Democracy and Dissent: The Challenge of International Rule Making

Democracy and Dissent M2466 - VIBERT PRINT.indd i 15/12/2010 08:09 M2466 - VIBERT PRINT.indd ii 15/12/2010 08:09 ...

Author: Frank Vibert

21 downloads 1133 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Democracy and Dissent

M2466 - VIBERT PRINT.indd i

15/12/2010 08:09

M2466 - VIBERT PRINT.indd ii

15/12/2010 08:09

Democracy and Dissent The Challenge of International Rule Making

Frank Vibert Visiting Fellow, ESRC Centre for the Analysis of Risk and Regulation, London School of Economics and Founder Director of the European Policy Forum – a London based think tank, UK

Edward Elgar Cheltenham, UK • Northampton, MA, USA

M2466 - VIBERT PRINT.indd iii

15/12/2010 08:09

© Frank Vibert 2011 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: 2010932044

ISBN 978 1 84980 920 7 (cased) ISBN 978 1 84980 921 4 (paperback)

04

Typeset by Servis Filmsetting Ltd, Stockport, Cheshire Printed and bound by MPG Books Group, UK

M2466 - VIBERT PRINT.indd iv

15/12/2010 08:09

Contents List of figures and boxes List of tables Acknowledgements List of acronyms

vi vii viii ix

Introduction 1 ‘How false were our postulates’ 2 Managing strain – styles of international rule making 3 Analytic frameworks 4 The choice of venue 5 The choice of instruments 6 The sources of failure 7 Diagnosing the democratic deficit 8 Challenge systems and the rule makers 9 Dissonance and democracy 10 Conclusions

1 17 36 58 78 94 118 141 167 190 205

Appendix A: Transaction costs and styles of organising Appendix B: Definitions of selected cognitive terms

216 219

Bibliography Index

221 247

v

M2466 - VIBERT PRINT.indd v

15/12/2010 08:09

Figures and boxes FIGURES Figure 2.1 Figure 2.2 Figure 5.1

Typology of international rule making International coordinating groups Orders of preference in negotiating and implementing rules and regulations

45 48 112

BOXES Box 5.1 Box 5.2 Box 5.3 Box 5.4 Box 8.1 Box 8.2 Box 8.3

Varieties of strategic motivation: climate change Escape clauses in the FCTC Options in the event of non-compliance: pharmaceuticals in the EU The use of intermediary instruments: the case of Art. 8 of the WHO’s FCTC Competitive evaluation in practice: the case of EMEA/SAWP Process tracing in EFSA scientific opinions Audit: pharmacovigilance in the EU

106 108 111 114 177 178 184

vi

M2466 - VIBERT PRINT.indd vi

15/12/2010 08:09

Tables Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 4.1 Table 4.2 Table 6.1 Table 6.2 Table 8.1 Table 8.2 Table 8.3 Table 8.4 Table 8.5

Selected ‘classic’ international organisations (by date of establishment) Selected ‘new’ venues Selected principles-based instruments Transaction costs and styles of organisation Organisational forms and organisational logic Examples of professional inter-institutional relationships Classes of cognitive failure Elite group cognitive dispositions and bias Intergovernmental Panel on Climate Change: key actors IPCC inputs and outputs IPCC report preparation processes IPCC use of confidence measures Challenge techniques against bias

39 41 43 52 85 90 129 132 173 174 175 180 186

vii

M2466 - VIBERT PRINT.indd vii

15/12/2010 08:09

Acknowledgements First, I would like to thank Professor Bridget Hutter, Director, ESRC Centre for the Analysis of Risk and Regulation (CARR), LSE, for inviting me as Visiting Fellow at CARR from October 2008 to September 2010. This enabled me to write the book as well as to benefit greatly from the help of colleagues at CARR. She also kindly commented on a draft of Chapter 5. I would also like to thank Professor Richard Rose, University of Aberdeen, who gave general encouragement and commented on an early outline. Professor Edward Page, Beatrice and Sydney Webb Professor at LSE, read the first draft in its entirety and gave me valuable advice. Professor Mark Thatcher, Professor of Comparative and International Politics, LSE, also gave general advice and commented in particular on a first draft of Chapter 3. Professor Christopher Hood, Gladstone Professor of Government and Fellow of All Souls College, Oxford, gave helpful advice on the approach to regulatory ‘failure’. Among others in the academic community who commented I should like to mention Professor Peter M. Haas, University of Massachusetts, Amherst. Among practitioners who commented I would like to thank in particular Dr Max Watson, Fellow, Wolfson College, Oxford, Cari Votava (World Bank) and Martin Summers.

viii

M2466 - VIBERT PRINT.indd viii

15/12/2010 08:09

List of acronyms AMAN ARM ASEAN BCBS BIS CHMP CRU EFSA EMEA EPA EU FAO FATF FCTC FIU FSB FSF GAAP GAO GATT GMO GOARN IAEA IAIS IANA IASB IASC IBRD ICANN ICAO ICC ICJ

Israeli Defence Forces’ Directorate of Military Intelligence Adjustable Rate Mortgage Association of Southeast Asian Nations Basel Committee on Banking Supervision Bank for International Settlements Committee for Medicinal Products for Human Use Climatic Research Unit, University of East Anglia European Food Safety Authority European Medicines Agency US Environmental Protection Agency European Union Food and Agriculture Organization Financial Action Task Force on Money Laundering Framework Convention on Tobacco Control Financial Intelligence Units Financial Stability Board Financial Stability Forum Generally Accepted Accounting Principles US General Accounting Office General Agreement on Tariffs and Trade Genetically Modified Organism Global Outbreak and Alert Response Network, World Health Organization International Atomic Energy Agency International Association of Insurance Supervisors Internet Assigned Numbers Authority International Accounting Standards Board International Accounting Standards Committee International Bank for Reconstruction and Development (World Bank) Internet Corporation for Assigned Names and Numbers International Civil Aviation Organization International Criminal Court International Court of Justice ix

M2466 - VIBERT PRINT.indd ix

15/12/2010 08:09

x

IFAD IFC IFRS IHR ILO IMF IMO IOSCO IPCC ISO ITO ITU MERCOSUR MMoU NGO NRC NSG OECD OEEC OIE OPEC PICT PRTR ROSC UK UN UNCTAD UNEP UNESCO UNHCR UNICEF UNIDO UNWTO UPU US

M2466 - VIBERT PRINT.indd x

Democracy and dissent

International Fund for Agricultural Development International Finance Corporation International Financial Reporting Standards International Health Regulations, World Health Organization International Labour Organization International Monetary Fund International Maritime Organization International Organization of Securities Commissions Intergovernmental Panel on Climate Change International Organization for Standardization International Trade Organisation International Telecommunication Union Mercado Común del Sur [Common Market of South America] Multilateral Memorandum of Understanding Non-Governmental Organisation US National Research Council Nuclear Suppliers Group Organisation for Economic Co-operation and Development Organisation for European Economic Co-operation World Organisation for Animal Health Organization of the Petroleum Exporting Countries Project on International Courts and Tribunals Pollutant Release and Transfer Register Reports on the Observance of Standards and Codes (World Bank). United Kingdom of Great Britain and Northern Ireland United Nations United Nations Conference on Trade and Development United Nations Environment Programme United Nations Educational, Scientific and Cultural Organization United Nations High Commissioner for Refugees United Nations Children’s Fund United Nations Industrial Development Organization United Nations World Tourism Organization Universal Postal Union United States of America

15/12/2010 08:09

List of acronyms

WHO WIPO WMO WTO

M2466 - VIBERT PRINT.indd xi

xi

World Health Organization World Intellectual Property Organization World Meteorological Organization World Trade Organization

15/12/2010 08:09

M2466 - VIBERT PRINT.indd xii

15/12/2010 08:09

Introduction The institutions and procedures for making international rules and regulations are in need of renewal. The existing system is prone to failure. It is also undemocratic. This book is about the principles of design for the international order that could make it both less vulnerable to failure and more consistent with democratic norms. The message of the book is that specific governing rules are needed to discipline the rule makers.1 The current system relies on self-disciplining by expert elites when it comes to the formulation of the substance of new international rules. It relies on self-restraint by governments when it comes to implementing them. Neither self-discipline nor self-restraint is enough. Self-discipline is not good enough to guard against failures in systems of rule making. Self-restraint is not good enough to guard against undemocratic impositions of rules. Governing rules are therefore needed to enforce external disciplines and restraints on the rule makers. The governing disciplines and restraints needed are those that allow for challenge and dissent. An organised system for challenge is the foundation for good regulation and rule making. The possibility of meaningful dissent is also a foundation of democratic legitimacy. Both have to be embedded within a system of governing rules. Only then will the current system of international rule making become less vulnerable to

1 The term ‘rules’ can be used in two different senses. First, there are rules (sometimes referred to as ‘meta-rules’) that define the rules of the game and that comprise the constitutional rules that govern the behaviour of those with power in a system of government. Second, there are rules (or regulations) that are made by those with power (the rule makers) within the game or within a constitutional and political order. In order to reduce confusion and, since not all governing rules of a political order are put into constitutional form, ‘meta’ or ‘constitutional’ rules are referred to in the text as ‘governing rules’ unless a specific constitutional context is indicated. Unless governing rules are specified, the terms ‘rules’ and ‘regulations’ are used interchangeably to refer to policies endorsed at the international level, designed to shape behaviour and outcomes both at the level of national governments as well as at the level of the general public and usually transmitted through rule-making bodies in national or regional jurisdictions. Baldwin defines a ‘rule’ as ‘A general norm mandating or guiding conduct or action in a given type of situation’ (Baldwin 1995: 7). Black defines regulation as ‘sustained and focussed attempts to change the behaviour of others in order to address a collective problem or attain an identified end or ends, usually through a combination of rules or norms and some means for their implementation and enforcement, which can be legal or non-legal’ (Black 2008: 139).

1

M2466 - VIBERT PRINT.indd 1

15/12/2010 08:09

2


failure. Only then will it be possible for international rule making to gain democratic legitimacy.2

THE REFORM OF INTERNATIONAL RULE MAKING An increasing number of the rules and regulations that affect the lives of nations, individuals and communities are being made at the international level. There are compelling reasons why this should be so. Climate change, health pandemics, scientific innovations and closely inter-connected markets do not recognise national or regional boundaries. At the same time, despite the end of the Cold War, the world remains confrontational, divided by values, interests and identities. The further development of a rule-based international system of behaviour thus remains vital in order both to tackle global problems and to avert global confrontations. The way it develops will not only frame the general conditions that impact on individual lives (such as vulnerability to pandemics or nuclear proliferation) but also the detail of lives (such as the type of pension fund individuals may be able to invest in). In recent years the making of new international rules has undergone a fundamental change.3 The architecture of international organisations with their universal membership set up at the end of the Second World War remains in place. But new venues are being used for rule making, new rule makers are involved, new types of rules are being made and they are being

2 The traditional term ‘international’ is used in preference to a term such as ‘global’ in order to avoid unfruitful debate about what is global or not (for a discussion of definitional issues surrounding the terms global and globalisation see McGrew 2005: 207–11). It is also used in preference to the term ‘transnational’ that is used in order to suggest the connections that cross boundaries but that are not necessarily global and, in addition, that do not necessarily involve states (see for example, Djelic and Sahlin-Andersson 2006 and Slaughter 2003). However, the use of the more common term ‘international’ does not create a presupposition that nations or states are the only actors in rule making that crosses boundaries or that there is a bright line that divides the international from the domestic. The term international is also used to include intercontinental actors such as the EU, ASEAN and MERCOSUR that may be active in rule making. Rule making within such regional groupings is not however the focus of the discussion. 3 The term ‘international rule making’ can be defined narrowly to refer only to those rules that are formulated, agreed and promulgated at the international level (for example the rules agreed in Basel on the capital adequacy of banks) and more broadly to those rules that may originate within national jurisdictions but which reflect perceived international standards of behaviour (for example corporate governance standards). The discussion that follows applies to any rules that are derived in whole or in part from external sources, recognising that at the edges the distinction between what is of domestic origin and what is of international origin is blurred. The term ‘international order’ is used to denote the aggregate of international institutions and rules that are intended to affect public and private behaviour.


15/12/2010 08:09

Introduction

3

spread in new ways. Alongside the old architecture a new one has come into being. The new architecture for international rule making is largely the product of improvisation rather than the result of thinking about the principles of design. Thinking about the international order as an integrated whole has been put to one side in favour of a pragmatic approach to practical problems. The end result is a mix of old and new venues and old and new styles of rule making. It is a mix that is convenient for the rule makers. However, it disguises the failures, reduces the accountability, and undermines the legitimacy of international rule making. The international financial crisis that unfolded in 2008 has exposed the vulnerabilities of the existing system. The crisis brought with it a massive destruction of wealth and a setback to an unprecedented period of global economic growth that had seen tens of millions of people lifted out of crushing poverty. It was however more than an economic and financial cataclysm. It was also a massive failure in the way that international rule making and international institutions have evolved in recent years. One of the most important functions of any system of financial regulation and rule making is to maintain public confidence in the financial system and to maintain the flow of credit. In this, the system failed across much of the world. Financial regulators were not the only ones whose failure led to the 2008 financial crisis – governments and markets also failed. However the regulatory failure is particularly troubling because the new international architecture and instruments have been largely pioneered in the financial sector – so also have new approaches to rule making such as ‘risk-based’ regulation and ‘principles-based’ regulation. The pioneering is not coincidental. Finance is where a fully international market has first come into existence, where any strains are most quickly transmitted and where the need for new structures and techniques for international governance have been most acutely felt. The lessons behind the failure extend well beyond the financial sector and present a challenge for all areas of international rule making. Other areas of international rule making, such as in environment and health, have also been changing in ways analogous to those in the financial sector. The mix of institutions and instruments involved in the financial sector is not unrepresentative of the current styles of working in other areas of international rule making. In 2008 it was international rule making in the financial area that failed. Unless the lessons are learnt, future years could bring failure in other equally important areas such as health or climate change or security. The failures may be failures of omission where rules that are needed are not made. Or they may be failures of commission where the rules that are made are fundamentally flawed. The 2008 financial crisis contained regulatory and rule-making failures of both types.


15/12/2010 08:09

4


A TWIN CHALLENGE The challenge to the current approach to international rule making is partly about why it failed in one of its most important practical tasks. The mix of venues and approaches to rule making has been justified as representing a pragmatic response to practical questions. This justification has been undermined. The challenge is also about the connection, or lack of connection, of the current approach to international rule making to democratic values. What the rule makers have claimed to have gained in seeking pragmatic and flexible ways to solve problems through international cooperation, citizens have lost in an erosion of basic democratic protections.4 In the 2008 international financial crisis, national governments shifted the blame onto international forces ‘beyond their control’. But the international arena that was meant to provide a means of weathering such challenges and failed to do so, itself offers no democratic means of accountability. On the contrary, the co-mingling of international venues, old and new, together with the co-mingling of old and new instruments for disseminating rules, creates an impenetrable jungle of acronyms. Citizens in democratic countries often do not know where salient rules have been made or who the rule makers are. Other procedural protections that are crucial for democracies have also been lost. Citizens are poorly informed about what procedures have been followed in making the rules, have little or no opportunity to influence the making of the rules, and lack the means to protest effectively if they disagree with them or to find remedies if they damage their own interests. In democratic theory the citizen is king and the rule makers are subjects. In today’s world, the making of international rules sees the rule makers as kings and citizens as subjects. When rule making fails, there is no redress. For citizens, faith in the democratic protections offered by their own societies is shaken; so too is their faith in international rule making. The twin challenges can be seen as twin deficits – a democratic deficit combined with an effectiveness deficit. They are not new but they have become more evident and are going to become more, not less, important in future. They need to be corrected. It is time therefore to think again about the principles of design in international rule making.

4 Dryzek comments ‘Collective choice in the international sphere is at best only a thinly democratic affair, at worst thoroughly undemocratic’ (Dryzek 1999: 30).


15/12/2010 08:09

Introduction

5

PRINCIPLES – FORM AND PROCESS In thinking about the principles of design there is a long-standing debate between those who emphasise the importance of ‘form’ and those who stress the importance of ‘processes’.5 An emphasis on form means looking at the design and role of institutions, the scope and spheres of their authority, at the form of instruments they use to express authority, and at the relationship between forms of authority and forms of instrument. An emphasis on process means looking at the procedures institutions and authorities employ in order to formulate the rules and achieve their objectives. It also means looking at how different actors behave in the different settings in which rules are made and applied. It means examining the different ways in which the authority of the different actors in rule making is grounded. Both perspectives involve empirical analysis. Both also involve normative analysis. In the context of post war international rule making it is a debate that started at the time that the post war architecture was originally established during the Second World War. Its salience continues. The two perspectives provide a unifying theme for much of the discussion that follows. They provide a common thread from the early post war history to current questions about the choice of venues and choice of instruments in international rule making. They link the discussion of the reasons for failures in rule making to the discussion of the reasons for the lack of respect for democratic norms.

THE ANALYTIC FRAMEWORK The analysis undertaken in this book juxtaposes two different frameworks that refract the debate about form and process. The first framework is that provided by theories of multi-level governance. The second framework is drawn from what is known as diffusion theory. The first framework, multi-level governance, focuses on forms and spheres of authority in the modern world. In particular it tries to combine newer, so-called ‘horizontal’ forms of authority, wielded by networks of 5 In later discussion, the start of the post Second World War debate is attributed to the functionalism of Mitrany. It is discussed from a different perspective by Easton (1990). The debate has a long historical background going back to late theories of kingship (where Bodin for example emphasised the importance of distinguishing between different forms of monarchical authority) and to early democratic theory that looked to processes embodied in notions such as ‘contract’ that were deemed necessary in order to establish legitimate authority.


15/12/2010 08:09

6


officials, lawyers, experts from the natural and social sciences, and civil society actors, with traditional, more hierarchical forms of authority wielded by governments, their legislatures, courts and executive agencies. The term ‘governance’ is preferred to ‘government’ in order to recognise that governments are not the only sources of authority and to signal that the new forms of authority cohere as a system.6 Multi-level governance offers an approach both to the analysis of the effectiveness of rule making across different jurisdictions, including the international, and to the analysis of its democratic legitimacy.7 Unfortunately, the claims made on its behalf are greatly exaggerated. In making the rules for the international financial system, governments have had the use of an extensive range of different organisational forms with different spheres of authority. Nevertheless, the financial crisis has shown that the current system of international rule making is still prone to failure. In addition, far from helping to pinpoint democratic responsibility, multiple spheres of authority and multiple actors often help to conceal where the real power lies and who the real decision takers are. The analysis in this book therefore also utilises a second framework that focuses on processes rather than form. The second framework chosen to explore the processes involved in international rule making is taken from what is known as ‘diffusion theory’, originally used by sociologists to explore the adoption of innovations in areas such as plant technology and medicine. In this application the adoption of a new international rule or regulation is treated as another kind of innovation.8 The diffusion framework involves a major distinction between three stages of international rule making, between the different actors playing the pivotal role at each stage and between the different types of reasoning deployed by the key actors at each stage.

6 Theories of multi-level governance draw on a number of different sources of inspiration including international relations theory, studies of federalism and studies of the European Union. For a description of the sources see Hooghe and Marks (2003). 7 The term ‘effectiveness’ is generally used in the discussion in preference to ‘efficiency’ as the term for characterising the performance of a rule or regulation. ‘Efficiency’ implies an economist’s definition of achieving a given end at least cost and is appropriate in the limited context of a discussion of the economics of rule making. However, rules and regulations are particularly about changing behaviour and additional criteria are also apposite. See the discussion of ‘effectiveness’ in Young (1999b: 21–27). 8 Black defines innovation in the field of regulation as ‘the use of new solutions to address old problems, or new solutions to address “new” (or newly constructed) problems’ (Black 2005: 4). Walker (1969: 881) defines innovation in terms of ‘a program or policy which is new to the states adopting it, no matter how old the program may be or how many other states have adopted it’. Either definition can be used.


15/12/2010 08:09

Introduction

7

The first stage is that of the formulation of rules where the substantive content of the rules is determined. It is a stage dominated by experts and bureaucratic elites and by reasoning that reflects their professional disciplines.9 Since experts often disagree, the venues and procedures used at this stage are those conducive to achieving consensus among peers. The experts form not only communities of knowledge but also communities of practice. The second stage is the adoption or endorsement of the proposed rules. It is a stage still dominated by governments and involves a different type of reasoning that reflects their own strategic calculations in deciding whether or not to endorse a proposed rule and, if so, in what form. The third stage is the acceptance of the rules. This is the stage where electorates and citizens are the central actors in democratic societies. In their own reasoning they may well be distrustful of what both governments and experts have to say.10 These distinctions are not absolute and are often blurred in practice. The experts focusing on the content of rules will keep an eye on what governments may be willing to endorse and adopt. Governments in their turn, in endorsing the proposed rules, may keep an eye on what their electorates may accept. Despite the simplifications, the distinctions nevertheless have hugely important advantages for both empirical and normative analysis. The two perspectives are discussed in greater detail later in the book. They are not mutually exclusive. Each offers important insights into why international rule making may fail and why there is a lack of consistency with democratic norms. It is important that the analytic perspective provides for both. In addition, the use of the two frameworks provides for an interdisciplinary approach.11 However, what is key is that the two approaches lead in different directions in diagnosing both the sources of failure in rule making and the origins of the democratic deficit. As a result they also lead to quite different prescriptions about the remedies for each.

9 An ‘expert’ can be defined as someone recognised in their field as competent or authoritative in providing advice on the design of a public policy. 10 These distinctions are based on Ryan and Gross (1943). In their pioneering case study of the diffusion of hybrid corn in Iowa in the 1930s, the laboratory scientists developed the hybrid seed, the seed company salesmen and farm agencies endorsed its superior qualities and the farmers were the accepters. 11 In particular it bridges one of the main divides in the social sciences between sociology and the assumptions about rationality commonly used in other social science disciplines. Benvenisti and Hirsch (2004) note that the distinction between rational choice and sociological analysis constitutes one of the major dividing lines in social sciences scholarship.


15/12/2010 08:09

8


DIAGNOSING FAILURE In the multi-level governance framework, failures in international rule making flow from a variety of forms of ‘mismatching’ and ‘misalignments’. For example, a task may be misassigned to the wrong kind of organisation, or there may be a misalignment between the spheres of authority of an organisation and the scale of collective good to be provided, or a mismatching between the substance of a measure and the legal form in which it is conveyed for implementation. By contrast, in the diffusion framework the issues centre on the first stage of rule making. They centre in particular on the weaknesses of what are known as ‘epistemic communities’ or elites. It is the epistemic elites who play a determining role both in framing the problem to be solved and in putting forward proposals on the substantive content of rules that purport to meet the problem. The advantages of epistemic elites centre on the use of technocratic, evidence-oriented venues. The disadvantages crystallise around the shortcomings of consensus between experts and peer approval as validation for evidence-based rule making. There are extremely strong theoretical reasons for thinking that like-minded groups are especially vulnerable to certain kinds of cognitive failings.12 These failings lead to defective rules. The new venues try to avoid a sterile type of political contestation. However, a different type of contestation is necessary for assurance that the proposed rules are indeed evidence based. Without procedures and institutions that enable the substance of the rules to be challenged, rule making can go fundamentally wrong. The two accounts of failure are therefore quite different. The account offered by the multi-level governance framework centres on what might be classed as a category of ‘executive failure’ where governments are at fault for the misalignments, mismatches and mis-assignments. The account offered by the diffusion framework centres on cognitive failure among the elites – failure that stems from the methods elites employ in going about making the rules. In looking at these two possible sources of failure there are pragmatic reasons for thinking that cognitive failure is ‘prior’. Failings in the original substantive choices made by the experts will not be corrected by changing the forms of organisation, reassigning roles, redefining spheres of authority or by altering the nature of the instruments through which the rules are conveyed. 12 Cognitive failure can be defined as failure in the method of mobilising and interpreting available information and knowledge, in making causal inferences, and in estimating uncertainties and probabilities.


15/12/2010 08:09

Introduction

9

The diffusion framework also makes clear why governments – elected or authoritarian, do not provide the necessary kind of contestation to correct for any initial failures in the substantive content of a rule. The attention of governments is largely centred on the strategic issues of whether or not to accept the proposed rules. These strategic interests, such as which other governments are likely to accept the rule and to what extent they can be counted on to implement any agreement, are different in their focus from the type of challenge and contest needed in arriving at the substance of the proposed rules.

DIAGNOSING THE DEMOCRATIC DEFICIT The two frameworks also lead to very different accounts of the sources of the democratic deficit in international rule making. For the architects of the post war international institutions, the problem for democracies was the simple absence of international rules of behaviour. The problem is different now that international rule making has been restored. According to the multi-level governance framework, democratic rule making involves a combination of traditional forms of political authority, organised vertically, with newer forms of authority organised horizontally. Governments with their traditional instruments of authority can act simultaneously at more than one level of decision taking. They can incorporate the preferences of their own voters at whatever level they choose to act – whether international, national or sub-national. At the same time the networks of the many other actors, in addition to governments, who are involved in rule making at all levels of government provide a form of decentralised governance. These horizontally organised forms of authority provide for diversified voices in policy making, including the voice of civil society. A democratic deficit occurs when the connections between these different forms of authority fail to respect such democratic norms as transparency, representativeness and participation. The diffusion framework provides a quite different analysis of the democratic deficit that centres on ‘role reversal’, ‘dissonance’ and the treatment of dissent. ‘Role reversal’ focuses on the way in which citizens experience international rules at the receiving end. In democratic theory the citizen is the ‘principal’, the instigator of collective rule making, while experts and governments are their agents and representatives. In practice, in international rule making the roles are reversed. In today’s world of international rule making, experts propose, governments dispose and the citizen is impotent. Citizens are placed in the position of being the accepters of rules made by


15/12/2010 08:09

10


others they may not be able to identify, in places they may not be able to trace. What ‘dissonance’ refers to in this context is the reasoning that citizens apply to the decisions that are presented to them by experts and governments. The kind of reasoning that citizens engage in as accepters of rules is not necessarily the same as the kind of reasoning that is used by those experts who decide on the substance of the rules. Neither is it the same as the kind of strategic reasoning that lies behind the decisions of governments to endorse the rules. Nor is it just about the effectiveness of the proposed rule. Experts talk about the substantive content of the rule and governments offer narratives in order to frame their endorsement within a context that they hope their electorates will accept and comply with. But electorates may have their own and sometimes dissenting views about whether or not a rule is good for them themselves. For democracies a problem arises whenever the much more diverse styles of reasoning, sentiment and intuition that electorates draw on in assessing the impact of international rules on their lives run counter to the different types of reasoning found convincing by experts and governments. Such a divergence will not always happen, but when it occurs it raises the basic issue of how dissent is treated in democracies. The way in which dissent to proposed international rules can be accommodated within a democratic framework is thus seen as the fundamental issue according to the diffusion framework.

‘LEAST STANDARDS’ The criteria that lie behind and emerge from the diagnosis of the sources of failure and the roots of the democratic deficit incorporate minimum standards or the ‘least standard’ in each case. The least that can be expected from experts is that they respect the evidence. They must therefore observe epistemic standards that are about respect for evidence in the way in which they frame, interpret and analyse policy problems and in the way in which they make recommendations about appropriate solutions. The least that can be expected from democratic systems of government is that they give opportunities for the expression of dissent and the voicing of minority views. A critical dividing line between democracies and authoritarian governments is that authoritarian regimes suppress dissent. Democracies may differ in the way they treat dissent – majorities may be allowed to override minorities or, alternatively, democracies may offer some form of power sharing or constitutional protection. But, in one way or another, provision that respects the place for dissent is the distinguishing feature of democracies.


15/12/2010 08:09

Introduction

11

There are more expansive and demanding standards that could be applied both to experts and to democratic systems. For example, experts might be subject to rules governing conflicts of interest or remuneration or funding. Similarly, higher standards might be applied to democracies, such as criteria for representation or criteria addressing the substantive goals of democratic politics such as social justice. But higher standards require their own further justification. Moreover, if adopted as part of the starting framework they risk slanting the entire subsequent analysis.13

REMEDIES The diffusion framework not only leads to a different diagnosis for the failures of international rule makers and for what lies at the root of the democratic deficit, it also offers a different perspective on the potential remedies. The need to give central place to how dissent is to be treated in any democratic system of international rule making focuses the debate on the relative merits of participatory politics as the vehicle for ‘socialising’ differences in comparison to the rival attractions of governing rules that protect different voices. According to the first approach, dissent marks out the international policies that have to be taken out of expert and government dominated arenas and brought into the public political arena. According to the second approach, dissent marks the areas where different discourses need their own autonomous space. The first approach calls for some kind of political transformation at the international level. The second approach calls for comprehensive principles, grounded independently of the political process, that can mediate relationships between overlapping jurisdictions including the international. The diffusion framework leads eventually however towards a third approach to the treatment of dissent and that is to introduce specific governing rules.. A mandatory period for reflection, debate and review after the conclusion of any international agreement, in whatever form the agreement takes, and before formal ratification, would provide citizens with new space for consultation, challenge and review, including judicial review. The same approach applies to remedying the epistemic vulnerability of the rule makers. In order to respect the evidence, the experts need to operate within a formalised system of challenge that applies throughout

13

Moravcsik (2004) warns against unrealistic or ‘utopian’ standards.


15/12/2010 08:09

12


the regulatory cycle so as to ensure that their treatment of the evidence has not simply been validated by peer review and consensus. The challenge system must always make explicit space for competing views about the problem, for tracking the key decision points where choice is narrowed down, for indicating the uncertainties attached to any decision and for auditing the causal thinking behind the decisions or conclusions. Both types of governing rule need to be externally enforced. Rather than respecting signs of dissent, governments are all too often content to obfuscate the sources and origins of the rules and rely on their powers of coercion to impose the rules. Public opinion is treated at best as something to be manipulated and at worst as an inconvenience to be ignored. Rather than exposing the way they frame problems and ground their preferred solutions, experts also prefer to operate without systematic challenge within a more comfortable world marked by peer approval. The public is asked to rely on the self-restraint of governments and the self-discipline of elites. It is not enough. Externally imposed governing or constitutional rules are needed.

TRADE-OFFS? There is a very traditional view that democratic forms of government involve compromising with effectiveness – that effectiveness and democracy do not go hand in hand. The same point is sometimes made in connection with international rule making – that a system that achieves effectiveness does it necessarily at the expense of democratic values.14 The diffusion framework questions this presumption. It suggests that there is a different connection between democratic values and the effectiveness of international systems of governance. The connection is that the foundation for effectiveness in rule making and the foundation for democratic procedures both require governing rules that entrench opportunities for organised challenge.

SPECIFIC GOVERNING RULES A number of influences are coming together that call for renewal of the architecture for international rule making. The failures in respect of international financial regulation have prompted calls for a ‘New Bretton

14 Dahl (1994) offers one account of the tension between effectiveness and democratic values.


15/12/2010 08:09

Introduction

13

Woods’ – a reference to the conference in 1944 that established the post war rules of financial behaviour and the International Monetary Fund (IMF) and the World Bank. The rise of India, China and other centres, including in the Middle East, also suggests the need to rebalance existing arrangements that were drawn up 70 years ago predominantly by just two powers – the US and the UK – with the US in the driver’s seat. There are new scientific challenges, new security challenges. A necessary precondition for the renewal of the international order is however the need to rebuild confidence in international rule-making procedures. The principal thrust of the analysis in this book is that the sources of the systemic failures in international rule making that we see today, and the sources of the democratic deficit as well, are symptoms of a failure to give proper weight to processes relative to the time and attention devoted to form. This means that the two key deficiencies of the existing system – its proneness to the epistemic failings of elites and the lack of respect paid by governments to dissent – have to be corrected. An approach that targets specific procedural weaknesses with specific governing rules seems to lack ambition. Certainly, it is much less ambitious than approaches that call for some kind of political transformation of international rule making or that call for the application of overarching general principles to mediate different international discourses. But while the approach recommended in this book is limited and specific it can also be seen as the first step in a much broader, longer term agenda. It would provide a first step towards bringing constitutional disciplines to international rule making – to experts and elites as well as to governments.

THE DISCUSSION The first chapter sets the historical scene and briefly accounts for how we arrived at the present arrangements for international rule making. It describes a collision between the assumptions on which the international architecture for the post war world was based with the post war reality. The architecture was built on false premises. The collision with reality created a compelling imperative to improvise in international rule making. With improvisation came a neglect of normative standards. At the same time, in both the construction of the original architecture and in the improvised changes that followed, the emphasis was on form. The chapter also describes the start of the debate in the 1940s about the role of processes in relation to form. The second chapter sets the scene in a different way by reviewing the international rule-making structures as they now operate. It compares


15/12/2010 08:09

14


two styles of rule making. In practice, both are intermingled. It is the resulting mix of venues and instruments that characterises contemporary international rule making. It is in this mix that rule making has failed and democratic norms appear to have been lost. The chapter also looks briefly at whether the current mix of institutions and instruments involved in international rule making is stable or whether a new universal organising logic is likely to emerge – possibly built around a combination or ‘hybrid’ of old and new styles of rule making. The analysis draws on two different types of theorising – the transaction costs approach of economists and congruence theory from political science. They suggest continuing flux and the use of differing mixes of venues and instruments. They also suggest that the criterion for judging the mix should not be formulated in terms of a goal of stability but in terms of a system for ‘managing strain’. Against this description of the present scene and how we got there, the third chapter explores the two different frameworks for analysing the nature of the shortcomings in international rule making. The key elements in the prevailing approach of ‘multi-level governance’ that focus on forms of authority are set out. So too are the key elements in the process-oriented framework drawn from diffusion theory. Both are deployed in the subsequent analysis. In order to carry forward the analysis of the potential sources of failure in international rule making and the sources of the democratic deficit, the next two chapters (Chapters 4 and 5) examine more thoroughly the factors that determine the mix of institutions and the mix of instruments used in international rule making. In the case of the mix of institutions, explanations that focus on form identify the different logic behind different forms, why one kind of venue might be preferred to another and how they come to be combined. Explanations that centre on processes look to a common world of expert elites. In this world, the mix reflects fluid boundaries between professional disciplines and shifting relationships from cooperation to competition. In the case of the mix of instruments used in international rule making, an account that emphasises form attributes the use of a wider mix of instruments to the need to achieve flexibility in the scale or reach of government in different policy areas and to the widening participation of non-governmental actors with authority in policy making. An account that focuses on process looks at the different kinds of reasoning used by governments in their approaches to negotiating an international rule with other governments and in negotiating implementation with the targets of the rules. The mix of instruments arises as a result of the need to link together the two settings and the two styles of reasoning.


15/12/2010 08:09

Introduction

15

Chapter 6 evaluates the sources of failure in the current system of international rule making identified in the earlier discussion of venues and instruments. It distinguishes between three possible sources of failure. One possible source is ‘executive failure’. Accounts of rule making that focus on form identify types of failure that fall within this category. A second possible source of failure stems from shortcomings in what is labelled ‘organisational culture’. A third source lies in cognitive failings. The chapter discusses these distinctions and the relationship between these three sources of failure. It sets out the particular characteristics defining epistemic elites that make them so vulnerable to particular types of cognitive failings. Chapter 7 discusses the democratic deficit in the light of the earlier discussion. It looks at three ways of viewing the underlying issues. The first looks at the issues raised by the need for ‘conciliation’ between the different styles of reasoning used by different actors at different stages of the rule-making process and the different rationalities involved in rule making. Conciliation methods used in international rule making are crude. ‘Muddling through’ and what is known as ‘satisficing’ remain the norms.15 The differences with conciliation processes used within national jurisdictions reintroduce the case for achieving greater congruence between the methods used in international rule making and the methods used in national contexts. Congruence, therefore, provides a second way of viewing the issues. The discussion distinguishes between institutional congruence (where the focus is on looking for analogous institutional arrangements at different levels of government) and value congruence where the emphasis is on aligning values. The third perspective focuses on ‘dissonance’ – the reasons why electorates may reject, for their own good reasons, the reasoning of elites and governments. Dissonance puts the spotlight on how democratic dissent is to be treated in international rule making. The discussion continues by looking at the potential remedies for both the managerial deficit and the democratic deficit. Chapter 8 examines the key features of an institutionalised challenge system that goes beyond peer review in order to ensure that unelected bodies observe more stringent principles and procedures in their rule making. The old international architecture has been pushed to one side partly because it had built in an unproductive kind of contestation. However a new and different kind of contestation remains necessary over the regulatory cycle. Chapter 9 looks at how to treat dissonance and dissent as a foundation

15 ‘Satisficing’ means taking decisions in order to respond to time pressures to ‘do something’ while leaving the underlying issues, including normative issues, unresolved.


15/12/2010 08:09

16


value in a democratic framework. The case for a political transformation of the international level is discussed along with approaches embracing comprehensive governing rules that try to provide for over-arching mediating principles between different types of discourse. The case for the introduction of specific governing or constitutional rules is set out. A brief concluding chapter summarises the analysis.


15/12/2010 08:09

1.

‘How false were our postulates’1

INTRODUCTION At 10.00 a.m. on Sunday December 14, 1941, US Secretary of the Treasury Henry Morgenthau phoned Harry Dexter White, his Director, Monetary Research (and effectively his chief international economic advisor) and asked him to prepare a memorandum and plan for setting up an InterAllied Stabilisation Fund. The fund was to be used during the war to give monetary aid to the allies, to provide the basis for post war international monetary stabilisation arrangements and to provide a post war international currency.2 Meanwhile in the UK Treasury, John Maynard Keynes had prepared his first proposal for an ‘international currency union’ in September 1941 and revised it in November. Thus started serious thinking between the two key players about the architecture for post war international financial and monetary arrangements. One week later, on December 22, US Secretary of State Cordell Hull sent a memorandum to President Roosevelt proposing the formation of an Advisory Committee on Post-War Foreign Policy that would work ‘In the inescapably interrelated fields of general security, limitation of armaments, sound international economic relationships, and other phases of international cooperation, the implementation of which is essential to securing world peace and to economic progress’.3 Roosevelt approved the proposal ‘on, or around’ December 28. In his memoirs Hull noted, ‘We were now launched on our post war work with a definite and prominent committee organized with Presidential authority’.4 It was this group that

1

Acheson (1969: 923). US archives. Unless otherwise stated, references to the US archives are to the general records of the US Treasury and State Department from the end of 1941 to the summer of 1945, including the weekly reports of the State Department to the President and also the daily reports from the Secretary of State to the President during the San Francisco conference and in particular to the chronological files of Harry Dexter White (boxes 8–14). 3 See U.S. Department of State (1950: 63–64). 4 Hull (1948: 1633). 2

17


15/12/2010 08:09

18


led the preparatory work for what was to become the United Nations organisation.5 In the Treasury, White moved quickly. He prepared a first statement of principles by January 6, 1942 that was discussed informally at a meeting in Rio de Janeiro by him and Sumner Welles with Latin American ministers at a Hemispheric meeting later that month. From the outset there was a debate between the US Treasury and State Department as to which countries should be included in planning for the post war. In practice, as far as post war trade, financial and monetary arrangements were concerned the debate was almost exclusively between the US and the UK. A similarly close relationship between the two countries prevailed in respect of post war international security arrangements where, however, the planning group for what became the United Nations organisation also included Russia and (less actively) China.6 The period of design and preparation for the post war architecture lasted until the summer of 1945 (when the San Francisco conference that drew up the United Nations charter was successfully concluded) or slightly longer if detailed proposals for trade arrangements are included.7 Three features stand out about the preparation from this distance of time. The first is the brevity of the design period – just under four years – in circumstances where reflection had to be interspersed between the many other more pressing operational imperatives of winning the war. The second feature was the dominance of the thinking of just two countries – the US and the UK. This was partly because of the closeness of the war-time alliance.8 It was also partly because what came to be known 5 Sumner Welles (Roosevelt’s closest aide at the State Department until his enforced resignation in September 1943) chaired the first meeting on February 12, 1942. His group prepared a report for a ‘United Nations Authority’ in April 1942 but it was put aside by Roosevelt until planning was once again authorised in October 1942 (see S. Welles 1951 and B. Welles 1997). 6 In a memorandum for the President dated December 19, 1944, the Secretary of State noted that, ‘In our view, there is no hope of turning victory into enduring peace unless the real interests of the United States, the British Commonwealth, the Soviet Union and China are harmonised and unless they agree and act together’ (US archives). 7 A detailed proposal on an international trade organisation was not prepared until November 1945 and negotiations on international trade arrangements continued until 1948. 8 A secret aide-memoire on post war arrangements from the British government dated August 4, 1943 and agreed to by the US stated, ‘His Majesty’s Government continues as in the past to regard as of the greatest importance the attainment of prior agreement on such matters between the United States and United Kingdom governments before they are discussed in a wide international field’. The desire for this close relationship was not one sided. A State Department memo to the President dated January15, 1945, stated ‘Unless we and the British get together urgently, while the political and trade situation is favourable, on adequate measures in the field of commercial policy (including cartels and commodity arrangements) there is a grave danger that our whole foreign economic program may be undermined’ (US archives).


15/12/2010 08:09

‘How false were our postulates’

19

as ‘the Keynes plan’ and ‘the White Plan’ were the only alternatives on the table in respect of financial, economic and trading principles. There was no time to debate additional alternatives. More prosaically, the dominance was also partly a result of the logistical difficulties of bringing together a broader representation of international technical expertise during the war. A meeting of technical experts representing 19 countries was held at Atlantic City shortly before the 1944 Bretton Woods conference that established the post war monetary rules and the two key institutions, the World Bank and the IMF, but this did not change the underlying reality.9 In the case of the United Nations, the preparatory conference was limited to the ‘Big Four’ at Dunbarton Oaks. It too settled most issues of principle, with the important exception of voting arrangements in the Security Council, before the convening of the San Francisco meeting.10 The third feature was the extraordinary ambition of the officials in the US and the UK. An international police force to enforce the decisions of the UN Security Council, an international currency, a global central bank, an international commodity stabilisation organisation, a food and agriculture organisation, an international anti-cartel organisation and an international full employment organisation, were all seriously discussed along with the principles and rules to govern trade, monetary and financial behaviour and the more general conduct of relations between nations.11 The international sharing of scientific information was later added too.12 Most of the bodies discussed came into being, albeit in modified form. 9 Professor Raymond Mikesell, then a key member of White’s team, commented later in a memoir that Bretton Woods was only a ‘drafting meeting’ and that, ‘the commissions and committees at Bretton Woods presented a façade of democratic procedure, but the outcome had been largely predetermined by the US and UK delegations’ (Mikesell 1994: 34). In support of this view, the instructions to the American delegates to Bretton Woods dated June 8, 1944, read, ‘Modification of the joint statement of principles may be made by the head of the delegation . . . if they do not fundamentally alter the principles’ [sic] (US archives). 10 Anglo-American cooperation on post war security arrangements was equally as close as on monetary arrangements despite the ruffling of feathers by Churchill on such matters as UN trusteeship arrangements where the State Department note dated February 9, 1945 on the 6th formal meeting at Yalta recorded ‘the Prime Minister interrupted with great vigor to say that he did not agree with one single word of the report on trusteeship . . . He said that under no circumstances would he ever consent to forty or fifty nations thrusting interfering fingers into the life’s existence of the British Empire. As long as he was Prime Minister, he would never yield one scrap of this heritage. He continued in this vein for some minutes’ (US archives). 11 Mikesell comments in reference to the Bretton Woods institutions, ‘One is astounded, however, by the number of powers and functions initially included for these organisations’ (Mikesell 1994: 7). 12 The possible high point of ambitions for after the war was expressed in the Agreed Declaration by the President of the United States, the Prime Minister of the United Kingdom and the Prime Minister of Canada in November 1945 that proposed setting up a United Nations body for the control of atomic science. ‘We believe that the fruits of scientific


15/12/2010 08:09

20


The fundamental weakness in the design for the post war world was not however a result of haste, nor of UK and US dominance, nor of lack of diversified technical expertise, nor of over-reaching ambition. It came instead from erroneous assumptions, shared by both the British and the Americans, about likely conditions after the war’s end. This chapter discusses those assumptions and the improvisations that ensued as they were shown to be false. The overturning of those assumptions and the reactions to a different world, provide an essential background to an understanding of the possible sources of future failings in rule making and to the reasons for a failure to respect democratic values.

THE FOUNDING ASSUMPTIONS The design for peacetime was built around three key assumptions. The first was the expectation that all countries would conclude from the experience of the 1930s that ‘go it alone’ policies had become impossible and would accept their interdependence in ‘one world’. The second was that the political breakdowns of the 1930s and the road to war were rooted in economic conditions and that if those economic conditions were addressed, peace and democracy would follow. The third assumption was that both Roosevelt’s New Deal and the successes of war time mobilisation had demonstrated that governments could manage markets and plan effectively. There was a belief among both the politicians and the technocrats that this experience could be extended and carried forward to manage international problems in the post war period. ‘One World’ Realism American foreign policy has long oscillated between a visionary idealism and isolationism. It is a history of which the American leaders in the Second World War were acutely aware. In particular they were scarred by the isolationism that had prevented America from entering the League of Nations following the First World War. They needed to find new clothes for continued American post war international involvement and leadership that appealed also to American self-interest. They found the clothing in a new kind of ‘realism’ that emphasised the inter-connectedness of countries and the impossibility of American disengagement. research should be made available to all nations, and that freedom of investigation and free interchange of ideas are essential to the progress of knowledge’ (see Dennett and Turner 1948: 547).


15/12/2010 08:09


21

The new realism was most forcefully expressed by Wendell Willkie (Roosevelt’s republican opponent for the Presidency) in a fierce riposte to the isolationists in his party. In One World written in 1943, he proclaimed, ‘when I say that peace must be planned on a world basis, I mean quite literally that it must embrace the earth’.13 In An American Program, written a year later, he appealed to the interdependence of countries. ‘We are not living in several worlds . . . Whatever we do at home constitutes foreign policy. And whatever we do abroad constitutes domestic policy. This is the great new political fact’.14 Cordell Hull, whose ‘non-partisan’ approach to post war arrangements was instrumental for obtaining congressional approval for US participation in the UN, was also fiercely opposed to any return to American isolationism. As he phrased it in his memoirs, ‘The world is with us, here and now, and all about us. It is in our front yard, in our back yard, at our side porch’.15 The same message was presented by Henry Morgenthau in his opening address to the 1944 Bretton Woods conference – ‘The transcendent fact of contemporary life is this . . . that the world is a community’. It was a view also fully endorsed by Roosevelt’s successor, President Harry S. Truman in his State of the Union address in January 1946: Advances in science, in communication, in transportation have compressed the world into a community. The economic and political health of each member of the world community bears directly on the economic and political health of each other member . . . Our ultimate security requires more than a process of consultation and compromise. It requires that we begin now to develop the United Nations Organisation as the representative of the world as one society.16

In practical terms, ‘one world’ realism meant that America looked to the new international institutions to include all countries in the world in their membership (other than the defeated powers). Roosevelt also looked to the ‘Big Four’ (the United States, the United Kingdom, Russia and China) to continue their cooperation in the post war world, through their informal meetings as well as through their permanent membership in the proposed UN Security Council. In his view they would be able to continue to exert leadership in world affairs and to act as the world’s police.17 13

Willkie (1943: 146). Willkie (1944: 22). 15 Hull (1948: 1739). 16 Dennett and Turner (1948: 14). Quoted also in Koenig (1956: 267). 17 Hull recorded that ‘The President favoured a four-power establishment that would police the world with the forces of the United States, Britain, Russia and China. All other nations, including France, were to be disarmed. He felt that the four nations had functioned well during the war, and he wanted this relationship to continue’ (Hull 1948: 1642). 14


15/12/2010 08:09

22


The Interconnection Between the Economic and the Political A second intellectual foundation for the international framework was a belief in a strong causal relationship between economic and social advance and the restoration and nurturing of democracy. From the outset of their cooperation, both the United States and the United Kingdom were agreed on the premise that the architecture could not just focus on political differences between nations but also had to address economic and social conditions as well.18 Hull expressed the view in classic free trading or ‘Cobdenite’ terms: ‘Ages of civilization have taught us that international commerce promotes material welfare, peace and advancement’.19 The more general view, on both sides of the Atlantic, paid greater attention to the experience of the 1930s. Full employment was seen as a condition for returning to the ideal of a multilateral trading system and freedom from want as a precondition for democracy to flourish. Roosevelt himself spoke forcefully in favour of an integrated approach to post war international arrangements. Roosevelt’s Under Secretary and later Secretary of State, Edward Stettinius also emphasised how one part of the institutional design was related to the other and to the whole.20 Managerialism A third pillar for post war arrangements was provided by a huge confidence in the ability of politicians and centralised bureaucrats to manage economies and markets. It is difficult now to recall just how widespread the faith in managed economic and trading arrangements was among those involved in establishing the post war institutions. Some wartime American leaders, including Henry Wallace, saw the future in terms of a coming together of Russian-style planning with the interventionism of the American New Deal.21 It was more commonly a belief spawned by the

18 The Atlantic Charter setting out Anglo-American aims for peacetime and agreed between Churchill and Roosevelt in August 1941 stated in its fifth article that ‘They desire to bring about the fullest collaboration between all nations in the economic field with the object of securing, for all, improved labor standards, economic advancement and social security’. The eighth and final article contained an oblique reference to a future world organisation for keeping the peace. 19 Hull (1948: 1735). 20 See speech to the Council on Foreign Relations, Chicago April 4, 1945 in Goodrich and Carroll (1947: 32–33). 21 In The Century of the Common Man Wallace wrote ‘Somewhere there is a practical balance between economic and political democracy. Russia and the United States both have been working towards this practical middle ground’ (Wallace 1943: 37). See also Culver and Hyde (2000: 292).


15/12/2010 08:09


23

success of wartime mobilisation techniques. It also reflected the perceived success of Roosevelt’s New Deal, which had involved radical interventions in the market, particularly in agriculture, steered by Henry Wallace.22 In Roosevelt’s message to the joint session of Congress in March 1945 following the Yalta summit he said: I know that the word ‘planning’ is not looked upon with favour in some quarters. In domestic affairs, tragic mistakes have been made by reason of lack of planning; and, on the other hand, many great improvements in living, and many benefits to the human race, have been accomplished as a result of adequate, intelligent planning . . . the same will be true in relations between nations.23

A belief in the virtues of managed markets was shared by the technocrats and notably by both Keynes and White as they sought to avoid a return to the competitive devaluations and beggar-my-neighbour trade policies of the inter war years and looked for a path between free market economics and Soviet-style planning. Treating potentially divisive questions about trade preferences, or sterling balances, or the acute short-term dollar scarcity, as issues that could be solved by technocratic managers also appeared to them as a way to avoid becoming bogged down in financial and economic frictions between the two countries. Agreement on this fundamental point of departure has been obscured by the way in which analysts have focused on the differences between the ‘Keynes Plan’ and the ‘White Plan’ – the differences being given added piquancy by the knowledge now available that White was passing information to the Soviet Union.24 Certainly there were important areas of disagreement. The Keynes plan for a Clearing Union had two fundamental features that did not survive. The first was the ‘banking principle’ that the new international institution could create money and act as the world’s central bank. The second was his proposal for an international currency (Bancor) that could act as more than a unit of account (Unitas), as envisaged in White’s plan, in order to become a genuine medium of exchange. British historians have tended to play up the virtues of the Keynes

22 Before the war, Wallace (1936, 1964) had argued for a ‘Council for the General Welfare’ that would advise the government on how to achieve the general good. After the war he wrote, ‘To win the peace of abundance within the framework of free enterprise requires even more planning than was required to win the war’ (Wallace 1945: 85). 23 Burley attributes a dominating influence to the lessons Roosevelt and other American leaders drew from the New Deal: ‘the domestic origins of the specific contours of the post war international order lie in the historical experience of one liberal state; the United States during the New Deal’ (Burley 1993: 129). 24 For an assessment of the evidence see Craig (2004).


15/12/2010 08:09

24


plan and to make invidious comparisons between Keynes ‘the rapier’ and White ‘the blunderbuss’.25 It is a false comparison. What mainly divided them were the different national interests they had to defend. Keynes was acutely afraid of Britain’s enfeebled economic outlook after the war, its uncertain trading prospects and its need for continuing financial support. White was equally concerned to limit the liabilities of the US in the post war period and for debtor countries to exercise discipline.26 Keynes regarded White as the best of the American officials he had to deal with because he focused on the substance of the issues. In his turn, White was not overawed by the reputation that Keynes carried with him within the British delegation.27 Asked by Roosevelt as to whether Keynes was ‘friendly’, White replied that ‘Keynes was an extremely able and tough negotiator with, of course, a thorough understanding of the problems that confronted us, but when not negotiating or discussing points of difference that he was quite friendly’.28 Keynes was also well aware that the US held all the cards.29 Post war monetary and financial arrangements were White’s rather than Keynes’s. The most significant feature of the relationship was the belief shared by both in the capacity of enlightened officials to manage the complexities of markets. For Keynes, the ‘banking principle’, that he struggled unsuccessfully to establish, reflected a view that a global central bank could act on the world scale in the same way as a central bank acted in a domestic market. Interwar experience had convinced him more generally that free market thinking was fundamentally flawed. ‘To suppose that there exists some smoothly functioning automatic mechanism of adjustment which preserves equilibrium if only we trust to methods of laissez-faire is a doctrinaire delusion which disregards the lessons of historical experience without having behind it the support of sound theory’, he wrote in

25

For example, Skidelsky (2000: 309). See the recollections of Edward M. Bernstein, White’s key lieutenant, as recounted to Black (1991: 39). It was Bernstein whom Keynes is alleged to have referred to as a ‘Ghetto rat’ – a remark that, according to Bernstein, had it been known to the Treasury or to Morgenthau, would have led to the end of discussions with Keynes (ibid.: 39). 27 Keynes did not at that time carry the reputation in the US that he was later to acquire. Salant sees the 1960s as being the high point of Keynes’s reputation in the US and concludes, ‘It is clear that Keynes had no direct influence on policy in the United States and, until perhaps 1938 or 1939, very little indirect influence. His influence later was on the intellectual atmosphere, and there it was immense’ (Salant 1989: 51). Salant also notes that the first American economic textbook written from a Keynesian perspective did not appear until 1947. 28 Memorandum of a conference of the White House, May 18, 1944 (US archives). 29 Bernstein recalled, ‘He [Keynes] knew that in the end we could do as we pleased’ (Black 1991: 39). 26


15/12/2010 08:09


25

an internal UK Treasury memorandum.30 He envisaged his international authority acting together with an ‘Anti-Depression Board’ in order ‘to exercise contractionist or expansionist inflows on the system as a whole’.31 It was probably this thinking that lay behind the discussion in October 1943 between the US and the UK as to ‘whether there shall be instituted an international organisation specifically charged with the duty of promoting and coordinating policies designed to maintain high levels of employment throughout the world’.32 In the event, such a special purpose body was not created on the grounds that full employment would involve the coordination of all aspects of economic policy. But it found expression in the creation of the Economic and Social Council of the UN that was intended to fulfil this over-arching economic and social coordinating function.33 Keynes was deeply committed to the eventual restoration of free trade and unimpeded current account transactions. At the same time he coupled this with a proposal for an international body, ‘Commod Control’, to intervene in commodity markets both to manage scarcities and to stabilise price fluctuations.34 In urging a buffer stock scheme to stabilise prices he also conceded that the regulation of production might be required as well.35 His scheme aimed not only to dampen price fluctuations but also to maintain a long-term equilibrium between supply and demand, ‘at a price which provides to the majority of primary producers a standard of life in reasonable relation to the standards of the countries in which they live.36 ‘Is not centralised international action capable of effecting a vast improvement of the system . . . ?’ wrote Keynes rhetorically.37 White shared Keynes’s belief in the importance of an international financial system that would underpin an expansionary world trading system. At 30

Memorandum dated September 8, 1941 (Keynes 1980: 22). Ibid.: 60. US State Department note of US/UK discussions dated October 16, 1943 (US archives). 33 At the San Francisco conference the powers of the Council were limited by the US desire to avoid giving it the authority to intervene too deeply into the domestic policies of UN members. However, US expectations about the role of the Council in monitoring international employment conditions were reflected in its November 1945 proposals on international trading arrangements. 34 In commenting on criticisms from the Bank of England to his proposals for international agreements on commodity prices and quantities, Keynes wrote, ‘I can only plead guilty of aiming at a plan which does take a middle course between unfettered competition under laisser-faire conditions, and planned controls which try to freeze commerce into a fixed mould’ (Collected Writings, Minute to Sir Richard Hopkins, April 15, 1942; Keynes 1980: 111). 35 Ibid.: 114. 36 Ibid.: 136. 37 Memorandum on the International Control of Raw Materials, 5th Draft, April 15, 1942 (ibid.: 111). 31 32


15/12/2010 08:09

26


the same time he also shared Keynes’s reservations about the merits of free market doctrines. In his one substantial academic study he concluded in cautious academic tones that the neoclassical theory of balance of payments adjustment ‘Explains what happens only under certain given conditions seldom found’.38 Both saw freely floating exchange rates as a danger leading to competitive devaluations and both supported a system of settled exchange rates instead.39 Both Keynes and White also opposed the free movement of long-term capital. White’s study had concluded in even more cautious language, ‘the assumption that capital exports benefit both the lending country and the world at large is not unassailable’ and less cautiously, ‘Some measure of intelligent control of the volume and direction of foreign investments is desirable’.40 In his own first statement of principles, in January 1942, White drew a careful distinction between ‘productive capital’ and other (less desirable) capital movements. In his turn, Keynes wrote, ‘I share the view that central control of capital movements, both inward and outward, should be a permanent feature of the post-war system’.41 Keynes in particular accepted that the need to control long-term capital movements would mean in practice that comprehensive exchange controls would be needed.

IMPROVISATION Assumptions Destroyed Each of these three intellectual building blocks was destroyed in the post war world as events actually unfolded. In his memoirs Present at the Creation, Dean Acheson, Truman’s Secretary of State 1949–53, summed up what he called ‘the struggle through illusion to policy’. He noted ruefully: Our preliminary ideas appeared more and more irrelevant to the developing facts and the attitudes, purposes and capabilities of others on the scene . . . They were founded on a refurbished and strengthened League of Nations, which assumed continued cooperation of the wartime alliance in banishing war and the use of force. Economic arrangements – even the new ideas of Maynard Keynes – were to be brought into conformity with the classical goals of

38

White (1933: 306). IMF par values were deliberately set at rates that would encourage exports (see Black 1991: 61). 40 White (1933: 311). 41 Keynes (1980: 52). 39


15/12/2010 08:09


27

removing obstructions from the free movement of goods, people and funds as means of expanding trade and development. And economic development was to take on an evangelistic character in support of social justice and democratic institutions’.42

The reality, as Acheson quickly realised, was very different – ‘how false were our postulates’, he wrote. The first pillar to crumble was the belief in ‘one world’ realism and the possibility for the Big Four to act as the world’s police. Within one month of President Truman’s January 1946 ‘one society’ speech, George Kennan, acting head of the US embassy in Moscow, dispatched his famous ‘Long Telegram’ warning that the Soviet Union was set on a deadly struggle.43 In March, Winston Churchill, in Fulton, Missouri, spoke of an ‘iron curtain’ that had descended across Europe. By 1950 a policy of ‘political containment’ turned to one of military containment.44 The over-arching vision of a world with international institutions enabling it to solve problems as ‘one society’ had been shattered almost as soon as it had been articulated. The onset of the Cold War and the struggle between two rival visions of how to organise economic and social activities also put paid to the idea of any automatic connection between economic prosperity and democracy. However, the belief in the possibility of centrally managed markets was not subject to the immediate and fatal challenges of the Cold War and took much longer to crumble. Socialist countries continued to put their trust in the virtues of managed systems. At the same time, the Bretton Woods institutions functioned without the membership of communist countries.45 Nevertheless, over the next 30 years there occurred a loss of belief in the ability of centralised bureaucratic managers either to know what to do or how to do it. Market oriented systems had begun to assert 42

Acheson (1969: 922). The long telegram concluded, ‘In summary, we have here a political force committed fanatically to the belief that with US there can be no permanent modus vivendi, that it is desirable and necessary that the internal harmony of our society be disrupted, our traditional way of life be destroyed, the international authority of our state be broken, if Soviet power is to be secure’ (see Kennan 1968: 552). This cable did not come out of the blue. US Ambassador to Moscow, Averell Harriman had warned much earlier in April 1945 that ‘we must clearly recognise that the Soviet program is the establishment of totalitarianism ending personal liberty and democracy as we know it’. His warning was ignored (US archives). 44 National Security Council resolution 68 dated April 14, 1950 set out the containment policy. It recommended that the US, ‘develop a level of military readiness which can be maintained as long as necessary as a deterrent to Soviet aggression, as indispensable support to our political attitude towards the USSR, as a source of encouragement to nations resisting Soviet political aggression, and as an adequate basis for immediate military commitments and for rapid mobilization should war prove unavoidable’ (recommendation (a)). 45 Russia and its satellites in Central and Eastern Europe decided not to join and the Czech republic withdrew its membership. 43


15/12/2010 08:09

28


their superiority over command and control systems of economic management within states. Half-way houses such as ‘indicative planning’ models also lost their appeal. The loss of credibility applied equally to would-be international systems and managers. When the Bretton Woods system of stable but adjustable exchange rates finally collapsed in 1973, the subsequent failure of the Committee on International Monetary Reform to agree on a new framework in 1974 marked the end of grand managerial visions. The immediate cause of the collapse of the negotiations on a new international economic framework was the shock of the oil price rise by the Organization of the Petroleum Exporting Countries (OPEC). However, other factors were at work and became increasingly prominent as the century wore on. Instead of an international economy driven by trade flows combined with capital controls as envisaged by Keynes and White, capital movements and shifts in portfolio preferences of investors came increasingly to be the drivers in the system. Monetarist theory increasingly supplanted Keynesian theory, but the kind of price targeting it espoused allowed asset price bubbles to go unchecked, risk premiums to be dangerously depressed and huge savings imbalances to emerge. The Bretton Woods framework appeared increasingly irrelevant and sometimes even unhelpful. Other kinds of institution were needed in a world driven by private capital flows. Replacing the Assumptions The reaction to the crumbling of the assumptions on which the architects of the post war institutions had built, as Acheson again noted, was to ‘place the strategic approach to practical objectives, concretely and realistically conceived ahead of generalizations, even those wearing the cloak of idealism’.46 In practice ‘the strategic approach to practical objectives’ meant an almost complete disavowal of the assumptions on which the design of the post war architecture had been based. Instead of ‘one world’ realism with its concomitant belief in institutions with universal membership, there was a turn towards setting up, or using, clubs limited to selected ‘like-minded’ members. Instead of an approach that emphasised the inter-connections between economic and social problems and political problems, there was a movement to disaggregate problems and a switch to ‘depoliticised’ venues. Instead of a belief in the possibilities of the central management

46

Acheson (1969: 924).


15/12/2010 08:09


29

of economies, trade and financial markets there came a reliance on the free flow of capital, floating exchange rates and the free flow of goods and services. In addition, the goal of binding international obligations came to be replaced by less prescriptive forms of international understandings.47 In short, the vast ambitions and grand designs of the original architects were replaced by piecemeal improvisation.48 From ‘One World’ Realism to ‘Like-mindedness’ The switch away from ‘one world’ realism and its embrace of a universal approach to the membership of international organisations, came with the establishment of the Organisation for European Economic Co-operation (OEEC) in 1947. Set up to administer Marshall Aid for European reconstruction it bypassed the World Bank, the IMF and the UN Economic Commission for Europe. Although an invitation to discuss Marshall Aid was sent to the Soviet Union it was couched in terms that Russia would reject and OEEC’s membership was restricted to America’s Western European allies.49 The significance of the establishment of OEEC was that it replaced universalism by ‘like-mindedness’. Instead of normative criteria that emphasised maximum inclusiveness for membership in international organisations the emphasis switched to ‘what works’. What was seen to work were organisations whose members shared common interests and a like-minded approach to problems and their solutions. Those who judged ‘what works’ were initially governments. Increasingly the judgements reflected the world of like-minded professional groups. Disconnecting the Political and the Economic The founders of the post war system emphasised the advantages of general purpose institutions with broad agendas because interwar experience had impressed on them the importance of inter-linkages between issues, and because in their view an integrated approach to the solving of economic problems would also serve political ends. In practice, the advent of the Cold War undermined the idea of any straightforward coupling of democratic political values and economic and

47 The general exception to this was a continued search for international commodity agreements in order to stabilise prices and incomes of producers. 48 Even in the case of the UN system, Cox (1996: 526) states, ‘It has more of a segmented than a hierarchically coordinated structure’. 49 See the discussion in Acheson (1969: 306–14) and Kennan (1968).


15/12/2010 08:09

30


social values. Although the atrocities of Stalin had yet to be revealed and the atrocities of Mao Tse Tung had yet to come, communism challenged both the nature of the connection between economic advancement and democracy and the ultimate destination of post war societies. Instead of an integrated approach to problem solving that emphasised the interconnectedness of the issues, the advantages of disaggregating problems and problem solving institutions became quickly apparent. The speedy reversal from a comprehensive approach to a disaggregated approach is best illustrated by the rise of the Bank for International Settlements (BIS) located in Basel. Established in 1930 as a club for central bankers, it was confronted by an agreed recommendation for its abolition at the Bretton Woods conference in 1944 that established the World Bank and IMF with their more sweeping and encompassing agendas.50 In practice, it not only escaped abolition, but by 1950 had re-established its role as a venue for central bankers when it was appointed as agent for the European Payments Union. It became of pivotal importance for international banking with the establishment of the Basel Committee for Banking Regulation and Supervision in 1974. Since 1999 it has provided support for the Financial Stability Forum (FSF; now Board) that provides an umbrella for a number of separate international regulatory bodies dealing not only with banking stability, but also with securities market regulation, insurance supervision, deposit insurance and accounting standards. It is an irony of history that the organisation the architects of the post war architecture wished to eliminate (the BIS) has emerged at the centre of the new global architecture for international finance while the centrality of the institutions they did create (the World Bank and IMF) has faded. The resurgence of the BIS and the turn to specialised bodies for international problem solving not only reflected a turn away from holistic visions of joined-up international policy making served by over-arching institutions but also was indicative of a desire to ‘depoliticise’ international institutions. The post war bodies with a universal membership had a disconcerting ability to bring out political divisions. This was especially true of the United Nations itself. But it also affected those bodies such as the World Bank, the IMF, the General Agreement on Tariffs and Trade/World Trade Organization (GATT/WTO) and the Food and Agriculture Organization (FAO), where divisions between the industrialised economies and developing countries, between rich and poor, and between borrowers and suppliers of resources quickly became established. These political divisions within the universal institutions created

50

For the history of the BIS at Bretton Woods see Tonioli (2005: 267–82).


15/12/2010 08:09


31

incentives to look for alternative venues that favoured technocratic approaches. The BIS had its origin precisely in a desire to depoliticise the issues surrounding German reparations after the First World War and establishing specialised bodies with technocratic aims has been seen as the way to avoid politicisation.51 The Decline of Centralised Management In order to help central managers to act effectively, the preference of the makers of the post war system had been to favour international agreements that were binding for all participants. The issue as seen by the original architects was how to reconcile their desire for binding rules of international behaviour with the kind of national autonomy that both the United States and the United Kingdom took for granted as desirable. The debate about how to reconcile binding international rules of behaviour with national autonomy centred in part around the peacekeeping functions of the United Nations. Here, it was resolved, for the major powers, by the arrangements for permanent members to possess veto powers in the Security Council.52 It also centred around a more arcane discussion about the powers of the IMF and the latitude of countries to make exchange rate changes.53 The failure of the US administration to satisfy the Senate that it had achieved such a reconciliation in the case of the proposed International Trade Organization (ITO) to manage international trading rules led to the defeat of the ITO legislation in Congress. As a result, the post war world operated initially under a ‘temporary’ trade regime (GATT) that has been called ‘a network of simultaneous bilateral

51 Kapstein claimed in 2006, ‘One of the great “successes” of financial supervisors over the past thirty years has been to de-politicize the systemic risk environment and to transform crisis management into a technocratic exercise, thereby making financial shocks somewhat easier to manage, by reducing the number and type of players involved in decision-making’ (Kapstein 2006: 15). 52 In discussing the proposed veto power with his liaison group of Senators, Hull stated that the veto power ‘is in the document primarily on account of the United States. It is a necessary safeguard in dealing with a new and untried world arrangement. Without it, the United States would not have anywhere near the popular support for the postwar organization as with it in, any more so perhaps than in 1920’ (Hull 1948: 1662). 53 For Keynes the freedom for countries to set their own interest rates, given capital controls, was critical in respecting national sovereignty while involving binding rules of behaviour. The issue of discretion versus rules was also critical to the exchange rate regime. In the last resort, if countries had to adjust their exchange rate because of wage/price differentials, the IMF had to give its consent after a consultation process. But, except when conditions of ‘fundamental disequilibrium’ had arisen, members of the IMF undertook a binding obligation to try to maintain stable exchange rates (see Keynes 1980: 149).


15/12/2010 08:09

32


reciprocal agreements’, whose rules reflected American demands.54 GATT subsequently oversaw eight rounds of multilateral tariff and trade negotiations before being converted into the WTO in 1995. The decline in confidence in centralised management eroded the belief that international rules could be set out in a binding and precise way that covered the huge variety of actual circumstances. The incentive has been to look for new and less prescriptive forms of rule making at the global level. The loss of confidence has also encouraged a search for decentralised forms of international organisation with greater involvement by the private sector. The collapse of the original assumptions on which the institutional arrangements for international rule making were based fundamentally shapes the system we see today. The current system now consists of a blend of the original architecture designed during the Second World War together with a multiplicity of newer organisations that reflect the turn towards what were seen to be more practical responses to practical questions – a turn to limited memberships, specialised terms of reference, ‘depoliticised’ settings, and a reliance on declarations of principle as much as on binding obligations. There is a blend of both different types of organisation and different types of instrument. This history of assumptions overturned and of a brief period of design replaced by a long period of improvisation provides reason enough why the international system of rule making might be prone to failure and to overlook democratic values. But, in addition, the history also provides an important insight into where, more precisely, the thinking may have gone astray.

FORM AND PROCESS The sweeping ambitions of the architects of the post war system were criticised at the time by, among others, David Mitrany, a political scientist, now best known as an advocate of a ‘functional’ approach to post war arrangements and a critic then of America’s ‘two sidedness’ and ‘sentimental attitude’ to democratic principles.55 He had two fundamental criticisms. First, he questioned whether democracy and the rule of law based on the nation state were consistent with the idea of a universal rule of law and democratisation at the international level. Second, he ques-

54 55

See the discussion in Goldstein (1998). Mitrany (1946).


15/12/2010 08:09


33

tioned the link between the effectiveness of arrangements for international cooperation and democratic values. He saw a ‘trade-off’ between the two. Both questions continue to resonate in contemporary debates. The Rule of Law Even Henry Wallace, the most visionary of the American wartime leaders, was aware of the potential for conflict between democratic values within states and the international rule-making system the allies were trying to establish in the post war world.56 However, the founding fathers of the post war international system thought they could avert any such conflict. By including all countries in international institutions, by carefully calibrating the scope of the authority of international institutions and by pursuing an integrated agenda where prosperity would underpin democracy, they believed that they were pursuing a consistent set of goals. Mitrany was a long-standing internationalist. Where he diverged from the architects was in seeing a basic conflict between national governments based on territory and an international rule-based order. His fundamental aim was for an international system that would break the link between authority and territory.57 The post war history of the spread of democracy seems at first sight to refute the conflict seen by Mitrany. In the first decade of the twenty-first century more people are living in democracies than at any previous point in human history.58 Some important countries, notably China, are exceptions and continue with authoritarian and repressive regimes. It is also the case that many democracies are far from perfect in their practices. Backsliding to more authoritarian politics, or to rule by corrupt and selfserving elites, remains an ever-present risk. Nevertheless, out of a current total world population of around 6.8 billion, roughly two thirds live in countries that can be generously classified as democratic.59 However, the spread of democracy at the level of nation states makes the question of its relationship to democratic values at the international level even more pertinent. If citizens in democratic countries have reason to believe that international rule making is, in some way, in conflict with 56 Discussing the role of the proposed United Nations organisation he wrote, ‘the aim would be the maximum of home rule that can be maintained along with the minimum of centralized authority that must come into existence to give the necessary protection’ (Wallace 1942). 57 Mitrany (1943, 1975). 58 See the discussion in Markoff (2009). 59 Estimate based on countries that Freedom House estimates were ‘free’ or ‘partly free’ in 2009. According to the same source 2.3 billion people live in countries that are ‘not free’.


15/12/2010 08:09

34


the democratic values they enjoy in their own country then there is a valid normative reason to resist the growth of international rule making. It also suggests that there is a choice to be made. Either democratic countries need to be able to insist on the primacy of their own democratically made laws, rules and regulations, or there is a need to look for ways of transforming the international rule-making system so that democratic values are reflected at the international level itself. Effectiveness and Democracy Mitrany’s belief in the inherent conflict between national orders of government and an international order also led him to criticise the desire of the architects to promote democracy as part and parcel of the thinking behind the new institutions. He wanted to decouple international cooperation from the support of democracy. He advocated instead a disaggregated ‘functional’ approach with a focus on ‘specific practical arrangements’ that would eventually lead to more generalised international cooperation and democratic government at the international level. This aspect of Mitrany’s critique of the architecture appears to have been vindicated.60 Disaggregation of problems and problem solving has become the international norm. Moreover, notwithstanding the 2008 international financial crisis, it has been a generally successful approach. On the economic front, unprecedented levels of economic prosperity have been reached, based on a massive expansion of international trade and capital flows and huge national investments in education and health. The focus has shifted from the perils of unilateral beggar-thy-neighbour policies to fears about the implications of ever-closer interconnections between countries – the fear of ‘globalisation’. With the exception of oil, attention has shifted from the economics of scarcities to fears of ‘excess’ – the implications of growing world production and consumption for the biosphere. Moreover, Mitrany’s functionalist perspective has been widely adopted in looking at the development of regional cooperation arrangements in Europe (which ironically he opposed because he favoured global cooperation). In Europe an initial focus on limited and specific practical

60 Not all of the aspects of Mitrany’s brand of functionalism have been vindicated by subsequent developments. For example he accepted the assumption that international institutions should aim for universal membership of all countries. He also accepted the prevailing belief in the ability of central managers to manage international markets through binding rules. He thus favoured providing international institutions with supra-national executive powers.


15/12/2010 08:09


35

arrangements for coal and steel has led progressively to the wider goal of a Common Market, the more ambitious harmonisation of the single market to the kind of political union reflected in the Lisbon Treaty. Process The basic questions raised by Mitrany about the relationship between democracy organised at the national level and policy making organised at the international level, and about the relationship between the objective of effectiveness and the objective of democracy in international rule making remain at the centre of current debates about international rule making. However, Mitrany’s own prescriptions and assertions about the relationships do not necessarily have to be accepted. Even in Europe the functionalist approach has not led to a smooth transition from an economic union to a democratic political union. On the contrary, a technocratic approach to building the Union has arguably been at the root of its problems in the transition to a democratic polity. What however is of lasting value in Mitrany’s critique is the fundamental importance of thinking about processes. The original architects had focused single-mindedly on the form and scope of authority of the institutions they were setting up. They paid much less attention to process. Mitrany suggested that process was vital both for the effectiveness of the arrangements and for the attainment of democratic values. In the analysis that follows a framework is used that gives equal weight to process and to form. Contrary to Mitrany’s assertions, the analysis does not suggest that there is an inherent conflict between democracy and the rule of law at the national level and the international level. Nor is there any necessary trade-off between the effectiveness of international rule making and democratic values. Before looking at the analytic frameworks suitable for exploring the relationship between form and process the next chapter looks in greater detail at the contemporary mix of institutions and instruments used in current styles of international rule making.


15/12/2010 08:09

2.

Managing strain – styles of international rule making

International rules now emanate from a bewildering array of institutions and in an equally bewildering variety of forms.1 The most important venues are often small and obscure. The source of rules that affect the well-being of individuals and communities can be traced to their origin only with difficulty by anyone other than the cognoscenti. Bodies such as the Financial Action Task Force on Money Laundering (FATF), the Nuclear Suppliers Group (NSG), or even the Intergovernmental Panel on Climate Change (IPCC) are not household names, but their actions range from preventing the financing of terrorism to preventing the export of equipment that could be used for the production or delivery of nuclear weapons, and to assessing whether and to what extent human activities might contribute to climate change. In addition, the form in which rules are transmitted from the international level where they are conceived, to the national and local level where they are applied, is equally diverse and bewildering. Moreover, in their passage from the international venue where they have been agreed, to the regional, national and local levels where they are applied, many of the rules change their nature, becoming precise where they were flexible, binding where they were exhorting ‘best practice’. It is a world that is comprehensible only to experts and specialists. In order to start describing the present state of international rule making, this chapter first compares two ‘ideal’ styles of international rule making (they are ‘ideal’ only in the sense of being simplified and abstract). One model centres on bodies with a universal membership whose form and functions rest on international treaties. In the discussion below it is labelled the ‘classic’ model because it reflects the kind of body that the original architects of the post war system envisaged. The other model is network based and relies on much less formal rule-making techniques.

1 Jepperson (1991: 145) defines an institution as ‘a social order or pattern that has attained a certain state or property’. He distinguishes between three carriers – a formal organisation (the sense used in this chapter) a regime of codified rules and a culture of customary rules.

36


15/12/2010 08:09


37

It reflects the kind of improvised venues that have grown up around the original architecture. It is labelled the ‘new’ model.2 The chapter turns secondly to look at the mix of rule making that results from the mingling of these two different styles of rule making. In practice the division between what belongs to the ‘classic’ model and what belongs to the ‘new’ is blurred. In the real world, contemporary rule making involves both old and new venues, both old and new techniques. Coordination mechanisms are described. But the analysis cautions against reading into the system any over-arching organisational logic. The chapter looks thirdly at the issue of the stability of the current mix of venues and instruments.3 Power relationships in the world are shifting and this is stimulating institutional change and bringing new coordinating venues into prominence. The question is whether this change could see the emergence of a new kind of ‘hybrid’ international institution that combines features of the ‘classic’ and the ‘new’ in ways that auger increasing stability for the future and a possible new consensus about organising principles. The question of stability is examined from two different perspectives. The first looks at the different transaction costs associated with the different styles of international rule making. The second looks at the congruence of international institutions with more familiar styles of organisation. Both perspectives suggest that international rule making will continue to make use of a changing and unstable mix of venues and instruments. Taken together these two approaches also suggest a different way of looking at the present structure of international rule making. The test is not stability, nor conformity to an ‘ideal’ form nor convergence on new hybrid forms. Instead, we should look on the arrangements as a fluid search for ways of ‘managing strain’ in international relationships.

TWO MODELS Contemporary descriptions of structures for international rule making depict two worlds. One is a world inhabited by a ‘classic’ style of

2 Koenig-Archibugi (2002 and 2005) suggests making institutional distinctions based on degrees of ‘publicness’, delegation and inclusiveness. Rosenau (2002) suggests a processoriented classification based on six distinctions between top-down, network, bottom-up, side-by-side, market governance and ‘Mobius’-web governance. Both alternatives add to measurement difficulties and to complexity without offering compensating advantages. 3 The term ‘instrument’ refers to the form in which a rule or regulation is conveyed – for example as a treaty or as a recommendation. They are sometimes graded on a scale of ‘soft’ to ‘hard’ depending in part on whether they are legally binding or not. Definitions and grading are refined and discussed further in Chapter 5.


15/12/2010 08:09

38


international organisation of the type established at the end of the Second World War. The other is a world inhabited by a ‘new’ style of organisation – the network organisation – allegedly more suited to the modern world. Both can be distinguished as ‘ideal types’. The reality however is that ‘pure’ types of either model cannot be seen.4 The ‘Classic’ Model The classic model of an international rule-making body has four defining features.5 First, it aims for universal membership, drawing in all countries in the world. Second, the organisation is set up as a ‘permanent’ body, intended to last for the indefinite future. Third, its terms of reference are comprehensive in their field. Fourth, the mode of operation is prescriptive – that is to say that countries joining the organisation agree to accept obligations that are binding on them. 6 Bodies that fit broadly within these criteria are shown in Table 2.1, together with their dates of establishment. The oldest date back to the late nineteenth century and the surge of new bodies following the Second World War can be clearly seen. Also visible is the fall in the number of new bodies created of this type. The decline in the creation of new bodies can be seen as a reflection of the occupancy of many of the fields of international public policy. But, as discussed below, it also reflects the fact that by the end of the period new venues were being set up for international rule making that were not of this traditional style. The Key Features The four key features of the classic model adopted since 1940 were rooted in the lessons that the architects drew from the interwar experience. The goal of universal membership in international organisations in the post second world war reflected a determination to avoid the failings of the League of Nations, notably the absence of the US as well as a normative principle (discussed later) of inclusiveness. In 1944/45 universal

4 Ahrne and Brunsson (2011) distinguish between a ‘true’ organisation based on membership, hierarchy, rules, monitoring and sanctions, a ‘genuine’ network that lacks these elements and an institution that regularises some elements of social interaction in a stable form. According to this terminology this chapter distinguishes between two types of institution or what is later referred to as ‘partial organisations’. 5 Peverhouse et al. (2005) suggest that the broadest and simplest definition of an international organisation is a body that is a formal entity, has states as members and possesses a permanent secretariat. 6 These distinguishing features are similar to those used by Aggarwal and Dupont (2005).


15/12/2010 08:09


Table 2.1

39

Selected ‘classic’ international organisations (by date of establishment)

Pre 1914

1915–39

1940–59

1960–80

1981–

ITU WMO UPU WIPO

ILO ICJ UNWTO OIE

UN IBRD IMF FAO UNESCO UNICEF ICAO IMO WHO WTO UNHCR IAEA

UNCTAD UNIDO

ICC*

UNEP IFAD

Notes: WMO, WIPO, ICJ, UNWTO, WTO listed by date of establishment of predecessor organisation. For details on acronyms see the list on pp. ix–xi. *USA not a member of ICC.

membership meant that about 45 states were involved in the membership negotiations for the main post war organisations. With the explosion since the Second World War in the number of states in the world, membership in these bodies now comprises between 150 and 200 states.7 The post war institutions were envisaged as part of a permanent infrastructure for international rule making because the intent was to show that the founding members were committed to an enduring system of international governance and not just to a system to settle immediate post war problems. Thus the institutions were able to offer long-term employment possibilities, analogous to those offered to civil servants in a domestic context, were provided with international immunities and privileges, and were equipped with an executive head. Periodic ministerial meetings representing the entire membership were intended to set and review strategic goals while smaller executive boards or councils were established to provide oversight on behalf of the member states on a more frequent basis.8 The comprehensive terms of reference of the main institutions were

7

As at mid-2010 there were 192 members of the UN. In the case of UNESCO the original Executive Board consisted of individuals elected by the membership for their distinction in the organisation’s substantive fields. This was changed in 1954 so that the Board would consist of government representatives. 8


15/12/2010 08:09

40


also a reflection not only of the ambitions of the founders of the post war system but also of their commitment to a long-term vision. The classic international organisation came with binding obligations on its members. This feature was also as a reaction to the interwar period. It was felt that many of the problems leading up to the Second World War could have been averted if binding rules of behaviour had been in place. Thus binding obligations were expressed either in the founding treaties or ‘constitutions’ of the classic organisations and were conditions of membership, or in supporting treaties that were agreed by members concurrently with the founding of the organisation. Among the best known are the exchange rate obligations of members of the IMF under the Bretton Woods system, trade agreements under the GATT/WTO, the Nuclear Test Ban Treaty under the International Atomic Energy Agency (IAEA), the Chicago Convention that accompanied the founding of the International Civil Aviation Organization (ICAO) and the Refugee Convention of the UN High Commissioner for Refugees (UNHCR) that is seen as having helped lay the basis for an international human rights law. Moreover, these binding obligations have been backed by surveillance or monitoring arrangements. Again a leading historical example is provided by the IMF, where the provisions of ‘Article IV’ have provided the basis for surveillance of the exchange rate policies of members since its founding (policies were modified in 1977 and 2007). Other more recent examples include the trade disputes settlement mechanism of WTO, the nuclear facilities inspection system of IAEA and the audit system of the International Maritime Organization (IMO). The ‘New’ Model The key features of the ‘new’ international rule makers all stand in strong contrast to the defining features of the classic model. First they have selective memberships rather than universal membership. Second, they rely on managing a network of officials from national capitals rather than on building up a permanent bureaucracy. Third, their tasks are specialised rather than comprehensive. Fourth, their main instrument is not the binding obligation but the statement of policy principles.9 The extent to

9 Winter draws attention to the multitude of meanings that can be attached to the concept of ‘principles’. He suggests distinguishing between principles of policy and principles of law and between principles of law and rules of law. ‘Principles of law are to be distinguished from rules of law; they are basic ideas informing rules, and whilst rules are conclusive, principles are open for balancing with opposing principles’ (Winter 2006: 603–604).


15/12/2010 08:09


Table 2.2

Selected ‘new’ venues

Venue

Date established

No. of members

OEEC/OECD BCBS/BIS NSG IOSCO IASB IPCC FATF IAIS FIU ICANN FSF/FSB

1949/1961 1974 1975 1983 1989 1989 1989 1994 1995 1998+ 1999

30 55 (27)++ 46 114 (18)* 16 194 (31)** 33 190 (9–24)*** 106 (15)**** 21 53 (24)*****

+ ++ * ** *** **** *****

41

ICANN operates IANA that dates back to the 1970s. From 27 countries. IOSCO’s technical committee (its key rule-making body) comprises 18 members. IPCC’s bureau consists of 31 members. IAIS members are drawn from insurance supervisors in 140 countries; its executive committee comprises a minimum of 9 and a maximum of 24 members. FIU, the organisation of Financial Intelligence Units, is known as the Egmont Group. Its key coordinating committee, the Egmont Committee has 15 members. Originally from 12 countries and widened to 23 (plus Hong Kong) in 2009.

which these policy principles subsequently enter into law and practice, and the way in which they do so, are left open. Table 2.2 gives a selection of the new venues that broadly fit within these criteria starting with the Organisation for Economic Co-operation and Development (OECD) and including the Basel Committee on Banking Supervision in the BIS. The basis for membership selection is the model set by the pre-war BIS – shared interests, outlook and professional position. The apparent exceptions to this selective approach include the International Association of Insurance Supervisors (IAIS) whose large numbers partly reflect the fact that, in the US, responsibility for insurance supervision rests with the individual states of the union, and the International Organization of Securities Commissions (IOSCO) and the Egmont Group. However, in each of these cases, the crucial work is done by a small subcommittee of members indicated in Table 2.2. The IPCC is included among this group, even though its membership is open to all members of the United Nations Environment Programme (UNEP) and World Meteorological Organization (WMO) (bodies of a classic type), because it operates on a network basis with only a small bureau and administrative staff.


15/12/2010 08:09

42


In contrast to the classic international institution with its comprehensive ambitions in the field, the new style of body makes a virtue out of specialisation. The new venues also rely usually on small secretariats and the key unit is the committee that brings together networks of national officials. OECD, the post war forerunner of this style of organisation is exceptional in that it has a substantial staff of its own (2500 mainly contract staff) and it also has permanent government representatives (ambassadors) to help provide strategic direction and oversight. However its main work is done through committee structures that draw on national officials. The BIS has a staff of around 500 (compared with 2400 in the IMF); others operate with only small secretariats and the NSG without a central secretariat. The pre-war BIS also established a precedent in involving the private sector.10 The private sector is brought in through a variety of organisational forms. The OECD brings in the private sector through its committee structure, IOSCO through a system of associate membership, while the International Accounting Standards Board (IASB) is a private foundation overseen by a group of trustees. It is important to note however, that the role of the private sector is not a uniform one. In the case of the IASB the private sector is the principal rule maker but depends on governments to endorse and enforce the rules. At the other end of the spectrum – NSG and FATF – government officials draw up the rules and the private sector is co-opted into their enforcement – through, for example, their reporting of dual use or suspicious transactions. Other bodies such as IAIS or IOSCO involve a more equal partnership of both officials and the private sector working together to draw up the rules. The Internet Corporation for Assigned Names and Numbers (ICANN) is different again – a ‘not for profit’ corporation that operates with an international government advisory committee and on the basis of an ‘Affirmation of Commitments’ by the US Department of Commerce designed to underpin ICANN as a ‘private sector led’ policy model. The final key feature of the new model is the preference for acting through agreements on principles rather than through binding treaty obligations or conventions. These agreements take a variety of names including accords, concordats, core principles, declarations, frameworks, recommendations, understandings, standards and guidelines. A selection of the most important, together with their original date of promulgation (disregarding updates and revisions) is given in Table 2.3. The IPCC is a partial exception to this general picture since its work has provided not only the basis for the UN Framework Convention on Climate Change

10

The founding members included two private banks.


15/12/2010 08:09


Table 2.3 OECD

BIS/ BCBS NSG IOSCO

IASB FATF IAIS

FSF/B

43

Selected principles-based instruments Recommendations on Regulatory Reform (1997) Principles of Corporate Governance (1999) Declaration on International Investment and Multinational Enterprise (2000) Principles for Regulatory Quality and Performance (2005) Concordat on Cross Border Banking Supervision (1983) Basel Capital Accord (1988) Guidelines on Nuclear Transfers (1978) Dual Use Guidelines (1992) Objectives and Principles of Securities Regulation (1998) Multilateral Memorandum of Understanding Concerning Consultation, Cooperation and the Exchange of Information (MMoU) (2005) 8 International Financial Reporting Standards (1989–) 41 International Accounting Standards 40 Recommendations (against money laundering) (1990) 9 Special Recommendations (against terrorist financing) (2001) Insurance Concordat (1999) Insurance Core Principles (2003) Common Framework for the Supervision of Internationally Active Insurance Groups (2010) Compendium of Standards (12 key standards endorsed by FSF as key for sound financial practices (various issuers/various dates)

that entered into force in 1994, which enshrines principles, but also the follow-up Kyoto Protocol that entered into force in 2005 and contained binding commitments.

THE MIX OF VENUES AND FORMS OF EXPRESSION Ideal Typology The distinctions made above between what belongs to the ‘classic’ model and what belongs to the ‘new’ is an abstraction (what has been referred to as an ‘ideal’ type). In the real world the distinctions are much less clear cut. Each of the defining features of the ‘new’ organisations can be found among the ‘classic’ institutions. In respect of universal membership, traditional organisations have had to find ways of restricting membership in negotiating agreements. For example, GATT/WTO is famous for its ‘Green Room’ negotiations where


15/12/2010 08:09

44


the negotiating group is limited to 20–40 of the most important members. Similarly, the G10 group of finance ministers meets alongside the semiannual ministerial meetings of the Bretton Woods organisations. It is also not the case that all the classic organisations have large permanent staffs. Some traditional international organisations such as the ITU, UPU, IMO, WMO and WHO (see List of acronyms, pp. ix–xi) were set up with the management of national networks central to their activities from the very start. In addition, private sector participation is also a feature of some. For example it has always been a component of the ITU, ILO, WIPO and UNWTO, and the World Bank group also has its private sector arm (the IFC). The creation of specialised affiliates or sub-organisations has also provided a way to divide up comprehensive terms of reference into more specialised categories and disciplines. For example the ITU has a specialised Radio Regulations Board, the WMO eight specialised technical commissions and the WTO three specialised councils for trade in goods, trade in services and trade related aspects of intellectual property rights. Similarly, WHO set up in 2000 a specialised Global Outbreak and Alert Response Network (GOARN) for reporting and monitoring outbreaks of selected diseases that could have important international implications. The use of non-binding instruments such as declarations of principle also has a long history of use by bodies of the ‘classic’ style. For example, the ILO’s 1998 Declaration of Fundamental Principles and Rights at Work is essentially a compilation and distillation of the most important recommendations and declarations of principle made since its founding in 1919. Since 1960 UNESCO has relied much more on non-binding recommendations and declarations than on binding conventions. Other examples where principles play a key role include the 1963 FAO/WHO Codex Alimentarius about food standards, guidelines and codes of practice. The IMF also provides an example of an organisation that originally relied on prescription to enforce global monetary rules and has shifted to a principles-based approach. As mentioned above, under the original Bretton Woods agreement the IMF monitored compliance with fixed exchange rate obligations under mandatory so-called Article IV consultations. Under the latest surveillance regime, agreed in 2007, surveillance takes place on an ad hoc basis and is conducted in the light of four guiding principles.11 Presenting the world of international rule making as one that comprises two distinct types of institution – a ‘classic’ form of venue for rule making 11 The first principle about the avoidance of manipulation of exchange rates constitutes an obligation on members; the other three principles about exchange rate intervention policies are recommendations rather than obligations.


15/12/2010 08:09

45

INSTRUMENTS


Binding

WTO Trade Agreement

INSTITUTIONS

FATF special recommendations

‘CLASSIC’

IMF/IBRD ROSCs

‘NEW’

IOSCO Self-Regulation

Principles

Figure 2.1

Typology of international rule making

with its classic style of binding obligation and a ‘new’ more specialised form of venue relying on declarations of principles – is thus a huge simplification. There is no entirely straightforward historical narrative where one form of institution has been superseded by another and one style of rule making has been overtaken by another. Neither do the ‘pure’ forms or ‘ideal types’ exist in reality.12 In practice the picture is much more complicated. The classic venues co-exist alongside the new venues and the new instruments co-exist alongside traditional instruments, and both venues and instruments are co-mingled. It is the resulting mix that stamps today’s scene of international rule making. Co-mingling The present style of co-mingling in international rule making is illustrated in Figure 2.1 above. It gives examples of venue and instrument along two different axes – the classic venue with its universal membership and the

12 According to Ahrne and Brunsson’s definitions (2011), we live in a world largely composed of ‘partial organizations’.


15/12/2010 08:09

46


new venue with its restricted membership on the horizontal axis, and the classic instrument, the binding obligation, and the new instruments (declarations of principles) on the vertical axis. On the horizontal axis classic bodies such as the IMF, WTO and WHO belong on the left while the new bodies such as the BCBS, IASB, IOSCO belong on the right. On the vertical axis the classic instruments such as a treaty or convention stand at the top while new instruments such as accords or recommendations for self-regulation are at the bottom. An example of rule making that combines these two measures is given within each quadrant. These examples are simplified but provide a first approximation at representing the reality of international rule making: ●

●

●

●

In the top left-hand quadrant, an organisation with universal membership (WTO), where members enter into and enforce binding international trade agreements with a formal mechanism for resolving disputes, provides an example where a classic venue continues to be used together with classic instruments. In the bottom left-hand quadrant, classic international institutions (IBRD and IMF) promulgate through their Reports on the Observance of Standards and Codes (ROSC) the new style of instrument (principles-based standards and codes in the area of financial sector regulation, supervision and market integrity) developed largely by the new specialised network organisations. In the top right-hand quadrant, an example of one of the new venues (FATF) is a source of new instruments in the form of ‘special recommendations’ that become binding obligations when translated into the domestic laws of members. In the bottom right quadrant a new style venue (IOSCO) promulgates the new style of instrument (for example a self-regulatory set of principles for private equity firms to mitigate potential conflicts of interest).

The case shown in Figure 2.1, where classic bodies (the IMF and IBRD) disseminate through their ROSC programs the financial standards and codes partly developed in the new venues (such as IOSCO and IASB), makes the further point about how venues and instruments come to be intermingled. This reflects the practical reality of overlapping functional boundaries. The interlinking of venues and instruments is one reason why international rules may change their character from the place of inception. For example principles of best practice (ROSC) may become entwined with IBRD and IMF ‘conditionality’ in cases where the adoption of the principles becomes part of a package of sticks and carrots that accompanies financial assistance from the Bretton Woods organisations.


15/12/2010 08:09


47

The end result is a very confused picture of international rule making. As a result, strategic coordination has now become a major issue of its own. Strategic Coordination The loss of faith in any over-arching institutional design for the world as ‘one society’ and the multiplication of different specialised venues has raised important questions about how to achieve strategic coordination in international rule making. The answer has been found largely in the ‘new’ model. It involves the formation of small groups with shared interests and outlooks, avoids large secretariats in favour of reliance on networks of officials and operates through suasion rather than obligation.13 For a long time the most important of these groups, with the longest history, has been the G7 that brings together heads of state of the seven leading market economies and democracies. Roosevelt had always leant towards a system where the ‘big four’ (the US, Russia, the UK and China) acted as the enforcers of global rules. As early as 1953 Winston Churchill suggested that strategic questions were better settled through a revival of the wartime meetings of the allied troika than through the United Nations. Ad hoc summits prevailed until 1975 when they were replaced by regular meetings of the leaders of the six major market oriented countries. G6 became G7 with the joining of Canada in 1976 and G8 with the participation of Russia in 1998. The agenda of the G7 varies according to the burning issue of the day, for example a possible breakdown in international trade talks, or according to the need to give impetus to a strategic goal, such as measures to address climate change. In order to support such a varied agenda the G7 at head of state level is supported by ministerial groupings that vary depending on the subject and that also may meet regularly. In the aftermath of the 2008 international financial crisis the G7/8 grouping has been overshadowed by the G20 that includes other major powers such as China, India and Brazil, following a similar fluid agenda and with similar ministerial support. Some observers see the shift as reflecting a basic change in relative economic and political power in the world and see the G20 as the new supreme coordinating body and the G7 possibly fading away. This prospect is considered further below in the discussion of the stability of institutional arrangements.

13 The G77 grouping of developing countries with about 130 members is the exception to small groups.


15/12/2010 08:09

48


G10 FSB*/G20 Switzerland†† Australia Hong Kong†† Singapore††

G7** United States† Canada Japan United Kingdom† EU Presidency Italy Germany † France

Netherlands††

Belgium Sweden Notes: * Financial Stability Board ** G7 meets as G8 when Russia is included *** G24: Country observer **** G77 operates as a larger group of developing countries with about 130 members † Permanent member of the UN Security Council †† Not member of G20

Figure 2.2

Saudi Arabia, Korea, Indonesia, Russia**†, Turkey

South Africa, Agrentina, Brazil, China***†, India, Mexico

Algeria, Colombia, RD Congo, Ivory Coast, Egypt, Ethiopia, Gabon, Ghana, Guatemala, Iran, Lebanon, Nigeria, Pakistan, Peru, Philippines, Sri Lanka, Syria, Trindad & Tobago, Venezuela

G24****

International coordinating groups

On the financial and monetary side, coordination initially centred on the G10 central bank governors who have been meeting since 1962 at the BIS along with a G20 group meeting since 1999. In addition the FSF (1999) has brought together financial market supervisors of the G10 alongside finance ministers and central bankers. The FSF was converted in 2009 into a ‘Board’ in the wake of the crisis and its membership expanded to include all members of the G20. Developing countries, feeling that their views were excluded by the G10 processes, set up their own grouping of G77 in 1964 that now has about 130 members. Because a group of 77 members was already ‘too large’ to achieve cohesion, a more restricted group of developing countries, the G24, was established in 1971. Not only is the membership tightly controlled in the various G formations but they have also avoided setting up their own bureaucracies. They can draw on bodies such as the OECD, BIS, and the Bretton Woods organisations, but the predominant mode of operation is to rely on national officials. Whether the FSB will establish a significant secretariat of its own remains to be seen. The overlapping architecture of the different groupings is shown in Figure 2.2 above.


15/12/2010 08:09


49

Organising Logic? In looking at the picture described above it is difficult to see any overarching organising principles.14 The conception of the founders of the post war system, of an architecture given coherence by the UN and by its Security Council and Economic and Social Council, has gone, and nothing other than improvised attempts at problem solving and coordination seem to have replaced it. There have been attempts to identify a coherent typology of relationships between the mix of institutions described above.15 Such typologies may be useful in some sectors of international rule making. There is however a danger in suggesting a logic that may misrepresent the reality. For example, the IMF and IBRD both count as specialised agencies of the UN suggesting some kind of ‘nested’ relationship within the UN system. In reality both keep a deliberate distance. Similarly, it is tempting to see the organisations around the BIS as a ‘cluster’. However, this detracts from another important aspect of their activities – the partially overlapping and competing relationship with the Bretton Woods organisations. The overall picture thus remains a confused and confusing one.

STABILITY The current mix of venues and instruments used in international rule making is not only confusing, it also appears inherently unstable. Multiple venues are involved in rule making and multiple forms of instrument used for conveying the rules. Coordination mechanisms look haphazard and weak in the face of recent and foreseeable future challenges. At the same time, shifts in the traditional balance of power between states and regions also suggest instability ahead. Russia’s place in the world is being revived, it has become commonplace to project the continued economic rise of countries such as China and India long into the future and, if the terms of trade continue to favour commodity producers, other economies too, such as Brazil, will rise in prominence. Taken together, it seems plausible that such new, or revived, powers will turn 14 Braithwaite and Drahos (2000: 7) suggest that there are no ‘master mechanisms’ and only ‘webs of influence’. 15 Young distinguishes between relationships that are ‘embedded’ in over-arching general principles, ‘nested’, where specific arrangements are folded into a broader framework, ‘clustered’ where different institutions combine into a coherent package and ‘overlapping’ where regimes are founded without reference to each other and yet impact on each other (Young 1996).


15/12/2010 08:09

50


away from the elites and the clubs that have dominated international rule making in recent years towards a different style of rule making.16 Continued instability is not the only possible outlook. Another possibility is that a different kind of stability will emerge. This would be built around a ‘hybrid’ style of organisation that combines features of the ‘classic’ organisation with features of the ‘new’. Stronger coordinating groups might also emerge in a new form. The prototype of the possible new kind of ‘hybrid’ is the IPCC that combines features of the classic institution (such as universal membership) with the network features of the new (a small bureau and staff) and is associated with both styles of instrument (a framework convention expressing principles together with a protocol containing binding obligations). Possibly this kind of body might represent the institution and style of the future. The candidate for the key coordinating body is, as mentioned above, the G20. It too can be seen as a kind of ‘hybrid’ – it has a small exclusive membership but at the same time it has a greater claim than for example G10 to be representative of a universal membership. The basic question posed by the seeming instability and complexity of the present scene is therefore whether some over-arching logic will be reestablished, or whether the complexity of current arrangements is inevitable and will persist.17 The prospects for a continuation of a fluid mix of rule-making styles as against the prospects for a new emergent, general logic are analysed below in terms of transaction costs and in terms of congruence. These two perspectives have been chosen partly in order to echo the historical debate about the relative importance of form in relation to process. Congruence is about form – and in its original construction about the way in which institutions with political authority needed to mirror patterns of authority experienced in social institutions outside politics in order to achieve stability. The transaction cost perspective is about process – the costs involved in different approaches to achieving and maintaining contractual agreements. The Transaction Costs Perspective The ‘transaction costs’ perspective looks at existing arrangements in terms of their costs and benefits. It is an economist’s perspective.18 The unit of

16 Gruber (2005: 127) emphasises the potentially destabilising impact of changes in power relationships because in his view most international agreements ‘afford some signatories substantially greater gains than they afford others’. 17 Rosenau (2000) is one observer who suggests that complexity is inevitable. 18 For a recent overview see Macher and Richman (2008).


15/12/2010 08:09


51

analysis is the contract. In the context of international rule making the contract is an international agreement. States will interact in order to try to reach agreements in order to satisfy their particular preferences and to gain from exchange. What are termed ‘transaction costs’ are not always costs in a literal sense, although they may be.19 The term stands for any costs involved in arriving at the content of an international agreement, specifying its terms, monitoring its observance and enforcing compliance. The approach assumes that there are multiple possibilities in how to treat each of these components – how and where to arrive at the content of agreements, how to specify their content and how to monitor and enforce an agreement.20 The multiple venues and instruments described above are thus to be expected. The venues form the equivalent of a market place and can be explained in the same terms as business firms in the market place – as devices to economise on transaction costs. Traditionally the analysis of transaction costs has divided the costs between those that occur before the agreement is reached (such as the negotiation costs) and those that occur afterwards (for example the costs to monitor and enforce an agreement). This distinction has its limits in respect of coordination costs that refer both to the need to coordinate action both before an agreement is reached (for example in order to bring different types of expertise or government viewpoints together) as well as to the need to coordinate action in the period following an agreement (for example where both the public and private sector are involved in the implementation of an agreement or because there is ambiguity and room for interpretation in the agreement).21 In Table 2.4 therefore coordination costs are shown in the middle, straddling the divide.22 Following these main categories of transaction costs Table 2.4 offers a rough guide to where the net advantage is likely to lie – with the classic model or with the new.23 (Appendix A contains a more detailed discussion in order to support the allocations shown in the table.) What Table 2.4 shows is the advantages of co-existence between differ-

19 ‘Transaction costs are the economic equivalent of friction in physical systems’ (Williamson 1985: 19). 20 For a discussion of definitions see Demsetz (1969) and Epstein and O’Halloran (1999). 21 For a discussion of incomplete contracts see for example Dixit (1998) and Williamson (1996: 37) – ‘all complex contracts are unavoidably incomplete’. 22 Gruber (2005) treats coordination costs as particularly important in the period before an agreement because of the need to narrow down the range of possible equilibrium points in a bargaining situation. 23 The ‘net’ advantage takes into account not just the transaction costs but also the transaction gains (see the discussion in Dunoff and Trachtman 1999).


15/12/2010 08:09

52


Table 2.4

Transaction costs and styles of organisation

Before agreement Institutional Information gathering from specialists from decentralised sources from market Negotiating

Classic

X

Coordination Across expertise Between governments Between public/private

Classic X X

After agreement Monitoring/surveillance Implementing/compliance by governments by private sector

Classic X

‘New’ X X X X X ‘New’

X ‘New’

X X

Note: X indicates likely net cost advantage. For further detail see Appendix A.

ent types of venue.24 The ‘new’ style of organisation has a clear advantage in terms of its institutional costs (lower setting up and running costs), and in lowering information costs within a field, but the classic style has clear advantages in monitoring and surveillance between governments (because every country is included in the membership). In other cases the cost advantages are more evenly divided. For example the new style of organisation may have an advantage in coordinating private and public actors and likeminded governments. But in the case of complex issues where experts from different disciplines have to be coordinated there may be an advantage with a classic organisation (in other words a specialised venue may be better at reducing uncertainty in a particular field but a classic organisation better at dealing with complexity extending across several fields). A key example of evenly divided cost considerations is shown in the table under negotiating costs where an X is shown in both columns. This is because the new model may provide an easier, lower cost negotiating context since it brings together a select group. But, if the rules need

24

See also Bueno de Mesquita and Stephenson (2006) on this point.


15/12/2010 08:09


53

to apply beyond the select group, there are additional costs involved in extending the agreement that may sometimes tilt the balance back towards the use of the classic organisation as the negotiating forum. The conclusion to be drawn from the perspective of transaction costs is that a mix of the two styles appears to offer a combination of the best of both worlds. At first sight this might give support to the idea that institutional arrangements will begin to converge around a new style of ‘hybrid’ such as the IPCC that combines, within a single organisation, features of both the ‘classic’ and the ‘new’. However it also supports the contradictory view that rule makers benefit most from the opportunity to use whichever style of organisation is most suited to a particular problem. The analogy with the market also suggests instability in the sense that different costs and benefits associated with different organisations will always continue to encourage a search for the lowest cost venue.25 Instability is further suggested by a very different perspective from political science focusing on ‘congruence’. Congruence ‘Congruence’ looks at the relationship between the way authority is wielded in international arrangements compared with the way it is wielded in more familiar national arrangements. Congruence theory has its origin in a desire to explain the reasons for stability or instability in democratic regimes.26 The key proposition was that high democratic governmental performance required a high degree of congruence (defined as close resemblance) among governmental authority patterns and specified nongovernmental authority patterns (such as in families or political parties or trade union organisations).27 This original field of interest has been superseded by other concerns among political scientists who study democracy. Questions such as the process of democratisation itself and the causes of backsliding, or the dynamics of democracy’s spread or retreat, have overtaken questions about stability. But the basic suggestion that there is a relationship between the form politics takes on the governmental level and the form politics takes in non-governmental social structures remains influential. For example, it can be seen in the study of notions of ‘social capital’ and

25 Alchian and Demsetz (1972) see the market place for venues as a device for enhancing competition between input resources. 26 For an account see Eckstein (1980). 27 ‘A government will tend to be stable if its authority pattern is congruent with the authority patterns of the society of which it is a part’ (Eckstein 1992: 188).


15/12/2010 08:09

54


in the (contested) relationship between civil society organisations and democracy.28 More recently there has been a revival of interest in congruence theory in the somewhat different form of interest in the relationship between beliefs about democracy and the practice of democracy.29 It is discussed later in this different form in terms of ‘value congruence’. If the congruence hypothesis is applied outside the original field of enquiry to the field of international rule making it would suggest that a system of international rule making will achieve stability only when it mirrors the form of patterns of authority familiar within national states. There will be incentives to change, or instability, where there is a lack of congruence with more familiar patterns of power. This hypothesis can be contrasted with a different hypothesis about change connected with what is known as ‘isomorphism’ – the theory that there is a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions.30 According to this theory what has to be explained is the ‘startling homogeneity’ in organisational forms. The theory attributes this in part to the need for organisations to have relationships with states and in part to the extent of professionalisation in an organisational field.31 If isomorphism constitutes a valid hypothesis, the need for relationships with states and the professionalisation of international bodies will together propel international institutions towards a convergence of forms. Today’s apparent instability will thus be transient. A difficulty with applying the isomorphism hypothesis in the context of the stability or instability of international institutions is precisely that it assumes that the outcome of the enquiry will be towards convergence, or in other words towards eventual stability. Congruence theory does not make any such assumption. It puts forward a more neutral hypothesis about the direction of change. In addition, both a transaction cost approach and congruence theory make less restrictive assumptions about what relationships in the environment are relevant for influencing institutional change. Relationships with states and professionalisation are not the only factors conditioning form. Moreover, as will be discussed later, professionalisation itself may be consistent with unstable forms. Isomorphism is however 28

See, for example, Putnam et al. (1993). See for example Dalton and Shin (2006a) and Schlosser (2009: 54), who finds support for the idea of a systematic relationship between mass beliefs and levels of democracy. 30 This definition is taken from DiMaggio and Powell (1991c: 66). 31 An organisational field is defined as ‘those organizations that, in the aggregate, constitute a recognized area of institutional life: key suppliers, resource and product consumers, regulatory agencies and other organizations that produce similar services or products’ (ibid.: 64–65). 29


15/12/2010 08:09


55

referred to again later in the narrower context of looking at the relationships between professional bodies. What gives the idea of congruence additional plausibility when applied to international rule making is that there seems to be a powerful heuristic (short-cut method of calculation) at work that leads people in general, and politicians in particular, to want to see arrangements at the international level that mirror arrangements within states even if the settings are different. It is a recognised phenomenon in the case of discussion about arrangements within the EU.32 Arguably the founders of the post war system had in mind a kind of mirroring of domestic hierarchical authority structures when they saw the UN, its General Assembly, Security Council and Economic and Social Council and the International Court of Justice as the supreme organs for global governance, with other international bodies playing a subordinate role. Subsequently, the greater use of networked organisations, a greater mingling of private alongside public authority and a much greater use of specialised regulatory organisations can also be seen in both spheres. In addition, in each sphere there is a need to distinguish, as congruence theory does, between symbolic or ceremonial structures in patterns of authority as distinct from the ‘effective’ bodies. Part of what has happened in historical terms at the international level is that bodies such as the UN General Assembly, or the annual meetings of the ministers at the Bretton Woods organisations, intended to be effective, have become ceremonial, while bodies that are effective (such as the BIS) avoid publicity. However, these actual and imagined parallels between international institutional structures and the more familiar structures within states, do not suggest stability or the convergence on some new ‘hybrid’. Within states there is no settled division of power between different forms of authority or between public and private authority or about how specialised regulatory bodies fit within or between the conventional branches of government. In addition, adjacent authority structures (such as family structures) important in congruence theory, have changed radically and continue to change. If, therefore, international institutional arrangements are likely to mirror developments within states, the prospects are for continued change. The original programme of empirical research into ‘congruence’ led in an unexpected direction. It came to emphasise the rarity of the conditions for stability. If congruence theory is applied to international rule making the conditions for stability also look correspondingly rare.

32

See for example the discussion in Majone (2005).


15/12/2010 08:09

56


The Perspectives Combined Taken together, the suggestion both from congruence theory and from a transactions cost perspective is that the present system of international rule making will remain unstable. The transaction cost perspective underlines the variety of reasons for governments, elites and the institutions themselves to prefer the existing mix of arrangements to continue. There is an advantage in being able to call on more than one style of rule making, to mix and match different venues for making the rules and different instruments for conveying them and to keep searching for new forms that lower the cost of ‘frictions’ in international rule making. At the same time, from a completely different perspective, congruence theory also suggests instability. This is because there will be continuing dissatisfaction with international rule making unless it comes closer to the kind of rule making that is familiar and intelligible from examples in a domestic context but, at the same time, the domestic template is itself an unstable one. Both congruence theory and the transaction cost perspective also suggest looking at the question of stability from a rather different perspective, and one that does not make the normative assumption that stability is necessarily a good quality in international rule-making institutions and procedures. As originally formulated, congruence theory led to the normatively unpalatable conclusion that ‘semi authoritarian government . . . is bound to be the most stable of all possible governments’.33 If applied to international arrangements it would imply a similarly normatively disquieting perspective on the relationship between the effectiveness of international systems and democratic values by suggesting that a stable system of international rule making will also be a semi-authoritarian one. In transaction cost theory there is no presumption that stability is a quality to be sought. On the contrary. Change will be a constant in a competitive market place for minimising ‘frictions’. In thinking more about the goals of international institutional arrangements there is an alternative suggestion that can be drawn from congruence theory and that is to think in terms of ‘managing strain’. In congruence theory this meant avoiding the complete breakdown of a normative order and managing the co-existence of different forms of order, including contradictory norms of conduct.34 The international order can also be seen, just as much as any domestic political system, as about man-

33

Eckstein (1992: 224). Easton (1965: 99–135) also discusses in similar terms the importance for political systems of being able to respond to ‘stress’. 34


15/12/2010 08:09


57

aging the co-existence of different forms of order, including contradictory norms, and how to avoid the breakdown of the overall order. ‘Managing strain’ provides an umbrella for a number of different concepts involved in thinking about international rule making – the management of risk; navigation through uncertainties, including uncertainties of behaviour; bringing together different approaches to political and social order; and managing tension between different value and belief systems. A transaction costs perspective suggests that ‘managing strain’ does not involve a search for an ideal form of either the classic or the new style; nor a search for a fresh ideal organising logic that combines both the ‘classic’ and the ‘new’. Instead, it suggests that we should accept fluid forms able to deal with changing problems, the mobilisation of different forms of knowledge and the capacity to reduce frictions between different views of economic, social and political order. *

*

*

The discussion so far has set out the current arrangements for international rule making, the historical background, and the prospects for the continued use of a mix of institutions and instruments in the absence of any new or convergent over-arching organisational logic. The next chapter looks at two different frameworks for analysing the vulnerability of the present system to failure in its rule making efforts and for diagnosing the source of the democratic deficit.


15/12/2010 08:09

3.

Analytic frameworks

The outstanding characteristic of international rule making is that it blurs all conventional boundaries and demarcation lines. It blurs geographical boundaries because policy proximity between policy makers becomes as important as geographical proximity; it blurs political and jurisdictional lines between what is domestic and what is international because policies are drawn from both internal and external sources; it blurs divisions between what is state and what is private because the private sector exercises authority alongside the state; it blurs distinctions between what belongs to international public law and what belongs to private law because they may be used as substitutes; it blurs what is law and what is not law because instruments such as codes may have legal effect without taking legal form; and it blurs professional distinctions as a fast-changing world forces professional boundaries to become more open and permeable. This chapter therefore examines in greater detail the two frameworks for analysing international rule making, referred to in the introduction, each of which avoids over-reliance on conventional distinctions. The first framework examined is known as ‘multi-level governance’. This framework addresses in particular the blurring of political boundaries between different jurisdictions, between what is domestic and what is international and between what is public and what is private. The second framework explored is provided by the series of analytic distinctions that underlie what is known as diffusion theory. This places the emphasis on a different set of distinctions – that of the different actors involved at different stages in international rule making, their different roles and their different styles of reasoning and organisational settings. They have been chosen as the frameworks for the analysis not only because they address the blurring of many conventional boundary demarcations but also because they reflect the two sides of the historical debate about post war international rule making. The first framework centres on questions of the form of modern systems of government; the second on processes. Thus they offer different ways into making judgements about the strengths and weaknesses of current international rule-making arrangements. Each also provides a different foundation for reasoning

58


15/12/2010 08:09

Analytic frameworks

59

about the success or failure of international rule-making systems and the presence or absence of democratic values. Both frameworks provide for an interdisciplinary approach and can draw on a variety of theories and models.1 The focus and unit of analysis are however different. Multi-level governance focuses on forms of authority. The unit of analysis is the network as it is embedded in a larger jurisdiction such as a federal state or a regime that crosses international boundaries. Alternatively, the unit may be the larger jurisdiction as it is modified by the network. By contrast, the unit of analysis in diffusion theory is the policy decision. The focus is on the characteristics of the actors in their particular decision-taking setting.2 Both frameworks reject the main alternative point of departure, which is to continue with a traditional focus on the nation state and to treat the state as the unit of analysis.3 The multi-level governance framework rejects the state-centred approach because, in contemporary democratic societies, authority is wielded by non-state actors outside the traditional institutional framework of the state. The diffusion framework rejects the state-focused approach because the use of the state as the unit of analysis compresses the stages of the policy process, the different actors who play key roles at different stages and their particular styles of reasoning. Instead of different styles of reasoning there becomes one compressed logic – the logic of state power or state interest. Instead of different actors, such as epistemic communities, who have a role in framing policy problems and their solutions, there is one actor – the state.4 The role of citizens has similarly to be compressed. A questionable assumption is made that states reflect societal preferences in a ‘bottom-up’ process.5 1 Ostrom (1999) makes a distinction between frameworks that help identify the elements we need to consider and which specify the unit of analysis, between the theories which link elements of the framework to particular questions and between models which make precise assumptions about particular variables in the analysis. 2 Manning (1986: 1292) refers to a process of ‘natural decision making’: ‘The aim is to connect decision points . . . with the cognitive frames which locate the relevant field . . . and internally pattern it’. This type of approach has part of its roots in theorising about the ‘social construction’ of reality, (Berger and Luckman 1966). For applications see Hawkins (2002) and Vaughan (1996). 3 For a recent example of a state-focused analysis see Drezner (2007). 4 Drezner rejects the role of epistemic communities on the grounds that the outcomes of the policy process rarely reflect the evidence as viewed by epistemic communities (ibid.: 20–21). 5 Drezner asserts that the exercise of voice by interests within the state ‘creates demands for action that must be addressed by the government in power’ (ibid.: 49; my italics). According to Drezner, preferences are aggregated by the state in a two-step process. See also Slaughter (2004b) for an analogous ‘two-level’ view of policies as a ‘bottom-up’ process. Legro (1996) criticises the ‘two-step’ assumption for failing to take proper account of the formation of preferences. Savage and Weale (2009) not only point to the lack of empirical


15/12/2010 08:09

60


Although both frameworks reject a state-centred account of international rule making, their different focus and different units of analysis lead them to offer completely different accounts of failures of effectiveness and the democratic deficit in international rule making.

THE MULTI-LEVEL GOVERNANCE PERSPECTIVE The multi-level governance perspective offers an approach to both the question of how to ensure the democratic legitimacy of rule making that crosses political boundaries, whether international or within a federal state, and how to ensure the effectiveness of such rule making.6 It focuses on forms of authority that emerge when conventional units of government demarcated by territory break down. In particular it highlights shifts in ‘horizontal’ relations between state and society (for example the importance of civil society organisations) and in ‘vertical’ relations (between hierarchical units of government). As mentioned earlier, the multi-level governance perspective has been developed from a variety of sources, including international relations theory, EU studies, studies of federalism and studies of how public services are delivered within multi-level national systems.7 The EU is sometimes held out as the representative example of a new form of governance to which other systems will evolve as international rule making increases in importance and possibly as a model for international governance more generally.8 Under this approach the term ‘governance’ is preferred to ‘government’. Although the term ‘governance’ can be used in a number of different senses, its core meaning is to imply that authority in modern systems of government is wielded by non-state actors as well as conventional branches of the evidence in support of a two-step process but also point out the rigorous set of normative conditions that would need to be met. 6 Marks and Hooghe (2004: 16) suggest that the different literatures on multi-level governance all ‘agree that the dispersion of governance across multiple jurisdictions is both more efficient than, and normatively superior to, central state monopoly’. 7 One particular source arising from dissatisfaction with traditional federalism studies has been ‘intergovernmental relations’ theory that aimed to get away from the assumption that the national level was inherently superior to state, local and municipal levels in delivering public services and allowed for non-state actors such as interest groups to be included in the analysis. (See Wright 1978 for a description and a history dating back to the 1930s in the US.) 8 Colomer (2007: 95) defines governance in the EU as implying ‘multiple levels of institutions of which none – neither the Union, nor each of the states, nor any smaller unit – can be considered to be “sovereign” anymore’. For a discussion of the EU as a model for others see Kahler and Lake (2003).


15/12/2010 08:09

Analytic frameworks

61

state, such as legislatures, executives and judiciaries, and that authority is wielded through informal as well as formal mechanisms.9 The term also implies that these new forms of authority cohere as a ‘system’.10 The Approach to Effectiveness In its emphasis on ‘governance’ the multi-level system incorporates many of the general assumptions about effectiveness associated with what is termed the ‘new governance’ – particularly the importance of the ‘new public management’ and ‘good governance’.11 However, many of the prescriptions of the ‘new governance’ already look dated and the claims of multi-level governance to offer a superior approach to the effectiveness of policy making rest on two more fundamental features. First, it offers ‘scale flexibility’. This means that by combining horizontal and vertical elements, the unit of government can be adapted to the size and the boundaries of a particular problem. Second, it offers network efficiencies – the benefits that accrue to bringing together decentralised sources of information and knowledge, notably that held within the private sector. Scale flexibility enables multi-level governance theory to recognise the plasticity of the boundaries of public policy. In today’s world the appropriate reach of public policies rarely fits boundaries determined by territory. International policies reach into national and local jurisdictions; conversely, local policies may also have implications for other localities, the national level and even have international consequences. In particular, scale flexibility takes into account what are known as policy ‘externalities’. Most policies have spill-over effects from the particular jurisdiction for which they are framed and the most appropriate level of government is the one that minimises what is termed ‘negative spill-over’ into other jurisdictions. Scale flexibility also allows jurisdictions to be adapted to the scope and scale of providing for different collective goods (such as public transport links) and common goods (such as clean air). Scale flexibility thus allows for jurisdictions to be ‘custom-designed’.12 Because of the emphasis it places on horizontal authority structures, the multi-level governance framework has absorbed earlier network analysis

9

See Bache and Flinders (2004b) and Rosenau (2004). Rosenau (2000:188). 11 See Rhodes (1997: 46–47) for a definition of the ‘new governance’. 12 See Marks and Hooghe (2004) and the discussion of ‘common goods’ in Mayntz (2002). 10


15/12/2010 08:09

62


with its emphasis on the importance of decentralised ‘nodes’ of authority.13 In particular it incorporates much policy network analysis.14 Network efficiencies arise because governments are not always the key actors in settling policy. The pivotal actors in international policy making are the officials, technical experts, lawyers, scientists and other professionals from the private as well as the public sector who in previous times would have occupied themselves with purely domestic agendas. They are linked together through their own social and professional networks. They add to the effectiveness of governance because they can communicate directly to other members of their network without having to go through government hierarchies and they can introduce privately sourced information.15 The Approach to Democratic Norms In claiming normative superiority, multi-level governance also relies heavily on the network element. Networks appear to provide a decentralised approach to governance where power is disaggregated and exercised through non-coercive means rather than through the power of the state. This component moderates any over-reliance on hierarchical models where power resides simply with governments. It also allows for incorporating the participation of non-governmental actors.16 In this area multilevel governance incorporates that component of the theorising under the ‘new governance’ referred to above that also emphasises the role of new actors wielding authority in new ways and involving new relationships with traditional sources of authority.17 In addition, theories of multi-level governance also emphasise the values that are shared within networks. For example, judiciaries that link 13 For a review of the development of network analysis see Rhodes (1997) and Thatcher (1998). 14 See Bache (2008). Marcussen and Olsen (2007) note that there are two strands to network analysis – social network analysis and policy network analysis – and that multilevel governance has incorporated more of the policy network analysis. Rhodes (1997: 29) defines a policy network as ‘a link between the micro-level of analysis, dealing with the role of interests and government in particular policy questions, and the macro-level of analysis, which focuses on broader questions about the distribution of power within modern society’. 15 In the context of international rule making, Murphy and Yates cite the International Organization for Standardization (ISO), a private network organisation that achieves authority through mobilising a voluntary consensus, as an example of ‘an alternative to our ineffective UN system of intergovernmental organization’ (Murphy and Yates 2009: 68). 16 Dryzek asserts, ‘The most appropriate available institutional expression of a dispersed capacity to engage in deliberation that helps determine the terms of discourse in the international system is the network’ (1999: 48). 17 See for example De Búrca and Scott (2006).


15/12/2010 08:09

Analytic frameworks

63

together will reinforce common procedural norms important to the rule of law and increasingly take into account substantive norms developed elsewhere.18 Scientists and economists also spread their own procedural disciplines. The ISO is an example of a network organisation that has branched out from setting technical standards to setting standards in areas such as quality management, corporate social responsibility and the environment.19 While networks and civil society organisations provide a ‘bottom-up’ element to theories of multi-level governance, there remains the question of the role and place of the hierarchical elements of government. In their treatment of the state, theories of multi-level governance acknowledge the continuing importance of the role of governments. But that role is given new clothes and new life by emphasising how governments are able to wield power and influence at different levels at the same time.20 The government of a centralised state will simultaneously determine what lower levels of government can do, will act at the national level and will represent the state at the international level and at the European level in the case of EU member states. Even in the case of federal states, where the power of a central government may be constitutionally limited, it is still likely to wield influence over lower level jurisdictions and it is the federal government that will usually have the authority to act internationally. Despite the attractions and plausibility of theories of multi-level governance, the claims have to be treated with a great deal of caution. The framework does not provide a reliable guide either to the effectiveness of international rule making or to the analysis of its democratic qualities. Effectiveness Achieved? First, the foundations on which the claimed advantages of scale flexibility stand are not as secure as they might seem. ‘Externalities’ have been described as an ambiguous concept and an ambiguous guide to policy.21 This is both because it is difficult to find any public policy that does not have some external effect and because of the possibility of very different views as to whether the effect is positive or negative. Taxation policy 18

Slaughter (2004a: 91) refers to ‘judges judging judges’. Murphy and Yates (2009: 68). Putnam provides the theoretical model (1988). Kapstein (1994) provides a practical example in the context of international financial regulation of how a government can help make rules internationally but apply them in its own jurisdiction without the intervention of others through the doctrine of ‘home state control’. 21 See Dunoff and Trachtman (1999). 19 20


15/12/2010 08:09

64


provides a clear example where some countries argue that those with low taxation policies undermine the welfare systems of countries with high tax policies while those with low tax policies argue that lower taxation would improve welfare.22 The issue of whether a spill-over has a negative or positive effect may also be a question of degree and of timing. For example, portfolio theory emphasises the advantages of diversification because risks can be spread if markets are not closely correlated. However, as the recent financial crisis illustrates, the more that investors diversify, the more closely markets become interlinked. Risks become correlated instead of being diversified. What starts as a positive effect (diversification) becomes a negative effect (vulnerability to the risk of contagion).23 The policy implications for rule makers thus change over time and because so many policies may acquire some external effects it becomes increasingly difficult to argue on empirical grounds that many if not most policies should not be dealt with internationally. There is a similar ambiguity attached to concepts of common and collective goods. There are few goods that are either purely private or purely public and lead to clear prescriptions about the scope and scale of public provision and public policy. In theory, a system of governance that is ‘scale flexible’ should be able to adjust to the changing boundaries of policy externalities and changing approaches to common and collective goods. In practice, governance is likely to be a one-way street. Jurisdictions rarely give up powers once they have been acquired.24 The claimed advantages are displaced by the inefficiencies of scale and the dislocations associated with overlapping actors in multiple jurisdictions.25 Second, in addition to the ambiguities of ‘scale flexibility’, there are reasons to be cautious about the efficiencies claimed for networks. Networks can vary widely in their characteristics.26 It is a term whose origin lies in part in attempts to describe the many varieties of institutional form 22

For a discussion of fiscal policy interdependence see Franzese and Hays (2008). See Gallegati et al. (2008). 24 Beer notes that while in theory governments could give up policies that have a negative spill-over impact they are unlikely to do so: ‘government is more likely to add a program modifying the burdensome effects, while maintaining the main activity’ (Beer 1973: 76). 25 A study of multi-level regulatory governance in Canada suggests that it can imply ‘At a minimum, twelve or thirteen jurisdictions’ (Doern and Johnson 2006: 18). 26 Ahrne and Brunsson (2008: 4) complain about ‘linguistic chaos’ – ‘Almost any interaction among organizations seems to be possible to describe as a network’. Case studies also indicate the very different content and degrees of ‘partial institutionalisation’ that different networks may have – from networks with dense personal interactions to ‘flow’ connections where there is little personal interaction (see Knorr Cetina 2005). 23


15/12/2010 08:09

Analytic frameworks

65

that lie between the theoretically pure forms of a market and hierarchies.27 Not all forms will offer efficiencies.28 There is also an inadequate consideration of how flows of information are managed in a network.29 Network theory also leaves a blank page about the characteristics of the venues that bring together members of the networks. Rule-making venues are more than simply convenient meeting places. There are reasons, discussed later, why particular venues are chosen for negotiating the content of rules. The behaviour associated with or characteristic of those venues has important implications for the nature and quality of the decision taking. A further weakness concerns the ‘node’ – the key organisational unit according to network theory.30 The ‘node’ is the particular location of a principal in the network. However, some nodes carry more weight and influence than others and it is important to know why. For example, Brussels, Rome, Madrid, London, New York and Washington are all nodes in networks of accountants, banking and security market supervisors and regulators but they do not carry an equal weight. The network is camouflaging different types and degrees of authority and different relationships between authorities. It is not describing it. The approach to the effectiveness of rule making in theories of multilevel governance, relying on claims about the treatment of policy externalities and the superiority of network organisations is thus not as compelling as it seems at first sight. The allocation of policy tasks becomes much more a question of discretionary political judgement than is allowed for by ‘custom design’ and ‘scale flexibility’. In addition, analysis of the way in which the substance of the tasks is approached has to go far beyond the assertion of network efficiencies. Networks can be seen as part of a shift in organisational structures but do not adequately describe what is happening.31 There are also reasons to be cautious about the claims of theories of multi-level governance to offer insights into the democratic qualities of systems of governance. 27 See Podolny and Page (1998) for scepticism about any generalised functional efficiencies claimed for networks. 28 According to one well-known theory, ‘weak ties’ are likely to generate more information than strong ties (Granovetter 1973) and according to another, strongly connected networks may make it more difficult to close deals or to adopt a multiple lens in analysing a problem (Mizruchi and Brewster Stearns 2001). 29 Burt (1992) emphasises the importance of ‘structural holes’. 30 Castells (1996). 31 Powell (2001: 45) therefore avoids the term in talking about a new logic ‘that is built around project-based work and team organization; flatter, more horizontal organizations that rely on long-term interdependent relations with external parties; and extensive efforts to leverage capabilities across a wide range of activities’.


15/12/2010 08:09

66


Democratic Norms? One reason for caution is that there appears to be a fundamental disjunction between the values encapsulated in networks and democratic norms. Networks value privacy, informality and ties within peer groups.32 Democratic norms involve transparency, due process and representation. Within democratic forms of government there must be space for many different varieties of organisation with their different norms. But if power to make public policy resides within networks the way in which the different norms connect up or relate to each other must be made clear.33 The importance of being clear about the norms of different actors applies particularly to international rule making. For example, when international NGOs base their activities on the moral ‘rightness’ of their cause it is important to recognise that their views are not necessarily representative of ‘civil society’ more generally, nor their factual claims necessarily evidence based. Neither is it safe to assume that the disciplines that apply to non-state actors in their national context will apply also in the international context. Actors are not immune to changes in the setting and context in which they operate. Because of these problems, in order to try to bolster the normative claims of network governance, the advocates of the virtues of networks shift their ground. Instead of appealing to the inherent normative qualities of networks, the appeal shifts to externally derived and imposed norms that can be applied to networks.34 The weak attachment of network theory to democratic norms conceals a fundamental confusion about what the underlying approach to legitimacy and accountability really is.35 For some multi-level governance theorists there is an assumption that governments still provide the crucial connection between international rule making and domestic decision taking.36 32 Podolny and Page (1998: 59) define a network as any collection of actors ‘that pursue repeated, enduring exchange relations with one another and, at the same time, lack a legitimate organizational authority to arbitrate and resolve disputes that may arise during the exchange’. This definition highlights the problem of the basis for authority in such groups. 33 Cutler et al. (2003b) note that the emergence of private power enhances the capacity of some actors to gain power over others. Peters and Pierre (2004: 85) refer to a ‘Faustian bargain’ where the core values of democratic government are traded for accommodation, consensus and purported increased efficiency. Keohane and Nye (2003: 403) argue that policy networks are by definition elitist and ‘there is no guarantee of democratic accountability’. Ahrne and Brunsson (2008) suggest that networks lack legitimacy because they have no authoritative centre. 34 See for example Slaughter (2004a: 259). 35 See Papadopoulos (2007). 36 Marcussen and Olsen (2007) refer to the need for networks to be democratically ‘anchored’, including being controlled by democratically elected politicians.


15/12/2010 08:09

Analytic frameworks

67

Others look beneath the inter-governmental appearances of international policy making. They look at the broad cultural underpinnings that help give civic organisations their authority in policy formation.37 The link between these two approaches to legitimacy needs to be made clear.38 In practice the relationship between horizontal and vertical structures and the justification of the mix is often disguised.39 What may look like horizontally organised networks may in practice operate under the threat of intervention by a body at a higher level in the hierarchy – the so-called ‘shadow of hierarchy’.40 Similarly, the claim that network organisations allow for grass roots involvement in multi-level politics by ‘civil society’ is also questioned because involvement may be determined by official ‘gatekeepers’.41 When this kind of intermingling of authority occurs, responsibility and accountability are blurred and legitimacy weakened. Accounts of multi-level governance also ignore two important characteristics of international rule making that apply to governments as conduits for democratic norms. First, the ambiguity about where rules are best made means that governments have considerable discretion about where to legislate. Their choice is likely to favour the point of least resistance rather than the point where democratic expression is most vigorous. In addition, because policy spill-over is so pervasive between different jurisdictions, governments often will look for ways of sharing responsibility. This too can be a means of concealment.42 Second, the actual negotiating setting in which governments operate is also key to any analysis. The most basic feature of any international negotiation on rule making is that no single government, democratic or not, can exert so much power that only the preferences of its own electorate can be reflected. In almost all situations any single government will have to choose negotiating objectives that take into account the negotiating objectives of other governments. Its eventual policy choices also will be influenced by those of other governments. The different preferences of different elected governments may overlap. Moreover, it is in this area

37

Boli and Thomas (1999). Linseth (2002) argues against trying to have it ‘both ways’. 39 Börzel and Heard-Lauréote (2009) draw a distinction in the case of the EU between governance ‘by’ networks and governance ‘in’ networks. They suggest that the role of private actors has been overstated. 40 For a discussion of the ‘shadow of hierarchy’ see Héritier and Lehmkuhl (2008). 41 See Maloney and van Deth (2008b). 42 Slaughter (2004a: 255–7) appeals to the EU principle of ‘subsidiarity’ as a way to demarcate responsibility and to ensure ‘bottom-up’ policy making. Unfortunately, in the EU context the principle operates largely at the discretion of the EU authorities themselves. 38


15/12/2010 08:09

68


of overlapping preferences that international agreements may be found. But a straightforward link with the most preferred choices of any single electorate will have been broken except through happy coincidence. What is more, because domestic rule making is increasingly linked with international rule making, those international rules, even if they do not fully reflect domestic preferences, will increasingly shape domestic policy. The logic of the negotiating context must therefore be central to any analysis of the role of governments as transmission belts for democratic expression. The cautions that apply to multi-level governance theory thus concern both its empirical and its normative claims. They do not, however, necessarily lead to the conclusion that the multi-level framework should be discarded. The new forms of governance are important. The reasons for caution do nevertheless suggest, at the very least, that an alternative framework is needed to get at characteristics of international rule making that are not reached through the multi-level governance framework. Such an alternative framework is provided by diffusion theory.

A DIFFUSION APPROACH What is known as diffusion theory has its roots in sociological studies of the adoption of innovations.43 Two classics in the field concern the adoption of hybrid corn and the adoption of a new class of medicine. Adoption can be either spontaneous or managed or both. Political scientists have borrowed from the approach and findings to apply it to such subjects as the spread of democracy in the world.44 The idea that the adoption of democracy goes in ‘waves’ is probably the best known of such adaptations.45 The association of the spread of democracy with waves is a contested borrowing and one that possibly ignores the authors of the original

43 ‘Diffusion refers to the spread of something within a social system’ (Strang and Soule 1998: 266). 44 Following a pioneering application in political studies by Walker (1969) relating to the spread of legislative programmes between the states in the US, more recent investigations include a study of the spread of the uniform sales act in the US (Smythe 2008), the study of the spread of state regulations in the US on the sale of securities (Mahoney 2003) and the study of the spread of workers’ compensation laws in the US (Fishback and Kantor 1998). Braithwaite and Drahos (2000: 585–93) refer to diffusion techniques for the spreading of global regulations. In Black et al. (2005) case examples of diffusion in regulation are also discussed. In all of these studies the focus is on empirical findings about the pattern and reasons for the spread of a law rather than on the conceptual framework of the analysis. 45 Huntington (1991). See also Markoff (1996).


15/12/2010 08:09

Analytic frameworks

69

findings that hybrid corn had been adopted by farmers in a wave pattern, who cautioned against assuming that such patterns necessarily applied in other fields.46 The Structure of Analysis What is relevant to the analysis of the effectiveness and democratic quality of international rule making from diffusion theory is not the findings about particular patterns of diffusion but the structure of the analysis. In particular the analysis distinguishes between three classes of actors, each playing a distinctive role at different stages in the diffusion process, each following their own logic, and each drawing on their own sources of reasoning in their respective organisational and social context. In the original diffusion model the first class of actors was the scientists who, at the first stage of innovation, came together to develop the hybrid corn or the pharmaceutical product. Their distinctive style of reasoning is scientific. In other fields a different type of expertise might be drawn from. For example the spread of so-called ‘poison pill’ defences against hostile take-overs in the corporate world might be attributed to groupings of corporate finance experts or lawyers. This class of actors operates within a scientific or professional or expert organisational context. The second class of actors, at the second stage of the diffusion process, comprises those who endorse the innovation and promote its adoption. In the original diffusion models they comprised both officials (such as farm agency officials or health authorities) and private sector actors such as seed company representatives or pharmaceutical company representatives. They endorse the findings of the scientists and have their own incentives and strategic reasons to spread them. The third class of actors, at the final stage in the original models, is those who accepted the innovations in their actual working lives – the farmers and the doctors. Again they followed their own logic for the uptake of what had been developed and endorsed by others. For farmers it seems as though local knowledge drawn from the neighbourhood (fellow farmers) was critical, while for doctors it seems to have reflected professional

46 Ryan and Gross (1943) specifically warned against assuming that their findings necessarily applied to other fields. Meyer et al. (1999) suggest that diffusion between states is likely to be an idiosyncratic process. Bell and Staeheli (2001) also take a sceptical view of ‘waves’ of democracy. For a supportive view that there are such waves see Elkink (2011). Gleditsch and Ward (2008: 296) straddle the divide, ‘We believe our results allow us to fairly reject the idea that institutional change is driven entirely by domestic processes’.


15/12/2010 08:09

70


position (so-called ‘structural’ motivation rather than just ‘cohesion’ with other doctors in general).47 It is this general framework and its distinctions that are relevant to international rule making. The making of new international rules and regulations is one kind of innovation, and their diffusion between groups and across geographical boundaries is central to international rule making. The idea that the policy process can be usefully divided into stages is also helpful.48 It is well established as a methodology in the diffusion framework.49 There are further analogies to be drawn between the different roles of different actors and the different styles of reasoning that they employ at different stages in the diffusion process and in different organisational contexts. Adapting the Model In the case of new international rules, three groups of actors are critical at three stages of decision taking. First, there are those who take the lead in determining the substantive content of the new rules. They are usually experts or bureaucratic elites. Often they are officials, but increasingly they are drawn from the private sector or academia. The logic they follow is that of their professional discipline – that of the natural sciences or the social sciences or the law. They frame the policy problem to be solved, look for ‘solutions’ and they analyse and may choose between alternative approaches.50 At their best they secure the evidence base of policy making: at their worst they can be used to close down debate about the evidence. This comprises the first stage of international rule making. Second, there are governments or the representatives of governments who endorse the content of the rules and take the decision that they should be adopted internationally and in their own jurisdiction. Their role in the 47 See Coleman et al. (1966) and Burt (1987). However, advertising may be important in the medical context; see Van den Bulte and Lilien (2001). 48 Sabatier (1999) questions breaking the policy process into stages. His critique is refuted in Deleon (1999). 49 Rogers (1995) distinguishes between five main steps in the diffusion–decision process – knowledge gathering, persuasion, decision making, implementation and confirmation. The three-stage categorisation used above treats persuasion and confirmation within the three stages. Rogers (1995: 188) comments, ‘Stages exist in the innovation-decision process. The evidence is most clear cut for the knowledge and decision stages and somewhat less so for the persuasion stage’. 50 Strang suggests, ‘The sciences and professions are central to the modernizing project. They are devices for turning local and parochial practices into universally applicable principles that can “rationally” be adopted by all sorts of superordinate authorities, implemented by subordinate ones, and copied by modern entities everywhere’ (Strang and Meyer 1993: 503).


15/12/2010 08:09

Analytic frameworks

71

decision process is to mobilise action. In cases where membership in the rule-making bodies is limited to only a select number of governments or official bodies, such as central banks or health authorities or regulators, mobilising action involves getting the endorsement of a wider group of governments or officials who may not have participated in the original discussions. The logic at this second stage of endorsement and adoption is different from the reasoning that goes into the substantive content of the rules. The experts at the first stage are dealing with the uncertainties about the diagnosis of the problem and alternative solutions to it. Governments and officials at the second stage are dealing with the uncertainties associated with other actors – how other governments will react.51 As will be discussed later, behaviour will range from governments that want to join in taking a lead in international rule making to those who wait to follow others, and from those who see new rules as an opportunity to impose a standard on others to those who see new rules as an opportunity to learn from others.52 Third, there are citizens or electorates. Their involvement at the final stage in the diffusion process is likely to be fleeting, except in cases of rules they recognise as immediately salient to their own particular interests. In practice, the attention span given to politics by citizens is limited, participation is sporadic, many feel that the subject matter of politics is too complicated for them and the recognition of rules that are salient may be faulty. Nevertheless, in democratic theory citizens have the final say in accepting, rejecting or trying to alter the decision making of their governments. Despite the well-known limitations to participatory or deliberative models of democracy, citizens still seem attached to the idea of democracy as the ‘least bad’ alternative. In judging at this final stage what is salient and whether or not it is acceptable, citizens are likely to look to different sources of knowledge again. It may be from neighbours (like the farmers and hybrid corn) or from peer groups and the media (as with doctors and new medicines) or 51 Brunsson (2007) makes the important distinction that decision making should not just be equated with choice. The experts at the first stage can be seen as making choices between alternatives, but governments at the second stage are reacting to the alternatives chosen by the experts and the task for governments is to make their decision in the light of their expectations about the decisions of other governments. 52 Peter Haas (1993: 188) has argued against the distinction between epistemic elites and governments on the grounds that, ‘As epistemic communities obtain and consolidate influence in different governments, national preferences and policies will come to reflect the epistemic beliefs’. Blurring the distinction however probably attributes too much power and influence to epistemic elites on policy outcomes and underplays the many different ways in which policy makers in government relate to knowledge (see the discussion in Weiss 1979).


15/12/2010 08:09

72


it maybe from other types of political or social affiliation (such as with a political party or an NGO). The critical point, discussed later, is that the type of reasoning is not limited to that which has been deployed by scientists or economists in formulating the content of, for example, an environmental rule, nor is it the same as the type of strategic reasoning that may have been deployed by governments in endorsing a rule. In the same way as farmers wanted to see for themselves whether the adoption of hybrid corn made sense to them and did not necessarily trust the word of the scientists, private seed company representatives or official farm agencies, so people have a well-honed distrust of what politicians say is in their best interest and some distrust too of scientists and other experts.53 Comparing the Two Frameworks The importance of the diffusion framework compared with the multi-level governance framework is that it leads to a very different account of the effectiveness of international rule making and where it can go wrong. It also points to a very different account of the democratic deficit in international rule making and where it lies. As far as the effectiveness of international rule making is concerned, the multi-level governance framework places the emphasis on getting spheres of authority correctly aligned with the boundaries of policy and correctly combining vertical forms of authority with network efficiencies. It points in the direction of various forms of misalignments, misassignments and mismatching between spheres of authority, policies and instruments, as the likely source of policy failure. By contrast, the diffusion framework focuses attention on the group characteristics of those deciding on the substance of the rules, the organisational setting in which they make their decisions and the methods of reasoning they deploy. Thus, at the first stage of policy making, the diffusion framework focuses attention on the kinds of reasoning used by professional experts and elites in framing problems and their solutions and the settings that encourage a certain kind of behaviour among experts. This steers the analysis towards looking at the strengths and weaknesses of such groups and their settings. Hence, the diffusion framework opens up a new 53 The lack of trust seems to apply particularly to politicians. According to the 2007–08 Citizenship Survey of the UK’s Office of National Statistics (prior to the 2009 scandal over the expense claims of MPs) 31% of people in England trusted parliament ‘a fair amount’ and only 4% ‘a lot’. For a more general discussion see Dalton (2008: 237–47). An Ipsos MORI report dated October 5, 2009 found that 70% of those polled generally trusted scientists to tell the truth, much higher than the 13% who trusted politicians but lower than the 92% who trusted doctors and 88% who trusted teachers.


15/12/2010 08:09

Analytic frameworks

73

and different line of enquiry into what can go wrong with international rule making and why it is prone to systematic failure. The diffusion framework also sheds a different light on the question of the democratic deficit. The deficit is not seen to arise from the fact that experts are so dominant in decisions on the substance of rules at the first stage of the policy process. Democracies need to be able to mobilise knowledge and expertise no less than undemocratic societies. Nor is it seen to arise from the continuing centrality of national governments in international rule making at the second stage in the diffusion framework. Governments remain the dominant unit around which political life is organised throughout the world. What however is seen to be key is the question of democratic acceptance or dissent from the proposed rules when citizens become the centre of attention at the third stage of the process. The diffusion framework highlights the fact that the reasoning employed by citizens when confronted by new international rules is not necessarily going to be the same as that used either by experts in their reasoning about the substance of the rules or by governments in their reasoning about whether or not to endorse the proposed rules. It points therefore in the direction of looking at how dissonance is treated whenever the democratic voice gives expression to a different style of reasoning and diverges from the voice of government. The Role of Citizens In both the multi-level governance framework and the diffusion framework there is an important issue about how to model the role of citizens in politics and in particular their role in the formation of policy preferences. Both frameworks reject the state-centred view that governments can be assumed to reflect the view of their citizens from the ‘bottom up’ through some ready process of preference aggregation. However, beyond this point the assumptions are different. The multi-level framework assumes that citizens are actively engaged in civic activities and that through the associations of civil society they become actively engaged also in politics. Civil society organisations and other private networks thus become the transmission belts for the voice of the citizen. When they exercise authority they exercise it on behalf of the citizen. By contrast, in the diffusion framework there is no assumption that private actors such as companies or civil society organisations are necessarily aligned with citizen views. On the contrary, the model of participation sees companies as looking out for their own interests (for example in


15/12/2010 08:09

74


selling seeds or medical products) with arguments that farmers or doctors may or may not accept or be influenced by. In addition, the diffusion framework does not make any normative assumption about the degree of engagement of the citizen with the rulemaking process. Great importance is attached to social networks (for example those that bring doctors together) but no assumption is made as to whether or not these connect with politically active networks and, if they do, whether they would be reflective of anything other than a particular professional interest. This approach is consistent with the assumption that citizens act according to what is termed ‘bounded rationality’. This means that they focus on politics only when they see for themselves that something is salient to them. The empirical evidence suggests that strong assumptions about the connection between involvement in voluntary associations, civic activism and political involvement should be avoided. There is evidence of some association between engagement in voluntary activities and civic activism.54 There is also evidence of a connection between civic activism and political interest.55 However, this does not necessarily carry over into political engagement in a multi-level system. The disconnect has been shown in a case study of the European Union where political engagement as measured for example by voter turn-out at European elections is low.56 Preference Formation This different view of the role of citizens has implications for preference formation. The multi-level governance framework attempts to restore the idea of ‘bottom-up’ policy making through the deliberate creation of space for horizontal networks and through recognising the authority of private actors alongside the authority of governments. The assumptions of the diffusion model however are consistent with the view that international rule making is usually experienced in practice, in the real world, as a ‘topdown’ process for citizens.57 This view reflects an important distinction 54 According to the 2007–08 Citizenship Survey, 27% of people in England participate in formal volunteering at least once a month and 39% said that they had carried out at least one form of civic activity (such as signing a petition) in the 12 months before the survey (source: ONS Citizenship Survey 2007–08). 55 Van Deth (2010). 56 Van Deth (2010) concludes that, taking the individual characteristics of activists into account, neither the nature of voluntary activity nor objective organisational features appear to be particularly relevant for engagement in European affairs. 57 As long ago as 1973 Beer noted that science and technocracy ‘shifts the initiative in government from the economic and social environment of government to government itself. The old pressure group model that found the origin of laws and programs in demands arising


15/12/2010 08:09

Analytic frameworks

75

about the processes of policy making prior to decisions by governments and those that follow after. The distinctions between the different stages of the policy process in the diffusion framework enable the processes of preference formation to be divided between a pre-decision period when policy problems are framed and alternatives weighed and an after-the-decision period when the ‘solutions’ adopted by governments are actually experienced by citizens. When it comes to the framing of policy at the first stage of policy making, the diffusion framework allows for multiple actors to be involved. Policy ‘problems’ often emerge from adverse events, framed by media treatment of those events or by the reaction of those immediately affected by them, from framing by professional elites, including NGO elites, and from framing by governments themselves. The views of citizens are only one element, usually after some disaster has stirred public opinion. The role of experts is crucial. However, when it comes to the third stage of the policy process, after the experts have examined alternative policy approaches and governments have negotiated agreements on what to do, citizens are placed in the position of receivers rather than makers of policy. Before a decision has been made, citizen involvement in preference formation is one of many and not necessarily the most important element; after the decision has been made, citizens experience international rule making as a top-down process. Experience as a Second Round Effect Citizen experience of international rule making is also usually an attenuated one. In this connection there is a useful distinction to be made between the addressees of a rule, the targets of a rule and citizens themselves.58 In the case of international rules an addressee is usually a government or more often an agency of a government – for example a financial regulator. They are the ones with responsibility for implementing the rule. The target is the group whose behaviour the rule is trying to modify in the first instance – for example financial intermediaries taking excessive risks or money launderers. The way in which citizens usually experience international rules is through second round effects. They are not often the direct addressees of rules or the direct targets. These second round effects

outside government does not hold for technocratic politics. In technocratic policy making the pressures and proposals arise within government and its associated circles of professionals and technically trained cadres. In a democratic society, of course, the electorate must be informed and its consent won, but in recent times it tends less and less to be the source of policy initiatives’ (Beer 1973: 76). 58 See Börzel (2002) for the distinction between addressees and targets.


15/12/2010 08:09

76


might for example include having to prove personal identity for routine financial transactions as a product of anti-money laundering or antiterrorist financing rules or a rise in domestic utility bills as governments agree on deriving energy from costly renewable sources. Salience In these second round effects the salience of an international rule for an individual is often difficult to detect. An example can be given from the trend in recent years for corporations to bring an end to defined-benefit personal pension schemes where the pension benefits are related to the final years’ remuneration of an employee. In this case, professional investors, lawyers and accountants saw poor quality corporate accounting as a ‘problem’, and international accounting experts prevailed on governments to introduce rules to insist that assets were ‘marked to market’. Individual citizens experienced this regulation as salient only when informed that final salary pensions from their employer were being terminated (as a result of the company having to mark down the value of their assets and having to mark up the cost of their obligations in order to comply with the mark to market regulation). In this case citizens experience an adverse event on a matter of personal as well as social importance, they may be unaware of the international agreement on marking to market and unable to recognise the international body of accounting experts responsible for framing the policy problem and its solution (IASB). Their governments shelter behind an international agreement. The diffusion framework suggests that the experience of top-down policy making cannot be wished away by making normative assumptions about politically engaged citizens involved in ‘bottom-up’ processes of preference formation. In looking for the origins of the democratic deficit in international rule making there is a very basic difficulty to resolve. In democratic theory citizens are in some way or other the principals in the system. In practice, they experience international rule making at the receiving end of policies created elsewhere by rule makers they may not even recognise.59 The diffusion framework therefore allows for the potential dissonance between experts, rule-making governments and citizens to be explored as one of the fundamental issues underlying international rule making. *

*

*

59 According to the Citizenship Surveys conducted by UK’s Office of National Statistics only about 20% of people over the period 2001–09 feel that they could influence decisions affecting Britain (source: ONS Citizenship Survey April–September 2009).


15/12/2010 08:09

Analytic frameworks

77

The two analytic approaches that have been discussed in this chapter illustrate opposite sides of the original debate about the relative importance of form and process.60 Form and process should not however be seen as alternatives or as opposites. Both perspectives are important. In the further investigation of the reasons for the vulnerability of international rule making to failure and the reasons for the democratic deficit both frameworks are used. In order to carry forward the discussion, the next two chapters look at the mix of venues used in international rule making and the mix of instruments. Both are implicated in accounts of failures in rule making and in accounts of the democratic deficit. In looking at the role of different venues and instruments, considerations of both form and process have their place and the analysis incorporates both perspectives.

60 For an account of how attention has switched from a focus on institutional form to process in the area of public management, see Barzelay and Gallego (2006).


15/12/2010 08:09

4.

The choice of venue

INTRODUCTION An extraordinary mix of venues is used in international rule making. In May 1944 Keynes was aghast to learn that the United States intended to invite 44 countries to participate in the Bretton Woods conference that he was to co-chair. He complained to a friend and colleague that it would be ‘the most monstrous monkey house assembled for years’ and listed 21 of those 44 countries that in his view would not be able to contribute to the work of the conference.1 Yet, at an earlier point, when he had to defend the proposed international agreements in front of a political audience, he sang a different tune. He referred approvingly to the need for ‘consciousness of consent’ – meaning that it was desirable that all countries that were going to sign up to post war financial and monetary arrangements knew full well what they were doing and had agreed to the commitments they would be undertaking.2 Keynes’s ambiguity illustrates two very different sentiments about the form of venue for reaching agreement on international rules. In wishing the conference to be kept small and limited to those who could contribute to the substance of the discussion, he was expressing a preference for a venue where the experts could have fruitful talks. In referring to the importance of the ‘consciousness of consent’ he was enlisting a different principle – that a venue that included everybody was normatively ‘appropriate’ for an agreement that was intended to include all countries. Within three years of the Bretton Woods conference a still different approach to the choice of the form of a venue was illustrated when it became clear that the Soviet Union wanted to play by its own rules. When it came to administering aid for post war recovery, as mentioned earlier, the United States set up a limited agreement for Western Europe under the auspices of OEEC. In this case the self-interest of non communist countries came ahead of the principle of general inclusion. This chapter looks at two different accounts of why one venue may be

1 2

Keynes’s letter to Sir David Waley, May 30, 1944 (Keynes 1980: 42). Speech in the House of Lords, May 18, 1943 (ibid.: 269).

78


15/12/2010 08:09

The choice of venue

79

chosen for rule making rather than another and how the mix of venues emerges. The first looks at the different logics referred to above that lie behind different forms – the logic of appropriate form, expertise and selfinterest. From this perspective the mix simply reflects different organising logics. The second approach centres on the common sociological characteristics of those who inhabit the different venues. What is common is that they all belong to what are termed ‘epistemic communities’.3 Whether they inhabit a venue such as the OECD, that serves the interest of a limited group of about 30 countries, or the IMF with its universal membership of about 190, or an expert network such as the technical committee of IOSCO with around 18 members, there are overriding social and professional characteristics shared by the rule makers.4 This perspective gives a different account of the reasons for the institutional mix. The mix reflects the fluid ‘ecologies’ or fields of a professional world. The chapter also discusses the directions in which the two accounts of the mix point for a diagnosis of the sources of failure in rule making and for a diagnosis of the democratic deficit. Both accounts have a role in diagnosing the sources of possible failures in international rule making. For example, the different mix of professional venues may lead to frictions and lack of cooperation so that failures arise in the proper appreciation of a policy problem. Failure may also arise if governments assign a problem to a body with universal membership when an expert body is needed. Both accounts also have a role in diagnosing the possible sources of the democratic deficit. Bodies with universal membership that can claim to be representative of a full range of interests and values may not be effective or decisive actors compared with smaller groups with self-interested memberships. At the same time, expert bodies employ specialised kinds of reasoning associated with their particular discipline in order to reach their conclusions. This style of reasoning may not be consonant with the more diverse kinds of reasoning relied on by citizens and electorates in order to reach their decision about whether or not a policy is acceptable.

3 Ernst Haas (1990: 41) defines an ‘epistemic community’ as one ‘composed of professionals (usually recruited from several disciplines) who share a commitment to a common causal model and a common set of political values. They are united by a belief in the truth of their model and by a commitment to translate this truth into public policy, in the conviction that human welfare will be enhanced as a result’. This general definition is refined further in later discussion in this chapter. The limited concept of an epistemic community should be distinguished from the much more expansive concept of an ‘episteme’ that includes elements such as legitimacy and fairness norms (Adler and Bernstein 2005). 4 Peter Haas (2004) asserts, ‘epistemic communities are the transmission belts by which new knowledge is developed and transmitted to decision-makers’.


15/12/2010 08:09

80


On the surface these two accounts complement each other. The chapter closes however by identifying the very different underlying diagnoses they offer of where the problems of effectiveness and the democratic deficit really lie.

ORGANISATIONAL LOGIC Overview The three different approaches to the choice of form of venue in international agreements and rule making referred to above correspond to three different organisational logics.5 The first to be discussed reflects what is termed the logic of ‘appropriate design’. It is an approach that emphasises the normative elements in international rule making such as inclusiveness or representation. It gives the greatest priority to venues where all countries in the world are represented, such as the UN General Assembly, or other bodies set up with the intention of providing for universal membership, such as the Bretton Woods institutions. The second approach looks for institutions to serve the self-interest of the rule-making governments rather than to meet normative criteria such as inclusiveness.6 The outcome of this approach is likely to be special purpose venues such as the OEEC or G7 or G10 representing the wealthiest market economies, or the G24 representing the interests of emerging economies. It reflects what is termed the logic of ‘rational design’ where what is ‘rational’ is the pursuit of self-interest.7 The third centres on designing venues that are suitable for discussions among professionals and experts.8 These are venues such as the BCBS or IASB that bring together central bankers or accountants. Membership is limited to those with the relevant expertise, similar occupations and shared problems.

5 Peter Haas (1993) distinguishes between five different typologies. They involve subdivisions of the three discussed in this chapter. 6 In a variation of this approach Elsig (2007) discusses the EU’s choice of venue for trade negotiations in terms of the rational self-interest of the European Commission where selfinterest is defined in terms of ‘autonomy’. 7 Koremenos et al. (2001b: 762) state that the basic assumption is that, ‘states use international institutions to further their own goals, and they design institutions accordingly’. 8 March and Olsen (1998) make a basic distinction between the logic of ‘appropriateness’ and the logic of ‘consequences’. Expert venues follow the logic of consequences.


15/12/2010 08:09

The choice of venue

81

Appropriateness The logic of appropriate design guided America’s leaders in their approach to the original design for the post war world. The political and normative side of international rule making was uppermost in their mind. Their central concern was to achieve a lasting peace. In order to achieve this overriding goal they wanted all countries in the world committed to rules of good behaviour. In order to get this commitment they emphasised the substantive goals for peacetime organisations. They also emphasised the organising principle of ‘inclusiveness’. Roosevelt expressed the substantive goals for the post war world in terms of ‘security’. The one supreme objective for the future . . . can be summed up in one word: security. And that means not only physical security which provides safety from attacks by aggressors. It means also economic security, social security, moral security – in a family of nations.9

The commitment of all countries to the post war order was thus to be cemented by the appeal of its substantive goals and by the appeal of organisations set up to achieve such goals as the restoration of income growth, the erasing of fears of starvation, addressing the physical and mental toll of refugees and the aim to achieve fair income distribution for commodityproducing nations. At the time the founders were criticised for not going far enough in specifying the normative principles behind the post war order. As a result the UN Declaration of Fundamental Human Rights was added in 1948 to the UN Charter in order to correct this perceived omission. Its contents were intended ‘as a common standard of achievement for all peoples and all nations’.10 The appeal to substantive goals was not enough in itself. The founders, and particularly the United States, were also deeply wedded to the principle of ‘inclusiveness’ – the idea that membership in international institutions should be based on country representation and that all countries other than the defeated powers should be represented. The debate about inclusive country representation played out in a number of ways in American attitudes – in its list of invitations to the Bretton Woods conference that so much dismayed Keynes, in its acceptance of Argentina

9 Annual message of the President to the Congress, January 11, 1944 (Goodrich and Carroll 1947: 20). 10 Preamble.


15/12/2010 08:09

82


at the San Francisco conference that founded the UN, in its rejection of Russian proposals that the UN should represent the component states of the Soviet Union and in its internal discussions about the shape of the United Nations. Here, the original proposal drafted by Sumner Welles envisaged dominant regional organisations with a loose overall umbrella. Cordell Hull however prevailed in his view that the central organisation of states had to be the key body with regional commissions as subordinate offshoots.11 The reasons for the aim to include all countries were partly historical – the lessons that American leadership drew from the failure of the League of Nations and their desire to get away from the organising alternative of empire. Churchill might protest, as he did when he misunderstood the proposed provisions for UN Trusteeship mentioned earlier. But he had no choice. No American leader saw the US itself as the new imperial power. The principle of inclusiveness was however not just about history. It also reflected a view about consistency between norms – if rules of behaviour were to apply to all then they needed to be made in institutions that represented all. In addition, it reflected a norm of fairness – if the rules were to apply to all, each country needed to give its consent (as Keynes recognised). Thus, the founders of the post war system looked to align their substantive goals with consistent norms for organisational arrangements. The founders saw no conflict between a normative approach to international organisations and the effectiveness of such bodies. On the contrary, their concern was that countries would feel under no obligation to observe post war rules of behaviour unless they had been involved in giving their consent to the rules. Nevertheless it is the question of effectiveness that has proved to be the Achilles heel of the norms of appropriate design. Rational Design Whatever the historical and normative justifications, the principle of inclusiveness, central to the concept of ‘appropriate’ design, carries within it one huge inbuilt disadvantage. It introduces into international organisations with universal memberships all the divisions in interests, perceptions and values that exist in the world. Insofar as the organisations themselves provided a pulpit and platform for attitudes to be expressed, they provided an incentive for differences to be magnified and exaggerated. Thus the

11

See Hull’s memoirs (1948) for his view of regionalism versus country representation.


15/12/2010 08:09

The choice of venue

83

outbreak of the Cold War led to an immediate challenge to the principle of inclusiveness and a decision with the creation of the OEEC to follow a different organising logic – that of ‘self-interest’. The OEEC was established to serve the interest of the United States by not giving aid to a country (the Soviet Union) rapidly becoming perceived as a threat and potential enemy. It also served the self-interest of the recipients, by not diluting the amount of aid on offer by its dispersion to an unfriendly neighbour and to other countries in Central and Eastern Europe coming under its sway. Even though the major divide between communism and marketoriented economies has now disappeared with the collapse of the Soviet Union, many other differences remain that have continued to lessen the attractiveness of international organisations with universal and inclusive memberships as venues for rule making – differences for example between emerging markets and developed market economies, between democracies and authoritarian regimes, between rich and poor nations, between advocates of a multi-polar world and those who accept the United States as a hegemonic power. The need to find more ‘like-minded’ settings than those provided by universalist international organisations has therefore persisted. Like-mindedness can be provided by shared interests. It has been suggested that bodies that reflect the norms of appropriate design can still serve the self-interested goals of its members because there is always the possibility of agreeing on practical steps even if there is disagreement about the underlying justification for them.12 It has also been suggested that ‘self-interest’ is impossible to define in a highly uncertain world and that it is better to agree on a stable principle such as inclusiveness rather than to make guesses and constant adjustments about forms that might seem more effective.13 In practice however, ‘rational design’ finds continued expression in self-interested groupings such as the G7, G10, G24 and G77 as well as in the continued existence of the OECD (the successor organisation of the OEEC). The weakness of ‘rational design’ has not been in its departure from the normative approach of the original architects but because organisations serving a narrow set of interests also have their own practical limitations. A pursuit of self-interest tends to lead to international problems being viewed from the perspective of a zero sum game, where one side can only gain at the expense of the other. The divisions of the Cold War meant that a large part of international relations were viewed in precisely this way and in the even more adversarial way through the lens of ‘threat’.

12 13

Appiah (2006). See Wendt (2001).


15/12/2010 08:09

84


However, even in the circumstances of the Cold War there were many areas where these perceptions were seen as exaggerated and damaging. There was a need for alternative venues where the mutual gains to be had in negotiating agreements and coordinating actions could be appreciated and exploited but where the disagreements built into organisations with a universal membership could also be avoided. The result was a turn to a different type of organisation built on consensus rather than self-interest and deliberately ‘depoliticised’ – the expert organisation. Expertise The prototype for the expert organisation was the BIS whose history has already been referred to. It offered above all a ‘depoliticised’ setting where originally a small group of concerned central bankers could discuss German reparation payments after the First World War without the intervention of political considerations. The leitmotifs of such organisations include a division of labour and expertise between specific organisational tasks, ‘depoliticisation’ that aims to separate practical tasks from ideology and politics, limited membership and a belief in the possibility of cooperative agreements through consensus. Expertise is crucial to each of these aims. Expertise defines the task, expert knowledge is used to separate ‘non-controversial’ practical approaches from political value judgements, expertise defines the members and enhances the possibilities for cooperation and it is expertise that helps identify and build consensus solutions.14 The venues are likely to be small and specialised partly because it becomes much easier to establish practical goals in smaller rather than larger organisations.15 The idea that organisations such as the BIS can be ‘depoliticised’ can be criticised. Such organisations can however be considered as ‘depoliticised’ in limited but important senses. They aim to distinguish between the technical aspects of a question and the value judgements that may inform the eventual decision. They focus on the technical aspects. They keep the politicians at a distance from the technical venues. The participants see themselves as representing their profession or organisation rather than as representing a government. In the real world of international organisations examples can be found of each of these main types of approach to the choice of a venue. Illustrative examples of each are shown in Table 4.1. 14

See the discussion in Ernst Haas (1964). See the argument in Simon (1964) that goal formation in large organisations involves ‘radical simplification and approximation’. 15


15/12/2010 08:09

The choice of venue

Table 4.1

85

Organisational forms and organisational logic

Appropriate design

Self-interest

Expert venues

IMF/IBRD WTO UNESCO

OECD G7/G10/G24 ILO

BIS IOSCO IASB

In practice the demarcation lines are not as clear-cut as the table suggests. As already mentioned in respect of the ITU, WMO, WTO and WHO, many universalist organisations that follow the logic of appropriate design establish expert groups and committees that follow an expert logic. They may also adapt their organisational form to better reflect national self-interest (for example by changing membership shares or quotas or voting weights). There is also the impact of changing real world tasks and challenges that blurs boundaries between the different logics.16 Nevertheless there remains a difference between the decision making characteristics of different international venues. It fundamentally affects where rule making takes place. The Mix It follows from this account that the institutional mix that can be seen today in international rule making reflects in a straightforward way these differences in form and the competing logics that lie behind the differences.17 To this can be added the opportunistic behaviour of the venues themselves to expand their core business, to form alliances or to promote new activity streams and other forms of interaction.18 Even if there remains a debate about what precise weight to give to the influence of different institutional logics it is difficult to deny its real world importance. A recent example of how a mix of institutional arrangements develops is provided by the discussions about the possible desirability of a ‘New Bretton Woods’. It might be thought that the venue for such discussions would be found in the Bretton Woods organisations 16 For example March and Olsen (1998: 965) argue that ‘It seems very likely that rather little of the experimentation in international organization occurs because of a conscious organizational intent to experiment. It occurs . . . as an unintentional by-product of instrumental action’. 17 Koremenos et al. (2001b) offer a variety of hypotheses linking choices of venues to the kind of problem that is to be tackled. 18 See Kanter (1991) and Oberthür and Gehring (2006a) for different accounts of interinstitutional behaviour.


15/12/2010 08:09

86


themselves. The logic of appropriateness would suggest that any major change in global arrangements should be discussed by all countries. In practice the forum chosen for setting the main guidelines for policy after the international financial crisis has been the G20. This choice represents a logic of self-interest by the major economic powers in the world. At the same time, the choice of the G20 is not the end of the story since the G20 has limited capacity to undertake the work itself. For professional back-up it can look to some of the expert professional venues such as IOSCO clustered around the BIS. However, the IMF also lays claim to expertise in relevant areas such as the relationship between macro policies and micro policies and has staked its place. The end result of institutional rivalries is a sharing of tasks between the FSF and IMF and other institutions, with FSF membership being enlarged to reflect the membership of the G20 rather than the G10 and renamed as the Financial Stability Board (FSB). There is a completely different way of looking at this whole scene. The alternative approach focuses on the common characteristics of the actors within the institutions, What stands out above all else is the dominance of professionals and professional relationships.19 According to this perspective, the common elements uniting the actors transcend the kinds of differences in institutional logic described above. Each of the varieties of venue discussed earlier would claim that it is professional in its approach – regardless of the particular type of professionalism involved, whether that of agricultural experts or health experts or trade experts and regardless of whether they belong to a organisation with universal membership or one with a limited membership. The bureaucrats who inhabit international organisations are recognisable as a tribe – not just by the size of their pay checks but by the way they look, the way they talk, the way they behave and by the way they approach problem solving.

19 Abbott (1988: 8) defines professions in ‘loose’ terms as ‘Exclusive occupation groups applying somewhat abstract knowledge to particular cases’. DiMaggio and Powell (1991c: 70) define ‘professionalisation’ as ‘the collective struggle of an occupation to define the conditions and methods of their work, to control “the production of the producers” and to establish a cognitive base and legitimation for their occupational autonomy’. Freidson (2001: 12) defines ‘professionalism’ in similar terms: ‘Professionalism may be said to exist when an organized occupation gains the power to determine who is qualified to perform a defined set of tasks, to prevent all others from performing that work, and to control the criteria by which to evaluate performance’.


15/12/2010 08:09

The choice of venue

87

PROFESSIONALISM The Terrain When governments established the institutions and rules for the post Second World War world and as markets were re-established, a huge terrain was created for professional skills and expertise. It was not new terrain, but it was an enormously expanded terrain.20 Economists, agricultural experts, trade experts, lawyers, accountants and auditors, nutritionists, human and animal health experts, central bankers, environmentalists and many other professions and disciplines all found a vastly expanded international arena for their services. They were presented with some new venues by governments and helped establish others for themselves. They dominate the world of international institutions, its culture and flavour. Viewed from this perspective, the world of international organisations is primarily about institutionalising professional expertise and mobilising knowledge. Form is secondary. Professional boundaries also overwhelm the divisions between the public and private sector. In a venue such as the Egmont Group the public sector defines the rules but the private sector is at the forefront of enforcement. In a venue such as the IASB, the private sector leads the formulation of the rules but governments take the lead on enforcement. Epistemic Elites As mentioned above, this world of professionalism has been characterised as one dominated by ‘epistemic communities’ or elites. There are four defining and unifying characteristics of such elites which carry across all professional disciplines.21 First, the experts share certain principled beliefs. This means that they share a desire to base rule making on evidence that is acceptable to the profession to which they belong rather than on political sentiment and believe that an ‘evidence base’ exists that can be distinguished from political values.22 As already mentioned, ‘depoliticisation’ is limited and the 20 Fourcade (2006) identifies three factors behind the internationalisation of professions; first, the growth of an international market for professional services, second, the diffusion of norms that destroyed barriers in previously localised professions and third, the emergence of international jurisdictions. 21 These distinctions are based on Peter Haas (2010). 22 This evidence base may not be explicitly articulated but derived instead from the knowledge that comes from doing the same things in their role as, for example, central bankers or market regulators (see Collins 1974).


15/12/2010 08:09

88


rules the experts come up with on the basis of the evidence may be deeply political in their implications. But this they can leave to the politicians. In the meantime the experts are likely to believe that a depoliticised approach will contribute to the general welfare. Second, the experts share causal beliefs. This means that there will exist a closeness between the participants in agreeing on how to frame and define the problem to be solved, a common understanding of how it might be approached and above all a shared sense of where the solutions might lie combined with a willingness to look for those solutions. Third, they have common notions of validity. This means that they feel able to differentiate between claims about the world – between those that are valid and permissible as evidence, or conclusions purportedly drawn from evidence, and those that are not valid or permissible. Fourth, they share a sense of being involved in a common problemsolving enterprise that brings them together in policy proximity despite possibly representing different disciplines within a field or having different perspectives within a field. Policy proximity unites the club or college with a sense of shared problems, with shared solutions and with shared benefits from cooperation. Taken together these unifying characteristics enable such groups to be collegial and ‘like-minded’ in their approach to tasks and to be consensus oriented. These qualities enable them to see eye to eye in defining policy problems, their solutions and in arriving at an agreed approach to them. The reason for the salience of like-mindedness to rule making is that it makes the gains to be had from cooperation much easier to compare.23 Participants see themselves coming from more or less the same starting points in perceptions of whatever is the problem and likely to see the potential gains from agreement in more or less the same terms. There is less need to introduce more contested terms into an agreement such as what is fair or equal. The actors share professional mind sets, disciplines and background values and can empathise with each other. An emphasis on the common features of epistemic elites appears to downplay the extent to which experts often disagree about problems, their diagnosis and solutions. However they operate within ‘communities of practice’ where the incentives to cooperate, to look for points of agreement and to reach consensus are strong.24 The rewards often come from

23 Binmore (2005: 31) notes that from a game-theoretic perspective, ‘like-mindeness’ increases the chances that the maximum pay-off from any bargaining will be chosen among the different possible equilibrium points. 24 For the term ‘community of practice’ see DiMaggio and Powell (1991c). See also Bowker and Starr (2000).


15/12/2010 08:09

The choice of venue

89

cooperation. Differences and disputes that may rage between academic experts become suppressed when communities of practice have an incentive to act together. Nevertheless there are frictions. It is these frictions that also help account for the institutional mix seen in international rule making. Accounting for the Mix – Professional Ecologies At first sight the emphasis on common features of the behaviour of epistemic elites appears to reduce the relevance of the divisions between institutional forms discussed earlier.25 However, in practice the professions add to the mix as much as they subtract from it. Professionals have their own divisions and their own ways of relating or what has been termed their own ‘ecologies’.26 They have taken the lead in establishing many of the newer ‘network’ organisations and they have their own professional dynamics that stretch across and add to other institutional divides. It is these professional dynamics that also contribute to the mix of institutions that can be seen today. The key point about professions and professional behaviour is that boundaries are fluid and attempts from within a profession to ‘control’ the boundaries or to exercise ‘authority’ over entry, footnoted earlier, are likely to be doomed.27 Boundaries are permeable and any one discipline is susceptible to influence and penetration by another.28 Economics provides one of the most fluid examples as, over the last half century, it has used the methodology of economics to explore questions in political science, incorporated game theory and rational choice theory, cooperated with scientists in environmental economics and co-opted the work of psychologists in the development of behavioural economics. Accountancy provides another.29 The behaviour of professional disciplines is thus about competition and conflict over the control of work.30 It is about division and compartmentalisation but also about alliances, co-option and cooperation. It is about 25 DiMaggio and Powell (1991c) treat professionalism as one of the normative forces, along with coercive and mimetic forces, behind the convergence on common forms. 26 Abbott (2005) defines an ecology in terms of three components: actors, locations and a relation associating one with the other (professions with their tasks). In his account the relational process is prior. 27 Abbott (2001) refers to professional disciplines as a ‘set of communities dividing and subdividing’. 28 A slightly different account of how professionals organise themselves, compete and transform themselves is offered by the theory of ‘fields’. See Fligstein (2001). 29 See Miller (1998). 30 Abbott (1988: 19).


15/12/2010 08:09

90


Table 4.2

Examples of professional inter-institutional relationships

Actors

WHO/OIE

WMO/UNEP

FSB/BIS/IOSCO IASB/IAIS

IMF/FSB

Policy arena Behaviour

Pandemics Cooperation

Climate change Alliance

Financial rules Co-option

Surveillance Competition

entrepreneurial opportunities for professions.31 What this means is that professional churning has its own impact on the way international rule making takes place and the way venues form and interact, as well as on the substantive content of the rules.32 Table 4.2 illustrates various examples of inter-institutional relationships in different fields of international rule making. What it suggests is that the mix of forms and venues seen in international rule making today is likely to reflect not just the different logics of form as seen by governments but also different types of professional behaviour. In the table, human health experts in the WHO cooperate with animal health experts in the OIE in the assessment of the dangers associated with possible pandemics; UNEP and WMO form an alliance through the IPCC in the assessment of climate change; FSB co-opts the work of a number of specialised financial organisations in drawing up the overall principles of financial best practice, and the IMF and FSB are shown in competition in multilateral surveillance.

APPRECIATION AND DECISIVENESS The two accounts given above that help explain the variety and mix of venues used in international rule making can be used to illustrate different types of difficulties that may lead to failures in rule making. The account that looks at professional interrelationships focuses attention on possible failures of appreciation. Situations are misread, misinterpreted or misunderstood because the institutional boundaries created by professions may stand in the way of the right mix of professional assessments. The account that looks at the logic of forms focuses on possible failures in decisiveness. A situation is appreciated correctly but an institution cannot respond decisively because, for example, the logic of appropriateness means that conflicting interests cannot be resolved inside the institution. 31

Burt (1992). Barnett and Finnemore (2004: 158) also identify the professional motivations behind international organisations as ‘agents in their own change’. 32


15/12/2010 08:09

The choice of venue

91

Failures of Appreciation An example of a possible failure of appreciation is given by the 2008 international financial crisis. One aspect of the crisis was that the link between ‘macro’ economic risk (the appearance of various asset bubbles such as in housing) and ‘micro’ risk (the adoption of risky funding and asset acquisition policies by individual financial institutions chasing higher yields in a low interest rate environment) was not properly assessed. The question that arises is whether the professional boundaries between the securities market regulators in the IOSCO, the central bankers in the BIS and the macroeconomists in the IMF stood in the way of an overall appreciation of the imminent crisis. In the aftermath of the crisis, in order to better address the problem of coordination across boundaries, the FSB has been upgraded in importance. Another example is given by the health sector. Novel forms of virus that cross boundaries between species require the bringing together of both animal health experts and human health experts in order for the risks to human health of any new form such as found in H5N1 (avian flu) or H1N1 (swine flu) to be properly assessed. The expertise is divided between two groups of professionals – those in the WHO and those in the OIE. Any failure to cooperate across these professional boundaries will result in failure to assess the risks accurately. The avian flu epidemic was arguably not initially assessed accurately, in part because of less than optimal cooperation prevailing between the time of the original isolation of the virus in geese in 1996 and the first human fatality in 2003.33 Failures of Decisiveness Failures of decisiveness are highlighted by accounts of the logic of forms. In particular there is a well-known and long-standing tension associated with the logic of appropriate form. Because, as mentioned earlier, universal membership in an organisation brings with it all the divisions and interests in the world it may be very difficult for such an organisation with universal membership to act decisively even if the situation has been appreciated in the full. Marshall Aid provides the classic example. There was little question that a rapid infusion of capital was needed into Europe after the war, but decisive action through the UN was hugely complicated by the outbreak of the Cold War. The history of the UN provides many other examples of failure to act in the face of overwhelming evidence of the need to act.

33 A formal framework agreement was put in place between the two organisations in 2004 to try to ensure that cooperation takes place.


15/12/2010 08:09

92


In one sense these two ways of looking at the possible sources of failure are complementary and both accounts may be valid. The mix and multiplicity of venues lead to failures both of appreciation and of decisiveness. But underlying the two accounts are profoundly different ways of looking at failure. Failures to align boundaries between institutions and tasks in order to achieve a proper appreciation of a situation or a decisive response to it can be seen as what is termed in later discussion a type of ‘executive’ failure. The institutions themselves, or the member governments, fail to recognise or to overcome boundary difficulties. Failures of appreciation can however also be seen as symptomatic of a very different underlying type of failure – cognitive failure – where it is not the professional boundaries of an organisation that are at the root of the problem, but the way in which professionals approach their task within the organisations. They are vulnerable to bias both in the way in which they build consensus and in the way that they validate that consensus. Democratic Values The two accounts of the mix of venues used in international rule making also lead to a different perspective on the nature of the disjunction with democratic norms. A focus on the logic of forms immediately invites consideration of a trade-off between effectiveness and democratic values. Bodies following the logic of ‘appropriate’ form with fully inclusive membership are not bodies that can act decisively when needed. Conversely, if decisive action is needed then consideration of norms such as fairness or representation or inclusiveness seem to become secondary. A focus on the mix of venues as a reflection of a competitive and fluid professional world invites no such trade-off. International rule-making bodies are to a large extent seen as ‘knowledge bodies’. Knowledge bodies derive their legitimacy from the professional standards, procedures and principles of scientific and social investigation that they follow. Professionalism in social and scientific behaviour is not primarily about the norms of representation or fairness. Only if a lack of adequate representation leads to significant omissions in diagnostic information, or to inadequate consideration of the different inferences that can be drawn from diagnostic information, is representation or inclusiveness an issue. It is the integrity of the process and the investigator rather than fairness that is the relevant norm.34 According to this view, the search for the sources of the democratic

34

See the discussion in Vibert (2007).


15/12/2010 08:09

The choice of venue

93

deficit must look much beyond the internal characteristics of international venues to other factors in the policy process. Professionalism is relevant not because of the way in which it is organised but because of the way in which it communicates. It uses particular types of reasoning that may not be shared either by governments or by citizens.35 Eventually these different styles of reasoning have to be brought together. When thinking about how this is to be achieved, institutions come back into the analysis. This is because people are accustomed in their national settings to see a variety of institutions helping to bring together professional reasoning with broader societal concerns and with their own concerns. The types of organisation that have been set up internationally may not, however, provide the kind of institutional support that people expect or are familiar with. *

*

*

The next chapter looks at the instruments in which international rules are conveyed. Transmission processes are a further potential area where rules may fail. In addition, transmission also plays a key part in the account of democratic deficits. Once again two different perspectives are compared. One emphasises the importance of the differences in form that instruments can take, the other emphasises the contexts and processes that lie beneath the forms. Yet again these two different accounts lead to different diagnoses both about where rule making can go wrong in transmission and about where the democratic deficit actually lies.

35 An example of professional communication cited later is that of the 2006 BIS guidelines on capital standards. As far as citizens and most politicians are concerned these could have been written in ancient Egyptian hieroglyphs for the amount of information conveyed to a non-specialist.


15/12/2010 08:09

5.

The choice of instruments1

ARITHMETIC OR ALGEBRA When Joseph Stalin met Anthony Eden, the British Foreign Secretary, in Moscow in December 1941 to start discussions on cooperation and post war goals, Stalin expressed a firm desire for ‘arithmetic rather than algebra’.2 The notorious scrap of paper presented by Churchill to Stalin in October 1944 dividing Europe into areas of control according to percentages of influence was one outcome.3 Many appalling charges can be laid against Stalin, but in his choice of arithmetic rather than algebra he was pinpointing an issue that permeates all international agreements. It is about how to express an international commitment – with symbolism or with action, with precision or with flexibility, with specificity or with ambiguity and room for interpretation. It is an issue that applies to treaties dealing with war and peace. It is an issue that applies equally to questions of regulating the global environment and international financial markets. In current debates the possible need for binding numerical commitments permeates discussions about international agreements on measures that might help mitigate climate change.4 In the Cold War era, game theorists sided with Stalin. They too argued that precision was superior to flexibility. This was because they equated precision with the credibility of the commitment of the signatories to the terms of the agreement.5 The end of the Cold War has led to a complete recasting of this debate in terms relating to the uses of ‘soft power’.6 Soft

1 Shelton (2000: 5) defines ‘instruments’ as the varieties of texts in which norms, obligations and standards are contained. 2 Quoted in Smyser (1999: 6). 3 Percentage divisions included Hungary 50% Moscow/50% West; Romania 90% Moscow/10% West. 4 Rayner and Malone (1998: 113) comment, ‘It is even worth questioning whether targets and timetables are the best way to achieve emissions reductions at all. Essentially they represent the end-of-pipe approach to pollution at the global scale. This approach is increasingly recognised as obsolete for environmental management at local and national level. Why does it remain the favoured tool for global environmental governance?’ 5 See the discussion in Shepsle (1991: 246) ‘Discretion is the enemy of optimality, commitment its ally’. 6 See Nye (2004), the inventor of the term, for a discussion.

94


15/12/2010 08:09

The choice of instruments

95

power points to the wider range of instruments available for pursuing international goals when the focus shifts from the use of force or instruments such as economic sanctions backed by the threat of force. In this new, less confrontational scenario, ‘hard’ law remains the benchmark of the credibility of an international commitment but ‘soft’ law allows for compromising over time and for collective learning.7 Using this kind of terminology, a treaty will be seen as ‘hard’ and a recommendation or statement of principles as ‘soft’. This chapter looks in greater detail into the way in which participants choose to frame and transmit international agreements on rules and regulations. Traditionally, international rules and regulations have come either from treaties between states or from ‘customary’ international law – defined as unwritten rules that are binding on states but where generally applicable treaties do not exist.8 Nowadays, as mentioned earlier, international agreements are expressed in a huge variety of forms in addition to treaties. Treaties themselves, referred to above as ‘hard’ instruments, also can and do contain many forms of ‘soft’ obligations.9 The array of instruments employed and the terminology used to describe them are as bewildering as the mix of venues that are involved in their formulation. What is even more confusing is that agreements may also change their legal form as they move from the international level to the national and local. Form, Authority and the Logic of Context The chapter offers two different accounts of the mix. One relates the choice of the form of agreement to the new styles of authority in systems of governance. In particular, it is argued that the need to link traditional vertical forms of authority with the newer horizontal forms of authority creates a need for new instruments.10 Hard law can be chosen when vertical authority is strong; softer instruments may be chosen where the scope of vertical authority is less clear. The other account centres on two different types of reasoning that enter into international agreements. There is one type of reasoning that is about reconciling the very different types of motivation that lead countries to the bargaining table about new international rules of behaviour. Governments bring different degrees of motivation to the bargaining

7

Abbott and Snidal (2000). Byers (1999). 9 Shelton (2000). Downs et al. (1996: 383) note, ‘states choose the treaties they make from an infinitely large set of possible treaties’. 10 See Shelton (2000) and Mörth (2004). 8


15/12/2010 08:09

96


table and different goals that have to be mutually adjusted in order for an agreement to be reached. For example in the case of negotiations relating to climate change, India and China may not wish to see their growth prospects hampered by new investment burdens. On the other hand, European countries may wish to lead the way towards a low carbon economy and do not want to see their competitive position undercut. One way or another, common ground has to be found in order for an agreement to be reached. There is also another type of reasoning that refers to the setting for the implementation of any agreement. Governments do not only have different goals, they also have different views as to what they are prepared to do themselves in order to implement their goals and what they want others to do. Implementation is a central concern in any agreement on international rules of behaviour because an agreement becomes largely meaningless if there is no prospect of it being put into practice. Concerns about implementation can be set aside only if an agreement is seen worthwhile simply as a declaration of future intent or in order to stand on its own merits just as a statement of principle. The type of reasoning that goes into aligning strategic goals and the type of reasoning that goes into implementation are different. However, in order to reach any agreement, different goals and different approaches to implementation both have to be brought into alignment. According to this account the use of a wider variety of instruments reflects the difficulties in bringing these two types of reasoning together. Once again, the two accounts lead in different directions for diagnosing both the nature of the management deficit and the source of the democratic deficit in international rule making. A focus on form points to failures that stem from any mismatch between the choice of the instrument and the form of authority needed or desirable. The failures may be ones of effectiveness where, for example, the substance of an agreement cannot be achieved because the form of instrument is ‘too soft’ and does not allow for authoritative implementation. Or, the failures may reflect a lack of consonance with democratic norms where, for example, soft instruments simply disguise where and how authority is really being exercised. A focus on the need to link together the different reasoning that enters into negotiations provides a totally different perspective on the reasons why internationally agreed rules fail. According to this account the kind of reasoning that motivates governments in negotiating international agreements differs from the kind of reasoning of experts that goes into choosing the substantive content of the rule. The reasoning that citizens might apply when they experience the rules is different again. According to this account the origins both of failures in effectiveness and of the democratic deficit


15/12/2010 08:09


97

can be identified in the problems of ‘translation’ or ‘transfer’ as an agreement moves from one context to another – from experts to governments, from governments to the targets of the regulations and finally to citizens themselves.

NEW FORMS OF AUTHORITY: NEW INSTRUMENTS In recent years there has been both a much wider range of instruments used for expressing international agreements and, at the same time as described earlier, patterns of authority have changed as horizontal networks and private actors have gained in influence. It is natural to link the two developments together and to hypothesise a straightforward causal relationship – new forms of authority have called for new forms of instrument.11 The terminology of ‘hard’ and ‘soft’ seems to lend itself equally naturally as a way to characterise the wider range of instruments involved and their relationship with different types of authority. Measurement One of the main difficulties in establishing the relationship between new forms of authority and the greater use of soft instruments arises from the difficulty of measuring what is ‘hard’ and what is ‘soft’. The names and labels attached to different kinds of instruments are themselves extensive and confusing. They range from treaties to agreements, to conventions, protocols, declarations, codes, agreed principles, recommendations, concordats, memorandums of understanding and guidelines. These and other terms are all used as though the distinctions were clear and established. It is however generally agreed that these terms do not give clear guidance as to whether an international agreement is ‘hard’ or ‘soft’.12 The idea that there is a continuum from hard to soft that can be designated by these terms is also challenged.13 How ‘hard’ or how ‘soft’ is a question of form, content and context. Labels can be misleading.

11 See for example Mörth (2004) who links the use of soft law to the use of networks. Schwartz (2005) links the use of ‘guidelines’ to a desire to encourage ‘self-governance’. Chinkin (2000) notes an association between soft law and the rise of non-state actors with authority, both as a source of soft law and as a target of soft law. 12 See Chinkin (2000: 37): ‘It is axiomatic that neither the form nor the nomenclature of an instrument is determinative of its legal status’. 13 ‘It is impossible to draw a unidimensional continuum from weak to strong enforcement and to place each of the instruments at some specified and stable point’ (Bennett and Raab 2006: 214).


15/12/2010 08:09

98


In order to try to overcome these difficulties there are attempts to combine different features of an agreement into a common measure. One well-known suggestion is that a continuum from hard to soft measures can be mapped along three different and independent dimensions – the degree of obligation imposed, the precision with which the rules are framed and the extent to which their implementation and interpretation are left to third parties (such as an independent disputes settlement panel) rather than to the contracting parties themselves.14 But degrees of obligation may still rest in the eyes of the beholder, precision can be spurious and implementation cannot avoid different compliance contexts.15 Perhaps the most that can be said for the terminology of ‘hard’ and ‘soft’ is that it stands as a kind of shorthand for indicating the range of interpretative latitude that surrounds an agreement. Notwithstanding these difficulties of measurement, there remains the hypothesis that a link can be established between the use of a greater variety of different instruments, whatever terminology is used to describe them, with the need to blend vertical and horizontal forms of authority. Causality The plausibility of a direct and straightforward causal connection between new forms of governance and the use of new instruments rests on two aspects of multi-level governance. One aspect concerns ‘scale flexibility’ and the need to define new relationships between different levels of authority. Such a need arises whenever new types of policy spill-over have to be addressed, or new collective or common goods provided, that blur traditional policy responsibilities. When this occurs there will be sensitivities in both higher and lower levels of a hierarchy and a possible desire to avoid hierarchical connotations altogether. Even in the case of hierarchical relationships, lower levels may not want to be seen to be subordinated in the new area or bypassed as other actors forge new links. Higher levels may also be wary of confrontation with other jurisdictions. The result may be for all sides to look for new policy instruments and new terminologies that achieve cooperation in practice while putting more fundamental jurisdictional questions aside or in abeyance. For example in the case of IOSCO’s fight against cross-border securities market misconduct, the instrument of agreement is a ‘Multilateral Memorandum of Understanding’ (MMoU) – a new style of instrument. 14

Abbott et al. (2000). In the context of the EU, Falkner and Treib (2008) refer to four different worlds including ‘the world of dead letters’. 15


15/12/2010 08:09


99

The use of the MMoU avoids any hierarchical connotations. It also suggests a degree of voluntariness in compliance. In practice however, signatories to the MMoU may well have to change their own traditional legal frameworks in order to meet the requirements of the MMoU and may face pressures from other signatories if they do not. A different example that also attempts to avoid hierarchical connotations, drawn from the EU’s ‘open method’ of coordination, is discussed later below. The other aspect concerns the relationship between public authority and private authority.16 Again there are sensitivities on both sides. When private actors exercise authority, regardless of whether they are nongovernmental organisations (NGOs) wanting new standards of behaviour from multinational corporations or professional or private interests wanting new accounting standards, they may wish to preserve their seeming independence from government. They want to exercise their authority in their own way for their own reasons. Governments on their side may also be reluctant to be seen by the public as too close to any particular private interests, including so-called ‘civil society’ interests. When private interests are the target of government measures, governments may not wish to appear heavy handed and to wish to respect the way they exercise authority on their own. In each situation there is motivation on each side to look for new instruments and new terminology to achieve practical results while leaving aside ‘constitutional’ questions about who exactly is wielding authority on behalf of whom. The terminology of the new instruments such as guidelines, principles, recommendations or codes is particularly appealing in this context. Tensions in the Choice of Instruments At first sight this account of the links between the use of a wider range of instruments and rise of new forms of authority is highly plausible. But there are a number of tensions in the account. They centre around the need to align the legal content of the instrument with the form of authority needed to stand behind the implementation and enforcement of the rule or regulation.17 The first source of tension is that horizontal actors, including private actors, may lack authority and credibility to monitor compliance and to exact penalties for non-compliance. The need for hard law comes back 16

Chinkin (2000). See the discussion in Diver (1983) and Vervaele (1999). Zürn refers to the problem of compliance between vertically and horizontally organised sources of authority as the ‘Achilles heel’ of multi-level governance (Zürn 2005: 5 and 37). 17


15/12/2010 08:09

100


in, together with the accompanying need for vertical authority providing institutional mechanisms for monitoring, inspection and enforcement of an administrative, judicial or quasi-judicial kind.18 For example, the IASB, a private network body, can agree on the substance of international accounting measures but it depends on the legal authority of governments and their agencies to implement and enforce its agreements.19 The tension was referred to earlier in reference to the so-called ‘shadow of hierarchy’. What this means in the context of implementation is that even where apparently soft instruments such as ‘principles’ or ‘standards’ are used, they may be no more than a ‘front’ or veneer. Behind the soft instruments lies the iron fist of traditional authority. If the soft instruments work or are respected, it is because of a threat lying behind them that hard instruments can and will be used if necessary. Soft instruments may thus be the rather deceptive face of hard instruments in the background.20 A second aspect of the difficulty of postulating clear relationships in the field of enforcement between the need to connect with horizontal actors and the use of new instruments can be illustrated by the case of fisheries conservation. In this sector, NGOs are highly active, both national and international. Private actors are also involved, such as non-government scientists who estimate fish stocks as well as other private interests (the fishermen and fishing companies). Government is involved at various levels (state and federal in the US and the EU institutions and member states in the EU). So are their agencies. The key however to bringing together these different actors representing both vertical and horizontal authority, private and public authority in fishery conservation experiments seems to be hard law. It seems to be hard law that is needed to set up protected marine fisheries reserves, hard law that establishes various forms of property rights regimes for the buying and selling of fishing permits that enable the reserves to be managed effectively by private or NGO bodies, and hard law that enables NGOs and private bodies to enforce contracts.21 In other words, new actors with authority, including private and NGO

18

See the discussion in Zürn (2005). Tamm Hallström (2004: 138–9) found that the ISO and the International Accounting Standards Committee (IASC) used four different strategies in order to exercise authority (achieve compliance). Only one of these (persuading users of the benefits of their standards) was of a purely voluntary nature. The other three strategies (through ways of organising, through borrowing authority from others and through using the enforcement powers of others) all potentially or actually involved relationships with those with traditional, hierarchical authority. 20 Gunningham and Sinclair (2002: 205) note that ‘Many less interventionist strategies are unlikely to succeed if they are not underpinned by direct regulation’. 21 For a description of various conservation strategies in this sector see Clover (2004). 19


15/12/2010 08:09


101

actors, may be just as dependent on old instruments as the old vertical authorities. A third source of tension in the enforcement area in establishing clear links between new forms of governance and legal form concerns behavioural incentives for the targets of rulemaking. The idea of scale flexibility tends to leave out of account the need to weigh the behavioural incentives of the different authorities whose behaviour is to be brought into alignment in new policy spaces. Legal theorists argue that the more important the content of the rule and the more that an agreement requires the signatories to depart from behaviour they would otherwise have engaged in, the more important it is that the obligations are expressed in a precise, binding form, that the obligations are monitored and that sanctions and penalties follow from non-implementation. Such conditions make it more likely that vertical forms of authority will be needed.22 However, the assertion of vertical forms of authority in new policy areas is likely, in its turn, to reintroduce clashes between jurisdictions over competing claims to authority. The tension can be illustrated by the case of the EU where there has been a long-standing debate over whether the new form of the so-called ‘open method’ of coordination matches the substantive changes needed in the policies pursued by EU member states in order to compete effectively on international markets. The debate has rumbled on because the open method relies on techniques such as review and persuasion (associated with horizontal forms of coordination and policy making between member states) rather than deadlines, sanctions and penalties (relying on the exercise of centralised, vertically organised authority by the EU institutions). In this example, the choice of instrument for a new area of collective policy making precisely reflects sensitivity about the exercise of vertical authority in areas where authority is disputed, in this case between the central EU bodies and the member states. Yet the choice of new instruments in this case seemed to some observers as completely mismatched with the fundamental importance of the ability to compete and to the many changes required in member state behaviour. The effectiveness of the ‘open method’ also depends on other factors that are relevant to compliance such as long-term relationships between the signatories to an agreement and the reputation that the signatories put on the line when they make a public commitment to a goal. These and other factors may 22 See the discussion in Downs et al. (1996). Abbott and Snidal (2000: 429) assert ‘states should use hard legal commitments as assurance devices when the benefits of cooperation are great but the potential for opportunism and its costs are high’.


15/12/2010 08:09

102


encourage compliance with substantively onerous rules without the need for penalties and sanctions.23 Yet for many observers the ‘stick’ of reputational shaming and the ‘carrot’ of role modelling in the open method are not sufficient to achieve EU objectives. What this case illustrates is that the behavioural changes involved in trying to achieve scale flexibility in new forms of governance cannot easily be overcome by the use of new instruments. That recent years have seen a growth in the use of a much wider set of instruments for expressing international agreements and, at the same time, a greater variety in the actors with authority in governance, is not in dispute. What is at issue is in what ways the two developments are associated – as a way to express new forms of authority, or as a way to camouflage old forms of authority; as a way of incentivising behaviour to form new relationships; or as a way of disguising old behaviours that maintain old relationships. Because the association is far from straightforward there are questions about just how much it helps in identifying the possible ways in which rules fail and democratic norms get ignored. A focus on the different contexts in which rules are agreed and applied provides a completely different perspective. According to this account, the choice of instruments is all about the circumstances of adoption and enforcement.

LINKING LOGICS – ENDORSEMENT AND IMPLEMENTATION When parties to an agreement, public or private, governmental or nongovernmental, choose the form their agreement will take, that form will in part reflect the ease or difficulty they have encountered in reaching a common understanding about a problem and what to do about it. It will also reflect their views on implementation – their own willingness or ability to implement the agreement and their assessment of the willingness and ability of other participants to implement the agreement. The focus of the reasoning is quite different. One is about negotiations that try to align the variety of motivations that bring governments to the bargaining table. The other is about the negotiations that will eventually take place ‘downstream’ about compliance between governments and their agencies with the targets of the agreement. The two settings and negotiating logics have to be brought together. The difficulties in doing so provide a process-

23

See Weiss (2000).


15/12/2010 08:09


103

oriented rationale behind the growth of a much wider range of instruments used in international rule making. Endorsement In an ideal world, rules and regulations would be agreed through a process of organised world-wide dissemination, starting with the framing of the problem by all interested parties, followed by an expert review and assessment of all the empirical evidence, further followed by agreement by governments in the light of a common appreciation of the facts and common norms shared by all countries. The practice is far different. When countries decide to endorse an international agreement the most important characteristic of their behaviour is that it is based on strategic considerations valid to that particular government and one that is contingent on the behaviour of other governments.24 Diffusion theory emphasises the variety of such strategic reasoning.25 The choice of instruments used to express internationally agreed rules will reflect these varieties. Varieties of Strategic Motivation When countries act strategically to endorse a proposed international rule, it has been suggested that they act according to four main possible motivations.26 First, countries may act from competitive reasons, either to try to establish a position of competitive advantage or to avoid having a position undermined. For competitive reasons they may also act in order to try to demonstrate power and leadership – persuaded partly by the substance of what they are endorsing but also by an opportunity to set an agenda, to state a problem in their terms, to get first mover advantage in proposing their solution and more generally to show their importance as an actor on the international stage. For example, in pressing for the adoption of IASB international accounting standards, the EU was trying to avoid a situation where US Generally Accepted Accounting Principles (GAAP) became the international norm by default. Second, some countries may want to emulate other countries. By so doing, they may wish to demonstrate to others that they share certain common perceptions and norms – for example about the importance of combating the depletion of scarce natural resources or the degradation of 24

Peter Haas (2000). See Strang and Soule (1998). 26 These are drawn from Simmons et al. (2008). See also Simmons and Elkins (2003) for a discussion of the differences between competitive motivation and the desire to emulate. 25


15/12/2010 08:09

104


the natural environment. Thus, under President Obama, the United States was quick to signal that it shared the concern of many other countries about climate change. For some countries emulation may be about following a leader, or joining a bandwagon, or adopting a fad.27 Others will simply not wish to be left out.28 Third, in any international negotiation some countries will come under coercive pressure from other participants. For example, within the EU, interests will differ according to the issue and some member states will be pressured in directions they would not otherwise have wished to go; so too may some non-member countries with important relations with the EU.29 Fourth, there may be a genuine interest in an international learning process. For example, the spread of international financial regulation has owed much to the perceived leadership of regulators in the US and the UK whose practices, until the recent crisis, other countries have wanted to learn from. Different assumptions can be made about the dynamics of the learning process.30 For example, countries may be persuaded by the weight of evidence behind a proposal, including evidence that is not available to them as a single state. They may also wish to learn from practices that appear to have been adopted successfully elsewhere.31 They will likely be attracted by easily transferable models that rely on principled approaches rather than detailed rules.32 Again there are a variety of motivations that apply to international learning.33 The Circumstances of Adoption Strategic behaviour is also intimately related to the circumstances of adoption.34 This may involve the behaviour associated with the need to reach the thresholds required for agreements in different international 27 Some observers emphasise the importance of ‘informational cascades’ where it becomes optimal for a participant to follow others without regard to its own information (Bikhchandani et al. 1992). 28 Gruber (2000). 29 For a further discussion of the varieties of motivation see Cole (1989) and Braun and Gilardi (2006). 30 There is a distinction between accounts that use a rational learning model and those that use bounded models. See Meseguer (2006) for a discussion. 31 The rather stringent conditions that need to be met for successful lesson learning are set out in Rose (1993). 32 Schwartz (2005). 33 See Dolowitz and Marsh (2000) for a discussion. International lesson learning is not limited to transnational lesson learning. International organisations also play their own role – for example the OECD in the area of governance. Finnemore (1993) discusses UNESCO in the context of science policy. 34 Granovetter (1978).


15/12/2010 08:09


105

organisations – for example the OECD operates through a consensus rule while in the case of the IMF an important decision requires an 85 per cent majority. Countries can manoeuvre and modify their positioning order to form part of a winning majority or a blocking minority. Agreement on one issue may also be linked to agreement on another, leading to the wellrecognised phenomenon of log-rolling or issue-trading. Formal voting rules are not always helpful guides to the necessary thresholds. In many international organisations, both old and new, there is often a desire to avoid votes. Votes formalise division as well as agreements. Formal divisions become part of the public record, complicate relationships between governments and leave ‘legacy costs’ that governments will usually wish to avoid. In addition, even if a formal threshold is achieved it may not be meaningful, if important governments do not participate. These different motivations are illustrated in Box 5.1 using the case of the background to international agreement on measures to combat fears of human contributions to climate change. Excluding the impact of threshold levels, the box illustrates each of the varieties of motivation and reasoning discussed above – power and competition, emulation and common norms, coercion and lesson learning. Through the voices of their leaders the EU tries to project power and competitiveness, the US tries to show that it shares common norms, a smaller EU member admits to coercion and Japan expresses a desire not to be left out and to learn from others. What the varieties of motivation and the varieties of circumstance leading to the endorsement of a proposed rule mean is that the connection between the substance of a problem, the alternative policies put forward by experts to address the problem and the eventual decisions actually taken by governments becomes attenuated. Attempts have been made to define the circumstances that favour substantive content over content diluted by political bargaining.35 However, in general a ‘loose coupling’ may be the best that can be expected.36 One way of trying to visualise this process is to think in terms of a pyramid of negotiating preferences.37 The most preferred solution for governments (at the bottom of the pyramid) is where they all come together to agree on precise and mutually binding agreements in response to a commonly agreed problem with a commonly agreed ‘solution’. However, as the need arises in the actual negotiations to take into account that not all 35 36 37

See Peter Haas (2007: 798–802). Cole (1989). The idea of the pyramid is taken from Ayres and Braithwaite (1992).


15/12/2010 08:09

106


BOX 5.1 VARIETIES OF STRATEGIC MOTIVATION: CLIMATE CHANGE Competitive Reasoning/Power and Leadership ‘Building our own low carbon economy offers us the chance to create thousands of new British businesses, hundreds of thousands of new British jobs and a vast new export market in which Britain can be a world leader.’ Gordon Brown, British Prime Minister, November 19, 2007. ‘We can lead the world in moving towards a more sustainable economy while preserving European competitiveness, and we will do it.’ José Manuel Barroso, President of the European Commission, Brussels, October 28, 2009. ‘Europe has become the greatest laboratory for inventing tomorrow’s “green technologies.” It is because we were the first to set ourselves unparalleled objectives that European technologies will tomorrow be the most advanced.’ Nicolas Sarkozy, President of France, September 10, 2009. Emulation/Common Norms ‘It is true that for too many years, mankind has been slow to respond or even recognize the magnitude of the climate threat. It is true of my own country as well. We recognize that. But this is a new day. It is a new era. And I am proud to say that the United States has done more to promote clean energy and to reduce carbon pollution in the last eight months than at any other time in our history.’ US President Barack Obama, Speech to UN, September 22, 2009 (New York Times). Coercion ‘We need an agreement on one objective: global warming must not exceed 2 degrees Celsius. To achieve this we need the readiness of all countries to accept internationally binding obligations. We cannot afford missing the objectives in climate protection that science tells us have to be met.’ Angela Merkel, Chancellor of Germany, November 4, 2009, Speech to US Congress.


15/12/2010 08:09


107

‘The train can’t be stopped and I consider that a huge mistake.’ Vaclav Klaus, President of the Czech Republic, September 20, 2009 (Reuters). International Learning Carbon trading: ‘We don’t want a situation where, a few years from now, we suddenly realise that every other country is doing it, so we rush to implement mandatory carbon trading and make mistakes because we have no experience. If other countries are doing it, Japan has to do it too.’ Yasuo Fukuda, Prime Minister of Japan, June 10, 2008 (Financial Times). ‘The latest scientific research on climate change confirms our worst fears . . . Australia is determined to be on the front foot in global efforts to tackle climate change. We know that our nation is more exposed to the impact of climate change than perhaps any other developed economy.’ Kevin Rudd, Prime Minister of Australia (PM’s blog, July 2009).

participants will see the problem in the same way, the agreement is likely to have to become more and more flexible in its content in order to accommodate the different motivations. Further flexibility may also be driven by the need to achieve thresholds for final acceptance. The loosest forms of agreement are the ‘last resort’ where any agreement is better than none. In practical terms flexibility is achieved mainly by the use of escape clauses that provide varying degrees of ‘wriggle room’ to the parties to an agreement, or by clauses that offer assistance in meeting an otherwise over-burdensome obligation. The WHO’s Framework Convention on Tobacco Control (FCTC) illustrates both techniques. It incorporates a variety of escape clauses to modify a list of detailed principles and undertakings by the contracting parties. The FCTC also contains assistance clauses in order to help some Parties to meet their obligations.38 The escape clauses are shown in Box 5.2. In this example, the escape clauses can be seen as providing the flexibility that enabled a large number of countries to sign up to the general principles of the Treaty (168 parties ratified the Convention).

38 The use of implementation facilitating mechanisms in an agreement in order to maximise endorsements is discussed in the context of the environment by Stokke et al. (2005).


15/12/2010 08:09

108


BOX 5.2

ESCAPE CLAUSES IN THE FCTC

Each Party shall, in accordance with its capabilities . . . (General Provisions Art. 5). The Parties shall, within means and resources at their disposal . . . (General Provisions Art. 5) Each Party shall, where approved by competent national authorities . . . (Art. 9) Each Party shall . . . in accordance with its national law . . . (Art. 11) Each Party shall, in accordance with its constitution or constitutional principles . . . (Art. 13) Each Party shall . . . taking into account national circumstances and priorities . . . (Art. 14). Source:

WHO.

The ordering of negotiating preferences described above can also be described in terms of a ‘race to the bottom’ or reaching agreement in terms of the ‘lowest common denominator’(where flexibility represents the bottom). However, such descriptions place a negative normative interpretation on the ordering that may not be justified. The ordering can be explained more simply, without negative connotations, in terms of the need to bring together different attitudes of different participants to a problem that may itself be perceived in different ways. Implementation and Enforcement The degree of flexibility in an agreement is driven not only by a need to capture the variety of motivations and circumstances of the negotiations around the acceptance and endorsement of an agreement but also by the need for governments to take into account the setting in which the agreement is to be interpreted and implemented by themselves and by other participants in the agreement. In practice, in moving from internationally agreed rules to their actual application on the ground there is always likely to be present what has been called ‘predictable incoherence’.39 But the nub of the issue is that implementation has its own logic that exerts its own influence on the shape of an agreement and the choice of instruments. 39

Sunstein et al. (2002).


15/12/2010 08:09


109

Explanations of the implementation logic behind the choice of instrument offer two different focal points for the analysis. The first focus is on different types of government attitude or capability surrounding enforcement.40 The second focus looks further beyond governments to those agencies usually responsible for compliance and to their relationship with the targets of the rule or regulation.41 The explanations are mutually consistent despite their different focus. Attitudes and Capabilities According to the first explanation that focuses on governments, three main situations can be distinguished. In one situation the implementing government is a reluctant enforcer because it is not fully persuaded of the rightness of the measure concerned – or there may be domestic resistances to be overcome. In the second situation the compliance problem revolves around limitations in a country’s capacity to implement rather than any opposition to the substance of the rule. In the third situation there is an innate willingness on the part of governments to implement measures and the international agreement is put in place simply to encourage and to benchmark progress. According to this account, the choice of the different instruments is linked to this typology. In the first situation of a reluctant government, the measure will need to take on a precise and binding form so that it is clear that noncompliance will be followed up by probable legal or financial or other sanctions. For example, the IMF or World Bank may choose to split their loans into tranches, if they fear resistance to the conditions associated with them, so that access to a subsequent tranche is contingent on the borrowing country’s observance of the conditions attached to previous tranches. In the second kind of situation an appropriate implementing measure might take the form of declarations of principle that allow for governments to find their own way to implement the intent in practice. For example, governments whose countries host important securities markets may all agree that insider trading should be banned. However, the capacity of particular regulators to implement a ban may be impeded by a variety of factors ranging from keeping up with the technical difficulties of detection, to opaque structures of shareholdings, to the inter-connection between business elites and political elites. A declaration of principle could allow governments to choose from approaches that range from tackling 40

Falkner et al. (2005) offer this kind of model. Compliance can be defined as ‘Rule-consistent behaviour of those actors to whom a rule is formally addressed and whose behaviour is targeted by the rules’ (Börzel 2002). 41


15/12/2010 08:09

110


the politico-business culture, to civil actions to criminal prosecution, or simply to choose how they can best blend their own tactics. In the third situation, the rule can be exhortatory because an agreed set of recommendations may be all that is needed in order to achieve international cooperation and the coordination of policies. For example, the FATF has put in place an agreed set of ‘special recommendations’ to combat money laundering and terrorist financing. Since all members are agreed on the need for measures they do not need to be spelt out in treaty form. Negotiated Compliance In many cases, the attempt to capture the subtleties of actual implementation settings by focusing just on governments is likely to fail to capture the complexities of the actual settings on the ground. Accordingly, other accounts provide a focus that looks beyond the attitudes of governments in order to examine how those actually responsible for implementation manage the relationship with the targets of the regulation in practice. From this perspective the overriding characteristic of implementation in practice is the need to achieve a practical balance between accommodation and sanction and to be able to mediate and achieve a working, cooperative interaction between regulator and regulated.42 This calls for fluid, negotiable relationships.43 Possibly, in a harmonious world, regulators and regulated might come together in ‘interpretative communities’ that would reduce the need for sanctions.44 At the other extreme, in a world that relied entirely on litigation to resolve disputes, sanctioning and resort to law would become the standard practice. However, in practice, in a world where relationships between regulators and the targets of regulation usually fall between these two points there is no simple compliance model. Instead it has been suggested that there is at best a pyramid of enforcement capabilities and regulatory strategies. According to this pyramid, the most preferred route to achieve compliance is one that involves persuasion. Further up the pyramid come warnings and civil penalties. At the top, criminal law and punishment are the last resort.45 An example of such an ordering of regulatory enforcement options is provided by the EU in the case of possible non-compliance by companies

42 See Hawkins (1984), Edelman (1992) and Hutter (1997) for empirically based accounts of this sort. 43 In their case study of the poultry industry Hirschauer and Zwoll (2008) note the importance of accommodative regulatory practices by the German food regulators. 44 Black (1997). 45 For a more detailed description of such pyramids see Ayres and Braithwaite (1992).


15/12/2010 08:09


BOX 5.3

111

OPTIONS IN THE EVENT OF NON-COMPLIANCE: PHARMACEUTICALS IN THE EU

Education and facilitation Inspection Warning Naming non-compliant market authorisation holders Urgent safety restriction Variation of marketing authorisation Suspension of the marketing authorisation Revocation of the marketing authorisation Source: Rules Governing Medicinal Products in the European Union, Vol. 9A, September 2008.

with the terms and conditions of pharmaceutical authorisations. The compliance options open to the European Commission, European Medicines Agency (EMEA) and the competent authorities in the member states are shown in Box 5.3. What is suggested by the idea of a pyramid of strategies is that options such as education, inspection and warning will be the most preferred (at the bottom of the pyramid) and suspension and revocation of an authorisation the least preferred (at the top). These different ways of focusing the settings in which compliance with a regulatory agreement takes place are mutually supportive.46 In practice, even under the first account when the focus is on government attitudes, the application of sanctions will still likely involve preceding warnings before sanctions are actually applied, as well as movement towards the type of negotiated compliance described in the second. What therefore the two accounts have in common is the view that at its heart compliance is about management and negotiation. Connecting the Settings For each government negotiating with other governments the strategic motivations that enter into endorsement and the different strategic

46 Tallberg (2003) argues in favour of combining management techniques with enforcement possibilities.


15/12/2010 08:09

112

Democracy and dissent Intermediary instruments

ORDER OF PREFERENCE

Last resort

Flexibility

Law

Binding with escape clauses

Warnings and penalties

Precise and binding

Flexibile enforcement

NEGOTIATING THE RULE

IMPLEMENTING THE RULE

First preference

Figure 5.1

Orders of preference in negotiating and implementing rules and regulations

concerns that go into securing compliance with an agreement have to be connected.47 The difficulty in making the connection arises because the different reasoning calls for different preferred strategies. In one setting, in aligning motivations with other governments behind a decision to endorse a rule, there is a first preference for binding and specific obligations. In the other setting, in negotiating compliance with the targets of rules, there is a first preference for flexibility. In one setting, flexibility is the last resort. In the other, law is the last resort. The differences between the two strategies therefore have to be bridged. The main way of building the bridge is through the use of intermediary instruments. The two settings with their different orders of preference is illustrated from a theoretical perspective in Figure 5.1 above. What Figure 5.1 also illustrates is the potential need for intermediary instruments in order to provide the link between the negotiating setting and negotiating preferences with the implementation setting and implementation preferences. The use of intermediary instruments means that an agreement in one form is likely to be succeeded by follow-up agreements in another. Figure 5.1 suggests that it is a mistake to search for a connection

47 Duina and Blithe (1999) emphasise the importance of ‘fit’. Abbott and Snidal (2004) suggest a form of connection that identifies different strategic ‘pathways’.


15/12/2010 08:09


113

between the two settings through the means of some consolidated instrument that reflects a particular point on a hypothetical scale between ‘hard’ and ‘soft’. Instead, the use of intermediary instruments means that nested below framework agreements that may be flexible and based on general principles come ‘guidelines or ‘guidance notes’ that define how the principles are to be interpreted, that in turn lay the basis for laws that may be specific and binding. Similarly, conventions are followed by protocols, general ‘recommendations’ by specific or ‘special’ recommendations, agreements or concordats on principles or recommendations by more binding obligations.48 The WHO’s Framework Convention on Tobacco Control (FCTC) illustrates the techniques for connecting the two settings. The Convention, negotiated in 2003 and that entered into force in 2005, is presented by the WHO, in its foreword to the FCTC, as a ‘global trend setter’. While the FCTC is a principles-based agreement with escape clauses already described, the FCTC is also deliberately structured to provide for follow-up guidelines and protocols that elaborate much more precisely how exactly the principles contained within the Convention are to be implemented. To date, as of mid-2010, four sets of guidelines have been agreed and one protocol is under negotiation. The development of these further guidelines and protocols restrict interpretive latitude and establishes the basis for national legislation, including sanctions to be used as a ‘last resort’. Box 5.4 (overleaf) shows the steps involved in the creation and use of intermediary instruments in the case of one of the articles of the FCTC. The example given in Box 5.4 also illustrates one of the confusing elements in the variety of instruments used in international rule making – the way in which agreements change their form. An international agreement starts as an agreement that represents a flexible statement of principle and ends as binding domestic law. Or, using shorthand terminology, it starts as soft and ends as hard.

MISMATCHING AND TRANSLATION The range of instruments used in international rule making and regulation can thus be looked at in two very different ways. One perspective tries to relate the choice of instrument to the form of authority structures. The other perspective relates the choice to the different settings behind

48

See Baldwin (1995: 188).


15/12/2010 08:09

114


BOX 5.4 THE USE OF INTERMEDIARY INSTRUMENTS: THE CASE OF ART. 8 OF THE WHO’S FCTC Step 1. The Convention establishes a declaration of principle with an escape clause. Art. 8 (2). ‘Each Party shall adopt and implement in areas of existing national jurisdiction as determined by national law and actively promote at other jurisdictional levels the adoption of effective legislative, executive, administrative and/or other measures, providing for protection from exposure to tobacco smoke in indoor workplaces, public transport, indoor public places and, as appropriate, other public places.’ (Italics added for escape clause) Step 2. The Convention provides for guidelines to be negotiated subsequently in order to interpret how the principle is to be applied. Art. 7. ‘The Conference of the Parties shall propose appropriate guidelines for the implementation of the provisions of the Articles.’ (Arts 8–13) Step 3. The subsequent guidelines use additional statements of principle in order to define what latitude (if any) is allowed in the interpretation of Art. 8. ‘Effective measures to provide protection from exposure to tobacco smoke as envisioned by Art. 8 . . . require the total elimination of smoking and tobacco smoke in a particular space or environment in order to create a 100% smoke free environment.’ (Principle 1) ‘Article 8 . . . creates an obligation to provide universal protection by ensuring that all indoor public places, all indoor workplaces, all public transport and possibly other (outdoor or quasi-outdoor) public places are free from exposure to second-hand tobacco smoke.’ (Principles; italics in original) ‘Each Party should strive to provide universal protection within 5 years of the WHO Framework Convention’s entry into force for that Party.’ (Principles)


15/12/2010 08:09


115

Step 4. The guidelines prescribe further legislation and for its content. ‘Legislation is necessary to protect people from exposure to tobacco smoke.’ (Principle 3) ‘Effective legislation should impose legal responsibilities for compliance on both affected business establishments and individual smokers, and should provide penalties for violations, which should apply to businesses and, possibly smokers.’ (Duty of compliance) ‘The legislation should specify fines or other monetary penalties for violations.’ (Penalties) ‘In addition to monetary penalties, the legislation may also allow for administrative sanctions, such as the suspension of business licences, consistent with the country’s practice and legal system. These “sanctions of last resort” are rarely used, but are very important for enforcing the law against any businesses that choose to defy the law repeatedly.’ (Penalties) Source:

WHO: FCTC Guidelines for Implementation, 2009.

endorsement and implementation strategies and the different preferences governments have for dealing with each setting. The need for a wide range of instruments arises because of the different characteristics of the two settings and the need for intermediary instruments to link them together. There is common ground between the two accounts. Both accept as a permanent feature of the landscape that international agreements will continue to take a wide variety of forms. Both accept that authority will be exercised in a wide variety of ways. They also agree that the way in which rules are expressed has direct consequences for both the effectiveness and the observance of democratic norms. Words with well-defined and universally accepted meanings encourage compliance and they also meet the democratic norm of transparency. Rules that are applicable to concrete situations without excessive difficulties help enforcement, but at the same time they also help meet the democratic norm that laws should apply equally and consistently to similarly situated persons.49 These criteria apply to all forms of instrument. However, the two accounts lead in very different directions when it comes to looking for the sources of failure in international rule making 49

Diver (1983).


15/12/2010 08:09

116


and to the reasons for failures to respect democratic norms. According to the first account both failure and disrespect for democratic norms arise from various kinds of mismatching. According to the second perspective both a reason for failure and a possible source of the democratic deficit arise from the difficulties of translating the kind of reasoning used in one context to the reasoning used in another. Mismatching Theories of multi-level governance establish a connection between democratic values and the use of new instruments because the new instruments are seen as providing a way to moderate the exercise of traditional vertically organised forms of authority. It follows that failures to observe democratic values flow from choices favouring instruments requiring vertical forms of authority when such are not needed, or from choosing instruments that are benign sounding but that disguise hard realities. Effectiveness is also about the appropriate matching of instrument to authority. If instruments are used that convey little sense of obligation in circumstances when firm authority is needed then policies will fail. Translation The accounts that emphasise the need to link the endorsement setting with the implementation setting point in a different direction for the possible sources of failure. As far as the effectiveness of an agreement is concerned, one possible source of failure lies in the culture of ‘negotiated compliance’ described above. Such a culture may tolerate lapses in observance that lead rules to become ineffective or for risks of breakdown to accumulate.50 A further key issue is that of translation. The difference between the kind of reasoning by governments that goes into a decision to accept and endorse a proposed rule creates first of all a gap with the kind of reasoning by experts that goes into choices about its substance. As mentioned above, only a ‘loose coupling’ may be possible. The managerial logic involved in compliance is next likely to create a further gap when rules are shaped to achieve implementation. Thus what starts as an agreement between experts on substance is translated next into an agreement by governments

50 The idea that deviance from rules may become ‘normalised’ and eventually lead to the breakdown of an order can be traced back to the 1958 study by Gresham Sykes of the social order within a prison. In the account given by Sykes reciprocity between the rule enforcers and the targets of the rules is key. See also Vaughan (1996) and Barnett and Finnemore (2004).


15/12/2010 08:09


117

in a form calculated to win endorsement and then further translated to fit the context for negotiated compliance with the targets of the rule. In this process a gulf can be created between what the evidence calls for in the substance of an agreement and what actually appears eventually on the ground. Effectiveness is a likely casualty. The process of linking these different settings in an international agreement also points to a source of the democratic deficit. Citizens are at a huge distance from each of the different kinds of reasoning that has gone into the formation of an international rule. They are far from the elites who debated the substance of the rule and also far from the strategic concerns of governments negotiating with other governments.51 They are also distant from the logic involved in issues relating to compliance. What the translation of a rule between these different settings means for citizens is that although citizens may have views about the substance of an international rule, views too about the negotiating attitudes of their government and other governments, views in addition about the addressees and targets of the rules, what they see and experience as salient to themselves comes at the end of a very long and remote process. In the case of the FCTC the eventual impact on citizens will be immediately salient and the reasons for bans on smoking may be well understood even if the international agreement is unrecognised. However, in many cases of international rule making the origins and logic are likely to be obscure and distant and their salience not so immediately evident. *

*

*

This discussion of the bewildering variety of different venues and instruments used in international rule making has identified many different possible explanations of what can go wrong in international rule making and why democratic norms seem to get lost. The discussion continues by pulling these possible explanations together so that the reasons both for failure and for the democratic deficit can be more fully explored.

51 Berry and Berry (1999) suggest that public pressure may sometimes lie behind the desire of one government to emulate another.


15/12/2010 08:09

6.

The sources of failure

In 1996 the first case of avian flu (H5N1) was identified in farmed geese in China. In 1997 the first reported case of human infection was reported in Hong Kong and in 2003 the first human fatality, also in Hong Kong. In May 2004 the World Health Organization (WHO) warned about ‘high mortality’ rates and ‘pandemic potential’. In July 2004 it announced that avian flu risked becoming a global pandemic.1 As it turned out, by the end of 2009 about 280 people had died. While this was tragic it was not mortality on the scale originally warned against. The WHO has been criticised for sounding a false alarm and scaremongering. In contrast to the alarm sounded in 2004, the WHO warning in April 2009 about the risks of a global pandemic from ‘swine’ flu (H1N1) was handled in a more cautious and considered way. The WHO had learnt from the earlier episode and in response to criticism had revised the scale of alerts it went through before declaring a pandemic in June 2009.2 Even so it has been criticised again because the scale took account of the international spread of the virus but not the severity of the illness or its mortality (death) rate. In addition, in the case of H1N1 the morbidity (infection) rate may also have been overestimated.3 By the end of 2009 the WHO estimates about 13,000 people had died worldwide. In its defence, the WHO can fall back on the ‘precautionary principle’ – better safe than sorry.4 It has been estimated that between 1973 and 1 In its report on the ‘Assessment of Risk to Human Health Associated with Outbreaks of Highly Pathogenic H5N1 Avian Influenza in Poultry’ dated May 14, 2004, the WHO reported, ‘First the H5N1 strain has demonstrated its capacity to infect humans and cause severe disease, with high mortality . . . A second and graver concern is that a new virus subtype with pandemic potential could emerge . . . ’ In its subsequent July 8, 2004 ‘Current Evaluation of Risks to Humans from H5N1’, the WHO stated ‘the virus has the potential to acquire the ability to spread easily from human to human, and thus trigger a global influenza pandemic’. 2 The WHO’s ‘Pandemic Preparedness Guidelines’ that introduced phased alerts were first issued in 1999 and revised in 2005. The Guidelines were updated again in 2009. 3 ‘WHO official denies exaggeration about dangers of swine flu pandemic’ (Washington Post, January 14, 2010). 4 Testifying before a hearing in front of the Council of Europe a WHO official stated, ‘We are under no illusions that this response was the perfect response. But we do not wait until these global virus outbreaks have developed and we see that lost of people are dying. What we try and do is take preventive actions’ (BBC, January 26, 2010).

118


15/12/2010 08:09


119

2003 more than 30 previously unknown diseases emerged associated with viruses and bacteria.5 There is therefore ample reason for precaution. But the underlying issue is about whether or not the WHO has been employing the best epidemiological knowledge, brought the best animal health expertise together with the best human health expertise, has accurately mapped the spread of the viruses and their ability to cross between animals and humans, and has properly assessed the risks of their mutation into new and more virulent forms. What happened in 2004 represents one kind of failure in international rule making and experience with H1N1 will probably suggest that there are still further lessons to be learnt. Accordingly, the WHO’s 2009 pandemic preparedness guidelines clarified the description of the phased alerts and called for the WHO to provide an assessment of severity based on indicators such as the fatality rate and unexpected mortality patterns and the WHO guidelines are to be further reviewed.6 This chapter starts by discussing definitions of failure. Typically failure is defined by outcomes that do not match the original expectations, goals or intentions. There are problems in trying to define failure in this way and therefore an alternative is explored that focuses on ‘dispositions’. Second, the chapter assesses the different ways of accounting for failure. Earlier discussion of theories of multi-level governance had pointed the finger in the direction of failures that may occur when the scope of authority is misaligned with the scale of governance needed, or tasks wrongly assigned or poorly coordinated, or the instrument chosen to convey a rule mismatched with the form of public or private authority needed. These are classified later as examples of ‘executive failure’. In the diffusion framework the discussion had pointed to alternative or additional sources of failure – for example, what might go wrong in the transfer of rules when translated from one context to another – a form of failure that is classified later as a type of ‘cultural’ failure. This alternative analytic framework also singled out the need to look further at the strengths and weaknesses of epistemic elites who play a crucial role in determining the substantive content of a rule. Here the issue is of possible ‘cognitive’ failures. The discussion in this chapter continues by looking further at cognitive failure among rule-making elites. This is because the processes involved 5

Source: National Advisory Committee on SARS and Public Health, Canada (2003). In January 2010 the WHO Executive Board approved a proposal from the Director General to convene the International Health Regulations (IHR) Review Committee to review both the H1N1 response and the revised international health regulations intended to help the world to prevent and respond to acute public health risks that have the potential to cross borders and to threaten people worldwide. The Review Committee has 29 members selected from experts rostered under the IHR structure and other WHO expert committees. Its final report is due in May 2011 (WHO, April 12, 2010). 6


15/12/2010 08:09

120


in the first stage of rule making – the determination of the substance of a rule – come prior to everything else. If the substance of the rule is wrong then errors are transmitted whichever instruments are used subsequently to convey the rule, whatever forms of authority are employed and no matter how many different cultural contexts have to be crossed in order for the rule to be accepted and applied. At this first stage, the key question in possible cognitive failures centres on the susceptibility of epistemic and bureaucratic elites to particular types of bias in the way they go about assessing information and causalities. Bias leads to defective content.

VARIETIES OF FAILURE Failure comes in many guises. One way is to define it in terms of the blameworthiness of the regulator. The actions of the WHO in 2004 in publicising the risk of an avian flu pandemic can be seen as failures of ‘commission’ – statements were made that may have triggered others to take actions that were not justified in the circumstances and where the WHO was to blame. In the 2008 financial crisis the actions of the financial regulators can be seen, by contrast, as failures of ‘omission’. They did not take action at an earlier point when action might have headed off the crisis and they therefore share the blame for what happened. However, the recognition of failure does not always involve blame. This is in part because blame may be difficult to allocate and partly because ‘witch hunts’ may lead to the suppression of information by those involved regarding what really happened. Probably, therefore, the most common starting point for defining failure is when the outcomes of rules do not produce the effects originally anticipated or intended, or even produce harm. Defining ‘failure’ in terms of outcomes that diverge from the original expectations, goals or intentions is open to the objection that the definition reflects the assumptions of ‘rational choice’ theory. In other words failures are being defined as outcomes that either violate the norms of economic rationality that look at the substantive goals of rules and regulations, or violate the rationality of psychology that looks at the reasonableness of the procedures employed in establishing or implementing the rules.7 This approach is contested by those who regard what counts as ‘failure’ as, at least in part, as ‘socially constructed’.8 According to this alternative perspective what is counted as a success or a failure is going to be sensitive to

7 8

See the discussion of the two rationalities in Simon (1986). Berger and Luckman (1966).


15/12/2010 08:09


121

more subjective criteria, to the different rationalities of different observers and to different appraisals at different points of time.9 If the social construction of failure is to be taken into account, then definitions of failure need to take into account a range of different perspectives including the social and political contexts in which judgements are made. One suggested way of doing this is to think in terms of ‘layers’ of failure. For example, the most publicised allegations of failure come when a disaster or catastrophe occurs, when the finger of blame can be pointed at some person or organisation, when a faulty rule or procedure can be identified and when the recipient of the blame is unpopular to start with.10 Thinking about these different layers leads to more inclusive definitions of failure.11 The difficulty with more inclusive definitions of failure is that they become increasingly subjective and lose focus. There is therefore appeal in trying to define failure in ways that sidestep the theology of rational choice versus social construction. This can be achieved by defining failure in terms of dispositions. The dispositions are those that lead to failure in what ever ways outcomes are looked at. There is another advantage in looking at failure in terms of dispositions. Dispositions impact the regulatory process from the start. Outcomes look at the end of the process. The key dispositions, discussed below, that lead to failure are those of over-ambition, unresponsiveness and over-generalisation. Over-ambition In everyday life the term ‘failure’ is used to describe a situation where a person’s aims or goals are not attained – perhaps because problems arise on the way that cannot be overcome. Over-ambition triggers such failures because it leads to unrealistic goals or lack of appreciation of the obstacles to attaining them. While over-ambition in this personalised sense is relevant to regulatory failure it has also been suggested that the resolution of a problem is not the most common form of policy making. Rule making by governments is often guided by a concern about what is broadly appropriate rather

9 Rayner and Malone (1998: 28) assert, ‘The mainstream social science . . . are overly dependent on a narrow concept of rationality and an approach to policy as the means for making the real world conform to a rational model. The dominant rational-actor approach is in many respects a normative framework masquerading as an analytic one’. 10 Bovens and ’t Hart (1995) refer to these as the four different ‘layers’ of failure. 11 In this light, Bovens and ’t Hart (1995: 15) define public policy ‘fiascos’ as, ‘A negative event that is perceived by a socially and politically significant group of people in the community to be at least partially caused by avoidable and blameworthy failures of public policy makers’.


15/12/2010 08:09

122


than by setting goals that focus strictly on the consequences of a policy or a rule.12 This implies that policies and rules will never be completely successful in attaining given ends in the real world. According to this account, effectiveness is about ensuring that a policy or rule can be changed without undue costs. This perspective is also consistent with the view that the best that policy makers and regulators can hope for is to achieve a ‘fit’ between their core policy orientations and societal and political conceptions of what is appropriate. Failure occurs when the ‘fit’ is disrupted.13 Over-ambition is a disposition that leads to failure according to these perspectives because it is likely to lead to rules that are over-prescriptive, difficult to change and less likely to achieve the broad ‘fit’ required. The regulatory dynamics can be illustrated by the case of the introduction of GM crops into the EU. In this case the authorities are ambitious to see the development of science-based EU industries but have a problem with the societal ‘fit’ because of popular resistance. They turn to approving limited trials under highly prescriptive conditions as the regulatory ‘solution’. No one is satisfied with the policy.14 Unresponsiveness A second type of disposition that leads to failure is when the rule makers are resistant to ‘feedback’. This means that they fail to respond to outside warnings, new information or changes in circumstances or to changing behaviour by the targets of the rules.15 In some cases unresponsiveness might be a result of the receipt of over-generalised warnings or of a plethora of signals that create selection difficulties in what to take seriously and what to ignore. However, in the case of the 2008 international financial crisis, there were advance warnings of a specific nature that capital adequacy rules were ‘pro-cyclical’, that regulators might be downplaying the change in market conditions associated with the growth of derivatives and that the relationship between so-called ‘macro risk’ and ‘micro risk’ was not being properly assessed.16 The warning that market participants were acting without due diligence because of an excess of ‘animal spirits’

12

See March and Olsen (1989). Alink et al. (2001). 14 On March 2, 2010 EU Health and Consumer Policy Commissioner, John Dalli stated, ‘Responsible innovation will be my guiding principle when dealing with innovative technologies’. He announced a ‘reflection’ on how to combine a European Genetically Modified Organism (GMO) authorisation system with the freedom of member states to decide on cultivation of GMOs (source: EU Commission, IP/10/222 dated March 2, 2010). 15 See Barnett and Finnemore (2004). 16 See the warnings summarised in Kapstein (2006). 13


15/12/2010 08:09


123

was also downplayed. In this case the unresponsiveness seems to have stemmed not from inattentiveness but partly from the resistance of the rule makers to these warnings (perhaps, for example, because of their overattachment to the particular risk-based approach to capital adequacy they had adopted) and partly from their over-confidence in their own abilities to assess and handle the risks in the situation. Whatever the reasons, the unresponsiveness led to failure to head off the crisis. Former US Senator Daniel P. Moynihan gave a much earlier example of this kind of failure in describing what went wrong with the community action programmes of the War on Poverty programme of President Lyndon Johnson. According to Moynihan’s account, the policy makers were unable to respond to warning signals as the programme veered off course. In his view the inability to respond arose because different policy makers had very different views of what community action programmes entailed, became wedded to their particular viewpoint and could not take into account information that did not fit their perspective. The end result, according to Moynihan, was that ‘the program was carried out in such a way as to produce a minimum of the social change its sponsors desired, and bring about a maximum increase in the opposition to such change, of the kind they feared’.17 Over-generalisation Finally, there is failure in the sense of dispositions that lead to over abstraction and over-generalisation. There is always likely to be some risk of failure in moving from the imaginary world of the bureaucrat or politician to the imperfections of the real world, from a generalised context to a specific one and from the drawing board to the vagaries of actual human behaviour. Thus, in the real world, rules are unlikely ever to meet their original objectives fully.18 However, the rate of failure will be greater the more that there has been over-generalisation in framing the problem and its ‘solution’. A case study of ‘excessive abstraction and over-generalisation’ has been provided by a study of deforestation in the Himalayas. The study found that deforestation did not consist of a single problem, based on the general concept of a downward spiral caused by forest use exceeding forest growth, but stemmed from a plurality of problems that had to be defined and differentiated in much more precise ways.19 Over-generalisation led to a misdiagnosis of the situation and inappropriate ‘remedial’ actions. 17 18 19

Moynihan (1969: xiii). See Barnett and Finnemore (2004) for failures of ‘universalism’. Thompson et al. (1986).


15/12/2010 08:09

124


In this and other such cases of over-generalisation, the further failure would consist of failing to learn the lessons – in other words, overgeneralisation followed by unresponsiveness. The dispositions outlined above – over-ambition or over-prescription, unresponsiveness to new information, and over-generalisation or abstraction, can be taken together as key dispositions that lead to systemic failure in rule making.20 They may not provide an exhaustive list but there is empirical support for their selection. They define failure in terms of the attitudes and approaches of the rule makers that lead to failure in the many different ways that outcomes may be judged. They focus further questions on why regulators and rule makers may suffer from such dispositions.

ACCOUNTING FOR FAILURE Corporate Failure In the world of catastrophes linked to corporate behaviour, failure has been attributed to three main causes.21 First there are ‘executive’ failures. These are the failures made notorious by cases, such as ENRON in the US or Northern Rock in the UK, where a headstrong executive pursues a disastrous strategy unchecked by any of the defences put in place by corporate governance structures – such as executive boards or audit committees. A second source of failure is organizational ‘culture’.22 For example an internal, ‘team’ culture may become embedded in an organisation that ‘normalises’ uncritical habits or a certain level of risk taking that leads to disaster. The April 2010 BP ‘Deepwater Horizon’ oil spill in the Gulf of Mexico may turn out to have been a disaster of this sort. Third, there are ‘cognitive’ failures. These occur when people within an organisation misread or misinterpret diagnostic information relevant to the firm or its products in an uncertain and competitive world. Each of these sources of failure can be associated with the group 20 Systemic failing in this sense can be distinguished from failures that stem from the complexities of trying to understand the world as a total system. For a discussion of the world as a complex system see Harrison (2006). 21 These distinctions have been taken and adapted from Perrow (2007). 22 In discussing the Challenger launch disaster Vaughan defines ‘culture’ as ‘A set of solutions produced by a group of people to meet specific problems posed by the situations that they face in common. These solutions become institutional, remembered and passed on as the rules, rituals and values of the group’ (Vaughan 1996: 64). Martin (2002) includes material as well as ideational manifestations of ‘culture’, and brings together perspectives that allow for different or ambiguous interpretations as well as aspects that will be shared by most members.


15/12/2010 08:09


125

of dispositions that lead to failure identified earlier. Thus executive failure may be associated with an over-ambitious CEO, inclined to overgeneralise and unresponsive to warnings (or actively discouraging warnings) from persons or bodies set up to provide a check. Organisational cultural failures can also be associated with unresponsiveness or overgeneralisation or over-ambition. In the case of cognitive failure the association is particularly with over-abstraction and over-generalisation but may also include over-ambition or unresponsiveness. Examples of each of these associations are given in the course of the further discussion including from the US Federal Reserve (executive failure associated with unresponsiveness), the IMF (cultural organisational failure associated with over-generalisation) and the BCBS (cognitive failure associated with over-abstraction). One reason why these different sources of failure are not often distinguished from each other is that social influences cut across each.23 Thus, cognitive failings are not always easy to distinguish from organisational cultural failings, or cultural failings from executive failings. Indeed, one well-known description of a type of failing, known as ‘group think’, combines each source of failing into one syndrome.24 Another type of social perspective that brings together both cognitive failure and failure flowing from the cultural setting focuses on the neglect of ‘local knowledge’.25 There is a long-standing recognition that some rule makers may, for social and cultural reasons, ignore significant particularities of the setting in which a regulation is to be applied, as well as experience cognitive failure, because crucial pieces of local knowledge may not have been appreciated and may not have been framed in standard epistemic form.26 Another reason why the different sources may not be distinguished has been because of a tendency to treat cognitive failure as part of the ‘sunk costs’ of an organisation.27 Methodological approaches may indeed be set by, or constrained by, an organisational setting and examples are discussed later. However, there are fundamental methodological approaches, principles and assumptions that are externally derived – notably from the 23

‘Sensemaking is a social process’ (Weick 1995: 39). It is defined by Janis as the mode of thinking that people engage in when ‘they are deeply involved in a cohesive in-group, when the members’ striving for unanimity overrides their motivation to realistically appraise alternative courses of action’ (Janis 1982: 9). 25 What is termed ‘local knowledge’ has three components – it is typically experience based, holistic and time and context specific (see Failing et al. 2007: 49). 26 Lansing (1991, 2007) provides an interesting case of epistemic knowledge clothed in ritual form. 27 For example DiMaggio and Powell state, ‘Institutions do not just constrain options: they establish the very criteria by which people discover their preferences. In other words, some of the most important sunk costs are cognitive’ (1991a: 11). 24


15/12/2010 08:09

126


professions and disciplines of the participants in an organisation. These external principles and the way they are put into practice by experts cannot just be treated as the ‘sunk costs’ of an organisation. As long as the inter-relationship between the categories is recognised, the distinctions between executive failure, organisational cultural failure and cognitive failure remain useful and valid. The distinctions can be adapted and applied to the world of international rule making. They demarcate between failings that relate to individuals or unified actors such as boards or governments, failings that relate to the setting in which rules are framed and applied and failings that relate to method. Executive Failure Executive or individual failures, in the context of international rule making, point initially in the direction of poor leadership in international rule-making bodies, and/or poorly functioning boards and committees set up to supervise and help steer such bodies and their management. Executive failure, in the context of international rule making, can also be interpreted in a much broader sense as referring to the failings of governments when, operating as boards of governors or as members of coordinating bodies such as the G20, they take executive decisions relevant to international rule making, such as deciding what tasks should be entrusted to which organisation or when, in negotiating an agreement, they decide on a particular type of instrument or combination of instruments. This type of failing embraces most of the analysis of the shortcomings in the effectiveness of systems of governance identified in terms of the mismatchings, misalignments and misassignments pinpointed within the multilevel governance framework. Governments may misattribute tasks between organisations so that mistakes of appreciation or of indecisiveness occur. They may further fail to coordinate actions between different venues or different jurisdictions. They may mismatch divisions of authority required in order to achieve scale flexibility. Governments may also misalign the content of an instrument with the authority needed for its implementation. Organisational Cultural Failure Cultural failings linked to organisation refer to the setting in which the rules are made or applied.28 Such failings exist within international rule-making 28 Turner and Pidgeon (1997: 102) suggest that ‘patterns of shared culture and shared assumptions about decision-making are essential for the functioning of any organization, since they make for economies of communication’.


15/12/2010 08:09


127

bodies. They occur when international organisations develop internal habits of thinking and acting that may be damaging to their rule-making activities. The symptoms arise when people within the organisation begin to displace any feeling of personal responsibility for a decision to the organisation itself and to value the maintenance of easy personal relationships over the discomfort of challenge. They come to public notice with the use of phrases such as ‘The Washington consensus’ to refer to ingrained ways of approaching problems within the Bretton Woods institutions. The phrase suggests, rightly or wrongly, that a standard diagnosis and response are given by the IMF and the World Bank to any and every financial crisis in developing countries no matter what their distinguishing features. Cultural failings apply to the newer style of network organisation just as much as to the older style of universal membership organisation with their permanent staffs.29 Another type of cultural organisation failure may be relevant in the case of regulatory attitudes towards compliance. As discussed earlier, the diffusion framework pointed to the importance of ‘negotiated compliance’ – where regulators give priority to suasion and negotiated change rather than to sanction, where reciprocity between regulator and the target of regulation is central to compliance and where law is the last resort. There is some evidence that, when such an approach becomes ingrained, an accumulation of risks may result, as for example happened in the case of risks in the US subprime mortgage bond market.30 It is sometimes asserted that the US is less vulnerable to such risks than, for example, European countries, because it has a more adversarial legal tradition resulting in a greater willingness to resort to law and legal remedies.31 However, the 2008 financial crisis suggests that compliance culture may have played a role on both sides of the Atlantic. For example, ahead of the 2008 financial crisis triggered in the mortgage market, US regulators were concerned about home mortgages being extended to bank customers without adequate checks on their ability to repay. The regulators used suasion and ‘jaw-boning’ to discourage the loan originators from such practices but, at the same time, allowed banks to offer exotic mortgages to customers that neither they, nor the investors who came to hold them after their repackaging, were likely to understand.32 In this case the give and 29 Shiller states, ‘There is a social cognition, referenced by consecration, ritual and symbols, that is unique to each interconnected group of people’ (1999: 1331). 30 See Sykes (1958) for a discussion of the breakdown of the negotiated prison order. 31 See Solomon (2008) for the view that the adversarial tradition affects the style of US governance. 32 So-called Option ARMs (adjustable rate mortgages) offered four payment options in any given month involving reset and recast provisions to initial offering terms that could conceal large future payment jumps. By end 2009 about half were in default.


15/12/2010 08:09

128


take of reciprocal relationships led the regulators to give lenders mixed signals about the acceptability of mortgage lending practices. As a result, risks accumulated. A culture of negotiated compliance also exists at the international level. For example if the conditions attached to IBRD or IMF loans are not complied with, the first response of the Bretton Woods organisations will be to modify the conditions or the objectives, while sanction, in the form of loan cancellation, will be the last resort. However it is not clear at the international level how far such behaviour will lead to an accumulation of risks that breed a wider crisis. The risks seem to relate as much or more to the difficulty of achieving the compliance of large member countries to rule-based behaviour. For example, China’s manipulation of its currency that lies behind some of the largest global payments imbalances seems to be something that the IMF is unwilling to confront. This is not a question of organisational culture but the inherent difficulty in dealing with large members that flout the rules of the organisation. Cultural sources of failure exist, however, in a different sense that applies to the issue identified earlier of ‘translation’.33 In this case the diagnosis of failure relates to the transformation of the content of rules as they pass from the world of epistemic elites to the world of negotiating governments and finally to the world of the regulated and to citizens. These are different social worlds with their distinct contexts, their own styles of reasoning and their own imperatives. As discussed earlier, rules and regulations are reframed and reshaped as they move from one context to the next.34 The alterations may be justified at each stage in terms of making a rule more understandable to a different audience or more applicable to particular settings. However, such changes offer no structured defence against over-ambition, unresponsiveness or over-generalisation and may exacerbate any initial problem. For example, it is a well-recognised problem that when governments get together to negotiate an agreement they may seek to impress external audiences that they are ‘doing something’ by papering over differences in the way the agreement may be interpreted or implemented through the use of ambiguous wording.35 What this may mean 33 Sahlin-Andersson and Engwall discuss ‘translation’ in the context of knowledge transfer. They note more generally about the translation process that ‘ideas are shaped, framed, reshaped and reframed as they flow’ (2002a: 23). 34 Sunstein et al. (2002) identify the translation problem with moving from a normative judgement to a practical judgement. 35 In the context of the EU the problem is recognised in debates about ‘transposition’ where the different worlds are expected to be brought together in a coherent form through the subtleties of legal drafting.


15/12/2010 08:09


Table 6.1

129

Classes of cognitive failure

Heuristic

Self-deception

Emotion-based

Categorisation Availability Anchoring

Confirmatory Action-induced Framing

Attribution Herding Discounting

For definitions of each see Appendix B.

is that content that may have been over-ambitious to start with may also become over-generalised. Cognitive Failure The issue of cognitive failure, or failures of method, as a possible source of failure in international rule making was identified earlier in the discussion of the vital role of epistemic elites at the first stage in the making of international rules. At this first stage they play a leading role in framing problems and analysing alternative responses, and in choosing between alternative solutions. The strength of epistemic elites lies in their expertise and knowledge, both practical and theoretical. Their potential weakness lies in their vulnerability to certain well-known types of cognitive failings. What this means is that the way they go about gathering information and knowledge, making causal inferences and estimating the uncertainties and probabilities to be attached to the information and knowledge is vulnerable to distinctive flaws. Cognitive theory distinguishes between three broad classes of cognitive failings – those that stem from the use of simplified rules of thumb (heuristic simplification) those that relate to various kinds of self-deception that lead to over-confidence and those that are emotion based (such as favouring the familiar).36 What these failings do is to introduce bias into the diagnosis and response to a problem. A number of common types of cognitive failing leading to bias are shown in Table 6.1 (above), grouped according to these broad classes. They are discussed later in greater detail. Behavioural theory emphasises that bias is not necessarily irrational. For example an emotion-based bias in favour of what is familiar may represent a saving in terms of the extra effort needed to gather information

36

See Daniel et al. (2002).


15/12/2010 08:09

130


on what is less familiar.37 Similarly, consistently over-confident and overoptimistic corporate earnings forecasts may be a rational way for an investment analyst to deal with incomplete and uncertain information because it may make the management of the corporate client more willing to disclose additional information.38 Bias therefore is not about irrationality in rule making. It is about incomplete or distorted evidence, analysis and content leading to defective policy making. In the context of the international rule-making bodies it constitutes a potentially fatal defect. Assessing the Sources of Failure Each of these alternative sources of failure, the potential executive failures identified in accounts of multi-level governance, and the potential cultural and cognitive failings identified in the diffusion framework, are likely to play a key role in real-world failings. But of these three sources there is a compelling reason to look further at the possible association of epistemic elites with cognitive bias. This is partly because cognitive bias among experts is more difficult to recognise and defend against. It is also because it occurs at the earliest stage of policy making and sets in train a process that is likely to be difficult to reverse. None of the sources of failure discussed above is easy to correct but some are more difficult than others. Executive failings associated with poor leadership by management or boards can be corrected because heads of international bodies are subject to re-election or rotation. Boards also can be recomposed through the re-election and rotation of members. The mechanics are easily identifiable even if, in the real world, the process can be slow and politicised. There is also a fairly readily identifiable defence against cultural failings in an organisation. The safeguard lies in what is termed ‘substitutability’.39 If governments feel that particular venues are becoming a closed culture, offering over-predictable and over-standardised responses to a variety of issues, then they can look to alternative venues for rule making or they can create a new one. Institutions are substitutable. There are real world examples. The IPCC was set up to provide the venue for assessing 37 Cornell and Welch (1996) and Grinblatt and Keloharju (2001) provide examples drawn from different fields. 38 Lim (2001). 39 The importance of ‘substitutability’ is discussed by Drezner (2007) from the perspective of power politics and by Most and Starr (1984) as a more general phenomenon. Most and Starr warn against analysing policies within domain-specific boxes. Governments can substitute policies and achieve similar results through different processes. This general observation applies also to the choice of international institutions.


15/12/2010 08:09


131

climate change and its causalities in part because of a lack of confidence in the cultural orientation of UNEP. The decline of the Bretton Woods organisations relative to specialist organisations also may owe something to negative perceptions in the past of the ‘Washington consensus’. Again the actual processes of redress may be slow and politicised but the means are readily identifiable. Other forms of executive mismatching (such as choosing the wrong instrument with which to convey the rules) and cultural failures associated with a compliance culture or with ‘translation’ are much more problematic to deal with. But there remains an important sense in which cognitive failings are ‘prior’. If the way in which a problem has been framed is distorted, the way in which evidence is analysed is slanted and the way in which alternatives are assessed is biased then there is little or no hope of correction at later stages. Governments do not have the knowledge themselves to challenge the experts. In any case, as previously discussed, the focus of their attention is on the attitude of other governments to a proposed rule. Furthermore, no amount of attention to the way in which the instruments to convey the rules are crafted subsequently is likely to correct for major errors of substance at the earliest point in the rule-making process. No amount of juggling with jurisdictional boundaries will make up for rules that have defective content. No shuffling of institutional roles will help unless it is accompanied by a change in the way in which the rules are made. It thus becomes a necessary precondition for correcting other sources of failure that the sources of cognitive failure have first been recognised and addressed.

ELITE BIAS Reasons for Concern Governments make extensive and growing use of expert regulatory bodies in their national settings. But there are reasons for thinking that cognitive failings are potentially a more serious problem among rule makers at the international level. International organisations are further removed from public scrutiny, their roles often obscure. Those outsiders who do follow the activities of international bodies closely, whether international aid experts or climate experts, often share the same professional interests and objectives. Board members sent by member governments to watch over a permanent international organisation of the traditional variety may also share the organisation’s objectives and even the most detached will face the incentive to ‘work with management’. Members of the newer network


15/12/2010 08:09

132


organisations belong to the same bureaucratic or professional communities and thus share the same mindsets. Many international organisations see their role as advocates for a cause; their cause may be ethically of the highest order – the prevention of pandemics or the eradication of poverty – but advocacy can undermine analysis. As will be discussed later, there is also a lack of the checks and balances that apply within national settings, including not only elected legislatures but also a coherent system of judicial review. There are certain types of bias against which expert groups should be well defended. For example economists should not suffer from certain types of discounting failure such as money illusion in their analysis or behaviour. Similarly, regulatory experts routinely think in discounted terms, debate together about what are the appropriate discount rates to apply in different settings and could be expected to avoid time inconsistencies in their analysis. Nevertheless there remain reasons for concern. The fundamental reason for concern flows from the cognitive characteristics of epistemic elites identified earlier and the possible association of these leading characteristics with particular forms of cognitive failing. These associations are illustrated in Table 6.2 (below). Shared principled beliefs can predetermine the way a policy question is framed and the way in which options are weighed. Common notions of validity may lead to over-confidence in the handling of evidence and to various forms of herd behaviour that reduce the quantity and quality of evidence that needs to be assessed. Shared causal beliefs can lead to confirmation bias that rejects contradictory views and evidence and to attribution bias that tilts analysis towards a favourite causal theory even in the absence of evidence. The collegial sense of sharing in a common problem-solving venture may lead to further bias in the way problems and their solutions are presented to governments and the public through the choice of unrepresentative examples (availability bias) as well as confirm organisations in the rightness of their traditional chosen paths (actioninduced bias). Table 6.2

Elite group cognitive dispositions and bias

Shared principled beliefs Framing Anchoring

Common notions of validity Categorisation Herding

Shared causal beliefs Attribution Confirmatory

Common problem-solving venture Action-induced Availability


15/12/2010 08:09


133

The Table 6.2 illustrates the potentially dangerous associations between elite characteristics and classes of bias. Each of the possible associations between the characteristics of epistemic elites and particular kinds of cognitive failing shown in the table can potentially lead rule making in the wrong direction by predisposing experts groups to certain kinds of bias. The associations and the risks attached are therefore discussed further below. Shared Principled Beliefs When experts share principled beliefs they use a frame of reference acceptable to their profession that provides a way of analysing a problem by reference to evidence rather than to political values. For example they use various branches of science to discuss the causalities of climate change and they use the tools of market economics to discuss adaptation to climate change. Thus adaptation policies may get discussed in terms of correcting for ‘market failures’.40 What shared principled beliefs can lead to is bias in the way the problem is framed and in the use of what are termed ‘anchoring’ and ‘reference points’. Framing bias is where the initial formulation of a policy problem is set out in ways that steers the subsequent analysis and choice of options.41 It includes both what the experts do to describe the problem and what they do to construct the subsequent interpretation.42 Anchoring refers to when beliefs are attached to initial values, leading to an inattentiveness to new evidence. What are termed ‘reference points’ also draw attention to starting predispositions that affect the subsequent analysis. In this case the bias is about the weighing of alternatives in the analysis. The experts do not value the gains and losses of a particular alternative from a neutral (linear) weighting of the probabilities of success but from an initial ‘reference point’ that reflects their willingness to take risks with the policy choice when information is missing.43 An example of anchoring can be seen in the international financial crisis when the US Federal Reserve and US regulators were inattentive to evidence of a housing bubble in the US. The Federal Reserve seems to have had a predisposition to see bubbles as a phenomenon to be corrected by 40 For example the 2006 Stern Review on the Economics of Climate Change commissioned by the UK Treasury asserted ‘Climate change presents a unique challenge for economics: it is the greatest and widest-ranging market failure ever seen’ (2006: 1). 41 See Rabin (1998) for a discussion. 42 Kahneman and Tversky (2000) discuss the extreme sensitivity of choices to the way they are formulated. 43 Camerer (1998).


15/12/2010 08:09

134


the market and not the target for correction by regulators and supervisors. An example of the use of reference points can be seen in climate change debates in the approach to calculating the costs and benefits of adaptation policies. If the starting point is that we are looking at ‘the greatest and widest-ranging market failure ever seen’ then it is likely that there will be a greater intervention in market mechanisms in the absence of relevant information on those costs and benefits. Common Notions of Validity When experts hold common notions of validity, the association is with fallacies of categorisation or representativeness. Sets of instances may be classified uncritically and the importance of the class of instance may then be over-emphasised. Such fallacies can lead to over-confidence in the treatment of data and in the models used in evaluating and assembling the data. There is lack of due regard either to the degree of abstraction of the assumptions used in the construction of models or to the validity of drawing real-world conclusions from the simplified relationships incorporated within the models. For example in the case of the 2008 financial crisis it is alleged that many of the financial models used in the market and by regulators to calculate risk exposures and the pricing of risk were based on limited historical data drawn from what was an untypically benign and unrepresentative base period. Thus, the modelling of mortgage risk seems to have been conditioned by the fact that the US had never experienced house price falls at the national level since before the Second World War. There was inattentiveness to longer time series and to experience outside the US. The problem was an even wider one. The BCBS allowed the use of short-term time series when banks were calculating the adequacy of their capital under the risk weighted capital requirements of the Basel regime.44 Common notions of validity may also lead to herding. The herding instinct values thinking ‘within the box’ and discourages or discriminates against views that lie outside the consensus. What it means is that those doing the analysis become less responsive to what the data are telling them themselves and this in turn means that the analysis they pass on to others is also less informative. The end result may be a severe reduction in the information used in the analysis, not because the information is not

44 See Bank for International Settlements, Basel II: International Convergence of Capital Measurement and Capital Standards: A Revised Framework – Comprehensive Version, Part 2: Paras 133–8 and 178–81 and Annex 4 section E (June 2006).


15/12/2010 08:09


135

available but because it is ignored.45 A particular form of herding occurs when experts with incomplete information simply rely on others believed to have superior knowledge or information.46 Again, in the international financial crisis many smaller financial centres may simply have followed the regulatory leaders in the larger UK and US centres because they felt the big centres were better informed even in relation to their own context. International bodies such as the IMF may have acted in the same way in following the policy orientation of the US Federal Reserve. Shared Causal Beliefs Shared causal beliefs may be associated with one of the most damaging and prevalent forms of bias in expert groups – confirmatory bias. Confirmation bias is about seeking evidence in order to confirm a preexisting belief.47 It arises when there is a need to gather information from many different sources. Instead of considering each piece of information for its own worth, confirmation bias occurs when the experts start their analysis with an initial hypothesis and are inattentive to contradictory evidence or misread additional evidence as confirming the starting hypothesis. Confirmation bias has a number of undesirable features. It leads to over-confidence in the hypothesis initially favoured. It leads to a selective approach to framing the question to be examined, to the gathering of the evidence, to the analysis and to the conclusions drawn from the analysis. In addition, the bias is not necessarily overcome by the provision of further information. On the contrary more information may exacerbate the bias.48 Because it shapes the analysis of the evidence it may lead people to believe in a hypothesis that is actually wrong. It can also amplify policy differences between those with different starting points (so-called ‘ideological amplification’). An example of ‘ideological amplification’ can be found in the current debate about measuring the contribution of human activities to climate change. Both sides of the debate see confirmation bias at work and their positions become more polarised. Another related kind of bias that may accompany shared causal belief is what is called ‘attribution bias’. This occurs when experts assign a greater

45

See Banerjee (1992). Kuran and Sunstein (1999). 47 Rabin and Schrag (1999: 38) define it as follows: ‘A person suffers from confirmatory bias if [sic] he tends to misinterpret ambiguous evidence as confirming his current hypothesis about the world’. 48 Rabin and Schrag (1999). 46


15/12/2010 08:09

136


impact to a causal theory than to diagnostic data.49 Experts want to be able to give a causal explanation to what they observe and they may turn to a shared view of causation as a shortcut through the facts.50 Causal beliefs take on their own life in the face of empirical challenge. Common Problem Solving Members of expert groups often live in the same world as their colleagues, for example in the world of central banking or securities regulation or public health, where they share experiences and knowledge of a particular world, face the same set of problems, look at the same set of regulatory choices and are often united in what they want from any new rules or regulations. These commonalities may also be associated with additional cognitive failings. One type of problem is in the way expert groups relate to those outside the group. Because they may be dealing with highly technical issues that are difficult for outsiders to grasp, there is an incentive to oversimplify in order to explain and persuade. This leads to what is called ‘availability bias’ where an unrepresentative sample or image is used for the purpose of conveying an argument.51 Pictures of ice cap melt or flooding are powerful persuaders but they may not represent the facts or causalities. Availability bias is likely to occur at the end of the decision-making cycle when experts are trying to persuade governments or the public of the rightness of their diagnosis of the problem and the aptness of their preferred solutions. It is also something likely to be exploited by governments in persuading the public that they have adopted the right course.52 There is another type of cognitive failing that may affect groups that are embarked on common problem-solving ventures that may have an even more pervasive effect on the decision-making cycle. That is ‘actioninduced’ bias.53 When elites have been pursuing a certain path for some time they may develop a vested interest in declaring it a ‘success’ even in the face of contrary evidence.54 In the case of the 2008 international

49

Tversky and Kahneman (1982). Nisbett et al. (1982). See also Ross and Anderson (1982). 51 In theory a different representation of the same choice problem should yield the same preference (the invariance principle). In reality invariance principle failures are ‘ubiquitous’ (Tversky and Kahneman 1986). 52 Lindblom (1990) refers to the use of the ‘compelling’ or ‘seductive’ empirical observation as a means to achieve convergence on a complex social issue. 53 Weick (1988: 307) argues more expansively that ‘Action precedes cognition and focuses cognition’. 54 Daniel et al. (2002). 50


15/12/2010 08:09


137

financial crisis the risk-based approach to the capital adequacy of banks and to other financial institutions being implemented under Basel II had been under negotiation since 1999 and this may have made it more difficult for regulators to consider further changes. Being vested in their preferred approach may also have diverted the attention of both regulators and the targets of regulation from other dimensions of risk such as liquidity risk. Action-induced bias may also cause expert groups to act defensively. Here they may share a bias with elected governments – the fear of being blamed. In the case of governments, blame avoidance can take a variety of forms. It may take the form of ‘passing the buck’ – often to a regulatory body or sometimes to another branch of government. It may take the form of legislative hyperactivity where the government passes legislation – any legislation – in the desire to show that they are doing something. Rule-making bodies act in analogous ways. They may approve or commend a spate of regulation in order to demonstrate that they are on top of the game or the problem. In November 2008 the G20 meeting identified 49 different regulatory actions that governments and regulators claimed would lead the world out of the international financial crisis and prevent a repetition. Regulators also may be susceptible to what is termed ‘negativity bias’ where they point to the many things that can go wrong in a situation and issue generalised warnings so that they can claim ‘we told you so’, or in order to justify measures not otherwise justified. Weighing the Association The discussion in this chapter started by defining failure in terms of regulatory and rule-making dispositions that are over-ambitious or overprescriptive, unresponsive to new or challenging sources of information or over-generalised. What the discussion suggests is that the characteristics that define the epistemic elites who frame the problems to be solved by international rules and regulations, who gather and assess the evidence, construct and evaluate the models used for analysing the evidence and who present the policy alternatives to governments (or choose among them themselves before presenting their preferred option to governments) can all be associated with well-recognised forms of cognitive failing. These cognitive failings will likely lead rule makers to suffer from these dispositions of over-ambition, unresponsiveness and over-generalisation that are associated with failure in rule making. The associations should ring alarm bells. They provide ample reason why international rule making may fail at the very first stage when epistemic elites frame the problem and formulate the response. Cognitive bias among epistemic elites is extremely difficult to prove.


15/12/2010 08:09

138


There are many anecdotes and much circumstantial evidence but a lack of hard empirical evidence on the way expert groups go about their reasoning at the international level. There is a lack too of the kind of ‘process tracing’ (discussed in a later chapter) that would allow observers to follow decision trails. Moreover, the construction of ‘counter-factuals’ – what would have happened if different rules had been put in place – is difficult. Nevertheless, there is a persuasive theoretical case for thinking further about the methods employed in the making of international rules. Laboratory testing is suggestive of the problem.55 The case studies mentioned earlier relating both to deforestation in the Himalayas and to the social programmes of the Johnson administration provide indirect evidence. There is also evidence, introduced later, that some international institutions with a specific mandate to be attentive to diagnostic data and the way in which it is treated, such as the IPCC, acknowledge the problems. The different types of cognitive bias may also be difficult to distinguish from each other and this can lead to uncertainty and dispute as to where the real causes of failure lie. For example the case of the inattentiveness of supervisors and regulators to the housing bubble in the US and UK was attributed in the earlier discussion to ‘anchoring’. However, the fact that their prior beliefs were not challenged by others (with the exception of Spain, which did take some precautions against its own housing bubble) can also be seen as an example of herding. It is partly for the reason of possible uncertainties about the source of cognitive bias that, in later discussion of the defences against the various forms of cognitive failure, the defences are presented as a ‘set’. A procedure such as the need to express confidence levels in findings may be a specific defence against herding and at the same time a requirement for a competitive evaluation is a specific defence against anchoring. Taken together they work in the same direction. Both require alternative diagnoses and evaluations to be considered and thus together help address both sources of cognitive defect. Any misattribution between the two sources of possible bias will not matter because both will be caught. Bias: the People, the Governments and the Elites Political theory has long recognised that systems of government need to be able to counteract bias. In the early days of modern democratic practice it 55 Koehler et al. (2002) note from their experiments that experts systematically miscalculate probabilities and suggest that this is not because they are intrinsically biased but reflects the reliance of experts on non-diagnostic information. The use and misuse of non-diagnostic information is precisely the problem in many instances of cognitive bias.


15/12/2010 08:09


139

was the fear of herd behaviour by majorities of citizens that received attention. In the light of experience, democratic theory now warns against too ready an assumption about where bias exists. This is because experience with democracy suggests that it is not merely majorities that can engage in biased behaviour but also governments. The pendulum has now swung so far that it is often governments that are identified as the arch manipulators in systems of government and the electors the targets of manipulation. The reason for this is that electors are believed particularly vulnerable to credulity risk.56 The average citizen pays relatively little attention to politics and is thought to rely heavily on short-cut methods of arriving at decisions on, for example, which way to vote. A particularly common way in which governments are said to attempt to deflect criticism or to claim credit involves exploiting the attribution biases of the public (an over-readiness to look to particular causations in areas of public concern, such as blaming crime on immigrants). The recently established practice that governments should not be in charge of the value of a currency reflects the strength of the temptation for governments to take advantage of money illusion in the populace and because of the inability of governments to deal with time inconsistencies that comprise a particular example of discounting bias. Now that international rule making has risen in importance an equally pressing need is to defend against elite or expert sources of bias and the credulity of governments. It is citizens who bear the cost for defective rule making and who need the defences. The question is not about how to throw the elites out.57 It is a question about looking further at the procedures that govern the way they go about their tasks. The issue goes beyond the effectiveness of the rule makers and the procedures that can guard against faulty rule making. The issue is also about the legitimacy of the rule makers. They are unelected and the justification for the authority they wield in modern societies is rooted in the principles and procedures they follow in helping democracies to gather the best information available on policy problems and to mobilise relevant knowledge. If these procedures are distorted by bias then their legitimacy is undermined. Their legitimacy is not a question of what is termed ‘output legitimacy’ – the idea that such bodies are legitimate if their output is deemed good

56 People are subject to ‘credulity risk’, ‘when examining an information event or value indicator, they do not discount adequately for the incentives of others to manipulate the signal’ Daniel et al. (2002: 177). 57 Moynihan (1969) concluded that sociologists should not be involved in formulating public policy but only in evaluating it. Lindblom also advocates the transfer of social problem solving ‘away from scientific, technical, political and economic elites’ (1990: 220).


15/12/2010 08:09

140


and acceptable to public opinion.58 The key question is about the way in which they go about their work regardless of the successes or failures of the product. *

*

*

Bias is pervasive in societies. It impacts not only the judgements of the epistemic elites who formulate the content of proposed international rules but also the judgements of governments who decide whether to endorse and adopt the proposed rules and the judgements of individual citizens as well who have to decide whether to accept or reject the rules. It is not the existence of bias that is the problem. It is the need to recognise it and to be able to guard against the adverse consequences. Before the discussion of possible defences against the failings of elites as rule makers the next chapter looks at the sources of the democratic deficit in modern rule making.

58 Eckstein and Gurr (1975) dismiss the concept of output legitimacy except in times of crisis. Bellamy (2010) rejects it on the ground that ‘input’ processes are prior in importance.


15/12/2010 08:09

7.

Diagnosing the democratic deficit

THE LACK OF SIMPLE IDENTITIES When Presidents Roosevelt and Truman commended the post war international bodies to the American people they made a simple equation between the role of the new institutions and the interests of democratic electorates. Because the new bodies were essential for peacekeeping and for removing the sources of tension such as trade and exchange rate disputes that had nearly extinguished democracy in Europe, American leaders could claim that there was a necessary connection between the rule-making activities of the new bodies and the continued survival of democracy. American leaders could thus claim an identity or congruence between the values attached to the new organisations and fundamental American values.1 The actual practice of international rule making has refuted this simple equation between international rule making and democracy. In a world where there are deep divisions about fundamental values, bodies such as the UN cannot always be relied on to act in ways to uphold democratic values. Moreover, in a world where international rule making has spread into a huge variety of policy areas, many of which have belonged traditionally to the domain of domestic politics, it is not possible to relate particular policy issues to general over-arching questions of war or peace, or to the survival or eclipse of democracy within countries. Nor is it easy to link national and international policy making together through less direct ways of making value judgements. In the context of domestic policy debate, political parties will often suggest ways in which a particular approach to a particular policy issue can be ‘nested’ within

1 In his Special Message to Congress (February 20, 1948) transmitting his Annual Report on US participation in the United Nations, Truman wrote ‘The accompanying Report describes the efforts made by this Government to constructive achievement in the United Nations during the past year . . . these efforts were directed above all to assuring that the principles of the United Nations would be given full effect. The aim of our policy in matters not falling within the United Nations but rather within the direct United States relations with other governments, was to uphold the same basic principles. These principles are fundamentally those to which we have traditionally given allegiance’ (Public Papers of the Presidents of the United States, Truman 1948: 153).

141


15/12/2010 08:09

142


the approach taken by the party to broader values – for example particular tax policies might be linked to broader claims about the need for ‘social justice’. But these claims are often at the core of political contest. Thus, claims about the way in which one set of values can be nested within another are likely to be equally contested at the international level. Moreover, at the international level, electoral contest involving political party debate about values is absent. There is therefore no easy way of establishing the connection between international rule making and the way in which values are expressed within a democratic state. In the absence of any simple equation between the activities of international rule-making bodies and the interests and values of democratic electorates, the two frameworks used for discussing the international rulemaking scene point towards different diagnoses of the democratic deficit. According to one it is a question of paying much greater attention to the norms governing the connection between different types of authority. According to the other it is a more fundamental question about bringing together different types of discourse.2 The Two Perspectives As discussed earlier, the multi-level governance framework draws attention to the importance of the connections between horizontal and vertical forms of authority and between public and private authority. Accordingly, the approach to the democratic deficit emphasises forms of connection that respect democratic norms of representation, participation and transparency. Engineering the connections necessary to observe these norms may go in either direction – towards private actors and horizontal connections or towards official actors and vertical connections. An example of the importance of norms of participation and representation in the context of international rule making can be seen in the new (2009) governance arrangements that apply to the IASB. In this case there was concern that a private rule maker (also a horizontal actor) was not properly representative or sufficiently connected to the governments and agencies (with vertical authority) that enact and implement its rules. At the same time there has been a desire to respect the private character of the IASB and the Trustees of its governing foundation (the IASC). The ‘solution’ has been to create an ‘oversight’ or monitoring board that provides a 2 In this context ‘discourse’ can be defined as a discursive field that comprises the fundamental category within which thinking can take place, that establishes the limits of discussion and that defines the range of problems that can be addressed (see Wuthnow 1989: Introduction).


15/12/2010 08:09


143

link between the Trustees and public authorities (the securities and financial regulators) and to widen the geographical representativeness of the IASB itself whose numbers are increased from 14 to 16. The same norms of participation, representation and transparency can be applied to the design of instruments, to the exercise of ‘vertical’ authority and connections between jurisdictions and to the many ways identified in earlier discussion in which authority in international rule making can be disguised. By contrast, the diffusion framework points the finger at very different aspects of international rule making not explicitly considered in the multilevel framework. The first centres on the different modes of reasoning used by the epistemic elites and governments and the criteria that citizens might use themselves to judge a policy. Somehow the different discourses have to be brought together and reconciled if the voice of citizens is to count. A second difference concerns the role of institutional arrangements. They play a key role in bringing different discourses together. However, according to the diffusion account, the issues centre around the ‘congruence’ between national systems of government and international systems, or its lack, rather than around vertical and horizontal connections. The institutions needed to help achieve conciliation at the international level are not congruent with those available at the national level. Since congruence in institutional arrangements would be difficult to achieve, congruence in terms of values re-enters the discussion. The return of value congruence does not rely on the simple identities claimed by Roosevelt or Truman. Instead it relies on the use of universal discourses such as the language of the rule of law. The third aspect highlighted by the diffusion framework centres on the gulf in distance rather than the gulf in discourse. The way in which international rule making attempts to achieve congruence with democratic discourse at the national level by promoting the idea of universal discourses tends to exacerbate the problem of the distance between international rule makers and citizens. The distance highlights the issue of ‘role reversal’ – the implications of the way in which citizens experience international rules as receivers rather than as democratic principals. The discussion thus moves to the question of the dissonance that arises when what governments propose in their international rule making diverge from what citizens want for themselves. There is room for both frameworks and for both accounts of the democratic deficit. However, this chapter looks particularly at the issues identified within the diffusion framework. As the example of the IASB Monitoring Board given above shows, the kinds of norms and connections envisaged under the multi-level governance framework do not overcome


15/12/2010 08:09

144


the problem of role reversal. A group of public regulators monitoring a group of private regulators does not change the fact that citizen experience is usually at the receiving end. Neither does this kind of connection nor the extension of representation on the IASB address the problems posed by differences in the styles of reasoning used by different participants in the rule-making process or situations where citizens may wish to apply their own standards of reasoning. The language used by both the private and public bodies is technical and professional. Citizens may approach the issues differently. This chapter therefore looks particularly at these more fundamental issues. Approaches to Role Reversal and Differences in Reasoning In order to assess how different styles of reasoning are brought together in international rule making, the discussion looks first at ‘conciliation’ processes. Conciliation emphasises that policy making, regardless of whether it takes place within national jurisdictions or whether it takes place internationally or in a mix of jurisdictions, involves, at its heart, the need to bring together incommensurate values and methods of reasoning. It identifies ways of doing this. Second, the discussion leads to the issue of congruence. Congruence in this context is not about the stability of systems of rule making discussed earlier, but about achieving correspondence between national methods of conciliation and international methods. The discussion of congruence distinguishes between institutional congruence (the use of similar institutional or constitutional arrangements at the international level as at other levels) and value congruence (the way in which the expression of common values may be used as an alternative to common institutional arrangements).3 Third, the discussion takes up the issue of the distance between the different actors in the policy-making process and the way in which citizens experience rule making at the receiving end. The problem is couched in terms of dissonance and divergence. Divergence refers to the democratic problem that arises when the policy preferences of governments diverge from those of citizens. What dissonance means in this context is that, even if there is no such divergence, nevertheless citizens still have to find their own reasons for accepting the choices made on their behalf. If 3 For an example of value congruence in the context of the EU, Mair and Thomassen (2010) refer to a congruence between the left/right divides of national parties and parties in the European Parliament as making up for the institutional differences between the role of national parliaments and the European Parliament.


15/12/2010 08:09


145

divergence and dissonance cannot be removed then dissent is the outcome. Democratic societies always afford a place for dissent in their governing rules.4 Authoritarian regimes suppress it. According to this perspective, at the heart of the democratic deficit in international rule making lies the treatment of dissent.5

CONCILIATION Defining the Problem Conciliation is about the bringing together of the different kinds of rationalities involved in policy making. Many different intellectual disciplines are involved, those of the natural sciences and the social sciences and law. So too are normative and ethical modes of reasoning. In the account given so far of the different styles of reasoning used in international rule making it is the epistemic elites that are likely to use the disciplines of the natural or social sciences while governments may be swayed by normative concerns such as ‘leadership’ or emulation. Citizens may rely on a much wider and more intuitive style of reasoning in reacting to policies that they judge as salient to themselves. The difficulty of bringing together the different disciplines involved in policy making applies within a broadly defined field such as the social sciences.6 It applies in an even more intractable form between fields. For example, the different modes of reasoning used in the natural sciences to talk about scientific content, in economics to talk about efficiency or incentives, or in law to talk about procedural norms, or in politics to talk about values such as social inclusion are not directly translatable into

4 Even in the case where majorities are allowed to override minorities, democratic theory emphasises the critical importance of ‘losers’ consent’ for the maintenance and vitality of the political system (see Anderson et al. 2005). 5 Barnett and Duvall (2005a: 22) suggest that the issues should be looked at through the perspective of ‘power’. Dissonance approaches the problem from the opposite end. As they acknowledge, ‘power and resistance are mutually implicated because the social relations that shape the ability of actors to control their own fates are frequently challenged and resisted by those on the “receiving end”’. In the context of the exercise of power at the international level the analysis of ‘power’ carries with it huge difficulties about how power is to be measured and an equally large normative baggage associated with views on the countries and doctrines that are seen as internationally powerful. 6 Easton and Schelling (1991: 23) noted, ‘There is no general theory about the best way to integrate the major areas of knowledge in the social sciences, let alone in the humanities’. Despite a growth in writing on inter-disciplinarity (referred to later) this judgement remains valid.


15/12/2010 08:09

146


each other.7 The way in which they are brought together also has to be persuasive to citizens. In a national context the gulf between styles of reasoning becomes visible in relation to major regulatory issues – such as whether a new airport imposes ‘unacceptable’ environmental harms or a new nuclear plant an ‘unacceptable’ health or contamination risk – but it applies also to smaller and local issues, such as whether to impose speed limits near a school in order to reduce an ‘unacceptable’ level of child fatalities.8 In these and other cases there is evidence from economists, scientists of different kinds, behavioural and other experts and there are value judgements about what is ‘acceptable’ or not. One kind of judgement may inform another – for example the costs of a new medical intervention may be extremely high in relation to quality of life or life expectancy improvements.9 Nevertheless, the ultimate choice about how to balance these different considerations in a final decision involves bringing together ethical discourse with the different kind of reasoning used by economists, and the reasoning of economists is different again from the way in which the question will have been looked at by behavioural and other scientists. The difficulties of conciliation apply in a similar fashion to international rule making and regulation where equally different styles of reasoning also need to be brought together. Accountants may reason that ‘marking to market’ is the best way to assure international investors that corporate accounts are ‘true and fair’. Politicians may however feel that the costs in terms of the volatility of the accounts are ‘too high’. Scientists may reason that the risks from genetically modified food products are ‘unproven’ but politicians may argue that unless their safety is proven then the risks to society of allowing for international trade in GM products are ‘unacceptable’. The people may agree or disagree with either and have their own way of viewing the issues. Common Methods of Conciliation There are two leading accounts of the conciliation process that apply both to policy decisions in a national context and to policy decisions in an international setting. Each account reflects the operational pressures that bear 7

Snellen (2002) distinguishes between these four main rationalities. Rayner (1986) looks at four different rationalities in a single hospital setting involving different attitudes to risk. 9 The growth of common metrics, such as cost/benefit analysis and the construction of composite rankings to measure performance across fields or situations, as a way to reduce the problems of incommensurability in public policy making is discussed in Espeland and Stevens (1998). 8


15/12/2010 08:09


147

on policy makers in the real world. In one account, the bureaucrats and or the politicians simply ‘muddle through’ and make their best overall judgement in the circumstances.10 In order to arrive at this overall judgement the politicians engage in what has been called ‘partisan mutual adjustment’ where they grade and trade values and outcomes and try to take all values into account.11 In the other account, policy decisions respond to the necessities of time and place rather than to a logic of ‘consequences’ that links problems to choices and choices to preferences.12 According to this account, when a policy problem arises, governments will do whatever they think is necessary in order to satisfy the demands of the occasion. ‘Satisficing’ has its own kind of rationality.13 What underpins these two accounts is an emphasis on how far actual methods of conciliation will diverge from ideal methods. In an ideal world, evidence will determine what the problem is, what the alternative solutions are, and governments will select the alternative that is most consonant with shared norms and with the preferences of their citizens. In practice, government decision making is often time pressured and ‘events driven’, actual decisions ‘muddle through’ incommensurable discourses and a government’s main concern may simply be to avoid being blamed by the public for mistakes of commission or omission. Each account has a good deal of real-world plausibility. There are however important shortcomings to both these methods of conciliation. The shortcomings apply equally to decision taking at the international level and at the national level. A policy that reflects a practice of ‘muddling through’ is difficult to defend if the reasonableness of the decision has to be justified. Under challenge, the assumption that incommensurate reasons can be ‘graded and traded’ will be shown to be false. The ‘satisficing’ approach has a different weakness. It appeals to decision makers under time pressures because it encourages them both to avoid taking fundamental decisions that might be difficult to reverse as well as to leave substantive differences unresolved. But fundamental decisions do sometimes have to be made and substantive differences do sometimes have to be resolved. Nor do governments consistently like to give the appearance of being forever driven by the forces of external events at either the national or international level. 10

Lindblom (1959). Lindblom (1965: 226–45). See the discussion in March and Olsen (1989). 13 The satisficing model is a development of the earlier ‘garbage can’ model that also had the properties that the resolution of a problem is not the most common style of making decisions and that the process is sensitive to load (the number and difficulty of the decisions) (see Cohen et al. 1972). 11 12


15/12/2010 08:09

148


Boundary Objects The need to maintain reasonableness across different fields has suggested a different approach to conciliation that proposes that different scientific or rule-making ‘communities of practice’ can find common ground through identifying those ‘boundary objects’ that inhabit each community and satisfy the informational requirements of each.14 For example, a concept such as ‘sustainable development’ brings together both economists and scientists. Similarly, viewing regulatory issues through a lens such as ‘risk’ has the appeal of providing a term that makes sense to scientists and engineers as well as to economists and ethicists. The use of terms and concepts that cut across different professional disciplines has however important limitations. First a term that has a clear meaning in one context may have an ambiguous or disputed application in another. For example, the concept of sustainability may have a clear meaning in the biological or environmental sciences (such as in relation to forest resources). Its applicability in economics is however more limited and debatable because, for example, of the importance of the price mechanism as a way of signalling scarcities, as well as the importance of technical progress, innovation and changes in consumer preferences. At the time of the Napoleonic wars a British admiral always travelled in England with his pockets stuffed with acorns. His concern was the sustainability of Britain’s supply of oak for building the navy on which Britain’s military strength depended. His concern could be set aside with the development of iron for use in shipbuilding. Second, a boundary term may provide a possibly misleading focus. In earlier discussion it was suggested that the term ‘risk’ did not cover other important dimensions in international rule making such as the need to manage strain, and could conceal conflicts between values. The question of focus also arises in much more limited regulatory contexts. For example, Britain’s aviation safety regulators keep track of what are termed ‘near misses’ that occur when aircraft fly too close to each other. Such a system appears to focus on controlling the risk of major accidents from mid-air collisions. However it also serves another purpose. Since mid-air collisions are highly infrequent, reporting ‘near misses’ provides an alternative measure through which safety regulators can tell the public that they are vigilant on the job on a regular basis. The combination of risk management and organisational purpose may not be for the best. It may distract from whether or not near misses are actually a good indicator of

14

See Bowker and Starr (2000) for this kind of account, and in particular pp. 297–326.


15/12/2010 08:09


149

the safety performance of aviation regulators (compared for example with clear lines of reporting, hours on duty etc.). What this means is that communities of practice that employ boundary terms have at the same time to be acutely aware of the political and organisational dimensions and the limitations of what they do.15 Even in theory it is a difficult set of requirements for communities of practice to meet for themselves on their own. Their own motivation may run in the opposite direction to exaggerate the usefulness of their chosen boundary concepts. In practice, conciliation has to bring in a range of actors. Boundary terms such as ‘sustainable development’ can be seen from a slightly different perspective as an example of the kind of innovation that occurs in interdisciplinary studies that can also help bridge the divide between different styles of reasoning. However, interdisciplinary studies have their own logic and the problem of conciliation with other types of logic remain.16 According to a different account (labelled ‘transdisciplinary’) when there is a goal to satisfy the public interest, the ‘dogmatically’ correct use of rules of investigation and testing of theories must give way to other means of handling the connection between the world of science and the real world.17 However, this does not address the question of how to judge what satisfies the public interest, or who is to make the judgement, or how the public interest is to be defined. Streaming The shortcomings attached to ‘muddling through’, to ‘satisficing’ and to relying on communities of practice to develop boundary terms and concepts, have been an important driver of a different approach to conciliation. This involves what is termed below as ‘streaming’. Streaming entails systems of government in segmenting their decision taking by establishing specialised bodies in each area of expertise and adapting the traditional branches of government to the growth of specialised bodies. Specialisation offers advantages in information gathering and in linking theoretical knowledge to practise within a professional field or discipline. At the national level specialisation is seen in the extraordinary growth of unelected regulatory bodies and other agencies outside the central departments of government. At the international level, as already discussed, it has been seen in the growth of special purpose venues outside

15 16 17

Bowker and Starr (2000). See the discussion in Barry et al. (2008). Balsiger (2004).


15/12/2010 08:09

150


the traditional universal organisations with their comprehensive terms of reference. At the same time, the growth of specialised and increasingly fragmented fields of discourse places an even greater burden on the role of other branches of government in conciliation. At the national level the conciliation of specialised fields of discourse depends on each branch of government adapting its traditional role and playing to its own strengths. The process can be seen as involving three main steps. First, the new specialised bodies can clarify the evidence and knowledge base behind a policy in their own area. The specialised bodies that employ particular types of reasoning enhance their own role by emphasising the validity and relevance of their style of discourse and method of reasoning. As a result many routine policy decisions and actions can go ahead without societies having to engage in substantial normative debate where underlying normative differences have to be resolved. As mentioned earlier, common metrics such as the language of cost/benefit analysis employed by specialised bodies also provides a means of comparing different kinds of cost and benefit in ways that will often forestall extensive normative debate. Where differences persist, streaming leaves the different methods of reasoning intact, more open to public review and scrutiny and their guardianship possibly strengthened. Second, the judicial branch, supplemented by many quasi-judicial tribunals and appeals bodies, is better able to focus on the procedural aspects of policy making. It can look at such aspects as the adequacy of consultation, whether or not all the evidence has been considered or some material factor omitted, or whether or not the decision is proportionate to the problem. Again this offers the possibility that decisions can be taken without reopening or re-examining the substantive investigation and without making a judgement on any underlying normative issue. Third, the political branch, such as elected assemblies, can pay special attention to normative debate. Normative issues will have been clarified and better defined by the other branches. Politicians can thus focus on such aspects. They can look at the way in which the statutory terms of reference of a specialised body are set up, the way public interest obligations are defined, and they can engage in general political debate about what is acceptable or not, or undertake more focused scrutiny. The political branch in systems of government is increasingly ill-adapted to the details of problem solving in public policy. It is able however to serve the ‘expressive’ purposes of politics. The tendency within national systems of government for normative issues to be handled within politics and for the evidence base to be handled outside politics by specialised bodies is arguably good for all forms of discourse including the normative.


15/12/2010 08:09


151

Because the advantages of distinguishing between different specialised discourses have become increasingly accepted in national contexts, the traditional branches of government have begun to adapt their own roles in the conciliation process. Some accounts of the adaptation use principal/ agent theory to explain the way in which the political branch can still oversee specialised agencies. Newer accounts suggest that what we are seeing constitutes a new balance of powers.18 The adaptation that is occurring within national systems is much more difficult to achieve at the international level. When it comes to the making of international rules and regulations the ‘satisficing’ standard or ‘muddling through’ standard still largely prevail. International organisations themselves engage heavily in normative debate; governments may involve themselves in technical matters; procedural standards are often unclear and decisions often appear to respond simply to the pressure of events or to highly artificial deadlines.19 Any desire to see a greater correspondence between the conciliation procedures that prevail in the international domain and the conciliation procedures followed in national jurisdictions brings back onto centre stage the issue of congruence. In this context, ‘congruence’ is not a theory about the stability of systems of government, it is about whether or not the international system could and should develop methods of conciliation that mirror the methods of conciliation now becoming commonplace at the national level.

CONGRUENCE The earlier discussion had defined ‘congruence’ in terms of general patterns of authority. In this further discussion, the notion of patterns of authority is broken up in order to make a distinction between congruence in terms of institutional arrangements and congruence in terms of values.20 18

For a discussion see Vibert (2007). The G20 declaration issued on November 15, 2008 on the international financial crisis contains the following example of artificial deadlines: ‘We are committed to taking rapid action . . . We instruct our Finance Ministers . . . to initiate processes and a timeline to do so. An initial list of specific measures is set forth in the attached Action Plan, including high priority actions to be completed prior to March 31, 2009’. 20 Dalton and Shin (2006a) sub-divide value congruence into congruence between social and political values, social and market values, political and economic values and citizen values and institutional structures. Held makes an oblique reference to the distinction by talking about congruence in terms of policies as well as electoral mechanisms: ‘There has been an assumption at the heart of liberal democratic thought concerning a “symmetrical” and “congruent” relationship between political decision-makers and the recipients of 19


15/12/2010 08:09

152


Institutional congruence points to the way in which, in order to achieve congruence with national processes of conciliation, international rule making needs to supplement its array of institutions that engage in specialised professional discourse with the judicial and political branches of government playing roles congruent with those that bring together the different discourses in national settings.21 In practice, international rule making challenges the relevance of institutional congruence and has come to rely instead on the relevance of ‘value congruence’. The claim is that the values expressed at the international level can be shown to be congruent with the key discourses used within democracies at the national level. In so doing, value congruence provides an alternative to institutional congruence.22 The instrument for demonstrating value congruence at the international level is the use of discourses with claims of universal validity. The simple equations about value congruence made by Presidents Roosevelt and Truman may not be possible or plausible. However, the use of discourses with claims to universal validity offers a more sophisticated approach. In this context, universal discourses serve three purposes. First, they provide forms of discourse that cross different jurisdictional boundaries. They provide a common language that all jurisdictions recognise and use themselves in their internal discourse. Second, they can be used as a means to bring together different ways of reasoning in public policy in an authoritative manner. By bringing together different styles of reasoning under a common style of discourse they aim to achieve conciliation at the international level. Third, they provide an avenue for claims about legitimacy for what is being done at the international level in ways that do not depend on congruence in institutional arrangements. The question is whether the use of universal discourses can serve these purposes or whether the claims over-reach.

political decisions. In fact, symmetry and congruence . . . are assumed at two crucial points: first between citizen-voters and the decision-makers whom they are, in principle, able to hold to account; and similarly, between the “output” (decisions, policies etc.) of decisionmakers and the constituents – ultimately, “the people” in a delimited territory’ (Held 1995: 224). 21 Hurrelmann and DeBardeleben (2009: 235) discuss institutional congruence more narrowly in terms of the electoral mechanisms to ensure government ‘of’ the people. They make a distinction between the congruence of people affected by a decision, people selecting their representatives and people choosing the decision makers with congruence between the electoral process and the ‘imagined’ communities of the electorate. 22 It can be seen as a way to overcome what Hurrelmann and DeBardeleben (2009: 238) refer to as ‘the congruence dilemma’ – the gap that arises between imagined communities at the national level and decision taking at higher levels (in their discussion at the EU level).


15/12/2010 08:09


153

Institutional Congruence There have always been practical difficulties standing in the way of envisaging arrangements at the international level that would mirror the judicial and elected branches of government at the national. As far as the elected branch is concerned theoretical alternatives range from the possibility of setting up some kind of international representative assembly, voted into place through some form of international voting procedures, to the creation of venues where parliamentarians or national legislators could sit alongside international organisations.23 Because either of these approaches would involve great leaps of faith, as well as great practical difficulties, it has also been suggested that international organisations should simply better represent countries and their populations by adjusting their membership shares.24 There are similar practical difficulties standing in the way of developing a judicial branch of government at the international level that can perform the same role as judiciaries at the national. While there has been a proliferation of courts, appeal bodies and tribunals that in some ways echoes what has taken place in national contexts, achieving an overall coherence in terms of access, remedies and agreed principles of adjudication remains a distant goal.25 Networking between courts seems the easiest option but the limitations of network theory have been discussed earlier.26 The institutional difficulties can be seen as part of a more general difficulty of pulling together the national and the international political spaces.27 The suggestion that conciliation at the international level can be achieved through a congruence of values bypasses institutional difficulties. Value Congruence In the context of a political system the expression of values plays three distinct roles. First, debate about relevant values occupies much of the 23 See Archibugi (2003) for discussion of the first option and Kapstein (2006) for the second. 24 See Ramachandran et al. (2009). This assumes that the governing and executive boards of international organisations can be seen as representing governments that in turn represent their peoples. 25 A mapping exercise kept current by PICT (the Project on International Courts and Tribunals organised by New York University and London University) suggests that there are around 100 plus courts and tribunals making up the international judiciary. For a discussion of the fragmentation, see Wellens (2004). 26 Slaughter (2004a) describes an emergent order of courts. 27 Zürn (2000) also highlights the issue of congruence and the incongruence between social and political spaces in international governance.


15/12/2010 08:09

154


subject matter of participatory politics. Second, the expression of values can be seen as a way for a system of government to mobilise authority for what it does. Third, the expression of values can be seen as a means of achieving legitimacy for a political system (a means of justifying the authority exercised). In democratic theory the idea that the expression of societal values can serve these different purposes goes back at least as far as Rousseau. In a contemporary context the idea that the expression of values provides the essence of debate in expressive politics is part of the discussion of the new separation of powers mentioned above.28 The association with authority may also be linked to participatory politics.29 In relation to legitimacy, the expression of values can be seen in nonelectoral terms by providing a political system with ‘support’ through encouraging the notion of a ‘common good’ or feelings of community.30 ‘Support’ in this context is about support for the system of government or rule making as distinct from support for a particular measure. The association of values with legitimacy can also be expressed in terms of ‘consent’.31 While again consent is often conceived in a democratic context as electoral consent, it can also be conceived as consent to a programme or cause.32 This separates the idea of consent from the process of the appointment of those who exercise authority. Thus, in the context of international rule making, values that generate support for a system or consent to a cause provide a route to authority and legitimacy that does not rely on the institutions of democracy – votes, parties and parliaments. Universal Discourses and Value Congruence In the context of international rule making the plausibility of value congruence gravitates around the use and development of universal languages

28

Vibert (2007). This view is associated in particular with David Easton: ‘what distinguishes political interactions from all other kinds of social interactions is that they are predominantly oriented towards the authoritative allocation of values for a society . . . an allocation is authoritative when the persons oriented to it consider that they are bound by it’ (Easton 1965: 50). 30 Easton suggests that values can instill a sense of ‘diffused support’ that acts as an alternative to the ‘structured regulation’ of support achieved by institutional or constitutional means (1965: 124–5). 31 Beetham (1991: 16) argues that evidence of consent to a particular power relationship is one of the three pillars of legitimate power. His two other pillars – conformity to established rules and rules that are justified by reference to beliefs shared by rulers and the ruled – also refer to different forms of value congruence. 32 Beetham (1991) distinguishes between electoral consent and what he calls consent through ‘mobilisation’. 29


15/12/2010 08:09


155

of discourse to serve the dual purpose of an authoritative conciliation of the different types of discourse used in international rule making and the legitimacy of the rules. The appeal of unifying discourses for international rule makers does not simply rest on the need to base their authority and legitimacy on grounds other than on missing branches of government and missing electoral systems. If governments have to negotiate new rules, persuade other governments, or sort out disputes, some kind of generally acceptable reasoning has to be adopted between governments. At the same time, a lack of trust and differences of viewpoints and beliefs between governments may make the informality of ‘muddling through’ both less attractive and less viable. Governments will also feel under pressure not only to give reasons for their preferences to other governments but also to the epistemic elites and to their own electorates. In order to meet these different demands in the context of international rule making, the proposed discourse has not only to claim universal applicability across different jurisdictions. It also has to meet three further criteria. First it must provide a practical form of reasoning applicable to most areas of policy making. Second, it must offer a form of causal reasoning so that a ‘logic of consequences’ is observed. Third, it must include a strong normative element so that normative reasoning is also included and value congruence can be demonstrated.33 Thus, as well as claiming universal relevance, the unifying discourse must be able to combine these different styles of reasoning applicable to rule making within its own encompassing style.34 If these conditions can be met then it may be possible for international rule makers to appeal to support for and consent to what they do in a manner that does not depend on achieving institutional congruence. Legitimacy as well as authoritative conciliation can be claimed. The Candidates At the time when the international system was established in the 1940s one style of reasoning that appeared able to fulfil the necessary criteria was that of economics. It claimed its laws were universal. With so much of international rule making seen as being tied to economic questions its style

33 Cashore (2002) suggests that legitimacy in international rule making depends on pragmatic, moral and cognitive persuasiveness. 34 It is because of the difficulties of meeting these criteria that Snellen (2002) argues against any ‘super-rationality’ and in favour of the view that the best that can be hoped for is co-existence between different types of rationality.


15/12/2010 08:09

156


of reasoning also looked widely applicable in a practical sense. It offered causal reasoning and, because improved economic conditions were seen as leading to desirable political results, there was a strong normative element. Today, economic reasoning is still important in rule making and, since the collapse of communism, economic discourse appears to have adopted the universal language of ‘market liberalism’. However, there is no agreement among economists either about the institutions of the market or on the boundaries between the state and the market. The only unity economics can claim is methodological. In today’s world the new claimant to provide a universal discourse is environmentalism. The problem it speaks to is universal; its message has widespread practical applications from individual human reproductive behaviour and individual consumer behaviour to many collective action or common good problems; it puts forward strong causal theorising organised not only around the theorising of the physical sciences but also around economic concepts such as market failures. It also offers a compelling normative narrative. Finally, it offers a form of ‘cultural or sacred canopy’ – the vision of a ‘garden of Eden’ unspoilt by human intervention, an ‘original sin’ of greed, and a glimpse into the ‘apocalypse to come’ as global warming reaches the tipping point towards a future where earth becomes too hot for human habitation.35 The main rival to environmentalism as the language of universal discourse is the one with the longest tradition – the language of the international rule of law. Its normative principles, embodied in such documents as the UN Charter of Fundamental Human Rights, claim to be universal and to ‘trump’ other normative principles. The compass of the law is also applicable to all areas of public policy, and it offers its own style of causal reasoning. It can employ the reasoning of science in arriving at judgments.36 At the same time, by emphasising the importance of procedural principles, such as what is to count as material evidence, how to treat expert evidence or how to measure what is proportionate, it provides courts at the international level with the same means available to courts at the national level to settle disputes that arise in other forms of discourse such as in economics or science without entering into the substance of a scientific debate. According to some accounts, an effective legal system

35 By encompassing these different rationalities environmentalism goes well beyond the kind of ‘scientisation’ Drori and Meyer (2006) see as a global form of rationalisation and corresponds more to their idea of science being used beyond instrumental purposes in order to provide a cultural canopy. (Drori et al. 2003). 36 Benvenisti (2004) suggests that judges can base their decisions on the best available scientific evidence as a way of reaching ‘neutral’ and efficient norms’.


15/12/2010 08:09


157

can establish a ‘virtuous circle’ that provides for a neutral interpretation of rules, respect for those rules and a greater willingness of citizens to play by the rules.37 Weaknesses of Value Congruence There are evident attractions in thinking that in an inter-connected world there are forms of discourse that can show how rule making at the international level conforms with values expressed at the national. Nevertheless, there are severe limitations to the use of discourses with claims to universal relevance and applicability as a means of achieving conciliation and legitimacy in international rule making. First, there is a question, illustrated by the examples above, as to whether the discourse should engage with the substantive values of politics or with the procedural values of democracy. Environmentalism engages with the substance of much policy debate within countries; the language of the international rule of law deals more with procedural norms. In the context of democratic legitimacy it seems that procedural values congruent with democratic beliefs must also be associated with the discourse even if they are not specifically those about voting and elections.38 Unless procedural requirements are met, it is difficult to judge whether there is ‘diffused’ support for the system of rule making.39 It is also difficult to judge whether or not the particular cause is a common one that achieves the degree of consent that might bring legitimacy through noninstitutional or non-electoral means.40 If however, as in the case of the language of the international rule of law, engagement is with procedural values, then congruence with the substance of political debate in democratic societies may be lost as debate may undermine claims about consent and a common cause. There is no easy way of resolving these dilemmas.41

37

See for example the account in Alter (2001: 217–21). Beetham (1994: 29) notes, ‘the freedom of speech, association, assembly and movement, the right to due legal process . . . are not something specific to a particular form of democracy called “liberal democracy” they are essential to democracy as such, since without them no effective popular control over government is possible’. 39 See the discussion in Dalton and Shin (2006b) and Haerpfer et al. (2009). Dalton and Shin also question the durability of any democratic system unless there is congruence with democratic values (2006a: 9). 40 Bellamy (2010: 5) emphasises the importance of ‘input’ processes ‘that ensure all positions obtain an equal hearing’. 41 Reus-Smit (1999: 33) argues for giving priority to procedure: ‘The existence of a generally accepted norm of pure procedural justice is a prerequisite for ordered social relations, domestically and internationally. Unless there is a minimal, baseline agreement among society’s members about how rules of coexistence and cooperation should be formulated, 38


15/12/2010 08:09

158


Second, in relation to the use of universal discourses as a means to achieve an authoritative conciliation of different discourses, there is a question about ‘precedence’. In order to act as an authoritative discourse that brings other discourses together, the proposed universal discourse has to claim some kind of precedence over other discourses. However, the claim of precedence can weaken or endanger those others. The danger that the choice of a unifying discourse may expose other discourses to the risk of serious abuse has long been recognised in the case of economics. For example, a standard criticism is that the discourse of economics impoverishes moral discourse. The criticism may not be valid. Adam Smith wrote about both. But it is one probable reason for the decline in the standing of economics as a unifying discourse. The danger is particularly acute when the discourse takes on the flavour of a cause or an ideology – as in the case of Marxist economics. Environmentalism is not the new Marxism.42 However, some of the same warnings apply. By framing the specific problem of the human contribution to global warming within the more general issue of ‘climate change’ there is a risk that all evidence of change may be counted as confirmatory evidence of a particular causality; by emphasising the negative externalities of individual actions, all behaviour becomes a matter of social concern and the strong normative messages may weaken other disciplines including the methodology of the social and natural sciences. In other words there is a tension between the epistemic requirements of a universal language that brings together different types of reasoning in an authoritative way and the use of a universal language to mobilise consent or support for a cause. The language of the international rule of law itself is also not immune to criticism when it is presented as a form of ‘super-rationality’ that stands over other forms of reasoning. It has been suggested that the law itself has to adjust its own practices and procedures in order to recognise the validity of different discourses and that any aspirations to provide a single set of universally agreed and applied legal norms are ‘doomed from the outset’.43 A third area of weakness also relates to the authority of universal

no basis exists for collective action or the resolution of conflict, let alone the formulation of substantive principles of justice’. 42 Wuthnow (1989) suggests that the components of an ideology such as Marxism are, first, that they thematise certain features of the social environment, second, set them in opposition to alternative visions of cultural authority and, third, supply figurations that mediate between present and idealised realities. On this definition some would see environmentalism as an ideology. 43 See Teubner (2000) and Fischer-Lescano and Teubner (2004).


15/12/2010 08:09


159

discourses. In particular it relates to a long-standing question about the relationship between higher level norms and higher level powers. The language of universal norms seems to imply that it is justified to use powers at the international level in order to implement those norms. It thus invites a confusion with norms expressed at the top with powers that come down from on high.44 There is no necessary connection, but it may encourage rule-making practices that go in that direction.45 What these areas of difficulty in value congruence share is that each, in their different ways, highlights the problem of distance between the rule makers and the ruled. By downplaying procedural values, important in themselves in a democratic context, it becomes very difficult to judge claims that the values expressed by the rule makers are congruent with the values held by people or that there is a general consent or diffused support. The attribution of an overriding authority to a particular discourse, questionable in itself, does not offer an equal hearing to people who may ascribe an equal or higher value to alternative styles of discourse. The potential confusion between higher level norms and higher level powers devalues the sense in which in democracies people are the ultimate source of political legitimacy. Common languages such as those of economics or environmentalism or the rule of law have a valuable role to play in international rule making in the sense that they apply as a common or shared means of discourse across different jurisdictional boundaries. However it is one thing to recognise their role as a facilitator and it is quite another to take the position that they provide the kind of congruence between values that would enable international rule makers to claim democratic authority and legitimacy even in the absence of missing branches of government. Instead their use highlights the problem of distance. Thus they switch the spotlight to the implications for democratic norms of the reversal of roles that takes place in international rule making – the position of citizens as receivers of rules made and endorsed by others rather than their being, in one sense or another, the principals in the system to whom the rule makers defer.

44 The concept of a hierarchy of norms and its association with a hierarchy of legal authority and powers can be attributed to Hans Kelsen and his search for the ‘basic norm’ from which authority could be derived without infinite regression between norms. For a brief discussion of Kelsen in this context see Vibert (1995: 57–9). 45 Even if a hierarchy of powers is preferred Easton (1990) suggests that the relationship between higher and lower authority does not have to be a determining one but can simply be ‘facultative’.


15/12/2010 08:09

160


DISSONANCE AND DIVERGENCE Role reversal can be viewed from two perspectives. The first is a statecentred perspective that identifies ‘divergence’ as the essential problem for democracies. Divergence occurs whenever the policy preferences of the state differ from those of citizens. The second is a societal-centred perspective that looks at how individuals behave when having to choose between alternative policy preferences. It distinguishes between decision-making processes before a decision is actually taken and those after. In the ‘after’ period, feelings of cognitive dissonance have to be resolved in order for the decision to sit comfortably and not to be re-opened. When citizens are modelled as receivers of policy after decisions have been taken, their feelings of dissonance have to be resolved in order for a policy to receive acceptance. If divergence occurs or dissonance is not resolved the result is that citizens dissent from the rules. State Preferences and Divergence The state-centred perspective is relevant because it is governments that negotiate international agreements or endorse the proposals of expert bodies. Sometimes this relevance is expressed in terms of ‘pooling points’.46 A pooling point is the point at which arguments on an international agreement cease and the form and content of a rule are affirmed. It may be viewed as the position where the agreement is actually reached and countries ‘pool sovereignty’. The state-centred perspective assumes that the state can be viewed as an entity composed of all those public officials, elected and non-elected, who are involved in policy making and who are endowed with society-wide decision-making authority. The state is seen as a largely autonomous actor with its own preferences.47 In cases where there is no divergence between state preferences and societal preferences, the state can follow its own wishes and also can try to forestall any future divergence. Even if there is a divergence the state can try to shift public opinion. According to this account the difficulty from the perspective of democratic procedures is when public officials use their authority to free them-

46

Lake and McCubbins (2006: 366–7). Nordlinger (1981: 1) from whom this account is drawn asserts, ‘The preferences of the state are at least as important as those of civil society in accounting for what the democratic state does and does not do’. He defines state preferences as those with the weightiest support of public officials behind them (ibid. 15). 47


15/12/2010 08:09


161

selves from societal restraints and rely on the inherent powers of the state to take action and to override divergent views. From a theoretical perspective there are good reasons why state preferences could regularly diverge from societal. There may be different views about what is important. Even if there is an agreed view about the importance of an issue, its salience may still be seen in different ways. Against this theoretical possibility of a pervasive problem, there are however practical reasons to assume that in modern democratic societies policy divergence should not occur to any great extent for any prolonged periods of time. Governments can be thrown out of office if they consistently fail to represent policy preferences over a long period of time and they have many ways of sampling public opinion in order to keep abreast of changing public attitudes. Empirical research indeed suggests that there is policy responsiveness although it may vary across domains.48 Nevertheless, when divergence occurs, even if it is relatively rare, there is a democratic problem. Citizens and Dissonance What can be viewed from a state-centred approach as a limited problem of ‘divergence’ can be viewed from a societal approach as a more pervasive problem of ‘dissonance’. Dissonance provides a different perspective on policy responsiveness. In decision-making theory dissonance arises as a natural consequence of the process of having to choose between alternatives. According to the theory, there is likely to be some regret associated with any choice, both because some alternatives will have to have been put aside and because the actual choice may turn out to be mistaken. In order to actually arouse this feeling of dissonance what is termed an ‘aversive event’ is needed.49 Although the theory has evolved in the context of studies of cognition, dissonance is not so much about inconsistency between cognitions as such but with the production of a consequence that is unwanted.50 In other words, citizens may not want to know in detail how epistemic elites have arrived at their conclusions or governments theirs – but they will be aroused if the consequences of those ways of reasoning are perceived as unwanted or against their own interest. When dissonance is looked at in the context of individual choice, the 48

See Wlezien (2004). Cooper and Fazio (1984: 232) define an aversive event as ‘an event that blocks one’s self-interest or an event that one would rather not have occur’. 50 Cooper and Fazio (1984: 234). 49


15/12/2010 08:09

162


theory suggests that the individual will act in ways to reduce dissonance – by looking for evidence that reaffirms the choice and provides reassurance that the right choice was indeed made, and by rejecting contradictory evidence.51 When this is generalised to society as a whole it means that even where public opinion is in broad agreement with government policies there may still be regret associated with the choices made and a continued search for confirming reasons. Equally important, if there is divergence or disagreement with the preferences of government, public opinion may look for evidence that confirms its own preferences and will reject contrary evidence.52 Confirming Preferences Dissonance theory highlights two important aspects of accepting a decision after it has been made. The first is that even good decision making in the period leading up to a decision may be followed by dissatisfaction afterwards. In the context of public policy this means that even if the rulemaking bodies have proposed measures that are soundly grounded on the best available evidence, even if governments have endorsed and commended them to their legislatures and justified them in terms of their consistency with their own values, and even if electorates had been involved at each of these prior stages, electorates will still need further reasons of their own in order to accept the proposed rules. Governments will try to reduce divergence and dissonance by selecting evidence that supports their choice and by suppressing contradictory indications. Citizens will be asked to accept what has been done in their name and told there is no other negotiable alternative. However, in the acceptance process, citizens may not trust the selective evidence produced by governments and may resent being told that ‘there is no alternative’.53 The second aspect is that the reduction of dissonance with the original decision does not involve reopening the particular logic followed by the experts or by governments but involves looking selectively for evidence that is confirmatory according to the individual’s or community’s own criteria. It goes beyond the notion of accountability where the experts and 51 See Festinger (1959 and 1964). For a more recent discussion of cognitive consistency see Johansson-Stenman and Svedsäter (2008). 52 Anderson et al. (2005: 26–27) cite the support given to such cognitive consistency theories in the context of political behaviour. 53 For evidence on distrust in government see Pattie et al. (2004) and Dalton (2008). Dalton notes ‘People express more confidence in non political institutions of government, such as the judicial system or the civil service, than in the institutions of representative democracy’ (ibid.: 247).


15/12/2010 08:09


163

the governments would be asked to justify their choices in their terms.54 The reduction of dissonance is about addressing a perception that a measure is possibly against an individual’s own interest or desire. In the example of the acceptance of hybrid corn by the farmers of Iowa, the accepters (the farmers) were not prepared to accept the scientific reasoning of the experts (the laboratory scientists) until they had seen the results in their own fields and they were not prepared to accept the word of the government agencies and seed companies (the endorsers) until they had seen their own peers prove that the new seed achieved superior yields. In the end, when their own experience and that of their peers proved persuasive, hybrid corn was rapidly accepted on a voluntary basis. Other accounts of the adoption of the new corn suggest that the farmers did have some regrets about accepting the new seed because the mature corn lacked the aesthetic qualities of the old and the associated social rituals of crowning a ‘corn queen’. Any such feelings were however clearly eliminated when the farmers confirmed for themselves the improved cash flow from higher yields and when their own acceptance was confirmed by the behaviour of others.55 In this example, the reasoning behind acceptance can be related to a simple goal of achieving higher yields and greater income. In many cases the requirements needed to overcome dissonance cannot be established so easily. It is more likely that acceptance is related to meeting a set of requirements.56 Accounts suggest the importance within the set of immediate personal experience, reliance on others in a social network who can be trusted and reliance on evidence that is visible and local.57 It is a set that echoes the reasoning of the Iowa farmers. The International Rule Making Context In the context of international rule making, the potential for divergence between state preferences and popular preferences is probably greater than in the case of national policies, and the difficulty of overcoming feelings of dissonance in the public is also likely to be greater. Governments do not act in the same way at the international level as they do at the

54

See Bovens (2007) for the limited definition. American Vice President Henry A. Wallace was a pioneer developer of the new seeds. The social rituals associated with the traditional corn and the social resistances to the adoption of the new are described by his biographers Culver and Hyde (2000). 56 Snellen (2002). 57 See Wolman and Page (2002). 55


15/12/2010 08:09

164


domestic level.58 As discussed earlier they employ strategic reasoning for negotiating with other governments that they do not need in a domestic setting. In addition, the need to link the international negotiating setting to the implementation setting across a variety of domestic circumstances adds to the difficulties of rule formation and implementation encountered within a single jurisdiction. There are further differences between the domestic and the international context. In domestic electoral terms the voice of the people is often a negative one of rejecting governments and policies rather than of proposing policies for governments to follow.59 However in the case of domestic policies electorates can vote with their feet if their concerns are consistently ignored. As a last resort they can move to another jurisdiction. When it comes to international rule making ‘exit’ is not an option since most countries have signed up. Thus, only voice counts. At the same time the power of ‘voice’ is much weaker. Any one government can blame the rule on other governments who can be punished through the ballot box only by their own citizens and are otherwise out of reach. Dissent expressed in any one jurisdiction is also likely to be ignored by others.60 Dissonance and the Democratic Foundation The relationship of divergence and dissonance to voice suggests that there is a straightforward connection between the role reversal problem in international rule making and theories of deliberative democracy that emphasise the importance of voice. According to these accounts of the democratic process, deliberation and discourse are seen as the way to achieve social learning, to take into account multiple contested discourses and to achieve conciliation.61 If this were the case a transition could be made to considering the role of deliberative democracy in approving international rules. The transition cannot be made in this way. First, voice does not necessarily resolve questions of dissonance or divergence. There is relatively little empirical evidence that debate changes people’s minds so that a feeling of dissonance can be changed simply by discussion with others with different views. Differences in initial conditions in terms of information 58 See Chayes and Chayes (1995) for one discussion of why the international context differs from the domestic. 59 One of the earliest findings of studies of electoral behaviour was that voters vote against rather than for. 60 In the EU this leads to the situation where voters are asked to vote again if they initially disagree with the proposals of governments. 61 See Dryzek (2006).


15/12/2010 08:09


165

and opinions of a group may be determining.62 In addition, there is evidence that debate may harden previously held positions.63 Second, the motivation behind dissonance may not always be that which can be expressed and theorised in debate. Not all knowledge is theorised or expressed.64 Moreover dissonance reflects perceptions rather than facts. In the case of the farmers mentioned above, it was the practical example set by other farmers who were respected in the community that seems to have been the key to acceptance. Third, as mentioned above, dissonance theory detaches the feelings of regret that accompany decisions from the debate about whether or not the original decision was good or not. Even good decisions will be accompanied by feelings of discomfort. Removing dissonance is therefore about selecting confirmatory evidence that is seen as salient to perceptions of the individual or group. Debate may throw up contradictory evidence. One possible approach to solving the problem of dissonance is provided by game theoretic approaches to ‘asymetrical policy making’.65 In this version the conflict between a government and public dissonance can be changed from a ‘zero sum’ game where one side can gain only at the expense of the other to a ‘positive sum’ game. What achieves this transformation is the supposed role of the floating voter who is assumed to ‘respond to the dispute’. This approach has the advantage that neither the appeal to the floating voter, nor the response, depends on the particular rationality assumptions of discourse democracy – the floating voter might respond to appeals in a variety of forms. However, the weakness in the account is the ‘deus ex machina’ assumption of the floating voter ready to be conjured up as a tie breaker or moderator either by the government or by the expression of public dissonance. Dissonance that cannot be resolved leads to dissent. Divergence is also about dissent. Thus at the heart of the problem of the democratic deficit in international rule making lies dissent. Citizens sit at the receiving end of policies made by others from which they may dissent. The issue thus 62

See Shiller (1995). Sunstein (2008: 2) notes, ‘Like-minded people tend, after discussion with one another, to end up thinking a more extreme version of what they thought before they started to talk’. 64 What has been popularised as the ‘wisdom of crowds’ and is now gathered in various kinds of prediction markets can be traced back to Hayek (1945: 521) when he wrote, ‘today it is almost heresy to suggest that scientific knowledge is not the sum of all knowledge . . . there is beyond question a body of very important but unorganized knowledge which cannot possibly be called scientific in the sense of knowledge of general rules: the knowledge of the particular circumstances of time and place’. 65 Scharpf defines asymmetrical policy making as a situation where, ‘only the government is capable of adopting and implementing effective policy choices and the opposition must choose between challenging those choices or letting them go’ (Scharpf 1997: 185). 63


15/12/2010 08:09

166


becomes one of how to treat dissent. How it is treated lies at the core of democratic practice within countries. How it is treated in international rule making lies at the core of bringing international rule-making practices into conformity with democratic norms. *

*

*

Before continuing the discussion of the treatment of dissent in democratic practice, the next chapter looks at the question of protection against the cognitive failings of epistemic elites.


15/12/2010 08:09

8.

Challenge systems and the rule makers

On October 6, 1973, Syria and Egypt launched a simultaneous attack on Israel in what came to be known as the Yom Kippur War. Israel had mobilised its forces only ten hours in advance of the attack. By the next day Israel’s main defence force in the south had only 103 tanks left out of its original force of 290 and in the north only 25 out of 77.1 On October 9 (the ‘low point’ for Israel and ‘high point’ for Egypt and Syria) Israel was considering ‘desperate measures’.2 The war is relevant to international rule making because it constitutes an important example of policy failure in a different context. Subsequent analysis of why Israel failed to anticipate the attack illustrates each of the sources of failure identified earlier. There was executive failure in the sense that the Head (Major General Eli Zeira) of military intelligence (AMAN) has been characterised as a ‘dominant personality’ who allegedly suppressed beliefs and signals that challenged his view that an attack was a low probability.3 There was organisational cultural failure in the sense of ‘group think’ that infected some key intelligence committees.4 Above all there was a failure of method or what has been labelled earlier as ‘cognitive failure’. AMAN’s working method was to produce a single ‘research opinion’ that omitted any debate or contrary opinion within AMAN. Moreover the Research Branch within AMAN held a monopoly in making national intelligence assessments, so its assessments went unchallenged.5 In this particular example of cognitive failure, Israel’s unpreparedness was not the result of insufficient information. There were numerous warnings of an attack from highly trusted and reliable sources. What was crucial was the way in which the intelligence was interpreted or misinterpreted. The example not only illustrates one variety of the kind

1

Rabinovich (2004). Ibid. 3 Bar-Joseph (2000). 4 Bar-Joseph (2005). 5 ‘On the eve of war there was not a single person or agency that could present an alternative to AMAN’s dominant estimates’ (Bar-Joseph 2005: 242). 2

167


15/12/2010 08:09

168


of cognitive failure discussed earlier but also suggests where some of the defences against methodological failure might lie – particularly in respect of the need for the competitive evaluation of data. This chapter discusses how to protect institutions and ultimately the citizen, against the cognitive failure of experts in the rule-making context. Earlier discussion described how the new venues and actors involved in international rule making aim to get away from a political environment. By depoliticising the setting, the collegial and technocratic virtues of epistemic elites can flourish. At the same time those same virtues make international venues, both old and new, vulnerable to other epistemic weaknesses – specific forms of cognitive failings leading to defective rule making. The question thus is what kind of institutional defences can be erected to guard against leading types of cognitive failure. In the case where the failure in rule making can be attributed to executive failings in the sense of mistakes by governments, there are wellestablished benchmarks for a relevant debate. Theories of competitive federalism, or about how to define the optimum domain for public policies, might provide guidance on issues of scale, or the division of authority, so that mismatching or misalignments can be reduced.6 Similarly, in the case of failings that can be attributed to the culture of an organisation there is a well-established cultural theory perspective that links failure to particular ways of organising.7 However, in the case of failures of method associated with various forms of cognitive failure, there is a lack of an established model for the analysis of institutional remedies beyond the general recognition that incentives in the knowledge world have to be institutionalised.8 Moreover, the attitude of the rule makers themselves to institutionalising defensive procedures may well be ambivalent at best. This is because any defence against cognitive weaknesses involves further formalising procedures and making them more rule governed, effortful, expensive and slow.9 The rule makers thus see themselves as trapped between operational pressures that call for speed of response and methodological rigour that calls for more time to be taken in preparing a response. More rigorous and transparent procedures also bring out the uncertainties in regulatory decision making and, particularly when under political pressures to do something, regulators may prefer to narrow their choices to the simple question

6 For example, the framework provided by Tiebout’s ‘pure theory’ of local expenditures (1956). 7 For the use of this kind of model see Hood (1998). 8 Jegadeesh and Titman (1995). 9 Kahneman and Frederick (2002).


15/12/2010 08:09


169

of whether to act or not to act rather than to nuance their choices.10 Governments also often do not want to be presented with uncertainties. They want answers or recommended courses of action. At present, expert regulatory groups use two main techniques in order to achieve quality control over what they do. The first is impact assessment. This involves gathering evidence and views grouped mainly around an assessment of the estimated costs, benefits and social impact of a proposed measure. However, while impact assessment provides a useful way of marshalling evidence, it does not in itself guard against cognitive failings and is itself susceptible, for example, to confirmatory and framing bias.11 At its worst, impact assessment is no better than a way of organising bias. It is therefore not considered further in the discussion below. The second form of quality control is peer review. This chapter outlines its main features. It is a method that has long had its critics. The chapter therefore goes on to describe defences against cognitive failure that go beyond peer review. They involve more formalised and extensive methods of challenge.12

PEER REVIEW Peer review can be defined as a method for screening the quality of empirical and theoretical research and analysis by those with similar or higher authority in the field in which the research is submitted.13 It has a very long history but currently is best known in two fields, in the refereeing of academic articles for inclusion in journals and in the award of grants for research funding.14 It also has a long established role in the review of professional performance in, for example, clinical settings. It is a powerful technique because its role in the award of grants brings it in at the first stage of epistemological enquiry when the problem to be examined is

10

Collins and Evans (2002: 246) state that regulators ‘have no use for small uncertainties’. See Vibert (2005). 12 A further technique to help ensure quality that may become of greater importance in future is the use of regulatory experiments. However, any experiment involves what is termed an ‘experimental gap’ – the distance between the experiment and the real world. Cognitive failings can enter into the way the experiment is set up or in the way in which results are interpreted. Thus, experiments can add to knowledge but also need to guard against cognitive failings (see Lezaun and Millo 2006). 13 Merton (1996) suggests that peer recognition and collegial esteem act as a kind of ‘property right’ for scientists because they provide a means of validating the worth of work without being directly compensated for any contribution to knowledge. 14 The US General Accounting Office investigation of peer review in federal science agencies noted, ‘Peer review has been used to judge the quality of science for over three centuries’ (US GAO 1999: 3). 11


15/12/2010 08:09

170


defined, performance review gives it a continuing role during the period of enquiry and its role in the refereeing of articles brings it back at the last stage when findings are disseminated.15 The Criticisms There is also a history of criticism. Criticism has in part been about method. For a procedure that is so central to academic endeavour the validity of peer review has been under-examined.16 The criticism is also about outcomes.17 There is a common view that peer review of journal submissions does not stop the publication of low quality analysis.18 In the social and natural sciences a particular aim behind the peer review of journal submissions is to evaluate or guarantee the significance of data and the soundness of its interpretation. Findings should be ‘reproducible’ even if, for example because of data confidentiality, not exactly ‘replicable’.19 However, a recent case study found that the peer review process does not always review the accuracy of data used in articles or computational methods or generally request replication of data. Nor does it necessarily insist that the authors make available their data and methods for inspection so that independent checking is possible.20 Peer review of grants is also alleged to favour established lines of research rather than novel thinking and novel approaches.21 Peer review of performance may also serve a defensive purpose to protect professional reputation.22 It may also be used to protect professional boundaries. There are additional reservations about peer review when it is employed in the rule making and regulatory process. The process of selection of peers may be biased; there can be a blurring of what is independent exter-

15

For a description of its uses see US GAO (1999). A recent investigation of how grant-giving panels work in the US concluded that ‘peer review processes themselves should be subject to further examination’ (see Lamont 2009). 17 See, for example, Schutz (2004). 18 However, an investigation of the peer review of manuscripts carried out by H.D. Daniel did not support this allegation (see Daniel 1993). 19 See the discussion in the Workshop Report of the US NRC (2003: 12–17). 20 See McCullough and McKitrick (2009) for a critical review of practices. 21 Cole’s case study of grant giving at the US National Science Foundation concluded, ‘There is so much disagreement among equally qualified reviewers of a proposal that whether or not a proposal is funded is to a large extent a result of luck in the program director’s choice of reviewers’ (Cole 1992: 84). 22 For a review in the medical field, see Lang (1999). 16


15/12/2010 08:09


171

nal review and what is internal assessment; there can also be blurring of what is science and what is policy.23 Despite all these criticisms, peer review continues in use. Publishers and grant makers need some method of review and examiners equally need some professionally approved benchmark for evaluating the performance of experts. There seems to be a lack of alternatives to peer review that would not introduce greater degrees of arbitrariness in judgements.24 The response to criticism has mainly taken the form of endorsing the idea that in theory peer groups should be diverse.25 In practice, diversity is often avoided in favour of like-mindedness. Regulators and Peer Review Regulators and rule makers also continue to rely heavily on peer review. They do so in four different ways. First, where regulators are looking for scientific evidence surrounding an issue, regulatory guidelines may stipulate that evidence and analysis must come from peer-reviewed sources, or that there must be special justification where this is not the case.26 This is a way of validating data and the specification of models. Second, where regulators are looking to relate evidence to a body of theory, they are likely to turn to the academic, peer-reviewed literature for the theoretical link.27 This is a way of validating causal reasoning. Third, regulators turn to peer reputation and measures of peer esteem (for example, the number of citations) for the validation of the credentials of experts they employ or use. Finally, regulators turn to their professional peers in the much broader sense of approval for their own conduct and performance. In this way the soundness of the methodologies they use is also validated. In the case of international rule making, peer review procedures vary in their degree of rigour. They range from the internal review procedures 23

See Jasanoff (1987). Lamont’s study found that ‘It is impossible to reach a definite, evidence-based conclusion concerning the system as a whole’ (2009: 7) and concluded that ‘biases are unavoidable . . . but if panellists . . . make considerable sacrifices to do a good job, they contribute to a relatively meritocratic system’ (ibid.: 243). 25 A 1999 report on the peer review program of the United States EPA stated, ‘the important goal is achieving balance in the spectrum of biases represented in the review group’ (US RSAC: 11). 26 See for example, Office of Management and Budget: Bulletin for Peer Review dated December 16, 2004. 27 The European Commission states in connection with economic evidence submitted in competition cases that ‘An economic or econometric analysis should explicitly discuss whether the theory or technique has been generally accepted in the scientific community’ (EU DG Competition; ‘Best Practices for the Submission of Economic Evidence and Data’, dated January 2, 2010). 24


15/12/2010 08:09

172


by fellow economists of draft economic reports prepared by the staffs of the IBRD and IMF, to the externally oriented peer review procedures of the WHO/ILO/UNEP in the case of their risk assessments of exposure to chemicals, and the similar external review procedures of the IMO in the case of its various scientific groups. The WHO draws from a roster of experts chosen by governments in drawing up and reviewing its International Health Regulations (IHR). A highly formalised example is provided by the case of the Intergovernmental Panel on Climate Change (IPCC) that was specifically set up to provide the world with a clear scientific view of the current state of climate change and its potential environmental and socioeconomic consequences. The example of the IPCC also illustrates the difficulties in combining peer review standards with practical policy making. In the case of the IPCC the first set of difficulties arises with the choice of experts – those who compose the Bureau that leads the process of developing the clear scientific view and the experts who provide the substance of the reports. The selection guidelines for experts requires that all nominees ‘shall have relevant scientific, technological or socio-economic expertise’, but the nominations come from governments whose policy-making concerns may be reflected (in the case of the Fourth Assessment about 600 experts were selected from about 4000 nominations). The policy concerns of governments are also reflected in the selection of the choice of members for the Bureau where the guidelines stipulate that, ‘The overall composition of the IPCC Bureau and the Bureaux of any Task Forces of the IPCC shall reflect broad geographic representation with due consideration for scientific and technical requirements’. In order to ensure broad geographic representation, regional quotas have been agreed. This balancing of expertise and policy considerations in the selection process is shown in Table 8.1 (opposite).28 The second area of potential difficulty concerns the material used in IPCC reports and the way in which it is presented. Here the IPCC makes a distinction between peer-reviewed material (to be used, ‘as far as possible’) and other material. In the case of non-peer-reviewed material the guidelines stipulate that the data must be made available to experts and reviewers. The IPCC also draws a distinction between different types of report.

28 Quotations in the text on these procedures and the material shown in the tables on IPCC processes are drawn from The Principles Governing IPCC Work, Appendix A, ‘Procedures for the Preparation, Review, Acceptance, Adoption, Approval and Publication of IPCC Reports’ (adopted 2003, as amended 2008) and from the Decision Framework for Special Reports, Methodology Reports and Technical Papers (adopted 2003, as amended 2008).


15/12/2010 08:09


Table 8.1

173

Intergovernmental Panel on Climate Change: key actors

Key actors

Chosen by

Chair of Bureau Bureau members (31)

Member governments Member governments (from regional quotas) Bureau from nominations by governments Member of Bureau + independent expert from list provided by governments

Coordinating and lead authors (prepare drafts) Review editors (assure balance)

In particular a distinction is made between Assessment Reports (and the Technical and Special Reports that may back them) and Synthesis Reports and Summaries for Policy Makers (used in both Assessment Reports and Synthesis Reports). In this way a distinction is made between the underlying material and the policy inferences drawn from the material. These distinctions are shown in Table 8.2 (overleaf). The difficulty is that these classifications still do not make entirely clear the standards that have been applied to the evidence or their conformity with standards such as transparency and reproducibility.29 The third area of difficulty in distinguishing between expert judgements and the concerns of policy makers is in the report preparation process itself. Here there is a distinction in IPCC processes between the role of governments in Assessment Reports and the role of governments in the Synthesis Report and its Summary for Policy Makers (SPM). In the case of Assessment Reports governments may comment on the draft and accept or reject the SPM. In the case of the Synthesis Report the drafting is in the hands of the IPCC Chair and a team selected by the Bureau, governments comment on the draft and the Panel must approve (line by line) the SPM. The Panel also has a role in conforming afterwards the underlying report with their agreed SPM. The policy concerns of governments thus

29 The Report (the Oxburgh Report dated April 12, 2010) of the International Panel set up by the University of East Anglia to examine the research of the Climatic Research Unit (CRU) that feeds its analysis into the IPCC, examined 11 peer-reviewed publications of the CRU. It noted in relation to deriving evidence of climate change from tree rings (dendroclimatology) used by the IPCC, ‘the potential for misleading results arising from selection bias is very great in this area. It is regrettable that so few professional statisticians have been involved in this work because it is fundamentally statistical. Under such circumstances there must be an obligation on researchers to document the judgemental decisions they have made so that the work can in principle be replicated by others’ (para. 6 p. 3).


15/12/2010 08:09

174

Table 8.2


IPCC inputs and outputs

Inputs

Outputs

Peer-reviewed material (as far as possible) Selected non-peer-reviewed material IPCC supporting material Unpublished material (available to experts and reviewers)

Technical papers (for topics included in Special and Assessment Reports for which ‘objective scientific/technical perspective essential’) Special Reports (assess specific issues) Assessment Reports (provide full scientific and technical assessment of each Working Group) Synthesis Reports (synthesise and integrate materials from Assessment and Special Reports)

play a potentially larger role in the preparation of Synthesis Reports. The two procedures are shown in Table 8.3 (opposite). The IPCC is at the forefront of international organisations in trying to base regulatory policy on evidence. Nevertheless, the example of the IPCC illustrates the difficulty of providing quality assurances about the integrity of the inputs used in its assessments as well as the difficulty of separating expert processes, based on peer review and peer approval, from policy processes that may introduce non-scientific considerations. The complexities do not invalidate peer-review processes. They do however suggest that peer review by itself cannot be relied on to make transparent the biases that may accompany the defining characteristics of the epistemic elites at the centre of international rule making – their shared principled beliefs, their common notions of validity, the shared causal beliefs and their sense of being engaged in a common problem-solving venture.

BEYOND PEER REVIEW Peer review is likely to continue to play an ongoing role in the validation of rule-making standards. The issue is not to replace it but to improve it and to look for supplementary means of defending against the weaknesses of epistemic elites in rule making. There are four types of institutional defence that are particularly important. They involve the use of competitive evaluation, process tracing procedures, the assessment of confidence levels and the deployment of continuing audit to examine the quality


15/12/2010 08:09


Table 8.3

175

IPCC report preparation processes

Stages (a) Assessment Reports Preparation of drafts First Review Revised Draft Second Review Final Draft Acceptance Approval of summary for policy makers (SPM) Acceptance of summary for policy makers

(b) Synthesis Report Draft Review Revised Draft Approval of summary for policy makers Review of underlying report

Adoption of underlying report

Actors Lead authors Experts Lead authors Experts and governments Lead authors and review editors Working group (WG) WG (line by line agreement) Panel (accepts report represents a comprehensive, objective and balanced view of matter and cannot change it)

IPCC Chair in lead + team agreed by Bureau Experts and governments IPCC chair + team agreed by Bureau Panel (line by line agreement) Panel (checking for conformity with SPM and consistency with underlying Assessment Reports) Panel (section by section endorsement)

of causal reasoning after a regulatory choice has been made. They offer important additional elements of challenge and review. They can themselves be seen as a set of practices grounded on principle. Each is discussed below. Competitive Evaluation Competitive evaluation involves the use of two or more independent teams to evaluate evidence and any modelling of evidence at the starting point of a regulatory assessment. It aims to address two aspects of the assessment where elites may have cognitive failings in the form of ‘framing bias’ discussed earlier. The first is the vulnerability of like-minded groups of experts with shared principled beliefs to start their evaluation of evidence from a predetermined ‘reference point’ or to ‘anchor’ their attentiveness to new data in an initial value. The second is the risk that like-minded groups


15/12/2010 08:09

176


will construct an interpretation of evidence in ways that confirm the prior belief shared between the experts.30 One approach to what might loosely be called competitive evaluation is illustrated by the BCBS in its revisions to its capital adequacy framework following the financial crisis. In this case the preparation for what is termed ‘Basel III’ involves the BCBS studying the impact of revisions on individual banks and the banking industry while a different group (the BCBS, FSB, BIS and IMF) assess the link between higher capital standards and broader economic growth. Together, the two groups combine a ‘top-down’ assessment with a ‘bottom-up’ granular assessment. A much more structured approach to competitive evaluation is adopted by the European Medicine Agency (EMEA). Box 8.1 highlights the key points in competitive evaluation used by the EMEA. The EMEA relies systematically on such a method in its authorisation procedures in order to achieve its mandate for ensuring the safety of medicinal products and treatments for human and animal use in the European Union while at the same time ensuring that products reach patients as quickly as possible. The particular example relates to the procedures to be followed by the EMEA’s standing working party (SAWP) set up to provide pre-authorisation scientific advice to businesses so that regulatory uncertainties about future approvals can be reduced. It reflects the EMEA’s experience with its use of competitive evaluation methods in its authorisation procedures. The example given in Box 8.1 shows that competitive evaluation is used not to replace peer review but to increase the degree of challenge involved. It also shows that consensus is still prized. At the same time the procedure allows for the expression of dissenting views. Process Tracing Process tracing serves two distinct purposes. First, it can be used as a research method to supplement statistical analysis. Second, it can be used to trace the path of a decision and to identify those key decision or branching points that may have foreclosed one decision and steered the analysis towards the choice ultimately taken.31 For the latter purpose the regulatory process is divided into key decision stages so that the component parts of the overall decision that may emerge at the end of the process can be identified and traced back. Most regulators face important ‘choice

30 31

See Kahneman and Tversky (2000) for a distinction between two types of framing bias. For a detailed discussion of process tracing see George and Bennett (2005).


15/12/2010 08:09


177

BOX 8.1 COMPETITIVE EVALUATION IN PRACTICE: THE CASE OF EMEA/SAWP ● ●

● ● ●

●

●

●

For any scientific advice or protocol assistance procedures the chair of SAWP shall appoint two coordinators. For protocol assistance, if the request includes issues relating to the demonstration of significant benefit, a third coordinator shall be appointed. Each appointed coordinator shall form their assessment team with external experts and/or internal assessors. Whenever possible the scientific or protocol advice shall be adopted by consensus of SAWP members. Discussion meetings shall be organised where SAWP does not agree with any important aspect of the programmes proposed by the applicant or when no consensus within SAWP members can be reached. When the need for a discussion meeting is agreed, the coordinators and other SAWP members may nominate additional experts to attend. The meeting shall be open to all SAWP members. If a consensus cannot be reached, the advice shall reflect the majority of SAWP members, while taking account of other positions. The advice should reflect the preferred option. SAWP members shall act as peer reviewers for the final letters of scientific advice.

Source:

EMEA/CHMP/SAWP/69686/04 Rev. 7, dated June 25, 2009.

points’ in their decision process. Process tracing records these choice points and makes explicit not only the empirical evidence that was influential at these choice points, but also any otherwise implicit assumptions made at these points. The main purpose of process tracing is to try to combat the vulnerability of expert groups that have common notions of validity to confirmatory bias and to attribution bias. As discussed earlier, both lead to over-prediction from uncertain data and to over-confidence in a starting hypothesis. Unlike the procedure for competitive evaluation, process tracing is not just about getting more information for a decision because more information may be interpreted in ways to confirm an original


15/12/2010 08:09

178


BOX 8.2 PROCESS TRACING IN EFSA SCIENTIFIC OPINIONS Identification of precise question to be addressed. Identification of relevant sources of information. Discussion of how information is evaluated. Identification of key issues leading to opinion. Listing of references and documentation. Identification of members of panel and subgroup. Identification of any minority opinion. Minority opinion shall be attributed to authors. Minority opinion shall include supporting argumentation. Source: Advice from the EFSA Scientific Committee on a general format for scientific opinions of the EFSA, Dated September 1, 2003.

view.32 The focus is on the integrity of the process itself, the reliability of the inferences drawn from data and the validity of the causal connections made at key points. The division of the rule-making process into stages forces each key assumption or assessment or input into an eventual decision to be explicitly identified. The example given earlier of the way in which the IPCC divided its reports into Technical Papers, Special Reports, Assessment Reports and Synthesis Reports can be seen as a crude form of process tracing. However, it does not make choice points and key judgements about the quality of data or the reliability of different types of methodological analysis fully explicit and transparent. An example of a more refined approach to process tracing, drawn from the European Food Safety Authority (EFSA) for reports that provide the support for EFSA Scientific Opinions, is given in Box 8.2. The way in which these different stages are handled can be illustrated by a recent Scientific Opinion of the EFSA about the extent of food poisoning (human campylobacteriosis) in the EU that can be attributed to the handling, preparation and consumption of broilers (poultry specifically reared for meat production).33 In this particular case the identification of relevant sources of information involved identifying the methodologies for

32

See Rabin and Schrag (1999). Scientific Opinion from EFSA Panel on Biological Hazards, EFSA Journal (2010) 8 (1), 1437. 33


15/12/2010 08:09


179

the attribution of human illness to specific sources and a reporting of the findings that stemmed from the use of these different methodologies. In discussing how information was evaluated the report compared the results and the strengths and weaknesses of the different methodologies as well as discussing the quality and availability of the data. In coming to their conclusion that the handling, preparation and consumption of broilers might account for 20 to 30 per cent of human cases of campylobacteriosis, the report identified the four key pieces of evidence that led to this conclusion. It also advised care in relying on this conclusion because of the uncertainties in the data. The report highlighted the lack of representative data for the EU and gave a wide estimate of the possible incidence of poisoning (which may be not less than 2 million and possibly as high as 20 million cases per year in the EU). In view of the uncertainties, the recommendations of the report focused on improving the data base, better surveillance and on developing methodologies for more accurate source attribution. In this case there was no minority opinion. This particular example also shows why techniques such as process tracing are not always welcome to politicians or even to regulators themselves. The wide estimate of the possible incidence of campylobacteriosis may be seen by politicians as making it unclear as to whether they should or should not be taking action and they may press for more precise estimates than the evidence and analysis allow. Process tracing helps preserve the integrity of the analysis against this kind of pressure. Confidence Reporting Confidence reporting involves more formal methods of rating uncertainties in the decisions of rule makers. It applies to two aspects of regulatory decision making. First, ratings can be applied to indicate the reliability or otherwise of the data used in an assessment of the problem and its solution. It involves combating over-confidence in data arising from the use of heuristics such as categorisation or availability bias by reverting to a more formal statistical concept.34 Second, ratings can be applied to the uncertainties to be attached to the causal reasoning of the rule makers, the way in which they construct models and the way in which they draw conclusions from the evidence.35 It is a way of trying to mitigate herding instincts

34

See Gilovich and Griffin (2002). Simon draws attention to the importance of prior knowledge for the making of causal attributions and notes that ‘determining the causal mechanisms that control phenomena is an inductive task. No finite body of data can point unequivocally to one particular causal model and exclude all others’ (Simon 1979: 79). 35


15/12/2010 08:09

180


among experts, to test for heterogeneity in model-building assumptions and to ensure that minority views are given a weight in the assessment process.36 It is also a way of trying to allow for the unexpected and unanticipated consequences that often accompany regulatory decisions because it forces the decision makers to consider the possibility in advance that they may have got the assessment wrong and the proposed regulatory action incorrectly specified.37 Table 8.4 (opposite) illustrates the treatment of uncertainties by the IPCC in the case of its Fourth Assessment Report in 2007.38 The table shows the three different metrics used by the IPCC. The first uses confidence terminology itself in order to show the level of confidence of the authors in their scientific understanding of climate change and the extent to which human activities contribute to it. The second measures the likelihood of specific results using a probabilistic assessment scale. The third represents a qualitative approach that combines measures of agreement among the experts (from high levels to low) with their judgement on the amounts of evidence (from limited to much) in support of their agreements. The use of the qualitative metric is intended to reflect the difference between the disciplines of the natural sciences and those of the economic and social sciences. The IPCC uses the different metrics shown in Table 8.4 for different purposes. Confidence levels are intended to reflect the authors’ collective judgement on the ‘structural’ uncertainties that arise from an incomplete understanding of the processes that control particular values or results. For example, the SPM for Working Group I states (p. 3) that, ‘There is very high confidence (9 out of 10 chances) that the net effect of human activities since 1750 has been one of warming’. Likelihood estimates are intended to address ‘value’ uncertainties that arise from the incomplete determination of particular results. For example the SPM for the Synthesis Report states (p. 19) that, ‘Unmitigated climate change would, in the long run, be likely (>66% probability) to exceed the capacity of natural, managed and human

36 Nisbett et al. (1982) suggest that people do not necessarily respond to a reported consensus in itself. It becomes a problem where it leads to the disregarding of data or an alternative hypothesis. 37 Fischhoff (1982) suggests that regulatory bodies should list the reasons why they may be wrong. Along the same lines Ethan Kapstein suggested in connection with international financial regulation and well before the 2008 crisis erupted, that, ‘central bankers and financial supervisors must speak more openly about the contemporary risk environment and make clear what it is they do not know or understand’ (Kapstein 2006: 14). 38 The material in Table 8.4 and the examples in the surrounding text are taken from the IPCC 2007 Fourth Assessment Report drawing on material in the Synthesis Report (Summary for Policy Makers), the Summary for Policy Makers for Working Groups I and III and the ‘Guidance Notes for Lead Authors of the IPCC Fourth Assessment report on Addressing Uncertainties’ (dated July 2005).


15/12/2010 08:09


Table 8.4

181

IPCC use of confidence measures

(a) Confidence scale Degree of confidence* Very high High Medium Low Very low

(b) Likelihood scale Likelihood 9 out of 10 8 out of 10 5 out of 10 2 out of 10 1 out of 10

* in chances of being correct

(c) Qualitative scale Levels of agreement ▼

◄

High agreement Medium agreement Low agreement

Limited evidence Limited evidence Limited evidence

Virtually certain Very likely Likely About as likely as not Unlikely Very unlikely Exceptionally unlikely

>99% >90% >66% 33–66%

Democracy and Dissent: The Challenge of International Rule Making

Rule of Law and Democracy

Gag Rule: On the Suppression of Dissent and the Stifling of Democracy

Organised Crime and the Challenge to Democracy

Inventing Leadership: The Challenge of Democracy

Democracy and the Rule of Law (Cambridge Studies in the Theory of Democracy)

Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule

Dissent

Dissent

Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule

The Ayatollahs' Democracy: An Iranian Challenge

Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule

The Challenge of Conflict: International Law Responds (International Humanitarian Law)

The Challenge of Conflict: International Law Responds (International Humanitarian Law)

Dissent, Injustice, and the Meanings of America

Dissent, Injustice, and the Meanings of America

Making things public: atmospheres of democracy

First Democracy: The Challenge of an Ancient Idea

Making Things Public: Atmospheres of Democracy

Negotiating democracy: transitions from authoritarian rule

Dissent, Injustice, and the Meanings of America

The Rule of Nine

The Rule of Three

World of Our Making: Rules and Rule in Social Theory and International Relations (Studies in International Relations)

Jihad vs. McWorld: Terrorism's Challenge to Democracy

The rule of four

Property and the Making of the International System

The Rule Of Four

The Rule of Four

The Rule Of Four

The Rule of Four

Democracy and Dissent: The Challenge of International Rule Making

Rule of Law and Democracy

Gag Rule: On the Suppression of Dissent and the Stifling of Democracy

Organised Crime and the Challenge to Democracy

Inventing Leadership: The Challenge of Democracy

Democracy and the Rule of Law (Cambridge Studies in the Theory of Democracy)

Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule

Dissent

Dissent

Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule

The Ayatollahs' Democracy: An Iranian Challenge

Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule

The Challenge of Conflict: International Law Responds (International Humanitarian Law)

The Challenge of Conflict: International Law Responds (International Humanitarian Law)

Dissent, Injustice, and the Meanings of America

Dissent, Injustice, and the Meanings of America

Making things public: atmospheres of democracy

First Democracy: The Challenge of an Ancient Idea

Making Things Public: Atmospheres of Democracy

Negotiating democracy: transitions from authoritarian rule

Dissent, Injustice, and the Meanings of America

The Rule of Nine

The Rule of Three

World of Our Making: Rules and Rule in Social Theory and International Relations (Studies in International Relations)

Jihad vs. McWorld: Terrorism's Challenge to Democracy

The rule of four

Property and the Making of the International System

The Rule Of Four

The Rule of Four

The Rule Of Four

The Rule of Four

Recommend Documents