: 7–22. Bovaird, T. (2008) ‘Emergent Strategic Management and Planning Mechanisms in Complex Adaptive Systems’, Public Management Review, 10, 3: 319–340. Bovens, M., ‘t Hart, P. & Kuipers, S. (2006) ‘The Politics of Policy Evaluation’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Bovens, M., ‘t Hart, P., Peters, B. G., Albaek, E., Busch, A., Dudley, G., Moran, M. & Richardson, J. (2001) ‘Patterns of Governance: Sectoral and National Comparisons’ in M. Bovens, P. ‘t Hart & B. G. Peters (eds.) Success and Failure in Public Governance (Cheltenham: Edward Elgar). Boyd, R., Richerson, P., Meinzen-Dick, R., De Moor, T., Jackson, M., Gjerde, K., Harden-Davies, H., Frischmann, B., Madison, M., Strandburg, K., McLean, A. & Dye, C. (2018) ‘Tragedy Revisited’, Science, 362, 6420: 1236–1241. Boyle, K. (2005) Media and Violence (London: Sage). Braithwaite, J. (2006) ‘The Regulatory State’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Brennan, G. (1996) ‘Selection and the Currency of Reward’ in R. Goodin (ed.) The Theory of Institutional Design (Cambridge: Cambridge University Press). Brewer, G. & deLeon, P. (1983) The Foundations of Policy Analysis (Chicago: Dorsey Press). Bridgman, P. & Davis, G. (2003) ‘What Use is a Policy Cycle? Plenty, if the Aim is Clear’, Australian Journal of Public Administration, 62, 3: 98–102. Brinkley, C. (2019) ‘Hardin’s Imagined Tragedy is Pig Shit: A Call for Planning to Recenter the Commons’, Planning Theory https://doi.org/10.1177/1473095218820460 Brinkmann, S. (2009) ‘Facts, Values, and the Naturalistic Fallacy in Psychology’, New Ideas in Psychology, 27, 1: 1–17. Buchanan, J. (1988) ‘Market Failure and Political Failure’, Cato Journal, 8, 1: 1–13, http://www.cato.org/pubs/journal/cj8n1/cj8n1-1.pdf Budge, I. (2006) in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Bulmer, S., Dolowitz, D., Humphreys, P. & Padgett, S. (2007) Policy Transfer in the European Union (London: Routledge). Bulmer, S. & Paterson, W. (2013) ‘Germany as the EU’s Reluctant Hegemon? Of Economic Strength and Political Constraints’, Journal of European Public Policy,

20, 10: 1387–1405. Burgess, M. (2006) Comparative Federalism: Theory and Practice (London: Routledge). Burns, J. (1963) The Deadlock of Democracy (Englewood Cliffs, NJ: PrenticeHall). Busch, P. & Jörgens, H. (2005) ‘The International Sources of Policy Convergence: Explaining the Spread of Environmental Policy Innovations’, Journal of European Public Policy, 12, 5: 860–884. Cairney, P. (1997) ‘Advocacy Coalitions and Political Change’, in G. Stoker & J. Stanyer (eds.) Contemporary Political Studies 1997 (Belfast: Political Studies Association) http://www.informaworld.com/index/792976384.pdf Cairney, P. (2002) ‘New Public Management and the Thatcher Health Care Legacy’, British Journal of Politics and International Relations, 4, 3: 375–398. Cairney, P. (2006) ‘Venue Shift Following Devolution: When Reserved Meets Devolved in Scotland’, Regional and Federal Studies, 16, 4: 429–445. Cairney, P. (2007a) ‘A “Multiple Lenses” Approach to Policy Change: The Case of Tobacco Policy in the UK’, British Politics, 2, 1: 45–68. Cairney, P. (2007b) ‘Using Devolution to Set the Agenda? Venue Shift and the Smoking Ban in Scotland’, British Journal of Politics and International Relations, 9: 73–89. Cairney, P. (2008) ‘Has Devolution Changed the British Policy Style?’ British Politics, 3, 3: 350–372. Cairney, P. (2009a) ‘Implementation and the Governance Problem: A Pressure Participant Perspective’, Public Policy and Administration, 24, 4: 355–377. Cairney, P. (2009b) ‘The “British Policy Style” and Mental Health: Beyond the Headlines’, Journal of Social Policy, 38, 4: 1–18. Cairney, P. (2009c) ‘The Role of Ideas in Policy Transfer: The Case of UK Smoking Bans since Devolution’, Journal of European Public Policy, 16, 3: 471–488. Cairney, P. (2012a) Understanding Public Policy: Theories and Issues 1st edn (London: Red Globe Press). Cairney, P. (2012b) ‘Complexity Theory in Political Science and Public Policy’, Political Studies Review, 10, 3: 346–358. Cairney, P. (2012c) ‘Intergovernmental Relations in Scotland Before and After the SNP’, British Journal of Politics and International Relations, 14, 2: 231–249. Cairney, P. (2013a) ‘Standing on the Shoulders of Giants: How Do We Combine the Insights of Multiple Theories in Public Policy Studies?’ Policy Studies Journal, 41, 1: 1–21. Cairney, P. (2013b) ‘What is Evolutionary Theory and How Does it Inform Policy Studies?’ Policy and Politics, 41, 2: 279–298. Cairney, P. (2015a) ‘How Can Policy Theory Have an Impact on Policy Making?’ Teaching Public Administration, 33, 1: 22–39. Cairney, P. (2015b) ‘Sabatier’s Advocacy Coalition Model of Policy Change’ in E. Page, S. Balla & M. Lodge (eds.) Oxford Handbook of the Classics of Public

Policy and Administration (Oxford: Oxford University Press). Cairney, P. (2016a) The Politics of Evidence-based Policymaking (London: Palgrave Pivot). Cairney, P. (2016b) ‘Storytelling for Policy Change: Promise and Problems’ Paul Cairney: Politics & Public Policy https://paulcairney.wordpress.com/2016/11/01/storytelling-for-policy-changepromise-and-problems/ Cairney, P. (2016c) ‘Policy Concepts in 1000 Words: The Intersection between Evidence and Policy Transfer’ Paul Cairney: Politics & Public Policy https://paulcairney.wordpress.com/2016/02/10/policy-concepts-in-1000-wordsthe-intersection-between-evidence-and-policy-transfer/ Cairney, P. (2017a) ‘5 Images of the Policy Process’, Paul Cairney: Politics & Public Policy https://paulcairney.wordpress.com/2017/07/10/5-images-of-thepolicy-process/ Cairney, P. (2017b) ‘Evidence-based Best Practice is More Political Than it Looks: A Case Study of the “Scottish Approach”’, Evidence and Policy, 13, 3: 499–515. Cairney, P. (2017c) ‘How to Write Theory-Driven Policy Analysis’ Paul Cairney: Politics & Public Policy, https://paulcairney.wordpress.com/2017/11/13/how-towrite-theory-driven-policy-analysis/ Cairney, P. (2018a) ‘Three Habits of Successful Policy Entrepreneurs’, Policy and Politics, 46, 2: 199–217. Cairney, P. (2018b) ‘Evidence-Based Policymaking and the “New Policy Sciences”’, Paul Cairney: Politics & Public Policy, https://paulcairney.wordpress.com/2018/10/25/evidence-based-policymakingand-the-new-policy-sciences-2/ Cairney, P. (2018c) ‘Why Don’t Policymakers Listen to Your Evidence?’, Paul Cairney: Politics & Public Policy https://paulcairney.wordpress.com/2018/03/26/why-dont-policymakers-listen-toyour-evidence/ Cairney, P. (2018d) ‘Teaching Evidence-Based Policy to Fly’, Paul Cairney: Politics & Public Policy https://paulcairney.wordpress.com/2018/10/11/teachingevidence-based-policy-to-fly-how-to-deal-with-the-politics-of-policy-learning-andtransfer/ Cairney, P. (2019a) ‘Evidence and Policy Making’ in A. Boaz, H. Davies, A. Fraser & S. Nutley (eds.) What Works Now? (Bristol: Policy Press). Cairney, P. (2019b) ‘Policy Styles in the UK: Majoritarian UK Versus Devolved CConsensus Democracies?’ M. Howlett & J. Tosun (eds.) Policy Styles and Policy-Making: Exploring the National Dimension (London: Routledge). Cairney, P. (2019c) ‘Fostering Evidence-informed Policy Making: Uncertainty Versus Ambiguity’, National Collaborating Centre for Healthy Public Policy (Québec: NCCHPP) http://www.ncchpp.ca/41/What_s_New_.ccnpps? id_article=1930

Cairney, P. (2019d) ‘The UK Government’s Imaginative Use of Evidence to Make Policy’, British Politics, 14, 1: 1–22. Cairney, P. (2019e) ‘The Transformation of UK Tobacco Control’ in M. Compton & P. ‘t Hart (eds.) Great Policy Successes (Oxford: Oxford University Press). Cairney, P., Fischer, M. & Ingold, K. (2016a) ‘Hydraulic Fracturing Policy in the UK: Coalition, Cooperation and Opposition in the Face of Uncertainty’ in C. Weible, T. Heikkila, K. Ingold & M. Fischer (eds.) Comparing Coalition Politics: Policy Debates on Hydraulic Fracturing in North America and Western Europe (London: Palgrave Macmillan). Cairney, P. & Geyer, R. (2015) ‘Introduction’ in R. Geyer & P. Cairney (eds.) Handbook on Complexity and Public Policy (Cheltenham: Edward Elgar). Cairney, P., Harvey, M. & St Denny, E. (2017a) ‘Constitutional Change, Social Investment and Prevention Policy in Scotland’ in M. Keating (ed.) A Wealthier, Fairer Scotland (Edinburgh: Edinburgh University Press). Cairney, P. and Heikkila, T. (2014) ‘A Comparison of Theories of the Policy Process’ in C. Weible & P. Sabatier (eds.) Theories of the Policy Process 3rd edn (Chicago: Westview Press). Cairney, P., Heikkila, T. & Wood, M. (2019a) Making Policy in a Complex World (Cambridge: Cambridge University Press). Cairney, P., Ingold, K. & Fischer, M. (2018a) ‘Fracking in the UK and Switzerland: Why Differences in Policymaking Systems Don’t Always Produce Different Outputs and Out-comes’, Policy and Politics, 46, 1: 125–147. Cairney, P. & Jones, M. (2016) ‘Kingdon’s Multiple Streams Approach: What Is the Empirical Impact of This Universal Theory?’, Policy Studies Journal 44, 1: 37–58 Cairney, P., Keating, M. and St Denny, E. (2018c) ‘How to use evidence to identify, learn from, and transfer policy success’ (IMAJINE Report D6.1 Conceptual Framework for Empirical Research) (Brussels: European Research Council). Cairney, P. & Kwiatkowski, R. (2017) ‘How to Communicate Effectively with Policymakers: Combine Insights from Psychology and Policy Studies’, Palgrave Communications, 3, 37 https://www.nature.com/articles/s41599-017-0046-8 Cairney, P. & Mamudu, H. (2014) ‘The Global Tobacco Control “Endgame”: Change the Policy Environment to Implement the FCTC’ Journal of Public Health Policy, 35, 4: 506–517 doi: 10.1057/jphp.2014.18 Cairney, P. & McGarvey, N. (2013) Scottish Politics 2nd edn (London: Red Globe Press). Cairney, P., McHarg, A., McEwen, N. & Turner, K. (2019b) ‘How to Conceptualise Energy Law and Policy for an Interdisciplinary Audience: The Case of PostBrexit UK’, Energy Policy, 129, June: 459–466. Cairney, P. & Oliver, K. (2017) ‘Evidence-based Policymaking Is Not Like Evidence-Based Medicine, So how Far Should You Go to Bridge the Divide Between Evidence and Policy?’ Health Research Policy and Systems, doi: 10.1186/s12961-017-0192-x

Cairney, P. & Oliver, K. (2019) ‘How Should Academics Engage in Policymaking to Achieve Impact?’ Political Studies Review https://doi.org/10.1177/1478929918807714 Cairney, P., Oliver, K. & Wellstead, A. (2016b) ‘To Bridge the Divide between Evidence and Policy: Reduce Ambiguity as Much as Uncertainty’, Public Administration Review, 76, 3: 399–402. Cairney, P. & Rummery, K. (2018) ‘Feminising Politics to Close the EvidencePolicy Gap: The Case of Social Policy in Scotland’, Australian Journal of Public Administration, 77, 4: 542–553. Cairney, P., Russell, S. & St Denny, E. (2017b) ‘The “Scottish Approach” to Policy and Policymaking: What Issues Are Territorial and What Are Universal?’ Policy and Politics, 44, 3: 333–350. Cairney, P. & St Denny, E. (2020) Why Isn’t Government Policy More Preventive? (Oxford: Oxford University Press). Cairney, P., Studlar, D. & Mamudu, H. (2012) Global Tobacco Control: Power, Policy, Governance and Transfer (Basingstoke: Palgrave Macmillan). Cairney, P., Timonina, I. & Stephan, H. (2018b) ‘If the UK Government is “All Out for Shale”, How Can the Scottish Government Prohibit Fracking? A Two-Level ACF Explanation’, ACF workshop, ECPR Nicosia, April. Cairney, P. & Weible, C. (2015) ‘Comparing and Contrasting Peter Hall’s Paradigms and Ideas with the Advocacy Coalition Framework’ in M. Howlett & J. Hogan (eds.) Policy Paradigms in Theory and Practice (Basingstoke: Palgrave). Cairney, P. & Weible, C. (2017) ‘The New Policy Sciences: Combining the Cognitive Science of Choice, Multiple Theories of Context, and Basic and Applied Analysis’, Policy Sciences, 50, 4: 619–627. Cairney, P. and Wellstead, A. (2019) ‘The Role of Trust in Policymaking’, Paper to the International Conference on Public Policy, Montreal, https://www.ippapublicpolicy.org//file/paper/5d039240e1560.pdf Cairney, P. & Widfeldt, A. (2015) ‘Is Scotland a Westminster-style Majoritarian Democracy or a Scandinavian-style Consensus Democracy? A comparison of Scotland, the UK and Sweden’, Regional and Federal Studies, 25, 1: 1–18. Cairney, P. & Yamazaki, M. (2018) ‘A Comparison of Tobacco Policy in the UK and Japan: If the Scientific Evidence is Identical, Why Is There a Major Difference in Policy?’ Journal of Comparative Policy Analysis, 20, 3: 253–268. Cairney, P. & Zahariadis, N. (2016) ‘Multiple Streams Analysis: A Flexible Metaphor Presents an Opportunity to Operationalize Agenda Setting Processes’ in N. Zahariadis (eds.) Handbook of Public Policy Agenda-Setting (Cheltenham: Edward Elgar). Caldwell, D. (1977) ‘Bureaucratic Foreign Policy-Making’, American Behavioral Scientist, 21: 87–110. Cammack, P. (2004) ‘What the World Bank Means by Poverty Reduction, and Why it Matters’, New Political Economy, 9, 2: 189–211.

Campbell, J. (2002) ‘Ideas, Politics and Public Policy’, Annual Review of Sociology, 28: 21–38. Capoccia, G. & Kelemen, D. (2007) ‘The Study of Critical Junctures: Theory, Narrative, and Counterfactuals in Historical Institutionalism’, World Politics, 59, 3: 341–369. Carey, G., Dickinson, H. & Cox, E. (2018) ‘Feminism, Gender, and Power Relations in Policy – Starting a New Conversation’, Australian Journal of Public Administration, 77, 4: 519–524. Carey, G., Dickinson, H. & Olney, S. (2019) ‘What Can Feminist Theory Offer Policy Implementation Challenges?’ Evidence & Policy, 15, 1: 143–159. Carey, G. & Harris, P. (2015) ‘Developing Management Practices to Support Joined-up Governance’, Australian Journal of Public Administration, 75, 1: 112– 118. Carter, E. & Whitworth, A. (2015) ‘Creaming and Parking in Quasi-Marketised Welfare-to-Work Schemes’ Journal of Social Policy, 44, 2: 277–296. Chakrabarti, R. & Sanyal, K. (2017) Shaping Policy in India (Oxford: Oxford University Press). Chalmers, A. (1999) What is this Thing Called Science? 3rd edn (Buckingham: Open University Press). Chan, K. N. & Zhao, S. (2016) ‘Punctuated Equilibrium and the Information Disadvantage of Authoritarianism: Evidence from the People’s Republic of China’, Policy Studies Journal 44, 2: 134–155. Chappell, L. (2006a) ‘Comparing Political Institutions: Revealing the Gendered “Logic of Appropriateness”’, Politics & Gender, 2, 2: 223–248. Chappell, L. (2006b) ‘Moving to a Comparative Politics of Gender?’ Politics & Gender, 2, 2: 221–263. Chappell, L. & Mackay, F. (2017) ‘What’s in a Name? Mapping the Terrain of Informal Institutions and Gender Politics’ in G. Waylen (ed.) Gender and Informal Institutions 23–44 (London: Rowman Littlefield International). Chappell, L. & Waylen, G. (2013) ‘Gender and the Hidden Life of Institutions’, Public Administration, 91, 3: 599–615. Christopoulos, D. & Ingold, K. (2015) ‘Exceptional or Just Well Connected? Political Entrepreneurs and Brokers in Policy Making’, European Political Science Review, 7, 3: 475–498. Church, C. & Dardanelli, P. (2005) ‘The Dynamics of Confederalism and Federalism: Comparing Switzerland and the EU’, Regional and Federal Studies, 15, 2: 163–185. Chwaszcza, C. (2008) ‘Game Theory’ in D. Della Porta & M. Keating (eds.) Approaches and Methodologies in the Social Sciences: A Pluralist Approach (Cambridge: Cambridge University Press). Clancy, P. (1996) ‘The New Zealand Experiment: A Canadian Perspective’, Electronic Journal of Radical Organization Theory, 2, 1, http://www.mngt.waikato.ac.nz/ejrot/Vol2_1/clancy.pdf

Cobb, R. & Elder, C. (1972) Participation in American Politics: The Dynamics of Agenda-Building (Boston: Allyn & Bacon). Cohen, M., March, J. & Olsen, J. (1972) ‘A Garbage Can Model of Organizational Choice’, Administrative Science Quarterly, 17, 1: 1–25. Cohn, C. (ed.) (2013) Women and Wars: Contested Histories, Uncertain Futures (London: Wiley). Colebatch, H. (1998) Policy (Buckingham: Open University Press). Colebatch, H. (2006) ‘Mapping the Work of Policy’ in H. Colebatch (ed.) Beyond the Policy Cycle: The Policy Process in Australia (Crow’s Nest, NSW: Allen & Unwin). Common, R. K. (1998) ‘Convergence and Transfer: A Review of the Globalisation of New Public Management’, International Journal of Public Sector Management, 11, 6: 440–450. Compton, M. & ‘t Hart, P. (2019) Great Policy Successes: How Governments Get It Right in a Big Way at Least Some of the Time (Oxford: Oxford University Press). Cook, B. (2018) ‘Old and New: Street-level Bureaucracy’, Administration and Society, 50, 8: 1071. Cooper, B. (2017) Beyond Respectability: The Intellectual Thought of Race Women (Chicago: University of Illinois Press). Cox, M., Arnold, G. & Villamayor Tomás, S. (2010) ‘A Review of Design Principles for Community-based Natural esource Management’, Ecology and Society 15, 4: 38 http://www.ecologyandsociety.org/vol15/iss4/art38/ Craib, I. (1992) Anthony Giddens (London: Routledge). Craig, C. (2019) How Does Government Listen to Scientists? (London: Palgrave Macmillan). Crenson, M. (1971) The Un-politics of Air Pollution: A Study of Non-decisionmaking in the Cities (London: Johns Hopkins Press). Crespy, A. & Schmidt, V. (2014) ‘The Clash of Titans: France, Germany, and the Discursive Double Game of EMU Reform’, Journal of European Public Policy, 21, 8: 1085–1101. Crossley, S. (2017) In Their Place: The Imagined Geographies of Poverty (London: Pluto Press). Crow, D. & Jones, M. (2018) ‘Narratives as Tools for Influencing Policy Change’, Policy & Politics, 46, 2: 217–234. Crowley, K., Stewart, J., Kay, A. & Head, B. (2019) Reconsidering Policy: Complexity, Governance and the State (Bristol: Policy Press). Curry, D. (2015) Network Approaches to Multi-level Governance: Structures, Relations and Understanding Power Between Levels (London: Palgrave Macmillan). Curtin, J. (2018) Women and Trade Unions: A Comparative Perspective (London: Routledge). Dahl, R. (1957) ‘The Concept of Power’, Behavioral Science, 2, 3: 201–215.

Dahl, R. (1958) ‘A Critique of the Ruling Elite Model’, The American Political Science Review, 52, 2: 463–469. Dahl, R. (1961) Who Governs? Democracy and Power in an American City (New Haven, CT: Yale University Press). Daniell, K. & Kay, A. (2017) Multi-level Governance: Conceptual Challenges and Case Studies from Australia (Acton, ACT: ANU Press). Davidson, B. (2017) ‘Storytelling and Evidence-based Policy: Lessons from the Grey Literature’, Palgrave Communications, 3: 17093, doi:10.1057/palcomms.2017.93 Davis, O., Dempster, M. and Wildavsky, A. (1966) ‘A theory of the budgetary process’, American Political Science Review, 60, 3: 529-547 Davidson, D. (1980) Essays on Actions and Events (Oxford: Clarendon Press). Dawkins, R. (1976) The Selfish Gene (New York: Oxford University Press). Dawson, R. & Robinson, J. (1963) ‘Inter-Party Competition, Economic Variables and Welfare Policies in the American States’, Journal of Politics, 25: 265–289. Day, P. & Klein, R. (2000) ‘The Politics of Managing the Health Service’ in R. Rhodes (ed.) Transforming British Government Vol. 1 (London: Macmillan). DBERR (Department for Business, Enterprise and Regulatory Reform) (2008) Understanding the Public Services Industry: How Big, How Good, Where Next? (The Julius Report) http://www.berr.gov.uk/files/file46965.pdf de Leeuw, E. (2017) ‘Engagement of Sectors Other than Health in Integrated Health Governance, Policy, and Action’, Annual Review of Public Health, 38: 329–349. Dearing, J. W. & Rogers, E. M. (1996) Agenda Setting (London: Sage). deLeon, P. (1978) ‘A Theory of Policy Termination’ in J. May & A. Wildavsky (eds.) The Policy Cycle (London: Sage). deLeon, P. (1999) ‘The Missing Link Revisited’, Review of Policy Research, 16, 3– 4: 311–338. deLeon, P. & deLeon, L. (2002) ‘Whatever Happened to Policy Implementation? An Alternative Approach’, Journal of Public Administration Research and Theory, 12, 4: 467–492. Della Porta, D. & Keating, M. (2008a) ‘Introduction’ in D. Della Porta & M. Keating (eds.) Approaches and Methodologies in the Social Sciences (Cambridge: Cambridge University Press). Della Porta, D. & Keating, M. (2008b) ‘How Many Approaches in the Social Sciences? An Epistemological Introduction’, in D. Della Porta & M. Keating (eds.) Approaches and Methodologies in the Social Sciences (Cambridge: Cambridge University Press). Deutsch, K. (1970) Politics and Government (Boston: Houghton Mifflin). Dewan, T., Dowding, K. & Shepsle, K. (2009) ‘Editors’ Introduction: Rational Choice Classics in Political Science’ in K. Dowding, T. Dewan & K. Shepsle (eds.) Rational Choice Politics, Volume 1 (London: Sage).

Dickinson, H. & Sullivan, H. (2014) ‘Towards a General Theory of Collaborative Performance: The Importance of Efficacy and Agency’, Public Administration, 92, 1: 161–177. Dietz, T., Ostrom, E. & Stern, P. (2003) ‘The Struggle to Govern the Commons’, Science 302, 5652: 1907–1912. DiMaggio, P. & Powell, W. (1983) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, American Sociological Review, 48, 2: 147–160. Dinan, D. (2004) Europe Recast: A History of European Union (Boulder, CO: Lynne Rienner). Dodge, J. (2015) ‘Indication and Inference: Reflections on the challenge of mixing paradigms in the Narrative Policy Framework’, Critical Policy Studies, 9, 3: 361– 367. Dodge, J. (2017) ‘Crowded Advocacy: Framing dynamic in the fracking controversy in New York’, VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 28, 3: 888–915. Dolowitz, D. (2000) ‘Introduction’, Governance, 13, 1: 1–4. Dolowitz, D. (2003) ‘A Policy-Maker’s Guide to Policy Transfer’, Political Quarterly, 74, 1: 101–108. Dolowitz, D. (2010) ‘Learning, Information and Policy Transfer Process: What do Governments Learn when Shopping for Ideas?’, Conference paper, Policy Learning and Policy Transfer in Multilevel Systems, Edinburgh, January. Dolowitz, D., Greenwold, S. & Marsh, D. (1999) ‘Policy Transfer: Something Old, Something New, Something Borrowed, But Why Red, White and Blue?’, Parliamentary Affairs, 52, 4: 719–730. Dolowitz, D. & Marsh, D. (1996) ‘Who Learns What From Whom: A Review of the Policy Transfer Literature’, Political Studies, 44: 343–357. Dolowitz, D. & Marsh, D. (2000) ‘Learning from Abroad: The Role of Policy Transfer in Contemporary Policy-Making’, Governance, 13, 1: 5–24. Dommett, K. & Flinders, M. (2015) ‘The Centre Strikes Back: Meta‐governance, Delegation, and the Core Executive in the United Kingdom, 2010–14’, Public Administration, 93, 1: 1–16. Douglas, H. (2009) Science, Policy, and the Value-free Ideal (Pittsburgh, PA: University of Pittsburgh Press). Dowding, K. (1991) Rational Choice and Political Power (Aldershot: Edward Elgar). Dowding, K. (1994) ‘The Compatibility of Behaviouralism, Rational Choice and “New Institutionalism”’, Journal of Theoretical Politics, 6, 1: 105–117. Dowding, K. (1995a) ‘Interpreting Formal Coalition Theory’ in K. Dowding & A. King (eds.) Preferences, Institutions and Rational Choice (Oxford: Oxford University Press).

Dowding, K. (1995b) ‘Model or Metaphor? A Critical Review of the Policy Network Approach’, Political Studies, 43, 2: 136–158. Dowding, K. (1995c) The Civil Service (London: Routledge). Dowding, K. (1996) Power (Buckingham: Open University Press). Dowding, K. (2000) ‘How Not to Use Evolutionary Theory in Politics: A Critique of Peter John’, British Journal of Politics and International Relations, 2, 1: 72–80. Dowding, K. (2001) ‘There Must be End to Confusion: Policy Networks, Intellectual Fatigue, and the need for Political Science Methods Courses in British Universities’, Political Studies, 49, 1: 89–105. Dowding, K. (2003) ‘Resources, Power and Systematic Luck: A Response to Barry’, Politics Philosophy Economics, 2: 305–312. Dowding, K. (2009) ‘Luck, Equality and Responsibility’, Critical Review of International Social and Political Philosophy, 12, 4: 561–582. Dowding, K. (2010) ‘Rational Choice Theory’, in M. Bevir (ed.) Sage Handbook of Governance (Beverly Hills, CA: Sage). Dowding, K. & James, O. (2004) ‘Analysing Bureau-Shaping Models: Comments on Marsh, Smith and Richards’, British Journal of Political Science, 34, 1: 183– 189. Dowding, K. & King, A. (1995) ‘Introduction’ in K. Dowding & A. King (eds.) Preferences, Institutions and Rational Choice (Oxford: Oxford University Press). Downs, A. (1967) Inside Bureaucracy (Boston: Little, Brown). Downs, A. (1972) ‘Up and Down With Ecology: The “Issue-Attention Cycle”’, The Public Interest, 28, Summer: 38–50, http://www.anthonydowns.com/upanddown.htm Drew, J. (2019) ‘How Losers Can Turn into Winners in Disputatious Public Policy: A Heuristic for Prospective Herestheticians’, Australian Journal of Political Science, 67, 1: 167–182. Driscoll, A. & Krook, M. L. (2009) ‘Can There be a Feminist Rational Choice Institutionalism?’ Politics & Gender, 5, 2: 238–245. Dror, Y. (1964) ‘Muddling Through – “Science” or Inertia?’, Public Administration Review, 24: 153–157. Dubois, E. (2016) ‘Political Business Cycles 40 Years After Nordhaus’, Public Choice, 166, 1–2: 235–259. Duffy, D. (2017) Evaluation and Governing in the 21st Century (London: Palgrave Pivot). Duggett, M. (2009) ‘The Return of the Westminster Supermodel’, Public Money & Management, 29, 1: 7–8. Duina, F. (1997) ‘Explaining Legal Implementation in the European Union’, International Journal of the Sociology of Law, 25: 155–179. Dunleavy, P. (1985) ‘Bureaucrats, Budgets and the Growth of the State: Reconstructing an Instrumental Model’, British Journal of Political Science, 15, 3: 299–328.

Dunleavy, P. (1990) ‘The Westland Affair: Theories of the State and Core Executive Decision Making’, Public Administration, 68, 1: 29–60. Dunleavy, P. (2019) “The Bureaucracy” as an Interest Group’ in R. Congleton, B. Grofman & S. Voigt (eds.) The Oxford Handbook of Public Choice (Oxford: Oxford University Press). Dunleavy, P., Margetts, H., Bastow, S. & Tinkler, J. (2006) ‘New Public Management is Dead – Long Live Digital-Era Governance’, Journal of Public Administration Research and Theory, 16, 3: 467–494. Dunlop, C. (2014) ‘The Possible Experts: How Epistemic Communities Negotiate Barriers to Knowledge Use in Ecosystems Services Policy’, Environment and Planning C: Government and Policy, 32, 2: 208–228. Dunlop, C. (2017a) ‘Policy Learning and Policy Failure: Definitions, Dimensions and Intersections’, Policy & Politics 45, 1: 19–37. Dunlop, C. (2017b) ‘Pathologies of Policy Learning: What are they and How do they Contribute to Policy Failure?’, Policy & Politics 45, 1: 3–18. Dunlop, C. (2017c) ‘The Irony of Epistemic Learning: Epistemic Communities, Policy Learning and the Case of Europe’s Hormones Saga’, Policy and Society, 36, 2: 215–232, doi: 10.1080/14494035.2017.1322260 Dunlop, C. & Radaelli, C. (2013) ‘Systematising Policy Learning: From Monolith to Dimensions’, Political Studies Review, 61, 3: 599–619. Dunlop, C. & Radaelli, C. (2015) ‘Overcoming Illusions of Control: How to Nudge and Teach Regulatory Humility’ in A. Alemanno & A. Sibony (eds.) Nudge and the Law (London: Bloomsbury). Dunlop, C. & Radaelli, C. (2016) ‘Policy Learning in the Eurozone Crisis: Modes, Power and Functionality’, Policy Sciences, 49, 2: 107–124. Dunlop, C. & Radaelli, C. (2017) ‘Learning in the Bath-tub: The Micro and Macro Dimensions of the Causal Relationship between Learning and Policy Change’, Policy and Society, 36, 2: 304–319, doi: 10.1080/14494035.2017.1321232 Dunlop, C. & Radaelli, C. (2018) ‘The Lessons of Policy Learning: Types, Triggers, Hindrances and Pathologies’, Policy & Politics, 46, 2: 255–272. Dunlop, C., Radaelli, C. & Trein, P. (2018) ‘Introduction: The Family Tree of Policy Learning’ in C. Dunlop, C. Radaelli & P. Trein. (eds.) (2018) Learning in Public Policy (London: Palgrave Macmillan). Durant, R. & Diehl, P. (1989) ‘Agendas, Alternatives, and Public Policy: Lessons from the U.S. Foreign Policy Arena’, Journal of Public Policy, 9, 2: 179–205. Durnová, A. (2018) ‘Understanding Emotions in Policy Studies through Foucault and Deleuze’, Politics and Governance, 6, 4, 95 -102 Durose, C., Needham, C., Mangan, C. & Rees, J. (2017) ‘Generating “Good Enough” Evidence for Co-production’, Evidence & Policy, 13, 1: 135–151. Durose, C. & Richardson, L. (eds.) (2015) Designing Public Policy for Coproduction (Bristol: Policy Press). Durose, C., Richardson, L., Combs, R., Eason, C. & Gains, F. (2012) “Acceptable Difference”: Diversity, Representation and Pathways to UK Politics’,

Parliamentary Affairs, 66, 2: 246–267. Dussauge-Laguna, M. (2012a) ‘On The Past and Future of Policy Transfer Research: Benson and Jordan Revisited’, Political Studies Review, 10: 313– 324. Dussauge-Laguna, M. (2012b) ‘The Neglected Dimension: Bringing Time Back into Cross-national Policy Transfer Studies’, Policy Studies, 33, 6: 567–585. Dye, T. (1966) Politics, Economics and the Public (Chicago: Rand McNally). Easton, D. (1953) The Political System (New York: Albert Knopf). Easton, D. (1965) A Systems Analysis of Political Life (New York: Wiley). Elazar, D. (1971) ‘From the Editor: The Themes of a Journal of Federalism’, Publius, 1, 1: 3–9. Eller, W. & Krutz, G. (2009) ‘Editor’s Notes: Policy Process, Scholarship and the Road Ahead: An Introduction to the 2008 Policy Shootout!’, Policy Studies Journal, 37, 1: 1–4. Emejulu, A. (2016) Community Development as Micropolitics (Bristol: Policy Press). Emejulu, A. (2018) ‘On the problems and possibilities of feminist solidarity’, IPPR Progressive Review, 24, 4: 267–273. Emejulu, A. (2019) ‘Can Political Science Decolonise? A Response to Neema Begum and Rima Saini’ Political Studies Review, 17, 2, 202–6. Emejulu, A. (2019) ‘Can Political Science Decolonise? A Response to Neema Begum and Rima Saini’, Political Studies Review, doi: 10.1177/ 1478929918808999 Engeli, I., Green-Pederson, C. & Larsen, L. T. (eds.) (2012) Morality Politics in Western Europe (Basingstoke: Palgrave Macmillan). Epp, D. (2017) ‘Public policy and the wisdom of crowds’, Cognitive Systems Research, 43: 53–56. Erikson, J. (2017) Criminalising the Client: Institutional Change, Gendered Ideas and Feminist Strategies (London: Rowman & Littlefield). Etheredge, L. (1981) ‘Government Learning: An Overview’, in L. Long (ed.) The Handbook of Political Behavior, Volume 2 (New York: Pergamon). Etzioni, A. (1967) ‘Mixed Scanning: A “Third” Approach to Decision Making’, Public Administration Review, 27: 385–392. European Commission (2011) ‘Policy cycle, EU policy cycle’, http://ec.europa.eu/governance/better_regulation/glossary_en.htm#_P Evans, A. (2018) ‘The Perils of Male Bias: Alice Evans replies to yesterday’s “Sausagefest”’, From Poverty to Power 11 January https://oxfamblogs.org/fp2p/the-perils-of-male-bias-alice-evans-replies-toyesterdays-sausagefest/ Evans, E. (2014) The Politics of Third Wave Feminisms (London: Palgrave Macmillan). Evans, M. & Barakat, S. (2012) ‘Post-war Reconstruction, Policy Transfer and the World Bank: The Case of Afghanistan’s National Solidarity Programme’, Policy

Studies, 33, 6: 541–565. Evans, M. & Davies, J. (1999) ‘Understanding Policy Transfer: A Multi-Level, MultiDisciplinary Perspective’, Public Administration, 77, 2: 361–385. Everett, S. (2003) ‘The Policy Cycle: Democratic Process or Rational Paradigm Revisited?’, Australian Journal of Public Administration 62, 2: 65–70. Exadaktylos, T. & Radaelli, C. (2009) ‘Research Design in European Studies: The Case of Europeanization’, Journal of Common Market Studies, 47, 3: 507–530. Eyestone, R. (1977) ‘Confusion, Diffusion, and Innovation’, American Political Science Review, 71, 2: 441–447. Fabbrini, S. (2013) ‘Intergovernmentalism and its Limits: Assessing the European Union’s Answer to the Euro Crisis’, Comparative Political Studies 20, 10: 1–27. Fabbrini, S. & Gilbert, M. F. (2000) ‘When Cartels Fail: The Role of the Political Class in the Italian Democratic Transition’, Government and Opposition 35, 1: 27–48. Fadlallah, R., El-Jardali, F., Nomier, M., Hemadi, N., Arif, K., Langlois, E. V. & Akl, E. A. (2019) ‘Using Narratives to Impact Health Policy-making: A Systematic Review’, Health Research Policy and Systems, 17, 26: 1–22. Fairclough, N. (2015) Language and Power 3rd edn (London: Routledge). Faling, M., Biesbroek, R., Karlsson-Vinkhuyzen, S. & Termeer, K. (2019) ‘Policy Entrepreneurship Across Boundaries: A Systematic Literature Review’, Journal of Public Policy, 39, 2: 393–422. Falkner, G., Hartlapp, M., Leiber, S. & Treib, O. (2004) ‘Non-Compliance with EU Directives in the Member States: Opposition through the Backdoor?’, West European Politics, 27, 3: 452–473. Feeny, D., Berkes, F., McCay, B. & Acheson, J. (1990) ‘The Tragedy of the Commons: Twenty-two Years Later’, Human Ecology 18, 1: 1–19. Feiock, R. (2009) Metropolitan governance and institutional collective action, Urban Affairs Review, 44, 3: 356–377. Feiock, R. (2013) ‘The Institutional Collective Action Framework’, Policy Studies Journal 41, 3: 397–425. Feldman, E. & Bayer, R. (2004) ‘Introduction’ in E. Feldman & R. Bayer (eds.) Unfiltered: Conflicts Over Tobacco Policy and Public Health (London: Harvard University Press). Fenger, M. & Klok, P. (2001) ‘Interdependency, Beliefs, and Coalition Behavior: A Contribution to the Advocacy Coalition Framework’, Policy Sciences 34, 2: 157– 170. Ferreira, F. (2015) ‘The international poverty line has just been raised to $1.90 a day, but global poverty is basically unchanged. How is that even possible?’ The World Bank https://blogs.worldbank.org/developmenttalk/international-povertyline-has-just-been-raised-190-day-global-poverty-basically-unchanged-howeven Finer, S. (1972) Adversary Politics and Electoral Reform (London: Wigram).

Finkelstein, L. (1995) ‘What is Global Governance?’, Global Governance 1, 1: 367–392. Fischer, F. (1998) ‘Beyond Empiricism: Policy Inquiry in Postpositive Perspective’, Policy Studies Journal, 26, 1: 129–146. Fischer, F. (2003) Reframing Public Policy (Oxford: Oxford University Press). Fischer, F. & Forester, J. (eds.) (1993) The Argumentative Turn in Policy Analysis (London: UCL Press). Fischer, F., Torgerson, D., Durnová, A. & Orsini, M. (2015) ‘Introduction to Critical Policy Studies’ in F. Fischer, D. Torgerson, A. Durnová & M. Orsini (eds.) Handbook of Critical Policy Studies (Cheltenham: Edward Elgar Publishing). Flinders, M. (2008) Delegated Governance and the British State (Oxford: Oxford University Press). Flinders, M. (2009) ‘The Politics of Patronage in the United Kingdom: Shrinking Reach and Diluted Permeation’, Governance, 22, 4: 547–570. Flinders, M. & Skelcher, C. (2012) ‘Shrinking the Quango State’, Public Money & Management, 32, 5: 327–334. Fonow, M. & Cook, J. (2005) ‘Feminist Methodology: New Applications in the Academy and Public Policy’, Signs, 30, 4: 2211–2236. Foucault, M. (1977) Discipline and Punish (Harmondsworth: Penguin). Freeman, G. (1985) ‘National Styles and Policy Sectors: Explaining Structured Variation’, Journal of Public Policy, 5, 4: 467–496. Freeman, G. (2006a) ‘Politics and Mass Immigration’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Freeman, R. (2006b) ‘Learning in public policy’ in M. Moran, M. Rein & R. E. Goodin (eds.) Oxford Handbook of Public Policy (Oxford: Oxford University Press). Freeman, R. (2009) ‘What is Translation?’ Evidence & Policy, 5, 4: 429–447. Freidenvall, L. & Krook, M. L. (2011) ‘Discursive Strategies for Institutional Reform: Gender Quotas in Sweden and France’ in M. L. Krook & F. Mackay (eds.) Gender, Politics and Institutions (Basingstoke: Palgrave Macmillan). Friedman, M. (1953) ‘The Methodology of Positive Economics’ in Essays in Positive Economics (Chicago: University of Chicago Press). Friedman, T. (2000) The Lexus and the Olive Tree: Understanding Globalization (New York: Anchor Press). Frischmann, B. (2018) ‘The Tragedy of the Commons, Revisited’, Scientific American, 19th November https://blogs.scientificamerican.com/observations/thetragedy-of-the-commons-revisited/?amp Fuchs, G. (2015) ‘Building the Agenda for Carbon Dioxide Capture and Storage: Limits of EU-Activism’ in J. Tosun, S. Biesenbender & K. Schulze (eds.) Energy Policy Making in the EU (London: Springer). Gains, F. & Stoker, G. (2009) ‘Delivering “Public Value”: Implications for Accountability and Legitimacy’, Parliamentary Affairs, 62, 3: 438–455.

Galligan, B. (2006) ‘Comparative Federalism’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Gamble, A. (1990) ‘Theories of British Politics’, Political Studies, 38, 3: 404–420. Gangopadhyay, P. (ed.) (2017) Economics of Globalisation (London: Routledge). Gava, R. (2016) ‘Financial Regulation and Agenda Dynamics: Impacts of the Global Financial Crisis’ in N. Zahariadis (ed.) Handbook of Public Policy Agenda-Setting (Cheltenham: Edward Elgar). Genieys, W. & Smyrl, M. (2008a) ‘The Problem of Policy Change’ in W. Genieys & M. Smyrl (eds.) Elites, Ideas and the Evolution of Public Policy (Basingstoke: Palgrave Macmillan). Genieys, W. & Smyrl, M. (2008b) ‘Competing Elites, Legitimate Authority, Structured Ideas’ in W. Genieys & M. Smyrl (eds.) Elites, Ideas and the Evolution of Public Policy (Basingstoke: Palgrave Macmillan). Genieys, W. & Smyrl, M. (2008c) ‘Accounting for Change: The Role of Programmatic Elites’ in W. Genieys & M. Smyrl (eds.) Elites, Ideas and the Evolution of Public Policy (Basingstoke: Palgrave Macmillan). Geva-May, I. (2004) ‘Riding the Wave of Opportunity: Termination in Public Policy’, Journal of Public Administration Research and Theory, 14, 3: 309–333. Geyer, R. (2012) ‘Can Complexity Move UK Policy beyond “Evidence-Based Policy Making” and the “Audit Culture”? Applying a “Complexity Cascade” to Education and Health Policy’, Political Studies, 60, 1: 20–43. Geyer, R. & Rihani, S. (2010) Complexity and Public Policy (London: Routledge). Giddens, A. (1984) The Constitution of Society (Cambridge: Polity). Giest, S. (2017) ‘Overcoming the Failure of “Silicon Somewheres”: Learning in Policy Transfer Processes’, Policy & Politics, 45, 1: 39–54. Gigerenzer, G. (2001) ‘The Adaptive Toolbox’ in G. Gigerenzer & R. Selton (eds.) Bounded Rationality: The Adaptive Toolbox (Cambridge, MA: MIT Press). Gigerenzer, G. (2015) Risk Savvy: How to Make Good Decisions (London: Penguin). Givel, M. (2015) ‘What’s the Big Deal? Complexity Versus Traditional US Policy Approaches’ in R. Geyer & P. Cairney (eds.) Handbook on Complexity and Public Policy (Cheltenham: Edward Elgar). Glencross, A. (2014) Politics of European Integration: Political Union or a House Divided? (London: Wiley). Gluckman, P. (2016) ‘The Science-policy Inter-face’, Science, Sep 2, 353, 6303: 969. Goggin, M., Bowman, A., Lester, J. & O’Toole, L. (1990) Implementation Theory and Practice: Towards a Third Generation (Boston: Little, Brown). Goldfinch, S. & Wallis, J. (2010) ‘Two Myths of Convergence in Public Management Reform’, Public Administration, 88, 4: 1099–1115. Goldsmith, M. & Page, E. (1997) ‘Farewell to the British State?’ in J. Lane (ed.) Public Sector Reform (London: Sage).

Goodin, R. & Tilly, C. (2006) The Oxford Handbook of Contextual Political Analysis (Oxford: Oxford University Press). Gormley-Heenan, C. & Birrell, D. (2016) Multi-level Governance and Northern Ireland (London: Palgrave Macmillan). Goyal, N. & Howlett, M. (2018) ‘Lessons Learned and Not Learned: Bibliometric Analysis of Policy Learning’ in C. Dunlop, C. Radaelli & P. Trein (eds.) Learning in Public Policy (London: Palgrave Macmillan). Gramsci, A. (1971) Selections from Prison Note-books (London: Lawrence & Wishart). Gray, C. (2000) ‘A “Hollow State”?’ in R. Pyper & L. Robins (eds.) United Kingdom Governance (London: Red Globe Press), pp. 283–99. Gray, G. & Jones, M. (2016) ‘A Qualitative Narrative Policy Framework? Examining the Policy Narratives of US Campaign Finance Regulatory Reform’, Public Policy and Administration, 31, 3: 193–220. Greenaway, J., Smith, S. & Street, J. (1992) Deciding Factors in British Politics (London: Routledge). Greener, I. (2002) ‘Understanding NHS Reform: The Policy-Transfer, Social Learning and Path-Dependency Perspectives’, Governance, 15, 2: 161–183. Greener, I. (2005) ‘The Potential of Path Dependence in Political Studies’, Politics, 25, 1: 62–72. Green-Pederson, C. & Walgrave, S. (eds.) (2014) Agenda Setting, Policies, and Political Systems (Chicago: Westview Press). Greenwood, J., Pyper, R. & Wilson, D. (2001) New Public Administration in Britain (London: Routledge). Greer, P. (1994) Transforming Central Government: The Next Steps Initiative (Buckingham: Open University Press). Greer, S. (2004), Territorial Politics and Health Policy (Manchester: Manchester University Press). Grey, J. (1973) ‘Innovation in the States: A Diffusion Study’, American Political Science Review, 67, 4: 1174–1185. Greer, S. & Jarman, H. (2008) ‘Devolution and policy styles’ in Trench, A. (ed.) The State of the Nations (Exeter: Imprint Academic), 167–197. Griggs, S. (2007) ‘Rational Choice Theory in Public Policy: The Theory in Critical Perspective’ in F. Fischer, G. Miller & M. Sidney (eds.) Handbook of Public Policy Analysis (London: CRC Press). Guerrina, R., Haastrup, T., Wright, K., Masselot, A., MacRae, H. & Cavaghan, R. (2018) ‘Does European Union Studies Have a Gender Problem? Experiences from Researching Brexit’, International Feminist Journal of Politics, 20, 2: 252– 257, doi: 10.1080/14616742.2018.1457881 Haas, E. B. (1958) The Uniting of Europe: Political, Economic, and Social Forces 1950–1957 (Stanford, CA: Stanford University Press). Haas, E. B. (1975) The Obsolescence of Regional Integration Theory (Institute of International Studies, University of California, Research Series No. 25).

Haas, P. M. (1992) ‘Introduction: Epistemic Communities and International Policy Coordination’, International Organization, 46, 1: 1–35. Habermas, J. (1984) The Theory of Communicative Action (Boston: Beacon Press). Haidt, J. (2001) ‘The Emotional Dog and Its Rational Tail’, Psychological Review 108, 4: 814–834. Haidt, J. (2007) ‘The New Synthesis in Moral Psychology’, Science, 316: 998– 1002. Haidt, J. (2012) The Righteous Mind: Why Good People are Divided by Politics and Religion (New York: Pantheon). Hajer, M. (1995) The Politics of Environmental Discourse: Ecological Modernization and the Policy Process (Oxford: Oxford University Press). Hale, D. (1988) ‘What is a Policy, Anyway? And Who’s Supposed to Make it? A Survey of the Public Administration and Policy Texts’, Administration & Society, 4: 423–452. Hale, T. (2018) Catalytic Cooperation (Oxford: BSG Working Paper Series). Hall, B. L. & Tandon, R. (2017) ‘Decolonization of Knowledge, Epistemicide, Participatory Research and Higher Education’, Research for All, 1, 1: 6–19. Hall, P. (1986) Governing the Economy (Oxford: Oxford University Press). Hall, P. (1993) ‘Policy Paradigms, Social Learning, and the State: The Case of Economic Policymaking in Britain’, Comparative Politics, 25, 3: 275–296. Hall, P. & Taylor, R. (1996) ‘Political Science and the Three New Institutionalisms’, Political Studies, 44, 4: 936–957. Hardin, G. (1968) ‘The Tragedy of the Commons’, Science, 162, December: 1243– 1248. Harsanyi, J. (1986) ‘Advances in Understanding Rational Behavior’ in J. Elster (ed.) Rational Choice (Oxford: Blackwell). Hay, C. (2002) Political Analysis: A Critical Introduction (London: Red Globe Press). Hay, C. (2004a) ‘Theory, Stylized Heuristic or Self-fulfilling Prophecy? The Status of Rational Choice Theory in Public Administration’, Public Administration, 82, 1: 39–62. Hay, C. (2004b) ‘“Taking Ideas Seriously” in Explanatory Political Analysis’, British Journal of Politics and International Relations, 6: 142–149. Hay, C. (2006a) ‘Constructivist Institutionalism’, in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Hay, C. (2006b) ‘Globalization and Public Policy’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Hay, C. (2009) ‘King Canute and the “Problem” of Structure and Agency: On Times, Tides and Heresthetics’, Political Studies, 57, 2: 260–279.

Hay, C. & Wincott, D. (1998) ‘Structure, Agency and Historical Institutionalism’, Political Studies, 46: 951–957. Hayes, M. T. (2001) The Limits of Policy Change: Incrementalism, Worldview, and the Rule of Law (Washington, DC: Georgetown University Press). Haynes, P. (2008) ‘Complexity Theory and Evaluation in Public Management’, Public Management Review, 10, 3: 401–419. Head, B. (2008) ‘Three Lenses of Evidence-Based Policy’, Australian Journal of Public Administration, 67, 1: 1–11. Heclo, H. (1974). Modern Social Politics in Britain and Sweden (New Haven, CT: Yale University Press). Heclo, H. (1978) ‘Issue Networks and the Executive Establishment’, in A. King (ed.) The New American Political System (Washington, DC: American Enterprise Institute). Heichel, S., Pape, J. & Sommerer, T. (2005) ‘Is there Convergence in Convergence Research? An Overview of Empirical Studies on Policy Convergence’, Journal of European Public Policy, 12, 5: 817–840. Heikkila, T. & Andersson, K. (2018) ‘Policy Design and the Added-value of the Institutional Analysis Development Framework’, Policy and Politics, 46, 2: 309– 324. Heikkila, T. & Cairney, P. (2018) ‘Comparison of Theories of the Policy Process’ in C. Weible & P. Sabatier (eds.) Theories of the Policy Process 4th edn (Chicago: Westview Press). Heikkila, T. & Carter, D. (2017) ‘Common Pool Resources’, Oxford Bibliographies, doi: 10.1093/OBO/9780199363445-0011 Heikkila, T. & Gerlak, A. (2013) ‘Building a Conceptual Approach to Collective Learning: Lessons for Public Policy Scholars’, Policy Studies Journal, 41, 3: 484–512. Heikkila, T. & Gerlak, A. (2018) ‘Working on Learning: How the Institutional Rules of Environmental Governance Matter’, Journal of Environmental Planning and Management, https://doi.org/10.1080/09640568.2018.1473244 Heikkila, T., Pierce, J., Gallaher, S., Kagan, J., Crow, D. & Weible, C. (2014) ‘Understanding a Period of Policy Change: The Case of Hydraulic Fracturing Disclosure Policy in Colorado’, Review of Policy Research, 31, 2: 65–87. Heikkila, T., Schlager, E. & Davis, M. (2011) ‘The Role of Cross‐scale Institutional Linkages in Common Pool Resource Management’, Policy Studies Journal 39, 1: 121–145. Heikkila, T. & Weible, C. (2017) ‘Unpacking the Intensity of Policy Conflict: A Study of Colorado’s Oil and Gas Subsystem’, Policy Sciences, 50, 2: 179–193. Henry, D., Ingold, K., Nohrstedt, D. & Weible, C. (2014) ‘Policy Change in Comparative Contexts: Applying the Advocacy Coalition Framework Outside of Western Europe and North America’, Journal of Comparative Policy Analysis, 16, 4: 299–312.

Henry, G. T. & Mark, M. M. (2003) ‘Beyond Use: Understanding Evaluation’s Influence on Attitudes and Actions’, American Journal of Evaluation, 24, 3: 293– 314. Herd, P. & Moynihan, D. (2018) Administrative Burden (New York: Russell Sage). Herweg, N., Huß, C. & Zohlnhöfer, R. (2015) ‘Straightening the Three Streams: Theorising Extensions of the Multiple Streams Framework’, European Journal of Political Research 54, 3: 435–449. Herweg, N., Zahariadis, N. & Zohlnhöfer, R. (2018) ‘The Multiple Streams Framework: Foundations, Refinements, and Empirical Applications’, in C. Weible & P. Sabatier (eds.) Theories of the policy process 4th edn (Chicago: Westview Press). Heyward, C. (2007) ‘Revisiting the Radical View: Power, Real Interests and the Difficulty of Separating Analysis from Critique’, Politics, 27,1: 48–54. Hill, K. (1997) ‘In Search of Policy Theory’, Policy Currents, 7, 1: 1–9, http://apsapolicysection.org/vol7_1/71.pdf Hill, M. (2005) The Public Policy Process 4th edn (Harlow: Pearson). Hill, M. (2009) The Public Policy Process 5th edn (Harlow: Pearson). Hill, M. & Hupe, P. (2002) Implementing Public Policy (London: Sage). Hill, M. & Hupe, P. (2009) Implementing Public Policy 2nd edn (London: Sage). Hill, M. & Varone, F. (2016) The Public Policy Process (London: Routledge). Hindess, B. (1988) Choice, Rationality and Social Theory (London: Unwin Hyman). Hindess, B. (1996) Discourses of Power: From Hobbes to Foucault (Oxford: Blackwell). Hindmoor, A. (2003) ‘Public Policy: The 2002 Spending Review and Beyond’, Parliamentary Affairs, 56, 2: 205–218. Hindmoor, A. (2006a) Rational Choice (London: Red Globe Press). Hindmoor, A. (2006b) ‘Public Choice’ in C. Hay, M. Lister & D. Marsh (eds.) The State: Theories and Issues (London: Red Globe Press). Hindmoor, A. (2010) ‘“Major Combat Operations Have Ended?” Arguing about Rational Choice, 1994–2009’, British Journal of Political Science, 41, 1: 191– 210. Hindmoor, A. & Taylor, B. (2015) Rational Choice 2nd edn (London: Red Globe Press). Hjern, B. (1982) ‘Implementation Research – the Link Gone Missing’, Journal of Public Policy, 2, 3: 301–308. Hjern, B. & Porter, D. (1981) ‘Implementation Structures: A New Unit of Administrative Analysis’, Organizational Studies, 2: 211–227. Hoberg, G. (1991) ‘Sleeping with an Elephant: The American Influence on Canadian Environmental Regulation’, Journal of Public Policy, 11, 1: 107–131. Hoberg, G. (1996) ‘Putting Ideas in Their Place: A Response to “Learning and Change in the British Columbia Forest Policy Sector”’, Canadian Journal of Political Science, 29, 1: 135–144.

Hoberg, G. (2001) ‘Globalization and Policy Convergence: Symposium Overview’, Journal of Comparative Policy Analysis: Research and Practice, 3: 127–132. Hofferbert, R. (1974) The Study of Public Policy (Indianapolis: Bobbs-Merrill). Hogwood, B. (1987) From Crisis to Complacency (Oxford: Oxford University Press). Hogwood, B. (1992a) Trends in British Public Policy (Buckingham: Open University Press). Hogwood, B. (1992b) ‘Ups and Downs: Is There an Issue-Attention Cycle in Britain?’, Strathclyde Papers on Government and Politics, 89 (Glasgow University of Strathclyde). Hogwood, B. (1997) ‘The Machinery of Government 1979–97’, Political Studies, 45, 4: 704–715. Hogwood, B. & Gunn, L. (1984) Policy Analysis for the Real World (Oxford: Oxford University Press). Hogwood, B., Judge, D. & McVicar, M. (2001) ‘Agencies, Ministers and Civil Servants in Britain’ in B. G. Peters & J. Pierre (eds.) Politicians, Bureaucrats and Administrative Reform (London: Routledge). Hogwood, B. & Peters, B. G. (1982) ‘The Dynamics of Policy Change: Policy Succession’, Policy Sciences, 14: 225–245. Hogwood, B. & Peters, B. G. (1983) Policy Dynamics (New York: St Martin’s Press). Holland, K., Morton, F. & Galligan, B. (1996) Federalism and the Environment (Westport, CT: Greenwood Press). Holliday, I. (2000) ‘Is the British State Hollowing Out?’, Political Quarterly, 71, 2: 167–176. Holzinger, K. & Knill, C. (2005) ‘Causes and Conditions of Cross-National Policy Convergence’, Journal of European Public Policy, 12, 5: 775–796. Hood, A. & Waters, T. (2017) Living standards, poverty and inequality in the UK: 2016–2017 to 2021–2022 (London: Institute for Fiscal Studies) http://dx.doi.org/10.1920/re.ifs.2017.0127 Hood, C. (1976) The Limits of Administration (London: Wiley). Hood, C. (1983) The Tools of Government (London: Macmillan). Hood, C. (1995) ‘The “New Public Management” in the 1980s: Variations on a Theme’, Accounting, Organizations and Society, 20, 2/3: 93–109. Hood, C. (2007) ‘Public Service Management by Numbers’, Public Money and Management, 27, 2: 95–102. Hood, C. (2010) The Blame Game: Spin, Bureaucracy, and Self-preservation in Government (New Jersey: Princeton University Press). Hood, C. & Margetts, H. (2007) The Tools of Government in the Digital Age (London: Red Globe Press). Hood, C., Scott, C., James, O., Jones, G. & Travers, T. (1999) Regulation Inside Government (Oxford: Oxford University Press).

Hooghe, L. & Marks, G. (2001) Multi-level Governance and European Integration (London: Rowman & Littlefield). Hooghe, L. & Marks, G. (2003) ‘Unraveling the Central State, But How? Types of Multi-level Governance’, American Political Science Review, 97, 2: 233–243. Hooghe, L. & Marks, G. (2004) ‘Contrasting Visions of Multi-level Governance’ in I. Bache & M. Flinders (eds.) Multi-level Governance (Oxford: Oxford University Press). Hopkins, R., Powell, A., Roy, A. & Gilbert, C. (1997) ‘The World Bank and Conditionality’, Journal of International Development, 9, 4: 507–516. Hopper, T., Lassou, P. & Soobaroyen, T. (2017) ‘Globalisation, Accounting and Developing Countries’, Critical Perspectives on Accounting, 43, March: 125– 148. Hoque, K., Davis, S. & Humphreys, M. (2004) ‘Freedom to Do What You Are Told’, Public Administration, 82, 2: 355–375. Horak, M. & Young, R. (eds.) (2012) Sites of Governance: Multilevel Governance and Policy Making in Canada’s Big Cities (Montreal: McGill-Queen’s University Press). Horgan, G. (2004) ‘Inter-institutional Relations in the Devolved Great Britain: Quiet Diplomacy’, Regional and Federal Studies, 14, 1: 113–135. Horn, M. (1995) Political Economy of Public Administration (Cambridge: Cambridge University Press). Housden, P. (2014) ‘This is us: A Perspective on Public Services in Scotland’, Public Policy and Administration, 29, 1: 64–74. Howard, C. (2005) ‘Policy Cycle: A Model of Post-Machiavellian Policy Making?’ Australian Journal of Public Administration, 3: 3–13. Howlett, M. (1998) ‘Predictable and Unpredictable Policy Windows’, Canadian Journal of Political Science, 31: 495–524. Howlett, M. (2000) ‘Managing the Hollow State. Procedural Policy Instruments and Modern Governance’, Canadian Public Administration, 43, 4: 412‒431. Howlett, M. (2011) Designing Public Policies: Principles and Instruments (London: Routledge). Howlett, M. & Ramesh, M. (2003) Studying Public Policy: Policy Cycles and Policy Subsystems (Oxford: Oxford University Press). Howlett, M., Ramesh, M. & Perl, A. (2009) Studying Public Policy: Policy Cycles and Policy Subsystems (Oxford: Oxford University Press). Howlett, M. & Tosun, J. (eds.) (2019) Policy Styles and Policy-making: Exploring the National Dimension (London: Routledge). Huang, L., He, R., Yang, Q., Chen, J., Zhou, Y., Hammitt, J. K., Lu, X., Bi, J. & Liu, Y. (2018) ‘The Changing Risk Perception Towards Nuclear Power in China After the Fukushima Nuclear Accident in Japan’, Energy Policy, 120, September: 294–301. Hudson, B. (1993) ‘Michael Lipsky and Street-level Bureaucracy: A Neglected Perspective’ in M. Hill (ed.) The Policy Process: A Reader (Hemel Hempstead:

Harvester Wheatsheaf). Hueglin, T. & Fenna, A. (2006) Comparative Federalism: A Systematic Inquiry (Peterborough, ON: Broadview Press). Hughes, O. (2017) Public Management and Administration 5th edn (London: Red Globe Press). Hunter, C. & Nixon, J. (1999) ‘The Discourse of Housing Debt: The Social Construction of Landlords, Lenders, Borrowers and Tenants’, Housing, Theory and Society 16, 4: 165–178. Hunter, F. (1953) Community Power Structure (Chapel Hill, NC: University of North Carolina Press). Hunter, F. (1980) Community Power Succession (Chapel Hill, NC: University of North Carolina Press). Hupe, P. & Hill, M. (2007) ‘Street-Level Bureaucracy and Public Accountability’, Public Administration, 85, 2: 279–299. Hwang, S. & Song, H. (2019) ‘Policy Transfer and Role of Policy Entrepreneur in International Aid: Exploring International Development Cases of Korea and Vietnam’, Policy Studies, 40, 1: 1–20. Ingold, K., Fischer, M. & Cairney, P. (2017) ‘Drivers for Policy Agreement in Nascent Subsystems: An Application of the Advocacy Coalition Framework to Fracking Policy in Switzerland and the UK’, Policy Studies Journal, 45, 3: 442– 463. Ingold, K., Fischer, M., Heikkila, T. & Weible, C. (2016) ‘Assessments and Aspirations’ in C. Weible, T. Heikkila, K. Ingold & M. Fischer (eds.) Policy Debates on Hydraulic Fracturing (London: Palgrave Macmillan). Ingold, K. & Gschwend, M. (2014) ‘Science in policy-making: Neutral Experts or Strategic Policy-makers?’, West European Politics, 37, 5: 993–1018. Ingold, K. & Varone, F. (2012) Treating Policy Brokers Seriously: Evidence From the Climate Policy’, Journal of Public Administration Research and Theory, 22, 2: 319–346. Ingram, H. & Schneider, A. (2005) ‘Introduction: Public Policy and the Social Construction of Deservedness’ in A. Schneider & H. Ingram (eds.) Deserving and Entitled: Social Construction and Public Policy (Albany: State University of New York Press). Ingram, H., Schneider, A. & deLeon, P. (2007) ‘Social Construction and Policy Design’, in P. A. Sabatier (ed.) Theories of the Policy Process, 2nd edn (Boulder, CO: Westview Press). Inter-Parliamentary Union (2018) ‘Women in National Parliaments’ http://archive.ipu.org/wmn-e/classif.htm Jachtenfuchs, M. (2001) ‘The Governance Approach to European Integration’, Journal of Common Market Studies, 39, 2: 245–264. Jacobsen, J. K. (1995) ‘Much Ado About Ideas: The Cognitive Factor in Economic Policy’, World Politics, 47, 2: 283–310.

James, O. (2001) ‘Business Models and the Transfer of Businesslike Central Government Agencies’, Governance, 14, 2: 233–252. James, O. & Lodge, M. (2003) ‘The Limitations of “Policy Transfer” and “Lesson Drawing” for Public Policy Research’, Political Studies Review, 1: 179–193. Jann, W. & Wegrich, K. (2007) ‘Theories of the Policy Cycle’, in F. Fischer, G. Miller & M. Sidney (eds.) Handbook of Public Policy Analysis (London: CRC Press). Jasanoff, S. (1986) Risk Management and Political Culture: A Comparative Analysis of Science in a Policy Context (New York: Russell Sage Foundation). Jasanoff, S. (2008) ‘Speaking Honestly to Power’, American Scientist, 6, 3: 240. Jasanoff, S. & Simmet, H. R. (2017) ‘No Funeral Bells: Public Reason in a “Posttruth” Age’, Social Studies of Science, 47, 5: 751–770. Jenkins, W. (1978) Policy Analysis (London: Martin Robertson). Jenkins-Smith, H., Nohrstedt, D., Weible, C. & Ingold, K. (2018) ‘The Advocacy Coalition Framework: An Overview of the Research Program’, in C. Weible & P. Sabatier (eds.) Theories of the Policy Process (London: Routledge), 145–182. Jenkins-Smith, H. & Sabatier, P. (1993a) ‘The Study of Public Policy Processes’ in P. Sabatier & H. Jenkins-Smith (eds.) Policy Change and Learning: An Advocacy Coalition Approach (Boulder, CO: Westview Press). Jenkins-Smith, H. & Sabatier, P. (1993b) ‘The Dynamics of Policy-Oriented Learning’ in P. Sabatier & H. Jenkins-Smith (eds.) Policy Change and Learning: An Advocacy Coalition Approach (Boulder, CO: Westview Press). Jenkins-Smith, H. & Sabatier, P. (1993c) ‘Methodological Appendix’ in P. Sabatier & H. Jenkins-Smith (eds.) Policy Change and Learning: An Advocacy Coalition Approach (Boulder, CO: Westview Press). Jenkins-Smith, H. & Sabatier, P. (1994) ‘Evaluating the Advocacy Coalition Framework’, Journal of Public Policy, 14, 2: 175–203. Jervis, R. (1998) System Effects: Complexity in Political and Social Life (Princeton, NJ: Princeton University Press). Jessop, B. (2004) ‘Multi-level Governance and Multi-level Metagovernance’ in I. Bache & M. Flinders (eds.) Multi-level Governance (Oxford: Oxford University Press). Jobert, B. & Muller, P. (1987) L’Etat en Action: Politiques Publiques et Corporatismes (Paris: Presses Universitaires de France). John, P. (1998) Analysing Public Policy (London: Continuum). John, P. (1999) ‘Ideas and Interests; Agendas and Implementation: An Evolutionary Explanation of Policy Change in British Local Government Finance’, British Journal of Politics and International Relations, 1, 1: 39–62. John, P. (2000) ‘The Uses and Abuse of Evolutionary Theory in Political Science: A reply to Alan McConnell and Keith Dowding’, British Journal of Politics and International Relations, 11: 89–94. John, P. (2003) ‘Is There Life After Policy Streams, Advocacy Coalitions, and Punctuations: Using Evolutionary Theory to Explain Policy Change?’ Policy

Studies Journal, 31, 4: 481–498. John, P. (2006) ‘The Policy Agendas Project: A Review’, Journal of European Public Policy, 13, 7: 975–986. John, P. (2011) Making Policy Work (London: Routledge). John, P. (2012) Analysing Public Policy 2nd edn (London: Routledge). John, P. (2018) How Far to Nudge? Assessing Behavioural Public Policy (Cheltenham: Edward Elgar). John, P. & Margetts, H. (2003) ‘Policy Punctuations in the UK: Fluctuations and Equilibria in Central Government Expenditure Since 1951’, Public Administration, 81, 3: 411–432. Jones, B. (1994) Reconceiving Decision-making in Democratic Politics: Attention, Choice and Public Policy (Chicago: University of Chicago Press). Jones, B. (1999) ‘Bounded Rationality’, Annual Review of Political Science, 2: 297–321. Jones, B. (2003) ‘Bounded Rationality and Political Science: Lessons from Public Administration and Public Policy’ Journal of Public Administration Research and Theory, 13, 4: 395–412. Jones, B. & Baumgartner, F. (2005) The Politics of Attention (Chicago: University of Chicago Press). Jones, B., Baumgartner, F., Breunig, C., Wleizen, C., Soroka, S., Foucault, M., Francois, A., Green-Pedersen, C., Koski, C., John, P., Mortensen, P., Varone, F. & Walgrave, S. (2009) ‘A General Empirical Law of Public Budgets: A Comparative Analysis’, American Journal of Political Science, 53, 4: 855–873. Jones, B. & Thomas, H. (2017) ‘The Cognitive Underpinnings of Policy Process Studies’, Cognitive Systems Research, 45: 48–51. Jones, B., Thomas, H. & Wolfe, M. (2014a) ‘Policy Bubbles’, Policy Studies Journal, 42, 1: 146–171. Jones, C. (1970; 1984) An Introduction to the Study of Political Life 3rd edn (Berkeley: Duxberry Press). Jones, M. (2009) ‘Phase Space: Geography, Relational Thinking, and Beyond’, Progress in Human Geography 33, 4: 487–506. Jones, M. (2014) ‘Communicating Climate Change: Are Stories Better than “Just the Facts”?’ Policy Studies Journal, 42, 4: 644–673. Jones, M. (2018) ‘Advancing the Narrative Policy Framework?’ Policy Studies Journal, 46, 4: 724–746. Jones, M. & Crow, D. (2017) ‘How can We Use the “Science of Stories” to Produce Persuasive Scientific Stories?’ Palgrave Communications, 3: 53 https://doi.org/10.1057/s41599-017-0047-7 Jones, M. & Jenkins-Smith, H. (2009) ‘Trans-Subsystem Dynamics: Policy Topography, Mass Opinion, and Policy Change’, Policy Studies Journal, 37, 1: 37–59. Jones, M. & McBeth, M. (2010) ‘Narrative Policy Framework: Clear Enough to be Wrong?’ Policy Studies Journal, 38, 2: 329–353.

Jones, M., McBeth, M. K. & Shanahan, E. A. (2014b) ‘Introducing the Narrative Policy Framework’ in M. D. Jones, E. A. Shanahan & M. K. McBeth (eds.) The Science of Stories: Applications of the Narrative Policy Framework in Public Policy Analysis (New York: Palgrave Macmillan), 1–26. Jones, M. & Peterson, H. (2017) ‘Narrative Persuasion and Storytelling as Climate Communication Strategies’, Oxford Research Encyclopedia of Climate Science. doi: 10.1093/ acrefore/9780190228620.013.384 Jones, M., Peterson, H., Pierce, J., Herweg, N., Bernal, A., Raney, H. & Zahariadis, N. (2016) ‘A River Runs Through It: A Multiple Streams MetaReview’, Policy Studies Journal 44, 1: 13–36. Jones, M. & Radaelli, C. (2015) ‘The Narrative Policy Framework: Child or Monster?’, Critical Policy Studies, 9, 3: 339–355. Jones, M. & Radaelli, C. (2016) ‘The Narrative Policy Framework’s Call for Interpretivists’, Critical Policy Studies, 10, 1: 117–120. Jones, M. & Song, G. (2014) ‘Making Sense of Climate Change: How Story Frames Shape Cognition’, Political Psychology, 35, 4: 447–476. Jones, R., Pykett, J. & Whitehead, M. (2013) Changing Behaviours: On the Rise of the Psychological State (Cheltenham: Edward Elgar). Jordan, A. G. (1981) ‘Iron Triangles, Woolly Corporatism and Elastic Nets: Images of the Policy Process’, Journal of Public Policy 1, 1: 95–123. Jordan, A. G. (1990a) ‘Policy Community Realism versus “New” Institutionalist Ambiguity’, Political Studies, 38: 470–484. Jordan, A. G. (1990b) ‘The Pluralism of Pluralism: An Anti-theory?’, Political Studies, 38, 2: 286–301. Jordan, A. G. (2005) ‘Bringing Policy Communities Back In? A Comment on Grant’, British Journal of Politics and International Relations, 7: 317–321. Jordan, A. G. & Cairney, P. (2013) ‘What is the “Dominant Model” of British Policy Making? Comparing Majoritarian and Policy Community Ideas’, British Politics, 8, 3: 233–259. Jordan, A. G., Halpin, D. & Maloney, W. (2004) ‘Defining Interests: Disambiguation and the Need for New Distinctions?’ British Journal of Politics and International Relations, 6, 2: 195–212. Jordan, A. G. & Maloney, W. (1995) ‘Policy Networks Expanded: A Comment on Cavanagh, Marsh and Smith’, Public Administration, 73: 630–634. Jordan, A. G. & Maloney, W. (1996) ‘How Bumble Bees Fly: Accounting for Public Interest Participation’, Political Studies, 44, 4: 668–685. Jordan, A. G. & Maloney, W. A. (1997) ‘Accounting for Subgovernments: Explaining the Persistence of Policy Communities’, Administration and Society, 29, 5: 557–583. Jordan, A. G. & Richardson, J. J. (1982) ‘The British Policy Style or the Logic of Negotiation?’ in J. J. Richardson (ed.) Policy Styles in Western Europe (London: Allen & Unwin).

Jordan, A. G. & Richardson, J. J. (1987) British Politics and the Policy Process (London: Allen & Unwin). Jordan, A. G. & Schubert, K. (1992) ‘A Preliminary Ordering of Policy Network Labels’, European Journal of Political Research, 21, 1: 7–27. Jordan, A. G. & Stevenson, L. (2000) ‘Redemocratizing Scotland. Towards the Politics of Disappointment?’, in A. Wright (ed.) Scotland: The Challenge of Devolution (Aldershot: Ashgate). Jordan, A. G. & Turnpenny, J. (eds.) (2015) The Tools of Policy Formulation (Cheltenham: Edward Elgar). Judge, D. (1993) The Parliamentary State (London: Sage). Judge, D. (2005) Political Institutions in the United Kingdom (Oxford: Oxford University Press). Kahneman, D. (2012) Thinking Fast and Slow (London: Penguin). Kam, C. D. (2005) ‘Who Toes the Party Line? Cues, Values, and Individual Differences’, Political Behavior 27, 2: 163–182 https://doi.org/10.1007/s11109005-1764-y Karp, J. & Banducci, S. (2008) ‘Political Efficacy and Participation in Twenty-Seven Democracies: How Electoral Systems Shape Political Behaviour’, British Journal of Political Science, 38, 2: 311–334. Kaufman, H. & Jones, V. (1954) ‘Review: The Mystery of Power’, Public Administration Review, 14, 3: 205–212. Kay, A. (2006) The Dynamics of Public Policy (Cheltenham: Edward Elgar). Keating, M. (2005) The Government of Scotland (Edinburgh: Edinburgh University Press). Keating, M. & Cairney, P. (2012) ‘Policymaking, Learning and Devolution’, Regional and Federal Studies, 22, 3: 239–250. Keating, M., Cairney, P. & Hepburn, E. (2009) ‘Territorial Policy Communities and Devolution in the United Kingdom’, Cambridge Journal of Regions, Economy and Society, 2, 1: 51–66. Keating, M., Cairney, P. & Hepburn, E. (2012) ‘Policy Convergence, Transfer and Learning in the UK Under Devolution’, Regional and Federal Studies, 22, 3: 289–307. Keating, M., Stevenson, L., Cairney, P. & MacLean, K. (2003) ‘Does Devolution Make a Difference? Legislative Output and Policy Divergence in Scotland’, Journal of Legislative Studies, 9, 3: 110–139. Keleman, R. D. (2004) The Rules of Federalism: Institutions and Regulatory Politics in the EU and Beyond (Cambridge: Harvard University Press). Kenny, M. (2007) ‘Gender, Institutions and Power: A Critical Review’, Politics, 27, 2: 91–100. Kenny, M. & Mackay, F. (2009) ‘Already Doin’ It for Ourselves? Skeptical Notes on Feminism and Institutionalism’, Politics & Gender, 5, 2: 271–280. Kenny, M. & Mackay, F. (2018) ‘Feminist and Gendered Approaches’ in V. Lowndes, D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science

4th edn (London: Red Globe Press). Keohane, R. & Ostrom, E. (1995) Local Commons and Global Independence (London: Sage). Kernick, D. (2006) ‘Wanted – New Methodologies for Health Service Research. Is Complexity Theory the Answer?’ Family Practice 23: 385–390. Kerr, P. (2002) ‘Saved from extinction: evolutionary theorising, politics and the state’, British Journal of Politics and International Relations, 4, 2: 330–338. Kettell, S. & Cairney, P. (2010) ‘Taking the Power of Ideas Seriously: The Case of the 2008 Human Fertilisation and Embryology Bill’, Policy Studies, 31, 3: 301– 317. Kettl, D. (2006) ‘Public Bureaucracies’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Kettl, D. F. (1993) Sharing Power: Public Governance and Private Markets (Washington, DC: The Brookings Institution). Kiely, R. (2017) ‘Marxism: A Global Perspective’ In V. Lowndes, D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science 4th edn (London: Red Globe Press). Kikeri, S., Nellis, J. & Shirley, M. (1992) Privatization: The Lessons of Experience (Washington, DC: The World Bank) http://wwwwds.worldbank.org/servlet/main? menuPK=64187510&pagePK=64193027&piPK=64187937&theSitePK=523679& entityID=000178830_9810191113028 King, A. & Crewe, I. (2014) The Blunders of Our Governments (London: Oneworld). King, G., Schneer, B. & White, A. (2017) ‘How the News Media Activate Public Expression and Influence National Agendas’, Science, 358, 6364: 776–780. Kingdon, J. (1984) Agendas, Alternatives and Public Policies (New York: HarperCollins). Kingdon, J. (1995) Agendas, Alternatives and Public Policies 2nd edn (New York: HarperCollins). Kislov, R., Wilson, P., Cummings, G., Ehrenberg, A., Gifford, W., Kelly, J., Kitson, A., Pettersson, L., Wallin, L. & Harvey, G. (2019) ‘From Research Evidence to “Evidence by Proxy”? Organizational Enactment of Evidence‐Based Health Care in Four High‐Income Countries’, Public Administration Review, Early View https://doi.org/10.1111/puar.13056 Kitschelt, H. & Streeck, W. (2003) ‘From Stability of Stagnation: Germany at the Beginning of the Twenty-First Century’, West European Politics, 26, 4: 1–34. Kjaer, A. (2004) Governance (Cambridge: Polity). Klijn, E. (2008) ‘Complexity Theory and Public Administration: What’s New?’ Public Management Review, 10, 3: 299–317. Knaggård, A. (2015) ‘The Multiple Streams Framework and the Problem Broker’, European Journal of Political Research, 54, 3: 450–465.

Knill, C. (2013) ‘The Study of Morality Policy: Analytical Implications from a Public Policy Perspective’, Journal of European Public Policy, 20, 3: 309–317. Knill, C. & Tosun, J. (2012; 2019) Public Policy: A New Introduction 1st and 2nd edn (London: Red Globe Press). Koebele, E. (2019a) ‘Cross-Coalition Coordination in Collaborative Environmental Governance Processes’, Policy Studies Journal, Early View doi: 10.1111/psj.12306 Koebele, E. (2019b) ‘Integrating Collaborative Governance Theory with the Advocacy Coalition Framework’, Journal of Public Policy, 39, 1: 35–64. Kooiman, J. (1993) ‘Socio-political Governance: Introduction’ in J. Kooiman (ed.) Modern Governance (London: Sage). Kooiman, J. (2003) Governing as Governance (London: Sage). Koski, C. & Workman, S. (2018) ‘Drawing Practical Lessons from Punctuated Equilibrium Theory’, Policy and Politics, 46, 2: 293–308. Kratochwil, F. (2008) ‘Constructivism’ in D. Della Porta & M. Keating (eds.) Approaches and Methodologies in the Social Sciences (Cambridge: Cambridge University Press). Krauss, E. & Pekkanen, R. (2004) ‘Explaining Party Adaptation to Electoral Reform: The Discreet Charm of the LDP?’, Journal of Japanese Studies, 30, 1: 1–34. Kriesi, H., Adam, S. & Jochum, M. (2006) ‘Comparative Analysis of Policy Networks in Western Europe’, Journal of European Public Policy 13, 3: 341– 361. Kuhn, T. (1970) The Structure of Scientific Revolutions (Chicago: University of Chicago Press). Laffan, B. (1997) ‘From Policy Entrepreneur to Policy Manager: The Challenge Facing the European Commission’, Journal of European Public Policy 4, 3: 422– 438. Lakoff, G. (2014) The All New Don’t Think of an Elephant (White River Junction, VT: Chelsea Green). Lam, W. F. & Chan, K. N. (2015) ‘How authoritarianism intensifies punctuated equilibrium: The dynamics of policy attention in Hong Kong’, Governance 28, 4, 549–570. Lamprinakou, C., Morucci, M., Campbell, R. & van Heerde-Hudson, J. (2016) ‘All Change in the House? The Profile of Candidates and MPs in the 2015 British General Election’, Parliamentary Affairs, 70, 2: 207–232. Larsen, J., Vrangbaek, K. & Traulsen, J. (2006a) ‘Advocacy Coalitions and Pharmacy Policy in Denmark’, Social Science and Medicine, 63, 1: 212–224. Larsen, T., Taylor-Gooby, P. & Kananen, J. (2006b) ‘New Labour’s Policy Style: A Mix of Policy Approaches’, Journal of Social Policy, 35, 4: 629–649. Lasswell, H. (1936) Politics: Who Gets What, When, How (Cleveland, OH: Meridian Books).

Lasswell, H. (1956) The Decision Process: Seven Categories of Functional Analysis (College Park, MD: University of Maryland Press). Laver, M. (1997) Private Desires, Political Action (London: Sage). Lavis, J., Ross, S., Hurley, J. et al. (2002) ‘Examining the Role of Health Services Research in Public Policymaking’, Milbank Quarterly, 80, 1: 125–154. Lawlor, A. & Crow, D. (2018) ‘Risk-Based Policy Narratives’, Policy Studies Journal, 46, 4: 843–867. Le Grand, J. (2003) Motivation, Agency and Public Policy: Of Knights and Knaves, Pawns and Queens (Oxford: Oxford University Press). Leach, W. & Sabatier, P. (2005) ‘To Trust an Adversary: Integrating Rational and Psychological Models of Collaborative Policymaking’, American Political Science Review, 99, 4: 491–503. Legrand, T. (2012a) ‘The Merry Mandarins of Windsor: Policy Transfer and Transgovernmental Networks in the Anglosphere’, Policy Studies, 33, 6: 523– 540. Legrand, T. (2012b) ‘Overseas and Over Here: Policy Transfer and Evidence Based Policy-making’, Policy Studies, 33, 4: 329–348. Lejano, R. (2015) ‘Narrative Disenchantment’, Critical Policy Studies, 9, 3: 368– 371. Lerner, D. & Lasswell, H. (1951) The Policy Sciences (Stanford, CA: Stanford University Press). Levin, K., Cashore, B., Bernstein, S. & Auld, G. (2012) ‘Overcoming the Tragedy of Super Wicked Problems: Constraining Our Future Selves to Ameliorate Global Climate Change’, Policy Sciences, 45, 2: 123–152. Lewis, O. & Steinmo, S. (2008) ‘Taking Evolution Seriously’ (Florence: European University Institute) http://spot.colorado.edu/~steinmo/TakingEvolution.pdf Lewis, P. (2013) ‘Policy Thinking, Fast and Slow: A Social Intuitionist Perspective on Public Policy Processes’, American Political Science Association 2013 Annual Meeting. SSRN: http://ssrn.com/abstract=2300479 Lieberman, R. C. (2002) ‘Ideas, Institutions and Political Order: Explaining Political Change’, American Political Science Review, 90, 4: 691–712. Lijphart, A. (1999) Patterns of Democracy (New Haven, CT: Yale University Press). Lindblom, C. (1959) ‘The Science of Muddling Through’, Public Administration Review, 19: 79–88. Lindblom, C. (1964) ‘Contexts for Change and Strategy: A Reply’, Public Administration Review, 24, 3: 157–158. Lindblom, C. (1965) The Intelligence of Democracy (New York: Free Press). Lindblom, C. (1968) The Policy-Making Process (Englewood Cliffs, NJ: PrenticeHall). Lindblom, C. (1977) Politics and Markets (New York: Basic Books). Lindblom, C. (1979) ‘Still Muddling, Not Yet Through’, Public Administration Review, 39: 517–525.

Linder, S. & Peters, B. G. (2006) ‘Coming to Terms with Intramural Differences over Policy Frameworks’, Policy Sciences 39, 1: 19–40. Lindquist, E. (2004) ‘Organizing for Policy Implementation: The Emergence and Role of Implementation Units in Policy Design and Oversight’, Journal of Comparative Policy Analysis: Research and Practice, 8, 4: 311–324. Lipsky, M. (1980) Street-Level Bureaucracy (New York: Russell Sage Foundation). Litfin, K. (2000) ‘Advocacy Coalitions Along the Domestic-Foreign Frontier: Globalization and Canadian Climate Change Policy’, Policy Studies Journal, 28, 1: 236–252. Lodge, M. & Gill, D. (2011) ‘Toward a New Era of Administrative Reform? The Myth of Post-NPM in New Zealand’, Governance, 24, 1: 141–166. Lodge, M. & Weigrich, K. (2012) Managing Regulation (London: Red Globe Press). Lomas, J. & Brown, A. (2009) ‘Research and Advice Giving: a Functional View of Evidence-Informed Policy Advice in a Canadian Ministry of Health’, Milbank Quarterly, 87, 4: 903–926. Long, N. (1958) ‘The Local Community as an Ecology of Games’, American Journal of Sociology, 64, 3: 251–261. Loomba, A. (2007) Colonialism/Postcolonialism 2nd edn (London: Routledge). Lovenduski, J. (1998) ‘Gendering Research in Political Science’, Annual Review of Political Science, 1, 1: 333–356. Lovenduski, J. (2011) ‘Foreword’ in M. L. Krook & F. Mackay (eds.) Gender, Politics and Institutions: Towards a Feminist Institutionalism (Basingstoke: Palgrave Macmillan). Lowi, T. (1964) ‘An American Business, Public Policy, Case-Studies, and Political Theory’, World Politics, 16, 4: 677–715. Lowi, T. (1972) ‘Four Systems of Policy, Politics and Choice’, Public Administration Review, 32, 4: 298–310. Lowi, T. (1988) ‘Comment’, Policy Studies Journal, 16: 725–728. Lowndes, V. (2010) ‘The Institutional Approach’ in D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science (London: Red Globe Press). Lowndes, V. (2014) ‘How Are Things Done Around Here? Uncovering Institutional Rules and their Gendered Effects’, Politics & Gender, 10, 4: 685–691. Lowndes, V. (2017) ‘Institutionalism’ in V. Lowndes, D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science 4th edn (London: Red Globe Press). Lubell, M. (2013) ‘Governing Institutional Complexity: The Ecology of Games Framework’, Policy Studies Journal, 41, 3: 537–559. Lubell, M. (2015) ‘Collaborative Partnerships in Complex Institutional Systems’, Current Opinion in Environmental Sustainability, 12: 41–47 http://dx.doi.org/10.1016/j.cosust.2014.08.011 Lubell, M., Henry, A. D. & McCoy, M. (2010) ‘Collaborative Institutions in an Ecology of Games’, American Journal of Political Science, 54, 2: 287–300. Lukes, S. (1974) Power: A Radical View (London: Macmillan).

Lukes, S. (ed.) (1986) Power (New York: New York University Press). Lukes, S. (2005) Power: A Radical View 2nd edn (London: Red Globe Press). Lukes, S. & Haglund, L. (2005) ‘Power and Luck’, European Journal of Sociology, 46, 1: 45–66. Lundin, M. (2007) ‘When Does Cooperation Improve Public Policy Implementation?’, Policy Studies Journal, 35, 4: 629–652. Lustick, I. (2011) ‘Taking Evolution Seriously’, Polity, 43, 2: 179–209. Lynas, M. (2008) ‘Six Degrees, but no PhD’ Guardian, 18 June, http://blogs.guardian.co.uk/books/2008/06/six_degrees_but_no_phd.html Machin, A. (2004) ‘Comparisons between Unemployment and the Claimant Count’, Labour Market Trends February 2004: 59–62. Mackay, F. (2004) ‘Gender and Political Representation in the UK: The State of the “Discipline”’, The British Journal of Politics & International Relations, 6, 1: 99– 120. Mackay, F. (2014) ‘Nested Newness, Institutional Innovation, and the Gendered Limits of Change’, Politics and Gender, 10: 549–571. Macleavy, J. & Gay, O. (2005) The Quango Debate, House of Commons Research Paper 05/30 (April). MacLeod, M. & Emejulu, A. (2014) ‘Neoliberalism with a Community Face? A Critical Analysis of Asset-Based Community Development in Scotland’, Journal of Community Practice, 22, 4: 430–450. MacRae, H. & Weiner, E. (eds.) (2017) Towards Gendering Institutionalism: Equality in Europe (London: Rowman & Littlefield). Mahoney, C. (2008) Brussels Versus the Beltway: Advocacy in the United States and the European Union (Washington, DC: Georgetown University Press). Mahoney, J. & Thelen, K. (eds.) (2010) Explaining Institutional Change: Ambiguity, Agency, and Power (Cambridge: Cambridge University Press). Mair D., Smillie L., La Placa G., Schwendinger F., Raykovska M., Pasztor Z., van Bavel R. (2019) Understanding our political nature: How to put knowledge and reason at the heart of political decision-making (Luxembourg: Publications Office of the European Union). Majone, G. (1980) ‘Policies as Theories’, Omega 8, 2: 151–162. Majone, G. (1989) Evidence, Argument and Persuasion in the Policy Process (New Haven, CT: Yale University Press). Majone, G. (1993) ‘The European Community Between Social Policy and Social Regulation’, Journal of Common Market Studies, 31, 2: 153–170. Majone, G. (1994) ‘The Rise of the Regulatory State in Europe’, West European Politics, 17, 3: 77–101. Majone, G. (1996) Regulating Europe (London: Routledge). Majone, G. (2006) ‘Agenda Setting’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Malbon, E., Alexander, D., Carey, G., Reeders, D., Green, C., Dickinson, H. & Kavanagh, A. (2019) ‘Adapting to a Marketised System: Network Analysis of a

Personalisation Scheme in Early Implementation’, Health & social care in the community, 27, 1: 191–198. Maloney, W., Jordan, G. & McLaughlin, A. (1994) ‘Interest Groups and Public Policy: The Insider/Outsider Model Revisited’, Journal of Public Policy, 14, 1: 17–38. Maloney, W. A. & Richardson, J. J. (1995a) Managing Policy Change in Britain: The Politics of Water (Edinburgh: Edinburgh University Press). Maloney, W. A. & Richardson, J. J. (1995b) ‘Water Policy-Making in England and Wales: Policy Communities Under Pressure?’ in H. Bressers, L. J. O’Toole Jr & J. J. Richardson (eds.) Networks For Water Policy: A Comparative Perspective (London: Frank Cass). Mamudu, H., Cairney, P. & Studlar, D. (2015) ‘Global Public Policy: Does the New Venue for Transnational Tobacco Control Challenge the Old Way of Doing Things?’ Public Administration, 93, 4: 856–873. Maor, M. (2014) ‘Policy Bubbles: Policy Over-reaction and Positive Feedback’, Governance, 27, 3: 469–487. March, J. & Olsen, J. (1984) ‘The New Institutionalism: Organizational Factors in Political Life’, The American Political Science Review, 78, 3: 734–749. March, J. & Olsen, J. (1996) ‘Institutional Perspectives on Political Institutions’, Governance, 9, 3: 247–264. March, J. & Olsen, J. (2006a) ‘Elaborating the “New Institutionalism”’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). March, J. & Olsen, J. (2006b) ‘The Logic of Appropriateness’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Marchildon, G. (2001) ‘Royal Commissions and the Policy Cycle in Canada’, Public Lecture at the Saskatchewan Institute of Public Policy, http://www.uregina.ca/sipp/documents/pdf/ssgm.pdf Marcussen, M. (2005) ‘Central Banks on the Move’, Journal of European Public Policy, 12, 5: 903–923. Marinetto, M. (2003) ‘Governing Beyond the Centre: A Critique of the AngloGovernance School’, Political Studies, 51: 592–608. Marks, G. (1993) ‘Structural Policy and Multi-level Governance in the EC’, in A. Cafruny & G. Rosenthal (eds.) The State of the European Community: The Maastrict Debate and Beyond (Boulder, CO: Lynne Rienner). Marks, G. & Hooghe, L. (2000) ‘Optimality and Authority: A Critique of Neoclassical Theory’, Journal of Common Market Studies, 38, 5: 795–816. Marsh, D. (1991) ‘Privatisation under Mrs Thatcher’, Public Administration, 69, 4: 459–480. Marsh, D. (2008) ‘Understanding British Government: Analysing Competing Models’, British Journal of Politics and International Relations, 10, 2: 251–269.

Marsh, D. (2011) ‘The New Orthodoxy: the Differentiated Polity Model’, Public Administration, 89, 1: 32–48. Marsh, D. & Evans, M. (2012) ‘Policy Transfer: Into the Future, Learning from the Past’, Policy Studies, 33, 6: 587–591. Marsh, D. & McConnell, A. (2010) ‘Towards a Framework for Establishing Policy Success’, Public Administration, 88, 2: 564–583. Marsh, D. & Rhodes, R. A. W. (eds.) (1992a) Implementing Thatcherite Policies (Buckingham: Open University Press). Marsh, D. & Rhodes, R. A. W. (1992b) ‘Policy Communities and Issue Networks: Beyond Typology’ in D. Marsh & R. A. W. Rhodes (eds.) Policy Networks in British Government (Oxford: Oxford University Press). Marsh, D., Richards, D. & Smith, M. J. (2001) Changing Patterns of Governance in the United Kingdom (Basingstoke: Palgrave Macmillan). Marsh, D., Richards, D. & Smith, M. J. (2003) ‘Unequal Plurality: Towards an Asymmetric Power Model of British Politics’, Government and Opposition, 38: 306–332. Marsh, D., Richards, D. & Smith, M. J. (2004) ‘Understanding and Explaining Civil Service Reform: A Reply to Dowding and James’, British Journal of Political Science, 34, 1: 189–192. Marsh, D. & Smith, M. (2000) ‘Understanding Policy Networks: Towards a Dialectical Approach’, Political Studies, 48, 1: 4–21. Marsh, D. & Smith, M. (2001) ‘There is More Than One Way to do Political Science: On Different Ways to Study Policy Networks’, Political Studies, 49, 3: 528–541. Marsh, D., Smith, M. & Richards, D. (2000) ‘Bureaucrats, Politicians and Reform in Whitehall: Analysing the Bureau-Shaping Model’, British Journal of Political Science, 30, 3: 461–482. Martin, C. J. & Swank, D. (2004) ‘Does the Organization of Capital Matter?’, American Political Science Review, 98, 4: 593–611. Martin, C. J. & Thelen, K. (2007) ‘The State and Coordinated Capitalism’, World Politics, 60, 1: 1–36. Massey, A. (2001) ‘Policy, Management and Implementation’ in S. Savage & R. Atkinson (eds.) Public Policy Under Blair (London: Red Globe Press). Massey, A. (2010) ‘Lessons from Africa: New Public Management and the Privatization of Kenya Airways’, Public Policy and Administration, 25, 2: 194– 215. Mastenbroek, E. (2005) ‘EU Compliance: Still a Black Hole’, Journal of European Public Policy, 12, 6: 1103–1120. Matland, R. (1995) ‘The Implementation Literature: The Ambiguity-Conflict Model of Policy Implementation’, Journal of Public Administration Research and Theory: J-PART, 5, 2: 145–174. Matthews, F. (2016) ‘Letting Go and Holding on: The Politics of Performance Management in the United Kingdom’, Public Policy and Administration, 31, 4:

303–323. Maxwell, S. (1999) ‘The Meaning and Measurement of Poverty’, Overseas Development Institute Poverty Briefing, http://www.odi.org.uk/publications/briefing/pov3.html May, P. (1992) ‘Policy Learning and Failure’, Journal of Public Policy, 12, 4: 331–354. Mayne, R., Green, D., Guijt, I., Walsh, M., English, R. & Cairney, P. (2018) ‘Using Evidence to Influence Policy: Oxfam’s Experience’, Palgrave Communications, 4, 122: 1–10 doi: 10.1057/s41599-018-0176-7 Mazey, S. & Richardson, J. (2006) ‘Interest Groups and EU Policy-Making: Organisational Logic and Venue Shopping’ in J. Richardson (ed.) European Union: Power and Policy-Making 3rd edn (London: Routledge). McBeth, M. (2014) ‘Preface: The Portneuf School of Narrative’ in M. D. Jones, E. A. Shanahan & M. K. McBeth (eds.) The Science of Stories (New York: Palgrave Macmillan). McCann, E. & Ward, K. (2012) ‘Policy Assemblages, Mobilities and Mutations: Toward a Multidisciplinary Conversation’, Political Studies Review, 10, 3: 325– 332. McCann, E. & Ward, K. (2013) ‘A Multi-Disciplinary Approach to Policy Transfer Research: Geographies, Assemblages, Mobilities and Mutations’, Policy Studies, 34, 1: 2–18. McConnell, A. (2000) ‘Local Taxation, Policy Formation and Policy Change: A Reply to Peter John’, British Journal of Politics and International Relations, 2, 1: 81–88. McConnell, A. (2010) Understanding Policy Success: Rethinking Public Policy (London: Red Globe Press). McConnell, A. (2018a) ‘Hidden Agendas: Shining a Light on the Dark Side of Public Policy’, Journal of European Public Policy, 25, 12: 1739–1758. McConnell, A. (2018b) ‘Rethinking Wicked Problems as Political Problems and Policy’ Problems, Policy & Politics, 46, 1: 165–180. McCool, D. (1995) Public Policy Theories, Models and Concepts: An Anthology (Englewood Cliffs, NJ: Prentice-Hall). McCool, D. (1998) ‘The Subsystem Family of Concepts: A Critique and a Proposal’, Political Research Quarterly 51, 2: 551–570. McEwen, N. & Petersohn, B. (2015) ‘Between Autonomy and Interdependence: The Challenges of Shared Rule after the Scottish Referendum’, Political Quarterly, 86, 2: 192–200. McEwen, N., Swenden, W. & Bolleyer, N. (2012) ‘Intergovernmental Relations in the UK: Continuity in a Time of Change?’, The British Journal of Politics and International Relations, 14, 2: 323–343. McGarvey, N. & Cairney, P. (2008) Scottish Politics (London: Red Globe Press). McGinnis, M. (2011) ‘An Introduction to IAD and the Language of the Ostrom Workshop: A Simple Guide to a Complex Framework’, Policy Studies Journal, 39, 1: 169–183.

McKay, D. (2000) ‘Policy Legitimacy and Institutional Design: Comparative Lessons for the European Union’, Journal of Common Market Studies, 38, 1: 25–44. McLean, I. (1987) Public Choice (Oxford: Basil Blackwell). McLean, I. (2002a) ‘Review Article: William H. Riker and the Invention of Heresthetic(s)’, British Journal of Political Science, 32, 3: 535–558. McPherson, A. M. & Raab, C. D. (1988) Governing Education: a Sociology of Policy Since 1945 (Edinburgh: Edinburgh University Press). Meier, K. (2009) ‘Policy Theory, Policy Theory Everywhere: Ravings of a Deranged Policy Scholar’, Policy Studies Journal, 37, 1: 5–11. Mény, Y. (2014) ‘Managing the EU Crises: Another Way of Integration by Stealth’, West European Politics 37, 6: 1336–1353. Merry, M. (2016) ‘Constructing Policy Narratives in 140 Characters or Less: The Case of Gun Policy Organizations’, Policy Studies Journal, 44, 4: 373–395. Merry, M. (2018) ‘Narrative Strategies in the Gun Policy Debate: Exploring Proximity and Social Construction’, Policy Studies Journal, 46, 4: 747–770. Meseguer, C. (2003) ‘Learning and Economic Policy Choices: A Bayesian Approach’, EUI Working Paper RSC No. 2003/5 (San Domenico: European University Institute), http://cadmus.iue.it/dspace/bitstream/1814/1847/1/03_05.pdf Meseguer, C. (2005) ‘Policy Learning, Policy Diffusion, and the Making of a New Order’, Annals of the American Academy of Political and Social Science, 598, March: 67–82. Mettler, S. & SoRelle, M. (2018) ‘Policy Feedback Theory’ in C. Weible & P. Sabatier (eds.) Theories of the Policy Process 4th edn (Chicago: Westview Press). Mewhirter, J., Lubell, M. & Berardo, R. (2018) ‘Institutional Externalities and Actor Performance in Polycentric Governance Systems’, Environmental Policy and Governance, 28, 4: 295–307. Michener, J. (2019) ‘Policy Feedback in a Racialized Polity’, Policy Studies Journal, 47, 2: 423–450. Mills, C. W. (1956) The Power Elite (New York: Oxford University Press). Mintrom, M. & Norman, P. (2009) ‘Policy Entrepreneurship and Policy Change’, Policy Studies Journal, 37, 4: 649–667. Mintrom, M. & Vergari, S. (1996) ‘Advocacy Coalitions, Policy Entrepreneurs and Policy Change’, Policy Studies Journal, 24, 3: 420–434. Mitchell, M. (2009) Complexity (Oxford: Oxford University Press). Mitleton-Kelly, E. (2003) ‘Ten Principles of Complexity and Enabling Infrastructures’ in E. Mitleton-Kelly (ed.) Complex Systems and Evolutionary Perspectives of Organisations (Amsterdam: Elsevier). Mondou, M. & Montpetit, E. (2010) ‘Policy Styles and Degenerative Politics: Poverty Policy Designs in Newfoundland and Quebec’, Policy Studies Journal 38, 4: 703–722.

Mooney, C. Z. (1999) ‘The Politics of Morality Policy: Symposium Editor’s Introduction’, Policy Studies Journal, 27, 4: 675–680. Moran, M. (2005) Politics and Governance in the UK (London: Red Globe Press). Moran, M. (2006) ‘Economic Institutions’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Moravcsik, A. (2012) ‘Europe After the Crisis: How to Sustain a Common Currency’, in Foreign Affairs, May/June, 54–68. Moravscik, A. (1993) ‘Preferences and Power in the European Community: A Liberal Inter-governmentalist Approach’, Journal of Common Market Studies, 31, 4: 473–524. Morreira, S. & Luckett, K. (2018) ‘Questions Academics Can Ask to Decolonise Their Classrooms’ African Skies 14 November https://folukeafrica.com/questions-academics-can-ask-to-decolonise-theirclassrooms/ Moyson, S. (2017) Cognition and Policy Change: The Consistency of Policy Learning in the Advocacy Coalition Framework, Policy and Society, 36, 2: 320– 344, doi: 10.1080/ 14494035.2017.1322259 Moyson, S., Raaphorst, N., Groeneveld, S. & Van de Walle, S. (2018) ‘Organizational Socialization in Public Administration Research: A Systematic Review and Directions for Future Research’, The American Review of Public Administration, 48, 6: 610–627. Moyson, S. & Scholten, P. (2018) ‘Theories on Policy Learning: Existing Approaches and Future Challenges’, in N. F. Dotti (e), Knowledge, Policymaking and Learning for European Cities and Regions (Cheltenham: Edward Elgar). Moyson, S., Scholten, P. & Weible, C. (2017) ‘Policy Learning and Policy Change: Theorizing their Relations from Different Perspectives, Policy and Society, 36, 2: 161–177, doi: 10.1080/14494035.2017.1331879 Mulaudzi, P. A. (2017) ‘The Clash of Ideologies: Notions of Multiparty (Liberal) Democracy versus African Systems’, International Journal of African Renaissance Studies – Multi-, Inter- and Transdisciplinarity, 12, 1: 44–54. Munshi, D. & Kurian, P. (2005) ‘Imperializing Spin Cycles: A Postcolonial Look at Public Relations, Greenwashing, and the Separation of Publics’, Public Relations Review, 31, 4: 513–520. Murji, K. (2017) Racism, Policy and Politics (Bristol: Policy Press). Nair, S. & Howlett, M. (2017) ‘Policy Myopia as a Source of Policy Failure: Adaptation and Policy Learning Under Deep Uncertainty’, Policy & Politics, 45, 1: 103–118. National Research Council (1986) Proceedings of the Conference on Common Property Resource Management, April 21 –26, 1985 (Washington, DC: National Academies).

Newig, J. and Fritsch, O. (2009) ‘Environmental governance: Participatory, multi‐ level–and effective?’, Environmental Policy and Governance, 19, 3: 197–214. Newig, J. & Koontz, T. (2014) ‘Multi-level Governance, Policy Implementation and Participation: The EU’s Mandated Participatory Planning Approach to Implementing Environmental Policy’, Journal of European Public Policy, 21, 2: 248–267. Newman, J. (2019) ‘The Role of Uncertainty in Regulating E-Cigarettes: The Emergence of a Regulatory Regime, 2005-15’, Politics & Policy, 47, 2: 407–429. Newman, J. & Bird, M. (2017) ‘British Columbia’s Fast Ferries and Sydney’s Airport Link: Partisan Barriers to Learning from Policy Failure’, Policy & Politics, 45, 1: 71–85. Newman, J. & Head, B. (2017) ‘Wicked Tendencies in Policy Problems: Rethinking the Distinction between Social and Technical Problems’, Policy and Society, 36, 3: 414–429 http://dx.doi.org/10.1080/14494035.2017.1361635 Newton, K. & van Deth, J. (2010) Foundations of Comparative Politics (Cambridge: Cambridge University Press). Newton, K. & Van Deth, J. (2016) Foundations of Comparative Politics 3rd edn (Cambridge: Cambridge University Press). Nicholson-Crotty, S. & Meier, K. (2005) ‘From Perception to Public Policy: Translating Social Contructions into Poicy Designs’, in A. Schneider & H. Ingram (eds.) Deserving and Entitled: Social Construction and Public Policy (Albany: State University of New York Press). Nilsson, M., Jordan, A., Turnpenny, J., Hertin, J., Nykvist, B. & Russel, D. (2008) ‘The Use and Non-use of Policy Appraisal Tools in Public Policy Making: An Analysis of Three European Countries and the European Union’, Policy Sciences, 41, 4: 335–355. Niskanen, W. (1971) Bureaucracy and Representative Government (Chicago: Aldine Atherton). Niskanen, W. (1975) ‘Bureaucrats and Politicians’, Journal of Law and Economics, 18, 3: 617–643. Nobel Media, A. B. (2009) ‘Elinor Ostrom – Facts’ https://www.nobelprize.org/prizes/economic-sciences/2009/ostrom/facts/ Norris, P. (1997) ‘Choosing Electoral Systems: Proportional, Majoritarian and Mixed Systems’ International Political Science Review, 18, 3: 297–312. Norris, P. & Lovenduski, J. (1995) Political Recruitment: Gender, Race and Class in the British Parliament (Cambridge: Cambridge University Press). Nutley, S., Powell, A. & Davies, H. (2013) What Counts as Good Evidence (London: Alliance for Useful Evidence) http://www.alliance4usefulevidence.org/assets/What-Counts-as-Good-EvidenceWEB.pdf Nutley, S., Walter, I. & Davies, H. (2007) Using Evidence: How Research can Inform Public Services (Bristol: The Policy Press).

O’Flynn, J. (2007) ‘From New Public Management to Public Value: Paradigmatic Change and Managerial Implications’, Australian Journal of Public Administration, 66, 3: 353–366. O’Neill, F. (2000) ‘Health: The “Internal Market” and Reform of the National Service’ in D. Dolowitz (ed.) Policy Transfer and British Social Policy (Buckingham: Open University Press). O’Toole, B. & Jordan, G. (1995) Next Steps (Aldershot: Dartmouth). Oaksford, M. & Chatter, N. (2007) Bayesian Rationality (Oxford: Oxford University Press). OECD (Organisation for Economic Co-operation and Development) (1995) Governance in Transition: Public Management Reforms in OECD Countries (Paris: OECD). Office for National Statistics (2013) ‘Middle Income Households, 1977–2011/12’ (London: ONS) https://webarchive.nationalarchives.gov.uk/20160107030736/http://www.ons.gov. uk/ons/dcp171776_341133.pdf Oleman, J. (1999) ‘Unified Government, Divided Government and Party Responsiveness’, American Political Science Review, 93, 4: 821–835. Oliver, K. & Cairney, P. (2019) ‘The Dos and Don’ts of Influencing Policy: A Systematic Review of Advice to Academics’, Palgrave Communications, 5, 21: 1–11. Oliver, K. & Faul, M. (2018) ‘Networks and Network Analysis in Evidence, Policy and Practice’, Evidence and Policy, 14, 3: 369–379. Oliver, K., Innvar, S., Lorenc, T., Woodman, J. & Thomas, J. (2014a). ‘A systematic review of barriers to and facilitators of the use of evidence by policymakers’, BMC Health Services Research, 14, 1: 2 http://www.biomedcentral.com/14726963/14/2 Oliver, K., Lorenc, T. & Innvær, S. (2014b). ‘New Directions in Evidence-based Policy Research: A Critical Analysis of the Literature’. Health Research Policy and Systems, 12: 34 http://www.biomedcentral.com/content/pdf/1478-4505-1234.pdf Oliver, K. & Pearce, W. (2017) ‘Three Lessons from Evidence-based Medicine and Policy: Increase Transparency, Balance Inputs and Understand Power’, Palgrave Communications, 3: 43, doi:10.1057/s41599-017-0045-9 Olsen, J. (2009) ‘Change and Continuity: An Institutional Approach to Institutions and Democratic Government’, European Political Science Review, 1, 1: 3–32. Olson, M. (1965; 1971) The Logic of Collective Action, 1st and 2nd printing (Cambridge, MA: Harvard University Press). Olson, M. (1982) The Rise and Decline of Nations (New Haven, CT: Yale University Press). OPSR (Office of Public Services Reform) (2002) Better Government Services, http://archive.cabinetoffice.gov.uk/opsr/documents/pdf/opsr-agencies.pdf

Ormond, D. & Löffler, E. (1998) ‘New Public Management: What To Take And What To Leave’, III Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Madrid, 14–17, October, http://bvc.cgu.gov.br/bitstream/123456789/1758/1/Derry+Ormond.pdf Osborne, S. (2006) ‘The New Public Governance?’ Public Management Review, 8, 3: 377–387. Osborne, S. (2010) The New Public Governance: Emerging Perspectives on the Theory and Practice of Public Governance (London: Routledge). Ostrom, E. (1990) Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press). Ostrom, E. (1998) ‘A Behavioral Approach to the Rational Choice Theory of Collective Action: Presidential Address, American Political Science Association, 1997’, American Political Science Review, 92, 1: 1–22. Ostrom, E. (1999) ‘Institutional Rational Choice. An Assessment of the Institutional Analysis and Development Framework’ in P. Sabatier (ed.) Theories of the Policy Process (Boulder, CO: Westview Press). Ostrom, E. (2006) ‘Converting Threats into Opportunities’, PS: Political Science and Politics 39, 1: 3–12 http://www.apsanet.org/imgtest/psjan06eostrom.pdf Ostrom, E. (2007) ‘Institutional Rational Choice’ in P. Sabatier (ed.) Theories of the Policy Process 2nd edn (Cambridge, MA: Westview Press). Ostrom, E. (2008) ‘The Challenge of Common-pool Resources’, Environment: Science and Policy for Sustainable Development 50, 4: 8–21. Ostrom, E. (2009a) ‘Design Principles of Robust Property-rights Institutions: What Have We Learned?’ in G. Ingram & Y. Hong (eds.) Property Rights and Land Policies (Cambridge, MA: Lincoln Institute of Land Policy), 25–51. Ostrom, E. (2009b) ‘A General Framework for Analyzing Sustainability of SocialEcological Systems’, Science, 325, 5939: 419–422. Ostrom, E. (2011) ‘Background on the Institutional Analysis and Development Framework’, Policy Studies Journal, 39, 1: 7–27. Ostrom, E., Cox, M. & Schlager, E. (2014) ‘An Assessment of the Institutional Analysis and Development Framework and Introduction of the Social-Ecological Systems Framework’ in P. Sabatier & C. Weible (eds.) Theories of the Policy Process 3rd edn (Boulder, CO: Westview Press). Ostrom, E. & Crawford, S. (2005) ‘Classifying Rules’ in E. Ostrom Understanding institutional diversity (Princeton, New Jersey: Princeton University Press). Ostrom, V. & Ostrom, E. (1971) ‘Public Choice: A Different Approach to the Study of Public Administration’, Public Administration Review, 31, 2: 203–216. Owusu-Mensah, I. (2015) ‘Promoting Local Governance in Ghana: The Role of Akan Queen Mothers’, The Journal of Pan African Studies, 8, 9: 98–114. Pacheco-Vega, R. (2014) ‘The Impact of Elinor Ostrom’s Scholarship on Commons Governance in Mexico’, Policy Matters, 19, April: 23–34. Pacheco-Vega, R. (2015) ‘Integrating under-represented scholars into my syllabi’, Raul Pacheco-Vega, PhD 18 July

http://www.raulpacheco.org/2015/07/integrating-under-represented-scholarsinto-my-syllabi/ Page, E. (2000) ‘Future Governance and the Literature on Policy Transfer and Lesson Drawing’, ESRC Future Governance Workshop, 28 January. Page, E. (2006) ‘The Origins of Policy’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Page, E. (2018) ‘Foreward’ in C. Dunlop, C. Radaelli & P. Trein (eds.) Learning in Public Policy (London: Palgrave Macmillan). Pahl-Wostl, C. & Knieper, C. (2014) ‘The Capacity of Water Governance to Deal with the Climate Change Adaptation Challenge: Using Fuzzy Set Qualitative Comparative Analysis to Distinguish between Polycentric, Fragmented, and Centralized Regimes’, Global Environmental Change, 29: 139–154. Palier, B. (2005) ‘Ambiguous Agreement, Cumulative Change’ in W. Streeck & K. Thelen (eds.) Beyond Continuity: Institutional Change in Advanced Political Economies (Oxford: Oxford University Press). Park, C., Wilding, M. & Chung, C. (2014) ‘The Importance of Feedback: Policy Transfer, Translation and the Role of Communication’, Policy Studies, 35, 4: 397–412. Park, J. & Sovacool, B. K. (2018) ‘The Contested Politics of the Asian Atom: Peripheralisation and Nuclear Power in South Korea and Japan’, Environmental Politics, 27, 4: 686–711. Parker, C., Stern, E., Paglia, E. & Brown, C. (2009) ‘Preventable Catastrophe? The Hurricane Katrina Disaster Revisited’, Journal of Contingencies and Crisis Management, 17, 4: 206–220. Parkhurst, J. (2017) The Politics of Evidence (London: Routledge). Parsons, S. (2005) Rational Choice and Politics: A Critical Introduction (London: Continuum). Parsons, T. (1938) ‘The Role of Ideas in Social Action’, American Sociological Review, 3, 5: 652–664. Parsons, W. (1995) Public Policy (Aldershot: Edward Elgar). Partelow, S. (2018) ‘A Review of the Social-ecological Systems Framework: Applications, Methods, Modifications, and Challenges’, Ecology and Society 23, 4: 36 https://doi.org/10.5751/ES-10594-230436 Pepin-Neff, C. (2019) Flaws: Shark Bites and Emotional Public Policymaking (London: Palgrave Macmillan). Peters, B. G. (2005) Institutional Theory in Political Science: The ‘New Institutionalism’ 2nd edn (London: Continuum). Peters, B. G. (2016) ‘Institutional Analysis: Progress and Problems’ in H. Keman & J. Wold-endorp (eds.) Handbook of Research Methods and Applications in Political Science (Cheltenham: Edward Elgar), 127–140. Peters, B. G., Capano, G., Howlett, M., Mukherjee, I., Chou, M. H. & Ravinet, P. (2018) Designing for Policy Effectiveness: Defining and Understanding a Concept (Cambridge: Cambridge University Press).

Peters, B. G. & Hogwood, B. W. (1985) ‘In Search of the Issue-Attention Cycle’, Journal of Politics, 47, 1: 238–253. Peters, B. G. & Pierre, J. (2004) ‘Multi-level Governance and Democracy’ in I. Bache & M. Flinders (eds.) Multi-level Governance (Oxford: Oxford University Press). Peters, B. G., Pierre, J. & King, D. (2005) ‘The Politics of Path Dependency: Political Conflict in Historical Institutionalism’, Journal of Politics, 67, 4: 1275– 1300. Petticrew, M. & Roberts, H. (2006) Systematic Reviews in the Social Sciences (Oxford: Blackwell). Pidd, H. & Goldenberg, S. (2011) ‘Germany Suspends Power Station Extension Plans as Nuclear Jitters Spread’, Guardian, 14 March, http://www.guardian.co.uk/environment/2011/mar/14/germany-japan-nuclearindustry Pielke Jr, R. (2007) The Honest Broker: Making Sense of Science in Policy and Politics (Cambridge: Cambridge University Press). Pierce, J., Giordono, L., Peterson, H. and Hicks, K. (2019) ‘Common approaches for studying advocacy: Review of methods and model practices of the Advocacy Coalition Framework’, The Social Science Journal (in press) https://doi.org/10.1016/j.soscij.2019.06.005 Pierce, J., Peterson, H., Jones, M., Garrard, S. & Vu, T. (2017) ‘There and Back Again: A Tale of the Advocacy Coalition Framework’, Policy Studies Journal, 45, S1: S13–46. Pierce, J., Siddiki, S., Jones, M., Schumacher, K., Pattison, A. & Peterson, H. (2014a) ‘Social Construction and Policy Design: A Review of Past Applications’, Policy Studies Journal, 42, 1: 1–29. Pierce, J., Smith-Walter, A. & Peterson, H. (2014b) ‘Research Design and the Narrative Policy Framework’ in M. D. Jones, E. A. Shanahan & M. K. McBeth (eds.) The Science of Stories (New York: Palgrave Macmillan). Pierre, J. & Peters, B. G. (2000) Governance, Politics and the State (London: Red Globe Press). Pierre, J. & Stoker, G. (2000) ‘Towards Multi-level Governance’ in P. Dunleavy et al. (eds.) Developments in British Politics 6th edn (London: Red Globe Press). Pierson, P. (1993) ‘When Effect Becomes Cause: Policy Feedback and Political Change’, World Politics, 45, 4: 595–628. Pierson, P. (2000a) ‘Increasing Returns, Path Dependence, and the Study of Politics’, The American Political Science Review, 94, 2: 251–267. Pierson, P. (2000b) ‘Introduction: Investigating the Welfare State at Century’s End’ in P. Pierson (ed.) The New Politics of the Welfare State (Oxford: Oxford University Press). Pierson, P. (2000c) ‘Post-Industrial Pressures on the Mature Welfare States’ in P. Pierson (ed.) The New Politics of the Welfare State (Oxford: Oxford University Press).

Pollack, M. (2001) ‘International Relations Theory and European Integration’, Journal of Common Market Studies, 39, 2: 221–244. Polsby, N. (1960) ‘How to Study Community Power: The Pluralist Alternative’, Journal of Politics, 22, 3: 474–484. Polsby, N. (1968) ‘On Intersections between Law and Political Science’, Stanford Law Review, 21, 1: 142–151. Polsby, N. (1980) Community Power and Political Theory 2nd edn (New Haven, CT: Yale University Press). Poteete, A., Janssen, M. & Ostrom, E. (2010) Working Together: Collective Action, the Commons and Multiple Methods in Practice (Princeton, NJ: Princeton University Press). Poulantzas, N. (1986) ‘Class Power’ in S. Lukes (ed.) Power (New York: New York University Press). Preiser, R., Biggs, R., De Vos, A. & Folke, C. (2018) ‘Social-Ecological Systems as Complex Adaptive Systems: Organizing Principles for Advancing Research Methods and Approaches’, Ecology and Society 23, 4: 46 https://doi.org/10.5751/ES-10558-230446 Pressman, J. & Wildavsky, A. (1973) Implementation (Berkeley: University of California Press). Pressman, J. & Wildavsky, A. (1979) Implementation 2nd edn (Berkeley: University of California Press). Princen, S. (2007) ‘Advocacy Coalitions and the Internationalization of Public Health Policies’, Journal of Public Policy, 27, 1: 13–33. Pülzl, H. & Treib, O. (2007) ‘Implementing Public Policy’ in F. Fischer, G. Miller & M. Sidney (eds.) Handbook of Public Policy Analysis (London: CRC Press). Putnam, R. (2001) Bowling Alone: The Collapse and Revival of American Community (London: Simon & Schuster). Pykett, J., Jones, R. & Whitehead, M. (2017) Psychological Governance and Public Policy (London: Routledge). Quiggin, J. (2006) ‘Economic Constraints on Public Policy’, in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Rabbi, M. F. & Sabharwal, M. (2018) ‘Dynamics of Nuclear Energy Policies, in India: A Case Study on the Emergence of Nuclear Safety Regulatory Authority’, Indian Journal of Public Administration, 64, 4: 664–685. Radaelli, C. (1995) ‘The Role of Knowledge in the Policy Process’, Journal of European Public Policy, 2, 2: 159–183. Radaelli, C. (1999) ‘Harmful Tax Competition in the EU: Policy Narratives and Advocacy Coalitions’, Journal of Common Market Studies, 37, 4: 661–682. Radaelli, C. & Dunlop, C. (2013) ‘Learning in the European Union: Theoretical Lenses and Meta-theory’, Journal of European Public Policy, 20, 6: 923–940. Radin, B. (2000) Beyond Machiavelli: Policy Analysis Comes of Age (Washington, DC: Georgetown University Press).

Radin, B. (2013) Beyond Machiavelli: Policy Analysis Reaches Midlife (Washington, DC: Georgetown University Press). Ramgotra, M. (2015) ‘On Teaching Political Theory to Undergraduates’, PSA Women and Politics Specialist Group December 21 https://psawomenpolitics.com/2015/12/21/on-teaching-political-theory-toundergraduates/ Ramia, G. & Carney, T. (2010) ‘The Rudd Government’s Employment Services Agenda: Is it Post-NPM and Why is that Important?’ Australian Journal of Public Administration, 69, 3: 263–273. Rawat, P. & Morris, J. (2016) ‘Kingdon’s “Streams” Model at Thirty: Still Relevant in the Twenty First Century?’ Politics and Policy, 44, 4: 608–638. Rayner, J., Howlett, M., Wilson, J., Cashore, B. & Hoberg, G. (2001) ‘Privileging the Sub-sector’, Forest Policy and Economics 2: 319–332. Read, M. (1996) The Politics of Tobacco: Policy Networks and the Cigarette Industry (Aldershot: Avebury). Reger, J. (2004) ‘Organizational “Emotion Work” Through Consciousness-Raising: An Analysis of a Feminist Organization’, Qualitative Sociology, 27, 2: 205–222. Reich, R. (1988) ‘Introduction’ in R. Reich (ed.) The Power of Public Ideas (Cambridge, MA: Ballinger). Rhodes, R. A. W. (1994) ‘The Hollowing Out of the State’, Political Quarterly, 65: 138–151. Rhodes, R. A. W. (1997) Understanding Governance (Buckingham: Open University Press). Rhodes, R. A. W. (2000) Transforming British Government, Vol. 1 (London: Palgrave Macmillan). Rhodes, R. A. W. (2006a) ‘Old Institutionalisms’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Rhodes, R. A. W. (2006b) ‘Policy Network Analysis’ in M. Moran, M. Rein & R. Goodin (eds.) The Oxford Handbook of Public Policy (Oxford: Oxford University Press). Rhodes, R. A. W. (ed.) (2009) The Australian Study of Politics (Basingstoke: Palgrave Macmillan). Rhodes, R. A. W. (2013) ‘Political Anthropology and Civil Service Reform: Prospects and Limits’, Policy & Politics, 41, 4: 481–496. Rhodes, R. A. W., Binder, S. & Rockman, B. (2006) ‘Preface’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Richards, D. & Smith, M. (2002) Governance and Public Policy in the UK (Oxford: Oxford University Press). Richards, D. & Smith, M. (2004) ‘The “Hybrid State”’ in S. Ludlam & M. Smith (eds.) Governing as New Labour (London: Red Globe Press).

Richardson, J. J. (1982a) ‘Convergent Policy Styles in Europe?’ in J. J. Richardson (ed.) Policy Styles in Western Europe (London: Allen & Unwin). Richardson, J. J. (1982b) Policy Styles in Western Europe (London: Allen & Unwin). Richardson, J. J. (2000) ‘Government, Interest Groups and Policy Change’, Political Studies, 48: 1006–1025. Richardson, J. J. (2008) ‘Policy-making in the EU: Interests, Ideas and Garbage Cans of Primeval Soup’, in J. J. Richardson (ed.) European Union: Power and Policymaking, 3rd edn (London: Routledge). Richardson, J. J. (2018) ‘The Changing British Policy Style: From Governance to Government?’ British Politics, 13, 2: 215–233. Richardson, J. J., Gustafsson, G. & Jordan, G. (1982) ‘The Concept of Policy Style’ in J. J. Richardson (ed.) Policy Styles in Western Europe (London: Allen & Unwin). Richardson, J. J. & Jordan, G. (1979) Governing Under Pressure: The Policy Process in a Post-Parliamentary Democracy (Oxford: Robertson). Richardson, J. J. & Jordan, G. (1983) ‘Over-crowded Policymaking: Some British and European Reflections’, Policy Sciences, 15, 3: 247–268. Richardson, J. J., Maloney, W. A. & Rüdig, W. (1992) ‘The Dynamics of Policy Change: Lobbying and Water Privatization’, Public Administration, 70: 157–175. Ridgeway, C. (2011) Framed by Gender (Oxford: Oxford University Press). Riker, W. (1980) ‘Implications from the Disequilibrium of Majority Rule for the Study of Institutions’, American Political Science Review, 74, 2: 432–446. Riker, W. (1982) Liberalism Against Populism (San Francisco: Freeman). Riker, W. (1986) The Art of Political Manipulation (New Haven, CT: Yale University Press). Rittel, H. & Webber, M. (1973) ‘Dilemmas in a General Theory of Planning’, Policy Sciences, 4, 2: 155–169. Roberts, N. & King, P. (1991) ‘Policy Entrepreneurs: Their Activity Structure and Function in the Policy Process’, Journal of Public Administration Research and Theory 1, 2: 147–175. Robson, J., Davidson-Hunt, I., Delaney, A., Lichtensetein, G., Magole, L. & Mead, A. (2014) ‘Remembering Elinor Ostrom’, Policy Matters, 19, April: 1–126. Rochefort, D. A. & Cobb, R. W. (1994) ‘Problem Definition: An Emerging Perspective’ in D. A. Rochefort & R. W. Cobb (eds.) The Politics of Problem Definition (Kansas City: University Press of Kansas). Roemer, J. (1988) Free to Lose: An Introduction to Marxist Economic Philosophy (London: Radius). Room, G. (2011) Complexity, Institutions and Public Policy (Cheltenham: Edward Elgar). Room, G. (2016) Agile Actors on Complex Terrains: Transformative Realism and Public Policy (London: Routledge).

Rosamond, B. (2000) Theories of European Integration (London: Red Globe Press). Rose, R. (1984) Do Parties Make a Difference? (London: Macmillan). Rose, R. (1986) ‘Steering the Ship of State: One Rudder but Two Pairs of Hands’, Appendix 15 of HC 92-II, Civil Servants and Ministers: Duties and Responsibilities, Seventh Report from the Treasury and Civil Service Committee, Session 1985–86. Rose, R. (1987) Ministers and Ministries: A Functional Analysis (Oxford: Clarendon Press). Rose, R. (1990) ‘Inheritance Before Choice in Public Policy’, Journal of Theoretical Politics, 2, 3: 263–291. Rose, R. (1991) ‘What Is Lesson-Drawing?’ Journal of Public Policy, 11, 1: 3–30. Rose, R. (1993) Lesson-Drawing in Public Policy (New York: Chatham House). Rose, R. (2005) Learning From Comparative Public Policy: A Practical Guide (London: Routledge). Rose, R. & Davies, P. (1994) Inheritance in Public Policy: Change Without Choice in Britain (New Haven, CT: Yale University Press). Ross, D. (2005) Economic Theory and Cognitive Science: Microexplanation (Cambridge, MA: MIT Press). Rutazibwa, O. & Shilliam, R. (2018) Routledge Handbook of Postcolonial Politics (London: Routledge). Sabatier, P. (1986) ‘Top-down and Bottom-up Approaches to Implementation Research: A Critical Analysis and Suggested Synthesis’, Journal of Public Policy, 6, 1: 21–48. Sabatier, P. (1987) ‘Knowledge, Policy-oriented Learning, and Policy Change’, Knowledge, 8, 4: 649–692. Sabatier, P. (1988) ‘An Advocacy Coalition Framework of Policy Change and the Role of Policy-Oriented Learning Therein’, Policy Sciences, 21, 2–3: 129–168. Sabatier, P. (1991) ‘Towards Better Theories of the Policy Process’, PS: Political Science and Politics, 24, 2: 147–156. Sabatier, P. (1993) ‘Policy Change over a Decade or More’ in P. Sabatier & H. Jenkins-Smith (eds.) Policy Change and Learning: An Advocacy Coalition Approach (Boulder, CO: Westview Press). Sabatier, P. (1998) ‘The Advocacy Coalition Framework: Revisions and Relevance for Europe’, Journal of European Public Policy, 5, 1: 98–130. Sabatier, P. (1999) Theories of the Policy Process (Boulder: Westview Press). Sabatier, P. (2007a) ‘The Need for Better Theories’ in P. Sabatier (ed.) Theories of the Policy Process 2nd edn (Cambridge, MA: Westview Press). Sabatier, P. (2007b) ‘Fostering the Development of Policy Theory’ in P. Sabatier (ed.) Theories of the Policy Process 2nd edn (Cambridge, MA: Westview Press). Sabatier, P., Hunter, S. & McLaughlin, S. (1987) ‘The Devil Shift: Perceptions and Misperceptions of Opponents’, The Western Political Quarterly, 40, 3: 449–476.

Sabatier, P. & Jenkins-Smith, H. (1993) ‘The Advocacy Coalition Framework: Assessment, Revisions and Implications for Scholars and Practitioners’ in P. Sabatier & H. Jenkins-Smith (eds.) Policy Change and Learning: An Advocacy Coalition Approach (Boulder, CO: Westview Press). Sabatier, P. & Weible, C. (2007) ‘The Advocacy Coalition Framework: Innovations and Clarifications’, in P. Sabatier (ed.) Theories of the Policy Process (Boulder, CO: Westview Press). Sandbu, M. (2015) Europe’s Orphan; The Future of the Euro and the Politics of Debt (Princeton NJ: Princeton University Press). Sanders, D. (2010) ‘Behaviouralism’ in D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science (London: Red Globe Press). Sanders, D. (2017) ‘Behavioural Analysis’ in V. Lowndes, D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science 4th edn (London: Red Globe Press). Sanders, E. (2006) ‘Historical Institutionalism’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Sanderson, I. (2009) ‘Intelligent Policy Making for a Complex World: Pragmatism, Evidence and Learning’, Political Studies, 57: 699–719. Sandler, T. (1997) Global Challenges: An Approach to Environmental, Political and Economic Problems (Cambridge: Cambridge University Press). Sandler, T. (2004) Global Collective Action (Cambridge: Cambridge University Press). Sandler, T. & Enders, W. (2004) ‘An Economic Perspective on Transnational Terrorism’, European Journal of Political Economy, 20, 2: 301–316. Sarmiento-Mirwaldt, K. (2013) ‘Can Multiple Streams Predict the Territorial Cohesion Debate in the EU?’ European Urban and Regional Studies, doi: 10.1177/0969776413481984 Schain, M. & Menon, A. (eds.) (2007) Comparative Federalism: The European Union and the United States (New York: Oxford University Press). Scharpf, F. (1988) ‘The Joint-Decision Trap: Lessons from German Federalism and European Integration’, Public Administration, 66, 3: 239–278. Scharpf, F. (1997) ‘Introduction: The Problem-solving Capacity of Multi-level Governance’, Journal of European Public Policy, 4, 4: 520–538. Schattschneider, E. E. (1960) The Semi-sovereign People (Fort Worth, TX: Harcourt Brace, 1975 edition). Schlager, E. (1995) ‘Policy Making and Collective Action: Defining Coalitions within the Advocacy Coalition Framework’, Policy Sciences, 28, 3: 243–270. Schlager, E. (2004) ‘Common-pool resource theory’ in R. Durant, D. Fiorino & R. O’Leary (eds.) Environmental Governance Reconsidered (Cambridge, MA: MIT Press), 145–176. Schlager, E. (2007) ‘A Comparison of Frameworks, Theories, and Models of Policy Processes’ Theory’ in P. Sabatier (ed.) Theories of the Policy Process vol. 2

(Cambridge, MA: Westview Press). Schlager, E. & Cox, M. (2018) ‘The IAD Framework and the SES Framework’ in C. Weible & P. Sabatier (eds.) Theories of the Policy Process 4th edn (Chicago: Westview Press). Schmidt, V. (2006) ‘Institutionalism’ in C. Hay, M. Lister & D. Marsh (eds.) The State: Theories and Issues (London: Red Globe Press). Schmidt, V. (2008) ‘Discursive Institutionalism: The Explanatory Power of Ideas and Discourse’, Annual Review of Political Science, 11: 303–326. Schmidt, V. (2009) ‘Discursive Institutionalism: The Explanatory Power of Ideas and Discourse’, Annual Review of Political Science, 11: 303–326. Schmidt, V. (2010) ‘Taking Ideas and Discourse Seriously: Explaining Change through Discursive Institutionalism as the Fourth “New Institutionalism’”, European Political Science Review, 2, 1: 1–25. Schmidt, V. & Radaelli, C. (2004) ‘Policy Change and Discourse in Europe: Conceptual and Methodological Issues’, West European Politics, 27, 2: 183– 210. Schneider, A. & Ingram, H. (1993) ‘Social Construction of Target Populations: Implications for Politics and Policy’, American Political Science Review, 87, 2: 334–347. Schneider, A. & Ingram, H. (1997) Policy Design for Democracy (Kansas: University of Kansas Press). Schneider, A. & Ingram, H. (2005a) ‘A Response to Peter deLeon’, Public Administration Review 65, 5: 638–640. Schneider, A. & Ingram, H. eds. (2005b) Deserving and Entitled: Social Construction and Public Policy (Albany: State University of New York Press). Schneider, A. & Ingram, H. (2019) ‘Social Constructions, Anticipatory Feedback Strategies, and Deceptive Public Policy’, Policy Studies Journal, 47, 2: 206–236. Schneider, A., Ingram, H. & DeLeon, P. (2014) ‘Democratic Policy Design: Social Construction of Target Populations’ in (eds.) P. Sabatier & C. Weible Theories of the Policy Process (Boulder: Westview Press). Schneider, A. & Sidney, M. (2009) ‘What is Next for Policy Design and Social Construction Theory?’, Policy Studies Journal, 37, 1: 103–119. Schofield, J. (2004) ‘A Model of Learned Implementation’, Public Administration, 82, 2: 283–308. Schofield, J. & Sausman, C. (2004) ‘Symposium on Implementing Public Policy’, Public Administration, 82, 2: 235–248. Schultz, K. (1995) ‘The Politics of the Business Cycle’, British Journal of Political Science, 25, 1: 79–99. Schumpeter, J. (1942) Capitalism, Socialism and Democracy (London: Allen & Unwin). Scott, J. W. (1986) ‘Gender: A Useful Category of Historical Analysis’, American Historical Review, 91, 5: 1053–1075.

Seager, A. (2008) ‘Ministers Risk Missing Key Labour Target’ Guardian, 11 June, http://www.guardian.co.uk/society/2008/jun/11/socialexclusion.children1 Sebők, M. & Berki, T. (2018) ‘Punctuated Equilibrium in Democracy and Autocracy: An Analysis of Hungarian Budgeting between 1868 and 2013’, European Political Science Review, 10, 4: 589–611. Sementelli, A. (2007) ‘Distortions of Progress: Evolutionary Theories and Public Administration’, Administration & Society, 39, 6: 740–760. Shaffer, R. (2017) ‘Cognitive Load and Issue Engagement in Congressional Discourse’, Cognitive Systems Research 44: 89–99. Shanahan, E., Adams, S., Jones, M. & McBeth, M. (2014) ‘The Blame Game: Narrative Persuasiveness of the Intentional Causal Mechanism’ in M. D. Jones, E. A. Shanahan & M. K. McBeth (eds.) The Science of Stories (New York: Palgrave Macmillan). Shanahan, E., Jones, M., McBeth, M. & Lane, R. (2013) ‘An Angel on the Wind: How Heroic Policy Narratives Shape Policy Realities’, Policy Studies Journal, 41, 3: 453–483. Shanahan, E., Jones, M., McBeth, M. & Radaelli, C. (2018a) ‘The Narrative Policy Framework’ in C. Weible & P. Sabatier (eds.) Theories of the Policy Process 4th edn (Chicago: Westview Press). Shanahan, E., Raile, E., French, K. & McEvoy, J. (2018b) ‘Bounded Stories’, Policy Studies Journal, 46, 4: 922–948. Shankardass, K., Muntaner, C., Kokkinen, L., Shahidi, F., Freiler, A., Oneka, G., Bayoumi, A. & O’Campo, P. (2018) ‘The Implementation of Health in All Policies Initiatives: A Systems Framework for Government Action’, Health Research Policy and Systems, 16, 26: 1–10. Sharkansky, I. (1972) Public Administration 2nd edn (Chicago: Markham). Sharman, J. (2008) ‘Power and Discourse in Policy Diffusion: Anti-Money Laundering in Developing States’, International Studies Quarterly, 52, 3: 635– 656. Sharman, J. (2010) ‘Dysfunctional Policy Transfer in National Tax Blacklists’, Governance, 23, 4: 623–629. Shepsle, K. (2006) ‘Rational Choice Institutionalism’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Shepsle, K. & Bonchek, M. (1997) Analyzing Politics: Rationality, Behaviour and Institutions (London: Norton). Shugart, M. (2006) ‘Comparative Executive-Legislative Relations’ in R. Rhodes, S. Binder & B. Rockman (eds.) The Oxford Handbook of Political Institutions (Oxford: Oxford University Press). Sicilia, M., Guarini, E., Sancion, A., Andreani, M. & Ruffini, R. (2016) ‘Public Services Management and Co-production in Multi-level Governance Settings’, International Review of Administrative Sciences, 82, 1: 8–27.

Simmons, R. (2018) ‘Using Cultural Theory to Navigate the Policy Process’, Policy and Politics, 46, 2: 235–253. Simon, H. (1957) Models of Man: Social and Rational (New York: John Wiley). Simon, H. (1960) The New Science of Management Decisions (Englewood Cliffs, NJ: Prentice-Hall). Simon, H. (1976) Administrative Behavior 3rd edn (London: Macmillan). Simon, H. (1983) Reason in Human Affairs (Oxford: Blackwell). Sinclair, J. (2004) ‘Globalization, Supranational Institutions and Media’, in J. D. H. Downing, D. McQuail, P. Schlesinger & E. Wartella (eds.) The Sage Handbook of Media Studies (Thousand Oaks, CA: Sage). Slaughter, A. (2004) A New World Order (Princeton, NJ: Princeton University Press). Sloman, S. & Fernbach, P. (2017) The Knowledge Illusion: Why We Never Think Alone (London: Penguin). Smith, A. (2000) ‘Policy Networks and Advocacy Coalitions: Explaining Policy Change and Stability in UK Industrial Pollution Policy?’, Environment and Planning C: Government and Policy 2000, 18, 1: 95–114. Smith, G. & May, D. (1993) ‘The Artificial Debate Between Rationalist and Incrementalist Models of Decision Making’ in M. Hill (ed.) The Policy Process: A Reader (Hemel Hempstead: Harvester Wheatsheaf). Smith, K. (2002) ‘Typologies, Taxonomies and the Benefits of Policy Classification’, Policy Studies Journal, 30, 3: 379–395. Smith, K. (2013) Beyond Evidence Based Policy in Public Health: The Interplay of Ideas (Basingstoke: Palgrave Macmillan). Smith, K. & Larimer, C. (2009) The Public Policy Theory Primer (Boulder, CO: Westview Press). Smith, M. (2009) Power and the State (London: Red Globe Press). Smith, V. (2018) Bargaining Power Health Policy-making from England and New Zealand (London: Palgrave Macmillan). Dolowitz, D., Keeley, M. & Medearis, D. (2012) ‘Stormwater Management: Can We Learn From Others?’, Policy Studies, 33, 6: 501–521. Sørensen, E. & Torfing, J. (2009) ‘Making Governance Networks Effective and Democratic Through Metagovernance’, Public Administration, 87, 2: 234–258. Soss, J. (2005) ‘Making Clients and Citizens: Welfare Policy as a Source of Status, Belief, and Action’, in A. Schneider & H. Ingram (eds.) (2005) Deserving and Entitled: Social Construction and Public Policy (Albany: State University of New York Press). Spivak, G. ‘Can the Subaltern Speak?’ in C. Nelson & L. Grossberg (eds.) Marxism and the Interpretation of Culture (London: Macmillan). St Denny, E. (2016) Explaining the Emergence and Gradual Transformation of Policy Regimes: The Case of Contemporary French Prostitution Policy (19462016), PhD thesis (Nottingham: Nottingham Trent University).

Stein, M. & Turkiewitsch, L. (2008) ‘The Concept of Multi-Level Governance in Studies of Federalism’, Paper Presented at the 2008 International Political Science Association International Conference, Montréal, May, http://montreal2008.ipsa.org/site/images/PAPERS/section3/RC%2028%20Stein %20Turewitsch%203.4.pdf Steinberger, P. (1980) ‘Typologies of Public Policy: Meaning Construction and the Policy Process’, Social Science Quarterly, 61, 2: 185–197. Steinmo, S. (2008) ‘Historical Institutionalism’ in D. Della Porta & M. Keating (eds.) Approaches and Methodologies in the Social Sciences (Cambridge: Cambridge University Press). Stensöta, H., Wängnerud, L. & Svensson, R. (2015) ‘Gender and Corruption: The Mediating Power of Institutional Logics’, Governance, 28, 4: 475–496. Stephens, J., Hansson, A., Liu, Y., de Coninck, H. & Vajjhala, S. (2011) ‘Characterizing the International Carbon Capture and Storage Community’, Global Environmental Change, 21, 2: 379–390. Stephenson, P. (2013) ‘Twenty Years of Multi-level Governance: “Where Does It Come From? What Is It? Where Is It Going?”’, Journal of European Public Policy, 20, 6: 817–837. Stern, P., Dietz, T., Dolˇsak, N., Ostrom, E. & Stonich, S. (2002) ‘Knowledge and Questions After 15 Years of Research’ in E. Ostrom, T. Dietz, N. Dolsak, P. Stern, S. Stonich & E. Weber (eds.) The Drama of the Commons (Washington, DC: National Academies). Stoeltje, B. J. (1997) ‘Asante Queen Mothers’, Annals of the New York Academy of Sciences, 810, 1: 41–71. Stoker, G. (2004) Transforming Local Governance: From Thatcherism to New Labour (London: Red Globe Press). Stoker, G. (2010) ‘Translating Experiments into Policy’, The ANNALS of the American Academy of Political and Social Science, 628, 1: 47–58. Stoker, G. & Marsh, D. (2002) ‘Introduction’ in D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science (London: Red Globe Press). Stolfi, F. (2010) ‘Testing Structuralist and Interpretative Explanations of Policy Change: The Case of Italy’s Budget Reform’, Governance, 23, 1: 109–132. Stone, D. (1989a) ‘Causal Stories and the Formation of Policy Agendas’, Political Science Quarterly, 104: 281–300. Stone, D. (1999) ‘Learning Lessons and Transferring Policy across Time, Space and Disciplines’, Politics, 19, 1: 51–59. Stone, D. (2000) ‘Non-Governmental Policy Transfer: The Strategies of Independent Policy Institutes’, Governance, 13, 1: 45–62. Stone, D. (2002a) Policy Paradox: The Art of Political Decision Making, rev. edn (London: Norton). Stone, D. (2004) ‘Transfer Agents and Global Networks in the “Transnationalization” of Policy’, Journal of European Public Policy, 11, 3: 545– 566.

Stone, D. (2010) ‘Private Philanthropy or Policy Transfer? The Transnational Norms of the Open Society Institute’, Policy & Politics, 38, 2: 269–287. Stone, D. (2012) ‘Transfer and Translation of Policy’, Policy Studies, 33, 6: 483– 499. Stone, D. (2017) ‘Understanding the Transfer of Policy Failure: Bricolage, Experimentalism and Translation’, Policy & Politics, 45, 1: 55–70. Stone, D. A. (1989b) ‘Causal Stories and the Formation of Policy Agendas’, Political Science Quarterly, 104: 281–300. Stone, D. A. (2002b) Policy Paradox: The Art of Political Decision Making rev. edn (London: Norton). Strassheim, H. (2019) ‘Behavioural Mechanisms and Public Policy Design: Preventing Failures in Behavioural Public Policy’, Public Policy and Administration, Early View doi: 10.1177/0952076719827062 Streeck, W. & Thelen, K. (2005) ‘Introduction: Institutional Change in Advanced Political Economies’ in W. Streeck & K. Thelen (eds.) Beyond Continuity: Institutional Change in Advanced Political Economies (Oxford: Oxford University Press). Strøm, K. (1990) Minority Government and Majority Rule (Cambridge: Cambridge University Press). Studlar, D. (2002) Tobacco Control: Comparative Politics in the United States and Canada (Peterborough, ON: Broadview Press). Studlar, D. (2004) Tobacco Control Policy in a Shrinking World: How Much Policy Learning?’ in E. Vigoda-Gadot & D. Levi-Faur (eds.) International Public Policy and Management: Policy Learning and Policy Beyond Regional, Cultural and Political Boundaries (New York: Marcel Dekker). Studlar, D. (2007a) ‘What Explains Policy Change in Tobacco Control Policy in Advanced Industrial Democracies?’, Paper to PSA Conference, Bath, April, http://www.psa.ac.uk/2007/pps/Studlar.pdf Studlar, D. (2007b) ‘Ideas, Institutions and Diffusion: What Explains Tobacco Control Policy in Australia, Canada and New Zealand?’, Common-wealth and Comparative Politics, 45, 2: 164–184. Studlar, D. & Cairney, P. (2014) ‘Conceptualizing Punctuated and Non-Punctuated Policy Change: Tobacco Control in Comparative Perspective’, International Review of Administrative Sciences, 80, 3: 513–531. Studlar, D. & Cairney, P. (2019) ‘Multilevel Governance, Public Health and the Regulation of Food: Is Tobacco Control Policy a Model?’ Journal of Public Health Policy, doi: 10.1057/ s41271-019-00165-6 Sullivan, H. (2011) ‘“Truth” Junkies: Using Evaluation in UK Public Policy’, Policy & Politics, 39, 4: 499–512. Swann, W. & Kim, S. (2018) ‘Practical Prescriptions for Governing Fragmented Governments’, Policy and Politics, 46, 2: 273–292. Tangney, P. (2018) ‘Between Conflation and Denial: The Politics of Climate Expertise in Australia’, Australian Journal of Political Science, doi:

10.1080/10361146.2018.1551482 Taylor, D. & Balloch, S. (2005) ‘The Politics of Evaluation: An Overview’, in D. Taylor & S. Balloch (eds.) The Politics of Evaluation: Participation and Policy Implementation 1–17 (Bristol: The Policy Press), 1–17. Teisman, G. & Klijn, E. (2008) ‘Complexity Theory and Public Management’, Public Management Review, 10, 3: 287–297. Tenbensel, T. (2015) ‘Complexity and Health Policy’ in R. Geyer & P. Cairney (eds.) Handbook on Complexity and Public Policy (Cheltenham: Edward Elgar). Thaler, R. & Sunstein, C. (2008) Nudge: Improving Decisions about Health, Wealth and Happiness (London: Yale University Press). Thelen, K. (2004) How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan (Cambridge: Cambridge University Press). Thelen, K. & Steinmo, S. (1992) ‘Historical Institutionalism in Comparative Politics’, in S. Steinmo, K. Thelen & F. Longstreth (eds.) Structuring Politics: Historical Institutionalism in Comparative Analysis (Cambridge: Cambridge University Press). Topp, L., Mair, D., Smillie, L. & Cairney, P. (2018) ‘Knowledge Management for Policy Impact: The Case of The European Commission’s Joint Research Centre’, Palgrave Communications, 4: 87, https://doi.org/10.1057/s41599-0180143-3 Torres, F. (2013) ‘The EMU’s Legitimacy and the ECB as a Strategic Political Player in the Crisis Context’, Journal of European Integration, 35, 3: 287–300. True, J. (2003) ‘Mainstreaming Gender in Global Public Policy’, International Feminist Journal of Politics, 5, 3: 368–396. True, J. L., Jones, B. D. & Baumgartner, F. R. (2007) ‘Punctuated Equilibrium Theory’ in P. Sabatier (ed.) Theories of the Policy Process 2nd edn (Cambridge, MA: Westview Press). Tsebelis, G. (1990) Nested Games: Rational Choice in Comparative Politics (Berkeley: California University Press). Tsebelis, G. (2002) Veto Players: How Political Institutions Work (Princeton, NJ: Princeton University Press). Tuckett, D., Mandel, A., Mangalagiu, D., Abramson, A., Hinkel, J., Katsikopoulos, K., Kirman, A., Malleret, T., Mozetic, I., Ormerod, P. & Smith, R. E. (2015) ‘Uncertainty, Decision Science, and Policy Making: A Manifesto for a Research Agenda’, Critical Review, 27, 2: 213–242. Tuckett, D. & Nicolic, M. (2017) ‘The Role of Conviction and Narrative in Decisionmaking Under Radical Uncertainty’, Theory and Psychology, 27, 4: 501–523. Tummers, L. L., Bekkers, V., Vink, E. & Musheno, M. (2015) ‘Coping During Public Service Delivery: A Conceptualization and Systematic Review of the Literature’, Journal of Public Administration Research and Theory, 25, 4: 1099–1126. Turnbull, N. & Hoppe, R. (2018) ‘Problematizing “Wickedness”: A Critique of the Wicked Problems Concept, from Philosophy to Practice’, Policy and Society,

June, 29, 1–23 https://doi.org/10.1080/14494035.2018.1488796 Turnpenny, J., Jordan, A., Benson, D. & Rayner, T. (2015) ‘The Tools of Policy Formulation: An Introduction’ in A. Jordan & J. Turnpenny (eds.) The Tools of Policy Formulation (Cheltenham: Edward Elgar). United Nations Population Division (2017) World Population Prospects: The 2017 Revision (New York: UN) https://population.un.org/wpp/Publications/Files/WPP2017_VolumeI_Comprehensive-Tables.pdf van der Heijden, J., Kuhlmann, J., Lindquist, E. & Wellstead, A. (2019) ‘Have Policy Process Scholars Embraced Causal Mechanisms? A Review of Five Popular Frameworks’, Public Policy and Administration, Early View doi: 10.1177/0952076718814894 Vilpišauskas, R. (2013) ‘Eurozone Crisis and European Integration: Functional Spillover, Political Spillback’, Journal of European Integration, 35, 3: 361–373. Vis, B. (2019) ‘Heuristics and Political Elites’ Judgment and Decision-Making’, Political Studies Review, 17, 1: 41–52. Wade, R. (2002) ‘US Hegemony and the World Bank: The Fight over People and Ideas’, Review of International Political Economy, 9, 2: 215–243. Wade, R. (2006) ‘The Invisible Hand of the American Empire’, Ethics and International Affairs, 17, 2: 77–88. Walker, J. L. (1969) ‘The Diffusion of Innovations among the American States’, American Political Science Review, 63, 3: 880–899. Walsh, J. (2000) ‘When Do Ideas Matter? Explaining the Successes and Failures of Thatcherite Ideas’, Comparative Political Studies, 33, 4: 483–516. Wang, Y. (2016) ‘Homology and Isomorphism: Bourdieu in conversation with New Institutionalism’, The British Journal of Sociology, 67, 2: 348–370. Ward, H. (1987) ‘Structural Power: A Contradiction in Terms?’ Political Studies, 35, 4: 593–610. Ward, H. (2002) ‘Rational Choice’ in D. Marsh & G. Stoker (eds.) Theory and Methods in Political Science (London: Red Globe Press). Ward, H. & Weale, A. (2010) ‘Is Rule by Majorities Special?’, Political Studies, 58, 1: 26–46. Watts, R. (2007) ‘The United Kingdom as a Federalized or Regionalized Union’, in A. Trench (ed.) Devolution and Power in the United Kingdom (Manchester: Manchester University Press). Waylen, G. (2014) ‘Informal Institutions, Institutional Change, and Gender Equality’, Political Research Quarterly, 67, 1: 212–223. Weale, A. (2001) ‘Deliberative Democracy: Science Advice, Democratic Responsiveness and Public Policy’, Science and Public Policy, 28, 6: 413–421. Weaver, R. & Rockman, B. (1993) ‘Assessing the Effects of Institutions’ in R. Weaver & B. Rockman (eds.) Do Institutions Matter? (Washington, DC: The Brookings Institution).

Weber, M. (with G. Roth & C. Wittich) (1978) Economy and Society: An Outline of Interpretive Sociology, Vol. 1 (London: University of California Press). Webster, D. (2002) ‘Unemployment: How Official Statistics Distort Analysis and Policy, and Why’, Radical Statistics Journal, 79, http://www.radstats.org.uk/no079/webster.htm Weible, C. & Cairney, P. (2018) ‘Practical Lessons from Theories’, Policy and Politics, 46, 2: 183–197. Weible, C. & Carter, D. (2015) ‘The Composition of Policy Change: Comparing Colorado’s 1977 and 2006 Smoking Bans’, Policy Sciences, 48, 2: 207–231. Weible, C. & Heikkila, T. (2017) ‘Policy Conflict Framework’, Policy Sciences, 50, 1: 23–40. Weible, C., Heikkila, T., deLeon, P. & Sabatier, P. (2012) ‘Understanding and Influencing the Policy Process’, Policy Sciences, 45, 1: 1–21. Weible, C., Heikkila, T., Ingold, K. & Fischer, M. (2016) ‘Introduction’ in C. Weible, T. Heikkila, K. Ingold & M. Fischer (eds.) Policy Debates on Hydraulic Fracturing (London: Palgrave Macmillan), 1–27. Weible, C. & Ingold, K. (2018) ‘Why Advocacy Coalitions Matter and Practical Insights About Them’, Policy & Politics, 46, 2: 325–343. Weible, C., Nohrstedt, D., Ingold, K., Heikkila, T., Fischer, M. & Cairney, P. (2018) ‘Advocacy Coalitions and Policy Change: Comparative Insights from Fracking Politics across Ten Countries’, Paper to ECPR General Conference, Universität Hamburg, 22–25 August. Weible, C., Pattison, A. & Sabatier, P. (2010) ‘Harnessing Expert-based Information for Learning and the Sustainable Management of Complex Socioecological Systems’, Environmental Science and Policy, 13, 6: 522–534. Weible, C. & Sabatier, P. (2005) ‘Comparing Policy Networks: Marine Protected Areas in California’, Policy Studies Journal, 33, 2: 181–204. Weible, C. & Sabatier, P. (2018) Theories of the Policy Process 4th edn (Chicago: Westview Press). Weible, C., Sabatier, P., Jenkins‐Smith, H., Nohrstedt, D., Henry, A. D. & DeLeon, P. (2011) ‘A Quarter Century of the Advocacy Coalition Framework: An Introduction to the Special Issue’, Policy Studies Journal, 39, 3: 349–360. Weible, C., Sabatier, P. & McQueen, K. (2009) ‘Themes and Variations: Taking Stock of the Advocacy Coalition Framework’, Policy Studies Journal, 37, 1: 121– 141. Weible, C. & Schlager, E. (2014) ‘Narrative Policy Framework: Contributions, Limitations, and Recommendations’, in M. D. Jones, E. A. Shanahan & M. K. McBeth (eds.) The Science of Stories (New York: Palgrave Macmillan). Weiss, C. (1976) ‘Using Social Science for Social Policy’, Policy Studies Journal, 4, 3: 234–238. Weiss, C. (1977a) ‘Research for Policy’s Sake: The Enlightenment Function of Social Research’, Policy Analysis, 3: 531–545.

Weiss, C. (1977b) Using Social Research in Public Policy-Making (Lexington: D. C. Heath). Weiss, C. (1979) ‘The Many Meanings of Research Utilization’, Public Administration Review, 39, 5: 426–431. Weiss, T. (2009) ‘What Happened to the Idea of World Government’, International Studies Quarterly, 53, 2: 253–271. Welch, S. & Kennedy-Pipe, C. (2004) ‘Multi-level Governance and International Relations’, in I. Bache & M. Flinders (eds.) Multi-level Governance (Oxford: Oxford University Press). Wellstead, A. (2017) ‘Plus ça change, plus c’est la même chose? A Review of Paul Sabatier’s “An Advocacy Coalition Framework of Policy Change and the Role of Policy-oriented Learning Therein”’, Policy Sciences, 50, 4: 549–561. Wellstead, W., Cairney, P. & Oliver, K. (2018) ‘Reducing Ambiguity to Close the Science-Policy Gap’, Policy Design and Practice https://doi.org/10.1080/25741292.2018.1458397 Werritty, A. (2006) ‘Sustainable Flood Management: Oxymoron or New Paradigm?’ Area, 38, 1: 16–23. Westney, E. D. (1987) Imitation and Innovation: The Transfer of Western Organizational Patterns in Meiji Japan (Cambridge MA: Harvard University Press). Weyland, K. (2005) ‘Theories of Policy Diffusion: Lessons from Latin American Pension Reform’, World Politics, 57, 2: 262–295. Wheeler, N. (2000) Saving Strangers (Oxford: Oxford University Press). White, J. (2015) ‘Emergency Europe’, Political Studies, 63, 2: 300–318. WHO (World Health Organisation) (2018) ‘WHO Framework Convention on Tobacco Control’, http://www.who.int/fctc/cop/about/en/ WHO (World Health Organization) (2008) ‘WHO Framework Convention on Tobacco Control’, http://www.who.int/fctc/en/ Wildavsky, A. (1980) The Art and Craft of Policy Analysis (London: Macmillan). Wilder, M. (2017) ‘Whither the Funnel of Causality?’ Canadian Journal of Political Science, 49, 4: 721–741. Wilks, S. & Wright, M. (1987) ‘Conclusion: Comparing Government-Industry Relations: States, Sectors, and Networks’, in S. Wilks & M. Wright (eds.) Comparative Government-Industry Relations (Oxford: Clarendon Press). Williams, J. (1985) ‘Redefining Institutional Racism’, Ethnic and Racial Studies, 8, 3: 323–348. Wilson, D. (2003) ‘Unravelling Control Freakery: Redefining Central-Local Government Relations’, British Journal of Politics and International Relations, 5, 3: 317–346. Wilson, J. (1980) American Government: Institutions and Policies (Lexington, MA: D. C. Heath). Wistow, G. (1992) ‘The Health Service Policy Community: Professionals Preeminent or Under Challenge?’, in D. Marsh & R. A. W. Rhodes (eds.) Policy

Networks in British Government (Oxford: Oxford University Press). Witting, A. (2017) ‘Insights from “Policy Learning” on How to Enhance the Use of Evidence by Policymakers’, Palgrave Communications, 3: 49 https://doi.org/10.1057/s41599-017-0052-x Wittneben, B. B. (2012) ‘The impact of the Fukushima Nuclear Accident on European Energy Policy’, Environmental Science & Policy, 15, 1: 1–3. Wolfe, M. (2012) ‘Putting on the Brakes or Pressing on the Gas? Media Attention and the Speed of Policymaking’, Policy Studies Journal, 40, 1: 109–126. Wolfinger, R. (1960) ‘Reputation and Reality in the Study of “Community Power”’, American Sociological Review, 25, 5: 636–644. Woodruff, D. (2016) ‘Governing by Panic, Politics and Society: The Politics of the Eurozone Crisis’, Politics and Society, 44, 1: 81–116. Woodward, A. (2004) ‘Building Velvet Triangles: Gender and Informal Governance’, in T. Christiansen & S. Piattoni (eds.) Informal Governance in the European Union (Cheltenham: Edward Elgar), 76–93. Workman, S., Shafran, J. & Bark, T. (2017) ‘Problem Definition and Information Provision by Federal Bureaucrats’, Cognitive Systems Research, 43: 140–152. World Bank (2008) ‘Understanding Poverty’, http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPOVERTY/0,conten tMDK:2015 3855~menuPK:373757~pagePK:148956~piPK: 216618~theSitePK:336992,00.html Wu, X., Howlett, M., Ramesh, M. & Fritzen, S. (2017) The Public Policy Primer 2nd edn (London: Routledge). Yanow, D. (1996) How Does a Policy Mean (Washington, DC: Georgetown University Press). Zafonte, M. & Sabatier, P. (2004) ‘Short-Term Versus Long-Term Coalitions in the Policy Process: Automotive Pollution Control, 1963–1989’, Policy Studies Journal, 1: 75–107. Zahariadis, N. (1999) ‘Ambiguity, Time and Multiple Streams’, in P. Sabatier (ed.) Theories of the Policy Process (Boulder, CO: Westview Press). Zahariadis, N. (2003) Ambiguity and Choice in Public Policy (Washington, DC: Georgetown University Press). Zahariadis, N. (2007) ‘The Multiple Streams Framework’, in P. Sabatier (ed.) Theories of the Policy Process (Cambridge, MA: Westview Press). Zahariadis, N. (ed.) (2016) Handbook of Public Policy Agenda Setting (Cheltenham: Edward Elgar). Zampini, G. (2018) ‘Evidence and Morality in Harm Reduction Debates: Can We Use Value-neutral Arguments to Achieve Value-driven Goals?’ Palgrave Communications, 4, 62: 1–10. Zapata-Barrero, R., Caponio, T. & Scholten, P. (2017) ‘Theorizing the “Local Turn” in A Multi-Level Governance Framework of Analysis: A Case Study in Immigrant Policies’, International Review of Administrative Sciences, 83, 2: 241–246.

Zhu, X. (2008) ‘Strategy of Chinese Policy Entrepreneurs in the Third Sector: Challenges of “Technical Infeasibility”’, Policy Sciences, 41: 315–334.

Index actors (definition) 2 advocacy coalition framework (ACF) applications outside the US 131, 181–3, 226, 244 applications and revisions 174, 183–5, 187, 209, 221, 240 belief systems and power 13–4, 52, 92, 174, 176–8, 186 case studies description of 9, 170–5, 180 flow diagram 109, 173 policy-oriented learning 178–9, 210 stability, instability, continuity and change 233 agency structure and 9–10, 14, 15, 50, 52, 54, 94–5, 97, 108, 237 agenda setting 4, 7–8, 19, 25–8, 40, 52, 57, 60, 66, 97, 148, 150, 154–5, 168, 188, 191, 195–7, 205, 208, 220, 225, 235 problem definition 156–7, 159 punctuated equilibrium theory 13–14, 162, 167 ‘second face’ of power 42–3 air pollution policy 44–6, 52, 127, 160 Allison, G. 2, 238 Althaus, C. 6, 18, 26, 33 ambiguity 60, 81, 92, 121, 155, 197, 201, 203, 226 Annesley, C. 89 Arrow problem (impossibility theorem) 117 Australia 2, 26, 33, 77, 92, 131, 142–3, 152, 216, 241 Axelrod, R. 116, 192 Bacchi, C. 18, 22, 24, 28, 88 Bache, I. 131–2, 135, 137, 139–42, 145, 204 Bachrach, P. and Baratz, M. 19, 38–9, 42–4, 52 Baumgartner, F. 25, 61, 73, 92, 105, 145, 148–67, 192, 195, 234, 243 behavioural public policy 110, 112 beliefs (to address bounded rationality) 174 belief systems 170, 175, 221 Bennett, C. 207, 209, 215–16, 218, 220 Bias (mobilization of) 42–3, 53 Blyth, M. 86–7, 190 bottom-up implementation 30–1, 106, 119, 186

bounded rationality 2, 5–8, 11, 13–14, 55–9, 61, 63–6, 73–4, 92, 100, 111, 113, 127, 130, 144–6, 149, 167, 168, 174, 178, 187, 203, 207–8, 212, 218, 224, 226– 8, 231–2, 238, 240, 247 Brexit 131, 137, 219 budget changes 164–5, 167, 169 bureaucracies or bureaucratic politics 71, 78–9, 85–6, 104, 149, 167 business 20, 24, 29–30, 40, 43–4, 46, 68, 77–8, 98–9, 106, 116, 126, 134, 136, 152, 160, 176, 185, 200, 204, 215–16, 245 Canada 34, 72, 92, 142–3, 152, 167, 182, 216–17, 219, 241 central banks 90, 215 centralization of power 7, 39, 77, 167, 225, 228 Chappell, L. 88–9, 200 China 167, 183, 204, 242–4 civil service 8, 49, 78, 89, 135–6, 146, 185 Cobb, R. 155–8 coercive policy transfer 129, 217, 218–9 cognitive bias 21, 60, 66, 69, 175, 231, 233 collective action problems 84, 87, 102, 110–29, 176, 185, 212, 236 combining theories 227–8, 237–8, 243 common pool resource (CPR) 110–11, 115, 117–21, 123–5, 128 communicative discourse 87 community power debate 13, 38, 40, 49, 53 Comparative Agendas Project (CAP) 149, 166–7 comparative analysis 181–2, 185 complex system 14–5, 28, 54, 94–5, 104, 107–9, 124, 133, 147, 149–50, 162, 167, 233 comprehensive rationality 1, 4–5, 7, 13, 34, 56–8, 61–4, 73, 113, 140, 147, 195–6, 203, 207, 235 see also bounded rationality; rationality concentration of power 1, 8, 90, 143 conditionality 218–19 confederal systems 76 consensus democracies 63, 90, 184, 243 constructivist institutionalism 75–6, 86–7 consultation 26, 30, 91, 151, 211, 223 continuity 12–14, 24, 65, 73, 87, 91, 103, 145, 147–8, 169, 171, 177, 180, 186–7, 195, 228, 230, 233, 245 contradictory policies 24, 32, 57, 129, 160, 195 convergence, policy 207–8, 215 coordinative discourse 87 core executive 89, 135–6, 146 corporatism 34, 77, 152

counterfactual 44, 46, 225 Crenson, M. 23, 42, 44–6, 51–2 crises 9, 12, 99, 102, 133, 138, 157–8, 171 critical juncture 82–3, 88, 106, 233 Dahl, R. 38, 40–1, 45–6, 172 hegenerative policy 57, 69, 71–2, 74 Demography 96 Denny, E. 22, 81, 158, 198 dependent variable 189 design principles (Ostrom) 111, 121 devil shift 176, 178 diffusion of policy 13, 65, 161, 166, 207, 213–15, 226, 240, 244–5 diffusion of power 7–8, 39, 131, 134–5, 138–9, 143, 145–6, 202, 243 direct coercive transfer 218 discursive institutionalism 79, 86–7 disproportionate information processing 149, 162, 166–7, 169, 233–4 dispute resolution 78, 144 distributive policy 21 Dodge, J. 69, 171 Dolowitz, D. 65, 207, 213, 216–20, 222 Dowding, K. 51–2, 79, 84, 103, 109, 113–14, 116, 152 Downs, A. 159 Downsian mobilizations 159–60 drug policy 71, 198 Dunlop, C. 12, 107, 207, 209–11, 213, 223–4, 226, 235 Dye–Sharkansky–Hofferbert (DSH) approach 97, 108–9, 214 ecology of games 112, 115, 120, 126 economic context 2, 36, 98, 226 economic institutions 78–9 electoral systems 72, 77, 79, 90–1, 135, 184, 220 elites 38, 40–2, 46, 63, 87, 97–9, 101, 155, 205 elitism 38, 40, 42, 205 Emejulu, A. 49–50 emergence 105–6 empirical institutionalism 76, 81, 89 emulation 73, 145, 214–15, 217, 219, 226 entrepreneurs, policy 54, 103, 109, 158, 188–9, 199–201, 204, 206, 216, 237 environment (policymaking) 1–2, 7, 9–11, 14, 16–17, 35–6, 54, 61, 64, 72, 94, 101, 103, 108–9, 119, 133, 137, 167, 170–1, 188, 199, 225–7, 232, 234, 238, 240–1, 244 environmental policy 143, 144, 160, 181

epistemic communities 191, 216–7 epistemic violence 50, 88 epistemology 38, 50, 86 European Union (EU) 14, 32, 77, 83–4, 89–92, 130–1, 134, 137–9, 142–6, 152, 167, 181–2, 202, 204, 213, 216, 219–20, 241 evaluating outcomes 119 evaluation 4, 17, 26–8, 32–3, 68, 142, 172, 180, 197, 208, 222, 225, 229, 235–6, 243 evidence based policymaking (EBPM) 3, 28, 34, 56, 58, 61, 71, 73, 103, 200, 211, 227–8, 235 evolutionary theory 101–2 external shocks 13–14, 53, 170–1, 179, 180–1, 183, 210, 221, 233–4 externalities 126–7, 219 false consciousness 45 fast and frugal heuristics 60, 231 federalism 7, 130, 142–4, 146 federal and quasi-federal systems 76, 78, 92, 131, 142–3, 146, 166 feedback 65, 70, 101, 104–5, 116, 149, 162, 177–8, 220–1, 223, 233, 235, 240 feedforward 70 Feminism 14, 49–50, 75–6, 79, 88–9, 234 Feminist institutionalism 14, 88–9 Feminist research 50, 88, 234 first order change 193 first-past-the-post systems 75, 77, 135 fitness landscape 107–8 Flinders, M. 131–2, 134–5, 137, 139–42 focusing events 61, 148, 157–8, 197, 201 formal rules 75–6, 78–81, 88, 92, 102, 231 formulation of policy 4, 7, 16–17, 21, 26–8, 57, 103, 161, 229 normative debate 53, 140 Foucault, M. 21, 47, 52 fragmentation 40–2, 53, 127, 129, 135–6, 143, 196 framework (definition) 25 Framework Convention on Tobacco Control (FCTC) 215, 244 framing 9, 60, 66–7, 70, 147–9, 156, 158, 179, 182, 188, 191, 197, 221 France 83, 91, 167, 182, 204, 241 free riding 110, 121 functionalism 85, 138 funnel of causality 97 Gains, F. 89, 137 game theory 14, 110–13, 115–16, 123, 128, 192, 238

garbage can model 58–9, 195–6, 199, 201–2 Gary, Indiana 44, 46, 52 gender 14, 39, 49–50, 76, 79, 88–9, 93, 158 general punctuation hypothesis 25, 105, 162, 164, 166, 169 geography 67, 214, 217 Germany 77, 83, 89, 92, 143, 161, 167, 182, 204, 217, 241 Gigerenzer, G. 55, 60, 231 global governance 133 globalization 11, 82, 98–100, 108–9, 133, 215 global North 15, 93, 99–100, 227–8, 241 global South 93, 227–8, 243 good governance 134, 218 governance multi-level see multi-level governance 119, 130–2, 136, 141–6, 152 as a problem 131, 136 governmental agenda 145, 155 Greece 219 group–government relations 8 Haidt, J. 59, 69 Hall, P. 52, 80–2, 87, 149, 181, 193, 195, 208 health care policy 17, 30, 72, 96, 99, 198 Heclo, H. 34, 151, 153–4, 172, 209 hegemonic project 219 hegemony 45, 70, 219 Heikkila, T. 35, 101, 111, 117–23, 125, 128, 178, 180, 208, 212, 226, 232–3, 235, 237, 240 heresthetic 65–6, 160, 191 Herweg, N. 202, 204–5 heuristic 7, 24, 59–60, 69–70, 208, 212, 223, 227, 231 Hindmoor, A. 58–9, 66, 87, 100, 112–14, 116–17, 223 historical institutionalism 81–2, 91, 106, 240 HIV policy 198 Hjern, B. 30–1 Hofferbert, R. 96–7, 214 Hogwood, B. 7, 17, 26, 28–9, 32–4, 48, 64, 96, 100–1, 134, 136, 156, 159 hollowing out of the state 136 Hood, C. 21, 28, 32, 136–7, 215 Hooghe, L. 126, 131, 133–4, 140–2, 145–6, 217 Howlett, M. 18, 28, 31, 91, 198, 208–9, 229 hyperincremental 65, 149, 164–5, 233 ideal-type 1, 4–5, 13, 16–17, 29, 48, 56, 73, 90, 135, 141, 146, 166

ideas (definitions) 190–1 inertia 56, 60, 63–4, 73–4, 77, 82, 100–1, 106, 143, 164, 199 multiple streams analysis 12, 27, 58, 65, 93, 103, 145, 188–9, 191, 195, 201, 203, 205, 220–1, 233 ideologies 11–12, 54, 163, 175, 188, 190 immigration 83, 137, 157 implementation 3–4, 7, 9, 16–17, 26–32, 36, 57, 89, 96, 106, 140, 143, 148, 153, 161, 175, 179, 184, 186, 210, 214, 219–20, 222, 226, 229, 237, 244, 246 inappropriate policy transfer 222 incommensurability 239 incomplete policy transfer 222 incrementalism 5, 23, 55–7, 61–5, 73–4, 100, 164–6, 169, 232–3 independent variable 189, 192 India 161, 183, 243–4 indirect coercive transfer 219 informal institutions 76, 89, 96, 234 informal rules 10, 75–6, 78–9, 81, 85, 88, 92–3, 102, 113, 225, 232, 242 information processing 58, 105, 149, 162–3, 166–7, 169, 233–4, 236, 244 Ingram, H. 6, 18, 48, 70–2, 74, 88, 157 inheritance before choice 100, 108 initial conditions, sensitivity to 82, 105, 106 innovation, policy 13, 102, 108, 140, 143, 145, 179, 214, 246 instability 12, 14, 84, 103, 105–6, 117, 147–50, 152–3, 169, 180, 186–7, 195, 228, 233, 246 Institutional Analysis and Development Framework (IAD) 11, 14, 80, 84, 106, 110– 11, 117–29, 177–8, 209, 237 Institutional Collective Action 84, 112, 120, 125 institutional isomorphism 86 institutional rational choice 11, 113, 119 institutionalism (new): constructivist 86–7 empirical 81, 89 feminist 14, 88–9 historical 81–2, 91, 106, 240 network 63, 81, 89 new 5, 75–6, 80–1, 88, 91, 149, 212, 233, 237 normative 85, 87 rational choice 84 sociological 85 institutions defining 75–6 formal 76, 89, 91–3, 96, 142, 144–5 informal 76, 89, 96, 234

issue attention cycle 154, 159 Ingold, K. 152, 171, 173, 176–7, 180–2, 184, 200, 210, 235 interest groups 8, 14–15, 23, 26, 30, 33, 36, 61, 78, 83, 91, 132, 139–40, 144–6, 148, 151–3, 162, 168, 172–4, 181, 190, 192–3, 199–200, 211, 214, 222, 242 interests ideas and 189–90, 192 intergovernmentalism 138–9 internal shocks 171, 180 International Monetary Fund (IMF) 218–19 international policy communities 181, 215–16 international relations 138 intra-departmental conflict 31 intransitive 66, 84, 117 iron triangles 152–3, 172, 186 issue attention cycle 159 issue networks 152–4 Italy 91 Japan 92, 98, 123, 217, 242 Jasanoff, S. 4, 59, 223 Jenkins-Smith, H. 172–7, 179–81, 183–5, 190, 210 John, P. 5, 7, 17, 21–2, 25, 34, 57, 85, 91, 96–7, 99–103, 112, 114–15, 128, 157, 167, 189–90, 200, 203, 205, 208, 232, 237–8 Jones, B. 25, 61–5, 73, 105, 145, 147–69, 192, 195 Jones, M. xviii, 6, 60, 66–9, 200, 202–4, 242 Jordan, G. 8, 21, 30, 34, 57, 64, 91, 133–4, 136, 145, 148–54, 172, 185, 213, 216– 17, 19–20 judiciary 7, 76–7, 134 Kahneman, D. 59–60, 231 Kenny, M. 79, 88–9 Kingdon, J. 65, 72, 93, 103, 152, 155, 157, 188–90, 195–202, 205, 221 Knaggård, A 200 Koski, C. 61, 157, 160, 163–4, 167, 235–6 Kuhn, T.S. 194 learning, policy 12–13, 15, 31, 127, 129, 177–8, 184, 186, 191, 207–9, 211–13, 223–4, 226, 231, 235 across coalitions 179, 184, 210 within coalitions 171, 179, 210 legitimation 4, 26–7 leptokursis 165 lesson-drawing 207, 213–14, 218, 224

negative lessons 12, 220 Lijphart, A. 75, 90, 184 Lindblom, C. 5, 56–7, 61–3, 73–4, 99, 107, 211 linearity (and non-linearity) 57–8, 73, 104–6, 124, 162, 194, 196, 199, 203 Lipsky, M. 29–30 Lowi, T. 17, 20–2 Lowndes, V. 76, 79–81, 88 Lubell, M. 126–7, 231 luck 28, 39, 51, 200 Lukes, S. 38, 44–6, 48, 51 Mackay, F. 88–9 maintenance, policy 27, 33, 101 Majone, G. 136, 143, 172, 183, 190, 197 majoritarian systems 63–4, 90–1, 243 Mamudu, H. 89, 157, 193, 230, 244 manipulation 37, 39, 46, 53, 55, 65–6, 112, 186, 203, 231 March, J. 79, 80, 85, 177 Margetts, H. 21, 100 markets 20, 99, 112, 132, 134, 136, 146 Marks, G. 126, 131–3, 134, 139–42, 145–6, 217 Marsh, D. 29, 32, 65, 97, 135–6, 150–2, 190, 207, 213, 216–20, 222, 224, 245 Marxism 98 material groups 176 mature or nascent sub-systems 184–5 McBeth, M. 68–9 McConnell, A. 2, 22–3, 32–4, 57, 155, 163, 222, 238 mean 165 measurement 1, 12–13, 32, 38, 49, 53, 80, 158, 230 memes 102–3 mental health policy 47 meta-analysis 238 metanorms 192 methodological individualism 10, 84, 113 methodological pluralism 239 methodology 38, 53, 86, 88 mixed scanning 62 mobilization of bias 42–3, 53 models 1, 4–5, 12–13, 17, 24–6, 34, 78, 90, 105, 110–13, 118, 121, 186, 207, 220, 224, 226, 240, 246 motivation 23, 41 Moyson, S. 80, 180, 208–10 multi-level governance (MLG)

comparing the EU and US 142–3 comparing MLG and other theories 142–6 federalism and 143–4 governance problem 131, 136 key descriptions 119, 130, 132 policy networks and the EU 152 types of 141 and the Westminster model 143, 146 multinational corporations (MNCs) 99, 156, 216 multiple perspectives 15–16, 238 multiple streams analysis (MSA) 12, 15, 27, 65, 93, 103, 109, 145, 188–9, 195, 200–5, 220, 226, 233, 237–8, 240, 244 how the streams come together 189, 196 policies stream 197 politics stream 198, 204 problems stream 196–7 mutual adjustment (partisan) 62–3, 211 narrative (versus frames) 67 narrative policy framework (NPF) 6, 13, 28, 55–6, 66–9, 72, 74, 157, 233, 240 narratives of policy development 6, 16–17, 22, 24, 36, 230, 233–4, 239 nascent sub-systems 184 negative feedback 104–5, 149, 162, 220 negative lessons 12, 220 negotiated agreements (ACF) 171, 184 neofunctionalism 138 nested games 115 network institutionalism 63, 76, 81, 89 networks (policy) 1, 8, 14–15, 41, 52, 65, 89, 132–3, 137, 144–5, 150–2, 172, 174, 191, 198–9, 215–16, 242, 245 new institutionalism see institutionalism 5, 10, 13–14, 75–6, 80–1, 88, 91, 149, 212, 233, 237 new public management (NPM) 8, 78, 106, 134, 215 new right ideology 83 New Zealand 90, 216–17, 220 nobel 111, 120 nodality 21 non-decision making 38–9, 42–4, 49, 53 non-linear dynamics 104–5, 162 normal distribution 165–6 norms 76, 85–6, 88–9, 116, 118, 191–2, 211 rational choice theory 10, 13, 84, 103, 110, 112–13, 128 normative institutionalism 85, 87, 177

norms 76, 85–6, 88–9, 116, 118, 191–2, 211 nuclear policy 98 obligated transfer 218 O’Flynn, J. 134, 215 oligarchy 38, 41 Oliver, K. 33, 59, 104, 152, 223, 225, 236 Olsen, J. 79–80, 85, 87, 177 Olson, M. 71, 114 ontology 86 operationalization of power 41 opportunity cost 61 organizational choice 195 organized anarchy 195–6, 201–2 Ostrom, E. 25, 76, 78, 80, 111, 114–25, 231, 242 outcomes (policy) 3, 7, 14, 17, 28–32, 36, 38, 51, 76, 95–6, 116, 130–1, 137–40, 144, 171, 201, 229–30, 233, 247 outliers 165, 244 paradigms 11–13, 52, 87, 188–91, 193–4, 205, 221, 226, 242 parallel processing 62, 154 Pareto efficiency (optimality) 117 parliamentary systems 26, 77, 202 partisan mutual adjustment 62–3, 211 party competition 202, 205 party systems 77 path dependence 81–3, 87–8, 106, 164, 242 People of Colour 71 pesticides policy 25, 153, 161–2, 168 Peters, B.G. 21, 26, 32–3, 64, 76, 79–80, 82, 84–7, 92, 101, 132, 142, 159, 177 Pierce, J. 69–70, 172, 181–2, 242 pluralism 38, 41–2, 56, 63, 78, 92, 152, 181, 205, 239–40 policies stream 197 Policy Agendas Project (and the CAP) 147, 149, 164, 166 policy areas 8–9, 15, 20, 22–3, 25, 42, 53, 63, 100, 138, 141, 144–5, 164, 166, 170, 172, 181, 202, 220, 240, 245 policy brokers 171, 173–4, 177, 180, 183, 185 policy change first, second and third order change 101, 122, 193–5 policy communities 8–9, 22, 63–4, 133, 148, 152–4, 160, 172, 186, 190–1, 193 punctuated equilibrium 5, 8, 13–14, 25, 27, 56, 65, 81, 87, 103–4, 131, 147, 149– 50, 160, 162, 166–7, 169, 220, 233, 237 policy convergence 207–8, 215

policy core beliefs 175, 178–81, 183–4, 186 policy cycle 6–7, 12–13, 16–17, 19, 25–8, 33, 36, 57–8, 172, 186, 195–6, 199, 223, 229, 232, 234–5, 240 policy formulation 21, 26–8, 161, 22 policy maintenance, succession and termination 27, 33 policy diffusion 65, 161, 207, 214 policy entrepreneurs 54, 103, 109, 158, 188–9, 199–204, , 15, , 216, 237 policy environment 2, 9, 11, 58, 95, 97–8, 103, 171, 202, 234 policy feedback theory 82, 101, 240 policy instruments 17, 22–3, 26, 36, 142, 193–4, 220, 225, 230 policy issues 7–8, 11, 16, 24–5, 28, 53, 59, 67, 73, 88, 127, 138, 144, 148, 152, 156–7, 160, 166, 169, 173, 175, 196 policy learning 12–13, 31, 127, 129, 177–8, 184, 186, 191, 207–13, 223–4, 226, 231, 235 Epistemic learning 213 Reflexive learning 211, 224 Bargaining (policy learning) 211, 213, 224, 226 Hierarchy (policy learning) 211, 223–6 policy maintenance 101 policy monopolies 9, 147–8, 150, 159–60, 162, 168–9, 172, 233, 245 policy networks 7–8, 14–15, 41, 52, 65, 89, 131–3, 137, 144–5, 150–2, 172, 174, 191, 198–9, 215–16, 242, 245 problem definition 22, 58, 145, 149, 154, 156, 159, 160, 168, 203–4, 217, 223, 235 policy shootout 239 policy styles 63, 72, 76, 81, 89, 91–2, 229–30 policy sub-systems 144, 150, 163, 171–4, 186, 228 policy succession 33, 100–1, 108, 246 policy termination 26–7, 33, 108, 246 policy tools 13, 20–2, 110, 229, 236 policy transfer: coercive 129, 217–9 convergence 207–8, 213, 215 countries 217 defining 207–8 diffusion 65, 161, 207, 214 lesson-drawing 207, 213–14, 218, 224 voluntary 218–19 policy transfer window 221 policy windows 200–2 political feasibility 103, 198 political parties 77, 83, 89, 92, 96–7, 173 party systems 77, 90, 92, 185 political stratum 41, 46

politics stream 200–1, 204–5 Polsby, N. 40–2, 45–6 polycentric governance 7, 31–2, 112, 119, 125, 141, 145 positive feedback 65, 104–5, 149, 162, 221 poverty 52, 70, 72, 158 power centralization of 7–8, 21, 30, 57, 90, 102, 131–2, 135, 140–1, 143, 146, 167, 185, 202, 225, 228, 242, 244–5 community power debate 13, 38, 40, 49, 53 definitions 39–43 and ideas 27, 52, 186, 189, 191–2 the right to exercise power 47 ‘second face’ of 38, 42–4, 46 structural 51–2, 94 third dimension of 38, 44–5, 47, 50, 52 practical lessons 234 preferences 4–5, 45–8, 57–9, 65–6, 80–1, 84–5, 92, 113, 116–17, 126, 175, 178, 191, 195–6, 203, 211, 236 prescriptive policy analysis 6 see also normative dimension 4–6, 27, 30 presidential systems 77 prisoner’s dilemma 14, 113–15, 128 privatization 29, 49, 99, 134, 136, 159, 203–4, 215, 218 problems stream 196–7 processing fluency (familiarity with information) 60 proportional systems 77, 99, 184 psychological governance 21 psychology (policymaker) 6, 56, 228, 234 gut feeling 59 fast thinking 55, 64–5, 127 emotion 6, 55, 58–60, 64, 66, 68–70, 72–4, 127, 155–6, 167, 174, 176, 178, 208–9, 231, 234–5 public attention 23, 43–4, 156, 159 public expenditure 20, 96, 100, 164 public goods 111–12, 125 public health 98, 112, 156–7, 161, 181, 193, 198 punctuated equilibrium 8, 13–14, 25, 27, 56, 65, 81, 87, 103–4, 131, 147, 149–50, 160, 162, 166–7, 220, 233, 237 agenda setting 4, 7–8, 13–5, 19, 25–8, 40, 42–3, 52, 57, 60, 66, 97, 148, 150, 154–7, 159, 162, 167–8, 188, 191, 195–7, 205, 208, 220, 225, 235 case studies 160–2 general punctuation hypothesis 25, 105, 162–4, 166, 169 generalizability 202

policy monopolies 9, 147–8, 150, 159–60, 162, 168–9, 172, 233, 245 policy process as a complex system 104–8 problem definition 156–60 quangos 7, 133, 136, 140, 146 quasi-federal systems 77, 131, 135, 142–4, 146, 166, 185 quasi-markets 20, 134, 136, 146 questions, research 59, 117, 236 race to the bottom 83, 99 Radaelli, C. 12, 69, 87, 107, 172, 184, 190, 207, 209–11, 224, 226, 235, 238 radical change 13, 34, 61–5, 73, 108, 162, 166 rational choice institutionalism 84 rational choice theory (RCT) 10, 13, 84, 103, 110, 112–13, 128 rationality deficit 133 RCTs (randomised control trials) 223–5 realism 43 receptivity to ideas 93, 154, 160, 162, 204, 226 redistributive policy 21, 116, 229 regulatory policy 21, 23 regulatory state 78, 136, 143 Rhodes, R. 29, 79–80, 106, 132–3, 135–6, 138, 140, 142, 143, 150–2, 190 Richardson, J. 8, 21–2, 24, 30, 34, 57, 63–4, 91, 100, 133–4, 139, 145, 150–2, 154, 267, 172, 185, 190–1, 202 Riker, W. 65–6, 84, 117 Rose, R. 19, 21, 64, 100–1, 134, 207, 209, 214, 216–18, 220, 222, 224 Rules (formal v informal) 10, 75–93, 102, 113, 225, 231, 232, 242 Sabatier, P. 7, 25, 28–9, 34–5, 64, 68, 170–81, 183–5, 190, 192, 210, 228, 236, 239–41 satisficing 56, 58, 73 Schattschneider, E.E. 43, 67, 149, 159, 161 Schlager, E. 67, 82, 101, 117, 119–24, 128, 176–7 Schmidt, V. 79–80, 82, 86–7, 138 Schneider, A. 6, 18, 43, 48, 60, 69–72, 74, 157 Scotland 91, 141, 225 secondary aspects 175–7, 179–80, 183–4, 186 ‘second face’ of power 38, 42–4, 46 sectors 8, 134, 172 selective attention 155, 157, 162 self-governance 46, 119 self-interest 68, 102, 113–14, 127, 156, 175–6, 191, 196 self-organization networks 124

semi-presidential systems 77 serial processing 154 Shanahan, E. 67 shared beliefs 11, 13, 52, 86, 176–7, 188, 190, 216 shocks 13–14, 53, 170–1, 179–81, 183, 210, 221, 233–4 Simon, H. 5, 56–8, 73, 154 social choice theory 117 Social Construction and Policy Design (SCPD) 6, 28, 55–6, 69, 72, 89 social construction theory 70–1, 157 social-Ecological Systems (SES) 112, 123–4 social factors 48, 96, 102, 209 sociological institutionalism 85 solutions chasing problems 198–9 stages heuristic 7 standard deviation 165 statistical analysis 165 Stone, D. (causal stories) 157 Stone, D. (learning and transfer) 200, 208, 215–16, 218–22 strange attractors 104–6 street-level bureaucrats 30 see also Lipsky structure: and agency 9–10, 14–15, 50, 52, 54, 94–5, 97, 108, 237 structural determinism 96–8, 100, 107 structural power 51–2, 94 Studlar, D. 87, 150, 158, 161, 193 subaltern 50 sub-sectors 8, 134, 145, 172 sub-systems, policy 144, 145, 150, 163, 171–4, 186, 228 see also policy communities; policy networks succession, policy 33, 100–1, 108, 246 supranational institutions 12, 138–9, 200, 216 Sweden 89, 91, 182, 217 Switzerland 7, 77, 90–1, 123, 131, 140–1, 143, 182 synthesis 220, 238 systems theory 96, 108 technology, prompting policy change 58, 96, 98, 212, 217 technical feasibility 198, 204 termination, policy 26–7, 33, 42, 108, 246 third order change 193–4, 195 timeframe of policy analysis 230 tit-for-tat strategy 116

tobacco policy 23–5, 142, 144, 143, 157, 161, 193, 215, 230, 244 top of the agenda 154–5, 199 top-down implementation 30–2, 61, 107, 135, 140, 222, 243 tragedy of the commons 14, 110–11, 114, 124, 128 transactions costs 176 transfer see policy transfer transitive preferences 117 treasure 21 trial and error learning (trial-and-error) 200, 209, 231 triggering events 157 see also framing events trust 122, 176–7 Tsebelis, G. 115, 145, 211 uncertainty 197, 211, 213 unicameral legislatures 77, 90 uninformed policy transfer 22 unintended outcomes 179 unitary systems 77, 92, 185 US Steel 44, 46 utility 5, 57–8, 113–14, 127 valence issues 157 velvet triangle 89, 152, 234 venue shift 145, 148, 154, 160, 166, 168 venue shopping 92, 103, 156, 159–61, 174, 204 viruses, ideas as 12, 189, 190–1, 205, 221 voluntary policy transfer 208, 217–19, 224 Walker, J.L. 207, 214 Waylen, G. 81, 88–9 Weible, C. 3, 11, 35, 58–9, 67, 94, 104, 171–3, 176–8, 180–5, 191–2, 195, 230, 234–5, 243–4 Weiss, C. 33, 175, 208, 210 welfare state 78–9, 82–3, 99–100, 217, 242 Wellstead, A. 3, 172, 183, 197 Western 50, 78, 91, 167, 188, 191, 198, 203, 228, 236, 241, 243, 245 Westminster model 75, 77, 130–1, 135, 138, 143, 146 wicked problems 22 Women of Colour 49–50, 88 Workman, S. 61, 92, 157, 160, 163–4, 167, 235–6 World Bank 158, 216, 218–19 world views 175, 188–90, 205

Zahariadis, N. 154–5, 197–8, 201–4