Handbook on the Political Economy of Health Systems 9781800885066, 1800885067

Citation preview

HANDBOOK ON THE POLITICAL ECONOMY OF HEALTH SYSTEMS

Handbook on the Political Economy of Health Systems Edited by

Joan Costa-Font Professor of Health Economics, London School of Economics and Political Science, UK

Alberto Batinti Assistant Professor of Economic Policy, La Sapienza University of Rome, Italy

Gilberto Turati Professor of Public Finance, Department of Economics and Finance, School of Economics, Università Cattolica del Sacro Cuore, Italy

Cheltenham, UK • Northampton, MA, USA

© Joan Costa-Font, Alberto Batinti and Gilberto Turati 2023

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Control Number: 2023935506 This book is available electronically in the Economics subject collection http://dx.doi.org/10.4337/9781800885066

ISBN 978 1 80088 505 9 (cased) ISBN 978 1 80088 506 6 (eBook)

EEP BoX

Contents

List of contributorsviii PART I 1

INTRODUCTION TO THE POLITICAL ECONOMY OF HEALTH SYSTEMS

An introduction to the Handbook on the Political Economy of Health Systems2 Alberto Batinti, Joan Costa-Font and Gilberto Turati

PART II

CONSTITUTIONAL DESIGN OF HEALTH CARE SYSTEMS

2

Health and healthcare as a human right Hiroaki Matsuura

12

3

Do democratic regimes exhibit ‘better’ health outcomes? Alberto Batinti and Joan Costa-Font

27

4

Soviet communism and later-life health and health care Joan Costa-Font and Anna Nicińska

42

5

Federalism and tax-financed healthcare: economic advantages, dilemmas, and solutions Roger D. Congleton

6

The coordination in European Union healthcare after Covid-19 Marco Buso, Massimo Bordignon, Rosella Levaggi and Gilberto Turati

7

Efficiency and equity effects of healthcare decentralization: evidence from Italy Caterina Ferrario, Rosella Levaggi and Massimiliano Piacenza

57 70

85

8 Political economy of health care insurance expansion in Mexico David G. Lugo-Palacios and Alejandro Sanders Villa

112

9

130

Weird health care for WEIRD societies? Hartmut Kliemt

PART III POLITICAL INSTITUTIONS, POLITICAL MARKETS AND HEALTH CARE DECISIONS 10

Medical composition of cabinets and the health care system Joan Costa-Font, Nicolas Marchi and Debra Winberg

147

11

Women politicians and public health Sonia Bhalotra and Mariana Lopes da Fonseca

157

v

vi  Handbook on the political economy of health systems 12

Community and civic participation effects on health and well-being Luke Munford and Daniel Gray

177

13

Ideology and health spending Bernd Theilen

200

14

Lobbying, health, and healthcare Nathaniel Z. Counts and Vinu Ilakkuvan

213

15

Healthcare corruption Martin McKee, Eleanor Hutchinson and Dina Balabanova

236

16

Provider power and healthcare systems Mason Barnard, Irini Papanicolas and Peter Smith

247

17

Health effects of trade policy and corporate interest groups Pepita Barlow

270

18

Institutional quality and health outcomes Giacomo De Luca, Domenico Lisi, Marco Martorana and Luigi Siciliani

292

PART IV THE POLITICAL ECONOMY OF HEALTH CARE FINANCING AND POLICY REFORMS 19

Towards a general political economy of private supplementary health insurance 319 Claudio Lucarelli and Mark Pauly

20

Political economy of public financing of health in low- and middle-income countries Sumit Mazumdar and Rodrigo Moreno-Serra

21

Political economy of health system reform: evidence from Spain Guillem López-Casasnovas

370

22

The pharmaceutical patent system and access to medicines Valbona Muzaka

380

PART V

339

POLITICAL ECONOMY OF HEALTH CARE DURING COVID-19

23

Covid-19 and the interest group approach to government Peter T. Leeson and Henry A. Thompson

397

24

Political preferences and nudging for healthcare: evidence from Covid certificates412 Mario Cesare Nurchis, Luca Salmasi and Gilberto Turati

25

Multilevel governance in the first wave of Covid-19 Marta Angelici, Paolo Berta, Joan Costa-Font and Gilberto Turati

430

Contents  vii PART VI LESSONS FROM THE HANDBOOK 26

The political economy of health systems: research space, goals and lessons Alberto Batinti, Joan Costa-Font and Gilberto Turati

447

Index454

Contributors

Marta Angelici is Lecturer in the Department of Social and Political Sciences at Bocconi University, Italy. Dina Balabanova is Professor of Health Systems and Policy in the Department of Global Health and Development at the London School of Hygiene and Tropical Medicine, UK. Pepita Barlow is Assistant Professor of Health Policy at the London School of Economics and Political Science, UK. Mason Barnard, MPhil, is a student in the Department of Sociology, Princeton University, USA. Alberto Batinti is Assistant Professor at the Department of Political Science of La Sapienza University of Rome, Italy. Paolo Berta is Associate Professor in Social Statistics at the Department of Statistics and Quantitative Methods, University of Milano-Bicocca, Italy. Sonia Bhalotra is Professor of Economics at the University of Warwick, UK. Massimo Bordignon is Professor of Public Economics and former Director of the Department of Economics and Finance at the Università Cattolica del Sacro Cuore, Milan, Italy. Marco Buso is Assistant Professor in the Department of Economics and Management “Marco Fanno” at the University of Padua, Italy. Roger D. Congleton is Truist Distinguished Chair of Free Market Thought and Professor of Economics at West Virginia University, USA. Joan Costa-Font is Professor of Health Economics in the Department of Health Policy at the London School of Economics and Political Science, UK. Nathaniel Z. Counts is the Senior Vice President of Behavioral Health Innovation for Mental Health America and Clinical Assistant Professor in the Department of Psychiatry and Behavioral Sciences at the Albert Einstein College of Medicine, USA. Giacomo De Luca is Professor in the Department of Economics and Management at the Free University of Bozen, Italy. Caterina Ferrario is Associate Professor in the Department of Economics and Management at the University of Ferrara, Italy. Daniel Gray is Lecturer in the Department of Economics at the University of Sheffield, UK. Eleanor Hutchinson is Associate Professor of Anthropology and Public Health in the Department of Global Health and Development at the London School of Hygiene and Tropical Medicine, UK. viii

Contributors  ix Vinu Ilakkuvan is the founder of POP Health and Lecturer at the Milken Institute School of Public Health at George Washington University, USA. Hartmut Kliemt is Guest Professor of Behavioral and Institutional Economics at the University of Giessen, Germany. Peter T. Leeson is the Duncan Black Professor of Economics and Law at George Mason University, USA. Rosella Levaggi is Professor of Public Economics at the University of Brescia, Italy. Domenico Lisi is Senior Assistant Professor of Economic Policy in the Department of Economics and Business at the University of Catania, Italy. Mariana Lopes da Fonseca is Assistant Professor of Economic Policy at the University of St. Gallen, Switzerland. Guillem López-Casasnovas is a Professor and the Founder of the Research Centre on Health and Economics at Pompeu Fabra University of Barcelona, Spain. Claudio Lucarelli is Associate Professor of Healthcare Management at the Wharton School at the University of Pennsylvania, USA. David G. Lugo-Palacios is Assistant Professor of Health Economics in the Department of Health Services Research and Policy at the London School of Hygiene and Tropical Medicine, UK. Nicolas Marchi is a medical student at Trinity College Dublin, Ireland. Marco Martorana is Professor of Privacy Law at Universitas Mercatorum, Italy. Hiroaki Matsuura is Provost and Vice President for Academic Affairs and Professor of Health Economics and Demography at Shoin University, Japan. Sumit Mazumdar is a Research Fellow with the Centre of Health Economics at the University of York, UK. Martin McKee is Professor of European Public Health in the Department of Health Services Research and Policy at the London School of Hygiene and Tropical Medicine, UK. Rodrigo Moreno-Serra is Associate Professor and Co-Team Leader of the Global Health Team at the University of York, UK. Luke Munford is Senior Lecturer in Health Economics in the Department of Health Sciences at the University of Manchester, UK. Valbona Muzaka is Senior Lecturer in International Political Economy at King’s College London, UK. Anna Nicińska is Assistant Professor with the Faculty of Economic Sciences & Digital Economy Lab at the University of Warsaw, Poland. Mario Cesare Nurchis is a Researcher in the Department of Woman and Child Health and Public Health at the Università Cattolica del Sacro Cuore, Rome, Italy.

x  Handbook on the political economy of health systems Irini Papanicolas is Professor of Health Services, Policy and Practice, at Brown University School of Public Health, USA. Mark Pauly is Professor Emeritus of Healthcare Management at the Wharton School at the University of Pennsylvania, USA. Massimiliano Piacenza is Professor of Economics in the Department of Law and Political, Economic and Social Sciences at the University of Piemonte Orientale, Italy. Luca Salmasi is Associate Professor of Economic Policy at the Department of Economics and Finance, Università Cattolica del Sacro Cuore, Rome, Italy. Alejandro Sanders Villa is a student at the London School of Economics and the London School of Hygiene and Tropical Medicine, UK. Luigi Siciliani is Professor in the Department of Economics at the University of York, UK. Peter Smith is Emeritus Professor of Health Policy at the Imperial College Business School, UK. Bernd Theilen is Professor in the Department of Economics at Rovira i Virgili University, Spain. Henry A. Thompson is Assistant Professor of Economics at the University of Mississippi, USA. Gilberto Turati is Professor of Public Finance at the Department of Economics and Finance at Università Cattolica del Sacro Cuore, Rome, Italy. Debra Winberg is a PhD candidate at Tulane University, USA.

PART I INTRODUCTION TO THE POLITICAL ECONOMY OF HEALTH SYSTEMS

1. An introduction to the Handbook on the Political Economy of Health Systems Alberto Batinti, Joan Costa-Font and Gilberto Turati

1.

A FRAMEWORK OF THE POLITICAL ECONOMY OF HEALTH AND HEALTH CARE

Although many would agree with the statement that “effective health care is at the heart of staying healthy and caring for those sick”, it is far more difficult for most of us to understand how important rules constraining and incentivising health care activity are for efficient health care, and more specifically to keep the costs of medicines under control, and to motivate physicians and nurses in the pursuit of their daily duties. Country-specific rules influencing the constraints and incentives of health care activity are often defined as “health care institutions”.1 Such rules can take the form of policies when they refer to specific interventions, or reforms when such incentives and constraints change. More generally, the system of rules influencing how health care is provided and financed are widely heterogenous across countries, and define what we know as “health systems”. Health systems differ by several features, including the way physicians and medicines are regulated and paid, whether they have a single payer or multiple payers for health care, as well as the barriers current and future patients face to access health care. This book will focus on studying how institutions affect the activity, outcomes, and financing of health systems as they are widely understood. Health systems are in continuous reform to respond to new evidence and challenges (see Chapter 21 to understand barriers to reform in public systems), and such reforms are interdependent (countries learn from each other, and react to each other’s reforms). Similarly, health systems are shaped by changes in the social values and ideologies (e.g., the shared belief in the primacy of market choice over government provision can explain the heavy reliance of health care on markets in the United States and other health systems), which add to other economic and financial constraints a country sets itself (e.g., austerity cuts weakened the preparedness of nursing homes during the Covid-19 pandemic in Europe, or lower reliance on primary care gate-keeping limits the effectiveness of public health interventions). A recent example of the role of regulation in health care at the time of writing this book is the US Supreme Court decision to overturn the 1973 Roe vs. Wade ruling on federally sanctioning the right to abortion, and the significant consequences this decision will have on maternal and child health. Another paradigmatic example can be found in the series of decisions (such as lockdowns, curfews, mask wearing, etc.) made by governments during the various waves of the COVID-19 pandemic, as well as the vaccination campaigns that followed (see Chapter 24). At the core of all these examples it is possible to identify what can be defined as “public choices”, namely collective decisions made by citizens and their representatives that gave rise to a series of associated political processes, which can prove critical for population health.

2

An introduction to the Handbook  3

2. THIS HANDBOOK’S LIFE COURSE AND MISSION This Handbook is written to add to the growing field of the political economy of health and health care – an area of an increasing interest, especially after the pandemic. However, the idea of editing and writing some chapters of this Handbook pre-dated the pandemic. It resulted from informal discussions about the importance of political processes in health care decision-making after presenting papers at events in the same sessions, especially at the European Public Choice Society annual meetings. Informal discussions during coffee breaks became ideas written down in papers, some of which have been turned into book and journal articles, but it was not until 2017–18 that we seriously conceived the idea of pushing for the development of a new research area in the health economics and policy field, namely the “political economy of health and health care”. We began by laying some basic ideas in the book The Political Economy of Health and Healthcare: The Rise of the Patient Citizen (2020, CUP). The book’s main idea is that each of us is a patient-citizen, interested as patients in accessing high-quality health care services, and as citizens, we attempt to influence how health care is provided and financed, and we intend to do so by engaging in political processes and designing health care regulations, demanding the recognition of the right to health care, and generally shaping the rules and the constraints that impact those health care services the patient citizen has access to. This idea is not necessarily a description of how health systems operate, but a framework for how they can be examined. We hope that this Handbook becomes a significant step towards the development of the field of the political economy of health and health care. We have collected contributions from very diverse scholars in economics, policy, politics, medicine, anthropology, law, and sociology, offering a heterogeneous set of views about what we mean by the “political economy of health and health care”. The message, however, is common and is that politics matters in health, even though the aspects that we focus on and the methods used differ by disciplinary area. This is far more than stating that health policymaking affects the health of citizens, which is obvious, but that health policymakers act according to incentives and constraints that can be designed and re-designed to influence citizens’ health care actions, and ultimately their health. Patient-citizens are endogenously affected by health care institutions, and collectively they can contribute to influence the type of health care regulation and financing at play (e.g., by electing a government that expands universal insurance coverage). As market and non-market players respond to different incentives and constraints, the analysis of the effect of institutions on health and health care becomes far from trivial, and understanding the processes at hand requires pooling different perspectives, to favour the cross-pollination between fields of analysis of health care. The mission of this Handbook is to bring together key theoretical and empirical academic and policy debates of what is progressively becoming known as the “political economy of health systems”. The main theoretical debates in the “political economy of health systems” originate from various scholarly groups, which share the common goal of understanding the institutional mechanisms, constraints, and determinants that influence decision-making in the health and health care domains. This includes scholars in public health and medical schools, new groups of health economists working on political economy, political scientists increasingly paying attention to political economy aspects of health care, public finance scholars interested in incorporating the role of institutions in their analysis, and historical institutionalists interested in policy reform in the health sector, in addition to public choice scholars

4  Handbook on the political economy of health systems and different groups of policy analysts, political scientists, and sociologists specialising in international and comparative political economy, and political psychologists, public policy, and management scholars interested in the influence of incentives and constraints on health care activity. Certainly, the fragmentation of the debate is the result of the disciplinary backgrounds of its researchers, but it does not have to result in fragmented scientific inquiry. Rather, different disciplines may specialise in the application of various methods and concentrate on answering specific questions where their disciplinary approaches provide an advantage. The obvious advantage of interdisciplinary approaches is the proliferation of various and competing theories often lacking with single disciplinary analysis. Indeed, the heterogeneity of worldviews can only enrich the search for empirical regularities and theories of how institutions influence health systems. This book is a modest step in this direction, and more specifically towards the creation of a space for discussion of the various approaches. This Handbook is a collection of contributions from all streams of the political economy debate, and provides a unique state of the art not found in other textbooks. It is intended to be an interdisciplinary, accessible, and non-technical contribution for a broad audience, including practitioners as well as academics. Scholars from these fields have been invited (though not everyone invited had the time to contribute). Inevitably, surely, we have failed to invite many relevant contributors, an omission that is not intentional and is purely driven by time constraints and ignorance on our part, and for which we hope we can be forgiven. The main mission of this Handbook is to highlight the importance of the political economy of health care as an area of research inquiry, and to discuss what questions we have the most agreement on, as well as pointing out areas where research in political economy still has to delve into. We can draw some main emerging themes from the highly interdisciplinary body of perspectives, which are classified into four broad sections of the Handbook, described in the following section.

3.

BOOK STRUCTURE AND CHAPTERS OVERVIEW

The book is divided into four sections (excluding the introduction and the conclusions): (i) the constitutional design of health care systems; (ii) political institutions, political markets, and health care decisions; (iii) the political economy of health care financing and policy reforms; and (iv) the political economy of health care during Covid-19. We will now briefly go over some of the highlights of the chapters in each of these four parts, providing an overview of the issues covered in the Handbook. The Constitutional Design of Health Care Systems The first part of the Handbook consists of eight chapters (Chapters 2–9) that broadly speaking spell out the main discussion issues in the constitutional political economy of health. Constitutions establish the “rules of the game”: they are first level decisions that influence how patient-citizens express their voices through voting, as well as how citizens’ voices translate into actual legislation. Finally, constitutions establish individuals’ rights by defining what the patient citizen can expect from the State, who should provide these services, and who should pay for them. Constitutional principles, as well as other laws, and regulations also define

An introduction to the Handbook  5 whether a health system should be private, public, or mixed, alongside social norms in each society. The book begins with a chapter by Hiroaki Matsuura discussing the recognition of health and health care as a constitutional right in Chapter 2, “Health and healthcare as a human right”, from both a formal and substantive standpoint. It includes methodological reviews as well as empirical findings on the economics and social consequences of making health a constitutional right. First, the author investigates the historical evolution of the human right to health, as well as the geographical spread of a constitutional right to health throughout various countries from the late 18th century to the present. Then he moves to an analysis of the difference between considering the right to health from an ethical or a legal perspective and the ensuing intellectual debate, including the normative discussion of the respective roles of a country’s courts and government towards the health and health care of the citizens. The following section discusses the economic and redistribution effects of including the constitutional right to health and health care in the framework of constitutional economics. The final section provides an up-to-date review of the literature on the economics of the right to health, and explores both methodological and empirical issues related to the right to health. Chapters 3 and 4 deal with the effect of autocratic (including communist) and democratic regimes in terms of outcomes and health care services. In Chapter 3, “Do democratic regimes exhibit ‘better’ health outcomes?”, Alberto Batinti and Joan Costa-Font examine the various mechanisms driving democratic and non-democratic regimes, influencing health outcomes through institutional change and policies, as well as the specific environmental and cultural effects that democracies and non-democracies bring about. The authors contend that while most of the evidence suggests a positive relationship between democracy and health, studies attempting to retrieve causal effects are more limited and exhibit some differences depending on the specific form of variation examined and, especially, the measure of health outcome considered, such as mortality and heights. More importantly, democracies seem to influence different forms of health inequalities, and the effects differ by gender and socioeconomic status; in some cases, they can give rise to health inequality traps when a majority impose their health policies on the rest of the population. Furthermore, democracies can exert a differential influence on the old-age population, and on the mental and long-term health of a society by affecting the social environment and the culture individuals live in. Finally, the authors provide an assessment of the effect of democracies in confronting communicable diseases such as Covid-19, and more specifically document that although the evidence suggests that autocracies are stricter in their mobility restrictions, they do not necessarily show better outcomes. Chapter 4, “Soviet communism and later-life health and health care” is a contribution from Joan Costa-Font and Anna Nicińska. This chapter explores the health and health care effects of exposure to Soviet communism and provides an up-to-date overview of the literature on the topic. Such exposure was found to hurt health in general, and longer exposure harmed trust and confidence in public institutions. This also had a detrimental impact on the early 1990s regime transitions, which had a negative impact on health outcomes. The chapter then examines the formal and informal components of the Soviet system’s health institutions, as well as the failures and successes of the communist regime’s public health and health care policies. Turning to positive effects, it then explores the role of health education provided in former communist countries in influencing the health gradient and discusses both methodological and empirical issues in measuring the causal effect of communist education on health.

6  Handbook on the political economy of health systems Chapters 5–8 address the issue of government vertical architecture and the complex set of problems that arise in these contexts. Roger D. Congleton contributes a chapter on federalism and constitutional design, Chapter 5, “Federalism and tax-financed healthcare: economic advantages, dilemmas, and solutions.” The author explores the issue of health care financing by comparing unitary and decentralised systems of governance, and stresses the matching and informational competitive advantages of the latter, especially in the presence of informational asymmetries, uncertainty, and health care goods and services heterogeneity. Moreover, an additional advantage of decentralisation is that rent-seeking might be reduced. The chapter concludes by stating that decentralised systems, while not perfect, produce health care more efficiently and with less rent-seeking, and by incentivising useful innovation in the delivery and management of health care resources. Chapter 6, “The coordination in European Union healthcare after Covid-19”, is a joint contribution by Marco Buso, Massimo Bordignon, Rosella Levaggi, and Gilberto Turati. This chapter draws on evidence from data envelopment analysis to produce measures of efficiency which may support the European Union’s (EU’s) larger role in health care policies. In general, and considering the overall spending, which is heavily reliant on curative medicine (a publicly provided private good), the authors do not find support for the EU playing a larger coordinating role in health care policies. However, the answer changes when specific sub-functions of European health systems, such as procurement and prevention, are considered. A larger role for the EU in these specific policy areas may be justified, forecasting financial savings from effectively managing externalities and financial savings from a more centralised procurement system. The findings are also extended to the EU’s role in procuring vaccines for European citizens, who overwhelmingly supported the initiative. With Chapter 7, “Efficiency and equity effects of healthcare decentralisation: evidence from Italy”, Caterina Ferrario, Rosella Levaggi, and Massimiliano Piacenza contribute a perspective account of the first- and second-generation fiscal federalism models, focusing on the main trade-offs between welfare and asymmetric information. They then move from general theories of decentralisation to theories of decentralisation applied to the health care sector, and review the literature on the role of health policies at the central and local levels, with special emphasis on the Italian experience, but not exclusively. Finally, they draw some relevant conclusions and identify the next steps, especially considering the problems that emerged during the Covid-19 pandemic in Italy, and how to restructure the health institutions in Italy in the aftermath of the pandemic. In Chapter 8, “Political economy of health care insurance expansion in Mexico”, David G. Lugo-Palacios and Alejandro Sanders Villa examine the evolution of Mexico’s public health care system, categorising it into three major stages: (i) the decentralisation process, (ii) the expansion of health care insurance coverage within the context of a decentralised system, and (iii) the recentralisation of the health system. The authors discuss the drivers of such major health care reforms and identify the key players involved in the design of Mexico’s health care system. Part I’s concluding contribution addresses two intriguing issues at the boundaries of State intervention as defined in a country’s constitution. Indeed, Hartmut Kliemt’s Chapter 9, “Weird health care for WEIRD societies?”, discusses how medical technological progress may increase political demand for public intervention, with the private sphere potentially crowded out by the public. This is especially true when constitutions define the right to health rather than the right to health care. If the State must intervene whenever a lifesaving intervention is available, as the lifesaving option is made possible and expanded by technological innovation, the sphere of public intervention automatically expands. This increasingly protective State

An introduction to the Handbook  7 may alter a country’s material constitution. The author uses kidney allocation for transplants as an example. Political Institutions, Political Markets, and Health Care Decisions The second bloc of the Handbook includes nine chapters (Chapters 10–18) and explores the role of political institutions (like a parliament or cabinet) in health care. More specifically, the composition of political institutions in charge of defining and enforcing laws, and how it influences policymaking and the allocation of resources. Another set of chapters focuses on the role of external factors that can influence the behaviour of political institutions; in this case, we use the term “political markets” to refer to both competitive and non-competitive forces (such as elections or the process of lobbying). Chapters 10 and 11 focus on two stimulating features of composition: the first is the role of ideology, and the second the role of politicians’ gender. Chapter 10, “Medical composition of cabinets and the health care system”, by Joan Costa-Font, Nicolas Marchi, and Debra Winberg, provides a broad discussion of the role of cabinets in health care. Drawing on some previous work, and a leadership dataset, they discuss the influence of different cabinet compositions in health care decisions and suggest that the composition of cabinets is important, consistent with the idea that the backgrounds of decision makers influence their views and priorities, and hence affect the decision-making outcomes. In Chapter 11, “Women politicians and public health”, Sonia Bhalotra and Mariana Lopes da Fonseca explore the literature on how women’s political participation affects policymaking in terms of redistribution, public health intervention, schooling, and policy outcomes aimed at improving the health of mothers and children. They also discuss challenges for empirical research in the field, such as identification issues and designs based on quota introduction, close elections in mixed-gender contexts, and textual analysis. Finally, they examine the supply- and demand-side (discrimination) factors that prevent women from entering politics. In contexts where politicians’ selection matters, the chapter concludes by emphasising the importance of gender-specific policy preferences (more redistribution and more spending on children and maternal health). Luke Munford and Daniel Gray explore the role of social capital as a determinant of health in Chapter 12, “Community and civic participation effects on health and well-being”. First, the chapter provides a current review of the literature on social capital and health, as well as the definition and relevant dimensions of social capital discussed in the literature. It then summarises the empirical evidence and mechanisms that mediate the channel from social capital to health. It also looks at two case studies from the United Kingdom, one on the role of community engagement in improving health and the other on the impact of political engagement and voting behaviour on mental health. Finally, the authors identify a number of interventions that could be implemented to improve health and well-being. Chapters 13–17 explore “political markets”, including competitive market structures and non-competitive ones, tackling issues like political electoral cycles, lobbying, corruption, and the role of health care providers as a special interest group. More specifically, Chapter 13, “Ideology and health spending”, by Bernd Theilen, examines the political demand for health care, reviewing the role of ideology on the allocation of health care resources, and specifically the influence of political electoral cycles in driving health spending. He documents that ideology (or partisanship) matters and there is no convergence towards a single platform which is characteristic of the median voter model. This makes for policy platforms that can

8  Handbook on the political economy of health systems be distinguished between left- and right-wing governments and heterogeneous trade-offs between spending and inflation. Additionally, the chapter discusses methodological issues in analysing the (causal) effect of changes in government ideology on health care spending, and then examines the empirical literature based on international comparisons, and on decentralised economies single-country studies. Chapter 14, “Lobbying, health, and healthcare”, by Nathaniel Z. Counts and Vinu Ilakkuvan, examines the effects of lobbying as influencing the political demand for health care. The chapter proceeds with a discussion of the main features of lobbying models, and their application in the realm of health care policymaking, providing a summary of findings and pointing the reader to data for future research and data sources that can be used to do research in the field of health care policy lobbying. Finally, it goes over the role that lobbying might have in promoting health equity and other important health-related policy objectives. Chapter 15, “Healthcare corruption”, by Martin McKee, Eleanor Hutchinson, and Dina Balabanova, digs into corruption. The authors contextualise corrupt behaviour in the relevant asymmetric information framework, in the presence of misaligned incentives, and possible disparities in social status between health care workers and patients. All these factors justify survey-evidence showing the health sector as one of the most corrupt in cross-country and sectoral comparisons. The chapter discusses the central challenges in defining corruption, capturing corrupt behaviour, and defining anti-corruption policies in the health care sector. The authors observe that, despite these challenges, anti-corruption policies made relevant progress in recent years, and they are becoming a priority in the policy agenda of international organisations. Chapter 16, “Provider power and healthcare systems”, by Mason Barnard, Irini Papanicolas, and Peter Smith, provides a very interdisciplinary perspective on the analysis of medical providers’ behaviour and their political demand for health policies. The discussion proceeds by looking at the incentives and constraints faced by health care personnel in influencing health care policies throughout various channels. Finally, the chapter examines how constraints, levels, and incentives on the part of providers have operated in the context of the Affordable Care Act and Covid-19. Chapter 17, “Health effects of trade policy and corporate interest groups”, by Pepita Barlow, examines the political economy of trade policies, and the role that they have played in influencing the health of the population. The author discusses the cases when there has been a positive or a negative influence and explores the relationship between globalisation and global health systems. The structure of the chapter starts with discussing contemporary global trade features, moves into an exploration of the evidence showing how institutions, trade agreements, and policies impact health systems, and concludes with the political economy of health systems involved in trade policies. Finally, Chapter 18, “Institutional quality and health outcomes”, by Giacomo De Luca, Domenico Lisi, Marco Martorana, and Luigi Siciliani, takes a broader perspective and discusses the relationship between the “quality” of political institutions and health outcomes. The authors first provide a theoretical framework, followed by empirical evidence relative to Italy. They show how higher institutional quality in some regional administrations does improve hospitals’ health outcomes in terms of reductions in mortality rates related to heart attacks, hip fractures, and stroke.

An introduction to the Handbook  9 The Political Economy of Health Care Financing and Policy Reforms A core part of the Handbook refers to the four chapters (Chapters 19–22) exploring central issues related to health care financing and health policy reforms. Chapters 19 and 20 discuss the political economy of public and private health insurance. Chapter 19, “Towards a general political economy of private supplementary health insurance”, by Claudio Lucarelli and Mark Pauly, takes a theoretical approach and explores the political economy of private health insurance supplementation in a majoritarian democracy. The authors provide a theoretical analysis and several case studies to offer factual evidence of what are the most likely political equilibria. More specifically, they show how the presence of private insurance supplementation for higher-income people does make lower-income people better off. Chapter 20, “Political economy of public financing of health in low- and middle-income countries”, by Sumit Mazumdar and Rodrigo Moreno-Serra, introduces the fundamental of the analysis of economic policies, theories, and approaches to the political economy of health at different stages of policymaking. They then look at the role of actors, mechanisms, and strategies influencing the policy process towards universal health care, especially regarding low- and middle-income countries. Finally, they offer an illustration of the political constraints driving the financing of the Indian health care system. In Chapter 21, “Political economy of health system reform: evidence from Spain”, Guillem López-Casasnovas uses the example of the Spanish health care system to provide an analysis of the institutional determinants of health care reforms in universal and tax-funded systems. He discusses what we have learned from the political economy of reform in national health service systems. The chapter identifies a series of constraints to health care reform, drawing on the Spanish experience after setting up a national health system. These include the effects of provider competition, and especially devolving most of the health policymaking to regional governments. In the concluding chapter of Part III, Chapter 22, “The pharmaceutical patent system and access to medicines”, Valbona Muzaka tackles the important issue of reforming the current patent systems for drugs. The author offers a historical perspective of the development of the international regulatory framework for patented drugs and sees as a crucial watershed the Trade-related Aspects of Intellectual Property Rights (TRIPS) Agreement in 1995. The chapter considers the high international costs for patented drugs deriving from the TRIPS, and the resulting problems in accessing drugs, especially in low- and middle-income countries, providing an analysis of the pros and cons of patents. The author also discusses the most recent attempts to modify the current policies in terms of alternative means for sustaining medical R&D, and other initiatives brought forwards within the World Health Organization regarding the accessibility of patented drugs on the part of low- and middle-income countries. Political Economy of Health Care During Covid-19 The last core part of the Handbook includes three chapters (Chapters 23–25) and deals with the political economy of health care since the onset of the Covid-19 pandemic. The pandemic has originated a great deal of research at the intersection of politics and health care. Here we take up three main issues: the role of special interest groups in affecting policy choices, the impact of nudging in the midst of contradictory media messages and polarised political positions on vaccinations, and the impact of governance centralisation/decentralisation on health

10  Handbook on the political economy of health systems outcomes. Chapter 23, “Covid-19 and the interest group approach to government”, by Peter T. Leeson and Henry A. Thompson, translates interest group theory into political economy issues raised by Covid-19. The chapter explores, from a public choice perspective, how private interests influenced lockdown measures and the allocation of Covid-19 resources on the part of governments, and discusses the potentially perverse consequences of these influences when policymaking in public health is undermining rather than promoting the health of vulnerable people. The authors also supply ideas on how research in the non-Covid area might inspire future investigations on the public choice of Covid-19. Chapter 24, “Political preferences and nudging for healthcare: evidence from Covid certificates”, by Mario Cesare Nurchis, Luca Salmasi, and Gilberto Turati, discusses the role of nudging in health policymaking when political preferences are heterogenous and correlated with individual choices. Covid-19 vaccination policy in Italy and Spain is taken as a testing ground, where no-vax groups resisted governments’ decisions on vaccination, and vaccine hesitancy was present as a result. The authors find that nudge interventions put forward by the Italian and the Spanish governments affected the uptake of vaccinations in the two countries. However, more importantly, citizens’ responses to vaccination policy measures varied according to both political preferences and the type of nudging measures put in place. Chapter 25, “Multilevel governance in the first wave of Covid-19”, by Marta Angelici, Paolo Berta, Joan Costa-Font, and Gilberto Turati, explores and compares the role of central and subnational policymaking in addressing the Covid-19 pandemic, considering as case studies Italy and Spain. The chapter exploits the different approaches of the two countries to face the first wave of the pandemic, and it offers comparative evidence of the relative advantages of a decentralised policy approach to respond to pandemics, which fosters regional cooperation as well as regional autonomy. The authors document that keeping governance decentralized produces better results than exclusively centralised responses to a global threat when the effects of the policy choices are highly uncertain. Finally, in this Handbook’s Chapter 26, “The political economy of health systems: research space, goals and lessons”, you will find not only a summary of the main insights of the Handbook but also some ideas on the more pressing topics in the political economy of health systems, to be addressed in the years to come. We make some suggestions to guide future research, and humbly point out some of the areas where there is a higher need, in our view, for further research.

NOTE 1. North Douglas, C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

PART II CONSTITUTIONAL DESIGN OF HEALTH CARE SYSTEMS

2. Health and healthcare as a human right Hiroaki Matsuura

1. INTRODUCTION The right to health is recognized as a fundamental human right in numerous international, regional, and national instruments (Backman et al., 2008). Today, most countries have ratified at least one international treaty that imposes obligations on governments regarding the right to health.1 In about a third of the countries, a right to health is explicitly enshrined in a national constitution or other forms of superior law (Backman et al., 2008). However, the right to health is broader than just the right of access to healthcare. The right to health entails not only the provision of timely and appropriate healthcare, but also the fair distribution of other determinants of health. According to General Comment 14 of the United Nations’ Committee on Economic, Social and Cultural Rights, the right to health embraces a wide range of non-medical determinants of health, such as food and nutrition, housing, access to clean water and sanitation, safe and healthy working conditions, and a healthy environment (UN, 2000). The right to health includes specific components which are legally enforceable through the judicial system, but similarly, it also imposes some ethical obligations on individuals and society to ensure such a right is protected. At the turn of the twentieth century, like other human rights, a right to health was an ethical right without legal protection. However, during the twentieth century, a clear trend of the codification of such rights was established. Today, the right to health is legally binding on governments in countries where a right to health is recognized in ratified international treaties or national constitutions. In Latin America, South Africa, and other parts of the world, courts have taken a more active role in redressing the failures of governments’ health policies by using this right since the 1990s. However, the trend of codification and judicialization of such a right does not undermine the value of a right to health as an ethical right. In fact, scholars such as Daniels have argued that a legal right to health without ethical deliberation is problematic because it does not solve disagreements and uncertainties about the scope and limits of such rights (Daniels, 2009). Other scholars, such as Sen, have argued that an ethical right would supplement the implementation and compliance with a legal right to health by requiring individuals, governments, and non-state players to internalize public ethical norms (Ruger, 2006; Sen, 2009). This chapter provides an overview of the role of a right to health that is enshrined in national constitutions or other forms of supreme laws in the resource allocation of health and healthcare. A claim that a certain right is important enough to be seen as a constitutional right is also a claim that such a right is likely to be ethically acceptable for a member of society after reasoned scrutiny. This ethical demand also affects resource allocation of health and healthcare beyond the judicial and legislative route. The rest of the chapter is organized as follows. Section 2 discusses the historical development of the human right to health. Section 3 discusses the nature of a legal right to health and its relation to an ethical right to health. Section 4 investigates why the literature on the right to health has not emerged in the field of health economics and other related fields until 12

Health and healthcare as a human right  13 quite recently. Section 5 summarizes three ways a right to health in national constitutions is integrated into economic analysis, and Section 6 concludes the chapter.

2.

HISTORICAL DEVELOPMENT OF THE HUMAN RIGHT TO HEALTH

The history of the right to health is a complex entanglement of political aspirations and ethical demands, but there is a clear trend of codification and, to some extent, the judicialization of such rights throughout the twentieth century. Before the twentieth century, like other human rights, the right to health or healthcare was merely an ethical norm or, at most, was considered to be one of the objectives of social progress (Ishay, 2004). The right to health or to healthcare first appeared in the political arena in Europe in the late eighteenth century. During the French Revolution, revolutionaries regarded the right to health as a right of citizens based on the idea of the social contract (La Berge, 2002). However, these early reformers faced political, religious, and professional opposition once the revolutionaries seized power (Weiner, 1993). In the end, the term “health” was eventually excluded from the Declaration of the Rights of Man and of the Citizen in 1789 and from the French Constitutions of 1791, 1793, and 1795. On the other side of the Atlantic, the term “right to health” first appeared in 1796 during congressional debates over quarantine laws in U.S. politics. However, the term did not appear in any constitutional documents until the Wyoming Constitution of 1890 recognized the duty of the legislature to protect and promote health and morality in order to promote the well-being of the people (Visscher, 1972; Leonard, 2009). In 1936, the Soviet Union became the first country to provide a guarantee of access to healthcare in its constitution at the national level, although Article 120 of the constitution focused exclusively on access to healthcare rather than on the right to health in general. In the post-war period, the concept of a right to health spread widely from the constitutions of the Soviet Union to the constitutions of democratic countries, but was termed “a right to health” rather than “a right to healthcare”. The Constitution of the Italian Republic enacted in 1947 saw the first democratic nation recognize the right to health at a national level, followed by other European countries such as Portugal, Spain, Belgium, and Poland. In 1948, the Universal Declaration of Human Rights (UDHR) was adopted by the 56 members of the United Nations. Article 25 of the UDHR provides for an “adequate” level of medical care and other necessary social services for health and well-being. Furthermore, the International Covenant on Economic, Social and Cultural Rights (ICESCR) was adopted as a legally binding treaty in 1966. Ambitiously, the ICESCR moves the goal from an “adequate” level of medical care to the “highest attainable standard”. Finally, the Declaration of Alma Ata (1978) and General Comment 14 of the United Nations Committee on Economic, Social and Cultural Rights (2000) further address this issue and provide a more detailed list of what the right to health should include. With greater recognition of this right by the international community, a constitutional right to health spread from Europe to several Latin American and African countries in the last two decades of the twentieth century. This historical trend determines the pattern of which country has a constitutional right to health. In the analysis based on five different legal origin classifications, following La Porta and others, Matsuura found that, except for South Africa, only countries with French commercial law or Soviet socialist law origins gave citizens a constitutional right to health or to healthcare. English-, German-, and

14  Handbook on the political economy of health systems Scandinavian-legal-origin countries were highly unlikely to have recognized a constitutional right to health (La Porta et al., 2008; Matsuura, 2013). In Latin America and South Africa, the courts took more active roles in redressing the failures of government health policy by using this right from 1990 on (Hogerzeil et al., 2006). Successful cases were primarily connected to constitutional provisions of this right, which are supported by international treaties (Hogerzeil et al., 2006). While some see this right as a powerful tool in the struggle against poverty and health inequality, others criticize that this rights-based litigation has distorted the allocation of health resources and should be used as a last resort (Zuniga et al., 2013; Daniels et al., 2015). This will be reviewed in the next section.

3. LEGAL vs ETHICAL RIGHT TO HEALTH At the turn of the twentieth century, like other human rights, a right to health was an ethical right without legal protection. Jeremy Bentham famously criticized human rights as a natural right described in France’s Declaration of the Rights of Man and of the Citizen since rights are only created by a society’s law (Bentham, 1838). In this sense, he insisted that this right must be the “child of law”. Amartya Sen rejected this claim and argued for treating human rights as an ethical demand for justice and equity in the distribution of material and non-material goods (Sen, 2009). He defended the position of Herbert Hart, who argued that this ethical demand served not as a “child of law” but more as a “parent of law” in guiding legislation. He also emphasized that this ethical right will supplement the implementation and compliance with a legal right to health beyond the legislative route. Even where a legal right to health is successfully established, this ethical demand also requires individuals, governments, and non-state players to internalize public ethical norms to enhance implementation and compliance with the right to health (Ruger, 2006; Sen, 2009). On the other hand, Daniels believed that a right to health or healthcare is justified only if it can be reasoned by an acceptable, general theory of distributive justice for health because a legal right to health alone does not solve disagreements about the scope and limits of such rights. For Daniels, a right to health is understood as a right to maintain an individual’s health within a normal functioning range that does not impair opportunities in one’s daily life. In his early work, Daniels expanded the Rawlsian concept of fair equality of opportunity to obligate the decent minimum level of healthcare based on the constraints that diseases and injuries place on individual opportunity (Daniels, 1985). In his later work, he expanded his theory to include broader determinants of health that are necessary to maintain an individual’s normal functioning beyond healthcare (Daniels, 2007). He acknowledged that this general theory is not always helpful, especially when resources are constrained to maintain everyone’s normal functioning. When we need to sacrifice someone’s normal functioning, government, as society’s designated agent, has to decide on the resource allocation of health and healthcare through a fair and accountable decision-making process (Daniels, 2007). In his view, our rights are not violated if this resource allocation is made by a fair, publicly accountable decision-making process that resolves ethical conflict regarding the allocation of healthcare resources. In this sense, courts are not an appropriate place to decide resource allocation because they cannot act as a society’s designated agent (Daniels, 2009). Moreover, the resource allocation of healthcare resources requires considerable expertise relevant to deciding how to allocate resources effectively and fairly. However, unlike governments, the courts lack the information

Health and healthcare as a human right  15 and expertise to assess competing treatment claims and to consider what meeting those claims requires (Daniels, 2009). For this reason, Daniels emphasized the division of labour between governments and the courts regarding the allocation of healthcare resources. He argued that judicial enforcement of a legal right to health appropriately works if courts limit their role to protecting the needs of right-holders against a government’s priority-setting decisions or engaging with policy makers and scrutinizing their decisions to demand transparency, fairness, and evidence (Daniels et al., 2015).

4.

A RIGHT TO HEALTH IN THE CONTEXT OF TRADITIONAL DISCIPLINES: FROM HEALTH ECONOMICS TO HUMAN RIGHTS INDICATORS RESEARCH

In the real world, economic activities and human rights are intimately connected. This is especially true for the right to health and other social and environmental rights. In Organisation for Economic Co-operation and Development (OECD) countries, governments spend a significant portion of their national budget on health, education, the environment, and social protection, which are closely linked to the protection of social and environmental human rights described in their constitutions (OECD, 2021). However, economists had long neglected human rights and their potential effects on governmental spending and other economic outcomes until Ben-Bassat and Dahan published their paper on the relationship between social rights and outcomes in 2008. In his book On Ethics and Economics, Sen attempted to answer why human rights have not been centred in the analysis of economics. He explained that basic economic concepts such as endowment, exchange, and contract involve various types of rights. Under utilitarian tradition, these rights were viewed as entirely instrumental to achieving other goods, in particular utilities, and had no intrinsic value attached (Sen, 1987). However, this does not explain why a right to health or other social and environmental rights have been ignored in economic analyses. In fact, having instrumental values is sufficient reason for economists to analyse these rights because these rights can still improve overall social welfare, which economists care about. The rest of this section will investigate why a right to health has not been centred on health economics or other related fields until quite recently. 4.1

In the Context of Health Economics

Health economics in the context of health policy deals with the effective distribution of healthcare and other determinants of health, with recent literature having a focus on equal distribution. In the United Kingdom, where cost-effectiveness analysis plays a major role in health policy, the concern for equity has informed the development of various methods of distributional cost-effective analysis to provide information about the distribution of costs and health benefits (Cookson et al., 2017). Empirical studies in this area rarely distinguish between egalitarian and prioritarian frameworks, with most studies looking at the cost and health benefits among low-income people and other vulnerable populations. In the United States, where insurance is a critical determinant of access to healthcare, the health consequences of being uninsured have been intensively studied (McWilliams, 2009). In both cases, access to

16  Handbook on the political economy of health systems healthcare for low-income people and other vulnerable populations has been well studied, but a right to health has not been central to the analysis. There are both theoretical and empirical reasons why a right to health has not been of key focus in the literature. From the theoretical side, health is just one of many inputs of the utility function in economics. Unlike human rights, economists do not have a pluralistic concept of value since utility is the only measure of a thing’s value. In this framework, health cannot be special goods whose consumption should be ethically protected over others. Second, which is also closely related to the first point, the theory of compensating differentials has been well established in economics. For example, value-of-statistical-life literature uses compensating differentials, which are the additional ex ante payments that are intended to compensate workers for high-risk jobs. However, both the ethical and legal right to health requires post hoc compensatory justice in the case of violation, but the occurrence of a violation is, in the first instance, strictly prohibited. From the empirical side, health is not something that can be easily measured. In recent years, there have been significant efforts to measure health using standardized and validated instruments such as Quality-adjusted Life Years or Disability-adjusted Life Years (Murray et al., 2002). However, we are not confident enough to rely on these measures to determine the threshold of a decent minimum level of health, implied by an ethical and legal right to health. It is worth noting that we can still obtain reliable data on the “consumption” of healthcare, which may be helpful to determine the threshold of a decent minimum level of healthcare. However, “demand” for healthcare is seldom, unexpected, and largely unpredictable. This is because most people who are healthy do not need healthcare at a given moment, so it is difficult to specify a decent minimum level of healthcare use for all members of society. On the other hand, demand for food or water is relatively stable over time because they have no close substitutes and are used in daily life. This is a reason why we have economic research on the right to food and water but not the right to health or to healthcare (see the next section). When a right to health is enshrined in national constitutions or other superior laws, the existence of a right to health is directly observable by researchers. In this case, one might be interested in the potential effects of such a right on health systems and their outcomes. However, there is some distance between national constitutions and actual health outcomes. There must be more proximal health legislations that make constitutional promises to become real between them. In the context of the United States, health economists have studied the effects of legislation on guaranteeing access to healthcare such as Medicaid, Medicare, and the Hill–Burton requirements of charitable care on health outcomes rather than the effect of a right to health in the U.S. and state constitutions (Finkelstein, 2007; Almond et al., 2011; Finkelstein et al., 2012). Much of the literature on access to healthcare outside the United States has also focused on the role of more proximal health legislations rather than a right to health in national constitutions. Finally, there may be more fundamental reasons that explain a lack of attention to the role of a right to health in health economics. A right to health is not recognized at the level of national constitutions in the United States and the United Kingdom, where health economic research is most active.2 For this reason, health economists in these two countries do not express interest in the role of a constitutional right to health. However, the right to health has also been largely ignored in health economics outside of these countries. In recent years, health economists have shown increased engagement with the capability approach originally developed by Amartya Sen and Martha Nussbaum. Both founders admit-

Health and healthcare as a human right  17 ted that health is an important dimension of capability. Life, bodily health, and bodily integrity are included in Nussbaum’s list of ten central capabilities, while Sen recognized that “health is among the most important conditions of human life and a critically significant constituent of human capabilities which we have reason to value” (Sen, 2002, p. 660; see also Nussbaum, 2001). With similar reasoning as Daniels, capability theorists believe that everyone should have the capability to obtain a decent and sufficient level of health because this is a necessary condition for “flourishing”, and a lack of health is deemed as a corrosive disadvantage (Culyer, 2001). For these reasons, researchers in this area are more likely to address the right to health than researchers who engage in different topics in health economics. In practice, however, the use of the capability approach in health economics has focused mainly on the development of capability questionnaires. Several capability-based questionnaires have been proposed for use in healthcare, including the Oxford capability instruments, the Adult Social Care Outcome Toolkit, and the ICEpop Capability Measures (Coast et al., 2015). Such questionnaires have shifted the focus from achieved functionings (which Daniels cared about) to the freedom that a person has the right to achieve different aspects of well-being in their life. The literature of the capability approach in health economics is likely to contribute to defining the scope and limit of an ethical right to health, but it is unlikely to contribute to an understanding of how a legal right to health in the real world works towards improving health outcomes. 4.2

In the Context of Empirical Health Laws

In recent years, the empirical literature on health laws has flourished in the area of law and public health (Mello and Zeiler, 2007). Traditionally, such studies have been carried out in the field of law and economics, but due to the medical and legal expertise needed in this area, more scholars with a background in law, medicine, and public health have engaged in this type of research. Researchers in this area have covered most aspects of the legal system related to health systems, including contract, tort, criminal, and administrative law, but have rarely worked on constitutional law, including a legal right to health enshrined in constitutions. One reason for this is that much of the research in this area has been concentrated in the United States due to the availability of legal and medical data as well as state-by-state variations in healthcare law. However, the U.S. Constitution contains no direct reference to health, and so the attention of researchers to state constitutional provisions of health and healthcare is only a recent phenomenon. For this reason, the constitutional right to health has not been well studied in the area of empirical health law. 4.3

In the Context of Constitutional Economics

Although the early political economy discussion on constitutional design has focused on the normative side of constitutions (Voigt, 2020),3 more recent literature has focused on empirical research investigating the constitutional effects on economic policy and performance. After the breakup of the Soviet Union, the focus shifted towards the effect of different democratic regimes (see Chapter 3 in this book). These studies have focused on different types of democratic regimes based on different electoral rules (proportional vs majority), different forms of government (presidential vs parliamentary), and different degrees of judicial independence enshrined in national constitutions (La Porta et al., 2004; Persson and Tabellini, 2004). In general, national constitutions outline not only the ways in which the government operate but

18  Handbook on the political economy of health systems highlight the fundamental rights and freedoms that people can enjoy. In the mid-2000s, Blume and Voigt analysed the effect of civil and political rights and some parts of workers’ rights on economic outcomes such as investment and productivity (Blume and Voigt, 2007). Due to incomplete datasets, they derived their human rights variables using factor analysis from the combined data of civil and political rights indicators and economic rights indicators. It is hard to understand their results because they investigated the relationship between factors they derived from a handful of variables and economic outcomes. However, they identified that the factors representing emancipatory rights and basic human rights could affect economic outcomes such as investment and productivity differently. To my knowledge, Ben-Bassat and Dahan were the first to collect constitutional data of social rights that included the right to social security, education, health, housing, and workers’ rights (Ben-Bassat and Dahan, 2008). Their data were cross-sectional, covered a smaller number of countries (68 countries), and were not based on expert judgements. However, their results mostly hold in the literature following their studies (Matsuura, 2013; Heymann et al., 2014; Edwards and Marin, 2015; Ben-Bassat and Dahan, 2016; Chilton and Versteeg, 2017; Minkler and Prakash, 2017; Cammeraat, 2021). Data availability for comparative human rights law seemed to be the major limiting factor for early research efforts in the study of a right to health and other social and economic rights. 4.4

In the Context of Human Rights Indicators Research

Debates over the way to track human rights progress or violations over time have preoccupied scholars for many years, but the use of quantitative indicators to track the progress of social and environmental rights at an international level is seen as a relatively recent phenomenon. In 1990, Danilo Turk, former United Nations Special Rapporteur on the Realization of Economic, Social and Cultural Rights, urged the use of quantitative indicators in the promotion and respect of economic, social, and cultural rights (Turk, 1990). In 2012, the Office of the United Nations High Commissioner for Human Rights (OHCHR) adopted the framework of human rights indicators, consisting of three different types of indicators: structural, process, and outcome (OHCHR, 2012). Structural indicators measure the acceptance and commitment to human rights. They reflect ratification and the enactment of legal instruments, and the existence of basic institutional mechanisms deemed necessary for the implementation of human rights. Process indicators measure duty bearers’ “efforts” to transform their human rights commitments into the desired results. These indicators assess policy measures taken to implement commitments on the ground. Finally, outcome indicators capture results that reflect the level of enjoyment of human rights in each context. The OHCHR has periodically updated its Human Rights Indicators Tables, and the most updated version included the global indicators for the Sustainable Development Goals (SDGs) in the table (OHCHR, 2021). Tracking the progress of the SDGs is the most ambitious and comprehensive data collection project at a global level in human history. The 232 global indicators, which aim to track the progress of 17 goals and 169 targets, are complemented by indicators at the regional and national levels developed by each UN Member State. These 232 indicators are then integrated into human rights indicators by the OHCHR. Table 1 shows the illustrative indicators on the right to enjoy the highest attainable standard of physical and mental health prepared by the OHCHR. Among the health-related global indicators listed in

Health and healthcare as a human right  19 Goal 3 of the SDGs, 16 indicators, including child and maternal health ([3.1.1][3.2.1][3.2.2] [3.7.2]), and prevalence ([3.3.1][3.3.2][3.3.3][3.3.4]) and mortality ([3.4.1][3.42][3.6.1][3.9.1] [3.9.2][3.9.3]) of specific diseases and injuries and risk factors ([3.5.2] [3.a.1]), are classified as outcome indicators. Twelve indicators, including human resource to health ([3.1.2][3.c.1]), patients with neglected diseases and antimicrobial-resistant (AMR) infections ([3.3.5] [3.d.2]), health service coverage ([3.5.1][3.7.1][3.8.1][3.b.1][3.b.3]) and health-related expenditure ([3.8.2], [3.b.2]), and International Health Regulations capacity and preparedness ([3.d.1]), are classified as process indicators. However, no indicator is classified as a structural indicator. In addition, four outcome indicators, nine process indicators, and two structural indicators outside of Goal 3 are added to the table. In total, 13 targets are listed in Goal 3 of the SDGs, but none of the targets includes all three types of indicator. It is clear that the SDGs’ indicators are concentrated in the process and outcome indicators. However, among the 13 targets included in Goal 3, only target 3.1 (maternal mortality), 3.5 (substance and alcohol abuse), and 3.7 (family planning) have both process and outcome indicators in a consistent manner. This means that the SDGs’ indicators do not provide enough information to understand how structural indicators, including a constitutional right to health, translate into process and outcome indicators at the country level. Although no systematic efforts are made to collect structural indicators at the international level, in the mid-2000s Backman and others collected a wide variety of indicators related to the right to health, which covers major structural indicators listed in Table 1. Such indicators include a status of a right to health recognized in ratified international treaties, national constitutions and other forms of superior law, and domestic laws (Backman et al., 2008). Later, Matsuura expanded this data set into a panel format of a right to health in ratified international treaties and national constitutions (Matsuura, 2013). Systematic collection of all three types of human rights indicators will contribute to an understanding of how a right to health in ratified treaties, national constitutions, or domestic laws work towards the improvement of a health system and population health. However, the limited availability of these indicators is still a major constraint for studies in this area.

5.

ECONOMICS OF THE “RIGHT TO HEALTH”

Although the study of a right to health had been quite limited until the mid-2000s, as more data on comparative human rights law become available, there has been a growing common interest in the role of a right to health in health economics and other related fields. It is fair to say that the early interest in integrating a right to health as an ethical right into these fields has given way to the interest in integrating this right as a legal right into these fields. This section overviews the recent progress of the study of a right to health in the context of health economics or, more broadly, the study of resource allocation of health and healthcare. There are at least three ways to incorporate a right to health in the analysis of resource allocation (Matsuura, 2014). This section reviews these three approaches in turn: a right to health as a social objective, a right to health as a constraint, and a right to health as a legal, political, and social instrument.

20  Handbook on the political economy of health systems 5.1

“Right to Health” as a Social Objective

The first approach is in regard to health outcomes, implied by a right to health and as an objective that society should work towards. As social welfare is not directly measurable, economists often opt to indirectly measure social welfare by using a list of indicators that are considered to capture relevant aspects of human well-being or a composite index that combines these indicators. Health outcomes are often selected as one or more of such indicators because the fact that a right to health is important enough to be included in a list of human rights helps implied health outcomes to be included in a list of indicators. In the former approach, the list of indicators reflects our value of what is important for social progress. The best-known example of this is the United Nations Millennium Development Goals and, more recently, the SDGs. As we saw in the previous section, health-related indicators are well included in the SDGs’ global indicators. It is worth noting that some of these indicators are classified into process indicators. A right to health includes a right to healthcare. In the OHCHR’s framework, healthcare is “efforts” to achieve decent health outcomes. For example, target 3.8 aims to achieve universal health coverage without suffering financial hardship, which is implied by a right to access healthcare. This is measured using two indicators, including coverage of essential health services and the proportion of people with a large household expenditure on health. Theoretically, where outcome indicators suggest achieved outcome is below the level expected, process indicators can be used diagnostically to explore the reasons. However, in the development goals, the relationship between outcome and process indicators is quite ambiguous and both types of indicators are used as social objectives that society should work towards. In the latter approach, one can interpret such a composite index as a proxy of social welfare or summary statistics of some components of social welfare. The best-known example of this is the Human Development Index (HDI) and, more recently, the Multidimensional Poverty Index (MPI) developed by the United Nations Development Program (UNDP), which aggregates a number of indicators in the dimensions of health, education, and standard of living (UNDP, 2020). Health is measured by life expectancy in the HDI, while deprivation in health is measured by household undernutrition and under-5 mortality in the MPI. Such composite indicators are often criticized because they implicitly assume trade-offs between components. For example, the same HDI can be achieved with less life expectancy and more per capita gross domestic product. Since human rights do not allow for such an exchange, the former approach is more desirable from the viewpoint of a right to health. The right to health includes not only the provision of timely and appropriate healthcare but also the fair distribution of a wide variety of other determinants of health. However, this right alone does not provide adequate information on which indicators are better included than the others. As the number of indicators included in a list must be kept to a minimum, the list of indicators must be agreed through fair and publicly accountable deliberation. In this sense, a right to health as a desirable social objective is more an issue of ethical rather than legal decision-making. 5.2

“Right to Health” as a Constraint

The right to health is interpreted to encompass a minimum consumption requirement of healthcare the government is obligated to fulfil when maximizing aggregated welfare. However, as

Health and healthcare as a human right  21 discussed in the previous section, the fact that demand for healthcare is seldom, unexpected, and largely unpredictable makes it difficult to treat a right to health as a constraint. On the other hand, demand for food or water is stable and largely predictable. There is substantial literature on consumption behaviour with subsistence consumption constraints. Jefford and Shah have subjected the human right to water to an analysis that builds upon assumptions of a neoclassical framework (Jeffords and Shah, 2013). Chavas also incorporated a minimum level of food consumption required to maintain life in an analysis of calculating the cost of food insecurity (Chavas, 2017). Although it is difficult to analyse a right of access to healthcare as a constraint, there is a study that analyses the right to equal quality of treatment. Cleary and others used mathematical programming techniques to achieve the priorities of HIV care while maintaining equitable treatment for equal conditions (Cleary et al., 2010). They calculated efficiency loss due to the constraint guaranteeing first-line antiviral treatment for all patients under a low-resource setting. 5.3

“Right to Health” as a Legal, Political, and Social Instrument

The last approach considers the constitutional right to health as a legal, political, or social instrument for achieving desirable health outcomes, implied in the right to health. Taking a right to health as the statement of a desirable social objective or constraint is an ethical issue. However, if we regard a right to health as an instrument for achieving desirable health outcomes, we have an empirical question that can be addressed with data. This type of research has rapidly grown over the last 15 years due to more availability of human rights indicators and has now covered not only the constitutional right to health but also a constitutional right to an adequate living, a constitutional right to education, and a constitutional right to the environment (Ben-Bassat and Dahan, 2008; Matsuura, 2013, 2015, 2016, 2018, 2019; Heymann et al., 2014; Edwards and Marin, 2015; Ben-Bassat and Dahan, 2016; Jeffords and Minkler, 2016; Chilton and Versteeg, 2017; Minkler and Prakash, 2017; Cammeraat, 2021). There is a reason why the constitutional right to health has been of focus in this empirical literature. An early study found that the effects of a right to health in ratified international treaties are limited. For example, Palmer and his colleagues investigated the relationship between the cumulative numbers of treaty ratification and health indicators such as maternal, infant, and child mortalities using cross-sectional data. They found no systematic association between them (Palmer et al., 2009). The results were also confirmed by the analysis using panel data (Matsuura, 2019). For the constitutional right to health, Ben-Bassat and Dahan were the first to analyse cross-sectional data of constitutional commitments to social rights. They found that a right to health was correlated with a reduction in infant mortality, but not life expectancy and health expenditure (Ben-Bassat and Dahan, 2008). Matsuura looked at the constitutional right to health in 157 countries and examined the effect of introducing such a right on infant and under-5 mortality using cross-country panel data for the period 1970 to 2007 (Matsuura, 2013). The study found that infant and under-5 mortality rates decreased significantly after countries introduced a right to health provision in their national constitutions. Moreover, this study investigated the potential role of institutions. The health-improving effect of a right to health seems to be much larger in countries with good governance and those with a high democracy score, reflecting both political competition and the rule of law. The latter study

22  Handbook on the political economy of health systems found no health-improving effect for maternal health outcomes even in countries with good governance (Matsuura, 2019). Another avenue of research focused on the protective or preventive effects of the presence of the constitutional right to health on the surge of child deaths at the time that natural and man-made disasters strike. The study only found a protective effect against international armed conflict, but this result is far from robust (Matsuura, 2018). Further research is needed based on more detailed regional data of disasters and health outcomes. The other type of research was conducted in specific regional areas, with one study focusing on state constitutions of the United States as opposed to the national level (Matsuura, 2015). The United States is a nation with typical British-common-law origins, which means that it does not guarantee health or healthcare as a fundamental right in its constitution. However, the federal constitutional design allows 15 state constitutions to contain provisions relating to health and healthcare (Leonard, 2009). Using this variation, the study found that the introduction of stronger constitutional commitment, which imposes a duty on the state to provide healthcare, was associated with a reduction in state infant mortality rates. Also, explicit targeting of the poor in the right to healthcare in the constitution is associated with a reduction in infant mortality rates. Finally, the health-improving effects of a constitutional right to health are found mainly from health improvements in the non-white population. This means that the right to health has an effect of reducing disparities as well as raising average population health. Other studies focused on Latin American countries, where human rights litigations are particularly active. As mentioned earlier, since the mid-1990s, the courts in these regions have actively redressed the failures of education, health, welfare, and environmental policies. Hundreds of thousands of reported decisions based on these rights have already been published (Hogerzeil et al., 2006). By using retrospective fertility surveys at the individual level from the Demographic and Health Surveys, Matsuura examined the health-improving effects of introducing a right to health into national constitutions in 15 Latin American countries (Matsuura, 2016). The study found that introducing a right to health into national constitutions is associated with a 2.6 per cent reduction in infant deaths among poor mothers, but was not associated with a decrease in infant deaths among the general population. These two studies also explain the relationship between a right to health and other social and environmental rights enshrined in national constitutions. The achievement of a right to health is closely related to, and dependent on, the realization of other social and environmental human rights. In fact, there is well-established evidence and recognition that education, adequate living (welfare), and a clean environment are considered determinants of health (WHO, 2008). On the other hand, these rights must be financed through competing government budgets like health and healthcare. This means there is a possibility that the implementation of one right may hinder the achievement of another. However, these two studies found that the rights to education, welfare, and environment are not associated with a reduction in infant mortality. In addition, these rights are not associated with an increase or decrease in total government spending and its composition related to these rights. Therefore, from the viewpoint of health, introducing rights to education, welfare, and environment into national constitutions will not benefit child health, but at the same time, it will not hinder public health activity because introducing these rights does not lead to a decrease in the healthcare budget of the government. These two studies also inform the potential pathway through which the right to health works to improve infant health among poor households. The study found that the effects of a right to health are robust after allowing for total government spending and its composition, which indicates that the allocation of government health spending – rather than an absolute amount

Health and healthcare as a human right  23 of spending induced by the constitutional right to health – might be the main mechanism to reduce infant deaths among vulnerable populations such as the non-white population or poor mothers. More recent research focused on the health-deteriorating effect of the violation of a right to health. Colombia is one of the most active human rights litigation spots in Latin America. From 2010 to 2016, more than 675,000 judicial claims (tutelas) were filed against health insurers. In Colombia, these claims are costless and can be filed with any judge within the local jurisdiction. The judge is also mandated to return a decision within ten days. On average, 80 per cent of health-related claims were decided in favour of the claimants. A recent paper by Bhalotra and Fernández Sierra (2021) used these judicial claims as a proxy for unmet demand for healthcare and found that an increase in judicial claims led to a restriction of healthcare use and an increase in mortality rates. They also found that this health-deteriorating effect is more prominent among vulnerable groups, such as individuals aged over 50, women, and the low-income population. In sum, the literature found that the constitutional right to health is health-improving, and a violation of this right is health-deteriorating. The effect is particularly larger among vulnerable population groups, but such an effect is also dependent on the quality of governance in a country and the level of commitment and strategies described in the constitutional provision. There is still a question of how to interpret the health-improving effects of a right to health. In some countries, like Colombia, the courts have taken an active role in redressing violations of a right to health, but even in the United States, where courts are not at all active in this area, we still find the health-improving effects. As Sen discussed, the fact that a right to health is important enough to be included in national constitutions not only fuels political processes and guides health legislation, it also enhances a social commitment of citizens to the right to health through public discussion, social monitoring, investigative reporting, and social work (Sen, 2009). The estimated health-improving effects capture all of these effects combined. On the other hand, there is no evidence that a legal right to health has distorted the allocation of health resources and proved harmful for population health. Thus, on average, the constitutional right to health works to improve overall population health without known side effects.

6. CONCLUSION This chapter summarized the role of the right to health enshrined in national constitutions or other forms of supreme laws in the resource allocation of health and healthcare. Recognizing a constitutional right to health creates a legal obligation of governments to respect this right, but also a catalyst for political and social changes beyond legislative routes if such a right is to be internalized as an ethical norm by individuals, governments, and non-state players. Like other human rights, a right to health has not been centred in economic analysis, but over the last 20 years, studies on the right to health as legal, political, and social instruments have flourished due to the increasing availability of data. As Sen addressed, economists do not place intrinsic value on a right to health and other human rights. However, it is evident that having instrumental values is good enough for these rights to attract the attention of economists. Future research is expected to enrich the pool of growing knowledge of the effects of a constitutional right to health and other human rights as instruments for improving real-world health outcomes.

24  Handbook on the political economy of health systems

NOTES 1.

These international treaties include, but are not limited to, the International Covenant on Economic Social and Cultural Rights (ICESCR), the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW), the Convention on the Rights of the Child (CRC), the International Convention on the Elimination of All Forms of Racial Discrimination (ICERD), the International Covenant on Civil and Political Rights (ICCPR), and the Convention Against Torture and other Cruel Inhuman or Degrading Treatment or Punishment (CAT). 2. The U.S. Constitution contains no direct reference to health and has never been interpreted to grant any legal protection of individual rights to health and healthcare. In the United Kingdom, a right to health is recognized in the Human Rights Act since there is no codified constitution. 3. According to Buchanan, constitutional economics has been developed “as an integral, but distinguishable, part of the sub-discipline of public choice” in the latter half of the twentieth century (Buchanan, 1990, p. 3).

REFERENCES Almond, D., Currie, J., & Simeonova, E. 2011. Public vs. private provision of charity care? Evidence from the expiration of Hill-Burton requirements in Florida. Journal of Health Economics, 30(1), 189–99. Backman, G., Hunt, P., Khosla, R., Jaramillo-Strouss, C., Fikre, B.M., Rumble, C., Pevalin, D., Páez, D.A., Pineda, M.A., Frisancho, A., Tarco, D., Motlagh, M., Farcasanu, D., & Vladescu, C. 2008. Health systems and the right to health: an assessment of 194 countries. Lancet, 372(9655), 2047–85. Ben-Bassat, A., & Dahan, M. 2008. Social rights in the constitution and in practice. Journal of Comparative Economics, 36(1), 103–19. Ben-Bassat, A., & Dahan, M. 2016. Constitutional commitment to social security and welfare policy. Review of Law & Economics, 12(1), 165–201. Bentham, J. 1838. Anarchical fallacies. In: Bowring, J. (ed.), The Works of Jeremy Bentham. Edinburgh: William Tait, vol. 2. Bhalotra, S., & Fernández Sierra, M. 2021. The right to health and the health effects of denials. Working Paper. Documento CEDE. Blume, L., & Voigt, S. 2007. The economic effects of human rights. Kyklos, 60(4), 509–38. Buchanan, J.M. 1990. The domain of constitutional economics. Constitutional Political Economy, 1(1), 1–18. Cammeraat, E. 2021. The effect of constitutional commitment to social security on social expenditure schemes. Review of Law & Economics, 17(1), 193–222. Chavas, J.-P. 2017. On food security and the economic valuation of food. Food Policy, 69, 58–67. Chilton, A., & Versteeg, M. 2017. Rights without resources: the impact of constitutional social rights on social spending. Journal of Law and Economics, 60(4), 713–48. Cleary, S., Mooney, G., & McIntyre, D. 2010. Equity and efficiency in HIV-treatment in South Africa: the contribution of mathematical programming to priority setting. Health Economics, 19(10), 1166–80. Coast, J., Kinghorn, P., & Mitchell, P. 2015. The development of capability measures in health economics: opportunities, challenges and progress. Patient, 8(2), 119–26. Cookson, R., Mirelman, A.J., Griffin, S., Asaria, M., Dawkins, B., Norheim, O.F., Verguet, S., & Culyer, A.J. 2017. Using cost-effectiveness analysis to address health equity concerns. Value in Health, 20(2), 206–12. Culyer, A.J. 2001. Equity: some theory and its policy implications. Journal of Medical Ethics, 27(4), 275–83. Daniels, N. 1985. Just Health Care. Cambridge: Cambridge University Press. Daniels, N. 2007. Just Health: Meeting Health Needs Fairly. Cambridge: Cambridge University Press. Daniels, N. 2009. Is there a right to health care and, if so, what does it encompass? In: H. Kuhse & P. Singer (eds), A Companion to Bioethics, 2nd edn. Oxford: Blackwell.

Health and healthcare as a human right  25 Daniels, N., Charvel, S., Gelpi, A.H., Porteny, T., & Urrutia, J. 2015. Role of the courts in the progressive realization of the right to health: between the threat and the promise of judicialization in Mexico. Health Systems & Reform, 1(3), 229–34. Edwards, S., & Marin, A.G. 2015. Constitutional rights and education: an international comparative study. Journal of Comparative Economics, 43(4), 938–55. Finkelstein, A. 2007. The aggregate effects of health insurance: evidence from the introduction of medicare. Quarterly Journal of Economics, 122(1), 1–37. Finkelstein, A., Taubman, S., Wright, B., Bernstein, M., Gruber, J., Newhouse, J.P., Allen, H., Baicker, K., & Oregon Health Study Group. 2012. The Oregon health insurance experiment: evidence from the first year. Quarterly Journal of Economics, 127(3), 1057–1106. Heymann, J., Raub, A., & Cassola, A. 2014. Constitutional rights to education and their relationship to national policy and school enrolment. International Journal of Educational Development, 39, 121–31. Hogerzeil, H.V., Samson, M., Casanovas, J.V., & Rahmani-Ocora, L. 2006. Is access to essential medicines as part of the fulfilment of the right to health enforceable through the courts? Lancet, 368(9532), 305–11. Ishay, M.R. 2004. The History of Human Rights: From Ancient Times to the Globalization Era. Berkeley: University of California Press. Jeffords, C., & Minkler, L. 2016. Do constitutions matter? The effects of constitutional environmental rights provisions on environmental outcomes. Kyklos, 69(2), 294–335. Jeffords, C., & Shah, F. 2013. On the natural and economic difficulties to fulfilling the human right to water within a neoclassical economics framework. Review of Social Economy, 71(1), 65–92. La Berge, A. 2002. Mission and Method: The Early Nineteenth-Century French Public Health Movement. Cambridge: Cambridge University Press. La Porta, R., López-de-Silanes, F., Pop-Eleches, C., & Shleifer, A. 2004. Judicial checks and balances. Journal of Political Economy, 112(2), 445–70. La Porta, R., López-de-Silanes, F., & Shleifer, A. 2008. The economic consequences of legal origins. Journal of Economic Literature, 46(2), 285–332. Leonard, E.W. 2009. State constitutionalism and the right to health care. University of Pennsylvania Journal of Constitutional Law, 12, 2010. Matsuura, H. 2013. The effect of a constitutional right to health on population health in 157 countries, 1970–2007: the role of democratic governance. Program on the Global Demography of Aging Working Papers 10613. Matsuura, H. 2014. Does the constitutional right to health matter? A review of current evidence. CESifo DICE Report, 12(2), 35–41. Matsuura, H. 2015. State constitutional commitment to health and health care and population health outcomes: evidence from historical US data. American Journal of Public Health, 1–5(Suppl 3), e1–e7. Matsuura, H. 2016. Constitutional social and environmental human rights and child health outcomes in Latin American countries. WIDER Working Paper Series. World Institute for Development Economic Research. Matsuura, H. 2018. The protective effect of the constitutional right to health during natural and man-made disasters. Proceedings from the 2018 International Conference on Sustainable Development 2018. New York. Matsuura, H. 2019. Exploring the association between the constitutional right to health and reproductive health outcomes in 157 countries. Sexual and Reproductive Health Matters, 27(1), 168–80. McWilliams, M.J. 2009. Health consequences of uninsurance among adults in the United States: recent evidence and implications. Milbank Quarterly, 87(2), 443–94. Mello, M.M., & Zeiler, K. 2007. Empirical health law scholarship: the state of the field. Georgetown Law Journal, 96, 649. Minkler, L., & Prakash, N. 2017. The role of constitutions on poverty: a cross-national investigation. Journal of Comparative Economics, 45(3), 563–81. Murray, C.J., Salomon, J.A., Mathers, C.D., Lopez, A.D., & World Health Organization 2002. Summary Measures of Population Health: Concepts, Ethics, Measurement and Applications. Geneva: World Health Organization. Nussbaum, M.C. 2001. Women and Human Development: The Capabilities Approach. Cambridge: Cambridge University Press.

26  Handbook on the political economy of health systems OECD. 2021. Government at a Glance. Paris: Organisation for Economic Co-operation and Development. OHCHR. 2012. Human Rights Indicators: A Guide to Measurement and Implementation. Geneva: United Nations. OHCHR. 2021. Human Rights Indicators Tables, Updated with the Sustainable Development Goals Indicators. Geneva: United Nations. Palmer, A., Tomkinson, J., Phung, C., Ford, N., Joffres, M., Fernandes, K.A., Zeng, L., Lima, V., Montaner, J.S., & Guyatt, G.H. 2009. Does ratification of human-rights treaties have effects on population health? Lancet, 373(9679), 1987–92. Persson, T., & Tabellini, G. 2004. Constitutional rules and fiscal policy outcomes. American Economic Review, 94(1), 25–45. Ruger, J.P. 2006. Toward a theory of a right to health: capability and incompletely theorized agreements. Yale Journal of Law and the Humanities, 18(2), 273–326. Sen, A.K. 1987. On Ethics and Economics. Oxford: Blackwell. Sen, A.K. 2002. Why health equity? Health Economics, 11, 659–66. Sen, A.K. 2009. The Idea of Justice. Cambridge: Harvard University Press. Turk, D. 1990. Realization of Economic, Social and Cultural Rights: Progress Report. Geneva: United Nations. UN. 2000. General Comment 14, the Right to the Highest Attainable Standard of Health. New York: United Nations. UNDP. 2020. Human Development Report 2020: The Next Frontier – Human Development and the Anthropocene. New York: United Nations. Visscher, M.B., ed. 1972. Humanistic Perspectives in Medical Ethics. Amherst: Prometheus Books. Voigt, S. 2020. Constitutional Economics: A Primer. Cambridge: Cambridge University Press. Weiner, D.B. 1993. The Citizen-Patient in Revolutionary and Imperial Paris. Baltimore: Johns Hopkins University Press. WHO. 2008. Closing the Gap in a Generation: Health Equity through Action on the Social Determinants of Health – Final Report of the Commission on Social Determinants of Health. Geneva: World Health Organization. Zuniga, J.M., Marks, S.P., & Gostin, L.O. 2013. Advancing the Human Right to Health. Oxford: Oxford University Press.

3. Do democratic regimes exhibit ‘better’ health outcomes? Alberto Batinti and Joan Costa-Font

1.

DEMOCRACY AND HEALTH

Individuals living in democracies, unlike those exposed to autocracies and dictatorships, typically exhibit equally before the law, and are provided important spheres of autonomy to determine their life courses, even when they might still face inequality in such choices. Indeed, citizens in democratic regimes hold some legal and social instruments to control power and are represented directly or indirectly in institutions which they elect. Their preferences, including their trade-offs between health and other social programs, are taken into consideration and contribute to the formation of government actions. Different country-specific democratic institutions can vary in how effectively equal individuals are, how strictly they apply rules of law and how they aggregate preferences (e.g., the type of electoral processes). What defines democratic countries is that their institutions are formed through forms of direct representation, also entailing competition among policy platforms (Downs, 1957) and political elites (Schumpeter, 1947), which in turn widens the potential set of policy options to attain as much political support possible. Yet, does it pay off in terms of health? That is, do individuals exposed to democratic institutions exhibit better health outcomes? If so, what health outcomes are more sensitive to democracy exposure? What underlying mechanisms explain a health effect of political regimes? Providing a satisfactory answer to such questions is difficult insofar as democratic regimes are far from simple to characterize. This is because democracy is part result, part cause of the wider social influences affecting a population’s health. Democratic institutions create environments that are part of what epidemiologists call ‘social determinants of health.’ In public health, economics and politics the relationship between democracy and health outcomes has received increasing attention in several social sciences. The last few years have seen an increasing amount of research dedicated to how democratic regimes1 and regime transitions impact health and health policies. However, such research is very heterogeneous in the set of chosen outcomes, measures of democracy and especially the consideration of different potential mechanisms in place. This chapter provides a contribution on the role of democratic regimes as institutional determinants of health, and specifically discusses the scholarly work examining how democracy affects health outcomes. In doing so, it provides an update of recent reviews of the literature in Costa-Font et al. (2020) and McGuire (2020), to which we refer the reader. More specifically, we discuss evidence of the effect of different measures of democracy on health outcomes that have been examined in the literature, and the potential reasons for the heterogeneity of the effects found across outcomes. The literature spreads across different fields, yet perhaps more often than desired, disciplines might fail to speak to each other, even when estimates are complementary, or consistent with different explanations. 27

28  Handbook on the political economy of health systems The next two sections elaborate on the different mechanisms underpinning the effect of democracy on health outcomes. Section 4 specifically examines the effect of the most recent literature documenting the effect of democracy on human height. Section 5 then provides an analysis on the issues of identification and mechanisms in examining the effect of democracy on health. The next section discusses how democracy influences health in a pandemic, and a final section concludes.

2.

HOW DOES DEMOCRACY INFLUENCE HEALTH?

The influence of democracy on the health of the population can be mediated by several mechanisms. 2.1

Democratic Environments

Democratic regimes influence the set-up of informal institutions such as social environments and cultures in which children grow up and individuals live. It is apparent that democracies tend to brew specific cultures with further attention to individuals’ needs, and as we discuss later social values of diversity and pluralism influence the types of aspirations individuals have, and their demands. For instance, democracies typically exhibit stability of employment relationships (as the instability of recessions weakens democracies) and the functioning of markets, and a positive impact on wages for blue-collar workers (Rodrik, 1999). Hence, democracies provide the means to set up reforms conducive to environments that reduce the stress individuals face in day-to-day life. Individuals living in democracies tend to have a better sense of agency (Benz & Frey, 2008) that comes from the subjective rights that a democratic regime provides for individuals, which include key, important household choices such as housing, basic social services and representation, both in companies through trade unions and directly in different levels of government. 2.2

Institutional Constraints

Democracies not only encompass different environments and informal institutions, but also entail a system of constitutionally defined institutions (whether codified or not), including a set of well-defined human and economic rights that protect individuals, among them the right to health care, which is discussed in Chapter 2. However, among those institutions, one could include electoral rules, the division of powers and the judiciary. Such rules guiding the allocation of power in a society can influence health by allowing individuals to report to the courts when others are infringing on their right to health care (as discussed in Chapter 2), and to select incumbents that are more likely to deliver the health care programs that align with one’s own preferences. Such institutions can influence the functioning of the health systems, the extent of corruption as well as health system priorities. That is, how ‘health’ goals are traded off against other alternatives goals in society (e.g., industrial or business interests, see Chapter 17). Insofar as the government’s power in a democracy is representative of the people’s preferences, it becomes more responsive to the needs and claims of the population.

Do democratic regimes exhibit ‘better’ health outcomes?  29 2.3

Policy Processes

Democratic regimes exhibit distinct policy processes, which typically involve veto players and inputs (including information) from a larger share of stakeholders than autocratic regimes. Democracy increases the number of agents influencing the policy process, and hence involves a larger number of different social groups that would have otherwise been neglected in non-democracies expressing their political demands for health care and other social determinants of health. Hence, it gives rise to a different system of collective decision making that brings about formal mechanisms of accountability, representation and participation, which are in place so that decisions on health care are the expressions of constituents’ demands (Besley, 2006; Besley & Kudamatsu, 2006). Increased participation by the segments of the population that are more likely to benefit the most from health policies and interventions enables improvements to the same health policies and proposals through a fair and competitive electoral process, and overall makes politicians more accountable to the wider polity of a country, with the potential for constructing a virtuous cycle of implementation, political feedback from the parties involved and consequential improvements and changes to the proposals at hand. However, while most of the literature focuses on the channel going from the constitutional design and policy process pertinent to a democracy and the policies for health, two important notes need to be made: (i) while health care access and the level of spending in public health have been seen as the main mediators from democracy to health, there are certainly other important mediating factors related to the psychosocial benefits of living in a democracy; (ii) not all democracies are the same, as different chapters in this book illustrate. It is possible that formal democracies fail to deliver effective and beneficial health policies, if any. The presence of informational limits, veto players, regulatory capture and rent-seeking from powerful interest groups in health care can give rise to dysfunctional outcomes which reduce, to some extent, the advantage of democracies compared to stable autocratic regimes, based on a top-down system of command and control (Geloso et al., 2020). For an up-to-date review of literature within the public choice tradition and government failures in public health policy making, see Leeson and Thompson (2021). 2.4

Cultural Norms and Values

Democracies typically encompass more different values and cultural norms than non-democracies. This is important insofar as values play the role of informal institutions which are intertwined with the efficient functioning of democratic institutions (Besley & Persson, 2019). Gorodnichenko and Roland (2017) show evidence that individualistic (as opposed to collectivistic) values are more akin to democracies. This explains why the values of tolerance and preference for diversity are more common in democracies that in autocracies. Experienced tolerance and autonomy might exert health returns in early life, and especially on the mental health and stress of populations. Individuals’ agency, as Benz and Frey (2008) illustrate, can be essential for individuals’ well-being.

30  Handbook on the political economy of health systems 2.5

Civil Society

Typically, democracies exhibit larger and more diverse civil society (see Chapter 12, which discusses, for example, the role of social capital), namely an heterogenous range of groups and institutions that are independent of the state, including the mass media, non-governmental organizations, think tanks, and research and education units, among others. Such institutions tend to surveil the way power is used, reporting abuses and typically taming and exposing corrupt practices (see Chapter 15). This is especially important in health care, where health systems often establish close links with the industry (see Chapter 14).

3.

DEMOCRACY AND MORTALITY

Past reviews of the literature have pointed out a burgeoning literature from different fields in the social sciences, using heterogenous techniques of analysis, data and identification strategies. Klomp and de Haan (2009) and Muntaner et al. (2011) provided early reviews of the literature on this topic. More recently, Chapter 5 in Costa-Font et al. (2020), and McGuire (2020), provided further and comprehensive updates on the literature. Both reviews suggest that, with some exceptions, there is a positive and robust conditional correlation, even causality, between democratization and measures of health status. Typically, the different studies employ measures of life expectancy and child mortality. While tracing a neat separation between early and recent studies is always subject to some arbitrary choice, we try here to give some guidance on the early features and on how these have been further explored according to the dimensions of the early research. Such early research focused on: (i)

Mostly using one measure for democracy. Recent trends point towards using multiple definitions and data sources for the assignment of a given country’s democratic status, even using and combining sources to reach less noisy measures of democracy, according to different methods. (ii) Investigating mechanisms in terms of moderating factors (distal vs proximal), but also trying to decompose the democracy idea into components. Further, investigating in terms of mediation and pathways from democracy to health, where early studies pointed mostly to the implementation of health policies covering the poorer sectors of the population, the health of children and infants, and also looking at alternative and more direct pathways; for example, the effect of diminishing psychosocial stressors within a political system that improves the agency of the citizens through the promotion of the accountability, representation and selection of policy makers. (iii) Given the observational nature of the data, refining and proposing alternative identification strategies aimed at justifying the causal interpretation of the regressions’ coefficients. (iv) Expanding on the use of dependent variables for health outcomes, from child/infant mortality and life expectancy at birth, towards cause-specific mortality rates and life expectancy at different ages, using biomarkers like height and weight, as we are going to discuss in the next section.

Do democratic regimes exhibit ‘better’ health outcomes?  31 Regarding this last point, recent literature has mainly focused on evaluating the impact of democratization on alternative measures of health. Below, we expand on exploring the contributions in the last two years, which involves literature exploring the use of human height as a retrospective marker of health fit and the role of political regimes in determining heights. However, while many aspects of this nexus have been unpacked and explored in depth, our aim with this contribution is to indicate future paths of research on the topic and offer possible venues for additional investigations within the democracy–health nexus.

4.

DEMOCRACY AND HUMAN HEIGHT

Most of the literature on democracy and height has focused on using infant/child mortality or life expectancy at birth. A few exceptions looking at measures of adult health do exist; recently Bollyky et al. (2019) explore life expectancy at 15 and for several communicable and non-communicable diseases for a panel of 170 countries from 1970/80 to 2016, given that non-communicable diseases (like cardio, chronic conditions and cancer, among others) are likely the area of greater interest for adult health, at least before the Covid-19 pandemic. More recently, it has focused on a measure of health, namely adult height, which has become a yardstick in evaluating standards of living and well-being. Heights have been used as a key marker of the impressive ‘technophysio evolution’ recorded in the last 150 years (Fogel & Costa, 1997). This rapid and non-genetical evolution stems from technological advances which granted a historically unprecedented control over the environment, and which allowed a drastic increase in biological performance in terms of body size and longevity in a relatively (for genetics) short amount of time. An important mediator had been the improvement in nutrition and standards of living, leading to large improvements in health biometrics in several populations around the world. Adult heights have thus become an established measure of the ‘biological standard of living’ (Komlos & Snowdon, 2005) and ‘retrospective measure of well-being’ (Silventoinen et al., 2000; Steckel, 1995, 2009). ‘Height plasticity’ before adulthood follows a common pattern, with a peak of height velocity during childhood and a second smaller peak during adolescence. After those phases, and in the absence of specific adverse health conditions, adult height is formed and can be considered fixed before shrinking starts after about 50 years of age (Beard & Blaser, 2002). Second, it has been found that the difference between genetically (potentially) predetermined and actual height is caused by psychosocial and environmental stressors occurring before adulthood (Eveleth et al., 1976; Stinson, 1985). Moreover, existing research estimates about 20 percent of height variability depends on environmental factors (Stunkard et al., 1986; Silventoinen et al., 2000), to which male height seems to be particularly reactive (Stinson, 1985). Hence, the height an individual achieves at the age of 20 is considered an established retrospective marker of human health and welfare during childhood, and, to a lesser extent, during adolescence (Tanner 1987; Steckel, 1995, 2009). Using heights has its own advantages by both allowing one to capture the wider contribution of institutional transitions to health before adult age and lending results that are easier to interpret and compare. First, the use of heights complements other standard measures of infant/childhood mortality and captures the wider health status of those who survive infancy and childhood and marks the health gains accrued until the beginning of adulthood. Second, using heights also has an advantage over life expectancy at birth, which reflects health shocks distributed over the

32  Handbook on the political economy of health systems whole lifespan and whose interpretation might not be easy (Batinti et al., 2022). Moreover, in cross-country studies where there is a large portion of low- and middle-income countries, most of the gains in life expectancy at birth are just reflected in the declining mortality rates for children and infants, with the result that the joint use of both measures can hardly be interpreted as independent findings or robustness tests. Third, heights measures tend also to be constant across countries and over time, unlike the incidence of specific diseases or causes of death, which are subject to revisions that might alter previous empirical results (McGuire, 2020). Finally, this means also that the timing of democratic reforms can be linked more tightly to a critical period when health is most influenced, unlike the other measures of well-being (Komlos & Snowdon, 2005), given that the distribution of height plasticity along the pre-adult life cycle follows precise patterns with some differences in the timing between males and females. An early study using a cross-sectional sample of about 120 countries (Frongillo & Hanson, 1995) finds that anthropometric measures are mostly explained by sex, ethnicity, food security, maternal and child care, health services and environment, but not by political institutions and political ideology. Similar findings are in Shandra et al. (2004). However, to the extent that some of these variables are mechanisms through which political transitions affect heights, these could be considered bad controls and likely to produce spurious results. On the other hand, later research documented a relationship between institutional transitions and height. Studies on German unification (Komlos & Kriwy, 2003; Hiermeyer, 2008) documented converging heights of males in East and West Germany after the reunification; Hiermeyer (2008) finds also that the convergence has been accompanied by an increase in heights’ inequality in the former East Germany. Similarly, Pak (2004) reported a divergence in heights after the division between North and South Korea. Costa-Font and Gil (2008) have shown a height increase in Spain occurring just after the transition to democracy. One potential natural experiment lies in examining changes in democracy resulting from the transition economies. Costa-Font and Kossarova (2019) explore the heights dynamic in the Czech Republic’s and Slovakia’s breaking up from the former Czechoslovakian federation during the transition from the Soviet bloc. Overall, the transition is accompanied by an increase in population stature in both the newly formed countries. Consistently with Stinson (1985), the effect is found among men and not women. More recently, Adserà et al. (2019) use a panel of countries transitioning from communism. They find shorter heights for those born right around the years of the transition from the Soviet Union; however, those losses have been followed by a rapid recovery regarding those born a few years after the transition. Marein (2020) studies heights of adults born during and a few years after the 1898 annexation of Puerto Rica by the USA. Again, it is found that the rate of increase in male height in Puerto Rico has been more than twice the average for Latin America and the Caribbean. Taking a historical perspective, Batinti et al. (2022) study the effect of franchise extensions on heights, extending the work of Hatton (2014) and Hatton and Bray (2010) on historical males’ heights for 15 European countries for the period 1860–1980. They found that the waves of democratizations increased height about by 0.7 cm. Moreover, the inclusion of a female franchise accounts for an even larger effect, to 1 cm. Finally, reduced inequality and increased expenditure on social and health services are the main mediators. Testing of mechanisms shows that a reduction in inequality of income distribution (which might be related to a more equal distribution of nutrition) and access to universal health care are the two most important mechanisms through which democracy might have affected heights. There is also evidence of

Do democratic regimes exhibit ‘better’ health outcomes?  33 an increasing effect when measures of democratic capital are used. Finally, results are robust in a battery of tests including instrumental variable (IV) regressions, event and sensitivity studies, and placebo treatments. Recently, Batinti and Costa-Font (2022) use individual data on self-reported stature. The estimates reported suggest that democracy increases average male stature, but has no effect on female stature, and estimates vary from a minimum increase of 1.33 cm to a maximum of 2.4 cm. Moreover, the paper focus on the additional height premium from being exposed during adolescence. Also, in this case, a positive association is found when a full battery of democracy scores increases from childhood to adolescence. Consistently with Batinti et al. (2022), they find a robust correlation between the democratic status of a country and male individual heights.2 The effects are mainly resulting from removing the psychosocial stressors which affect males but are not found to be significant for females’ heights. Batinti et al.’s paper also offers new insight and tests regarding the additional effect on height during adolescence, finding evidence of a second peak exactly around the adolescent years (earlier for females, later for males). Consistently with the dynamics of height velocity, the tests find an additional 40 to 50 percent increase in the estimates when the adolescent age effect is considered. This might account for the differences between the infant/childhood and pre-adult overall effect which is measured by heights, and which also includes adolescence and the post-survival additional effects. These results thus add an interesting insight when related to the psychosocial pathways from democracy to health; in fact, it is plausible to assume that during adolescence individuals (especially male individuals) are progressively more exposed to the direct psychosocial stressors which are less mediated by the family environment, producing additional evidence for a direct effect of democracy on health. Batinti et al. (2022) find that most of the effect is concentrated in the lower deciles of the height distribution, suggesting that democratization both increases average height and reduces height inequality for males. This reduction is consistent with the overall increase in average heights for males if shorter heights are the most affected by democratization and by the improvements in standards of living and nutrition that are promoted by democratization through the channel of increasing growth and development (Acemoglu et al., 2019). Further studies might focus more on the link between stress and bones’ growth, providing even more proximal mechanisms joining democracy to bone growth. Future research could focus more on how psychosocial stressors impacted by a democratic society do in turn influence the human biology of the population; in other terms, how social stressors trigger physiological reactions, which can be detrimental for bones’ growth and health in childhood and adolescence. See, for example, experimental and interdisciplinary evidence on mice from Wippert et al. (2017).

5.

ISSUES FOR FURTHER STUDY

5.1

Causality and Correlation

Possibly one of the important divisions in the literature is whether studies do attempt to retrieve correlations, or do they exploit some form of exogenous variation in democracy exposure. Hypotheses in social sciences, to mention some, see democratic transitions as the result of deeper historical, social and psychological forces which have gradually modified individual

34  Handbook on the political economy of health systems attitudes, cultural traits and both informal and informal institutions, building the prerequisites for a democratic transition. It is thus very likely that some of these factors might have also influenced the health state of a population, together with favoring the transition towards a more democratic political regime. For example: According to the modernization theory (Lipset, 1959), democracy is the result of social and economic progress, which in turn increased the demand for civil liberties and transitions from the de facto power of rising classes towards a more established de jure power regarding the new emerging sectors of the population; for example, with the extension of the franchise. (ii) The middle-class hypothesis: connected to modernization, the idea is that both economic and social progress, and also the distribution of resources and the individual attitudes of the middle class towards more democratic regimes, increased the demand for democratic transitions. (iii) According to the Revolutionary Threat hypothesis (Robinson & Acemoglu, 2012), when autocratic regimes experience a too-inequal distribution of resources, lower classes can organize and threaten a revolution. If the threat is credible, dominant elites might choose to extend the franchise instead of facing the possible upheavals brought about by a revolution. (iv) In Lizzeri and Persico (2004), the progressive inclusion of disenfranchised segments of the population is also the result of a calculation of the elites to reshuffle the articulation of political power. (v) Parasite stress hypothesis (Schaller & Murray, 2011): the idea that parasites’ prevalence and cross-sectional variability is an indicator, and likely a cause, of individual attitudes towards individualism, in-group vs out-group decision making and personality traits, which in turn condition the preferred political system as more authoritarian instead of more democratic. (i)

Most studies establish an overall positive association between democracy and height, but is it causal? This is the same question advanced when more traditional measures are used. For example, standard instruments for democratization are geographical (Acemoglu et al., 2019; Persson & Tabellini, 2012), but certainly geographical instruments might have pitfalls to correct for. In one of the robustness tests, Batinti et al. (2022) propose as an instrument the decolonization process involving European countries between the nineteenth and twentieth centuries. Other papers use Difference in Difference (Adserà et al., 2019), but there is certainly space to adopt other techniques like Synthetic Control Methods (as done by Bollyky et al., 2019) for finding geographical or other types of discontinuities. However, democratization remains essentially a process affected by reverse causality and confounders, only to cite some endogeneity issues. 5.2

Democracy and Health Inequality

Democracy might have heterogeneous effects on individuals’ socioeconomic position and gender, which suggests that it can affect health inequalities in addition to health outcomes. For instance, if there is an overall reduction in nutrition quality, the negative effects on heights are more prominent for poor people (the majority) than wealthy ones, with the result of a more than proportional reduction in heights for the poorer classes. At the same time, the overall

Do democratic regimes exhibit ‘better’ health outcomes?  35 effect will be exacerbated if nutrition is already quite unequally distributed to begin with. Moreover, as income levels increase and with them the baselines for having adequate nutrition, then income levels are protective against major deprivation and severe health and well-being effects. If democracy fosters economic development (Robinson & Acemoglu, 2012), the amelioration of the standards of living and nutrition from economic growth (McKeown et al., 1972) increases height levels and reduces inequality if a substantial portion of the additional wealth created is distributed to the poorer segments of the population. A similar effect could be induced by the removal of psychosocial stressors. In this case, however, research (see above) reveals that there is a gender differential, with the male population benefiting more than the female, with the result of increased gender dimorphism, which is a form of inequality. This, however, is often found to be an overall improvement in the population, with females being not affected but not negatively affected. Thus, democracy can influence the priority of reducing disparities in health between the average individual in a country, neglected populations when they are politically mobilized and individuals with poorer health (e.g., individuals that suffer from chronic conditions or disability). Some of this literature examines changes in individual-specific variables such as socioeconomic status (e.g., income, education, occupation), indicating that changes in the way income is distributed or the educational attainment of the population would produce important changes in the distribution of health outcomes (Winkleby et al., 1990). Nonetheless, democracy might also lead to a ‘health inequality trap’ where democratic decisions fail to adequately address the needs of minorities and are captured by dominant elites instead (Powell-Jackson et al., 2011). In less socially cohesive societies, pressure by interest groups might not be interested in policy responses leading to growing, or at least stagnating, inequalities in health (Powell-Jackson et al., 2011). 5.3

Democracy and the Well-Being of the Elderly

Democracies can influence life expectancy, which suggests that they deliver some improvement in innovation that can expand the lifespan of individuals, with the result of increasing the proportion of the elderly in the population. As development proceeds, the focus shifts gradually from child mortality to the standard of living of those in an advanced stage of life. This might as a result give rise to a higher proportion of older-aged people in a society. Hence, democracy can be instrumental for the aging of populations, in terms of improving life expectancy after 50, and reducing the mortality and morbidity of cardiovascular disease, cancer, chronicity and other non-communicable diseases which are the main threats to health and health quality later in life. This might also result in slowing down the natural shrinking process after 50. Clearly the Covid pandemic shifted the narrative, which was previously centered on the double disease transition, and brought back to the fore the role of communicable disease in influencing standards of living in more developed, high-income economies. 5.4

Democracy and Mental Health

While the main and more consistent mechanism across the literature mechanisms from democracy to health is an increase in social welfare, health spending and access to health care resources, democracies can influence the stability of the social environment, and reduce the stress where individuals operate. As a result, this increases the priority that mental health

36  Handbook on the political economy of health systems has as a policy area compared to other competing areas. Recent research is more and more interested in evaluating the role of psychosocial stressors as a mediator between institutions and health. 5.5

Parasite Stress Hypothesis

According to the parasite stress hypothesis (Schaller and Murray, 2011), regional differences in pathogens’ prevalence affects cross-cultural differences and attitudes towards the preferred political regime. Higher pathogen prevalence is associated with, and possibly causes, changes in political attitudes and support towards a more authoritarian regime. Thus, there is a reverse channel from communicable disease to political regime, a channel mediated by cultural adaptation also involving political attitudes and views.

6.

DEMOCRACY AND COMMUNICABLE DISEASES

Most democracy studies in the literature refer to its effect over non-communicable health inputs which can be influenced via reducing stress and improving the access to health care. However, one of the important questions in examining the heterogeneous effects of democracies on outcomes can be examined in the context of unexpected health shocks, where government must make quick decisions. Is democracy helpful for making the quick decisions needed? Does democracy, by prioritizing the health of the population above other goals, help to legitimize lockdowns and other policy restrictions? 6.1

Political Regimes, Covid-19 Policies and Health Outcomes

Related to the above point, there is research on the political economy of pandemics looking at: how autocracies and democracies reacted to the diffusion of Covid-19, and how the policy implemented and the virus itself will affect the heights of adults-to-be a few years from now; how future heights are mediated by the lockdown and restriction policies enacted by democracies and autocracies; how democracies experience different trade-offs between lives and livelihoods in ordinary times than autocracies, and how this trade-off is impacted during major health outbreaks like pandemics; and finally, how this trade-off changes when moving the analysis from the short- to long-term perspective. We will see that the causal pathways are multiple, and not all of them go towards the same direction, which makes the long-term consequences of pandemics in democracies and autocracies complex to foresee in obvious ways. 6.2

Democracies and Fatal Trade-Offs

Democracies can influence important fatal trade-off decisions in the context of pandemics. Indeed, Troesken (2015) explains how democracies face more constraints when facing communicable disease in terms of preserving individual liberties. Central to the story is the increasing urbanization which weakened social control in small towns. In small towns anti-pandemic remedies are enforced by moral suasions, but in urban settings there is a need for more coercion and enforcement on the part of the state as there is less social peer pressure and control. At the same time, individualistic societies are less willing to obey government restrictions that

Do democratic regimes exhibit ‘better’ health outcomes?  37 are not shared, and so feel imposed and limiting regarding individual liberties. In this light, the spread of scientific knowledge, mostly promoted in liberal democracies at the beginning of the twentieth century, favored the transition from communicable to non-communicable disease, and to some extent this favored keeping, and perhaps enhancing, individual liberties, and had a relevant self-sustaining role for democratic rule. Recent research from Koyama (2021) and Desierto and Koyama (2020) elaborates on the lives–livelihoods trade-off during pandemics from a public choice perspective of public health interventions. Government reactions to the spread of pandemic disease are determined by two important constraints: incentives’ alignment among the policy makers, and informational limits on what is the best policy to be adopted.3 Public health interventions are often justified based on correcting the ‘negative externalities.’ Economic theory suggests that these externalities can be corrected by market adjustments and expansions, or by private bargaining within social norms. However, in the case of the wide spread of novel pathogens (pandemics), there are also large informational limits at the beginning, especially the knowledge of the transmissibility and harm of the virus, which can make the computation of the externalities challenging. So pandemic measures are to be taken under risk, but most of the time under uncertainty, which makes the scope and severity of the potential externalities difficult to assess. In addition, it is important to consider the impact on the health systems, which might get crowded out and transmit additional negative externalities towards other non-infected patients in terms of lower quality and longer waiting, all the way to exclusion from access. There is also a trade-off between limiting ‘negative global externalities’ and the governance problems that arise in terms of cooperation and failure to achieve global collective action, even the coordination of governments and agencies, and problems of asymmetric information, which can be crucial during phases of global uncertainty. 6.3

Democracies, Incentives Alignment and Information Constraints

Health decisions in a pandemic depends on the political context and political institutions, including who are the constituents of reference (winning coalitions), what are the personal incentives of politicians and bureaucrats, how they interact within the context of rules for decision making laid out by legislation, and their informational constraints characterized by the nature and structure of the knowledge they own, if decisions must be made within deterministic, probabilistic or uncertain contexts. All the aforementioned characterizes the opportunity costs perceived by each party and the resulting trade-offs. Then it is a matter of understanding what the salient ones are. Note that the personal incentives can naturally also be shaped by honest and genuine interest to try to do what is best for the people, but this is one variable among others and might not be the crucial one in many contexts. This might be because of the presence of competing incentives, and, even when incentives are aligned, the knowledge problem implies that not all the possible ramifications of the consequences of the policy choices can be predetermined, and there are often unintended consequences players need to deal with. According to Koyama (2021), historically, public health initiatives became more prominent as scientific progress clarified the nature and transmission mechanisms of pathogens among the population. When the knowledge problem was sensibly reduced and there was a clear scientific basis on which to fight the spread of contagious disease, policy was effective to a great extent. In other terms, and by introducing a temporal perspective, autocratic regimes score better in a static setting. Autocracies dispose of more effective top-down decision making and

38  Handbook on the political economy of health systems face less unhappiness from their population if polities are culturally and politically more oriented towards authoritarianism and collectivism. But dynamically, democratic/liberal regimes improve the diffusion of scientific knowledge and promote more prosperous societies, to abate and win against parasitic diseases through the discovery of new drugs and vaccinations which require much less sacrifice of individual liberties, shifting from preventative to curative measures. For example, one can still get Covid-19 after being vaccinated, but the vaccine acts as an ex ante curative medicine for protection from the symptoms, and literally, it does not prevent being infected and even experiencing possibly milder symptoms. So, in the long run, the trade-off between liberty and health safety is likely to be reversed. Moreover, the intensity of the trade-off depends on social norms and social capital which moderates the relationship between politicians and the governed. Analyzing the population mobility in Italy after the country’s very first Covid outbreak, Durante et al. (2021) find that, before the introduction of a mandatory national lockdown, provinces of Italy with higher social capital experienced larger reductions in population mobility even in the absence of any legal obligation to do so. This is in part explained as social capital and is related to civic responsibility, but even in the case of national obligations, regions with higher social capital might trust public authorities and that the limitations put in place are adequate, or it might also involve the trust that citizens have that those reductions in mobility will be withdrawn and individual liberties restored as soon as the dangers of harmful contagion are removed. Desierto and Koyama (2020) propose a static model explaining anti-Covid policy making from a political economic perspective based on the selectorate model. Each regime faces a trade-off between lives and livelihoods, but the size of this trade-off is determined by the dimension and composition of the winning coalition of a ruler, which might or might not reflect the trade-off regarding what would be best for the overall population. If the winning coalition has more interest in preserving health than the economy, stricter measures will be tolerated; vice versa, lockdowns and stringent policies will be seen with scorn and lightly implemented by the rulers so to maintain support and political survival. Early empirical evidence on how autocracies or democracies reacted to the pandemics is mixed. For example, Frey et al. (2020) do not find an autocratic advantage, but say that both types of regimes reacted with pros and cons. Autocratic regimes implemented stricter lockdown policies but, perhaps more interesting, have not been as successful in restraining the mobility of citizens as democracies. So, they might not have been effective in implementing the restrictions. Thus, democracies were more successful, overall, in restricting people’s mobility. Is it instead found that stringency was more effective in more collectivistic societies than individualistic ones? The mix of more collectivist and democratic countries produced the most effective responses to Covid-19 in terms of reducing geographic mobility. But cultural traits, social capital and forms of governments are very likely to be correlated, with democracies being more likely tilted towards individualism, and autocracies towards collectivism. Similar mixed evidence is presented in Karabulut et al. (2021), where it is found that while democracies experienced higher infection rates, they also had lower death rates from Covid. Overall, empirical research is still at an early stage and as more and better data becomes available, identification designs will be refined, and further mediating and moderating factors will be tested, allowing further exploration and understanding. For example, one factor that could be included in the estimates is the positive externality effects of having the most effective vaccines now available produced in democracies, in a relatively short term of 8–10 months after the declaration of the global pandemic, which should be included in the count of the

Do democratic regimes exhibit ‘better’ health outcomes?  39 democratic response to pandemics. It could also be said that part of this innovative effort was motivated by paying more attention to individual liberties than collective security through the control of the population’s mobility. Of course, this is ongoing research, which will be further explored after the end of the pandemic, to explore both the short- and long-term effects of Covid and the differential effects that the pandemic had on different types of political regimes.

7. CONCLUSION This chapter has examined the effect democracies play in influencing health outcomes. We have discussed the different mechanisms at play, influencing effects via institutional change and policies, and environmental and cultural effects that democracies entail. While evidence indicates a positive correlation between democracy and health, studies that attempt to retrieve causal effects are more limited and exhibit some discrepancies depending on the specific form of variation examined and, in particular, the measure of health outcomes such as mortality and height. More importantly, democracies seem to influence different forms of health inequalities, and effects differ by gender and socioeconomic status; in some cases, they can give rise to health inequality traps when a majority imposes their health policies on the rest of the population. Furthermore, democracies can exert a differential influence on old-age populations, and on the mental and long-term health of a society by affecting the social environment and the culture individuals live in. Finally, we have provided an assessment of the effect of democracies in confronting communicable diseases such as Covid-19, and more specifically we document that although the evidence suggests that autocracies are stricter in their mobility restrictions, they do not necessarily show better outcomes. Taken together, the tentative evidence suggests that the effect of democracy on health depends on the role of different mechanisms at play.

NOTES 1. Some of these dimension that define democracies can be identified and countries can be classified under different indexes, such as the Polity IV, Freedom House or VDem. 2. Regressions control for a battery of individual- and country-level controls, and are robust to several specifications and measures of democracy from four different sources (Polity, BMR, V-Dem, Machine Learning Democratic Indexes), and different construction of the variables, both multicategories indicators or dummies. 3. When government is putting in place public health policies, it faces similar constraints to market players. These constraints are based on (i) incentives’ alignment among agents and between principals and agents (incentives problem), which becomes more problematic as the number of agents and principals increases; and (ii) informational constraints – i.e., choices must be made in the contexts of risks and uncertainty.

REFERENCES Acemoglu, D., Naidu, S., Restrepo, P., & Robinson, J.A. (2019). Democracy does cause growth. Journal of Political Economy, 127(1), 47–100. Adserà, A., Pozza, F.D., Guriev, S., Kleine-Rueschkamp, L., & Nikolova, E. (2019). Transition from plan to market, height and well-being. IZA Institute of Labor Economics Discussion Paper No. 12658.

40  Handbook on the political economy of health systems Batinti, A., & Costa-Font, J. (2022). Does democracy make taller men? Cross-country European evidence. Economics and Human Biology, 45, 101117. Batinti, A., Costa‐Font, J., & Hatton, T.J. (2022). Voting up? the effects of democracy and franchise extension on human stature. Economica, 89(353), 161–90. Beard, A.S., & Blaser, M.J. (2002). The ecology of height: The effect of microbial transmission on human height. Perspectives in Biology and Medicine, 45(4), 475–98. Benz, M., & Frey, B.S. (2008). Being independent is a great thing: Subjective evaluations of self‐ employment and hierarchy. Economica, 75(298), 362–83. Besley, T. (2006). Principled Agents? The Political Economy of Good Government. Oxford University Press on Demand. Besley, T., & Kudamatsu, M. (2006). Health and democracy. American Economic Review, 96(2), 313–8. Besley, T., & Persson, T. (2019). Democratic values and institutions. American Economic Review: Insights, 1(1), 59–76. Bollyky, T.J., Templin, T., Cohen, M., Schoder, D., Dieleman, J.L., & Wigley, S. (2019). The relationships between democratic experience, adult health, and cause-specific mortality in 170 countries between 1980 and 2016: An observational analysis. The Lancet, 393(10181), 1628–40. Costa-Font, J., & Gil, J. (2008). Generational effects and gender height dimorphism in contemporary Spain. Economics & Human Biology, 6(1), 1–18. Costa-Font, J., & Kossarova, L. (2019). Transitions that matter: Czechoslovakia’s break up and human stature. International Journal of Environmental Research and Public Health, 16(24), 5050. Costa-Font, J., Turati, G., & Batinti, A. (2020). The Political Economy of Health and Healthcare: The Rise of the Patient Citizen. Cambridge University Press. Desierto, D., & Koyama, M. (2020). Health vs. economy: Politically optimal pandemic policy. Journal of Political Institutions and Political Economy, 1(4), 645–69. Downs, A. (1957). An Economic Theory of Democracy. Harper & Row. Durante, R., Guiso, L., & Gulino, G. (2021). Asocial capital: Civic culture and social distancing during COVID-19. Journal of Public Economics, 194, 104342. Eveleth, P.B., Tanner, J.M., Eveleth, P.B., Tanner, J.M., & Chang, W.H. (1976). Worldwide Variation in Human Growth (Vol. 8). Cambridge: CUP Archive. Fogel, R.W., & Costa, D.L. (1997). A theory of technophysio evolution, with some implications for forecasting population, health care costs, and pension costs. Demography, 34(1), 49–66. Frey, C.B., Chen, C., & Presidente, G. (2020). Democracy, culture, and contagion: Political regimes and countries responsiveness to Covid-19. Covid Economics, 18, 1–20. Frongillo, E.A., & Hanson, K.M.P. (1995). Determinants of variability among nations in child growth. Annals of Human Biology, 22(5), 395–411. Geloso, V., Berdine, G., & Powell, B. (2020). Making sense of dictatorships and health outcomes. BMJ Global Health, 5(5), e002542. Gorodnichenko, Y., & Roland, G. (2017). Culture, institutions, and the wealth of nations. Review of Economics and Statistics, 99(3), 402–16. Hatton, T.J. (2014). How have Europeans grown so tall? Oxford Economic Papers, 66(2), 349–72. Hatton, T.J., & Bray E. (2010). Long run trends in the heights of European men, 19th and 20th centuries. Economics and Human Biology, 8(3), 405–13. Hiermeyer, M. (2008). The trade-off between a high and an equal biological standard of living: Evidence from Germany. Economics & Human Biology, 6(3), 431–45. Karabulut, G., Zimmermann, K.F., Bilgin, M.H., & Doker, A.C. (2021). Democracy and COVID-19 outcomes. Economics Letters, 203, 109840. Klomp, J., & de Haan, J. (2009). Is the political system really related to health? Social Science & Medicine, 69(1), 36–46. Komlos, J., & Kriwy, P. (2003). The biological standard of living in the two Germanies. German Economic Review, 4(4), 459–73. Komlos, J., & Snowdon, B. (2005). Measures of progress and other tall stories. World Economics, 6(2), 87–135. Koyama, M. (2021). Epidemic disease and the state: Is there a tradeoff between public health and liberty? Public Choice, 195, 145–67. https://​doi​.org/​10​.1007/​s11127​-021​-00944​-4.

Do democratic regimes exhibit ‘better’ health outcomes?  41 Leeson, P.T., & Thompson, H.A. (2021). Public choice and public health. Public Choice, 195, 5–41. https://​doi​.org/​10​.1007/​s11127​-021​-00900​-2. Lipset, S.M. (1959). Some social requisites of democracy: Economic development and political legitimacy. American Political Science Review, 53(1), 69–105. Lizzeri, A., & Persico, N. (2004). Why did the elites extend the suffrage? Democracy and the scope of government, with an application to Britain’s ‘Age of Reform’. Quarterly Journal of Economics, 119(2), 707–65. Marein, B. (2020). Economic development in Puerto Rico after US annexation: Anthropometric evidence. Economics & Human Biology, 38, 100892. McGuire, J.W. (2020). Democracy and Population Health. Cambridge: Cambridge University Press. McKeown, T., Brown, R.G., & Record, R.G. (1972). An interpretation of the modern rise of population in Europe. Population Studies, 26(3), 345–82. Muntaner, C., Borrell, C., Ng, E., Chung, H., Espelt, A., Rodriguez‐Sanz, M., … & O’Campo, P. (2011). Politics, welfare regimes, and population health: Controversies and evidence. Sociology of Health & Illness, 33(6), 946–64. Pak, S. (2004). The biological standard of living in the two Koreas. Economics & Human Biology, 2(3), 511–21. Persson, T., & Tabellini, G. (2012). Macroeconomic Policy, Credibility and Politics. Abingdon: Routledge. Powell-Jackson, T., Basu, S., Balabanova, D., McKee, M., & Stuckler, D. (2011). Democracy and growth in divided societies: A health-inequality trap? Social Science & Medicine, 73(1), 33–41. Robinson, J.A., & Acemoglu, D. (2012). Why Nations Fail: The Origins of Power, Prosperity and Poverty. London: Profile. Rodrik, D. (1999). Democracies pay higher wages. Quarterly Journal of Economics, 114(3), 707–38. Schaller, M., & Murray, D.R. (2011). Infectious disease and the creation of culture. Advances in Culture and Psychology, 1, 99. Schumpeter, J. (1947). Capitalism, Socialism and Democracy. New York: Harper and Brothers. Shandra, J.M., Nobles, J., London, B., & Williamson, J.B. (2004). Dependency, democracy, and infant mortality: A quantitative, cross-national analysis of less developed countries. Social Science & Medicine, 59(2), 321–33. Silventoinen, K., Kaprio, J., Lahelma, E., & Koskenvuo, M. (2000). Relative effect of genetic and environmental factors on body height: Differences across birth cohorts among Finnish men and women. American Journal of Public Health, 90(4), 627. Steckel, R.H. (1995). Stature and the standard of living. Journal of Economic Literature, 33(4), 1903–40. Steckel, R.H. (2009). Heights and human welfare: Recent developments and new directions. Explorations in Economic History, 46(1), 1–23. Stinson, S. (1985). Sex differences in environmental sensitivity during growth and development. American Journal of Physical Anthropology, 28(S6), 123–47. Stunkard, A.J., Sørensen, T.I., Hanis, C., Teasdale, T.W., Chakraborty, R., Schull, W.J., & Schulsinger, F. (1986). An adoption study of human obesity. New England Journal of Medicine, 314(4), 193–8. Tanner, J.M. (1987). Growth as a mirror of the condition of society: Secular trends and class distinctions. Pediatrics International, 29(1), 96–103. Troesken, W. (2015). The pox of liberty. In The Pox of Liberty. Chicago: University of Chicago Press. Winkleby, M.A., Fortmann, S.P., & Barrett, D.C. 1990. Social class disparities in risk factors for disease: Eight-year prevalence patterns by level of education. Preventive Medicine 19(1): 1–12. Wippert, P.M., Rector, M., Kuhn, G., & Wuertz-Kozak, K. (2017). Stress and alterations in bones: an interdisciplinary perspective. Frontiers in Endocrinology, 8, 96.

4. Soviet communism and later-life health and health care Joan Costa-Font and Anna Nicińska1

1. INTRODUCTION Striking differences in crude health indicators between post-communist countries and Western Europe still prevail, 30 years from the dissolution of the Soviet Union (cf. Figure 4.1). The differences also concern health-related behaviors (including but not limited to alcohol consumption) alongside health policy institutions. Indeed, communist health care systems were characterized by universal coverage of (poor-quality) care that was inefficiently organized and exerted limited responsiveness to patient needs. This chapter provides an overview of the institutional legacies of Soviet communism in health systems. Firstly, we examine the effects of the exposure to Soviet communism, controlling for other relevant factors. Secondly, we discuss possible mechanisms responsible for these effects that were pivotal for Soviet regimes. We include in the analysis Central and Eastern European countries from the Soviet bloc (being under the political influence of or belonging to the Union of Soviet Socialist Republics (USSR), or Soviet Union, in the previous century) and Asian USSR republics, if data are available. The Asian republics varied deeply with respect to cultural and historical heritage as well as pre-existing economic ties with Europe, which should be kept in mind when comparing communist institutions between European and Asian parts of the Soviet bloc. Furthermore, Poland, the Czech Republic and Slovakia were historically treated as Central rather than Eastern Europe (Orenstein, 2001), and the ‘in-between’ identity seems to be the best description of these countries, belonging neither to the European Union nor the Soviet Union in the Cold War era. During the Soviet period, the health care system was organized around the premises of Nikolai Semashko, People’s Commissar of Public Health between 1918 and 1930. The health care system was state financed and provided universal coverage under a multi-tier model coordinated by a district physician. Formally, there was no private health insurance nor private hospitals, and Semashko’s health care system operated in all Soviet countries. However, certain differences occurred, especially as far as the informal institutions are concerned, including the prevalence of private medical practices and informal access to services. Relatively developed infrastructure and human resources in the Soviet health system in the 1950s deteriorated over time and underwent a major transition in the 1990s that led to various structures of mixed private and public health care service providers and funding. Although the exposure to communism entails a wide range of interactions with formal and informal institutions, in this chapter we focus on two major centrally planned elements of Soviet regimes, because they vary deeply from other regimes, especially those of the Global West and North, and as such they might be driving the specific communist repercussions on health and health production. These are the health care system (both its formal and informal parts) and the education system. The role of the former under communism, and during the 42

Soviet communism and later-life health and health care  43

Source: Work Bank, map by Fede-5-19 – Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index. php?curid=83639711.

Figure 4.1

Life expectancy at birth in European countries

transformation to the free market, is relatively well examined, while for the latter many questions remain open. To what extent well-documented knowledge on the effects of health care and education systems in Western countries is representative for their communist counterparts is an important policy question considering the welfare state and market failures in post-communist regions. Specifically, whether the same mechanisms contribute to the universal phenomenon of positive correlation between socio-economic status (SES) and health, often referred to simply as ‘the gradient,’ in various institutional environments is unclear (Galama et al., 2018; Lleras-Muney, 2022). Moreover, SES is most often defined as a combined measure of individual education, occupation and income. Education seems to be responsible for most of the channels through which SES might affect health (Grossman, 2005; Kaestner, 2013), because it largely affects income and occupation (and, unlike other components of SES, is observable for entire populations). However, the role of education and human capital is different in centrally planned economies than in free markets, which provides an opportunity for the investigation of the health impact of education systems. We close the chapter with conclusions and areas for further research.

44  Handbook on the political economy of health systems

2.

SOVIET COMMUNIST INSTITUTIONS AND THEIR IMPACT ON HEALTH

Soviet communism was a politico-economic regime based on the collectivist planning of human needs, introduced first in 1922 in the Soviet Union (USSR) with Lenin and Stalin in power. After the Second World War, the Soviet bloc of allied countries controlled by Stalin was formed (Figure 4.2). Formally, the Soviet Union was dissolved in 1991, which was preceded by three significant events of 1989: the fall of the Berlin Wall in Eastern Germany, Ceausescu’s execution in Romania and free democratic elections in Poland. Before that, the state ran by the Communist Party had control over all institutions, and citizens were subject to dispossession and political persecution for disagreeing with official Marxist–Leninist ideology, while their wealth could be and often was nationalized. Nationalized buildings were used for public purposes, adapted to serve as schools or hospitals.

Source: Wikipedia, Eastern bloc.

Figure 4.2

Regime transition in Central and Eastern Europe

The Cold War with the West was frequently used as a tool for propaganda regarding Soviet communism. Despite intense indoctrination heavily present in all public institutions, the attitudes opposing the regime, and protests, were regularly present in many countries (even leading to uprisings in East Germany, Czechoslovakia, Romania and Poland). The persistent underground resistance against the Soviet regime led to the transition into democratized political systems and free market economies. The entire process has had a profound impact on various spheres of private and public affairs, including health care and education systems. Strong evidence of the negative impact of communism on health is documented, frequently referring to comparisons between East and West Germany in the research on effects of the exposure to Soviet communism, although some methodological concerns were recently raised by Becker et al. (2020). In the 1960s life expectancy was higher in East Germany than West,

Soviet communism and later-life health and health care  45 and by the 1970s the numbers had reversed (Nolte et al., 2000). Longevity in Eastern Germany declined dramatically in the 1990s, especially for males. Small improvements took place before the second half of the 2000s, when longevity began to improve steadily. The infrastructure and human resources in the post-communist public health care sectors remained scarce, especially in Asian parts of former Soviet republics, which inhibited the speed of closing the gap in health status between formerly communist countries and developed countries. The case of East Germany has many similarities with other formerly communist countries. However, the effects of exposure to communism and the transition from a centrally planned to a free market economy are heterogeneous in the economic development of post-communist countries. In some Eastern European countries life expectancy at birth has improved after communism due to changes in diets and cardiovascular health care and behaviors (Zatonski et al., 1998). Other countries exposed to communism exhibited a reduction in life expectancy after transition due to increased alcohol consumption (Connor et al., 2004), poor diets (Stuckler et al., 2009) and a deterioration of health care systems (Popov, 2017). For example, male life expectancy in Russia dropped by 6.2 years between 1990 and 1994 (see Figure 4.1). One of the better-documented legacies of the Soviet politico-economic regime is the deterioration of generalized trust and confidence in public institutions. We observe this effect also with respect to trust in medical personnel and hospitals (Costa-Font et al., 2023), referring to the Wellcome Global Monitor (WGM) data on more than 100 countries from all continents.

Source: Photo by Alieksiej Sharovij, author’s family archives.

Figure 4.3

A medicine student in public hospital in Chelyabinsk, Russia, 2006

46  Handbook on the political economy of health systems The said mistrust is accompanied by a significantly lower assessment of the current state of health services in the formerly communist countries, which we document using the European Social Survey (ESS). On a scale from 1 to 10, health services in post-communist countries are graded at 4.03 on average, whereas in Western Europe they are 5.66 on average. One might expect that the subjective perception of health care quality be larger by 40 percent in the Western Europe results from the communist impact on health care. Interestingly, the relative difference between communist and non-communist European countries in the assessment of current health system condition is even larger than the respective difference in self-reported health (by 18 percentage points). To what degree these perceptions reflect actual differences is a question that cannot be answered without understating the impact of regime transformation on formal and informal institutions relevant to health.

3.

THE TRANSFORMATION TO FREE MARKETS AND ITS IMPACT ON HEALTH

The demise of Soviet communism entailed institutional discontinuity related to formal institutions directly linked to both health care systems and prevention programs, as well as in other domains of public life. For example, the introduction of free markets and products’ marketing resulted in an increase in tuberculosis, and in turn also lung cancer morbidity, among women targeted by tobacco advertisement companies (McKee and Fister, 2004), and the development of a smoking habit (Hinote et al., 2009). Moreover, the (hyper)inflation (of 640 percent in 1989 Poland) and structural unemployment (Rostowski et al., 1998) were a source of particular stress to many individuals in post-communist countries, whose culturally promoted tools for stress management referred often to behavioral risks such as alcohol, nicotine and sugar consumption. The deep changes in SES and relative income as well as the enormous volatility of the economic environment had a direct effect on health (Bobak et al., 2000). The sense of instability and work-related stress (Jenkins et al., 2005; Salavecz et al., 2010) was predominant, which contributed to mental and cardiovascular health deterioration. In the cultures of formerly communist countries, alcohol and nicotine consumption were widely accepted ways of dealing with stress for men and also women (Kosiński, 2005; Valuch, 2021), while eating habits were far from healthy in the Soviet bloc, where food shortages and regimentation, malnutrition and hunger did not end with the post-war period. Taking into account the fact that alcohol consumption had been a long-lasting problem in Soviet Russia, the demise of the 1985–88 Gorbachev Anti-Alcohol Campaign entailed by the transformation contributed to the major raise of male mortality in Russia (Bhattacharya et al., 2013). Between 1990 and 1994 it rose by 40 percent, which translated to a drop in life expectancy of about 6 years for men (from 1989 to 1994). The transformation process was deeply heterogeneous between formerly communist countries. The pre-existing institutions made the Soviet bloc heterogeneous from its very beginning (Grosfeld and Zhuravskaya, 2013; Becker et al., 2020). The lands of the former Russian empire, where communist ideas had been seriously discussed in elite circles and popularized among workers for decades before the revolution, varied deeply from the Prussian and Hapsburg empires, where democratic and free market institutions had been present before. Importantly, some of the republics dependent on the Soviet Union gained political independ-

Soviet communism and later-life health and health care  47 ence only in 1918 after the Treaty of Versailles. Moreover, religious denominations, having a significant impact on capital and labor relations, although predominantly Christian, differed deeply between the lands, with Orthodox, Roman Catholic and Protestant being the main rites, accompanied by specific sets of minorities (such as Muslims and Jews). Thus, the pre-existing institutions varied significantly between countries in which communism evolved or was introduced. A recent study by Lecce and Ogliari (2019) shows that these factors are relevant for the processes of institutional adaptation and transformation. In some countries, democratic and market institutions were literally a novelty (Roland, 2004). In others, the privatization that took place during the transition to free markets was in fact a re-privatization (Sachs, 1992; Balcerowicz, 1994). In most of the countries, the exit from communism entailed structural unemployment, hyperinflation and an overall turbulent environment (Guriev and Zhuravskaya, 2009) as well as the deterioration of public goods provision and scrutiny of socio-political activity during communism.

4.

FORMAL AND INFORMAL INSTITUTIONS RELEVANT FOR HEALTH

The Soviet system of health care focused initially on the control of communicable diseases (Rechel et al., 2014). Mass vaccinations and malaria surveillance, accompanied by sanitary control of water supplies, hygienic disposal of waste and sewage, as well as pasteurization of milk, were the main elements of population health policy. The funding for health was relatively low, as health was not considered capital, and the entire sector was perceived as unproductive, in contrary to heavy industries and agriculture. Deteriorating infrastructure and poor qualifications of medical workers mark the entire communist period (Feshbach and Rubin, 1991). The health care systems during the transition to capitalism deteriorated even further, because of the deep economic crisis and lack of coordination in the reforms applied to the fragmented elements of public health systems as well as unanticipated effects of the rapid and unregulated development of private health care services (Millard, 1995). The emphasis put on restructuring health care systems and gradually separating its funding from the state budget was most pronounced in the Czech Republic from the beginning of the transformation process (Dyba et al., 1994). The inefficiency of the system deepened in the 1990s (cf. Figure 4.3) despite some efforts by governments to reallocate resources, including human capital and equipment, as well as in subsidizing the public sector with the introduction of payments for medicines (Millard, 1995). As a result of the parallel presence of private and public services, private services (often delivered in public hospitals by their employees using rented public infrastructure after hours) crowded out a substantial portion of public health care, where informal payments were not only widely accepted but also openly requested. Despite a certain heterogeneity in the initial systems between post-communist countries (Marrée and Groenewegen, 1997; Romaniuk and Semigina, 2018), the quality of services and access to them became even more unequally distributed between regions, and urban and rural areas, as well as between the rich and the poor, in most of the formerly communist countries. Governments failed to plan and instill appropriate reforms in a comprehensive and accurate way and to regulate the rapidly developing private sector. A decision to promote competition within the public sector between various suppliers contributed to its deterioration (Millard,

48  Handbook on the political economy of health systems 1995). Many patients were deprived of access to general practitioners, specialized medical treatments and medicines during the first decade of the transformation. Further reforms continued over the next decades, leading to more stable systems, with a major role for the private sector, and improved supply of and access to health care – for example, with respect to treatment of cardiovascular diseases prevailing in the region (Movsisyan et al., 2020) – though in certain domains such as mental health care the deterioration continued (Mundt et al., 2012). Lack of coordination between health care and other public systems is still present in the post-communist countries, and the discussion on system efficiency continues (Kujawska, 2017). The inefficiencies due to heavily decentralized authority with limited budgetary responsibilities were the main target of reforms during the regime transformation. Although the health care insurance provided ‘universal’ coverage, the package of services was very limited, and out-of-pocket money (usually informal payments) provided the large source of health financing in certain countries. Individuals from formerly communist countries were deeply attached to the concept of free public health care and hence exhibited an unwillingness to contribute financially with direct fees for public health care; at the same time, some of them are willing to pay for private health care services and also pay informally for public health care services (Tymowska, 2001). There is a well-entrenched culture of informal payments in exchange for health care services, and corruption, which is found to impact the health of the populations as it reduces the income available for other key health inputs such as food (Mavisakalyan et al., 2021). Informal payments appear more common among higher-wealth individuals, and in environments exhibiting a low quality of care and poor resources, including medicines’ availability and long waiting times (Habibov and Cheung, 2017). Specifically, some studies use Hirschman’s (1970) theory of exit, voice and loyalty to explain the phenomenon, as it expresses an ‘inxit’ when exit and voice pathways are canceled (Gaal and McKee, 2004). The proportion of individuals who believe in referring to informal payments in order to receive public medical services reached 37 percent in the post-communist countries in 2016, whereas in the control group (comprised of Albania, Bosnia and Herzegovina, Greece, Italy, Kosovo, Macedonia and Serbia) only 11 percent believed so, based on the Life in Transition Survey (LITS), as shown in Table 4.1. Using the same data, we find the prevalence of informal payments in formerly communist countries is larger than in the control group, and in the case of public health care systems reaches almost 40 percent among individuals that used any public services in the 12 months prior to the interview, exceeding the respective number in the control group by 10 percentage points. The significance of informal institutions was critical in Soviet communist regimes, because in the absence of free markets, the distribution of power was related to an individual’s place in the formal political structures, while the access to privilege depended on individual social and family networks (Filtzer, 2013). This duality was learnt early in one’s life when children were exposed to communist educational institutions, starting from the age of seven at the latest. Schools in the Soviet bloc were responsible for the development of ‘new Soviet men’ (Gerovitch, 2007) and training suitable for the centrally planned economy. Under Stalin’s reign, the Marxist–Leninist ideological content was present in every class in primary education. Starting from the late 1950s and 1960s, Communist Party guidelines for so-called politechnicization reformed the system by shifting the focus from the indoctrination to the development of technical skills needed for heavy industries’ growth. The vague idea of bringing school closer to real life was implemented with the slogans of technological progress combining physical

Soviet communism and later-life health and health care  49 Table 4.1

The shares of individuals believing that informal payments are needed to obtain public medical health care services and referring to them in the last 12 months

  Country

% Believing informal payments are needed

% Making informal payments in last 12 months

Mean

(St.Dev)

Mean

(St.Dev)

Armenia

0.82

(0.38)

0.43

(0.50)

Azerbaijan

0.55

(0.50)

0.72

(0.46)

Belarus

0.54

(0.50)

0.19

(0.40)

Bulgaria

0.50

(0.50)

0.42

(0.50)

Czech Rep.

0.31

(0.46)

0.20

(0.42)

Estonia

0.21

(0.40)

0.11

(0.32)

Georgia

0.21

(0.41)

0.09

(0.30)

Hungary

0.67

(0.47)

0.33

(0.48)

Kazakhstan

0.43

(0.49)

0.53

(0.50)

Kyrgyz Rep.

0.77

(0.42)

0.61

(0.50)

Latvia

0.35

(0.48)

0.28

(0.45)

Lithuania

0.45

(0.50)

0.29

(0.46)

Moldova

0.84

(0.37)

0.55

(0.50)

Mongolia

0.49

(0.50)

0.39

(0.50)

Poland

0.43

(0.49)

0.11

(0.32)

Romania

0.68

(0.47)

0.52

(0.51)

Russia

0.55

(0.50)

0.49

(0.50)

Slovak Rep.

0.41

(0.49)

0.15

(0.37)

Tajikistan

0.41

(0.49)

0.67

(0.47)

Ukraine

0.41

(0.74)

0.59

(0.50)

Uzbekistan

0.30

(0.46)

0.41

(0.51)

Source:  EBRD (2022).

and intellectual effort that translated into a solid vocational training curriculum throughout the entire Soviet bloc, mainly at the secondary and also tertiary ISCED (International Standard Classification of Education) levels (Sowtis, 1991). Another practical change in the education system aimed at preparing keyworkers for future communist economies entailed two years of compulsory physical work (in factories or kolchozes) as a part of secondary education. The values of socialism, anti-religion, atheism, anti-individualism, being anti-family, respect for physical labor, technological development and gender equality were promoted in all schools during the entire period of communism. The superiority of communism over capitalistic exploitation, along with the appreciation of folklore culture, was a constant element of the education system under communism, though after the curricula reforms from the 1980s, these elements of school curricula were treated less seriously than before, particularly as compared to the Stalinist period. The lack of alternatives to free public schooling with a uniform curriculum makes the experience of attaining formal education under communism a unique one, especially in comparison to Western education. Despite numerous reforms to the education systems, the systems in Western and communist Europe differed with respect to the following characteristics throughout the whole communist period: tight state control over curricula, prevalence of non-public and religious schools, length of school day and homework load, amount and type of extra-curricular activities, and approach to physical education (Kurian, 1988). Regional differences between rural and urban areas as well as the absence of religious education were

50  Handbook on the political economy of health systems characteristic to Soviet countries. Most importantly, the entry ages and duration of the compulsory education differed, with later entry to the education system and a shorter compulsory period of education under Soviet communism. The reforms in the middle of the previous century extending compulsory education took place throughout the entire continent, starting from Western Europe and spreading slowly to the Soviet bloc, but deep differences in the curricula contents reflecting the values and social role attached to the education system remained. Some of these differences between Western and Eastern European approaches to education systems have remained after the transition to free markets, not to mention the country-specific cultural contents of school curricula. A major difference of the Soviet system was actually the equal-for-all-pupils opportunity to form new elites by attaining tertiary education (Zajda, 1984; Coumel, 2009). Latvia is a good example of the imposition of the Soviet education system and its performance (McKee and Fister, 2004), despite the fact that in some countries (especially Romania and East Germany) a certain independence from Soviet guidelines was possible. The actual installment of a Soviet standard in education throughout the entire region took place in 1949, once the post-war societies in Europe reached a certain stabilization. By this time almost all children aged seven were in schools and compulsory education lasted seven years. In that period, until 1958, the focus was placed on basic skills and intense Marxist–Leninist indoctrination. The end of Stalinism marked a substantial shift in communist societies, which facilitated reforms to the education system that were formally introduced to all Soviet republics in 1958, taking the opportunity to extend the compulsory education by one more year. The reforms were introduced gradually, and curricula were adapted so that the first cohorts of eighth graders were in school in 1962/63. Similar changes took place in other countries belonging to the Soviet bloc, such as Poland. The next substantial reforms to education systems, entailing, among other changes, the reduction of the entry age to six rather than seven, were planned in the 1980s and were disrupted by the regime crisis leading to its final demise. In all formerly communist countries structural reforms to education systems took place in the 1990s, introducing substantial heterogeneity between countries, but reducing the disparity between the east and west of Europe.

5.

HEALTH EDUCATION GRADIENT IN EASTERN EUROPE

Exposure to specific communist institutions during formative years can influence health behaviors (Pyle, 2021), yet changes to the health care system during the transition to capitalism are likely to affect later-life health in a particular way. The ways in which formal education could affect later-life health are numerous. Firstly, it is still unclear whether the education effects on health are causal, or if the direction of the relationship is not reversed. In addition to the direct impact of education per se, it seems that individuals’ educational attainment enhances better decision-making and healthier choices. The latter can be attributed to more efficient health production due to reduced behavioral risks, enhanced pro-health behaviors and increased medical literacy. Finally, educational attainment can be endogenous to health-relevant preferences. Specifically, human capital formation obtained through formal education might be associated with an individual’s discount factor valuing the distant future more, and exposure to physical education might play a crucial role in the formation of pro-healthy habits, such as physical activity.

Soviet communism and later-life health and health care  51 Indeed, physical education and physical fitness were at the heart of the ‘Soviet man’ ethos, which translated into a substantial amount of physical education in the compulsory education curricula as well as extra-curricular activities in schools (Riordan, 1980). When the physical education in the Western education was focused on scouting talent with a professional potential, the communist approach promoted, in addition to professional sport, recreational physical activities at all levels of formal education. Pupils and students were granted access to sports facilities and equipment during classes and after school hours, were introduced to various sport disciplines and could borrow equipment from schools. Extra-curricular sport classes, free of charge, were very popular, as were inter-school competitions. In sum, the promotion of physical activities in Soviet education played an important role. Physical strength was maintained and further developed in vocational occupations. Educational gradients in self-reported health assessment and other dimensions of health among SHARE (Survey on Health, Ageing and Retirement in Europe) respondents aged 50-plus point to significant differences in the gradients in the formerly communist and Western European countries, despite there being hardly any difference between the regimes as far as the measurements of grip strength among individuals who could have spent a substantial portion of their lives under communist regimes (cf. Figures 4.4 and 4.5).

Source: SHARE (2022).

Figures 4.4 and 4.5

Grip Strength (log) and self-reported health by education level

Our understanding of education’s effects on health seems to support the causal relationship for a subset of developed countries where the educational reforms from the past extending the length of compulsory education can be used as an instrumental variable in the ordinary least square (OLS) analysis or for substantially large samples as a discontinuity in the research discontinuity design. However, the underlying mechanisms remain largely unknown. Moreover, they are documented for one politico-economic regime only, namely matured free market democracies. Furthermore, the exogenous shocks to compulsory schooling laws yielding positive effects of compulsory schooling in the first half of the twentieth century on health are found for the US and UK populations (Lleras-Muney, 2002, 2005; Mazumder, 2008), but analyses for other times and spaces (though still limited to developed countries) yield ambiguous results (Brunello et al., 2013). The consensus in the economic literature is that causal education effects are vastly heterogeneous between countries (Galama et al., 2018). Politico-economic regimes seem to be one

52  Handbook on the political economy of health systems possible candidate for explaining the heterogeneity. A study by Brunello et al. (2013) uses a relatively rare set-up of more than one country, including one formerly communist country (the Czech Republic), and refers to health-related behaviors as a mechanism of education effects (smoking, drinking, Body Mass Index, exercising).

Source: SHARE (2022).

Figures 4.6 and 4.7

Mobility limitations and daily living limitations by education level

Figures 4.6 and 4.7 confirm that health is worse in Eastern than Western Europe, but the gradient is stronger in the East, which might suggest that the difference is driven mainly by the less-educated sub-population, and higher education in formerly communist countries brings more returns with respect to health capital in later life. However, the mechanisms might be different to those in the global West, as in Eastern Europe we do not observe a positive association between educational attainment and refraining from behavioral risks such as alcohol and nicotine consumption. Physical activity is a pro-health activity positively correlated with education attainment in both the West and East. The association between engaging in any type of physical activity (mediocre or vigorous) and self-reported health seems to be almost the same in the East and in the West in an examined SHARE sample (cf. Figure 4.9). Figure 4.8 reveals that in fact,

Source: SHARE (2022).

Figures 4.8 and 4.9

Mediocre activity (1–5) and physical activity (0–1) by education level

Soviet communism and later-life health and health care  53 the effect in the West is driven by mediocre physical activity such as gardening or going for a walk. In contrast, vigorous activities entailing more physical effort prevail in the East. These behavioral traits might be driven by the differences in education systems related to the intensity and content of physical education, as the role of physical education in compulsory schooling curricula and free availability of extra-curriculum sport activities were at the core of Soviet education (Riordan, 1980; Rowley, 2006). Several studies show that exposure to physical education at younger ages has an effect on full-lifespan engagement in physical activities (Kirk, 2005; Hamar et al., 2006) in line with recent experimental studies focused on young adults (Kaushal et al., 2017, 2018).

6. CONCLUSION The disintegration of Soviet communism led to the development of direct health effects as well as institutional effects that could indirectly impact on health and the functioning of health systems. Indeed, the end of communism brought about an erosion in trust regarding health care institutions which changed how individuals approach the delivery of health care, and more specifically the use of informal payments and the proliferation of forms of corruption that further enhanced the privatization of the system. Some health systems developed a unique structure, especially in countries of the former Soviet Union’s so-called Semashko system and its remnants in Central-Eastern Europe characterized by significant under-investment and large inequality, especially after the transition to capitalism, which resulted is large disparities in health, and health care use, across the population. Indeed, evidence suggests that income- and education-related inequalities in health are comparable between Eastern and Western European countries (Jutz, 2020). Yet, the end of Soviet societies led to direct effects on healthy lifestyles, and especially on schools’ investment in physical activity. The effects of Soviet investment in physical education are still visible today, as we have shown in this chapter.

NOTE 1. The research was supported by the Polish National Science Centre, from grant no. 2019/33/B/ HS4/00387.

REFERENCES Balcerowicz, L. (1994). Understanding postcommunist transitions. Journal of Democracy, 5(4): 75–89. Becker, S.O., Mergele, L., and Woessmann, L. (2020). The separation and reunification of Germany: Rethinking a natural experiment interpretation of the enduring effects of communism. Journal of Economic Perspectives, 34(2): 143–71. Bhattacharya, J., Gathmann, C., and Miller, G. (2013). The Gorbachev anti-alcohol campaign and Russia’s mortality crisis. American Economic Journal: Applied Economics, 5(2): 232–60. Bobak, M., Pikhart, H., Rose, R., Hertzman, C., and Marmot, M. (2000). Socioeconomic factors, material inequalities, and perceived control in self-rated health: Cross-sectional data from seven post-communist countries. Social Science & Medicine, 51(9): 1343–50.

54  Handbook on the political economy of health systems Brunello, G., Fabbri, D., and Fort, M. (2013). The causal effect of education on body mass: Evidence from Europe. Journal of Labor Economics, 31(1): 195–223. Connor, J., Norton, R., Ameratunga, S., and Jackson, R. (2004). The contribution of alcohol to serious car crash injuries. Epidemiology, 15(3): 337–44. Costa-Font J., Garcia-Hombrados J., and Nicińska A. (2023). The institutional origins of vaccine distrust: Evidence from former-Soviet countries. PLOS ONE, 18(3): e0282420. doi: 10.1371/journal. pone.0282420. Coumel, L. (2009). The scientist, the pedagogue and the party official: Interest groups, public opinion and decision-making in the 1958 education reform. In M. Illic and J. Smith (eds), Soviet State and Society Under Nikita Khrushchev, pp. 84–103. Routledge. Dyba, K., et al. (1994). Stabilization and transition in Czechoslovakia. In O.J. Blanchard, K.A. Froot and J.D. Sachs (eds), The Transition in Eastern Europe, Vol. 1, pp. 93–122. University of Chicago Press. EBRD (2022). Life in Transition Survey (LITS). https://​www​.ebrd​.com/​what​-we​-do/​economic​-research​ -and​-data/​data/​lits​.html. Feshbach, M., and Rubin, A. (1991). Health care in the USSR. In J. Adam (ed.), Economic Reforms and Welfare Systems in the USSR, Poland and Hungary, pp. 68–84. Springer. Filtzer, D. (2013). Privilege and inequality in communist society. In S. Smith (ed.), The Oxford Handbook of the History of Communism, pp. 505–21. Oxford University Press. Gaal, P., and McKee, M. (2004). Informal payment for health care and the theory of ‘INXIT’. International Journal of Health Planning and Management, 19(2): 163–78. Galama, T.J., Lleras-Muney, A., and Van Kippersluis, H. (2018). The effect of education on health and mortality: A review of experimental and quasi-experimental evidence. National Bureau of Economic Research Working Paper 24225. Gerovitch, S. (2007). ‘New Soviet man’ inside machine: Human engineering, spacecraft design, and the construction of communism. Osiris, 22(1): 135–57. Grosfeld, I., and Zhuravskaya, E. (2013). Persistent effects of empires: Evidence from the partitions of Poland. CEPREMAP Working Papers 1311. Grossman, M. (2005). Education and non-market outcomes. National Bureau of Economic Research Working Paper 11582. Guriev, S., and Zhuravskaya, E. (2009). (Un)happiness in transition. Journal of Economic Perspectives, 23(2): 143–68. Habibov, N., and Cheung, A. (2017). Revisiting informal payments in 29 transitional countries: The scale and socio-economic correlates. Social Science & Medicine, 178: 28–37. Hamar, P., Peters, D.M., Van Berlo, K., and Hardman, K. (2006). Physical education and sport in Hungarian schools after the political transition of the 1990s. Kinesiology, 38(1): 86–93. Hinote, B.P., Cockerham, W.C., and Abbott, P. (2009). Post-communism and female tobacco consumption in the former Soviet states. Europe-Asia Studies, 61(9): 1543–55. Hirschman, A.O. (1970). Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and states, Volume 25. Harvard University Press. Jenkins, R., Klein, J., & Parker, C. (2005). Mental health in post-communist countries. BMJ, 331(7510): 173–4. Jutz, R. (2020). Health inequalities in Eastern Europe: Does the role of the welfare regime differ from Western Europe? Social Science & Medicine, 267: 113357. Kaestner, R. (2013). The Grossman model after 40 years: A reply to Peter Zweifel. European Journal of Health Economics, 14(2): 357–60. Kaushal, N., Rhodes, R.E., Meldrum, J.T., and Spence, J.C. (2018). Mediating mechanisms in a physical activity intervention: A test of habit formation. Journal of Sport and Exercise Psychology, 40(2): 101–10. Kaushal, N., Rhodes, R.E., Spence, J.C., and Meldrum, J.T. (2017). Increasing physical activity through principles of habit formation in new gym members: A randomized controlled trial. Annals of Behavioral Medicine, 51(4): 578–86. Kirk, D. (2005). Physical education, youth sport and lifelong participation: The importance of early learning experiences. European Physical Education Review, 11(3): 239–55. Kosiński, K. (2005). Z historii pijaństwa w czasach prl „peerelowskie” wzory picia alkoholu. Polska 1944/45-1989: studia i materia ly, 7: 267–305.

Soviet communism and later-life health and health care  55 Kujawska, J. (2017). The efficiency of post-communist countries’ health systems. In U. Comite (ed.), Advances in Health Management, pp. 93–111. IntechOpen. Kurian, G.T. (1988). World Education Encyclopedia, Vol. 3: Major Countries – Sudan–Zambia, Middle Countries – Afghanistan–Zimbabwe, Minor Countries – Andorra–Yemen, Appendixes. Facts on File Publications. Lecce, G., and Ogliari, L. (2019). Institutional transplant and cultural proximity: Evidence from nineteenth-century Prussia. Journal of Economic History, 79(4): 1060–93. Lleras-Muney, A. (2002). Were compulsory attendance and child labor laws effective? An analysis from 1915 to 1939. Journal of Law and Economics, 45(2): 401–35. Lleras-Muney, A. (2005). The relationship between education and adult mortality in the United States. Review of Economic Studies, 72(1): 189–221. Lleras-Muney, A. (2022). Education and income gradients in longevity: The role of policy. Technical report, National Bureau of Economic Research. Marrée, J., and Groenewegen, P.P. (1997). Back to Bismarck: Eastern Europe Health Care Systems in Transition. Avebury. Mavisakalyan, A., Otrachshenko, V., and Popova, O. (2021). Can bribery buy health? Evidence from post-communist countries. Journal of Comparative Economics, 49(4): 991–1007. Mazumder, B. (2008). Does education improve health? A reexamination of the evidence from compulsory schooling laws. Economic Perspectives, 32(2): 2–16. McKee, M., and Fister, K. (2004). Post-communist transition and health in Europe. BMJ, 329(7479): 1355–6. Millard, F. (1995). Changes in the health care system in post-communist Poland. Health & Place, 1(3): 179–88. Movsisyan, N.K., Vinciguerra, M., Medina-Inojosa, J.R., and Lopez-Jimenez, F. (2020). Cardiovascular diseases in central and eastern Europe: A call for more surveillance and evidence-based health promotion. Annals of Global Health, 86(1): 21. Mundt, A.P., Frančišković, T., Gurovich, I., Heinz, A., Ignatyev, Y., Ismayilov, F., Kalapos, M.P., Krasnov, V., Mihai, A., Mir, J., et al. (2012). Changes in the provision of institutionalized mental health care in post-communist countries. PLoS One, 7(6): e38490. Nolte, E., Shkolnikov, V., and McKee, M. (2000). Changing mortality patterns in East and West Germany and Poland, II: Short-term trends during transition and in the 1990s. Journal of Epidemiology & Community Health, 54(12): 899–906. Orenstein, M.A. (2001). Out of the Red: Building Capitalism and Democracy in Postcommunist Europe. University of Michigan Press. Popov, V. (2017). Mortality and Life Expectancy in Post-Communist Countries. https://​pages​.nes​.ru/​ vpopov/​documents/​Mortality​%20and​%20life​%20expectancy​%20in​%20post​.pdf. Pyle, W. (2021). Russia’s ‘impressionable years’: Life experience during the exit from communism and Putin-era beliefs. Post-Soviet Affairs, 37(1): 1–25. Rechel, B., Richardson, E., and McKee, M. (2014). Trends in health systems in the former Soviet countries. European Journal of Public Health, 24(suppl 2). Riordan, J. (1980). Sport in Soviet Society: Development of Sport and Physical Education in Russia and the USSR, Volume 22. CUP Archive. Roland, G. (2004). Understanding institutional change: Fast-moving and slow-moving institutions. Studies in Comparative International Development, 38(4): 109–31. Romaniuk, P., and Semigina, T. (2018). Ukrainian health care system and its chances for successful transition from soviet legacies. Globalization and Health, 14(1): 1–11. Rostowski, J., et al. (1998). Macroeconomic Instability in Post-Communist Countries. OUP Catalogue. Rowley, A. (2006). Sport in the service of the state: Images of physical culture and Soviet women, 1917–1941. International Journal of the History of Sport, 23(8): 1314–40. Sachs, J.D. (1992). Privatization in Russia: Some lessons from Eastern Europe. American Economic Review, 82(2): 43–8. Salavecz, G., Chandola, T., Pikhart, H., Dragano, N., Siegrist, J., Jöckel, K.-H., Erbel, R., Pajak, A., Malyutina, S., Kubinova, R., et al. (2010). Work stress and health in Western European and post-communist countries: An East–West comparison study. Journal of Epidemiology & Community Health, 64(1): 57–62.

56  Handbook on the political economy of health systems SHARE (2022). Survey of Health, Ageing and Retirement in Europe. https://​share​-eric​.eu/​. Sowtis, D. (1991). Soviet industrial strategy and reforms in vocational education, 1984–88: Policy implications and implementation. Comparative Education, 27(1): 23–33. Stuckler, D., King, L., and McKee, M. (2009). Mass privatisation and the post-communist mortality crisis: A cross-national analysis. The Lancet, 373(9661): 399–407. Tymowska, K. (2001). Health care under transformation in Poland. Health Policy, 56(2): 85–98. Valuch, T. (2021). Everyday Life under Communism and After: Lifestyle and Consumption in Hungary, 1945–2000. Central European University Press. Zajda, J. (1984). Recent educational reforms in the USSR: Their significance for policy development. Comparative Education, 20(3): 405–20. Zatonski, W.A., McMichael, A.J., and Powles, J.W. (1998). Ecological study of reasons for sharp decline in mortality from ischaemic heart disease in Poland since 1991. BMJ, 316(7137): 1047.

5. Federalism and tax-financed healthcare: economic advantages, dilemmas, and solutions Roger D. Congleton

1.

INTRODUCTION: THE LOGIC OF THE SUBSIDIARITY PRINCIPLE

The subsidiarity principle suggests that tax-financed services should be provided at the lowest level of government that is able to realize all or most economies of scale in production (Oates 1972). The standard argument focuses on the better match between local demand and local supply under decentralized forms of government policy making. In addition, decentralization provides a variety of informational and efficiency advantages. The results generated by the policy decisions of lower levels of government provide useful information—“yardsticks”— that can be used to improve both voter and expert assessments of best practices. These, in turn, can be used to improve the policies in place (Salmon 1987, 2019). Moreover, citizen mobility puts pressures on governments to provide and finance services efficiently, which tends to improve the quality and extent of the services provided while reducing tax burdens (Tiebout 1956). In cases in which politically active interest groups have significant influence over the policies adopted, game theory suggests that rent-seeking losses from such activities tend to be lower under federalism than under unified governance (Wärneryd 1998). All these lines of argument—and their supporting empirical work—imply that decentralized provision of most government services yields better outcomes for voters than those associated with centralized supply. Healthcare could be an exception to that rule, however, because it has many special properties. Tax-financed healthcare is not a single service—as elementary education or roads arguably are—but a variety of services with differences in both economies of scale and in the nature of citizen demands for them. It is not uniformly provided to a citizenry, but rather the services received vary with a citizen’s state of health. Other demands for tax-financed services also tend to be conditional—as demands for highways and bicycle paths, for example, are largely conditional on the ownership of cars and bicycles—but their conditioning factors tend to be well-understood constants rather than unforeseeable random events. In addition, there is more informational asymmetry in markets for healthcare than for many other government services—although it is not unique in that respect. The demand for many kinds of healthcare tends to be driven by expert (doctor) opinion. The typical patient has only a vague idea about the best methods for addressing a heart problem or specific form of cancer. In this, healthcare demands are similar to others for which expert opinions are deferred to, as true of defense spending, higher education, and auto repair. In all such cases, the quality of expert advice is rarely obvious to the nonexperts who make use of their recommendations. Such informational asymmetry problems are increased by the fact that an immediate need for treatment for an unforeseen malady may eliminate opportunities to collect additional information after a health problem emerges. Experts often have different interests in healthcare than 57

58  Handbook on the political economy of health systems their recipients. For suppliers, healthcare is often a matter of personal income and convenience, whereas for patients, it may be a matter of life and death. This chapter demonstrates that these and other properties of healthcare do not eliminate the case for the decentralized provision of tax-financed healthcare services but do require some of the arguments to be modified. For example, differences in the economies of scale involved in treating different maladies imply that different types of healthcare services may be most appropriately provided by different levels of government. Informational problems imply that both centralized and decentralized systems of provision are likely to work better for services that are well understood by patients/voters than for those for which deference to experts is necessary. However, in general, decentralized provision of healthcare services is superior to centralized provision, as is true of many other government services, partly because it provides more useful information about the quality of expert services than centralized systems do. For the purposes of this chapter, it is assumed that the aim of a tax-financed healthcare system is to provide broader access to healthcare than that associated with supply in private markets. The analysis focuses on single-payer systems that are financed with a tax on wage income. Other methods of provision and tax finance tend to have more or less similar effects on the menu of services provided and the efficiency with which they are provided when they are ultimately driven by electoral outcomes; thus, the overview developed herein also sheds light on other tax-financed healthcare systems than the one focused on. It is also assumed that all the states or counties making healthcare decisions are of sufficient size to realize all economies of scale in the provision of the relevant services, and also sufficient in number that useful information is generated about the healthcare services provided. The analysis focuses on a single type of healthcare and separately considers services that are well understood by patients/voters and those that are not. This approach implies that the menu of services and service levels, in effect, is selected one service at a time. Although this method ignores complementarities between healthcare services, it allows the main features of electorally driven healthcare systems to be characterized with just a few pages of straightforward analysis and prose.1

2.

A POINT OF DEPARTURE: THE SIMPLE ECONOMICS OF A SIMPLE, TAX-FINANCED SINGLE-PAYER SYSTEM FOR SINGLE WELL-UNDERSTOOD TYPES OR CATEGORIES OF HEALTHCARE

We start with the easy cases—those in which healthcare services are well understood by both patients and healthcare providers. In such settings, well-informed choices can be made about the services demanded. 2.1

A Short Digression on Private Markets for Well-Understood Forms of Healthcare

The logic of private markets implies that products produced without significant economies of scale or information problems tend to be produced at levels that equate long-run market demand with long-run market supply. At the equilibrium, the products sold are produced at their lowest average cost and the organizations producing them are efficiently sized to do so.

Federalism and tax-financed healthcare  59 This minimizes costs for consumers while providing an adequate return on both human and physical capital for those investing in the production and distribution of the health-related goods and services sold. In cases in which no or few externalities are generated by the production or consumption of healthcare, the equilibrium also maximizes (expected) social net benefits (expected consumer surplus plus profits) and is Pareto efficient. An illustration of such an equilibrium is provided at point (P*, Q*) in Figure 5.1A and B. Healthcare services of this type are simply a subset of many services with these properties.

Figures 5.1A and B

Competitive equilibrium, and competitive equilibrium with single-payer system

However, under other normative theories, such as utilitarianism, it can be argued that private markets undersupply healthcare services because of the effects of income on purchases of both critical and noncritical healthcare services. The “poor” purchase fewer services than maximizes aggregate utility or advances other similar normative theories, as with Rawls’ difference principle. Together, such normative theories and the tax systems used to finance government-provided healthcare can create voter demands to expand healthcare beyond private levels. A single-payer system has several advantages over other methods for doing so. For example, it can rely upon market competition and incentives to generate the lowest sustainable prices paid for healthcare services. In such cases, there is no reduction in the efficiency with which healthcare services are provided, although there is an effect on the magnitude of expenditures, as illustrated in Figure 5.1B. Expenditures increase from the area of rectangle P*Q* to that of P’Q’. In competitive markets for well-understood healthcare services, a single-payer system can be adopted without undermining the efficiency with which healthcare was formerly provided, although this requires government monitoring and other billing practices to be as efficient as those provided by private insurance companies. Long-run supply slopes upward because healthcare markets are relatively large and so attracting significantly additional labor and capital to those markets requires somewhat higher wage rates and rates of return. This implies that average costs tend to increase as one switches from private provision to tax-financed provision, although there is no reduction in efficiency. Long-run average costs are still minimized

60  Handbook on the political economy of health systems for the service level provided as long as consumers continue to shop for the best values in healthcare or healthcare administrators do so for them. Tax-financed expenditures may be deemed normatively attractive to voters that have internalized utilitarian, Rawlsian, or egalitarian ideals. Furthermore, tax-financing tends to be favored by voters for whom tax finance reduces their personal cost for healthcare. The latter tends to be true of voters with below-average income or greater-than-average healthcare needs. Together, such voters may be sufficient to generate majority support for single-payer systems, even though average costs are increased by such systems. Average costs, however, do not rise in every case. They are most likely to rise in cases in which all relevant markets are competitive, informational asymmetries are minimal, and healthcare demands for inputs are relatively large. In noncompetitive healthcare markets, healthcare costs may be reduced through administered prices. Such cases are analyzed in Section 3. 2.2

An Advantage of Decentralized Single-Payer Systems over Unitary Systems

When the menu of healthcare services provided by the tax-financed system is determined by electoral pressures, the menu of healthcare services funded tends to reflect the median voter’s demand for healthcare services, given the tax system adopted and his or her own expected future health. These vary, for example, with age, regional health risks, and personal income. As a consequence, regional governments (state, province, or city) tend to provide somewhat different menus of services (or insurance coverage), because of differences in health expectations among regional median voters. In cases in which voters are well informed about the risks and services of interest and the menu of services reflects electoral pressures, each region’s services are optimal for the median voter of that region, and they are preferred by majorities both to those of other regions and any uniform menu of services different from that in their own region.2 This property was termed the “decentralization theorem” by Oates (1972). Figure 5.2 illustrates its essential logic. To make the illustration as clear as possible, the two median voters are assumed to face the same marginal cost for additional healthcare services that treat the malady of interest, as would be true of competitive markets for the same service. State median voters, however, have different marginal benefits from the service of interest because of differences in age, base health, population density, risk aversion, and so forth. The national median voter is likely to prefer a service level between those levels because he or she is roughly the median of state medians. Note, however, that an intermediate level between the two ideals of the median voters would impose losses on the median voter of state A equal to area I and losses on the median voter of state B equal to area II. This implies that majorities of voters in each state would prefer their own state’s policies to that of the other state and also to any uniform national service levels between them. Federal and other decentralized methods of supply provide net benefits that are not available in unitary systems in which all citizens receive the same (conditional) healthcare services. This result does not hinge on mobility among states or superior local information; it simply requires sufficient voter information and electoral competition that tax-financed healthcare services converge toward the median voter’s ideal in every state, province, or county.3

Federalism and tax-financed healthcare  61

Figure 5.2

3.

The geometry of the decentralization theorem

MONOPSONY POWER UNDER A TAX-FINANCED SINGLE-PAYER SYSTEM

Even better outcomes from single-payer systems are possible in cases in which significant monopoly power initially exists for a subset of healthcare services or inputs. In such cases, tax-financed systems can potentially reduce—rather than increase—the average costs of healthcare services through various kinds of price administration. Although authority to regulate service prices can be abused, let us first consider cases in which it is not—that is to say, cases in which it is used to advance voter interests rather than to favor service and input providers. In cases in which monopoly power exists for a subset of healthcare providers or services, it is possible to “force” prices down without significantly reducing the supply of services. The reason is that some providers are realizing “rents” from the services they sell; that is, they are realizing prices or rates of return above those required to attract them to the healthcare sector. Reducing prices for services provided by competitive markets will tend to induce resources to leave the healthcare sector for others that realize better prices and rates of return. This will either diminish the healthcare services available or diminish their average quality. However, where monopoly (or other rents) are being realized, the average cost of tax-financed medical services can be reduced by imposing administered prices on the monopolized or “protected” subset of the inputs or final services. Figure 5.3 illustrates how an idealized form of price administration works. A healthcare regulator replaces monopoly price W* with administered price W’, which can increase supply of a formerly monopolized type of labor (from L* to L’) while lowering its average cost to patients (from W* to W’). Opportunities to use such powers include services previously provided by cartels, machines and drugs protected by excessively long patents, payments for

62  Handbook on the political economy of health systems immobile resources such as land, and services where licensing reduces the ability of practitioners to provide medical services in different states or countries.

Figure 5.3

Monopoly and price administration

National governments can use this method of reducing the cost of medical services somewhat more effectively than states can, because national borders tend to be less porous (more regulated) than state borders, although states can also engage in such practices. Administered prices reduce average costs without affecting the quality of services only in cases where suppliers were previously receiving monopoly profits or rents. Variety of pricing policies adopted within federal systems, thus, provides useful information about the extent to which prices for various inputs can be reduced without affecting quality. (There is often more competition in a given market than is obvious to nonexperts.)

4.

THE CONSEQUENCES OF REGULATING PRICES IN COMPETITIVE MARKETS

When administrative pricing is applied to inputs that are provided through competitive markets, the average quality of healthcare services tends to decline. Figures 5.4A and B help to illustrate why this tends to be the case. Figure 5.4A characterizes the upper bound, K(w), of the quality of the labor inputs available for a specific service at wage w. Normally, medical suppliers attempt to hire the best persons willing to work at a given wage. This tends to produce medical staff for various positions from a relatively narrow band of quality near the upper bound of the quality of applicants attracted by an advertised wage rate. As wage rates decrease, the upper bound decreases. For example, in Figure 5.4A and B, a decrease in the wage on offer from w* to w’ reduces the upper bound of the quality of applicants from K(w*) to K(w’).

Federalism and tax-financed healthcare  63 When monopsony power depresses wages for persons not earning rents, the upper end of the distribution of the quality of applicants declines because such policies do not depress wages in other fields for which the same talents are valuable. Moreover, as indicated by Figure 5.4A and B, lower wages tend to reduce the extent to which the persons employed are willing to work, which implies that more employees are required to produce the services previously provided (e.g., before wages were lowered by healthcare administrators). When such administrative price strategies are used at the same time that the production of healthcare services is being increased, the quality effects tend to be larger because the upper bound of the pool of persons willing to work at the administered wage (w’) has diminished because of lower wages, whereas an increase in services requires more persons to be hired from that pool. The reduction in wages decreases the upper bound of the pool of applicants, and the necessity of hiring additional medical staff lowers the lower bound of the pool of “acceptable” applicants. Both of these effects reduce the average quality of those employed and the services produced, other things (such as technology) being equal.

Figures 5.4A and B

Quality/quantity tradeoffs and Q of labor of given quality supplied at a given wage

What federalism does when pricing authority is applied to previously competitive markets is to provide evidence of how the quality of services is affected by differences in pricing policies. The effects of different state price reductions on the quality of staff or other inputs and thereby on final services will be observed, and those data can be used to determine optimal tradeoffs. Whether the cost savings in formerly competitive markets are sufficient to offset the lower quality of healthcare services is likely to vary by service. If higher wages had previously attracted higher-quality inputs into the healthcare sector without materially improving the quality of final services (e.g., without increasing the longevity or base health of the average person of a given age in the community), voters may prefer somewhat reduced quality at a significantly lower tax cost. The relative shares of gross domestic product spent on healthcare services in Scandinavia (10–11 percent) and in the United States (17.8 percent) according to Organisation for Economic Co-operation and Development (OECD). These statistics suggest that monopoly rents are commonplace in the United States and/or that reductions in the average quality of healthcare personnel have only minor effects on the quality of healthcare services. Longevity

64  Handbook on the political economy of health systems in Scandinavian countries is somewhat higher than that in the United States, both for life as a whole and at age 60. World Health Organization (WHO) data for 2020 provide the following average longevities at age 60: Norway, 25.1; Sweden, 24.5; Finland, 24.2; Denmark, 23.6; and United States, 23.1.4

5.

UNDER-CONTRACTING FOR SERVICES AND TWO-PART PRICING FOR HEALTHCARE SERVICES

Another somewhat related method of reducing the tax price of single-payer systems is to contract with fewer or smaller health providers than sufficient to provide timely healthcare for all persons eligible for the tax-financed services. Under-contracting generates congestion of healthcare facilities and relatively long waiting times for many of the services provided. The latter is a common feature for the most resource-intensive services funded by national healthcare systems, whether single-payer or government-provided. This reduces the annual flow of healthcare services provided, but also reduces the taxes necessary to fund them. In effect, such a strategy creates two-part pricing of healthcare services: the tax part discussed in the previous sections and waiting time for healthcare services. Such two-part pricing may be in the interest of voters who are willing to bear longer waiting times for services in exchange for lower taxes, as might be the case for low-income voters when single-payer systems are financed with a proportional tax on labor income. In such cases, electoral pressures would generate the mix of waiting times and higher taxes that is approximately ideal for the median or pivotal voter in the relevant electorate, although other voters would prefer different combinations of waiting time and price. Poorer and less busy voters, for example, might prefer a lower tax price and even longer waiting time. Those with a higher opportunity cost for waiting would prefer a higher tax price and shorter queues. Such variation in assessments of waiting times provides another possible rationale for federal supply. Because the price/waiting tradeoffs of moderate voters vary among regions, the overall pattern of pricing and waiting times would tend to generate net benefits for individual state median voters, similar to those shown in Figure 5.2. Congestion also reduces healthcare costs somewhat by discouraging the use of healthcare facilities by persons with less than clear health needs. (In effect, waiting is a copayment.) Information about relevant tradeoffs might also affect the resource allocations of hospital administrators.

6.

THE DILEMMA OF EXPERTS IN TAX-FINANCED SINGLE-PAYER SYSTEMS

Unfortunately, relatively few areas of healthcare are well understood by persons without medical training. Markets for the commonplace treatments of colds, minor physical injuries, and inoculations are well understood by most patients and voters, but other more technical areas of service—the ones that tend to be most expensive, partly for reasons discussed in this section—tend to be ones in which providers have significant expertise and have to be trusted by their patients to provide reliable advice about the best treatments. In settings in which a good deal of informational asymmetry exists, one cannot simply assume that competitive markets are efficient in the sense that economists use that term. Nor can it be assumed that

Federalism and tax-financed healthcare  65 the expert administrators of a single-price system will adopt the best compensation system for every service. Unfortunately, expert service providers rarely have strong reasons to provide their services at least cost because of the manner in which services are paid for. Self-serving advice and errors will be commonplace in many choice settings. Medical experts often overprescribe their services (Green 2014, Makary et al. 2017). In service areas in which significant informational asymmetries exist, single-payer systems that rely upon market forces to control costs tend to reinforce problems already present in market-based healthcare systems. When bills in a single-payer system are simply sent to the government, this replaces patient budget constraints with the far larger and more elastic budget constraints of government(s). Because the “sky” would otherwise be the limit, given the lack of well-functioning markets in such areas, single-payer systems have to limit payments for expert services provided in order to keep healthcare costs at or below those of private markets.5 In a unitary government, there is just one administered price schedule instead of the multiplicity typical in insurance markets. This reduces transactions costs but may distort medical practices because medical providers will attempt to find the most profitable niches in those administrative prices schedules. Within unitary systems, those prices cannot be taken to be signals of need, as would be the case in competitive markets when consumers are themselves experts on what they need. Federalism allows fee-for-service schedules to generate information about how different pricing and compensation schedules affect outcomes (such as longevity and waiting times), which allows more finely grained schedules to be worked out that better advance voter interests. If patients and circumstances were identical in all states, such adjustments would gradually induce convergence to a single ideal pricing system. When voters and circumstances are not the same (and elections are open and competitive), state administrative price schedules tend to converge toward schedules that best advance each state’s moderate voter health and economic interests.

7.

ADVANTAGES AND DISADVANTAGES OF VOTER/ PATIENT MOBILITY

To this point, it has been assumed that voters/patients are all immobile, in which case the advantages of federalism emerge from its ability to take account of differences in healthcare demands among regional voters and from useful information generated by differences in regional healthcare policies. This subsection drops the assumption of voter immobility and considers the effects of both permanent and temporary medical migration, given all the above. In the analysis that follows, patients may migrate to other places with better healthcare services either temporarily or more or less permanently in the pursuit of care available elsewhere but not in proximity to their current homes. Tiebout’s (1956) strong normative case for a federal system is grounded on citizen mobility. His classic paper demonstrates that, if there is enough competition among local governments for residents (who bring their tax payments and taxable assets with them) and if residents chose among locations based on the fiscal packages (taxes and services) provided, the result is an equilibrium in which each community’s demand for services is homogeneous (because of migration); each community’s services is produced at lowest cost (because competition for residents among governments); and each community’s fiscal package is ideal for its residents

66  Handbook on the political economy of health systems (because of in- and out-migration). Possible externalities among governments are ignored in Tiebout’s analysis (as they usually are with respect to competition in private markets). In this limiting case of inter-governmental competition, the governmental supply of local public goods and other services is Pareto efficient and resembles that of perfectly competitive markets. However, problems arise from mobility when persons can obtain healthcare services in communities other than those in which they pay taxes. Suppose, for example, that person A lives in community L, which provides relatively low services, and commutes to community H, which provides relatively high services. A pays taxes for healthcare services in community L, where healthcare services are adequate on his or her relatively healthy days, but he or she prefers the services in community H on those days and weeks when he or she is in ill health. If community H pays the medical expenses of A, taxpayers in H are subsidizing A’s healthcare, and their taxes for healthcare services necessarily increase. This tends to induce residents of H to migrate to communities like L. As a consequence, the tax base of H decreases and its ability to provide high service levels declines. In the limit, communities of type H disappear. In such cases, there is, in effect, a race to the bottom that eliminates type-H communities and their associated high service levels. The opposite is also possible. Suppose instead that community L is required to reimburse community H for A’s expenses. In this case, taxes rise for community L. This reduces the attractiveness of community L for those not taking advantage of the services in community H, which weakens interest in living in low-cost, low-service towns like L. In the limit, out-migration may cause communities of type L to disappear. Similar issues are associated with emergency treatments when a person travels to a location other than that in which he or she pays taxes and has the misfortune to suffer a medical emergency. Such payment and eligibility systems undermine the Tiebout-type of equilibrium because they create a mismatch between tax rates and services received. Medical free-riding problems are less likely in unitary systems than in federal systems because medical migration and other travel is more likely to be intra-national than international. Both temporary medical migrations and the need for emergency treatments during travel thus undermine the normative case for federalism unless some reasonable solution for such problems can be found—or it turns out that such problems occur very infrequently. Note that simply “forcing” A to pay taxes in community H would solve this problem, because H’s service level is actually A’s true demand. This would effectively make A a resident of community H rather than of community L and eliminate the free-riding problem. Unfortunately, this requires knowing everyone’s true long-run demand for healthcare services—which is beyond the ability of medicine and economics at this point. Alternatively, person A might be eligible for services in community H only if he or she purchases a supplemental insurance policy that covers the extra cost of services in high-service communities. Community L would reimburse other locales for demands by its residents according to its own price and service schedules. Supplementary insurance solves the free-riding problem without requiring as much information about A’s long-term health risks, because estimating aggregate or average risks tends to be easier than estimating individual risks.

Federalism and tax-financed healthcare  67

8.

MEDICAL RENT-SEEKING

To this point, it has been assumed that both local and national tax-financed healthcare services are determined by electoral pressures. However, this is not always the case. The design and management of healthcare systems requires significant expertise. As a consequence, the task of choosing the details of a healthcare system is normally delegated to specialists and their advisers, much as highway construction is. Such delegation has advantages, but it also weakens the link between elections and healthcare systems. This potentially allows special interest groups to influence the extent of healthcare provided or the administrated fees received by input owners. Olson’s (1971) analysis suggests that well-organized groups with relatively large interests (stakes) in healthcare policy decisions are the most likely to be able influence the decisions of policy makers and the advice given to them by their advisers. In Olson’s terminology, healthcare providers are “privileged” relative to other larger “latent” groups that fail to organize. Not all such efforts conflict with the interests of patients/voters. Just as advertising in private markets is sometimes informative and useful, so too is the lobbying that takes place within a legislature and bureaucracy. However, there are clearly cases in which the interests of organized interest groups run counter to the interests of patients/voters, and where the information provided tends to be biased to advance the group’s narrow interests. For example, all input providers would like higher prices for their services and products, whereas all patients/voters would prefer lower prices—other things, including quality, being equal. Efforts to increase reimbursement rates might be justified by a politically active interest group; for example, be justified by providing production cost data that are higher on average than actual costs are. If persuasive, such lobbying efforts would produce economic rents for the persons benefiting from increased reimbursement rates. It would also raise the average cost of healthcare services and the associated taxes necessary to finance a single-payer system. In addition to such direct costs of higher prices for particular healthcare services, resources are consumed by the process of seeking such rents. Insofar as the costs of lobbying are paid for out of revenues generated by selling inputs to healthcare systems, this tends to further increase the average costs of single-payer and other tax-financed healthcare services. The stakes are larger in unitary systems than decentralized ones because pricing and other policy decisions affect much larger markets and therefore have larger impacts on the wage levels and profits of practitioners and input suppliers. This “prize effect” tends to increase investment in rent-seeking in centralized systems relative to that of decentralized systems. Models of rent-seeking effort developed by Wärneryd (1998) suggest that such costs tend to be lower in federal and other decentralized policy making systems than in unitary ones, which implies that average costs tend to be lower for this reason as well as others developed above.

9.

CONCLUSIONS: FEDERALISM AND TAX-FINANCED HEALTHCARE SERVICES

The term “federal” has been used throughout this chapter to describe systems in which policy making is independently undertaken by several levels of government and local taxpayers pay for the “last” dollar of the healthcare services provided, as was true of, for example, tax-financed healthcare services in the United States and Scandinavia during the second half

68  Handbook on the political economy of health systems of the 20th century. In this respect, the term “federal” is used somewhat differently than it is by political scientists, who tend to focus on the constitutional structure of a government. No “state” or “regional” chamber of a legislature is required for a healthcare system to be federal in the economic sense used here. What is required is that state or provincial governments or agencies are able to organize, finance, and regulate the provision of healthcare services and that local taxes fund the “last dollar” spent in those systems. The latter reduces a state or provincial authority to free riding on the taxes paid by persons living outside the state or province of interest. It has also been presumed that either all services can be economically provided by the relevant states or provinces or that responsibilities for funding and managing particular healthcare services have been delegated to particular levels of government in accord with economies and diseconomies in production. This assumption may seem far-fetched to some readers, but it should be kept in mind that some of the best-run tax-financed healthcare systems are found in Scandinavian countries with relatively small populations and that their healthcare services are often controlled by subunits of governments (what might be termed “counties” in Great Britain or the United States). The populations of Norway and Finland are approximately 5 million each, that of Denmark is approximately 6 million, and that of Sweden is approximately 10 million. Many states and cities in larger countries exceed national populations in Scandinavia and, therefore, should be able to realize all the economies of scale in healthcare realized in those countries. Given these assumptions and their supporting facts, this chapter has argued that decentralized systems of tax-financed healthcare tend to achieve better results than unitary ones. They do so for several reasons. First, services are better matched to local demands. Variation in the extent of services and their manner of production tends to reflect the interests of moderate voters in the healthcare-providing districts when electoral pressures are decisive. Second, variations in the extent and production of healthcare services provide data through which both voters and experts can better assess the efficiency of local healthcare services and directions for improvements. In contrast, within a unitary state, only a single mode of finance and production is typically observed, so it is impossible to determine the relative merits of alternative methods of reimbursement, delivery, or combinations of services. Third, there are reasons to expect less rent-seeking in decentralized systems than in centralized ones. The potential rents that might be realized are lower because relevant markets are smaller. The analysis provided in this chapter does not rely on the idealized competition used in Tiebout’s (1956) classic article. Although mobility can provide additional economic and political pressures to increase the efficiency with which healthcare services are provided, it also causes problems for decentralized financing and provision of healthcare. Healthcare demands for most voters are infrequent and unpredictable; so it is relatively easy for most persons to live in a low-healthcare-service community and travel to a high-service community during periods of illness. This weakens the linkage between tax payments and services provided, which distorts voter demands for both. Such problems can be solved with various forms of supplemental insurance, but the need for such insurance is less associated with unitary systems because they are funded by national rather than local taxes. Overall, the political economy of tax-financed healthcare implies that decentralized systems tend to produce healthcare more efficiently with less rent-seeking and more opportunities for useful innovation than unitary systems. They are not perfect, but they have political and economic properties that make them superior to unitary systems.

Federalism and tax-financed healthcare  69

NOTES 1.

2. 3. 4. 5.

The simplest possible tools from economics are used to illustrate the logic of the analysis. For those interested in a mathematical analysis of median voter demands for tax-financed healthcare service, see Congleton et al. (2017) or Batinti and Congleton (2018). Those papers analyze unitary, not federal systems; however, the results fully characterize median voter demands and provide the basis for the intuitive microeconomics-based arguments developed herein. Perfect information is not required for this, although most voters must have representative samples of the relevant information. For more on this, see, for example, Congleton (2007). Mathematical models that demonstrate why state demands for healthcare tend to vary with age, location, information, and ideology are found in Congleton (2021). Empirical evidence that federal systems respond better to health emergencies is found in Shvetsova et. al (2021). A useful table of WHO data on longevity is available at: https://​en​.wikipedia​.org/​wiki/​List​_of​ _countries​_by​_life​_expectancy. Evidence of the effects of informational asymmetries on healthcare costs is, perhaps, most clearly demonstrated by the premiums that specialized doctors receive over general practitioners, whose training is similar to that of specialists but whose services are less understood by laypersons and often likely to be matters life or death, both of which tend to increase the bargaining power of specialized service providers. The OECD’s recent study of the remuneration of and migration of doctors clearly shows the premiums received by specialists. See www​.oecd​-ilibrary​.org/​sites/​ 0acc1895​-en/​index​.html​?itemId​=/​content/​component/​0acc1895​-en.

REFERENCES Batinti, A., & Congleton, R.D. (2018). On the codetermination of tax-financed medical R&D and healthcare expenditures: Models and evidence. European Journal of Political Economy, 54, 175–88. Congleton, R.D. (2007). Informational limits to democratic public policy: The jury theorem, yardstick competition, and ignorance. Public Choice, 132(3–4), 333–52. Congleton, R.D. (2021). Federalism and pandemic policies: Variety as the spice of life. Public Choice. https://​doi​.org/​10​.1007/​s11127​-021​-00915​-9. Congleton, R.D., Batinti, A., & Pietratonio, R. (2017). The electoral politics and the evolution of complex healthcare systems. Kyklos, 70(4), 483–510. Green, E.P. (2014). Payment systems in the healthcare industry: An experimental study of physician incentives. Journal of Economic Behavior & Organization, 106, 367–78. Makary, M.A., Overton, H.N., & Wang, P. (2017). Overprescribing is major contributor to opioid crisis. BMJ, 359, j4792. Oates, W.E. (1972). Fiscal Federalism. Harcourt Brace Jovanovich. Olson, M., Jr (1971). The Logic of Collective Action: Public Goods and the Theory of Groups, with a new preface and appendix (Vol. 124). Harvard University Press. Salmon, P. (1987). Decentralization as an incentive scheme. Oxford Review of Economic Policy, 3(2), 24–43. Salmon, P. (2019). Yardstick Competition among Governments: Accountability and Policymaking when Citizens Look Across Borders. Oxford University Press. Shvetsova, O., Van Dusky-Allen, J., Zhirnov, A., Adeel, A.B., Catalano, M., Catalano, O., Giannelli, F., Muftuoglu, E., Rosenberg, D., Sezgin, M.H., & Zhao, T. (2021). Federal institutions and strategic policy responses to COVID-19 pandemic. Frontiers in Political Science, 3. www​.frontiersin​.org/​ article/​10​.3389/​fpos​.2021​.631363. Tiebout, C.M. (1956). A pure theory of local expenditures. Journal of Political Economy, 64(5), 416–24. Wärneryd, K. (1998). Distributional conflict and jurisdictional organization. Journal of Public Economics, 69(3), 435–50.

6. The coordination in European Union healthcare after Covid-19 Marco Buso, Massimo Bordignon, Rosella Levaggi and Gilberto Turati

1. INTRODUCTION The issue of whether health policies should be centralized at the national level or decentralized to sub-national governments closer to citizens is the focus of a large literature, both theoretical and empirical, which has been surveyed by Ferrario, Levaggi and Piacenza (Chapter 7, this volume). However, besides reviving this debate on the optimal governance of single countries’ healthcare systems, the Covid-19 pandemic has prompted a new discussion about the optimal governance of healthcare systems between countries, calling for coordinated actions across countries and for a global governance. Many scholars have argued that, when facing a global health threat, such as a new virus causing severe health problems, like Covid-19, countries should give up part of their sovereignty in favor of international institutions that are asked to play an important role in coordinating local health and economic responses (e.g., Costa-Font et al., 2020). In this chapter we focus on the governance of health systems at a supra-national level. Our analysis concentrates on the case of the European Union (EU). As is well known, the EU is a union of 27 European countries (called Member States, MS from now on) established by the Maastricht Treaty since 1993. At present, the EU is a largely incomplete construction. It is certainly more than a customs union, with supra-national institutions such as the Commission, the EU Parliament and the EU Court of Justice, but it is not yet a federation. The main achievement of the EU across all MS has been on economic grounds, the establishment of an internal single market regulated by standardized laws and a common regulatory policy. A sub-group of MS further enhanced their cooperation, establishing a European monetary union, governed by a common central bank (the European Central Bank) now even entrusted with banking sector supervision. Roughly the same subset of EU countries has also eliminated internal borders with the Schengen Agreement. However, since the EU is not a federation, the bulk of competences remain national, implying fiscal policies that are also decided at national level. Indeed, policy coordination outside the areas of the single market and the single currency has proven to be difficult for the European Commission, the governing body of the EU. As for healthcare, according to Article 168 of the Treaty on the Functioning of the European Union, the EU shall complement national policies to prevent physical and mental illness/diseases, promote healthier lifestyles, facilitate access to better and safer healthcare, encourage cooperation between MS, deal with cross-border threats, keep people healthy throughout their lifetimes, and harness new technologies and practices. Hence, it is clear that the current role of the EU in healthcare is almost negligible, and mainly focused on the determinants of health rather than the provision of healthcare services, which is left entirely in the hands of MS. 70

The coordination in European Union healthcare after Covid-19  71 Given the current legislative framework, our aim in this chapter is to analyze whether a reallocation of healthcare policies from MS to the EU could lead to efficiency gains in healthcare systems. We start our analysis by considering whether the pandemic has caused a shift in the preferences of European citizens toward more coordination in healthcare. We then move to a normative analysis, considering benchmarking exercises both on the full set of functions concerning healthcare at the MS level, and on more specific sub-functions which can be influenced by coordinated actions at a higher level of governance. Using a standard benchmarking technique, data envelopment analysis (DEA), we show that there are no reasons to support the view that the whole governance of healthcare systems should be placed at a supra-national level. However, we find that the reallocation of some specific sub-functions, such as procurement and prevention, may allow MS to save financial resources and improve their performance. Our results hold in normal times but are even more relevant during a health crisis, such as a pandemic. When dealing with global threats characterized by relevant spillovers across MS, the involvement of the EU governing body may guarantee the effective implementation of national policies. In our empirical exercise, we show that the governance and the timing of policies are fundamental for an effective management of the crisis. In this regard, it is important to stress that the current decision-making process in the EU is often very difficult and time-consuming. This may explain the initial difficulties met by the European Commission in the implementation of the plan to procure vaccines against Covid-19 for all MS. To improve the effectiveness of the EU action in the future, it will be necessary to simplify procedures and allow EU institutions to take more autonomous decisions, by explicitly enlarging the set of functions attributed to the EU in terms of healthcare policies. The process of reallocation to the EU of some specific functions goes toward the construction of an EU health union. The chapter is organized as follows. Section 2 reports what European citizens think about an EU health union. Section 3 presents the DEA exercise and discusses the results. To further understand the potential gains from a coordinated action across MS, Section 4 analyzes the performance of the Covid-19 vaccine procurement plan developed by the European Commission. Section 5 concludes.

2.

POSITIVE ANALYSIS: A CHANGE IN PREFERENCES TOWARD AN EU HEALTH UNION?

The opinions of European citizens are regularly polled by the EU institutions through the Eurobarometer surveys, which aim to provide high-quality data for experts in public opinion and the general public alike. Given the limited role played by the EU in terms of healthcare policies, which are largely in the hands of MS (and sub-national governments in many of them), it does not come as a surprise that Eurobarometer surveys seldom enquire about this issue. However, since the beginning of the Covid-19 pandemic, the European Parliament has conducted three Eurobarometer surveys also dealing with topics related to healthcare: the standard semiannual Eurobarometer survey in spring 2021, the flash Eurobarometer survey “Attitudes on vaccination against Covid-19” in May 2021, and the special Eurobarometer survey “Future of Europe” in autumn 2021 (European Commission, 2021a, 2021b, 2022). Taken together, these surveys provide a consistent view in terms of a change in preferences by European citizens toward a larger role for the EU on healthcare policies.

72  Handbook on the political economy of health systems Consider first the results from the standard semiannual Eurobarometer survey in spring 2021. These results are reported in Figure 6.1. They show interesting patterns since the onset of the pandemic at the beginning of 2020 with respect to the trust in the EU. After several years of stability, 49 percent of Europeans trust the EU and 46 percent now have a positive image of the EU; in both cases, we observe a 6-point increase from the first to the second survey in 2020. On the contrary, trust in national governments (36 percent) and national parliaments (35 percent) decreases in the same period. In line with these trends, the support for a “European economic and monetary union with one single currency” has reached the highest level since 2004: 79 percent of citizens in the euro area, 70 percent of citizens in the EU.

Figure 6.1

QA6b of the standard Eurobarometer 94 (winter 2020–2021)

Interestingly, besides standard questions, the 2021 survey included specific sections on the Covid-19 pandemic. Considering the measures to fight the pandemic crisis taken at the EU level, 49 percent of Europeans are not satisfied (a 5-point increase since the previous survey in 2020), while 43 percent are satisfied (a 2-point decrease since the previous survey in 2020). People who are not satisfied are from large (Germany and France) or central European countries (Luxembourg, Belgium, Austria, Czech Republic, Slovakia), with the exception of Greece. However, considering measures taken by national governments, the percentage of people who are not satisfied is even larger, reaching 56 percent of Europeans and marking a 19-point increase since the previous survey in 2020. In line with these results, when looking at the confidence in the EU ability to make the right decisions in the future to tackle the pandemic, 59 percent of Europeans trust the EU (a 3-point decrease since the previous survey in 2020), while 39 percent do not trust the EU. As before, people who do not trust generally come from large and central European countries, while the highest percentage (55 percent) is reached in Greece.

The coordination in European Union healthcare after Covid-19  73 Concerning the priorities in the EU response to the Covid-19 pandemic, Figure 6.2 reports the most voted options by respondents. The most voted option by 36 percent of respondents is “Ensuring rapid access to safe and effective vaccines to all EU citizens”; but “Develop a European health policy” is the third most voted option by about one-quarter of the respondents. This last result is important since it shows that the higher trust in EU institutions translates also in the support of a larger role by the EU in healthcare policies, those more directly related with the Covid-19 crisis.

Figure 6.2

QA11 of the standard Eurobarometer 94 (winter 2020–2021)

The results of the standard survey have been confirmed in the Eurobarometer survey “Future of Europe,” where the “Risks related to health” are seen as the main global challenge for the future of the EU by 34 percent of respondents (the second most voted option after climate change and environmental issues). Moreover, in the same report, 22 percent of respondents mentioned “A common health policy” as the element considered the most helpful for the future of Europe (the second most voted option after comparable living standards). Finally, it seems clear that one of the reasons behind this change in preferences by EU citizens toward EU institutions is driven by the role the EU has played since the beginning of the crisis to ensure “access” to vaccines for all EU citizens. In Figure 6.3, results of the flash Eurobarometer survey show that the large majority of Europeans (70 percent) agree or tend to agree on the statement that the EU is playing a key role in providing access to vaccines. On the contrary, opinions on the way national governments and the EU have handled the vaccination campaign are mixed: respondents who are satisfied (dissatisfied) are, respectively, 46 percent

74  Handbook on the political economy of health systems (49 percent) for national governments and 47 percent (45 percent) for the EU. However, vaccination strategies are influenced by attitudes toward the vaccine. At the time of the survey, 37 percent of respondents were already vaccinated against Covid-19 and 32 percent would like to get vaccinated as soon as possible, while 9 percent would never get vaccinated. People who are not vaccinated mentioned as key reasons for not being vaccinated the belief that Covid-19 vaccines have not yet been sufficiently tested and worries about the side effects of the vaccines.

Figure 6.3

3.

Question of the flash Eurobarometer survey “Attitudes on vaccination against Covid-19” (May 2021)

NORMATIVE ANALYSIS: SHOULD THE EU HAVE A LARGE ROLE IN HEALTHCARE POLICIES?

In order to discuss the reallocation of competences between EU MS and the EU, we consider three steps: (1) we analyze the performance of MS healthcare systems before the Covid-19 pandemic, by computing the budgetary waste rate of MS in the health sector;1 (2) we discuss the performance of MS in specific health functions whose reallocation at the EU level could allow the reaping of additional benefits; (3) we analyze MS’ response to the Covid-19 pandemic given the inputs of their healthcare systems. 3.1

The Strategy and Data

For comparing health system performances of MS, we apply DEA, a standard benchmarking technique that allows one to estimate a “best practice” production frontier, defined as the minimum quantity of resources needed to attain a desired level of output (input-oriented); or, alternatively the maximum level of output that can be achieved given a fixed level of input

The coordination in European Union healthcare after Covid-19  75 (output-oriented). With respect to alternative benchmarking techniques, this approach has the advantage that it does not require one to assume any functional form for the production frontier, allowing one to derive the efficiency score for each Decision Making Unit (DMU), which corresponds in our analysis to MS. Efficient DMUs are those identifying the best practices. Moreover, DEA allows one to test for returns to scale characterizing the production set, and to decompose total technical efficiency into scale efficiency and pure technical efficiency (Banker et al., 1984; Simar and Wilson, 2000; Marselli and Vannini, 2004; Daraio and Simar, 2007). However, the DEA does not allow one to directly test the presence of spillover effects between MS, one of the key arguments, according to the literature on fiscal federalism, to define the optimal level of (de-)centralization of public goods (Tiebout, 1956; Oates, 1972, 2008; Boadway and Tremblay, 2012). Although DEA is primarily used to assess the efficiency of private firms in using inputs to produce outputs, this approach has been extended by several authors to evaluate the public spending performance of government in different sectors such as education and health (Herrera and Pang, 2005; Afonso and St. Aubyn, 2005; Sutherland et al., 2007; St. Aubyn et al., 2009; Afonso and Kazemi, 2017). In these exercises, the level of public spending in each sector is used as an input, while measures like student test scores and life expectancy are used as outputs or outcomes. Following these studies, here we first apply DEA to assess and compare the efficiency of health systems in the EU (27 MS) before the Covid-19 shock. Using a standard interpretation in the literature (e.g., Greene, 2004; Kumbhakar, 2010; Piacenza and Turati, 2014), we consider that inputs are consumed to produce intermediate healthcare services (output) that are used to improve citizens’ health (outcome). Therefore, in the analysis we implement two models: the first relates one input with one output, the second relates one input with one outcome. We use as input the ratio of public health spending (PPP) to global domestic product, or GDP (at market prices) (HS); as output the number of discharges per 1,000 inhabitants (disch); as outcome Life Expectancy (LE). From here onwards, the two models will be called, respectively, the input–output model and input–outcome model. To run these exercises, we collect data from Eurostat, and we use the average value of each variable for the period 2011–2018.2 Since we are interested in computing the budgetary waste rate of each MS, we adopt an input-oriented approach that derives the frontier finding the minimum amount of input needed to achieve a desired level of output/outcome. As a second step of our analysis, we investigate in more detail the two sub-functions “prevention” and “procurement of medical goods,” which clearly play a crucial role during a pandemic. To answer our research question – i.e., whether a reallocation of competences at the EU level can be beneficial in terms of more efficient health services for citizens – we decide to focus on output rather than outcome measures that depend on additional factors besides the health system organization. Thus, we use as input the ratio of either prevention or procurement health spending (PPP) to GDP (at market prices) (ProcS, PrevS), while as outputs we use some specific indicators: medical equipment per 100,000 inhabitants (including equipment like computed tomography scanners, magnetic resonance imaging units, gamma cameras, angiography units, lithotriptors, positron emission tomography (PET) scanners, radiation therapy equipment, mammographs) as procurement output (tech); and the percentage of people aged over 65 vaccinated against seasonal influenza as prevention output (vacc). From here onwards, the two models will be called, respectively, the procurement model and prevention model. To

76  Handbook on the political economy of health systems run these specific exercises, we use the same database built for the general analysis and we still adopt an input-oriented approach. Finally, after assessing the budgetary waste rate of health systems during normal times, we focus on the Covid-19 crisis. Our aim is twofold: (1) identify which health systems were more prepared to deal with the emergency and (2) assess how efficient were countries’ responses to the health crisis. Using the input-oriented DEA, we develop two models. In the first model we use as input variables reflecting structural differences in healthcare systems such as the number of doctors and the number of nurses per 1,000 inhabitants in the last available year (2018) (doct-nurs), while as output variables we use the (inverse of the) number of deaths per 100,000 inhabitants (deaths). In the second model we keep the same output, while as input we use the Oxford Covid-19 Government Response Tracker (OxCGRT), an indicator summarizing governments’ reaction to the pandemic (Hale et al., 2020).3 The underlying assumption is that policies such as stay-at-home orders are costly in terms of GDP losses (Deb et al., 2020). From here onwards, the two models will be called, respectively, the Covid-health model and Covid-response model. In running these additional exercises, we use data on health systems coming from Eurostat and data reflecting the Covid-19 situation from the European Centre for Disease Prevention and Control (ECDC) and from the OxCGRT website (Hale et al., 2020). Differently from before (when our interest was to assess the added value from moving competences at the EU level), in this case – given our aim to evaluate governments’ reactions to the pandemic – we include the United Kingdom (UK) in the sample of countries.4 We run a general analysis where we use as Covid data the average values over the entire pandemic period (from January 1, 2020, to December 14, 2020), and two specific analyses where we consider separately the first (from January 1 to June 30) and the second (from July 1 to December 14) wave of the pandemic. After applying the Simar and Wilson tests (2002), we run the analysis assuming that the technology is characterized by global Variable Returns to Scale (VRS).5 As a robustness check, we run the analysis using alternative outputs or outcomes. As we discuss below, the main conclusions of the analysis still hold using different variables. 3.2 Results In this section we study and discuss efficiency of health systems in normal times. We will then focus on the Covid-19 pandemic to investigate the resilience capacity of health systems in time of crisis. 3.2.1 Health systems efficiency in normal times We first apply the DEA to the general health function. We start by characterizing returns to scale: applying the tests developed by Simar and Wilson (2002), we find that the technology is globally VRS in both models.6 Assuming VRS, we then compute efficiency scores for the two models (Figures 6.4 and 6.5). Looking at the figures it is possible to note that there are countries that are efficient in producing outputs (i.e., Austria, Germany), but not outcomes, as well as countries very efficient in terms of outcomes (i.e., Italy, Spain), but not in terms of outputs; indeed, the rank correlation between the two models is quite low (0.43). This result is not surprising as it is well known from the literature that health outcomes depend on other factors besides health

The coordination in European Union healthcare after Covid-19  77 systems, such as citizens’ (un)healthy behaviors (Nixon and Ulmann, 2006; Álvarez-Gálvez and Jaime-Castillo, 2018; Palmer et al., 2019).

Figures 6.4 and 6.5

Input–output model and input–outcome model

Finally, DEA allows us to decompose total technical efficiency (TTE) into pure technical (PTE) and scale efficiency (SE). Considering the input–output model, and computing the population-weighted mean of efficiency scores, we find that, on average: TTE is equal to 0.57, PTE is equal to 0.43 and SE is equal to 0.81. However, considering the input–outcome model, we find that, on average: TTE is equal to 0.57, PTE is equal to 0.47 and SE is equal to 0.79. Both models exhibit high levels of SE, suggesting that a reallocation of competences to a different scale would not be desirable. This result is not surprising as the largest share of total spending is devoted to curative care services that are characterized by very limited economies of scale.7 As a second step of the analysis, we implement the procurement model and prevention model.8 As before, after applying the two tests developed by Simar and Wilson, we conclude that the technology is globally VRS considering both the procurement and the prevention sub-function.9 Figure 6.6, reporting population-weighted means of efficiency scores, proposes a comparison between procurement and prevention sub-functions, and the general health function in terms of TTE, PTE and SE. From the figure we can extrapolate two main insights. First, efficiency is generally lower in the procurement and prevention sub-functions than in the general health function. Second, differently from results concerning total health spending, SE is much lower when looking at prevention and procurement. In addition, from the Simar and Wilson tests on returns to scale, we know that the VRS assumption is accepted against the Constant Returns to Scale (CRS) and the Non-Increasing Returns to Scale (NIRS) assumptions (Simar and Wilson, 2002). Following this evidence, we can conclude that the reallocation of some prevention and procurement competences to an institution operating at a larger scale could allow MS to save financial resources, while producing the same level of outputs. These results

78  Handbook on the political economy of health systems are in line with the literature (e.g., Bandiera et al., 2009; Baldi and Vannoni, 2017) and confirm recommendations included in the document prepared by the expert panel for the European Commission on the effective ways of investing in health (EXPH, 2016).

Figure 6.6

Efficiency scores: a comparison

3.2.2 Health systems’ efficiency in time of crisis To test the resilience of health systems in time of crisis, we run a DEA analysis using Covid-19 data. Results in terms of efficiency scores for the Covid-health model and Covid-response model are reported, respectively, in Figures 6.7 and 6.8.

Figure 6.7

Covid-health model

The coordination in European Union healthcare after Covid-19  79 Figure 6.7 shows that countries that experienced the worst results in terms of deaths per 100,000 inhabitants (Italy, Spain, France) have a lower number of doctors or nurses per 1,000 inhabitants, while some countries that have succeeded in limiting the number of deaths, such as Germany, Finland and Denmark, are the ones with the highest number of doctors and nurses per 1,000 inhabitants.10 This result suggests that countries may decide to tolerate some inefficiencies in their health systems to create a buffer of healthcare inputs to use in time of crisis.

Figure 6.8

Covid-response model

Figure 6.8 shows that countries with high levels of deaths per 100,000 inhabitants (like Italy or Spain) were almost forced to implement stringency policies, which are costly in terms of GDP losses, resulting in a low level of efficiency. The most efficient countries are those that applied less stringent policies with respect to countries like Italy and Spain, obtaining the same (or even lower) levels of outputs in terms of mortality, such as Sweden; or those that applied the same level of stringency policies obtaining better outputs, such as Germany and Denmark. Figure 6.8 also shows some differences between the first and the second wave of the pandemic. For instance, countries such as Sweden, Greece and Germany performed well during the first wave, and they recorded a bad performance during the second wave of the pandemic. Following the recent literature on the topic, we are interested in better understanding what may affect the level of efficiency experienced by countries when applying stringency measures. Thus, we perform a second stage regression to control for the determinants of efficiency scores computed through the input–outcome model. We apply the two-stage estimator developed by Simar (2007); a bootstrap procedure that yields Standard Errors and confidence intervals that are not biased by the correlation with the estimated efficiency scores. According to the most recent literature (e.g., Bjørnskov and Voigt, 2020), we consider as potential determinants of efficiency scores: the World Bank Governance Indicator11 and, for each wave of the pandemic, the number of days between the date with the highest level of the stringency index and the date with the highest number of deaths. As additional controls we include the

80  Handbook on the political economy of health systems population percentage of daily smokers, overweight and elderly people and a dummy variable indicating Eastern European countries. From the regression analysis, we can extrapolate two relevant results. First, the level of the World Bank Governance Indicator (thus, the higher quality of institutions) positively affects the efficiency of stringency measures, a result in line with findings by Bjørnskov and Voigt (2020). This means that MS characterized by a high level of the World Bank Governance Indicator could have achieved the same or better outcomes of MS characterized by a low level of the same index, while applying less stringent measures.12 This result is also in line with De Luca et al. (2021), who identify a positive effect of institutional quality on the appropriateness of healthcare provision in Italian hospitals (see also Chapter 18 in this volume). Second, during the second wave of the pandemic, efficiency scores are negatively affected by the time span between the introduction of stricter measures and the peak in the number of deaths. This suggest that stringency measures should be used as soon as possible to reduce the peak of deaths, which influence the duration of those same measures. Finally, comparing Figures 6.7 and 6.8, we can note that results are very different. Indeed, the rank correlation between the two models is equal to 0.2095 (0.0038 first wave, 0.1951 second wave). This result, together with the previous evidence, may suggest that among EU MS, countries with a good governance or with over-capacity of healthcare inputs (doctors, nurses) are also the ones that had the opportunity to implement less stringent policies, limiting GDP losses.

4.

THE EU VALUE-ADDED: COVID-19 VACCINE PROCUREMENT AS A FIRST TEST OF EU POLICIES

At the beginning of this chapter, we raised the question: “Should the EU have a larger role in healthcare policies?” According to our empirical analysis, if we consider the aggregate spending in health, the answer is no. Indeed, the largest fraction of total spending is used for curative care, a sub-function that is characterized by limited economies of scale and limited spillovers, and that depends on local characteristics (population density, age profile, etc.). However, when looking at specific sub-functions, such as procurement and prevention, the answer changes. Reallocating, partially or totally, these competences at a supra-national level may allow MS to save financial resources and improve their performance. In this regard, a first relevant test of the value-added that might be gained by giving more competences to the EU is given by the common strategy recently developed by the European Commission to procure the Covid-19 vaccine for all MS. The plan of the European Commission had two main objectives: first, accelerate the development, manufacturing and deployment of safe and effective vaccines against Covid-19; second, the financing of purchase agreements with vaccine manufacturing companies. Thanks to a common procurement strategy, MS can exploit economies of scale, thus achieving significant cost savings, and they do not need to compete one against each other to get the needed doses of vaccine. Since June 2020, the EU has started to sign preliminary arrangements with pharmaceutical companies to ensure the future supply of vaccines. By December 2021, four vaccines had been authorized by the European Medicines Agency (EMA): BioNtech–Pfizer, Moderna, AstraZeneca, Johnson & Johnson; the EU concluded agreements with these companies for

The coordination in European Union healthcare after Covid-19  81 the procurement of up to 3.6 billion doses (2.4 billion from BioNtech–Pfizer, 460 million from Moderna, 400 million from AstraZeneca, 400 million from Johnson & Johnson). In addition, the EU signed further preliminary arrangements with three companies (Sanofi–GSK, CureVac, Novavax) for the purchase of vaccines against Covid-19 once they were authorized by the EMA.13 Compared to other developed countries, such as the UK and the USA, the EU launched its procurement strategy with some delay, but it reached most of its initial objectives after the first year. By the beginning of August 2021 more than 70 percent of EU citizens had received at least one dose of vaccine among adults, while more than 60 percent had received the complete cycle of vaccination. Clearly, we do not have a counterfactual to evaluate the improvement brought about by a common EU procurement, and we do not even have information to estimate the saving of costs as much information is still confidential. However, looking at world experiences, it is evident that middle- and low-income countries had serious problems in accessing the vaccine due to the unfair competition with richer countries. As emphasized by the president of the European Commission, Ursula von der Leyen, during a speech at the European Parliament (February 10, 2021), the common procurement strategy avoided the possibility that some countries in the EU would be left without doses of vaccines (“empty-handed”), an event that would have meant the end of the European unity. Comparing final achievements and failures, we can conclude that this important test for a common procurement strategy was fulfilled. However, it is important to understand what went wrong and what needs to be improved. First, the starting delay was due to a problem of governance. The presence of a supra-national EU together with an intergovernmental EU makes the decision-making procedures time-consuming and difficult (Bordignon, 2020). Second, differently from other countries, such as the USA, the EU did not invest enough financial resources to promote the local development and manufacturing of vaccines. Besides being beneficial for the local economy, the presence of internal production could have accelerated the vaccination process and made the supply of vaccines less dependent on the multinational pharmaceutical companies. The lesson for the future is twofold. First, the EU should be more autonomous from MS in taking decisions, especially in an emergency situation when time is crucial. Second, the EU should create an institution like the Biomedical Advanced Research and Development Authority (BARDA) in the USA for dealing with situations of health crises. The existence of such an institution with the needed spending autonomy could have made the implementation of the investment plan developed by the European Commission more effective. The European Commission has already taken some steps in this direction by establishing the new European Health Emergency Preparedness and Response Authority (HERA), which in February 2022 launched its first work plan with €1.3 billion for preparedness and response to health emergencies in 2022. Contributions to the HERA budget come from EU4Health (€275 million), Horizon Europe (€395 million) and UCPM/rescEU14 (€630 million), and are used to finance six tasks/areas: threat assessments and intelligence gathering, promoting advanced R&D of medical countermeasures and related technologies, addressing market challenges and failures and boosting the Union’s open strategic autonomy, ensuring the provision of medical countermeasures, strengthening knowledge and skills, and the international dimension (see the HERA Work Plan 202215).

82  Handbook on the political economy of health systems

5. CONCLUSIONS In the 2021 Eurobarometer surveys, Europeans collected data on preferences for the development of a European health policy. Respondents to the surveys indicate that the EU should establish a strategy to prevent and be prepared for facing health crises in the future and invest more money to develop treatments and vaccines. In this chapter we use a benchmarking technique to estimate whether a reallocation of health policies from a national to the EU level may be beneficial for MS. Using total spending in health, the analysis shows that economies of scale are very limited; thus the health function should be kept at a national or local level. However, when looking at some specific sub-functions, such as procurement and prevention, the analysis concludes that a reallocation to the EU level may generate efficiency gains, even without accounting for positive spillovers that heavily characterize prevention. Therefore, the study supports the idea that the EU should be more involved in health policies, especially for what concerns the preparation for future health crises. The empirical analysis has also shown that in a situation of emergency the governance and the timing of decisions is fundamental to obtain successful outcomes. The slow decision making was the main problem of the European Commission in adopting the common plan to procure vaccines against Covid-19 for all MS. A precondition for the future creation of a health union is the autonomy in decision making of EU institutions from MS.

NOTES 1. This is the same approach used by Saulnier et al. (2020) in the report for the European Parliament Improving the Quality of Public Spending in Europe: Budgetary ‘Waste Rates’ in EU Member States. In the report, the waste rate is defined as the amount of resources that could have been saved if a particular policy area and relative funding was allocated to the European level rather than remaining in the hands of MS. 2. For what concerns the input–output model, we do not consider in the analysis data from Bulgaria because the number of discharges in Bulgaria is available only for the year 2018; thus the country is not comparable with other MS for which we consider an average value of the last eight years. 3. OxCGRT collects publicly available information on 18 indicators of government responses. Eight of the policy indicators record information on containment and closure policies; four of the indicators record economic policies, such as income support to citizens or provision of foreign aid; and six of the indicators record health system policies such as the Covid-19 testing regime or emergency investments into healthcare. 4. The UK is comparable to European countries for what concerns the reaction to the pandemic. However, excluding the UK from the analysis does not alter our main conclusion. 5. In both the Covid-health model and Covid-response models, the p-values of the null hypotheses that global technology is CRS (test 1) or NIRS (test 2) are very small (0.00), meaning that VRS is the correct assumption for the technology. 6. Simar and Wilson (2002) propose two tests. The first test compares the hypothesis that the technology is globally constant returns to scale (CRS) versus the hypothesis that the technology is variable returns to scale (VRS). The second test compares the hypothesis that the technology is globally not-increasing returns to scale (NIRS) versus the hypothesis that the technology is VRS. In the input–output and input–outcome models, both the p-value of the null hypothesis that global technology is CRS (test 1) and the p-value of the null hypothesis that the technology is NIRS are very small, meaning that VRS is the correct assumption for the technology.

The coordination in European Union healthcare after Covid-19  83 7.

Nonetheless, it is important to remark that in the analysis we do not test the role of spillover effects between MS, an element that could have suggested the reallocation of competences at the EU level, even in the presence of limited returns to scale. However, most of the spending in modern healthcare systems is for curative care services, characterized by limited spillovers, and not preventive care services. 8. Results related to these two models are available upon request from the authors. 9. In both models, the p-value of the null hypothesis that global technology is CRS (test 1) is very small (0.00), meaning that CRS is not an appropriate assumption. Similarly, also the null hypothesis that the technology is NIRS is rejected (p-value 0.00), meaning that VRS is the correct assumption for the technology. 10. These results are not reported here for brevity but are available upon request from the authors. 11. The World Bank Worldwide Governance Indicator is obtained by summing up for each country the score obtained in the different dimensions surveyed by the World Bank: voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law and control of corruption (Sapir 2020). The World Bank Governance Indicator is available for over 200 countries over the period 1996–2020, and it is particularly suggested for cross-country analyses. 12. For instance, comparing two countries with high and low levels of the governance indicator, such as Denmark and Italy, the latter experienced more deaths per 100,000 inhabitants although its level of the OxCGRT is higher than in Denmark. 13. These amounts refer to the first agreements signed during 2021. Thereafter, the EU has signed further agreements with the same and other companies. 14. EU4Health, Horizon Europe and the EU Civil Protection Mechanism (UCPM)/rescEU are three different EU programs connected to the tasks assigned to HERA, the new authority designed to strengthen the EU’s health emergency response and preparedness. 15. See https://​health​.ec​.europa​.eu/​system/​files/​2022​-02/​hera​_work​-plan​_2022​_en​_0​.pdf.

REFERENCES Afonso, A., and Kazemi, M. (2017) “Assessing Public Spending Efficiency in 20 OECD Countries”. In: Boekemeier, B., and Greiner, A., Eds, Inequality and Finance in Macrodynamics, Cham, Springer, 7–42. Afonso, A., and St. Aubyn, M. (2005) “Non-parametric approaches to education and health efficiency in OECD countries”, Journal of Applied Economics, VIII(2): 227–46. Álvarez-Gálvez, J., and Jaime-Castillo, A.M. (2018) “The impact of social expenditure on health inequalities in Europe”, Social Science & Medicine, 200: 9–18. Baldi, S., and Vannoni, D. (2017) “The impact of centralization on pharmaceutical procurement prices: The role of institutional quality and corruption”, Regional Studies, 51(3): 426–38. Bandiera, O., Prat, A., and Valletti, V. (2009) “Active and passive waste in government spending: Evidence from a policy experiment”, American Economic Review, 99(4): 1278–308. Banker, R.D., Charnes, A., and Cooper, W.W. (1984) “Some models for estimating technical and scale inefficiencies in data envelopment analysis”, Management Science, 30(9): 1078–92. Bjørnskov, C., and Voigt, S. (2020) This Time Is Different? – On the Use of Emergency Measures During the Corona Pandemic, ILE Working Paper Series, No. 36, University of Hamburg, Institute of Law and Economics (ILE), Hamburg. Boadway, R., and Tremblay, J. (2012) “Reassessment of the Tiebout model”, Journal of Public Economics, 96(11/12): 1063–78. Bordignon, M. (2020) “La crisi pandemica e il futuro delle istituzioni europee”, Osservatorio Monetario, 2/2020, Università Cattolica del Sacro Cuore. Costa-Font, J., Turati, G., and Batinti, A. (2020) The Political Economy of Health and Healthcare: The Rise of the Patient Citizen, Cambridge: Cambridge University Press. Daraio, C., and Simar, L. (2007) Advanced Robust and Nonparametric Methods in Efficiency Analysis: Methodology and Applications, New York, Springer.

84  Handbook on the political economy of health systems Deb, P., Furceri, D., Ostry, J.D., and Tawk, N. (2020) “The economic effects of COVID-19 containment measures”, VOXEU/CEPR, https://​voxeu​.org/​article/​economic​-effects​-covid​-19​-containment​ -measures. De Luca, G., Lisi, D., Martorana, M., and Siciliani, L. (2021) “Does higher institutional quality improve the appropriateness of healthcare provision?”, Journal of Public Economics, 194: 104356. European Commission (2021a) “Eurobarometer survey: Attitudes on vaccination against Covid-19”, European Commission, June. European Commission (2021b) “Standard Eurobarometer 95”, European Commission, September. European Commission (2022) “Eurobarometer survey: Future of Europe”, European Commission, January. EXPH (2016) Report on Access to Health Services in the European Union, Brussels, Expert Panel on Effective Ways of Investing in Health. Greene, W. (2004) “Distinguishing between heterogeneity and inefficiency: Stochastic frontier analysis of the World Health Organization’s panel data on national health care systems”, Health Economics, 13(10): 959–80. Hale, T., Webster, S., Petherick, A., Phillips, T., and Kira, B. (2020) “Oxford COVID-19 Government Response Tracker”, Blavatnik School of Government. Herrera, S., and Pang, G. (2005) Efficiency of Public Spending in Developing Countries: An Efficiency Frontier Approach, World Bank Policy Research Working Paper 3645. Kumbhakar, S. (2010) “Efficiency and productivity of world health systems: Where does your country stand?”, Applied Economics, 42(13): 1641–59. Marselli, R., and Vannini, M. (2004) L’efficienza tecnica dei distretti di Corte d’Appello italiani: aspetti metodologici, benchmarking e arretrato smaltibile, Working Paper CRENoS 200409, Sardinia, Centre for North South Economic Research, University of Cagliari and Sassari. Nixon, J., and Ulmann, P. (2006) “The relationship between health care expenditure and health outcomes”, European Journal of Health Economics, 7: 7–18. Oates, W.E. (1972) Fiscal Federalism, New York, Harcourt Brace Jovanovich. Oates, W.E. (2008) “On the theory and practice of fiscal decentralization”, Institutional Foundations of Public Finance: Economic and Legal Perspectives, 165: 176. Palmer, R.C., Ismond, D., Rodriquez, E.J., and Kaufman, J.S. (2019) “Social Determinants of Health: Future Directions for Health Disparities Research”, American Journal of Public Health, 109: S70–S71. Piacenza, M., and Turati, G. (2014) “Does fiscal discipline towards subnational governments affect citizens’ well-being? Evidence on health”, Health Economics, 23(2): 199–224. Sapir, A. (2020) “Why has COVID-19 hit different European Union economies so differently?” Policy brief, Contribution 2020/18, Bruegel. Saulnier, J., et al. (2020) Improving the Quality of Public Spending in Europe: Budgetary ‘Waste Rates’ in EU Member States, Brussels, European Parliamentary Research Service. Simar, L. (2007), “Estimation and inference in two-stage, semi-parametric models of production processes”, Journal of Econometrics, 136: 31–64. Simar, L., and Wilson, P.W. (2000) “Statistical inference in nonparametric frontier models: The state of the art”, Journal of Productivity Analysis, 13: 49–78. Simar, L., and Wilson, P.W. (2002) “Non-parametric tests of returns to scale”, EJOR, 139(1): 115–32. St. Aubyn, M., Pina, A., Garcia, F., and Pais, J. (2009) “Study on the efficiency and effectiveness of public spending on tertiary education”, European Economy – Economic Papers 2008–2015, 390, Brussels, Directorate General Economic and Financial Affairs, European Commission. Sutherland, D., Price, R., Joumard, I., and Nicq, C. (2007) Performance Indicators for Public Spending Efficiency in Primary and Secondary Education, OECD Economics Department, Working Paper 546, 2007. Tiebout, C.M. (1956) “A pure theory of local expenditures”, Journal of Political Economy, 64(5): 416–24.

7. Efficiency and equity effects of healthcare decentralization: evidence from Italy Caterina Ferrario, Rosella Levaggi and Massimiliano Piacenza

1.

ECONOMIC THEORY OF DECENTRALIZATION

Public policy making at national and supranational level must rely on delegated choices in which autonomous decision makers are charged with the responsibility for a specific competence, which may concern either policy design or policy implementation, or both. Policy implementation at decentralized levels may vary from goods or services production to market regulation. A significant body of economic literature has investigated the rationale for economic decentralization, providing both normative and positive perspectives on this policy issue. These contributions are generally classified into two strands, commonly known as firstand second-generation models of fiscal federalism. 1.1

First-Generation Fiscal Federalism

Traditional literature on fiscal federalism, or, as it is generally referred to, first-generation theory of fiscal federalism (Oates, 1972, 2005; King, 1984), suggests that under certain circumstances, decentralization increases public sector efficiency. Therefore, under a normative perspective, decentralization is justified for efficiency reasons (Rosen, 1988; Tresch, 2002). Efficiency is enhanced when expenditure decisions concerning local public goods are left to the tier which is better informed on local preferences, allowing for differentiation in local provision (Oates, 1972), while intergovernmental grants might be used for equity and efficiency reasons. Furthermore, fiscal federalism should increase efficiency by inducing some interjurisdictional competition among political powers resulting from “voting with the feet” (Tiebout, 1956) or yardstick competition (Besley and Case, 1995). In addition, efficiency and effectiveness may also stem from diversity and experimentation in public policies allowed by decentralization (the so-called “laboratory federalism”; Oates, 1999, p. 1132). Although the efficiency gains from decentralization are well recognized, in practice for them to be fully exploited, local governments would need their own adequate financial resources, and this opens up the problem of identifying the appropriate tax sources for decentralized levels of government, the so-called “tax assignment” problem (McLure, 1983). Intergovernmental grants may contribute to finance local governments’ expenditure, when their own tax revenues are not enough. Grants also fulfil additional functions, namely: correction of externalities (spillover grants), fiscal equalization (grants pursuing horizontal equity), achievement of minimum service levels (according to grantors’ decisions and preferences), insurance for external shocks affecting decentralized governments, compensation of vertical fiscal imbalances (revenue-sharing grants) and improvement of the overall tax system (Oates, 1999). 85

86  Handbook on the political economy of health systems Overall, first-generation models of fiscal federalism generally depend on stringent assumptions, primarily demand heterogeneity, the absence of economies of scale in public production and benevolent social planners. Relaxing these hypotheses does not guarantee that their main results still hold. Nevertheless, these models are highly significant in that they first identified an economic rationale for decentralization and opened the way to a discussion of the relative advantages of decentralization on economic grounds. 1.2

Second-Generation Fiscal Federalism

Second-generation theory of fiscal federalism (Oates, 2005) is more recent than first-generation models and has more of a positive rather than normative approach, although the distinction is not always clear-cut (Weingast, 2009). Second-generation models acknowledge that centralization allows greater coordination, but decentralization may be efficiency-enhancing (Alonso et al., 2008). However, the success of fiscal federalism in terms of efficiency gains is conditional to a wide array of factors which should be better investigated and understood. These factors are primarily linked to economic agents’ information and to fiscal and political incentives facing local and central politicians as well as local lobbies. These factors become significant because second-generation models relax the key assumption that politicians are benevolent social planners and maximize social welfare (Weingast, 2009) and analyse a setting of imperfect information and control. In this new framework, the comparative advantages of decentralization must be assessed, taking into account public officers’ objective functions and the fiscal and political incentives they face. For this purpose, second-generation fiscal federalism builds on results from different strands of economic literature, besides agency theory, economics of information, the new theory of the firm, organization theory and the theory of contracts (Oates, 2005). Under this perspective, the success of fiscal federalism is affected by important agency problems (e.g., rent seeking, lobbying) and this marks a significant difference from first-generation models. In first-generation models, citizens (the agents) have private information on their demand function for local public goods and decentralization could solve the problem of preference revelation for central government (the principal), allowing for optimal provision of local public goods by subnational jurisdictions. In second-generation models, the relevant asymmetry of information is rather between local politicians (the agents), from one side, and central government and citizens-voters (the principals), on the other side. Inefficiencies may arise due to local politicians’ rent-seeking behaviour or by their capture by local lobbies (Bardhan, 2002), although rent seeking may be limited by yardstick competition and mobility (Besley and Case, 1995). Therefore, the comparative advantages of decentralization depend on the information the agents possess; for instance, about specific parameters (Levaggi and Smith, 1994; Snoddon and Wen, 2003; Akai and Mikami, 2006; Levaggi and Levaggi, 2016). Finally, the demand for decentralization may itself depend on the choice of the instruments the regulator is using to reduce horizontal fiscal imbalance (Levaggi and Menoncin, 2017). These problems have been widely studied in literature and suggest a trade-off between autonomy and control: the local level is better informed than the centre about the relevant parameters that affect welfare and it can strategically use such information. Central government should then balance the improvement in welfare from decentralization with the cost deriving from asymmetric information. At central level the problem is essentially that of finding the best equilibrium between autonomy, control (Levaggi, 2002, 2007, 2010) and the

Efficiency and equity effects of healthcare decentralization  87 need to reduce the drawbacks of asymmetry of information. In fact, on the one hand the local government (being better informed on preferences) is the better tier to choose local services (Petretto, 2000); however, such information may be used to pursue interests that are in contrast with either national welfare, or even with the maximization of welfare altogether (Wildasin, 2004; Crivelli and Staal, 2013). Even with benevolent governments, local initiatives’ coordination may be needed because of the presence of spillovers (Besley and Coate 2003; Ogawa and Wildasin, 2009). Empirical evidence is, however, less clear-cut and has revealed some drawbacks of decentralization; for instance, the increase in the number and size of bureaucracies (Reverte-Cejudo and Sánchez-Bayle, 1999; Repullo, 2007). In this new framework, which accounts for self-interest and opportunistic behaviours, moral hazard problems may arise. Local officials may take advantage of soft budget constraints (Weingast, 2009) and expand local expenditures beyond the efficient level, if the central government cannot commit not to bail out their deficits. In addition, efficiency gains are highly dependent on the shape and extent of local governments’ fiscal interactions, both horizontal and vertical. Horizontal interactions among local governments may take the form of either tax competition or yardstick competition, or finally may be connected with externalities. When competition is for mobile factors of production, the suboptimal outcome may be a race to the bottom, where tax levels decrease and mobile factors’ owners appropriate all efficiency gains. Vertical competition involves different levels of government, and may concern tax and expenditures or intergovernmental transfers. Vertical competition may specifically arise when fiscal arrangements contemplate tax base sharing, when taxes levied by other levels of government have an impact on local demand and when goods and services provided by other levels of government are complements or substitutes to those locally offered. In second-generation models, the role of fiscal transfers also changes. They are no more simple tools designed to compensate vertical and horizontal fiscal imbalances and thus to enhance efficiency and equity, but rather the design of intergovernmental transfer schemes becomes critical to align incentives between local and central policy makers under asymmetric information (agency problems), in a context where the importance of local government tax autonomy is recognized. In particular, second-generation theory of fiscal federalism emphasizes the importance of local revenues for subnational governments’ financing, arguing that the higher the alignment of spending and funding responsibilities obtained through a greater tax autonomy, the stronger the electoral accountability of local politicians and their incentives to efficiently manage spending (see, for example, Qian and Weingast, 1997; Oates, 2005; Weingast, 2009; Boetti et al., 2012; Eyraud and Lusinyan, 2013). In addition, the design of transfer schemes is analysed for its potential to mitigate spillover problems, help control local debt growth (Martell and Smith, 2004; Huber and Runkel, 2008) and help solve problems of mobility and efficiency. Therefore, the key role of intergovernmental transfers with a hierarchical governance is acknowledged. Transfers may be either vertical (generally top-down, but in principle also bottom-up) or horizontal (within the same level of government). Vertical transfers may be used to compensate for vertical fiscal imbalances, to equalize local fiscal endowments, to internalize externalities and spillovers, to create the right incentives for subnational governments and to control subnational spending, to provide insurance for lower levels of government against external shocks, or to pursue redistributive objectives. A final relevant issue for second-generation theory concerns the wider institutional framework, which is deemed of critical importance. As Weingast acknowledges (2009, p. 280): “local governments exist within a complex set of institutional arrangements, with political,

88  Handbook on the political economy of health systems legal-constitutional, and economic aspects”, which shape individual actions and interactions, and therefore affect decentralization outcomes and the performance of decentralized governments. The institutional arrangements and political processes behind decentralization are widely studied (Rubinchik-Pessach, 2005; Mookherjee, 2015) and have an impact on the relative efficiency of decentralized systems. For instance, they shape the way conflicts of interest between citizens of different jurisdictions are solved (Besley and Coate, 2003) and define the voting rules that make decentralization welfare-improving (Rubinchik-Pessach, 2005). In conclusion, second-generation models of fiscal federalism analyse a wide array of problems and instances, in a growing body of literature, to an extent that it has been maintained that there is a need for a new political economy of fiscal federalism (Inman and Rubinfeld, 1997) and to systematize all these contributions. Both first- and second-generation theories of fiscal federalism, however, suffer from two main limitations. First, these models are generally biased towards efficiency concerns. The comparative advantages of decentralization are mostly assessed under an efficiency perspective. Conversely, equity considerations are generally disregarded, though according to some authors decentralization can lead to an uneven geographical distribution of its benefits (e.g., Martinez-Vazquez and McNab, 2003). However, evidence from empirical studies is scarce and results are mixed (McKinnon, 1997; Qian and Weingast, 1997; Cheshire and Gordon, 1998; Shankar and Shah, 2003; Gil et al., 2004; Rodríguez-Pose and Ezcurra, 2010; Ferrario and Zanardi, 2011; Di Novi et al., 2019). A second important point is that, although sophisticated in its modelling approach, most of the literature models decentralized decisions assuming that the good to be provided is a local public good with spillovers (significant exceptions are Wildasin, 2001; Ogawa and Wildasin, 2009). In the presence of a (local) public good, government intervention is justified either because a market does not exist or for efficiency reasons. The prevailing focus on public goods and the disregard for other types (private, common, club, merit goods, etc.) is a legacy of the early literature on fiscal federalism. That literature was indeed developed along these lines, mostly because they reflected the actual scope for government at that time and the characteristics of most of the goods and services traditionally produced by government. However, later on, with the development of the welfare state, the nature of government has changed and, with it, also the goods that are produced. Nowadays, most of the services produced at local level are either impure public goods or merit goods (Levaggi and Levaggi, 2016). The former are both private goods (positive correlation between utility and quantity actually bought) and public goods (increasing utility for the entire amount produced), and they are more efficiently produced under decentralization or a limited form of fiscal federalism (Levaggi, 2007); the latter are private goods whose consumption is financed by the government for equity/redistribution purposes. Examples are refuse collection, local transport, sporting facilities, healthcare and education (Cornes and Sandler, 1994). Merit goods can be made available to a specific community either by producing them or by allowing people to receive them outside the local authority boundaries. The main examples of goods falling into this category are education and healthcare, for which public provision is in practice driven more by equity than by efficiency concerns and also follows the need to ensure equal citizenship rights to all citizens in a country. Fiscal federalism in this context has particular characteristics that have not yet received due attention. Contrary to local public goods, although the price of merit goods is subsidised through general taxation, the quantity to be produced is set by the market. Government can use only indirect instruments to produce an optimal allocation, but the quantity of information necessary to achieve a first-best result is

Efficiency and equity effects of healthcare decentralization  89 usually not accessible and only second-best solutions can be attained. In the following these specific characteristics of healthcare provision under decentralization are analysed, but, before that, the reasons for healthcare decentralization are discussed.

2.

HEALTHCARE DECENTRALIZATION: THEORETICAL PERSPECTIVES

2.1

The Economic Rationale for Healthcare Decentralization

One of the key issues in fiscal decentralization is the optimal allocation of government functions to the different levels of government, the so-called “assignment problem” (Oates, 2005; Boadway et al., 2008). Efficiency criteria are pivotal for first-generation fiscal federalism, which emphasizes demand heterogeneity, technology of production, externalities and fiscal competition as the key factors to be taken into account when assigning competencies. Second-generation models have added agency problems to this list (Tommasi and Weinschelbaum, 2007); for instance, lobbying behaviour under different institutional configuration (Bordignon et al., 2008). Healthcare provision is quite often decentralized (Saltman et al., 2007; Costa-Font and Greer, 2013), either in federal countries (e.g., Canada and Australia) or in regional states (e.g., Spain), or even in unitary but decentralized countries (e.g., Sweden). However, the institutional setting of decentralized healthcare varies extensively across countries (Levaggi and Smith, 2005; Saltman et al., 2007) and the impact of decentralization on healthcare system performance is widely debated (Schneider, 2003; Saltman et al., 2007; OECD, 2014; Toth, 2014; Cavalieri and Ferrante, 2016; Marchildon and Bossert, 2018). Decentralization may concern both the revenue and the expenditure side. As for expenditure, central government may transfer only spending competencies to decentralized governments (spending decentralization) or also legislative powers (political devolution). On the revenue side, fiscal autonomy may be granted to decentralized governments by assigning them their own tax revenue (fiscal decentralization). Therefore, in many countries, public health policy results from the interaction of various levels of governments (Greer and Da Fonseca, 2015; Greer et al. 2016). However, it has not always been the case. Before the development of the modern welfare state, healthcare was provided at the local level by charity or religious associations, and later by mutual funds. With the birth of the modern welfare state, healthcare was generally centralized, to better pursue universality and uniformity of provision, in a context where the major concern was effectiveness in ensuring basic care to all citizens. However, pre-existing providers of healthcare did not cease to operate, and in many countries healthcare services were centrally planned but locally organized. Later, due to the over-expansion of public sector welfare spending, healthcare reforms became a critical issue and the advantages of decentralization under an efficiency perspective were recognized. At the same time, decentralization was advocated for it could allow differentiation, competition, higher accountability of decision makers and better alignment of services offered to local needs and specificities, together with experimentation. However, from a theoretical perspective, the significant externalities from local healthcare services should be considered when assessing the advantages of decentralization. In addition, efficiency arguments for decentralization generally apply to pure public goods, while healthcare services are rather merit goods. For these goods there is seldom

90  Handbook on the political economy of health systems a comparative advantage in local production (Levaggi and Menoncin, 2014). Rather, the reasons for devolution may rest in a reduction of solidarity among jurisdictions (Calsamiglia et al., 2006; Dirindin, 2019). Besides efficiency, and given the merit good nature of healthcare, equity concerns have always been highly significant. In many countries healthcare policies are informed by the principle of universality: access to public health services is ensured to all inhabitants regardless of personal characteristics (gender, age, employment, economic and social conditions) or geographic location. Thus, both social and geographical equity may be pursued. The latter is a significant goal in countries characterized by stark geographical differences (e.g., Magnussen and Martinussen, 2013, for Scandinavian countries). Ensuring equity may in fact be more challenging as differences in healthcare standards between geographic areas of the nation become a key issue, together with the more widespread concern for differences within each area, which has more to do with social equity. When healthcare provision is decentralized, geographical equity becomes an even more critical stance and there is an important role to play for the central government in ensuring that differences between geographical areas are minimized. Therefore, in healthcare provision, both efficiency and equity concerns are significant. While the advantages of decentralization under an efficiency perspective are well investigated (e.g., for the Italian NHS, Bordignon and Turati, 2009; Francese et al., 2014; Piacenza and Turati, 2014; Cavalieri and Ferrante, 2016), the properties of decentralization in terms of equity enhancement are rather disregarded by the economic literature. It is, however, clear that in a decentralized setting, equity concerns make interregional grants for health services necessary, especially in order to prevent different fiscal endowments from causing unequal access to services and quality of care across regions. Grants are also needed to compensate for cross-border provision, which is common in healthcare. Further, Levaggi (2007) shows that competition may reduce differences. When levels and quality of services are highly different across areas, equity may be partially restored if patient mobility is allowed, with reimbursement schemes for receiving territories. In this framework, patient mobility should improve quality and allow cost containment. Further, Puy (2007) shows that there is compatibility between mobility and redistribution. Although in a different setting, Kessler and Lülfesmann (2005) show that a trade-off between policies’ allocative and redistributive effects may arise. Finally, citizens’ attitudes towards equity, or solidarity, may be positively correlated with the degree of decentralization in their country. Specifically, Calsamiglia et al. (2006) show that the higher citizens’ preferences for solidarity, the lower the degree of decentralization in the country. However, Ferrario and Zanardi (2011) show that, given a certain degree of healthcare decentralization, a reduction in solidarity (in terms of lower central government equalizing grants) significantly widens between-regions inequality. 2.2

The Problems of Healthcare Decentralization

Fiscal federalism in the context of healthcare provision has specific characteristics that deserve a careful analysis. These refer to the role of intergovernmental transfers and equity concerns. With a hierarchical governance, grants are generally required, primarily to ensure sufficient resources to decentralized governments (vertical fiscal imbalances) and to compensate for territorial differences in the fiscal base (horizontal fiscal imbalances). By its nature, the equalization grant plays a very important role, especially due to the merit good nature of healthcare,

Efficiency and equity effects of healthcare decentralization  91 which requires ex ante redistribution of resources in contexts where income is unevenly distributed, to prevent differences in access and quality of care across regions. Second, spillovers and cross-border provision have financial implications. Generally, they give rise to financial agreements that need to be regulated. Third, the very existence of grants may reduce fiscal responsibility of decentralized governments that may over-expand their healthcare budget. Fourth, in a decentralized context, diversification of services gives rise to the problem of ensuring equity for all citizens regardless of their residence. Besides equity within territories, between territories equity also becomes an important issue if health is conceived as a citizenship right. As regards the first point, one key issue in the health sector is the optimal grant structure. Grants, besides providing sufficient financial means, may be used to align incentives between central and decentralized governments. Snoddon and Wen (2003) investigate the optimal grant structure in a context of strategic interdependence of central government’s choice of grants and local governments’ responses. In addition, Huber and Runkel (2008) and Martell and Smith (2004) analyse the interaction between central government grants and debt policy, showing that under certain circumstances transfers can limit local government debt. A second issue is the need for coordination between levels of government due to the existence of spillovers and cross-border provision for healthcare services. Generally speaking, decentralization is welfare-enhancing compared to centralized uniform provision when policies exhibit limited spillovers and individuals show heterogeneous preferences (Besley and Coate, 2003). However, when it comes to decentralized health policies, as explained by Levaggi and Menoncin (2013), it is often the case that coordination is necessary for at least two reasons: spillover effects (deriving from the public good characteristics of the services produced) as well as the contractual agreement for mobility-related service supply (deriving from the merit good aspect of the good produced). Coordination may entail the design of compensating transfers, either horizontal – between decentralized governments – or vertical, top-down from central government. Therefore, vertical fiscal transfers are essential to top up local resources and guarantee local service provision if the local tax base and local tax revenues are not sufficiently wide. However, the very existence of vertical fiscal transfers may induce opportunistic behaviour by decentralized governments whenever they face soft budget constraints (Kornai, 1986; Kornai et al., 2003; Rodden et al., 2003, p. 14; Levaggi, 2018). There is significant empirical evidence on problems of soft budget constraints in intergovernmental relations (Rodden et al., 2003) and on the critical role played by expectations (Pettersson-Lidbom and Dahlberg, 2003; Rodden, 2006; Bordignon and Turati, 2009; Pettersson-Lidbom, 2010; Piacenza and Turati, 2014). If decentralized governments have expectations that the central government will transfer additional resources and foot their health bills, this may negatively affect the efficient allocation of resources. Local governments may become less fiscally responsible, causing an over-expansion of the local health budgets, inefficient expenditures and an undesirable negative impact on general government fiscal balances (Kornai et al., 2003). There is evidence that soft budget constraints are widely occurring in healthcare policies (Kornai, 2009), because the central government has an interest in preventing decentralized government from “failing” to provide essential healthcare services (Wildasin, 1997; Bordignon, 2006), especially if health policy is a shared competency and therefore the central government can be held responsible for the failure (Bordignon and Turati, 2009). This obviously feeds expectations of future bailouts. Institutional and political variables have a significant influence on the formation of

92  Handbook on the political economy of health systems expectations. For instance, there is evidence that richer and more autonomous regions have lower expectations (Rodden, 2002, 2006) and that central government fiscal discipline reduces bail-out expectations (Bordingon and Turati, 2009). As for political factors, according to Bordignon and Turati (2009), when regional and central governments are politically aligned then decentralized governments show higher fiscal discipline. Specific mechanisms have been proposed to prevent financial irresponsibility of subnational governments and reduce inefficiencies. The first solution is an increase of decentralized governments’ fiscal autonomy, through local taxes, which reduces financial dependency and increases accountability of local politicians. Second, fiscal rules to constrain expenditure and deficit of central or local governments, or both (Grembi et al., 2016; Schakel et al., 2018). Third, the central government may impose some kind of “administrative subordination” on decentralized governments that show financial irresponsibility. This may take the form of technical controls (e.g., appointing an accountant to supervize decentralized budgeting) or recovery plans imposed on decentralized governments (e.g., the Italian “Piani di Rientro”; Bordignon et al., 2020) or even enforcing temporary administration of health policy and health recovery plans by a commissioner appointed by the central government (Ben-Bassat et al., 2016; Aimone Gigio et al., 2018; Depalo, 2019). However, if excess spending is not the result of inefficiencies, but due to insufficient local financial means to finance the needed services, then hardening the budget constraint may reduce social welfare (Piacenza and Turati, 2014). A related point made by Levaggi and Menoncin connects the merit-good nature of healthcare to the existence of soft budget constraint problems in healthcare policy: Merit and impure public goods for which decentralisation is sought seldom present a comparative advantage in being produced locally. The reasons for devolution in this case may be determined by a reduction in solidarity among jurisdictions. We believe that this aspect related to gainers and losers may also explain the onset of soft budget constraint policies (Kornai et al., 2003), one of the less desirable effects of fiscal federalism. In a federal system where wealthier local authorities may be the ultimate gainers from devolution, less wealthy ones may try to reduce their losses by running into a deficit. (2011, p. 2)

When healthcare is decentralized, if countries exhibit stark differences in terms of local economic development, tax base and/or local needs, and if equalizing grants (or even deficit spending) are not enough to ensure equity between territories, then citizens’ right to healthcare may be hindered. There is, however, mixed evidence on the impact of fiscal decentralization on within- and between-region inequalities, in terms of both access to services and health outcomes (Chandra et al., 2012; Skinner, 2012; Costa-Font and Turati, 2018; Di Novi et al., 2019). In this framework, patients’ mobility may help reduce differences in the standards of healthcare affordable to residents of different areas of the same country, and also enhance efficiency in provision (Ferrario and Zanardi, 2011). Finally, there is a concern that decentralization in healthcare, rather than enhancing quality and responding to local needs and preferences, may give rise to a “race to the bottom” among decentralized governments (Oates, 1999; Costa-Font and Rico, 2006) and may hinder service appropriateness. However, Costa-Font and Greer (2013) recognize that it is not decentralization per se that affects health policies’ efficiency or effectiveness, but rather the outcomes from decentralization depend on the institutional design and how it shapes incentives, democratic accountability and the effective political power of each level of government. For instance, the design of the financial framework is critical, as it shapes the extent of potential

Efficiency and equity effects of healthcare decentralization  93 policy divergence and the extent of “coercive isomorphism” (Costa-Font and Greer, 2013; Francese et al., 2014).

3.

HEALTHCARE DECENTRALIZATION IN ITALY

As for decentralization of health policy, Italy is an interesting case study for many reasons. First, the public sector has long played a major role in healthcare provision. Second, the public National Health Service performs rather well in terms of outputs and outcomes as well as for cost containment. Third, although Italy is a unitary country, in the past 30 years intergovernmental relations have been deeply transformed by fiscal devolution reforms towards a model of regional governance. In particular, public healthcare is decentralized on both the revenue and the expenditure side, while central government retains a role in defining mandatory services, enacting framework legislation and topping up regional resources to ensure sufficient financial means to all regions. And finally, decentralization of service organization and management to the regional level has allowed the development of differentiated models of public health provision, an example of laboratory federalism (Oates, 1999). According to the Italian Constitution, health is a fundamental individual right and is of collective interest. The National Health Service, or NHS (Servizio Sanitario Nazionale, or SSN), a public national health insurance service, is based on principles of universal coverage and provision of a comprehensive range of health services largely free at the point of use, and is financed through general taxation. It is decentralized on both the revenue and the expenditure side. As for expenditure, according to Article 117 of the Constitution, healthcare is the responsibility of both central and regional governments, with different roles and competencies. Generally speaking, central government is responsible for defining the fundamental principles of national health policy, for setting the minimum level of services to be offered to all citizens and for deciding the amount of resources that every year make up the National Health Fund, or NHF (Fondo Sanitario Nazionale). Regional governments autonomously organize and manage healthcare services in their jurisdiction, and also define the rules related to the involvement of private providers. On the revenue side, regional governments have autonomous tax sources to finance healthcare services, topped-up by their share of the NHF. Co-payments may be required for outpatient services and for pharmaceuticals. The current structure of the public health service is the result of various reforms implemented in the last 40 years which alternatively focused more on improving efficiency or on equity issues in service provision and resulted in regions now having political, administrative, fiscal and organizational responsibility for healthcare provision. The health sector reforms were part of a wider transformation of intergovernmental relations in Italy, inspired by fiscal federalism principles and pursuing higher devolution of revenue and expenditure. 3.1

The Birth of the National Health Service, 1978

According to Article 32 of the Constitution, “The Republic safeguards health as a fundamental right of the individual and as a collective interest, and guarantees free medical care to the indigent.” Despite this constitutional provision, it was only in 1958, 11 years after the Constitution was approved, that the Health Ministry was established for the first time (law 296/1958).

94  Handbook on the political economy of health systems Then, in 1968 autonomous public institutions providing hospital care (enti ospedalieri) were introduced. However, healthcare provision rested on a fragmented system of more than 100 different health insurance funds, of a corporative nature, which did not cover the whole population nor offered uniform services, and which also ran into large deficits. In 1978 a major reform radically transformed healthcare, by abolishing health insurance funds and establishing the Italian NHS (law 833/1978), a universal and comprehensive public fund financed by sickness contributions, levied on gross wages or pensions, and by central government general revenues. By law, the Italian NHS is “the complex of functions, structures, services and activities directed to promoting, maintaining and recovering the physical and mental health of the whole population, without distinctions based on individual or social conditions, and ensuring equality of all citizens with respect to the service” (Art. 1, law 833/1978). Equity was therefore a central concern in this reform, and with the establishment of the NHS, the constitutional provisions defining health as a fundamental right of the individual came into force. The Italian NHS was set out to provide universal and comprehensive health protection and to guarantee equal access to uniform health services for all, irrespective of income or location. In addition, the main aims of the NHS included controlling health spending and ensuring public democratic control. As for its organization, it had a multi-layered structure, involving all levels of Italian government, each one with a specific function. The central government had a planning role, in accordance with regional governments, through the drafting of a national health plan that would define the goals of the NHS and the levels of healthcare for the entire country, the amount of financing of the NHF and the criteria for its apportionment to regions; regional governments had responsibilities for planning, organizing and managing services in their jurisdiction. Administrative responsibilities and service provision was the task of local health units (Unità Sanitarie Locali, or USL), public bodies involving one or more municipalities, responsible for providing services directly or by contracting with private producers according to the regional government framework for private providers’ involvement. Since 1978 three major reforms have significantly transformed the structure and operation of the NHS, by introducing elements of an internal market, by giving managerial autonomy to local health units and public hospitals, and by increasing regional governments’ autonomy and responsibilities. 3.2

The First Wave of Reforms, 1992–1993

In 1992–1993, a first wave of reforms produced a significant transfer of managing and funding competencies, as well as financial resources, to the regional level of government. Reforms were implemented for the primary purpose of improving efficiency and controlling expenditure, which had been increasing steadily since 1978. At the same time, universal coverage and equity of access were safeguarded. As for efficiency, a separation between service producers and purchasers was sought, based on the experience of the British NHS, where quasi-markets were introduced in 1990. For this reason, local health units were transformed into publicly owned firms named local health firms (Aziende Sanitarie Locali, or ASL), with their own management and budget. Hospitals were then separated from ASL and became hospital firms (Aziende ospedaliere). This business-like transformation implied two main changes: first, budgets needed to be efficiently managed; second, a new managerial organization was required. However, as key managing positions of local health firms were appointed by regional governments, these firms remained strongly exposed to political pressures (although

Efficiency and equity effects of healthcare decentralization  95 no more from municipalities). In a quasi-market framework, local health firms were responsible for purchasing services for citizens from different producers, from general practitioners, to publicly owned hospital firms, to private providers, and competition was expected to boost the efficient allocation of public funds. Contracts with providers made use of a prospective payment system based on diagnosis-related groups (DRGs), later partly substituted with a budget-based approach, used in particular with private producers. The separation of functions has also meant a revolution in the concept of public healthcare. While in the previous system cost was about the only measure of service provision, with the reform outputs are measured and providers are paid on the basis of what they provide. The 1992–1993 reform also concerned the financing side: sickness contributions were assigned to regional governments, thus attaining some degrees of fiscal decentralization, in an attempt to align funding and expenditure responsibilities at the regional level and to increase regional government accountability. This initial form of fiscal decentralization was accompanied by a law provision explicitly forbidding central government from bailing out future regional deficits (Art. 13 legislative decree 502/92, later modified by legislative decree 517/93). In addition to sickness contributions, the NHF financed through general taxation continued to contribute to cover NHS financial needs, by topping up regional funds, but with a new equalizing function, due to large regional inequalities in the distribution of sickness contributions. The ex ante definition of the amount of resources making up the NHF is a critical factor that contributes to controlling expenditure. 3.3

The Second Wave of Reforms, 1999

The 1999 reforms were more concerned with equity, but also efficiency was fostered mostly through changes affecting the financing side. Equity was pursued by introducing the so-called essential levels of care (Livelli Essenziali di Assistenza, or LEA), mandatory and uniform services to be provided on the whole national territory (legislative decree 229/99), defined by the central government with the agreement of regional governments. At the same time Supplementary Funds of the NHS were introduced, to cover both expenses for services excluded from the NHS and expenses for co-payments required for services included in the NHS. Finally, a step back from the quasi-market model and the search for competition introduced in 1992–1993 was made by requiring more co-operation between ASL and providers, especially private providers. To further improve regions’ responsibility and accountability, in 1999 regions’ tax autonomy was increased by abolishing sickness contributions and substituting them with two new regional taxes: a tax on productive activities (Imposta regionale sulle attività produttive, or IRAP) and a surcharge on personal income tax (Addizionale IRPEF). This was consistent with the wider fiscal federalism reform underway in Italy in those years (and not yet completed). Finally, the reform reinforced the role of regional governments in healthcare services organization and management. Each region, besides providing the LEAs, could also autonomously enlarge its offer and finance it with its own resources. This gave way to a wider regional differentiation in service provision. 3.4

The Third Wave of Reforms, the Early 2000s

In the early 2000s, newly approved legislation affected the NHS, but generally produced a consolidation of trends already defined in the previous two decades. Specifically, the

96  Handbook on the political economy of health systems management role of regional governments in healthcare was consolidated and their fiscal autonomy reinforced. These trends reflected the prevailing political orientation towards higher decentralization of government functions. The most important act was the 2001 constitutional reform, which modified intergovernmental relationships and designed a more decentralized institutional asset for Italian government (Constitutional Law n. 3/2001), with a significant impact also on the NHS. In particular, the reformed Article 117 of the Constitution, on the assignment of legislative powers, defines an explicit list of matters on which the central government has exclusive legislative powers. It then identifies areas of concurrent central/regional competence and finally states that regions have legislative powers in all subject matters that are not expressly covered by state legislation. The protection of health is a shared responsibility of central and regional governments (Art. 117, par. 2). Central government (the Ministry of Health) defines framework legislation and has responsibility for essential levels of care; that is, for the “determination of the basic level of benefits relating to civil and social entitlements to be guaranteed throughout the national territory” (par. 1, letter m). The central government also defines the total amount of resources needed to finance LEAs, and ensures that they are effectively attained. Regional governments have management responsibilities: they organize regional health policy, guarantee that essential levels of care are offered and ensure quality and performance of their health systems. Local health authorities (a total of 148 all over the country) are in charge of service provision in a geographically defined area, but citizens can freely access healthcare services wherever they prefer. Overall, regional governments enjoy large autonomy, but they have to comply with the framework legislation defined by central government. The second process affecting the NHS was the ongoing fiscal federalism reform, pursuing decentralized governments’ higher financial autonomy (and responsibility). In 2000 regions’ own tax sources were enlarged thanks to the increase in revenue-sharing on the Gasoline Tax and the introduction of revenue-sharing on the Value Added Tax. The need for a proper equalizing formula to apportion the NHF to the regions was solved by introducing the concepts of standard needs and standard costs that would drive the allocation of funds. Law 42/2009 on fiscal federalism prescribed “the gradual overcoming … of the historical spending criterion in favour of standard financial needs” (Art. 2, par. 2, letter m), and that “[expenditures] are determined with respect to standard costs” (Art. 8, par. 1, letter b). Criteria to compute standard costs were defined in 2011 (legislative decree 68/2011), but the implementation process proved to be long and complex and today the allocation criteria are essentially based on weighted capitation. A critical issue in the financing of regional health expenditure is how regional deficits are dealt with. Initially, central government simply bailed out regional health deficits, an approach obviously producing distorted incentives. To limit regional deficits, in 2005 regions’ access to the so-called premium share of the NHF was conditioned to having a balanced budget (law 311/2004) or, in case of deficits, access was granted only if the region identified the causes beneath the deficit and defined a so-called Operation Programme (Programma Operativo) to restructure and improve the regional health service. In 2007 so-called recovery plans (Piani di Rientro) were introduced to improve the accountability of regional governments (law 296/2006). This was seen as a form of re-centralization, also because of the controlling role assumed by the Ministry of Finance. Recovery plans are formal agreements between central government and a region that runs health deficits (initially above 7 per cent, reduced to 5 per cent since 2010), committing them to restore economic and financial balance while continuing

Efficiency and equity effects of healthcare decentralization  97 to ensure the essential levels of care. Recovery plans are three years long and may be extended for a further three-year period. If deficits are not wiped out, a commissioner ad acta (the president of the regional government) is charged with the implementation of the plan and austerity measures are introduced, such as personnel reduction and tax rate increases. Finally, the economic crisis that hit the world economy in 2008 forced governments to introduce austerity measures in response to increasing government budget deficits. In Italy, as in many other countries, these also included measures in health policy, in terms of both resources and governance, such as controls and rationalization of health expenditure (expenditure caps, standard costs), reorganization of hospital services, revision of public procurement rules and procedures, and strengthening of controls on regional deficits.1 Under cost-containment policies, NHS under-financing and understaffing worsened, also compared with the main European Union (EU) countries (Neri and Mori, 2017). Under the EU’s and financial markets’ pressures for balanced budgets, the central government took a stance to limit health deficits, which were run by southern regions in particular. This implied a partial re-centralization of policy making and restrictions to regional autonomy, which affected regions in an asymmetric way: most southern regions had to implement recovery plans, aimed at restoring fiscal discipline. Since 2007 all southern regions plus Lazio have been constantly under recovery plans, Puglia “only” since 2010 (the only exception is Basilicata, which so far has not undergone any recovery plan). Conversely, in the centre north of Italy recovery plans were introduced only exceptionally. The Liguria region was under a recovery plan in 2007–2009 and Piemonte in 2010–2015. Therefore, central and northern regions retained organization autonomy, and have even asked for increased decentralization. 3.5

The Italian NHS Today

In Italy, healthcare policy results from the interaction of the central and regional levels of government and is ruled by a complex set of institutional and political arrangements, structured according to the reforms that have taken place since 1978 and especially in the last 30 years. The main steps in this process were the constitutional reform of 2001, which also defined the general framework for the implementation of health policies, and ensuing law provisions, primarily law 42/2009 and legislative decree 68/2011. Total health expenditure in 2019 in Italy amounted to 8.7 per cent of gross domestic product (GDP), of which public health expenditure was equal to 6.4 per cent (from 4.7 per cent in 1980). Both values are below the Western European average and significantly less than expenditure in the major European countries, which amount to, respectively, 11.7 per cent and 9.9 per cent in Germany; 11.2 per cent and 9.4 per cent in France; 10.3 per cent and 8 per cent in the UK; and 9 per cent and 6.4 per cent in Spain (OECD, 2020). In addition, over the period 2010–2019, cuts to the NHS amounted to more than €37 billion. The Italian NHS is considered among the most efficient in the world (it ranked fourth according to Bloomberg, 2018) and Italy is one of the healthiest countries (Bloomberg, 2019; WHO, 2019). Over the years, health expenditure has been near-constantly growing, from an initial 4.7 per cent of GDP in 1980, and has constantly produced deficits. In the early 1990s effective measures were taken to curb expenditure, in an effort to meet the Maastricht Treaty parameters. By the early 2000s health expenditure had been reduced to 5.5 per cent of GDP but then started growing again, at an average yearly growth rate of 5 per cent. In 2009 it reached 7.1 per cent of GDP. Especially after the 2008 economic and financial crisis, austerity measures were

98  Handbook on the political economy of health systems introduced, aiming to control public expenditure. As a result, since 2009 public healthcare expenditure has constantly declined, to reach 6.4 per cent of GDP in 2019. In these years, the nominal growth rate of public funds was on average equal to 0.9 per cent, while average inflation rate was 1.07 per cent (thus in real terms public healthcare expenditure has declined on average by 0.8 percentage points) and per capita expenditure reduced from 1.893 to 1.746 euros (Fondazione Gimbe, 2019). This reduction was mainly concentrated in southern regions, where expenditure cuts were made according to recovery plans. Savings were obtained essentially from curbing personnel expenditure (for instance, by freezing personnel turnover), and reducing hospital beds and pharmaceutical expenditure. However, these measures inevitably impacted on public healthcare, and produced an increase in out-of-pocket payments (WHO, 2012) and probably unevenly affected access to care (Torfs et al., 2021). For instance, hospital beds’ reduction implied that their number reached 34.9 per every 10,000 inhabitants in 2019 (the number was 51 in 2000, 45 in 2005, 40 in 2009), while in Germany and France they were respectively 80 and 60 in that same year. However, the Italian national average results from significantly different territorial data: in 2019 the number of hospital beds per every 10,000 inhabitants was 37.5 in the northern regions, 34.3 in central regions and 31.6 in southern ones (Istat, 2022). In 2019 the average occupancy rate (78.7 per cent) was higher than the European average, a clear indicator of stress on hospital services, which makes them vulnerable to sudden increases in demand. Further, Italy had one of the lowest numbers of intensive care beds in Europe: nine per every 10,000 inhabitants in the north of the country, eight in the centre–south. Despite expenditure cuts, according to various indicators the health of the Italian population has improved over the last few decades. Average life expectancy at birth reached 83.57 in 2020, an indicator that is constantly improving, although with important regional and socio-economic differences. Both adult and infant mortality rates have fallen significantly (Istat, 2022). And it is still true that Italy is doing quite well in the group of heavily regulated public systems characterized by a stringent budget constraint (Joumard et al., 2010, p. 51). However, in almost all demographic and health indicators, there are marked regional differences moving from north to south (Turati, 2013). For instance, southern regions are not effectively supplying the set of constitutionally guaranteed essential services (Di Novi et al., 2019, p. 2). These regional disparities explain the significant interregional mobility of patients in Italy, which accounted for approximately 4.1 per cent of total public health expenditure in 2018 and is characterized by a net structural patients flow from southern to northern regions (Fondazione Gimbe, 2020, p. 10). Various factors affect patients’ interregional mobility, such as the effectiveness and efficiency of each regional healthcare service, the presence of reference centres for specific diseases, waiting lists, diagnostic services and availability of treatments, and perceived or real quality of assistance (Berta et al., 2021; De Curtis et al., 2021; Ricci et al., 2021). Patients’ mobility reflects the difficulties for southern regions in addressing the accountability process (Bruzzi, 2012), and interregional mobility reimbursements impose a burden on these regions’ budgets. 3.6

The Italian NHS: Equity and Efficiency Evaluations of Fiscal Decentralization

The Italian NHS pursues both equity and efficiency objectives: it aims to guarantee all citizens uniform access to healthcare while attaining efficiency in service provision and containing costs. Decentralization to regional governments of competencies for healthcare organization

Efficiency and equity effects of healthcare decentralization  99 served the purpose of improving effectiveness in service delivery while fiscal decentralization pursued accountability and responsibility in health spending. Data reported above show that since the early 1990s the Italian NHS has successfully contained costs. In addition, fiscal decentralization contributed to a reduction of within-regions inequality (Di Novi et al., 2019). However, public healthcare provision is affected by problems under both an efficiency and an equity perspective. As for equity, the wide differences in the availability and quality of services among different regions, in particular along the North– South divide, and the significant interregional patient mobility (primarily from southern to northern regions) suggest that between-regions equity is far from being attained both in access to healthcare and, as a consequence, in citizens’ health levels. The North–South gap is hard to reduce: regional disparities in health services are a legacy of the past, and the empirical literature has found mixed evidence of decentralization effects on them (De Belvis et al., 2012; Toth, 2014; Blöchlinger et al., 2016; Costa-Font and Turati, 2018; Di Novi et al., 2019). Significantly, cross-region inequality cannot be ascribed to regions’ different financial endowments. In fact, the Italian NHS is characterized by a significant ex ante redistribution, aiming at ensuring comparable financial resources to all regions. Under fiscal decentralization, regions enjoy their own tax revenues, but these are topped up by central government equalizing grants. Rather, between-regions inequality is a clue regarding across-regions differences in the degree of efficiency and effectiveness in using financial resources. A critical issue is therefore the identification and correction of sources of inefficiency in health expenditure, which hinder ex post redistribution, despite the ex ante significant positive redistribution (Turati, 2013, p. 61) and despite the mandatory uniform essential services (LEAs) across the country. The previously discussed results of the theoretical and empirical literature on grants in a financially decentralized setting may contribute to the explanation. According to second-generation fiscal federalism, decentralization is effective in attaining efficiency when regions enjoy fiscal autonomy and spend their own revenue sources (Oates, 2005; Boetti et al., 2012). Conversely, when decentralization of competencies is not matched by devolution of fiscal revenues and local financing heavily relies on central government grants (vertical fiscal imbalances), incentives for decentralized government accountability are lost (Weingast, 2009; Eyraud and Lusinyan, 2013). In Italy this happens especially for southern regions, where the share of their own revenues in total health expenditure is rather low. In addition, distorted incentives may arise, fostering excess inefficient expenditures, if the central government cannot commit to avoid future bailouts of decentralized governments’ deficits (soft budget constraints). For Italy there is empirical evidence that regional financing suffers from problems of soft budget constraints (Bordignon and Turati, 2009) and that moral hazard has a negative impact on the efficient allocation of health expenditure (Piacenza and Turati, 2014). Fiorani and Meneguzzo (2008) maintain that regions acted as if central government would repay their health-related debts in the case of default, and the model proposed by Levaggi and Menoncin (2013) is consistent with this assumption. Some empirical works focusing on this phenomenon have reached interesting empirical conclusions for Italy. First, there is evidence that the link between ex ante resources and ex post expenditure (so absence of excess inefficient spending) was stronger when regional expectation of future central government bailouts were lower. In addition, there is evidence that regional governments of the same political party as the central government were less inclined to run deficits (Bordignon and Turati, 2009). Finally, central government fiscal discipline effectively hardened subnational governments’ budget constraints in the years 1993–2006. More significantly, fiscal discipline

100  Handbook on the political economy of health systems did not negatively impact on citizens’ health (as a consequence of a potential reduction of needed services), it only affected the inefficient component of regional governments’ expenditure (Piacenza and Turati, 2014). The complex system of intergovernmental relations and the need for coordination to prevent deficits and guarantee a minimum level of services have prompted central government’s interventions. Since 2007, the central government has addressed regional inefficiencies by imposing on regions running large deficits a form of “administrative subordination” through recovery plans. They appear to have been effective in containing expenditures, restoring sound financial conditions and reducing unjustified and inappropriate provision of services (Atella et al., 2019; Bordignon et al., 2020). However, evidence is less clear-cut regarding their impact on health outcomes and access to healthcare services. For instance, Bordignon et al. (2020) do not find any significant effect of recovery plans on citizens’ health and on the perceived quality of services, while, using a different methodology, Depalo (2019) estimates an increase in mortality rates in some regions, and Arcà et al. (2020) estimate a 3 per cent rise in avoidable deaths, a reduction in hospital capacity and a rise in south-to-north patient migration as a consequence of recovery plans. A comprehensive study by Aimone et al. (2019) stresses that recovery plans achieved mixed economic results, because some regions still suffer from deficits. This study also points out that recovery plans’ impact on service quality was even more mixed. Among the regions under recovery plans, only one could guarantee the essential levels of care (Piemonte), while two regions were never able to meet these standards (Calabria and Campania). The alternative strategy to foster efficiency – i.e., increasing fiscal autonomy – is constrained by the limited availability of taxes that can be efficiently decentralized according to fiscal federalism theoretical prescriptions on optimal tax assignment. It is also constrained by the unequal distribution of tax bases, due to the relevant economic differences across Italian regions.2 Therefore, despite the devolution of tax sources to regional governments under the fiscal federalism reform, in Italy the central government has continued to allocate grants to finance regional health policy for two reasons: to top up regions’ insufficient revenues (vertical fiscal imbalances) and to equalize different regional fiscal capacities (horizontal fiscal imbalances). As a result, after fiscal devolution, the relative amount of funds for each region’s health budget did not change, while the composition of sources of finance across regions became significantly different, according to the regional distribution of the tax base: more endowed regions receive limited grants, while those with a limited tax base heavily rely on grants (Di Novi et al., 2019, p. 3). Significantly, efficiency and accountability of health policy is higher in regions that rely more on their own resources, compared to those that are more dependent on central government transfers. This is not enough to assume a causal nexus between the two; in fact, a limited local tax base and limited local efficiency may both be caused by a third factor – for instance, different socio-economic conditions (Putnam et al., 1993). Surely further empirical tests should be required to clarify this issue, but it is nevertheless interesting to note that evidence from Italy is consistent with fiscal federalism theoretical conclusions that, as the share of transfers reduces and that of regions’ own taxes increases, decentralized governments’ accountability increases and so does the efficiency and effectiveness of public policies (Weingast, 2009; Eyrand and Lusinyan, 2013). For instance, Francese et al. (2014) show that regions having access to their own significant revenues to fund health expenditure are more accountable towards citizens-voters and hence perform better in terms of controlling the inappropriateness of treatments. Northern Italian regions, enjoying a wider tax

Efficiency and equity effects of healthcare decentralization  101 base and relying less on central government grants, perform better. This also shows that decentralization does not produce accountability for all, but only for the more fiscally endowed regions. Italy illustrates how, in a country characterized by stark regional economic differences, decentralization may work well for some and much less for some others. This suggests the efficiency of implementing some form of asymmetric federalism, which grants spending autonomy only to those regional governments characterized by low degrees of vertical fiscal transfers and higher accountability. The Italian Constitution, after the 2001 reform, envisages the possibility for asymmetric federalism (Art. 116, par. 3): “Additional special forms and conditions of autonomy … may be attributed to other Regions by State Law, upon the initiative of the Region concerned, after consultation with the local authorities.” Actually, since 2017 some northern regions (Lombardia, Veneto, Emilia-Romagna) have started the procedure to obtain larger autonomy in an asymmetric way. The process is now at a halt and there are specific problems that need to be dealt with. But surely in healthcare a kind of asymmetric federalism has already been implicitly implemented, as practically all southern regions in the last 10 years have been under central-government-controlled administration through recovery plans. The demand for greater autonomy by some regions is partly a local response to alleged inefficiencies and lack of effectiveness of the central government, but it also reflects a cultural change that has weakened the focus on solidarity, in Italy and abroad (Dirindin, 2019), although solidarity is a founding principle in the Italian Constitution (Art. 2 and Art. 119). However, in practice, implementing asymmetric federalism is not straightforward: there are major issues that need to be resolved under a public finance perspective. First, universalism is at risk – there is the problem of how under asymmetric federalism the constitutional right to health can be equally guaranteed to all citizens, the risk being that inequalities could further increase; second, a balanced budget should be attained; third, criteria to define the amount of resources needed to finance new decentralized functions should be identified; and finally, the appropriate tax instruments have to be used to finance decentralized fiscal autonomy (Zanardi, 2019).

4.

CONCLUDING REMARKS AND FUTURE PERSPECTIVES FOR ITALY

Given the marked regional economic differences and the uneven development of regional tax bases and tax autonomy (Brandolini and Torrini, 2010; Daniele and Malanima, 2011; Bordignon, 2017; Bank of Italy, 2018; Cannari et al., 2019), the implementation of healthcare decentralization in Italy faced different problems and produced different outcomes across regions, displaying a clear North–South asymmetry. These disparities resulted in an unintended form of asymmetric federalism in healthcare. Evidence from Italy, showing a clear correlation between wealthier local economies and wider regional tax bases on the one side, and autonomous, diverse, more efficient and equitable regional health systems on the other, seems to support the theoretical conclusions that decentralization of public functions can be effectively implemented only under fiscal autonomy which favours regional government responsibility. This seems to imply that effective decentralization requires a set of economic and institutional conditions, and it cannot be equally effective with highly diverse economic bases. Although these conclusions concern a developed country, they are consistent with

102  Handbook on the political economy of health systems findings from the extensive literature on decentralization and development, which provides evidence that decentralization often falls short of its expected outcomes if implemented in the context of low economic and institutional development (Bardhan, 2002; Schragger, 2010), that the fiscal system needs to be carefully designed (Shah, 2004), that a well-functioning democracy is required to obtain benefits from decentralization (Bardhan and Mookherjee, 2006), and that local governance and the process of implementing decentralization can be as important as the design of the system in affecting service delivery outcomes (Ahmad et al., 2005). Therefore, institutions, accountability, governance and capacity play a key role for the success of decentralized policies, and there is scope for further empirical analyses of evidence from Italy under this perspective. Similar concerns for the attainments of decentralization have arisen in developed countries and prompted some forms of re-centralization (Prud’homme, 1995; Treisman, 2007; Terlizzi, 2019). In the health sector, since the first years of the 21st century the role of the central government has strengthened (Saltman, 2008; Tediosi et al., 2009). In healthcare, limited and selective re-centralization was a response to healthcare systems’ financial sustainability problems, equity and accessibility concerns, and territorial differences in service provision (Mauro et al., 2017). In the Italian policy debate, re-centralization is sometimes proposed as a way of ensuring higher nationwide equity and efficiency, because the territorial socio-economic divide and the uneven financial endowments may produce a zero-sum game, where the gains from decentralization for the better-off regions are counterbalanced by the losses in performance of the weaker ones (Pavolini and Vicarelli, 2012). In practice, since 2006, within the country’s decentralized regional health system an unintentional re-centralization process has been underway, through recovery plans that targeted the weaker regions, or in extreme cases, through centrally appointed Commissioners charged with ensuring the attainment of central government’s targets (Mauro et al., 2017). The recent outbreak of the Covid-19 pandemic offered additional arguments in favour of re-centralization of healthcare, as a way of ensuring effective and timely responses (Greer et al., 2020a, 2020b) and of improving institutional coordination. In Italy a strong centralized response was advocated (Bosa et al., 2021) and a “state of emergency” was declared: power and authority were concentrated within the central government, which directly intervened in the management of the health system (Mauro and Giancotti, 2021). In turn, the central government drew on the advice of the Covid-19 emergency commission and scientific advisory panel (Comitato tecnico scientifico), which was set up early on in the pandemic. In this framework, although different regions adopted different management models to address the health emergency, in accordance with the differences in their health systems (Mauro and Giancotti, 2021), regional governments were subordinated to central government decisions, which pursued coordinated strategies. The issue is whether this approach is a form of re-centralization in response to the severe limits of a decentralized system, which cannot properly respond to sudden and severe threats to public health. According to Greer et al. (2020b), re-centralization, or between-government centralization, as they name it, was only a temporarily response, with the central government making more policies than usual. In fact, in the second wave of Covid, centralization was lower and decentralized governments regained more autonomy. As for Italy, the prominent role of the central government during the pandemic is consistent with the country’s institutional design of healthcare decentralization. In fact, healthcare is a shared responsibility of central and regional governments. In addition, there is an implicit supremacy clause (Onida, 2020; Poggi and Sobrino, 2021) in that a transfer of powers to the central gov-

Efficiency and equity effects of healthcare decentralization  103 ernment, limited in time and scope, is provided for by the Constitution in exceptional cases; for instance, when public health is threatened, as happens in a pandemic. According to Article 120 (par. 2), the central government can act for bodies of the regions, metropolitan cities, provinces and municipalities … in case of severe danger for public safety and security, or whenever such action is necessary … to guarantee the basic level of benefits relating to civil and social entitlements, regardless of the geographic borders of local authorities.

Further, regional and central governments have shared competencies on health policy (Art. 117 of the Constitution), but law 883/1978, which introduced the NHS, foresaw the possibility for the Health Minister to enact ordinances on health matters, of a contingent and urgent nature, with effectiveness on the whole national territory (Art. 32). Therefore, the centralized response to the pandemic and the temporary centralization of healthcare management is not at all in contrast with the Italian decentralized health system (Piperno, 2020). Rather, central government healthcare measures introduced in response to the pandemic were made necessary by some existing weaknesses of the Italian healthcare system, caused by both national health policy and some regional models of healthcare (Mauro and Giancotti, 2021). These weaknesses were sharply disclosed by the Covid emergency. At the outbreak of the pandemic in early 2020, hospital capacity collapsed rapidly, and hospitals were quickly unable to offer proper care to all those who needed it. This was certainly due to the dramatic spread of the highly infective and aggressive SARS-Cov-19. But it was also caused by policies and choices implemented long before with regard to NHS funding, primary healthcare organization and hospital management (Costa-Font et al., 2020; Buzelli and Boyce, 2021; Plagg et al., 2021). At the national level, these measures included cuts in NHS funding and cost containment through the reduction of hospital beds (Piacenza et al., 2010; De Belvis et al., 2012; Ministero della Salute, 2017; Pecoraro et al., 2021; World Bank, n.d.). At the decentralized level, measures encompassed some regions’ hospital-centric organization of care matched by a neglect of primary care, delivered by general practitioners (Kurotschka et al., 2021). The latter included, for instance, dismantling territorial services and constraining the number of general practitioners, whose number per inhabitant is comparatively much lower than in other EU countries (Kringos et al., 2015). These choices resulted in an overuse of emergency departments, which had to substitute insufficient primary care (Berchet, 2015; Garattini et al., 2016). Further, the specific form of managed competition implemented in some regions may have exacerbated the impact of the pandemic by providing the wrong incentives to hospitals (Costa-Font et al., 2020). This was, for instance, the case in Lombardy, where the insufficient development of primary care and territorial services caused excess pressure on hospitals from patients who had no other option for receiving care (Plagg et al., 2021), and where swift coordination was hindered by the region’s hospitals’ decentralized competition model (Costa-Font et al., 2020). This prompted a faster collapse of hospital capacity, already constrained by the limited number of both ordinary and intensive care beds, which had been constantly cut over the years (Plagg et al., 2021). In Italy, therefore, the NHS and the different regional health systems face some weaknesses, which the pandemic has amplified (Costa-Font et al., 2020; Dirindin, 2020; Vicarelli and Neri, 2021). There is therefore space for decentralized health systems’ improvement, for both their efficiency and equity outcomes. The more efficient northern regions, under healthcare decentralization, had reduced within inequality and further increased their health system’s effi-

104  Handbook on the political economy of health systems ciency; however, they now need to rethink their approach to primary care, territorial services and hospital-managed competition in order to foster the efficiency, effectiveness and equity of their regional health systems. The case for southern regions is more complex, as they first need to achieve effective fiscal autonomy, but the limited tax base seems to be a strong obstacle on the way to accountable, efficient and equitable regional health systems. The North–South economic gap is therefore mirrored by a North–South divide in the challenges that regional health systems need to address to further improve equity and efficiency under healthcare decentralization. For both northern and southern regions, one rather disregarded strategy to achieve equity and efficiency gains would be that of fully exploiting the potential from the so-called “laboratory federalism” (Oates, 1999). The current 21 models of regional health service which make up the Italian NHS provide a rich array of experiences. Effective benchmarking and cross-fertilization would safeguard local autonomy and, at the same time, provide regions with opportunities to revise their approaches and access potentially useful best practices to improve their health services, taking full advantage of the experimentation allowed by decentralization (Marchitto, 2001; Scognamiglio, 2014). Benchmarking is time- and resource-consuming, and its implementation requires specific competencies, but it offers significant opportunities, as proved by some international experiences (Northcott and Llewellyn, 2005; Thonon et al., 2015). Far from being a case for re-centralization, the central government in its coordinating role could become a catalyst for benchmarking across the NHS, an exercise from which service improvements and benefits for public finances may arise for all participating regions, thanks to a win–win strategy of exploiting the gains from decentralization.

NOTES 1. For a synthetic review, see Camera dei Deputati (2022). Many different legislative measures contributed to the end of controlling expenditures, among them: legislative decree 98/2011 on expenditure caps for medical devices (revised by legislative decree 95/2012 and law 228/2012), legislative decree 95/2012 (spending review) on standard prices for healthcare procurement, reduction of hospital beds; a reduction of services purchases from private providers; law 66/2014 on public procurement and legislative decree 78/2015 on the revision of NHS contracts with third parties and on the reduction of inappropriate hospitalizations and in-hospital rehabilitations. 2. The tax base of IRAP, the regional tax that finances health spending, is a proxy of the regional GDP, which is highly unevenly distributed across Italian regions, much more than other tax bases; for instance, that of a consumption tax, which therefore would have been more suitable to finance regional governments.

REFERENCES Ahmad, J., Devarajan, S., Khemani, S., and Shah, S. (2005), Decentralization and Service Delivery, Policy Research Working Paper 3603, Washington, DC, World Bank. Aimone, L., Camussi, S., and Mancini, A.L. (2019), The Effect of Financial Recovery Plans on Interregional Disparities on Health Services, Bank of Italy, www​.bancaditalia​.it/​pubblicazioni/​altri​ -atti​-convegni/​2019​-piani​-rientro​-deficit​-sanitari/​Aimone​.pdf (last accessed 4 February 2022). Aimone Gigio, L., Alampi, D., Camussi, S., Ciaccio, G., Guaitini, P., Lozzi, M., Mancini, A.L., Panicara, E., and Paolicelli, M. (2018), La sanità in Italia: il difficile equilibrio tra vincoli di bilancio e qualità

Efficiency and equity effects of healthcare decentralization  105 dei servizi nelle Regioni in Piano di Rientro, Questioni di economia e finanza, Occasional Papers, 427, Rome, Bank of Italy. Akai, N., and Mikami, B. (2006), Fiscal decentralization and centralization under majority rule: A normative analysis, Economic Systems, 30 (1), 41–55. Alonso, R., Dessein, W., and Matouschek, N. (2008), When does coordination require centralization? American Economic Review, 98 (1), 145–79. Arcà, E., Principe, F., and Van Doorslaer, E. (2020), Death by austerity? The impact of cost containment on avoidable mortality in Italy, Health Economics, 29 (12), 1500–16. Atella, V., Belotti, F., Bojke, C., Castelli, A., Grašič, K., Kopinska, J., Piano Mortari, A., and Street, A. (2019), How health policy shapes healthcare sector productivity: Evidence from Italy and UK, Health Policy, 123 (1), 27–36. Bank of Italy (2018), L’economia delle regioni italiane (dinamiche recenti e aspetti strutturali), Economie regionali, 23, Rome, Bank of Italy. Bardhan, P. (2002), Decentralization of governance and development, Journal of Economic Perspectives, 16 (4), 185–205. Bardhan, P., and Mookherjee, D. (2006), Decentralization, corruption and government accountability, in S. Rose-Ackerman and H.R. Luce (eds), International Handbook on the Economics of Corruption, Cheltenham, UK and Northampton, MA, USA, Edward Elgar Publishing, pp. 161–88. Ben-Bassat, A., Dahan, M., and Klor, E.F. (2016), Is centralization a solution to the soft budget constraint problem? European Journal of Political Economy, 45, 57–75. Berchet, C. (2015). Emergency Care Services: Trends, Drivers and Interventions to Manage the Demand, OECD Health Working Papers, 83, Brussels, OECD. Berta, P., Guerriero, C., and Levaggi, R. (2021), Hospitals’ strategic behaviours and patient mobility: Evidence from Italy, Socio-Economic Planning Sciences, 77, 101030. Besley, T., and Case, A. (1995), Incumbent behavior: Vote-seeking, tax-setting, and yardstick competition, American Economic Review, 85 (1), 25–45. Besley, T., and Coate, S. (2003), Centralized versus decentralized provision of local public goods: A political economy approach, Journal of Public Economics, 87 (12), 2611–37. Blöchlinger, H., Bartolini, D., and Stossberg, S. (2016), Does Fiscal Decentralisation Foster Regional Convergence? OECD Economic Policy Paper, September 2016, No. 17, Brussels, OECD Library. Bloomberg (2018), These are the economies with the most (and least) efficient health care, www​ .bloomberg​.com/​news/​articles/​2018​-09​-19/​u​-s​-near​-bottom​-of​-health​-index​-hong​-kong​-and​ -singapore​-at​-top (last accessed 4 February 2022). Bloomberg (2019), These are the world’s healthiest nations, www​.bloomberg​.com/​news/​articles/​2019​ -02​-24/​spain​-tops​-italy​-as​-world​-s​-healthiest​-nation​-while​-u​-s​-slips (last accessed 4 February 2022). Boadway, R., Buettner, T., and Wildasin, D.E. (2008), New directions in fiscal federalism: Introduction, Journal of Public Economics, 92 (12), 2285–7. Boetti, L., Piacenza, M., and Turati, G. (2012), Decentralization and local governments’ performance: How does fiscal autonomy affect spending efficiency? FinanzArchiv, 68 (3), 269–302. Bordignon, M. (2006), Fiscal decentralization: How to harden the budget constraint, in P.J. Wierts and S. Deroose (eds), Fiscal Policy Surveillance in Europe, Basingstoke, Palgrave Macmillan, pp. 109–36. Bordignon, M. (2017), Italy: Fiscal federalism and the division between North and South, Politica economica, Journal of Economic Policy, 20 (3), 269–88. Bordignon, M., Colombo, L., and Galmarini, U. (2008), Fiscal federalism and lobbying, Journal of Public Economics, 92 (12), 2288–301. Bordignon, M., Coretti, S., Piacenza, M., and Turati, G. (2020), Hardening sub-national budget constraints via administrative subordination: The Italian experience of regional recovery plans, Health Economics, 29 (11), 1378–99. Bordignon, M., and Turati, G. (2009), Bailing out expectations and public health expenditure, Journal of Health Economics, 28 (2), 305–21. Bosa, I., Castelli, A., Castelli, M., Ciani, O., Compagni, A., Galizzi, M.M., Garofano, M., Ghislandi, S., Giannoni, M., Marini, G., and Vainieri, M. (2021), Corona-regionalism? Differences in regional responses to COVID-19 in Italy, Health Policy, 125 (9), 1179–87. Brandolini, A., and Torrini, R. (2010), Disuguaglianza dei redditi e divari territoriali: l’eccezionalità del caso Italiano, Rivista delle Politiche Sociali, 3, 37–58.

106  Handbook on the political economy of health systems Bruzzi, S. (2012), Healthcare Regionalisation and Patients’ Mobility: The Challenges for a Sustainable Italian Healthcare Service, Pavia, Jean Monnet Centre. Buzelli, M.L., and Boyce, T. (2021), The privatization of the Italian national health system and its impact on health emergency preparedness and response: The COVID-19 case, International Journal of Health Services, 51 (4), 501–8. Calsamiglia, X., Garcia-Milà, T., and McGuire, T.J. (2006), Why do differences in the degree of fiscal decentralization endure?, http://​ideas​.repec​.org/​p/​ces/​ceswps/​_1877​.html (last accessed 4 February 2022). Camera dei Deputati (2022), La spending review sanitaria, Servizio studi, Rome, 30 June, https://​temi​ .camera​.it/​leg18/​temi/​tl18​_controllo​_spesa​_sanitaria​.html (last accessed 4 February 2022). Cannari, L., Viesti, G., and Zanardi, A. (2019), Interregional disparities in Italy: Structural changes and public policies – A brief introduction, Politica economica, Journal of Economic Policy, 2, 159–72. Cavalieri, M., and Ferrante, L. (2016), Does fiscal decentralization improve health outcomes? Evidence from infant mortality in Italy, Social Science & Medicine, 164 (C), 74–88. Chandra, A., Cutler, D., and Song, Z. (2012), Who ordered that? The economics of treatment choices in medical care, in M.V. Pauly, T.G. McGuire and P.P. Barros (eds), Handbook of Health Economics, Vol. II, Amsterdam, North Holland, pp. 397–431. Cheshire, P., and Gordon, I. (1998), Territorial competition: Some lessons for policy. Annals of Regional Science, 32 (3), 321–46. Cornes, R., and Sandler, T. (1994), The comparative static properties of the impure public good model, Journal of Public Economics, 54 (3), 403–21. Costa-Font, J., and Greer, S.L. (2013), Federalism and Decentralization in European Health and Social Care, Basingstoke, Palgrave Macmillan. Costa-Font, J., Levaggi, R., and Turati, G. (2020), Resilient managed competition during pandemics: Lessons from the Italian experience during COVID-19, Health Economics, Policy and Law, 17 (2), 212–9. Costa-Font, J., and Rico, A. (2006), Vertical competition in the Spanish National Health System (NHS), Public Choice, 128 (3), 477–98. Costa-Font, J., and Turati, G. (2018), Regional healthcare decentralization in unitary states: Equal spending, equal satisfaction? Regional Studies, 52 (7), 974–85. Crivelli, E., and Staal, K. (2013), Size, spillovers and soft budget constraints, International Tax and Public Finance, 20 (2), 338–56. Daniele, V., and Malanima, P. (2011), Il divario Nord-Sud in Italia 1861–2011, Rubbettino, Soveria Mannelli. De Belvis, A.G., Ferrè, F., Specchia, M.L., Valerio, L., Fattore, G., and Ricciardi, W. (2012), The financial crisis in Italy: Implications for the healthcare sector, Health Policy, 106 (1), 10–16. De Curtis, M., Bortolan, F., Diliberto, D., and Villani, L. (2021), Pediatric interregional healthcare mobility in Italy, Italian Journal of Pediatrics, 47, 139. Depalo, D. (2019), The side effects on health of a recovery plan in Italy: A nonparametric bounding approach, Regional Science and Urban Economics, 78, 103466. Di Novi, C., Piacenza M., Robone S., and Turati G. (2019), Does fiscal decentralization affect regional disparities in health? Quasi-experimental evidence from Italy, Regional Science and Urban Economics, 78, 103465. Dirindin, N. (2019), Proposte di autonomia differenziata e rischi per il Servizio Sanitario Nazionale, Politiche Sociali, 3, 533–6. Dirindin, N. (2020), Le politiche sanitarie alla luce della pandemia, Parolechiave, 2, 51–9. Eyraud, L., and Lusinyan, L. (2013), Vertical fiscal imbalances and fiscal performance in advanced economies, Journal of Monetary Economics, 60 (5), 571–87. Ferrario, C., and Zanardi, A. (2011), Fiscal decentralization in the Italian NHS: What happens to interregional redistribution?, Health Policy, 100 (1), 71–80. Fiorani, G., and Meneguzzo, M. (2008), I Piani Regionali di Rientro: Analisi dinamica sistemica, MECOSAN, 68, 27–44. Fondazione Gimbe (2019), Il definanziamento 2010-2019 del SSN, Report 7/2019, www​.gimbe​.org/​ osservatorio/​Report​_Osservatorio​_GIMBE​_2019​.07​_Definanziamento​_SSN​.pdf (last accessed 4 February 2022).

Efficiency and equity effects of healthcare decentralization  107 Fondazione Gimbe (2020), La mobilità sanitaria interregionale nel 2018, www​.gimbe​.org/​osservatorio/​ Report​_Osservatorio​_GIMBE​_2020​.02​_Mobilita​_sanitaria​_2018​.pdf (last accessed 4 February 2022). Francese, M., Piacenza, M., Romanelli, M., and Turati, G. (2014), Understanding inappropriateness in health spending: The role of regional policies and institutions in Caesarean deliveries, Regional Science and Urban Economics, 49, 262–77. Garattini, L., Curto, A., and Freemantle, N. (2016), Access to primary care in Italy: Time for a shake-up? European Journal of Health Economics, 17 (2), 113–6. Gil, C., Pascual, P.P., and Rapún Gárate, M. (2004), Regional economic disparities and decentralisation, Urban Studies, 41 (1), 71–94. Greer, S.L., and Da Fonseca, E.M. (2015), Decentralization and health system governance, in E. Kuhlmann, R.H. Blank, I.L. Bourgeault and C. Wendt (eds), The Palgrave International Handbook of Healthcare Policy and Governance, London, Palgrave Macmillan, pp. 409–24. Greer, S.L., Jarman, H., and Baeten, R. (2016), The new political economy of healthcare in the European Union: The impact of fiscal governance, International Journal of Health Services: Planning, Administration, Evaluation, 46 (2), 262–82. Greer, S.L., Jarman, H., Rozenblum, S., and Wismar, M. (2020a), How are countries centralizing governance and at what stage are they doing it? COVID-19 Health System Response Monitor: Cross-Country Analysis, WHO, European Commission, European Observatory on Health Systems and Policies, 19 April 2020, https://​analysis​.covid19healthsystem​.org/​index​.php/​2020/​04/​19/​how​-are​-countries​ -centralizing​-governance​-and​-at​-what​-stage​-are​-they​-doing​-it/​ (last accessed 20 January 2022). Greer, S.L., Jarman, H., Rozenblum S., and Wismar, M. (2020b), Who’s in charge and why? Centralization within and between governments, Eurohealth, 26 (2), 99–103. Grembi, V., Nannicini, T., and Troiano, U. (2016), Do fiscal rules matter? American Economic Journal: Applied Economics, 8 (3), 1–30. Huber, B., and Runkel, M. (2008), Interregional distribution and budget institutions under asymmetric information, Journal of Public Economics, 92 (12), 2350–61. Inman, R.P., and Rubinfeld, D.L. (1997), Rethinking federalism, Journal of Economic Perspectives, 11 (4), 43–64. Istat (2022), Rapporto Annuale sulla situazione del Paese 2022, www​.istat​.it/​storage/​rapporto​-annuale/​ 2022/​Rapporto​_Annuale​_2022​.pdf (last accessed 4 February 2022). Joumard, I., André, C., and Nicq, C. (2010), Healthcare Systems: Efficiency and Institutions, OECD Economics Department Working Paper 769, Brussels: OECD Library. Kessler, A.S., and Lülfesmann, C. (2005), Tiebout and redistribution in a model of residential and political choice, Journal of Public Economics, 89 (2–3), 501–28. King, D. (1984), Fiscal Tiers: The Economics of Multi-Level Government, London, Allen & Unwin. Kornai, J. (1986), The soft budget constraint, Kyklos, 39 (1), 3–30. Kornai, J. (2009), The soft budget constraint syndrome in the hospital sector, International Journal of Health Economics and Management, 9 (2), 117–35. Kornai, J., Maskin E., and Roland, G. (2003), Understanding the soft budget constraint, Journal of Economic Literature, 41 (4), 1095–136. Kringos, D.S., Boerma, W.G., Hutchinson, A., and Saltman, R.B. (eds) (2015), Building Primary Care in a Changing Europe: Case Studies, London, WHO Regional Office for Europe. Kurotschka, P.K., Serafini, A., Demontis, M., Serafini, A., Mereu, A., Moro, M.F., Carta, M.G., and Ghirotto, L. (2021), General practitioners’ experiences during the first phase of the COVID-19 pandemic in Italy: A critical incident technique study, Frontiers in Public Health, 9, doi: 10.3389/ fpubh.2021.623904. Levaggi, L., and Levaggi, R. (2016), Devolution and grant-in-aid design for the provision of impure public goods, SpringerPlus, 5, 282, doi: 10.1186/s40064-016-1919-9. Levaggi, R. (2002), Decentralised budgeting procedures for public expenditure, Public Finance Review, 30 (2): 273–95. Levaggi, R. (2007), Fiscal federalism in the presence of impure public goods, Studi economici, 91, 5–23. Levaggi, R. (2010), From local to global public goods: How should we write the utility function, Economic Modelling, 27 (5), 1040–42.

108  Handbook on the political economy of health systems Levaggi, R. (2018), Public finance and soft budgets, in J.H. Hamilton (ed.), Oxford Research Encyclopedia of Economics and Finance, Oxford, Oxford University Press, https://​oxfordre​.com/​economics/​view/​ 10​.1093/​acrefore/​9780190625979​.001​.0001/​acrefore​-9780190625979​-e​-50 (last accessed 4 February 2022). Levaggi, R., and Menoncin, F. (2011), Devolution in the provision of merit and impure public goods: Who gains, who loses, https://​ssrn​.com/​abstract​=​1194923 (last accessed 4 February 2022). Levaggi, R., and Menoncin, F. (2013), Soft budget constraints in healthcare: Evidence from Italy, European Journal of Health Economics, 14 (5), 725–37. Levaggi, R., and Menoncin, F. (2014), Healthcare expenditure decisions in the presence of devolution and equalisation grants, International Journal of Healthcare Finance and Economics, 14 (4), 355–68. Levaggi, R., and Menoncin, F. (2017), Would less regional income distribution justify the present call for devolution? International Tax and Public Finance, 24 (5), 780–99. Levaggi, R., and Smith, P.C. (1994), On the intergovernmental fiscal game, Public Finance, 49 (1), 72–86. Levaggi, R., and Smith, P. (2005), Decentralization in healthcare: Lessons from public economics, in P.C. Smith, L. Ginnelly and M. Sculpher (eds), Health Policy and Economics: Opportunities and Challenges, Maidenhead, Open University Press, pp. 223–47. Magnussen, J., and Martinussen, P.E. (2013), From centralization to decentralization, and back: Norwegian healthcare in a Nordic perspective, in S.L. Greer and J. Costa-Font (eds), Federalism and Decentralization in European Health and Social Care, Cham: Springer, pp. 101–20. Marchildon, G., and Bossert, T. (eds) (2018), Federalism and Decentralization in Healthcare: A Decision Space Approach, Toronto, University of Toronto Press. Marchitto, F. (2001), Benchmarking nella pubblica amministrazione. Una metodologia per il miglioramento continuo, Milan, Franco Angeli. Martell, C.R., and Smith, B.M. (2004), Grant levels and debt issuance: Is there a relationship? Is there symmetry? Public Budgeting and Finance, 24 (3), 65–81. Martinez-Vazquez, J., and McNab, R.M. (2003), Fiscal decentralization and economic growth, World Development, 31 (9), 1597–616. Mauro, M., and Giancotti, M. (2021), Italian responses to the COVID-19 emergency: Overthrowing 30 years of health reforms? Health Policy, 125 (4), 548–52. Mauro, M., Maresso, A., and Guglielmo, A. (2017). Health decentralization at a dead-end: Towards new recovery plans for Italian hospitals. Health Policy, 121 (6), 582–7. McKinnon, R.I. (1997), The logic of market-preserving federalism, Virginia Law Review, 83 (7), 1573–80. McLure, C.E. (1983), Tax Assignment in Federal Countries, Canberra, Australian National University. Ministero della Salute (2017), Annuario Statistico del Servizio Sanitario Nazionale, Rome, Ministero della Salute. Mookherjee, D. (2015), Political decentralization, Annual Review of Economics, 7, 231–49. Neri, S., and Mori, A. (2017), Crisi economica, politiche di austerità e relazioni intergovernative nel Ssn. Dalla regionalizzazione al federalismo differenziato? Autonomie locali e servizi sociali, 40 (2), 201–20. Northcott, D., and Llewellyn, S. (2005), Benchmarking in UK health: A gap between policy and practice? Benchmarking: An International Journal, 12 (5), 419–35. Oates, W.E. (1972), Fiscal Federalism, New York, Harcourt Brace Jovanovich. Oates, W.E. (1999), An essay on fiscal federalism, Journal of Economic Literature, 37 (3), 1120–49. Oates, W.E. (2005), Toward a second-generation theory of fiscal federalism, International Tax and Public Finance, 12 (4), 349–73. OECD (2014), OECD Reviews of Healthcare Quality: Italy 2014 – Raising Standards, Paris, OECD Publishing. OECD (2020), COVID-19 and Fiscal Relations across Levels of Government, Paris, OECD Publishing. Ogawa, H., and Wildasin, D.E. (2009), Think locally, act locally: Spillovers, spillbacks and efficient decentralized policy making, American Economic Review, 99 (4), 1206–17. Onida, V. (2020), Perché le clausole di supremazia non servono, Il Sole 24ore, 27 March. Pavolini, E., and Vicarelli, G. (2012), Is decentralization good for your health? Transformations in the Italian NHS, Current Sociology, 60 (4), 472–88.

Efficiency and equity effects of healthcare decentralization  109 Pecoraro, F., Luzi, D., and Clemente, F. (2021), The efficiency in the ordinary hospital bed management: A comparative analysis in four European countries before the COVID-19 outbreak, PLoS One, 16 (3), doi: 10.1371/journal.pone.0248867. Petretto, A. (2000), On the cost-benefit of the regionalisation of the National Health Service, Economics of Governance, 1 (3), 213–32. Pettersson-Lidbom, P. (2010), Dynamic commitment and the soft budget constraint: An empirical test, American Economic Journal: Economic Policy, 2 (3), 154–79. Pettersson-Lidbom, P., and Dahlberg, M. (2003), An Empirical Approach for Evaluating Soft Budget Constraints, Working Paper No. 28, Uppsala, Department of Economics, Uppsala University. Piacenza, M., and Turati, G. (2014), Does fiscal discipline towards subnational governments affect citizens’ well-being? Evidence on health, Health Economics, 23 (2), 199–224. Piacenza, M., Turati, G., and Vannoni, D. (2010), Restructuring hospital industry to control public health care expenditure: The role of input substitutability, Economic Modelling, 27 (4), 881–90. Piperno, S. (2020), Ripensare il regionalismo in tempi di pandemia? Centro studi sul federalismo, Commenti, 174, 8 April. Plagg, B., Piccoliori, G., Oschmann, J., Engl, A., and Eisendle, K. (2021), Primary healthcare and hospital management during COVID-19: Lessons from Lombardy, Risk Management and Healthcare Policy, 14, 3987–92. Poggi, A., and Sobrino, G. (2021), La Corte, di fronte all’emergenza Covid, espande la profilassi internazionale e restringe la leale collaborazione (ma con quali possibili effetti? nota a Corte Cost., sentenza n. 37/2021), Osservatorio Costituzionale, 4, 231–61. Prud’homme, R. (1995), The dangers of decentralization, World Bank Research Observer, 10 (2), 201–20. Putnam, R.D., Leonardi, R., and Nanetti, R. (1993), Making Democracy Work: Civic Traditions in Modern Italy, Princeton, Princeton University Press. Puy, M.S. (2007), Skill distributions and the compatibility between mobility and redistribution, Regional Science and Urban Economics, 37 (3), 345–62. Qian, Y., and Weingast, B.R. (1997), Federalism as a commitment to reserving market incentives, Journal of Economic Perspectives, 11 (4), 83–92. Repullo, J.R. (2007), Gasto sanitario y descentralización: ¿saldrá a cuenta haber transferido el INSALUD?, Presupuesto y Gasto Público, 49 (4), 47–66. Reverte-Cejudo, D., and Sánchez-Bayle, M. (1999), Devolving health services to Spain’s autonomous regions, British Medical Journal, 318 (7192), 1204–5. Ricci, A., Barzan, E., and Longo, F. (2021), How to identify the drivers of patient inter-regional mobility in Beveridgean systems? Critical review and assessment matrix for policy design & managerial interventions, Health Services Management Research, 34 (4), 258–68. Rodden, J. (2002), The dilemma of fiscal federalism: Grants and fiscal performance around the world, American Journal of Political Science, 46 (3), 670–87. Rodden, J. (2006), Achieving fiscal discipline in federations: Germany and the EMU, in P. Wierts, S. Deroose, E. Flores and A. Turrini (eds), Fiscal Policy Surveillance in Europe, New York, NY: Palgrave Macmillan, pp. 137–60. Rodden, J., Eskeland, G.S., and Litvack, J. (2003), Fiscal Decentralization and the Challenges of Hard Budget Constraint, Cambridge, MA, MIT Press. Rodríguez-Pose, A., and Ezcurra, R. (2010), Does decentralization matter for regional disparities? A cross-country analysis, Journal of Economic Geography, 10 (5), 619–44. Rosen, H. (1988), Fiscal Federalism: Quantitative Studies, Chicago, University of Chicago Press. Rubinchik-Pessach, A. (2005), Can decentralisation be beneficial? Journal of Public Economics, 89 (7), 1231–49. Saltman, R.B. (2008), Decentralization, re-centralization and future European health policy, European Journal of Public Health, 18 (2), 104–6. Saltman, R.B., Bankauskaite, V., and Vrangbæk, K. (eds) (2007), Decentralisation in Healthcare Strategies and Outcomes, European Observatory on Health Systems and Policies Series, Maidenhead, Open University Press and McGraw-Hill Education. Schakel, H.C., Wu, E.H., and Jeurissen, P. (2018), Fiscal rules, powerful levers for controlling the health budget? Evidence from 32 OECD countries, BMC Public Health, 18 (1), 300.

110  Handbook on the political economy of health systems Schneider, A., (2003), Decentralization: Conceptualization and measurement, Stud Comp Int Dev, 38 (3), 32–56. Schragger, R.C. (2010), Decentralization and development, Virginia Law Review, 96 (8), 1837–910. Scognamiglio, A. (2014), Il benchmarking nella pubblica amministrazione: misurazione delle performance, confronto e condivisione di esperienze e risorse per migliorare la qualità dei servizi offerti al Cittadino, Tigor. Rivista di scienze della comunicazione e di argomentazione giuridica, 6 (1), 160–64. Shah, A. (2004), Fiscal Decentralization in Developing and Transition Economies: Progress, Problems, and the Promise, Policy Research Working Paper, 3282, Washington, DC, World Bank. https://​ openknowledge​.worldbank​.org/​handle/​10986/​14114 (last accessed 4 February 2022). Shankar, R., and Shah, A. (2003), Bridging the economic divide within countries: A scorecard on the performance of regional policies in reducing regional income disparities, World Development, 31 (8), 1421–41. Skinner, J. (2012), Causes and consequences of regional variations in healthcare, in M.V. Pauly, T.G. McGuire and P.P. Barros (eds), Handbook of Health Economics, Vol. II, Amsterdam, North Holland, pp. 45–93. Snoddon, T., and Wen, J.F. (2003), Grants structure in an intergovernmental fiscal game, Economics of Governance, 4 (2), 115–26. Tediosi, F., Gabriele, S., and Longo, F. (2009), Governing decentralization in healthcare under tough budget constraint: What can we learn from the Italian experience?, Health Policy, 90 (2–3), 303–12. Terlizzi, A. (2019), Explaining health system decentralization and recentralization, in Health System Decentralization and Recentralization, London, Palgrave Macmillan, pp. 43–78. Thonon, F., Watson, J., and Saghatchian, M. (2015), Benchmarking facilities providing care: An international overview of initiatives, SAGE Open Medicine, 3, doi: 10.1177/2050312115601692. Tiebout, C. (1956), A pure theory of local expenditures, Journal of Political Economy, 64 (5), 416–24. Tommasi, M., and Weinschelbaum, F. (2007), Centralization vs. decentralization: A principal-agent analysis, Journal of Public Economic Theory, 9 (2), 369–89. Torfs, L., Adriaenssens, S., Lagaert, S., and Willems, S. (2021), The unequal effects of austerity measures between income-groups on the access to healthcare: A quasi-experimental approach, International Journal for Equity in Health, 20, 79. Toth, F. (2014), How healthcare regionalisation in Italy is widening the North–South gap, Health Economics, Policy and Law, 9 (3), 231–49. Treisman, D. (2007), The Architecture of Government: Rethinking Political Decentralization, Cambridge, Cambridge University Press. Tresch, R. (2002), Public Finance: A Normative View, 2nd edn, San Diego, Academic Press. Turati, G. (2013), The Italian Servizio Sanitario Nazionale: A renewing tale of lost promises, in J. Costa-Font and S. Greer (eds), Federalism and Decentralization in European Health and Social Care, London, Palgrave Macmillan, pp. 47–66. Vicarelli, G., and Neri, S. (2021), Una catastrofe vitale? Le scelte di politica sanitaria per far fronte al Covid-19, Politiche Sociali, 8 (2), 233–53. Weingast, B.R. (2009), Second generation fiscal federalism: The implications of fiscal incentives, Journal of Urban Economics, 65 (3), 279–93. WHO (2012), Health Policy Responses to the Financial Crisis in Europe, Policy Summary 5, Copenhagen, WHO Regional Office for Europe and European Observatory on Health Systems and Policies. WHO (2019), State of Health in the EU: Italy Country Health Profile 2019, European Observatory on Health Systems and Policies, Paris, OECD and World Health Organization. Wildasin, D.E. (1997), Externalities and Bailouts: Hard and Soft Budget Constraints in Intergovernmental Fiscal Relations, Policy Research Working Paper, 1843, Washington, DC, World Bank. Wildasin, D.E. (2001), Externalities and Bailouts: Hard and Soft Budget Constraints in Intergovernmental Fiscal Relations, Public Economics 0112002, Munich, University Library of Munich, http://​ideas​ .repec​.org/​p/​wpa/​wuwppe/​0112002​.html (last accessed 4 February 2022). Wildasin, D.E. (2004), The institutions of federalism: Towards an analytical framework, National Tax Journal, 57 (2), 247–72.

Efficiency and equity effects of healthcare decentralization  111 World Bank (n.d.), Hospital beds (per 1,000 people): Italy. https://​data​.worldbank​.org/​indicator/​SH​ .MED​.BEDS​.ZS​?end​=​2018​&​locations​=​IT​&​most​_recent​_year​_desc​=​false​&​start​=​1960​&​view​=​chart (last accessed 4 February 2022). Zanardi, A. (2019), Audizione sulla definizione delle intese ai sensi dell’art. 116, 3° c., della Costituzione, UPB, 10 Luglio 2019.

8. Political economy of health care insurance expansion in Mexico David G. Lugo-Palacios and Alejandro Sanders Villa

1. INTRODUCTION In 2003, following 20 years of decentralisation reforms, Mexico conducted a major reform that transformed its health care system to extend health care insurance coverage to more than 50 million uninsured Mexicans. This reform created Seguro Popular (or Popular Health Insurance) – the subsidised insurance-based component of the reform – which offered free access at the point of delivery to an explicit set of health care interventions to the population not covered by any other public insurance scheme. In 2020, under a new political regimen that took office in 2018, Seguro Popular was substituted by a new model recentralising the public purchasing and delivery of health services to the population not covered by the Mexican social security system. This chapter first presents a description of the main characteristics of the Mexican health care system, and then an overview of the series of health reforms conducted in Mexico during the last 40 years that have shaped the non-social security institutions and the services they provide. This reform overview is divided into three sections: (1) the decentralisation process, (2) the expansion of health care coverage via Seguro Popular and (3) the recentralisation of the health system.

2.

THE MEXICAN HEALTH CARE SYSTEM

The General Health Law establishes in its fifth article that the national health system is formed by the federal and local entities of the Public Administration and by the physical and moral persons in the social and private sectors that provide health services with the goal of complying with the right of health protection included in the fourth article of the Mexican Constitution (Ley General de Salud, 2021). The most distinctive feature of the Mexican health system is its subdivision into various sub-systems where each sub-system replicates the fundamental health system activities for its affiliated populations: stewardship and revenue raising, as well as the purchasing and provision of health services (OECD, 2016). Therefore, in practice, each sub-system operates as a distinct health system, with little coordination of functions across them and with each institution owning and managing its own medical facilities, and employing its own health workforce (OECD, 2016). These sub-systems can be classified into three categories: social security institutions, non-social security institutions and a big, heterogeneous and largely unregulated private sector. The origin of social security in Mexico dates from 1943, during the Second World War, when the Instituto Mexicano del Seguro Social (IMSS) was established following the promul112

Political economy of health care insurance expansion in Mexico  113 gation of the Social Insurance Law. Original IMSS beneficiaries were mostly salaried private sector employees with permanent contracts, mostly working in the manufacturing industry, and some service-sector workers (González Block et al., 2020). Agricultural workers, who constituted two-thirds of the economically active population, were excluded from IMSS, although future schemes would eventually provide a differentiated level of coverage to the uninsured population. IMSS-Coplamar (now IMSS-Bienestar1), created in 1979, is one of these schemes, which is administered by IMSS and runs mostly in rural areas. It is funded through general taxation and offers health care services in first-level medical clinics and in second-level rural hospitals. Labour unions and key employers affected by the creation of IMSS protested against the Social Insurance Law. In particular, IMSS medical employees went on strike as they resisted becoming salaried employees of IMSS and thus losing their status as independent liberal professionals (González Block et al., 2020). To address their demands, IMSS offered generous salaries above industry standards to their physicians and, after negotiating with key employers, committed to providing medical services at standards above those offered to employees in the market (González Block et al., 2020). Also in 1943, the federal government decided to merge the Department of Health and the Ministry of Health and Assistance, both funded through general taxation, as a response to the concerns of public officials and influential physicians whose institutes and hospitals had been excluded from social insurance (González Block et al., 2020). During this time, agreements with state governments were created or strengthened to provide health care for the vulnerable population not covered by the social security. In this way, from its origin, the Mexican health care system was created as a fragmented public health system where labour status determines the sub-system from which each Mexican is entitled to receive health care. Ever since, there has been an important gap between the depth and quality of the services provided by social security institutions and those offered by non-social institutions (OECD, 2016). Figure 8.1 illustrates the complexity of the Mexican health system. Currently, the social security institutions, covering salaried workers in the formal sector of the economy and their dependants (approximately 60.4 million people, accounting for 48 per cent of the total Mexican population), are IMSS, Instituto de Seguridad y Servicios Sociales de los Trabajadores del Estado (ISSSTE) and the ones providing services to the workers of the National Mexican Oil Company (PEMEX), the Mexican Army and the Mexican Navy, as well as similar schemes that cover employees from local governments and some students from public universities (INEGI, 2021). These institutions provide ‘complete medical care’ (i.e., no explicit package of services), including prescribed drugs, without any co-payment. In addition to health services, the package of social security benefits includes a system of pensions, protection against occupational risk and child care centres, as well as housing and recreational services. The social security sub-systems are funded through payroll contributions by the employer and the employee with an additional allocation financed by general taxation. Social security institutions are responsible for their own funds and resource allocation decisions. The Mexican population not covered by the social security system (i.e., the unemployed, non-salaried workers, the self-employed and informal workers) is entitled to receive preventive and curative health care services only at the facilities managed by non-social security institutions, such as state health services (SESA – state health ministries), the national health institutes (INSALUD) and the federal reference hospitals (HFR) managed by the Ministry of

114  Handbook on the political economy of health systems Health (MoH), as well as in the medical units of IMSS-Bienestar. These institutions are mostly financed through general taxation. Users of private health services belong to both the beneficiaries and non-beneficiaries of the social security schemes and receive medical care in heterogeneous private hospitals and medical clinics. Typically, the use of these services implies a high expenditure that can be financed in two ways: directly from users (out-of-pocket expenditure) or through private insurance companies, with the latter only covering about 8 per cent of the Mexican population (González Block et al., 2020).

Source: Adapted from Frenk et al., 2003.

Figure 8.1

Overview of the Mexican health system

In 2003, the General Health Law was reformed to establish the System of Social Protection in Health (SPSS), which introduced new financial rules to fund population-based interventions and personal health care interventions, the latter financed through Seguro Popular, the subsidised insurance-based component of the SPSS that offered free access, at the point of delivery, to an explicit set of health care interventions listed in the Universal Catalogue of Essential Services (CAUSES), as well as some treatments defined as catastrophic in financial terms2 (González-Pier et al., 2006; Knaul et al., 2012). Seguro Popular was operated by the 32 state health ministries and its beneficiaries served by SESA, INSALUD and HFR. In 2012, the federal government declared that universal health insurance coverage was achieved, with Seguro Popular covering 51.8 million people and the social security institutions more than 69 million people.3 This achievement was also recognised by the World Health Organization (WHO) and the international community (The Lancet, 2012). However, in many cases, this coverage was just nominal and not real, in the sense that not all Mexicans had effective access to health care and not even all were formally registered with a health care institution (Fundación Mexicana para la Salud, 2012). Despite the ruling political party chang-

Political economy of health care insurance expansion in Mexico  115 ing, Seguro Popular continued its operation during the following presidential administration from 2012 to 2018. However, in January 2020, Seguro Popular was terminated and replaced by the National Institute of Health for Wellbeing (INSABI). INSABI was designed as a centralised health system with integrated financing and delivery aiming to provide universal and free access to health services and medicines to the population not covered by the social security system. While INSABI legally guarantees all basic primary and secondary care needs, it is less clear regarding the access to specialised care (Reyes Morales et al., 2020). Unlike Seguro Popular, INSABI was created without explicit rules for allocating financial resources for health care, and its implementation followed political and ideological principles rather than technical planning and evidence from past reforms evaluations (Ochoa-Moreno, 2020; Reich, 2020). Two years after having launched INSABI, in early 2022, President López Obrador announced a new change in the expansion of health care strategy, transferring the responsibility of its implementation from INSABI to IMSS-Bienestar. The unmet political and technical expectations of INSABI and the possibility of accelerating the strategy through a more centrally controlled institution – like IMSS – help explain the reversal of the original strategy of the López Obrador administration. As described below, this was not the first time that IMSS was envisioned as the channel to expand health care insurance in Mexico.

3.

THE DECENTRALISATION PROCESS IN THE MEXICAN HEALTH CARE SYSTEM (1980–2018)

Mexico is organised as a federal republic comprising 32 states. The Mexican Constitution assigns health functions to both the federal and the state governments. The federation has the capacity to issue guidelines, execute actions in matters of national importance (such as health regulation and disease control) and provide health care services through the hospitals directly managed by the MoH. States are responsible for health service provision within their jurisdictions, complemented by federal actions that go beyond their borders (González Block et al., 2018). The relationship between the federal and states government regarding the financing, purchasing and provision of the health care services to the population not covered by social security institutions was subject to a long series of decentralisation reforms during the last 40 years aimed at expanding access to health care as well as improving the quality throughout the system (OECD, 2016). This decentralisation process was not exclusive to the health sector as it was part of the ‘decentralisation of national life’ which was a key principle in the De la Madrid administration (Soberón-Acevedo and Martínez-Narváez, 1996). By transferring responsibilities to the state and local governments, the federal government intended to improve the quality and efficiency of public services, such as education and health care (Soberón-Acevedo and Martínez-Narváez, 1996). For its study, the decentralisation process in the health sector can be divided into three waves: ● First wave: 1982–1988 ● Second wave: 1994–2000 ● Third wave: 2000–2018

116  Handbook on the political economy of health systems 3.1

First Wave of Decentralisation (1982–1988)

When Miguel de la Madrid became president in the early 1980s, Mexico was going through a serious economic crisis which motivated the reduction of federal government expenditure across sectors. This motivated different austerity measures which were smoothed by transferring spending responsibilities to the state governments (Autrique-Echeveste, 2012). In this setting, the right to health protection was added to the Mexican Constitution and in 1984 the General Health Law was promulgated. This law provided the legal basis for the first attempt to decentralise health services offered to the population not covered by the social security system. This resulted in the creation of 14 state health services (as only 14 states signed the decentralisation agreements with the federal government) that were given the management responsibility of some first- and second-level hospitals and medical clinics, as well as the administrative responsibility of integrating IMSS-Coplamar services into their jurisdiction (Merino, 2003). A gradual decentralisation approach was preferred to a complete and immediate approach due to important budget constraints and capacity differences across states (Soberón-Acevedo and Martínez-Narváez, 1996). However, the administrative decentralisation of the health sector was incomplete as it was only limited to the provision of health care services without transferring financing responsibilities to the state level (Autrique-Echeveste, 2012). This meant that states could not raise their own revenues nor plan their budgets and relied almost completely on transfers from the federation. Therefore, most – if not all – health planning decisions were taken at the federal level and states were merely operators of the federal plans. While this gradual decentralisation stage focused on the most industrialised states with higher-than-average per capita incomes, greater financing capacity and the highest levels of social security coverage, one of the main limitations of this decentralisation attempt was precisely the huge diversity in management and infrastructure capacity across states. Hence, the national health and finance authorities faced a dilemma between strengthening and decentralising local health systems. The solution was to strengthen the state systems up to the point to which they could be decentralised (Soberón-Acevedo and Martínez-Narváez, 1996). This wave was brought to a halt by adverse economic developments. Political unrest due to the economic environment vetoed change at a time when federal resources were at a historic low, making it impossible for the government to fund the transitory costs of the reform (OECD, 2016). During the Salinas de Gortari administration (1988–1994), local health systems were strengthened, but without transferring further responsibilities to the state governments as occurred before and after this period (Merino, 2003). 3.2

Second Wave of Decentralisation (1994–2000)

In the mid-1990s, the decentralisation process continued with the transfer of more functions and responsibilities to states alongside the corresponding resources in order to complete the decentralisation and strengthen the state health services (OECD, 2016). These efforts were based on the national agreement signed in 1996 amongst the federal government, the 32 states and two labour unions. This agreement considered four strategies: (1) the decentralisation of health services for the uninsured population, (2) the design of state health services, (3) coverage expansion by offering a basic package of health services and (4) quality improvement through better coordination amongst institutions (Autrique-Echeveste, 2012).

Political economy of health care insurance expansion in Mexico  117 During this stage, the federal fund for health services (FASSA) was created, comprising earmarked fiscal resources in the federal budget for the health care of the uninsured population. FASSA transfers to the states were based on infrastructure, workforce, operation costs and agreed investments. This was an important step towards the financing of the new decentralised health system as before the creation of FASSA, all the federal funds for health care were managed by the federal government through the MoH. In this period, states assumed responsibility for the direct operation of primary and secondary care and hence a decentralised public entity within each state was created to be in charge of the operation and supervision of devolved health care facilities as well as the management of the federal transfers. Regarding financing responsibilities, states acquired autonomy for budget planning and spending while agreeing to increase financing of health care from their own resources. These enhanced responsibilities and increased federal transfers can be seen as the response from the federal government to the pressure induced by increasing political competition at the national and state levels resulting from – at that time – the most important legislative and regional electoral defeats suffered by the then hegemonic PRI party since its creation at the end of the Mexican revolution.4 However, all this was done without addressing the capacity differences across states that motivated the gradual decentralisation approach of the 1980s, and thus the new resource allocation rules preserved the existing inequalities across states (Autrique-Echeveste, 2012). Other limitations of the second wave of the decentralisation process are the following: first, the regulatory function of the federal MoH was not clearly defined, which restricted the ability of the federal government to implement national health policies (Autrique-Echeveste, 2012). Second, there were no clear incentives for the states to deliver their commitments, namely the effective coverage of the basic package of health services or increasing their contributions to health care financing. Third, there were no clear transparency and accountability mechanisms as states were not really asked to provide detailed reports to the federal government on the way they spent federal resources for health care. 3.3

Third Wave of Decentralisation (2000–2018)

The first alternation between political parties at the presidential level in Mexico occurred in 2000. When President Vicente Fox – from the PAN political party – took office that year, the Mexican health system was characterised by low health expenditure (5.1 per cent of gross domestic product, or GDP), high out-of-pocket expenditure and a lack of financial protection on health for the population not covered by the social security system, and the federal transfers to states contributed to preserving the inequalities in the supply of health care. In this context, in 2003, the General Health Law was reformed to establish the System of Social Protection in Health, which introduced new financial rules to fund both population-based and personal health care interventions, the latter being financed through Seguro Popular. The financial structure of Seguro Popular was based on a tripartite scheme with contributions for every beneficiary based on a non-risk-adjusted capitation scheme coming from three different sources: the federal government, the state governments and the beneficiaries, with beneficiaries in the lowest income deciles being exempted from these fees. Importantly, restrictions on the funding of human resources, medicines, public health and infrastructure were clearly defined.

118  Handbook on the political economy of health systems Federal contributions to Seguro Popular were transferred from the federal Ministry of Finance (MoF) to each state treasury and then from state treasuries to state health ministries. Within each state health ministry, a new body – the State Regimen for Social Protection in Health, or REPSS – was created and charged with allocating federal and state funds to the state-level provider network. In 2014, corruption scandals and state government inefficiency led to these purchasers being re-established as decentralised semi-autonomous units of state governments, and they were ordered by the General Health Law to directly receive at least 50 per cent of the Seguro Popular transfers (González Block et al., 2020). While the health services covered by Seguro Popular were still provided by the state health services, having two different bodies in charge of the provision of health care and of its financing represented an important step towards the separation of purchasing and provision of health services. However, the lack of effective accountability for Seguro Popular’s funds sent to states, along with ideological and political factors, motivated the elimination of Seguro Popular and the creation of INSABI in early 2020 (Reich, 2020). Anti-corruption was indeed one of the founding principles of INSABI as the new political regimen argued that corruption could be addressed by centralising control over purchasing and service delivery (Reich, 2020).

4.

SEGURO POPULAR: EXPANDING HEALTH CARE COVERAGE IN A FRAGMENTED AND QUASI-​ DECENTRALISED SYSTEM

Before the 2003 health reform, Mexicans not covered by the social security system accessed federally and state-run health care facilities funded mostly through historical supply-based budget allocations without explicit entitlements. Health care for this population was not comprehensive and patients receiving it needed to pay a recovery fee at the point of delivery. Despite this fee being means-tested and heavily subsidised, it still contributed to the high proportion of out-of-pocket financing of health care, which represented more than 50 per cent of total health expenditure in 2000 (Knaul et al., 2012). Shortages of medicines were common, thus resulting in uninsured families incurring additional out-of-pocket payments (Frenk et al., 2006). These payments, and those made to private providers that were relied on when the public sector failed, sometimes drove Mexicans into poverty (Lakin, 2010). In parallel, the population covered by the social security system received comprehensive health care from well-financed and vertically integrated federal institutions (Knaul et al., 2012). The start of the Fox administration in 2000 faced four important challenges in the health sector: (1) a general low level of investment in health, (2) inefficient health care financing due to the predominance of out-of-pocket payments, 3) an inequitable distribution of resources amongst health care institutions and amongst geographical regions and (4) no shared responsibilities in health care financing between the federal government and the 32 state governments (Secretaría de Salud, 2006). Consequently, a health reform was designed with five main objectives. First, to offer financial protection in health to the population without social security. Second, to create a culture of pre-payment through public and subsidised insurance with progressive premiums to reduce out-of-pocket expenditure. Third, to reduce the number of families driven into poverty as a result of covering their health care expenditures. Fourth, to increase the percentage of Mexico’s GDP that was allocated to health spending. Finally, to

Political economy of health care insurance expansion in Mexico  119 adjust the formula for distributing federal health funds to states (Secretaría de Salud, 2006; Lakin, 2010). The analysis of the 2003 health reform presented here focuses on its adoption, its implementation and its subsequent elimination. This analysis, deliberately, does not discuss the impact on health outcomes, equity and efficiency as this has been extensively covered in the literature (Knaul et al., 2012; Bosch and Campos-Vazquez, 2014; Knaul et al., 2018; Nikoloski and Mossialos, 2018; Conti and Ginja, 2020; Costa-Font et al., 2021; del Valle, 2021; Colchero et al., 2022). 4.1

Adoption of the Health Reform

Two contextual factors created a window of opportunity for the adoption of a health reform pursuing the expansion of health care coverage in Mexico (Gómez-Dantés et al., 2015). First, the epidemiological transition observed in Mexico at the start of the new millennium (Vos et al., 2020). The escalation of the noncommunicable diseases in the national burden of disease exerted additional pressure on the health system and demanded additional financial resources as well as the design of new models of care to address the increased health care needs of the uninsured population. Second, Mexico was also facing a democratic transition that led to the beginning of political alternation at the national level for the first time in more than 70 years. The democratisation process in Mexico opened political space for a major health reform focused on the democratisation of health (Gómez-Dantés et al., 2015). These enabling factors are, however, not sufficient conditions for the expansion of public health care coverage. The opportunity created by these factors needed to be exploited by political actors in the design of a reform proposal that could be incorporated into the national policy agenda (Gómez-Dantés et al., 2015). In the Mexican case, these efforts were led by Julio Frenk, the Minister of Health in the Fox administration, who built a team with strong technical and political skills who negotiated the support of key groups inside and outside the federal government. The original plan for the reform was to subsidise the participation of the uninsured population in IMSS (the largest social security institution in Latin America, covering salaried workers in the private sector and their families) in an effort to also increase the integration of the health system (Ortiz, 2006). However, the initiative to shift the budget to a demand-based allocation of health resources and to introduce a separation of purchaser and provider, along with explicit contracts between public purchasers and both public and private providers, clashed with IMSS’s vision and interests. The demand-driven finance model, in which the distribution of the budget would be based on the use of services by patients (as opposed to installed capacity) and separating provider and purchase functions, would have transformed IMSS internal organisation (Lakin, 2010). It would have required IMSS to compete with public and private health care providers and it would have radically altered labour relations within IMSS (Lakin, 2010). Faced with resistance to extending social security coverage to the uninsured through an IMSS reorganisation, the minister of health and his team designed a new insurance scheme, the Seguro Popular. Like the IMSS main insurance scheme, Seguro Popular would be funded by a tripartite scheme in which employer contributions would be substituted with state contributions. However, as opposed to the coverage offered by the existing social security institutions

120  Handbook on the political economy of health systems in Mexico, the Seguro Popular would cover an explicit set of health services and would be based on the separation of the provision and purchasing functions. To demonstrate the feasibility of Seguro Popular and to gain political support, a pilot study was launched in five states in 2001 (Secretaría de Salud, 2006). However, the main opponents of the reform did not find the evidence compelling and continued to oppose the programme (Lakin, 2010). There were three main opponents of the reform, two of them within the federal government (Lakin, 2010; Gómez-Dantés et al., 2015). The IMSS director general opposed Seguro Popular even when the new scheme did not directly involve social security anymore. IMSS opposition was based on the argument that this new insurance scheme would divert resources from social security to Seguro Popular, would not address the health system fragmentation and would encourage informality in the Mexican workforce (Levy, 2008). A second opposing force to the reform within the federal government was the MoF, which based its opposition on fiscal grounds. The MoF was worried about the financial implications of the reform and wanted to condition its implementation on the approval of a fiscal reform that would raise more government revenues (Ortiz, 2006; Lakin, 2010; Gómez-Dantés et al., 2015). However, a previous fiscal reform initiative in the Fox administration had failed to be enacted, which meant that this condition was difficult to meet. The most important opposing force to the reform outside the federal government was represented by a large fraction of Mexico’s largest left-wing political party at that time (PRD), led by Andrés Manuel López Obrador, then Mayor of Mexico City and who would become President of Mexico in 2018. López Obrador, along with Mexico City’s minister of health, opposed the reform on ideological grounds. They argued that the annual premium that beneficiaries were asked to pay was incompatible with the idea of health care as a social right (Ortiz, 2006; Lakin, 2010; Gómez-Dantés et al., 2015). They also opposed Seguro Popular as they were concerned that it would undermine a free health care programme that PRD had started in Mexico City and the party did not want to lose the ability to take credit for that programme (Lakin, 2010). The political strategy to face this opposition was to present the reform not as a government or party project, but as a unified project – a project where the federal government and all the political parties in Congress were represented, made amendments and shared the credit for the reform (Ortiz, 2006). This process required intense lobbying and persuasion by the minister of health and his team within and outside the federal government for which the support from President Fox and the majority of the legislators of the political party in power was crucial (Sparkes et al., 2019). During the negotiations to pass the reform initiative into law, two decisions shaped the implementation of Seguro Popular. First, while the initiative presented to the Congress considered subsidised insurance premiums to be paid by affiliated families, legislators argued that the poor should be exempt from premiums altogether and agreed that the bottom two deciles of income would not pay to affiliate to Seguro Popular (Lakin, 2010). Second, in order to smooth the budget impact of the reform, the MoH and the MoF agreed that Seguro Popular was going to be implemented gradually with annual increases in its coverage until 2010, when it was expected to reach all its target population across the 32 states (Ortiz, 2006). In this way, Seguro Popular officially began on 1 January 2004.

Political economy of health care insurance expansion in Mexico  121 4.2

Implementation of Seguro Popular

In January 2004, once its pilot stage was completed, Seguro Popular started to be gradually rolled out across the country through collaboration agreements between the federal and state governments. The programme was intended to reach areas with the greatest need first, but in practice the roll-out was arbitrary, determined mainly by logistical and political factors, rather than need (Ochoa-Moreno, 2020). By 2007, the programme had already reached 98 per cent of the municipalities in Mexico (del Valle, 2021). This process involved intense political negotiation between the federal government and the states, who defined the timing to sign the collaboration agreement and the order of deployment of the programme in the different municipalities of their state (Ochoa-Moreno, 2020). The last collaboration agreement to be signed was that of Mexico City in July of 2005 after years of public opposition from its mayor and health minister. Two defining elements of the implementation of Seguro Popular were its eligibility criteria and its financing guidelines. The main eligibility criterion to affiliate to Seguro Popular was to not be covered by any other public insurance scheme. However, in practice, the lack of coordination between sub-systems and the absence of a reliable electronic registry impeded the adherence to this criterion; previous studies have indeed documented multiple coverage amongst Seguro Popular beneficiaries (Fundación Mexicana para la Salud, 2012). The target population definition in each state thus was not the product of a technical calculation, it was instead largely influenced by the negotiating capacity of each state. Each year, state and federal governments negotiated the target number of households (or individuals) that the state would enrol over the year with a maximum total number capped by the estimated number of eligible beneficiaries in the state. The operational guidelines of Seguro Popular established a non-risk-adjusted capitation payment for each person affiliated to the programme. The federal government provided most of the funding through two components: the ‘social contribution’ (equivalent to 3.92 per cent of the annual minimum wage) and the ‘federal solidarity contribution’ equivalent to 1.5 times the social contribution. States also provided solidarity contributions equivalent to 0.5 times the social contribution. All these contributions were allocated to states per affiliated household (or individual5) in each state. The collaboration agreements between the federal and state governments set the rules of states’ contribution to Seguro Popular, which was originally meant to pay for health services. However, as Seguro Popular rolled out, states proved reluctant to make these payments and the federal government was in a weak position to force them to pay up as it wanted to exploit the programme’s appeal in advance of the 2006 presidential election (Lakin, 2010). This situation led to non-transparent federal–state negotiations in which states ultimately reduced the effective state contribution as they were allowed to credit part of their contribution from previous infrastructure investments (Lakin, 2010). State governments, therefore, had clear incentives to encourage their population to affiliate to Seguro Popular. On the one hand, the amount of resources transferred from the federal government depended on the number of registered beneficiaries: the higher the affiliated population, the more federal resources were transferred. On the other hand, while, in principle, state contributions to Seguro Popular would also increase with higher affiliation, state governments could credit their contribution with a broad definition of health expenditure and not necessarily by allocating new resources to the system. The relative ambiguity of the state’s financial obligation implied that the investment from the federal government would end up substituting

122  Handbook on the political economy of health systems state investment (crowding-out effect), that it would not follow the same priorities as expected by Seguro Popular and that the funding would be insufficient to ensure effective coverage. In each state, REPSS were responsible for pooling federal and state funds and purchasing services on behalf of the enrolled population from public and private providers. The separation of purchasing and provision of health services was seen as a critical way to promote efficiency and allocate resources according to the population’s needs (Nigenda et al., 2015). However, REPSS were not consolidated as originally planned. In practice, most REPSS were located within the state ministry of health, which meant that REPSS were accountable to the state health minister and governor, thus violating the desired autonomy. Additionally, the mechanism for transferring resources from the federal government to the state’s health system was initially mediated by the state ministries of finance and state treasuries, making funds vulnerable to being (illegally) diverted for other purposes of greater political interest to the governor in turn. Therefore, the limited accountability mechanisms combined with the misaligned incentives of the entity that provided most of the resources (federal government) and the one that exercised most of them (state governments) facilitated the lack of transparency in the use of resources and the dilution of responsibility between the federal government and the state governments. Both parties were interested in politically capitalising on the programme’s benefits (Pueblita, 2013), but also in shifting to their counterpart – be it the federal financier or the local implementer – the responsibility for the limitations and deficiencies. Due to heterogeneity in the use of resources across states during the initial years of Seguro Popular, in 2008, regulations for the use of federal resources were incorporated into the annual federal–state agreements (Knaul et al., 2012). In this way, the 32 state governments – who were responsible for spending 89 per cent of the total resources of Seguro Popular – could not spend more than 40 per cent of the total federal allocation on personnel and no more than 30 per cent on pharmaceuticals, and no less than 20 per cent needed to be spent on preventive activities. Some states, however, did not adhere to these rules (Nigenda et al., 2015). In principle, states that did not comply with the budget limits had to return to the federal government the amount of resources that surpassed the budget ceilings before they could receive their subsequent federal contributions. In practice, while the non-adherence did generate tensions in some federal–state relations, Seguro Popular did not have the legal means to force states to behave according to general national norms (Nigenda et al., 2015). In 2014, in an effort to address the lack of transparency and the uncontrolled expenditure of resources, the federal government promoted a reform that transformed the REPSS into public decentralised bodies with their own legal personalities and patrimonies (Chemor Ruiz et al., 2018). In addition, new financial accounts were established in the federal treasury to allocate 50 per cent of the total Seguro Popular resources which had to be spent mainly on the acquisition of pharmaceuticals, health-related materials and outsourcing services, as well as on salary payments. Furthermore, in 2016 new penalties and the punishment of imprisonment were established for cases of illegal acts associated with Seguro Popular (Chemor Ruiz et al., 2018). However, despite these efforts to enhance transparency, the lack of appropriate accountability mechanisms meant that complaints filed by the federal MoH did not always translate into timely legal action and a clear commitment to the rule of law. Despite these challenges, Seguro Popular achieved important improvements. First and foremost, through Seguro Popular, more than 50 million uninsured people (~43 per cent of Mexico’s national population) were offered financial protection regarding health. A number of studies have found that Seguro Popular reduced catastrophic and impoverishing health

Political economy of health care insurance expansion in Mexico  123 expenditures (Knaul et al., 2018; Nikoloski and Mossialos, 2018). Existing evidence also suggest positive effects on health outcomes, including a decline in maternal and child mortality (Knaul et al., 2012; Celhay et al., 2019; Conti and Ginja, 2020). With respect to health inequalities and regional inequalities, the results have been mixed though (Autrique-Echeveste, 2012; Costa-Font et al., 2021). For example, on the one hand, inequality in the supply of health care for the uninsured population across the 32 states increased. Those with more and better infrastructure at the beginning of the decentralisation process improved their situation much more than the worse-off states, which are usually those where most of the poor population live (Autrique-Echeveste, 2012). On the other hand, differences in per capita health expenditure across states and across sub-systems have decreased significantly during the last 20 years (Knaul et al., 2012; Chemor Ruiz et al., 2018). While the path followed by Seguro Popular may not be the only way to achieve an insurance expansion in an upper-middle-income country like Mexico, given the Mexican political and economic landscape at the time, it was feasible, arguably successful and represented an important step towards achieving universal health coverage. Following a landslide victory in the 2018 presidential elections, López Obrador – the main opponent of Seguro Popular since its origin – eliminated the programme shortly after taking office. This elimination, however, does not seem to have been due to the technical analysis of any of the Seguro Popular results, but to a fundamentally political decision consistent with a movement towards the ‘recentralisation of national life’ and with a generalised sense of urgency to eliminate and replace symbolic institutions from the past.

5.

THE RECENTRALISATION OF THE HEALTH SYSTEM

In December 2018, López Obrador became the President of Mexico and announced a shift towards the recentralisation of the public purchasing and delivery of health services to the population not covered by the Mexican social security system. The recentralisation argument was based on the premise that it would enable a more efficient procurement of supplies and simplify processes. To achieve this, in 2019 and 2020, a series of administrative and legal changes, including a constitutional reform, was introduced to substitute Seguro Popular with a new federal agency, the National Institute of Health for Wellbeing (INSABI). This new institution, which began operations on 1 January 2020, was entrusted and empowered to negotiate new collaboration agreements with each of the 32 state governments. The reform, however, was not enforced nationwide as states had to agree to roll back Seguro Popular and affiliate to INSABI. Unaffiliated states would not transfer their decentralised responsibilities to the federal government at the potential cost of not increasing their resources and being deprioritised from the federal agenda. Nevertheless, initially, nine out of the 32 states (known as the ‘Federalist Alliance’) did not affiliate to INSABI due to political differences with the federal government. It was during these uncertain times that the Covid-19 pandemic hit the country in March 2020. The delayed implementation of an incomplete recentralising reform weakened states’ capacity to respond at a crucial time. This health crisis deepened the challenges of a chronically underfunded and segmented health system, with states now heavily relying on the federal government for the response, while INSABI – the entity in charge of implementing public health actions – did not have operational guidelines until August 2020.

124  Handbook on the political economy of health systems Paradoxically, against the new system’s foundational motivation, the reconfiguration has led to centralised but less transparent purchasing mechanisms. This disruption has also been conducive to managerial and coordination frictions that have complicated the acquisition and distribution of medical supplies, raising a new and meaningful group against the new system headed by the families affected by the medicine shortages (Agren, 2021; Das, 2021). The lack of strong leadership at the federal level combined with unclear and incomplete operational guidelines for INSABI6 led to a health governance crisis. This crisis, coupled with the deficiencies of the new system highlighted by the Covid-19 pandemic, precipitated the most recent change in the federal health care strategy. In March 2022, the President of Mexico announced that IMSS-Bienestar (the programme founded in 1979 and renamed each presidential term) would be in charge of the coordination and implementation of the current administration’s health universalisation strategy. This new strategy faces two major challenges. First, the need to scale up IMSS-Bienestar. As of March 2022 only 11 million people were affiliated to the programme and it did not operate in 13 out of 32 states (IMSS-Bienestar, 2022). Moreover, even in the states where it currently operates its coverage is very limited. Second, this new strategy requires new coordination agreements between the federal government and each of the 32 state governments.

6.

SUMMARY AND CONCLUDING REMARKS

The review of the Mexican efforts to universalise health protection sheds light on how different sources of influence, from the long-term trends to specific conjunctures, have shaped the Mexican health system. In this way, the system has been defined by the interaction amongst broader forces such as decentralisation processes, financial and political constraints, economic and democratic developments, cross-cutting policy decisions (commonly driven by ideology and electoral strategy) and the tensions within the public administration. Figure 8.2 illustrates the variables and interactions described in previous sections that help explain the continued fragmentation that characterises the Mexican health system from its origin as well as the different institutional configurations that the system has had over time. Economic conditions such as level of formality, income distribution, economic performance and budget availability have influenced the building efforts of the universal health protection in Mexico. The diagram also shows a relationship in the other way: different health insurance systems are conducive to divergent labour market conditions and varying incentives for economic growth. This tension has fuelled enriching debates between public health and economic approaches commonly represented by the MoH and the MoF or the social security institutions. Additionally, it is crucial to recognise the significant weight of political constraints and that the labour unions’ interests as well as the corporativist arrangements have disseminated across the political parties’ system, the government and the health (insurance) institutions. Moreover, the ideological influence, such as a ‘Bismarckian ethos’ and the symbolic dimension of breaking with the institutions representative of the past, must be considered to understand the ‘foundational’ and ‘pull-down’ milestones in the Mexican health system history. The timeline presented in Figure 8.3 helps identify the defining moments described in this chapter as well as the potential sources of influence that contribute to explaining how they developed. It should be noted that the structural architecture of the current Mexican health system, characterised by its profound fragmentation, has not changed since its origin in 1943.

Political economy of health care insurance expansion in Mexico  125

Figure 8.2

Sources of influence and shaping forces of the Mexican health system’s configuration

In this year, IMSS and the antecedent of the current MoH were founded in the context of tight fiscal constraints and a generalised trend towards social security schemes. However, there have been incremental efforts to offer protection to the population without social security. The first mechanism to implement this objective relied on the same corporatist logic of gradually incorporating newly organised population groups into the same structure. This dynamic found its limitations when the assumptions of rapid and widespread urbanisation and formalisation were not fulfilled. Then, when an expected budget relaxation was allowed, increasing the public financing on social expenditure, IMSS-Coplamar emerged to address the rural and vulnerable populations. The following attempts to continue the universalisation of health protection in Mexico, such as the Coverage Expansion Programme, now hosted at the MoH, were shaped by a necessary decentralisation process driven by the economic crisis of 1994 and the increasing democratic pressures. In 2000, in the context of the first political alternation, the new government made a bold attempt to accelerate these efforts to protect the right to health by enabling additional funding from general taxes. Given the competing topics in the policy agenda and the limited legislative power in that term, the advocacy efforts to convince stakeholders about the social and political virtues of Seguro Popular deserve recognition and academic study. The continued (but not necessarily inertial) pursuit of universal health coverage, building on previous efforts, was abruptly stopped by the new administration that took office in 2018. The sources of this change are well founded on the social expectation and clear electoral mandate of eradicating corruption from the public administration in Mexico and addressing the deep social inequity. A proportion of this animosity was fuelled by visible cases of misuse

126  Handbook on the political economy of health systems

Figure 8.3

Selected contextual and Mexican Health System milestones

of resources in the health system, particularly at the state level, where accountability mechanisms failed to prevent corruption. However, until now the replacement institutions have failed to offer higher levels of responsiveness, accountability, equity or integration within the system. The government’s urgency to eradicate the institutional and technical infrastructure with a relevant political symbolism of the past may explain this regressive result and increased uncertainty. The anti-corruption narrative has also fuelled a new re-centralisation trend across the Mexican State in favour of the presidential agenda. Nevertheless, it is striking that this reconcentration of power, broad social support and relative economic stability at the beginning of the term was not used to propose an ambitious reform that finally transformed the architecture of the health system. The agenda’s relevance that public health acquired globally during the Covid-19 pandemic was also insufficient to trigger a decisive comprehensive redesign to address the systems’ fragmentation and synergise economic development incentives with social protection. If these unique favourable conditions do not eventually translate into a profound transformation of the health system, this period will be remembered as the ‘historic missed opportunity’.

Political economy of health care insurance expansion in Mexico  127

ACKNOWLEDGEMENTS We are grateful to Ana Bernal Stuart, Antonio Chemor Ruiz, Eduardo González-Pier, Santiago Levy and José Narro Robles for their helpful comments. Responsibility for errors and omissions remains with the authors.

NOTES 1. The public denomination of the subsequent versions of this programme has changed according to the name of the main social programme operating during each presidential term; i.e., ‘IMSS – Solidaridad’, ‘IMSS-Oportunidades’ and ‘IMSS-Prospera’, and today ‘IMSS-Bienestar’. 2. Catastrophic health expenditures are the ones that represent more than 30 per cent of household’s income net of food spending. 3. Adding up to more than 120 million people when the total population in Mexico was at that time 112.3 million people. This is clear evidence of multiple coverage. According to FUNSALUD (2012), more than 14 per cent of Seguro Popular beneficiaries were also covered by social security institutions. 4. The four most important political parties in Mexico during the period under analysis are PRI (centre), PAN (right), PRD (left) and most recently MORENA (left). 5. From 2004 to 2009 the funding of Seguro Popular was done on a family basis. However, from 2010, the federal and the state government’s contributions were computed on an individual basis, while the beneficiary contributions were paid by families. 6. For example, FASSA resources – key to the financing structure of the second wave of decentralisation – are still decentralised and directly transferred to state governments.

REFERENCES Agren, D. 2021. Lack of medicines in Mexico. Lancet, 398, 289–90. Autrique-Echeveste, D. 2012. Análisis de los efectos de la descentralización del Sistema de Salud en México, para la atención de la población sin seguridad social, sobre la distribución interestatal de variables selectas de oferta de servicios de salud. Dissertation, Licenciatura en Economía, ITAM, DF, Mexico. Bosch, M., & Campos-Vazquez, R.M. 2014. The Trade-Offs of Welfare Policies in Labor Markets with Informal Jobs: The Case of the ‘Seguro Popular’ Program in Mexico. American Economic Journal: Economic Policy, 6, 71–99. Celhay, P., Martinez, S., Muñoz, M., Perez, M., & Perez-Cuevas, R. 2019. Long-Term Effects of Public Health Insurance on the Health of Children in Mexico: A Retrospective Study. Lancet Global Health, 7, e1448–e1457. Chemor Ruiz, A., Ratsch, A.E.O., & Alamilla Martínez, G.A. 2018. Mexico’s Seguro Popular: Achievements and Challenges. Health Systems & Reform, 4, 194–202. Colchero, M.A., Gómez, R., & Bautista-Arredondo, S. 2022. A Systematic Review of the Literature on the Impact of the Seguro Popular. Health Research Policy and Systems, 20, 42. Conti, G., & Ginja, R. 2020. Who Benefits From Free Health Insurance: Evidence from Mexico. Journal of Human Resources. doi: 10.3368/jhr.58.3.1117-9157R2. Costa-Font, J., Cowell, F.A., & Saenz de Miera, B. 2021. Measuring Pure Health Inequality and Mobility during a Health Insurance Expansion: Evidence from Mexico. Health Economics, 30, 1833–48. Das, M. 2021. Shortage of Cancer Drugs in Mexico. Lancet Oncology, 22, 1216. del Valle, A. 2021. The Effects of Public Health Insurance in Labor Markets with Informal Jobs: Evidence from Mexico. Journal of Health Economics, 77, 102454. Frenk, J., González-Pier, E., Gómez-Dantés, O., Lezana, M.A., & Knaul, F.M. 2006. Comprehensive Reform to Improve Health System Performance in Mexico. The Lancet, 368, 1524–34.

128  Handbook on the political economy of health systems Frenk, J., Sepúlveda, J., Gómez-Dantés, O., & Knaul, F. 2003. Evidence-Based Health Policy: Three Generations of Reform in Mexico. The Lancet, 362, 1667–71. FUNSALUD (Fundación Mexicana para la Salud). 2012. Universalidad de los Servicios de Salud. Una propuesta de FUNSALUD. DF: FUNSALUD. Gómez-Dantés, O., Reich, M.R., & Garrido-Latorre, F. 2015. Political Economy of Pursuing the Expansion of Social Protection in Health in Mexico. Health Systems & Reform, 1, 207–16. González Block, M., Cahuana-Hurtado, L., Ávila-Burgos, L., & Orozco, E. 2018. Decentralization of Health Policy and Services in Mexico. In: Marchildon, G.P., & Bossert, T. (eds) Federalism and Decentralization in Health Care. Toronto: University of Toronto Press, 174–99. González Block, M., Reyes Morales, H., Cahuana-Hurtado, L., Balandrán, A., Méndez, E., & Allin, S. 2020. Mexico: Health System Review. Copenhagen: WHO. González-Pier, E., Gutiérrez-Delgado, C., Stevens, G., Barraza-Lloréns, M., Porras-Condey, R., Carvalho, N., Loncich, K., Dias, R.H., Kulkarni, S., & Casey, A. 2006. Priority Setting for Health Interventions in Mexico’s System of Social Protection in Health. The Lancet, 368, 1608–18. IMSS-Bienestar. 2022. IMSS-Bienestar. Cobertura. www​.imss​.gob​.mx/​imss​-bienestar. INEGI. 2021. Población con afiliación a servicios de salud por entidad federativa según institución, 2020. https://​inegi​.org​.mx/​app/​tabulados/​interactivos/​?pxq​=​Derechohabiencia​_Derechohabiencia​_02​ _822ebcc5​-ef41​-40c1​-9901​-22e397025c64. Knaul, F.M., Arreola-Ornelas, H., Wong, R., Lugo-Palacios, D.G., & Méndez-Carniado, O. 2018. Efecto del Seguro Popular de Salud sobre los gastos catastróficos y empobrecedores en México, 2004–2012. Salud Pública de México, 60, 130–40. Knaul, F.M., González-Pier, E., Gómez-Dantés, O., García-Junco, D., Arreola-Ornelas, H., Barraza-Lloréns, M., Sandoval, R., Caballero, F., Hernández-Avila, M., Juan, M., Kershenobich, D., Nigenda, G., Ruelas, E., Sepúlveda, J., Tapia, R., Soberón, G., Chertorivski, S., & Frenk, J. 2012. The Quest for Universal Health Coverage: Achieving Social Protection for All in Mexico. The Lancet, 380, 1259–79. Lakin, J.M. 2010. The End of Insurance? Mexico’s Seguro Popular, 2001–2007. Journal of Health Politics, Policy and Law, 35, 313–52. Levy, S. 2008. Good Intentions, Bad Outcomes: Social Policy, Informality, and Economic Growth in Mexico. Washington, DC: Brookings Institution. Ley General de Salud. 2021. Nueva Ley publicada en el Diario Oficial de la Federacion (DOF) el 7 de febrero de 1984. Última reforma publicada DOF 01-06-2021. Ciudad de México: Ley General de Salud. Merino, G. 2003. Descentralización del sistema de salud en el contexto del federalismo. Caleidoscopio de la salud: de la investigación a las políticas y de las políticas a la acción. DF: FUNSALUD. Nigenda, G., Wirtz, V.J., González-Robledo, L.M., & Reich, M.R. 2015. Evaluating the Implementation of Mexico’s Health Reform: The Case of Seguro Popular. Health Systems & Reform, 1, 217–28. Nikoloski, Z., & Mossialos, E. 2018. Membership in Seguro Popular in Mexico Linked to a Small Reduction in Catastrophic Health Expenditure. Health Affairs, 37, 1169–77. Ochoa-Moreno, I. 2020. Assessing Universal Coverage: Mexico’s Seguro Popular. PhD, University of York. OECD. 2016. OECD Reviews of Health Systems: Mexico 2016. Paris: OECD Publishing. Ortiz, M. 2006. El Seguro Popular: Una crónica de la democracia mexicana. DF: FCE. Pueblita, J.C.R. 2013. Screening Seguro Popular: The Political Economy of Universal Health Coverage in Mexico. CID Research Fellow and Graduate Student Working Paper Series 2013.61. Reich, M.R. 2020. Restructuring Health Reform, Mexican Style. Health Systems & Reform, 6, e1763114. Reyes Morales, H., Dreser-Mansilla, A., Arredondo-López, A., Bautista-Arredondo, S., & Ávila-Burgos, L. 2020. Análisis y reflexiones sobre la iniciativa de reforma a la Ley General de Salud de México 2019. Salud Pública de México, 61, 685–91. Secretaría de Salud. 2006. Sistema de Protección Social en Salud. Elementos conceptuales, financieros y operativos. In: FCE (ed.) Biblioteca de la Salud. DF: Secretaría de Salud. Soberón-Acevedo, G., & Martínez-Narváez, G. 1996. La descentralización de los servicios de salud en México en la década de los ochenta. Salud Pública de México, 38, 371–8. Sparkes, S.P., Bump, J.B., Özçelik, E.A., Kutzin, J., & Reich, M.R. 2019. Political Economy Analysis for Health Financing Reform. Health Systems & Reform, 5, 183–94.

Political economy of health care insurance expansion in Mexico  129 The Lancet. 2012. Mexico: Celebrating Universal Health Coverage. The Lancet, 380, 622. doi: 10.1016/ S0140-6736(12)61342-7. Vos, T., Lim, S.S., Abbafati, C., Abbas, K.M., Abbasi, M., Abbasifard, M., Abbasi-Kangevari, M., et al. 2020. Global Burden of 369 Diseases and Injuries in 204 Countries and Territories, 1990–2019: A Systematic Analysis for the Global Burden of Disease Study 2019. The Lancet, 396, 1204–22.

9. Weird health care for WEIRD societies? Hartmut Kliemt

1.

INTRODUCTION AND OVERVIEW It is therefore, on opinion only that government is founded; and this maxim extends to the most despotic and most military governments, as well as to the most free and most popular. (Hume, 1985, iv)

The subsequent article conceives of an extended sphere in which ‘stability of possession, its transference by consent and the performance of promises’ (Hume’s ‘natural law(s)’; Hume 1739, Bk III, part ii, sec xi) enables ‘private’ decisions. That their concrete specification of ‘natural law’ politically confers extensive powers of decision-making on individual citizens (‘rights of contracting’) is an essential characteristic and one of the ‘secrets’ of the success of so-called Great Societies of the classical liberal kind. Though the underlying overall political equilibrium of today’s Western, Educated, Industrialized, Rich and Democratic (WEIRD) societies (Henrich 2020) has always been under siege and is threatened at present by rather dramatic changes in communication and opinion formation, WEIRD societies still manage to separate an extended private law sphere from a public policy sphere.1 However, more recently a new threat seems to have arisen from the ascent of modern bio-medical technologies: Public responsibility for health care in combination with the deeply entrenched opinion that a legitimate WEIRD state must protect all citizens equally against ‘imminent grave risks for life and limb’ may further the growth of health-related welfare state regulation and redistribution to an extent that puts the ‘liberal’ political equilibrium at risk. If in a society, due to a lack of advanced medical care, chronically ill patients cannot survive for extended periods of time, then this implies that the members of such a society will suffer less from chronic afflictions. Therefore, citizens of societies with primitive health care may on average enjoy better rather than worse health during their relatively shorter lives than citizens of societies with a more developed health care system. This is as such not an argument against expanding health care. But if the fundamental coercive power of the state is invoked to prioritize public health care guarantees over other public and private concerns, a positive feedback loop that leads to an unsustainable growth of public health care can emerge. To be sure, there are cases in which technological progress in medicine may lead to cost reductions and/or, by restoring the ability to participate productively in the exchange nexus of the private law society, will ‘pay for itself.’2 Yet, quite independently of the question of how purely economic costs and benefits of medical progress affect public finances, there is the fundamental regulatory risk that the expansion of public health responsibilities will not be in line with the fundamental social organizational principle of subsidiarity according to which [t]he legitimate object of government is to do for a community of people whatever they need to have done but cannot do at all, or cannot do so well for themselves in their separate and individual capacities. In all that people can do individually as well for themselves, government ought not to interfere. (Abraham Lincoln, Speech, 1854)3

130

Weird health care for WEIRD societies?  131 So much for a first introductory assault on some of the issues of the constitutional political economics of public health care guarantees in WEIRD societies that will be addressed subsequently. Section 2 will focus on the so-called ‘rule of rescue’ which demands that a legitimate WEIRD state guarantees feasible rescue of its citizens when they are exposed to imminent survival-relevant threats.4 In Section 3, central aspects of medical progress in treating end-stage renal disease (ESRD) will be used to illustrate how central imposition of specific ethical views may systematically suppress the beneficial effects of self-organization and exchange in health care provision. Section 4 concludes with some tentative remarks on the essential importance of addressing issues of public health care in terms of procedural demands for guarantees as opposed to consequentialist demands for collective benefits.

2.

SISYPHUS SYNDROME, RULE OF RESCUE AND INSURANCE

Medical progress expresses itself not only in successfully restoring health by medical means but also, and increasingly so, by keeping chronically ill patients alive. Though in terms of extended length and improved quality of life medical progress is altogether desirable, it will with increasing frequency not lead to a restoration of self-sustaining health but to a chronic condition that requires continuous further intervention, often increasingly so. The treatment of ESRD can be used to illustrate concretely what is at stake here. 2.1

Public Health Care Provision and the ‘Protective’ State

Though quality and expected time of life with one healthy kidney is comparable to living with two kidneys, the expected lifespan without kidneys is very brief unless the patient gets access to a renal replacement therapy (RRT; i.e., dialysis or transplantation). Without transplantation of a kidney, only dialysis can prevent imminent death in the case of end-stage renal failure (ESRF), and treatment must continue if the patient is not to die at short notice. In that regard health care workers are in the situation of Sisyphus, who, after rolling his stone uphill, has not done his job but must roll it up another time: In the case of dialysis, after restoring metabolic equilibrium sufficiently by administering a round of treatment, deterioration towards a life-threatening metabolic disequilibrium state requires further dialysis or transplantation of a deceased- or living-donor kidney.5 In practically all WEIRD societies publicly financed and regulated guarantees of providing life-saving dialysis for all citizens are implemented. The public at large regards using the state’s fundamental coercive powers to tax and to regulate as legitimate when it comes to guaranteeing inclusive equal access to dialysis in a sufficiently ‘affluent’ affluent society. Though there may be some means-testing, ultimately rescue guarantees are publicly provided and conditional only on citizenship as such (sometimes presence in a state’s territory is deemed a sufficient condition). In consequence, access to dialysis is (ultimately) guaranteed independently of willingness and ability to pay for service.6 As demanded by the citizenry at large, the political aim is to guarantee that dialysis services can be offered to every patient whose life can be rescued by dialysis (should the patient choose to make use of the services). Public triage decisions, which discriminate between who is to be

132  Handbook on the political economy of health systems treated and who not, need not be made if (and only if) sufficient dialysis capacities are on offer (ultimately) unconditionally: 1. In WEIRD societies even most foes of the expansion of the welfare state endorse guaranteeing equal access to dialysis for all citizens by means of coercive regulation and taxation and thereby accept basic welfare state activities. 2. Once a WEIRD state is held publicly responsible by its citizenry for preventing imminent death through dialysis services it can hardly terminate the treatment of a patient, nor can it – as long as the provision is financially and technically feasible – stop offering the same services to future patients who do not yet suffer from ESRD. As far as the legitimacy of a coercively financed state is concerned, it deserves to be emphasized that an adherent of ‘rule of law’ guarantees implicitly endorses the legitimacy of a minimum welfare state. After all, regulatory and redistributive welfare state elements are present already in guaranteeing the services of a so-called ‘protective’ state of the nightwatchman type:7 It is an egalitarian minimum welfare state in that it provides equal service guarantees and coercive regulations of rule of law for all citizens on the basis of coercive taxes.8 Since fundamental health care guarantees can obviously be provided within the rule of law framework of the state, it is merely open whether relying on the fundamental coercive power of the state beyond providing rule of law can be justified without violating the principle of subsidiarity. But no principal borderline is transgressed. That in the emergent notional continuum of welfare states citizens can legitimately demand public help beyond the enforcement of a minimal rule of law framework is most plausible with respect to guaranteeing rescue in cases of imminent serious threats to life and limb. This view is indeed deeply entrenched in basic opinions concerning the legitimacy of the political order, opinions that seem to be very widely shared among citizens of WEIRD societies. It is expressed in voicing their opinions and in casting their votes in expressive voting behavior (Brennan and Lomasky 1993). 2.2

The Rule of Rescue and the Veil of Insignificance

According to the ‘rule of rescue,’ the richer a society, the more imminent and serious the threat to life and limb, the more identifiable the endangered persons are,9 the more urgent it is according to prevalent opinions of legitimacy of the state to discharge the public responsibility that endangered persons be rescued by the state (respectively its agents). Behind the veil of individual insignificance of their individual vote (Kliemt 1986), voters will tend to express the demand that state agencies of WEIRD societies must guarantee to the feasible extent that concrete persons be rescued from imminent dangers to their corporeal integrity independently of their ability and willingness to pay, and if necessary at the taxpayers’ expense (Brennan and Lomasky 1985, 1989). Providing such guarantees does not imply that expenses must be paid without means-testing or that no co-payments must be required. Central is the assumption of a residual (subsidiary) responsibility of the public at large to see to it that fundamental rescue services to the feasible extent prevent people from ‘dying on the streets.’ States of WEIRD societies have used their power to regulate and to tax for putting public infrastructure in place for such purposes: In the case of a flood people will be rescued from drowning, in the case of an accident people will be transported to the emergency room without anybody first checking their ability to pay (respectively their insurance status); a person who suffers from a life-threatening heart condition that can be controlled by a surgical intervention but who

Weird health care for WEIRD societies?  133 cannot pay for that intervention is granted access to it from the public purse on occasion of a cardiac arrest, etc.10 Public opinion and the voters of WEIRD societies hold the state responsible for seeing to it that it is guaranteed that the preceding demands are met. To discharge the responsibility that is assigned to it by the citizenry, the state should use its fundamental coercive power to raise revenue, to buy or produce services, to impose regulations that hold private entities responsible for providing the services – e.g., it can oblige emergency rooms to admit patients in cases of imminent serious risks to life and/or limb – or both.11 The preceding applies to all basic state services occasioned by imminent and urgent life-threatening emergencies, not merely to those arising in connection with health care. However, there are relevant distinctions that should be made between different classes of emergency service guarantees. First, health care guarantees are special in that they are much more expensive than protections against freezing to death, drowning, etc. Second, the state in its role as (residual) guarantor of access to services may choose to charge the beneficiaries of urgent rescue operations not at all or may charge them for the expenses in full after rescue, or choose some reimbursement between these extremes. As argued next, in the case of health guarantees an essential difference between non-financial and financial guarantees may be seen to arise once we take into account different reimbursement schemes. After rescue the urgency of services is over and charging for services provided ex post is a viable option. 2.3

Guarantees of Rescue and Paying for It

A person who has been rescued ‘on credit’ may survive but be financially ruined afterwards. The latter will be the case if the citizen had not provided for this contingency by accumulating sufficient wealth or by buying sufficient insurance coverage for the financial risks of illness on a private market.12 Such possibilities are often neglected in the public discussion of public health care guarantees. But there is a crucial difference between the health aspects and the financial aspects of rescue guarantees. In the narrow sense the rule of rescue demands of the state only to prevent imminent loss of life or limb. Financial obligations that may arise in the wake of state interventions are not necessarily covered.13 It is wrong to claim that guaranteed rescue from serious threats to life and limb can be had only in a system that also covers the financial risks related to such threats. The narrow rule of rescue focuses only on the health dimension, whereas a more comprehensive rule of rescue can also cover financial risks. For instance, it is well conceivable that a person who is rescued at high cost and has not insured herself against the financial consequences is afterwards alive but must – like other persons who went broke – live on financial aid and a minimum guaranteed income. With respect to subsidiarity, it is quite plausible that the narrow guarantee of rescue from imminent serious threats to corporeal integrity can in a broad class of cases be provided only centrally by public action. In view of transaction costs of private organizations, effective and efficient provision may require coercive state action to maintain certain rescue capacities ‘on stand by’ so that they can be discharged instantaneously in emergencies. It is much less plausible that the latter type of argument equally applies for protections against financial risks. The obvious economic advantage of restricting guarantees to health itself is that strong incentives arise for better-off individuals to insure privately against financial risks of illness in

134  Handbook on the political economy of health systems ways that bring the efficiency gains of insurance markets into play. Nevertheless, supporters of the comprehensive rule of rescue prevail in public opinion. They demand the elimination of inequalities in access to health care altogether. But the urgency argument that supports centralized political action with respect to imminent damages to health does not extend to the ability to ‘pay for service.’14 2.4

Rations of Equal Access to State-of-the-Art Health Care

Covering the financial risks of guaranteed access to the same state-of-the-art standard of medical care to all citizens is widely demanded for reasons of extended egalitarian ideals of ‘social justice.’ Such egalitarian opinions add political traction to the expansionist tendencies of public health care in democracies. In the eyes of many citizens of WEIRD societies, the obligation to pay for received services is equitable only in the case of well-to-do recipients.15 Health ethicists exacerbate the expansionist tendencies arising from such widely shared opinions by framing the failure to guarantee equal access to any form of health care ‘expected to be of net benefit’ to patients as the ‘withholding of care.’16 By this apparently innocuous conceptual move the focus is not on guaranteeing a standard of publicly financed care that is expected to be of net benefit for patients – which clearly is an act of giving – it is directed entirely on what is not included in the care guarantees. Even if the guaranteed standard is rather high, whenever it does not guarantee everything expected to be of net benefit withholding (additional) care takes place. The omission of providing something that has not been promised as withholding something latently suggests that there was in the first place a legitimate claim to receive what goes beyond the guaranteed rations.17 Yet, “the reason for wishing that a certain right were established is not that right – want is not supply – hunger is not bread” (Bentham 1843, pp. 501–2). To remove the unwarranted stigma of illegitimately withholding something, ‘rationing’ must be – more reasonably – understood as an active provision of limited quantities of care expected to be of net benefit to patients. Pre-specified quantities or rations of care – of a certain quality – are ‘given’ at the moment of service below market clearing prices and this is exactly what is demanded by beneficiaries of, say, health maintenance organization (HMO) services.18 Rational persons who are endowed with sufficient funds will be willing to pay risk-equivalent premiums for contracts that guarantee access to standardized rations of care on private insurance markets.19 In sum, properly understood private health insurance – offered at risk-equivalent premiums – is a scheme of guaranteeing access to rations of treatment to be provided if the diagnosis of a diagnosis–treatment pair is fulfilled. Providers are paid for guaranteeing access to rations of care (below market clearing prices when the acute demand or need arises) and certainly not for withholding them. The central implication of such a rationing as limited giving as opposed to a rationing-as-withholding approach is psychological. It becomes possible to legitimately set limits to health care guarantees.20 Let us turn to an extended concrete example illustrating the preceding.

Weird health care for WEIRD societies?  135

3.

PRIVATE AND PUBLIC RATIONING: THE CASE OF ESRD

If state-of-the-art medical services would have to be guaranteed to the extent to which they are expected to be of net benefit to patients, then the whole gross domestic product (GDP) of a developed nation would potentially have to be spent on medical services with some positive marginal benefit for the patient (think of rescue helicopters narrowly spaced, specialized clinics for burn victims, additional fMRI scans). Unlimited guarantees of care cannot credibly be given, not even by the state. Limited guarantees of access to rations of care can be credibly given also by private institutions. The state may be helpful in enhancing credibility by regulatory enforcement services and via subsidies. So let us see how this spells out in the case of ESRD treatment. 3.1

Guaranteed Treatment of ESRD, Private and Public

In a world without publicly guaranteed dialysis services, people would be willing to pay for insurance covering treatment of ESRD if offered by an HMO. In the moment of need of dialysis, a person who had bought claims to get access to such services on a private insurance market has a claim to receive services of a defined quality and quantity without additional payments. The buyer of an HMO contract is not only ensuring financial risks. The buyer is also assured that at the point in time when the services are needed, they will be provided without further payment according to contractually pre-specified criteria describing the medical condition (diagnosis). That is, at the time of need it is, in the case of an HMO contract, not a matter of paying for services, of negotiation on the spot, etc., but of ‘care’ provision according to contractually pre-specified, ‘standardized’ terms or claims. Behind the veil of uncertainty about whether she will suffer from certain afflictions, uncertain about future market prices and her own future ability to pay, the buyer of a ‘plan’ offered by an HMO acquires a rather comprehensive assurance that predefined rations of care (e.g., treatments according to standardized diagnosis treatment pairs for ESRD) will be provided according to her needs in the moment of need rather than ability and willingness to pay on that occasion.21 Within the context of the private law society the promise of an HMO exhausts itself in an obligation to deliver limited rations:22 Whenever the contractually defined condition that captures the situation of ‘medical need’ is fulfilled, an enforceable claim to a specified treatment effort without further negotiation exists.23 Guaranteed rations are limited from below – treatment corresponding to a diagnosis in the contractually specified diagnosis–treatment pair must be provided. This makes the contract potentially attractive for the patient. Guarantees of rations are limited from above, too – additional care, even if it is expected to be of benefit for the patient, need not be provided. Only this way of limiting the promise from above makes it potentially attractive for HMOs to offer such a contract on a market. As the preceding shows, in developed legal orders there can and will in principle be markets on which competing institutions offer contracts (subscriptions) that guarantee participation in ongoing rationing schemes. These ‘plans’ integrate risk spreading (financial insurance) with care guarantees (assurance of care). They can cover even such rather extreme conditions like ESRD on a regular insurance market for those willing and able to pay for them. Therefore, the aim of guaranteeing that all people have in one way or another access to standardized health

136  Handbook on the political economy of health systems care could be met by private market supply provided that the rations of care are sufficiently standardized and subsidized (possibly by a health demogrant). At least if the concept of a market would be used as widely as, for instance, in Roth (2016) the state could – in line with Lincoln’s principle of subsidiarity – show restraint in using its fundamental coercive power to regulate and to tax in creating an appropriate market. For example, it would be possible to design a public auction platform on which standardized guarantees of care provision would be auctioned off to the lowest bidder. More concretely, vectors containing all available health-relevant predictive information about patients could be auctioned to those institutions that would be willing to guarantee the politically required level of care given the vector information. All sorts of bells and whistles could conceivably be added, including the possibility to switch between plans of providers for a private co-payment so as to assure some competition among providers, for example.24 For such a scheme to work it would be crucial, however, that public politics would be willing to specify explicitly what is not part of the guarantees for which the community adopts responsibility. Though this seems to require merely openly acknowledging what is done in practice anyway, this seems to be politically infeasible in many WEIRD societies.25 To the extent that it is infeasible to explicitly state which kinds of services will not be provided, it will be impossible to utilize the potential of the private law, private contract and exchange society to render the health care services that its citizenry demands more efficiently and more equitably than under a scheme of implicit exclusion of services. In the case of ESRD the states of practically all WEIRD societies have adopted the public responsibility of guaranteeing access to dialysis services of a defined quantity and quality to all citizens if certain criteria of urgency and expected benefit are met. But state involvement changes in a fundamental way once we turn to transplantation of human kidneys as the presently only alternative RRT. 3.2

For and Against Exchange in Organ Allocation

While at the American Enterprise Institute, Leon Kass – who later became the chief bioethical advisor to President George W. Bush – remarked that human organs should not be bought and sold like Wurlitzers (musical instruments).26 In his What Money Can’t Buy (2012), Michael Sandel, certainly not a friend of free enterprise, also argued that human organs should not be traded for money. Despite their different political affiliations, Kass and Sandel both appeal to a widely shared resentment against money to discredit organ exchange. But this appeal is superficial. The deeper problem concerns the respect for the autonomy of individuals and their voluntary exchanges (whether involving money or not). As the subsequent discussion concretely illustrates, basic bioethical issues are predominantly not regarded as matters of the private law society – and with rather negative effects on such values as respect for inter-individual value pluralism and individual autonomy. 3.2.1 Towards institutionalizing kidney exchange Initially, efforts to realize human organ transplantations failed because the causes of the rejection of transplanted organs were not well understood (Hamilton, Starzl, and Barker 2012).27 In the case of kidneys, after the quality of the match between human lycocyte antigens (HLA) was identified as a crucial predictor of success and the immunosuppressive drug

Weird health care for WEIRD societies?  137 cyclosporin had been introduced in the 1980s, expected survival times after transplantation dramatically increased. Transplantation of kidneys of deceased donors established itself as a regular medical procedure. University hospitals began to systematically ‘harvest’ organs of potential donors whose death had been diagnosed.28 While bodily functions of the deceased were upheld, close relations were asked for their consent to explant organs. After consent had been obtained organs were explanted and then implanted.29 Two medical and one broadly legal aspect of the process deserve particular emphasis: First, kidneys of deceased donors become available stochastically. Second, once available, kidneys of deceased donors have to be transplanted at short notice (they cannot be preserved for more than 48 hours).30 Third, after diagnosing death, somebody needs to be entitled to dispose of the organs. Initially, close relations of the deceased were treated as entitled to transfer the organs to the agents of the (nascent) transplant centers whose agents in turn – in the act of implantation – would transfer them to the recipient.31 Yet, in view of the basic norm that the autonomy of patients is to be respected as a supreme value in legitimate health care provision, participants in the process of realizing organ donations of deceased donors became increasingly concerned about how the will of potential donors could be taken into account. Typically, there was no living will specifying how to dispose of the organs post-mortem. Ways to assure truthful representation of donor intent that did not exist in writing were initially unavailable.32 As an ‘approximate’ solution to the legal problem the rights of disposal were treated as restricted to endorsing donation or to veto it, and the close relations of the deceased were treated as if entitled to exercise that restricted right. After campaigns in favor of organ donation, people commenced to state their will concerning the use of their organs in the case of their own death.33 But due to this it became rather implausible that they were not entitled also to express who should receive their organs. Nevertheless, almost nowhere may willing donors condition their donation in a reciprocal manner. They may not state that those potential recipients who had been willing donors before the time that they needed an organ themselves should have priority in access to the organs of the donor. Had this rather natural extension of regular rights of bequest been accepted, the mechanisms of the private law society could and would in all likelihood have run their course and a mutuality of donors willing to donate for others in exchange for their willingness to donate would have emerged.34 On the path not taken, large HMOs and other health care providers could have made such reciprocal exchanges of commitments part of their health plan.35 Even public health care systems could have incorporated reciprocity as part of their rules and could have used it to prioritize access to scarce human organs in ways dependent on autonomous contribution decisions of future patients at the time when they did not know of their own later ESRD.36 In a WEIRD world in which respect for the autonomy of the individual (patient) is embodied in the basic protective rules of the legal order, such a course of action should have seemed compelling. Yet, the ‘stability of possession, its transference by consent and the performance of promises’ (Hume 1739, Bk III, xii) otherwise characterizing the private law subsystem of society was not extended to human organs (not even in a constrained way that would rule out monetary payments). The preceding observation does not merely concern a peculiarity of organ transplantation. It seems rather a symptom of a general syndrome that threatens the equilibrium of WEIRD rule of law societies in general: the familiar conflict between empowering individuals to

138  Handbook on the political economy of health systems self-organize voluntarily in autonomous contractual agreements – the private exchange/private law society – and the desire that certain ‘ethical’ principles be imposed on all individuals independently of whether or not they voluntarily agree to them.37 The general support for the collective responsibility to implement institutions expressive of the rule of rescue is expressive of anti-individualist tendencies in the case of ESRD as well: 1. The public at large thinks of human organs entirely as a common pool resource that should be strictly used according to the criteria of urgency and greatest expected future medical benefit of needy patients and therefore independently of previous behavior of the recipient (a variant of the rule of rescue). 2. The public at large rejects full rights of disposal – including the right to sell ‘futures’ concerning one’s own organs for money. Deeming monetary rewards for organ donation repugnant, it feels obliged to close off what it regards as a slippery slope towards ordinary markets.38 The typical Smithian economist and more generally any adherent of the ideals of subsidiarity may lament the prevalence of (1) and (2). Yet the underlying opinions of legitimacy impose constraints on health policies in that they support erecting coercive obstacles to organizing health care by exchanges to which individuals otherwise would voluntarily agree in pursuit of their own plural values. What might have been possible had the forces of free inter-personal exchange run their course can be illustrated by the spontaneous evolution of inter-center kidney exchange, but it also illustrates the arbitrary character of prevailing anti-exchange sentiments in the case of individuals. 3.2.2 The Eurotransplant kidney exchange practice In the early days transplant centers saw themselves primarily as agents of the interests of their clients; i.e., the patients on their waiting list. To serve them better they sought to acquire full rights of disposal over organs, which they used to establish a rather elaborate system of organ exchange between transplant centers.39 Given the physiological and other constraints of kidney transplantation, finding good – in particular good HLA – matches between donor and recipient tissue on the waiting lists of single transplant centers was often exceedingly difficult. Forming networks of organ exchange among transplant centers was the obvious way to having access to a larger pool of organs and waiting patients, allowing for better matches.40 The mutual advantages that could be realized by exchange led the European transplant centers to form a club, the Eurotransplant network (comprising originally Austria, Benelux, Germany and the Netherlands – then Slovenia joined, and also Hungary and Croatia).41 Organs were treated as a common pool resource and successfully managed as such (Ostrom 1990). Typically, kidneys were passed on in exchange for the promise of receiving a kidney back from the receiving or another center at a later point in time.42 For comparing a partly private, exchange-based and a fully public organization of health care it deserves to be emphasized that the exchange of kidneys of deceased donors could emerge only because the de facto title of centers was acknowledged in the countries in which the participating centers were located. That the interaction was ongoing and information concerning behavior of the centers was systematically collected was sufficient to render the realization of the positive network effects by private initiative sustainable. As if led by an invisible hand the exchange activities of centers served the ‘common good’ more broadly understood. So why not let individuals do the same?

Weird health care for WEIRD societies?  139 3.2.3 Incoherent anti-organ-exchange sentiments Eurotransplant emerged despite the fact that public opinion was fundamentally hostile to inter-individual exchange of human organs (respectively individual exchanges of promises and ownership concerning human organs). According to widely expressed opinion, cadaveric organs have to be donated without a quid pro quo as an unselfish gift. This closes off directed donations for a quid pro quo. Unconditional non-directed donation is required in the case of deceased donor donation, while in some countries, like Germany, undirected donation is nevertheless forbidden in the case of living donations. In Germany, it is demanded that living donations be directed by individual donors to specific close relations – even so-called cross-over living donations of kidneys require that all participants build up some kind of personal relationship, not only with the recipient for whom they originally intended to donate. Since the last case is perhaps the most mysterious one, a more specific look may be warranted: Imagine two pairs of spouses (A1, A2), and (B1, B2). A1 is a willing donor for A2, who suffers from ESRD; B1 is a willing donor for B2, who suffers from ESRD. Due to tissue (in)compatibility A1 and B1 are not suitable donors for their respective spouses, but A1 could donate for B2, while B1 could donate for A2. The organ exchange is serving exclusively the purpose of finding better matches for organs that are altruistically donated for a close relation but cannot be transplanted because of an incompatibility between donor and recipient.43 Since this situation is exactly analogous to the one that led to organ exchange in Eurotransplant, it seems quite unintelligible unless one has taken a course in deep German philosophy. But closer to earth, one may also recall that those who have not been willing donors during their lifetime may not be excluded from equal access to organs. To demand that even those who have explicitly rejected, during their lifetime, donating their organs after death are to be treated exclusively according to need (urgency and expected benefit) amounts economically to an invitation to free-ride and ethically to endorsing a variant of ‘organ communism.’44 Using terms like ‘organ communism’ to describe present practices of organ allocation can be defended on conceptual grounds.45 Such honest description may be a first step towards imposing some conceptual order on the practical mess of organ allocation and health care in general.

4. CONCLUSIONS The yearning for living in a world in which the individual is cared for by the community (in the modern world typically by the state and its agents) and the desire for autonomous individual self-determination (in the modern world typically by market participation in the company of strangers)46 are pulling in two opposite directions. The latter expresses itself in support for Smithian exchange-based organization of health care according to the Smithian maxim Give me that which I want and you shall have this which you want.

The former mobilizes support for an organization of health care according to the Christian/ Marxian maxim From each according to ability to each according to need.

140  Handbook on the political economy of health systems The contexts in which the two maxims reign can be kept apart to some practical extent by the political separation of the private law and the public care sphere. Yet, the dynamics of medical progress and the political demand for equal participation of all citizens in it threaten that separation. Invoking the principle of subsidiarity in the abstract when fixing general rules will be helpful but will not be sufficient, because in the games that are played within the chosen rules concrete rather than statistical lives will be at risk. In concrete cases, the rule of rescue will tentatively demand to overrule any commitment to limit the provision of feasible care.47 Acknowledging all this, health care services – though to a large extent publicly guaranteed in WEIRD societies – are in principle characterized by (easy) excludability from and rivalry in consumption. They can in principle be provided by voluntary exchange on markets. Yet, even if markets may as a matter of empirical prediction on average (statistically) be expected to satisfy fundamental survival interests of individuals better than non-market alternatives, this will not satisfy the demand for guarantees. Public opinion and deeply entrenched convictions of citizens of WEIRD societies will tend to evaluate basic political institutions not (only) according to average expectations of representative individuals behind some fictitious veil or other but according to how they deal with concrete (tragic) scarcity.48 They will deem political governance structures legitimate only if they generate and sustain guarantees, claims and rights that operate in concrete situations at least almost lexicographically pre-ordered to generally expected consequences.49 The WEIRD welfare state seeks to guarantee all citizens (mostly) independently of their ability to pay in the moment of acute need equal protection against imminent infringements of what they regard as ‘rights.’ Guaranteed protection against fundamental violations of so-called ‘negative’ rights is expected to be provided by the legal staff (police and courts). Beyond this the institutionalization of guaranteed protection of citizens against imminent threats of starvation, drowning, freezing, dying from accidents and viable rescue from acute threats of failing health are also demanded (legal guarantees of access to dialysis in the case of ESRD being a typical example). Adam Smith’s observation that it is one of the great merits of exchange on markets that participants are not dependent on the benevolence of the butcher for their meat, for example, applies analogously to guarantees of care. Citizens desire to be independent from the benevolence of others when it comes to the protection of their fundamental interests, be they of the negative protective rights type or the basic share rights in rescue services. There is a demand for guarantees on which actors can individually rely (almost) no matter what. At the same time, when it comes to tragic scarcity in which not all but only some concrete members of a political unit or community can be rescued from imminent threats, the basic communist intuition that need (urgency and expected benefit) should rule the use of the fundamental coercive power of the state and its agents seems to dominate. The abstract tension between procedural demands for equal guarantees of rights and consequentialist demands for stochastically better average outcomes will not go away in any constitutional political approach, but it may be tweaked more to the individualist than the collectivist side of the path along which political institutions of WEIRD societies will evolve.50

Weird health care for WEIRD societies?  141

NOTES 1. Henrich (2020) is an interesting evidence-based ‘narrative’ of the emergence of WEIRD societies in what has been called the co-evolutionary process of ‘genes, mind and culture’ (Lumsden and Wilson 1981). The path towards creating the depoliticized subsystem is plausibly accounted for in a stylized potential explanation by North, Wallis and Weingast (2013); it may also be noteworthy that the German politician and legal theorist Franz Böhm coined the term ‘private law society’ (Böhm 1966). 2. The ascent of the transplantation of human kidneys in the treatment of ESRD is one of the examples in which medical progress brought about a superior and less costly treatment. 3. This thought also played a prominent role in later Catholic doctrine; see Pius XI (1931) and Kliemt (1997). 4. It lays out some factors that politically promote an expansion of direct public responsibility for health care provision even where adherence to the subsidiarity principle and the implied priority of an exchange-based organization would suggest otherwise. 5. Of course, vital processes are generally subject to Sisyphus’ predicament in that after eating we need to eat again, after breathing we need to take another breath, etc. The crucial distinction is that the regular means of maintaining a healthy metabolic equilibrium state are fundamentally less costly than the means that are necessary to sustain life in many cases of states of chronic illness like ESRD. 6. In the specific case at hand the regulatory and redistributive coercive power of the state may be used to impose regulations like mandatory membership in a health plan and/or to raise taxes that directly finance dialysis services. In any event institutionalized obligations and rights do not exist solely because there is a demand that they should exist. Listening to ethicists and political philosophers, one gets the impression that they are of the opinion that what they think is a demand justified by some a priori reasoning therefore becomes an institutionalized right as an empirical fact (a posteriori). 7. That the presence of distributive and regulatory dimensions of a welfare state is constitutive of statehood and government as such is acknowledged and analyzed as a breach of anarchical law in Nozick (1974). Whether or not one accepts Nozick’s original position, the equal legal services are provided employing the fundamental coercive power of the state in regulated ways and essentially independently of willingness and ability to pay; for some more extensive comments, see Brennan and Kliemt (2019). 8. Citizens of WEIRD societies who endorse the rule of law and – reasonably – believe that the rule of law in large numbers’ interaction per se requires a state can in fact accept their state as legitimate only if it is at minimum a nightwatchman state. The nightwatchman state is simply the minimal extreme point of a continuum of welfare states which reach up to the most comprehensive nanny state; on this, see Kliemt (1993, 2015). 9. The distinction between statistical and concrete lives allows for shades of grey; see Calabresi and Bobbit (1978). 10. It should be noted that so-called preventive medicine even if it were as effective as is often assumed would still have less priority as a public health responsibility since it lacks the urgency of cases prioritized according to the ‘rule of rescue.’ 11. Even if empirically or as a matter of statistical fact as many or more ‘acts of rescue’ might under certain circumstances be predicted to occur without direct government involvement, the citizens will demand that the state provides guarantees. One should be careful not to discredit such demands of equal guarantees that come along with citizenship as ‘merely symbolic.’ If the citizens do not restrict themselves to consequences but demand the implementation of a guarantee to reach them as expressive of their values, a value-neutral economist must remain silent on the legitimacy of such ends (Robbins 1935). The present chapter shares in many regards John C. Goodman’s views and excellent policy analyses but differs with respect to the symbolic aspects of guarantees and the opinions of legitimacy supporting them (Goodman 2015). 12. It is not claimed that reimbursement should take place. That would require additional arguments; in particular, if the victim did not voluntary agree to receiving the service, for example. 13. On a two-tier system that approximates the relevant distinctions, see Breyer and Kliemt (2015).

142  Handbook on the political economy of health systems 14. The assurance of guaranteed access to a minimum standard of living can be left to the regular provisions for cases of private bankruptcy, for example, while otherwise full liability reigns. 15. Though proportional or progressive tax contributions to providing the state-sponsored guarantees of care should be counted in by adherents of fairness principles then. 16. Typical of such views, Allen Buchanan states ‘Rationing – which means the withholding of care expected to be of net benefit – occurs throughout every health care system and is unavoidable’ (1997, 335–6). 17. Yet, having no handkerchief, the use of which would be of net benefit, does not give rise to a claim of getting access to it even if there is one in the possession of another person. Only if there is a prior claim is the other person who is not granting me access to the use of the item withholding it. 18. Withholding in the proper sense of that term presupposes an obligation to give, and this obligation cannot be created by simply demanding that something be given. 19. Given reasonable risk preferences the very poor and the very rich may not have rational reasons to acquire insurance at risk-equivalent premiums, but such aspects can be left aside as irrelevant for the central points about rations of care that are to be made here. 20. For early statements of an extended debate, see Callahan (1987) and Barry and Bradley (1991). 21. Though the promise of guaranteed access to service is limited, HMO insurance has the advantage of providing a hedge against price fluctuations concerning the costs of provision or the wealth of the buyer of insurance. Beyond this a long-term HMO contract may also be seen as a commitment device that actors can use as an instrument of rational self-management. 22. There may be certain bona fide negotiable clauses in health care contracts, but the implicit obligations incurred by them will be limited, including in particular progress in medical technology and the ‘needs’ arising from it. 23. The diagnosis–treatment pairs can be defined according to some conventional standard of guidelines familiar from evidence-based medicine and its mechanisms of quality control for a given state of the art. 24. In fact, rather than seeking to erect barriers for such ‘bad risks,’ providers might specialize in chronic illnesses like diabetes to realize the gains from developing a comparative advantage in treatment of this central affliction while offering the complementary services they would have to guarantee as well either themselves or through outsourcing them. 25. For some more detail, see Kliemt (2006). 26. Kass’ positions must be taken very seriously even if one might disagree with them. The opinions Kass voices are quite clearly representative of one of the two moral tribes (Greene 2013) present in American (and other WEIRD) and other societies: on the one hand, there are those who are against regulating markets in ordinary work and business life while favoring regulation of most other realms of private life, and on the other, those who oppose regulation of private life while endorsing it for ordinary business across the board. Haidt (2013) provides a rather balanced view of some of the moral-psychological underpinnings of this split. 27. That genetics played a role was relatively clear from the fact that a first rather singular successful living kidney transplantation took place between identical twins in 1954. 28. I restrict myself to the German case for reasons of space, because I know it at first hand and because it is particularly instructive and typical in several regards that will hopefully become obvious from my discussion. 29. With improving capabilities to prevent ischemic damage to the explanted organ the time of safe so-called cold ischemia could be extended to 48 hours (faster still being better, of course). 30. Extending the time of medically acceptable cold ischemia by developing new procedures and technologies was also important for facilitating organ exchange. 31. Of the elements necessary to give rise to a private law interaction, ‘stability of possession, its transference by consent and the performance of promises’ (Hume 1739, Bk III, xii), the first two were clearly assumed to be present and acknowledged as such. 32. In the case of donations of corpses for medical teaching and research, consent was often given during the deceased’s lifetime, but not in the case of organ donation. 33. In the course of time donor cards and/or transplantation registers were introduced, and opting-in – the default rule being a non-donor, as in Germany – and respectively opting-out rules – the default being a donor, as in Austria – were implemented.

Weird health care for WEIRD societies?  143 34. Such a mutuality was suggested, for instance, by the eminent biomedical researcher Joshua Lederberg ten days after the first heart transplant; see on this Lederberg (1967), Gubernatis and Kliemt (2000) and Kliemt (2021). 35. See Marlies Ahlert (2004) on possible threshold problems that might lead to club failure in this context. 36. In view of respect for the autonomy of the human individual, it would seem prima facie plausible to treat the corpse as something that a person could dispose of during their lifetime in ways that name the intended purpose and recipient. In view of an otherwise – at least in WEIRD societies – fairly common resentment against free-riders, prioritizing potential donors as recipients would have been a most natural way of extending current practices. 37. The various social contract fictions seek to camouflage the basic conflict between plural values by presenting imposed rulings as if they were agreement based. But in the real world dissent prevails. 38. The larger ‘medical community’ was – like the public at large – inclined to prefer a situation in which human cadaveric organs are treated as a common pool resource and allocated according to criteria of highest medical benefit. 39. Somewhat strangely in view of the fact that organs were acquired as unselfish gifts of donors rather than in exchange for a quid pro quo. 40. The focus was centrally on human lycocyte antigens; a so-called full house match would increase the predicted five-year survival time of the transplant by some 8 percent as compared with a transplantation involving more than one mismatch. The significant difference in survival time between a full house match (about 7 percent as compared to a case with one mismatch) made it desirable to realize full house matches. As it turned out, such full house matches could be realized under prevailing distributions of genetical traits in Europe in perhaps up to 25 percent of cases provided that the pool of waiting individuals was large enough, but certainly could be realized in 20 percent of the cases with a large pool. 41. On clubs, of course, see Buchanan (1965) and more comprehensively Cornes and Sandler (1996). 42. Since due to cold ischemic injury storing kidneys for extended periods of time is not viable, such exchanges served the double purpose of directly improving matching at the time and indirectly as an imperfect remedy for the inability to preserve kidney explants over time. 43. For more on cross-over kidney donation, see the elegant treatment in Roth (2016). 44. Only a combination of treating organs as common property and communism in allocating them can completely disentangle the provision of care ‘according to need’ from the requirement to contribute according to ability. A socialist rather than communist rule that prioritizes access to organs proportional to the time of one’s own willingness to donate before the onset of ESRD will not do. In that case those endowed with two healthy kidneys would be treated as able and expected to contribute by declaring their willingness to donate during their lifetime and failing to contribute could be excluded if not having made use of this ability. 45. On the religious roots of communist thinking in Western thought more recently, see the impressive account in Bovens and Lutz (2019), and on the empirical motivational side Bowles (2017). 46. Alluding to Seabright (2010). 47. Restrictions of capacities are viable since capacities typically cannot be increased instantaneously; fewer intensive care unit capacities, fewer rescue helicopters, fewer specialized units for treating severe burn injuries are examples of such viable limitations. Not making use of available capacities will have such a high symbolic potential that there are strong political temptations not to make the capacities available in the first place. Likewise, not licensing drugs at all instead of merely not financing them will be tempting. These and many similar (ab)uses of regulatory powers of the state may conceivably emerge for such reasons. 48. On such veils, see originally Vickrey (1948) and later, of course, James M. Buchanan and Tullock (1962), Harsanyi (1976) and Rawls (1971); but in health care the rule of rescue leads to piercing any veil in practice. 49. True enough there is an ‘almost’ proviso involved; for a concrete example, see Ahlert (2001) regarding organ allocation. 50. Excellent analyses and practical proposals of how to tweak institutions in the desired direction can be found – with a focus on the American situation – in Goodman (2015), but I believe that these arguments underestimate the power of the demand for guarantees or guaranteed rations of care.

144  Handbook on the political economy of health systems

REFERENCES Ahlert, Marlies. 2001. ‘A Lexicographic Decision Rule with Tolerances: The Example of Rule Choice in Organ Allocation’. Analyse & Kritik 23 (2): 191–204. Ahlert, Marlies. 2004. Public and Private Choices in Organ Donation. Working Paper. Barry, Robert L., and Gerard V. Bradley. 1991. Set No Limits: A Rebuttal to Daniel Callahan’s Proposal to Limit Health Care for the Elderly. Urbana: University of Illinois Press. Bentham, Jeremy. 1843. ‘Anarchical Fallacies’. In The Works of Jeremy Bentham, vol. 2, ed. John Bowring. Edinburgh: William Tait. https://​oll​.libertyfund​.org/​title/​bowring​-the​-works​-of​-jeremy​ -bentham​-vol​-2. Böhm, Franz. 1966. ‘Privatrechtsgesellschaft and Marktwirtschaft’. ORDO 17: 75–151. Bovens, Luc, and Adrien Lutz. 2019. ‘From Each According to Ability; To Each According to Needs’. History of Political Economy 51 (2): 237–57. Bowles, Samuel. 2017. The Moral Economy: Why Good Incentives Are No Substitute for Good Citizens. Reprint. New Haven: Yale University Press. Brennan, Geoffrey, and Hartmut Kliemt. 2019. ‘Private and Political Exchange: Nozick’s Club and Buchanan’s Communitarian Contractarianism’. In Demokratie und Entscheidung: Beiträge zur analytischen politischen Theorie, eds. Karl Marker, Annette Schmitt and Jürgen Sirsch. Wiesbaden: Springer VS, 107–29. Brennan, Geoffrey, and Loren Lomasky. 1985. ‘The impartial spectator goes to Washington’. Economics and Philosophy 1 (2): 189–211. Brennan, Geoffrey, and Loren Lomasky. 1993. Democracy and Decision. Cambridge: Cambridge University Press. Brennan, H. Geoffrey, and Loren E. Lomasky. 1989. Large Numbers, Small Costs: Politics and Process – New Essays in Democratic Thought. Cambridge: Cambridge University Press. Breyer, Friedrich, and Hartmut Kliemt. 2015. ‘“Priority of Liberty” and the Design of a Two-Tier Health Care System’. Journal of Medicine and Philosophy 40 (2): 137–51. Buchanan, Allen. 1997. ‘Health-Care Delivery and Resource Allocation’. In Medical Ethics, ed. Robert M. Veatch. New York: Jones and Bartlett, 321–62. Buchanan, James M. 1965. ‘An Economic Theory of Clubs’. Economica 32 (125): 1–14. Buchanan, James M., and Gordon Tullock. 1962. The Calculus of Consent. Ann Arbor: University of Michigan Press. Calabresi, Guido, and Philip Bobbit. 1978. Tragic Choices: The Conflicts Society Confronts in the Allocation of Tragically Scarce Resources. New York: Norton. Callahan, Daniel. 1987. Setting Limits: Medical Goals in an Aging Society. New York: Simon & Schuster. Cornes, Richard, and Todd Sandler. 1996. The Theory of Externalities, Public Goods, and Club Goods. 2nd ed. Cambridge: Cambridge University Press. Goodman, John C. 2015. A Better Choice: Healthcare Solutions for America. Oakland: Independent Institute. Greene, Joshua David. 2013. Moral Tribes: Emotion, Reason, and the Gap between Us and Them. New York: Penguin. Gubernatis, Gundolf, and Hartmut Kliemt. 2000. ‘A Superior Approach to Organ Allocation and Donation’. Transplantation Proceedings 70 (4): 699–707. Haidt, Jonathan. 2013. The Righteous Mind: Why Good People Are Divided by Politics and Religion. London: Penguin. Hamilton, David, Thomas E. Starzl and Clyde F. Barker. 2012. A History of Organ Transplantation. Pittsburgh: University of Pittsburgh Press. Harsanyi, John C. 1976. Essays on Ethics, Social Behavior, and Scientific Explanation. Theory and Decision Library, vol. 12. Dordrecht: D. Reidel. Henrich, Joseph. 2020. The Weirdest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous. London: Allen Lane. Hume, David. 1739. A Treatise of Human Nature. Oxford: Clarendon. Hume, David. 1985. Essays: Moral, Political and Literary. Indianapolis: Liberty Fund.

Weird health care for WEIRD societies?  145 Kliemt, Hartmut. 1986. ‘The Veil of Insignificance’. European Journal of Political Economy 2 (3): 333–44. Kliemt, Harmut. 1993. ‘On Justifying a Minimum Welfare State’. Constitutional Political Economy 4 (2): 159–72. Kliemt, Hartmut. 1997. ‘Subsidiarity as a Guiding Principle of Constitutional Design and of Constitutional Adjudication’. In Normative Systems in Legal and Moral Theory, eds. Ernesto Garzón Valdés, Werner Krawietz, Georg Henrik von Wright and Ruth Zimmerling. Berlin: Duncker & Humblot. Kliemt, Hartmut. 2006. ‘Ethische Konflikte im Gesundheitswesen’. Perspektiven der Wirtschaftspolitik 7 (Mai): 27–48. Kliemt, Hartmut. 2015. ‘The Defensive State’. Independent Review 20 (1): 11–21. Kliemt, Hartmut. 2021. ‘Virginia Virtue, Virginia Vice’. In The Soul of Classical Political Economy: James M. Buchanan from the Archives, eds. Peter J. Boettke and Alain Marciano. Advanced Studies in Political Economy. Arlington: Mercatus Center. Lederberg, Joshua. 1967. ‘Heart Transfer Poses Grim Decisions: Moribund Patient’s Trust is at Stake’. Washington Post, December 10: B1. Lumsden, Charles J., and Edward O. Wilson. 1981. Genes, Mind, and Culture: The Coevolutionary Process. Cambridge: Harvard University Press. North, Douglass C., John Joseph Wallis, and Barry R. Weingast. 2013. Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Reprint. Cambridge: Cambridge University Press. Nozick, Robert. 1974. Anarchy, State, and Utopia. New York: Basic Books. Ostrom, Elinor. 1990. Governing the Commons. New York: Cambridge University Press. Pius XI. 1931. Quadragesimo Anno (May 15). www​.vatican​.va/​content/​pius​-xi/​en/​encyclicals/​ documents/​hf​_p​-xi​_enc​_19310515​_quadragesimo​-anno​.html. Rawls, John. 1971. A Theory of Justice. Oxford: Oxford University Press. Robbins, Lionel. 1935. An Essay on the Nature and Significance of Economic Science. London: Macmillan. Roth, Alvin. 2016. Who Gets What – And Why: Understand the Choices You Have, Improve the Choices You Make. London: Harper Collins. Sandel, Michael. 2012. What Money Can’t Buy: The Moral Limits of Markets. Open Market Edition. London: Allen Lane. Seabright, Paul. 2010. Company of Strangers. 2nd Revised Edition with a New Foreword. Princeton: University Press Group. Vickrey, William. 1948. ‘Measuring marginal utility by reactions to risk’. Econometrica 13 (4): 319–33.

PART III POLITICAL INSTITUTIONS, POLITICAL MARKETS AND HEALTH CARE DECISIONS

10. Medical composition of cabinets and the health care system Joan Costa-Font, Nicolas Marchi and Debra Winberg

1. INTRODUCTION Government decisions are not independent of individuals making such decisions, and cabinets are typically made up of ministers and the prime minister, playing the role of ‘core executive’ in setting priorities and coordinating the governmental policy-making process. Given that all government decisions eventually pass through the cabinet, ministers are political actors in making policy decisions (Costa-Font et al., 2020). Hence, the specific background and training of cabinet members can influence policy and regulations, including the priority that health care has compared to other policy matters, and the types of health policies and outcomes. Generally speaking, we argue that the influence of cabinets can be predicted by the influences of health-related ministers (also called ‘health secretaries’ in some countries) in a cabinet, especially to counteract the strength of the prime minister and the treasury or finance ministers in steering health care decisions, and the stability of such health ministers regarding crises in the country, as well as their visibility and accountability (Costa-Font et al., 2020). Health care is a government responsibility that is typically perceived as technical and typically a continuous driver of significant public health care spending growth, which means that high-level administrators play a central role as advisors, and warranting the continuity of policy decisions and their logic. Hence, ministers that are better versed in specific health care decisions can exert a differential influence. Some events, such as Covid-19 or other health crises, might have scaled up the role of health care; similarly the decentralized or federal structure of health care decision making in many countries means that health care decisions are often multilevel, and entail coordinated cabinets at different levels of government (see Congleton, this volume). At the highest level of executive government functioning, different motivations and influences can sway the health agenda and the policies pursued. This can have downstream effects on measurable health outcomes and the health services offered to the population. In this chapter, we focus on how the specific actors and dynamics of cabinets can sway health policy decisions. More specifically, we examine whether, at the individual level, the background of the health minister becomes particularly relevant. Some health ministers were formerly trained as doctors and possess a medical background. For this reason, we report evidence on whether a health minister’s professional background, and more specifically whether they have been trained as a medical doctor (MD),1 is associated with any specific health system outcomes. It is intuitive to ascertain that doctors, due to their similar training, are more likely to share similar values, and perhaps this is reflected in health policy decisions and spending. Understanding the effect of variables at the individual level provides a framework of analysis that can explain policy decisions. 147

148  Handbook on the political economy of health systems Health care decisions are subject to both political and economic influences. Interest group politics, elite preferences and institutional arrangements can all strongly influence health. Vatter and Wälti (2003) found in the Swiss federal state that both the supply and demand side of health were heavily influenced by political factors. Robert Alford (1975) argued that health industry expansion in the United States is due to power struggles between different groups seeking to control a greater market share of health resources. Electoral politics also tends to be responsive to voter interest in health care. Health care is a highly salient voter issue. In the United States, health care was the leading issue for potential voters in the 2016 presidential election (Pearl, 2018). In the United Kingdom, health care and the National Health Service was the third most important issue for voters in a 2014 poll (Ipsos MORI, 2014). Health expenditure, usage and outcomes affect the citizenry and are therefore politically important. Examining how health ministers and their backgrounds may be associated with different outcomes lends itself to a political economy analysis. The economic, educational and political leanings of a health minister may all play a role in how health is financed and delivered. The current literature in the political economy of health places great focus on system-level methods of analysis, including the broad effects of how equity (geographic and social) and economy (costs and policy) affect the inputs and outputs of health (Reisman, 1993). Similarly, Tuohy and Glied (2011) from a system-level standpoint examine the government’s role in the health care system and the political and economic forces that may affect government action. The influence of voting behavior, interest groups and institutions is also discussed (Tuohy & Glied, 2011). Like Reisman’s work, Glied and Tuohy use a system-level approach by examining health system outcomes as a function of government behavior. System-level analyses are used to discuss a wide range of health topics. For example, researchers use a political economy framework to examine the universal health coverage (UHC) reform. Using this framework, researchers focus on the role of the budget-making process, institutionalism, stakeholder engagement, political environment, regulatory processes and negotiation processes (Loffreda et al., 2021; Nannini, Biggeri & Putoto, 2021; Tangcharoensathien, Patcharanarumol, Kulthanmanusorn, Saengruang & Kosiyaporn, 2019). Significant work has also been done to understand how the political economy influences government responses to pandemics such as Ebola, Zika and Covid-19. This research highlights urban policy, neoliberal ideologies, economization, political incentives and voting processes (Allam, 2020; Maffioli, 2021; Patchin, 2021; Primrose, Chang & Loeppky, 2020). By explaining health system functioning by the state’s actions, the literature fails to explain the role of single political and economic actors. Few articles address this individual-level analysis, and the current literature focuses on very specific themes, including interest-level politics, corporatism and individual cabinet ideology. Interest-level politics greatly influence health policy, such as health care reform and UHC (Gray, Lowery & Benz, 2013; Sparkes, Bump, Özçelik, Kutzin & Reich, 2019). Analyses often discuss interest groups (Bambra, Smith & Pearce, 2019; Gray et al., 2013; Novignon, Lanko & Arthur, 2021; Sparkes et al., 2019). However, alternatively it is important to understand the actual composition of cabinets, and especially the characteristics such as the medical background of the individuals who are responsible for making health care decisions. This is the purpose of this chapter. The next section discusses the role of the professional background of cabinet members. Section 3 discusses the evidence of the Global Leadership Project dataset. Section 4 provides evidence of the medical composition of cabinets, and a final section concludes.

Medical composition of cabinets and the health care system  149

2.

THE PROFESSIONAL BACKGROUND OF CABINET MEMBERS

Decision-making individuals in health care are constrained by their past experiences and professional background, which can exert an influence on the type of information and knowledge they have, alongside the specific constraints (membership of, and loyalty to, a specific professional group) individuals face, as well as the group’s values and preferences. For instance, economics students tend to uphold different values to the rest of the population, making them less cooperative and more likely to be self-interested (Frank & González Block, 1992; Frank, Gilovich & Regan, 1996; Frank & Schulze, 2000). Diwan and Vartanova (2018) use the World Value Survey to examine the association between education and political preferences. Drawing on countries that have undergone a regime change, they find evidence that those educated under a democratic system have a larger electoral return in terms of support for education and are less conservative. Other research shows that the gender composition of cabinets matters, as frequently cabinets tend to reflect the preference of males (Krook & O’Brien, 2012; Bhalotra and Lopes da Fonseca, this volume), and the views of middle-class people and those with professional backgrounds, who are found to exhibit a negative correlation with views about expanding sickness- and employment-related benefits. Indeed, ministers and members of the cabinet are constrained by their electoral pledges and especially the manifestos and ideologies of their government alongside their budget. However, the socio-economic backgrounds of members of cabinets, at the margin, play a role in influencing policy. This occurs not only through knowledge, beliefs and preferences, but directly regarding the responsiveness of such policies. This is particularly true in welfare policies, such as those involving health and sickness (Alexiadou, 2022). That is, cabinets that have a higher number of cabinet ministers identified as liberal professionals are consistently associated with cuts in welfare generosity, whereas cabinets with a higher number of ministers identified as sociocultural professionals are consistently associated with positive changes in welfare generosity. Generally speaking, some research finds that in many European political systems organized by proportional representation, governments are just as likely to adopt policies that are closer to the preferences of higher-income citizens (Schakel et al., 2020). Many of the interest-level politics touch upon the idea of corporatism, a collectivist ideology advocating for the organization of society by corporate groups. A popular theoretical approach in the late 20th century, corporatism has been used to describe health equity, the rise of medical associations, and health care decision making (Frank & González Block, 1992; Zweifel & Eichenberger, 1992). More recently, corporatism frameworks have discussed health promotion and current health problems such as non-communicable diseases (Raphael, 2013; Sell & Williams, 2020). Although these individual-level analyses focus slightly on individual ideology and health, they still use somewhat of a system-level approach and do not look at specific individuals. On the other hand, research into cabinet-level ideology takes a truer individual-level approach. However, evidence is mixed on the impact of cabinet ideologies. While some studies found that ideology impacts health outcomes (Barnish, Tørnes & Nelson-Horne, 2018; Herwartz & Theilen, 2014), others found cabinet ideology had little impact on health outcomes or health care reform (Reeves, McKee, Basu & Stuckler, 2013; Tavits & Letki, 2009).

150  Handbook on the political economy of health systems A close examination of who the decision makers are and how they may affect equity and economy is also necessary. While the system-level lens is useful for broad explanation, current literature on the individual level is lacking. The individual-level data as presented in this work provides a more comprehensive analysis. This work seeks to bridge this gap by offering data and drawing conclusions based upon the professional backgrounds of the political elite who hold ministerial positions in health. Indeed, the professional background can influence decision making either through experience and knowledge pathways, or through its effects on social networks of ‘influence’ which pave the way to the development of corporatism. That is, ‘a system of interest and/or attitude representation, a particular model or ideal-typical institutional arrangement for linking the organized interests of civil society with the decisional structures of the state’ (Schmitter, 1974, p. 86). Consistently, organizations of common interest will seek to influence state policy making. Examples of such organized interests would include agriculture, the business community and the military. A minster from one of these industries can exert their power on policy making to strengthen their organization’s interests. In the case of health, ministers who are doctors can use their sphere of influence to maximize the position of doctors, their stance and their revenue in society by limiting the number of doctors in the system, and hence reducing the potential competition that could erode their earnings. Alternatively, doctors might have a better knowledge of the health system, which might convey respect and authority in making decisions that especially affect the human resources in the health system who are doctors themselves, alongside nurses and other health-related professionals. However, the evidence so far on the role of health decision makers with a medical degree is limited, and mostly limited to Germany (Pilny & Roesel, 2020), yet the social insurance constraints of the system might limit the extensive validity of the estimates. This chapter will attempt to add to the existing literature by examining wide cross-country evidence available.

3.

GLOBAL LEADERSHIP PROJECT DATASET

Many authors have compiled datasets on political elites. Two comprehensive datasets on political elites are the Global Leadership Project dataset (Gerring, Oncel, Morrison & Pemstein, 2019) and the Seki–Williams dataset (Williams & Seki, 2016). The Global Leadership Project dataset, while comprehensive, did not include a significant number of health ministers. Ministers’ background data for this chapter was therefore sourced from the Seki–Williams dataset. Additionally, official government web sources provided biographical and professional details on health ministers. The specific information extracted from the Seki–Williams dataset for ministers’ backgrounds included: (1) name, (2) gender, (3) country, (4) in-office tenure dates, (5) political party/affiliation and (6) portfolio area. The remaining information, such as political and medical background, was filled in through extensive biographical searches. The dataset could then be used to determine which individual oversaw health care as their portfolio area during specific years for specific countries. This then allowed specific outcome data to be aligned with specific ministers in specific countries over a set time (generally 1990–2018, with some variation). The output variable data, including health outcomes and health spending, was sourced exclusively from the Organisation for Economic Co-operation and Development (OECD, 2020).

Medical composition of cabinets and the health care system  151

4. EVIDENCE To determine if a relationship between background characteristics and health system outcomes exists, a linear regression model was used. The issue we examine is whether there is an effect seen on health policy and outcomes when the health minister has a medical background. Two significant findings were found from the data: 1. When the health minister is a doctor, there is a negative correlation with the number of physicians practicing per capita in OECD countries. 2. When the health minister is a doctor there is a negative correlation with health spending as a percentage of gross domestic product (GDP). No other significant findings were found regarding health minister medical background and health outcomes and spending. However, an analysis is currently underway to examine the effects of background on an even wider set of variables including health spending, health care outcomes, financing and government systems. The first set of findings examined the effect that medical doctors serving as health ministers may have on macro-level health outcomes and spending. In this first set, it appears that doctors in power have a negative effect on the number of physicians practicing in their country. Figure 10.1 suggests that in about 34 percent of the cabinets examined, the health minister is a medical doctor.

Note: Not all data was available for all years for each country. Source: Global Leadership Project Database.

Figure 10.1

Percentage of ministers of health who are doctors, 1990–2018

152  Handbook on the political economy of health systems

Note: Not all data was available for all years for each country. Source: Global Leadership Database.

Figure 10.2

Percentage of ministers of health who are doctors, by country, 1990–2018

Figure 10.2 reports the share of ministers who are doctors by country. Interestingly, in some countries, like the UK, in our data there is no health secretary who has been a doctor, while in Poland, the Czech Republic and the Slovak Republic it is mostly the norm that health ministers are doctors. This finding is thought-provoking. Intuitively, it might be assumed that doctors who hold power will want to increase the number of physicians in order to offer more health services to the population. Doctors tend to be altruistic and will want to maximize the health benefit to their society. Many OECD countries have supply-limited public health systems, so naturally doctors with authority and sway will want to increase the health care supply to meet demand. This, however, presupposes that all doctors will act altruistically and prioritize the benefit of society. Another theory provides more explanatory power for this finding. Some doctors may act as corporatists at the ministerial level. From a corporatist perspective, a health minister who is a doctor will decrease the number of doctors in society to drive demand for medical doctors and increase their desirability in society. The demand-side pressures and supply-side constraints could manifest as higher salaries and increased autonomy for practicing doctors. Gatekeeping is typical of interest groups, so it would follow that MDs at the executive level will exhibit such behavior. Other professions are represented at the cabinet level. Finance ministers often have banking backgrounds, education ministers may have teaching backgrounds, and so forth. Labor market protections are seen with professional guilds in other cabinet areas (and even in other health care professions, such as nursing). Therefore, it is plausible that health ministers with medical backgrounds are acting in a similar fashion as their cabinet-level colleagues. They ‘protect their turf’ by working to constrain the practicing number of doctors to increase their relative power in the health system. The second finding observed is that in countries where the health minister is a doctor there is an overall reduction in health spending as a proportion of GDP. This might be explained

Medical composition of cabinets and the health care system  153 by the technical expertise held by medically trained health ministers. In most cabinet areas, the executive will typically employ inputs from technical experts. In the case of the health ministry, this might include input from doctors, pharmacists and other professionals (Buse, Mays & Walt, 2012). When the minister is a doctor, there may be less pressure to seek external expertise as the minister will be seen as possessing the requisite technical knowledge. This has several implications, which may explain the reduction in relative health spending observed in the findings. Doctors will likely understand the relative value of lower-cost and more efficient forms of medical interventions. These ministers therefore may be more likely to invest in public health measures, expanded primary care offerings and immunizations. In both developing and developed states, expanded primary care services has been associated with better health outcomes, better access to health care and decreased rates of hospitalization (Shi, 2012). Additionally and understandably, increased vaccination rates tend to be associated with reduced mortality (Kristensen, Aaby, Jensen & Fine, 2000; Shann, 2000). Health ministers with this knowledge may very well understand how to more efficiently allocate health resources to decrease overall health spending. In this context, Cuba would serve as an instructive case study. In Cuba, health policy prioritizes preventative and primary care and pursues an aggressive public health strategy (Cooper, Kennelly & Orduñez-Garcia, 2006). This has directly translated into positive health outcomes and impressive results on health metrics, especially considering economic constraints facing the country. In several indices of health system performance, including life expectancy and malnutrition indices, Cuba fares better than the United States. Those with medical training will likely understand the immense benefit of early intervention, vaccination and primary care. With this knowledge in mind, health ministers who are doctors could bring down health spending by pursing a more public-health-focused policy agenda. At the cabinet level, health ministers hold the executive position over the entirety of the public health sector. Of course, variations exist in terms of the level of centralization of health services. However, broadly, they control the contents of the policy agenda. Health ministers who are doctors also hold an additional layer of influence as being seen as possessing advanced technical knowledge. As medically trained politicians, such health ministers may have greater flexibility in making unilateral decisions. They will be viewed as not requiring external expertise from technical advisors. This increased sway provides both fewer checks and an opportunity for the ideology of the minister to directly influence policy making. Doctors do not hold homogenous views. Having concentrated power at the cabinet level may allow these health ministers to more heavily influence agenda setting and policy outcomes. Individual backgrounds matter for elite politics and health ministers are certainly no different, especially when they are uniquely qualified as medical doctors.

5. CONCLUSION In this chapter, we have examined the relationship between individual-level characteristics of health ministers in OECD countries and the effects on the respective health systems. The medical background of each health minister – whether they were trained formally as a doctor – was compared against different outcomes. There were two main findings. First, we have documented that ministers of health who are doctors (medical cabinets) are associated with a lower number of physicians practicing per capita.

154  Handbook on the political economy of health systems Second, the presence of health ministers who are doctors is associated with lower health spending as a percentage of GDP compared to their non-medic counterparts. There are several explanations for these findings, including the role of corporatist attitudes limiting the expansion of the number of doctors in the system to reduce competition and keep doctors’ earnings well above a competitive equilibrium. Alternatively, it can be argued that medical doctors have a better understanding of lower-cost preventative measures that can improve population health. This allows them to more efficiently allocate health resources and lower overall spending and choose the right advisors and advocates to attain such a goal. Finally, another explanation is that a medical minister might face less resistance when making spending cuts in the health system. These findings are consistent with Besley’s (2005) claim that politicians’ characteristics influence policy, and specifically the individual networks, information and upbringing of health ministers may influence the resulting health spending and the availability of medical doctors in the system.

NOTE 1. An MD in most European countries, and unlike in the United States, is actually a degree beyond standard undergraduate medical training. It typically involves advanced research and is not undertaken (nor particularly common) amongst practicing doctors. An MD in the United States is equivalent to Europe’s MB BCh.

REFERENCES Alexiadou, D. (2022). Cabinet ministers and inequality. European Journal of Political Research, 61(2), 326–50. Alford, R.R. (1975). Health Care Politics, Ideological and Interest Group Barriers to Reform. Chicago, IL: Chicago University Press. Allam, Z. (2020). Surveying the Covid-19 Pandemic and Its Implications: Urban Health, Data Technology and Political Economy. San Diego, CA: Elsevier. Bambra, C., Smith, K.E., & Pearce, J. (2019). Scaling up: The politics of health and place. Soc Sci Med, 232, 36–42. Barnish, M., Tørnes, M., & Nelson-Horne, B. (2018). How much evidence is there that political factors are related to population health outcomes? An internationally comparative systematic review. BMJ Open, 8(10), e020886. Besley, T. (2005). Political selection. Journal of Economic Perspectives, 19(3), 43–60. Buse, K., Mays, N., & Walt, G. (2012). Making Health Policy. Maidenhead: McGraw-Hill Education. Cooper, R.S., Kennelly, J.F., & Orduñez-Garcia, P. (2006). Health in Cuba. International Journal of Epidemiology, 35(4), 817–24. Costa-Font, J., Turati, G., & Batinti, A. (2020). The Political Economy of Health and Healthcare: The Rise of the Patient Citizen. Cambridge: Cambridge University Press. Diwan, I., & Vartanova, I. (2018). Does education indoctrinate? The effect of education on political preferences in democracies and autocracies. Working Papers 1178, Economic Research Forum. Frank, B., & Schulze, G.G. (2000). Does economics make citizens corrupt? Journal of Economic Behavior & Organization, 43(1), 101–13. Frank, J., & González Block, M.A. (1992). Corporatism and health care: A comparison of Sweden and Mexico. Health Policy, 21(2), 167–80. Frank, R.H., Gilovich, T.D., & Regan, D.T. (1996). Do economists make bad citizens? Journal of Economic Perspectives, 10(1), 187–92.

Medical composition of cabinets and the health care system  155 Gerring, J., Oncel, E., Morrison, K., & Pemstein, D. (2019). Who rules the world? A portrait of the global leadership class. Perspectives on Politics, 17(4), 1079–97. Gray, V., Lowery, D., & Benz, J.K. (2013). Interest Groups and Health Care Reform across the United States. American Government and Public Policy. Washington, DC: Georgetown University Press. Herwartz, H., & Theilen, B. (2014). Health care and ideology: A reconsideration of political determinants of public healthcare funding in the OECD. Health Economics, 23(2), 225–40. .ipsos​ .com/​ en​ -uk/​ Ipsos MORI (2014). The Economist/Ipsos Issues Index – July 2014. www​ economistipsos​-mori​-issues​-index​-july​-2014. Krook, M.L., & O’Brien, Z. (2012). All the president’s men? The appointment of female cabinet ministers worldwide. Journal of Politics, 74(3), 840–55. Kristensen, I., Aaby, P., Jensen, H., & Fine, P. (2000). Routine vaccinations and child survival: Follow up study in Guinea-Bissau, West Africa – Commentary: An unexpected finding that needs confirmation or rejection. BMJ, 321(7274), 1435. Loffreda, G., Bello, K., Kiendrébéogo, J.A., Selenou, I., Ahmed, M.A.A., Dossou, J.P., … Bertone, M.P. (2021). Political economy analysis of universal health coverage and health financing reforms in lowand middle-income countries: The role of stakeholder engagement in the research process. Health Research Policy and Systems, 19(1), 143. Maffioli, E.M. (2021). The political economy of health epidemics: Evidence from the Ebola outbreak. Journal of Development Economics, 151, 102651. Nannini, M., Biggeri, M., & Putoto, G. (2021). Health coverage and financial protection in Uganda: A political economy perspective. International Journal of Health Policy and Management, 11(9), 1894–904. Novignon, J., Lanko, C., & Arthur, E. (2021). Political economy and the pursuit of universal health coverage in Ghana: A case study of the National Health Insurance Scheme. Health Policy Plan, 36(Supplement_1), i14–i21. OECD (2020). OECD.Stat. https://​stats​.oecd​.org/​Index​.aspx​?ThemeTreeId​=​9. Patchin, P.M. (2021). For the sake of the child: The economisation of reproduction in the Zika public health emergency. Transactions – Institute of British Geographers (1965), 46(1), 2–14. Pearl, R. (2018). Healthcare is the No. 1 issue for voters; a new poll reveals which healthcare issue matters most. Forbes. www​.forbes​.com/​sites/​robertpearl/​2018/​08/​13/​midterms/​?sh​=​5d5eb41b3667. Pilny, A., & Roesel, F. (2020). Are doctors better health ministers? American Journal of Health Economics, 6(4), 498–532. Primrose, D., Chang, R., & Loeppky, R. (2020). Pandemic unplugged: COVID-19, public health and the persistence of neoliberalism. Journal of Australian Political Economy, 85, 17–28. Raphael, D. (2013). The political economy of health promotion: Part 1, national commitments to provision of the prerequisites of health. Health Promotion International, 28(1), 95–111. Reeves, A., McKee, M., Basu, S., & Stuckler, D. (2013). The political economy of austerity and healthcare: Cross-national analysis of expenditure changes in 27 European nations 1995–2011. Health Policy, 115(1), 1–8. Reisman, D. (1993). The Political Economy of Health Care. London: Palgrave Macmillan. Schakel, W., Burgoon B., & Hakhverdian, A. (2020). Real but unequal representation in welfare state reform. Politics & Society, 48(1), 131–63. Schmitter, P.C. (1974). Still the century of corporatism? Review of Politics, 36(1), 85–131. Sell, S.K., & Williams, O.D. (2020). Health under capitalism: A global political economy of structural pathogenesis. Review of International Political Economy: RIPE, 27(1), 1–25. Shann, F. (2000). Non-specific effects of vaccines in developing countries. BMJ, 321(7274), 1423. Shi, L. (2012). The impact of primary care: A focused review. Scientifica, 2012, 432892. Sparkes, S.P., Bump, J.B., Özçelik, E.A., Kutzin, J., & Reich, M.R. (2019). Political economy analysis for health financing reform. Health Syst Reform, 5(3), 183–94. Tangcharoensathien, V., Patcharanarumol, W., Kulthanmanusorn, A., Saengruang, N., & Kosiyaporn, H. (2019). The political economy of UHC reform in Thailand: Lessons for low- and middle-income countries. Health Systems and Reform, 5(3), 195–208. Tavits, M., & Letki, N. (2009). When left is right: Party ideology and policy in post-communist Europe. American Political Science Review, 103(4), 555–69.

156  Handbook on the political economy of health systems Tuohy, C.H., & Glied, S. (2011). The Political Economy of Health Care. In The Oxford Handbook of Health Economics, eds Sherry Glied and Peter C. Smith. Oxford: Oxford University Press, 58–77. Vatter, A., and Wälti, S. (2003). Swiss Federalism in Comparative Perspective. Special Issue of the Swiss Political Science Review, 9(1), 1–25. Williams, L., & Seki, K. (2016). Seki-Williams Government and Ministers Data. Harvard Dataverse, V3. https://​doi​.org/​10​.7910/​DVN/​0UNUAM. Zweifel, P., & Eichenberger, R. (1992). The political economy of corporatism in medicine: Self-regulation or cartel management? Journal of Regulatory Economics, 4(1), 89–108.

11. Women politicians and public health Sonia Bhalotra and Mariana Lopes da Fonseca

1. INTRODUCTION In 2020, the global average for the proportion of women in parliament reached an all-time high of 25.5 percent. As Figure 11.1 shows, though, there is considerable heterogeneity between countries in the share of women in parliament despite the across-the-board increasing trend in female representation. The top three countries with the highest share of women in parliament are Rwanda, Cuba, and the United Arab Emirates, where women comprise at least 50 percent of the parliament.1 In contrast, three small Pacific countries do not have a single woman in parliament (Vanuatu, Papua New Guinea, Micronesia). In terms of executive positions, 9.5 percent of the heads of state and 6.7 percent of the heads of government are women. Thirteen countries have more women in ministerial positions than men, while in 12 countries women hold no ministerial positions. Despite the significant strides in women’s political representation, the gender gap in political empowerment remains the largest of the four gaps tracked by the World Economic Forum, with only 22 percent closed to date (WEF, 2021).2 At the same time, an increasing body of scholarly evidence demonstrates the importance of political identity in policy decisions (Besley, 2005) and a widening gender gap in political preferences (Edlund and Pande, 2002; Edlund et al., 2005). The provision of a critical basic need might be suffering from the underrepresentation of women: health. Maternal mortality as well as the death of children from preventable diseases, which could be averted by low-cost interventions, may be symptomatic of the low priority that has been put on maternal and child health services. According to the World Health Organization (WHO), in 2019 neonatal conditions were the fifth leading cause of death in the world and the number one cause of disability-adjusted life years. More generally, the United Nations (UN) reports that in 2017 less than half of the global population was covered by essential health services. Women not only favor more redistributive policies than men (Alesina and La Ferrara, 2005), but put a higher priority on children and health issues (Duflo, 2003; Powley, 2006). At the same time as the share of women in parliament and executive positions has increased, important advances have been made in public health provision and outcomes (Figure 11.2). For example, in Eastern Asia, Northern Africa, and Southern Asia, maternal mortality has declined by around two-thirds. This review aims at understanding whether there is a causal relationship between the increase in women’s political representation and the advances in public health provision. If this relationship proves causal, it is essential to further promote the inclusion of women in politics. At the current pace, it will take 145.5 years to attain gender parity in politics (WEF, 2021). According to the WHO, each year an estimated 6.2 million children and adolescents under the age of 15 years die, mostly from preventable causes (WHO, 2020). Every day, approximately 800 women die from preventable causes related to pregnancy and childbirth (WHO, 2023). 157

Figure 11.1

Women in parliament around the world: trend and distribution

Source: Inter-Parliamentary Union (via World Bank).

158  Handbook on the political economy of health systems

Women politicians and public health  159

Note: Raw trends in the number of countries with reserved seats for women, the percentage of women in parliament, and the log of the maternal mortality ratio. Source: Bhalotra et al. (2023), where data details are provided.

Figure 11.2

Trends in gender quotas, women in parliament and maternal mortality

The remainder of this chapter is structured as follows. Section 2 discusses the theoretical backdrop to our proposition: identity politics, the gender gap in preferences, and their implication for women’s political representation. Section 3 discusses the state-of-the-art methodologies used to identify the causal impact of women’s representation. Section 4 surveys the existing evidence on the effect of an increase in women’s representation on policymaking and policy outcomes in the health domain. Section 5 documents the existing evidence on the barriers to women’s political representation, while Section 6 discusses how to overcome these barriers. Finally, Section 7 concludes.

2.

WHY SHOULD GENDER MATTER IN POLICYMAKING

2.1

Identity Politics

While evidence on the extension of suffrage to women appears to have affected policymaking as predicted in a Downsian model with policies gravitating towards the preferences of the median voter (Lott and Kenny, 1999; Miller, 2008), mounting evidence disagrees with this theory of democracy. The favored alternative is a model where politicians’ identity matters (Besley, 2005). In this alternative model, known as the citizen-candidate model (Osborne and Slivinski, 1996; Besley and Coate, 1997), politicians cannot commit to policies, and credibility lies on selection; i.e., on the politicians’ identity. Policy decisions are affected, for example, by the policymakers’ caste (Pande, 2003), religion (Meyersson, 2014), geographic origin (Hodler and Raschky, 2014), ethnicity (Burgess et al., 2015), or family (Washington, 2008).

160  Handbook on the political economy of health systems According to the citizen-candidate model, politicians’ gender may also reshape policy decisions to reflect gender-specific preferences. In this case, the underrepresentation of women in politics is undesirable from a public policy perspective as policies would not address the choices and preferences of a significant part of the population. To study whether gender does affect policymaking, however, is difficult. The lack of randomization in the allocation of women to political positions creates empirical difficulties in identifying the causal impact of women on policy. For this purpose, it is necessary to disentangle constituents’ preferences from politicians’ gender. Chattopadhyay and Duflo’s seminal 2004 paper exploiting a natural experiment in India represents a breakthrough in the understanding of the impact of women’s political representation. 2.2

Gender Gap in Preferences

Lott and Kenny (1999) link the extension of suffrage to women in the US with an expansion of the welfare state, suggesting a gender gap in preferences for redistribution, with women having higher redistributive preferences. Experimental evidence supports this hypothesis with women giving more in lab-based dictator games (Eckel and Grossman, 1998) and being more concerned with inequality (Andreoni and Vesterlund, 2001). Survey-based studies also reveal consistent findings. Using data from a US survey, Alesina and La Ferrara (2005) document that women are more supportive than men of redistributive policies. Two additional survey-based studies even show this gender gap in preferences for redistribution widening both in the US (Edlund and Pande, 2002) and in Europe (Edlund et al., 2005), with women becoming more left-wing. Using the transcripts of the UK House of Commons debates as data, Bhalotra and Fernandez (2018) show that women are more likely to discuss families, health, and employment, while men are more likely to engage in and elaborate on debates about Europe and taxes (Figure 11.3). They also document that the gender difference in the frequency with which health is discussed outweighs the party difference. In line with the inferred gender gap in preferences for redistribution, increases in women’s representation are generally linked to advances in public goods provision (Chattopadhyay and Duflo, 2004), and investments in education and health (see, for example, Beaman et al., 2007; Clots-Figueras, 2011, 2012; Bhalotra and Clots-Figueras, 2014). These findings have fueled the hypothesis that women might make a difference in fixing the political failures in public health provision (Besley and Kudamatsu, 2006; Chaudhury et al., 2006), most obvious in developing countries. All in all, advances in women’s representation are expected to benefit public health provision for the following reasons. First, if women have higher preferences for redistribution, as discussed, they will tend to invest more in broad-based public services that serve everybody and are most used by the poor. Second, they place a greater weight on children’s welfare than men (Besley and Case, 2003; Duflo, 2003; Lundberg and Pollak, 1993; Powley, 2006). In many developing countries, children still die of preventable diseases (Black et al., 2003), which can be averted with low-cost interventions (Cutler et al., 2006; Lorentzon and Pettersson-Lindbom, 2021). Third, women disproportionately bear the costs of a weak public health system and, therefore, also disproportionately benefit from its improvement, particularly in terms of maternal and children health services (Bhalotra and Clots-Figueras, 2014;

Women politicians and public health  161

Source: Bhalotra and Fernandez (2018).

Figure 11.3

Participation index by political party and debate topic

Bhalotra et al., 2023). Fourth, women naturally understand maternal health risks better (Ashraf et al., 2021) and can thus make more informed decisions. 2.3

Women as Politicians

Differences in ability alone cannot explain the absence of women in typically male-dominated fields such as politics (Ellison and Swanson, 2010). Women who decide to enter politics and succeed are likely of above-average quality (Fox and Lawless, 2004). For example, competitively elected female mayors in Brazil are less likely to be corrupt (Brollo and Troiano, 2016). Indeed, different studies indicate positive self-selection of women politicians. In a contribution to a literature that has focused on women politicians changing the composition of policy, Baskaran et al. (2018) show that increasing the share of women in Indian state legislative assemblies leads to improved economic growth in the constituencies from which the women are elected. They also show that women are more effective at completing village road-building projects, and they identify lower rates of corruption among women as a mechanism. This is relevant as it suggests that any compositional changes, such as prioritizing health, need not come at the cost of other items.

162  Handbook on the political economy of health systems

3.

IDENTIFYING THE CAUSAL IMPACT OF WOMEN ON POLICYMAKING

3.1

Gender Quotas

In recent decades, many countries sought to rectify the existing gender imbalance in politics by establishing gender quotas (Figure 11.4).3 These reforms, encompassing legislative candidate quotas, reserved seats, and political party quotas, accelerated women’s entry into politics.4 It is within the context of affirmative action policies that researchers gather much of the evidence on the impact of the increase in women’s representation on policy decisions and outcomes.

Note: Geographic distribution of countries implementing a quota for reserved seats in parliament and candidate list quotas. Source: Data is compiled from Dahlerup (2005) and updated with information for recent years from the online quotaproject.org database.

Figure 11.4

Quota coverage: 1990–2015

Much of the evidence comes from India, which has a mandate reservation system at the village level. Since 1993, a gender quota reserves, in randomly allocated villages, a third of the seats in village councils and the leadership position for women. The randomness of the assignment mechanism allows researchers to attribute differences in policymaking and policy outcomes to the leader’s gender. Since Chattopadhyay and Duflo (2004), different authors have taken advantage of this setting to study various aspects of female representation. Additionally, researchers have also turned to quasi-experimental regression methods to explore (quasi-) random variation in female representation to identify the causal impact of women on policy. Within the context of the Indian mandated reservation practice, Beaman et al. (2007) document that female village leaders provide more public goods, while Chattopadhyay and Duflo (2004) find them catering to women’s preferences. In terms of outcomes, in reserved villages children are healthier, and the gender gap in school attendance and educational attainment declines more rapidly than in unreserved villages (Beaman et al., 2007, 2012). In other settings, authors leverage different quota implementation dates and institutional aspects to identify causal effects. For example, Clayton and Zetterberg (2018) use a time-series

Women politicians and public health  163 cross-sectional data on quota adoption to compare countries that did with those that did not introduce a gender quota. Bhalotra et al. (2023) leverage idiosyncratic variation in the implementation of parliamentary gender quotas across countries (Figure 11.5) to show that gender quotas lead to sharp drops in maternal mortality and identify several mechanisms. Bagues and Campa (2021), in turn, exploit population thresholds and the introduction of gender quotas in Spain to study the impact of the increase in female representation on local public finances.

Note: Timing of the implementation of reserved seats documented by geographic region. Source: Bhalotra et al. (2023).

Figure 11.5

Reserved seat quota timing: 1990–2012

However, affirmative action can result in a quality gap between genders which would reinforce negative biases if women who benefit from such measures are less qualified than the men they displace. This is unlikely as there should be enough high-quality, ambitious women willing to enter politics (Casas-Arce and Saiz, 2015). Indeed, extant scholarly evidence confirms that female politicians elected under affirmative action policies are at least as qualified as their competitively elected male and female peers and more qualified than the men they displace (Baltrunaite et al., 2014; Besley et al., 2017; Bagues and Campa, 2021).

164  Handbook on the political economy of health systems 3.2

Competitive Elections

Affirmative action, however, changes the nature of the political process and potentially the pool of candidates. Thus, it is also critical to assess the impact of competitively elected women. Research using close mixed-gender elections for the Indian state parliament allows a within-country assessment of the effect of women’s political representation, even if at different levels of government. In line with the gender gap in redistributive preferences and the evidence from studies exploring affirmative action policies, Clots-Figueras (2011, 2012) finds that an increase in the share of competitively elected women in Indian state parliaments leads to more investment into health and education. In terms of outcomes, studies find that such an increase in female representation also improves primary education attainment (Clots-Figueras, 2012) and economic growth (Baskaran et al., 2018) in districts that elect women. Outside India, Ferreira and Gyourko (2014) exploit mixed-gender close mayoral elections in the US to study the impact of the mayor’s gender on the size or composition of municipal government expenditure, which is indirectly relevant to the question of whether health spending changes. Baskaran and Hessami (2019), in turn, use female victories in mixed-gender races for the last council seat in Bavarian municipalities to explain the expansion of child-care provision. This is relevant because it demonstrates the mechanisms to explain why women leaders may be better at providing public health care.

4.

WOMEN AND PUBLIC HEALTH

A large body of scholarly evidence links women to increases in investment in public health. For example, in the same vein as Lott and Kenny (1999), Miller (2008) uses historical data on women’s suffrage rights in the US and a difference-in-difference approach to link women’s voting rights to an increase in public health spending. This spending fueled large door-to-door hygiene campaigns that are connected with a reduction in child mortality of between 8 and 15 percent in early-twentieth-century US. However, few studies explore the causal link between women’s political representation and public health provision. The rest of this section discusses studies examining the causal link between women’s political representation and public health provision (Table 11.1), relying on exogenous variation from the two sources discussed in the previous section: the introduction of gender quotas and close competitive elections. 4.1

Gender Quotas

Clayton and Zetterberg (2018) use time-series cross-sectional data from 139 countries during 1995–2012 to measure how gender quotas affect government spending priorities. The authors rely on a difference-in-difference approach, centered around quota adoption, to compare countries that did with those that did not introduce a gender quota. Results show that large increases in women’s parliamentary representation are followed by higher spending in public health. To assess the trade-off of this investment, the authors analyze spending on other categories and find that the increase in public health spending is offset by a decrease in military spending. Bhalotra et al. (2023), in turn, use a sample of up to 178 countries and harmonized time-series data on maternal mortality between 1990 and 2015 to analyze the impact of the introduction of gender quotas in 21 developing countries in the 1990s. Relying on an event study approach

Women politicians and public health  165 Table 11.1

Women politicians and public health

Study

Sample

Methods

Main result

Clayton & Zetterberg (2018)

139 countries, 1995–2012

Diff-in-diff

↑ women representation,

Bhalotra et al. (2023)

178 countries, 1990–2015

Event study,

↑ women representation,

diff-in-diff

↓ maternal mortality

Lippmann (2022)

French parliament, 2001–2007

RDD, diff-in-diff,

Women focus on child and

text analysis

health issues

Beaman et al. (2007)

Surveys from India

OLS

Women leaders improve

Clots-Figueras (2011)

16 Indian states, 1967–2001

RDD, IV

↑ women representation,

Bhalotra & Clots-Figueras (2014)

246 Indian districts, 1968–1999

RDD, IV

↑ women representation,

↑ public health spending

children’s well-being ↑ health provision ↓ infant mortality

Notes:  This table summarizes the evidence on women politicians and public health. The abbreviations diff-in-diff, RDD, OLS and IV refer to difference-in-difference, regression discontinuity design, ordinary least squares and instrumental variables methodologies.

and a difference-in-difference design, the authors leverage the timing of gender quota legislation across developing countries and document an increase in women’s representation and a reduction in maternal mortality (Figure 11.6). This decline is due to gender quotas leading to a drop in fertility, and an increase in skilled birth attendance, prenatal care utilization, and the schooling of young women. These outcomes were made possible by an increase in state health expenditure following the introduction of gender quotas. When studying whether these advances in maternal health had a detrimental effect on other health outcomes, the authors do not find an impact on male mortality in the reproductive age band, on tuberculosis, nor on infant mortality. The authors further show that gender quotas did not affect gross domestic product (GDP) nor the international development assistance for maternal health. The only study focusing on European politics, by Lippmann (2022), analyzes gender differences in lawmaking in the French parliament following the introduction of a gender quota in 2001 that produced a sharp increase in the share of women legislators. The study builds on text analysis tools to classify into topics 300,000 amendments defended by legislators between 2001 and 2007 in both parliament houses. Using a regression-discontinuity approach exploiting close mixed-gender races for the Lower House, the study finds that men and women work on different topics with women being more active on child and health issues (Figure 11.7). For the Upper House, a difference-in-difference strategy comparing the activity of districts affected with those unaffected by the quota (the quota only applies to districts that elect more than four senators) identifies an increased engagement in women’s issues in treated districts. In line with Clayton and Zetterberg (2018), the increased activity in women’s issues is offset by decreased activity on military issues. The remaining evidence on female representation and public health comes from India. Beaman et al. (2007) use data on the practice of seat reservation for women in village governments in India to assess the impact of increased female representation on children’s well-being. Because reservations were randomized across village councils, they identified causal impacts, showing improvements along two dimensions. Children in reserved villages in West Bengal have better access to drinking water, and children in reserved villages in Rajasthan are more

166  Handbook on the political economy of health systems

Note: Estimates of the impact of quotas on women in parliament and maternal mortality. Source: Bhalotra et al. (2023), where details of the data and the method are provided.

Figure 11.6

Gender quota impacts: de Chaisemartin and D’Haultfœuille’s DIDM estimator and placebos

likely to be immunized and attend government daycare centers. These improvements largely reflect the concerns raised by women in these two states. 4.2

Competitive Elections

Additionally, using data on female representation in Indian state legislatures, Clots-Figueras (2011) shows that female representation leads to higher investments in public health provision, and Bhalotra and Clots-Figueras (2014) document reductions in neonatal mortality in districts that elect women. Clots-Figueras (2011) uses data on the 16 largest states in India

Women politicians and public health  167

Notes: Each row corresponds to a topic. The outcome is a dummy that equals 1 if the legislator initiates at least one amendment on the topic considered. Each dot represents the coefficient associated with the variable Woman divided by the average of male legislators (scaled effect). Confidence intervals are represented at the 95 percent and 90 percent levels. Source: Lippmann (2022), where details of the data and the method are provided.

Figure 11.7

Gender differences in lawmaking in the Lower House

during 1967–2000 and an instrumental variables approach where the fraction of seats in the state won by women in a mixed-gender close election is the instrument for the fraction of seats in the state won by women. While, compared to male politicians, female politicians improve health provisions, as measured by the number of beds in hospitals and dispensaries, those in seats reserved for disadvantaged tribes and lower castes increase more than the general female politician. Bhalotra and Clots-Figueras (2014), in turn, use the same instrumental variable estimator approach to study antenatal and early postnatal services targeted at women and newborns. Using survey data on health variables and electoral data covering the 16 largest states in India from 1967–1998, results show that raising the share of seats held by women decreases neonatal and infant mortality. Women’s political representation increases the number of antenatal visits a woman makes, raises their probability of taking iron supplements, and of early breastfeeding,

168  Handbook on the political economy of health systems which is when the benefits are greatest. Additionally, women’s political representation reduces home deliveries, which are typically unattended by professionals. The focus on antenatal and early postnatal care is not only critical for mothers, as proven by the reduction in maternal mortality, but also for newborns and infants. In line with the fetal origins hypothesis (Barker and Osmond, 1986), growing evidence shows that health interventions are more productive the earlier in childhood they are made (Almond and Currie, 2011). Two studies exploring historical policy experiments in Sweden in the early twentieth century confirm that investments that improve antenatal health and promote infant care in the first months are very productive. Bhalotra et al. (2017) show that exposure to the program, which covered information and support for mothers and infant care through home visits and clinics, led to a decrease not only in infant death but also in the risk of dying by age 75. Additionally, leveraging quasi-random variation in eligibility for the program, Bhalotra et al. (2021) estimate that exposure to the program was associated with higher test scores in primary school for boys and girls, and gains in secondary schooling, employment, and earnings for girls. While studies find competitively elected women and those benefiting from affirmative action policies focusing on the same policy topics and producing similar outcomes in India, this experience may not extend to other countries. In fact, outside India, there is little evidence of the impact of women on policymaking or policy outcomes, neither for competitively elected women nor for those that benefit from affirmative action. However, recent studies by Baskaran and Hessami (2019) and Lippmann (2022) suggest that the lack of evidence may be related to the aggregate level of the outcome variables used in previous studies and the nature of the political office under analysis; i.e., executive rather than legislative. Thus, recent evidence suggests that female politicians are broadly pro-redistribution and prioritize investments that promote children’s well-being and health provision.

5.

BARRIERS TO WOMEN’S POLITICAL REPRESENTATION

In this section, we discuss two major supply- and demand-side explanations for the underrepresentation of women in politics: women’s unwillingness to compete and discrimination. 5.1

Supply Side: Competitiveness

One of the main explanations for the underrepresentation of women in politics as well as in many high-profile jobs and across whole professions relates to their unwillingness to perform in competitive environments. In an experimental setting, Niederle and Vesterlund (2007) show that in the absence of gender differences in performance, men are more than twice as likely as women to choose to enter a competition. These patterns translate into real-life decisions. In a setting involving the choice of academic tracks of secondary school students in the Netherlands, Buser et al. (2014) find that among boys and girls of similar ability, boys choose substantially more prestigious academic tracks, where prestige is positively correlated with competitiveness. And in a natural experiment involving job-entry decisions, Flory et al. (2015) show that women disproportionately shy away from competitive work settings. This gender gap in the willingness to enter a competition is mainly driven by two factors: male overconfidence and gender differences in preferences for competition (Niederle and

Women politicians and public health  169 Vesterlund, 2007). Men are overconfident about their relative performance, which leads them to engage in too much competition.5 Women are underconfident about their relative performance, which leads them to shy away from competition even in situations where they should embrace it (Niederle and Vesterlund, 2010). However, women’s preferences for competition crucially depend on the environment (Gneezy et al., 2003). For one, task matters. Gender-neutral and female-oriented tasks reveal no differences in preferences for competition: Dreber et al. (2014) show that adolescent girls and boys are equally likely to self-select into competition in a verbal task. Gender composition also appears to matter. Born et al. (2021) document that women are less willing to lead male- than female-majority teams, as in becoming a candidate in the election for the team leader. The authors also show that this unwillingness to lead is due to women’s beliefs of relative underperformance and their low expectations regarding electoral support in male-majority teams. Women’s low feelings of confidence and competence play a key role not only in the decision to enter a competition but also in their performance in competitive environments (Gneezy et al., 2003). In general, as the competitiveness of an environment increases, the performance and participation of men increase relative to that of women. Though this is not always the case. Women’s performance is higher in single-sex competitions as compared to the non-competitive environment, as shown in seminal work by Gneezy et al. (2003). Importantly, the gender gap in performance in competitive environments, when present, does not necessarily match the gender differences in non-competitive performance (Niederle and Vesterlund, 2010). This differential manner in which men and women respond to competitive environments contributes to women’s underrepresentation in fields like politics along two axes. First, selection alone causes very few women to win competitions. If women are less likely to compete, this not only reduces the number of women who enter but also the number of women who win competitions. Second, women may not perform to their maximal ability in mixed-sex competitions. Yet, research shows that there is much to gain from creating environments in which high-ability women are willing to compete. Women who choose competitive environments perform just a well as men in those settings (Croson and Gneezy, 2009).6 Moreover, Shurchkov (2012) documents that women perform better than men in low-pressure verbal tasks, where, unlike men, they prioritize quality over quantity. In addition to this unwillingness to perform in competitive environments, different features of political life may keep women from entering politics. Women may prefer other occupations with a better work–life balance, particularly given their usual roles in caring for the household, children, and older relatives. 5.2

Demand Side: Discrimination

The above evidence and discussion abstracts from another important explanation for the underrepresentation of women: discrimination. Discrimination against women in politics can have two sources: voter bias and party bias. Evidence on voter bias exists for different settings and is typically associated with stereotypes regarding gender roles, which have voters associating men with leadership positions and increased political efficacy to the detriment of women (Akerlof and Kranton, 2000). In Italian regions with less favorable attitudes towards working women, Gagliarducci and Paserman (2012) find female mayors more likely to see their mandates terminated earlier. In France, survey-based measures of voters’ unfavorable

170  Handbook on the political economy of health systems attitudes towards women and local gender earnings gaps correlate negatively with the share of female candidates in parliamentary elections (Le Barbanchon and Sauvagnat, 2021). Exposure to competitively elected women may attenuate these biases. Baskaran and Hessami (2018) exploit mixed-gender mayoral elections in Germany to show that exposure to female mayors diminishes anti-female bias among voters not only in female-led but also in neighboring municipalities. Similarly, by increasing voters’ exposure to female politicians, affirmative action policies can also break down negative stereotypes (Coate and Loury, 1993). Beaman et al. (2009) document that West Bengal villages where the council leader’s position was randomly reserved for women display lower negative biases regarding women’s political effectiveness. Following a term where the constituency was reserved for women, twice as many women stand for and win council seats in West Bengal (Beaman et al., 2009), and female candidates’ probability of success increases five-fold in Mumbai (Bhavnani, 2009). Similarly, in Italy, De Paola et al. (2010) document a lower voter bias against women and a continued increase in female representation in a set of municipalities affected by a gender quota reform in the past as compared to other unaffected municipalities. However, depending on the extent of gender bias, both competitively elected women as well as those benefiting from affirmative action policies may suffer from backlash for violating stereotypes regarding the role of women. Exploiting a setting with mandated political representation for women in India, Beaman et al. (2007) find that villagers prefer men even though villages reserved for women leaders provide more and at least as high-quality public goods. Bhalotra et al. (2018), in turn, find that following a woman’s victory in India’s state legislative assembly fewer women contest the elections, particularly in states characterized by entrenched gender bias. While parties may rationally respond to voters’ bias by fielding fewer women candidates, (male) party leaders may also decide to field fewer women at the expense of lower votes. This party bias typically follows Becker’s (1957) taste-based discrimination. Parties – historically dominated by men – discriminate against women independent of their quality or potential success among voters. Esteve-Volart and Bagues (2012) show that Spanish parties nominate fewer women than optimal and to positions where the likelihood of being elected is low. Nevertheless, the authors find that conditional on the position in the ballot and experience, female candidates attract more votes than male candidates. Also in Spain, Casas-Arce and Saiz (2015) find that parties affected by the introduction of a gender quota increased their vote share by more than their counterparts in the control group, suggesting that voters were happy with more balanced lists. In line with the mechanism in Becker (1957), both studies show that women benefit from more competitive settings. While affirmative action can address women’s hesitance to compete and promote voters’ exposure to female politicians attenuating anti-female bias, it does not address intrinsic party bias. Male-dominated parties may instead use gender quotas as means to maintain the status quo. Frechette et al. (2008) show that the French national assembly passed an ineffective gender quota in the (correct) expectation that it would increase male incumbency advantage. In Spain, parties circumvent gender quotas by strategically nominating females to poorer positions on the ballot, choosing women not based on their quality, and keeping powerful positions out of their reach (Esteve-Volart and Bagues, 2012; Bagues and Campa, 2021).

Women politicians and public health  171

6.

PROMOTING WOMEN’S POLITICAL REPRESENTATION

To further promote women’s inclusion, it is helpful to understand the underlying sources of the observed gender differences: the question of nature or nurture. On the one hand, studies with children and adolescents (see, for example, Dreber et al., 2011, 2014) and a study comparing patriarchal and matrilineal societies (Gneezy et al., 2009) provide support for the argument that stereotypes and societal structures are linked to the observed gender differences in competitiveness, suggesting that nurture matters. On the other hand, genetic or hormonal differences could cause women to be less competitive than men, suggesting that nature matters as well.7 Effective strategies to reduce the observed gender gaps must therefore encompass measures that address both nature and nurture. For example, reducing the competitiveness of the education system and labor markets addresses nature, while targeting socialization and education to eliminate this asymmetric treatment of men and women concerning competitiveness addresses nurture. In politics, affirmative action policies have been relatively successful at addressing both. As discussed, not only competitively elected women but also those benefiting from affirmative action policies can break down stereotypes and weaken negative biases regarding the effectiveness of women politicians (Beaman et al., 2009; Bhavnani, 2009; De Paola et al., 2010). Moreover, Beaman et al. (2012) show that female leaders are positive role models for both adolescent girls and their parents by raising their aspirations and the former’s educational attainment. Additionally, affirmative action policies also address nature by increasing the probability of single-sex competition, increasing women’s willingness to compete (Gneezy et al., 2003; Niederle et al., 2013). To overcome party bias, institutional measures may be necessary. Electoral rules have an impact on the share of women elected, with proportional representation providing a better ground than single-member district elections for the representation of women (Duverger, 1955). A recent study comparing closed- and open-list elections in Spanish municipalities shows that the former increases the share of women among candidates, councilors, and mayors (Gonzalez-Eiras and Sanz, 2021). Additionally, seat-reservations are a type of affirmative action policy that cannot be circumvented as well as gender parity laws that do not allow for violations. The recent introduction in Italy of double preference voting conditional on gender was also successful in increasing the share of female councilors (Baltrunaite et al., 2019).

7. CONCLUSION Women’s representation is important not only due to equity or fairness concerns but because politicians’ characteristics influence policy (Besley, 2005). Accordingly, if a specific population is overrepresented so must be their preferences. In the case of women, their chronic underrepresentation in politics likely translates into underrepresentation of their policy preferences and priorities. In line with this rationale, scholarly research finds significant effects of women politicians on policymaking and policy outcomes (for a review, see Hessami and Lopes da Fonseca, 2020). These effects are in line with women’s documented preferences for redistribution and their priorities regarding children and health provision (Edlund and Pande, 2002; Duflo, 2003; Chattopadhyay and Duflo, 2004; Alesina and La Ferrara, 2005; Edlund et al., 2005).

172  Handbook on the political economy of health systems The 2030 Agenda for Sustainable Development, adopted by all UN Member States in 2015, sets 17 goals, among which are gender equality and good health and well-being. Research shows that gender equality – to the extent that it promotes the political participation of women in leadership positions and decision-making bodies – has the potential to address some of the points under good health and well-being. Bhalotra et al. (2023) document how an increase in women’s political representation following the introduction of gender quotas in 21 developing countries led to a decrease in maternal mortality. Bhalotra and Clots-Figueras (2014) show how an increase in the share of competitively elected women in Indian states’ legislatures led to a decrease in neonatal mortality. Beaman et al. (2007) report women village heads promote investments in clean drinking water and immunization. Enhanced water infrastructure can greatly improve children’s health since water-borne and diarrheal diseases kill over half a million children under 5 yearly (WHO, 2017). Immunization, in turn, can protect children at risk of tropical diseases and hepatitis. According to the UN, the Covid-19 pandemic has interrupted childhood immunization programs in around 70 countries. In fact, evidence explicitly shows an impact on three of the five topics highlighted by the UN: child health, maternal health, immunizations, tuberculosis, and HIV. In addition to the evidence on outcomes addressing these specific targets, the remaining literature on women politicians and public health shows that increased female representation leads to more public health spending (Clayton and Zetterberg, 2018) and health provision (Clots-Figueras, 2011), as well as more activity on legislation addressing children and health (Lippmann, 2022), which could help target the remaining topics. Altogether, women politicians have a clear impact on the supply side of public health provision. However, in developing countries, there is a puzzling low demand for health that also needs addressing. Research shows that women’s political representation also addresses determinants of the demand for health. Women’s political representation improves education attainment (Beaman et al., 2007; Clots-Figueras, 2012; Bhalotra et al., 2023). Female politicians promote women’s property rights (Clots-Figueras, 2011). Potentially as a result of an expansion of contraceptive coverage and women’s schooling, an increase in women’s political representation leads to a decrease in fertility (Bhalotra et al., 2023). Whether similar improvements extend to developed countries is an open question. Lippmann (2022) reports that female legislators in the French parliament share much of the same priorities regarding children and health as their counterparts in developing countries. However, to our knowledge, there is no evidence regarding the impact of women’s political representation on policy choices or outcomes in Western countries. The margin for improvements in specific outcomes is lower in developed countries; however, maternal mortality, for example, is not a problem exclusive of the developing world. Maternal mortality in the US is much higher than in other rich countries and has been on the rise since 2000 (MacDorman et al., 2016).

NOTES 1. For details, see IPU (2021). 2. The World Economic Forum tracks gender gaps in the following dimensions: economic participation and opportunity, educational attainment, health and survival, and political empowerment. 3. For details, see the website of the International Institute for Democracy and Electoral Assistance (IDEA) at www​.idea​.int/​.

Women politicians and public health  173 4. For example, Mali saw the number of seats held by women triple (from 9.5 percent to nearly 28 percent) in the 2020 parliamentary elections thanks to a new quota law (IPU, 2021). 5. Male overconfidence also leads them to trade more excessively than women (Barber and Odean, 2001). 6. Similarly, gender differences in risk preferences among the general population do not extend to specific positions such as managers and entrepreneurs, which could be due to selection (Croson and Gneezy, 2009). 7. For a discussion, see Croson and Gneezy (2009) and the references therein.

REFERENCES Akerlof, G., and R.E. Kranton (2000). Economics and identity. Quarterly Journal of Economics 115(3), 715–53. Alesina, A., and E. La Ferrara (2005). Preferences for redistribution in the land of opportunities. Journal of Public Economics 89(5–6), 897–931. Almond, D., and J. Currie (2011). Killing me softly: The fetal origins hypothesis. Journal of Economic Perspectives 25(3), 153–72. Andreoni, J., and L. Vesterlund (2001). Which is the fair sex? Gender differences in altruism. Quarterly Journal of Economics 116(1), 293–312. Ashraf, N., E. Field, A. Voena, and R. Ziparo (2021). Maternal mortality risk and spousal differences in the demand for children. Mimeo. Bagues, M., and P. Campa (2021). Can gender quotas in candidate lists empower women? Evidence from a regression discontinuity design. Journal of Public Economics 194(5), 104315. Baltrunaite, A., P. Bello, A. Casarico, and P. Profeta (2014). Gender quotas and the quality of politicians. Journal of Public Economics 118, 62–74. Baltrunaite, A., A. Casarico, P. Profeta, and G. Savio (2019). Let the voters choose women. Journal of Public Economics 180, 104085. Barber, B.M., and T. Odean (2001). Boys will be boys: Gender, overconfidence, and common stock investment. Quarterly Journal of Economics 116(1), 261–92. Barker, D.J., and C. Osmond (1986). Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet 327(8489), 1077–81. Baskaran, T., S. Bhalotra, B. Min, and Y. Uppal (2018). Women legislators and economic performance. IZA Discussion Paper No. 11596. Baskaran, T., and Z. Hessami (2018). Does the election of a female leader clear the way for more women in politics? American Economic Journal: Economic Policy 10(3), 95–121. Baskaran, T., and Z. Hessami (2019). Competitively elected women as policymakers. CESifo Working Paper No. 8005. Beaman, L.A., R. Chattopadhyay, E. Duflo, R. Pande, and P. Topalova (2009). Powerful women: Does exposure reduce bias? Quarterly Journal of Economics 124(4), 1497–540. Beaman, L., E. Duflo, R. Pande and P. Topalova (2007). Women Politicians, Gender Bias, and Policy-Making in Rural India. Technical report, Unicef. Beaman, L., E. Duflo, R. Pande and P. Topalova (2012). Female leadership raises aspirations and educational attainment for girls: A policy experiment in India. Science 335(6068), 582–6. Becker, G. (1957). The Economics of Discrimination. University of Chicago Press. Besley, T. (2005). Political selection. Journal of Economic Perspectives 19(3), 43–60. Besley, T., and A. Case (2003). Political institutions and policy choices: Evidence from the United States. Journal of Economic Literature 41, 7–73. Besley, T., and S. Coate (1997). An economic model of representative democracy. Quarterly Journal of Economics 112(1), 85–114. Besley, T., O. Folke, T. Persson, and J. Rickne (2017). Gender quotas and the crisis of the mediocre man: Theory and evidence from Sweden. American Economic Review 107(8), 2204–42. Besley, T., and M. Kudamatsu (2006). Health and democracy. American Economic Review 96(2), 313–18.

174  Handbook on the political economy of health systems Bhalotra, S., D. Clarke, J.F. Gomes, and A. Venkataramani (2023). Maternal mortality and women’s political participation. Journal of the European Economic Association. Working paper, doi: 10.3386/ w30103. Bhalotra, S., and I. Clots-Figueras (2014). Health and the political agency of women. American Economic Journal: Economic Policy 6(2), 164–97. Bhalotra, S., I. Clots-Figueras, and L. Iyer (2018). Path-breakers? Women’s electoral success and future political participation. Economic Journal 128, 1844–78. Bhalotra, S., and M. Fernandez (2018). Gender, debate and rebellion in the UK House of Commons. Mimeo. Bhalotra, S., M. Karlsson, and T. Nilsson (2017). Infant health and longevity: Evidence from a historical intervention in Sweden. Journal of the European Economic Association 15(5), 1101–57. Bhalotra, S., M. Karlsson, T. Nilsson, and N. Schwarz (2021). Infant health, cognitive performance and earnings: Evidence from inception of welfare state in Sweden. Review of Economics and Statistics 104(6), 1138–56. Bhavnani, R.R. (2009). Do electoral quotas work after they are withdrawn? Evidence from a natural experiment in India. American Political Science Review 103(1), 23–35. Black, R.E., S.S. Morris, and J. Bryce (2003). Where and why are 10 million children dying every year? The Lancet 361(9376), 2226–34. Born, A., E. Ranehill, and A. Sandberg (2021). Gender and willingness to lead – does the gender composition of teams matter? Review of Economics and Statistics 104(2), 259–75. Brollo, F., and U. Troiano (2016). What happens when a woman wins a close election? Evidence from Brazil. Journal of Development Economics 122, 28–45. Burgess, R., R. Jedwab, E. Miguel, A. Morjaria, and G. Padro i Miquel (2015). The value of democracy: Evidence from road building in Kenya. American Economic Review 105(6), 1817–51. Buser, T., M. Niederle, and H. Oosterbeek (2014). Gender, competitiveness, and career choices. Quarterly Journal of Economics 129(3), 1409–47. Casas-Arce, P., and A. Saiz (2015). Women and power: unpopular, unwilling, or held back? Journal of Political Economy 123, 641–69. Chattopadhyay, R., and E. Duflo (2004). Women as policymakers: Evidence from a randomized policy experiment in India. Econometrica 72(5), 1409–43. Chaudhury, N., J. Hammer, M. Kremer, K. Muralidharan, and F.H. Rogers (2006). Missing in action: Teacher and health worker absence in developing countries. Journal of Economic Perspectives 20(1), 91–116. Clayton, A., and P. Zetterberg (2018). Quota shocks: Electoral gender quotas and government spending priorities worldwide. Journal of Politics 80(3). Clots-Figueras, I. (2011). Women in politics. Evidence from the Indian states. Journal of Public Economics 95, 664–690. Clots-Figueras, I. (2012). Are female leaders good for education? Evidence from India. American Economic Journal: Applied Economics 4, 212–44. Coate, S., and G.C. Loury (1993). Will affirmative-action policies eliminate negative stereotypes? American Economic Review 83(5), 1220–40. Croson, R., and U. Gneezy (2009). Gender differences in preferences. Journal of Economic Literature 47(2), 448–74. Cutler, D., A. Deaton, and A. Lleras-Muney (2006). The determinants of mortality. Journal of Economic Perspectives 20(3), 97–120. Dahlerup, D. (2005). Increasing Women’s Political Representation: New Trends in Gender Quotas. In J. Ballington and A. Karam (Eds), Women in Parliament: Beyond Numbers, pp. 140–53. International Institute for Democracy and Electoral Assistance. De Chaisemartin, C., and X. D’Haultfœuille (2020, September). Two-way fixed effects estimators with heterogeneous treatment effects. American Economic Review 110(9), 2964–96. De Paola, M., V. Scoppa, and R. Lombardo (2010). Can gender quotas break down negative stereotypes? Evidence from changes in electoral rules. Journal of Public Economics 94, 344–53. Dreber, A., E. von Essen, and E. Ranehill (2011). Outrunning the gender gap – boys and girls compete equally. Experimental Economics 14(4), 567–82.

Women politicians and public health  175 Dreber, A., E. von Essen, and E. Ranehill (2014). Gender and competition in adolescents: Task matters. Experimental Economics 17(1), 154–72. Duflo, E. (2003). Grandmothers and Granddaughters: Old-Age Pensions and Intrahousehold Allocation in South Africa. World Bank Economic Review. Duverger, M. (1955). The Political Role of Women. UNESCO. Eckel, C.C., and P.J. Grossman (1998). Are women less selfish than men? Evidence from dictator experiments. Economic Journal 108(448), 726–35. Edlund, L., L. Haider, and R. Pande (2005). Unmarried parenthood and redistributive politics. Journal of the European Economic Association 3(1), 95–119. Edlund, L., and R. Pande (2002). Why have women become left-wing: The political gender gap and the decline in marriage. Quarterly Journal of Economics 117(3), 917–61. Ellison, G., and A. Swanson (2010). The gender gap in secondary school mathematics at high achievement levels: Evidence from the American mathematics competition. Journal of Economic Perspectives 24(2), 109–28. Esteve-Volart, B., and M. Bagues (2012). Are women pawns in the political game? Evidence from elections to the Spanish senate. Journal of Public Economics 96, 387–99. Ferreira, F., and J. Gyourko (2014). Does gender matter for political leadership? The case of US mayors. Journal of Public Economics 112, 24–39. Flory, J.A., A. Leibbrandt, and J.A. List (2015). Do competitive workplaces deter female workers? A large-scale natural field experiment on job entry decisions. Review of Economic Studies 82(1), 122–55. Fox, R.L., and J.L. Lawless (2004). Entering the arena? Gender and the decision to run for office. American Journal of Political Science 48(2), 264–80. Frechette, G.R., F. Maniquet, and M. Morelli (2008). Incumbents’ interests and gender quotas. American Journal of Political Science 52(4), 891–909. Gagliarducci, S., and M.D. Paserman (2012). Gender interactions within hierarchies: Evidence from the political arena. Review of Economic Studies 79(3), 1021–52. Gneezy, U., K.L. Leonard, and J.A. List (2009). Gender differences in competition: Evidence from a matrilineal and patriarchal society. Econometrica 77(5), 1637–64. Gneezy, U., M. Niederle, and A. Rustichini (2003). Performance in competitive environments: Gender differences. Quarterly Journal of Economics 118(3), 1049–74. Gonzalez-Eiras, M., and C. Sanz (2021). Women’s representation in politics: The effect of electoral systems. Journal of Public Economics 198(C). Hessami, Z., and M. Lopes da Fonseca (2020). Female political representation and substantive effects on policies: A literature review. European Journal of Political Economy 63(C). Hodler, R., and P. Raschky (2014). Regional favoritism. Quarterly Journal of Economics 129(2), 995–1033. IPU (2021). Women in Parliament in 2020. Technical report, Inter-Parliamentary Union. Le Barbanchon, T., and J. Sauvagnat (2021). Electoral competition, voter bias and women in politics. Journal of the European Economic Association 20(1), 352–94. Lippmann, Q. (2022). Gender and lawmaking in times of quotas. Journal of Public Economics 20(March), 104610. Lorentzon, L., and P. Pettersson-Lindbom (2021). Midwives and maternal mortality: Evidence from a midwifery policy experiment in 19th-century Sweden. Journal of the European Economic Association 19(4), 2052–84. Lott, J.R., Jr., and L.W. Kenny (1999). Did women’s suffrage change the size and scope of government? Journal of Political Economy 107(6), 1163–98. Lundberg, S., and Pollak, R.A. (1993). Separate spheres bargaining and the marriage market. Journal of Political Economy 101(6), 988–1010. MacDorman, M.F., E. Declercq, H. Cabral, and C. Morton (2016). Recent increases in the US maternal mortality rate: Disentangling trends from measurement issues. Obstretics & Gynecology 128, 447–55. Meyersson, E. (2014). Islamic rule and the empowerment of the poor and pious. Econometrica 82(1), 229–69. Miller, G. (2008). Women’s suffrage, political responsiveness, and child survival in American history. Quarterly Journal of Economics 123(3), 1287–327.

176  Handbook on the political economy of health systems Niederle, M., and L. Vesterlund (2007). Do women shy away from competition? Do men compete too much? Quarterly Journal of Economics 122(3), 1067–101. Niederle, M., and L. Vesterlund (2010). Explaining the gender gap in math test scores: The role of competition. Journal of Economic Perspectives 24(2), 129–44. Niederle, M., C. Segal, and L. Vesterlund (2013). How costly is diversity? Affirmative action in light of gender differences in competitiveness. Management Science 59(1), 1–16. Osborne, M.J., and A. Slivinski (1996). A model of political competition with citizen-candidates. Quarterly Journal of Economics 111(1), 65–96. Pande, R. (2003). Can mandated political representation increase policy influence for disadvantaged minorities? American Economic Review 93(4), 1132–51. Powley, E. (2006). Rwanda: The Impact of Women Legislators on Policy Outcomes Affecting Children and Families. Technical report, UNICEF. Shurchkov, O. (2012). Under pressure: Gender differences in output quality and quantity under competition and time constraints. Journal of the European Economic Association 10(5), 1189–213. Washington, E.L. (2008). Female socialization: How daughters affect their legislator father’s voting on women’s issues. American Economic Review 98(1), 311–32. WEF (2021). Global Gender Gap Report. Technical report, World Economic Forum. WHO (2017). Diarrhoeal Disease. Fact Sheet, World Health Organization. WHO (2020). Children: Improving Survival and Well-being. Fact Sheet, World Health Organization. WHO (2023). Maternal Mortality. Fact Sheet, World Health Organization.

12. Community and civic participation effects on health and well-being Luke Munford and Daniel Gray

1. INTRODUCTION Worldwide, particularly in the past decade, there has been a renewed focus on policies concerned with the idea that community and its cohesion are important determinants of health and well-being and quality of life (QoL). For example, in 2010 the coalition government in the United Kingdom (UK) – formed of the Conservative Party and the Liberal Democrats – stressed the need for a ‘Big Society’ where individuals engage more with the assets and facilities available to them in their local community. More recently, in England, the English National Health Service (NHS) published a Long-Term Plan (NHS England, 2019) where it committed to championing ‘community approaches’. The underlying logic behind these community approaches was the notion that an individual’s situation can be improved by reducing loneliness and increasing social cohesion. The ultimate aim of these community approaches was to reduce demand on formal health and care services by improving the health of the population through non-clinical interventions such as leisure, education, and the arts. Across Europe, the promotion of social capital as a means to improve health and well-being outcomes is the basis for the Healthy City movement (WHO, 2020). Healthy Cities aim to put individual health outcomes at the forefront of social, economic, and political agendas, resulting in urban settings that support the health and well-being of the people that use them. This movement seeks to promote strong social relations and peer interactions which manifest themselves through, for example, community participation and community ownership of assets. As we subsequently explore in the first case study we consider, these elements are core to the idea of social capital. There has also been a call to continue to promote social-capital-enhancing public policies as communities attempt to recover from the effects of the Covid-19 pandemic. It is hypothesized, based on the current evidence base, that maintaining and possibly even strengthening existing social capital will lead to a quicker recovery (Wong and Kohler, 2020).

2.

WHAT IS ‘SOCIAL CAPITAL’ AND HOW DO WE MEASURE IT?

Social capital is a concept that has its origins in sociology and psychology. However, social capital does not have a widely accepted definition, and those definitions that are used are continually evolving. Several widely used definitions of social capital include: ● Putnam (1993, p. 167): ‘features of social organization, such as trust, norms, and networks that can improve the efficiency of society by facilitating coordinated actions’. 177

178  Handbook on the political economy of health systems ● Woolcock (1998, p. 155): ‘a broad term encompassing the norms and networks facilitating collective action for mutual benefit’. ● Cannuscio et al. (2003, p. 395): ‘as the resources available to individuals and groups through their social connections to their communities’. Portes argues that social capital encompasses all that is good about sociability and interacting with a community, but notes that ‘excessive extensions of the concept may jeopardize its heuristic value’ (Portes, 1998, p. 1). Moreover, Coleman (1994) notes that social capital is an ‘intangible concept’, which is a view held by many sociologists. Given that social capital is acknowledged to be an intangible multi-faceted construct, the majority of sociological work on social capital has focused on the conceptual understanding, as opposed to actually measuring it and determining how individuals can influence their stock of social capital. 2.1

Different Types of Social Capital

One of the most cited works on social capital is Putnam’s seminal book Bowling Alone: The Collapse and Revival of American Community (2000). In this book, Putnam argues that there are two distinct strands of social capital: bonding – or exclusive – social capital; and bridging – or inclusive – social capital. The former relates to strong social ties between homogenous individuals (i.e., within families and/or existing networks of friends) whereas the latter is concerned with attempting to expand social networks to include a more diverse social grouping. The multi-faceted nature of social capital makes it a diverse and complex concept. As such, the classification of social capital has become a source of major debate, with many authors expressing different viewpoints on the contributing dimensions of social capital (Hean et al., 2003). According to the classification widely adopted by the World Bank (Grootaert and Van Bastelaer, 2002), social capital can be divided into two categories: structural and cognitive social capital (Figure 12.1). Structural social capital comprises externally observed and relatively concrete social structures, such as community networks, institutions, and associations, and their rules and guidelines. On the other hand, cognitive social capital includes subjective and intangible elements, such as shared values, reciprocity, trust, attitudes, and norms of behaviour observed within the community or group. In an alternative taxonomy, Pridmore et al. (2007) categorized social capital into four categories: structural social capital, cognitive social capital, horizontal social capital, and vertical (linking) social capital (Figure 12.2). Horizontal social capital can be divided into bonding social capital and bridging social capital. ‘Bonding social capital’ refers to close group relationships such as family, neighbours, friends, and colleagues. ‘Bridging social capital’ is weak links between different occupational and ethnic backgrounds, including formal and informal social participation. In contrast, vertical (linking) social capital represents a hierarchy of unequal relationships based on differences in power and resource capabilities. Putnam’s theory explains that: bonding social capital is connections or perceptions between similar individuals, bridging social capital is connections between different individuals, and linking social capital occurs across institutionalized or formal authority gradients, considering power differences in social relationships. In a recent systematic review, Cai et al. (2021) summarized various studies on social capital and devised three classifications: structural social capital, cognitive social capital, and relational social capital. ‘Structural social capital’ refers to the connections among individuals in

Community and civic participation effects on health and well-being  179

Source: Grootaert and Van Bastelaer (2002).

Figure 12.1

Structural and cognitive social capital

social ties (Nahapiet and Ghoshal, 1998). Individuals can be expected to obtain information, jobs, and resources through networks and connections (Tsai and Ghoshal, 1998). ‘Cognitive social capital’ indicates resources that provide interpretations, representations, and meanings shared within a group and promote mutual understanding of norms in society and shared goals (Nahapiet and Ghoshal, 1998; Tsai and Ghoshal, 1998). ‘Relational social capital’ is based on interpersonal relationships built through individual interactions, which includes factors that influence individual behaviours such as trust, obligations, norms, identity, reciprocity, commitment, and communication. 2.1.1 Individual and ecological social capital In contrast to the above classifications, it has been argued that social capital has two aspects: characteristics of individuals (i.e., individual social capital) and groups of people (i.e., ecological social capital). Individual social capital is frequently measured by asking about an individual’s participation in community and group activities. It is also related to their perceptions of the quality of these relationships. On the other hand, ecological social capital is non-directive relationships, such as connections and social relationships, which can be beneficial regardless of group participation. This can be measured by asking community members individual social capital questions about civic engagement and trust and aggregating them at the community level (Table 12.1) (De Silva et al., 2007; Ehsan and De Silva, 2015). Sociological and political science definitions have tended to emphasize the aggregate-level perspective, albeit not exclusively. By contrast, economists have tended to focus more on an individualistic notion of social capital. Glaeser et al. (2002) emphasized the individual-level perspective by defining social capital as the individual’s social characteristics that enable private returns via interaction with others. They also point out that only if social capital is

180  Handbook on the political economy of health systems

Source: Pridmore et al. (2007).

Figure 12.2

Relationship between structural and cognitive social capital

an individual concept can we hope to rationalize it, and discuss its accumulation and its production by applying the economic toolbox. Guiso et al. (2008) define social capital as ‘good’ culture that is a set of beliefs and values that facilitates cooperation among the members of a community. According to Guiso et al. (2008), social capital can be thought of as an individual-level characteristic that can be accumulated and transmitted across generations and that is distinguished from human capital because its returns are contingent on the norms and beliefs of the other members of the community or society. 2.2

Measures of Social Capital Used in the Literature

Putnam (2000) used many variables to capture levels of social capital, including most notably the uptake of bowling in social and competitive environments as well as the frequency of visits to friends and family. Alternative proxy measures of social capital are listed in Table 12.1 and include concepts such as trust, feeling of community engagement (the focus of Case Study 1), actively playing a part in the local community, and voting (the focus of Case Study 2).

Community and civic participation effects on health and well-being  181 Table 12.1

Social capital at the individual and ecological level

Type

Measurement

Explanation

Individual-level

Trust

General trust in others, trusting others for specific tasks, and/or trust in institutions.

cognitive social

Social Cohesion

Getting along with neighbours, how well community members know one another, and/or the degree of awareness and supportiveness between neighbours.

capital (SC) Social Support

The perceived support in the neighbourhood, an example being whether neighbours are willing to help in theoretical situations.

Sense of Community

Feeling at home in community, neighbourhood attachment, and/or community integration.

Ecological-level

Trust

cognitive SC

The average level of general trust in the environment, and/or the average level of trust in politicians.

Individual-level

Group Membership

The level of participation in local or voluntary organizations.

structural SC

Engagement in Public

The level of involvement in local civic action, and/or informal social control, or the

Affairs

willingness to intervene in neighbourhood-threatening situations.

Social Support

The extent of the actual support or help received from neighbours.

Community Networks

Contact with family and friends, the informal social ties within neighbours, and/or the bridging social ties with others from different social groups.

Ecological-level

Group Membership

structural SC

The per capita membership of voluntary organizations, and/or the per capita number of public spheres.

Engagement in Public

The voting rates within the community.

Affairs

Source:  Adapted from Ehsan and De Silva (2015).  

The World Bank (through the book of Grootaert and Van Bastelaer, 2002) considers the following as proxies for social capital (grouped by classification): ● Horizontal associations ● Number and type of associations or local institutions an individual belongs to ● Extent of membership in local associations ● Extent of participatory decision making ● Extent of kin homogeneity within the association ● Extent of income and occupation homogeneity within the association ● Extent of trust in neighbours ● Extent of trust in government ● Extent of participation and trust in trade unions ● Perception of extent of community organization/cohesion ● Reliance on networks of support ● Percentage of household income from remittances ● Percentage of household expenditure for gifts and transfers ● Participation in civic and political society ● Percentage of population facing political discrimination ● Indices of intensity of political discrimination ● Percentage of population facing economic discrimination ● Indices of intensity of economic discrimination ● Percentage of population involved in separatist movement ● Indices of political rights ● Freedom House index of political freedoms

182  Handbook on the political economy of health systems ● Indices of democracy ● Indices of corruption ● Indices of government inefficiency ● Strength of democratic institutions ● Measures of human liberty ● Measures of political stability ● Degree of decentralization of government ● Voter turnout ● Political assassinations ● Constitutional government changes ● Coups ● Social integration ● Indicators of social mobility ● Measures of strength of ‘social tensions’ ● Ethnolinguistic fragmentation ● Riots and protest demonstrations ● Strikes/disruptions ● Homicide rates ● Suicide rates ● Other crime rates ● Prisoners per 100,000 people ● Illegitimacy rates ● Percentage of single-parent homes ● Divorce rate ● Youth unemployment rate ● Legal and political governance ● Quality of bureaucracy ● Independence of court system ● Expropriation and nationalization risk ● Repudiation of contracts by government ● Contract enforceability ● Contract-intensive money Of course, not all of the extensive list of possible proxies listed above are available to researchers, and not all make sense to analyse in certain contexts. Therefore, the choice of which indicators/measures of social capital to use are often driven by data availability and the exact nature of the underlying research question. Some authors also advocate the construction of indices of social capital that combine many individual facets using techniques such as principal component analysis. The definition of social capital in terms of values and beliefs resolves the debate as to whether social capital is an individual or community/ecological concept. By construction, a belief is an individual measure of social capital formed on the basis of information and priors available to those individuals. As these priors vary across individuals, so will the corresponding beliefs. At the same time, individual beliefs reflect at least imperfectly some latent fundamental traits of the society, namely the true community social capital.

Community and civic participation effects on health and well-being  183

3.

THE EFFECT OF SOCIAL CAPITAL ON HEALTH

There has been a vast amount of research in recent years on the relationship(s) between social capital and health. In fact, there is such an abundance that a recent paper conducted a systematic review of systematic reviews on the topic (Ehsan et al., 2019). The authors screened 4,941 possible titles on the topic, read 187 reviews, and retained 20 systematic reviews in their review of reviews. Ehsan et al. (2019) conclude that there is good evidence that social capital strongly predicts better mental and physical health and that social capital is protective against mortality. Additionally, there was a Special Issue of Social Science & Medicine in 2020 relating exactly to ‘social capital and health’. The editors note that social capital is ‘one of the most popular social science imports into public health’ (Moore and Carpiano, 2020). It is often argued that social capital, however measured, is a key social determinant of health as there are many ‘political levers’ that can be pulled by which to influence it. Given the plethora of empirical papers examining the relationship between social capital and health, we do not consider specific papers here. However, we signpost readers to the ‘review of reviews’ by Ehsan et al. (2019) as well as the Special Issue of Social Science & Medicine. We do, however, briefly discuss in the next paragraph the hypothesized mechanisms by which social capital can affect health. Social capital is suggested to influence health through several channels. Van der Gaag and Webber (2008) outline four main mechanisms. Firstly, groups with higher levels of cohesion are better equipped to participate in collaborative actions that benefit the wider community. Secondly, more cohesive groups are better placed to ensure positive social norms (of good health) are adhered to. Thirdly, reciprocity of exchanges between members of a social network helps promote beneficial health behaviours. Finally, social capital can help the diffusion of information that is beneficial for health between the group members. More broadly, potential mechanisms between social capital and health include reduced loneliness and a greater sense of belonging. For example, social capital can empower people to take more of an interest in themselves and their friends/peers and their community, which can lead to a ‘warm glow’ effect. It is important to note that there are potential negative aspects relating to social capital and health. An example of this would be incorrect diffusion of information between group members. For example, consider the anti-vax movement, where incorrect information may spread more quickly through more cohesive groups.

4.

TWO CASE STUDIES ON HOW SOCIAL CAPITAL CAN AFFECT HEALTH

In this section we outline in detail two case studies that demonstrate the link between social capital and health and well-being. Each case study considers a different aspect of social capital. Both case studies were conducted in the UK (or parts of the UK), but we argue that the results are generalizable to other countries and settings with similar political economies.

184  Handbook on the political economy of health systems 4.1

Case Study 1: The Effects of Engaging with Community Assets and Participating in Society on Health

4.1.1 Overview of community assets and how they are hypothesized to affect health Community assets have been, and continue to be, promoted as a way of reducing loneliness, increasing health and well-being, and reducing demand for formal health care services. These community assets include community centres, libraries, markets, and pubs and are defined as ‘the collective resources which individuals and communities have at their disposal, which protect against negative health outcomes and promote health status’ (McLean, 2011, p. 4). Community assets – the physical infrastructure – can foster many aspects of social interaction through, for example, ‘coffee mornings’ in community centres. Despite there being many theorized relationships between community assets and health, there is surprisingly little quantitative evidence. Based on the Pridmore et al. (2007) categorization of social capital, community assets belong to the ‘structural’ group as they aim to promote enhanced social networks and involvement in the community. They can also be viewed as horizontal social capital, but depending on who else participates in the community assets, they may be viewed as bonding (if similar people attend) or bridging (if different types of individuals attend). In the Grootaert and Van Bastelaer (2002) classification, community assets would be viewed as structural social capital. Foot and Hopkins (2010) argued that community assets can achieve a number of goals, including: (1) providing new ways of challenging health inequalities, (2) valuing resilience, (3) strengthening community networks and (4) recognizing local expertise. McKnight and Kretzmann (1993) introduced community asset mapping, which can be used to provide health promoters ‘with an understanding of how best to create the conditions required to maximise the potential for health’ (Morgan and Ziglio, 2007; p. 20). Kretzmann and McKnight (1996) introduced Asset Based Community Development, a system in which a community increases the health and well-being of its population using activities, skills, and assets of (lower-income) people and neighbourhoods. Mathie and Cunningham (2003) further developed this concept by stressing the importance of a sense of community belonging, social inclusion, and social capital. 4.1.2 Setting Data from a large-scale National Institute of Health Research (NIHR)-funded study ‘Comprehensive Longitudinal Assessment of Salford Integrated Care (CLASSIC)’ were used (Bower et al., 2018). CLASSIC was an evaluation framework designed to evaluate the Salford Integrated Care Programme (SICP). The SICP is a large-scale integrated care project to transform care for older people with long-term conditions and social care needs. The SICP aims to improve care via a number of mechanisms, including improved access to community assets to bolster social capital. Salford is a local authority in the North West of England, close to Manchester and part of ‘Greater Manchester’, a combined authority made up of ten local authorities (see Figure 12.3). The main area of residence is the city of Salford. According to Public Health England, Salford is among the 20 per cent most deprived districts in England, with lower life expectancy than the national average. Ninety-four per cent of residents are white. However, Salford has experienced many health care reforms in the recent past, particularly around older people. As

Community and civic participation effects on health and well-being  185 a result, Salford is the first ‘Age Friendly City’ and the Age Well campaign has experienced considerable success at promoting engagement and improving health outcomes.

Source: https://resin-cities.eu/greatermanchester/.

Figure 12.3

Map showing where Salford is located in England, and Greater Manchester

In this setting, three separate but related studies have analysed the relationship between community asset participation and individual health and well-being outcomes. Below, we outline the overall structure of the data and discuss variables of interest before outlining the methods used in each study and summarising the results. We finish by summarising the implications for policy based on this case study. In summary, the three studies we focus on are: 1. Munford et al. (2017), who examined cross-sectional relationships between community asset participation and health-related quality of life (HRQoL), self-reported health care utilization, and the combined ‘net benefit’. 2. Munford et al. (2020a), who examined the longitudinal benefits of starting to participate in community assets from a baseline position of non-participation. Individuals who started to participate in community assets were matched to a ‘control’ group of individuals who never participated, and their HRQoL, use of health care services (measured using linked administrative records) and net benefits were compared using double-robust techniques. Additionally, the symmetry of effects was assessed by comparing outcomes of people who stopped participating in community assets relative to a ‘control’ group of people who always participated. 3. Munford et al. (2020b), who examined the longitudinal benefits of starting to participate in community assets from a baseline position of non-participation. Individuals who started to participate in community assets were matched to a ‘control’ group of individuals who never participated, and their well-being and QoL were compared using propensity score matching.

186  Handbook on the political economy of health systems All three studies show a positive and statistically significant relationship between participation in community assets and health and well-being. Further, participation in community assets was shown to generate a substantial societal net benefit. 4.1.3 Data and variables of interest As part of the CLASSIC study, questionnaires were mailed to 12,989 individuals aged 65 years and older with at least one long-term health condition who were living in the Salford area between November 2014 and February 2015. These individuals were selected from the disease registers of 33 general practices within Salford. At baseline, 4,377 individuals (34 per cent) individuals responded with usable responses. These individuals were then sent follow-up questionnaires at 6, 12 and 18 months, resulting in four waves of data. At 18 months, responses were revived from 2,449 individuals (56 per cent of the baseline cohort). A flow chart showing response rates over time can be found in Munford et al. (2017, 2020a). 4.1.4 Community asset participation Individuals were asked in all four waves:1 ‘Have you attended or used any of the following community groups, activities and services in the last 6 months? (TICK ALL that apply)’ with these possible responses: ● ● ● ● ● ● ●

Group for the elderly or older people (e.g., lunch club) Education, arts, music or singing groups (including evening classes) Religious group or church organization Charity, voluntary or community groups Social club (including working men’s clubs, rotary clubs, etc.) Sports club, gym, exercise or dance groups Other group or organization

An individual was classified as a community asset participant if they ticked yes to one or more of the above. In each of the four waves (baseline, 6-month, 12-month, and 18-month follow-up) individuals were classified as ‘community asset participants’ if they reported attending any of the above groups, and ‘non-participants’ otherwise. Supplementary analysis also considered the intensive margin by examining the number of types of community asset that an individual reported participating in. 4.1.5 Health-related quality of life, health care utilization, and net benefit The questionnaire contained the EQ5D-5L, the five-level response option of the EQ5D (Brooks et al., 1996; Herdman et al., 2011). The EQ5D is a generic preference-based measure of HRQoL covering five domains (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression). Individuals in the baseline, 6-month, 12-month, and 18-month follow-up questionnaires completed the EQ5D-5L. Responses were converted to a single index utility value based on the crosswalk mapping tool of Van Hout et al. (2012), which maps from the five-level questionnaire onto the three-level questionnaire. This crosswalk tool is the National Institute for Health and Care Excellence’s (NICE’s) preferred method of obtaining utility values from the EQ5D-5L. In robustness analyses, the alternative algorithm for directly calculating utility scores from the EQ5D-5L was used (Devlin et al., 2018). Quality-adjusted life

Community and civic participation effects on health and well-being  187 years (QALYs) were calculated at the individual level using the area under the curve method assuming linear extrapolation of utility between time points (Hunter et al., 2015). In addition to information relating to EQ5D-5L, individuals in CLASSIC were asked about their health care utilization (Munford et al., 2017) and were also matched to their administrative health records using NHS Numbers, unique identifiers that allow individuals to be linked across various aspects of formal health and care services. Munford et al. (2020a) exploited this administrative linkage rather than self-reports as this allowed the construction of detailed information on the use of primary and secondary health care services (i.e., information on the use of both general practitioner (GP)/family doctor services and hospital services). Individual-level health care resource utilization over the study period was collected from two sources. The number of GP contacts in the previous 6 months was collected from electronic primary care databases. Hospital utilization was extracted from linked administrative patient records provided by the NHS, divided into emergency admissions (short stays ≤5, long stays >5 days), elective admissions, elective day cases, outpatient attendances, and accident and emergency (A&E) department attendances, as in Panagioti et al. (2018). These activities were then costed and assigned monetary values.2 A discount rate of 3.5 per cent was applied to the costs and benefits, as recommended by NICE. Finally, information on the effects of community asset participation on HRQoL and health care costs was combined to produce estimates of the societal value of a year’s participation in community assets using the net-benefit framework introduced by Stinnett and Mullahy (1998). The main threshold used was £20,000 per QALY, but other thresholds were used as robustness checks; see Munford et al. (2017) for further details. As well as considering health, Munford et al. (2020b) additionally considered the effects of community asset participation on QoL by using the Abbreviated World Health Organization Quality of Life questionnaire (WHOQOL-BREF), and its subdomains, as outcomes. The WHOQOL-BREF is an international, generic patient-reported outcome measure (PROM) developed by a collaboration of 15 cultures through the WHO that was designed for both healthy and sick people to self-report their QoL (Skevington et al., 2004). 4.1.6 Demographic and socioeconomic characteristics The questionnaire also collected a limited set of demographic and socioeconomic information, including gender and age (using a series of 5-year age categories, ranging from 65 to 69 years, up to 85+ years). The living situation of each individual was assessed, and responses included ‘live with spouse’, ‘live with other’ or the reference category ‘live alone’. Information on the following educational qualifications was also reported: ‘one or more Ordinary Level (O-Levels)/Certificate of Secondary Education (CSEs)/General Certificate of Secondary Education (GCSEs)’, ‘one or more A-Levels/AS-Levels’, ‘Degree’, ‘National Vocational Qualification (NVQ)’, ‘Trade qualifications’, ‘Professional qualifications’. An individual could tick multiple responses. The reference category was ‘no qualifications’. 4.1.7 Methods The data above was analysed in three separate studies by research teams based at the University of Manchester. The first – Munford et al. (2017) – used baseline data and cross-section analysis to explore the relationship between community asset participation and (1) HRQoL, (2)

188  Handbook on the political economy of health systems self-reported health care utilization, and (3) societal net benefits using multivariate linear regression models of the form: H i   CAi   X i   i where subscript i refers to individual and H is one of the measures of health, health care utilization, or net benefit as described above; CA is a dummy variable equal to 1 if an individual participated in community assets and zero otherwise; X is a vector of individual-level socioeconomic and demographic information, including, for example, age, gender, living arrangements, and education attainment, and ε is a standard error term. The initial analysis conducted in 2017 was cross-sectional and therefore is potentially susceptible to a number of possible biases. The subsequent analysis accounted for the longitudinal nature of the data by exploiting within-person changes. Munford et al. (2020a) considered the same set of outcomes but applied double-robust methods (Kreif et al., 2013).3 Double-robust estimation is a form of treatment effects estimator that accounts for observable factors that could influence treatment allocation; in this case the decision to start to participate in community assets. The method combines a propensity score model with a regression adjustment. The main sample comprised people who did not participate in community assets at baseline. The propensity score was obtained from a logistic regression of community asset participation in later waves on baseline covariates. The inverse of this propensity score was then used to weight the regression model for the outcome. As long as one model (treatment assignment model or outcome model) is correctly specified, the double-robust estimator produces unbiased results (Robins et al., 1994, 1995). If both models are correctly specified, then the double-robust estimator is both unbiased and efficient (Robins et al., 2007). More detail on the method, as well as the list of variables included in each model, is contained in Munford et al. (2020a). In additional analysis, the reverse of the above was considered (‘cessation analysis’), where the sample was made up of people who reported they did participate in community assets at baseline and ‘treatment’ was defined as those individuals who stopped participating in later waves compared to a ‘control’ group of people who always participated in community assets. In the third paper, Munford et al. (2020b) were interested in the effects of community asset participation on well-being and QoL. They used propensity score matching (PSM) to exploit the longitudinal nature of the data by considering within-person changes. Similar to Munford et al. (2020a), they considered both the uptake and cessation of community assets. More detail on the PSM method, as well as which variables were included in the model, is contained in Munford et al. (2020b). 4.1.8

Results of Case Study 1

Cross-sectional relationship between community asset participation and health outcomes (Munford et al., 2017) The multivariate regression analyses revealed that individuals who participated in community assets, relative to those who did not, had EQ5D scores that were 0.081 (95 per cent CI4 0.064 to 0.098) higher than non-attenders (Table 2; column 1 of Munford et al., 2017). When information on the presence of 23 specific health conditions was added to the model, this effect dropped to 0.063 (95 per cent CI 0.048 to 0.077). Controlling for whether the conditions were limiting reduced the effect to 0.039 (95 per cent CI 0.025 to 0.052). The authors compared this

Community and civic participation effects on health and well-being  189 effect to the size of the coefficients of a number of conditions which limited daily activity, and they showed that not attending community assets had an effect comparable with one-quarter of the size of having limiting back pain or osteoarthritis. The effect of diabetes was −0.053 (95 per cent CI −0.102 to −0.005), roughly 1.5 times the effect of community asset participation. The authors therefore argued that the effects of community asset participation on EQ5D scores were not trivial. The authors then estimated the effect of community asset participation on each of the five domains of the EQ5D. They show that participation in community assets increases the probability of reporting the highest level (‘best health’) for each of the five domains and decreases the probability of reporting the lowest levels (‘worst health’). The general pattern of results is similar across all five domains, with the greatest positive effect on usual activity, self-care, and mobility, and the smallest effects on pain/discomfort and anxiety/depression. When considering self-reported health care utilization, community asset participation led to a small and statistically insignificant reduction in utilization when the presence of health conditions was added to the models. Individuals who participated in community assets used NHS resources worth £75.17 (95 per cent CI −141.43 to −8.90) less than non-participants. However, this difference became statistically insignificant when information on underlying health conditions was added (a reduction of £33.61; 95 per cent CI −£91.10 to £29.88). However, when the authors combined the effects of benefits and costs into a measure of societal net benefit (by using the threshold of £20,000 per QALY as recommended by NICE), they estimated that community asset participation led to a positive societal net benefit equal to £763 (95 per cent CI 478.21 to 1047.75) per participant per year. Using alternative thresholds (such as £30,000 and £12,936) altered the point estimated but the societal net benefit remained large, positive, and statistically significant. Double-robust results of the relationship between community asset participation and health outcomes (Munford et al., 2020a) There was a positive and statistically significant effect of starting community asset participation on HRQoL. At the 6-month follow-up, this benefit was a 0.017 QALY gain (95 per cent CI 0.002 to 0.032) for individuals who started to participate in community assets compared with those who never participate. At the 12-month follow-up, the effect of starting to participate in community assets was a 0.030 QALY gain (95 per cent CI 0.005 to 0.054) and a QALY gain of 0.056 (95 per cent CI 0.017 to 0.094) at 18 months. The decision to start to participate in community assets reduced health and care costs in the 6-month period by £96 (95 per cent CI −£512 to £321), in the 12-month period by £283 (95 per cent CI −£926 to £359), and in the 18-month period by £453 (95 per cent CI −£1366 to £461). While these effects were in the direction expected (e.g., a reduction in costs), they were not statistically significant (at p.09)

Percentage private insurance

Gini

.0097

.245*

 

(.006)

(.108)

Notes:  * Significant at 5% level. Standard errors in parentheses.

there is strong evidence for substitution—with the magnitude of substitution particularly large in Latin America and the Caribbean, followed by Australia/New Zealand and European countries. There seem to be no countries where in aggregate private insurance complements public spending, though in some there is no statistically significant relationship. Hence the institutions and the behavior necessary to support a political process in which private insurance is tolerated or even subsidized seems present in most countries that currently have a significant presence of private insurance. These results reveal some patterns related to income distribution, with more substitution, which we explore explicitly in the next analysis. Table 19.2 shows the results from descriptive regressions with fixed effects that relate the private insurance percentage of spending in each country to its public insurance percentage. The results indicate significant negative relationships between the two measures, consistent with the hypothesis of substitution. The magnitudes are larger and closer to unity outside of Europe. If we hypothesize that average insured spending will be moderately larger for those who select the private sector, the coefficient values are about what we would expect. Given the results above, we now explore whether that presence was related to income and its distribution. Our primary hypothesis, suggested by the theory, is that there will be more opportunity for mutual gain by encouraging private substitute insurance in countries with a more sizeable segment with higher than average incomes, relative to countries in which income is distributed more evenly. Because the income distribution evolves slowly, we have little variation of the Gini coefficient over time, and therefore we focus our analysis on the cross-section of countries. Regression analysis of the presence of private insurance above the 9 percent threshold (a binary dependent variable), as well as the relative percentage covered by private insurance conditional on being above the threshold, were related to the Gini coefficient on post-tax household income, given the average gross domestic product (GDP) per capita. According to our results, income inequality, measured by the Gini coefficient, has a positive effect on the probability that the private sector is large enough (above 10 percent enrollment), although this coefficient is imprecisely estimated with a p-value of .109. It also has a positive impact on the percentage of health spending covered by private insurance. An additional point of income inequality, measured by the Gini coefficient, increases the percentage covered by private insurance by .245 percentage points. GDP per capita itself is inversely related to being above the threshold (the richest countries do not have much private insurance) but is unrelated to the size of the private insurance share for countries above the threshold.

Towards a general political economy of private supplementary insurance  333

5.

CASE STUDIES: US MEDICARE, AUSTRALIA AND NEW ZEALAND, CHILE, AND BRAZIL

5.1

Medicare in the United States

Persons over 65 and some other high-need patient populations are eligible for publicly financed care in the United States under the public Medicare program. The most common vehicle for coverage is a set of insurances (“Original Medicare”) financed by the federal government that cover inpatient, outpatient, and prescription drug expenses. Those insurances have deductibles and coinsurance requirements, the most important of which is 20 percent unlimited cost sharing on outpatient services. Private supplementary insurance (“Medigap”) is available to cover part or all of that cost sharing in a regulated setting. There is also a private Medicare Advantage voucher program that funds substitute private insurance. At present about 56 percent of the population takes Original Medicare, and of those not eligible because of very low income for Medicaid supplemental insurance, about 73 percent take Medigap. Only about 11 percent of those who are not poor take Original Medicare, the collectively chosen amount, as their only coverage; the others either take private Medicare Advantage or Medigap. Research shows that Medigap functions as a complementary insurance to Original Medicare in that people who take it have larger claims in Original Medicare than those who do not (Cabral and Mahoney, 2017). Medicare Advantage in contrast operates as almost a perfect dollar-for-dollar substitute, but such plans do add additional benefits, sometimes at additional premiums, and do cap total out-of-pocket spending. Both Medigap and Medicare Advantage are obviously supported by those who choose them, but their existence also appears at present to be a political equilibrium. Taxing Medigap premiums to reflect the increased spending it causes has been proposed but not enacted. These facts appear to be inconsistent with a median voter model since the median voter is not eligible for Medicare but must pay taxes to support both the increased spending in Original Medicare and the higher voucher value that generates for Medicare Advantage. One possibility is that voters expect in the future to benefit from this set of options (adjusted for survival to age 65), which could explain support for easy-to-supplement Medicare Advantage. Support for Medigap still poses a problem, however, since the current arrangement would be dominated by a tax on supplementary coverage whose proceeds are then used to reduce the premiums seniors must pay for outpatient and drug coverage. Perhaps the status quo is preferable to attempting the complex political bargain to lead to more appropriate cost sharing in Medicare. 5.2

Australia and New Zealand

Public policy in Australia encourages the substitution of private coverage for public spending. While supplementation of the intensity, timeliness, and amenity of the public system requires additional spending, that spending is encouraged by a combination of subsidies for those who do opt out and taxes on remaining in the public system for higher-income households that do not. About 40 percent of Australians have private coverage. The public system has relatively low cost sharing (in contrast to an otherwise similar system in New Zealand), so private coverage does not directly stimulate use of the public system. Easier access to testing and other

334  Handbook on the political economy of health systems services that may lead to high-cost episodes of public care appears to be offset by a substitution effect according to several research studies. Empirical analysis has generally found a cost offset on public spending for those who elect private health insurance in Australia, after controlling for other determinants of spending and for self-selection (Doiron and Kettlewell, 2018). While the calculations are not precise, it does appear that the cost of tax-financed subsidies just about equals in total the cost offset in the public system at the current rate of subsidy and take-up (Cheng, 2014). This observation suggests that at the margin the subsidy may not quite pay for itself so may not be set at the level that maximizes gain to the decisive voter. To our knowledge, there has been no analysis of how spending or amenity levels in the public sector are affected by varying amounts of private substitution. Public support for subsidized private coverage in Australia has been fairly strong, though left-of-center governments are less supportive of it. There is concern that private insurance may be subject to adverse selection largely based on age, which may challenge its market share in the future. Right-of-center governments have tended to favor increased subsidization as a way of maintaining the private share. As noted, the system of public provision in New Zealand is similar to that in Australia except that cost sharing is somewhat higher for outpatient or day-hospital care. Although private insurers in New Zealand have advocated for subsidies as a reward for opting out, research by Linda Blumberg (2006) suggests that the net effect of private coverage on public spending in cross-sectional analysis is near zero or slightly positive. There has been no strong policy advocacy for changing private health insurance coverage in New Zealand, though private insurers have advocated Australian-style subsidies and progressive politicians higher public sector spending. The private insurance systems in these two countries thus appear to be at levels of support roughly consistent with the median voter model. Ideological discomfort with unequal wait times or amenity levels could lead to changes, however, but only at the cost of more tax cost for the public system—but probably improvements in its quality as well. 5.3 Chile About 20 percent of the population in Chile opts out of the public insurance (FONASA), and enrolls in a private insurance plan (Isapre), for which they pay the full cost. In FONASA, people contribute 7 percent of their income up to a cap and obtain a unique plan, with different copays depending on income, where the lowest-income people have no copays, and the highest 20 percent. Given that coverage in the public system is always the same, the price of insurance is increasing as income increases, because the contribution of 7 percent of income is larger for higher earners, which encourages higher-income people to opt out of FONASA, thus avoiding the wage-based contribution, and enroll in a private plan at a premium that does not vary with income. A large number of different private plans are offered, whose premiums are risk-rated. People that opt out of the public system can take their contribution of 7 percent and use it to buy private insurance in an Isapre. In that sense, people who opt out of public insurance do not pay double, but take their contribution with them to buy a private plan. The premiums in the private sector are determined by risk, and are fully funded by the beneficiaries, who in addition to the 7 percent contribution, contribute an additional 3 percent of income on average. On the other hand, the premiums paid in FONASA fund about one-third of the

Towards a general political economy of private supplementary insurance  335 benefit, and the other two-thirds come from general taxes. Consistent with the analysis shown in Section 4, the private option was introduced in 1981, a time at which income inequality was larger than now, with a Gini coefficient above 0.56 (1987 is the oldest data available), compared to 0.43 in 2015. The percentage of people that opt out of FONASA has been fairly stable over the last 20 years, suggesting there is an equilibrium. That percentage was larger and above 25 percent in the early 1990s; at that time, and since 1986, the government provided a subsidy of 2 percent of income to low-income workers that wanted to opt out of the public system. This is very similar to the subsidies provided in Australia that were discussed in the previous case study. This subsidy was removed in 1999, and about 900,000 lower-income Isapre beneficiaries went back to the public system. Among the reasons that were given to remove the subsidy were the fact that it was poorly focused as it did not reach the lowest-income people, and that its design was regressive, as the 2 percent of a relatively higher income corresponded to a larger subsidy amount. Although these are true, the design of the policy could have been improved instead of repealing the subsidy. However, a more ideological view prevailed, generating a more inefficient outcome, as the evidence provided in this chapter has shown. A person who switches from FONASA to an Isapre reduces the cost paid in the public system but also reduces revenues to that system by the 7 percent contribution, which is about one-third of the total tax financed cost of that system, the rest being financed by general revenue taxation. Hence a lower-income person who leaves provides more benefit to the public system than a higher-income person, and this has been used as an argument to shut down the private sector to improve quality in the public system with the additional contribution. However, a large part (55 percent) of the 7 percent contribution goes to fund sick leave and not to health care; therefore, some recent calculations show that the policy of abolishing Isapres adds only an additional 10 percent of resources to the public system for an additional 20 percent of people. Other policy changes have been gradually introduced in the Chilean private system that contribute to more instability in the industry. Incremental steps towards community rating discourage healthy lower-income Isapre beneficiaries from participating in this market and encourage them to go back to the public system. In addition, the regulator has made it hard for Isapres to update their premiums, which in a context of increasing health care cost threatens their profitability and viability. At the time of writing this chapter, a new national constitution was being written, and it was not clear whether the option to opt out of the public sector and get insurance in the private sector would survive. 5.4 Brazil With the enactment of Brazil’s new constitution in 1988, health was defined as a universal right, giving rise in 1990 to the Unified Health System (Sistema Único de Saúde, SUS). The SUS has made substantial progress towards universal coverage, being the default option for all citizens and foreigners, including undocumented immigrants. It is funded by progressive income taxes, and therefore everyone contributes. There are important bottlenecks in the SUS, which prompts higher-income individuals to purchase private health insurance. In Brazil, private health insurance is supplemental to the SUS; in other words, people with private insurance continue to pay their taxes to fund the SUS and maintain their eligibility for the public SUS benefits, and pay extra for private health insurance premiums. The large

336  Handbook on the political economy of health systems majority of private health insurance in Brazil (90 percent) is employer-provided, with only 10 percent of insured with individual policies. About 25 percent of the population is covered by private health insurance, which strongly correlates with income: about 10 percent of people in the lowest quintile of income have private health insurance, whereas in the highest quintile 80 percent of people hold a private health insurance policy. The coverage in private health insurance is comprehensive, and therefore it duplicates the SUS coverage—but it provides more timely access to private facilities. For some complex procedures such as cancer care, people with private insurance might decide to use public facilities. In theory, the cost of that care should be covered by private health insurance; however, the information systems that integrate public and private utilization are still a work in progress, which means that in practice, private insurers pay a small fraction of utilization in the public sector. This is a contentious issue that is currently under review by the Brazilian Supreme Court. The premiums for private health insurance, and individuals’ spending on health care, are subject to a tax rebate (similar to Australia); in addition, health benefits provided by firms can be deducted from their taxable profits. The size of these tax breaks is considerable, which, according to Ocké-Reis and Gama (2016), correspond to 22 percent of total federal expenses in the SUS. These important tax rebates are indicative of a voting equilibrium in which high-income individuals are encouraged to take private insurance.

6.

CONCLUSIONS FOR HEALTH POLICY: IRONIES ABOUND

The political economy of private health insurance shows as usual that efficiency and equity often conflict, but it differs from the standard story in that what is more to the benefit of voters with lower incomes is often not obvious. One small but familiar ironic inefficiency is the analysis of redistribution in kind rather than in cash, as Gouveia (1997) had noted. Some lower-income people will prefer higher to lower health spending, but would prefer more money income to higher health spending. However, we conclude that for people with middle and upper incomes this situation will not occur because they will want to—and may succeed in trying to—add private spending on top of the exogenous public supply of private goods spending, so that public provision only functions as an income transfer. Low-income people might prefer money to higher medical spending that sets a lower bound to their ability to convert care into cash. Another potentially more important irony that we have identified in this chapter is that, despite continuous discussion of the harms from private insurance in the global public health and policy community, such insurance seems surprisingly stable in the countries where it has typically had a major presence. To a considerable extent private insurance is created by heterogeneity in preferences driven in some countries by a more uneven income distribution—but even in those countries so far the poor seem unwilling to outvote the rich to change that income or wealth distribution and thus destroy the support for private insurance. However, the greatest irony is that, despite the apparent inequity and unseemliness of private supplementation or substitution, an arrangement that allows those with means to get earlier access to higher-quality and higher-amenity care, such an institutional arrangement may actually make lower-income people better off than forbidding it. Abolishing private insurance may hurt the poor, and this paradox can be perplexing. Richardson (2009), for example,

Towards a general political economy of private supplementary insurance  337 decries the Australian scheme of permitting and even encouraging private insurance for the rich as reflecting objectionable and selfish distributional preferences in Australian society, but then recognizes the positive fiscal impact of private substitution on the public budget. The seemingly interminable controversy in the Canadian province of British Columbia over a physician who wants to offer a private insurance plan for delayed elective surgery, so far criticized by the courts as “un-Canadian,” paradoxically could permit more access to timely surgery to both rich and poor. As usual the outcome will depend in part on the behavioral facts about extent of substitution, slope of supply curves, and distribution of incomes and preferences, but also on the extent to which self-interested voting trumps ideology. There are some remaining puzzles. We do not have a plausible political economy explanation for the persistence in some countries of an institutional structure for the public sector that imposes cost sharing but then removes incentives to control moral hazard by permitting easy supplementation with private insurance to cover that cost sharing, and with private insurance largely purchased by those with higher incomes. The other puzzle is whether the phenomenon of many high-income countries, especially in Europe, with near-zero private insurance and high public spending, is politically stable. As long as the underlying more even distribution of income and wealth is stable, probably this system will persist, but if income becomes less equally distributed among voters (for example, because of immigration) things could change. Fundamental country-level value judgments about equity of resource distribution are most crucial. Private insurance provides the most additional and mutual benefit as a kind of corrective to a setting of unequal distribution of income combined with uniform public provision of the private good, medical care. Hence it has a role to play in some settings but not necessarily in all.

REFERENCES Besley, Timothy, and Stephen Coate. 1991. “Public Provision of Private Goods and the Redistribution of Income.” American Economic Review 81(4): 979–84. https://​econpapers​.repec​.org/​article/​aeaaecrev/​ v3a83ay​3a19913ai3​a43ap3a979​-84​.htm. Blumberg, Linda J. 2006. The Effect of Private Health Insurance Coverage on Health Services Utilisation in New Zealand. Wellington: Fulbright New Zealand. www​.fulbright​.org​.nz/​publications/​ 2006​-blumberg/​. Buchanan, James M., and Mark V. Pauly. 1970. “On the Incidence of Tax Deductibility.” National Tax Journal 23(2): 157–67. Cabral, Marika, and Neale Mahoney. 2017. “Externalities and Taxation of Supplemental Insurance: A Study of Medicare and Medigap.” Working Paper 19787. National Bureau of Economic Research. Cheng, Terence Chai. 2014. “Measuring the Effects of Reducing Subsidies for Private Insurance on Public Expenditure for Health Care.” Journal of Health Economics 33(January): 159–79. Doiron, Denise, and Nathan Kettlewell. 2018. “The Effects of Health Insurance on the Substitution Between Public and Private Hospital Care.” Economic Record 94(305): 135–54. Epple, Dennis, and Richard Romano. 1996a. “The Private Supply of Public Goods.” Journal of Political Economy 104, 57–84. Epple, Dennis, and Richard Romano. 1996b. “Ends Against the Middle: Determining Public Service Provision When There Are Private Alternatives.” Journal of Public Economics 62(3): 297–325. Gouveia, Miguel. 1997. “Majority Rule and the Public Provision of a Private Good.” Public Choice 93(3): 221–44. Ocké-Reis, C.O., and Gama, F.N.D. (2016). Radiografia do gasto tributário em saúde, 2003–2013. Nota Tecnica, IPEA, Brasilia.

338  Handbook on the political economy of health systems Richardson, Jeff R.J. 2009. “Steering Without Navigation Equipment: The Lamentable State of Australian Health Policy Reform.” Australia and New Zealand Health Policy 6: 27. Thomson, Sarah, Anna Sagan, and Elias Mossialos. 2021. Private Health Insurance: History, Politics and Performance. Cambridge: Cambridge University Press.

20. Political economy of public financing of health in low- and middle-income countries Sumit Mazumdar and Rodrigo Moreno-Serra

1.

INTRODUCTION: WHAT IS POLITICAL ABOUT HEALTH SYSTEM FINANCING?

A core ‘building block’ of health system functions (World Health Organization, 2007), the financing of health care services in any country involves crucial public policy choices. Governments across the world have traditionally been involved to varying extents in determining and implementing policies related to health, and have played a major role in shaping these. Political systems significantly influence how governments function in the realm of social policies, and how citizen voices are reflected in these policy decisions. Together with the historical evolutionary pattern of sociocultural and economic systems in any country, the political system also influences how institutions crucial to public policymaking and implementation are developed, function and transform. Policy decisions on public financing of health care determine how financial resources for the health system are raised and distributed across different health services and functions, catering to the needs of different population groups and economic organisations. As this affects how different sections of society and the economy benefit in different ways, the relative power and influence of these competing groups often determine the nature and content of these policy choices. Political determinants of public policy embody these crucial aspects of relative power, interests and incentives, which are inherently associated with policy choices, design and their implementation. The degree of political will and the existing incentives for setting and prioritising a reform agenda, along with the institutional capacities to implement it, are key to the success of any policy measures. It is hard to imagine any meaningful analysis of health system financing policies in any country without considering such key influences. The emergence of universal health coverage (UHC) as a fundamental aspirational goal for health systems worldwide has further augmented the influence of political factors for health financing policies and programmes (Fox and Reich, 2015; Greer and da Fonseca, 2016; McDonnell et al., 2019; Reich et al., 2016; Rizvi et al., 2020). UHC has been acknowledged to be a predominantly political choice, primarily because it involves difficult decisions on what services are covered to ensure equal access to all and protection against financial risks, subject to the resource constraints countries face (Ghebreyesus, 2019). With considerable variations across countries, these decisions have been largely dominated by debates on rethinking the public role in health financing and its wider implications across other social and economic sectors. For example, in most high-income countries (HICs) with public-sector-dominated health financing systems, the key questions involve sustaining the higher fiscal commitments that a move towards universalisation would entail in the face of spiralling health care costs, to balance such outlays with the conflicting demands across other social and economic sectors, without any significant alteration to the ‘equilibrium’ tax structure (Alesina and Perotti, 339

340  Handbook on the political economy of health systems 1995; Ghosh et al., 2013). At the other extreme, in most low- and middle-income countries (LMICs), commitments towards UHC necessitate a supportive health financing transition – higher public spending on health and a greater, proactive stewardship role of governments and ministries of health in financing to expand access, reduce out-of-pocket (OOP) payments and consequently mitigate risks of financial impoverishment. Often, this involves remarkable course correction in the relative roles of key health system players, particularly where the private sector predominates, hence bringing further interests around market shares and profit maximisation to the policy discussion arena. Models of political economy and institutional analysis provide useful insights for understanding the priority-setting mechanisms adopted by decision-makers and the effective implementation of these policy decisions, which must balance the diverse, often conflicting, interests of the different stakeholders (Goddard et al., 2006). As we discuss below, within the broader context of health policy choices aligned with UHC objectives, the political economy dimensions of health financing reforms relate to how the power to influence these decisions, and their expected impacts, vary between the different groups of actors. The balance of power between these decision-making and other interest groups is shaped by prevailing political, economic, institutional and sociocultural systems in any country and change over time. Understanding these processes and their underlying drivers is critical to assess the feasibility of any reforms, and to ensure their success and sustainability.

2.

THEORIES AND APPROACHES ON THE POLITICAL ECONOMY OF HEALTH POLICY

Standard theoretical approaches on the political economy of public policy typically begin by identifying a set of key actors relevant to the policy under question, followed by analysing the distribution of power they exert through different stages of the policy process, along with the varying incentives they face within alternative scenarios for action. For some of these actors, political or ideological affiliations are useful to explain the choice between alternative actions. The relative ease or difficulty of designing and implementing any specific policies, and often the likelihood of their effectiveness, can usually be explained in no small part by the prevailing institutions in any setting. Opportunities for adopting and pursuing any policy decisions, the optimal strategies that contribute to different stages of the policy process, the nature of the political regime or competition between rival political groups are all key ingredients in understanding the political economy of health policy. 2.1

Political Foundations of Policymaking

A useful starting point in any political economy analysis for health policy is to identify the key foundational or explanatory variables that influence different stages of the policy process. While alternative views can be found in the political science literature, there is general consensus on the heuristic relevance of the ‘four Is’: interests, institutions, ideas and ideologies (Fox and Reich, 2015; Reich, 1995; Shearer et al., 2016). Interests respond to the ‘who’ questions, by identifying the different stakeholders and the interests they represent – either as individuals or as collective groups – who stand to be affected by the policy decisions or its impacts. Interest groups represent their constituents and

Political economy of public financing of health  341 are considered as legitimate channels for influencing policy. Interest group theories help to understand how power is distributed across the different groups and how such relative power is exerted, considering how the policy reforms would affect these groups and their constituencies (Dür and de Bièvre, 2007; Flöthe, 2020). Accordingly, stakeholder mapping, and the analysis of relative power and interests across the key groups and individuals relevant to any policy, are useful and popular starting points for empirical political economy analysis (Bianchi and Kossoudji, 2001; Nash et al., 2005; Reich, 1995). In their most general conceptualisation, institutions have been defined as ‘the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interaction’ and ‘structure incentives in human exchange’ (North, 1990, p. 3). These constraints or rules primarily exist to reduce uncertainty in the interactions, provide stability and define legitimate incentives to the interacting agents or groups (Acemoglu et al., 2005). Institutions also provide a collective framework of how citizens interact with the state to negotiate how public resources are allocated or public services are provided.1 Institutions include both economic (e.g., property rights, contracts) as well as political types (e.g., forms of government, division of powers between different arms of the government such as the executive, legislature and the judiciary); these can also include both formal or codified rules as in most constitutional provisions, and informal rules involving social norms, cultural practices and customs. Understanding how institutions have evolved, their composition and how they influence different stages of policymaking is of pivotal importance, particularly (for example) in comparative analysis of why similar health policies implemented in alternative settings produce different results. Ideas and ideologies are related but distinct political concepts. Ideas include theories, conceptual models, norms, world views, frames, principled beliefs which influence and affect policymaking (Campbell, 2002). In discussing the role of ideas on how policies are developed, Campbell (1998) distinguishes between four types of ideas, namely programmes, paradigms, frames and public sentiments, depending on whether they act at the cognitive or normative level behind policymaking, or whether they influence policy by serving as either the background or the foreground of policy debates. Researchers have also considered how ideas interact with interests representing specific identities; for example, to determine political actions by pharmaceutical manufacturers on issues such as flexibilisation of regulatory norms for drug manufacturing or pricing (Reich, 1995). Finally, ideologies serve as the bedrock of political views that are relevant to any policy, encompassing ‘a broad range of beliefs about social reality’(Weber, 2019). They embody political viewpoints, usually differentiated as right- or left-wing, but also with related connotations such as conservatism, liberalism and populism. Ideologies help to justify or contradict policy decisions depending on political or philosophical views and opinions. In the health policy arena, historically one of most ideology-influenced debates has centred around the role of the public versus private sector in financing or providing health care. While neoliberal economic ideologies – associated more with pro-right-wing political ideals – have defended a greater role of market-based reforms and a lesser role of the state, so-called leftist political ideologies have underpinned policy decisions that favour increased public participation in the financing and provision of health care, which are usually justified by pointing to perceived deficiencies of market-based approaches to guide health policy decisions (Navarro, 1977).

342  Handbook on the political economy of health systems 2.2

Four Stages of Policymaking

A common framework outlining the process of policymaking includes four key stages: problem identification and issue recognition, policy formulation, policy implementation and, finally, policy evaluation (Buse et al., 2012). Kingdon (2014) maps these stages into a set of distinct processes, namely: setting the agenda, specification of available policy alternatives, an authoritative choice among these alternatives and implementation of the decision. Others have laid out these stages more generally as agenda-setting, designing the policy (particularly the technical aspects), policy adoption and implementation (Fox and Reich, 2015; Jann and Wegrich, 2007). We briefly discuss these stages below, identifying the political determinants relevant to each of these stages. 2.2.1 Setting the policy agenda What explains whether certain issues such as UHC or social protection receive greater attention from national leaders and policymakers than, say, housing or improving infrastructure such as roads? Why and when do political parties include promises for a new or revamped health insurance programme in their election manifestos? Answering these and similar questions requires understanding how an issue emerges and occupies importance on the policy agenda. An agenda involves ‘the list of subjects or problems to which government officials and people outside the government closely associated with these officials, are paying some serious attention at any given time’ (Kingdon, 2014, p. 3). Different theoretical explanations exist on how any issue gets to claim a place in the policy agenda, but the general consensus is that problems need to be (a) socially defined and (b) politically supported through different forms of political mobilisation (Birkland, 2015; Fox and Reich, 2015; Rochefort and Donnelly, 2013), thereby hinging crucially on the competitive negotiating powers of concerned agents, groups and organisations. An often-cited model of the process of agenda-setting is Kingdon’s multiple streams approach, which invokes an aquatic metaphor: as in a river receiving water from different streams, the agenda that gets accepted (or ‘flows’) necessarily requires a simultaneous supply from three streams (Buse et al., 2012; Kingdon, 2014). The problem stream refers to the existence of a clear, well-defined, specific and articulated problem for the decision-maker. Media – both traditional and the ‘new’ (or social) media – can play a significant role in highlighting some problems and drawing some legislative attention. The politics stream includes organised interest groups as well as other less-evident agents, such as academic researchers, who play active roles in contributing towards both the visibility or championing of a problem, and highlighting potential policy solutions. Finally, in the policy stream the decision-maker selects the policy agenda from the problem and politics streams based on a combination of different viability criteria, including economic and technical feasibility and anticipated public acceptability.2 The problem, politics and policy streams have ‘lives of their own’; in most cases they exist independently, and concerted policy action remains elusive. Only when these streams come together – through the opening of some ‘window of opportunity’ – do issues get translated into agendas that are taken up by the policymakers. Political or regime changes, economic shocks or similar crises often provide such windows of opportunity. It is important to note that agenda-setting does not essentially follow a sequential process, graduating from identifying policy problems to then seeking viable policy solutions. The multiple streams model suggests

Political economy of public financing of health  343 that streams can exist in parallel for a long time while awaiting an opportune moment of convergence that leads to the development of a policy (Buse et al., 2012). Moreover, there is no guarantee that events such as regime changes or economic shocks will necessarily bring about health or other social sector reforms. In a number of LMICs, recent health system reforms have originated through diverse and often unique contextual channels, including policy decision-makers coming from a technocratic background (Mexico, Turkey, Taiwan) or rights-based movements (El Salvador, Brazil) (Clark, 2015; Cotlear et al., 2015). 2.2.2 Influencing the design and content of policies Similarly to agenda-setting, policymakers’ decisions around the technical design of policies are influenced by different actors both within and outside the government, as well as both domestic and international groups of actors. For most LMICs, it has been appealing to adopt policy reform models that have been proven to be successful or popular elsewhere, but other reasons have been advanced for such externally influenced policy diffusion. Fox and Reich (2015, 2013) review the literature on policy diffusion and identify four primary explanations for how foreign models get adopted: influence of external pressure such as donor countries, or international financial institutions such as the US Agency for International Development and the International Monetary Fund (IMF); ‘normative imitation’ or imitation of policies adopted by pioneering countries including those of regional similarity (e.g., Western Africa or East Asia); rational learning, from the analysis of information available internationally about alternative policy models, with their relevance and efficacy; and ‘cognitive heuristics’ for when detailed information (or time to assess existing information) is unavailable, consisting of reliance on some existing beliefs and trust in other models, which may or may not have a past evidential basis. Deciding on technical aspects of policy reforms can also be carried out by technical experts usually having some affiliation with the ministries of health, or other related government departments. Technical experts can operate either independently or in a supportive role to policymakers, once broad decisions on the reform model have already been made. These technocrats have been noted to lead ‘change teams’ comprising of technical specialists, bureaucrats and other experts who are familiar with the technical aspects as well as the politico-administrative mechanisms through which policies are developed and implemented (Berman and Bossert, 2000; González-Rossetti and Bossert, 2000) These change teams – which often include members with university education and professional training in universities of Europe or North America – establish critical bridges between the core decision-making groups for a given policy reform, as well as with other stakeholders essential to build consensus on the reform contents (Campos and Reich, 2019). Domínguez (1997) identifies an interesting group of ‘technopols’ who combine a technocrat’s technical skills with a politician’s pragmatism and experience to deliver feasible and workable change solutions, usually in response to a specific problem. Change teams have been found to play instrumental roles in health financing reforms in several LMIC contexts including Turkey, Mexico, Chile, Colombia and Uganda (Berman and Bossert, 2000; Roberts et al., 2003). The recent emergence of health benefit packages (HBPs) as a key component of health financing reforms in several LMICs has underscored the importance of political determinants for decisions on which services are included under the HBPs, and under what terms (Glassman et al., 2016). For example, similar to the experience of financing reforms in Latin America, where the neoliberal ideologies of technocrats have been argued to represent a major reason

344  Handbook on the political economy of health systems for a greater role of the private sector in social health insurance programmes (Giedion et al., n.d.; González-Rossetti and Bossert, 2000), political and/or ideological affiliations of the technocrats and consultants have been suggested as a primary reason for the inclusion of a wide array of services in the Ghanaian National Health Insurance Scheme (NHIS) (Rajkotia, 2009). As in many LMICs local technical expertise remains scarce for evidence-based priority-setting exercises – which are fundamental to developing cost-effective HBPs – decisions on HBP contents are more likely to be driven by either external influences or preferences of local bureaucrats and politicians. 2.2.3 Policy adoption Any policy or reform agenda, even if with an agreed design and content, still needs to tackle the often-convoluted paths of adoption, navigating opposition and devising acceptable agreements. Similar to the earlier stages in the policy process, interest groups and political institutions are of great influence during policy adoption, often through introducing different veto points. An influential theoretical approach in political economy, veto point theories invoke the role of veto players – individual or collective actors whose agreement is required for a change in policy (Tsebelis, 1995) – in supporting or resisting any changes from the status quo. Key actors here are the interest groups, such as professional associations or trade unions, that seek to protect the interests of their members in the course of a policy change, attempting to influence public policy but not seeking political power (Buse et al., 2012). All political systems define political institutions tasked with devising arrangements by which political representatives can veto legislative decisions that formalise policy adoption. In a seminal analysis on the role of veto points in policy analysis, Ellen Immergut (1990) suggests that this involves a chain of decisions across different ‘interconnected political arenas’ (e.g., different levels of the political or party executives), where agreement is required at several points in this decision chain. Interest groups or veto players influence policy adoption through their influence on political representatives situated at these veto points (Immergut, 1990). Veto points refer to these ‘junctures in legislative and policy-design process where reform can be blocked’ (Fox and Reich, 2015, p. 1039), with their location and number determined by the institutional and political arrangements in a given jurisdiction. A higher number of veto points or players increases stability of existing (or status quo) policies; new policy reforms are likely to face more resistance towards adoption. Relative access of the interest groups to the veto players and political decision-makers, along with their respective powers, determine which groups can hinder or expedite the policy adoption process, by either reducing the number of veto points, bypassing some of them or ensuring a faster progression across the decision chain. While interest groups are key actors across the veto points, partisan politics and electoral competition also have significant influence on the process of policy adoption. The existence and extent of such influence depends on the nature of the government – parliamentary or presidential, single-, two- or multi-party systems – or having a unicameral or bicameral legislative system. In multi-party parliamentary systems with multiple dominant political groups, electoral competition tends to be higher and the role of partisan politics greater in influencing policy reforms, as compared to presidential or single-/two-party systems. In multi-party systems – in addition to the existence of veto points and the influence of interest groups – politicians can choose to be more cautious in their policy reform decisions, as they are simultaneously driven by the objective to ensure acceptability of these decisions by their electorates and not jeopardise their re-election prospects. Public opinion or demands, ideological affiliations,

Political economy of public financing of health  345 voter loyalties, clientelist or patronage politics, and intra-party discipline also affect policy adoption in various ways, including by introducing further veto points, and can influence how politicians vote for or against policy reform bills in legislatures. Leftist political parties are more likely to pose strong veto points or resist market-driven reform strategies, while for a more centrist political party such veto points are unlikely to exist on ideological grounds, but exceptions do exist. In Mexico, the Seguro Popular (SP) social health insurance reforms, which have been credited with significant expansion of financial risk protection among the poor (Chemor Ruiz et al., 2018), were introduced in 2003 by the conservative, centre-right Partido de Acción Nacional (PAN), but resisted by the left-wing Partido Revolucionario Institucional (PRI). Upon being re-elected to office in 2018, the new leftist government eliminated SP and introduced a new health insurance programme (INSABI, launched January 2020), contending claims by the predecessor government about universal coverage achieved under SP and increasing the role of the private sector under the INSABI reforms (Reich, 2020). Veto points can also involve different government departments or ministries relevant to implementation of policy reforms, particularly when such reforms are multisectoral in nature. Ministries of finance are natural veto player candidates for health financing reforms, but as shown in the case of Turkey’s Health Transformation Plan (HTP, introduced in 2003), this can also include other ministries such as labour and social security, branches of the executive such as the president’s office, and judiciary branches such as the Constitutional Court. In the Turkish HTP context, reform adoption was facilitated by long and careful negotiation with each of these institutional veto players, conducted by the health minister and his advisors (Sparkes et al., 2015, 2014). Partisan politics and legislative bargaining between different veto players and political representatives, particularly when there are more veto points denoting these critical decision ‘junctures’, are also more likely to accommodate policy decisions that preserve – or even augment – fragmentation in health system financing. This is not surprising because each powerful constituency, through use of its veto powers, is likely to resist unified systems (such as a single-payer health insurance system) that might be interpreted as relinquishing any status quo privileges, in favour of retaining fragmented, multi-payer systems or financing pools. In India, a subsidised social health insurance programme targeted for informal sector workers (and the poor) was introduced for the first time at a nationwide scale in 2008 by a coalition government led by a centrist party (Indian National Congress, or INC), but having the critical support of leftist parties. This health insurance scheme was expanded in a newer programme introduced in 2018 by the right-wing majority government (led by the Bharatiya Janata Party, or BJP). Neither programme accommodated the inclusion of other funding pools meant for civil servants and blue-collar formal sector workers, which operate under parallel payer arrangements, mostly due to strong veto points from the administrative bureaucracy and trade unions (see the case study below). 2.2.4 Implementing policies Even after a policy or reform agenda graduates from the conceptual stage through to decisions on its adoption, there is no straightforward guarantee that it will be implemented, including in its agreed form and contents. Caution has been advocated against ‘top-down’ considerations of policy implementation as a natural sequence of earlier stages in the policy process, and considering it as a largely managerial or administrative exercise (Buse et al., 2012). While policies can fail to take off for the want of adequate technical or managerial know-how, there is still

346  Handbook on the political economy of health systems an important influence of political factors – specifically regarding institutional characteristics and performance relevant to policy implementation – that needs to be accounted for while planning a policy or reform rollout (and later, for evaluating its performance). Fox and Reich note three distinct political economy dimensions that affect policy or reform implementation: structure of political institutions and available political time horizon, the politics of targeting and exclusion, and unintended consequences during implementation (Fox and Reich, 2013). The structure of political institutions and governance arrangements are of prime importance in influencing how easy or difficult it is to implement a policy. In federal or decentralised systems, differences in political affiliations – different parties being in power at the national and provincial or local levels – can create various roadblocks for implementation. This may include hampering adequate flow of resources, both financial as well as manpower or other key inputs; varying level of effort by local functionaries; and interest or support from local political elites (Cleaves, 1980; Grindle, 1980). Elements of discretion allowed in the policy legislations to local agents can also affect implementation, particularly when these agents have rent-seeking incentives related to own effort control, selection of beneficiaries or other forms of patronage, or want to satisfy local elites and politicians by altering original policy features. A rich literature underscores how nepotism and corruption following from such undesirable incentives can significantly affect policy implementation (see, for example, Hutchinson et al., 2019, and several other articles in response in the same issue of this journal). Policy implementation is also influenced by time horizons available to the political incumbent, which can act in different ways. Most health system reforms, including on financing, are time-intensive. Initial bottlenecks in implementation and scaling-up policy interventions imply that politicians need to allow for a prolonged gestational phase in most cases, before being able to leverage on their success for electoral dividends. At the same time, politicians and their parties may also choose to back longer-term policies which once implemented would be difficult to reverse by newer incumbents among their competitors. Finally, if reforms such as new or expanded HBPs enjoy popular support from the public or in the media during the agenda-setting or design processes, opposition politicians might find it difficult to resist its adoption and implementation. Diversions in the policy’s original contents and designs, particularly on elements related to the targeting of local communities, can also be politically motivated to influence ‘swing’ or opposition voters, or as an instrument of consolidating elitist or clientelist control (Bardhan and Mookherjee, 2017, 2012; Khemani, 2015) and penalising opponents. In health financing reforms, such as the introduction of health insurance programmes, this can involve including beneficiaries who might be otherwise ineligible for coverage under the programme, or withholding information on participation and benefits from potential beneficiaries who have different political loyalties. Redistributive financing policies that have no explicit targeting criteria, or that are largely universal in nature, face lower risk of such diversions and can promote desirable political competition (Frey, 2019). 2.3

Politics in Action: Actors, Mechanisms and Strategies Influencing the Policy Process

While the political foundations of policymaking and the different stages of the policy cycle are important to understand why particular policies are chosen over others, there are quite a few other practical questions that confront a policymaker. Different skills and strategies are often

Political economy of public financing of health  347 required to negotiate across different manifestations of power and interest associated with the policy and its potential impacts, and collaborations need to be established between different actors across policy arenas. 2.3.1 Policy actors A useful starting point for understanding policymaking in the health sector is the identification of the stakeholders associated with different stages or roles in the policymaking process. Stakeholder analysis has been acknowledged as a key element in policy analysis (for a detailed review, see Brugha and Varvasovszky, 2000), contributing to each of the distinct stages of the policy cycle discussed in the previous section. To understand the politics surrounding the role of these diverse actors across different stages, it is essential to identify their positions, roles, interests and power. Campos and Reich (2019) propose a framework that distinguishes six major categories of actors and the associated politics that affect health policy and reforms: interest group politics, bureaucratic politics, budget politics, leadership politics, beneficiary politics and external actor politics. Interest group politics is concerned with stakeholders with shared interests, such as health workers, pharmaceutical manufacturers or health insurers, who may vary considerably in the amount of power they have to influence different aspects of policymaking. As we have discussed earlier, it is hard to imagine any policy stage which is bereft of interest group influence, often involving an explicit contestation between the different interests represented. Bureaucratic politics refer to the different agencies within the government, including at different administrative levels such as national, provincial or local levels, particularly where there are formal divisions of administrative power across these. In many LMICs, this also involves politics – manifest in hierarchies of power – both within and between different civil service cadres, such as in general administration, technical or professional cadres (e.g., health services, engineering or public works, auditing). Budget politics, as the name implies, concerns mechanisms of allocation and expenditure of budgets across government departments. Here, questions of power and interest across the actors – ministries of finance and health, for instance – primarily involve the size of resource envelopes made available, reflecting the political priority attributed to the health sector, as well as the negotiating power of the health ministries in making a convincing case for higher and/or more flexible allocations. This closely parallels the bureaucratic politics among the relevant departments, but also involves more executive or political roles; e.g., between the ministers in charge of the finance and health portfolios, or between national and subnational units. In certain cases, wider macroeconomic questions for policy decisions – e.g., introducing taxes on tobacco products or carbonated drinks – introduce multiple actors across government ministries and industry groups in the negotiations. Leadership politics refers to the political will and commitment of executive and/or legislative leaders, often as the ‘face’ of the government or the political party in power. In most non-authoritarian settings, the key questions on leadership politics involve how leaders assess the political feasibility of policy decisions – that is, the ‘political benefits and costs of actions’ (Sparkes et al., 2019, p. 186) – and accordingly choose to prioritise or veto specific policy processes. Under incentives for re-election, leaders are also motivated by the relevance of the policy decisions to electoral cycles, and if applicable for legislative approvals, to mobilise bipartisan or coalition support.

348  Handbook on the political economy of health systems Beneficiary politics address how new policies alter existing benefits to relevant stakeholders and their distributional balance across key groups. Public opinion on any potential policy does affect how different beneficiaries react to it, but the core political foundations discussed earlier – ideological affiliations and preferences, as well as distinctive identities that influence interests – are of crucial importance. An example pertaining to the health domain may be how workers are organised or affiliated with trade unions with specific political ideologies (e.g., left- or right-wing), and accordingly react to policy proposals on issues such as enrolment, benefits and costs under a new social health insurance regime. Workers with non-permanent contracts, or those affiliated to centrist or right-wing trade unions, might be less resistive to market-oriented policies (e.g., with a higher co-payment), while more organised or left-wing unionised workers would be more likely to accept policies that have a strong public, subsidising role. Finally, external actor politics introduces the influence of foreign donors or other international agencies, as well as of non-government actors such as charitable foundations and non-profitable organisations, into the policymaking process. The power of international financial institutions is considered to be substantial in influencing policy decisions around liberalising the different interconnected markets for health care across LMICs. Under structural adjustment programmes spearheaded by the IMF since the 1990s, several countries in Asia and South America have increasingly adopted pro-market policy reforms often involving a greater participation of the private sector (Forster et al., 2020; Loewenson, 1993) or liberalised pharmaceutical pricing policies (Semin and Güldal, 2008). 2.3.2 Political strategies and mechanisms to influence policymaking Once the key actors relevant to distinct stages of the policy process are identified, understanding the optimal set of political strategies and mechanisms that connect their anticipated roles to actual actions is essential. Such political analysis is equally relevant for both ex ante understanding of the likely feasibility of any policy reform, as well as for the retrospective assessment of the success or failure of any policy adopted. These political strategies connect the political actors or players – as discussed above – with their relative position or interests with respect to the policy in question, their relative power to influence its course of adoption or rejection, and perceptions on how their respective constituencies or the general public would react to such decisions (Roberts et al., 2008). In Section 3, we present a more detailed discussion on the issues of power held and exerted by different groups and individuals to influence the policy process, but a few salient issues are worth noting here. Power of any actor to influence political strategies and decisions primarily follows from their access to and command over political resources (World Bank, 2008). These resources could be tangible or intangible, and have varying significance across different categories of policy actors. Tangible resources include money, organisational strength, people, votes and infrastructure (Roberts et al., 2008), and in most cases are highly interdependent. Financial resources are essential inputs to help mobilise and sustain organisational strength, particularly workers and supporters, for both partisan as well as non-partisan agents. Financial power is also highly relevant for lobbying with key decision-makers or canvassing any agenda among the general public, and so is availability of manpower that helps in carrying out such actions. For political agents, the electoral support of their voters is a major source of power and ensuring stability of such support often hinges on organisational strength. Infrastructural resources – office space, vehicles, means of communication – are all essential ingredients to maintain

Political economy of public financing of health  349 functional presence and influence for all agents, but more important for non-governmental actors who cannot rely on any entitled supply of these inputs. Intangible political resources that generate and sustain power can be diverse. Information on key aspects of the contents of any policy agenda, as well as on the likely supporters of and resistance to these, is a valuable resource lending power to actors and groups who are privy to them. Technical skills, competence, relevant expertise and familiarity with the policy context are all crucial resources that connect power to influence in policy decisions, and are again more relevant to particular groups as drivers of their power; e.g., among the bureaucratic or external actors above. Access to political leaders or other opinion makers such as the media is also an important resource that contributes to power, for both potential beneficiaries of the policy reform and other interest groups having a stake in the decisions (Roberts et al., 2008). Other informal sources of power – which we return to in the following section – such as legitimacy of any political regime or the degree of allegiance it can generate among the general public also play a major role in defining the remits of power retained by political actors. Power alone, however, cannot fully predict what political strategies different actors and groups are likely to adopt in real-life questions of decision-making. The core political foundations discussed at the beginning of this chapter – interests and ideological affiliations – are useful keys to anticipate actions that are likely to be taken, with positions actually adopted in the past in response to similar decision-making contexts also being important guides (Roberts et al., 2008). Here it is important to distinguish between, and rely more on, public positions rather than a priori or normative ‘expectations’ on how these groups are likely to behave. Illustrations of ‘deviations’ from expected positions include standpoints on market-oriented reforms by political groups with leftist ideologies, who are traditionally considered as opponents to such reforms. However, support from these groups for market-oriented reforms can be a strategic decision, as has occurred in Costa Rica with the objective of improving efficiency of the bureaucracy (Vargas and Muiser, 2013), or otherwise responding to external donor influence, for instance. Finally, public perceptions of policy decisions are crucial considerations for most policy actors and interest groups, not limited only to those in government office. Policy reforms that might be construed by the general public as any reduction in existing privileges or imposing additional costs are likely to be opposed, which political supporters of such reforms might find difficult to ignore. Non-governmental interest groups such as professional associations of physicians or pharmaceutical industry leaders also need to be sensitive to potential repercussions of their standpoint with respect to any policy reform, among both their own members and the general public. In most cases, particularly for decisions that have obvious and clear implications on access, utilisation or financing of health services among the public, policy stakeholders need to invest in canvassing the benefits to reduce resistance. Policy decisions for which public opinion has either not been elicited or effectively taken into consideration could be less susceptible to these considerations, and similarly may be bypassed in authoritarian contexts (Khan, 2018). The discussion so far has outlined the conceptual basis required for understanding the political foundations, determinants, actors, mechanisms and strategies that are essential in explaining health policy reforms. The essence of this complex analytical process is to be cognisant about two key factors: how power is derived and distributed across the major political actors, and how such relative distribution of power matches with their interests in the outcomes of the decisions adopted and the distribution of incentives that follow. As we shall see in the next

350  Handbook on the political economy of health systems section, it is the political settlement that arises out of this interactive process – drawing on both formal and informal sources of power, and its manifestation across individuals, groups, organisations and institutions in any country context – that anchors any policy reforms and is responsible for its successful adoption. More importantly, analysis of political settlements provides crucial answers about why policy choices that are successful in certain contexts do not necessarily guarantee similar results in others.

3.

POLITICAL SETTLEMENTS AND POLICYMAKING FOR PUBLIC FINANCING OF HEALTH SYSTEMS

Policy reforms around the nature and scope of public financing of health systems are often a product of the nature and scope of coalitions formed between the various sources of power relevant to the policy context. Policy decisions are a result of agreements between political elites representing their respective constituencies’ interests, and shaped by the capacity of relevant institutions such as the administrative bureaucracy. Political settlements reflect such balancing agreements, underpinning the supportive political ecosystem essential for successful policy implementation. Political settlements analysis (PSA) is a powerful and increasingly popular political economy approach that offers insights to understand why similar policies and institutions achieve different results in different contexts, and also why different policies and institutions have been found successful in addressing similar problems in different contexts (Khan, 2018). There are a few key definitions of concepts employed in the political settlements discourse. Institutions are rules of the game that set incentives, opportunities and limitations for individuals or organisations (Khan, 2018). Under the New Institutional Economics school of thought, which emerged prominently in the 1990s to explain uneven or rather slow-paced economic development across LMICs, institutions are defined as rules that evolve to solve particular ‘transaction’ problems involving any social or economic exchanges between agents under contracts, formally or informally defined (North, 1990; Preker et al., 2000). Policies are also rules like institutions, but are easier to amend or revoke. Similar to institutions, policies also influence social, political and economic outcomes but, more importantly, they alter the distribution of benefits in the society. Organisations are groups of individuals who work together in structured ways and are subject to the rules of interaction set by institutions in their transactions with other individuals or organisations (Khan, 1995). The state is a set of organisations that operate under rules that are more or less effective in different contexts, with the major distinction of being the only body in society that can legitimately enforce institutions and collect taxes, using force if necessary (Khan, 2018). Elites – who as we discuss below play significant roles in defining and maintaining political settlements – include socially, economically and politically powerful groups (Di John and Putzel, 2009). They include traditional elites such as the capitalist ‘owner-class’, but also traditional chiefs, landlords, religious leaders and other political organisations. The political settlements framework explains institutional performance through the relative power and capabilities of organisations that function under the institutions. In this context, political settlement refers to the distribution of organisational power that is stable and reproducible over time (Khan, 2018). PSA is valuable for both retrospective assessments of the drivers of successful policy reforms and diagnostics of which actors and mechanisms are likely to be supportive for the

Political economy of public financing of health  351 adoption of new policy decisions (Kelsall et al., 2016). We discuss below some salient features of the PSA and review a few recent applications for understanding health financing policy reforms in LMICs. 3.1

Political Settlements, Quality of Institutions and Policy Reforms

An increasingly popular approach in comparative analysis of institutions, PSA owes its development to the pioneering works of Mushtaq Khan (1995, 2018, 2010), and later by scholars at the Centre for Effective States and Inclusive Development (ESID)3 (Kelsall et al., 2016; Kelsall and Seiha, 2014; Lavers, 2019; Lavers and Hickey, 2016) at the University of Manchester. While developing typologies for variations in institutional development and characteristics, PSA researchers (Levy, 2015; Levy and Walton, 2013) have also drawn on thinking under the New Institutional Economics School inspired by Douglass North (North, 1995, 1990) and others (North et al., 2012; North and Weingast, 2000). PSA argues that policy reforms can only work by changing or adjusting to the distribution of power across affected organisations. Irrespective of its different motivations (e.g., new political mobilisations, technologies, economic opportunities or shocks) or facilitators (such as supportive ideology, leadership or external influence) discussed earlier, reforms can only be sustainable if the ‘changes in the distribution of power required for their effective operation had already come about, or could be brought about as a result of ongoing mobilizations’ (Khan, 2018, p. 639). Accordingly, while assessing the feasibility of any policy reform, it is essential to consider how the relevant organisations affected would respond, and bargain or contest to get the most out of it. 3.1.1 Political settlements and informal sources of power Khan’s framing of political settlements also offers important insights. First, as reforms are primarily driven by the distribution of power across organisations, a natural starting point in PSA is to identify the relevant powerful organisations. These powerful organisations are, however, not confined among economic or social elites, or ‘formal’ organisations such as the civil services, but also ‘informal’ organisations drawing their power from varying sources such as cultural identities, ethnic loyalties and traditions. Often these informal sources of power have strong historical origins. Second, political settlement does not necessarily imply a ‘settlement’; distribution of organisational power is rarely static and in equilibrium. Understanding political settlements can help in predicting how or whether reforms would change the dynamic in desirable directions and offer support or resistance to the policies intended. Third, absence of conflicts or violence is not a precondition for developing any political settlement. In contexts of protracted conflict, political uncertainty or instability, any balance of formal and informal distribution of power among powerful organisations can develop or lead to a political settlement that can provide grounds for designing and implementing reforms. Fourth, political settlements do not guarantee the inclusion of all groups, or even the most directly affected groups by the reforms, unless they represent unignorable interests in the power relations that are relevant for the settlement, and not represented otherwise. Also, there is no essential precondition of any political systems that are ‘ideal’ for arriving at any settlement: stable political settlements can be achieved by exclusionary, authoritarian regimes in what has been described as ‘illiberal peace’(Lewis and Sagnayeva, 2020).

352  Handbook on the political economy of health systems Empirically, through its focus on balance and distribution of organisational power, PSA is alike common political economy analysis (PEA) approaches such as power- or interest-group analysis, but with a key additional emphasis. PSA is concerned not just with the relative distribution of power, but also with ascertaining its sustainability or reproducibility over time in similar reform contexts. In addition, unlike most traditional PEA approaches, PSA emphasises the role of informal sources of power and how informal institutions exert power over their constituencies. In doing so, it incorporates important drivers of such informal powers; e.g., corruption or other forms of rent-seeking, including clientelist privileges that lend stability to the derived power. In fact, sources of organisational power are of key importance to understand its sustainability, its response to any reforms/institutional changes, and how ultimate outcomes are to be impacted (Khan, 1995). The possibility of institutional distortion and corruption, and the presence of effective checks and balances among state organisations, all influence the outcomes following from policy reforms. Examples related to health sector financing reforms include the ability of organisations to enforce the effective execution of contractual obligations, such as purchasing from private sector providers or aligning financing to results-based designs under performance-based reimbursement arrangements. 3.1.2 Typologies of political settlements Scholars have suggested a few typologies of frameworks to understand political settlements and how they transition from one stage to another, as institutions become mature and interactions more complex. Khan (1995, 2017) distinguishes between institutions in terms of their growth oriented-ness, and how formal or informal the inclinations for the dominant sources of power are. Countries in northern America, and most parts of northern and western Europe, have institutions that subscribe to a capitalist political settlement ensuring formally defined rights and institutions, with impersonality and adherence to formal systems of rules and political organisation. Other countries, including many LMICs, have clientelist political settlements where significant informal sources of power coexist with strong executive discretion, in both parliamentary democracy and authoritarian forms. Formal institutions, whenever and wherever present, also (mostly) do not operate with impersonality and rule-adherence. Most formal colonies or newly independent nations have pre-capitalist political settlements – stable, mostly formal sources of power, but with limited developmental or reformist agendas. These countries can and do take different routes transitioning into other settlements; a higher pace of redistribution of power and emergence of ‘productive investment’ leads to a move towards capitalist settlements, a slower pace leads towards clientelist forms. Finally, there are settlements in crisis marked by significant political instability, conflicts and violence. In these countries – e.g., current scenarios in West Asian and North African countries (such as Lebanon, Syria, Yemen, Sudan and most recently Ethiopia) – formal institutions and systems gradually collapse and informal power dominates. Similarly, other scholars have drawn on the concepts of ‘access orders’ to classify institutional arrangements in terms of the dominant forms of power and degree of impersonality (Kelsall et al., 2016; Lavers and Hickey, 2016; Levy and Walton, 2013). Originally due to North’s (1990, 1995) New Institutional Economics School, this distinguishes between limited access orders (LAO) and open access orders (OAO). The former relies on personalised elite bargains – agreements that set out negotiated distribution of power and resources among the elites (Putzel and Di John, 2012) – only allowing restrictive and personalised access to influence policy processes. OAOs, on the other hand, are characterised by openness of political

Political economy of public financing of health  353 orders and higher degree of impersonality. Transitions from LAO to OAO involve organisational complexity and impersonal institutions (Figure 20.1).

Source: Levy (2015).

Figure 20.1

Levy’s typology of political settlements

Predatory forms of dominant LAO settings (top-left quadrant in Figure 20.1) are characterised by mostly authoritarian governance, poor public service and bureaucratic performance, but still offer some opportunities for ‘islands of effectiveness’ (Levy and Walton, 2013), as the Rwandan experience of expanding social health insurance shows (see Section 3.2). Under developmental LAOs – which approximate the context of newly independent nations during the mid/late-20th century, but with mostly authoritarian political systems – reforms are initiated (and are mostly likely to succeed) to win over support from groups that can threaten political stability or to triumph over political opponents. In some cases, such reforms are essential to counteract shocks or crises, or as a precondition for essential external (donor) support, yet ensuring support from domestic groups is still critical to ensure that the reforms are institutionalised. In the developmental stages that follow, with growing impersonality in institutions, bureaucratic performance improves, allowing the emergence of a ‘dominant political leadership with developmental orientation’ (Levy and Walton, 2013, p. 16). Such progress is mostly through technocratic actions and the ruling elites have limited threat from less powerful elites. Competitive LAO settings can have either clientelist or more inclusive forms of institutions. Transitions to these settings are often supported by progressive social and political reforms. Under competitive clientelism – witnessed in several countries in Asia, Africa and Latin America (India, Kenya and Mexico, for instance) – institutional systems are defined by wider

354  Handbook on the political economy of health systems stakeholder involvement, but also persistently high importance of political connectivity and organisation, a common feature in early democracy systems. In most of these settings, denoting a shift from the bottom-left to the bottom-right quadrant in Figure 20.1, strong demands for redistribution gradually emerge and so do stronger, more transparent organisations. In both early and transitional stages, there are still strong incentives for ‘quick-fire’ or ad hoc reforms to gain electoral dividends, which we will return to in the case study in Section 4. How are such classifications of political settlements and the related analyses of the distribution of power important for health financing reforms? While the quality and effectiveness of institutions are integral to whether policy reforms are likely to succeed, political settlements are critical preconditions for health system reforms in clientelist or developmental settlements (Kelsall et al., 2016). In these contexts – but also equally in countries that have recently transitioned towards more competitive or programmatic settlements, but continue to have strong traits of dominant clientelist powers – a diagnosis of prevailing political settlements and their past trajectories is an essential aspect of policy reform analysis. For both ongoing or planned reforms, such as introducing new social insurance or purchasing services from private providers, it is important to understand (a) which organisations and groups wield power and influence for these reforms, (b) how important informal institutions and sources of organisational power are, and (c) what sort of institutional changes – primarily involving the underlying power relations – need to precede any reforms, to ensure greater adherence to and guide better monitoring and evaluation of the implemented programmes. The political settlements framework can also be useful to understand ‘common organisational features’ conducive to health financing reforms; e.g., subsidised or social health insurance, deciding on benefit packages or provider reimbursement arrangements. The framework helps to specify what facilitates the ‘political will’ considered as a central precondition for reforms, by identifying the sources of political power critical for reforms (content, scale, duration). This involves, among others, how reforms (a) align with the balance of organisational power among the political forces, (b) have clear political dividends that can consolidate power or ensure political ‘rents’, and (c) can create new political settlements but not disruptive changes that contradict the existing balance of power. In sum, PSA can guide how successful coalitions can be built and sustained that can inform, develop, adopt and internalise reforms, and how non-political organisations can align with political power to support and sustain reforms. 3.2

Political Settlements and Health Financing Reforms in LMICs

Although PSA is a relatively recent approach of applied PEA in health systems and policy research, a growing body of literature indicates its increasing popularity among scholars in this field. In a useful cross-country review, Kelsall et al. (2016) discuss political settlements in the context of the universal health coverage agenda and identify pathways through which political settlement in varying forms connects with UHC progress. In their view, dominant developmental settlements are generally characterised by strong political commitments if potential reforms are aligned to the powers and interests of dominant group(s); this would lead to what they term as ‘government-supporting’ reforms. Typically, state-centred or a higher public role in the reforms are preferred. In Vietnam, representing a dominant developmental political settlement with a communist single-party government, strong support for publicly funded health insurance reforms has led to significant expansion in financial risk protection.

Political economy of public financing of health  355 But results have been less than optimal due to the presence of corruption and rent-seeking tendencies of the politically powerful groups. In dominant-predatory settlements, such as in the Democratic Republic of Congo, Myanmar and several other authoritative regimes in northern Africa and West Asia, political commitment remains weak, and a ‘government-substituting’ settlement dominates with a higher dominance of market-based reforms. Kelsall et al. (2016) argue that several LMICs are either transitioning towards or have developed an elitist or inclusive competitive clientelist settlement. In the former (elitist) case, although some political or electoral competition exists, both the pathways and results are not very different from the dominant elitist settlement. Most benefits from reforms such as expanded health insurance or wider choice of providers – which continue to be driven by market-based or government-substituting reforms – are captured by the elites. In Indonesia, for example, the benefits of the health financing reforms have remained largely limited to expansion in service coverage, without comparable benefits in terms of reducing OOP expenditure. In more inclusive clientelist settlements, illustrated by the case of Kyrgyzstan, consensual understanding among different stakeholders both within and outside the government, as well as with external donors, has helped in sustaining progress across the key UHC goals (Kelsall et al., 2016). Common interests and agreements around the reform agenda by key political actors, facilitated to an extent by powerful international actors, helped. Bangladesh is another interesting example of a variant of the competitive clientelist settlement – perhaps uniquely among other countries achieving independence in the late 1960s and early 1970s – where some commendable consensus among the elites was established in the early post-independence period, leading to a political settlement favouring strong public health and other pro-poor programmes (Kelsall, 2020), with a resulting central featuring of health issues in electoral competitions. However, corruption and rent-seeking tendencies among functionaries at lower levels with approval and support from the elites and the politically powerful groups that include both emerging socio-religious groups and the strong non-governmental organisation (NGO) sector (Naher et al., 2020; Transparency International, 2015), or intra-group contests due to partisan affiliations among members of professional associations such as the Bangladesh Medical Association (Kelsall et al., 2016), often has the risk of undermining the achievements. For African contexts, Chemouni (2018) presents a good discussion on the political trajectories of the adoption of health insurance reforms in Rwanda, considered one of the LMIC success stories on UHC progress around financing goals (Saksena et al., 2011). Emerging from a long and bloody period of ethnic conflicts, Rwanda has demonstrated remarkable progress in expanding health insurance coverage during the last decade, relying primarily on a nationally implemented community-based health insurance (CBHI) programme, the Mutuelle de Santé. Several political factors have been considered as the key drivers of the adoption, prioritisation and gradual intensification of the reforms. First, a dominant political settlement – both horizontal (between elites and other politically powerful groups such as ethnic organisations) and vertical (between the elites and the non-elite citizenry) – converged into a concentration of power with Paul Kagame’s RPF party. Limited political opposition, both inter- and intra-party, and the availability of a longer time horizon helped in the stability of power. Even when the enrolment into the CBHI programme was made compulsory along with the mandated contributions, the clear authoritarian shift in the programme matched well with the prevailing dominant settlement. Although it is beyond the scope of the present chapter to discuss how different types of political settlements influence state capacity, it should be noted for the Rwandan context that the dominant political settlement has fed into strong state

356  Handbook on the political economy of health systems capacity – the ‘infrastructural power’ required to maintain adherence via command and control over grassroot bureaucracy. Second, an ideological opposition to reliance on external agencies and preference towards promoting community solidarity worked as initial motivations for piloting the reforms. Third, a combination of political expediency, consolidating power and regime stability were critical in lending political support for the reforms. Mutuelle de Santé was a useful binder for establishing legitimacy by the Tutsi RPF in rural areas mostly dominated by the Hutus and fostering cohesive identities. Fourth, a process of expanding coalitions by the higher levels of politico-administrative power, involving and co-opting local officials, and a gradual process of expanding professionalisation in a manner that allowed distribution of power without threatening the dominant political settlement, significantly improved the functioning of the CBHI programme. These local powers were skilfully accommodated: opportunities for seeking rents by local bureaucracy were linked with performance, which created incentives for the officials to expand CBHI enrolment. At the same time, risks of benefit capture by local elites were accounted for through recognition of leakages such as mis-identification of eligible beneficiaries by implementing better means of identification of beneficiaries and closer monitoring. Ethiopia also constitutes an interesting case for applying PSA to understand recent health financing reforms, particularly early successes in combining CBHI for the informal sector with a contributory social health insurance for the growing formal sector, leading to rapid increases in financial risk protection (Fenny et al., 2021; Lavers, 2019). Lavers (2019) characterises the political settlement in Ethiopia as one of competitive authoritarianism, where the ethno-nationalist EPDRF emerged as the dominant political power, but gradually accommodated other ethno-political groups in different power-sharing arrangements. During the early years of the ruling coalition, it promoted ideas of a developmental state involving social equity and progressivity, thereby lending strong support to CBHI programmes and countering formal sector resistance (Lavers, 2019). Apart from ideological affiliations and commitment to ideals such as self-reliance and community mobilisation, strong links between political actors represented by President Meles and Minister of Health Dr Tedros (later World Health Organization Director-General) were instrumental to expedite reforms and ensure adoption. However, weakening of the concentrated power and dominance in the post-Meles period – including the more recent resurgence of civil war and political instability – has resulted in delayed implementation of SHI reforms and renewed resistance by the formal sector.

4.

ILLUSTRATIVE EXAMPLE: POLITICAL DRIVERS OF HEALTH SYSTEM FINANCING IN INDIA

The earlier sections have laid out some key considerations of political and institutional factors associated with health system financing reforms, particularly on the public role in different stages of the policy cycle or for the key financing functions. We may now proceed to highlight some of the real-life political parameters and processes that are relevant for designing and implementing policies related to public roles in financing health care in LMIC contexts, using an illustration of recent health financing reforms in India. India is an interesting reference case to discuss political economy aspects of health sector reforms for several reasons. The coexistence of opposites in several sectors impinging on the health system – falling poverty but persisting income inequalities across regions and social groups (Alkire et al., 2020; Chancel and

Political economy of public financing of health  357 Piketty, 2019); a vast public health sector that is steadily losing ground to the private sector (de Costa and Diwan, 2007; Govt. of India, 2011; Kanjilal and Mazumdar, 2012; Patel et al., 2021), which itself varies hugely in terms of quality and capacities; a multi-party, parliamentary democratic system that often pits political arch-rivals in power in federal and provincial governments locking horns on key policy decisions (Bardhan, 1999; Frankel, 2006; Kohli and Singh, 2018) – along with stagnancy in public investments in health leading to significant disparities in access to and financing of health care (Balarajan et al., 2011; Sakthivel and Karan, 2012), present relevant scenarios to assess how political forces and processes influence policy measures and their likely results. Additionally, a discernible shift towards different variants of publicly funded health insurance programmes since the mid-2000s (La Forgia and Nagpal, 2012) by several provincial governments as well as the national government has led to realignment in the roles and influence of different actors in the policy reform process across the country. An analysis of the political economy aspects of all or most of the distinctive facets of public financing of the health system in India, however, lies beyond the scope of this chapter. This illustrative example focuses on two key issues that provide useful leads in understanding the major political economy parameters and mechanisms that have shaped relevant policy priorities and choices, using the PEA tools and concepts explained earlier. We briefly discuss how PEA can help understanding (a) the reasons for persistence of low levels of public spending on the health sector in India, and (b) the gradual predominance of publicly funded health insurance as the major policy strategy to extend financial risk protection, particularly among the socioeconomically vulnerable population groups. We review some of the important literature available to identify (a) the key actors and interests relevant to the policy decisions – for both new reforms or maintaining the status quo, (b) how relative manifestations of power – both formal as well as informal – held by these actors and interest groups anchors the political settlement that binds these decisions, and (c) political mechanisms such as dominant party ideologies and leadership, electoral results and legislations which are responsible for both stability or bringing about changes in such political settlements. 4.1

The Political Economy of Public Spending on Health in India

Public expenditure on health in India is one of the lowest among the LMICs and has changed little over the last three decades. At less than 1 per cent (0.96 in 2018; Global Health Expenditure Database4) of gross domestic product (GDP), government expenditure on health in India ranks 24th from the bottom; in comparison the government in China spends 3 per cent and in Brazil about 4 per cent of their respective GDPs on health (World Bank, op. cit.). This seem to be a paradox as the Indian economy has been experiencing one of the fastest growth rates of GDP globally since the 1990s. Public spending on health in India, similar to other federal countries, is met by both the national government as well as by the state governments with the revenues mostly accruing from general taxes (and the states’ shares in the national tax revenues), which leads to some variation in the levels of public spending on health across states. With some exceptions, states in southern India (led by Kerala) with better health system coverage and outcome indicators have higher public spending on health per capita, while those in the northern and eastern regions spend less. Interestingly, in most of the poorer northern and eastern states public spending tends to be higher than private (mostly OOP) spending on

358  Handbook on the political economy of health systems health, which has been partly explained by larger rural populations and a poorly developed private sector. Insights from political economy offer few explanations for such consistently low levels of public spending on health. Interest group theories argue how policies which dictate public investments across different competing sectors are influenced by the power and interests of the dominant proprietary classes. The ability of the elites (Bardhan and Mookherjee, 2017, 2012) to exert power and influence on policy decisions – both directly through groups or institutions they control, or by establishing coalitions with other dominant groups – has been suggested to explain higher allocations to certain sectors or supporting policies such as subsidies that benefit these powerful groups, instead of adopting more egalitarian policies. For example, fertiliser subsidies in India have steadily increased in the last couple of decades,5 most of which are cornered by the northern and western states with powerful farm lobbies. Preferences for such private- or group-based transfers in public spending decisions over public goods such as health and education are also indicative of typical supportive political settlements. Disproportionate influence of politically powerful interest groups on policy decisions is common in varying extents across almost all of the Levy’s settlement types discussed earlier (Kelsall et al., 2016), but it might be useful to note that scenarios such as the subsidy question above are typical of LAO systems, ranging between dominant developmental and elitist competitive clientelist orders. When different powerful elite groups are able to establish coalitions between themselves – rich farmers and businesses, for example – they have little pressure (or threats) from other, or lower-level (non-elite) groups or external stakeholders to support or implement progressive policies such as higher public social spending (Lavers and Hickey, 2016). Interestingly – and we shall return to this later to emphasise it – these groups may, however, acquiesce some devolution of privileges through different social protection policies among their constituencies or lower-level groups under certain scenarios. Keefer and Khemani (2005) explain public expenditure patterns or factors that influence public spending decisions that are more likely to benefit the poor vis-à-vis other alternative expenditure items in terms of the extent of political market imperfections. They identify three sources of such imperfections all acting against electoral support in favour of or incentivising pro-poor spending: lack of information among voters on politician performance; social fragmentation that encourages identity-based, rather than issue-based, voting; and lack of credibility of politicians’ promises to citizens (Keefer and Khemani, 2005). They and others empirically demonstrate that in a setting such as in India where politicians’ promises and reputation are not credible and voters are not organised around ‘common performance thresholds’ for the politicians, strong norms of political clientelism ensure that less is spent on public goods such as health, and more on transferring specific benefits to sections of the electorate (Keefer and Khemani, 2004; Keefer and Vlaicu, 2008). It is argued that as leading political parties in India cannot be differentiated significantly on economic policies, voting behaviour is predominantly influenced by the ability of the parties to consolidate their electorates based on existing social cleavages or identities, mostly through promises of using public resources for specific benefits such as quotas for different social groups in public jobs (Chhibber, 1999; Jaffrelot and Kalaiyarasan, 2019; Keefer and Khemani, 2004). In fact, a rich and distinctive strand of literature on the political economy of public goods provision suggests that in a socially or ethnically fragmented society, quality and quantity of public services being provided remains poor (Alesina et al., 2019, 1999; Miguel and Gugerty, 2005). In such fragmented or polarised societies, as in most regions of India, voters generally

Political economy of public financing of health  359 vote for candidates they most closely identify with because of social or other group-based affiliations, irrespective of their performance or policies. Such clientelist or other identities-related (castes, religion) loyalties reduce demand for public services, and are more oriented towards private benefits. In India, the adverse effects of polarisation or social fragmentation have also been noted in terms of lower inputs for public goods. Districts with higher shares of socially disadvantaged groups such as the Scheduled Castes (SC, also referred as Dalits) or religious minorities have fewer or less-reliable health centres or availability of health workers (Banerjee et al., 2007; Banerjee and Somanathan, 2007, 2005; Betancourt and Gleason, 2000). Higher ethnic or social fragmentation also perpetuates political fragmentation by incentivising narrow identity-based politics, leading to higher political competition between large numbers of contestants with lower winning vote margins. This further dissuades politicians from pursuing spending on public goods such as health with less-visible ‘deliverables’ to the voters, when compared against private benefits such as jobs or similar contracts, or other public goods such as infrastructure projects (Keefer and Khemani, 2005). However, Datta (2020) used state-level time-series data on electoral results and public expenditure across 16 major Indian states during 1991–2001 to find a positive correlation between voter turnout, level of political competition (measured by seat–share difference between the top two political parties in the state) and public expenditure on health (Datta, 2020). Explaining state-level variations in public spending on health in India, in terms of contemporary patterns in mobilisation of political forces, particularly the (re)emergence of identity politics, remains a fertile area for future research. 4.2

The Politics of Publicly Funded Health Insurance Reforms in India

Finally, we consider another key facet of recent health financing reforms in India – a distinctive preference towards various publicly funded, subsidised health insurance programmes, both centrally as well as pursued independently in a few states. A cornerstone of this recent policy paradigm is the Ayushman Bharat-Pradhan Mantri Jana Aarogya Yojana (AB-PMJAY, or roughly the Healthy India – Prime Minister’s Public Wellness Programme) launched in 2018 by the right-wing BJP government led by Prime Minister Narendra Modi, but it would be useful to trace the genesis of such programmes across India, which started about a decade earlier, from the mid-2000s. We highlight the key political issues involved in the evolution and predominance of these programmes in the financing policy agenda, especially how the politically sensitive issue of social security and welfare of the poor and other vulnerable groups such as workers with insecure, informal livelihoods aligns (or not) with such policy choices. The history of publicly funded (also referred by some observers as government-sponsored) health insurance programmes is relatively recent. Apart from general tax-funded publicly provided health services across the country, mostly available for free or for nominal user charges, health insurance coverage had been largely limited to private ‘mediclaim’ schemes, operated by the major public sector general or life insurance companies. Traditional forms of employment-linked social health insurance have also been limited, covering civil servants and other central government staff (under the Central Government Health Scheme, or CGHS, launched in 1954) and blue-collar workers in formal sector industrial establishments (under the Employees State Insurance Scheme, or ESIS, launched in 1948). In the early 2000s, these different insurance programmes contributed a trifling amount of the total health expenditure in the country – about 5.1 per cent in 2000–01, marginally increasing to 5.9 per cent in

360  Handbook on the political economy of health systems 2004–05. In 2003–04, ESIS and CGHS covered only about 35 million people, with another 15 million covered by different commercial public and private sector insurers (La Forgia and Nagpal, 2012). The high predominance of informality in the Indian labour market, with only a small fraction of employment provided by formal public and private enterprises, and general unaffordability of the high premiums required by the private insurance schemes were major reasons for keeping the coverage low. In 2009–10, however, the population coverage of the public insurance programmes experienced an unprecedented nearly eight-fold increase – to 243 million. While there was some healthy increase in ESIS coverage (from 31 to 56 million) (La Forgia and Nagpal, 2012), the game changer here was the introduction of India’s first large-scale subsidised health insurance programme – the Rashtriya Swasthya Bima Yojana (National Health Insurance Scheme, or RSBY, launched in 2008).6 Around the same time, two states, Andhra Pradesh and Tamil Nadu, also introduced two major state-level programmes, the Rajiv Aarogyasri (2007) and Kalaignar (‘great leader’) scheme, respectively. Political forces and processes played a leading role in setting the agenda for these pioneering initiatives at the outset and in providing crucial life support to them in the fledgling years and overcoming the teething problems. The return to power of the nationalist Congress Party defeating the right-wing alliance led by BJP in the parliamentary elections of 2004, with the crucial support of the leftist parties (putting up their best results so far), led to the formation of the United Progressive Alliance, based on a Common Minimum Programme which laid a strong emphasis on social security programmes for the poor. A National Advisory Council (NAC) comprising leading social scientists, activists and experts was formed, which in one of its key recommendations suggested a suitable insurance system aimed at the unorganised or informal sector workers.7 A combination of the strong demands and pressure from leftist parties and members within the NAC received positive support from Prime Minister Dr Manmohan Singh, himself a leading economist and credited with ushering the economic liberalisation programme in India in 1991, and RSBY was launched nationwide in April 2008. To highlight how RSBY was the result of successful convergence of different key interests and political support, Shroff et al. (2015) use Kingdon’s multiple stream framework discussed earlier. They suggest that the defeat of the BJP, claiming to have ushered in an era of economic boom, led to a resurfacing of the problem of insecurities of common people in the national political discourse (the ‘problem stream’). The reliance of the new government on the crucial political support from the left parties pushing for more redistributive, egalitarian policy reforms lent key political support to the reforms (the ‘politics stream’). Finally, the availability of competent and interested bureaucrats, the emergence of strong domestic private insurers as third-party administrators (TPA) and locally available expertise on strong information technology systems to process insurance claims ensured that once political decisions had been arrived at, the necessary wherewithal to implement the policy was readily available (the ‘policy stream’) (Shroff et al., 2015). The policy process of developing and implementing the RSBY programme also reflects characteristic attributes of a competitive clientelist political settlement mentioned earlier, where strong political competition between the two major political parties (the Congress and the BJP-led alliances) created grounds for some reforms on extending financial risk protection to the poor. However, strong elitist predominance by powerful interest groups ensured that the scope of the reforms remained rather narrow, in terms of both coverage of services as well as the population under the insurance programme. In fact, a key recommendation of the National Commission for Enterprises in the Unorganised Sector (NCEUS), 2009, noting that 92 per

Political economy of public financing of health  361 cent of the country’s workforce was engaged in the informal sector, was on establishing a contributory National Social Security Fund to support essential social security benefits, including expenses on medical care.8 However, illustrating the ‘veto points’ discussed earlier that can thwart policy recommendations, strong resistance from the corporate sector forced successive governments to ignore any such comprehensive social security reforms. The pro-elite bargain was made easier due to the passive role of the trade unions of leading political parties, which had a weak presence among the workers in the unorganised sector (Srivastava, 2010). These workers could be more easily mobilised under other clientelist social and ethnic identities, rather than their class-based demands. RSBY gradually lost its relevance with the informal or unorganised sector with the eligibility criteria being ‘officially’ listed as poor households,9 without any reference to their forms of employment. As a further underscore to the dilution of a comprehensive social security agenda involving social health insurance, administrative management of RSBY was also moved from the reluctant Ministry of Labour to the Ministry of Health in 2015, when the BJP-led National Democratic Alliance (NDA) government under Prime Minister Narendra Modi returned to power with a comfortable majority in the parliamentary elections of 2014. Nevertheless, in 2015 about 360 million families in 21 (out of 28) states were enrolled under RSBY, making it among the world’s largest health insurance programmes.10 While evaluation studies on the impact of RSBY have led to mixed results (Karan et al., 2017; Prinja et al., 2017; Sakthivel and Karan, 2012), and at best indicate that RSBY had rather modest, marginal impacts, it remains India’s pioneering attempt at national-level health insurance. Yet, the Congress made only pedestrian reference to RSBY in its national election manifesto for the parliamentary elections in 2014, in which it suffered a crushing defeat. The new NDA government, unlike its earlier term in office, wanted to posture with an inclusive, pro-poor development agenda and build a strong, progressive image of the new prime minister, Modi.11 He has been criticised for taking little action to prevent anti-Muslim riots while he was chief minister in Gujarat,12 a rich western Indian state, although investigations by the Supreme Court of India in 2010 did not find any culpable evidence. Under Modi, the BJP has, however, often been associated with a radical political agenda supporting Hindu (the majority religious group in India) nationalism (Harriss, 2015; Jaffrelot, 2015). In 2018, towards the end of its first term in office, the government declared the launch of a ‘new’ nationwide health insurance programme, which, similar to several other such social sector programmes, was named Ayushman Bharat-Pradhan Mantri Jana Aarogya Yojana after the prime minister, indicating the party’s high, centralised dependence on the image of Modi. PMJAY was similar to RSBY in many respects, with the major difference being the financial coverage ceiling, increased to INR 500,000 (£5000) per family, and an ambitious target of covering 100 million poor families. The launch of PMJAY was well timed before the parliamentary elections in 2019, and the BJP-led NDA hoped to reap electoral dividends from this and several other social schemes, mostly involving some definite entitlements to a section of the electorate, generally the poor. To a large extent, such a strategy was quite successful as the NDA returned to power with a landslide majority. Recent empirical research (Deshpande et al., 2019) suggests a positive impact of these welfare programmes, including PMJAY, on the BJP’s electoral success, which may signal a new form of ‘ethno-political majoritarianism’ (Chhibber and Verma, 2019) to consolidate political power in favour of the BJP. The availability of pre-existing and supportive convergence of political and bureaucratic support – much of it strategically sourced from the grounds prepared under the predecessor

362  Handbook on the political economy of health systems RSBY programme – were the key political drivers in the introduction and implementation of PMJAY. This second generation of health insurance reforms in India, with a much larger involvement of the private medical sector, a significantly higher financial envelope and wider population coverage, involves several other deeper political economy questions critical to the future and impact of the reforms. These are indicative of a rather transitory balance between dominant power and interests under the present arrangement, instead of a stable political settlement that can consolidate the programmes into a comprehensive reform agenda with potentially more sustainable health system gains. Without going into the complex details of these issues and their political linkages, a few brief indicative illustrations might be useful here. First, the PMJAY-type insurance paradigm – in such a wide scale both financially and operationally – adds further emphasis to the government’s role as a purchaser of services from the private sector. A small body of literature refers to the intricate political considerations associated with strategic purchasing, but remains scant in terms of mixed health system settings in LMIC contexts such as in India, where the public sector faces several constraints in enforcing effective contracts. While organised interests in the private hospital sector have always been strong in India, recent developments in their rapid growth,13 particularly in smaller cities, and early indications of concentration of ownership by major conglomerates such as the Apollo Hospitals Group, Narayana Group, Max Healthcare, Fortis, etc., in large cities raise new challenges for adequate political will and power to implement effective purchasing. Second, the over-emphasis on a hospitalisation-linked publicly financed health insurance programme as the cornerstone of health financing reforms also raises several concerns from the health systems viewpoint. Higher budget allocations for PMJAY have been to an extent at the expense of lower spending on other critical areas, such as the National Rural Health Mission (NRHM), which has been vital to strengthen primary-level health care and encourage useful synergies between local political and health service actors in planning and priority-setting. In its present scope, PMJAY risks increasing fragmentation in financing by ignoring most health services not requiring hospitalisation, and also has so far done little to expand coverage among informal sectors such as the new gig-economy workers. The pro-elite bias in the regulatory framework is evident as, notwithstanding recent legislation regarding new Labour Law reforms promising some improvements, these informal workers have not yet been included under formal social insurance arrangements as part of the ESIS. Third, a gradual rise of assertive regional political forces in major states such as West Bengal, Maharashtra and Tamil Nadu, for example, has been responsible for frequent political deadlocks around centrally funded schemes such as PMJAY, with these states preferring their own schemes. Such growth of strong sub-nationalism in Indian politics might lead to some realignment in respective local political settlements with a more diverse role of political ideologies, clientelism and strength of institutions across some politically powerful states. Finally, the electoral relevance of programmes like PMJAY might have a limited time horizon. But, with limited alternatives being offered by contending political forces, and with the risk of disturbing the fragile balance across powerful interests such as the insurance, hospital and pharmaceuticals sectors and the physicians’ lobbies currently maintained by a dominant political coalition, the programme may continue to exist without contributing effectively to the desired health system goals.

Political economy of public financing of health  363

5. CONCLUSION This chapter discusses some of the key theoretical foundations and concepts essential to understand political economy considerations of public financing of health services in LMIC contexts. We have discussed how different stages of the policy process – in financing but also equally applicable in other, wider aspects of health systems – are influenced by different actors and groups, their political ideologies and the relative power retained and exerted by them. We have introduced the political settlements framework and discussed its salient features, which offer several insights in classifying political economy and institutional systems observed in LMICs and understanding why some policy reforms are adopted and why some are not. Finally, we discuss – through a case study highlighting leading public health financing issues for India – how some of these concepts and frameworks can be useful in better explaining the evolution of policy decisions and ascertaining its likely trajectories in future. While the political economy of health policy has been a rich area of scholarship over the last few decades, the fast-changing landscape of public financing of the health sector in LMICs – particularly in the context of global commitments to universal coverage – involving the emergence of new political and economic forces adds to the further relevance of this field of study and calls for further empirical research. New and distinctive considerations typical to the political economy architecture of LMICs – informal power, imperfections in electoral systems, dominant political clientelism and identity politics, weak state capacities – make it imperative to adopt a grounded, realistic approach to explain policymaking and implementation in these contexts, not just retrospectively but with reasonable power to trace likely policy prognoses in the future.

NOTES 1. Institutions differ from organisations, which are collectives of individuals who agree to adhere by a certain set of common rules, or affiliate themselves to such common institutions and act towards common goals or principles. 2. In a related conceptualisation of the agenda-setting process, Peter Hall defines three such ‘viability’ conditions: economic, administrative, and political (Campbell, 2002). 3. For further details, see www​.effective​-states​.org. 4. See https://​data​.worldbank​.org/​indicator/​SH​.XPD​.GHED​.GD​.ZS​?most​_recent​_value​_desc​=​false. 5. See www​.faidelhi​.org/​general/​subsidy​-fert​.pdf. 6. RSBY provided for cashless hospitalised treatment for eligible families (maximum five members in a family) up to a maximum of Indian Rupees 30,000 (about £300) annually, in designated hospitals, both private and public. 7. Another key health sector policy reform introduced at the behest of the leftist parties and the NAC was the National Rural Health Mission (NRHM) in 2005, which remains India’s first national level reform initiative to decentralise health service administration and involve local elected governments in planning and managing lower-level health facilities. 8. See https://​pib​.gov​.in/​newsite/​PrintRelease​.aspx​?relid​=​75595. 9. Known in India as the Below Poverty Line (BPL) families. A headcount poverty ratio (and the number of poor) for each state is decided by the Central Government, and the states are responsible to identify the poor households, usually using periodic BPL census that adopts a means-testing approach. However, political affiliations and other social identities as well as a significant degree of local-level corruption remains major reasons for incorrect identification of the poor, which allows clientelism to subvert anticipated gains from different targeted programmes, such as subsidised health insurance. (See, for example, Bardhan and Mookherjee, 2012; Panda, 2015.)

364  Handbook on the political economy of health systems 10. See https://​pib​.gov​.in/​Pr​essRelease​IframePage​.aspx​?PRID​=1​ 513162. 11. See http://​cdn​.narendramodi​.in/​wp​-content/​uploads/​2014/​04/​Man​ifesto2014​highlights​.pdf. 12. See, for example, www​.theguardian​.com/​commentisfree/​2014/​apr/​07/​narendra​-modi​-massacre​ -next​-prime​-minister​-india. 13. To some extent, rapid growth of the organised private hospital sector has been aided by favourable fiscal and administrative policies of different governments, ranging from tax concessions to newer units and lowering of corporate taxes and allocation of lands at lower costs. As, contrary to expectations or even agreements with governments to provide subsidised treatment to the economically weaker sections (in Delhi, for example), the benefits of such favourable policies have been regressive. This is a further example of elitist bias and capture of public resources in settings where state capacity to enforce contracts remains low.

REFERENCES Acemoglu, D., Johnson, S., and Robinson, J.A., 2005. Institutions as a Fundamental Cause of Long-Run Growth, in: Aghion, P., and Durlauf, S.N. (Eds), Handbook of Economic Growth, vol. 1. Elselvier, pp.  385–472. https://​doi​.org/​10​.1016/​S1574​-0684(05)01006​-3. Alesina, A., Baqir, R., and Easterly, W., 1999. Public goods and ethnic divisions. Quarterly Journal of Economics 114(4), 1243–84. https://​doi​.org/​10​.1162/​003355399556269. Alesina, A., Gennaioli, C., and Lovo, S., 2019. Public goods and ethnic diversity: evidence from deforestation in Indonesia. Economica 86(341), 32–66. https://​doi​.org/​10​.1111/​ecca​.12285. Alesina, A., and Perotti, R., 1995. Fiscal expansions and adjustments in OECD countries. Economic Policy 10(21), 205–48. https://​doi​.org/​10​.2307/​1344590. Alkire, S., Oldiges, C., and Kanagaratnam, 2020. Multidimensional Poverty Reduction in India 2005/6–2015/16: Still a Long Way to Go but the Poorest are Catching Up. Oxford Poverty and Human Development Initiative Research Paper 54a. Balarajan, Y., Selvaraj, S., and Subramanian, S.V., 2011. Health care and equity in India. The Lancet 377(9764), 505–15. https://​doi​.org/​10​.1016/​S0140​-6736(10)61894​-6. Banerjee, A., Iyer, L., and Somanathan, R., 2007. Public Action for Public Goods, in: Schultz, T.P., and Strauss, J.A. (Eds), Handbook of Development Economics. Elsevier, pp.  3117–54. https://​doi​.org/​10​ .1016/​S1573​-4471(07)04049​-1. Banerjee, A., and Somanathan, R., 2007. The political economy of public goods: some evidence from India. Journal of Development Economics 82(2), 287–314. https://​doi​.org/​10​.1016/​j​.jdeveco​.2006​.04​ .005. Banerjee, A., and Somanathan, R., 2005. History, social divisions, and public goods in rural India. Journal of the European Economic Association 3(2–3), 639–47. Bardhan, P., 1999. The Political Economy of Development in India. OUP India. Bardhan, P., and Mookherjee, D., 2017. A theory of clientelistic politics versus programmatic politics. Boston University Department of Economics, Institute for Economic Development Working Papers Series, dp-317. Bardhan, P., and Mookherjee, D., 2012. Political clientelism and capture: theory and evidence from West Bengal, India. UNU-WIDER Working Paper, No. 2012/97. Berman, P.A., and Bossert, T.J., 2000. A Decade of Health Sector Reform in Developing Countries: What Have We Learned? Paper prepared for the DDM Symposium “Appraising a Decade of Health Sector Reform in Developing Countries”. Washington, DC, 15 March 2000. Betancourt, R., and Gleason, S., 2000. The allocation of publicly-provided goods to rural households in India: on some consequences of caste, religion and democracy. World Development 28(12), 2169–82. https://​doi​.org/​10​.1016/​S0305​-750X(00)00074​-7. Bianchi, R., and Kossoudji, S., 2001. Interest groups and organizations as stakeholders. Social Development Papers 35, i–30. Birkland, T.A., 2015. An Introduction to the Policy Process: Theories, Concepts, and Models of Public Policy Making, 4th ed. Taylor & Francis. https://​doi​.org/​10​.4324/​9781315717371.

Political economy of public financing of health  365 Brugha, R., and Varvasovszky, Z., 2000. Stakeholder analysis: a review. Health Policy and Planning 15(3), 239–46. Buse, K., Mays, N., and Walt, G., 2012. Making Health Policy, 2nd ed. Open University Press. https://​ doi​.org/​10​.1163/​9789004333109. Campbell, J.L., 2002. Ideas, politics, and public policy. Annual Review of Sociology 28, 21–38. https://​ doi​.org/​10​.1146/​annurev​.soc​.28​.110601​.141111. Campbell, J.L., 1998. Institutional analysis and the role of ideas in political economy. Theory and Society 27(3), 377–409. Campos, P.A., and Reich, M.R., 2019. Political analysis for health policy implementation. Health Systems and Reform 5(3), 224–35. https://​doi​.org/​10​.1080/​23288604​.2019​.1625251. Chancel, L., and Piketty, T., 2019. Indian income inequality, 1922–2015: from British Raj to billionaire Raj? Review of Income and Wealth 65(S1), S33–S62. https://​doi​.org/​10​.1111/​roiw​.12439. Chemor Ruiz, A., Ratsch, A.E.O., and Martínez, G.A.A., 2018. Mexico’s Seguro Popular: achievements and challenges. Health Systems and Reform 4(3), 194–202. https://​doi​.org/​10​.1080/​23288604​.2018​ .1488505. Chemouni, B., 2018. The political path to universal health coverage: power, ideas and community-based health insurance in Rwanda. World Development 106, 87–98. https://​doi​.org/​10​.1016/​j​.worlddev​.2018​ .01​.023. Chhibber, P., 1999. Democracy without Associations. University of Michigan Press. https://​doi​.org/​10​ .3998/​mpub​.23136. Chhibber, P., and Verma, R., 2019. The rise of the second dominant party system in India: BJP’s new social coalition in 2019. Studies in Indian Politics 7(2), 131–48. https://​doi​.org/​10​.1177/​ 2321023019874628. Clark, M.A. (2015). The New Left and health care reform in El Salvador. Latin American Politics and Society, 57(4), 97–118. www​.jstor​.org/​stable/​24765996. Cleaves, P.S., 1980. Political Power and Policy Design, in: Grindle, M.S. (Ed.), Politics and Policy Implementation in the Third World. Princeton University Press, pp. 281–303. Cotlear, D., Nagpal, S., Smith, O, Tandon, A., and Cortez, R., 2015. Going Universal: How 24 Developing Countries Are Implementing Universal Health Coverage from the Bottom Up. World Bank. https://​openknowledge​.worldbank​.org/​handle/​10986/​22011. License: CC BY 3.0 IGO. Datta, S., 2020. Political competition and public healthcare expenditure: Evidence from Indian states. Social Science & Medicine 244, 112429. https://​doi​.org/​10​.1016/​j​.socscimed​.2019​.112429. De Costa, A., and Diwan, V., 2007. “Where is the public health sector?” Public and private sector healthcare provision in Madhya Pradesh, India. Health Policy 84(2–3), 269–76. https://​doi​.org/​10​.1016/​j​ .healthpol​.2007​.04​.004. Deshpande, R., Tillin, L., and Kailash, K.K., 2019. The BJP’s welfare schemes: did they make a difference in the 2019 elections? Studies in Indian Politics 7(2), 219–33. https://​doi​.org/​10​.1177/​ 2321023019874911. Di John, J., and Putzel, J., 2009. Political Settlements. Issues Paper, Governance and Social Development Resource Centre, University of Birmingham. Domínguez, J.I. (Ed.), 1997. Technopols: Freeing Politics and Markets in Latin America in the 1990s. Penn State University Press. Dür, A., and de Bièvre, D., 2007. The question of interest group influence. Journal of Public Policy 27(1), 1–12. https://​doi​.org/​10​.1017/​S0143814X07000591. Fenny, A.P., Yates, R., and Thompson, R., 2021. Strategies for financing social health insurance schemes for providing universal health care: a comparative analysis of five countries. Global Health Action 14(1), 1868054. https://​doi​.org/​10​.1080/​16549716​.2020​.1868054. Flöthe, L., 2020. Representation through information? When and why interest groups inform policymakers about public preferences. Journal of European Public Policy 27(4), 528–46. https://​doi​.org/​10​ .1080/​13501763​.2019​.1599042. Forster, T., Kentikelenis, A.E., Stubbs, T.H., and King, L.P., 2020. Globalization and health equity: the impact of structural adjustment programs on developing countries. Social Science and Medicine 267, 112496. https://​doi​.org/​10​.1016/​j​.socscimed​.2019​.112496.

366  Handbook on the political economy of health systems Fox, A.M., and Reich, M.R., 2015. The politics of universal health coverage in low- and middle-income countries: a framework for evaluation and action. Journal of Health Politics, Policy and Law 40(5), 1023–60. https://​doi​.org/​10​.1215/​03616878​-3161198. Fox, A.M., and Reich, M.R., 2013. Political Economy of Reform. World Bank. Frankel, F.R., 2006. India’s Political Economy: The Gradual Revolution (1947–2004), 2nd ed. Oxford University Press. Frey, A., 2019. Cash transfers, clientelism, and political enfranchisement: evidence from Brazil. Journal of Public Economics 176, 1–17. https://​doi​.org/​10​.1016/​j​.jpubeco​.2019​.05​.002. Ghebreyesus, T.A.A., 2019. Health: A Political Choice. Delivering Universal Health Coverage, 2030. Ghosh, A.R., Kim, J.I., Mendoza, E.G., Ostry, J.D., and Qureshi, M.S., 2013. Fiscal fatigue, fiscal space and debt sustainability in advanced economies. Economic Journal. https://​doi​.org/​10​.1111/​ecoj​ .12010. Giedion, U., Bitran, R.A., Tristao, I., and Inter-American Development Bank Social Protection and Health Division, n.d. Health Benefit Plans in Latin America: A Regional Comparison. Inter-American Development Bank. Glassman, A., Giedion, U., Sakuma, Y., and Smith, P.C., 2016. Defining a health benefits package: what are the necessary processes? Health Systems and Reform 2(1), 39–50. https://​doi​.org/​10​.1080/​ 23288604​.2016​.1124171. Goddard, M., Hauck, K., Preker, A.S., and Smith, P.C., 2006. Priority setting in health: a political economy perspective. Health Economics, Policy and Law 1(1), 79–90. González-Rossetti, A., and Bossert, T., 2000. Enhancing the Political Feasibility of Health Reform: A Comparative Analysis of Chile, Colombia, and Mexico. Latin America and Caribbean Regional Health Sector Reform Initiative. Govt. of India, 2011. High Level Expert Group Report on Universal Health Coverage for India. Government of India. Greer, S.L., and da Fonseca, E.M., 2016. Decentralization and Health System Governance, in: Kuhlmann, E., Blank, R.H., Bourgeault, I.L., and Wendt, C. (Eds), The Palgrave International Handbook of Healthcare Policy and Governance. Palgrave Macmillan, pp. 409–24. https://​doi​.org/​10​ .1007/​978​-1​-137​-38493​-5​_25. Grindle, M.S., 1980. Political Constraints on Rural Development in Mexico, in: Grindle, M.S. (Ed.), Politics and Policy Implementation in the Third World. Princeton University Press, pp. 197–223. Harriss, J., 2015. Hindu nationalism in action: the Bharatiya Janata Party and Indian politics. South Asia: Journal of South Asian Studies 38(4), 712–8. https://​doi​.org/​10​.1080/​00856401​.2015​.1089826. Hutchinson, E., Balabanova, D., and McKee, M., 2019. We need to talk about corruption in health systems. Int J Health Policy Manag 8(4), 191–4. www​.ncbi​.nlm​.nih​.gov/​pmc/​articles/​PMC6499907/​ pdf/​ijhpm​-8​-191​.pdf. Immergut, E.M., 1990. Institutions, veto points, and policy results: a comparative analysis of health care. Journal of Public Policy 10(4), 391–416. Jaffrelot, C., 2015. The Modi-centric BJP 2014 election campaign: new techniques and old tactics. Contemporary South Asia 23(2), 151–66. https://​doi​.org/​10​.1080/​09584935​.2015​.1027662. Jaffrelot, C., and Kalaiyarasan, A., 2019. The political economy of the Jat agitation for other backward class status. Economic and Political Weekly 54(7), 29–37. Jann, W., and Wegrich, K., 2007. Theories of the Policy Cycle, in: Fischer, F., Miller, G.J., and Sidney, M.S. (Eds), Handbook of Public Policy Analysis. Taylor & Francis, pp. 43–62. Kanjilal, B., and Mazumdar, S., 2012. Transition in the Indian Healthcare Market, in: Bloom, G., Kanjilal, B., Lucas, H., and Peters, D. (Eds), Transforming Health Markets in Asia and Africa. Routledge, pp. 18–34. Karan, A., Yip, W., and Mahal, A., 2017. Extending health insurance to the poor in India: an impact evaluation of Rashtriya Swasthya Bima Yojana on out of pocket spending for healthcare. Social Science and Medicine 181, 83–92. https://​doi​.org/​10​.1016/​j​.socscimed​.2017​.03​.053. Keefer, P., and Khemani, S., 2005. Democracy, public expenditures, and the poor: understanding political incentives for providing public services. World Bank Research Observer 20(1), 1–27. https://​doi​ .org/​10​.1093/​wbro/​lki002. Keefer, P., and Khemani, S., 2004. Why do the poor receive poor services? Economic and Political Weekly 39(9), 935–43.

Political economy of public financing of health  367 Keefer, P., and Vlaicu, R., 2008. Democracy, credibility, and clientelism. Journal of Law, Economics, and Organization 24(2), 371–406. https://​doi​.org/​10​.1093/​jleo/​ewm054. Kelsall, T., 2020. Political settlements and the implementation of maternal health policy in the developing world: a comparative case study of Rwanda, Ghana, Uganda and Bangladesh. ESID Working Paper, No. 137. Kelsall, T., Hart, T., and Laws, E., 2016. Political settlements and pathways to universal health coverage. ODI Working Paper, No. 432. https://​cdn​.odi​.org/​media/​documents/​10382​.pdf. Kelsall, T., and Seiha, H., 2014. The political economy of inclusive healthcare in Cambodia. ESID Working Paper, No. 43. https://​doi​.org/​10​.2139/​ssrn​.2539172. Khan, M., 1995. State Failure in Weak States: A Critique of New Institutionalist Explanations, in: Harriss, J., Hunter, J., and Lewis, C. (Eds), New Institutional Economics and Third World Development. Routledge, pp. 71–86. Khan, M.H., 2018. Political settlements and the analysis of institutions. African Affairs 117(469), 636–55. https://​doi​.org/​10​.1093/​afraf/​adx044. Khan, M.H., 2010. Political Settlements and the Governance of Growth-Enhancing Institutions. SOAS Research Online. Khemani, S., 2015. Buying votes versus supplying public services: political incentives to under-invest in pro-poor policies. Journal of Development Economics 117, 84–93. https://​doi​.org/​10​.1016/​j​.jdeveco​ .2015​.07​.002. Kingdon, J.W., 2014. Agendas, Alternatives, and Public Policies, New International ed. Pearson. Kohli, A., and Singh, P., 2018. Politics in India: An Overview, in: Kohli, A., and Singh, P. (Eds), Routledge Handbook of Indian Politics. Routledge, pp. 1–19. La Forgia, G., and Nagpal, S., 2012. Government-Sponsored Health Insurance in India: Are You Covered? World Bank Publications. Lavers, T., 2019. Towards universal health coverage in Ethiopia’s ‘developmental state’? The political drivers of health insurance. Social Science and Medicine 228, 60–67. https://​doi​.org/​10​.1016/​j​ .socscimed​.2019​.03​.007. Lavers, T., and Hickey, S., 2016. Conceptualising the politics of social protection expansion in low income countries: the intersection of transnational ideas and domestic politics. International Journal of Social Welfare 25(4), 388–98. https://​doi​.org/​10​.1111/​ijsw​.12210. Levy, B., 2015. Governance reform: getting the fit right. Public Administration and Development 35(4), 238–49. https://​doi​.org/​10​.1002/​pad​.1734. Levy, B., and Walton, M., 2013. Institutions, incentives and service provision: bringing politics back in. ESID Working Paper, No. 18. Lewis, D.G., and Sagnayeva, S., 2020. Corruption, patronage and illiberal peace: forging political settlement in post-conflict Kyrgyzstan. Third World Quarterly 41(1), 77–95. https://​doi​.org/​10​.1080/​ 01436597​.2019​.1642102. Loewenson, R., 1993. Structural adjustment and health policy in Africa. International Journal of Health Services 23(4), 717–30. McDonnell, A., Urrutia, A.F., and Samman, E., 2019. Reaching universal health coverage: A political economy review of trends across 49 countries. ODI Working Paper, No. 570. Miguel, E., and Gugerty, M.K., 2005. Ethnic diversity, social sanctions, and public goods in Kenya. Journal of Public Economics 89(11–12), 2325–68. https://​doi​.org/​10​.1016/​j​.jpubeco​.2004​.09​.004. Naher, N., Hoque, R., Hassan, M.S., Balabanova, D., Adams, A.M., and Ahmed, S.M., 2020. The influence of corruption and governance in the delivery of frontline health care services in the public sector: a scoping review of current and future prospects in low and middle-income countries of south and south-east Asia. BMC Public Health 20, 880. https://​doi​.org/​10​.1186/​s12889​-020​-08975​-0. Nash, R., Hudson, A., and Luttrell, C., 2005. Stakeholder Analysis, in: Mapping Political Context: A Toolkit for CSOs. ODI, pp. 26–8. Navarro, V., 1977. Social class, political power, and the state and their implications in medicine. International Journal of Health Services 7(2), 255–92. www​.jstor​.org/​stable/​45140170. North, D.C., 1995. The New Institutional Economics and Third World Development, in: Harriss, J., Hunter, J., and Lewis, C. (Eds), The New Institutional Economics and Third World Development. Routledge, pp. 17–26.

368  Handbook on the political economy of health systems North, D.C., 1990. Institutions, Institutional Change and Economic Development. Cambridge University Press. North, D.C., Wallis, J.J., Webb, S.B., and Weingast, B.R., 2012. In the Shadow of Violence: Politics, Economics, and the Problems of Development. Cambridge University Press. North, D.C., and Weingast, B.R., 2000. Introduction: institutional analysis and economic history. Journal of Economic History 60(2), 414–7. Panda, S., 2015. Political connections and elite capture in a poverty alleviation programme in India. Journal of Development Studies 51(1), 50–65. https://​doi​.org/​10​.1080/​00220388​.2014​.947281. Patel, V., Mazumdar-Shaw, K., Kang, G., Das, P., and Khanna, T., 2021. Reimagining India’s health system: a Lancet Citizens’ Commission. The Lancet 397(10283), 1427–30. https://​doi​.org/​10​.1016/​ S0140​-6736(20)32174​-7. Preker, A.S., Harding, A., and Travis, P., 2000. ‘Make or buy’ decisions in the production of health care goods and services: new insights from institutional economics and organizational theory. Bulletin of the World Health Organization 78(6), 779–90. https://​apps​.who​.int/​iris/​handle/​10665/​268156. Prinja, S., Chauhan, A.S., Karan, A., Kaur, G., and Kumar, R., 2017. Impact of publicly financed health insurance schemes on healthcare utilization and financial risk protection in India: a systematic review. PLoS ONE 12(2), e0170996. https://​doi​.org/​10​.1371/​journal​.pone​.0170996. Putzel, James., and Di John, J., 2012. Meeting the Challenges of Crisis States. Crisis States Research Centre, London School of Economics and Political Science. Rajkotia, Y. (2009). National Health Insurance in Ghana: Politics, Adverse Selection, and the Use of Child Health Services. Unpublished PhD dissertation, Johns Hopkins University. https://​media​ .proquest​.com/​media/​hms/​ORIG/​2/​31AGH​?​_s​=​K3Z3EuhhjJejrCee​%2Fh273n5sC​%2Fs​%3D. Reich, M.R., 2020. Restructuring health reform, Mexican style. Health Systems and Reform 6(1), 1–11. https://​doi​.org/​10​.1080/​23288604​.2020​.1763114. Reich, M.R., 1995. The politics of health sector reform in developing countries: three cases of pharmaceutical policy. Health Sector Reform in Developing Countries 32(1–3), 47–77. Reich, M.R., Harris, J., Ikegami, N., Maeda, A., Cashin, C., Araujo, E.C., Takemi, K., and Evans, T.G., 2016. Moving towards universal health coverage: lessons from 11 country studies. The Lancet 387(10020), 811–6. https://​doi​.org/​10​.1016/​S0140​-6736(15)60002​-2. Rizvi, S.S., Douglas, R., Williams, O.D., and Hill, P.S., 2020. The political economy of universal health coverage: a systematic narrative review. Health Policy and Planning 35(3), 364–72. https://​doi​.org/​ 10​.1093/​heapol/​czz171. Roberts, M.J., Hsiao, W., Berman, P., and Reich, M.R., 2008. Political Analysis and Strategies, in: Getting Health Reform Right: A Guide to Improving Performance and Equity. Oxford University Press, pp. 61–89. Roberts, M.J., Hsiao, W.C., Berman, P., and Reich, M.R., 2003. Getting Health Reform Right: A Guide to Improving Performance and Equity. Oxford University Press. Rochefort, D.A., and Donnelly, K.P., 2013. Agenda-Setting and Political Discourse: Major Analytical Frameworks and Their Application, in: Araral, E., Fritzen, S., Howlett, M., Ramesh, M., and Wu, X. (Eds), Routledge Handbook of Public Policy. Routledge, pp. 189–203. Saksena, P., Antunes, A.F., Xu, K., Musango, L., and Carrin, G., 2011. Mutual health insurance in Rwanda: evidence on access to care and financial risk protection. Health Policy 99(3), 203–9. https://​ doi​.org/​10​.1016/​j​.healthpol​.2010​.09​.009. Sakthivel, S., and Karan, A.K., 2012. Why publicly-financed health insurance schemes are ineffective in providing financial risk protection. Economic & Political Weekly 47(11), 60–68. Semin, S., and Güldal, D., 2008. Globalization of the pharmaceutical industry and the growing dependency of developing countries: the case of Turkey. International Journal of Health Services 38(2), 379–98. https://​doi​.org/​10​.2190/​HS​.38​.2​.i. Shearer, J.C., Abelson, J., Kouyate, B., Lavis, J.N., and Walt, G., 2016. Why do policies change? Institutions, interests, ideas and networks in three cases of policy reform. Health Policy and Planning 31(9), 1200–211. https://​doi​.org/​10​.1093/​heapol/​czw052. Shroff, Z.C., Roberts, M.J., and Reich, M.R., 2015. Agenda setting and policy adoption of India’s national health insurance scheme: Rashtriya Swasthya Bima Yojana. Health Systems and Reform 1(2), 107–18. https://​doi​.org/​10​.1080/​23288604​.2015​.1034310.

Political economy of public financing of health  369 Sparkes, S.P., Bump, J.B., Özçelik, E.A., Kutzin, J., and Reich, M.R., 2019. Political economy analysis for health financing reform. Health Systems and Reform 5(3), 183–94. https://​doi​.org/​10​.1080/​ 23288604​.2019​.1633874. Sparkes, S.P., Bump, J.B., Özçelik, E.A., Kutzin, J., and Reich, M.R., 2014. A Political Economy Analysis of Turkey’s Health Transformation Program. World Bank. Sparkes, S.P., Bump, J.B., and Reich, M.R., 2015. Political strategies for health reform in Turkey: extending veto point theory. Health Systems and Reform 1(4), 263–75. https://​doi​.org/​10​.1080/​ 23288604​.2015​.1093063. Srivastava, R., 2010. Making the case for a rights-based approach. Economic and Political Weekly 45(42), 29–31. Transparency International, 2015. Bangladesh: Overview of Corruption and Anti-Corruption with a Focus on the Health Sector. Transparency International. Tsebelis, G., 1995. Decision making in political systems: veto players in presidentialism, parliamentarism, multicameralism and multipartyism. British Journal of Political Science 25(3), 289–325. Vargas, J.R., and Muiser, J., 2013. Promoting universal financial protection: a policy analysis of universal health coverage in Costa Rica (1940–2000). Health Research Policy and Systems, 11(1). https://​ doi​.org/​10​.1186/​1478​-4505​-11​-28. Weber, C. 2019. Ideology and values in political decision making. Oxford Research Encyclopedia of Politics. Retrieved 3 May 2023 from https://​oxfordre​.com/​politics/​view/​10​.1093/​acrefore/​ 9780190228637​.001​.0001/​acrefore​-9780190228637​-e​-998. World Bank, 2008. The Political Economy of Policy Reform: Issues and Implications for Policy Dialogue and Development Operations. World Bank. World Health Organization, 2007. Everybody’s Business: Strengthening Health Systems to Improve Health Outcomes – WHO’s Framework for Action. World Health Organization.

21. Political economy of health system reform: evidence from Spain Guillem López-Casasnovas

1. INTRODUCTION The Covid-19 pandemic has been a major disruptor of health care systems and has had both conjunctural and structural effects, but it also provides new opportunities to learn from an unexpected shock. Pandemics are not an unusual feature and health systems today face a series of complex challenges, including the need to constantly update their coverage and access gaps, and address concerns about a rapidly changing quality of care and workforce shortages, as well as the duty to provide integration and continuity of care, including the epidemiological shift in needs that involve multiple chronic diseases alongside emerging viral threats. Health systems struggle with growing health costs and the social demands to find ways to bend the cost curve while designing sustainable financing schemes. Technology is increasing the demand for more and better person-centered care, challenging established approaches to service delivery, alongside quality regulation and governance. All of these goals and challenges are conditioned by an aging population and a commitment to universal health coverage. New public health problems call for refocusing public financing of health systems, and rebalancing the State’s role in providing health care. This chapter discusses the political barriers encountered in mainstream health systems that are socially financed, such as national health service (NHS)-type systems. We draw on some of the evidence of the Spanish health care sector as a case study. The Spanish case is especially interesting given the degree of decentralization to cater to territorial diversity in needs and preferences, the nature of government institutions and the frequent confusion in political narratives between public provision and public production of health care services. Furthermore, the human resources of the system are grounded on the civil servant status of its employees, yet it allows physicians to work in both public and private practices (López-Casasnovas & Pifarré i Arolas 2021). This chapter argues that reforming such a system entails overcoming political barriers driven by its unique institutional features. This chapter follows Hart and Moore (2005) and Roberts et al. (2003) in describing the various issues at stake in health system reform from a unique perspective. We follow an institutionalist political economy approach. That is, we rely on the assumption that an analysis of health care reform requires a careful understanding of the institutional and political constraints that reformers face. We specifically inquire about which reforms are likely to succeed, and what actions can be taken to mitigate potential reform barriers which we describe below. The chapter puts forward the argument that health care reforms are the result of context-specific institutional advantages resulting from the way political markets operate, which in equilibrium give rise to a certain degree of path dependency. That is, reforms encompass first and foremost the identification of political opportunities considering the political 370

Political economy of health system reform  371 demand and supply for health policies (Costa-Font et al., 2020). Such political opportunities include the role of constraints or the social barriers to reform, including overcoming ideological, distributional concerns and addressing wider social determinants of health. Political decentralization, both functional and territorial, plays a central role in the Spanish case in steering experimentation and policy innovation and transfer (Costa-Font & Rico, 2006). The Spanish experience suggests that health care reforms in public health care systems need to overcome constraints beyond their broader macroeconomic environment (e.g., economic growth, technological research and development, and industrial activity), along the lines of narratives of inequality and social cohesion. The final part of this chapter discusses the specific factors that may act as health reform deterrents, and some proposals to change the dynamics of how health care systems operate, mainly a stronger attention to outcomes and evaluation.1 The rest of the chapter is organized as follows. Next, we outline the main types of health systems. Section 3 discusses the political factors in the public provision of health care, and Section 4 reports on the main drivers and constraints to health care system reforms in Spain.

2.

GENETICS OF HEALTH SYSTEMS: MORPHISMS AND POLYMORPHISMS

Public intervention and financing of health systems has typically given rise to two types of models, namely an NHS or a social health insurance system (SHIS). In an NHS model, health is pursued by “services” managed by an administrative organization with a “national” scope that aims to be cohesive. In contrast, an SHIS model is structured based on a network of diverse suppliers. These health system designs are defined by the following features: they are “systems” in a web of cogs that accommodates this diversity of suppliers and the heterogeneity of the individual affiliations; they play the role of an insurer as they cover a limited range of limited health contingencies; and they are “social,” because they need to serve all of society (they do not serve just individuals, they are not actuarially based, they are not adjusted to risk premiums). Both structures, NHS and SHIS, are legitimized through political processes, even when policy proposals are often not met by evidence. For instance, the desired intersectorality of the NHS is often a myth, and the same is true for the elimination of any sign of adverse selection or risks in the SHIS. Reforms in public health systems should account for the political or bureaucratic processes which are country-specific, the system’s capability to manage capital, access to contingency funds and a certain autonomy to anticipate cycles, how data is interpreted and managed, and how health care coordinates with other health services of interest. That said, the distinction between the two systems is not merely semantic; it also has a logic that allows the assignment of the different basic functions of the agents involved for an effective system operation.2 Two basic parameters also distinguish these functions: the implicit weight assigned to the political and technical components in each of the systems and the management logic that feeds their rationality.3 The systems could also be distinguished in terms of the balance (between costs and benefits) of priorities. In making policy choices, one should disentangle the tendency to fall prey to political demands which follow an electoral logic, and the management logic grounded on relative needs which tends to reject new policy proposals (“learning to say no”). All the above is a testimony to the difficulty of making health systems comparisons even when one accounts for the idiosyncrasies of their citizens. Benchmarking here seems to be

372  Handbook on the political economy of health systems impossible. In public systems, the debate encompasses the classic Bismarck and Beveridge characterizations of health systems according to their features, and the models that have emerged from them: (i) financing based on direct payment, taxes, social insurance and private insurance; and (ii) the preponderance of the financing criteria with regard to service access and eligibility for use. Thus, regarding what must be understood as the public opportunity for health system reform, one needs to bear in mind the concept of the fiscal burden against the private social cost resulting from public regulation. In general, the genetics of the different systems can be known from verifying how they have arranged the key functions of any health organization in terms of financing and service provision and production. While by no means an exhaustive explanation, what follows is illustrative of a system’s “polymers” (networks), “base molecules” (care functions), “cells” (centers) and “proteins” (incentives), enabling us to distinguish between systems that correspond to different trajectories, cultures and ideologies so as to avoid the consideration that the extremes of some are clearly “superior” to those of the rest. For this, the analysis of health systems can give rise to a few basic drivers: (i) those based on regulation, which confer responsibilities on third parties and, subsidiarily, only maintain public social security networks for the poor and the elderly, as in the case of the United States; (ii) those based on public provision and production that are structured in the form of national health services, as already spoken about, and treat health services as simply another administrative service; and (iii) those that opt for the social insurance form of health care, the so-called public provision and private production systems, frequently found in Continental European models. Considering the above, it is remarkable that, regardless of the historical paths of each health system, health service organizations in the majority of Western countries are undergoing some significant reform, which calls for an understanding of the drivers and barriers to reform.

3.

CONSTRAINTS TO THE REFORM OF PUBLIC HEALTH SYSTEMS

3.1

The Goals and Limits of Public Health Care Systems

Implicit in public provision of health care is the idea that health care is both a public responsibility and a financing obligation that must be addressed by the political decision-making process. A more extreme version refers to public production models (care delivery by civil servants), referring to the full integration of the health care value chain, namely when the public sector takes responsibility for every aspect, from planning to production of services as well as the definition and purchasing of the package of health care services and its allocation across the territory. The effectiveness of both public provision and production systems is tied to the possibility of integrating health care with no other obstacles than self-established policy directives and specific regulations and memos driving its implementation. The main advantage of such health systems lies in the role of collective purchasing of health care, which enables keeping costs down, and the pursuit of some uniformity in the provision of health care, even though this entails limited choice, common basic coverage of health care and an exclusive system for monitoring compliance. The latter are the common limitations of such systems.

Political economy of health system reform  373 3.2

The Misalignment of Incentives, Responsibilities and Surveillance

Health systems that rely on the public provision and production of health care have to confront the problems of a widespread misalignment of incentives, responsibilities and the factual management of the system. This feature is the result of the political incentives in place. The political processes produce a unique prioritization of health services, where the definition of specific services and the planning of the different needs (including the specific needs of the most fragile groups) are often misaligned. Such unique prioritization results, at times, independently from political attitudes in society which, in contrast, do not show evidence of major disagreement in the needs to be addressed. The origins of such misalignment can be tracked back to the fact that the political demand for health care is muddled by major institutional constraints such as the separation of health care responsibilities into the remit of different departments (e.g., the department of health, the treasury, the department of social care). Another set of constraints includes features that are more technical in nature, such as the inability of some health systems to define budgets that are specific to each health care program. Similarly, as described in Chapter 14 of this volume by Counts and Ilakkuvan, the demand for certain health programs is heavily influenced by lobbies and interest groups (e.g., the pharmaceutical industry, colleges of physicians), which explains why more efficient health care reforms do not always end up being implemented, or when they do, the implementation might be significantly influenced by the priorities of different interest groups rather than follow the wider public interest. 3.3

Limited Attention to Outcomes

An important limit to health care reforms comes from the relatively weak political influence of departments of health compared to other departments that make up different cabinets. Indeed, the political demand for health care reform often needs to confront the mismatch between health care reform narratives which are grounded on public interest and a population’s health objectives, alongside the operational narratives of health care bureaucrats which are very much grounded on health care activity goals. Typically, information systems play very limited attention to how health care reform proposals impact a population’s health. That is, the opacity of health care systems based on activity becomes an important limit to the capacity of patient citizens to judge the efficiency of health care reforms, which instead are grounded on improving health outcomes. Public systems typically face significant barriers in monitoring the effective access to basic health care to which people are entitled. While traditionally control systems focus on the cost of health care inputs, the logics of health care financing reforms are instead defined incrementally, so that outcomes are evaluated by each health program. 3.4

Equity in Access as a Barrier to Reform

Another important feature of publicly funded systems is the emphasis placed on their distributional effects. Indeed, “equity” in public health system provision is frequently the ultimate political argument underpinning NHS reforms, even when such equity is frequently inadequately defined, let alone measured or evaluated. However, even when policy proposals are budget-neutral (e.g., the introduction of small and capped co-payments), equity concerns can put a halt to them. Instead, political debates label them as attempts to privatize health care

374  Handbook on the political economy of health systems systems. Prejudices, economic interests, ideological biases and mediocre analysts plague such privatization concerns. Hence, health reforms must take into account the narratives, ideologies and prejudices of each country’s policy reform environment (see Theilen, Chapter 13 in this volume; Costa-Font et al., 2020). The prevalence of equity concerns is pervasive, even though the pursuit of health system equity differs from equality in access or provision of health care because fairness is a principle guiding policy rather than the result of the implementation of a mathematical algorithm. So far, we know too little about whether policy should be based on “equity of access” (not to be confused with equity in health care financing) or equity of outcomes. This distinction is significant because what is accessible is not always consumed, and what is consumed does not always result in the same outcome. The latter can be explained by differences in information access, particularly how rationing by waiting list works alongside a number of non-monetary barriers that are frequently overlooked. Indeed, what is effective in theory is frequently ineffective in individual and actual terms due to other unaccounted constraints (habitat, social support, loneliness, unemployment). In contrast to egalitarianism in access to care, full egalitarianism in outcomes might not be desirable in a democratic society insofar as it requires a degree of interference in individual decision-making, privileging collective goals above legitimate personal ones, hence restricting individuals’ liberties and civic rights. So, often, health policy in NHS systems is content to erect barriers and allow access by “get in line for what is available in the kitchen as many times as you are in the queue.” To understand how “equity” works in health care reform, one must look beyond the role of health care systems in providing financial protection against the catastrophic consequences of health care use. Although the distribution of private health expenditure (resulting from out-of-pocket spending, insurance premiums and co-payments) is easily observable by the patient citizen, it seldom serves as evidence to judge the distributional effects of health system reforms. Private spending analysis is only a proxy inequity if all observed differences in private spending are deemed “inequitable” or as resulting from otherwise “unmet needs.” However, this would only hold true if spending does not result from individuals’ choice, nor amenities or “wasteful utilization” which do not necessarily reflect an unequal use of the public health system. However, if the health system package of services is the result of a careful financial assessment and prioritization, the private health care expenditure on treatments outside such a catalogue reflects health care of lower value to individuals, and hence likely not to give rise to unfair inequalities. 3.5

Cost-Sharing and Private Practice as a Barrier to Health System Reform

One of the barriers of health system reform lies in the use of co-payments. This is a highly political matter, though as I will argue below, it is based on ambiguous arguments. Certainly, one should distinguish between cost-sharing and the use of user changes. Among the first we find the use of co-payments, for example, on effective treatments where the health system passes its costs down to its users, as is common with regard to new dental or mental health treatments in most public health systems. However, a co-payment exerts a similar effect to a tax deduction on private health insurance to increase the perceived quality of care (see Lucarelli & Pauly, Chapter 19 in this volume), and both refer to public choices regarding what is a “merit good,” and whether the patient citizen should pay the full cost of care or get

Political economy of health system reform  375 compensated for services offered at a lower quality of care. However, cost-sharing might well depend on the value of health care to a health system, and hence if a treatment is effective but at a high cost, the system may indeed abolish cost-sharing and regulating health care might be welfare-improving to ensure access by lower-income users.4 A similar phenomena on the supply side refers to the extent to which physicians should be able to work simultaneously in their private and public practices. The risk with the latter is creating a two-tier system where some individuals have better access to health care by going private.

4.

CONSTRAINTS TO HEALTH SYSTEMS REFORMS IN SPAIN

The nature and evolution of the health system in Spain has been characterized by the expansive universalist coverage. Universal coverage is recognized in the 1986 General Health Act, although it existed long before. That said, the economic crisis has occasionally questioned its foundations (the last unsuccessful reform in 2008 tried to limit the amount of coverage given to illegal immigrants). However, for a quality evaluation of the changes undergone by the health system in Spain, one must not get bogged down in conjunctural details. Considering the influence of the abovementioned political and institutional constraints on health system reform, the most outstanding institutional constraints to understand Spanish health system reform include the following. The Spanish health system is defined by the decentralized nature of its health care organization and a uniform expenditure per capita, which requires self-sufficiency in health care provision irrespective of the size of the autonomous region (even though populations range from 250,000 to 8 million people). The budget size of the local demand for health care explains why most regional-level jurisdictions are eager to make their own spending decisions irrespective of their fiscal responsibility (not having to contribute from their own revenues) (see López-Casasnovas, Costa-Font & Planas-Miret, 2005). Despite its decentralized nature, the health system’s autonomy is severely limited by political restrictions and framework legislation regulating the appointment of high-ranking officials, publicly setting caps to professional salaries, and heavy administrative intervention in expenditure processes that is not efficiency-driven. Although the health system is tax-based and follows the steps of an NHS-type system, it is still heavily rooted in a former social security model, whereby health care provision is paid for by affiliated employees, but that since the inception of democracy in the late 1970s offers a universal right to health care. Regional inequalities are persistent and refer to key dimensions of the service provision across regions (such as access, time and quality). However, such disparities become a double-edged sword, and although politically are presented as a rupture in the social cohesion goals, such narratives ignore the fact that at the same time they are an opportunity for policy innovation compared to a monolithic and unfirm provision of health care (see López-Casasnovas & Rico Gomez, 2000).5 Some regional health systems in Spain are rooted in historical pathways and culture, and have been structured alongside service providers other than strictly administrative bodies (such as health foundations, public–private consortiums and some social companies), led by civil society movements. In Catalonia, 60 percent of hospital beds are publicly financed and privately (out of the public system) owned. This is also the case but to a lesser degree in the Balearic Islands, the Basque Country and Navarra. However, empirical estimates suggest

376  Handbook on the political economy of health systems that devolution has not increased regional inequalities, unlike in other health systems such as the UK’s (Costa-Font & Perdikis, 2021). Indeed, in Spain one can observe evidence of policy interdependence fostering both collaboration and political differentiation between the central and regional governments.6 That said, some political discourses in Spain have been designed to report territorial differences in access to services, waiting lists and quality of care. However, they do tend to turn a blind eye to such things existing within regional disparities (López-Casasnovas et al., 2005). Spanish health care reform has been “trapped” in its status quo, which is commonly defined as path-dependent. Consistently with political market behavior, health system reforms have traditionally been defined by interest groups, generally professional corporations, trade unions and a number of key vested interests (industry, academy, policymakers, groups of patients). The final reform of the health system has resulted generally from negotiation and exchange between governments of the same and different jurisdictions with specific vested interests. Among them, health professionals are central to the system. Physicians are civil servants who receive salaries that are meagre in comparison with their counterparts in neighboring countries, though in line with median Spanish standards of living.7 These private–public dual employment arrangements are especially widespread in cities. In the public sector, professionals receive greater recognition from citizens and in some cases benefit from “diverting” the flow of patients towards private practices. Professionals in the public sector make their decisions assuming no financial risk from them. The exceptions to the rule are some jurisdictions that have experimented, albeit tentatively, with empowering general practitioners under capitated financing. Although the current consensus among all stakeholders is that the Spanish health system is heavily underfunded. Such consensus is dissipated when it comes to agreeing on how to attract extra finance to the system, either from taxes, complementary premiums, co-payments or direct prices. Typically, the financing conundrum is solved by raising taxes, as tax payments are perceived as more equitable, while co-payments are classified as regressive financial instruments. All the above is at the root of the difficulty in implementing financial reforms, leaving the system in a situation where the new is thwarted from being born and the old refuses to die (Congreso de Diputados, 2022). The latter explains why path dependency prevails, and reforms need to confront wider power competition that is grounded on a status quo (Rico & Costa-Font, 2005). This explains the difficulty in making changes (Artells 2014) and calls for balancing out the need of financial sustainability and the joint public and private responsibility. One way to address the latter is by pre-empting the effects of political health care markets, as we discuss in the following section.

5.

THE POLITICIZATION OF THE HEALTH SYSTEM

One of the features of the Spanish health system is its extreme politicization right from the outset. Health care managers are usually appointed based on their political affiliation rather than their technical competence, and often lack sufficient standing to drive or redirect actions (López-Casasnovas et al., 2005). The territorial decentralization is part of the political market for health care. Each regional government tries to place its health managers to serve the purpose of eroding its political adversaries. Coordination among them, and hence with the central government, is harder,

Political economy of health system reform  377 and when it takes place is commonly determined by the political leanings of each regional government. As a result, a coherent health system reform entails large transaction costs and improvisation; not only in determining what actions should be pursued, but also in quantifying its effects and impact. An illustrative example of the latter in Spain was the extensive decentralization that took place after 2001 as a political strategy to water down the differences between territories that since 1981 had been responsible for their regional health care. Indeed, the move was an attempt to implement a more uniform decentralization despite the fact that this would limit policy innovation and best practice diffusion that typically take place when regions engage in “learning by doing.” Rather than evaluating the effects of decentralization, regional autonomy and potential inequalities have been placed at the core of the political competition among mainstream parties, irrespective of the existing evidence in place. Hence, health system reform seems to result from proposals of political manifestos, with political narratives serving different interest groups as opposed to being driven by evidence of the impact of decentralization on efficiency or equity. That is, health system reform is limited by the politicization of the system itself.

6.

LESSONS LEARNED SO FAR

Although there is wide consensus among policy academics in favor of the evaluation, prioritization and cost effectiveness analysis of different reforms, political decision makers tend to prefer the opacity of a limited evidence-based policy, to a system based on a “fourth hurdle” where policies are systematically evaluated by health technology agencies. Indeed, politicians are reluctant to face up to the results of independent policy evaluation. Furthermore, when it comes to the analysis of health care priority setting (based on such evaluations), they fear it will be perceived as a negative discrimination against those who were “non-prioritised.” Similarly, in countries like Spain, the management of the health systems and specifically the appointment of staff is still subject to several legal barriers which make reform very inflexible. Still, to date, over-legalism and bureaucratization of the health system prevails over efficiency considerations. Primary health care and public health and the importance of the quality of hospital care are fundamental to attain value in the financing of the health sector. Among the obstacles to these developments, one can cite the dangers of traditional political responses such as study committees, consultation committees, national observatories, administrative secretaries and national centers of experts, all hurdles to the implementation of reform. Health care reform fails when the set-up dilutes more than it concentrates. When it passes the ball around rather than taking a shot at goal, which ends up with representatives of the status quo sitting alongside legitimate experts to ensure “that everyone is represented.” Often the outcome is that of seeking a consensus that gives rise to euphemistic proposals involving more investment. Such processes are at odds with a reliance on commission groups, namely a well-identified group of experts who, driven by prudence (and recommendations of “Do it, or explain why you are not doing it”), make an advisory ruling once the costs and results have been evaluated. For this, a draft of supposed consensuses is typically not required. Finally, it is worth mentioning that policy wishes are often confused with realities and aspirations acknowledged as being highly unlikely to be satisfied, despite commanding wide support bases.

378  Handbook on the political economy of health systems A recent example of the latter has been the Commission for Social and Economic Reconstruction of the Spanish State’s upper house of parliament, which has been gathering recommendations to provide a response to the current and future challenges of Covid-19 and its innumerable consequences. The initial worry of “it must not happen again” has now dissipated, and the temptation of the economic and political status quo has converged, once again, in short-termism. The Covid-19 pandemic has made it harder for health system reforms to rely on co-payments and has turned a blind eye to the growing weight of private health insurance and potential inequalities in access to health care. Cost-sharing arrangements are not “politically” affordable given existing political narratives. Health system reform calls for changing narratives and reaching broader consensus where concepts such as “better governance,” or “effective management,” are not just administrative reforms, but policy goals that are commonly not attained given the reform barriers in place.

NOTES 1. For more extensive detail, see López-Casasnovas and Pifarré i Arolas (2021, 2022). 2. Clearly, we are referring to who must be responsible for (i) health planning, (ii) financing, (iii) insurance coverage and (iv) the purchase and supply of services. 3. In general, a gradient of political and technical content can be established for each function, ranging from planning and financing (much more political) to service provision. 4. At any rate, in a public health system the ultimate responsibility always rests in the hands of political decision makers rather than private insurers. The latter are profit-driven and make a business of improving and complementing the package of care and quality that public health insurance schemes provide, even though private providers might influence the support for the public health systems (Costa-Font & Jofre-Bonet, 2008). 5. That said, Costa-Font and Ferrer-i-Carbonell (2021) show that the decentralization of the public health care system does indeed reduce the use of private health care and insurance, and generally strengthens the use of public health care, and can potentially reduce regional inequalities. 6. Some variants of the public–private collaboration have been adopted in some regions, most notably in the communities of Madrid, Valencia and Galicia. 7. Although with high variance, they are reasonable when adjusted regarding full- or part-time commitment, productivity and moonlighting in private health care.

REFERENCES Artells, J.J. (2014). La reforma del sistema sanitario español, entre lo deseable y lo posible. Delphi de stakeholders. Fundación Salud, Innovación y Sociedad. Fundación Novartis Barcelona. Congreso de Diputados (2022). Proyecto de Ley por la que se modifican diversas normas para consolidar la equidad, universalidad y cohesión del Sistema Nacional de Salud. www​.congreso​.es/​es/​iniciativas​ -organo​?p​_p​_id​=​iniciativas​&​p​_p​_lifecycle​=​0​&​p​_p​_state​=​normal​&​p​_p​_mode​=​view​&​_iniciativas​ _mode​=​mostrarDetalle​&​_iniciativas​_legislatura​=​XIV​&​_iniciativas​_id​=​121​%2F000110. Costa-Font, J., & Ferrer-i-Carbonell, A. (2021). Does Devolution Alter the Choice of Public Versus Private Health Care? Working Paper 1291, Barcelona School of Economics. Costa‐Font, J., & Jofre‐Bonet, M. (2008). Is there a “secession of the wealthy”? Private health insurance uptake and national health system support. Bulletin of Economic Research, 60(3), 265–87. Costa-Font, J., Levaggi, R., & Turati, G. (2020). Managed competition during pandemics: Lessons from the Italian experience during Covid-19. Health Economics, Policy and Law. e-pub ahead of print. doi:10.1017/S1744133120000353.

Political economy of health system reform  379 Costa‐Font, J., & Perdikis, L. (2021). Policy interdependence and the models of health care devolution: “Systems or federacies”? Regional Science Policy & Practice, 13(3), 492–500. Costa-Font, J., & Rico, A. (2006). Vertical competition in the Spanish national health system (NHS). Public Choice, 128(3), 477–98. Hart, O., & Moore, J. 2005. On the design of hierarchies: Coordination versus specialization. Journal of Political Economy, 113(4), 675–702. López-Casasnovas, G. (2015) El Bienestar Desigual. Editorial Península. López-Casasnovas, G., Costa-Font, J., & Planas-Miret, I. (2005). Diversity and regional inequalities in the Spanish system of health care services. Health Economics, 14(S1), 5221–35. López-Casasnovas, G., & Pifarré i Arolas, H. (2022). Health care systems: Organization and response to COVID-19 with a focus on Spain. Forum for Social Economics, 51(2), 175–91. López-Casasnovas, G., & Pifarré i Arolas, H. (2021). “The Spanish Health-Care System”, in Baltagi, B.H., and Moscone, F. (Eds) The Sustainability of Health Care Systems in Europe (Contributions to Economic Analysis, vol. 295), Emerald Publishing, pp. 189–207. López-Casasnovas, G., & Rico Gomez, A. (2000). Evaluación de las políticas de servicios sanitarios en el estado de las autonomías. Fundación BBVA Madrid. Rico, A., & Costa-Font, J. (2005). Power rather than path dependency? The dynamics of institutional change under health care federalism. Journal of Health Politics, Policy and Law, 30(1–2), 231–52. Roberts, M.J., Hsiao, W., Berman, P., & Reich, M.R. (2003). Getting Health Reform Right: A Guide to Improving Performance and Equity. Oxford University Press, Oxford.

22. The pharmaceutical patent system and access to medicines Valbona Muzaka

1. INTRODUCTION So absolute had its defence of intellectual property in global fora been that the recent announcement by the US that it would (partially) support proposals to suspend intellectual property (IP) rules for Covid-19 products and technologies during the pandemic was met with surprise everywhere. The proposal in question, made initially by India and South Africa in October 2020 at the World Trade Organization (WTO), and later supported by the vast majority of developing countries, was aimed at addressing shortages and numerous other problems associated with producing and procuring Covid-19 vaccines, diagnostic tools and therapeutic treatments. Because it implied that these problems were linked to, if not caused by, the high levels of IP protection – including patents, copyrights, trade secrets and other forms – afforded worldwide to pharmaceutical companies by the WTO Trade-Related Aspects of Intellectual Property Rights (TRIPS) Agreement, it was vehemently opposed by these companies and many developed countries, especially the US and UK, and the European Union (EU), that had helped bring TRIPS about. The U-turn performed by the US administration on this matter was surprising but has not materialised in real action yet; at the time of writing, negotiations over the TRIPS waiver continue at the WTO while the pandemic rages, Covid-19 vaccines continue to be in short supply and TRIPS protection standards remain untouched. Its achievement to date is that of representing the first time the US state has admitted in an international forum that there may be a problematic relation between IP protection standards as currently construed and access to medicines (including vaccines, therapeutics, etc.) around the world. The aim of this chapter is to explore this problematic relation from a political economy perspective, focusing on the pharmaceutical patent system. The WTO waiver encapsulates all IP forms relevant to the development, production and procurement of Covid-19 treatments and technologies; pharmaceutical patents thus represent only one element of the proposal, but they are essential to the issue of access to medicines and, indeed, to the proprietary pharmaceutical industry business model. So much so that proprietary pharmaceutical companies have probably done more than any other business sector to shape domestic and – following the emergence of the WTO TRIPS Agreement in 1995 – global patent regulation in their favour (Sell 2003, Muzaka 2021). Because they afford the owners the key privilege of setting prices, pharmaceutical patent rights are at the same time central to pharmaceutical companies’ profits and to patients’ and/or healthcare systems’ ability to access patented medicines, this conflict being central to the relationship between patents and access to medicines discussed in this chapter. Certainly, healthcare systems around the world are complex and heterogenous; nevertheless, affordable and equitable access to medicines is an indispensable part of their proper functioning everywhere, especially in low- and middle-income countries (LMICs) where up to 67 percent of total health expenditure is accounted for by expenses on medicines, mostly paid 380

The pharmaceutical patent system and access to medicines  381 out-of-pocket by consumers (Bigdeli et al. 2015). As we shall see, access to affordable medicines is not limited to this part of the world and affects patients in developed countries, too. Debates about access to medicines cover various issues, depending on who is debating and where; this chapter focuses on two of the main ones that are directly linked to the current pharmaceutical patent system, namely the high prices of patented medicines in world markets, and the relative absence in such markets of medicines that treat conditions predominantly affecting populations with low or no purchasing power. To do so, the next section provides an account of the state of the current pharmaceutical patent system, placing it in a historical context that stretches back to the late 19th century when both the modern pharmaceutical sector and the international patent system took their present shape. The most important change in the last century has been the coming into force of the WTO TRIPS Agreement in 1995 which ushered in a global pharmaceutical patent system whose main characteristics will also be covered in this section. The third section discusses the main arguments put forward in defence of the current pharmaceutical patent system and its many shortcomings, focusing on the two that are the focus in this chapter. Both the issue of medicines for neglected diseases and that of the unaffordability of patented medicines ushered in an ‘access to medicines’ campaign as soon as TRIPS entered into force. A considerable number of state and non-state actors mobilised in the late 1990s to counter what were seen as some of the most problematic aspects of TRIPS on medicines and public health, and did so with some success. But whatever their achievements, they did not rectify the fundamental changes to the pharmaceutical patent system that TRIPS had just brought about. Before the Covid-19 pandemic struck, the most notable element of this campaign was the continued efforts to negotiate a new medical R&D Treaty at the World Health Organization (WHO). Like the WTO TRIPS waiver proposal by India and South Africa in October 2020, this modest proposal was aimed at dealing only with medicines for neglected diseases. Nevertheless, again like the waiver proposal, it was seen as a threat to the current pharmaceutical patent system, which is why both proposals are considered briefly in the fourth section, embedding them in the broader ‘access to medicines’ campaign and its aftermath. The overall argument in the chapter is that the current pharmaceutical patent system is no longer justifiable.

2.

THE PHARMACEUTICAL PATENT SYSTEM BEFORE AND AFTER TRIPS

The nature of the current pharmaceutical patent system, and that of the patent institution of which it is part, is the outcome of a long series of social, economic and political factors whose account can stretch back to the monarchies and guilds of medieval Europe and, indeed, as far as ideas of intangible property in Roman law (Drahos 1996, May & Sell 2006). For our purposes, suffice to say that it was in the latter part of the 19th century that the patent system, the private corporation and the pharmaceutical sector began taking their modern form in Europe and in the US. Although significant legal, medical and technological advances had been gradually preparing the stage for its emergence, it was the development of the modern physical and life sciences in the 18th and 19th centuries and legal reforms regarding the administration of medicines in Europe in the 19th century that led to the shift away from the small, family-run apothecaries towards the emerging modern pharmaceutical companies (Braithwaite & Drahos 2000). One important turning point was the discovery in the mid-1800s that some simple

382  Handbook on the political economy of health systems organic chemicals derived from coal tar had important medical properties; the discovery and development of such chemicals became the building blocks of the chemical, dyestuff and pharmaceutical companies, primarily located in Germany and Switzerland and, to a lesser degree in France and the UK. By the end of the 19th century and up until World War I, the strength of the German and Swiss chemical-pharmaceutical companies – among them Hoechst and Bayer (German), to whom the emergence of the sector is attributed, and Ciba, Sandoz and Geigy (Swiss) – was undisputable (Achilladelis 1999). The patent institution was already a well-established one in Europe and in the US by the time the modern pharmaceutical companies emerged. Patents were privileges granted to individual inventors (or someone who introduced an imported invention), allowing them limited monopoly on the manufacture, use or sale of the invention (a process or a product) for a certain period of time (seven years was typical). As the European economic centre moved away from Italian city-states towards England, the refinement of the first patent statute developed in Venice in 1474 followed the same route and materialised in a number of legal innovations that, aiming primarily to proscribe the abuse of royal monopolies – a widespread practice in England and other European monarchies – forbade all monopolies with the exception of inventions on condition they encouraged the development or importation of innovations that benefitted English trade (Machlup 1958, Braithwaite & Drahos 2000). From these rudimentary beginnings, relatively distinct patent systems developed in various European countries and in the US through a mixture of norm borrowing and their adaptation and contestation in specific socio-economic and legal contexts. As economic competition among states grew with the onset of the Second Industrial Revolution, important changes in socio-economic institutional forms occurred more or less simultaneously (though not in the same manner) in them. The most relevant for our purposes is the rise of a new corporate model in Germany and the US that was predicated as much on industrial research, production and marketing as it was on the management of IP (May & Sell 2006). In light of the high growth prospects of sectors where this corporate model initially took hold – the electric and chemical ones – the US and the German state responded relatively rapidly (in 1871 and 1877, respectively) to corporate demands to amend patent laws so as to grant them the right to own IP which up to then was granted only to individuals (Fisk 2003). Effectively, corporations were granted IP rights for the work created collectively by their employees, a fundamental change that would have important implications for the pharmaceutical patent system. Prior to the emergence of the modern pharmaceutical companies in the late 19th century, nearly all inventions in the field were introduced by European academics and physicians, or in close collaboration with them (Achilladelis 1999). From the beginning of the 20th century onwards, the vast majority of patents – in pharmaceuticals and other sectors – came to be owned by big companies that since then have routinely used them as strategic business tools to control markets, limit competition and extend technological rents (Drahos & Braithwaite 2002). Although well established by then, the patent system offered different levels of protection to pharmaceuticals (and other sectors) across countries by the late 19th century. Interestingly, countries with strong pharmaceutical sectors – Germany and Switzerland – had offered weak if any IP protection for the chemical and dyestuff sector from where the modern pharmaceutical companies emerged and, thereafter, only offered patent protection for pharmaceutical processes (i.e., not products),1 eventually introducing full patentability for both in 1967 and 1977, respectively (Boldrin & Levine 2008). The variegated level of patent protection did

The pharmaceutical patent system and access to medicines  383 not preclude an aggressive patenting strategy by these companies in countries that offered stronger patent protection. The US, for instance, offered protection for both product and process patents for which the German chemical-pharmaceutical companies were grateful. Finding its chances of growth severely curtailed by the dominance of German companies, the American Pharmaceutical Association (unsuccessfully) lobbied to weaken US patent protection for chemicals and pharmaceuticals in the early decades of the 20th century. Publicly, the Association appealed to the moral argument that patents had to be issued primarily to enhance public benefit, an argument promptly abandoned when, following the demise of the German chemical and pharmaceutical industry during the two world wars, the US pharmaceutical sector emerged as a key player in the world pharmaceutical market. Companies’ concerns over variegated levels of patent protection across countries and particularly over the explicit prohibition of such protection to foreigners was one of the key reasons for their political mobilisation in the early 1880s, by all accounts a successful effort that concluded with the first ever international convention on patent protection, the 1883 Paris Convention for the Protection of Industrial Property (Drahos & Braithwaite 2002). It was successful especially because it included the national treatment principle according to which signatory states could not discriminate against foreigners. The 1883 Paris Convention ushered in a new international era of patent protection based on the principles of non-discrimination and national treatment, but signatory states still retained significant discretion over patent standard-setting and enforcement: by and large, they could design their own patent system as long as they did not discriminate against foreigners. This is one of the reasons why pharmaceutical patent protection continued to vary significantly for most of the 20th century; in the late 1980s, out of 98 members of the Paris Convention, half excluded pharmaceutical products from protection altogether and when they did offer it, duration varied from five years in some developing countries to 15–17 years in developed ones (WIPO 1988). This would constitute one of the main reasons why the proprietary pharmaceutical companies, especially the US ones, would mobilise alongside other high-tech sectors during the late 1970s and 1980s to secure stronger, global protection standards of the kind eventually enshrined in the 1995 WTO TRIPS Agreement. A final and equally important component of what would become the pharmaceutical patent system in the US and Europe in the 20th century related to the fundamental change in the role of the state post-World War II. The state had been crucial for the patent system since its rudimentary beginnings; whether patent privileges were granted by the monarch or, later, by an independent public authority, they could only exist and be upheld by the power of the state. The fundamental change did not occur in this aspect, although the legislative and enforcement arms of the state certainly expanded over time as far as IP provisions were concerned. Rather, it was in the central role the state was granted in guiding the process of socio-economic development post-World War II, although such role found notably different manifestations in practice. Two particular aspects stemming from this role are important here: the first relates to the substantial public investment in R&D, and the second to the institutionalisation of public healthcare systems. No account of the pharmaceutical sector or of the pharmaceutical patent system would be complete without taking fully into account the substantial contribution made by academia and major support from public funds: the undisputable success of the German pharmaceutical companies early on, the rise of US pharmaceutical sector post-World War II, and the re-emergence of the European pharmaceutical sector during the 1950s and 1960s would simply not have been achieved without the transfer of large sums of public

384  Handbook on the political economy of health systems money to fund basic research and close cooperation with scientists and academic institutions (Achilladelis 1999, Ramirez & Tylecote 1999). As will become clear later on, this contribution weakens companies’ justification for patent exclusivity as central to pharmaceutical R&D. Substantial as this contribution was and remains, it must be measured alongside the even larger contribution the emergence of socialised medicine and public healthcare systems in Europe and to a lesser extent in the US have made to pharmaceutical companies’ profits. Initiated as much as a response to the ‘evils’ of industrialisation as to the fears of revolutionary movements that were brewing in Europe in the 19th century, the expansion of early public health measures into the universal healthcare systems of Europe, and the widespread – though not universal – private health insurance in the US in the 20th century, ensured a guaranteed market for medicines (Rosen 1993). Unlike any other industrial sector, public funds underwrote the growth of the pharmaceutical sector and shielded it from the vagaries of business cycles; moreover, in countries with universal healthcare systems, it also occulted the most pernicious effects of granting patents on medicines: patients rarely knew and almost never paid in full the price of the medicine they needed, and patent and non-patented medicine prices were regulated (Danzon & Keuffel 2005). Where this has not been the case – in the developed world the most notable case is that of the US – the problematic aspects of the pharmaceutical patent system have also been clearest. The most important ones are high prices and lack of affordability: it is well known that US medicine prices are among the highest in the developed world and these are even higher for uninsured patients, a good number of whom go without because they cannot afford them. This and other shortcomings such as anticompetitive behaviour, malpractices and rent-seeking behaviour have been the subject of a number of investigations, reports and bills2 that have had little success in reforming the pharmaceutical patent system in the US. This failure, while bad news for US patients, would come to affect patients around the world because it was largely the US patent system that would provide the basis for what became the patent provisions of the WTO TRIPS Agreement. The reasons for this are many, but a key explanation can be found in the manner in which the crisis that hit the developed world in the 1970s was framed in US policymaking circles at the time. Essentially, the problem was largely framed as a loss of international competitiveness, opening the way for a dominant ‘solution’ to emerge that placed innovation and high-tech sectors at the forefront (Sell 2003). The demands of these sectors became politically more expedient, the most important of which related to strengthening IP protection in both the US and, importantly, abroad. The push to secure the latter could not be achieved without the former. The most radical changes occurring in the US patent landscape during this period include (1) the 1980 Supreme Court Diamond v. Chakrabarty decision, which expanded patent protection to genes and living matter modified by the emerging biotechnologies; (2) the 1980 Bayh–Doyle Act that opened the way to the appropriation and commercialisation of hitherto freely available academic research; (3) the creation in 1982 of the specialised Court of Appeals for the Federal Circuit (CAFC), which greatly expanded the reach of IP by lowering standards for patentability and expanding the definition of patentable subject matter to new areas; and (4) the 1984 Patent Term Restoration and Competition Act (Hatch–Waxman Act). The latter was a particularly important piece of legislation for proprietary pharmaceutical companies since, among other things, it extended the pharmaceutical patent term to compensate for regulatory ‘delays’ and offered a new type of protection, pharmaceutical data exclusivity, allowing originator companies to exclude any third party (e.g., a generic firm) from using pharmaceutical research data for up to five years.

The pharmaceutical patent system and access to medicines  385 Securing stronger IP protection and enforcement in the US was important, but not enough. Achieving the same across the world became a key demand of US high-tech (and entertainment) sectors whose claims of ‘theft’ and ‘piracy’ of their IP abroad was in sync with broader arguments that put the blame for US economic malaise on other countries playing ‘unfairly’ (Sell 2003). Armed with seemingly irrefutable arguments about ‘theft’ – legally invalid, for international law permitted variation in patent protection – and grotesquely inflated figures that put annual losses to the US high-tech sector abroad in the range of US$43–61 billion (Emmert 1990: 1327), these sectors were successful in getting the US state to support their agenda internationally, the most successful achievement being the 1995 WTO TRIPS Agreement that would singlehandedly usher in the much more restrictive global IP regime of today. Irrespective of the very different socio-economic conditions prevailing in them, most countries lost whatever policy spaces they had enjoyed until then in designing their own IP systems and were obliged for the first time in history to respect the high IP protection standards demanded by TRIPS. For the first time, too, this new IP regime came equipped with the binding dispute settlement and enforcement mechanism of the WTO. It is now a well-recognised fact that the large US pharmaceutical companies, alongside the European and Japanese ones, were a particularly active group in the coalition of high-tech business that mobilised for TRIPS (Sell 2003) and achieved the most through it. Seeking to rectify the variegated levels of patent protection worldwide, TRIPS establishes a twenty-year protection for products and processes granted to the owner of patents that prevents anyone (including inventors, if they are not the owners) within the granting territory from exploiting the protected process or product. The current pharmaceutical patent system is not global in the sense that once a patent is granted in one country, it is valid in all countries; patents have largely remained territorial in nature, but the obligation to protect pharmaceutical products and processes is legally binding on all the 164 countries party to the Agreement. This was a significant achievement for the proprietary pharmaceutical sector; although not yet economically significant on their own – they would decidedly become so during the economic boom of the 2000s – developing countries’ pharmaceutical markets were nonetheless a critical market frontier to bring under control. It was not African countries, whose rising populations had plenty of health afflictions but whose low purchasing power made them an uneconomic market, but rather developing countries in Asia and Latin America that were of concern, especially those that had managed to develop productive capabilities in the sector. As many of them could only produce final products out of active pharmaceutical ingredients, the ‘pirate’ label was applied more forcefully to the few, namely Argentina, Mexico, India, China and South Korea, who could manufacture generic versions of patented pharmaceuticals through reverse engineering (Kaplan & Laing 2005). Since 1995, and especially since 2005 when the TRIPS transitional period ended for developing countries (but not for least developed countries), production through reverse engineering of patented medicines is no longer an option. Whatever was said about the links between stronger global IP protection and innovation or development during the negotiation process, TRIPS represents a particular instance of the US and EU efforts to protect and extend the technological rents of their most promising high-tech sectors in ever wider geographical areas and for longer time periods. For the proprietary pharmaceutical sector, despite its arguments about the necessity of patent protection as the only way of guaranteeing that new and better medicines reach the market, the much stronger patent protection achieved through TRIPS mainly offered them new means to protect existing markets, enter new lucrative ones, limit competition and extract higher rents. TRIPS was by no

386  Handbook on the political economy of health systems means the end of the road in this respect. Putting direct pressure on governments of key developing countries to ‘improve’ patent protection and enforcement for pharmaceuticals regardless of their levels of socio-economic development or healthcare needs has been supplemented – and by all accounts this has been more effective – by relying on the weight of the US state and to a lesser extent of the EU to extract such concessions, primarily via dedicated IP chapters in preferential trade agreements. These agreements often introduce provisions that, like the ones in the US, expand the scope and term of pharmaceutical patents, provide stronger protection for pharmaceutical test data (data exclusivity), prohibit parallel importation, narrow the exception to patent rights and add conditions on the use of compulsory licences (Muzaka 2021).3

3.

EVALUATING THE CURRENT PHARMACEUTICAL PATENT SYSTEM

The interest of proprietary pharmaceutical companies in strengthening and expanding patent protection standards in the developing world has gone hand in hand with the considerable growth of these countries’ pharmaceutical expenditure; so rapid has this growth been in the top layer of these countries, especially China, India and Brazil, that the new label ‘pharmemerging countries’ had to be introduced to capture the novel phenomenon of growth rates in pharmaceutical markets in this part of the world outpacing those of the developed one. Despite such growth rates, it remains the case that the North American and European markets still account for around 71 percent of global sales (EFPIA 2020). Importantly, the ‘pharmemerging’ phenomenon – largely driven by the expansion of healthcare systems in key developing countries – has not altered ‘the 10/90 gap’, the term introduced in 1990 to highlight the fact that less than 10 percent of global pharmaceutical R&D resources were devoted to the health needs of developing countries where over 90 percent of all preventable deaths occurred (Global Forum for Health Research 2000). A study in 2002 found that of the 1,393 new chemical/ pharmaceutical entities (NCE) introduced in the market between 1975 and 1999, only 16 (1.1 percent) were for tropical diseases and tuberculosis that accounted for over 11 percent of the global disease burden (Trouiller et al. 2002). Despite the notable attention paid to global health issues in the following decade, and particularly to the issue of the neglected tropical diseases that affect around 1.7 billion people around the world, this picture did not change: of the 336 NCEs introduced between 2000 and 2011, only 4 (1.2 percent of the total) were for neglected diseases (Pedrique et al. 2013). There has been an uptick in the number of companies and volume of R&D dedicated to neglected tropical diseases since 2014; while the full picture is yet to emerge, movement remains rather slow regarding diseases and conditions that constitute public health priorities in low-income and less developed countries (MSF 2019). The continued lack of attention paid to the ‘diseases of the poor’ constitutes one of two main shortcomings of the pharmaceutical patent system and which the WHO R&D Treaty proposal, discussed later, was meant to address. The tendency to focus on diseases that are ‘lucrative’ is confirmed by industry data: over 70 percent of sales of new medicines launched in the last five years were accounted for by the US and European markets (EFPIA 2020). But skewed research priorities towards economically profitable health conditions (or non-conditions, as long as they can be framed in need of medication; e.g., baldness) is not the only problem stemming from the current pharmaceutical patent system: affordable access to existing medicines is also a challenging issue. The unaf-

The pharmaceutical patent system and access to medicines  387 fordability and high prices of on-patent medicines constitutes the other major consequence of the current pharmaceutical patent system. As the classic geographical spread of type I, II and III diseases4 has changed, patients everywhere, bar those who are insured, often face medicine prices that are well beyond their purchasing power. As noted earlier, this problem had been evident in the US for a while and has become a problem in the rest of the developed world where healthcare systems do not use a particular medicine as it is too expensive. Nevertheless, this problem is many times more pronounced in the developing world. Studies have repeatedly shown that patients there often face drug prices higher than those in the developed world, and certainly higher than their average per capita income (Scherer 2000). The lack of universal public health schemes in much of the developing world means that demand for drugs is highly sensitive to drug prices; that is, patients go without medication even when drugs to treat their condition exist. This was the case, for instance, for HIV/AIDS patients facing antiretroviral prices in the range of US$12,000–14,000 per person per year (ppy) in the early 1990s or for millions of hepatitis C patients who faced prices in the range of US$16,000–18,000 for a 48-week treatment in the first decade of the century (WHO 2010). Despite recent progress, only a small number of patients diagnosed with Hep C every year (around 7 percent in 2015) start treatment, and often treatments are not of the more effective and well-tolerated kind as these tend to be expensive (WHO 2017). Affordability is not a problem that only concerns relatively poorer patients; Gilead’s US$84,000 price tag for Sovaldi (a Hep C medicine) or US$1,000 for every single pill, makes it nearly 67 times more expensive than gold in weight terms (MSF 2016: 13). The ability to set prices is one of the key privileges the pharmaceutical patent system grants its beneficiaries and, where prices are not controlled or where no other sources for procuring a specific medicine exist, they make the most of it. In the former category the best illustration is provided by the US, where recently Martin Shkreli’s overnight 5,500 percent price hike for a 50-year-old drug provoked global outrage, even though price hikes have been a normal course of action for proprietary pharmaceutical companies: prices for proprietary prescription medicines doubled between 2010 and 2015 in the US (Crow 2016). In the second category would fall a large number of developing countries with insufficient domestic pharmaceutical production capabilities or those who have them but are restricted post-TRIPS to use such capabilities to produce generic versions of patented drugs. Examples can be drawn from numerous and exhausting price-negotiating rounds with proprietary pharmaceutical companies for priority medication; e.g., for antiretrovirals in Brazil during the 2000s that still left the Brazilian state paying higher prices for its AIDS programme than the lowest available international company price (Flynn 2013). Much lower prices for a number of important patented medicines have been available from generic producers in India, for instance, but as India and other countries introduced full pharmaceutical patentability in 2005, the supply of cheaper generic versions of patented medicines is bound to diminish over time. Faced with these issues, the typical response by industry representatives has been two-pronged: pharmaceutical patents and associated high prices are indispensable for funding medical R&D, whereas issues related to diseases disproportionally affecting developing countries are seen as ‘a social problem’ which industry ‘should not be expected to solve’ (Roy Vagelos, quoted in Silverstein 1999). Addressing these two arguments requires revisiting, even if only briefly, the raison d’être of the patent institution. Justifying private property, an immensely challenging task, is even harder for private intellectual property like patents because what is to be owned – that is, ideas, knowledge and other products of human intel-

388  Handbook on the political economy of health systems lect – are not scarce the way tangible goods are. What can be drawn from the centuries-long debates that have preoccupied lawyers, philosophers and economists over this issue is that the ‘social good’ has served as the very foundation for justifying the existence of private intellectual property (Ryan 1986). In other words, the social good is the raison d’être of the patent system. As noted earlier, even if early patents were often granted at the whim of the monarch, patent monopolies came to be justified on public interest grounds: the very first patent statute (Venice) rewarded patents on authorities’ assessment of social considerations and utility for the municipality, thus constituting an early version of the balance between public benefit and private rewards to encourage intellectual activity. In time, these considerations materialised in the utilitarian position that came to justify patents on account of them providing adequate incentives for the creation of socially valuable intellectual goods (Boyle 2008) such as medicines. Although individual rights claims have also been influential (Drahos 1996), it has been this utilitarian ethos that continues to underpin the patent system in most European states and the US, where the modern pharmaceutical patent system was first established. In the US Constitution (Article I, §8, Clause 8) IP protection is explicitly a matter of positive law (i.e., not natural law) justified in social grounds. Thomas Jefferson, one of the key authors, insisted that ‘Society may give an exclusive right … as an encouragement to men to pursue ideas which may produce utility, but this may or may not be done, according to the will and convenience of the society’ (Jefferson 1813). To be sure, the utilitarian position is not uncontested: innovation is driven by many motives rather than the award of an exclusive patent, and historically creative periods have often preceded rather than followed the appropriation of IP (Daly & Cobb 1989). Arguments that seek to justify patents on economic grounds are probably the weakest, insofar as patent excludability causes an underutilisation and inefficient use of knowledge (Heller & Eisenberg 1998). Nonetheless, this economically inefficient use of knowledge is also excused on utilitarian grounds; that is, in order to have more intellectual goods that serve society we may have to bear with this system, although this position stands on the contested premise that exclusive patents are the best incentives for intellectual activity. Two of the most prominent patent economists of the 20th century – Fritz Machlup and Edith Penrose – separately came to the conclusion that no economist could make a conclusive case for instituting a patent system (Penrose 1951, Machlup 1958). Not routinely discussed in utilitarian accounts but closely linked to the issue of patent exclusivity and the social good are two additional arguments. Because it gives exclusivity to use the product or process protected only to the owner, the patent system needs justification insofar as innovation is collaborative and cumulative; far from being creation from nothing, intellectual products are the result of collective human effort; that is, they are social products (Kinsella 2001). Secondly, and related, by giving the reward to a few, justification is needed why the patent-holder should have the exclusive right to most or all the market value the invention will garner, given that market value is a socially created phenomenon (Hettinger 1997). These two arguments are especially pertinent to the pharmaceuticals given the social nature (and value) of healthcare and the large contribution of public funds and institutions on research in general and medical research in particular. Publicly funded and academic research on pharmaceuticals tackles basic research that is also the riskiest and the most costly; in the US, for instance, publicly funded medical R&D claims around 22 percent of the total federal R&D budget, second only to defence (Schacht 2011: 19). Noting the difficulties in calculating the contribution of this kind of funding to medications that reach the market, a report by the

The pharmaceutical patent system and access to medicines  389 US National Institutes of Health – the largest public medical research agency in the world, with a budget of US$32 billion a year5 – noted that US taxpayer-funded research produced 55 percent of the published research projects leading to the discovery and development of the five most lucrative medicines in the market in 1995, with foreign academic institutions contributing another 30 percent (Public Citizen 2001: 8). A more recent US study found that publicly funded research was associated with each of the 210 new medicines approved for use between 2010 and 2016 (Mazzucato & Roy 2018: 108). Although the US state is the biggest funder of medical R&D, the considerable contribution of publicly funded R&D to the development of medicines that reach the market is a worldwide phenomenon. Such contribution weakens the most robust argument the pharmaceutical sector’s representatives put forward in defence of the pharmaceutical patent system: exclusive patents are justifiable because they provide the necessary financial incentives (via sky-high prices) to develop new drugs that would not otherwise be developed due to the increasingly lengthy and costly nature of the R&D process. The latter is reported to stand at over US$2 billion currently, a mythical figure which since it first emerged in the early 2000s up to its recent updates remains marred by accounting and methodological controversies, not to mention the fact that it is exclusively based on data furnished by pharmaceutical companies themselves (Goozner 2004, DiMasi et al. 2014). Of other studies, the lowest estimate is US$30 million (in 2013 US$), a significantly smaller amount (MSF 2016: 25). The only point that can be made with confidence is that the cost of pharmaceutical R&D development is unknown. As long as this remains the case, and as long as proprietary pharmaceutical companies appear in the most profitable companies’ lists, arguments about the necessity of patent protection for covering the exceptionally high costs and risky nature of medical R&D remain unfounded. Even if the argument is accepted, the fact remains that increasingly stronger pharmaceutical patent protection from the 1980s onwards has been accompanied by a lower, rather than higher, rate of pharmaceutical innovation. The absolute number of pharmaceutical patents has in fact increased, but this is an indication of poor-quality patents rather than of increased innovation (Jaffe & Lerner 2004). As a matter of fact, the (average) annual number of NMEs reaching the world market diminished from a high of 93 in the 1960s to an average of 31 in the 1996–2014 period (Gassman et al. 2018: 7). The trend in patent grants is the opposite: the number of patents granted worldwide in the field of pharmaceuticals and medical technology increased by over 500 percent between 1980 and 2014.6 The fall of NME numbers has also been accompanied by the increasing number of ‘me-too’ drugs in the market since the 1980s. These are versions of successful (i.e., profitable) existing drugs that claim patent protection on account of making incremental improvements on existing drugs, a patenting strategy also known as ‘ever-greening’. Clearly, ‘me-too’ drugs are competitive vis-à-vis other similar drugs in terms of usage, but they often enter the market at the same or an even higher price level than their nearest therapeutical substitute already in the market, offering patients none of the benefits expected from competition. Because they usually go through all the R&D clinical trial phases, ‘me-too’ drugs utilise pharmaceutical R&D resources in a manner that is both uneconomic and unethical, as is the use of resources wasted on fierce legal disputes over the merits of ‘me-too’ patents. Diseases and medicines needed to treat them certainly constitute a social problem and this is true not only for ‘diseases of the poor’ most manifestly neglected by the private pharmaceutical sector, but for all diseases regardless of their incidence and geographical spread. At the same time, the pharmaceutical patent system is also an institution with an undeniably

390  Handbook on the political economy of health systems social function and purpose. That it exists or is spread around the world, that it generates some medicines and many, many patents, is certainly no proof that it fulfils such purpose. It can only be said to have fulfilled this social function when, firstly, it successfully promotes the development of drugs that seek to treat all the various health conditions prevailing in the society. Now that the pharmaceutical patent system has become binding globally, the society to which it must serve has to expand to that level. Secondly, the social function of the pharmaceutical patent cannot be considered fulfilled if medication is not available in affordable terms; pharmaceutical innovation without access or affordability is meaningless from a societal point of view. Those defending the current pharmaceutical patent system seem unable to accept or find ways to address these serious shortcomings, focusing their efforts instead on thwarting any discussions or experiments with alternative arrangements.

4. ‘PATENT vs LIFE’ CONTESTS POST-TRIPS While the TRIPS negotiations were going on (1986–1994) without much consideration of the social purpose of the pharmaceutical patent system or the social consequences of extending it worldwide, public health concerns were entering the global agenda with considerable urgency, partly on account of the AIDS/HIV crisis being seen as an ‘exceptional disease’ that required exceptional action. The 1990s were full of meetings bringing together a considerable number of state and non-state actors intent on addressing pressing public health needs; the adoption of the United Nations (UN) Millennium Development Goals in 2000 was one outcome of these gatherings, as was the first ever UN Security Council session on global health that same year, followed by the creation of the most high-profile private–public partnership in the area of public health, the Global Fund to Fight AIDS, Tuberculosis and Malaria in 2002. Never before had such attention been paid to global health issues, nor such amounts been channelled towards improving it, especially when the going was good: sums dedicated to global health issues reached unprecedented levels just before the recent financial crisis (Schneider & Garrett 2009). These commendable efforts hardly made a dent into the rather more robust and effective spread of the pharmaceutical patent system worldwide post-TRIPS. Although concerns about the social ineffectiveness of the current pharmaceutical patent system had been voiced systematically since the early 1990s,7 most public health efforts from the mid-1990s onwards directed towards developing countries mainly sought to patch over some of the more visible and problematic gaps opened up by its global strengthening and expansion, such as negotiating prices or financing the purchase of certain existing but expensive medicines under patent protection. Seriously questioning the impact of strong pharmaceutical patents on the affordability and accessibility of patented HIV/AIDS medicines – and, later, of all medicines – in the developing world was taken up by a small but growing coalition of international health non-governmental organisations (NGOs) in the mid-1990s that became known as the Access Campaign. Closely related to it was a broader coalition of human rights NGOs whose concerns over the consequences of the new TRIPS Agreement went beyond health to include education, farmers’ rights and protection of biodiversity and of indigenous culture and knowledge. In response, the UN Human Rights Sub-Commission adopted a resolution on IP and human rights that explicitly pointed at real and potential conflicts between TRIPS obligations and the realisation of economic, social and cultural rights.8 This was only the first of a large number of

The pharmaceutical patent system and access to medicines  391 resolutions and reports the many bodies of the human rights regime have been only too keen to pass and adopt, all of which have been highly critical of the new IP arrangement. Of all the various contests that emerged post-TRIPS, the ‘patents vs patients’ campaign became the most politically charged (Muzaka 2011). This was so not only because of the emotive appeal and the sustained attention paid to global public health issues, but also because many developing countries’ governments were finding their policy options on public health matters limited, partly as a result of certain TRIPS obligations and partly due to the intense efforts of IP-reliant businesses and their home states to further ratchet up IP protection standards beyond TRIPS. The most notable success of the ensuing coalition between Access Campaign NGOs and key state actors from developing countries was the Declaration on TRIPS and Public Health at the WTO Doha Ministerial in 2001. This was hailed as a stunning political victory, but courageous and surprising as it may have appeared at the time, the Declaration could not and did not resolve the contradictory pulls underpinning the global expansion of the pharmaceutical patent system and the increased awareness about and dedication towards addressing acute global public health needs. By virtue of merely reiterating the TRIPS provisions and flexibilities related to pharmaceuticals and public health needs, the Declaration posed no real challenge and brought no transformative changes to the global pharmaceutical IP protection arrangement that TRIPS set in place. The WHO also became one of the fora where the ‘patents vs patients’ contests played out, first issuing a series of resolutions which encouraged member states to review options under international agreements (i.e., TRIPS) to ensure and safeguard access to essential medicines.9 Perhaps the most interesting idea to emerge during the 1990s – a hard law approach to changing the pharmaceutical R&D model, especially as it related to the ‘diseases of the poor’ first considered by an independent working group (the Neglected Diseases Group) set up by Médecins Sans Frontières (MSF) in 1999 – found its way into the WHO in the early 2000s. The idea of changing the pharmaceutical R&D model by decoupling the relationship between high prices and R&D costs gained support among an increasingly large group of health NGOs and activists who, following the first fully formed proposal by Tim Hubbard and Jamie Love in 2002, went on to make a series of proposals towards this end (Gombe & Love 2010). Simultaneously, and inspired by some of these ideas, a proposal by a group of developing countries led by Brazil and supported by many health NGOs was made to the WHO in 2003 to establish an independent commission to examine the relationship between patent protection, innovation and public health. The new commission set up for this purpose (CIPIH)10 in 2004 considered among others a new pharmaceutical patent model based on the idea of delinking the cost of pharmaceutical R&D from the price of medicines and expanding the definition of patent exclusivity to include access, as well as innovation. By the time the 2005 new global medical R&D Treaty proposal was made at the WHO by a group of scientists, public health experts, NGOs and state representatives,11 the idea was well established, but also appeared heretical to patent interests. As a result, a long succession of events and clashes unfolded at the WHO and outside it: Kenya and Brazil led calls at the WHO for a new and bold global framework for essential medicines R&D that eventually led to the 2008 WHO Global Strategy and Plan of Action agreeing to explore a range of incentive schemes for medical R&D, including prizes, patent pools and government funds (WHA 2008). The idea of a prize fund seemed to enjoy considerable support at the time, not least because Senator Bernie Sanders (and presidential candidate John Edwards) persisted with the idea of using prizes to stimulate R&D in the US, and notable economists such as Joseph Stiglitz wrote

392  Handbook on the political economy of health systems a number of widely read articles in support of prizes to reward medical R&D (Gombe & Love 2010). Despite this broad support, the 2010 Report of the WHO Expert Working Group on R&D was disappointing on this front for many developing countries (though it was welcomed by the US and the EU) and health NGOs and came under heavy criticism for being influenced by private pharmaceutical interests.12 In an unusual turn of events, the World Health Assembly restarted the process in May 2010; the new Consultative Group on R&D produced a bolder set of recommendations in April 2012, including recommendations for a binding Global Framework on R&D, open knowledge innovation frameworks, patent pools to share knowledge and prizes as incentives for drug discovery and development.13 The proposed Global Framework on R&D (also known as the medical R&D treaty) would not fundamentally alter the pharmaceutical patent model; focusing on neglected diseases that have historically attracted negligible private investment, it requested each country to make a minimum financial contribution (0.01–0.02 percent of gross domestic product) to a global R&D fund, allowing product developers to be paid to carry out the research and enabling products thus financed to be sold at generic prices immediately following their regulatory approval. Indeed, the special report recommending it made clear that it was not to be ‘a replacement for the existing intellectual property rights system but … a supplementary instrument where the current system does not function’ (CEWG 2011: point 12), namely an instrument to provide financing and coordination for R&D on neglected diseases that have historically attracted a negligible amount of private R&D funds. Nevertheless, high hopes were dashed in 2016 when the WHO Health Assembly concluded without giving the green light for negotiating the treaty, putting forward instead a rather weak commitment to consider in 2017 the possibility of convening another open-ended meeting to discuss the subject (WHA 2016). Since then, discussions in open-ended meetings have been comparably much easier for a WHO Observatory set up to map ongoing R&D and identify gaps (a process debated at the WHO since the 1990s) and rather more difficult on the medical R&D treaty proposal. Given its limited scope to neglected diseases and the fact that the US state already meets and indeed exceeds the R&D contribution requirement in the treaty proposal, resistance to it by the US and patent interests was closely bound to concerns that such a treaty would in time inspire other challenges to the patent system on which many vested interests (directly or indirectly) depend. Indeed, the US, and the proprietary pharmaceutical sector, which interpreted the modest scope of the treaty as a challenge to the patent system and their business model, were the most aggressive opponents of the treaty (Love 2009, Carter 2012). Overtaken by the Covid-19 pandemic that started in 2019, it appears unlikely the treaty will be negotiated and concluded soon, unless the change of position on the part of the US state regarding the negative link between IP and (Covid-19) medicines and treatments turns out to be more than the vague statement it has been so far. As noted at the start, negotiations for a TRIPS waiver for the duration of the pandemic have not produced anything concrete yet; it bears repeating that the waiver is limited in time and does not constitute a radical alteration of the pharmaceutical IP system. That said, the impact it is likely to have on company profits and, perhaps more importantly, the message that IP is a hindrance to access to medicines, which it implies, are deemed too damaging by large proprietary pharmaceutical companies and key developed states alike. Perhaps as a ruse to deflect attention from both the WHO medical R&D treaty and the WTO TRIPS waiver, an idea for a new international pandemic treaty has been recently mooted, despite the fact that many of its components already exist (e.g., the International Health Regulations treaty, adopted in 1969 and revised in 2005 by the

The pharmaceutical patent system and access to medicines  393 World Health Assembly). Notably, the new pandemic treaty appears to contain no language whatsoever about IP and its impact on the necessary medicines, treatments and technologies needed to address the next pandemic.

NOTES 1.

Because process patents only provide exclusivity for the method of making a medicine, they do not hinder other companies from making the same medicine through a different process. This kind of protection is often considered to have fewer negative effects than product patents. India’s patent law changes in 1971 forbade pharmaceutical product but not process patents, opening the way to the creation of one of the most vibrant generic pharmaceuticals in the world (a success that obviously depended on many other sectoral policies than just the said change). 2. An early investigation was that initiated by Senator Kefauver in 1959 on the monopoly prices charged by the pharmaceutical companies in the US, their excessive advertising spending and so on, although, due to the thalidomide tragedy, the 1962 Kefauver–Harris Amendments to the 1938 Food, Drug and Cosmetics Act focused largely on drug safety. For a rather more recent congressional investigation on alleged influence of the pharmaceutical industry in the 2010 Affordable Care Act, see Armstrong (2012). 3. Almost all US preferential trade agreements (PTAs) and some of the EU ones (especially those signed after the 2006 Global Europe Strategy, such as EU–South Korea) contain ‘TRIPs-plus’ provisions of one kind or another. All US PTAs from 2000 onwards (with the exception of the Jordan PTA) introduce obligations to protect pharmaceutical data, for instance, and most of them limit the grounds on which the partner can issue a compulsory licensing (e.g., the US PTAs with Jordan, Singapore, Australia and Vietnam). 4. Type I diseases occur in both rich and poor countries (e.g., cancer). Type II also occur in both sets of countries but with a substantial proportion of the cases in poor countries (e.g., tuberculosis, HIV/ AIDS). Type III are those overwhelmingly or exclusively incident in developing countries (e.g., malaria). 5. Data from the NIH website: www​.nih​.gov/​grants​-funding. 6. According to data elaborated from the WIPO database on patents: http://​ipstats​.wipo​.int/​ipstatv2/​ index​.htm​?tab​=​patent. 7. The Commission on Health Research and Development (CHRD) was established in 1987 to address concerns about mismatching research priorities related to diseases affecting developing countries; its 1990 report coined the ‘10/90 gap’ phrase. This was followed by the establishment at the WHO in 1994 of the Committee on Health Research Relating to Future Intervention Options that among other things recommended a systematic approach for the allocation of health research funds, prioritising maternal and child health, and microbacterial resistance, among others. The Global Forum for Health Research was created in 1997 to help correct the 10/90 gap through, in part, developing ways towards global health needs priority-setting. 8. Resolution 2000/7 of the 52nd Session of the Sub-Commission E/CN.4/Sub.2/Res/2000/7, 17 August 2000, available at www​.ohchr​.org. 9. See, for instance, WHA resolutions WHA 52.19 of 24 May 1999 and WHA 54.11 of 21 May 2001. 10. See the WHA Resolution WHA56.27 of 28 May 2003, establishing The Commission on Intellectual Property Rights, Innovation and Public Health (CIPIH). 11. See the proposal here: http://​test​.tacd​.org/​wp​-content/​uploads/​2013/​09/​TACD​-IP​-2005​-Letter​-to​ -WHO​-proposing​-Medical​-RD​-Treaty​.pdf. 12. See, for instance, ‘Expert report on R&D financing triggers inquiry consultations’, www​.twnside​ .org​.sg/​title2/​health​.info/​2010/​health20100102​.htm; see also ‘WHA agrees to roadmap to take forward work on R&D financing’, www​.twnside​.org​.sg/​title2/​intellectual​_property/​info​.service/​ 2010/​ipr​.info​.100509​.htm. 13. See www​.who​.int/​phi/​CEWG​_Report​_5​_April​_2012​.pdf.

394  Handbook on the political economy of health systems

REFERENCES Achilladelis, B. (1999) ‘Innovation in the pharmaceutical industry’, in R. Landau, B. Achilladelis and A. Scriabine (Eds), Pharmaceutical Innovation: Revolutionizing Human Health, Philadelphia, PA: Chemical Heritage Press. Armstrong, D. (2012) ‘Drugmakers’ Deal with Obama Said to Be Probed by House’, Bloomberg, 4 May. Bigdeli, M., Laing, R., and Tomson, G., et al. (2015) ‘Medicines and Universal Health Coverage: Challenges and Opportunities’, Journal of Pharmaceutical Policy and Practice 8: 8. Boldrin, M., and Levine, D. (2008) Against Intellectual Monopoly, Cambridge: Cambridge University Press. Boyle, J. (2008) The Public Domain: Enclosing the Commons of the Mind, New Haven, CT: Yale University Press. Braithwaite, J., and Drahos, P. (2000) Global Business Regulation, Cambridge: Cambridge University Press. Carter, Z. (2012) ‘Obama Administration Blocks Global Health Fund to Fight Disease in Developing Countries’, Huffington Post, 25 May. CEWG (WHO Consultative Expert Working Group on R&D) (2011) Financing and Coordination, Third Report, www​.who​.int/​phi/​cewg3​_report​_dec2011​.pdf. Crow, D. (2016) ‘Price of Branded Prescription Drugs in the US Doubles in 5 Years’, Financial Times, www​.ft​.com/​content/​900291d8​-ea04​-11e5​-888e​-2eadd5fbc4a4. Daly, H.E., and Cobb, J.B. (1989) For the Common Good, Uckfield: Beacon Press. Danzon, P., and Keuffel, E. (2005) Regulation of the Pharmaceutical Industry, paper presented at the National Bureau of Economic Research Conference on Regulation, 9–10 September 2005, Cambridge, MA. DiMasi, J.A., Grabowski, H., and Hansen, R. (2014) ‘Cost of Developing a New Drug’, Tufts Center for the Study of Drug Development, http://​csdd​.tufts​.edu/​files/​uploads/​Tufts​_CSDD​_briefing​_on​_RD​ _cost​_study​_​-​_Nov​_18​,​_2014​.pdf. Drahos, P. (1996) A Philosophy of Intellectual Property, Aldershot: Ashgate. Drahos, P., and Braithwaite, J. (2002) Information Feudalism: Who Owns the Knowledge Economy? London: Earthscan. EFPIA (2020) ‘The Pharmaceutical Industry in Figures’, European Federation of Pharmaceutical Industries and Associations, https://​efpia​.eu/​media/​554521/​efpia​_pharmafigures​_2020​_web​.pdf. Emmert, F. (1990) ‘Intellectual Property in the Uruguay Round: Negotiating Strategies of the Western Industrialized Countries’, Michigan Journal of International Law 11: 1317–99. Fisk, C. (2003) ‘Authors at Work: the Origins of the Work-for-Hire Doctrine’, Yale Journal of Law and Humanities 15: 1–69. Flynn, M. (2013) ‘Origins and Limitations of State-Based Advocacy: Brazil’s AIDS Treatment Program and Global Power Dynamics’, Politics and Society 41(1): 3–28. Gassman, O., Schuhmacher, A., Zedtwitz, M., and Reepmeyer, G. (2018) Leading Pharmaceutical Innovation: How to Win the Life Sciences Race, Third Edition, Cham: Springer. Global Forum for Health Research (2000) The 10/90 Report on Health Research, http://​announcementsfiles​ .cohred​.org/​gfhr​_pub/​assoc/​s14791e/​s14791e​.pdf. Gombe, S., and Love, J. (2010) ‘New Medicines and Vaccines: Access, Incentive to Investment, and Freedom to Innovate’, in G. Krikorian and A. Kapczynski (Eds), Access to Knowledge in the Age of Intellectual Property, New York, NY: Zone Books, 531–46. Goozner, M. (2004) The $800 Million Pill: The Truth about the Cost of New Drugs, Berkeley, CA: University of California Press. Heller, M.A., and Eisenberg, R.S. (1998) ‘Can Patents Deter Innovation? The Anti-Commons in Biomedical Research’, Science 280: 698–701. Hettinger, E.C. (1997) ‘Justifying Intellectual Property’, in A.D. Moore (Ed.), Intellectual Property: Moral, Legal and International Dilemmas, Oxford: Rowman & Littlefield. Jaffe, A., and Lerner, J. (2004) Innovation and its Discontents, Princeton, NJ: Princeton University Press. Jefferson, T. (1813) ‘Letter to Isaac McPherson’, http://​memory​.loc​.gov/​ammem/​collections/​jefferson​ _papers/​mtjser1​.html.

The pharmaceutical patent system and access to medicines  395 Kaplan, W., and Laing, R. (2005) Local Production of Pharmaceuticals: Industrial Policy and Access to Medicines – An Overview of Key Concepts, Issues and Opportunities for Future Research, New York, NY: World Bank. Kinsella, N.S. (2001) ‘Against Intellectual Property’, Journal of Libertarian Studies 15(2): 1–53. Love, J. (2009) ‘Hillary and Obama Set to Kill Medical R&D Treaty at WHO Meeting’, Huffington Post, 20 May. Machlup, F. (1958) ‘An Economic Review of the Patent System’, reprinted in F. Abbott, T. Cottier and F. Gurry (Eds), The International Intellectual Property System: Commentary and Materials, Part I, London: Kluwer Law. May, C., and Sell, S. (2006) Intellectual Property Rights: A Critical History, London: Lynne Rienner. Mazzucato, M., and Roy, V. (2018) ‘Rethinking Value in Health Innovation: From Mystification Towards Prescriptions’, Journal of Economic Policy Reform 22(2): 101–19. MSF (Médicins Sans Frontières) (2016) Living on the Edge: Time to Align Medical Research and Development with People’s Health Needs, www​.msfaccess​.org/​sites/​default/​files/​MSF​_assets/​ Access/​Docs/​R​&​D​_report​_LivesOnTheEdge​_ENG​_2016​.pdf. MSF (Médicins Sans Frontières) (2019) Are Pharmaceutical Companies Making Progress When It Comes to Global Health? https://​access​tomedicine​foundation​.org/​publications/​are​-pharmaceutical​ -companies​-making​-progress​-when​-it​-comes​-to​-global​-health. Muzaka, V. (2011) The Politics of Intellectual Property Rights and Access to Medicines, Basingstoke: Palgrave MacMillan. Muzaka, V. (2021) ‘Making Global Public Policy: Business and Global Patent Protection Standards’, in K. Ronit, T. Porter and A. Kellow (Eds), Handbook on Business and Public Policy, Cheltenham, UK and Northampton, MA: Edward Elgar Publishing, 124–39. Pedrique, B., Strub-Wourgaft, N., Some, C., et al. (2013) ‘The drug and vaccine landscape for neglected diseases (2000–11): a systematic assessment’, The Lancet 1(6): e371–e379. Penrose, E.T. (1951) The Economics of the International System, Baltimore, MD: Johns Hopkins Press. Public Citizen (2001), ‘Rx R&D Myths: The Case against the Drug Industry’s R&D “Scare Card”’, www​ .citizen​.org/​congress/​reform/​drug​_industry/​. Ramirez, P., and Tylecote, A. (1999) Technological Change in the Pharmaceutical Industry: A Literature Review from the Point of View of Corporate Governance, COPI Report, Sheffield University, prepared for EC DG XII. Rosen, G. (1993) A History of Public Health, Baltimore, MD: Johns Hopkins University Press. Ryan, A. (1986) Property and Political Theory, London: Blackwell. Schacht, W.H. (2011) ‘Federal R&D, Drug Discovery, and Pricing: Insights from the NIH-University-Industry Relationship’, US Congressional Research Office, RL 32324, www​.crs​.gov. Scherer, F.M. (2000) The Pharmaceutical Industry, Amsterdam: Elsevier. Schneider, K., and Garrett, L. (2009) ‘The End of the Era of Generosity? Global Health Amid Economic Crisis’, Philosophy, Ethics, and Humanities in Medicine 4(1): 1–7. Sell, S. (2003) Private Power, Public Law: The Globalization of Intellectual Property Rights, Cambridge: Cambridge University Press. Silverstein, K. (1999) ‘Millions for Viagra, Pennies for the Poor’, The Nation, 19 July. Trouiller, P., Olliaro, P., Torreele, E., et al. (2002) ‘Drug Development for Neglected Diseases: A Deficient Market and a Public-Health Policy Failure’, The Lancet, 359(9324): 2188–94. WHA (World Health Assembly) (2008) Global Strategy and Plan of Action on Public Health, Innovation and Intellectual Property. Document WHA 61.21 (24 May). WHA (World Health Assembly) (2016) Follow-Up of the Report of the Consultative Expert Working Group on Research and Development: Financing and Coordination, Sixty-Ninth World Health Assembly, Geneva, Switzerland. Document WHA 69.23 (28 May). WHO (World Health Organization) (2010) Global Alert and Response: Hepatitis C, Geneva: WHO, www​.who​.int/​csr/​disease/​hepatitis/​whocdscsrlyo2003/​en/​index4​.html. WHO (World Health Organization) (2017) Global Hepatitis Report, Geneva: WHO, https://​apps​.who​ .int/​iris/​bitstream/​handle/​10665/​255016/​9789241565455​-eng​.pdf​?sequence​=​1. WIPO (World Intellectual Property Organization) (1988) Existence, Scope and Form of Generally Internationally Accepted and Applied Standards/Norms for the Protection of Intellectual Property, GATT document MTN​.GNG/​NG11/​W/​24/​Rev1, Geneva: WIPO.

PART V POLITICAL ECONOMY OF HEALTH CARE DURING COVID-19

23. Covid-19 and the interest group approach to government Peter T. Leeson and Henry A. Thompson

1. INTRODUCTION Scholarly response to the Covid-19 pandemic has been immense. The first case of Covid in the United States was confirmed on 20 January 2020. By March, the Centre for Economic Policy Research (CEPR) launched Covid Economics: Vetted and Real-Time Papers, a publication that rapidly evaluates and disseminates economic research relevant to Covid-19. As of 9 April 2021, Covid Economics had disseminated 469 studies (CEPR.org 2021). Over a similar period, the National Bureau of Economic Research disseminated 390 pandemic-related working papers (NBER.org 2021). And a growing number of Covid-relevant studies appear in other economics working paper series or have been published in economics journals. Perhaps surprisingly, hardly any of this research considers governmental responses to Covid-19 from the perspective of the interest group approach to government. None of the recent NBER papers and only one of the recent CEPR papers, for example, draws on that approach explicitly. The situation, however, is beginning to change. The Southern Economic Journal, for instance, recently published a symposium dedicated to the political economy of the pandemic, and several of the contributions leverage the interest group approach to government or are closely connected to it. This chapter surveys the small but burgeoning literature at the intersection of Covid-19 and the interest group approach to government. Our survey highlights three themes: (1) private interests may influence how government-imposed lockdowns are implemented, (2) private interests may influence governmental allocation of Covid resources, and (3) governmental responses to Covid-19 may have perverse consequences, undermining rather than promoting the health of vulnerable people. The foregoing themes parallel those which Leeson and Thompson (2023, p. 6) identify in their survey of non-Covid scholarship that studies issues in public health from the perspective of the interest group approach to government: “(1) Public health regulations often are driven by private interests, not public ones. (2) The allocation of public health resources often reflects private interests, not public ones. (3) Public health policies may have perverse effects, undermining instead of promoting health-consumer welfare.” In the Covid context, however, only one study to date has touched on the second of these themes. To provide ideas for further work in that area, we consider research that examines the influence of private interests on governmental allocation of health resources in non-Covid contexts and suggest possible applications in the context of Covid. In focusing on the above themes, our survey excludes complementary work in political economy concerned with Covid-19. We do not consider research that examines how variation in governmental institutions may affect governmental responses to the pandemic.1 We do not consider scholarship that compares governmental and market responses to the pandemic.2 Nor 397

398  Handbook on the political economy of health systems do we consider work on recent stimulus bills ostensibly intended to mitigate the economic, as opposed to health, consequences of Covid-19.3 Our discussion directs attention narrowly to research relevant to the interest group approach to government that examines governmental responses to the pandemic’s health consequences.

2.

PUBLIC INTERESTS, PRIVATE INTERESTS, AND PUBLIC HEALTH

The economic rationale for government health intervention is, like the rationale for other interventions, grounded in the theory of market failure fathered by Arthur Cecil Pigou (1920).4 That theory identifies departures of unhampered markets from the perfectively competitive model and describes policies for their correction. The approach to government most often married to the theory of market failure may be called the “public interest” approach. According to it, observed interventions are motivated by and corrective of market failures. The public interest approach to government implies that interventions increase social welfare and produce the particular results they ostensibly seek; for example, by promoting the welfare of particular consumers. Kenneth Arrow (1963) pioneered the application of that approach to health. Since his seminal work, scholars inspired by Arrow have focused on “four broad categories of government action in healthcare markets, linked to corresponding market failures: healthcare as a merit good; informational gaps; infrastructure as a public good; and externalities” (Tuohy and Glied 2011, p. 58). In the context of Covid-19, the public interest approach conceives of externality correction as the primary motivation for governmental interventions (Leeson and Rouanet 2021). Covid is potentially deadly and highly contagious. Government thus responds by, for example, imposing lockdowns and subsidizing vaccinations to limit the disease’s spread. The public interest approach to government has a competitor: the “public-choice, or interest-group approach” (McCormick and Tollison 1981, p. 3; italics added). According to it, observed interventions are driven by private interests and redistribute wealth. George Stigler (1971, 1976), Sam Peltzman (1976), Robert McCormick and Robert Tollison (1981), and Gary Becker (1983) pioneered the interest group approach to government, whose principal building blocks—the logic of interest groups and rent seeking—were developed by Mancur Olson (1965) and Gordon Tullock (1967), respectively. The interest group approach to government conceives of politics as a market for wealth redistribution. That market’s participants are self-interested politicians (or bureaucrats) and citizens. Political influence requires costly political organization, and the cost of organizing politically varies over different combinations of citizens.5 The demand side of the market reflects combinations of citizens who can organize for less than a dollar to secure a dollar in transfers through favorable intervention. The supply side reflects combinations of citizens for whom it would cost more than a dollar to organize to prevent having a dollar transferred from them through unfavorable intervention. Politicians are market middlemen whose “arbitrage” efforts are remunerated by votes and campaign contributions—the “price” that transfer-demanders pay politicians to redistribute wealth to them from transfer-suppliers. Politicians thus maximize their remuneration by transferring wealth from combinations of citizens who resist the least to those who value transfers the most.

Covid-19 and the interest group approach to government  399 The interest group approach to government implies that interventions redistribute wealth to well-organized groups of citizens who anticipate large per capita gains from poorly organized groups of citizens who anticipate small per capita losses. A corollary is that interventions need not increase social welfare, may instead reduce it, and may produce particular results that are at odds with the outcomes the interventions ostensibly seek. The interest group approach is versatile. Scholars have used it to investigate a variety of public health policies such as food and drink regulations, medical licensing, public health insurance, biomedical research funding, vaccine distribution, pharmaceutical regulations, and mental healthcare, to name but a few (Leeson and Thompson 2023). Governmental responses to Covid-19 present a new public health policy arena for application of the interest group approach. We survey the modest but growing literature connected to that enterprise below. Our discussion summarizes but does not evaluate the validity of the arguments or findings in the studies it considers, which may be challenged as new data become available.

3.

PRIVATE INTERESTS AND COVID-19 LOCKDOWNS

Government-imposed lockdowns forbid “nonessential” businesses from operating and prohibit large social gatherings.6 The alleged goal of such regulations is public health: by limiting travel and interpersonal interactions, lockdowns curb the spread of Covid-19. Lockdown implementation, however, may be driven by politicians and well-organized interest groups seeking to benefit themselves rather than the public. Lockdowns confer net benefits on some, but they impose net costs on others and can be crippling to businesses designated “nonessential.” Such regulations are therefore redistributive by nature, creating motivation and opportunity for private interests to influence them. Desierto and Koyama (2020), for example, study variation in the severity of initial government lockdowns around the world. To explain that variation they develop a “selectorate model” according to which incumbent politicians maintain the support of coalitions of voters on the margin of defection by tailoring policies to those coalition members’ interests. Swing coalitions seek to either avoid Covid infection or avoid economic recession. More severe lockdowns redistribute wealth from the economically vulnerable to the medically vulnerable. Less severe lockdowns redistribute wealth in the opposite direction. The relative vulnerability of swing coalitions thus determines the severity of the lockdown imposed by a support-seeking political incumbent. Desierto and Koyama find support for their selectorate model in lockdowns across the globe. The primarily economic concerns of swing coalitions in Brazil, Sweden, the UK, and the US, for example, may account for these countries’ comparatively less severe lockdowns. In contrast, the infection vulnerability of swing coalitions in Argentina, China, and Russia may account for those countries’ comparatively more severe lockdowns. Across all countries, however, initial lockdown severity appears to reflect the private interests of members of key political coalitions rather than concern for public health. Pulejo and Querubín (2021) offer an alternative explanation for variation in lockdown stringency internationally. They argue that because voters hold politicians responsible for the economy’s health near elections, incumbent presidents who are seeking reelection have an incentive to implement more moderate lockdowns. By imposing less stringent lockdowns, incumbents avoid more severe economic contractions, thereby improving their reelection

400  Handbook on the political economy of health systems prospects whatever the consequence for public health. Pulejo and Querubín find that among countries with an incumbent president seeking reelection, lockdowns tend to be less severe where election day is nearer. Lockdown implementation thus seems to be concerned not with correcting externalities but with advancing the political careers of incumbent presidents facing reelection. Redford and Dills (2021) explore the influence of private interests on governmental lockdown designations of businesses as “essential” or “nonessential.” When the Covid-19 pandemic began, most US states designated liquor stores and marijuana dispensaries essential, exempting them from lockdown closure. There were, however, two notable exceptions: Pennsylvania and Massachusetts. The governors of both states implemented lockdowns in March 2020. Pennsylvania, however, initially designated liquor stores nonessential, while Massachusetts designated recreational marijuana dispensaries nonessential. Redford and Dills contend that these lockdown policies reflect interest group considerations. Liquor stores in Pennsylvania are owned by the state and operated by the Pennsylvania Liquor Control Board. Most employees of that board belong to a union: the United Food and Commercial Workers Local 1776 Keystone State. According to Redford and Dills, this union was especially well positioned to influence the state’s lockdown policies in its favor since, unlike other state policies, the state’s lockdown scope is controlled solely by the governor. Seizing the opportunity this situation presented, the union lobbied Pennsylvania’s governor to declare liquor stores nonessential businesses—thereby shutting them down—but to continue to pay liquor store employees at full rate while the lockdown lasted. The governor obliged. The result was a paid vacation for the union’s members courtesy of Pennsylvania taxpayers. Four weeks later, however, Pennsylvania’s governor introduced a policy that discontinued pay for all state employees who were not at work. The union abruptly changed its mind about the essentiality of liquor stores and lobbied to have them redesignated essential. The governor obliged again. The paid vacation had ended, but unlike the employees of many other businesses, liquor store employees could return to remunerative work. In Massachusetts it was not liquor stores but marijuana dispensaries whose subjection to lockdown policies was influenced by well-organized interest groups—or rather their absence.7 Marijuana dispensaries can be medical or recreational. Medical dispensaries require customers to show a state-registered medical marijuana card before they can make purchases. Recreational dispensaries do not. Most US states designated both medical and recreational marijuana dispensaries essential business. Massachusetts, however, designated recreational marijuana dispensaries nonessential. The reason, Redford and Dills argue, is that a key interest group favoring continued access to recreational marijuana that was present in other states was not present in Massachusetts: military veterans. Veterans cannot register for medical marijuana cards without losing their military benefits. Hence, recreational dispensaries are veterans’ primary source of marijuana. Because veterans comprise a much smaller proportion of the population in Massachusetts than in other states, veterans’ groups in Massachusetts failed to defend recreational dispensaries from nonessential designation. Daumann and Follert (2020) consider the influence of private interests on Covid policy in Germany when the pandemic began. Their concern is not lockdowns per se but the reaction of interest groups to governmental Covid responses more generally. Daumann and Follert argue that in nonemergency situations, interest groups can secure policies that redistribute wealth to their members by supplying politicians with selective information. Voters have weak incentives to acquire detailed information about the behavior of politicians, and politicians have

Covid-19 and the interest group approach to government  401 weak incentives to acquire detailed information about voters’ preferences. Politicians thus rely on the information available to them most cheaply: the information that interest groups provide them. Emergency contexts like the one that Covid-19 presents may exacerbate such reliance. The novelty of the virus and the initial uncertainty of pandemic policy efficacy compound the foregoing knowledge problems, creating additional opportunities for interest groups to seek rents and a corresponding flurry of additional rent-seeking activity. Daumann and Follert provide evidence of such a flurry in Germany at the pandemic’s outset. They draw on German newspaper articles that describe how interest groups deluged political officials with funding requests while Germany’s government was developing its Covid response.

4.

PRIVATE INTERESTS AND COVID-19 RESOURCE ALLOCATION

Governments are among the important distributors of Covid resources, from vaccines to ventilators to coronavirus tests. The public interest approach to government suggests that such resources are distributed optimally from the perspective of social welfare: political actors allocate Covid resources where their social value (accounting for equity considerations) is maximized. The interest group approach to government, in contrast, suggests that Covid resources are distributed optimally from the perspective of the distributors’ private welfare: political actors allocate those resources where their value to political actors is maximized. Only one paper to date considers the potential influence of private interests on governmental allocation of Covid resources. After discussing that paper, we consider possible further applications of the theme. Coyne et al. (2021) highlight four ways in which American Covid policy may deviate from socially optimal Covid policy because political actors or well-organized interest groups pursue their private interests. Two of those ways identify potential private-interest pressures on Covid resource allocation specifically. First, because federal, state, and local political actors are beholden to different constituents with different priorities, one-size-fits-all policies like national or state-level social distancing mandates may meet resistance from state and local officials whose reappointment depends on how well they serve the interests of their locale.8 Second, politicians at higher levels of government have authority and incentives to allocate Covid resources such as medical equipment to electorally important areas, and there is no reason to think that Covid resources have higher social value in those areas. Third, well-organized interest groups have strong incentives to lobby for Covid resource distributions and policies that benefit their members, and often what benefits their members may not benefit public health. Fourth, public health agencies have strong incentives to lobby the federal government for larger resource allocations at the expense of other health agencies, with the result that Covid resource distribution may reflect the relative effectiveness of health agencies’ lobbying efforts rather than the distribution that maximizes social welfare. Coyne et al.’s analysis is conceptual and thus does not empirically investigate the potential private-interest influences on governmental resource allocation that it identifies. Previous work that considers governmental allocation of health resources in non-Covid contexts, however, supports the likelihood of those influences in the Covid context as well. Below we review some of that work with the goal of highlighting possible applications to Covid.

402  Handbook on the political economy of health systems Cooper et al. (2020) study the influence of private interests on the allocation of public health resources via America’s Medicare Modernization Act of 2003. That law, which expanded Medicare to cover prescription drugs for seniors, reflected the largest expansion of the program in Medicare’s history. Its sweeping nature, however, made finding lawmaker support for enacting the law difficult. Reelection motives incentivize lawmakers to pass legislation that confers direct benefits on their constituents, for that is the kind of legislation for which lawmakers can claim credit. In general, lawmakers cannot plausibly claim credit for sweeping legislation like the Medicare Modernization Act. Few lawmakers therefore have a strong interest in supporting such legislation. If, however, sweeping legislation is modified to include pork9 that targets the constituents of otherwise disinterested lawmakers, the support of those lawmakers for sweeping legislation may be secured. Cooper et al. argue that Olsonian “selective benefits” for lawmakers propelled the passage of the Medicare Modernization Act. The act secured requisite legislative support only after the addition of Section 508, a provision that permitted hospitals to apply for larger Medicare payments. Cooper et al. find that hospitals represented by members of Congress who voted for the Medicare Modernization Act were more likely to receive more generous Medicare payments and tended to receive much larger payments than hospitals represented by their colleagues who voted against the Act. The addition of Section 508 thus constituted provision of the healthcare pork necessary to motivate sufficient “yea” votes. That pork, moreover, took on a life of its own. The payment increases enabled by Section 508 were meant to expire three years after enactment. Before they could do so, however, the hospitals that received them formed the Section 508 Hospital Coalition. Through large campaign contributions to relevant congressmen, the newly created interest group successfully secured the extension of supernormal payments. Cooper et al.’s insights might be usefully applied to study American Covid legislation. In March 2021 President Joe Biden signed into law the American Rescue Plan. That plan was worth $1.9 trillion, at least $428 billion of which was ostensibly intended to promote public health: “funding for vaccine distribution and testing to combat the Covid-19 pandemic, making policy adjustments to the Medicaid program, facilitating health insurance coverage, and providing more money for healthcare providers. The final bill also makes two narrowly focused technical Medicare payment changes” (Whitlock and Zimmerman 2021, p. 1). If the sweeping nature of the Medicare Modernization Act necessitated the addition of substantial pork to earn the support of a sufficient number of lawmakers, the still more sweeping—and expensive—American Rescue Plan seems likely to have done so as well. Hegde (2009) studies the influence of private interests on the allocation of public health resources via the National Institutes of Health (NIH), the primary government agency responsible for biomedical and health-related research in the United States.10 The NIH funds research conducted by successful grant applicants such as universities, disease research centers, and businesses. Health-related interest groups seek to channel these funds to their members and, for that purpose, lobby legislators who have influence over the NIH’s budget. Hegde observes that the NIH’s budget is appropriated by the House Appropriations Subcommittee for Labor, Health and Human Services, Education and Related Agencies. To protect NIH’s independence, its budgetary appropriations are not earmarked for specific investigators. Hegde, however, argues that health-related interest groups shape NIH budget appropriations nonetheless. Unable to earmark NIH support explicitly for specific research performers in their constituencies, budget-appropriating legislators steer support to those performers in the form of “soft

Covid-19 and the interest group approach to government  403 earmarks.” Such earmarks encourage the NIH to use appropriations to fund the biomedical research fields and projects in which the researchers in budget-appropriating legislators’ constituencies are engaged. Hegde finds that researchers in the states of appropriations’ subcommittee members receive more NIH research funding than researchers at unrepresented institutions. In a related paper, Hegde and Sampat (2015) examine interest group lobbying for soft earmarks in NIH budget appropriations. They find that more active lobbying by disease-specific interest groups is associated with larger numbers of soft earmarks for the diseases with which those interest groups are concerned. Similar rent seeking may have influenced the allocation of US government funding for coronavirus research. In March 2020, Congress allocated $10 billion to Operation Warp Speed, a program for accelerating vaccine development, manufacture, and distribution (GAO. gov 2021). Additionally, Congress allocated $3.6 billion to the NIH to fund research on Covid-19 (COVID19.nih.gov 2021). Public health concerns motivated this funding. But scholarly examination may find that private interests influenced its distribution. Ryan (2014) studies the influence of private interests on the US government’s allocation of the H1N1-virus vaccine following the unexpected spread of swine flu in 2009. That vaccine’s allocation was the responsibility of Department of Health and Human Services. Political oversight of the department’s vaccine distribution fell to the Committee on Energy and Commerce (in the House of Representatives) and to the Committee on Health, Education, Labor, and Pensions (in the Senate). Congressional dominance theory (Weingast and Moran 1983)—an extension of the interest group approach to government to encompass the control of bureaucrats—suggests that specialized congressional committees like those mentioned above will use their budgetary and oversight responsibilities to ensure that bureaucrats use their authority in ways that congressional committee members desire: in a manner that serves the committee members’ interests. Ryan finds support for that theory in the allocation of the H1N1 vaccine. States with Democratic representatives on the Committee on Energy and Commerce received disproportionately large allocations of the vaccine relative to states without committee representatives. Initial distributions of the vaccine were not allocated to states with larger at-risk populations (pregnant women, young adults, and young children). Furthermore, once the weekly distribution of vaccines increased sharply, the influence of committee membership on vaccine allocation declined sharply. As the value of being allocated additional vaccine fell, committee members’ interest in influencing its allocation did too. The United States is currently in the middle of a Covid-19 vaccine rollout.11 Early news reports suggest that vaccine development and distribution may have been subject to political influence (see, for instance, NYTimes.com 2021, WSJ.com 2021). Rigorous investigation of that possibility warrants scholarly analysis. Mendoza (2015) analyzes the influence of private interests on the allocation of public health resources via the essential health benefits mandate of America’s Affordable Care Act of 2010 (ACA).12 That health insurance mandate, known colloquially as Obamacare, required health insurance plans in the individual and small-group insurance markets to cover “essential health benefits.” Which health benefits are “essential” is not obvious, and which benefits are deemed essential affects the demand for the services of different healthcare providers. Healthcare provider interest groups thus have a strong incentive to see that the services they provide are designated as essential. The result for the ACA, Mendoza argues, was an essential health benefits definition that reflected interest group rent seeking. Lobbying by the American

404  Handbook on the political economy of health systems Chiropractic Association and the American Society of Plastic Surgeons, for example, resulted in chiropractic and prosthetic care being designated as essential. Camobreco (1996) examines the influence of private interests on the allocation of public health resources via optional Medicaid spending in the US states.13 Under Medicaid, states have the option to cover additional health services and individuals beyond those required by federal law. Camobreco finds that stronger hospital interest groups and stronger nursing interest groups were both associated with more optional Medicaid spending. Kousser (2002) also studies optional state Medicaid spending. He finds that the strength of physician, senior citizen, and medical industry interest groups is positively related to optional spending, as is Democrat control of the state legislature. Public opinion, in contrast, is unrelated to optional state Medicaid spending. Similarly, Grogan (1994) investigates optional state Medicaid policies with respect to recipient eligibility and programs. She finds that the strength of healthcare interest groups is positively related to policies that more generously define categories of eligible recipients and offer more generous programs. Pracht and Moore (2003) consider state Medicaid drug reimbursements to pharmacies. Pharmacies benefit from more generous reimbursement rates in the form of lower ingredient and dispensing costs. Pracht and Moore find that the share of state pharmacists who are members of the American Pharmaceutical Association is strongly and positively related to state reimbursement rates. The Covid-centric American Rescue Act, discussed above, expanded eligibility for subsidized health insurance under the Affordable Care Act and increased subsidies to premiums covered under the 1985 Consolidated Omnibus Budget Reconciliation Act (Whitlock and Zimmerman 2021). Scholarly analysis of these or other provisions in the American Rescue Act may uncover similar rent-seeking influences. Philipson and Posner (1993) consider the influence of private interests on the allocation of public health resources to combatting AIDS. In fiscal year 2019, the US federal government allocated more than $28 billion to domestic AIDS services, research, and programs (HIV. gov 2020). Three-and-a half billion dollars of America’s domestic AIDS budget is allocated to research and prevention alone. Philipson and Posner argue that government-funded AIDS research, subsidized HIV testing, and AIDS education programs in the United States reflect the efforts of AIDS interest groups to redistribute wealth from the mass of taxpayers to their members. They suggest that two major populations that stand to benefit significantly from public spending on AIDS—homosexual males, along with medical professionals and drug producers—satisfy the conditions for influential interest groups: their members are concentrated geographically, modest in number, tend to be well educated, and tend to have above-average incomes. Homosexual males, according to Philipson and Posner, seek public spending for AIDS because they constitute the population at high risk of contracting AIDS. Medical professionals and drug producers seek such spending because it increases the demand for their services and subsidizes their inputs. Well-organized AIDS interest groups may therefore be responsible for the fact that AIDS receives such large public health resource allocations. The federal response to Covid-19 has been much larger than its response to AIDS. According to the Committee for a Responsible Federal Budget, as of March 2021, Congress has distributed $128 billion in federal grants and funding to the health industry to help “various health care providers & agencies to respond to COVID-19” (COVIDmoneytracker.org 2021). While Covid threatens the health of far more Americans than AIDS, the virus is disproportion-

Covid-19 and the interest group approach to government  405 ately dangerous to the elderly, and governmental Covid expenditures are disproportionately beneficial to medical professionals, drug producers, and researchers. Future work might therefore examine the degree to which federal Covid expenditures reflect the advancement of public health versus the advancement of private interests.

5.

PERVERSE CONSEQUENCES OF GOVERNMENTAL RESPONSES TO COVID-19

When public health policies are designed to benefit private interests rather than the public, they are unlikely to promote public health and may even injure it, if not in the specific areas the policies target then in others. Such consequences are perverse. Below we survey work that considers perverse consequences of governmental responses to Covid-19.14 Lockdowns are intended to curb the spread of Covid-19 by limiting travel and interpersonal interactions. Some lockdown policies, however, may have the opposite effect. Redford and Dills (2021), for example, argue that Pennsylvania’s short-lived designation of liquor stores as nonessential, discussed above, encouraged more travel and less social distancing, potentially hastening the disease’s spread. Two factors account for this perverse effect. First, because neighboring states designated liquor stores essential, Pennsylvania’s nonessential designation rendered liquor in the former states relatively less expensive to Pennsylvanians. Pennsylvania’s lockdown policy thus encouraged its citizens to travel outside the state to purchase liquor, contributing to the virus’ potential spread across states. Second, when Pennsylvania redesignated liquor stores as essential four weeks into its lockdown, it did so without raising liquor prices. Pennsylvanians thus rushed to purchase liquor, resulting in crowded liquor stores and long lines that contributed to the virus’ potential spread within the state. Lee et al. (2021) study the effect of lockdowns and border closures on the spread of Covid-19 in India, Bangladesh, and Pakistan. These countries are among the largest suppliers of migrant workers in the world. When countries that employ migrant workers announce lockdowns, those workers—now prevented from working—are incentivized to return home before borders close. Migrants returning from abroad may bring infection with them, spreading Covid in their home countries. Lee et al. find that in India, Bangladesh, and Pakistan, districts that were home to more people working abroad tended to have higher rates of Covid infection after international border closures in March 2020. Anti-price gouging laws are popular regulations in times of crisis, and they have been popular with US state governments amid the Covid-19 pandemic. Proponents of these laws hope that by imposing price ceilings on goods used to cope with Covid—for example hand sanitizer and masks—and other goods, such as toilet paper, the regulations will promote public health through making the former goods more accessible and promote “fairness” through preventing “profiteering.” Chakraborti and Roberts (2020a) analyze the effect of anti-price gouging laws precipitated by the Covid-19 pandemic applied to hand sanitizer and toilet paper in the US states (see also Chakraborti and Roberts 2020b). In unregulated markets, when demand for a product surges unexpectedly as it did for hand sanitizer and toilet paper when the pandemic began, quantity demand may temporarily exceed quantity supplied, producing fleeting shortages. With time, price rises, restoring market equilibrium. Price ceilings, however, interfere with that adjustment process.

406  Handbook on the political economy of health systems Chakraborti and Roberts find that the consequences of this interference in the context of Covid-related price ceilings were two-fold and perverse. First, as conventional economic reasoning predicts, anti-price gouging laws exacerbated shortages of hand sanitizer and toilet paper, which became less accessible instead of more accessible. States with anti-price gouging laws tended to have more online shopping searches for hand sanitizer and toilet paper than states without such laws, implying that more severe physical-store shortages plagued the former states than the latter. Second, by exacerbating shortages, anti-price gouging laws encouraged consumers to hunt across physical stores for hand sanitizer and toilet paper, potentially facilitating the spread of the virus. States with anti-price gouging laws tended to have more intrastate travel than states without such laws, implying that the policy incentivized consumers to engage in more virus-spreading activity in the former states than in the latter. Perhaps the most widely recognized perverse consequence of governmental responses to Covid-19 is the potentially deleterious effect of lockdowns on mental health, which may induce loneliness, social isolation, or depression.15 Multiple studies find that government-imposed lockdowns have had such effects (see, for example, Armbruster and Klotzbücher 2020, Brodeur et al. 2020, Etheridge and Spantig 2020, Tubadji et al. 2020).16 Altindag et al. (2021), for instance, investigate the effect of Turkey’s lockdown on the mental health of the people that policy was supposed to benefit the most: the elderly. In March 2020 the Turkish government imposed a targeted lockdown on people aged 65 and older. The lockdown lasted approximately three months. Altindag et al. find that Turkish survey participants aged 65 and older reported more mental and physical symptoms of depression on average than those just below the cutoff age of 65. The authors attribute this finding to reduced social interaction and physical activity by the elderly pursuant to the lockdown.17 Another governmental response to Covid-19 whose potential for perverse consequences has received attention is governmental regulation of Covid tests, treatments, and vaccines. That regulation is intended to promote public health by protecting consumers from ineffective or dangerous healthcare products. But by preventing or delaying consumer access to effective products, such regulation may undermine public health instead. March (2021) studies the US Food and Drug Administration’s (FDA’s) gatekeeping role early in the Covid-19 pandemic. He reports that five weeks after the first confirmed Covid case in the United States, scarcely more than 40 laboratories across the country had FDA approval to administer Covid tests, with the result that just 1,235 US patients were tested for the virus. Eight weeks after the first Covid infection in the US, just 22 patients had FDA approval to be treated with remdesivir, one of the first experimental treatments found to mitigate Covid severity. Moreover, while the pharmaceutical company Moderna developed its Covid vaccine a week before the US confirmed its first Covid case, the FDA did not approve the vaccine for public use until eleven months later. It seems unlikely, then, that FDA regulation of Covid tests, treatments, and vaccines promoted public health more than it harmed public health. March attributes this perverse consequence to the incentives of FDA officials, who incur minimal costs for delaying approval of effective health products because the people thereby injured are rarely observable. In contrast, the people injured by a drug or treatment that the FDA approves are readily observable, exposing the agency to severe criticism. FDA officials thus are overly cautious gatekeepers. Other perverse effects resulting from governmental regulation amid the Covid-19 pandemic result not from governmental responses to the virus per se but from preexisting public health regulations that hamstring the responses of nongovernmental actors to the virus. An example

Covid-19 and the interest group approach to government  407 of that phenomenon is furnished by US certificate-of-need laws, which were in force in 36 states when the pandemic began. Those laws prohibit hospitals from expanding their facilities, equipment, or services without state permission, thereby securing hospitals in poor areas sufficient market share to remain profitable. While the ostensible purpose of this regulation is to promote public health by improving healthcare access for the poor, at least once Covid-19 emerged, the regulation became a detriment to public health. Ghosh et al. (2020) examine the effect of certificate-of-need laws on Covid and non-Covid mortality rates during pandemic-induced medical equipment shortages. By preventing hospitals from expanding their medical equipment in the face of the sudden, Covid-caused increase in demand for hospital services, certificate-of-need laws compelled hospitals to use their existing equipment more intensively. Thus in hospitals near or at capacity, doctors might, for example, redeploy ventilators from pneumonia patients to severe Covid patients, increasing the former’s chance of death. Ghosh et al. find that in states with hospitals near capacity that did not temporarily relax certificate-of-need laws, mortality rates were higher for diseases that share treatment procedures with Covid-19 (for example, septicemia, influenza, pneumonia, and chronic lower respiratory disease) than in states that relaxed certificate-of-need regulations.

6. CONCLUSION Despite the rapid growth of economic scholarship that considers Covid-19, little research has addressed governmental responses to the pandemic from the perspective of the interest group approach to government. That situation, however, is beginning to change. This chapter surveyed the small but burgeoning literature at the intersection of Covid-19 and the interest group approach to government. Our survey focused on three themes: (1) private interests may influence how government-imposed lockdowns are implemented, (2) private interests may influence governmental allocation of Covid resources, and (3) governmental responses to Covid-19 may have perverse consequences, undermining rather than promoting the health of vulnerable people. While, to date, only one study has addressed the second of these themes, we offered some ideas for further work in that area inspired by research that examines the influence of private interests on governmental allocation of health resources in non-Covid contexts. All three themes, however, would benefit from additional attention by economists persuaded of the productivity of the interest group approach to government. We hope this survey is useful to them.

NOTES 1.

See Boettke and Powell (2021), Candela and Geloso (2021), Choutagunta et al. (2021), Costa-Font et al. (2020a), and McCannon and Hall (2021). In the context of non-Covid public health interventions, see Costa-Font et al. (2020b). 2. See Leeson and Rouanet (2021). 3. See Duchin and Hackney (2020). 4. This section draws its discussion from Leeson and Thompson (2023). 5. That cost includes the expense of not only organizing per se but also, for example, obtaining information and controlling organizational free riding.

408  Handbook on the political economy of health systems 6. 7. 8. 9. 10. 11. 12. 13. 14.

15. 16.

17.

Allen (2021) persuasively argues that most recent studies on lockdown policy simultaneously overestimate the benefits of lockdowns and underestimate their costs due to reliance on inappropriate measures of lockdown benefits and incomplete measures of lockdown costs. Barrett (2006) studies the delayed—and nearly failed—global eradication of smallpox. He attributes the governmental (non)response that contributed to that delay to the absence of rent-seeking interest groups. Adolph et al. (2021) find that the political party of state governors and their electorates affected the timing of social-distancing mandates after the community transmission of Covid was first detected in February 2020. “Pork” comes from the idiomatic phrase in the US “pork barrel politics,” referring to the trading of favors for votes or campaign contributions, where “pork” is the government spending for local projects that benefits only a politician’s district or supporters. See also Ward and Dranove (1995) and Batinti (2016). At the time of writing, 131 million Americans have received at least one dose of Covid-19 vaccine (Covid.cdc.org 2021). See also Spithoven (2016), who studies ACA spending provisions. Medicaid is financed jointly by the federal and state governments, with every dollar appropriated by the latter matched by at least 50 cents from the former. Effects of this nature are commonly called “unintended consequences,” and sometimes that appellation may be correct (see Coyne et al. 2021 and Storr et al. 2021 for the “unintended consequences” framework applied broadly to government action during the Covid-19 pandemic). We prefer, however, to call such consequences “perverse” since the interest group approach to government suggests that they often are intended—or, if not exactly intended, are at least expected. Of course, the pandemic itself may also have deleterious consequences for mental health. See, for example, Beland et al. (2020), who examine those consequences in Canada. Cronin and Evans (2020) find that US nursing homes that more severely isolated residents to prevent Covid infection tended to have higher non-Covid mortality rates. The authors interpret this finding as evidence that extreme social isolation meant to protect nursing-home residents from the virus imposed on them an emotional toll that may have hastened their death. Mulligan (2020) investigates excess fatalities in the US not directly related to Covid-19 during the first nine months of the pandemic. He finds that about 30,000 such deaths occurred. Mulligan suggests that such fatalities may be “deaths of despair” driven by social-distancing and isolation measures. Cronin and Evans (2020) and Mulligan (2020), however, do not distinguish between the effects of voluntary lockdown-type measures and government-imposed lockdowns, so the extent to which governmental responses to Covid are responsible for these perverse health outcomes is unclear (see also Bianchi et al. 2021). Negative mental health consequences are not the only potential negative health consequences of lockdown policies, nor must such consequences be limited to the elderly. Takaku and Yokoyama (2021) find that school closures in Japan are associated with increases in the weight of children unable to attend school in person.

REFERENCES Adolph, C., Amano, K., Bang-Jensen, B., Fullman, N., & Wilkerson, J. (2021). Pandemic politics: Timing state-level social distancing responses to COVID-19. Journal of Health Politics, Policy and Law, 46(2), 211–33. Allen, D.W. (2021). Covid lockdown cost/benefits: A critical assessment of the literature. Mimeo. Altindag, O., Erten, B., & Keskin, P. (2021). Mental health costs of lockdowns: Evidence from age-specific curfews in Turkey. IZA Discussion Paper Series No. 14281. Armbruster, S., & Klotzbücher, V. (2020). Lost in lockdown? Covid-19, social distancing, and mental health in Germany. University of Freiburg, Wilfried Guth Endowed Chair for Constitutional Political Economy and Competition Policy Discussion Paper Series.

Covid-19 and the interest group approach to government  409 Arrow, K.J. (1963). Uncertainty and the welfare economics of medical care. American Economic Review, 53(5), 941–73. Barrett, S. (2006). The smallpox eradication game. Public Choice, 130(1–2), 179–207. Batinti, A. (2016). NIH biomedical funding: Evidence of executive dominance in swing-voter states during presidential elections. Public Choice, 168(3–4), 239–63. Becker, G.S. (1983). A theory of competition among pressure groups for political influence. Quarterly Journal of Economics, 98(3), 371–400. Beland, L., Brodeur, A., Mikola, D., & Wright, T. (2020). The short-term economic consequences of COVID-19: Occupation tasks and mental health in Canada. IZA Discussion Paper Series No. 13254. Bianchi, F., Bianchi, G., & Song, D. (2021). The long-term impact of the COVID-19 unemployment shock on life expectancy and mortality rates. NBER Working Paper 28304. Boettke, P.J., & Powell, B. (2021). The political economy of the Covid-19 pandemic. Southern Economic Journal, 87(4), 1090–106. Brodeur, A., Clark, A.E., Fleche, S., & Powdthavee, N. (2020). COVID-19, lockdowns and well-being: Evidence from Google trends. Journal of Public Economics, 193, 104346. Camobreco, J.F. (1996). Medicaid and collective action. Social Science Quarterly, 77(4), 860–76. Candela, R., & Geloso, V. (2021) Economic freedom, pandemics and robust political economy. Southern Economic Journal, 87(4), 1250–66. CEPR.org (2021). Covid Economics: Vetted and Real-Time Papers. 1 April 2020. Chakraborti, R., & Roberts, G. (2020a). Anti-gouging laws, shortages, and COVID-19: Insights from consumer searches. Journal of Private Enterprise, 35(4), 1–20. Chakraborti, R., & Roberts, G. (2020b). Learning to hoard? The effects of preexisting and surprise price-gouging regulation during the COVID-19 pandemic. SSRN Working Paper. https://​dx​.doi​.org/​ 10​.2139/​ssrn​.3672300. Choutagunta, A., Manish, G.P., & Rajagopalan, S. (2021) Battling COVID-19 with dysfunctional federalism: Lessons from India. Southern Economic Journal, 87(4), 1267–99. Cooper, Z., Kowalski, A.E., Powell, E.N., & Wu, J. (2020). Politics and health care spending in the United States. NBER Working Paper 23748. Costa-Font, J., Levaggi, R., & Turati, G. (2020a). Resilient managed competition during pandemics: Lessons from the Italian experience during COVID-19. Health Economics, Policy and Law, 17(2), 212–9. Costa-Font, J., Turati, G., & Batinti, A. (2020b). The Political Economy of Health and Healthcare: The Rise of the Patient Citizen. Cambridge: Cambridge University Press. COVID19.nih.gov (2021). Funding. 9 April 2021. Covid.cdc.org (2021). Covid data tracker. 18 April 2021. COVIDmoneytracker.org (2021). COVID money tracker: Tracking the COVID response. 9 April 2021. Coyne, C.J., Duncan, T.K., & Hall, A.R. (2021) Political economy of state responses to infectious disease. Southern Economic Journal, 87(4), 1119–37. Cronin, C.J., & Evans, W.N. (2020). Nursing home quality, COVID-19 deaths, and excess mortality. NBER Working Paper 28012. Daumann, F., & Follert, F. (2020). COVID-19 and rent-seeking competition. New Perspectives on Political Economy, 16(1–2), 52–69. Desierto, D., & Koyama, M. (2020). Health vs. economy: Politically optimal pandemic policy. Journal of Political Institutions and Political Economy, 1(4), 645–69. Duchin, R., & Hackney, J. (2020). Buying the vote? The economics of electoral politics and small business loans. COVID Economics: Vetted and Real-Time Papers, 62, 20–65. Etheridge, B., & Spantig, L. (2020). The gender gap in mental well-being during the Covid-19 outbreak: Evidence from the UK. ISER Working Paper Series. GAO.gov (2021). Operation Warp Speed: Accelerated COVID-19 vaccine development status and efforts to address manufacturing challenges. 9 April, 2021. Ghosh, S., Choudhury, A.R., & Plemmons, A. (2020). Certificate-of-need laws and healthcare utilization during COVID-19 pandemic. SSRN Working Paper. https://​dx​.doi​.org/​10​.2139/​ssrn​.3663547. Grogan, C.M. (1994). Political-economic factors influencing state Medicaid policy. Political Research Quarterly, 47(3), 589–622.

410  Handbook on the political economy of health systems Hegde, D. (2009). Political influence behind the veil of peer review: An analysis of public biomedical research funding in the United States. Journal of Law and Economics, 52(4), 665–90. Hegde, D., & Sampat, B. (2015). Can private money buy public science? Disease group lobbying and federal funding for biomedical research. Management Science, 61(10), 2281–98. HIV.gov (2020). Federal domestic HIV/AIDS programs and research spending. 12 May 2020. Kousser, T. (2002). The politics of discretionary Medicaid spending, 1980–1993. Journal of Health Politics, Policy and Law, 27(4), 639–71. Lee, J.L., Mahmud, M., Morduch, J., Ravindran, S., & Shonchoy, A.S. (2021). Migration, externalities, and the diffusion of COVID-19 in South Asia. Journal of Public Economics, 193, 104312. Leeson, P.T., & Rouanet, L. (2021). Externality and COVID-19. Southern Economic Journal, 87(4), 1107–18. Leeson, P.T., & Thompson, H.A. (2023). Public choice and public health. Public Choice, 195(1–2), 5–41. March, R.J. (2021). The FDA and the COVID-19: A political economy perspective. Southern Economic Journal, 87(4), 1210–28. McCannon, B., & Hall, J. (2021). Stay-at-home orders were issued earlier in economically unfree states. Southern Economic Journal, 87(4), 1138–51. McCormick, R.E., & Tollison, R.D. (1981). Politicians, Legislation, and the Economy: An Inquiry into the Interest-Group Theory of Government. Boston: M. Nijhoff. Mendoza, R.L. (2015). Rents from the essential health benefits mandate of health insurance reform. Health Marketing Quarterly, 32(2), 197–215. Mulligan, C.B. (2020). Deaths of despair and the incidence of excess mortality in 2020. NBER Working Paper 28303. NBER.org (2021). COVID-19. 1 April 2020. NYTimes.com (2021). Scientists worry about political influence over coronavirus vaccine project. 11 May 2021. Olson, M. (1965). The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge: Harvard University Press. Peltzman, S. (1976). Toward a more general theory of regulation. Journal of Law and Economics, 19(2), 211–40. Philipson, T.J., & Posner, R.A. (1993). Private Choices and Public Health: The AIDS Epidemic in an Economic Perspective. Cambridge: Harvard University Press. Pigou, A.C. (1920). The Economics of Welfare. London: Macmillan. Pracht, E.E., & Moore, W.J. (2003). Interest groups and state Medicaid drug programs. Journal of Health Politics, Policy and Law, 28(1), 9–39. Pulejo, M., & Querubín, P. (2021). Electoral concerns reduce restrictive measures during the COVID-19 pandemic. Journal of Public Economics, 198, 104387. https://​doi​.org/​10​.1016/​j​.jpubeco​.2021​.104387. Redford, A., & Dills, A.K. (2021). The political economy of substance misuse during the COVID-19 pandemic. Southern Economic Journal, 87(4), 1175–209. Ryan, M.E. (2014). Allocating infection: The political economy of the swine flu (H1N1) vaccine. Economic Inquiry, 52(1), 138–54. Spithoven, A. (2016). The influence of vested interests on healthcare legislation in the USA, 2009–2010. Journal of Economic Issues, 50(2), 630–38. Stigler, G.J. (1971). The theory of economic regulation. Bell Journal of Economics and Management Science, 2(1), 3–21. Stigler, G.J. (1976). The size of legislatures. Journal of Legal Studies, 5(1), 17–34. Storr, V.H., Haeffele, S., Lofthouse, J.K., & Grube, L.E. (2021). Essential or not knowledge problems and COVID-19 stay-at-home orders. Southern Economic Journal, 87(4), 1229–49. Takaku, R., & Yokoyama, I. (2021). What the COVID-19 school closure left in its wake: Evidence from a regression discontinuity analysis in Japan. Journal of Public Economics, 195, 104364. Tubadji, A., Boy, F., & Webber, D.J. (2020). Narrative economics, public policy and mental health. COVID Economics: Vetted and Real-Time Papers, 20, 109–31. Tullock, G. (1967). The welfare costs of tariffs, monopolies, and theft. Western Economic Journal, 5(3), 224–32.

Covid-19 and the interest group approach to government  411 Tuohy, C.H., & Glied, S. (2011). The political economy of health care. In S. Glied & P.C. Smith, eds. The Oxford Handbook of Health Economics, pp. 58–77. Oxford: Oxford University Press. Ward, M.R., & Dranove, D. (1995). The vertical chain of research and development in the pharmaceutical industry. Economic Inquiry, 33(1), 70–87. Weingast, B.R., & Moran, M.J. (1983). Bureaucratic discretion or congressional control? Regulatory policymaking by the Federal Trade Commission. Journal of Political Economy, 91(5), 765–800. Whitlock, R., & Zimmerman, E. (2021). American rescue plan act of 2021: Key healthcare provisions. McDermott+ Consulting. 1 April 2021. WSJ.com (2021). Andrew Cuomo investigation expands as state attorney general looks at aide’s calls. 11 May 2021.

24. Political preferences and nudging for healthcare: evidence from Covid certificates Mario Cesare Nurchis, Luca Salmasi and Gilberto Turati

1. INTRODUCTION In this chapter we discuss the use of nudging in healthcare policies, and the impact of nudging in the presence of different political/ideological preferences that might be correlated with individual choices. Among the many available examples, we concentrate on the case of Covid-19 vaccination, a salient recent political issue as no-vax groups extensively protest against government decisions in several countries. The concept of nudging was first formulated in 2008 and, since then, the academic literature in social sciences has investigated the influence of nudging strategies on individual decision-making. In broad terms, nudging is defined as “any attempt at influencing people’s judgement, choice or behavior in a predictable way, that is made possible because of cognitive boundaries, biases, routines, and habits in individual and social decision-making” (Hansen, 2016, p. 158). Cognitive limitations and biases represent hurdles for people to perform rational choices. Therefore, nudging strategies intervene regarding these hindrances by modifying the “choice architecture,” the social and physical environment where decisions are typically made (Hansen, 2016). The essence of the nudge theory is grounded on the principles of behavioral economics, challenging the standard economic assumptions based on individuals’ rationality, consistency and predictability concerning their choices, and a proper development of their opinions and belief structures when a new set of information is acquired. According to Hansen (2016), nudging techniques influence habitual behavior focusing on the individuals’ subconscious routines and biases characterizing their decision-making. The available evidence highlighting those nudging strategies is particularly suitable for public policy, given their low cost and the low amount of resources required for them to be sustained (see, for example, Yoong et al., 2020). In public policy, nudging has been a game-changer leveraging on specific interventions to steer decision-makers’ choices (Moseley, 2020). For effective policy-making, many governments, including those in the U.S., Canada, the U.K., Singapore, Australia, New Zealand, France and Italy, set up dedicated “nudging units,” also known as Behavioral Insights Teams, to enhance compliance with government policies, thus lowering social and government costs related to defiance against policies and regulations. For instance, some nudging units formulated innovative interventions, following the principles of the nudge theory, across several policy sectors, such as pension policy (in the U.S.; Beshears et al., 2005), tax compliance (in the U.K.; Behavioural Insights Team, 2012), agriculture (in Africa; Duflo et al., 2011), savings (Ashraf et al., 2006; Karlan et al., 2016) and education (Nguyen, 2008). Focusing on health policies, the use of nudging in recent years has been extensive. In particular, the adoption of nudging strategies in the healthcare sector is usually framed within the broad issue related to the inequalities in accessing healthcare services. There is abundant 412

Political preferences and nudging for healthcare  413 evidence in the literature on the factors affecting healthcare access. An up-to-date umbrella review mapped the global evidence on determinants influencing access to health services, in low-, middle- and high-income countries, investigating where the barriers to access lie (Dawkins et al., 2021). According to this review, the delay in the decision to seek care is mainly affected by demographic, socio-cultural, cognitive and financial factors. Additionally, another hurdle is represented by the distrust in public healthcare services (Santalahti et al., 2020). Hence, despite the public health systems’ efforts in ensuring the availability of services and treatments to the whole population, there is a burdensome issue concerning the lack of participation by a share of individuals who, voluntarily, decide not to access the services or not to take up the treatment. In this context, nudging strategies have been used to promote patients’ lifestyle choices such as diet (Rouyard et al., 2018), medication adherence (Reddy et al., 2017) and physical activity (Angellotti et al., 2019). But they are also widespread in the attempts to increase uptake rates of population-based screening programs, and the access to such programs. A literature review on nudging in screening found 109 articles describing nudging as including pre-set appointments, personalized letters, phone calls, monetary incentives, appeals to fear or framing of information (Hofmann & Stanak, 2018). A systematic review also demonstrated the effectiveness of nudges in improving the self-management of patients with chronic diseases (e.g., Moellenkamp et al., 2019). For instance, the SCAN randomized controlled trial showed the efficacy and cost-effectiveness of a three-step intervention, to improve cervical cancer screening uptake, based on nudging techniques (Firmino-Machado et al., 2019). Targeted invitation letters and individual consultative meetings were also found to be effective in cardiovascular screening (Dahl et al., 2021). Similarly, nudging has proved to be extremely useful for overcoming vaccine hesitancy. A systematic literature review on nudging towards vaccination showed the potentiality of implementing nudges, including reminders and re-calling people, tangible and intangible incentives, social norms and emotional associations (e.g., words, images and events), in mitigating vaccine hesitancy and increasing vaccine uptake (Reñosa et al., 2021). In this chapter, we focus on a clear and recent example of hesitance in accessing health services: the Covid-19 vaccine. Since the vaccine’s development and testing, supranational and national authorities put efforts into adopting policies as well as in investing large amounts of money to try to guarantee the worldwide 70 percent coverage target by mid-2022 as established by the World Health Organization (WHO, 2021). Over a year into the biggest vaccination campaign in history, more than 12 billion doses had been administered by June 22, 2022, across 184 countries. Nevertheless, the vaccine uptake has been heterogeneous across regions in the world, with the United Arab Emirates leading in the share of fully vaccinated people, followed by Portugal and Singapore, with clear differences also across high-income liberal democracies (Ritchie et al., 2020; Falkenbach & Willison 2022). In these countries, after the initial phase, characterized by a race towards vaccination by those most interested, it has become clear that a share of people was hesitant with respect to the Covid-19 vaccine. In this second phase of the campaign, strategies to promote vaccination varied between governments, ranging from binding constraints (like Covid-19 certificates, which were required to enter almost any public indoor spaces) to less coercive instruments (e.g., awareness campaigns on the health problems that Covid-19 might cause). Also, for Covid-19 vaccines, nudging has been proved to be useful. For instance, two recent sequential randomized controlled trials tested the effect of behavioral interventions on the uptake of Covid-19 vaccines through

414  Handbook on the political economy of health systems text-based reminders on different dates from the notification of vaccine eligibility (Dai et al., 2021). As for Covid certificates, an emerging literature has defined their adoption as a nudging technique to incentivize vaccinations across the population (Sotis et al., 2021; Spitale et al., 2022). In this chapter, we add to this literature by studying how the announcement of these measures by the government affected the uptake of vaccinations in two countries, Italy and Spain, that share a similar healthcare system. We also try to understand whether differences in the reactions to government measures vary according to political preferences. The remainder of the chapter is as follows: we first provide information on the institutional background and sketch a conceptual framework to understand how nudging can affect vaccine hesitancy. We then describe our empirical analysis, starting from presenting the data and the empirical methodology, and then discuss the results. Concluding remarks are collected in a brief final section.

2.

THE COVID CERTIFICATE: INSTITUTIONAL BACKGROUND AND CONCEPTUAL FRAMEWORK

Since 2020, countries were in the grip of several waves of Covid-19 infections, hospitalizations and deaths (Milman et al., 2021). Covid-19 is a new, deadly, viral infection (first appearing in China at the end of 2019) that affected the whole world, and required impressive and unprecedented measures by governments, which severely affected the global economy. However, at the end of 2021, the efforts by researchers to find a vaccine against Covid-19 (fueled by an unprecedented amount of public investment) paid off, with more than one vaccine authorized by public authorities in the U.S. and in Europe. Since authorization, mass production and mass vaccination campaigns have started in high-income countries, while poor countries have been lagging behind. At the beginning of the vaccination campaigns, in the middle of a race to obtain the vaccine by those most interested, the policy issue to be discussed was how to identify priority categories to whom the vaccine should be given first. But after a few months, it became clear in many countries that the policy issue would be how to combat vaccine hesitancy. The answer from several governments has been the introduction of Covid-19 certificates. Covid-19 certificates have been under assessment in many countries and mandated in others as a pre-requisite to perform some daily activities. Arguments for and against the adoption of certificates were mainly focused on their effectiveness in boosting vaccination rates as well as on ethical concerns (Sleat et al., 2021). The adoption was swift across many countries worldwide, and especially among the 27 European Union member states; the majority of them were requiring certificates by mid-2021. The main features of Covid-19 certificates vary across countries. Our analysis focuses on two European countries which differed in terms of policies related to the adoption and request of the certificates: Italy and Spain. In Spain, after the press conference by the Spanish prime minister at the end of May 2021 to present the certificate, the central government did not impose the Covid-19 certificate nationwide, with the exception of Galicia and the Canary Islands, which, at the beginning, required the proof of vaccination, testing or recovery for people to visit indoor public spaces. A few months after the announcement, some Spanish Superior Courts of Justice (i.e., the Canary Islands on July 29, 2021; Cantabria and Andalusia on August 6, Galicia on August 13 and the Balearic Islands on August 23) debated the compulsory requesting of Covid-19 certificates for leisure

Political preferences and nudging for healthcare  415 activities within the jurisdiction of the above-mentioned autonomous communities, therefore abrogating their mandatory use. As of December 2021, following the worsening of the burden of Covid-19, out of Spain’s 17 regions, 11 (i.e., the Balearic Islands on October 1, Aragon on November 24, Catalonia and Navarre on November 25, Galicia on November 26, the Canary Islands and the Valencia region on November 29, Basque Country on November 30, Andalusia on December 7, Murcia and Cantabria on December 9) received court approval to make the certificate a condition of entry in specific spaces. In Italy, the institutional scenario was radically different. Following the introduction of the national version of a Covid-19 certificate (called the Green Pass) in June 2021, the Italian prime minister announced on July 22 the mandatory use of certificates to access public transport, leisure activities and other public venues (e.g., museums, thermal baths, parks, restaurants, hotels, gyms, stadiums, amusement arcades, country festivals, religious ceremonies). On November 15, the Italian government announced the introduction of an enhanced form of Covid-19 certificate, known as the Super Green Pass, to be issued only to fully vaccinated people (i.e., those having received two doses of the vaccine) or to individuals fully recovered from Covid-19. The Super Green Pass was compulsory to access all the above-mentioned activities. As a result, individuals without the Covid-19 certificate could only access their workplace. However, at the end of December 2021, the Italian central government introduced mandatory vaccination for individuals aged 50 and over. The decision was taken amid difficulties in increasing uptake of the vaccine, given the higher risk of hospitalization for over-50s compared with other age groups. To understand the governments’ decision to introduce the Covid-19 certificate in the light of nudging, one should look at how economists traditionally frame individual choices. The standard way of thinking about an individual choice is to consider costs and benefits. These are more objective for some choices than for others. For instance, buying an ice cream in the summer implies clear costs (the price of the ice cream, the time and the effort to get it) and clear benefits (the joy of tasting a cold, sweet ice cream when it is hot), which should be easy to evaluate. But thinking about receiving a new vaccine, developed very rapidly when it usually takes many years, implies different evaluations. In these cases, when making their decisions, individuals form more subjective evaluations of costs and benefits, and their evaluations are influenced by beliefs, as well as by information received and collected by each individual. Consider individuals who have to decide whether to be vaccinated or not. They will agree to get the vaccine V if and only if their net utility U, given by their subjective benefits B minus costs C, is non-negative. One can think of two extreme cases: an individual scared by the virus will have large benefits from the vaccine and low costs, while an individual scared by the vaccine will consider low benefits and large costs; the first will get vaccinated, while the second will refuse the vaccine. These decisions are conditional on the set of information available at a given point in time I(t); hence, U(V|I(t)) = B(V |I(t)) − C(V |I(t)). Notice that costs and benefits are clearly influenced also by individual beliefs and by political/ideological preferences. For instance, many populist parties across Europe (like the Five Stars Movement [M5S] or the Northern League in Italy, or Podemos in Spain) argued that the long-term consequences of the vaccine were still unknown, and mass vaccination was a sort of large-scale experiment favoring big pharma companies. Believing in such statements was easier for people supporting these parties. Suppose now that individuals receive information from the government about a brand new digital Covid-19 certificate. The certificate is released by the government to all citizens receiv-

416  Handbook on the political economy of health systems ing the Covid-19 vaccine, and it allows them to access several public spaces, including public transport. As a substitute for vaccination, in order to get a (temporary) certificate, individuals are required to provide the results of swabs. Think back to the two cases, after receiving the information: the pro-vaccine individual will have additional reasons to support his choice, since the mandatory certificate will boost the benefits of receiving the vaccine, while costs are unchanged. On the other hand, the no-vaccine individual will now see an increase in benefits (because of the mandatory nature of the certificate) to be compared with costs. This is the nudge: the attempt to increase net utility of vaccination for “marginal” individuals – i.e., those for which the net disutility is small – by creating a small incentive. Notice that for individuals with strong ideological beliefs, nudging will not work, and the new set of information might also have the opposite effect: instead of increasing only the benefits of vaccination, it might also cause an increase in the costs of the choice, by fueling the uncertainty about the effects of vaccine. In Italy, some politicians belonging to populist parties have spoken loudly of the government being a health dictatorship imposing a vaccine with unknown consequences, which has already caused a number of deaths. These statements were never really confuted with official numbers by the government, leaving the evaluation of the costs and benefits of vaccination largely subjective.

3.

EMPIRICAL ANALYSIS: THE IMPACT OF COVID CERTIFICATE ANNOUNCEMENTS

3.1 Data To understand the impact of Covid-19 certificates in Italy and Spain as a nudging strategy, we need data on vaccinations in both countries. Data on vaccinations for Italian regions were retrieved from the Covid-19 Opendata Vaccini database. The dataset lists daily administrative information about vaccinations by region, age, class and gender, and distinguishes between first and second doses. It contains reliable data collected by the Italian Ministry of Health through the national vaccines registry. Comparable information, collected by the Spanish Ministry of Health, is available for the Spanish autonomous communities for the same period of time. Data on the regional population were obtained from the Italian National Statistical Institute (ISTAT) and the Spanish Statistical Office (INE). Using this information, we computed the vaccination rate for the two countries, defined as the ratio between the total number of first doses and the regional population (per 100,000 persons). To characterize the information set available for individuals, we used Google Trends to collect data on individuals’ interest regarding the digital Covid-19 certificate for the period from April 2021 to January 2022. Google Trends is a search trends tool displaying how frequently a specific search term is entered into Google’s search engine relative to the site’s total search volume over a given timespan. Google Trends yields keyword-related data, including search volume index and geographical information about search engine users. Google Trends was queried using “greenpass,” “certificato verde,” “certificazione verde Covid-19,” “super greenpass” and “obbligo vaccinale” as the main keywords to retrieve interest about the adoption of the digital Covid-19 certificate, the advanced form of the digital certificate and the mandatory vaccination for people aged 50 and over across the Italian regions. Meanwhile, “certificado verde,” “certificado Covid” and “pasaporte covid” were used to obtain the inter-

Political preferences and nudging for healthcare  417 est regarding the digital Covid certificate across the Spanish autonomous communities. This information is available at the regional level on a weekly basis. Furthermore, in order to investigate the impact of ideological and political preferences, we collected information about votes received by populist political parties (i.e., M5S in Italy and Podemos in Spain) during the latest political elections from the official ministerial sources. To measure the support for these political parties, we use a simple binary variable assigning a value of zero if the share of votes in region r is above the median, and 1 if it is below. 3.2

Empirical Modeling Strategy

We look at the effect of nudging on the outcome of interest; i.e., the (log of the) ratio between first doses of Covid-19 vaccine administered in region r during day d over the total population of region r. The treatment variable of interest is represented by the governments’ announcements in both countries. For the Italian government, we consider announcements about the introduction of: (i) the Covid-19 certificate (i.e., the standard Green Pass), (ii) the enhanced form of the Covid-19 certificate (i.e., the Super Green Pass) and (iii) mandatory vaccinations for specific age groups. For the Spanish government, we look at announcements about: (i) the introduction of the standard Green Pass, (ii) the verdicts of some Spanish Superior Courts of Justice abrogating the mandatory use of the standard Green Pass and (iii) the re-introduction of the standard Green Pass. Our aim is to measure how individuals’ attitude towards vaccinations change after each announcement, which is a test of the effectiveness of the nudging strategy. To this end, we rely on the Sharp Regression Kink Design (SRKD) estimator, which allows us to estimate the slope change in the outcome variable caused by the provision of new information (the “treatment” in statistical terms) to citizens. In this context, however, it is not easy to precisely define the threshold at which treatment is assigned, because there are several dates that could potentially be correct candidates. For instance, one can use the date of official press conferences; but, sometimes, news about the content of such conferences leaks out even before, because media anticipate salient information like the use of the Green Pass. In order to define when to set each threshold, we identify the week during which individuals’ interest about the Covid-19 certificate policy increases suddenly with respect to the average number of searches over the weeks immediately before. We assume that the first week where interest regarding a specific announcement increases suddenly represents the beginning of our post-treatment period. Nevertheless, since we consider multiple announcements that are consecutive over time, we need to assume that the announcement is effective in shaping vaccination rates over a certain number of weeks. Here we consider a window of six weeks centered on the announcement, corresponding to three weeks before and three weeks after the announcement. 3.3

Descriptive Evidence

Figures 24.1 and 24.2 show how interest about the digital Covid-19 certificate and vaccination rates varied in Italy and Spain during the period from April 2021 to January 2022. The dashed lines correspond to weeks recording abnormal variations in the interest towards each term, solid lines identify pre- and post-treatment periods considering a six-week window around the date of the announcement.

418  Handbook on the political economy of health systems Focusing first on Italy and the lower panel of Figure 24.1, we can immediately see that vaccination rates initially reached a peak of almost 750 per 100,000 individuals around the end of May, probably explained by the opening of the vaccination campaign to the high school students in their final year and also to all individuals regardless of their age, and then started to decline sharply until mid-July. Vertical bars in the upper panel of Figure 24.1 show the interest about the Green Pass (light gray bars), Super Green Pass (medium gray bars) and mandatory vaccinations (dark gray bars) proxied using Google searches for the terms described above. Until mid-July, Google searches are very low, proving that interest in Covid-19 certificates was very low in Italy. However, we can notice that, after the second week of July, interest in the term “Green Pass” increases suddenly. This variation is located in proximity to the first official press conference organized by the Italian government about the introduction of the basic version of the Covid-19 certificate. The same week, first-dose rates in the lower panel stopped decreasing, and started increasing again from the ensuing week. This increasing trend continued for the following three weeks, and then first doses declined again. The declining trend goes on till November, when one can notice an anomalous variation in Google trend searches about the Super Green Pass; also, in this case first-dose rates mildly increase, inverting the decreasing trend observed during the previous weeks. The increase in vaccinations seems to last for the next three weeks and then remains constant for the next couple of weeks, to finally increase again, at a faster pace, when interest in mandatory vaccinations increases during the last week of December. Considering Spain (Figure 24.2), we can see how in the period from April 1, 2021, until the first increase in searches for “Certificado Covid,” observed during the first week of June, first doses per 100,000 individuals move around an average value of 450–500. Then, they start to decline at a rate that increases after the last week of July, when some Spanish Superior Courts of Justice start abrogating the mandatory use of the Covid-19 certificate in Spain (as discussed in Section 2). Then, we observe an additional increase in the interest in “Certificado Covid” starting from mid-November – around this date some Spanish regions decided to re-introduce the mandatory use of the Covid-19 certificate – until the first week of December. First doses per 100,000 individuals increase as well, but with a three-week lag with respect to the interest in the term “Certificado Covid.” 3.4

Effect of Nudging on Vaccinations in Italy and Spain

Figure 24.3 shows estimates of the slope change in the rate of first doses per 100,000 individuals for Italy using the SRKD. The vertical dashed line represents the threshold to identify days after which new information is available. In the upper-left panel, the threshold is set corresponding to July 12, which represents the first day (Monday) of the week where we identified an anomalous variation in Google searches for the term “Green Pass.” Right before the threshold we can observe a decreasing trend in the rate of first doses per 100,000 individuals, whereas right after it we first observe a constant pattern that, a few days later, turns into an increasing dynamic in first doses. Focusing on the upper-right panel, we can observe what happens around the second threshold; i.e., the first day of the week where we identified an anomalous variation in terms of Google searches for the term “Super Green Pass.” Also in this case, we identify a variation in the trend of first doses per 100,000 individuals during the post-announcement period, that goes from 30 to 70 first doses per 100,000 individuals during the three weeks after the threshold. Lastly, we observe, again in Figure 24.3 (lower panel),

Political preferences and nudging for healthcare  419

Notes: This figure shows the amount of Google Trends searches for the following terms: (i) “Greenpass,” (ii) “Super Greenpass” and (iii) “Mandatory vaccinations” (upper panel) and the number of first doses per 100,000 individuals (lower panel) in Italy. Dashed lines correspond to weeks recording abnormal variations in the interest towards each term, solid lines identify pre- and post-treatment periods considering a six-week window around the date of the announcement.

Figure 24.1

Identification of pre- and post-announcement periods in Italy

a very similar pattern, with stable first doses before the threshold, which in this case identifies the introduction of mandatory vaccinations, and a boosting effect right after. In this case, first doses per 100,000 individuals pass from 60 to 140 after the government announcement. Figure 24.4 shows the estimates for Spain. The upper-left panel identifies what happens to first doses per 100,000 individuals around the first date; i.e., June 3, when the Spanish

420  Handbook on the political economy of health systems

Notes: This figure shows the amount of Google Trends searches for “Certificado Covid” (upper panel) and the number of first doses per 100,000 individuals (lower panel) in Spain. Dashed lines correspond to weeks recording abnormal variations in the interest towards each term, solid lines identify pre- and post-treatment periods considering a six-week window around the date of the announcement.

Figure 24.2

Identification of pre- and post-announcement periods in Spain

government introduced the mandatory use of the Covid-19 certificate. The rate of first doses increases sharply in the post-announcement period, first jumping from 450 to 600 per 100,000 individuals and, a few days later, settles on an average value of 500 per 100,000 individuals. Looking at the upper-right panel, we can observe how first doses per 100,000 individuals started to decline after the second relevant date – i.e., August 2 – and continued on the same trend over the following three weeks, exactly when some Spanish Superior Courts of Justice decided to abrogate the mandatory use of the Covid-19 certificate. Lastly, in the lower panel,

Political preferences and nudging for healthcare  421

Notes: The vertical dashed line illustrates the threshold needed to identify days after which new information is available, and it corresponds to July 12 in the upper-left panel, November 15 in the upper-right panel, and December 23 in the lower panel.

Figure 24.3

Green Pass effect on vaccination rates in Italy after key announcements

we observe an increase in first doses per 100,000 individuals after the re-introduction of the Covid-19 certificate in some autonomous communities. Here, we observe that first doses increase very quickly around mid-December, implying a possible lagged effect, explained by the fact that individuals may decide to delay as much as possible the first dose, expecting a possible intervention of the Spanish Superior Courts of Justice, which in this case did not abrogate the mandatory use of the Covid-19 certificate. Figure 24.5 provides a graphical overview of parameters estimated using the SRKD for different bandwidths (BW). Diamonds identify estimates adopting the optimal BW chosen according to the procedure proposed by Imbens and Kalyanaraman (2012), while triangles, circles and squares identify estimates obtained using BW of 7, 14 and 21 days, respectively. Looking at the upper panel, presenting results for Italy, we estimate a positive effect on first doses per 100,000 individuals considering the first announcement, which increases in size as we use larger BW. Considering the optimal BW (i.e., 4.7 days), we find that first doses increase by 3.57 per 100,000 individuals per day after the threshold. However, the coefficient is not statistically different from zero. Increasing the BW, we can also observe that the estimated effect increases and becomes statistically significant. The estimated change in slope is 13.27 and 23.31 when the BW is 14 or 21 days, respectively. Looking at the second announcement, we find that the effect of the announcement is positive for each BW, except for a BW of 21 days.

422  Handbook on the political economy of health systems

Notes: The vertical dashed line illustrates the threshold needed to identify days after which new information is available, and it corresponds to June 3 in the upper-left panel, August 2 in the upper-right panel, and November 26 in the lower panel.

Figure 24.4

Covid certificate effect on vaccination rates in Spain after key announcements

The effect is significant only when the BW is equal to 7 or 14 days, suggesting that this second announcement about the “Super Green Pass” does not seem to have been effective in boosting first doses. The last announcement about mandatory vaccinations for specific age groups is also always positive and statistically different from zero, but the size decreases for larger BW. The estimated effect ranges from 21.58 for the optimal BW to 4.01 for a BW of 21 days. The lower panel of Figure 24.5 presents estimates for Spain. Here most of the effects are statistically not significant. The only announcement that seems to have been able to produce a significant variation in terms of first doses per 100,000 individuals is the third one, on November 26, when some autonomous communities decided to re-introduce the Covid-19 certificate. The estimated effect ranges from 6.2 to 3.58 for the optimal BW and for the 14 days BW, respectively. When the BW is equal to 21 days, the effect is not statistically different from zero. To turn this evidence into numbers easier to interpret about the effect of nudging, we use information about total population, population already vaccinated and population infected at the day of each announcement to compute the share of additional individuals who got vaccinated following the announcements. These results are shown in columns (4) and (5) of Table 24.1.1 We provide a range of magnitudes, by considering the minimum and the maximum

Political preferences and nudging for healthcare  423

Notes: This figure plots a graphical overview of parameters computed adopting the SRKD for different BWs. Diamonds identify estimates computed choosing the optimal BW, whereas triangles, circles and squares identify estimates obtained using BW of 7, 14 and 21 days respectively.

Figure 24.5

Estimates of effect of Covid certificate announcements on vaccination trends in Italy and Spain

values for the estimated effects reported in column (3). Our computations show that the digital Covid certificate was able to increase vaccinations in Italy, among non-vaccinated individuals, in a range that spans from 2.01 percent to 6.93 percent. The two announcements that mostly increased such vaccinations are the first introduction of a Covid-19 digital certificate and the mandatory vaccinations for individuals over 50. The differential effect among announcements could be explained considering that the first announcement was probably able

424  Handbook on the political economy of health systems Table 24.1 Date

Magnitude of Covid-19 certificate effect on vaccination rates in Italy and Spain after key announcements Vaccinated population

τ min−max

Magnitude min−max

Magnitude min−max excluding infected

Italy 12 July 21

36,133,218

13.27–23.31

0.95–1.67

0.95–1.67

15 Nov. 21

46,930,060

0.76–1.69

0.14–0.32

0.14–0.32

23 Dec. 21

48,116,251

4.01–21.69

0.91–4.94

0.92–4.99

 

 

2.01–6.93

2.02–6.98 0–0

Total

Spain 3 June 21

19,038,135

0–0

0–0

2 Aug. 21

32,352,818

0–0

0–0

0–0

26 Nov. 21

38,288,327

0–6.2

0–2.26

0–2.26

 

 

0–2.26

0–2.26

Total

Notes:  The number of days amounts to 21. The number of positive unvaccinated individuals is computed by multiplying the number of current Covid-19 cases by the percentage of unvaccinated Covid-19 cases, resulting in 527,286 on July 12, 1,576,603 on November 15 and 5,608,975 on December 23 in Italy, and 0 across the three Spanish dates. The number of current Covid-19 cases are 40,426 on July 12, 120,875 on November 15 and 430,029 on December 23 in Italy, and 23,368 on June 3, 108,329 on August 2 and 38,729 on November 26 in Spain.

to target individuals who were moderately hesitant, and, being the first, it reached a larger number of individuals. Once moderately hesitant individuals are no longer present among the non-vaccinated, only a stronger intervention (i.e., mandatory vaccinations) was capable of affecting vaccination rates. For Spain, the situation is different, since only the last announcement seems to have been effective in increasing vaccinations. This result can be explained by the lower share of hesitant individuals and by the misalignment between the Spanish government and the Spanish Superior Courts of Justice during the first announcement, with the latter abrogating the mandatory use of the Covid-19 certificate. When the pandemic started to hit harder in terms of new cases and deaths, the government re-introduced the digital Covid certificate and was able to increase vaccinations effectively, without any interference from the Superior Courts of Justice. 3.5

The Effect of Ideological and Political Preferences

One can think that nudging is affected by the support for populist parties, which might influence the subjective evaluations of costs and benefits related to the choice of vaccination. To explore this issue, we consider our previous exercise using different sub-samples of regions and autonomous communities in Italy and Spain, defined according to the support for populist parties. Figures 24.6 and 24.7 show estimates of the SRKD when we split regions according to their support for populist parties. In particular, the upper-right panels of each figure show how first doses changed in Italy and Spain after the key announcements already described above for regions who recorded a share of votes for M5S or Podemos above the median during the latest available political elections (i.e., 2018 for Italy and 2019 for Spain). The lower-right panels of each figure show how first doses changed for regions located below the median in the support for populist parties. Each figure (on the upper- and lower-left panels) shows estimates for all regions to facilitate comparison. We focus on M5S and Podemos because they are identified

Political preferences and nudging for healthcare  425 as left-wing populist parties by Van Kessel (2015), Inglehart and Norris (2016) and Vittori (2017), and some of their elected members have adopted a strong rhetoric against vaccines. Looking at Italy (Figure 24.6), we do not notice significant differences between the variation in first doses for the entire population and for regions with higher (lower) support for M5S. Instead, we observe different behaviors in Spain. For instance, looking at Figure 24.7, it seems that regions with lower support for Podemos are those showing a stronger variation in terms of first doses after the first and the third announcements; i.e., the first introduction and the re-introduction of the standard Covid certificate in Spain. In addition, somewhat surprisingly, regions with higher support for Podemos are those where the verdicts of the Spanish Superior Courts of Justice abrogating the use of the standard Covid certificate affected fewer first doses, which increase significantly when we consider estimates with a bandwidth equal to 7 days. This is an unexpected result since regions which showed the highest consensus for Podemos are also those where the Spanish Superior Courts of Justice abrogated the mandatory use of the Covid certificate first.

Notes: This figure provides estimates of the SRKD dividing Italian regions on the basis of their level of support for M5S. The upper-right panel illustrates how first doses changed in Italy after the key announcements for regions who recorded a share of votes for M5S above the median during the latest political elections (i.e., 2018). The lower-right panel shows how first doses changed for regions located below the median. Each panel also shows estimates for the entire population for comparison against the upper- and lower-left panels.

Figure 24.6

Estimates of effect of Green Pass announcements on vaccination trend in Italy for the general population and by level of support for populist political parties in Italy

426  Handbook on the political economy of health systems

Notes: This figure provides estimates of the SRKD dividing Spanish autonomous communities on the basis of their level of support for Podemos. The upper-right panel illustrates how first doses changed in Spain after the key announcements for regions who recorded a share of votes for Podemos above the median during the latest political elections (i.e., 2019). The lower-right panel shows how first doses changed for regions located below the median. Each panel also shows estimates for the entire population for comparison against the upper- and lower-left panels.

Figure 24.7

4.

Estimates of the effect of Covid certificate announcements on vaccination trend in Spain for the general population and by level of support for populist political parties in Spain

DISCUSSION AND CONCLUSION

Our evidence supports the view that nudging can be an effective strategy to solve vaccine hesitancy, but it requires certain conditions. For Italy, we document that only the first and third announcements produced effects that are significant across different BW levels, meaning that, according to our estimates, they are the ones that most affected vaccinations. We can interpret the opposite direction in terms of magnitude of the estimated effects – increasing for the first and decreasing for the third announcement – as a measure of the speed with which each announcement was capable of affecting vaccinations. The first announcement was able to increase vaccinations slowly; i.e., the estimated effect increases with the BW, whereas the third one implied a faster increase in terms of vaccinations. This result can be explained considering the difference between the two measures adopted to improve vaccinations, the former associated with the use of the Covid-19 certificate to access some public places, the latter with the introduction of mandatory vaccination, prohibiting access to the workplace for individuals

Political preferences and nudging for healthcare  427 over 50 who are not vaccinated, implying severe consequences in terms of income in addition to a fine of 100 euros. On the contrary, for Spain, the introduction of the Covid-19 certificate seems to be less effective in increasing vaccinations, for two main reasons: first, the higher share of people already vaccinated at the date of the announcement, which means that nudging should be stronger to convince the marginal citizens to get vaccinated; second, the misalignment in terms of communication between different institutional actors, the government in Madrid on the one hand and the Courts of Justice in different autonomous communities on the other. In addition, mostly for the Spanish case, we do observe some interactions between the central government, the Courts of Justice and the prevalence of populist political parties at the local level in shaping both the policies (and the relative announcements) and the response from citizens. Besides adding to the literature exploring the impact of public policies on the Covid-19 pandemic, our research sheds light on the more general use of nudging in healthcare policies. Since the costs of nudging policies are relatively low, policymakers may seek to encourage the development of public policies aimed at offering the right incentive (or at creating the right hurdles) to “guide” people in choosing what is deemed to be the more beneficial option for their health. At the same time, specific nudge units, within central or local governments, should manage the proper implementation and surveillance of nudging strategies. However, as the case of the Covid-19 vaccination has shown, it would be of paramount importance for governments to work on people’s beliefs, making the evaluation of the costs and benefits of each specific choice more objective. This calls into question the role of government in making transparent information available to all citizens as a crucial means for the democratic process to work correctly. Official information would be also a way to combat fake news and misinformation, the so-called “infodemic,” which certainly played a role in influencing people’s beliefs regarding the health effects of vaccinations against Covid-19.

NOTE 1. In more detail: we calculate the ratio between the total number of additional first doses after each announcement and individuals not yet vaccinated. The numerator is obtained as the change in slope estimated after the announcement multiplied by 21 (i.e., the number of days over which the announcement is effective in increasing vaccination rates) and further multiplied by the total population in Italy or Spain divided by 100,000. The denominator is obtained as the population eligible for vaccination minus the population already vaccinated (column 4), and also further subtracting individuals who, on the day of the announcement, are infected and not vaccinated (column 5). Positive and not vaccinated individuals are calculated by multiplying currently infected individuals by the percentage of positive unvaccinated individuals. The percentage of positive unvaccinated individuals was retrieved by the Italian National Institute of Health (2022) and by the Spanish Ministry of Health (2022), for Italy and Spain respectively.

REFERENCES Angellotti, E., Wong, J.B., Pierce, A., Hescott, B., and Pittas, A.G. (2019), “Combining wireless technology and behavioral economics to engage patients (wibeep) with cardiometabolic disease: A pilot study”, Pilot and Feasibility Studies 5(1), 1–6. Ashraf, N., Karlan, D., and Yin, W. (2006), “Tying Odysseus to the mast: Evidence from a commitment savings product in the Philippines”, Quarterly Journal of Economics 121(2), 635–72.

428  Handbook on the political economy of health systems Behavioural Insights Team (2012), Applying Behavioural Insights to Reduce Fraud, Error and Debt, UK Cabinet Office. Beshears, J., Choi, J., Laibson, D., and Madrian, B.C. (2005), “Early decisions: A regulatory framework”, Swedish Economic Policy Review 12(2), 41–60. Dahl, M., Søndergaard, S.F., Diederichsen, A., Søndergaard, J., Thilsing, T., and Lindholt, J.S. (2021), “Involving people with Type 2 diabetes in facilitating participation in a cardiovascular screening programme”, Health Expectations 24(3), 880–91. Dai, H., Saccardo, S., Han, M.A., Roh, L., Raja, N., Vangala, S., Modi, H., Pandya, S., Sloyan, M., and Croymans, D.M. (2021), “Behavioural nudges increase Covid-19 vaccinations”, Nature 597(7876), 404–9. Dawkins, B., Renwick, C., Ensor, T., Shinkins, B., Jayne, D., and Meads, D. (2021), “What factors affect patients’ ability to access healthcare? An overview of systematic reviews”, Tropical Medicine & International Health 26(10), 1177–88. Duflo, E., Kremer, M., and Robinson, J. (2011), “Nudging farmers to use fertilizer: Theory and experimental evidence from Kenya”, American Economic Review 101(6), 2350–90. Falkenbach, M., and Willison, C. (2022), “Resources or trust: What matters more in the vaccination strategies of high-income liberal democracies?” Health Policy and Technology 11(2), 100618. Firmino-Machado, J., Varela, S., Mendes, R., Moreira, A., Lunet, N., Carmo, A., Cancela, A., Firmino, A., Ramos, A., Teixeira, A., et al. (2019), “A 3-step intervention to improve adherence to cervical cancer screening: The scan randomized controlled trial”, Preventive Medicine 123, 250–61. Hansen, P.G. (2016), “The definition of nudge and libertarian paternalism: Does the hand fit the glove?” European Journal of Risk Regulation 7(1), 155–74. Hofmann, B., and Stanak, M. (2018), “Nudging in screening: Literature review and ethical guidance”, Patient Education and Counseling 101(9), 1561–9. Imbens, G., and Kalyanaraman, K. (2012), “Optimal bandwidth choice for the regression discontinuity estimator”, Review of Economic Studies 79(3), 933–59. Inglehart, R.F., and Norris, P. (2016), “Trump, Brexit, and the rise of populism: Economic have-nots and cultural backlash”, HKS Working Paper No. RWP16-026. http://​dx​.doi​.org/​10​.2139/​ssrn​.2818659. Italian National Institute of Health (ISS) (2022), Report esteso iss. Covid-19: Sorveglianza, impatto delle infezioni ed efficacia vaccinale, Italian National Institute of Health (ISS). www​.epicentro​.iss​.it/​ coronavirus/​bollettino/​Bollettino​-sorveglianza​-integrata​-Covid​-19​_20​-aprile​-2022​.pdf. Karlan, D., McConnell, M., Mullainathan, S., and Zinman, J. (2016), “Getting to the top of mind: How reminders increase saving”, Management Science 62(12), 5–8. Milman, O., Yelin, I., Aharony, N., Katz, R., Herzel, E., Ben-Tov, A., Kuint, J., Gazit, S., Chodick, G., Patalon, T., et al. (2021), “Community-level evidence for SARS-CoV-2 vaccine protection of unvaccinated individuals”, Nature Medicine 27(8), 1367–9. Moellenkamp, M., Zeppernick, M., and Schreyogg, J. (2019), “The effectiveness of nudges in improving the self-management of patients with chronic diseases: A systematic literature review”, Health Policy 123(12), 1199–209. Moseley, A. (2020), “Nudging in public policy”, in Oxford Research Encyclopedia of Politics. https://​doi​ .org/​10​.1093/​acrefore/​9780190228637​.013​.949. Nguyen, T. (2008), “Information, role models and perceived returns to education: Experimental evidence from Madagascar”, unpublished manuscript. Reddy, A., Huseman, T.L., Canamucio, A., Marcus, S.C., Asch, D.A., Volpp, K., and Long, J.A. (2017), “Patient and partner feedback reports to improve statin medication adherence: A randomized control trial”, Journal of General Internal Medicine 32(3), 256–61. Reñosa, M.D.C., Landicho, J., Wachinger, J., Dalglish, S.L., Barnighausen, K., Barnighausen, T., and McMahon, S.A. (2021), “Nudging toward vaccination: A systematic review”, BMJ Global Health 6(9), e006237. Ritchie, H., Mathieu, E., Rodés-Guirao, L., Appel, C., Giattino, C., Ortiz-Ospina, E., Hasell, J., Macdonald, B., Beltekian, D., and Roser, M. (2020), “Share of people who completed the initial Covid-19 vaccination protocol”. Published online at OurWorldInData.org. https://​ourworldindata​.org/​ coronavirus.

Political preferences and nudging for healthcare  429 Rouyard, T., Leal, J., Baskerville, R., Velardo, C., Salvi, D., and Gray, A. (2018), “Nudging people with Type 2 diabetes towards better self-management through personalized risk communication: A pilot randomized controlled trial in primary care”, Endocrinology, Diabetes & Metabolism 1(3), e00022. Santalahti, M., Sumit, K., and Perkio, M. (2020), “Barriers to accessing health care services: A qualitative study of migrant construction workers in a southwestern Indian city”, BMC Health Services Research 20(1), 1–7. Sleat, D., Innes, K., and Parker, I. (2021), “Are vaccine passports and Covid passes a valid alternative to lockdown?”, British Medical Journal 375, 2571. Sotis, C., Allena, M., Reyes, R., and Romano, A. (2021), “Covid-19 vaccine passport and international traveling: The combined effect of two nudges on Americans’ support for the pass”, International Journal of Environmental Research and Public Health 18(16), 8800. Spanish Ministry of Health (2022), Actualizacion no 509. Enfermedad por el coronavirus (covid-19), Report 509, Spanish Ministry of Health. Spitale, G., Biller-Andorno, N., Germani, F., et al. (2022), “Concerns around opposition to the green pass in Italy: Social listening analysis by using a mixed methods approach”, Journal of Medical Internet Research 24(2), e34385. Van Kessel, S. (2015), Populist Parties in Europe: Agents of Discontent? Springer. Vittori, D. (2017), “Podemos and the Five-Star Movement: Populist, nationalist or what?” Contemporary Italian Politics 9(2), 142–61. WHO (World Health Organization) (2021), “Strategy to achieve global Covid-19 vaccination by mid-2022”. www​.who​.int/​publications/​m/​item/​strategy​-to​-achieve​-global​-covid​-19​-vaccination​-by​ -mid​-2022. Yoong, S.L., Hall, A., Stacey, F., Grady, A., Sutherland, R., Wyse, R., Anderson, A., Nathan, N., and Wolfenden, L. (2020), “Nudge strategies to improve healthcare providers’ implementation of evidence-based guidelines, policies and practices: A systematic review of trials included within Cochrane systematic reviews”, Implementation Science 15(1), 1–30.

25. Multilevel governance in the first wave of Covid-19 Marta Angelici, Paolo Berta, Joan Costa-Font and Gilberto Turati

1. INTRODUCTION How best to organize the territorial governance of the health system has been at the center of policy responses to new virus outbreaks in the past, such as the emergence of SARS in 2002, the MERS outbreak in 2012 and the Ebola outbreak in 2013. The balance of power between a highly centralized form of governance and more decentralized solutions has played a key role in these cases. Although the fight against Covid-19 is a global one that requires coordination at the highest possible level, crucial local knowledge about how best to address the challenges posed by a pandemic might not be used when decision-making is completely centralized (see Chapter 7 in this volume). This is of relevance in many healthcare systems in the European Union where health policy takes place at different levels of government. While coordination across borders is required at a Europe-wide level to face a pandemic, regional reactions might respond better to idiosyncratic needs; hence, a “one size fits all” approach might not always be the most efficient solution, especially when the impact of policies is highly uncertain, as in the presence of a completely new virus. Centralized governance entails a uniform response to counteract adverse effects of territorial self-interest (e.g., not circulating essential protective equipment). In contrast, advocates of decentralization put forward the role of innovation and low-cost experimentation when the optimal policy reaction is unknown, such as when a new virus is surrounded by uncertainty. Decentralized governance can still allow for some degree of coordination; for instance, via pandemic plans within and even between countries, and offers incentives to continue to provide information to central level authorities (Angelici et al., 2023). Cross-country and cross-regional coordination via pandemic plans allows for a swift exchange of information on the characteristics of the pathogen, alongside the setting-up of common standards to track its evolution and collect comparable data, and regulations to manage the actions of infected patients and prevent the spread of the disease further (including border closures and quarantines). While coordination across borders is required at a European-wide level to face a global pandemic, regional reactions are more flexible to respond to idiosyncratic needs; hence, a “one size fits all” approach might be a less efficient governance design when the impacts of policies are highly uncertain, as is the case in the presence of a completely new virus. This chapter discusses the effects of national (central) and sub-national (regional) reactions to Covid-19 in Italy and Spain. We add to the growing multidisciplinary literature (e.g., Bailey et al., 2020; Dodds et al., 2020; Casula and Pazos-Vidal, 2021). In the UK, the Coronavirus Act 2020 conferred new powers on devolved ministers. This allowed devolved administrations to make new regulations to tackle the pandemic. Although 430

Multilevel governance in the first wave of Covid-19  431 reactions were originally similar across the UK, Scotland and Wales implemented different policies with regard to restrictions on movement. The Scottish government even introduced additional legislation, the Coronavirus (Scotland) Act. However, coordination also took place, with the first ministers of the different countries participating in meetings of the Civil Contingencies Committee chaired by the Prime Minister. The main outstanding quarrels were around the implementation of a two-tier system (i.e., a system where some people would pay for care) and funding across regions. More specifically, we study the effect of multilevel governance of Covid-19 on several outcomes across Italy and Spain after the declaration of a state of emergency.

2.

ITALY AND SPAIN

Italy and Spain share common institutional backgrounds (e.g., decentralized healthcare systems), but differed in the governance of the first wave of Covid-19 pandemic (as discussed qualitatively by Casula and Pazos-Vidal, 2021). Both countries were hit hard by the pandemic, approximately around the same time: Spain was only a few weeks behind Italy in the spread of the virus. In May 2020, when the “first wave” was reaching an end and countries gradually re-opened their economies, reported cases in Italy (230,000) were comparable to those reported in Spain (240,000), and the same applies to deaths (33,000 and 29,000, in Italy and Spain respectively). However, despite sharing heavily decentralized health systems and hence important regional-level expertise, their central governments responded differently during the crisis. While the Spanish government centralized the purchase of healthcare equipment and imposed central-level coordination in all policy domains related to pandemic management, the Italian government did not enforce full coordination among the regional governments. In addition, even before a national lockdown was called in Italy, regional governments were allowed to differ in their policy priorities: Lombardy relied mostly on hospitals, while Veneto pushed on contact tracing. This was not possible during the first wave of the pandemic in Spain, given the centralization of power in the hand of the central government. Hence, the latter offers an opportunity to study the effects of this choice on relevant outcomes. Table 25.1 includes details of the policies and the interventions taken in the two countries. Given their different governance responses to the first wave of the Covid-19 outbreak, comparing evidence from Italy and Spain can be informative of the “balance of territorial power” allocation, and specifically, the welfare effects of healthcare (de-)centralization. Indeed, reliance on central-level coordination runs the risk of amplifying the effect of a policy when it succeeds, but especially when it fails. A uniform response across the entire national territory is still possible when effective cooperation takes place, as has been the case of the countries of the UK. In contrast, decentralized designs allow for experimentation in identifying a region-specific policy solution to face the spread of the virus. When the latter proves effective, then other regions can learn from such effects and adjust their own response. In Italy, despite bordering the Lombardy region Veneto experienced fewer than 20,000 cases, compared to about 80,000 in Lombardy during the first-wave emergency. In contrast, Madrid and Catalonia replied in the same manner, given the limited role for local policies allowed by the central government in Spain.

432  Handbook on the political economy of health systems Table 25.1

Policies adopted in Italy and Spain to manage the Covid-19 pandemic relative to different domains Italy

Spain

Health communication On 23 January 2020, the Task Force for the design and On 5 March 2020, the Ministry of Health (MOH) specified essential behavior concerning hand hygiene. Face masks were

development of a Covid-19 rapid response published the first

suggested only if there was a suspicion of somebody being sick

protocol establishing measures for early detection, surveillance

or if they were assisting somebody who was. The Department of and prevention. Since the initiation of a curfew (called the Civil Protection provided daily updates on the epidemiological

State of Alarm) on 14 March 2020, all ministries involved in

situation.

these operations (the ministries of Health, Economy, Internal

Specific information for each area was further detailed on

Affairs and Defence) were responsible for releasing the latest

regional government and local health authorities’ websites.

news on the implementation of the specific measures. The different regional governments, usually through the heads of their departments of health, debriefed local media about the local progression of the outbreak.

Physical distancing Italy was the first country in Europe to adopt restrictive physical The first community outbreak was declared on 3 March and social distancing measures, defining “red zones.” Within a

2020, prompting progressive physical distancing measures.

“red zone,” people were told to stay at home and not leave the

The Interterritorial Council of the National Health System

area, and all non-essential economic activities were temporarily

(a body composed of the Ministry of Health, or MOH, and

shut down. The first red zone was imposed to isolate two

representatives of regions, known as autonomous communities)

outbreaks in small villages in Lombardy and Veneto on 23

agreed on 9 March 2020 on new measures to slow down the

February 2020. On 8 March, the red zone regime was extended

spread of the new Coronavirus. The State of Alarm was declared

to all of Lombardy and some neighboring provinces. The entire

on 14 March. People over 6 years old had to wear face masks on

country was locked down on 11 March, about one and a half

transport. Similarly, masks had to be worn outdoors or in closed

months after the declaration of a “state of emergency for sanitary public spaces where a minimum distance of 1.5 meters could not reasons” on 31 January 2020.

be assured.

Isolation and quarantine Surveillance and monitoring mechanisms to detect cases coming Isolation was imposed upon those who had tested positive to Covid-19 but did not require hospitalization, and on those who

from China were activated on 23 January 2020. All suspicious

had had contact with somebody who had tested positive, to

cases were isolated upon arrival to Spain and potential contacts

inhibit the transmission of SARS-CoV-2 and avoid overloading

investigated.

the hospital system.

Starting on 25 February 2020, people with mild symptoms

Considering the different timing of the spread of the Covid-19

who would had been visiting high-risk areas or in contact with

epidemic across the Italian regions, regional presidents issued

infected people from those areas (China, South Korea, Singapore,

special decrees imposing quarantine on people coming from

Japan, Iran and Italy) were confined for two weeks, and potential

regions or provinces considered at higher risk.

contacts further investigated.

A crucial issue is that – at the beginning of the first wave,

A few days later, from 28 February, cases (possible or confirmed

in February 2020 – the testing policy was limited to people

Covid-19) with mild symptoms were requested to stay at home

returning from China and developing flu-like symptoms,

in self-isolation and were followed up by a home care team, via

according to national regulations (see below). The identification

telephone but with home visits or immediate hospitalization if

of the first patient in Lombardy on 20 February 2020 was the

needed.

result of a test outside national rules.

Multilevel governance in the first wave of Covid-19  433 Italy

Spain

Monitoring and surveillance Case definition has evolved in response to the evolution of the Contact tracing has been recognized as a sustainable public health tool for the prevention and containment of Covid-19.

outbreak but has always followed protocols proposed by the

However, the government was unable to use this strategy,

World Health Organization (WHO), which were updated by the

especially during the first wave.

European Centre for Disease Prevention and Control (ECDC) on

The central government also tried to develop contact-tracing

24 February 2020.

software (an app called Immuni). But, waiting for this app,

A nationwide seroprevalence study launched with the aim of

Italian regions autonomously developed several different digital

estimating the population’s level of exposure to the virus, and

solutions for the control and containment of infected citizens

whether the population had developed protective antibody

which, in most cases, were based on analyses of movements

titers. Regional health authorities took responsibility for the

and gatherings on the basis of anonymous data. These digital

deployment of the study. Primary care professionals, specifically

solutions were never really taken seriously in any of the regions. nursing staff, were in charge of the epidemiological survey and Lombardy launched a remote training course in order to train

sample extractions (carried out either at home or in primary care

interns and healthcare professionals specifically in contact

centers).

tracing, to monitor the infected, keep contacts under surveillance and solicit testing in case of risk exposure. However, this solution was made ineffective by the mounting cases during the first wave. Planning services From the second week of March 2020, as a response to the

On 1 March 2020, the MOH issued a regulation requiring all

regions to take action according to following recommendations/

increasing toll of cases, elective surgery and non-urgent

rules:

consultations were postponed. In turn, primary care centers also

● Increase by 50 percent the number of intensive care beds.

called off non-urgent consultations, canceled emergency care

● Increase by 100 percent the number of beds in pneumology and except for patients with respiratory symptoms, and implemented infectious diseases wards (these beds had to be equipped with

an e-prescription mechanism for chronic patients so they could

adequate assisted pulmonary ventilation systems).

get their prescriptions renewed automatically, and thus avoid

● Mainly use private contractors (private hospitals accredited with the NHS) for non-Covid-19 patients; however, in Lazio

a visit to primary care premises. From 14 March, after the declaration of the State of Alarm, the

and Lombardy, private contracted hospitals also increased their MOH was temporarily entitled to determine the best distribution capacity for Covid-19 patients. ● Re-allocate health professionals according to internal re-organizations and provide a short training program if required.

of technical resources, including those from the military forces and private health sector, and even use private business as hotels.

434  Handbook on the political economy of health systems Italy

Spain

Managing cases The first contact point measures changed with the evolution of This is a crucial domain to understand the differences in the outcomes across regions. According to national guidelines,

the outbreak and are somewhat different across Spain’s regions

patients having symptoms but not in a critical situation were

(autonomous communities). At the very beginning of the crisis, the

asked to stay at home and contact their general practitioner

usual pathways of care – that is, primary care centers and hospital

(GP). The GP was asked to make the diagnosis either

emergency departments – were the reception point for cases.

physically or virtually and, depending on the symptoms, he or

Shortly afterwards, mild cases were advised to stay at home and

she could request a nasopharyngeal swab that was generally

call the regular 24/7 emergency call centers, from which they could

taken at the patient’s home. Only patients who suffered from

obtain advice and/or the activation of a mobile unit to their home.

severe failure of the respiratory system were told to go to the

Most autonomous communities created a dedicated phone helpline

hospital; likely too late to be saved. The decentralization of

different from the regular 24/7 call center number, keeping the

the Italian healthcare system and the quasi-market models

latter for emergencies not related to Covid-19. Some autonomous

developed by regions has led to different behaviors among

communities also developed an online self-administered survey to

regions with regard to the choice of care settings.

help citizens figure out how likely it was for them to be infected.

Lombardy primarily managed cases by resorting to inpatient care (49 percent of positive cases were hospitalized as of 8 April 2020). Veneto, with a very low hospitalization rate of 21 percent (as of 8 April), resorted mainly to outpatient care. Hospitalization was almost entirely restricted to those requiring intensive care. Maintaining essential services On 17 May 2020, when the first wave was ending, the MOH

During the first wave, most regions defined their own plan for

the reorganization of hospital care. In fact, in almost all Italian

published guidance on how to program elective surgery

regions, intensive care unit (ICU) bed capacity was increased by during the transition period. So, elective surgery could start if over 50 percent, exceeding the amount suggested by the MOH

admitted Covid-19 patients were less than 5 percent of the total

and stopping elective care almost entirely. By 26 May 2020, the

hospital capacity and hospitals had separate areas for Covid

organization of hospital care was showing very heterogeneous

and non-Covid patients. It was recommended that pre-surgery

behaviors among regions, and none of them had issued specific

consultations were telematic, admissions should be done the

resolutions or acts relating to the reorganization of hospital care

same day of the intervention and early discharge assessment was

after the emergency. Only nine regions defined Covid-19 hospital encouraged. In addition, patients should have been interviewed networks, albeit with different levels of detail, and by 11 June

on clinical symptoms compatible with Covid-19 and tested with

the number rose to 10 out of 20. Lombardy, Liguria, Veneto and a PCR genetic test in the 72 hours prior to the surgery. They had Tuscany preferred not to set up Covid hospitals.

to wear face masks during their stay, as did their companions, and visits were severely restricted.

Multilevel governance in the first wave of Covid-19  435 Italy The first two confirmed cases of Covid-19 were represented

Spain Testing Laboratory testing for the diagnosis of the SARS-CoV-2 was

by two Chinese tourists traveling to Rome on 31 January 2020,

mandatory in two situations:

and the first cases among Italian residents were registered on 21

● a patient presenting clinical signs of acute respiratory infec-

February 2020 in Lombardy and Veneto. Since the beginning of

tion who was hospitalized or who met criteria for hospital

the mitigation phase (26 February), Italian national guidelines on testing followed WHO and ECDC guidelines stating that patients with symptoms could be tested if (i) they had contacts with a confirmed case, (ii) they came from areas where local

admission; ● a patient presenting clinical signs of acute respiratory infections of any severity who belonged to the health and social care workforce or to any other essential service.

transmission was ongoing or (iii) they were hospitalized.

Tests could only be requested by a medical doctor, and were

However, it was challenging to ensure full consistency with

approved once public health authorities checked whether the

this policy throughout the Italian territory. This was due to

patient fulfilled the testing criteria described.

the fact that the regions, which are responsible for health services’ delivery, organized the tracing and testing in different manners. During the first days of March 2020, as recommended by the Scientific Committee advising the national government, Lombardy tested only symptomatic patients. Going beyond national guidelines aligned with WHO and ECDC recommendations, Veneto adopted a different approach to testing, aiming towards mass population screening. The regional government approved on 16 March a plan for large-scale population testing.

436  Handbook on the political economy of health systems Italy

Spain

Governance Notwithstanding the availability, on paper, of an (old) National From 7 January 2020, when Covid-19 was identified as the Pandemic Plan, the national government decided not to follow

pathogen that caused the outbreak in Wuhan, China, the MOH,

its prescriptions and to use the Plan as a coordination device

through the Centre for the Coordination of Health Alerts and

during the first months of the Covid-19 crisis. On 5 February

Emergencies (CCAES, in Spanish), activated the Covid-19

2020, after the declaration of the state of emergency attributing

protocol in coordination with the Department of Health in

special powers to the Department of Civil Protection and the

each of Spain’s 17 autonomous communities. The royal decree

national government, a Scientific and Technical Committee was

declaring the State of Alarm on 14 March 2020 conferred full

instituted. Its purpose was (and still is) to provide guidance based responsibility to the Spanish government to implement measures on scientific evidence to the national government.

to deal with the Covid-19 crisis. The Prime Minister delegated

Notwithstanding the increased role of the central government,

competences to the ministers of Defence, Internal Affairs,

regions still retain decision-making autonomy regarding the

Transport, Mobility and Urban Matters, as well as to the MOH,

delivery and organization of health services, such as whether

in their respective areas of responsibility, led by the latter.

to conduct Covid-19 tests on the entire regional population.

In order to enhance coordination in the collection of

Different approaches were taken by the Lombardy and Veneto

epidemiological information, each regional Health Authority had

regions. Aside from being the two regions most affected by

to report a core set of indicators to the CCAES, from March 15,

the crisis, these two neighboring regions also share a similar

which included: epidemiological indicators (e.g., new confirmed

socioeconomic profile. They both imposed social distancing

cases, differentiating those diagnosed with PCR tests and those

measures and retail closures.

with rapid tests, and those with and without symptoms; cured

Veneto applied a more proactive strategy aimed at containment.

cases; deaths), utilization indicators (number of admitted and

It performed tests on asymptomatic cases at a very early stage

discharged patients, differentiating by the type of care provider),

and then traced potential positives. Once someone was found to

and supply indicators (number and occupation of beds in ICUs

be infected, their families and neighbors were tested in turn or, if and resuscitation units, and workforce available for service, testing kits were not available, they were quarantined. Home care especially ICU doctors and anesthesiology and reanimation and diagnoses were also strongly emphasized (for example, when professionals, including fourth- and fifth-year residency possible, samples were collected at home) and healthcare and

physicians, as well as any other healthcare professionals that

essential workers were specifically monitored.

could be summoned if required, including retired professionals

Lombardy, proportionally to its population, conducted half the

and physicians and nurses in their first years of training).

number of tests performed in the Veneto region and focused only on symptomatic cases. The regional government relied less on tracing, home care and monitoring healthcare workers. Another example were the regional measures on face masks adopted in Lombardy. National guidelines followed WHO recommendations and imposed their utilization only on health professionals and Covid-19 positive patients. However, Lombardy’s president issued an ordinance that made wearing protection over the mouth and face mandatory, starting from 5 April 2020.

Note:  Authors’ elaboration starting from information published by the WHO Covid-19 Health System Response Monitor.

Policy Reactions to Covid-19 in Italy The first Covid-19 case in Italy was officially identified on 20 February 2020, at a public hospital in Codogno, a small town close to Milan in Lombardy, thanks to the intuition of an anesthesiologist, who tested a 38-year-old patient against the national advice for Covid-19 testing. In fact, the Italian Prime Minister had declared a national emergency via an “emergency decree” on 31 January 2020, for a period of six months; but before this first case was detected, the people to be checked and tested were only those returning directly from China.

Multilevel governance in the first wave of Covid-19  437 A similar town was Vo’ Euganeo, an even smaller jurisdiction in the surroundings of Padua in Veneto, where an outbreak was discovered early on. Yet, both Codogno and Vo’ Euganeo were locked down into a red zone by the central government after 23 February 2020, which entailed temporary closures of all economic activities but for essential services, and stay-at-home orders for all the people residing in the area. On 8 March 2020, the entire Lombardy region, as well as a few provinces in the bordering regions of Veneto, Piedmont and Emilia Romagna, were locked into red zones too. Finally, the whole country was locked down in a national red zone a few days later, starting from 11 March 2020. After months of lockdown, a de-escalation of measures was started at the beginning of May, marking the end of the first wave of the Covid-19 pandemic in Italy. Despite a national lockdown being enacted, the evolution of the epidemic in Italy was regionally heterogeneous. Northern Italy was more exposed to Covid-19 infection compared to both the Center and South, where the spread of the new Coronavirus did not follow a similar growth. In Northern Italy, Lombardy was by far the most affected region, and one of the most affected in the world during the first wave. Conversely, in Veneto the evolution of contagion had been mitigated more. These are the two Italian regions that we consider as case studies below. The Italian National Healthcare System (NHS) has provided universal healthcare coverage since 1978, and it is financed with taxes, mostly collected at the central level. During the 1990s, several policy reforms transferred administrative and organizational responsibilities from the central government to the regional administrations, so that Italian regions have significant autonomy in organizing their own healthcare systems (Turati, 2013). This autonomy was also enjoyed during the pandemic despite the declaration of the national emergency, and it helps to explain the different policy patterns followed by Lombardy and Veneto. Among the regions, Lombardy has a population of 10 million residents, and it ranks among the most competitive areas in Europe. Public expenditure for healthcare services reached 20 billion euros in 2019, the year before the Covid pandemic. The healthcare system comprises approximately 150 hospitals generating 1.5 million discharges annually. A regional reform in 1997 radically transformed the healthcare system in Lombardy into a quasi-market in which citizens are free to choose the provider, regardless of its ownership (private or public). Unlike in other Italian regions, the healthcare system in Lombardy is entirely built on a clear separation between insurers (the Local Health Authorities, or LHA) and providers, on resources being allocated using a prospective payment system based on diagnosis-related groups (DRGs), and on the reimbursement for all the (public and private) providers within the regional accreditation system (Brenna, 2011). The unfolding of the Covid-19 pandemic led to a rise in hospitalization, allowing the virus to spread into hospitals, which forced authorities in the provinces of Bergamo and Brescia to convert entire hospitals into Covid-19 wards, increasing bed capacity in ICUs and moving physicians and nurses from their usual activity to care for patients affected by the Coronavirus. This policy of increasing ICU bed capacity was later adopted across the country. In contrast to Lombardy, Veneto presented a more centralized healthcare model, with the regional government capable of better coordinating with a top-down approach the choices of hospitals regarding the use of intensive-care beds or procurement of personal protective equipment (PPE). As for the pandemic, this model appeared more ready to deal with the epidemic outside the hospitals. Veneto addressed the Covid-19 epidemic with extensive testing of symptomatic and asymptomatic citizens, broad contact tracing around positive cases, quarantine

438  Handbook on the political economy of health systems for cases and suspected cases with daily telephone monitoring, detailed practical guidelines on home isolation, minimization of contacts with physicians and nurses, and limited hospital admissions for patients with major healthcare needs (Binkin et al., 2020). Policy Reactions to Covid-19 in Spain During the first wave, Spain had one of the highest numbers of Covid-19 cases in the world, after the United States. The first positive case was detected on 31 January 2020, but it was only from March that the diagnoses began to increase exponentially. As of 25 February 2020, cases in Spain had skyrocketed because people with pneumonia of unknown origin were tested for Covid-19. On the same day, four new cases related to the Italian cluster were confirmed in Spain. By 13 March 2020, cases had been confirmed in all 50 provinces of the country. A state of alarm and national lockdown was imposed on 14 March, and the central government was allocated full responsibility to coordinate and implement interventions to deal with the Covid-19 crisis. On 29 March 2020 it was announced that, beginning the following day, all non-essential workers were to stay home for the next 14 days. On 28 April, the government announced a plan for easing lockdown restrictions, but people were allowed out of their homes for short walks and individual sports only from 2 May, marking the end of the first wave in Spain. Unlike in Italy, Spain adopted a “command and control” approach in stark contrast to its healthcare decentralization in normal times. Indeed, the health system in Spain is comparable to the Italian one in all of its relevant design features: it is organized along the lines of a National Health System, and the governance of the system is decentralized at the regional level. Seventeen regions (called autonomous communities) have healthcare responsibilities with regard to providers’ organization and funding, and the system is funded by unadjusted block grants and, to a lesser extent, regions’ devolved and own taxes. So far, evidence documents that decentralized governance plays a role in lowering regional inequalities in healthcare use and in stimulating innovation (Costa-Font and Turati, 2018). However, strikingly, a newly appointed Minister of Health coordinated the command of the health system amidst the State of Alarm, which was declared on 14 March 2020. The decree centralized the purchase of medical equipment, and the suspension of flights from Italy. At the time of the first wave, healthcare policies were already highly heterogeneous across regions, since regional governments were run by different political coalitions. At the time of the first outbreak, the region of Madrid was run by a conservative coalition government which had engaged in a plan of significant healthcare privatization, and during the pandemic pushed ahead in outsourcing healthcare services to private for-profit providers. In contrast, Catalonia was run by a regional coalition while the central government was supported by a left-wing coalition with different regional supporters. Madrid was the focal point of the pandemic in Spain, followed by Catalonia, and we consider these in the two Spanish case studies below. Yet, although exposure to the pandemic differed by region considerably (e.g., besides Madrid and Catalonia, other heavily affected regions were the two Castiles, Basque Country, Navarra and Andalusia), as the speed of the pandemic differed by regions, the state of emergency and central-level coordination were imposed. In contrast, in the second and third wave, regional governments kept their own responsibilities. This provides some levels of policy variation when examining the effects of decentralization on relevant health outcomes.

Multilevel governance in the first wave of Covid-19  439

3.

EVIDENCE AT THE NATIONAL LEVEL

This section and the next draw on the empirical analysis in Angelici et al. (2023). Figure 25.1 displays the cross-country comparisons in terms of four Covid-19-related measures: number of infected cases, hospitalizations, admissions in ICUs and deaths. Data consider aggregate figures at the national level and all the measures are standardized rates in terms of population. To obtain a comparable scale for all plots, the number of cases and hospitalization rates are multiplied by 100, and admissions in ICU and deaths by 1,000. The values on the x-axis refer to days t0–t75, as defined by pairing the two time series.1 All figures reveal a consistent path: despite Spain having a population of about 47 million people, compared to about 60 million people in Italy, Spain recorded a higher number of confirmed cases, hospitalized patients, patients admitted in ICUs and deaths. More strikingly, while hospitalizations and admissions to ICU tail off after 30 days in Italy, they continue growing in Spain. This descriptive evidence points towards the better performance of a governance model allowing for regional differentiation of policies.

Figure 25.1

Evolution of Covid-19 first wave in Italy and Spain

One potential explanation of the differences between Italy and Spain lies in the stringency of measures implemented in Italy. Let us consider the Stringency Index produced by the Blavatnik School of Government (Wade et al., 2020). The index details the lockdown policies adopted by the countries, summarizing several pieces of information about containment efforts, including the following measures: school and workplace closures, canceling public events, limits on private gatherings, closing of public transport and restrictions on internal movement between

440  Handbook on the political economy of health systems cities/regions. The index is computed at the national level, and it goes from 0 to 100: a higher value of the Stringency Index suggests that the overall government response has become stronger. The comparison between Italy and Spain (see Table 25.2) in terms of the Stringency Index suggests that – although in the early days of the pandemic the two countries differed in the stringency of measures implemented to fight the pandemic – both countries ended up exhibiting similar values of the index. In the following analysis, we consider the overall index provided by the Blavatnik School of Government, instead of single specific policy domains included in the index. In fact, most of the measures (relative to school closures, international travel controls or canceling public events) were implemented early on in both countries. However, restrictions regarding workplaces or public transportation were applied later in Spain as compared to Italy. Hence, the few days of delay before central government in Madrid adopted harsh measures as compared to Italy might explain part of the difference in outcomes between the two countries (for more on this, see, for example, Montesò-Curto et al., 2020). The slight delay in response by the Spanish government with respect to the actions taken in Italy can also be gauged by looking at excess mortality in 2020 compared to mortality estimates in 2019. Information about overall mortality in Spain is gathered from the Spanish Mortality Monitor (MoMo, available at www​.isciii​.es). Spanish data are collected daily and include all-causes mortality obtained from the General Register of Civil Registers and Notaries of the Ministry of Justice, distributed among all the autonomous communities and including the 52 provincial capitals. For 2020, MoMo in Spain includes deaths from all causes from 3,929 computerized civil registries, representing 92 percent of the Spanish population. Daily data are available from 5 April 2018 up to 22 April 2020. The Italian Institute of Statistics (ISTAT) provides data about overall mortality in Italy. ISTAT focused on the municipalities with reliable data that show at least 10 deaths in the period 1 January to 31 May 2020 and that recorded a 20 percent increase in mortality in the period 1 March to 4 April 2020 compared to the average mortality for the same period in the years 2015–2019. ISTAT made available the data of 7,357 municipalities (out of a total of 7,904, or 93.1 percent), for which consolidation was possible until 31 May 2020, and covering 95 percent of the population resident in Italy. In Figure 25.2, the comparison between Spain and Italy is performed analyzing the first four months (January–April) of 2019 and 2020, and the mortality rate is computed by considering population in the two countries. It is clear that excess mortality is higher in Italy than in Spain. However, it is also evident that excess mortality in Spain was positive sharply after t0, while t0 in Italy is about 10 days before excess mortality becomes positive. Once again, this supports the view that the Spanish government was some days late in adopting the same measures as the Italian government. In addition, notice that mortality is also higher in Italy in 2019 with respect to Spain, suggesting that differences in the age structure of the population might affect the level of mortality, and the outcome of Covid-19. For instance, Islam et al. (2021) show that – accounting for the different age structures of European countries – excess mortality in Italy in 2020 is actually lower than that recorded in other European countries, including Spain, Belgium and the UK. A further and connected explanation of the differences observed in the number of cases and the number of excessive deaths, calling into question the role of governance, is that the pandemic was strongly concentrated in very few regions in Italy because of the adoption of severe measures early from the start of the crisis, while in Spain the region of Madrid remained open and contributed to the spread of the pandemic to other regions. To better understand the concentration and the evolution of the pandemic, we compute the Gini index on the number

Multilevel governance in the first wave of Covid-19  441 Table 25.2

Italy and Spain compared

National-level dataset

Min

Mean

  Cases

StdDev

Max

Spain 1,100.00

99,935.06

73,008.05

204,027.00

Hospitalizations

150.00

53,565.89

40,104.35

108,932.00

ICU Admissions

10.00

5,582.47

3,658.87

10,187.00

0.00

8,976.21

7,866.11

22,156.00

11.11

76.48

15.86

85.19

Deaths Stringency Index  

Italy

Cases

322.00

78,546.00

65,588.20

187,327.00

Hospitalizations

150.00

19,548.78

12,592.69

33,004.00

ICU Admissions

35.00

2,332.00

1,417.18

4,068.00

Deaths

10.00

9,418.88

8,817.17

25,085.00

Stringency Index Pooled regional-level dataset

68.52

85.64

9.50

93.52

Cases

Catalonia 49.00

31360.71

20187.03

55921.00

Hospitalizations

0.00

16896.64

10955.54

29497.00

ICU Admissions

0.00

1800.92

1096.02

2969.00

Deaths

0.00

3010.33

2215.99

6021.00

11.11

76.48

15.86

85.19

Stringency Index   Cases

Madrid 738.00

45027.55

22598.21

67049.00

Hospitalizations

0.00

28253.52

14978.29

42497.00

ICU Admissions

0.00

2615.75

1121.17

3617.00

Deaths

0.00

5388.11

3257.72

8931.00

11.11

76.48

15.86

85.19

Stringency Index  

Lombardy

Cases

240.00

41583.95

28346.66

81225.00

Hospitalizations

104.00

8258.83

4357.17

13328.00

ICU Admissions

25.00

798.07

411.85

1381.00

9.00

7182.80

5487.85

14924.00

68.52

85.64

9.50

93.52

Deaths Stringency Index  

Veneto

Cases

43.00

9445.96

6918.81

18671.00

Hospitalizations

19.00

1114.23

667.68

2068.00

ICU Admissions

7.00

169.99

111.59

356.00

Deaths

1.00

630.03

569.47

1643.00

68.52

85.64

9.50

93.52

Stringency Index

of deaths in each region and each week from t0 to t75. Results (not reported here for brevity) confirm a higher concentration of the Covid-19 pandemic in Italy than in Spain, which implies that a centralized governance allows for more homogeneous outcomes across regions, while decentralized governance enables the separation and identification of best practices from those regions that have adopted unsuccessful choices. The concentration index also shows a decreasing trend for Spain, suggesting even more homogeneous outcomes as the pandemic also makes progress in regions that were not hit by the virus at the beginning.

442  Handbook on the political economy of health systems

Figure 25.2

4.

Excess mortality 2019–2020 (January–April)

REGIONAL-LEVEL EVIDENCE

To better understand the role of regional patterns, we examine the regional trends of Covid-19 outcomes, selecting as case studies two of the most affected regions in the two countries under analysis, namely Lombardy and Veneto in Italy, and Catalonia and Madrid in Spain. For Italy, the importance of focusing on Veneto and Lombardy is well described by Binkin et al. (2020) in terms of the different approach to the Covid-19 epidemic in the two Italian regions. The authors showed that the community-based approach adopted in Veneto seems to be correlated with a limited rate of cases, hospitalizations and deaths, whereas the approach based on a strong hospitalization of positive cases adopted in Lombardy overwhelmed the healthcare system, with major consequences for the whole regional population. Similar arguments are also discussed by Costa-Font et al. (2020), who focus their attention on the different model of managed competition adopted by the two regions, with the one adopted by Lombardy more decentralized than the one adopted by Veneto. As for Spain, the importance of focusing on Madrid and Catalonia is supported by, for example, Legido-Quigley et al. (2020). The Madrid region was the epicenter of the crisis in Spain. Catalonia requested a complete shutdown of the region together with a full range of social distancing measures, but the royal decree declaring a national emergency contained controversial new measures attributing to the central government more and new powers over health services. The panels in Figure 25.3 are defined following t0–t75 at the national level. They compare the four regions in the two countries, standardizing all measures by the population in each region. The panel representing confirmed cases shows evidence that the two regions that were the focus of the pandemic in both countries (Lombardy and Madrid) reveal increasing trends in terms of confirmed cases,

Multilevel governance in the first wave of Covid-19  443 but Catalonia in Spain follows Madrid closely, while Veneto in Italy presents a very different pattern with respect to Lombardy.

Figure 25.3

Evolution of Covid-19 first wave in four regions in Italy and Spain

The panels regarding the number of hospitalizations and patients admitted in ICUs describe the trends in the two variables in each of the four regions. The two Spanish regions clearly stand above Lombardy and Veneto following very similar patterns, while Lombardy performs differently from Veneto. As for mortality, Lombardy exhibits much higher numbers than all the other regions; the trend in Madrid is very similar to that in Catalonia, while Veneto clearly follows a very different pattern with respect to Lombardy. This is consistent with the differential role of regional autonomy in Veneto and Lombardy, compared to a much more centralized management of the crisis in Spain.

5. CONCLUSION Decentralized government offers an advantage when responding to a pandemic. More specifically, we document a significant gap in the trends in cases, hospital and ICU admissions, and mortality in Italy and Spain, at both the national and the regional level. Our analysis indicates that the strong localization of the pandemic in Italy and the existence of regional autonomy explain the differences between the two countries.

444  Handbook on the political economy of health systems Coordination does play a role in solving potential collective action problems such as border closures. However, our evidence indicates that regional autonomy, via experimentation and local knowledge, can make a difference in the number of lives saved, as well as in avoiding unnecessary hospitalizations. The experience of the UK also suggests that regional autonomy is not at odds with coordination. A system that encourages regional cooperation, but that relies on regional autonomy, might therefore be beneficial for meeting the challenges posed by pandemics. This approach allows for the emergence of good practices compared to more centralized approaches, especially when regional needs and knowledge are largely heterogeneous. This lesson was ultimately learned in Spain, as in the later waves of the pandemic the country did not pursue such a centralized response, which helped improve health and social care outcomes.

NOTE 1. In more details, we consider t0 as the day when the two countries exhibited the same number of hospitalized patients, namely March 7th in Spain, and February 25th in Italy. See Angelici et al. (2023) for further details.

REFERENCES Angelici, M., Berta, P., Costa-Font, J., Turati, G. (2023). Divided we survive? Multilevel governance during the COVID-19 pandemic in Italy and Spain, Publius: The Journal of Federalism, 53(2), Spring, 227–50. Bailey, D., Clark, J., Colombelli, A., Corradini, C., De Propris, L., Derudder, B., Fratesi, U., Fritsch, M., Harrison, J., Hatfield, M., Kemeny, T., Kogler, D.F., Lagendijk, A., Lawton, P., Ortega-Argilés, R., Otero, C.I., and Usai, S. (2020). Regions in a time of pandemic, Regional Studies, 54(9), 1163–74. Binkin, N., Salmaso, S., Michieletto, F., and Russo, F. (2020). Protecting our health care workers while protecting our communities during the Covid-19 pandemic: A comparison of approaches and early outcomes in two Italian regions, medRxiv. https://​doi​.org/​10​.1101/​2020​.04​.10​.20060707. Brenna, E. (2011). Quasi-market and cost-containment in Beveridge systems: The Lombardy model of Italy. Health Policy, 103(2011): 209–18. Casula, M., and Pazos-Vidal, S. (2021). Assessing the multi-level government response to the Covid-19 crisis: Italy and Spain compared. International Journal of Public Administration, 44(11–12), 994–1005. Costa-Font, J., Levaggi, R., and Turati, G. (2020). Resilient managed competition during pandemics: Lessons from the Italian experience during COVID-19. Health Economics, Policy and Law, 17(2), 212–19. Costa-Font, J., and Turati, G. (2018). Regional healthcare decentralization in unitary states: Equal spending, equal satisfaction? Regional Studies, 52(7), 974–85. Dodds, K., Castan Broto, V., Detterbeck, K., Jones, M., Mamadouh, V., Ramutsindela, M., Varsanyi, M., Wachsmuth, D., and Yuan Woon, C. (2020). The Covid-19 pandemic: Territorial, political and governance dimensions of the crisis. Territory, Politics, Governance, 8(3), 289–98. Islam, N. Shkolnikov, V.M., Acosta, R.J., Klimkin, I., Kawachi, I., Irizarry, R.A., Alicandro, G., Khunti, K., Yates, T., and Jdanov, D.A. (2021). Excess deaths associated with Covid-19 pandemic in 2020: Age and sex disaggregated time series analysis in 29 high income countries. British Medical Journal, 373: n1137. Legido-Quigley, H., Mateos-García, J.T., Regulez Campos, V., Gea-Sánchez, M., Muntaner, C., and McKee, M. (2020). The resilience of the Spanish health system against the Covid-19 pandemic. The Lancet, 5(5), e251–e252.

Multilevel governance in the first wave of Covid-19  445 Montesò-Curto, P., Sanchez-Montesò, L., Maramao, F.S., and Toussaint, L. (2020). Coping with the Covid-19 pandemic in Italy and Spain: Lessons in response urgency. Journal of Global Health, 10(2), 020326. Turati, G. (2013). The Italian Servizio Sanitario Nazionale: A Renewing Tale of Lost Promises. In: Costa-Font, J., and Greer, S.L. (eds) Federalism and Decentralization in European Health and Social Care. Palgrave Macmillan, London, 47–66. Wade, A., Petherick, A., Kira, B., Cameron-Blake, E., Elms, J., Green, K., Hallas, L., et al. (2020). Variation in government responses to Covid-19. Blavatnik School of Government Working Paper, University of Oxford.

PART VI LESSONS FROM THE HANDBOOK

26. The political economy of health systems: research space, goals and lessons Alberto Batinti, Joan Costa-Font and Gilberto Turati

1.

THE RESEARCH SPACES THIS BOOK CONTRIBUTES TO

Notwithstanding the growing number of scientific contributions suggesting that politics matters for health care decision-making, the “political economy of health and health care” is not yet a well-established area of study, and it may still be too early to define with sufficient precision what is and is not the political economy of health systems. As we announced in the Introduction, a defining characteristic is that it is an area of inquiry that benefits from the contributions made by scholars from several disciplines, such as politics and international relations, economics (and especially health economics), public health, or sociology of health, that are all interested in questions connecting politics, health care, and health. Certainly, these disciplines differ in the approaches used to provide answers to these common questions, the type of methods adopted more than the question itself, and possibly the “rationality” assumptions relative to patients, managers, hospitals, or the ministry of health, as well as the units of analysis, whether individuals, organizations, or processes. This Handbook has attempted to present contributions that try to speak to all those disciplinary approaches. This book is not the first; in fact, it follows up on our previous book, where we attempted to provide some structure about how to think about issues that gravitate around the political economy of health care. However, to our knowledge, it is the first Handbook that contains an overview of the state of the art of the most relevant contemporary debates in political economy, including the effect of democracy, corruption, federalism, constitutional design, cabinet compositions, the role of ideology, and capture of the health system by medical interest groups. Compared to our previous book (Costa-Font et al., 2020), this Handbook is written for health care practitioners willing to expand their technical skills as well as researchers and policy scholars interested in specific political economy issues in health and health care. We have brought together contributions from a large network of expert scholars in the field. In this final chapter you will find our attempt to provide tentative answers on two main issues: what we think this book has accomplished, and what is missing but should still be included in the political economy of health care systems and should perhaps be the object of future work in the field.

2.

THE TWO MAIN GOALS OF THIS BOOK

This chapter serves to describe in some detail the two main goals we believe this book has served. First, this Handbook provides a cutting-edge discussion of some of the issues in the political economy of health and health care. That is, it successfully identifies the most relevant (and often most prolific) areas of analysis of the political economy of health systems. Clearly, 447

448  Handbook on the political economy of health systems this reflects our understanding of what the political economy of health systems is, including how collective decision-making occurs, and its relevance for today’s health policymaking and population health. However, this is only possible because a long list of scholars has contributed to the Handbook, each providing their perspective on a specific subject. However, the selected topics could be expanded, if we could commission more chapters and we had a different disciplinary composition. We begin by discussing how constitutional design influences health care systems, including the constitutional right to health, the quality of a country’s democracy, political regimes, and the vertical structure of governance. The Handbook has provided an analysis of the role of political institutions and political markets, discussing whether and how cabinet composition (for example, the presence of women or physicians), lobbying, and health professionals as a specific interest group influence health care policies and the design of health systems. Finally, the Handbook concentrates on the political economy of health care financing and reform, and it concludes with a section that delves into some of the political economy issues raised by the recent Covid-19 pandemic. The remainder of this chapter will discuss what, in our view, are the likely next steps for the study of the political economy of health care to follow, and we identify several topics that this book did not manage to discuss, but we hope they will soon be the matter of future research. Without a doubt, the field of the political economy of health and health care is larger and varies far more than the list of contributors. This is because this Handbook is limited in space, and because some authors were unable to prepare a chapter for this volume. Certainly, there may be work that we have not been able to intercept. In addition, Covid-19 has steered a new and rich research program on the role of trust in the health care systems (Besley and Dray, 2021), incentives to take up a vaccine, and the role of politics and, specifically, decentralized governance in responding to the challenges of global threats such as pandemics (Angelici et al., 2023). We have only included three chapters on Covid-19, as the pandemic has expanded the number of papers on health security, but the political economy of health is a far more balanced area examining health production, health care decision-making, and the functioning of health care markets, among other issues.

3.

THE POLITICAL ECONOMY OF HEALTH SYSTEMS: LESSONS LEARNED

The political economy of health systems is an interdisciplinary area of research with two potential approaches: a positive approach focused on understanding the effects of political and health care institutions, and a broad range of health-related outcomes (access to treatment, morbidity, mortality); alongside a normative approach focused on how to design political institutions to build better and more effective, efficient, and fairer health systems. Such a mission encompasses understanding if and how the right to health is defined in a country’s constitution, and the constitutional constraints to designing health care institutions, as well as the political incentives in the demand and supply of health care. The book has discussed the political constraints on financing health care, and how to use the tools of political economy in designing specific health policies, including steering social capital, encouraging vaccinations, or getting hospitals to deliver better health care. Recognizing a constitutional right to health creates a legal commitment that can catalyze political and social changes beyond the legislative route if such a right is internalized as an

The political economy of health systems  449 ethical norm by individuals, governments, and other stakeholders (Matsuura, 2013). However, the literature on the causal evidence of constitutional reforms is still in its infancy, and providing a better understanding of the mechanisms at play seems an important endeavor. This book has provided an updated overview of the literature on the effect of democracy on health (McGuire, 2020), drawing on different measures of democracy and health outcomes, which suggests that democratic institutions can influence different forms of health inequalities, though the effect is far from straightforward and very much dependent on the measure of democracy and health examined. Indeed, democracy can give rise to “health inequality traps,” and the effect seems to be more influential among the old-age population, and is more sensitive to certain health outcomes, such as mental health. However, democratic stability is also important since socio-political regime changes matter for health care and health. The effects of exposure to Soviet communism and the health effects of its disintegration led to the erosion of trust in health care institutions. These two effects condition even today how individuals engage with health care providers, and explain the use of informal payments and the proliferation of forms of corruption that have enhanced the privatization of the systems in some former communist states. Given the experience of many countries all around the world, a great deal of attention should be paid to “health care federalism,” defined as a constitutionally protected form of decentralized authority concerning the political and tax powers constraining the health system. In general terms, the literature shows that decentralized systems tend to achieve better outcomes and outputs under certain conditions. The reasons are the better matching of preferences and policies when median voters are heterogeneous and electoral incentives are salient; the provision of information to judge incumbents, very much in line with the idea of “interdependent governments”; and the reduction of the opportunity for rent-seeking, as potential rents are smaller at lower levels of government. If we focus on unitary states, like Italy or Spain, which experienced health system decentralisation, the literature shows that – although decentralization can be welfare-enhancing – it needs to be combined with some form of coordination, to address issues like spillover effects, patient mobility, the design of compensating transfers, and soft budget constraints. The equity and efficiency effects of health care decentralization depend very much on institutional design, as shown by using the example of Italy, a country characterized by a clear North–South divide along several dimensions. The importance of coordination in decentralized settings has been clearly emphasized by the experience of the recent Covid-19 pandemic. In this case, the issue of the vertical structure of governance of health systems cannot be discussed only considering single countries, but should consider supra-national institutions, like the European Union for many European countries. Consistently, recent survey data suggest a shift in preferences of Europeans toward a stronger role of Europe in health policymaking. Although economies of scale are very limited for many health policies’ functions (which suggests that they should not be scaled up to a European level), this is not the case for some specific sub-functions, such as procurement and prevention, where a European-wide authority can be welfare-improving, calling for further policy coordination of health care responsibilities at the European Union level. It is also important to understand what happens when decentralization is reverted, as the evidence of the tension between decentralization and re-centralization reveals in Chapter 8 in this Handbook, on the Mexican experience, but the evidence can be made extensive to other countries. The constitutionally defined right to health and health care implicitly influences the role of markets in the organization of health systems. From a theoretical standpoint, there is a tradeoff

450  Handbook on the political economy of health systems between care provided by the community and the demand for self-determination in so-called WEIRD (Western, educated, industrialized, rich, and democratic) societies. Theoretically, health care does not exhibit “public good” characteristics; hence, in principle, it could be provided by voluntary exchanges such as health care markets. However, the role of markets is limited due to the deeply entrenched convictions of citizens according to how they deal with concrete (tragic) scarcity, especially when this involves life and life-threatening decisions. Indeed, consistently with the mentioned tradeoff, the average citizen deems a political governance structure legitimate only if it gives rise to rights and claims that can be foreseen ex ante as a form of commitment to do everything that it is possible for any citizen when life-or-death situations appear. The role of social determinants of health, and in particular social capital, is considered as the network of positive enforcing social relationships which are protective of the health of individuals and operate beyond the standard role of formal medicine. Constitutions play a central role in defining political institutions, such as voting rules, the dynamics of parliamentary activity, and how the government functions, sets out the basic checks and balances that limit the government’s ability to function. The literature clearly indicates that the background (e.g., medical training), characteristics (e.g., gender), and preferences of legislators and policymakers (ideology) has an impact on the outcomes (maternal mortality) and outputs (spending per capita). For instance, considering the technical background of the members of cabinets, and more specifically by examining whether medical doctors deliver different outcomes than other technical cabinet compositions, descriptive analysis suggests evidence of a lower number of physicians practicing per capita, and lower health spending as a percentage of gross domestic product compared to non-medical doctors’ counterparts. Such evidence can be explained by standard corporatist behaviors to reduce potential competition in the delivery of health care, which might keep reimbursement well above competitive levels – though medical doctors are better at reducing health spending, due either to network effects, information, or status that legitimizes difficult spending decisions. Also, women’s representation does influence policy. Women politicians show different preferences for redistribution and prioritize children and maternal health; this is reflected in the significant effects of women politicians on policymaking and policy outcomes. Female politicians tend to reduce infant mortality and decrease neonatal mortality, even though some of such effects are constrained by the existence of adequate infrastructure. One of the ways political stakeholders can influence health care activity is by following a specific ideological manifesto. Indeed, Chapter 13 in this Handbook shows that ideology exerts an impact on health care spending. As expected, left-wing governments tend to spend more on health care than governments with different ideologies. However, this is just part of the story. For instance, after the Great Recession such effects were not totally true anymore, and evidence suggests that governments engaged in significant spending reallocations irrespective of the political ideologies of their cabinets. Other influences include the effect of the “electoral spending cycle” influencing many health system decisions such as the availability of prescription drugs and the allocations of beds and physicians. But there are several non-competitive factors pushing in different directions. One such factor is lobbying: the literature shows that the evidence on the effect of lobbying is muddled by the complexity of lobbying strategies, which are often difficult to unveil in practice. It is also important to mention that health care providers can use lobbying as leverage to push for certain policies, producing unintended effects for the patient-citizens. And lobbying is reinforced by profes-

The political economy of health systems  451 sional prestige, with physicians consulted for an opinion on what should be done when taking decisions regarding health care. Another non-competitive factor is corruption, which is the result of power and information imbalances as described in Chapter 15. There is limited research on corruption in health care, and further research on the effects of corruption can play a role in unmasking its detrimental effects on health systems. Corruption is one of the many dimensions through which institutional quality can be assessed. A wealth of available evidence suggests that better health outcomes are associated with better institutional quality. Hence, although furthering transparency and accountability has struggled to reach national and international policy agendas in the past, it does not mean that it cannot steer some change in the future. Chapter 17 in this Handbook shows that there is a role also for trade institutions and agreements as explanatory factors for detrimental effects on individuals’ health, via specific pathways and political-economic underpinnings. For instance, trade and investment agreements (TIAs) are responsible for the availability and consumption of unhealthy foods and drinks, and the associated health-related harm to society. Hence, the role of interest groups and corporate agreements cannot be overlooked when studying the determinants of population health. A great deal of discussion in the field of the political economy of health systems concerns the quest for universal care, and the interactions between public and private health care insurance. Chapter 19 shows that there are welfare effects of supplemental private insurance in a median voter framework. However, whether public health spending increases or not depends on the median voter’s demand elasticity. The empirical evidence based on World Health Organization data provides strong evidence of substitution between public and private insurance; and this result is even more likely in countries where income is more unevenly distributed. Some of these, however, are low- and middle-income countries (LMIC), characterized by informal power, imperfections in electoral systems, dominant political clientelism, and identity politics, as well as weak state capacities. All these factors should be considered to understand why some reforms have been implemented and some others not in these less-developed nations, where universal public care can have large marginal returns and can play a very important role in helping people to improve their quality of life. But difficulties in reforming public health care systems are also common in developed countries. The constraints to reform in these countries can be traced back to the weak political influence of health departments compared to the treasury in cabinets, and concerns about equity in the distribution of health resources. Difficult reforms are also those related to the pharmaceutical patent system, which clearly impact on the universal access to drugs, especially in LMIC. Governmental responses to Covid-19 offer a unique opportunity to study how changes in political processes and institutions (rules) influence health outcomes (such as Covid-19 deaths and cases). Lobbies and interest groups clearly played a role in shaping sub-optimal US health care policy and regulations, regarding Covid-19 restrictions and the allocation of Covid resources, including vaccines. As for vaccines, politics played a role also after the first phase of the vaccination campaigns ended, when the problem was not how to prioritize vaccinations, but how to convince hesitant citizens to get vaccinated and reach herd immunity. The evidence for two countries, Italy and Spain, suggests that government communication is crucial. Nudging policies based on digital Covid certificates were effective in Italy, while they were much less so in Spain, where the introduction of the certificates by the government was questioned by some regional courts. The role of regional authorities emphasizes potential differences at the local level in the presence of decentralized governance. The evidence, again

452  Handbook on the political economy of health systems for Italy and Spain and related to the first phase of the pandemic, suggests that when it is not clear what policy is the best to counteract the pandemic, allowing regional governments to test different solutions would be helpful also for the central government to understand what the best course of action is or, at least, what policies do not work well.

4.

WHAT IS NEXT IN THE POLITICAL ECONOMY OF HEALTH SYSTEMS?

We believe that this book has provided some insightful analysis that illustrate the state of the art, ideas, and evidence that political economy of health and health care is a growing field. However, even as it stands, almost every chapter already posits several questions to know more about. Clearly, there are several areas that due to space we could not cover in the book: these include, for instance, the role of bureaucracies and, more generally, a transaction costs approach to a health system. Similarly, other important areas to develop further, especially outside the United States, include a long list of questions on health care procurement, law enforcement, and medical malpractice which make up what can be defined as the law and economics of health care, which have been shown to be of increasing importance given the role of defensive medicine in the United States. Unfortunately, due to a matter of space these have not been addressed in this volume. The Handbook has not examined the effect of important political economy issues that population aging and intergenerational decision-making bring to the health system. So far most of the evidence on gerontocracies focuses on the area of pensions, and limited attention has been placed on health care. Another important growing area refers to the growing field of “behavioral political economy” of health care, which is possibly too early to include in a volume like this one, but already suggests important issues to expand upon following recent work on political adverse selection (Bursztyn et al., 2022). We hope to see important developments on this and other fields in the years to come. Finally, it is worth mentioning that, although the goal of this book has been to provide an account of the most important debates in the literature, it will always be an incomplete effort; however, we would regard this Handbook as a success if it has provided some inspiration to future scholars and policy audiences, and especially if it has managed to increase the attention paid to the political economy of health and health care, which is increasingly becoming a central research area in all social sciences.

REFERENCES Angelici, M., Berta, P., Costa-Font, J., & Turati, G. (2023). Divided we survive? Multilevel governance during the COVID-19 pandemic in Italy and Spain. Publius: The Journal of Federalism, 53(2), 227–50. Besley, T., & Dray, S. (2021). Institutions, trust and responsiveness: Patterns of government and private action during the COVID-19 pandemic. LSE Public Policy Review, 1(4): 10, 1–11. Bursztyn, L., Kolstad, J.T., Rao, A., Tebaldi, P., & Yuchtman, N. (2022). Political Adverse Selection (Working Paper No. 30214). National Bureau of Economic Research. Costa-Font, J., Turati, G., & Batinti, A. (2020). The Political Economy of Health and Healthcare: The Rise of the Patient Citizen. Cambridge University Press.

The political economy of health systems  453 Matsuura, H. (2013). The Effect of a Constitutional Right to Health on Population Health in 157 Countries, 1970–2007: The Role of Democratic Governance (Working Paper No. 10613). Program on the Global Demography of Aging. McGuire, J.W. (2020). Democracy and Population Health. Cambridge University Press.

Index

Abbott, A. 206 Abdulai, A. 273 Adserà, A. 32 agency of individuals 29 Ahlert, Marlies 143 Aimone, L. 100 Akerlof, G.A. 191 Alesina, A. 160, 206 Alford, Robert 148 Allee, T. 272 Angelici, M. 439 Arcà, E. 100 Ariste, R. 257 Arrow, Kenneth 398 Australia 89, 333–4 Austria 76 autocratic regimes 5, 36–9, 293 see also cabinet and parliamentary composition; corruption; democracies; Soviet Union and post-communist states Azfar, O. 296 Backman, G. 19 Bagues, M. 163, 170 Bai, J. 205 Banerjee, A.V. 295 Bangladesh 405 Barlow, P. 276–7 Baskaran, T. 161, 164, 168, 170 Batinti, A. 32, 33, 34, 69 Baumol, W.J. 259 Beaman, L.A. 162, 165, 170, 171, 172 Becker, G. 170, 398 Becker, S.O. 44 Béjaoui, A. 257 Beland, L. 207 Bellido, H. 205 Ben-Bassat, A. 15, 18, 21 Bennedsen, M. 226 Benoit, K. 204 Bentham, Jeremy 14 Benz, M. 29 Bertrand, M. 226–7 Besley, T. 154, 293 Bessho, S. 208 Bhalotra, S. 23, 160, 163, 164, 166, 168, 170, 172 Biden, Joe 402 Binkin, N. 442

Bjørnskov, C. 80 Blumberg, Linda 334 Blume, L. 18 Böhm, Franz 141 Bollyky, T.J. 31 Bombardini, M. 226–7 Bordignon, M. 92, 100 Born, A. 169 Bovens, Luc 143 Bowles, Samuel 143 Bowling Alone (Putnam) 178 Braendle, T. 207 Brandi, C. 279 Bray, E. 32 Brazil 161, 280, 335–6 Brunello, G. 52 Bulgaria 82 Buser, T. 168 cabinet and parliamentary composition barriers to women’s representation 168–71 coordination in decentralized systems 147 gender composition and quotas 149, 162–3, 170 gender quotas 162–3, 164–5, 170 ideology and health system outcomes 150, 151–4 ideology and professional background of 149–50, 153 influence of health-related ministers 147, 150, 151–4 policy priorities and gender 160–61, 164, 171, 207 role in health care, overview 7 women in, overview 157 see also gender; health policy Cai, W. 178–9 Callaghan, T. 229 Calsamiglia, X. 90 Cameron, D.R. 201 Camobreco, J.F. 404 Campa, P. 163 Campbell, J.L. 341 Campos, P.A. 347 Canada 89, 257, 258, 329 Cannuscio, C. 178 Carrion-i-Silvestre, J.L. 205 Caruana Galizia, Daphne 243 Casas-Arce, P. 170

454

Index  455 centralization Covid-19 and 437–8, 442–3 health care financing, overview 6 of healthcare, generally 89 in Italy 93–101, 437–8, 442 Mexico 112, 115, 123–4, 126 re-centralization 102–3, 112, 115, 123–4, 126 see also decentralization; federalism Chakraborti, R. 405–6 Chakraborty, D. 294 Chattopadhyay, R. 160, 162 Chaudhury, N. 294 Chavas, J.-P. 21 Chemouni, B. 355 child and infant mortality attributable to corruption 237 fetal origins hypothesis 168 right to health and 21–3 symptomatic of political priorities 157, 160 trade policies and 276–7 women in politics and 167–8 see also mortality rates Chile 334–5 China 257 Chown, J. 257 Christianson, J. 251 Clark, J.H. 207 Clayton, A. 162–3, 164 Clots-Figueras, I. 164, 166–7, 172 Coleman, J.S. 178 Colombia 23 Colombier, C. 207 communicable diseases 36–9, 119 see also Covid-19 community 177, 190–91 see also social capital Congleton, R.D. 69 constitutional design influencing role of markets 449–50 markets influenced by 449–50 national constitutions’ function 17–18 overview 4–5 right to health 5, 448–9 right to health enshrinement 13–14, 16, 19, 21–3, 335, 448–9 social rights and 18 see also centralization; decentralization; federalism Cooper, Z. 402 corruption Anti-Corruption Evidence (ACE) approach 242 causes of 241 child mortality and 237

defined 238–40 in health policymaking and implementation 346 in health systems 30, 48, 237–8, 244, 294–6 health systems, overview 8 institutional quality and 293, 294–6 research on 243 solutions for 241–2, 244, 294–5 temptation of, for health workers 236, 242, 294–5 Costa-Font, J. 27, 30, 32, 33, 92, 206 costs see health spending Covid-19 Affordable Care Act 403–4 American Rescue Plan 402, 404 centralized health systems and 437–8, 442–3 child immunization programs and 172 consequences of government responses, perverse 405–7 Coronavirus Act 2020 430–31 Covid certificates and nudging 414–16, 418–24, 425–7 decentralization vs centralization, overview 10 decentralized health systems and 102–103, 430–31, 437, 442–3 democracies’ response to, generally 36 efficacy of stringency measures 79–80 European Union’s response 72–4, 78–81 excess mortality 440, 443 in Germany 79, 400–401 government responses to, methodology for evaluating 76 health system weaknesses disclosed 103, 123, 124, 126 human height and 36 interest group approach to government 397–8 lockdowns 399–401, 405, 437, 438, 439–40 mental health consequences 406 in Mexico 123, 124 nudging and ideology in 424–5 Operation Warp Speed 403 pharmaceutical patent system and 380, 392–3 pressure on health systems 259 public health and policymaking, overview 10 public interest approach to government and 398 re-centralization and 102–3, 123, 124, 126 resource allocation 401–5 role of social capital in response to 37–8 trade and access to medical goods 280 in United Kingdom 430–31 vaccination campaigns 73–4, 413–14, 425–7 vaccine procurement 80–81, 82, 403

456  Handbook on the political economy of health systems see also Italy; Spain Cowling, K. 273, 275 Coyne, C.J. 401 Cruess, R.L. 253 Cruess, S.R. 253 Cuba 153, 157 Cunningham, G. 184 Cusack, T.R. 204 Dafny, L. 257 Dahan, M. 15, 18, 21 Daniels, N. 12, 14–15, 17 Datta, S. 359 Daumann, F. 400–401 decentralization advantages of 68, 86, 90, 104, 430, 442–4, 449 cabinet coordination 147 Covid-19 and 102–3, 430–31, 437, 442–3 decentralization theorem 60 disadvantages of 87, 90–93 economic rationale 89–90 equity in healthcare 90, 91, 92, 103–4, 449 fiscal competition 87 fiscal federalism and 85–9, 99 goods provisions 88 of health care, generally 57–8, 60, 89 health care financing, overview 6 health policy, overview 6 implementation, conditions for success 101–2, 104 implementation, obstacles to 346 in Italy 93–101, 437–8, 442 life expectancy and 98 medical migration and 65–6 medical rent-seeking and 67 in Mexico 112, 115–18, 121–2, 125 public healthcare expenditure and ideology 206–8 public sector efficiency 85, 86, 103–4, 449 reform and 371, 375–7 resource allocation 91 in Spain 89, 370, 371, 375–7 stimulating innovation 438 subsidiarity principle 57 tax assignment problem 85, 89 in United Kingdom 200 see also centralization; federalism Declaration of Alma Ata 13 De Doner, P. 201 De Haan, J. 30 De la Madrid, Miguel 115, 116 De Luca, G. 80, 296, 303 democracies as institutional determinants of health 27

communicable disease response 36–9 democratic environments 28–30 health goals and trade offs 28 health inequality and 34–5, 449 health inequality traps 449 health institutions and democratization 293 health outcomes, overview 5 heath spending in 293 human height and 31–3, 34–5 infant mortality rates 294 institutional constraints 28, 36, 37 lessons learned 449 life expectancy and mortality rates 30–31, 35, 293 mental health as policy area 35–6 participation by population 29 policy processes 29 private insurance and 323, 324–31 social science hypotheses regarding 33–4 types of democracies 29 see also cabinet and parliamentary composition; corruption; Covid-19; individual countries Democratic Republic of Congo 355 Denmark 79 Depalo, D. 100 De Paola, M. 170 Desierto, D. 37, 38, 399 Devries, P. 206 Dills, A.K. 400, 405 Di Tella, R. 295 Dithmer, J. 273 Diwan, I. 149 Domínguez, J.I. 343 Döring, H. 204 Downs, A. 208 Dreber, A. 169 Duflo, E. 160, 162 Durante, R. 38 Eastaugh, S.R. 251–2 educational systems 43, 48–53 efficiency decentralized healthcare and 85, 86, 103–4, 449 European Union 74–8, 79 fiscal and public sector efficiency 85, 86, 88, 98–101 in Germany 76 Italy’s NHS 76, 98–101, 103–4 Mexico’s healthcare reforms 122 recentralization 102 in Spain 76 Ehsan, A. 183 elderly 35

Index  457 Elsig, M. 272 Engelhardt, L. 204 Epple, D. 320, 325 Epple, E. 324, 326 equity and equality as barrier to reform 373–4 decentralization and equity in healthcare 90, 91, 92, 103–4, 449 health inequality 34–5, 207, 449 in Italy’s NHS 94, 95, 98–101, 103–4 lobbying’s role in health equity 213 patient mobility and 90 private insurance and 207, 330–31 recentralization 102 social determinants of health and 223–4 ESRD (end-stage renal disease) 131, 135–9 Esteve-Volart, B. 170 ethical rights 12, 14 right to health as 14 Ethiopia 356 European Union about 70 Covid-19 response 72–4, 78–81 decision-making process 71, 81 efficiency of health care systems 74–8, 79 pharmaceutical patent system 382–6 role in health care 6, 70–72, 73, 80–81, 82 trust in 72 Falk, A. 251 Fan, H. 277–8 federalism asymmetric federalism 100–101 fiscal federalism 85–9, 90–93, 95, 96, 99 health care federalism 67–8, 449 implementation, obstacles to 346 lessons learned 449 see also decentralization Fehr, E. 251 Feldmann, S.E. 226 Fernandez, M. 160 Fernández Sierra, M. 23 Ferrario, C. 90 Ferreira, F. 164 Fiorani, G. 99 Fisman, R. 227 Flory, J.A. 168 Follert, F. 400–401 Foot, J. 184 Fox, A.M. 343, 346 Fox, Vincente 117, 118, 120 France 165, 169–70 Francese, M. 100, 303 Frechette, G.R. 170 Frenk, Julio 119

Frey, C.B. 29, 38 Gagliarducci, S. 169 Gaitonde, R. 239 Gama, F.N.D. 336 gender citizen-candidate model and identity politics 160 competition preferences 168–9, 171 gender composition of cabinets 149 gender dimorphism as inequality 35 gender gap in political empowerment 157 parliamentary gender quotas 162–3, 170 policy priorities and 160–61, 164, 171, 207 voting and health 194 see also women Germany Covid-19 and 79, 400–401 healthcare reform 200 health system efficiency 76 history of pharmaceutical patent system 381–2 physicians strike 254 women in politics 170 Ghosh, S. 407 Gil, J. 32 Gillanders, R. 295 Glaeser, E. 179 Gleeson, D. 280–81 Glied, S. 148 Gneezy, U. 169 Goodman, John C. 141, 143 Gorodnichenko, Y. 29 Gouveia, M. 321, 325, 336 Gray, D. 191, 192–5 Greece 79 Green, B.N. 252 Greer, S.L. 92, 102 Grogan, C.M. 404 Grootaert, C. 184 Guiso, L. 180 Gurgur, T. 296 Gyourko, J. 164 Hackinen, B. 227 Haidt, Jonathan 142 Hart, Herbert 14 Hart, O. 370 Hatton, T.J. 32 health, measuring 16–17, 21–3 health care as voter issue 148 defined 20 health legislation and access 16 informational asymmetry 57–8, 60, 64–5

458  Handbook on the political economy of health systems measuring consumption of 16 resource allocation 14–15 see also health workers; insurance; providers; right to health health economics 15–17, 19–23, 225–6 see also health systems; political economy of health and health care health equity see equity and equality health ministers 147, 150, 151–4 see also cabinet and parliamentary composition health policy health economics of 15–17 lobbying 213–18 obstacles to implementation 346, 356–7 policy actors 347–8 political foundations 340–41 political settlements 351–6, 358 political strategies for influencing 348–50 quality and effectiveness of institutions 350–53 regulations and trade policies 282–3 role of ideas and ideologies 341 role of institutions 341, 350–53, 358–9 stages of policymaking 342–6 see also cabinet and parliamentary composition; decentralization; health spending; nudging health spending in democracies vs autocracies 293 explanatory variables for 202–3 Great Recession and 205 ideology and decentralized economies 206–8 in Italy 437 lobbying 252 party ideology and, theoretical background 201–2 party ideology’s impact on 202–8, 210 pharmaceutical patent system and 381, 386–7, 388–90, 391–2 political drivers of, in India 356–62 political settlements analysis and framework 350–56 private insurance and 331–2, 333–4 public financing of 9, 350–56 public healthcare expenditure (HCE), overview 200–202 rising costs of care 255–6 trade policies and 281 see also insurance; single-payer systems; tax-financed healthcare systems; universal health coverage; WEIRD societies health systems cabinet-member ideology and 149–50, 151–4

consolidation 256–7 formal vs informal payments 48 industry links 30 legitimized through political processes 371 single-payer 59–60, 61–3, 64–5, 67, 257 tax-financed systems 57–8, 59–60, 67, 68 types of 371–2 under-contracting 64 see also cabinet and parliamentary composition; centralization; corruption; Covid-19; decentralization; democracies; efficiency; federalism; insurance; lobbying; providers; reform; right to health; trade policies; individual countries health workers altruistic motives of 152, 236, 244, 251, 253, 297 corruption, temptation of 236, 242, 294–5 labor power and strikes 254–5, 257, 259 politics of providers 247 see also providers Healthy City movement 177 Hedge, D. 402 height 31–3, 34–5, 36 Hemphill, C.S. 257 Henrich, Joseph 141 Herwartz, H. 204, 206 Hessami, Z. 164, 168, 170 Hibbs, D.A., Jr. 201 Hiermeyer, M. 32 Hindriks, J. 201 Hirschman, A.O. 48 HIV/AIDS 258, 387, 390, 404 Hopkins, T. 184 Hsiao, A. 296 Hsiao, W.C. 257 Hubbard, Tim 391 Huber, B. 91 human rights 14, 18–19, 20 see also right to health Hutchinson, E. 239 identity politics 159–60 ideology, role of allocation of resources, overview 7–8 cabinet ministers 149–50, 151–4 in Covid-19 vaccinations 415–16, 417, 424–5 decentralization and 206–8 female politicians and policy priorities 160–61, 161f, 164, 165, 171, 207 Great Recession and 205 in health spending 201–8, 210

Index  459 insurance and 207 lobbying in policymaking 213–18, 450–51 neoliberalism 284 nudging and 424–5 party ideology 201–8 party ideology and 201–8, 210 redistributive policies 157, 160, 168, 171 Imbens, G. 421 Immergut, Ellen 344 India Covid lockdowns 405 healthcare reforms 345, 359–62 political drivers of health financing 356–62 women in politics 161, 162, 164, 165–7, 170 Indonesia 242, 355 inequality see equity and equality infant mortality see child and infant mortality infant mortality rates 21–3 democratization and 294 economic freedom and 294 Inglehart, R.F. 425 institutional quality corruption and 293, 294–6 health outcomes, overview 8 hospital quality and 297–9 impact on health outcomes, overview 292 Italy case study 304–9 literature review 293–6 mortality rates and 307 treatment choice and 299–303 insurance Australia and New Zealand case study 333–4 Brazil case study 335–6 Chile case study 334–5 cost-sharing 374–5 in democracies 323, 324–31 distribution of income and private insurance 330 guaranteeing access 134, 135–6 as healthcare determinant 15 health care equity and inequality 207, 330–31 India’s healthcare reforms 359–60 Medicare 333 normative theory of publicly provided goods 322–4 paying for service and right of rescue 133–4 private demand 319, 320–21 private insurance supplementation, overview 9 regional and country patterns 321–2 spending impact 331–2, 333–4 substitute insurance 327–8, 332 substitution alternatives 328–30 voter income 238–9, 322–3, 334–5, 336–7

in WEIRD societies 133–4 intellectual property protection 380, 384–5, 388 see also pharmaceutical patent system International Covenant on Economic, Social, and Cultural Rights (ICESCR) 13 Israel 254 Italy asymmetric federalism 100–101 Covid-19, generally 79, 102–3, 431, 436–8, 439–41, 442–3 Covid-19 certificates 415, 416 Covid-19 policy and nudging 418–24, 425–7 Covid-19 vaccinations 416–418 Covid-19 vaccinations and nudging 418–24 discrimination against women 169, 170 equity in NHS 94, 95, 98–101, 103–4 health care centralization 102, 437–8 healthcare decentralization 93–101, 437–8, 442 health spending 437 health system efficiency 76, 98–101, 103–4 institutional quality 304–9 life expectancy 98 right to health 93–4, 96 Jacobs, L.R. 229 Japan 208, 385 Jeffords, C. 21 Jensen, C. 204 Johnson, C.D. 252 Jones, P. 206 Jordan, J. 204 Joshi, N.K. 207 Kalyanaraman, K. 421 Kang, K. 227–8 Karabulut, G. 38 Kass, Leon 136, 142 Keefer, P. 358 Kelsall, T. 354, 355 Kenny, L.W. 160 Kessler, A.S. 90 Khan, M.H. 242 Khan, Mushtaq 351, 352 Khemani, S. 358 Kingdon, J.W. 342, 360 Kirkwood, J.B. 257 Klomp, J. 30 Kossarova, L. 32 Kousser, T. 228–9, 404 Koyama, M. 37, 38, 399 Kranton, R.E. 191 Kretzmann, J. 184 Krueger, P.M. 294 Kuciak, Jan 243

460  Handbook on the political economy of health systems Kudamatsu, M. 293, 294 Kusnirova, Martina 243 Kyrgyzstan 355 labor 254–5, 257, 259, 277–8 see also health workers La Ferrara, E. 160 Lagravinese, R. 296 Larcinese, V. 203 Latin America 22, 23 Latvia 50 Laver, M. 204 Lavers, T. 356 Lecce, G. 47 Lederberg, Joshua 143 Lee, J.L. 405 Leeson, P.T. 29, 397 Levaggi, R. 90, 91, 92, 99 Lewis, M. 241 licensing systems 252, 253 Lichand, G. 295 life expectancy 44–5, 46, 98, 131, 293 see also child and infant mortality; mortality rates Lindstrom, M. 296 Lippmann, Q. 165, 168, 172 Litschig, S. 295 Lizzeri, A. 34 lobbying Affordable Care Act 252–3 data sources for studying 229–31 defined 213 demand for health care, overview 8 expenditures in the US 252 healthcare association priorities 220–22 in healthcare sector, overview 219–23 health equity 213 Medicaid 228–9 modeling lobbying in health and healthcare 226–9 by patients 258 political economy of health and healthcare 225–6 provider lobbying power 251–3 resource allocation during Covid-19 401 revolving-door concerns 217 role in policymaking 213–18, 450–51 social determinants of health 223–5 see also health policy López Obrador, Andrés Manuel 115, 120, 123, 124 Lott, J.R., Jr. 160 Love, Jamie 391 Lülfesmann, C. 90 Lutz, Adrien 143

Machlup, Fritz 388 Malta 243 Manow, P. 204 March, R.J. 406 Marein, B. 32 markets health outcomes and free markets 46 influenced by constitutional design 449–50 interest group approach to government and 398–9 organ exchange and 136–9 private and public rationing 136 private insurance as guaranteeing access 134, 135–6 reform and 370–71 wages and 62–63 in WEIRD societies 135–9, 450 for well-understood forms of healthcare 58–59 see also insurance; trade policies Martell, C.R. 91 maternal mortality 157, 161, 163, 164–5, 172, 294 see also child and infant mortality; life expectancy; mortality rates Mathie, A. 184 Matsuura, H. 13, 19, 21, 22 McCormick, Robert 398 McGuire, J.W. 27, 30 McKay, A.M. 228 McKnight, J. 184 McNamara, C. 278 medical expertise, distrust of 258 medicalization 254 medical migration 65–6 medical progress 131 Mendoza, R.L. 403–4 Meneguzzo, M. 99 Menoncin, F. 91, 92, 99 mental health 35–6, 406 Metcalfe, D. 257 Mexico Covid-19 123, 124 decentralization 112, 115–18, 121–2, 125 healthcare reforms 112, 118–123 health care system, overview 112–15, 124–6 health system, overview 6 policymaking and political parties 345 re-centralization 112, 115, 123–4, 126 social security 112–14 Miller, G. 164 Mohseni, M. 296 monopolies and monopsonies 61–2, 63, 256–7, 382 Moore, J. 370

Index  461 Moore, W.J. 404 mortality rates Covid-19 excess mortality 440, 443 democratic regimes and 30–31 institutional quality and 307 maternal mortality 157, 161, 163, 164–5, 172, 294 trade policies and 278 see also child and infant mortality; life expectancy Mullahy, J. 187 Munford, L.A. 185, 187–9 Muntaner, C. 30 Myanmar 355 national healthcare systems see health systems; single-payer systems; tax-financed healthcare systems New Zealand 333–4 Niederle, M. 168 Nigeria 237, 242 non-communicable diseases 271, 274–5 Norris, P. 425 North, Douglass 141, 351 Nozick, Robert 141 nudging about 412–13 Covid certificates and 414–16, 418–24, 425–7 health policy and 412–13, 427 in policymaking, overview 10 role of ideology and 424–5 Nussbaum, Martha 16–17 nutrition 34–5, 273–5, 283 Oates, W.E. 60 Obama, Barack 200 Ocké-Reis, C.O. 336 Ogliari, L. 47 Oloomi, S. 207 Olson, M., Jr. 67 Olson, Mancur 398 On Ethics and Economics (Sen) 15 organ transplants and allocation 136–9 Pak, S. 32 Pakistan 280, 405 Palmer, A. 21 Panda, P. 276 pandemics 36–9, 430, 442–4 see also Covid-19 Paradiso, M. 296 parasite stress hypothesis 34, 36 Parsons, Talcott 257 Paserman, M.D. 169

patents see pharmaceutical patent system Peltzman, Sam 398 Penrose, Edith 388 Perron, P. 205 Persico, N. 34 pharmaceutical patent system cost of medicines 381, 386–7, 389–90 Covid-19 and 380, 392–3 European Union 382–6 evaluation of 386–90 global patent system (TRIPS) 381, 384, 385–6, 387, 390–92 health spending 388–9 history of 381–6 low-purchasing-power demographics 381, 386, 387, 389–90, 391 patents vs patients campaign 391 “pharmemerging countries” 386 prize fund for R&D 391–2 reform of, overview 9 social value of healthcare and 388–90 Philipson, T.J. 404 physical fitness 50–53 Pierce, J.R. 277 Pigou, Arthur Cecil 398 political economy of health and health care defined 447 lessons learned 448–52 see also cabinet and parliamentary composition; constitutional design; corruption; democracies; health systems; ideology, role of Pons-Novell, J. 206 Portes, A. 178 Posner, R.A. 404 Potrafke, N. 204, 205, 208 Pracht, E.E. 404 Pridmore, P. 178, 184 private insurance see insurance providers constraints on 255–8 health policies, overview 8 incentives and motivations 250–51, 262, 297–9 leverage over healthcare systems 251–5, 259 lobbying power 251–3 politics of 247, 259 provider power, overview 248–50 regulatory frameworks and 292 treatment choice incentives 299–303 see also health workers public health care constraints to reform of 372–5 gender and political representation 157

462  Handbook on the political economy of health systems interest group approach to government 397, 398–9 unsustainable growth of 130 women advancing public goods provision 160–61, 162, 164–8, 172 see also single-payer systems Pulejo, M. 399–400 Putnam, R.D. 177, 178, 180 Puy, M.S. 90 quality of life 177, 190–91 see also social capital Querubín, P. 399–400 Radin, D. 296 rationing access 134, 135–9 Rawanda 353, 355–6 Razvi, S. 294 Redford, A. 400, 405 reform constraints to 371, 372–8 decentralization and 371, 375–7 defined by interest groups 376 equity and 373–4 in Germany 200 in India 345, 359–62 institutional advantages and markets 370–71 insurance reforms 333, 359–60 in Mexico 112, 118–23 pharmaceutical patent system 9 post-communist reforms 47–8 public opportunity for 372 social barriers to 371 in Spain 9, 370, 375–8 Reich, M.R. 343, 346, 347 renal disease, end-stage (ESRD) 131, 135–9 resource allocation 14–15, 91, 122, 215, 401–2 role of ideology 7–8 Richardson, J.R.J. 330, 336–7 right to health about 12, 13–14 attempts to measure health 16 capability theory and 17 as constraint 20–21 as ethical right 12, 14 healthcare and 20 health economics and 15, 16, 19–23 human rights indicators and 18–19, 20 in Italy 93–4, 96 legal and constitutional enshrinement 13–14, 16, 19, 21–3, 335, 448–9 as legal, political, and social instrument 21–3 as legal right 14–15 in Mexico 112 resource allocation and 14–15, 19

as social objective 20 see also constitutional design; health systems Roberts, G. 405–6 Roberts, M.J. 370 Roesel, F. 208 Roland, G. 29 Romano, R. 320, 324, 325, 326 Rose, N.L. 257 Roth, Alvin 136 Runkel, M. 91 Rwanda 157 Ryan, M.E. 403 Saiz, A. 170 Salinas de Gortari, Carlos 116 Sampat, B. 403 Sandel, Michael 136 Schargrodsky, E. 295 Schott, P.K. 277 Schram, A. 275 Semashko, Nikolai 42 Sen, A. 12, 14, 15, 16–17, 23 Shah, F. 21 Shandra, J.M. 32 Shkreli, Martin 387 Shurchkov, O. 169 Shvetsova, O. 69 Simar, L. 76, 77, 79, 82 Singh, Manmohan 360 single-payer systems 59–60, 61–3, 64–5, 66, 67, 257 see also insurance; tax-financed healthcare systems Slovakia 243 Smith, P. 91 Snoddon, T. 91 social capital about and types of 177–80 civic engagement and health 191–5 Covid-19 and communicable disease response 37–8 effect on health, overview 183, 195–6 health outcomes, overview 7 Healthy City movement 177 participation in society and health 184–91 proxy measures for 180–82 social determinants of health 27, 223–5, 450 social welfare 20, 112–14, 398, 399 Solé-Ollé, A. 203 Sorribas-Navarro, P. 203 Soviet Union and post-communist states about 44 communicable diseases, control of 47 educational system 48–50, 51–3 free market and worse health outcomes 46

Index  463 Gorbachev Anti-Alcohol Campaign 46 guaranteed access to healthcare 13 health care reforms post-communism 47–8, 449 health care system’s organization 42, 47 health outcomes, overview 5 life expectancy and 44–5, 46 physical fitness and 51 post-communist deterioration in health 46, 449 post-Soviet transformation process 46–7 public institutions, mistrust of 45–6 socio-economic status and health correlation 43, 46 Spain constraints to health system reforms 375–8 Covid-19, generally 79, 431, 438, 439–41, 442–3 Covid-19 certificates 414–15 Covid-19 policy and nudging 418–24, 425, 427 Covid-19 vaccinations 416–18 Covid-19 vaccinations and nudging 418–24 decentralization 89, 370, 371, 375–7 health system efficiency 76 parliamentary gender quotas 163 politicization of health system 376–7 universal health coverage 375–6 women in politics 170 spending see health spending Spithoven, A. 228, 252 Starr, Paul 251 Stigler, George 398 Stiglitz, Joseph 391–2 Stinnett, A.A. 187 Stinson, A.J. 32 subsidiarity principle 57 Sweden 79, 89, 168 Switzerland 207, 242 Tabellini, G. 305, 307 Takaku, R. 208 tax-financed healthcare systems 57–8, 59–60, 67, 68 Theilen, B. 204, 206 The Social Transformation of American Medicine (Starr) 251 Thompson, H.A. 29, 397 Tiebout, C.M. 65–6, 68 Tollison, Robert 398 trade policies African Growth and Opportunity (AGOA) 276 Canada-US FTA (CUSFTA) 275 Covid-19 medical supplies and 280

employment and 277–8 environmental hazards and 278–9 global trade 270, 271 health impacts of 273–9 health services and policies 280–83 health spending 281 health systems, overview 270–72 Multi-Fibre Agreement (MFA) 278 neoliberal ideology and 284 non-communicable diseases and 271, 274–5 North American Free Trade Agreement (NAFTA) 272, 275, 282 political-economic influences on 284 political economy and health outcomes, overview 8 see also markets Trebbi, F. 226–7 Troesken, W. 36 Trump, Donald 200 Tullock, Gordon 398 Tuohy, C.H. 148 Turati, G. 92 Turk, Danilo 18 Turkey 406 United Arab Emirates 157 United Kingdom ‘Big Society’ 177 civic engagement and health 191–5 community approaches to health services 177 cost-effectiveness analysis of health policy 15 Covid-19 and 430–31 decentralization 200 Health and Social Care Act 200 health care as voter issue 148 investigating corruption 243 participation in society and health 184–91 physicians strike 254–5 provider lobbying 252 right to health and 16 Salford Integrated Care Programme (SICP) 184 United Nations 12, 13, 20 Universal Declaration of Human Rights (UDHR) 13 United States access to healthcare and legislation 16 Affordable Care Act 200, 228, 229, 252–3, 259, 403–4 American Rescue Plan 402, 404 Covid-19 and 399–400, 401–2, 403, 404 healthcare association priorities 220–22 health care as voter issue 148 healthcare inequality 207

464  Handbook on the political economy of health systems healthcare workers strike 254, 259 history of pharmaceutical patent system 381–6 insurance as healthcare determinant 15 lobbying 228–9, 252–3 maternal mortality 172 Medicaid 228–9 medical expertise, distrust of 258 Medicare 333, 402 monopsony in health systems 257 Operation Warp Speed 403 public healthcare expenditure and ideology 207 right to health and 13, 16, 22 Sunshine Act 257–8 see also insurance; lobbying Universal Declaration of Human Rights (UDHR) 13 universal health coverage 339–40, 375–6 policy process, overview 9 see also insurance; Mexico; right to health; single-payer systems Van Bastelaer, T. 184 Van der Gaag, M. 183 Van Kessel, S. 425 Vartanova, I. 149 Vatter, A. 148 Vesterlund, L. 168 Vian, T. 238, 241 Vietnam 354–5 Vittori, D. 425 Voigt, S. 18, 80 Von der Leyen, Ursula 81 wages 62–3 Wallis, John Joseph 141 Wälti, S. 148 Wärneryd, K. 67 Webber, M. 183 Weingast, B.R. 87–8, 141 WEIRD (Western, Educated, Industrialized, Rich, Democratic) societies bodily autonomy of patient respected 137 coercively financed health care 131–3

health systems, overview 6–7 insurance and paying for service 133–4 political equilibrium of 130, 137–8 private and public rationing 135–9, 450 public health care guarantees 130–31 rule of rescue 131–4, 138, 140 welfare state endorsed 132, 140 Wen, J.F. 91 Wiese, R. 205 Wilson, P.W. 76, 77, 82 Wippert, P.M. 33 women barriers to political representation 168–71 child mortality and female politicians 167–8 citizen-candidate model and identity politics 160 competition and 168–9, 171 competitive elections and 164, 169 diminishing anti-female bias 170 discrimination against 169, 170 gender gap in political empowerment 157 maternal mortality 157, 161, 163, 164–5, 172, 294 parliamentary gender quotas 162–3, 164–6 policy priorities 160–61, 164, 165, 171, 207 political representation and public health provision 7, 157, 164–8 as politicians 161 in politics in India 161, 162, 164, 165–7, 170 public goods provision and political representation 160–61, 162, 164–8, 172 redistributive policies favored 157, 160, 168, 171 voting and health 194 see also cabinet and parliamentary composition; gender Woolcock, M. 178 Zamboni, Y. 295 Zanardi, A. 90 Zetterberg, P. 162–3, 164 Zhang, Q. 279 Zhu, L. 207