Globalization and the Environment: Risk Assessment and the Wto

Globalization and the Environment: Risk Assessment and the WTO Edited by David Robertson and Aynsley Kellow Edward Elgar Publishing Limited

Globalization and the Environment

Globalization and the Environment Risk Assessment and the WTO

Edited by

David Robertson John Gough Professor of International Trade, Melbourne Business School, Melbourne University, Australia and

Aynsley Kellow Professor of Government, University of Tasmania, Australia

Edward Elgar Cheltenham, UK • Northampton, MA, USA

© David Robertson and Aynsley Kellow 2001 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, photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited Glensanda House Montpellier Parade Cheltenham Glos GL50 1UA UK Edward Elgar Publishing, Inc. 136 West Street Suite 202 Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Globalization and the environment : risk assessment and the WTO / edited by David Robertson and Aynsley Kellow. p. cm. Papers presented at a two-day meeting at the Melbourne Business School in Feb. 1999. Includes bibliographical references and index. 1. Free trade—Environmental aspects—Congresses. 2. International trade—Environmental aspects—Congresses. 3. Environmental policy—Congresses. 4. Foreign trade regulation—Environmental aspects—Congresses. 5. Risk management—Congresses. 6. Globalization—Environmental aspects—Congresses. 7. World Trade Organization. I. Robertson, David, 1935– II. Kellow, Aynsley J. (Aynsley John), 1951– HF1713 .G563 2000 333.7—dc21 00–060972 ISBN 1 84064 538 5 Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall

Contents List of figures, tables and boxes List of contributors Preface and acknowledgements Glossary and trade dispute cases

vii viii ix x

1 Introduction: accounting for risk in trade agreements David Robertson

1

PART I RISK AND THE WTO 2 Risk and the WTO Gary P. Sampson 3 Risk, the environment and MEAs Neil Byron 4 Risk management experience in WTO dispute settlement Thomas Cottier 5 Applying SPS in WTO disputes Joost Pauwelyn

15 27 41 63

PART II MANAGING RISK IN POLICYMAKING 6 Risk assessment and risk management in policymaking Marion Wooldridge 7 Allowing for risk in setting standards Michael J. Nunn 8 Public perceptions, risk communication and biotechnology Christine R. Deane

81 95 106

PART III NEGOTIATING EXPERIENCE WITH RISK 9 Accounting for risk in multilateral negotiations Aynsley Kellow 10 Reducing chemical risks: OECD and lead Kaye Dal Bon

v

119 135

vi

PART IV

Globalization and the environment

NATIONAL RISKS AND QUARANTINE STANDARDS

11 National risk management and the SPS agreement David Wilson and Digby Gascoine 12 Risk and protection for grains: analysis and quarantine Donald MacLaren 13 A business view of quarantine Lyall Howard

155 169 182

PART V MANAGING BIOTECHNOLOGY 14 Government regulations and genetically-modified organisms Nancy F. Millis 15 GM foods and global trade David Robertson PART VI

191 206

CONCLUDING SUMMARY

16 Trade and risk management: exploring the issues Ian Holland and Aynsley Kellow

229

Appendix Index

249 253

Figures, tables and boxes FIGURES 6.1 The major phases in risk management 12.1 Market effects of quarantine restriction 14.1 Surveillance system operating in Australia through IOGTR and GMAC 14.2 Steps required for an antibiotic resistance gene (R-gene) in food to be transferred to a pathogen

92 174 193 202

TABLES 14.1 GMAC activities in Australia 14.2 GMOs deliberately released in Australia to 1999 14.3 Some transgenes and host plants released in Australia from 1981–97

197 197 198

BOXES 15.1 Monarchs, milkweed and GM maize 15.2 The precautionary principle 15.3 European multifunctionality

vii

210 220 222

Contributors Neil Byron, Commissioner for the Environment, Productivity Commission, Melbourne. Thomas Cottier, Professor of European and International Economic Law, University of Berne. Kaye Dal Bon, Policy Analyst based in Canberra. Christine R. Deane, Visiting Fellow at the Bureau of Rural Sciences, Department of Agriculture, Fisheries and Forestry – Australia, Canberra. Digby Gascoine, Director, Policy and International Division, AQIS, Canberra. Ian Holland, Key Centre for Ethics, Law, Justice and Governance, Griffith University. Lyall Howard, Director, Trade and Quarantine, National Farmers’ Federation, Canberra. Aynsley Kellow, Professor of Government, University of Tasmania. Donald MacLaren, Associate Professor, Department of Economics, University of Melbourne. Nancy F. Millis, Professor Emeritus, Department of Microbiology and Immunology, University of Melbourne. Michael J. Nunn, National Offices of Animal and Plant Health and Food Safety, Department of Agriculture, Fisheries and Forestry – Australia, Canberra. Joost Pauwelyn, Legal Affairs Division, World Trade Organization, Geneva. David Robertson, John Gough Professor and Director of the Centre for the Practice of International Trade, Melbourne Business School. Gary P. Sampson, Professorial Fellow at the Melbourne Business School and Research Associate at the London School of Economics. Former Director for Trade and Environment, WTO Secretariat. David Wilson, Quarantine Development Unit, AQIS, Canberra. Marion Wooldridge, Director, Department of Risk Research, Ministry of Agriculture, UK.

viii

Preface and acknowledgements After the Uruguay Round reforms, the GATT and related agreements administered by the WTO have brought increasing trade liberalization and associated gains from increased efficiency in production and consumption. At the same time, the strengthening of trade rules in the WTO has increased the scope for disputes over interpretations of more extensive and complicated agreements and raised the profile of the dispute settlement process, one of the more farsighted innovations. One of the unforeseen consequences of the comprehensive WTO agreements has been controversy over risk. New agreements on technical standards and quarantine are designed to prevent these regulations from being used as surrogate protection as tariffs and other measures are dismantled. But changing national regulations to meet international standards introduces non-economic risks which need to be assessed and managed. Several high profile, controversial trade disputes have drawn attention to the problems associated with risk. Experts invited to give opinions to dispute panels have, if anything, fuelled the controversy. Because risk and the interpretation of WTO agreements have become contentious in international trade relations, the Centre for the Practice of International Trade convened a two-day meeting at the Melbourne Business School, in February 1999, on ‘Accounting for Risk in Trade Agreements’. The papers presented at the seminar are collected in this volume, with a final paper which highlights the key points made during the discussions. As Director of the Centre for the Practice of International Trade, I express my appreciation for the financial support provided by the Australian Quarantine Inspection Service (AQIS), Environment Australia, the Grains Development Research Council (GDRC) and Monsanto Ltd. Professor Aynsley Kellow, University of Tasmania, approached me with the idea for a multidisciplinary seminar on risk. His support and contributions are greatly appreciated. Among the many people who assisted with the organization and editing of this volume my special thanks must go to Gail Wissmann who has seen this project through the difficult process of editing the papers and preparing them for publication. Thanks also go to Sylvia Jones and Christina Rust for earlier work. David Robertson February 2000 ix

Glossary and trade dispute cases ADB AFFA ALOP ALR ANZFA AQIS BSE CAP CFC CITES CTE DFAT DPIE DSB DSU

EC EU FAO FCCC FDA GATT GM GMAC GMO GTRAP GTTAC IBC IPCS IPPC IPR ISO MAI MEA

Asian Development Bank [Ministry of] Agriculture, Fisheries and Forestry – Australia Appropriate level of protection Acceptable level of risk Australian and New Zealand Food Authority Australian Quarantine and Inspection Service Bovine Spongiform Encephalopathy (Mad Cow Disease) Common Agricultural Policy (EU) Chloro-fluoro carbon Convention on International Trade in Endangered Species Committee on Trade and Environment (WTO) Department of Foreign Affairs and Trade (Australia) Department of Primary Industries and Energy (Australia) Dispute Settlement Body (WTO) Understanding on Rules and Procedures Governing the Settlement of Disputes (Annex 2 of the Marrakesh Agreement Establishing the World Trade Organization) European Commission European Union Food and Agriculture Organization (United Nations) UN Framework Convention on Climate Change United States Food and Drug Authority General Agreement on Tariffs and Trade (1947, revised 1994) Genetically modified Genetic Manipulation Advisory Committee (Australia) Genetically modified organism Gene Therapy Research Advisory Panel (Australia) Gene Technology, Technical Advisory Committee Institutional Biosafety Committee International Programme on Chemical Safety International Plant Protection Convention (FAO) Intellectual Property Rights International Standards Organization Multilateral Agreement on Investment Multilateral Environmental Agreement x

Glossary

xi

MFN MNE NGO OECD OIE PIC

Most favoured nation treatment Multinational enterprise Non-government organization Organization for Economic Cooperation and Development International Office of Epizootics Convention on Prior Informed Consent for the Importation of Hazardous Chemicals PPMs Production and processing methods SPS Sanitary and phytosanitary [quarantine] TBT Technical barriers to trade TRIPs Trade related aspects of intellectual property rights UNCED UN Conference on Environment and Development (Earth Summit) UNCLOS UN Convention on Law of the Sea UNECE UN Economic Commission for Europe UNEP United Nations Environment Programme US EPA US Environmental Protection Agency WHO World Health Organization WTO World Trade Organization WWF Worldwide Fund for Nature

OTHER TECHNICAL TERMS Short Title Appellate Body

Basel Convention

Biodiversity Convention

Biosafety Protocol

Cairns Group

Full Details Recommendations from WTO dispute panels may be taken to appeal; an Appellate Body decision then goes to the Dispute Settlement Body for a consensus decision. Basel Convention on the Control of Transboundary Movements of Hazardous Wastes and their Disposal, adopted 1989, came into force 1992 (UNEP). United Nations Framework Convention on Biological Diversity, signed at the conclusion of Earth Summit, Rio de Janiero, 1992. A protocol to the Biodiversity Convention, covering trade and movements of GMOs (a.k.a. Cartegena Protocol). Comprises major agricultural export countries (excluding EU and US), led by Australia (Argentina, Chile, Thailand, Hungary, Canada, and so on); now 18 members.

xii

Codex Convention on Desertification Miami Group

Millennium Round

Tokyo Round Uruguay Round

Vienna Convention

Globalization and the environment

FAO/WHO Codex Alimentarius Commission UN Convention to Combat Desertification, 1995. Group of ‘like-minded’ countries who regard the Biosafety Protocol as an environmental treaty and seek to minimize its effects on trade (US, Canada, Australia, Argentina, Chile, Uruguay). Refers to new WTO negotiating round, intended to commence in 2000 after the Seattle ministerial meeting. Seventh round of GATT trade negotiations (1973–79). Eighth round of GATT trade negotiations which resulted in the Marrakesh Declaration, April 1994 and established the WTO. UN Vienna Convention on the Law of Treaties, 1969, came into force 1980.

TRADE DISPUTE CASES The following are the trade dispute cases most often cited in the text. These disputes (except for tuna-dolphin) have been handled by the WTO’s Dispute Settlement Body (DSB), its nominated panels and its Appellate Body. Details of the current status of these disputes can be found in the WTO’s Annual Report or its website: http://www.wto.org/dispute/dispute.htm. Australia – Measures affecting the importation of salmon. Complaint by Canada (WT/DS18). Hereinafter cited as salmon. European Communities – Measures affecting meat and meat products (hormones). Complaint by the United States (WT/DS26) and Canada (WT/DS48). Hereinafter cited as beef-hormones. European Communities – Regime for the importation, sale and distribution of bananas. Complaints by Ecuador, Guatemala, Honduras, Mexico and the United States (WT/DS27). Hereinafter cited as bananas. Japan – Measures affecting agricultural products. Complaint by the United States (WT/DS76). Hereinafter cited as agricultural products. United States – Import prohibition of certain shrimp and shrimp products. Complaint by India, Malaysia, Pakistan and Thailand (WT/DS58). Hereinafter cited as shrimp-turtle. United States – Measures affecting imports of woven wool shirts and blouses from India (WT/DS33). Hereinafter cited as United States – shirts and blouses.

Glossary

xiii

United States – Standards for reformulated and conventional gasoline. Complaints by Venezuela (WT/DS2) and Brazil (WT/DS4). Hereinafter cited as gasoline. United States – Tuna-dolphin. Complaint by Mexico against US restrictions on imports of tuna, based on US environment regulations to protect dolphins from tuna fishing nets. This case was handled in the GATT. In February 1991, Mexico asked for a panel according to GATT article XXII (nullification and impairment). The report in September 1991 was never implemented because the US and Mexico agreed to labelling standards. This case has aggravated tensions between the WTO and environmental groups. Hereinafter cited as tuna-dolphin. Six of these cases have quarantine or environmental dimensions; bananas is an unresolved issue over trade preferences and development assistance.

1.

Introduction: accounting for risk in trade agreements David Robertson

The WTO agreements extended the GATT principles of non-discrimination and transparency into new areas of international commerce, strengthening and clarifying trade rules and extending them to cover trade in agriculture and services, establishing new agreements on trade-related investment measures and intellectual property rights, and introducing an understanding on dispute settlement procedures. Subsequently, new objectives were added to meet developing countries’ special interests, and trade and the environment. This enlargement of trade subject to international rules, however, has increased tensions among the WTO membership, which has reached 135, and generated trade disputes over interpretations of the complex new agreements. The Uruguay Round final act contains much bureaucratic language and many ‘constructive ambiguities’ to mask differences, all of which were necessary for a negotiated text to be agreed. In fact, new tensions have surrounded the rapid growth in international trade and globalization in the past decade. The new WTO agreements were incomplete and some texts contained ambiguities which increased the scope for disputes among members and placed unexpected pressures on the new dispute settlement understanding. Moreover, the new dispute settlement procedures introduced unforeseen issues, including the concept of ‘risk’ which had to be assessed and allowed for in dispute hearings. A complaint of ‘unfair’ trade protection arising from quarantine or technical regulations required an assessment of the consequences of changing them to meet international standards, which would include an analysis of risk of disease or damage. The WTO is not equipped to assess scientific risk or to establish international standards, and now disputes include arguments about social, environmental and cultural risks too. This has placed severe pressures on dispute panels and the Appellate Body, and has exacerbated differences among members over interpretations of WTO agreements. The WTO has also become a target for ‘global civil society’. Since 1995, non-government organizations (NGOs) pursuing a range of environmental and social objectives in the international arena have attacked inter-governmental 1

2

Globalization and the environment

agencies, such as the WTO and OECD, declaring they are supra-national, probusiness and anti-democratic. The resulting demonstrations in Paris, Geneva, London and Seattle since 1998 have elicited very little response from governments. Such reluctance to confront the NGOs can only encourage them and weaken international agencies. A more subtle challenge is occurring through new procedures made possible by Uruguay Round agreements intended to shore up the multilateral trading system. The dispute settlement process has developed into a quasi-legal process which is constructing a case-law approach to interpretations of WTO provisions. This creation of ‘international law’ by bureaucrats is welcomed by NGOs, many of which are led by lawyers. These new legal verdicts are exactly what the NGOs seek through environmental treaties, such as the Kyoto Protocol, Montreal Protocol, Biosafety Protocol and so on. Agreements under United Nations’ auspices are declared to supersede national laws, and are used as standards with which to batter ‘politically incorrect’ governments into line. They provide a route to influence national policies without the hindrance of having to be democratically elected to national legislatures. Though many legal scholars deny that ‘international law’ is superior to national law – indeed, deny there is any such thing as international law – the acceptance of this approach by powerful NGOs confronts governments with a dilemma; to reject these ‘treaty’ laws and oppose NGOs, or to seek compromises which reduce sovereignty. Both require more courage than most OECD governments can find. The case-law approach employed in the WTO dispute settlement process has become one focus for NGOs. Any panel report or Appellate Body decision that suits their interest is regarded as a ‘law’ by NGOs. Each dispute referred to a WTO panel interests NGOs, even if the volume of trade affected is paltry. In this process risk measurement, assessment and management are scrutinized, and if necessary challenged. If scientific evidence does not suit their interests, social or cultural risks are emphasized to achieve the desired outcome, and any inconsistent evidence rejected or ignored. Since the WTO Secretariat is not equipped to undertake risk analyses, the dispute panels have to rely on outside experts, though this still requires them to choose between alternative opinions. This is a far cry from GATT’s negotiated settlements of disputes of earlier times. The role of risk in international trade relations and in the interpretation of agreements has become a battleground for NGOs to undermine the international economic agencies, and indirectly national governments, under the guise of countering the dangers of globalization. Ironically, multilateral environmental agreements – just as much a part of globalization – are viewed positively. At the same time, the question of risk raises difficult technical problems which go beyond the political dimension.

Introduction

3

WHAT IS RISK? Risk is evident in every aspect of life and it has to be taken into account at all levels of production, consumption, investment, leisure, sport and so on. Taking risk and overcoming it brings benefits; profits and growth, personal achievement, financial returns and so on. Current attitudes (probably drawn from economic prosperity and the welfare state) seek to minimize risk and suggest that any damage from taking risks is someone else’s responsibility (the litigious society). Progress depends on prudent risk taking, and risk takers benefit from making correct judgements, because few human activities do not involve some degree of risk. To understand the role of risk in economic and social decisions requires understanding of probabilities and alternative outcomes. Rejecting risk without some idea of probabilities of an event occurring, and the costs that may result from it, is to close opportunities unnecessarily. Analysis of risk in international trade requires assessment of probabilities of different scientific and economic outcomes, followed by an assessment of the social consequences of alternatives. Only when all the evidence and evaluations are available to policymakers can decisions be made about managing risk. Evidently, this depends on perceptions of risk by the community. A risk-averse community would place low value on potential returns from a risky venture. Collective or social decisions are likely to be more risk-averse than individual decisions, and wealthy individuals are likely to be more risk-averse than poor ones. In international agreements, participating countries may agree to harmonize standards of risk. This is most apparent in regional economic agreements (EU, NAFTA and so on) but harmonization is also present in some WTO agreements (TRIPs, GATS). Any compromise is likely to create conflicts with national policies in the more risk-averse countries. This causes problems over sovereignty, which becomes a reason for escape clauses and exceptions provisions in many agreements. This volume explores aspects of risk with special reference to the WTO, where some national instruments to reduce risks may conflict with international trade rules. Quarantine regulations, technical/product standards and environmental legislation in some circumstances may conflict with trade rules and principles, and result in trade disputes. How should differences in attitudes to risk be resolved? Should international commitments override national regulations? In what circumstances might national preferences prevail? How should risk assessments be made and judged? How should differences in social perceptions of risk be taken into account? The papers in this volume were presented at a seminar held at the Melbourne Business School in February 1999. That meeting brought together practition-

4

Globalization and the environment

ers and scholars from the areas of trade and risk management, from universities and business, from national government agencies and the WTO. The chapters are based on the invited presentations and remarks by discussants who were provided with the papers in advance. Several discussants’ remarks have become chapters in their own right. The discussion proceeded under Chatham House rules, which preclude attribution of comments, but the flavour of these discussions has been distilled into the concluding chapter. The range of papers indicates the many disciples and perspectives which are appropriate to the subject of risk and trade. There is inevitably a diversity of styles and approaches, as befits contributions from international lawyers, economists, political scientists, scientists and administrators.

STRUCTURE OF THE VOLUME The papers presented at the seminar are under five headings, plus a concluding summary. I. II. III. IV. V.

Risk and the WTO Managing risk in policymaking Negotiating experience with risk National risks and quarantine standards Managing biotechnology

Part I comprises four chapters which explain how the WTO agreements deal inadequately with risk relating to trade and environmental considerations, and review the difficulties this has created in the dispute settlement process. Gary Sampson, former Director for Trade and Environment, WTO Secretariat, notes in Chapter 1 that the WTO seeks to provide flexibility for national policies without allowing disguised barriers to trade. The comprehensive scope of the WTO means that risk in all its forms has become a controversial issue when it comes to interpreting agreements. Sampson begins by explaining the narrow focus of the WTO’s mandate on the environment. Once this is understood many environmental NGOs’ complaints are put into perspective. The Committee on Trade and Environment (CTE) is concerned strictly with trade and trade effects of environmental policies, in the context of the WTO’s agreements and trade rules. Risk enters into the work of the CTE because international environmental agreements can contain provisions on trade that conflict with WTO disciplines. In particular, efforts to get the precautionary principle incorporated into these treaties (for example, EU efforts in the Kyoto Protocol and the Biosafety Protocol) raise problems if trade discrimination results. Some instruments proposed for the

Introduction

5

Kyoto Protocol would require border tax adjustments and discrimination according to production processes that would not be GATT compatible. Controversies over interpretations of GATT article XX have been at the centre of the CTE negotiations since 1995. In the recent shrimp-turtle dispute, the Appellate Body appears to have shifted this interpretation to allow for environmental considerations under some circumstances. Immediate adoption of this decision by the US delegate to the Dispute Settlement Body (DSB) brought a significant change in WTO treatment of environmental policies. WTO treatment of risk is important also in the context of the agreement on sanitary and phyto-sanitary policies (SPS) and, more broadly, the agreement on technical barriers to trade (standards) (TBT), where the adoption of international standards may produce arbitrary decisions which could affect trade but would impose economic and non-economic costs on countries. The Canadian complaint about Australian quarantine restrictions on imports of fresh, chilled and frozen salmon (salmon case) required assessments of risks arising from changes. Such cases of increasing risks may have farreaching consequences. These new Uruguay Round agreements emphasize the national treatment requirements in GATT article III. The TBT agreement was invoked in 1998 by the US and Canadian delegates against an EU proposal to regulate labelling of GM foods, which they regarded as an unnecessary technical barrier. If pursued, this would require assessment of scientific, economic, social and political risks. Each country may consider it necessary to conduct its own risk analysis. This would create complex problems for dispute panels. (The results from disputes that have been referred to panels are examined by Cottier and Pauwelyn.) Sampson concludes that the WTO function is to avoid unjustified impediments to trade. But do existing provisions give sufficient flexibility to address complex risk issues? Sampson’s chapter establishes the WTO framework for dealing with risk and identifies the relevant agreements. In Chapter 2, Neil Byron takes up the discussion as viewed by environmental lobbies. Green groups believe that the WTO impedes desirable environmental strategies and they would wish to transform the WTO into the guardian of the Earth’s environment. This confrontation is one of the most dangerous because it could end up restoring trade protection and reducing environmental protection. Byron agrees with Sampson that the WTO has no role in assessing scientific risk, especially when the environment is involved, and it should not arbitrate on acceptable levels of risk or on risk management. However, the WTO is responsible for assessing whether its trade rules are being flouted using risk as an excuse. The precautionary principle often features in such risk assessments. The European Commission (EC) has become a strong advocate of the precautionary principle in many international agreements including the recent Cartagena Protocol on Biosafety. The EC position leans towards social deter-

6

Globalization and the environment

mination of risk and political processes that play down scientific estimates of risk. (The EC has published a ‘communication’ on use of the precautionary principle, which is an appendix to this report.) Byron analyses this controversial approach and exposes ambiguities in alternative definitions of the precautionary principle that cause many problems. He concludes, like Sampson (and later, Nunn), that there is no general acceptance of this principle as international law. Byron concludes that environmental groups seem to believe that trade measures administered by the WTO offer the best hope of enforcing their goals, and this accounts for the intensity of feeling generated against WTO rules requiring consistent and transparent procedures. Chapters 4 and 5 describe experience with the dispute settlement process and the uncertainties that have arisen between panel reports and re-interpretations of evidence by the Appellate Body. Thomas Cottier has participated in several dispute panels and has wide experience in the interpretations of WTO agreements. Joost Pauwelyn is a WTO official dealing with trade disputes. Both focus on disputes brought under the SPS agreement. Cottier separates scientific from social assessments of risk, and separates risk management from risk assessment, which is consistent with decisions by the beef-hormones panel. He calls for a suitable methodology for taking account of social and economic considerations in risk management, including social and political acceptance of risk, something not recognized in the SPS agreement. The need to adopt international standards has to begin with scientific evidence, as required by the SPS agreement. National authorities could be required to examine conformity of their measures with international standards to avoid disputes. Clarification of scientific interpretations, however, would help to eliminate contradictions between panel and Appellate Body reports. Trade remedies could be tightened by introducing compensation for delays in achieving international standards. Cottier acknowledges that conflicts in society over biotechnology products bring new problems and he calls for a review of the SPS agreement to take account of experience over the past five years. Pauwelyn traces the procedures and the evolution of case law in three trade disputes conducted using the SPS agreement – EU–US beef-hormones, Japan–US agricultural products and Australia–Canada salmon. The 1999 review of the SPS agreement drawing on this experience raised some matters of concern but no changes to the text were recommended to the ministerial meeting by the DSB. The type of amendments suggested by Cottier open the whole agreement to renegotiation, which was regarded as too radical in the lead-up to the Seattle ministerial meeting. Moreover, the case-law approach allows an evolutionary process to interpretation of the SPS agreement and panel procedures. For example, the EU beef-hormones dispute established that risk assessment should be based on scientific evidence and not involve social or

Introduction

7

political values. Once a risk assessment is made, it is up to the member government to judge an acceptable level of risk and, if necessary, an appropriate non-discriminatory SPS measure (ALOP). In many instances, the Appellate Body has reversed or qualified dispute panel decisions, which is causing confusions on interpretations and weakening the case-law approach. Adoption of dispute panel reports is delayed until an appeal has been held. When a decision is finalized by the DSB, implementation of the findings is invariably delayed. Decisions on compensation or retaliation create bad feeling. Whether trade sanctions against offenders are appropriate in terms of WTO principles on non-discrimination and trade liberalization is a fundamental question. Part II on managing risk takes a broader view of risk assessment and risk management in policymaking. The three chapters in this part cover definitions and perceptions of risk, and how they affect government policies. Marion Wooldridge is a quarantine expert in the UK Ministry of Agriculture and has given expert evidence to WTO dispute panels. In Chapter 6 she regards risk management as the key requirement of policymaking, to reduce uncertainties and anxieties in the community. Later she admits that risk taking is also the driving force for progress because risk takers earn high returns, but that definition is rather broader than her specific interest, which is quarantine – the protection of human, animal and plant health. If zero risk is not possible, how much risk is acceptable? This became a problem of perception, which Christine Deane later shows is a morass of alternative sources of information and opinion, complicated by competing interest groups and disinformation. Wooldridge posits that the first question is, ‘What is the acceptable level of risk?’ This is exactly where dispute panels find themselves in the SPS cases. This ‘judgement’ will determine the safeguards necessary (ALOP). In a review of requirements of international risk management, Wooldridge assesses favourably the principles of the SPS agreement. The traditional approach to risk assessment was to gather information on hazards and the consequent risks, which could then be compared with expected benefits from an activity and alternative safeguards. Dramatic changes in dissemination of information have shifted the emphasis of this approach to concern about ‘stakeholders’’ attitudes, which complicates risk assessments and any proposed policies. Wooldridge examines the stakeholder position and risk communication in detail. The political clout of ‘stakeholders’ (especially in the EU) evidently complicates decision making, and once policymakers open their deliberations to such groups it becomes difficult to exclude any group, however extreme or irrelevant. Ultimately, all opinions cannot be satisfied. In Chapter 7 Michael Nunn provides a very thoughtful and thorough chapter on risk analysis and identifies clearly the different stages in that process; risk

8

Globalization and the environment

assessment, risk management and risk communication. He relates risk analysis to trade and technical standards. Nunn defines risk assessment as the identification of the risks or hazards, the assignment of probability of each risk and an estimation of the consequences of a risk. Quantifying risk poses many problems even when limited to scientific evidence, and difficulties of estimation have encouraged resorting to qualitative approaches in quarantine disputes. Moreover, variability in scientific risk assessment causes concern, particularly when associated with environmental problems. Nunn argues that sensitivity analysis can assist by showing that often only a few critical points significantly affect overall probabilities. However, where uncertainty occurs there are always advocates for the precautionary principle, which means a conservative approach to risk. Such approaches become especially difficult when quantitative assessments of risk consider economic and social factors as well as scientific evidence. He concludes that in future, trade disputes will require risk assessment by multidisciplinary teams. Christine Deane reviews evidence on risk perceptions, which introduces the additional uncertainties of opinion and disinformation. Deane concentrates on public perceptions of risks from biotechnology which extend beyond scientific risk to incorporate social attitudes, personal values and education. Once again, the role of ‘stakeholders’ becomes significant. By surveying public opinion polls, Deane is generally optimistic that new technology is acquiring more acceptance in OECD countries, but much scepticism still exists in these communities. Scientists and the general public have different perceptions of risk, and with social values dominating socially active groups (‘civil society’), it means risk perceptions depend more on the characteristics of the public than on risk probabilities. This is revealed in public acceptance of biotechnology for medical purposes while it is rejected in GM foods. With public opinion and ‘stakeholders’ becoming so significant in risk evaluation, the opinion formers in society become important. Surveys show that ‘experts’ (scientists, economists, bureaucrats) are not trusted. TV programmes, consumer organizations, NGOs and newspapers are thought more reliable. Consultation is regarded as one device to change public opinion in the EU, and establishing an effective independent regulatory system is also valuable (compare the US Food and Drug Authority (FDA) with food safety worries in Europe). The third element that Deane identifies as significant in influencing public perceptions of risk is education. Consultation and dissemination of information are important, but education can be more fundamental. The concept of probability is not widely understood judging by media reporting of scientific results. Most people focus on personal effects should an identified risk materialize. A probability of infection arising from a vaccination of 1:1 000 000

Introduction

9

dissuades some mothers from having a child vaccinated, but no weight is given to the 1:10 000 risk of contracting the disease without such protection. The same misunderstanding of probabilities is apparent in smokers. This suggests basic education is very important for public risk perception. Part III introduces the politics of international negotiations. Aynsley Kellow (Chapter 9) explores the politics of risk and the balancing of conflicting interests in international negotiations. Kaye Dal Bon presents a history of extended OECD negotiations on reducing risks from lead. This experience indicates that low risk perceptions can differ and result in conflicting proposals, even when all the evidence is scientific. Aynsley Kellow draws on extensive research on multilateral environmental agreements (MEAs). He divides risk analysis into four stages: hazard identification, risk estimation, risk evaluation and risk management. The first three are covered by risk assessment in Nunn’s approach and Kellow combines risk management with risk communication. But Kellow focuses on socio-political factors which Nunn incorporates into risk assessment and emphasizes in risk communication. Kellow points out that in international negotiations there is a temptation to seek unanimity (or at least agreement) in science: a fault that could be levelled at the SPS agreement. But on environmental issues, the sociopolitical forces tend to be as significant as the scientific evidence: references are made to the Kyoto Protocol and the Basel Convention. Kellow argues that attempting to keep socio-economic interests out of risk management at the international level is to invite the contamination of science by interests. Rather, the evaluative aspects of risks need to be explicitly included at the appropriate stage to enhance negotiated agreements. This would reduce the chance of measures being used as trade impediments. Dal Bon’s review of OECD negotiations on lead contamination in Chapter 10 is a valuable case study. It shows how mistaken perceptions can arise in measurement of lead levels under different circumstances. Happily European misperceptions of lead levels were corrected in these extended OECD negotiations. Other attempts to condemn chemical uses may not be properly constrained, causing potential damage from protection for false environmental reasons (for example, restricting use of dioxin in developing countries to control mosquitos may satisfy OECD environmentalists but it does not help developing countries to control disease). Part IV draws on Australian experience with the SPS agreement to evaluate interactions between national policies and international agreements. The intensity of the quarantine debate in Australia over a broad range of import embargoes makes this a review with implications for many other countries. In Chapter 11, Wilson and Gascoine (AQIS) review the contents of the SPS agreement and its relationships with other WTO agreements and international standards organizations. They go on to examine quarantine risk analysis in

10

Globalization and the environment

Australia after the Nairn review of Australian Quarantine (1996) and compare these procedures with obligations in the SPS agreement, including references to the salmon dispute proceedings. The authors conclude that the SPS agreement has been beneficial and the gain real. Ambiguities remain in the drafting, but clarifications from experience are preferable to a substantial re-negotiation of the present text. Donald MacLaren (Chapter 12) undertakes an economic analysis of the SPS provisions in the context of bulk imports of feed grains into Australia in 1994. He casts doubt on the practicality of the SPS agreement because it assumes that scientific evidence is unambiguous and risk probabilities can be estimated. MacLaren argues that even if very small probabilities of damage exist, very high costs might occur. Governments are very sensitive to these, and unforeseen contingencies cannot be allowed for. All these considerations complicate national decisions on quarantine risks, while international assessments of these risks may place different weights on the risks and costs of outcomes. Chapter 13 contains a business view of quarantine regulations. Lyall Howard acknowledges the AQIS assessment of the SPS agreement and the ambiguities and gaps that need to be clarified. Some Nairn report submissions indicated that uncertainties surround risk assessments and MacLaren’s paper has formalized these problems in a theoretical analysis. The 1994 grains case analysed by MacLaren has provided new information on risks for future national assessments. Howard notes that a harmonized approach to quarantine regulation should benefit world trade, but the three chapters in this part also show a proper balance still has to be found between trade and SPS interests. Part V introduces the controversial subject of genetically modified organisms (GMOs). Nancy Millis (Chapter 14) is Chair of Australia’s Genetic Manipulation Advisory Committee and has had a distinguished career in microbiology. While acknowledging that proper testing of GM crops is necessary, Millis points out that much genetic engineering simply accelerates normal breeding within species that has been followed for decades (even centuries). Transgenic manipulation (splicing genes from different species) needs more careful monitoring but on the whole this is still in laboratories and strictly supervised. Testing of GM crops needs to be scrutinized, but opposing all GM plants on principle is irrational, especially when it is ‘within species’ rather than transgenic – and seems ridiculous when genetic engineering for medical purposes is welcomed. Professor Millis explains the complicated procedures and committees being introduced in Australia to monitor and evaluate GMOs. The draft legislation, which has yet to be discussed in Parliament, constructs a pyramid of screening bodies and review committees to assess GMOs at the experimental stage and before release for experimental or commercial use. This is an example of regulatory mechanisms that are being prepared by most national governments.

Introduction

11

Robertson’s Chapter 15, ‘GM foods and global trade’ draws attention to the major trade issues surrounding the GM foods issue. This was expected to form part of the general discussion of biotechnology in the new trade round. Interestingly, many of the risk issues and prejudices evident in WTO disputes are also present in the GM food debate. The volume concludes with a summary of the main points raised during the seminar and some assessment of problems. Globalization of economic activity and the adoption of a system of trade rules to facilitate integration have highlighted problems of risk evaluation, assessment and management. The papers here attempt to clarify some of the issues arising in the growing controversy over risk. The final chapter considers the issues surrounding risk management as globalization and the counterforce of ‘civil society’ confront.

PART I

Risk and the WTO

2.

Risk and the WTO Gary P. Sampson*

Risk assessment is the scientific determination of the relationship between cause and effect in situations where adverse effects may occur. While the process of identifying, analysing and assessing risk is the province of experts, risk management – the determination of an acceptable level of risk and the appropriate level of protection (ALOP) – involves elements of political judgement. In many instances, the importance that public opinion assigns to a specific risk appears to bear little relation to science-based risk assessments. For example, cigarette smoking poses a greater scientific risk to health than eating hormone-treated meat, yet in many European countries, the first is tolerated and the second is banned. How risks are managed can provide opportunities for arbitrary and unjustified restrictions on trade and commerce. The prevention of unwarranted discriminatory barriers to trade is the raison d’être for WTO rules and disciplines. The present challenge facing WTO members is implementing its component agreements to provide flexibility for domestic policies without creating disguised barriers to trade. Balancing these conflicting interests can lead to controversy and disputes in international trade. Moreover, public perceptions of risk and their management differ between WTO member countries. The purpose of this chapter is to address the manner in which risk assessment, risk management and public perceptions of risk bear on the work of the WTO. This depends on how the problem presents itself and how the WTO agreements deal with their trade dimension. The intention is not to be prescriptive, nor to cover all WTO agreements where risk is relevant. The aim is to identify key areas in the WTO where risk is important and to examine relationships between risk management and trade relations.

TRADE AND THE ENVIRONMENT The WTO Committee on Trade and Environment (CTE) was established by a decision of the Marrakesh ministerial meeting (April 1994). The CTE was given *

Professor Sampson is on leave from the WTO Secretariat in Geneva. Until January 1999, he was Director, Trade and Environment Division. This paper is part of a larger study on the Role of the WTO in Global Governance financed by the Ford Foundation.

15

16

Risk and the WTO

a broad mandate to identify the relationships between trade and environmental measures and to make recommendations on whether any modifications are required to permit a positive interaction between them. The CTE presented a comprehensive report to the 1996 WTO ministerial meeting in Singapore which established the trade and environment issues on the WTO agenda through the continuation of the Committee. While the CTE does not specifically address matters relating to risk management, it is a recurring theme. For example, the acceptability of using trade sanctions as a policy tool to deal with risks to the environment is dealt with under various headings. They may take the form of the acceptability of Multilateral Environment Agreements (MEAs) that provide measures to deal with environmental problems that are not consistent with WTO agreements (for example banning trade in endangered species and hazardous wastes). Another example is discussion of the conditions under which exceptions can be sought to employ measures that protect against risk to the environment which would otherwise breach WTO obligations. Two specific parameters guide the CTE’s work: (i)

the competence of the WTO is limited to trade and aspects of environmental policies which may result in significant trade effects; (ii) policy coordination to promote sustainable development must uphold and safeguard the principles of the multilateral trading system. In other words, there is no intention that the WTO should become an environmental agency, nor that it should review national environmental priorities, set environmental standards or develop global policies on the environment. These responsibilities belong to national governments and inter-governmental organizations designed for the task. The CTE work programme has focused on rules and disciplines of the multilateral trading system and trade-related aspects of environmental issues, including trade measures that may be employed according to MEAs. WTO provisions on non-discrimination, most-favoured-nation and national treatment (GATT articles I and III), and transparency (GATT article X) are relevant to any use of trade measures for environmental purposes. Beyond that, the exceptions clauses contained in GATT article XX allow members to give priority to public health and safety, conserving natural resources and national environmental goals ahead of its WTO obligations, as long as the measures used do not go beyond what is necessary, in terms of trade restrictions or discrimination.

Risk and the WTO

17

WTO RULES AND THE ENVIRONMENT An important challenge is to deal with the interface between trade policy and environmental policy designed to reduce risks to the environment. Measures to discriminate among products in international trade because of the way they are produced is prohibited through the interpretation of ‘like products’ in GATT article III on national treatment. For environmentalists, the life-cycle approach differentiates between products precisely on the grounds that they are produced differently. Different production methods constitute grounds for discriminating between products in environmental regulations. Measures taken to reduce greenhouse gas emissions as part of the Framework Convention on Climate Change (FCCC, 1997) provide one example of this conflict, which is not helped by the uncertainty surrounding the science of ‘climate change’. There are differing scientific findings relating to the risk of greenhouse gas emissions heating the upper atmosphere, and different national priorities over management of this risk. The FCCC is intended to establish institutional and economic mechanisms to provide negotiated reductions in global emissions of greenhouse gases. The FCCC addresses trade effects: ‘Measures taken to combat climate change, including unilateral ones, should not constitute a means of arbitrary or unjustifiable discrimination or a disguised restriction on international trade.’ It does not provide for any specific trade-related environmental measures. The Kyoto Protocol to the FCCC requires that the parties shall strive to implement policies and measures ‘in such a way as to minimize adverse effects ... on international trade’ (FCCC article 2:3). After the Kyoto agreement establishing emission reduction targets, the next step is to determine how emissions should be reduced. Several measures are already in use; including taxes on carbon content, mandatory or ‘voluntary’ emission standards, subsidies for environmentally friendly production processes, labelling and certification schemes and markets in emission permits. The Kyoto Protocol itself contains mechanisms for the fulfilment of commitments; Annex I countries are allowed to act within ‘bubbles’ to achieve their emission reduction commitments (that is, share responsibilities to achieve commitments). All these measures to reduce emissions will affect international trade, through costs of production and competitiveness on world markets. Individual producers who lose in this process will demand compensation for reduced competitiveness arising from different emission standards or taxation. Such measures will raise complex issues about WTO consistency; for example, border tax adjustments. Enforcement of environmental standards by discrimination between imported products according to production processes contradicts fundamental WTO principles. It has been argued also that preferential trading of emission entitlements between countries within ‘pollution’ bubbles may be subject to WTO roles relating to regional trading arrangements (GATT article

18

Risk and the WTO

XXIV). Any inconsistencies between the rights and obligations of the WTO and conditions in future global environmental treaties may lead to new disputes (Goldberg, 1995; Robertson, 1995). These conflicts over measures to reduce greenhouse emissions are complicated further by the assessment and management of risk. The FCCC gives formal recognition to the precautionary principle: Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation, taking into account that policies and measures to deal with climate change should be cost-effective so as to ensure global benefits at the lowest possible cost.

Effective consideration depends on the risk of damage and the uncertainty of pre-emptive actions. Disagreements about the risk that greenhouse gas emissions will alter climatic conditions will influence any assessment of risk. The WTO does not inhibit governments from taking action to minimize risks associated with the production and consumption within national boundaries, as long as foreign goods are treated no less favourably than domestic goods inside the country. They can be prohibited, regulated or subject to charges, according to the conditions of GATT article III. High levels of carbon dioxide emissions and energy-use may be penalized or managed by a government to reduce the risk of climate warming. The WTO permits governments to levy taxes according to method of production, as long as an imported product or service is not discriminated against according to the production process such as its energy intensity. An importing country should not discriminate in trade. An import cannot be treated differently from a ‘like product’ (that is, a product with the same physical characteristics) produced domestically (GATT article III). To be WTO consistent, products that have the same physical form have to be treated as ‘like products’, irrespective of how they have been produced abroad (for example, even if produced in an environmentally unfriendly manner). Despite the GATT/WTO principles of non-discrimination and transparency, there are provisions for ‘exceptions’ from WTO obligations. Under specific circumstances, WTO members can invoke measures that normally would constitute a breach of WTO obligations. The circumstances for exceptions are set out in GATT article XX: General Exceptions, which includes measures necessary to protect human, animal or plant life or health or to conserve natural resources. Measures taken under GATT article XX can be effective only within the boundaries of the implementing country. In the case of greenhouse gas emissions, the production process in an exporting country is outside the jurisdiction of the importing country. If such trade action is taken in the context of

Risk and the WTO

19

a legal instrument addressing reductions in greenhouse gas emissions, it would be inconsistent with the necessity test of GATT article XX(b) (if taken against a WTO member that is not a signatory of that environmental agreement). According to GATT article XX, any measure which breaches WTO obligations should not be used as a means of arbitrary or unjustifiable discrimination between countries, or a disguised restriction on international trade. Further, the measures must be necessary to protect the environment; that is, other means are not available. In the case of measures to restrict trade in products that emit too much carbon dioxide, they could be consistent with a multilateral environmental agreement (MEA). While the WTO does not have the protection of the environment as a primary aim, this commitment has been acknowledged in the preamble to the agreement establishing the WTO – members agreed to ‘protect and preserve the environment and to enhance the means for doing so’. In November 1998 in the US shrimp-turtle dispute, the report of the Appellate Body drew on the preamble of the WTO agreement to assert ‘we have not decided that the sovereign nations that are members of the WTO cannot adopt effective measures to protect endangered species, such as sea turtles. Clearly, they can and should.’ Furthermore, WTO members are free to adopt their own policies aimed at protecting the environment as long as, in doing so, they fulfil their obligations and respect the rights of other members under the WTO agreement. This interpretation regards principles expressed in the preamble as general legal commitments, rather than specific obligations of states. In making its ruling, the Appellate Body clearly assigned importance to promoting sustainable development and preserving the environment, something that appears only in the WTO preamble. In this case, however, there was agreement between all the parties to the dispute that turtles are an endangered species – that is, there is a serious risk of their extinction. What was not agreed to and which led to the sanctions by the United States is how the risk should be managed. WTO members recognize that the organization has no special expertise to deal with the risk assessment or management of environmental problems such as greenhouse gases. Nor is it equipped to make judgements on the most appropriate means to reduce greenhouse gas emissions. MEAs are the best instruments to coordinate policies to deal with global and trans-boundary environmental problems. Members of the WTO are, however, sensitive that trade measures taken using MEAs can contradict WTO rights and obligations. Only about 20 MEAs contain trade provisions: some, such as the FCCC and the Kyoto Protocol, could have important commercial and political consequences. The CTE has examined suggestions that the WTO should formally recognize how some measures designed to deal with risk and available within MEAs may be inconsistent with WTO provisions. If special treatment is given to any specific MEA, provisions would have to be made for WTO countries not

20

Risk and the WTO

wishing to join the MEA; a country may be unpersuaded by the scientific evidence, it may not have access to the necessary technology, or it may not agree with the objectives of the MEA. Society may simply not assign the same priority to the management of the risk in question.

ENVIRONMENT: THE PRECAUTIONARY PRINCIPLE The Rio Declaration (UNCED, 1992) defines the precautionary principle in similar fashion to the FCCC (p. 20). Some OECD governments and the European Commission (EC), in particular, regard the precautionary principle as accepted into customary international law. This suits environmental groups and their advocates (Sands, 1995). The Appellate Body of the WTO’s Dispute Settlement Board, however, has not accepted the precautionary principle as customary law (US–EU dispute reports on hormone-treated beef (beef-hormones)). The legal underpinnings of the precautionary principle are elusive and difficult to define. Precaution is a ‘culturally framed concept ... muddled in policy advice and subject to the whims of international diplomacy and the unpredictable public mood over the true cost of sustainable living’ (O’Riordan and Cameron, 1994). Because the precautionary principle is so flexible and can adapt to circumstances, it remains legally controversial. Precautionary measures require some scientific evidence before they are used and evidence that inaction would lead to irreversible damage or demonstrate harm to the environment (O’Riordan and Cameron, 1994). The precautionary principle depends on determining when the definition shifts the burden of proof towards protection of the environment, or heath and safety. This threshold can be high, as in ‘serious or irreversible harm to the environment’, or lower, as in ‘may cause harm to the environment’. The precautionary principle has been incorporated into some international environmental agreements.1 The treaty establishing the European Community incorporates it into European law, which was significant in the EU–US WTO dispute over beef-fed growth hormones (beef-hormones). The WTO panel and the Appellate Body ruled against the EU employing trade restrictions on hormone-treated beef because of insufficient scientific evidence (Pauwelyn, Chapter 5, this volume). The acceptance of the precautionary principle into the text of the UN Biodiversity Convention and the Biosafety Protocol is also controversial. The Biodiversity Convention states ‘where there is a threat of significant reduction or loss of biological diversity, lack of full scientific certainty should not be used as a reason for postponing measures to avoid or minimize such a threat’ (UNEP, 1994). This could become an important source

Risk and the WTO

21

of restrictions on trade in products using genetically modified organisms if it is extended to the bio-safety protocol. Notwithstanding the importance that some ascribe to the precautionary principle, there is no international consensus to accept it or agree on any one definition.2 This lack of consensus has already caused controversy in WTO disputes, which will intensify if the precautionary principle is inscribed into international agreements, such as the FCCC and the Biodiversity Protocol. The status of the precautionary principle in international treaties has immediate relevance to risk. If a precautionary approach can be invoked, the analysis and measurement of risk becomes less relevant because any risk becomes sufficient to justify preventative action, including trade protection. It is the role of the WTO to distinguish between legitimate trade protectionist measures and those that contravene WTO rules.

WTO PRODUCT STANDARDS AND REGULATIONS The core concept of non-discrimination as contained in the GATT 1994, and the interpretation of ‘like products’ and ‘non-product related production processes’ are criticized (and misunderstood) by some environmental groups. But if there is a legitimate basis to discriminate between ‘like products’, the WTO members have argued in the CTE that the grounds should be established by setting standards in an MEA. The agreement on technical barriers to trade (TBT agreement) addresses matters relating to risk assessment and management through assigning the responsibility to experts. The TBT agreement encourages the use of international standards developed by international standards organizations with the necessary technical competence. There is a fine line between the right of governments to adopt the standards they consider appropriate, and the use of standards that have trade protection as their intent. Internationally recognized standards are assumed to constitute acceptable non-trade measures. So the TBT agreement defers to specialized organizations to establish standards. According to the TBT agreement, the difference between a standard and a technical regulation rests on compliance and enforcement. While conformity with standards is voluntary, technical regulations are mandatory. If an imported product does not fulfil the requirements of a technical regulation, it will not be tradable. In the case of standards, non-complying imports will be allowed but they may not be acceptable to consumers.3 Most technical regulations and standards adopted are to protect human safety or health. Five principles govern the preparation, adoption and application of technical regulations, standards and conformity procedures in the TBT agreement:

22

Risk and the WTO

(a) members must respect the fundamental principles of non-discrimination and national treatment in their use of technical regulations, standards and conformity procedures; no preference may be granted to national producers in charging fees for testing, inspection, and so on; (b) members shall ensure that technical regulations, standards and conformity assessment procedures are not prepared, adapted or applied to create unnecessary obstacles to trade; (c) members, whenever appropriate, should use relevant standards or conformity assessment guidelines provided by international standards organizations; (d) members should, where possible, recognize foreign technical regulations as equivalent to their own, and consult with other authorities to seek mutual recognition of standards; (e) transparency should be encouraged through notification processes and the establishment of national information centres. Technical barriers to trade result from the adoption and application of different regulations and conformity assessment procedures. Differences may exist for legitimate reasons such as different tastes, income levels, geographical and climatic resources. The TBT agreement acknowledges legitimate divergences and provides for flexibility in the preparation, adoption and application of national technical regulations. The preamble to the TBT agreement notes ‘no country should be prevented from taking measures necessary to ensure the quality of its exports, or for the protection of human, animal, and plant life or health, of the environment, or for the prevention of deceptive practices, at the levels it considers appropriate’. TBT article 2.2 admonishes that technical regulations should not be prepared, adopted or applied with a view to, or with the effect of, creating unnecessary obstacles to trade. It should choose the approach which has the least restrictive impact on trade. In consequence, international standards, guidelines or recommendations should be adopted to avoid unnecessary obstacles to trade. Unnecessary obstacles to trade may result from regulations that are more restrictive than necessary to achieve a given policy objective. Risks of over regulation arise from inadequate consideration of available technical and scientific information, technology or product uses. TBT article 2.2 lists objectives that are legitimate, including national security requirements, prevention of deceptive practices, protection of human health or safety, protection of animal and plant life or health, and the environment. The TBT agreement embraces the GATT’s non-discrimination and national treatment obligations. TBT article 2.1 states that ‘in respect of their technical regulations, products imported from the territory of any member be accorded treatment no less favourable than that accorded to like products of national

Risk and the WTO

23

origin and to like products originating in any other country’. A similar provision applies to standardization. The TBT agreement encourages members to adopt international standards unless their use would be ineffective or inappropriate; for example, because of climatic or geographical conditions or fundamental technological problems (TBT article 2.4). Risk assessment is becoming relevant to the TBT. At the September 1998 TBT committee meeting, the United States and Canada expressed concern over EU proposals to regulate the labelling of products containing genetically modified soya or maize. They argued that such labels were unnecessary technical barriers to trade since no scientific reason existed to differentiate foodstuffs produced from genetically modified crops. This dispute has now become a fundamental issue of biotechnology that transcends several WTO agreements (Robertson, Chapter 15, this volume).

SANITARY AND PHYTOSANITARY MEASURES Before the Uruguay Round, quarantine measures to protect human, animal or plant life and health were allowed under GATT article XX(b). The complicated effects on trade, however, required more precise rules and the Agreement on the Application of Sanitary and Phytosanitary Measures (SPS) was negotiated in the Uruguay Round (Wilson and Gascoine, Chapter 11, this volume). The SPS agreement recognizes members’ sovereign rights to apply quarantine measures which may restrict trade in order to implement national laws protecting human, animal or plant life or health. In accordance with the nondiscrimination principle, such measures should apply to domestically produced food or local animal and plant diseases, as well as to products coming from other countries, and without discriminating among foreign sources of supply. Sanitary and phytosanitary measures should not become disguised restrictions to trade. The main purpose of the SPS agreement is to reduce arbitrariness of decisions by clarifying which factors should be taken into account when imposing quarantine measures. Measures to ensure food safety and animal and plant health should be based on the analysis and assessment of objective and accurate scientific data. This has proved harder to supervise than expected (Pauwelyn, Chapter 5, this volume). International standard-setting organizations offer ready-made yardsticks for WTO members, and most countries participate in these standards committees. The SPS agreement explicitly refers to three standard-setting organizations; the FAO/WHO Codex Alimentarius Commission, the International Office of Epizootics (OIE), and the international and regional organizations operating within the framework of the FAO International Plant Protection Convention

24

Risk and the WTO

(IPPC). These organizations provide expert advice to WTO dispute settlement panels. Before the WTO was established, however, international standards, guidelines, recommendations and so on, were adopted by governments on a voluntary basis. Since 1 January 1995, these norms have achieved new status according to the SPS agreement, although their use is not compulsory. One criticism of the SPS agreement is the adoption of standards that were not originally legally binding, but have acquired that status as part of the legally binding SPS agreement.4 When adopting these norms, a WTO member is presumed to be in full compliance with the SPS agreement. If a country decides to establish standards that exceed existing codex, OIE or IPPC standards, it requires scientific evidence to support the higher standards. A harmonized standard of risk management would be facilitated if all WTO members adopted international standards, but they do not exist for all goods and services produced and traded. Moreover, different circumstances (climate, income levels and so on) may require different standards. Each country needs to conduct its own risk assessment and determine its acceptable level of risk. The exporting country has to demonstrate that its domestic quarantine and technical standards are at least as good as those required by the importing country. Even so, under present agreements, WTO members should be guided by scientific principles when deciding their appropriate level of protection. The SPS agreement has generated some significant trade disputes and revealed that scientific evidence does not always provide clear-cut conclusions.5 Hence disputes over EU import bans on beef treated with growth hormones (beef-hormones), Australia’s quarantine barriers against imported fresh salmon (salmon), and Japan’s quarantine testing of imported fruit (agricultural products) have resulted in intricate legal battles in the WTO. (These three SPS disputes are examined in detail in the following chapters.) These cases are evolving new legal interpretations of Uruguay Round agreements that will influence applications of multilateral trade rules in future. A vital element in the scientific evidence concerns the risk associated with changes in the SPS measures, and what level of risk is acceptable. What is the risk a disease or pest might be introduced if standards are lowered? And what might be the economic/social costs of such an invasion? How might risk be assessed and then managed?

A NEW DIMENSION The principal goal of the WTO is to avoid unjustified or discriminatory barriers to trade. As new agreements to deal with non-tariff measures have been established, new difficulties of implementation have come to prominence. In

Risk and the WTO

25

particular, ‘risk’ has become contentious in the context of the environment, quarantine standards, technical standards and new products (such as those derived from genetically modified organisms). Member countries evaluate the risks differently, have different perceptions of risk and how it can be managed. The critical question is what constitutes an appropriate response to a particular risk? WTO concern is that protection from risk should not be disguised trade protection. How risks are managed can lead to abuse and existing GATT articles and other WTO agreements contain provisions to prevent such abuse, but can they cope with complex new multi-dimensional questions and conflicts between environmental and trade objectives, scientific and socio-economic goals, national and international commitments? The WTO must rule on the legitimacy of trade measures. But do the existing WTO provisions provide sufficient flexibility to address issues of an economic, social, cultural and environmental nature?

NOTES 1. This includes the preamble of the 1985 Vienna Convention, the Montreal Protocol, the Framework Convention on Climate Change, the London Dumping Convention, as well as the Rio Declaration of 1992. The Convention on Biological Diversity does not mention the principle by name, but defines its properties within the agreement. 2. This consideration is formally recognized in principle 7 of the Rio Declaration, which states that there is a ‘common but differentiated responsibility’ of states in resolving environmental problems of a global nature. 3. For example an energy-intensive refrigerator, that does not qualify for a national eco-label, may be admitted for sale, but environmentally sensitive consumers might simply abstain from purchasing it. 4. Requirements that motor vehicles carry reflective devices to prevent road accidents; labelling requirements for cigarettes, drugs and alcoholic beverages; electric insulation requirements; car emission standards, or limits on the use of certain dying and tanning materials. 5. The codex standard for minimum residue levels for growth promoting hormones were adopted by a vote of 33–29, with seven abstentions. The SPS agreement requires that the EU implement a full risk assessment for hormones if the minimum levels are agreed to by less than half of the codex experts.

REFERENCES Goldberg, D.M. (1995), ‘The Framework Convention on Climate Change’, in R. Houseman et al. (eds), The Use of Trade Measures in Select Multilateral Environment Agreements, UNEP. O’Riordan, T. and J. Cameron (1994), ‘The History and Contemporary Significance of the Precautionary Principle’, Interpreting the Precautionary Principle, London: Earthscan.

26

Risk and the WTO

Robertson, D. (1995), ‘Environmental Constraints on Development and International Rules’, in R. Garnaut, E. Grill and J. Reidel (eds), Sustaining Export Oriented Development, Cambridge: Cambridge University Press. Sands, Philippe (ed.) (1995), Principles of International Environmental Law, Manchester: Manchester University Press. UNCED (1992), Rio Declaration on Environment and Development; principle 15. UNEP (1994), Convention on Biodiversity; Preamble, UNEP/CBD/04/1, Geneva. WTO (1997), Taxes and Charges for Environmental Purposes: Border Tax Adjustment, WT/CTE/W/47, May, Geneva.

3.

Risk, the environment and MEAs Neil Byron

It is increasingly popular to paint a demonized image of the WTO as the ‘World Trans-Science Organization, a global meta-regulator’. It is allegedly resolving scientific issues such as carcinogenicity, decreeing what are acceptable levels of risk or scientific uncertainty, and making decisions about appropriate levels of public health and safety. In doing so, it stands accused of overriding the role and authority of national governments to protect their people and their lands. It might seem like exaggerated rhetoric, except for the fact that there are some who believe it, and these critics – or opponents – have some powerful allies, particularly amongst international environmental NGOs and on the fringes of secretariats to the MEAs. This suggests that the trade–environment debate suffers from a serious communication failure or worse. Some environmentalists apparently want to transform the WTO into the supreme guardian of the Earth’s environment. Their goals and standards for the international environmental agenda would come from MEAs. Those goals and standards would be incorporated into the WTO, making trade restrictions and bans the instrument of choice for environmental policy. The discussions and protests over genetically modified foods indicate the control consumers in developed countries would like to acquire over certain production processes. Sampson’s chapter takes us through a discussion of risk and rationality, noting the irrationality, or at least inconsistency, of many government decisions based on perceived public demands or opinions. The case is built up from general trade–environment linkages and the existing (and prospective) role and the various interpretations of the precautionary principle. This lays the basis for a discussion of the WTO’s current rules and processes regarding the environment. These in turn are then extended into the specific areas where risk assessment and response are important – namely technical product standards and regulations, and quarantine standards. The chapter articulates a consistent thread throughout the whole body of practice – consistency, transparency, the rule of principle over expediency, and the principles of national treatment and most favoured nation treatment. Thus Chapter 2 traces the logical development of the practice of risk assessment to its current state, but one missing element in the story is the 27

28

Risk and the WTO

author’s assessment of how processes might evolve from here, and what forces might shape that. Clearly, there is no role for the WTO in scientific risk assessment of environmental (or public health and safety) hazards and risks – the scientific determination of the relationship between cause and effect in situations where adverse effects can occur. Nor does it (or should it) attempt to arbitrate on what constitutes an acceptable level of risk, or the most appropriate risk management strategy. However, the WTO does have the responsibility and authority to rule on whether risk is being used as an ‘unjustified and discriminatory barrier to trade’. Differentiating between a (valid) reason and an excuse can be difficult, and this is precisely where the WTO stands. At the extremes of the spectrum of possible scenarios, it is easy to distinguish black from white, but in the middle are many shades of grey – plausible excuses for protectionism, and apparently strong cases for bending the rules. It seems that in many cases, the protagonists in the trade–environment debate do not have opposite (black/white) objectives, but rather have conflicting priorities. To caricature the conflicting objectives: (a) Pursuing gains from trade liberalization; environmental problems can and should be dealt with only in other fora. (b) Pursuing gains from trade liberalization; subject to a constraint of no serious environmental harm and no direct incompatibility with the pursuit of MEA objectives. (c) Pursuing environmental gains; recognizing a constraint of not damaging trade too much. (d) Pursuing environmental gains, at any price, whenever possible. People in the first and last of these groups have very little common understanding as a basis for communication. Most international discussions seem to be at the interface of (b) and (c) above. Different organizations quite legitimately have different focuses and responsibilities. In cases of conflict in policy coordination and priorities, the WTO’s position could hardly be anything but to uphold and safeguard the principles of the multilateral trading system, that is more like (b) above. The MEAs are, quite rightly, concerned solely with achieving environmental gains but, in practice, they have demonstrated awareness of the consequences of distorting world trade in the process of achieving their aims. Oxley (1999) has argued that a major reason why environmental organizations want to use the WTO’s strong and effective apparatus is the inherent weakness or fuzziness of enforcement measures within the MEAs. This question will be dealt with in the final section, asking whether trade measures even work in pursuit of MEA objectives.

Risk, the environment and MEAs

29

THE PRECAUTIONARY PRINCIPLE The precautionary principle is an increasingly important concept in environmental disputes. According to the Rio Declaration on Environment and Development:1 in order to protect the environment, the precautionary approach shall be widely applied by states according to their capabilities. Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation. Note that it does not say anything like the following interpretation which is commonly used or implied in many environmental circles: where there are threats of serious or irreversible damage (either domestically or in any other member state), all member states must take all possible measures to prevent environmental degradation. The Rio version of the precautionary principle does not demand action, it merely negates one possible reason for inaction. Other reasons for postponing measures may be acceptable; the measures should be cost-effective, and so on. However, the new EC definition (EC, 1998) is quite different from the Rio one: ‘The precautionary principle is an approach to risk management that is applied in circumstances of scientific uncertainty, reflecting the need to take action in the face of potentially serious risks, without awaiting the results of scientific research.’ (For the latest 2000 version see appendix to this volume.) This seems to authorize and legitimize governments taking action even without due cause. It reverses the onus and the direction of action. Nevertheless, the more positive aspect of the EC paper is that it does require that: • the implementation of an approach based on the precautionary principle should start with an objective risk assessment identifying at each stage the degree of scientific uncertainty; • measures based on the precautionary principle must be proportionate to the risk which is to be limited or eliminated; • measures based on the precautionary principle must include a cost/benefit assessment with an eye to reducing the risk to a level that is acceptable to all the stakeholders; • measures based on the precautionary principle must establish responsibility as to who must furnish the scientific proof needed for a full risk assessment; • measures based on the precautionary principle must always be of a provisional nature, pending the results of scientific research performed to furnish the missing data and performance of a more objective risk assessment.

30

Risk and the WTO

This is consistent with the interpretation used later in this volume by Wooldridge ‘when the manager believes on the basis of an initial examination of the evidence, that there is a risk of serious (and possibly irreversible) consequences contingent upon a given hazard, steps should be put in place to minimize that risk whilst a more thorough risk analysis is performed’. But this is quite different from either the Rio or the SPS articulation of the precautionary principle. Clearly the concept continues to evolve, along with elaborations of the conditions under which it may or may not be applied, but much of the controversy may well be due to the ambiguity of alternative definitions in current usage. As Sampson notes ‘The legal underpinnings of the precautionary principle are elusive and difficult to define’. How rigorous is this concept, or does the precautionary principle have so much flexibility in application that it could be little more than ‘smoke and mirrors’? How much weight in decision-making can be given to such a nebulous and qualified concept? Note the differences between the relatively strong wording of the precautionary principle in the Rio Declaration and the Convention on Biological Diversity and the much more moderate articulation, and requirements for evidential support in the SPS (Sampson, Chapter 2, this volume). To speculate, why is the onus of proof reversed – perhaps because the primary objective in the former was environmental and the drafters and signatories were principally not economists concerned with trade impacts? Sampson (p. 20) takes us through a discussion of whether the precautionary principle is now established as a principle of customary international law.2 Principles as expressed in preambles are general legal commitments. In making its shrimp-turtle ruling, the Appellate Body clearly assigned importance to promoting sustainable development and preserving the environment, something that appears only in preambular language. It also found that living, as well as non-living, exhaustible resources are embraced under article XX(g). Nevertheless, the USA applied a measure that was legitimate under XX(g) – in a manner that constituted ‘arbitrary and unjustifiable discrimination between countries’. The second order question is, if the precautionary principle is indeed now part of customary international law, how is it to be applied in practice? That is where Sampson focuses on the threshold level for the burden of proof regarding protection of the environment, or public health and safety (p. 20). This threshold can be high, such as will definitely cause ‘serious or irreversible harm to the environment’, or very low and general, as ‘may harm the environment’. Sampson points out that the European Commission has embraced, and attempted further to clarify and codify, the precautionary principle to make it more accepted as customary international law. The EC apparently believes that it can be used as an argument to ban foods (apparently in the absence of scientific evidence) that the public perceive to constitute a health risk. The

Risk, the environment and MEAs

31

United States still denies that the principle is customary law, while Canada takes the middle ground. One wonders what the current positions are on this matter in countries like China and India? (They have given little evidence of much concern over precautions except as the principle has been applied to impede their exports.) In short, there is still no consensus on accepting the precautionary principle as a basis for establishing obligations in national and international rules.

DISCRIMINATION ON THE BASIS OF PROCESS The two areas of the WTO in which exemptions could be made on environmental grounds (article XX paras (b) and (g)) concern product standards and regulations, and sanitary and phytosanitary standards. Each of these is now the subject of separate agreements, which elaborate and seek to clarify the more contentious trade–environment linkages. The fundamental trade–environment dispute revolves around the central tenet of the WTO (GATT article III) that ‘measures to discriminate against products in international trade because of the manner in which they are produced’ are not permitted. Yet this is precisely what environmentalists want to achieve – the ability to differentiate between, and selectively target, those products, processes and producers they consider dirty, or perhaps ‘unclean’, in many different senses. The greenhouse issue provides an excellent example, where a country’s management response to a widely perceived (but poorly defined) risk has potentially profound trade impacts. The Framework Convention on Climate Change (FCCC) does not provide for any specific trade-related environmental measures only that ‘Measures taken to combat climate change, including unilateral ones, should not constitute a means of arbitrary or unjustifiable discrimination or a disguised restriction on international trade.’ The Kyoto Protocol to the FCCC article 2:3 states that parties shall strive to implement policies and measures ‘in such a way to minimize adverse effects on international trade’. While the FCCC thus gives formal recognition of the precautionary principle, it also recognizes the potential pitfalls of pseudoprotection measures being introduced in the guise of measures to address the greenhouse/climate change concerns. The WTO does not inhibit governments from protecting against risks associated with the production and consumption of products produced within national boundaries. As Sampson notes, these can be prohibited, regulated, taxed or charged for any purpose thought appropriate by national governments. Similarly, the WTO has no problems with a government levying taxes according

32

Risk and the WTO

to how a product is produced within its territory, for example, domestic taxes on production methods. The important point is, however, that the manner in which a foreign product is produced is not a basis on which WTO rights and obligations are established. An imported product cannot be discriminated against only because the production process was energy intensive, for example. As Sampson argues, the underlying thesis is that should any country wish to influence the manner in which products are produced in other countries – however appropriate this may be thought by the importing country – this should not be translated into discriminatory trade measures.3 For example an imported product, produced by a process that emits more greenhouse gas than the importing country considers acceptable, cannot be treated differently from a ‘like product’ (that is, a product with the same physical characteristics) produced domestically or in another country, solely because of the production process. The exceptions clauses contained in GATT article XX do allow a WTO member legitimately to place its public health and safety and national environmental goals ahead of its WTO commitments. But this is subject to not going beyond what is necessary, in terms of trade restriction or discrimination, to achieve those policy goals. Exceptions can be sought providing the measure is necessary to protect human, animal or plant life or health (article XX(b)) or for measures related to the conservation of exhaustible natural resources (article XX(g)).4 This general exception is clearly designed to permit governments to maintain or implement laws they feel are necessary to preserve the lives of people, flora, fauna and exhaustible natural resources. Pressure for the use of trade sanctions comes from various sources. One is the rising environmental awareness and mobilization of consumers in developed countries. With varying degrees of success and geographical coverage, campaigns have been driven by large international non-governmental organizations on different real or perceived environmental threats: perhaps mostly against logging, but also against disposal of oil platforms in the sea, against the use of fur, in favour of cute animals, against hydro-electric dams, against oil exploration, against pesticides and fertilizers, against genetically modified food. Radical environmentalists tend to be mono-thematic and single issue campaigners and, accordingly, operate (consciously or not) with very partial analyses. Past experience in the GATT/WTO has been that the measures taken to protect fauna and flora applied within the boundaries of the country taking the action. This raises the question of whose flora, fauna, exhaustible resources, or public heath and safety is at risk? For example, could Australia discriminate against Polish cars because the steel manufacturing process there threatens Swedish lakes and fish with acid rain? Or French cars because they use energy produced by nuclear power stations that some believe constitute a health and

Risk, the environment and MEAs

33

safety risk to the French public? Now these are deliberate exaggerations, stretching the point, but they aim to illustrate the ‘slippery slope’ of discrimination on the basis of production process. Such differentiation really challenges the definition of ‘like product’ (GATT article III), especially when the manner of production cannot be scientifically detected in the finished product. Try opening two cans of tuna, with their labels removed, and see if you can tell which one is ‘dolphin-friendly’. Sampson also explains how, if a trade action is taken in accordance with an MEA, then the necessity test of article XX(b) comes into play. Snape (1999) in his comments on the shrimp-turtle case, notes that the word ‘necessary’ appears in article XX(b) and has been strictly interpreted in the past, but XX(g) does not contain the word ‘necessary’ in dealing with the exhaustion of natural resources (which, in the Appellate Board’s report, includes the extinction of species). The question of a special ‘window’ or provision for the MEAs has been raised – in effect a cross-linking of international instruments, where membership of an MEA would confer some special rights/obligations in the WTO, and vice versa. Sampson makes a strong case against that. Furthermore, it is not inconceivable that ‘strategic’ manoeuvres have been in play; for example that countries have (reluctantly) accepted trade provisions in the MEAs confident that the WTO/GATT would preclude them from ever being actually exercised or, in other cases, might provide an ‘escape clause’. The WTO has no special expertise to deal with environmental problems such as greenhouse gas emissions, and nor is it well placed to judge the most appropriate means to achieve targets such as reducing those emissions or conserving biodiversity. A consensus has emerged that MEAs are the best way of coordinating policy action to tackle global and trans-boundary environmental problems cooperatively. But as Oxley (1999) has argued, they tend to be quite ineffective in doing so. Rather than modifying WTO rules and procedures, through blanket exceptions, approval for selective sanctions and so on, those pursuing environmental objectives would be better advised to address the underlying causes of the environmental problems where they arise.5 Thus the core WTO concepts of non-discrimination and transparency as contained in the GATT since 1947, and the important interpretation of ‘like products’ and non-product related production processes, have evolved to the current position as responses to risk of environmental hazard or public health and safety. The logical conclusion is that if there is a legitimate basis to discriminate between like products, then the grounds should be established by those with the technical expertize to do so through setting standards in an MEA. This comes under the TBT agreement which encourages the use of international standards developed by international standardizing bodies technically competent to deal with such matters.

34

Risk and the WTO

However, there is a fine line between the right of governments to adopt the standards they consider appropriate, and the use of standards that have trade protection as their intent. The difference here is that internationally recognized product standards are rebuttably presumed not to constitute unnecessary barriers to trade.6 TBT article 2.2 lists policy objectives that may be deemed ‘legitimate’ including national security requirements; prevention of deceptive practices; protection of human health or safety; protection of animal and plant life or health, or the environment. Most technical regulations and standards adopted by countries are aimed at protecting human safety or health.7 Sampson summarizes the five principles governing the preparation, adoption and application of technical regulations, standards and conformity assessment procedures in the TBT agreement. The present or potential impact of trade clauses in an increasing number of MEAs warrants more study. Those which discriminate between countries are not compatible with WTO rules, and at least until the Appellate Body’s shrimpturtle decision, neither was discrimination between products on the basis of process. Those seeking the protection of the environment in a country other than the one imposing the measure, constitute an ‘extraterritorial’ reach that is also incompatible with free trade principles. Strictly, the gasoline dispute was not an environmental case. The dispute was not over the environmental goals pursued by the country which discriminated against imports and there was no discussion on whether those goals were being reached in the best way, or even whether they were being reached at all. The dispute was over the use of trade as an instrument to reach that goal. According to the panel report, WTO members are free to set their own environmental objectives, but trade discrimination is not an accepted instrument to pursue them. Nobody examined whether the original trade ban had any impact on cleaning the air of the cities in the country (USA) applying the ban.

SANITARY AND PHYTOSANITARY (SPS) MEASURES The final section of Sampson’s chapter deals with the more stringent measures for the protection of human, animal or plant health – clearly an area where very different perceptions of risks can legitimately exist, and lead to quite different responses, which in turn may have substantive trade impacts. Prior to the Uruguay Round, such protection could only be taken under GATT article XX(b). The new agreement on SPS measures applies to measures to protect against additives, contaminants, toxins or disease-causing organisms in foods as well as the spread of disease by pests or animals or plants.

Risk, the environment and MEAs

35

Like TBT, any sanitary and phytosanitary measures should not be disguised restrictions to trade. That is, they must accord with the national treatment and MFN principles. Again the main purpose of this agreement is to reduce the possible arbitrariness of governments’ decisions in this field by clarifying which factors should be taken into account when imposing health protection measures. In particular, measures should be based on the analysis and assessment of objective and accurate scientific data. Like the TBT, the SPS agreement encourages consistent and transparent decision making in setting up the appropriate level of protection; that potentially trade-restrictive measures be applied for no other purpose than ensuring food safety and animal and plant health, and not result in unjustified barriers to trade. A country may decide to establish protection levels that exceed international standards if it believes that the latter do not meet its acceptable level of protection. But subject to article 5.7 this decision, if challenged, must be supported by scientific evidence resulting from an objective assessment of the potential health risks involved. When introducing a more trade restrictive standard, members must be able to demonstrate that any deviation from existing standards (codex, OIE or IPPC) is based on sound scientific criteria. Countries must establish SPS measures on the basis of an evaluation of the actual risks involved. On the basis of the risk assessment exercise, governments must determine their ‘acceptable level of risk’ which commonly includes substantial safety margins as a precautionary measure. The exporting country has the burden of demonstrating that its domestic sanitary requirements are at least as good as those of the importing country in that they achieve the same level of health protection. Overall, WTO members should be guided by scientific principles when deciding their appropriate levels of protection, so that those decisions are not arbitrary or protectionist. The accepted level of risk should be a reflection of health protection, not a means to protect domestic producers from competition. Where the risks and the health consequences are similar, the accepted level of protection should be similar. The SPS agreement does not preclude countries taking precautionary measures in an emergency even when sufficient scientific evidence does not yet exist to support definitive measures. The BSE scare in 1996 was a good example. In the absence of sufficient scientific evidence, emergency bans were immediately introduced on a provisional basis. In the long run, governments must seek the additional information needed to carry out a more objective assessment of the risks involved, and adapt their measures accordingly. (Again, note the contrast with the Rio wording of the PP, and the move towards convergence with the EC 1998 definition discussed above.) The key point from the beef-hormones case was that although the chemical treatments posed no scientifically discernible risk to public health in the

36

Risk and the WTO

European Union, there was a strongly held public perception that the risk was real and serious. The dispute between the EC and North America over genetically modified organisms (and products which may contain them) centres on a standard which has the effect of impeding trade, justified by the precautionary principle, in spite of no scientific evidence of risk to public health and no evidence of the magnitude of the hazard. The onus thus seems to fall on the EC to provide a scientific risk assessment, and to demonstrate that its response is necessary, consistent and appropriate.

DO TRADE MEASURES WORK TO PURSUE THE MEAs’ GOAL OF ENVIRONMENTAL PROTECTION? What some economists seem to find particularly irritating about the trade-ban clauses of the MEAs, is that they seem to rely on an incomplete or ‘half-baked’ diagnosis of the problem. For example, since tropical deforestation is due primarily to conversion of forests to various forms of agriculture, total bans on the imports of tropical timber will not prevent tropical deforestation. If the trees cannot be sold they are burnt, in order to get at the land beneath. A (CITES) ban on trade in turtle-shell handicrafts will not stop people killing turtles for food, destroying their habitat, eating their eggs and so on. While turtle excluder devices may save as many as 700 turtles per year in North Queensland fisheries, is this a major accomplishment while just across the border 7000 per year are killed in the eastern islands of Indonesia, and this is not being addressed by anyone! Perhaps in addition to a ‘necessity’ test, there should also be a rationality test or an efficacy test? To sum up Hoffmann’s (1998) argument, it is not evident that trade restrictions are the best, or even a workable, instrument for environmental purposes. The Montreal Protocol is a good example of an MEA explicitly using traderestricting measures. Roughly, worldwide production of ozone-depleting substances fell by the mid 1990s to one-fifth of its 1988 peak (Van Vuuren and Bakkes, 1997, p. 27). There is no direct answer to the question of whether trade bans were crucial in meeting the Montreal Protocol’s phase-out requirements. Most analysts agree that the main factors contributing to the ‘success’ were: • the availability of technical alternatives for substitution of CFCs at costs that, albeit somewhat higher, were acceptable; • clear scientific evidence of the damage to the ozone layer, which was easy to publicize (for example, images of the Antarctic ozone hole using satellite photography); its potential health impact, skin cancer, was

Risk, the environment and MEAs

37

considered serious by the public, and almost became a general scare, as indicated by the worldwide rise in the sales of sun-blocking creams; • the Montreal Protocol comprises well-defined implementation schemes and evaluation procedures, as well as flexible implementation, such as the possibility for countries to meet their production and consumption limits jointly; • industrialized countries provided technical and financial assistance through the global enviornment fund and bilateral financial and technical assistance, supporting a change of consumption and production patterns away from ozone-depleting substances (Van Vuuren and Bakkes, 1997, pp. 27, 75–6). In addition, once developed countries banned the production of CFCs and alternative technology was available (even though somewhat more expensive), a powerful set of exporting interests was created that functioned as a pressure group in favour of those alternative technologies – the infamous ‘Baptists and Bootleggers Coalition’ (Yandle, 1998). Thus, international business interests probably magnified the environmental outcome sought, but the mechanism bears little relation to trade bans. There were other incentives to join, but it is probably fair to conclude like Brack (1996, p. 57) that ‘the combination of financial assistance [after 1990] and technology transfer plus the fear of trade discrimination created a powerful incentive amongst developing countries’. Once a given environmental measure is in place in one country, those industries or groups forced to apply it at home have a vested interest in pushing their international competitors along the same path. This is evident in the implementation of the Montreal Protocol, in the role played by large chemical concerns that invested heavily in CFC substituting technology. At a different level, it is also evident in the gains to domestic competitors from the dolphin/tuna ban imposed by the USA, or the more recent shrimp-turtle ban. For the time being, enforcement and dispute settlement mechanisms in MEAs remain extremely weak (Oxley, 1999). Voluntary commitments and exchange of information are emphasized, as well as domestic policies, such as the approval of ‘national strategies’ in response to international environmental agreements. In fact, trade measures are about the only coercive instrument in MEAs (von Moltke, 1996). That explains to some extent the enthusiasm of some environmentalists for ‘greening the WTO’ and the illusion that the dispute settlements mechanism of the WTO could be transformed to enforce environmental goals. This is a dangerous proposition and not just because it could reinforce protectionist trends in developed countries under the guise of environmental prudence. To mix the international environmental regime and the international trade regime, as they stand now, could only create an extremely disorderly and conflict-prone institution for global negotiations. The growing

38

Risk and the WTO

number and variety of multilateral environmental agreements in existence, and their rapid evolution through new protocols and amendments, is creating a web of ever more complex international environmental regimes.

CONCLUSIONS There is a consistent thread throughout the whole body of WTO practice – consistency, transparency, the rule of principle over expediency, and the principles of national treatment and MFN. This has been built up over 50 years in the GATT and the WTO, but has yet to consolidate for the MEAs (most of which are less than 10 years old). There is no shortage of contentious topics where legitimate differences of opinion exist firstly in the assessment of the risk and secondly in the appropriate response to that assessment. Environmental health For example: decision on reprocessing of spent nuclear fuel rods in Germany; motor vehicle emission standards; greenhouse gas emissions; the next generation of tuna-dolphin and shrimp-turtle cases; sustainably managed forests; excessive use of surface water and land degradation. Public health and safety For example: agricultural chemicals; antibiotics or growth hormones in meat; GMO foods not just maize and soya; pharmaceuticals; aircraft noise standards; certification and safety standards of motor vehicles and aircraft. Extra-territoriality For example: human rights, occupational health and safety concerns; and more generally under the ‘social justice’ rubric. In principle the WTO now has a process, a mechanism to deal systematically and consistently with such a plethora of prospective challenges. Again this reveals the advantage of a rules-based system – having established a set of principles and procedures, consistency and transparency should prevail over ad hoc and arbitrary discretion. There is a temptation for certain groups to wish to retain the ability to throw up their own protective shield in response to a national risk assessment (which of course could be perfectly fair and fully justifiable!) but at the same time to want to prevent others from using their shields to (unfairly) exclude exports. Small, open trading countries, however, are better served by consistent and rigorously enforced rules which prevent the arbitrary use of market power by large economies. As stated in the introduction, the essential feature of the system is the ability to differentiate, fairly and transparently, between a legitimate

Risk, the environment and MEAs

39

response to strongly held fears of catastrophic risks, and attempts to use exaggerated risk assessments and responses as a barrier to trade. Environmentalists are pressing urgently to address land degradation and conservation issues in countries like Australia and are willing to use any instrument that might assist in the battle. The ability to pressure competitors into accepting the same environmental standards that Australia has (or that they would like Australia to adopt) increases the political palatability of these measures. If that means using trade measures (boycotts, bans), then so be it! No doubt US fishermen were pleased when US trade pressures forced Mexican fisherman to meet the same dolphin-saving standards that were required of US tuna fisheries. It greatly complicates the arguments to assert ‘We have to force you to improve your environmental protection, to enable us to improve our standards, without suffering too much competition’. People arguing thus, do not realize how much it challenges the rule-based system, and how much the rules protect everyone, rather than constrain actions. Industries in rich countries sometimes find themselves in natural alliance with environmentalists to exclude what they claim to be ‘environmentally inferior products’ from the market (Cairncross, 1995). Some environmental groups in developed countries are campaigning for the introduction of tariffs and taxes to raise the price of imports that do not conform to certain standards. Unilateral definition of criteria to classify environmentally superior or inferior products will bring more trade conflicts. But will it bring improved environmental outcomes (Hoffmann, 1998)?

NOTES 1. United Nations (1992), Report of United Nations Conference on Environment and Development, Annex 1, Rio Declaration on Environment and Development, Principle 15. Rio de Janeiro, 3–14 June 1992. 2. In the preamble to the WTO, member countries agreed to ‘protect and preserve the environment and to enhance the means for doing so’. In its late-1998 ruling in a dispute concerning US trade measures against countries that do not employ devices to protect sea turtles from drowning in shrimp nets, the report of the Appellate Body drew on the preambular language of the agreement establishing the WTO to assert that ‘we have not decided that the sovereign nations that are members of the WTO cannot adopt effective measures to protect endangered species, such as sea turtles. Clearly, they can and should.’ The Appellate Body held further that ‘WTO members are free to adopt their own policies aimed at protecting the environment as long as, in doing so, they fulfil their obligations and respect the rights of other Members under the WTO agreement’. 3. Prison labour is a noteworthy exception, which may be used as a precedent in future. 4. There are other grey areas at the margin, for example discriminating between hand-made and machine-made goods (that is, on the basis of capital intensity) or against products made by sweatshop labour (on the basis of human rights and labour obligations) even though there may be no discernible difference in the products, even at the level of microscopic or chemical analysis. 5. Note that CITES is now to be reviewed in the light of evidence that it has been ineffective, even counter-productive, in such cases as banning the ivory trade in order to conserve African elephants.

40

Risk and the WTO

6. Unnecessary obstacles to trade may result from a regulation that is more restrictive than necessary to achieve a given policy objective, for example, because the objective can be achieved through alternative measures which have less trade-restricting effects, or the necessary effects of the regulation on trade are disproportionate to the risks the non-fulfilment of the objective would create. 7. Technical regulations are inherently mandatory – an imported product, which does not fulfil the requirements of a technical regulation, will not be allowed for sale. Conformity with standards is voluntary; non-complying imports will be allowed on the market, although they may be shunned by consumers.

REFERENCES Brack, Duncan (1996), International Trade and the Montreal Protocol, London: The Royal Institute of International Affairs and Earthscan Publications. Cairncross, Frances (1995), Green, Inc. A Guide to Business and the Environment, London: Earthscan. European Commission (1998), Guidelines for the Application of the Precautionary Principle, Brussels: Directorate General XXIV (Consumer Policy and Consumer Health Protection). Hoffmann, Helga (1998), Trade as Instrument of Environmental Policy: Does it Work? Presentation to the First World Congress of Environmental Economists, Venice, Italy, 25–7 June 1998. Oxley, Alan (1999), ‘MEAs and the WTO – the Real Trade and Environment Issue’, in Industry Competitiveness, Trade and the Environment, Melbourne: Productivity Commission. Sampson, Gary P. (1999), Trade, Environment and the WTO: A Framework for Moving Forward, ODC Policy Paper, Washington, DC: Overseas Development Council. Snape, Richard (1999), ‘Some Implications of the Shrimp-Turtle Decision (AB1998–4)’, in Industry Competitiveness, Trade and the Environment, Melbourne: Productivity Commission. Van Vuuren, D.P. and J.A. Bakkes (1997), ‘Agenda 21 Interim Balance, 1997’, Netherlands Institute of Public Health and the Environment, Bilthoven, The Netherlands, October (RIVM Report no. 402001008). von Moltke, Konrad (1996), International Environmental Management, Trade Regimes and Sustainability, Winnipeg: International Institute for Sustainable Development. Yandle, B. (1998), Bootleggers, Baptists and Global Warming, PERC Policy Series PS14 Bozeman MT, Political Economy Research Center.

4.

Risk management experience in WTO dispute settlement Thomas Cottier*

INTRODUCTION The 1995 WTO Agreement on the Application of Sanitary and Phytosanitary Measures (SPS), negotiated during the Uruguay Round, subjects the adoption and application of national or regional measures relating to the protection of human, animal and plant health to the WTO dispute settlement system (WTO, 1995; Barcel, 1994; Yukyan Shin, 1998). Until 1995, such measures were excluded from international legal scrutiny, as both GATT articles XI and XX(b) and the ‘standards’ agreement did not offer adequate guidance and had not assisted in solving longstanding issues. One unresolved case was the EC ban on the importation of hormone-treated beef from the United States. The persistence of this dispute motivated negotiations on a specific and separate instrument. Since coming into force, the SPS agreement has profoundly affected the international law on food security (WHO, 1998; FAO, 1998). During the Uruguay Round negotiations, work on the SPS agreement did not receive much publicity, despite the looming dispute over hormone-treated beef. The SPS negotiations were overshadowed by major topics, such as liberalization of agricultural trade, the elaboration of multilateral disciplines for services (GATS) and the establishment of a multilateral framework for intellectual property (TRIPs). Once adopted, however, the SPS agreement quickly moved to centre stage. Within four years, it had been tested in three major cases before dispute panels: the expected complaints of the United States and Canada against the EC ban on the importation of hormone-treated beef,1 the complaint by Canada against the Australian ban on the importation of fresh, chilled and frozen salmon2 and the complaint by the United States against Japan on measures affecting agricultural products.3 All these cases were appealed. By mid 1999, Appellate Body reports on beef-hormones,4 salmon5 and quarantine on agricultural products,6 and three arbitration awards relating to implementation had been produced.7 *

The author was a member of the Swiss negotiating team during the Uruguay Round and a member and chairman of several GATT and WTO panels, including EC – Measures concerning Meat and Meat Products (Hormones) and related arbitration.

41

42

Risk and the WTO

These disputes were complex legal undertakings. The panels and the Appellate Body were confronted, for the first time, with adjudicating politically sensitive and legally difficult matters under the new agreement, and required to assess extensive scientific evidence for the first time in the history of GATT/WTO. In terms of implementation of results, some are likely to remain controversial because they represent fundamental differences in attitudes to technology. The experience has been, and still is, a process of trial and error in an essentially new domain of WTO law. This is true for the cases and the case law. The SPS agreement itself will require amendment and improvement during the coming Millennium Round, in the tradition of gradually improving and developing the side agreements of the GATT after the Tokyo Round. The main feature of the SPS agreement consists in an approach which is essentially based on scientific evidence and assessment. Members are obliged under SPS articles 2.2, 3.3 and 5.1 and 5.2 to base their SPS measures on scientific evidence, with the exception of temporary precautionary measures under SPS article 5.7. The cases adjudicated so far show that science plays a predominant role. Yet, there are other societal, ethical and political factors at stake with which the present framework has difficulty coping in a systematic and organized manner. Indeed, transAtlantic tensions caused by the beefhormones dispute result less from scientific disagreement than from diverging societal and ethical perceptions in the public at large (see Deane, Chapter 8, this volume). A better framework to deal with such concerns and to differentiate them from economic protectionism needs to be developed. In future work, it will be necessary to provide adequate room and an appropriate methodology for assessing such arguments and concerns in dispute settlement. This is important not only for dealing with traditional issues of quarantine regulations, but more significantly in adjudicating on matters relating to genetically modified organisms in the new age of biotechnology (see Millis, Chapter 14, this volume; Robertson, Chapter 15, this volume). This chapter will consider a number of fundamental systemic issues arising from recent disputes and assess their implications for risk assessment and risk management in the WTO. It will offer some recommendations for further work in the concluding section. The focus is the relationship between the scope of national or regional autonomy in standard setting and administration of standards, and the requirements of the SPS agreement as they emerge from the dispute panels and Appellate Body reports.

THE CORE FUNCTION OF THE SPS AGREEMENT Enactment, monitoring and enforcement of appropriate food and health standards is a core function of government and has long been an important part

Risk management experience in WTO dispute settlement

43

of policing domestic agriculture, industry and imports, despite restricting effects on trade and market access for foreign agricultural products. This is uncontested and clearly reaffirmed by the preamble and article 2.1 of the SPS agreement. States have a right to take sanitary and phytosanitary measures necessary for the protection of human or plant life or health within their own jurisdictions. In practice, however, such measures frequently and equally provide protection for domestic markets (James and Anderson, 1998). With the process of liberalizing international trade in agricultural products and foodstuffs, such traditional powers are confronted with enhanced market access requirements. Sanitary and phytosanitary measures became increasingly apparent as other barriers are gradually reduced. It might also be that there will be a temptation to compensate for the reduction in traditional trade barriers by new and sophisticated SPS barriers in the future, relating in particular to GMOs. Given these risks on the one hand, and the legitimate interests protected by SPS measures on the other hand, law and policy in food and health standards has become a matter of balancing equally legitimate interests: health protection, market access and consumer interests. Essentially, it is a matter of detecting measures which are economically protectionist, rent-seeking interests and exercise more restrictions on trade than is necessary to achieve the goals of health protection. The SPS agreement, therefore, is at the centre of what is a constitutional approach to WTO rules, and their relationship to national or regional sovereignty and rule making (Jackson, 1999). The problem of national or regional autonomy in standard setting appears at several levels in the SPS agreement. Of course, the very existence of GATT (1994) and the SPS agreement entails a restriction on uninhibited national or regional standard setting. Yet, this still leaves open the substantive question of the proper relationship between existing international standards and recommendations to scientific evidence and the burden of proof and the impact of societal judgements. Procedurally, it is a matter of defining the appropriate level of investigation and research to be undertaken in order to satisfy the needs of the international system. What is the proper degree of control which the international system should exercise beyond traditional examination of necessity and proportionality under general exceptions (GATT article XX(b)). How have dispute panels and the Appellate Body dealt with this question under the new agreement? The record is mixed, and the proper balance still has to be found. The Appellate Body first offered in beef-hormones substantially more deference to the member country than the panel did. Subsequently, it showed somewhat stricter views than the dispute panel in complying with risk assessment requirements in the salmon case. In agricultural products, there is an essentially balanced relationship between the panel’s finding and the assessment by the Appellate Body. It essentially affirmed the panel’s reasoning and findings, and emphasized the

44

Risk and the WTO

procedural obligations under the SPS agreement. The following sections address some of these questions. It should be stated at the outset, however, that this chapter falls short of dealing with all the puzzling complexities of the cases involved. It is far from a complete analysis.

THE RELATIONSHIP OF INTERNATIONAL AND NATIONAL OR REGIONAL STANDARDS: THE BURDEN OF PROOF The scope of autonomous standard setting is firmly defined by the reading of SPS article 3 on harmonization. While SPS article 2.1 affirms the right of member states to enact measures, article 3.1 essentially states that they shall base such measures on international standards, guidelines and recommendations, where such instruments exist. According to article 3.2, these international standards are deemed to be necessary and prevail over market access rights, as they are presumed to be in accordance with the pertinent provisions of the SPS agreement and the GATT (1994).8 Measures in conformity with international standards of risk assessment undertaken by specialized international fora are likely to pose less potential for conflict with trading rules. It will be difficult to rebut the presumption of the need for such multilateral standards, just as it has been impossible to rebut the presumptions of nullification and impairment of unlawful measures in the context of GATT article XXIII.1(a).9 It is interesting to observe that the policy and approach expressed in SPS article 3.1 is comparable to the one adopted by the Appellate Body with respect to multilateral environmental agreements (MEAs). The Appellate Body indicated in shrimp-turtle that trade restrictions emanating from MEA agreements are to be respected by the WTO.10 Unlike most unilateral measures adopted by members and having transnational effects, they are very likely to prevail in dispute settlement. The situation may be different in the context of bilateral or regionally agreed health standards, to the extent it can be shown that they result in excessively high levels of protection. However, this would seem to be an exceptional constellation as SPS article 4 encourages the adoption of policies and agreements of equivalence and mutual recognition in bilateral relations. Overall, the SPS agreement clearly seeks to provide an incentive to use and to refer to multilateral standardization.11 It goes without saying that under the international or global approach, the autonomy of national or regional standard setting is inherently limited as both agreement and common ground need to be found in an appropriate institution, such as the FAO/WHO Codex Alimentarius Commission, or the International Office of Epizootics (OIE). The system therefore offers an escape route. SPS

Risk management experience in WTO dispute settlement

45

article 3.3 leaves members the option to maintain or adopt their own and higher standards under a number of conditions. This option, of course, is necessary where common and international standards are still lacking. But it allows governments to ignore international standards whenever they are deemed to be inadequate. Taking into account the compelling wording of SPS article 3.1, the panel on beef-hormones construed this provision to establish an exception to a commitment, in principle to use and rely upon lawfully adopted international standards in the first place (beef-hormones, paras 8.49, 8.89). The Appellate Body, on the other hand, strongly rejected this approach and ruled that the option to establish own levels of sanitary protection in article 3.3 ‘is an autonomous right and not an “exception” from a “general obligation” under article 3.1’.12 Conceptually, the change was important. It established and emphasized the equivalence of national standards under the SPS agreement. It barred international standards from achieving farreaching effects and from becoming binding on the WTO members by way of the SPS agreement (beef-hormones, para.165). It expressed the principle of in dubio mitius, that is, that in doubt an interpretation is adopted which puts a lesser burden on national sovereignty. The judicial policy of deference most clearly appears at this juncture of the report and perhaps was felt necessary because of the cries of anguish expressed upon publication of the dispute panel’s report. The shift was welcomed in particular by European commentators who found the ruling much too intrusive and detrimental to the WTO (Eggers, 1998; Hilf and Eggers, 1997). Subsequently, the view and perception of an autonomous right to set national standards was no longer contentious in salmon (Panel Report, note 2, para. 4.15) and in agricultural products (Panel Report, note 3, para. 8.13). The equivalence of international and national standards is settled in case law. Politically, the drive and incentive to work towards international standards – unlike in the field of environmental protection – has been lost. In a dispute settlement, the conceptual shift from an exception to a right and obligation on its own is of procedural importance. The prevailing interpretation shifts the burden of proof from the defending to the complaining party. While the beef-hormones panel report concluded that it is up to the defending party to demonstrate that all requirements of SPS article 3.3 (and thus of article 5) are met, it is now up to the complaining party to establish prima facie evidence first, not only with respect to the violation of SPS article 3.1, but also with respect to independent national measures adopted under article 3.3 of that agreement (beef-hormones, paras 104–9). In accordance with standards set out by the Appellate Body in United States – shirts and blouses,13 the complaining party bears the initial burden to demonstrate in law and facts that the defending party is in violation of its

46

Risk and the WTO

obligations. It is only upon the establishment of such prima facie evidence that the burden then shifts to the defendant to justify the measures under review. The Appellate Body stated the law in beef-hormones as follows: The initial burden lies on the complaining party, which must establish a prima facie case of inconsistency with a particular provision of the SPS agreement on the part of the defending party, or more precisely, of its SPS measure or measures complained about. When that prima facie case is made, the burden of proof moves to the defending party, which must in turn counter or refute the claimed inconsistency.14

In this first case, it was not quite clear what the conceptual shift really meant in practical terms. The Appellate Body refrained from setting down detailed requirements to be met by the complainant in order to pass the test of providing prima facie evidence. Since the Appellate Body is limited to reviewing legal issues, it was not in a position, nor did it find it necessary, to undertake an additional and detailed examination of evidence put forward by the claimants. Instead, it essentially relied upon the findings of the panel in that respect. It would seem from this case, as well as from salmon, that the prima facie evidence test is not too difficult to pass. In agricultural products, the Appellate Body reversed the findings of the dispute panel relating to the alternative, and less restrictive methods of testing (‘determination of absorption levels’) under SPS article 5.6. The panel deduced its finding from evidence before it while the approach was not specifically argued by the United States as a complainant (agricultural products, paras 125, 126). The Appellate Body held: Pursuant to the rules on burden of proof set out above, we consider that it was for the United States to establish a prima facie case that there is an alternative measure that meets all the three elements under article 5.6 in order to establish a prima facie case of inconsistency with article 5.6. Since the United States did not even claim before the panel that the ‘determination of sorption levels’ is an alternative measure which meets the three elements under article 5.6, we are of the opinion that the United States did not establish a prima facie case that the ‘determination of sorption levels’ is an alternative measure within the meaning of article 5.6.

From this ruling, two conclusions may be drawn. First, the prima facie evidence test can only be met if specific arguments to the point are explicitly made by the complainant. Secondly, panels are barred from stating in accordance with the principle of iura novit curia, short of specific arguments raised by the complainant. They cannot, in other words, find the law on their own initiative and draw from evidence generally before the panel on a particular issue if the complainant has not established a prima facie case.15 The ruling entails a limitation to a broad statement found in beef-hormones. Here, the Appellate Body held that a panel, within the legal claims made, is free to adopt

Risk management experience in WTO dispute settlement

47

its own arguments and legal reasoning to support its own findings and conclusions on the matter in dispute.16 It follows from agricultural products that such liberty only exists in as much as the complaining party has provided sufficient arguments to sustain a prima facie case. Finally, it still is an open question whether the complaining party needs to make a compelling case on the possibility of less restrictive measures. Is it sufficient to raise the argument, or is demonstration of viable alternatives within the legal system of the other member required? In practical terms, it would still seem that the important conceptual shift in the relationship of SPS article 3.1 and article 3.3 has had limited practical effect. Whatever the precise standards for prima facie evidence, it is clear that even under SPS article 3.3, the main burden in defending national measures and demonstrating that they have met all the conditions of that article 3.3 and other pertinent provisions of the SPS agreement remains with the defending party. The thrust of debate and discussion in the panel proceedings and reports thus relate to the arguments in defense of the measure. There is no evidence that the Appellate Body, based upon the shift, pursues a clear policy of deference visà-vis the defendant. Interestingly, the policies of deference inspired by beef-hormones and followed by the next panel on salmon were reversed. The dispute panel carefully sought to avoid findings on risk assessment where they were not absolutely necessary. The Appellate Body, on the other hand, found that the panel should have made such findings and stated itself that Australia had not based its measures on risk assessment. On this basis, the Appellate Body reversed the panel’s presumptions (salmon, para. 128). Discussions held at the Melbourne conference on the problem of deference and burden of proof inherent in the interpretation of SPS article 3, as it is currently framed, revealed disagreements on interpretation. It showed conceptual shortcomings and dilemmas in the present agreement. My view on the subject is that in beef-hormones there is not sufficient doubt as to the reading of article 3.1 and article 3.3. While article 3.1 requires members to apply international standards where appropriate recommendations exist, a member may introduce or maintain measures based upon a higher level of protection. It would seem from the wording and the context of SPS article 3 that the drafters intended to reinforce international standard setting with a view to preventing difficult trade disputes. At least, it would have been necessary to look into the travaux préparatoires, as required by article 32 of the Vienna Convention in cases where the interpretation of the text is considered to be contextually unclear. It is submitted that the relationship of international and national standards should be reviewed and clarified. Which is the best regulatory approach to bring about a reasonable balance between SPS and market access rights? What are the long-term political incentives which the system should provide in order to avoid severe trade disputes while at the same time continuing to protect life

48

Risk and the WTO

and health? These considerations, on the one hand, would tend to speak in favour of shaping SPS article 3.1 as a mandatory requirement and SPS article 3.3 as an exception to it. On the other hand, the current SPS article 5 and annex A leave members with full rights to determine the appropriate level of sanitary or phytosanitary protection (ALOP) or the acceptable level of risk (ALR). The existence of such rights, recognizing divergent societal values implies a right to set national standards, independently of existing international recommendations. This allows for levels of ALOP or ALR which are different from international standards. The Appellate Body argues conversely. Finally, one further explanation for the Appellate Body’s approach may be found in an implied argument that international bodies are not yet fully prepared to absorb the function of enacting in effect binding instruments in the field. This is of importance from the point of view of legitimacy and democracy, and should be further discussed. A presumption to apply international recommendations and base measures on these presupposes a high level of legitimacy of such standards. Legitimacy again depends on participation and representation of those elaborating such recommendations. In conclusion, bearing in mind the lessons from case law, the relationship of SPS articles 3.1 and 3.3 and SPS article 5 requires further thought and consideration in coming negotiations with a view to achieving a fully consistent regime. It is submitted that incentives to develop international standards should be created. Efforts need to go hand-in-hand with improving rule making and procedures in appropriate standard-setting bodies and institutions outside the WTO. This is yet another example of how institution building may transgress the scope of one particular international organization. One way to bring this about would be to establish procedural requirements in terms of transparency, participation, representation and rights for other organizations to have to comply with, such as the Codex Alimentarius (FAO/WHO), as a prerequisite for accepting global, and common recommendations as mandatory rules under SPS article 3.1. Discussions should also include whether more nuanced differentiations need to be made between binding standards and recommendations adopted in specialized fora and in dealing with those within the SPS agreement. A further and more fundamental change would be to shift the burden of showing compliance with international standards to the exporting producer, and not the importer. In dispute settlement, it would be up to the exporting country to demonstrate that all the relevant international standards had been complied with. Such compliance would establish a presumption of market access rights. The importing country could rebut such a presumption by demonstrating that the level of protection provided by international standards is inadequate to meet the national level of appropriate protection (ALOP) or the level of risk deemed acceptable (ALR).

Risk management experience in WTO dispute settlement

49

STANDARDS OF REVIEW A second and important aspect of the relationship between national and international systems is standards of review. What exactly should a panel or the Appellate Body examine and look into, and to what extent? Clearly, it is not enough for these bodies to make a de novo assessment, that is, to define proprio motu whether or not the measure is necessary to protect public health. Rather, it is currently a matter of assessing whether (i) the complaining party provides adequate prima facie evidence that the measures are inconsistent with the SPS agreement and, if yes, whether the defending party is in a position to demonstrate that it has complied with its requirements. In the absence of any defined standards, a panel is bound to examine the matter in accordance with DSU article 11. The panel therefore is obliged to make ‘an objective assessment of the matter before it, including an objective assessment of the facts of the case and the applicability of and conformity with relevant covered agreements’. This standard entails levels which differ, in particular between questions of law and questions of fact. Questions of Law and Facts On questions of law and interpretation of the provisions of the SPS agreement, the reports clearly show that this is entirely in the hands of the panels and the Appellate Body, although no reference could be found to this effect. The Appellate Body, in particular, undertakes to prescribe the proper reading of rights and obligations without deference to the parties, or the interpretation given to the provision by the panel (Cottier, 1998). The salmon panel was held to have examined the wrong measure, exceeded its terms of reference, and failed to focus on the ban on uncooked salmon. Accordingly, the Appellate Body reversed the decision (salmon, paras 97–105). Things are different for questions of facts and the assessment of scientific evidence. In SPS cases, this is of paramount importance. Measures in SPS article 3.3 need to be based on a scientific justification or the criteria of risk assessment in SPS article 5.1–8 (which are equally, but not exclusively, based on scientific justification, except for article 5.7). In terms of scientific evidence, the review by the panel was held to mean that the result of the risk assessment provided in SPS article 5.5 ‘must sufficiently warrant – i.e. reasonably support – the SPS measures at stake’ (beef-hormones, para. 193; salmon, Panel Report note 3, para. 8.94). This does not mean the standard should be limited to mere standard reasonableness (beef-hormones, para. 119), as it was suggested by the EC, with academic writings as the appropriate standard of deference. Instead, the panel has to assess whether the arguments and findings made are sustained by scientific evidence and sound policy considerations.

50

Risk and the WTO

In agricultural products (para. 79), the Appellate Body defined scientific justification within SPS article 3.3 as establishing a rational relationship between the SPS measure at issue and the available scientific information (Walker, 1998; Wirth, 1994). As the Appellate Body does not have jurisdiction over factual assessment, this is essentially a task for the panel. The Appellate Body does not intervene in correcting factual assessments except in cases of egregious and grave errors or misrepresentations which amount to a violation of DSU article 11, and thus constitute a legal issue over which the Appellate Body has jurisdiction. In agricultural products, it held: As we stated in our Report in European Communities – Hormones, not every failure by the Panel in the appreciation of the evidence before it can be characterized as failure to make an objective assessment of the facts as required by DSU article 11. Only egregious errors constitute a failure to make an objective assessment of the facts as required by DSU article 11.

The panels therefore bear a considerable responsibility when reviewing the factual aspects of the case. The following sections turn to the relevant standards of review in discharging this difficult task. Review of Risk Assessment Factors Within the application of SPS article 3.3, parties have to comply with the requirement of risk assessment set forth in article 5. The Appellate Body set forth the relevant factors as listed in the agreement (beef-hormones, para. 185). In the salmon case, it elaborated the relevant criteria to include (i) the identification of the disease, (ii) the evaluation of the likelihood of the establishment and spread of the diseases and of the biological and economic consequences, and (iii) evaluation of the likelihood (that is, the probability as opposed to mere possibility) of events occurring in the light of the measures which might be applied (salmon, para. 121). Under current definitions of SPS annex A (4), risk assessment comprises all the potential elements, scientific, economic, social and valuational. A distinction made by the parties and the beef-hormones panel to distinguish within SPS article 5.1 and annex A (4), between risk assessment (in a technical sense) and risk management was not upheld by the Appellate Body on grounds that the text did not provide a basis for such a distinction (beef-hormones, para. 181). The reason for the panelists adopting this implied distinction was their attempt to distinguish between scientific evidence and assessment, and those relating to political and economic considerations. Despite the fact that precise definitions are still lacking, such a distinction was considered useful, not only because it is generally used in the field,17 but also because the two groups require different

Risk management experience in WTO dispute settlement

51

standards of review. The distinction allows for a rational and systematic analysis of the measure at stake, and should be retained. The panel first examined the risk assessment on scientific grounds and intended to proceed to societal values and risk management where an existing risk can be affirmed. Risk assessment: review of scientific evidence The review of SPS measures in dispute settlement cases necessarily entails the involvement of an independent expert view regarding the assessment of scientific evidence. The beef-hormones panel sought and found a number of individual experts to conduct the hearings. The approach was upheld and referred to also by the salmon panel and agricultural products panel and upheld by the Appellate Body (beef-hormones, paras 146–54). The hearings were based on questions formulated in the light of legal requirements. The answers provided were evaluated in the light of the arguments and facts submitted by the defendant parties. The panels strongly relied upon expert views, and could not have rendered their assessment and recommendation without such independent advice of individual experts. Again, it is not a matter of asking for a de novo review by the experts, that is, how they themselves would best regulate the problem. Rather, it is a matter of asking whether the facts and findings submitted by the defendant could be seen as warranting the measure or not in light of the evaluation made by individual experts. While the assessment of scientific evidence per se is a matter of fact and essentially a matter left to the panel, the standards of review for scientific evidence are a question of law and may be defined by the Appellate Body. In the beef-hormones case, the panel relied upon extensive scientific evidence and concluded, on the basis of prevailing views in the scientific communities, that the EC failed to demonstrate a risk to health and life by the use of growth hormones if administered in doses according to good veterinary practices. It dismissed hypothetical evidence relating to potential cancerous effects, not yet established nor accepted by the scientific community. Yet, it held that evidence may evolve and change the situation in time (beef-hormones, note 1, paras 7.113, 7.116–7.199). The Appellate Body, however, adopted a more liberal attitude to relevant scientific evidence and dismissed the idea of relying upon prevailing views in the scientific community. It defined the standard as follows: article 5.1 does not require that the risk assessment must necessarily embody only the view of a majority of the relevant scientific community. ... In most cases, responsible and representative governments tend to base their legislative and administrative measures on ‘mainstream’ scientific opinion. In other cases, equally responsible and representative governments may act in good faith on the basis of what, at a given time, may be a divergent opinion coming from qualified and respected sources. (ibid., para. 194; restated in agricultural products, para. 77)

52

Risk and the WTO

This approach, again, implies a considerable amount of deference to WTO members. It leaves governments to choose from scientific evidence available. Yet, while it is true that science is not based on majority views (Wirth, 1994), and long-held prevailing views and paradigms may change due to novel insights necessarily put forth as minority views in the first place (Kuhn, 1970), there is a risk of adopting too broad a scope for scientific justifications of what perhaps may amount to an economically protectionist agenda. Allowing governments wide and open access to all scientific views and data from which they can pick and choose therefore requires some caution and containment. Such containment is inherent in the criteria applied to the relationship of scientific evidence and the risk for life and health discussed above. It is not enough to show the possibility of risk, but a tougher standard of probability of risk has to be shown under the criteria of likelihood discussed above. The Appellate Body qualified this criterion in salmon and agricultural products. ‘It is not sufficient that a risk assessment conclude that there is a possibility of entry, establishment or spread. ... A proper risk assessment ... must evaluate the “likelihood”, i.e. the “probability”, of entry, establishment or spread’ (salmon, para. 123; agricultural products, para. 78). Scientific evidence invoked therefore must meet these standards. Justification based upon far-fetched, speculative and uncorroborated scientific theories and data is excluded. It will be interesting to see how a future panel or the Appellate Body addresses new evidence submitted on health risk relating to the use of growth hormones and other substances under this test in any future assessment of the import ban of hormone-treated beef. Risk management: review of social, economic and ethical factors The standard of ‘sufficiently warranting the measure’ applies to scientific evidence. Different and more deferential standards apply, however, to societal and ethical goals inherent in the criteria set forth in SPS article 5. The Appellate Body ruled in the light of annex A art. 5, according to which the level of protection is described as being ‘deemed appropriate by the member’, and the right under the SPS agreement for members to determine the allowable level of risk, that it is entirely a matter to be defined by governments. The level set cannot be reviewed by the panel even under a standard of reasonableness. DSU article 11 does not allow a panel to substitute its own reasoning about the level of protection (Hilf and Eggers, 1997) (beef-hormones, para. 199). Members therefore enjoy substantial discretion in setting the goals of protection and therefore the import of valuational and hence societal and political considerations. It is important to stress that this can be done within a category of risk management. It was not the intention of the beef-hormones panel to exclude such considerations. Yet, it held that they should only take effect upon establishment of risks to life and health, on the basis of scientific evidence in the

Risk management experience in WTO dispute settlement

53

first place. The panel held: ‘[A]n assessment of risks is, at least for risk to human life or health, a scientific examination of data and factual studies; it is not a policy exercise involving social value judgments made by political bodies.’ The second aspect of a member’s decision to enact or maintain a sanitary measure relates inter alia, to the determination of the appropriate level of sanitary protection by that member against risks to human, animal or plant life or health which have been assessed in accordance with articles 5.1 to 5.3. This aspect is commonly referred to by the parties to this dispute as an essential part of risk assessment. The member wishing to impose a sanitary measure must decide the extent to which it can accept the potential adverse affects related to specific substance which have identified in the risk assessment. (ibid., note 3, paras 8.97 and 8.99; see also 8.8.101, 8.8.107, 8.8.110)

Member states therefore are free to set ambitious levels of protection. Subject to consistency requirements discussed below, they can even adopt a zero risk policy (Atik, 1997), provided, however, that in the first place some minimal risk to health based on scientific evidence can be shown. Members, therefore, are allowed to include political and ethical considerations under the existing agreements in setting their standards of tolerable risk. This also explains why standards more restrictive than those developed by international bodies can be applied. The international system will, in particular, examine whether the measures go beyond what is necessary to achieve the level of protection defined. The test of proportionality under SPS article 5.6 is particularly ripe for judicial review, but hardly effective under an adopted zero-risk policy. The absence of reviews relating to the appropriate level of protection gives maximum autonomy to national standard setting, but there may not yet be an appropriate balance between health and trade interests. If a government decided to operate on a zero risk basis, it is difficult to see how measures less than a ban on importation would be able to achieve this goal. It would seem that the only avenue to discipline excessive measures is the test on consistency discussed below. In future negotiations, the problem of adopting appropriate risk standards should be further studied with the possibility that members could be disciplined if their measures disguised restrictions of trade. It will be necessary to develop appropriate standards and tests, and review procedures for decisions on risk management. To this effect, a proper methodology based upon social sciences should be developed to assist panels and the Appellate Body. Such a methodology should be able to look into factors such as acceptance of new technologies and ethical considerations. For example, polls could be used. Expert hearings should not be limited to science, but extend to social scientists in order to assess to what extent a limitation is primarily motivated by legitimate political and ethical concerns. The present agreement does not sufficiently nor systematically respond to these concerns which need to form part of an integral constitutional approach to assessing SPS measures.

54

Risk and the WTO

Risk assessment and risk management may occur in the absence of scientific evidence, but strong political and societal inhibitions to grant market access for a product may arise from strong ethical resistance. Such conflicts have happened in the context of biotechnology, where appropriate labelling would leave choices to the consumer. In the WTO labelling falls under the TBT agreement, but this needs further development to meet new consumer sensitivities. The Review of Implied Procedural Obligations The panel on beef-hormones held that SPS article 5.1 agreement contains an implied procedural requirement to establish the necessary risk assessment. The Appellate Body held that there is no legal foundation for such a requirement (para. 188) and that this in fact would impair using existing data, including those produced by international organizations (para. 190). The reason for this implied distinction by the panel can be found in considerations of standards of review. In the light of the fact that the standards set forth for the purpose of risk assessment necessarily comprise a considerable leeway, and in part are entirely a matter for the member in the stage of risk management, the review by panels should primarily focus on procedural requirements in risk assessing and turn to scientific and political assessment on substance only to the extent necessary. Whenever the standards of substantive law leave a large degree of discretion, remedies and disciplines should be sought in procedural guarantees. This is compensation by process. The fact that procedures prescribed are affected contains an inherent guarantee that all considerations, scientific and societal, are being considered and measures are not adopted unilaterally on the basis of specific and often protectionist economic interests. The elimination of implied procedural requirements and thus a procedural approach, leaves substantive and procedural requirements mixed and their relationship presently unclear. They should be distinguished, and it seems that the case law increasingly does so. The Appellate Body recognized in salmon an implicit obligation to determine the appropriate level of protection, even without an explicit requirement to do so (salmon, para. 206). In agricultural products, the Appellate Body emphasized and examined procedural requirements to meet the conditions for provisional SPS measures (precautionary principle) in article 5.7 SPS agreement (agricultural products, paras 86–94). A future review of the agreement would benefit from clearly defined procedural obligations which would be subject to full review and thus provide a main guarantee that national risk assessments are undertaken in accordance with the prescriptions of the agreement. This would also be in line with experience and practices by national courts (Wirth, 1994, p. 843).

Risk management experience in WTO dispute settlement

55

THE PROBLEM OF CONSISTENCY Consistency in overall sanitary and phytosanitary policy and measures is most important and an effective test in the examination of a specific measure. SPS article 5.5 merely contains references to classical prohibition of arbitrary and unjustifiable distinctions in different situations which must not result in discrimination or disguised restrictions on international trade. The panels and the Appellate Body have struggled with this open textured provision in connection with SPS article 2.3. The standard of review applied to the (cumulative) criteria of this provision is not clear. The Appellate Body considered this to be a programmatic provision (beef-hormones, para. 210), but nevertheless engaged, as did the panel, in a fairly detailed and diverging comparison of the products which can hardly be achieved on the basis of a reading of the text alone. It affirmed the ‘warning signals’ stemming from different standards as identified by the panel on salmon (para. 160). Indirectly, the level of protection, to be freely defined by members, was limited by the test of consistency; for example, it would be contrary to SPS article 5.5 to adopt a zero-risk policy in one field, while not doing so in a comparable field. The criteria on consistency and comparison provide the most important test in assessing policies for their sectoral economic protectionism. At the same time, the criteria set forth in SPS article 5.5 do not provide much guidance and should be supplemented by additional guidelines. Political and ethical criteria, and a proper methodology discussed above, could perhaps best be placed in the context of an examination of the wider consistency of SPS policies in a member state. As ethical and political concerns (as opposed to economic ones) are not divisible, consistency of application throughout the policies of a member may provide a strong test worth considering in assessing the legitimacy of a contested measure.

IMPLEMENTATION AND ENFORCEMENT Finally, the relationship of sovereignty and standard setting is defined by prescriptions requiring the implementation of decisions by the DSB, taken on the basis of the panel and Appellate Body reports. Under current practices, these reports do not generally specify the measures and steps which should be taken in order to remedy a situation which is inconsistent with the obligations under the SPS agreement. DSU article 19.1 allows for, but does not require, a panel to specify the measures to be taken. Generally, this matter is left to the country concerned. In beef-hormones, it remains unclear whether the EC is obliged to remove the measure, or whether it may remedy the situation by undertaking

56

Risk and the WTO

the necessary operations of risk assessment in conformity with the SPS article 5. The arbitration award did not specify further on these obligations. It held that the Appellate Body report did not primarily emphasize the absence of a suitable specific risk assessment, but that the measure was not based on a risk assessment (beef-hormones, Arbitration Award note 9, para. 37). It denied the need for additional studies in special circumstances which would justify granting an implementation period beyond 15 months. At the same time, the arbitrator affirmed that the undertaking of further studies can form part of the process of implementation: It would not be in keeping with the requirement of prompt compliance to include in the reasonable period of time, time to conduct studies or to consult experts to demonstrate the consistency of a measure already judged to be inconsistent. That cannot be considered as ‘particular circumstances’ justifying a longer period than the guidelines suggested in article 21.3(c). This is not to say that the commissioning of scientific studies or consultations with experts cannot form part of a domestic implementation process in a particular case. However, such considerations are not pertinent to the determination of the reasonable period of time. (ibid., para. 39)

The arbitration did not find the EC to be under an obligation to reverse the measure, but granted 15 months, until 13 May 1999, to implement the findings. Since the EC failed to meet that decision, the US introduced 100 per cent tariffs on selected EU exports, which poses a serious new threat to the multilateral trading system. The absence of specific conclusions is not peculiar to the SPS agreement. Except for one case,18 the panels and the Appellate Body have refrained from specific findings. This leaves the matter with the defendant which may either implement the report, seek transitional compensation or accept the withdrawal of concessions by the complainant. Conducting additional risk assessments is likely to exceed the reasonable period of time granted and thus result in compensation or withdrawal of concessions in complex SPS cases. Experiences gained from the implementation of decisions in the banana cases19 further complicate matters as the current legal regime on implementation shows considerable shortcomings. The parallel avenues of seeking and adjudicating compensation and withdrawal of concessions in SPS articles 22.1 and article 21.5 proved to be difficult to reconcile and clearly need to be revised.20 These weaknesses, as well as the absence of specific recommendations and rulings, are further forms of deference to members. They may well serve national sovereignty in setting health and food standards unimpaired, but operate to the detriment of the credibility of the international trading system as a rule-based system. The balance between rule-based and power-oriented polices has now shifted to the phasing and implementation of decisions, and much work will be needed in the real world to improve predictability of the law and decisions.

Risk management experience in WTO dispute settlement

57

CONCLUSIONS The entry into force of the SPS agreement and subsequent case law clearly show that disciplines for risk assessment and risk management in the field of food and health standards have been considerably enhanced. Governments are required to provide solid justification, both scientific and legal, in order to sustain measures having trade restricting effects. The measures challenged so far in dispute settlements all failed to stand the test, mainly because the necessary trials, scientific assessments and examinations had not taken place. The thorough examination of measures by scientific experts, panels and the Appellate Body cannot and will not remain without implications and feedback for national and regional authorities. In order to prevent costly disputes, the very existence of the WTO dispute settlement system needs to allocate sufficient and adequate resources to the elaboration and administration of quarantine measures and to research with a view to collecting necessary scientific evidence in order to support adopted levels of protection. The experience in dispute settlement has brought to light some deficiencies and weaknesses in the SPS agreement which should be considered in the review process during the next round of trade negotiations. In the light of the experience of panels, the following proposals should be examined with a view to revising and developing further the SPS agreement. • Clarification of the relationships of national or regional and international food and health standards under the SPS agreement are needed. Legal security and predictability can best be achieved by using and promoting international standards. This approach, however, implies that international standards are legitimate. It is not for the WTO to establish international standards itself. The matter is left to appropriate international organizations. However, the SPS agreement may develop further criteria which have to be met in order to render international standards binding and compulsory under the agreement. Conceptually, a difference between binding standards and recommendations should be elaborated, the two categories having different legal effects. • The SPS agreement should be revised and clarified in order to bring about a clear distinction between risk assessment based on scientific evidence and risk management. The latter should comprise economic, social and other factors, as well as scientific evidence. A proper methodology referring to the social sciences should be developed in the context of risk management. In particular, this includes inquiries into the social and political acceptance of an existing risk. Standards of review should be framed accordingly, and examination of scientific evidence and social and political criteria should be undertaken in consecutive steps. Except

58

Risk and the WTO

for provisional measures applied under the precautionary principle under SPS article 5.7, the establishment of a risk based on scientific evidence in the first place should be retained. Giving up such a primary requirement would amount to the equivalent of inviting economically protectionist policies under the guise of risk management. In the absence of scientifically established risk, governments should resort to labelling foodstuffs according to the TBT agreement, leaving decisions about value with the consumer. • The SPS agreement should distinguish substantive and procedural requirements more clearly, and standards of review should be framed accordingly. Review could essentially rely upon procedural requirements while leaving value judgements to governments. Examining whether a member has undertaken the necessary studies and trials before introducing separate standards would facilitate the assignment of panels and assist rational policymaking at home. • Such procedural obligations should also comprise an examination of all criteria set forth by the SPS agreement by governments. A requirement to examine SPS consistency at home will contribute to the prevention of international disputes. • A revised SPS agreement should seek to prescribe more specific remedies, in particular compensation in cases of renewed scientific trials exceeding the reasonable period of time for implementation, but for a defined extension of time. The withdrawal of the measure should be recommended where it can be shown that compliance with existing international standards satisfies the appropriate level of risk as defined by the member. Given the complex issues involved, many might consider these efforts a bridge too far. We are living in a period of scepticism, just as biotechnology is being increasingly employed, and recourse to national sovereignty may seem appropriate to many. Yet, given the potential for serious trade disputes in the field of biotechnology and its underlying social and cultural problems, the first experiences under the SPS agreement should not be forgotten. The next step should be towards a better structured SPS agreement and towards clarification and improvement of its inextricable components.

NOTES 1. WTO (1997), EC Measures concerning Meat and Meat Products (Hormones) – Complaint by the United States – Report of the Panel, WT/DS26/R/USA, 18 August, 1997 [appealed] and EC Measures concerning Meat and Meat Products (Hormones) – Complaint by CanadaReport of the Panel, WT/DS48/R/CAN, 18 August 1997 [appealed]. Without special indication, quotations in this chapter refer to the WT/DS48/R/CAN (Canadian complaint).

Risk management experience in WTO dispute settlement

2. 3. 4. 5. 6. 7.

8.

9. 10.

11.

12.

59

All reports of panels and of the Appellate Body, as well as arbitration awards are located at http://www.wto.org/wto/dispute/panel.htm, and reprinted in The International Trade Law Reports (1996), London, Cameron May. WTO (1998), Australia – Measures Affecting Importation of Salmon, Report of the Panel, WT/DS18/R, 12 June 1998 [appealed]. WTO (1998), Japan – Measures Affecting Agricultural Products – Report of the Panel, WT/DS76/R, 27 October 1998 [appealed]. WTO (1998), EC Measures Concerning Meat and Meat Products (Hormones) – AB-1997-4 – Report of the Appellate Body, WT/DS26/AB/R, WT/DS48/AB/R, 16 January 1998 [adopted]. WTO (1998), Australia – Measures Affecting Importation of Salmon – AB-1998-5 – Report of the Appellate Body, WT/DS18/AB/R, 20 June 1998 [adopted]. WTO (1998), Japan – Measures Affecting Agricultural Product – AB-1998-8 – Report of the Appellate Body, WT/DS76/AB/R, 22 February 1999 [adopted]. WTO (1998), EC Measures Concerning Meat and Meat Products (Hormones) – Arbitration under article 21.3(c) of the Understanding on Rules and Procedures Governing the Settlement of Disputes, WT/DS26/15, WT DS48/13, 29 May 1998 [Award of the Arbitrator Julio LacarteMuro – granting a period of implementation up to May 13, 1999]; the date passed without implementation; WTO (1999), Australia – Measures Affecting Importation of Salmon, Arbitration under article 21.3(c) of the Understanding on Rules and Procedures Governing the Settlement of Disputes, WT/DS18/9 23 February 1999 [Award of the Arbitrator Said-ElNaggar – Granting a period of implementation up to 6 July 1999]; WTO (1999), European Communities – Measures Concerning Meat and Meat Products (Hormones) – Recourse by the United States to article 22.2 of the DSU, WT/DS26/19, 18 May 1999, and WTO (1999), European Communities – Measures Concerning Meat and Meat Products (Hormones) – Recourse by Canada to article 22.2 of the DSU, WT/DS48/17, 20 May, 1999 [Award pending and due July 13, 1999]. See beef-hormones, para. 170 ‘Under article 3.2 of the SPS agreement, a Member may decide to promulgate an SPS measure that conforms to an international standard. Such a measure would embody the international standard completely and, for practical purposes, converts it into a municipal standard. Such a measure enjoys the benefit of a presumption (albeit a rebuttable one) that it is consistent with the relevant provisions of the SPS Agreement and of the GATT 1994’. The approach is comparable to the so called New Approach in technical standardization in EC law. See United States – Measures Affecting Alcoholic and Malt Beverages, para. 5.6 DS23/R, 19 June 1992, BISD 39th Supp. 206, 270/271. WTO (1998), United States – Import Prohibition of Certain Shrimp and Shrimp Products – Report of the Appellate Body, WT/DS58/AB/R, 12 October 1998 para. 185. (Upon concluding that the unilateral measures imposed by the United States amounted, under the facts of the case, to a violation of article XX GATT (chapeau) the Appellate Body held ‘We have not decided that the protection and preservation of the environment is of no significance to the members of the WTO. Clearly, it is. We have not decided that the sovereign nations that are members of the WTO cannot adopt effective measures to protect endangered species, such as sea turtles. Clearly, they can and should. And we have not decided that sovereign states should not act together bilaterally, plurilaterally or multilaterally, either within the WTO or in other international fora, to protect endangered species or to otherwise protect the environment. Clearly, they should and do.’) See also FAO (1998, p. 13) (‘It is noteworthy that the SPS and TBT Agreements both acknowledge the importance of harmonizing standards internationally so as to minimize or eliminate the risk of sanitary, phytosanitary and other technical standards becoming barriers to trade’). Beef-hormones, para. 172 (also para. 104). In para. 104 the Appellate Body confirmed that exceptions are not to be construed in a narrow sense: ‘In much the same way, merely characterising a treaty provision as an “exception” does not by itself justify a “stricter” or “narrower” interpretation of that provision than would be warranted by examination of the ordinary meaning of the actual treaty words, viewed in context and in light of the treaty’s

60

13.

14. 15.

16.

17.

18.

19.

20.

Risk and the WTO object and purpose, or, in other words, by applying the normal rules of treaty interpretation.’ It should be voted that the Panel had no intention to construe article 3.3 narrowly. The obiter dictum of the Appellate Body is of importance and welcomed in light of previous functionalist practices to construe GATT exceptions, in particular article XX, in a narrow manner; see in particular United States Restrictions on Tuna, GATT BISD 39th supp. 155 para. 5.22 (1993), 30 ILM 1594, 1619 (1991). ‘[V]arious international tribunals, including the International Court of Justice, have generally and consistently accepted and applied the rule that the party who asserts a fact, whether the claimant or the respondent, is responsible for providing proof thereof. Also, it is a generallyaccepted canon of evidence in civil law, common law and, in fact, most jurisdictions, that the burden of proof rests upon the party, whether complaining or defending, who asserts the affirmative of a particular claim or defense. If the party adduces evidence sufficient to raise a presumption that what is claimed is true, the burden then shifts to the other party, who will fail unless it adduces sufficient evidence to rebut the presumption,’ WTO, United States – Measures Affecting Imports of Woven Wool Shirts and Blouses from India, WT/DS33/AB/R para. 14 [adopted 23 May 1997], see also agricultural products, para. 121. Beef-hormones, para. 98, see also Japan – Agricultural Products, para. 122. The Appellate Body clearly follows rules of evidence in common and Anglo-Saxon law, while disregarding the power of courts in civil law to assess evidence under an investigating or inquisitional principle. Claims and arguments need to be made and presented by parties in order to be taken into account by panels and the Appellate Body. The Appellate Body held: ‘Panels are inhibited from addressing legal claims falling outside their terms of reference. However, nothing in the DSU limits the faculty of a panel freely to use arguments submitted by any of the parties – or to develop its own legal reasoning – to support its own findings and conclusions on the matter under its consideration. A panel might well be unable to carry out an objective assessment of the matter, as mandated by article 11 of the DSU, if in its reasoning it had to restrict itself solely to arguments presented by the parties to the dispute’ (beef-hormones, para. 156). The distinction can be found in the FAO/WHO definition of ‘A Risk Management: The process of weighing policy alternatives in the light of the results of risk assessment and, if required, selecting and implementing appropriate control options, including regulatory measures’. Risk assessment policy, however, is defined broadly to be ‘Guidelines for value judgment and policy choices which may need to be applied at specific decision points in the risk assessment process’, Risk Management and Food Safety – Report of a Joint FAO/WHO Consultation, Rome 27–31 January 1997, FAO, Food and Nutrition Paper no. 65 (FAO: Rome 1997, p. 4). See also, Wirth (1994, pp. 817, 833), ‘Although by no means universally accepted, one approach that expressly acknowledges this dichotomy prescribes a bifurcation of the regulatory process into two phases: “risk assessment”, which in principle establishes the strictly scientific basis for regulatory action, and “risk management”, which is the multidisciplinary process of choosing regulatory measures.’ WTO (1996), United States – Restrictions on Imports of Cotton and Man-made Fibre Underwear, Report of the Panel, WT/DS24/R, 8 November 1996, paras 8.1–8.3. The panel ruled in light of article 19.1 DS: ‘We, consequently, recommend that the Dispute Settlement Body request the United States to bring the measure challenged by Costa Rica into compliance with US obligations under the ATC. We find that such compliance can best be achieved and further nullification and impairment of benefits accruing to Costa Rica under the ATC best be avoided by prompt removal of the measure inconsistent with the obligations of the United States. We further suggest that the United States bring the measure challenged by Costa Rica into compliance with US obligations under the ATC by “immediately withdrawing the restrictions imposed by the measure”’ (para. 8.3). See WTO (1997), European Communities – Regime for the Importation, Sale and Distribution of Bananas. Report[s] of the Panel, WT/DS27, 22 May 1997; WTO (1997), Appellate Body – European Communities – Regime for the Importation, Sale and Distribution of Bananas, Report of the Appellate Body, WT/DS27, 9 September 1997. While the United States requested authorization to suspend concessions under article 22.2 DSU, leading to arbitration under article 22.6 DSU and the granting of an annual amount of

Risk management experience in WTO dispute settlement

61

USD 191.4 million per year, Ecuador sought a review of new import measures by way of recourse to arbitration under article 21.5 DSU. The European Communities equally filed a complaint under this provision, arguing procedures under article 21.5 to be exclusive, thus impeding procedures to withdraw concessions under article 22.2 at the same time. The panel (composed of the same persons in all three simultaneous proceedings) denied such a request, and assessed the measures in all three proceedings simultaneously and necessarily disregarding diverging time periods of these procedures (article 22.6: 60 days; article 21.5: 90 days). The constellation caused considerable confusion, both procedurally and politically. See WTO (1999), European Communities – Regime for the Importation and Sale and Distribution of Bananas – Recourse to Arbitration by the European Communities under article 22.6 of the DSU – Decision by the Arbitrators, WT/DS27/ARB, 9 April 1999; WTO (1999), European Communities – Regime for the Importation, Sale and Distribution of Bananas – recourse to Article 21.5 by Ecuador – Report of the Panel, WT/DS27/RW/ECU, 12 April 1999; WTO (1999), European Communities – Regime for the Importation, Sale and Distribution of Bananas – Recourse to Article 21.5 by European Communities – Report of the Panel, WT/DS27/RW/EEC, 12 April 1999.

REFERENCES Atik, Jeffrey (1997), ‘Science and international regulatory convergence’, Journal of International Law, 17, 736–44. Barcel, John J. (1994) ‘Product standards to protect the local environment – the GATT and the Uruguay Round Sanitary and Phytosanitary Agreement’, Cornell International Law Journal, 27, 755. Cottier, T. (1998), ‘The WTO Dispute Settlement System: New Horizons’, Proceedings of the 92th Annual Meeting, Washington, DC: American Society of International Law, 89–91. Eggers, B. (1998), ‘Die Entscheidung des WTO Appellate Body im Hormonfall’, Europäische Zeitschrift für Wirtschaftsrecht, 9, 147–51. FAO (1998), FAO Technical Assistance and the Uruguay Round Agreements, 2nd edn, Rome: FAO, pp. 12–20. Hilf, M. and B. Eggers (1997), ‘Der Panel-Bericht im EG/US Hormonstreit: Anstoss zum grenzenlosen Weltbinnenmarkt für Lebensmittel oder Eigentor der WTO?’, Europäische Zeitschrift für Wirtschaftsrecht, 8, 559. Jackson, John H. (1999) reprinted, ‘The World Trade Organization: Constitution and Jurisprudence’, Chatham House Papers, London: The Royal Institute of International Affairs. James, Sallie and K. Anderson (1998), ‘On the Need for More Economic Assessment of Quarantine/SPS Policies’, Centre for International Economic Studies Seminar Paper 98–02, University of Adelaide, February 1998 (since published in the Australian Journal of Agricultural and Resource Economics, 42(4), 425–44). Kuhn, Thomas S. (1970), The Structure of Scientific Revolutions, 2nd edn, Chicago: The University of Chicago Press. Shin, Yukyan (1998) ‘An analysis of the WTO Agreement on the Application of Sanitary and Phytosanitary Measures and its implications in Korea’, Journal of World Trade, 32, 85–119. Walker, Vern R. (1998), ‘Keeping the WTO from becoming the “World Trans-science Organization” – science, uncertainty, science policy, and factfinding in the growth hormones dispute’, Cornell International Law Journal, 31, 251.

62

Risk and the WTO

WHO (1998), Food Safety and Globalization of Trade in Food: A Challenge to the Public Health Sector, Food Safety Issues (WHO/FSF/FOS, 8 Rev.1.). Wirth, David A. (1994), ‘The role of science in the Uruguay Round and NAFTA trade disciplines’, Cornell International Journal, 27, 817, 854. WTO (1995), ‘The results of the Uruguay Round of multilateral trade negotiations’, The Legal Texts, Geneva: WTO Secretariat, 69–84.

5.

Applying SPS in WTO disputes Joost Pauwelyn*

This chapter provides an account of how WTO dispute settlement panels and the Appellate Body have interpreted the SPS agreement. (The SPS agreement is explained by Wilson and Gascoine in Chapter 11, this volume.) Commentaries on specific cases, in particular the high profile beef-hormones dispute, have also been written (Hurst, 1998; Charnovitz, 1997, 1998; Wirth, 1998). The focus here is on how the rights and obligations of WTO members in the SPS agreement have been clarified by decisions reached by the dispute settlement body (DSB). SPS provisions not yet invoked in disputes are not dealt with. Dispute settlement reports are only one source of SPS interpretations but represent a consistent approach (Roberts, 1998). It is an objective summary which raises some unanswered questions, while providing some guidance to health and quarantine regulators by addressing the intersection of trade and health protection. This assessment will focus on the three major SPS disputes agricultural products,1 beef-hormones2 and salmon.3 The three pillars of the SPS agreement have been addressed (human health in beef-hormones; animal health in salmon; and plant health in agricultural products). The clarifications and instructive statements made in these three cases use case law. This gives prominence to legal precedence which is not always used in systems of national law and could lead to new conflicts in the future. The SPS committee, established to monitor and clarify the SPS agreement, comprises all 135 WTO members. It finished its first review of the SPS agreement in 1999. It emphasized that the SPS agreement had contributed to improving international trading relationships with respect to sanitary and phytosanitary measures and welcomed the fact that a number of SPS-related trade matters had been resolved following discussion at formal meetings of the committee, or bilaterally (SPS Committee Report, 1999). Certain concerns were raised but no formal changes to the text of the SPS agreement were recommended, except some minor amendments to notification of procedures. *

Any opinions expressed are strictly personal and not to be attributed to my employer the WTO Secretariat. I am grateful for useful comments from William Davey, Oliver Landwehr, Gabrielle Marceau and Erik Wijkstrom. The usual disclaimer applies.

63

64

Risk and the WTO

This chapter addresses certain developments in substantive rights and obligations under the SPS agreement, and describes some of the more procedural elements in SPS disputes. It concludes with an assessment of how reports have been implemented.

NOTES ON SUBSTANTIVE RIGHTS AND OBLIGATIONS UNDER THE SPS AGREEMENT What is an SPS Measure and When Does the SPS Agreement Apply? Whether a specific measure is regarded as an SPS measure is a subjective question. It depends on the aim or purpose of the regulation. If it is enacted with a view to protect human, animal or plant life or health, it is an SPS measure (SPS annex 1). So far, this subjective criterion has not been controversial. In all three disputes, the defending party argued that health protection was the purpose of the measure challenged.4 What if a defendant argues that health protection is only one of the objectives, that consumer concerns or moral issues are relevant also? What level of health protection is enough for the measure to be an SPS measure? What if a defendant claims that it is not protecting health? Can a panel examine this matter of intent? Once it is established that the measure is applied to protect life or health, two additional (more objective) criteria have to be met. First, the measure needs to protect against either (1) so-called ‘food-borne’ risks (human or animal life or health) or (2) pest or disease-related risks (human, animal or plant life or health). The distinction between these two categories of risk is important for an SPS definition. The kind of risk assessment to be conducted is different for each of these risk categories (see below).5 In beef-hormones, the alleged risk was food-borne and related to human life and health; the risk related to contaminants (hormone residues) in foods (meat and meat products). In salmon, Australia claimed to protect its fish against the introduction of some 24 salmon diseases. In agricultural products, Japan wanted to avoid the introduction of the codling moth, considered to be a plant pest, by banning fruit imports from the United States. The second criterion for the SPS agreement to apply, is that the measure needs to ‘directly or indirectly affect’ international trade (SPS article 1.1). So far this criterion has not caused any controversy. In dispute settlement, two decisions have been taken about when the SPS agreement applies. First, the SPS agreement was found to apply to all SPS measures affecting international trade.6 Before the SPS agreement, health regulations only had to be justified if a prior violation of one of the GATT principles had been found.7 GATT article XX(b) (allowing measures ‘necessary to protect human, animal or plant life or health’) is only activated once a

Applying SPS in WTO disputes

65

violation of the non-discrimination provisions (GATT articles I or III) has been established. Under the SPS agreement, all disciplines apply regardless of whether a GATT violation exists. The second decision found that the SPS agreement applies. The SPS agreement applies not only to SPS measures enacted after 1 January 1995; but also to SPS measures introduced before the WTO commenced operation and still in force.8 This means that many WTO members need to re-examine their existing SPS measures, in light of SPS obligations.9 SPS Measures Based on Science SPS measures cannot be maintained ‘without sufficient scientific evidence’ (SPS article 2.2). (The exception to this rule is the possibility to enact provisional measures according to the precautionary principle, discussed below.) The only dispute where this requirement of ‘sufficient scientific evidence’ has been explicitly addressed, was agricultural products: The ordinary meaning of ‘sufficient’ is ‘of a quantity, extent, or scope adequate to a certain purpose or object’. From this, we can conclude that ‘sufficiency’ is a relative concept. ‘Sufficiency’ requires the existence of a sufficient or adequate relationship between two elements, in casu, between the SPS measure and the scientific evidence … we agree with the Panel that the obligation … that an SPS measure not be maintained without sufficient scientific evidence requires that there be a rational or objective relationship between the SPS measure and the scientific evidence. Whether there is a rational relationship between an SPS measure and the scientific evidence is to be determined on a case-by-case basis and will depend upon the particular circumstances of the case, including the characteristics of the measure at issue and the quality and quantity of the scientific evidence.10

Applying this test to Japan’s requirement that approval to import certain fruits has to be sought separately for each fruit variety (Japan’s so-called varietal testing requirement), the panel (as well as the Appellate Body) concluded that the requirement was not maintained with sufficient scientific evidence.11 What is a ‘Risk Assessment’? SPS article 5 requires that quarantine measures must be based on a risk assessment. A risk assessment to protect human beings or animals against ‘foodborne’ risks is easier to conduct than a risk to protect against pests or disease risk. Risks are defined as follows: • The evaluation of the potential for adverse effects on human or animal health arising from the presence of additives, contaminants, toxins or disease-causing organisms in food, beverages or feedstuffs.

66

Risk and the WTO

• The evaluation of the likelihood of entry, establishment or spread of a pest or disease within the territory of an importing member according to the sanitary or phytosanitary measures which might be applied, and of the associated potential biological and economic consequences. Additional principles have been developed through case law: • a risk assessment can either be quantitative or qualitative;12 • a risk assessment does not need to establish a minimum magnitude of risk; a WTO member may determine that its acceptable level of risk is ‘zero risk’;13 • the risk being assessed must be an ascertainable risk; theoretical uncertainty is not the kind of risk to be assessed; the existence of unknown and uncertain elements does not justify a departure from the risk assessment requirement; • a risk assessment needs to be specific; a separate risk assessment must be conducted for each substance, a generic risk assessment for a class of substances is not enough;14 also, the studies of a risk assessment need to be specific enough to address the particular kind of risk at stake;15 • a country imposing an SPS measure does not have to conduct the required risk assessment itself; it can use assessments carried out by others or by international organizations.16 A crucial distinction was, however, found between the risk assessment required for food-borne risks and that for disease or pest risks: While [a risk assessment for ‘food-borne’ risks] requires only the evaluation of the potential for adverse effects on human or animal health, the [risk assessment for disease or pest risks] demands an evaluation of the likelihood of entry, establishment or spread of a disease, and of the associated potential biological and economic consequences. (Salmon, Appellate Body report, note 19)

It is, therefore, not sufficient for a risk assessment on disease or pest risks to conclude that there is a possibility of entry, establishment or spread of diseases or pests and associated biological and economic consequences. A proper risk assessment must evaluate the likelihood (that is, probability) of the entry, establishment or spread of diseases or pests and associated biological and economic consequences. This evaluation of probability can be expressed either qualitatively or quantitatively. In beef-hormones, the panel applying the food-borne risk assessment definition followed a two-step analysis. First, did the EC identify adverse effects? Second, did it evaluate the potential of these effects to occur (ibid.,

Applying SPS in WTO disputes

67

paras 129, 134)? The Appellate Body in beef-hormones concluded that the EC had not actually undertaken such a risk assessment and found that the EC had violated the SPS obligations (ibid., para. 208). In salmon and agricultural products, where the disease or pest risk assessment definition was applied, the three requirements were specified: (a) identify the diseases whose entry, establishment or spread is being prevented within its territory, as well as the potential biological and economic consequences associated with the entry, establishment or spread of these diseases; (b) evaluate the likelihood of entry, establishment or spread of these diseases, as well as the associated potential biological and economic consequences; (c) evaluate the likelihood of entry, establishment or spread of these diseases according to the SPS measures which might be applied.17 In salmon, the Appellate Body found that Australia had not met requirements (paras 129, 134). In agricultural products, it found that (at least in respect of apricots, pears, plums and quince) the third requirement was lacking (paras 113–14). In both cases, violations of SPS article 5.1 were found. In salmon, it was found that if a measure is not based on a risk assessment, it can be presumed not to be based either on scientific principles nor to be scientifically justified.18 In legal terms, a violation of the more specific SPS article 5.1 presumes a violation of the more general SPS article 2.2. What Evidence is Relevant to Assessing Risk? SPS article 5 sums up specific factors to be taken into account in a risk assessment. The main issue is whether ‘non-scientific’ factors, such as consumer concerns, cultural or moral preferences and social values (not explicitly mentioned in the SPS agreement) can be used to determine whether a risk exists. The panel on beef-hormones was of the view that ‘an assessment of risks is, at least for risks to human life or health, a scientific examination of data and factual studies; it is not a policy exercise involving social values made by political bodies’.19 In its opinion non-scientific factors should not be taken into account in risk assessment but in risk management; that is, in determining the acceptable level of risk and the selection of the SPS measures to be used. The Appellate Body reversed the panel decision to conclude: There is nothing to indicate that the listing of factors that may be taken into account in a risk assessment of SPS article 5.2 was intended to be a closed list. It is essential to bear in mind that the risk that is to be evaluated in a risk assessment under article 5.1 is not only risk ascertainable in a science laboratory operating under strictly controlled conditions, but also risk in human societies as they actually exist; in other

68

Risk and the WTO words, the actual potential for adverse effects on human health in the real world where people live and work and die. (Ibid., para. 187)

The Appellate Body added that the object and purpose of the SPS agreement should justify the examination and evaluation of all such food-borne risks for human health whatever their precise and immediate origin may be (ibid., para. 206). Even though most commentators hold the view that SPS article 5 includes a closed list of factors which ‘can’ be considered, it does suggest factors which ‘shall’ be taken into account without indicating other factors can also be considered. Resolving this confusion will be important for the effective operation of the SPS agreement. When is an SPS Measure Based on Acceptable Risk Assessment? The SPS agreement not only requires member countries to conduct a risk assessment (or to refer to a risk assessment conducted elsewhere), but they also have to base any measures on a risk assessment.20 Yet, it does not involve a subjective or procedural examination of the regulator’s decision-making process. The Appellate Body in beef-hormones argued: ‘The requirement that an SPS measure be “based on” a risk assessment is a substantive requirement that there be a rational relationship between the measure and the risk assessment.’ The Appellate Body noted that even minority opinions could provide a rational relationship: We do not believe that a risk assessment has to come to a monolithic conclusion that coincides with the scientific conclusion or view implicit in the SPS measure. The risk assessment could set out both the prevailing view representing the ‘mainstream’ of scientific opinion, as well as the opinions of scientists taking a divergent view … In most cases, responsible and representative governments tend to base their legislative and administrative measures on ‘mainstream’ scientific opinion. In other cases, equally responsible and representative governments may act in good faith on the basis of what, at a given time, may be a divergent opinion coming from qualified and respected sources. By itself, this does not necessarily signal the absence of a reasonable relationship between the SPS measure and the risk assessment, especially where the risk involved is life-threatening in character and is perceived to constitute a clear and imminent threat to public health and safety.21

The Appellate Body rejected the panel’s view that there was a procedural requirement imposed on WTO members ‘to submit evidence that it took into account a risk assessment when it enacted or maintained its sanitary measure in order for that measure to be considered as based on a risk assessment’.22 The rejection of such a procedural requirement opens the door for ex post facto jus-

Applying SPS in WTO disputes

69

tification of SPS measures (Hurst, 1998, p. 13). In practice, only evidence submitted at the time of the panel proceedings matters; not evidence that was available when the SPS measure was first enacted or later re-examined and maintained. Provisional SPS Measures can be Taken Under SPS article 5.7, members can enact provisional SPS measures: (a) they are imposed where ‘relevant scientific information is insufficient’; and (b) they are adopted on the basis of available pertinent information (agricultural products, Appellate Body report, para. 89). Such provisional measures may not be maintained unless: (c) the instigator seeks to obtain the additional information necessary for a more objective assessment of risk; and (d) reviews the … measure within a reasonable period of time. The Appellate Body in agricultural products, where the defendant argued that its measure was a provisional one, found these four conditions to be cumulative. If one is not met, the measure will be found to be inconsistent with the SPS agreement. The Appellate Body also defined the option of provisional measures as a qualified exemption from the obligation to base SPS measures on sufficient scientific evidence (ibid., para. 80). Hence the party imposing a provisional SPS measure has to prove that it meets all four requirements. The Appellate Body added that: neither SPS article 5.7 nor any other provision sets out explicit prerequisites regarding the additional information to be collected or a specific collection procedure. Furthermore, article 5.7 does not specify what actual results must be achieved; the obligation is to ‘seek to obtain’ additional information. However, article 5.7 states that the additional information is to be sought in order to allow the member to conduct ‘a more objective assessment of risk’. Therefore, the information sought must be germane to conducting such a risk assessment. (agricultural products, Appellate Body report, para. 92)

Applying this requirement to Japan’s varietal testing procedures, the Appellate Body agreed with the panel that Japan did not seek to obtain the additional information necessary for a more objective risk assessment. The Appellate Body noted that the ‘reasonable period of time’ to review the provisional measures depends on the specific circumstances of each case,

70

Risk and the WTO

including the difficulty of obtaining the additional information necessary for the review and the characteristics of the provisional SPS measure (ibid., para. 93). The Appellate Body concluded that although the obligation to review the varietal testing requirement had only been in existence since 1 January 1995, Japan had not reviewed its varietal testing requirement ‘within a reasonable period of time’. Although Japan’s measure might be considered to be provisional and adopted in accordance with requirements (a) and (b) of SPS article 5.7, it did not meet the other two requirements of that article (ibid., para. 94). In beef-hormones, the EC did not claim that its import ban was a provisional measure, but it invoked the precautionary principle under SPS article 5.1 (risk assessment). The Appellate Body agreed with the panel that the precautionary principle (other than that expressed in SPS article 5.7 on provisional measures) did not override the obligation to base SPS measures on a risk assessment (ibid., para. 125). On the status of the precautionary principle in international law, the Appellate Body noted: it is regarded by some as having crystallized into a general principle of customary international environmental law. Whether it has been widely accepted by members as a principle of general or customary international law appears less than clear … We note that … the precautionary principle, at least outside the field of international environmental law, still awaits authoritative formulation. (Ibid., para. 123)

This remains controversial in many international agreements. WTO Members Set Their Own Levels of Risk Once a proper risk assessment has been conducted and an ‘ascertainable risk’ is detected, the member concerned has to make a social judgement whether to accept that risk.23 This decision will be made in accordance with a government’s decision on the acceptable level of risk or ALOP. If it can accept the risk, no SPS measures are required. If it cannot, then an appropriate SPS measure will be selected to reduce the risk to an acceptable level. The acceptable level of risk is a crucial part in determining whether an SPS measure is WTO consistent. It is an important concept of the SPS agreement, after ‘sufficient scientific evidence’ and ‘risk assessment’ has been decided. In salmon, the Appellate Body argued that the determination of the appropriate level of protection (para. 5 of SPS annex A) is a prerogative of the member concerned and not of a panel or of the Appellate Body (ibid., para. 199). The beef-hormones panel referred to the process of determining and applying the acceptable level of risk as part of risk management, where social value judgements (risk perceptions and consumer tolerance) are crucial.24 The Appellate Body in salmon implied that a member’s acceptable level of risk could even be set at zero risk.25

Applying SPS in WTO disputes

71

How to Select SPS Measures with ‘Acceptable Level of Risk’ If an assessed risk is found to be unacceptable (based on ALOP), risk mitigating measures can be imposed to reduce the risk to that acceptable level. In respect of disease and pest risk, SPS measures that might be imposed should be examined in the risk assessment, remembering that measures should not be more trade-restrictive than required to achieve their appropriate level of quarantine protection. No Discrimination Among Markets or Products SPS article 2.3 includes the traditional GATT requirement of non-discrimination, provided in GATT article I and GATT article III. In addition SPS article 5.5 includes a prohibition on discriminations between situations or products, irrespective of their origin. In case law, three elements have been required for a violation of SPS article 5.5: (a) adopting different appropriate levels of sanitary protection in different situations; (b) evidence of arbitrary or unjustifiable differences in levels of protection (c) leading to ‘discrimination or a disguised restriction on international trade’.26 The most controversial element under the SPS objective of ‘consistency’ has been that an SPS measure results in discrimination or a disguised restriction on international trade. The Appellate Body in beef-hormones stressed that it is not enough for this to be met when there is an arbitrary or unjustifiable distinction between situations or products which results in a trade restriction. 27 In addition, acts of discrimination between imported and domestic production need to be established. For example, the Appellate Body after deciding that an arbitrary or unjustifiable distinction was involved in the way the EC treated hormones in cattle and antimicrobials in swine, concluded that no violation of SPS article 5.5 could be found.28 In salmon, on the other hand, the Appellate Body confirmed the panel’s finding of a violation of SPS article 5.5. Australia, by banning imports of salmon products but allowing imports of herring used as bait and live ornamental finfish, which represent similar risks of disease introduction, acted inconsistently with SPS article 5.5.29 Presumption of SPS Measure in Conformity with International Standards In beef-hormones, the Appellate Body stated that a member government wanting to impose an SPS measure has three options:

72

Risk and the WTO

(a) It can enact an SPS measure which conforms to an international standard. This meets all SPS disciplines, including the requirement of sufficient scientific evidence and risk assessment. (b) It may choose to base its SPS measure on an international standard. It will still be subject to full scrutiny under all SPS disciplines, in particular the obligation to base the measure on sufficient scientific evidence and a risk assessment. (c) It may decide to deviate from the international standard and to set its own different protection which requires a ‘scientific justification’.30 The beef-hormones panel found that any deviation from international standards is an exception to the general rule that SPS measures have to be based on international standards.31 It was up to the EC to prove that it met the conditions linked to that exception. In an important decision, the Appellate Body reversed the panel’s decision stating that the SPS agreement recognizes the autonomous right of a member to establish a higher level of sanitary protection than would be achieved by a measure based on an international standard. Transparency The SPS article 7 prescribes that all measures adopted (laws, decrees or ordinances) must be published promptly in such a manner that enables interested parties to become acquainted with them. In agricultural products, the Appellate Body noted that this obligation applied also to other instruments which are applicable generally and are similar in character to the instruments explicitly referred to (ibid., para. 105). The panel in that case had noted that a non-mandatory government measure is also subject to WTO provisions where compliance is necessary to obtain an advantage from the government. Accordingly, Japan’s varietal testing requirement, even though it was not mandatory, was found to violate the SPS agreement (agricultural products, Appellate Body report, para. 108).

NOTES ON WTO DISPUTE SETTLEMENT PROCEDURES RELEVANT TO SPS DISPUTES Terms of Reference A panel can only address those measures and legal claims sufficiently specified in the complaining party’s request for the establishment of a panel (that is, the

Applying SPS in WTO disputes

73

request on the basis of which the WTO’s Dispute Settlement Body (DSB) establishes the panel).32 This document sets out a panel’s terms of reference. Moreover, a panel can only make findings in respect of the specific products which have been referred to in the panel request.33 Measures, claims or products not sufficiently specified in the panel request cannot be subject to a panel finding or Appellate Body consideration. Complainants must specify in the panel request precise and complete details of measures on which it wants panels to make legal findings. Once a legal claim is validly before a panel, the complainant can make any argument or submit any evidence under that claim, even if the argument or evidence had not been mentioned in the panel request. The panel may use its own legal arguments to come to a conclusion. In its report it is not required to address explicitly all arguments or evidence submitted to it by the parties or the scientific experts (beef-hormones, Appellate Body report, para. 133). Traditionally, once a GATT panel had found a violation in respect of one legal claim it did not address other legal claims, referring to the principle of ‘judicial economy’. The WTO dispute settlement system, with its appeals processes, means that panels tend to examine more than one claim.34 This approach ensures that the Appellate Body disposes of necessary factual assessments by the panel in case it reverses the panel’s first finding of violation. The Appellate Body cannot make findings of fact (DSU article 17.6). Who Has to Prove that there is a Risk/No Risk? The issue of which party bears the burden of proof has been very prominent in SPS disputes (Pauwelyn, 1998, p. 227). In most SPS cases there is scientific uncertainty. Is it then for the country imposing the restriction to prove that there is a risk (or sufficient scientific evidence to support the restriction)? Or is it for the challenger to prove there is no health risk (not sufficient scientific evidence)? In the beef-hormones case, it was decided that the initial burden rests with the complaining party, which must establish a prima facie case of inconsistency with a particular provision of the SPS agreement by a defending party. If that prima facie case is made, the burden of proof moves to the defending party, which must in turn counter or refute the claimed inconsistency.35 Elsewhere, the Appellate Body argued that the party who asserts a fact, whether the claimant or the respondent, is responsible for providing proof. It is generally accepted in civil law, common law and most jurisdictions that the burden of proof rests upon the party who asserts the affirmative of a particular claim or defence.36 In essence, the party in an SPS dispute asserting a fact (for example, the existence of a relevant international standard), claim (for example that an SPS

74

Risk and the WTO

measure is not maintained with sufficient scientific evidence) or defence (that the measure challenged is a provisional one according to SPS article 5.7),37 bears the burden to prove it. However, that party does not always need to provide complete proof of the fact, claim or defence to be accepted by the panel. Once it is able to establish a prima facie case or presumption that it is correct, it will be up to the other party to rebut that case. The Appellate Body has defined a prima facie case as one which, in the absence of effective refutation by the defending party, requires a panel, as a matter of law, to rule in favour of the complaining party presenting it. The Appellate Body has argued that weighing of evidence by a panel falls outside the scope of appellate review.38 A Panel’s Investigation Panels have to make an objective assessment of the relevant facts. They cannot conduct a de novo review (that is, their own risk assessment). On the other hand, they cannot give complete deference to the findings of national authorities.39 In beef-hormones the Appellate Body referred to and applied the interpretative principle of in dubio mitius, in deference to the sovereignty of states. This principle states that if the meaning of a term is ambiguous, that meaning is to be preferred which is less onerous to the party assuming an obligation. Scientific Experts Panels are mostly composed of trade experts and academics. SPS disputes involve complex scientific questions. To make an objective assessment, all three SPS panels sought advice from scientific and technical experts using SPS article 11.2 and DSU article 13. These experts acted as individuals and were not required to put forward a consensus view. The panel questioned each individual expert, but their opinions were not binding on the panel. Parties to the dispute were able to comment on the experts’ answers and to meet with the experts and the panel to discuss answers and comments. In agricultural products, the Appellate Body noted that the comprehensive nature of a panel’s authority to seek information from external sources suggested that panels had a significant investigative authority. However, the Appellate Body also found that expert scientific evidence could not be used by the panel to establish an inconsistency with the SPS agreement if the complaining party itself had not first established a prima facie case of inconsistency. Apart from the scientists appointed by the panel, the disputing parties may themselves be assisted or represented by scientists of their choice. These

Applying SPS in WTO disputes

75

scientists were allowed to attend panel meetings as part of a participating delegation.

THE STATUS OF IMPLEMENTATION OF THE PANEL AND APPELLATE BODY REPORTS IN SPS DISPUTES The reports in all three SPS disputes discussed here have been adopted by the Dispute Settlement Body (DSB). The adoption of reports is automatic unless a consensus against them is found (DSU articles 16 and 17). As required, the EC, Australia and Japan expressed their intentions to implement reports finding against them, consistent with their WTO obligations. Losing parties to a WTO dispute have to comply ‘immediately’ with the recommendations and rulings adopted by the DSB. If it is impracticable, they are granted a period for implementation (15 months is the normal maximum; see DSU article 21.3). In beef-hormones and salmon, no agreement could be reached on an appropriate period, so it was determined by an arbitrator (member of the Appellate Body who had not been involved in the cases). In beef-hormones, the Arbitrator decided that no additional time could be added for the EC to conduct a new risk assessment. The deadline for implementation was set at 13 May 1999. In salmon, the Arbitrator confirmed that no time could be allowed to conduct additional studies. Since Australia had originally asked for 15 months, and argued that most of that period would be used to conduct risk assessments, the Arbitrator decided that less than 15 months was needed. Australia was granted eight months which terminated on 6 July 1998. In agricultural products, parties agreed that the reasonable period of time to implement the reports was nine months and twelve days; that is, by 31 December 1999. The EC, Australia and Japan were obliged to submit a status report to each meeting of the DSB on the progress made in the implementation phase, starting six months after the decisions. In beef-hormones, the EC conceded that it failed to implement the decisions by 13 May 1999. As a result, but only following an additional arbitration dispute run by the original panel, the USA and Canada were authorized to take ‘retaliatory’ action against the EC – in the form of a 100 per cent tariff increase imposed on certain EC products entering the USA/Canada – for an annual trade value of up to, respectively, US$116.8 and CDN$1.3 million. Salmon was also not settled by the expiration of the deadline of 6 July 1999. In that case, Canada requested authorization to take ‘retaliatory’ action worth annually CDN$45 million. Australia objected, arguing that its new measure

76

Risk and the WTO

was SPS consistent. In February 2000, the original panel decided, however, that it was not (even though large parts of Australia’s new regime were upheld).

NOTES 1. WTO (1999), Japan – Measures Affecting Agricultural Products – Report of the Appellate Body, WT/DS76/AB/R [not yet adopted]. This report was issued as a result of an appeal by both Japan and the United States against the report of the panel (Japan – Measures Affecting Agricultural Products – Report of the Panel, WT/DS76/R, circulated on 27 October 1998 [not yet adopted]). 2. Reports of the panels and Appellate Body on [WTO (1998)] EC Measures Affecting Meat and Meat Products (beef-hormones) WT/DS26/R/USA and WT/DS26/AB/R – panel as a result of a complaint by the United States [adopted on 13 February 1998]; and WT/DS48/R/CAN and WT/DS48/AB/R – panel as a result of a complaint by Canada. 3. Reports of the panels and Appellate Body on [WTO (1998)] Australia – Measures Affecting the Importation of Salmon, WT/DS18/R and WT/DS18/AB/R – complaint by Canada [adopted on 6 November 1998]. 4. Beef-hormones, panel reports, para. 8.22 (US panel) and para. 8.25 (Canada panel); salmon, panel report, para. 8.32; Agricultural products, para. 8.12. 5. The correct classification of the measure at issue (is it protecting against ‘food-borne’ risks or pest or disease risks?) was discussed before the panel on salmon (paras 8.31–8.37). The panel found that even though the measure in dispute might cover both categories of risk, the objectives for which it was being applied were more appropriately covered by the definition of an SPS measure applied to protect against disease risk (the Appellate Body, agreed with this). As a consequence, the panel also applied the stricter definition of risk assessment which corresponds to this kind of SPS measure (at para. 8.68). 6. Beef-hormones, panel reports, para. 8.36 (US panel) and para. 8.39 (Canada panel). 7. See WTO (1990), Thailand – Restrictions on Importation of and Internal Taxes on Cigarettes – Report of the Panel (DS10/R), 37S/200 [adopted on 7 November]. 8. Beef-hormones, panel reports, paras 8.24–8.28 (US panel) and paras 8.27–8.31 (Canada panel) and Appellate Body report, paras 128–30. 9. Beef-hormones, panel reports, para. 8.99 (US panel) and para. 8.102 (Canada panel) and Appellate Body report, para. 129; salmon, panel report, para. 8.56. 10. Agricultural products, Appellate Body report, paras 73 and 84 (footnote omitted). 11. Panel report, para. 8.43; Appellate Body report, para. 85. 12. Beef-hormones, Appellate Body report, paras 184 and 186; salmon, Appellate Body report, para. 124. 13. Ibid., respectively, in para. 186 and paras 124 and 125. 14. Beef-hormones, Appellate Body report, para. 201, confirms the panel’s finding in para. 8.257 (US panel) and para. 8.260 (Canada panel). The panel on salmon (para. 8.58) clarified, however, that some of the evidence and evaluations contained in a risk assessment for one product may be relevant for the risk assessment for other products, so that a completely new risk assessment for each product may not be necessary. 15. Beef-hormones required studies had to address the carcinogenic or genotoxic potential of hormone residues in meat derived from cattle to which hormones had been administered for growth promotion purposes; general studies showing the existence of a general risk of cancer were not enough (beef-hormones, Appellate Body report, para. 200). 16. Beef-hormones, Appellate Body report, para. 190 and salmon, Appellate Body report, footnote 68. 17. Salmon, Appellate Body report, para. 121 and agricultural products, Appellate Body report, para. 112. 18. Salmon, Appellate Body report, paras 137–8, panel report, paras 8.52 and 8.99. 19. Ibid., para. 8.94 (US panel) and para. 8.97 (Canada panel).

Applying SPS in WTO disputes

77

20. It was failing this requirement that the EC was found to violate SPS article 5.1 (panel report, para. 8.156 (US panel) and para. 8.159 (Canada panel); Appellate Body Report, para. 208). 21. Ibid., para. 194. It reiterated this view in respect of SPS article 2.2, discussed above, in agricultural products, para. 77. 22. Ibid., para. 8.113 (US panel) and para. 8.116 (Canada panel). 23. Beef-hormones, panel report, paras 8.160 ff. (US panel) and para. 8.163 ff. (Canada panel). 24. Ibid., paras 8.95 ff. and 8.160 ff. (US panel) and paras. 8.98 ff. and 8.163 ff. (Canada panel). The use of risk management was, however, rejected by the Appellate Body on grounds that it has no basis in the SPS agreement (op. cit., para. 181). 25. Op. cit., para. 125. See also beef-hormones, Appellate Body report, para. 186. 26. Salmon, panel report, para. 8.108; Appellate Body report, para. 140. 27. Appellate Body, para. 240, reversing the panel report, paras 8.203, 8.216 and 8.241 (US panel) and paras 8.206, 8.219 and 8.244 (Canada panel). 28. Op. cit., para. 245. Hurst (1998) argued ‘the EC’s good intentions save its measures from an article 5.5 violation’. The Appellate Body transformed the analysis under SPS article 5.5 ‘into an unprincipled inquiry into the intent of the measure’. 29. Op. cit., paras 178 and 240. It is interesting to note that a violation of article 5.5 was presumed to constitute also a violation of article 2.3 (salmon, panel report, paras 8.109 and 8.160; Appellate Body report, paras 178 and 240). It was added, however, that ‘given the more general character of article 2.3, not all violations of article 2.3 are covered by article 5.5’ (panel report, para. 8.109). 30. In agricultural products, the Appellate Body noted that ‘there is a ‘scientific justification’ for an SPS measure, within the meaning of article 3.3, if there is a rational relationship between the SPS measure at issue and the available scientific information’ (op. cit., para. 79). 31. Op. cit., para. 8.87 (US panel) and para. 8.90 (Canada panel). 32. Refer to articles 6.2 and 7.1 of the Understanding on Rules and Procedures Governing the Settlement of Disputes (DSU). In salmon, the panel considered the measure at issue to be certain regulations in so far as they prohibit the importation of fresh, chilled or frozen salmon and, from a sanitary perspective, the heat-treatment requirements imposed by Australia which resulted in the import ban on these salmon products (op. cit., paras 8.19 and 8.95). The Appellate Body reversed the panel in this respect, finding that ‘the SPS measure at issue in this dispute can only be the measure which is actually applied to the product at issue’. Since the product at issue was ‘fresh, chilled and frozen salmon’, the measure at issue was the import prohibition imposed on these products; not the heat-treatment requirements which, according to the Appellate Body, only apply to smoked or heat-treated salmon (op. cit., paras 101–5). For a limitation put on the legal claims validly before the panel: salmon, panel report, para. 8.28; agricultural products, panel report, para. 8.4. 33. Salmon, panel report, para. 8.20; Agricultural products, panel report, para. 8.6. 34. Beef-hormones, panel report, paras 8.163 ff. (US panel) and paras 8.166 ff. (Canada panel), where the claim under SPS article 5.5 (consistency) was addressed, even though a violation in respect of the article 5.1 claim (risk assessment) had already been found. The same occurred in salmon, where the panel also examined Canada’s article 5.6 claim (op. cit., paras 8.103 ff. and paras 8.161 ff.). In agricultural products, the panel made a finding of violation of claims under article 2.2, article 5.6 (reversed by the Appellate Body, op. cit., paras 204 and 213) and article 7. 35. Beef-hormones, Appellate Body report, op. cit., para. 98. 36. Report of the Appellate Body on United States – Measure Affecting Imports of Woven Wool Shirts and Blouses from India, adopted on 23 May 1997, WT/DS33/AB/R, p. 14. 37. Given that the Appellate Body in agricultural products, para. 80, found that ‘article 5.7 operates as a qualified exemption from the obligation under article 2.2 not to maintain SPS measures without sufficient scientific evidence’, article 5.7 is arguably an exception to be invoked by the defendant and for which the defendant bears the burden of proof. 38. Salmon, para. 261 and agricultural products, paras 98 and 136. 39. Beef-hormones, Appellate Body report, para. 117, referring to Article 11 of the DSU.

78

Risk and the WTO

REFERENCES Charnovitz, S. (1997), ‘The World Trade Organization, meat hormones and food safety’, International Trade Reporter, 14, 1781. Charnovitz, S. (1998), ‘Environment and Health under WTO dispute settlement’, The International Lawyer, 32(3), 901–22. Hurst, D. (1998), ‘Hormones: European Communities measures affecting meat and meat products’, European Journal of International Law, 9(1), 182 (www.ejil.org/journal/ Vol9/No1/sr1g.html [19/01/2000]) Pauwelyn, J. (1998), ‘Evidence, proof and persuasion in WTO dispute settlement’, Journal of International Economic Law, 1(2), 227–58. Roberts, D. (1998), ‘Preliminary assessment of the effects of the WTO agreement on sanitary and phytosanitary trade regulations’, Journal of International Economic Law, 1(3), 377–405. SPS Committee Report (1999), Review of the Operation and Implementation of the SPS Agreement, March (G/SPS/12). Wirth, D. (1998), ‘European Communities’ measures affecting meat and meat products’, American Journal of International Law, 92(4), 755–75.

PART II

Managing Risk in Policymaking

6.

Risk assessment and risk management in policymaking Marion Wooldridge*

INTRODUCTION: SOME THOUGHTS ON POLICY AND RISK Why do we Need Policies? Governments introduce policies to regulate human activities, presumably for the greater good. Free trade and market forces bring economic benefits to consumers and producers, and this is the basis of the WTO system. Therefore, if free trade and market forces were unfettered by any policies, would this allow everyone to benefit from advances in science and manufacturing, and improvements in trade facilities, as much as they wished and without coercion? What would happen without any policies to regulate imports, exports, methods of manufacture of the myriad different types of product available today, and uses of any new scientific discovery which might have a marketable application? There would be some advantages, for example: • consumers would have plenty of choice • manufacturers could utilize the cheapest production methods • new drugs, vaccines and technologies of all sorts could come quickly onto the market. But there would be no control over the outcomes from the use of these new products and processes. Many of these outcomes would be uncertain. The future, always unpredictable, would become much more so. With high levels of uncertainty and lack of knowledge, a common response is the generation of high levels of apprehension, followed possibly by fear and perhaps panic. This *

The final version of this paper has benefited from comments on an earlier draft offered by Gareth Davies MRCVS (Independent veterinary consultant to EU), R.A. (Robin) Bell MRCVS (Veterinary International Trade Division, MAFF), Ms Heather Blake and Mr Gary Larkman (International Trade and Tropical Foods Policy Division, MAFF), Dr Henri Belveze and Prof. Jean-Louis Jouve (EU Directorate General XXIV).

81

82

Managing risk in policymaking

is not very efficient, nor very comfortable – and is in fact the classic description of a risky situation, since risk refers to: • a situation where a number of different outcomes are possible • at least one outcome is unwanted • and which outcome(s) will occur is unknown. Therefore a key requirement of policymaking is risk management – to attempt to reduce the uncertainties (and anxieties) of the future. A Brief History of International Risk Management Trading started early in the history of the human race, but from the outset was not without its risks, real and perceived. The most obvious risks from contact with the tribes of strangers was probably plunder, theft and violence. If the risks of these were assessed as high, one obvious risk management method was to keep one’s own tribe armed at the ready. Increasing sophistication in international relationships and trade identified different potential risks, often closely connected with increasing travel. These included the spread of new and dangerous ideas, and the spread of disease – the latter by animals and goods as well as people. Often the barrier to disease was simply to put border or movement controls in place and ban the import of animals and goods, and the movement of sick people. This often works but it is restrictive, as all the benefits of trade are lost and other countries are likely to respond by taking the same approach. Today, rapid transport and mass transport means that infectious agents can easily travel around the world, safely within their hosts as undetected incubating diseases. Vectors of disease can be carried rapidly to distant continents, and even if they cannot adapt and survive, may live long enough to pass on disease. This is not restricted to tiny insects either; potentially rabid bats from the USA and other more exotic locations have happily arrived in the UK within aircraft cockpits and suitcases. Multinational companies mean that many individual products are also distributed worldwide, often rapidly and sometimes without great local awareness of their origins. This includes foods, vaccines and other biologicals, and drugs. These products may contain genetically modified (GM) products, toxins, residues and pollutants, which are then also widely distributed. Both direct and indirect effects can be hypothesized, the latter particularly from GMs and pollutants. The major drivers for this risky international movement in both goods and people is affluence, acquisitiveness and consumer demand. In summary, the scale, speed and potential global effects of international movements today make international risk management cooperation vital.

Risk assessment and management in policymaking

83

RISK MANAGEMENT PROBLEMS, PRINCIPLES AND PRACTICE Some Common Risk Management Problems Given that a hazard has been identified and the risk assessed then, if everyone agreed what to do about it, risk management would be greatly simplified. But this is seldom the case. This leads to the first common problem – a clash of interests. It is often the case that those who benefit most from whatever risky endeavour is proposed are not those who run the greatest risks. For example in the case of the import of a new line of valuable and expensive breeding animals, it is likely to be the entrepreneurial importers who stand to benefit most directly and rapidly if the venture succeeds. However, if an exotic, serious and virulent disease is imported along with these animals, then the whole industry stands to lose both directly in terms of sick or dead animals, and indirectly as it is also likely that the export status of that industry will suffer long-term difficulties too. So how do risk managers make their decisions? The second common problem lies in the fact that generally and in ideal circumstances nobody wants to run any risks at all (unless in the name of sport!) – but zero risk is impossible to achieve. It is always possible (however unlikely) to hypothesize a series of events that will lead from the hazard of interest to the unwanted outcome. Even if there has been no outbreak or case recorded of cattle disease X in country Y since records began, and the risk of disease from an import from that country seems zero, someone right now could be transporting into that country on their shoe the bug responsible for that disease. The first outbreak could be headline news next week; a low probability maybe, but always a possibility. Sometimes the risk pathway is much more obscure, and it is difficult to convince people that zero risk is impossible. Heaping on safeguards will not produce zero risk however much it reduces the magnitude. So how far should the safeguards go? Other problems in managing risks include the prevalence of a short-term viewpoint, sometimes politically driven (the ‘not whilst I’m in office’ syndrome), perhaps coupled with a highly visible media campaign to ‘do something’ to guarantee that a particular outcome cannot occur. For example, ‘That rail crash last week which killed 157 people and injured hundreds more was horrific – we must have an inquiry; it must never be allowed to happen again’. Obviously – and sadly – this is impossible; zero risk just does not exist. But an intensive or highly politicized campaign may lead to erroneous perceptions about the magnitude of the risk which is out of all proportion to any known facts; people start seeing every train as an instant death trap despite available statistics showing quite the reverse. Managers react to an apparent

84

Managing risk in policymaking

urgency; all trains must be fitted with automatic warnings, new brakes, speed controllers and video cameras on long periscopes for seeing 100 miles up the track. So when does urgent action become necessary? Risk Perception Risk perception is mentioned in the preceding paragraph. It is hypothesized that how a risk is perceived affects one’s attitude to whether that risk is acceptable or not. Studies have been done on what affects people’s perceptions of risk. Major factors include: • whether exposure is voluntary or involuntary (Fischhoff et al., 1978) (cigarette smoking versus unidentified genetically modified soya in food, for example); • whether there is any control over the risk or not (risks run in many sports would probably be unacceptable if the participants did not feel in control, and driving a car is the classic example); • whether the person is knowledgeable on the subject or not (for example a study which showed that in general scientists working in the nuclear industry thought nuclear risks were much lower than did the general public, and recent public discussions amongst genetic engineers suggest a similar attitude there) (Frewer et al., 1997; Hunt and Frewer, 1997). Even when information on a particular risk is available, the amount of trust given to the source of that information affects the perception of the level of risk; again studies have shown that (at least in the UK) (Frewer et al., 1994; Frewer et al., 1996): • information from government sources and scientists is not greatly trusted; • information from environmental pressure groups and consumer organizations is much more highly trusted; • the media also rate more highly than the government in terms of trust. Thus, if the ‘Daily Rag’ says the level of chemical Z in your milk is going to give everyone itchy-left-leg syndrome, and the government scientist says the probability of this is exceedingly remote, there will still be a lot of scratching going on. Other major factors involved in risk perception are the severity of the consequences, and cultural differences. Risk Management Principles Some of the major problems in risk management have been highlighted, and this has resulted in a series of questions, but as yet no answers. However, risk

Risk assessment and management in policymaking

85

managers must make decisions; even deciding to do nothing, which may be appropriate in some circumstances, is a decision. To aid their decision making, risk managers need information from many sources. But what is the fundamental decision which they need to make? Clearly, it cannot be in the first instance; ‘Which safeguards shall we put in place?’, but rather ‘What is the acceptable level of risk in this situation?’ This is a policy decision potentially influenced in differing degrees by many factors including the presence or absence of international agreements. Once this level has been decided upon (the difficult part), the risk manager selects safeguards to achieve risks at or below those accepted levels, and this again then becomes part of the policy in force. In some circumstances where consequences are considered so severe that no risks can be reasonably described as acceptable, the level chosen may be more appropriately described as a tolerable risk level. But what constitutes an acceptable level of risk? How is that level decided upon? And who else might be involved in that decision? Information Needed by the Risk Manager Effective risk management relies on information: • the risk manager needs to be aware of hazards, including new and emerging hazards, the risks resulting from those hazards, their magnitude, consequences, and which groups bear those risks; • the risk manager also needs to know what the benefits from the proposed activity are likely to be, as well as the magnitude of those benefits and who is most likely to receive them; • information on the costs of each potential course of action open to the risk manager are necessary, as well as who actually pays those costs. The information described so far might be considered to be the ‘facts’ of the case (although that word ‘facts’ must be used with care; bias creeps in almost everywhere). But one other area of information is crucial: • the risk manager needs information on what ‘the stakeholders’ (see below) actually want, and their perceptions of the risks under consideration. For the ‘factual’ part of the information, risk managers may call for formal risk assessments, either qualitative or quantitative, and formal cost–benefit and risk–benefit analyses. They may use expert systems to help predict the outcome of certain courses of action. For hazard identification, surveillance systems may be in place, dedicated ‘early-warning’ monitoring units such as that for the US Department of Agriculture at Fort Collins may have been set up, and

86

Managing risk in policymaking

‘what-if ...?’ brainstorming think-tanks may be utilized. But what about information from the stakeholders? Who are these stakeholders? How do we get information from them? And why should they be involved anyway? Is it not enough to have only the ‘facts’? Stakeholders Stakeholders refers to any individual, group or body involved with, affected by, or interested in a risk or a risk management decision, and this includes those generations not yet born (particularly relevant in environmental and genetic issues). Take as an example the proposed new import of meat product X, intended for human consumption, into country Y. ‘Private’ stakeholders include: • the potential consumers of the product (often including, but certainly not fully represented by, pressure groups with their own interests and agendas), • the potential importers and retailers of the product, • the farmers who supply the meat in the country of origin, • the farmers with similar livestock in the importing country, • the health professionals, both human and veterinary, who will have to deal with any human or animal health problems resulting from this import. ‘Public’ stakeholders include: • the government of importing country Y, which is responsible for legislation on national safety and international trade, and which in most countries (whatever the system of government) will also have at least an eye on its own popularity ratings; • the governments of the international trading partners of country Y, where at the least a perception of ‘liberal trade’ is necessary if trading is to be successfully maintained; • regulatory and standard setting bodies connected with that international trade (and specifically the World Trade Organization (WTO)) if any international agreements have been entered into through those bodies as a result of a desire to trade (examples here include the Sanitary and Phytosanitary Agreement (SPS), and the Technical Barriers to Trade Agreement (TBT)). And there may well be other groups one could identify as stakeholders here. From this, it might reasonably be concluded that in the realm of international trade, almost everybody in the world can be considered a stakeholder. This is

Risk assessment and management in policymaking

87

particularly so if a decision reached in respect of one import or trade issue has the legal ability to set a precedent for future trade with other countries. Having identified some stakeholders, how do we go about involving them in the risk management and decision-making process? For this we need, not surprisingly, something that has been given the name of risk communication. At first sight this might sound simple, but like most risk management issues there are complexities. By definition, where there is a risk, there is uncertainty, and for some groups of people in some situations this can be a very difficult concept to convey effectively, particularly perhaps if they have a poor grounding in scientific method and a quantitative risk estimate is being given in isolation and out of context. Closely bound up with this is the issue of terminology, particularly specialized or scientific terminology. Most people want to know if the air is ‘safe’ to breathe, the water ‘safe’ to drink, the food ‘safe’ to eat and the transport ‘safe’ to use. Since we have already seen that there is no such thing as zero risk, what is the meaning of the word safe? P = 1 × 10–3? P = 1 × 10–20? It seems unlikely that most people think in those terms. Perhaps in any case it means different things to different people. As previously discussed, prior media reports, trust and the way in which risk has already been perceived will have affected communication on risk perceptions. But any type of risk communication will influence risk perception in some way; the simple fact that a hazard is being discussed for some people may result in the perception that ‘It must be very risky or why is anyone bothering with it?’. And existing risk perception in its turn leads to a bias in how incoming information and communications are ‘processed’, as does the level of trust in the risk communicator or the body or group which they represent. Any stakeholder who participates in any form of risk communication, whether formally or informally, has some motive for doing so (even if only curiosity) or they would not take part. With any issue we can be pretty certain that some stakeholders actually want to ensure a particular outcome and this is likely to affect the way they take part in communications. So maybe risk communication is not as easy as it first appeared. It is clear that risk communication must be a two- (or even multi-) way process, but the format depends to a certain extent upon the purpose of that communication, which differs at different stages of the risk management process. The major purpose may be, for example, to obtain information from scientists or consumers on estimated or acceptable risk levels, costs and benefits, or to give information to consumers (and get feedback) on the results of these analyses, or to discuss management options and safeguards with a range of stakeholders. Potential methods include: • working parties both national and international with some or all of policymakers, technical experts, consumer and industry representatives selected in some way;

88

Managing risk in policymaking

• • • • •

joint meetings between policymakers and specific interest groups; public meetings, focus groups and ‘juries’; dedicated helpdesks and interactive websites; leaflets, both targetted and freely available; more lengthy discussion documents.

But given this multiplicity of possibilities, how is the best and most appropriate method selected in any particular circumstance? The recognition of the importance of risk communication is relatively recent, and assessing the best approach in each type of situation is an area of active research. Examples include: • in the UK, the Health and Safety Executive have recently asked for proposals for a project to investigate how best to select ‘experts’ for working parties, and once selected how best to elicit and use the information they are able to give; • the Waste Policy Institute at Blacksburg, Virginia, USA have a unit devoted to the very applied aspect of devising public communication strategies and assessing how to maximize their effectiveness. This ranges from methods for the successful organization of public meetings to reviewing and improving the design of simple information pages for their website; • the Institute of Food Research in the UK has a risk communication research unit which is more academic in nature, and projects include those looking at how the media affects risk perception, and how risk perception affects decision making. Members of my department also collaborate with them in the latter research area. One idea largely intended to improve international risk communication which is currently being floated in the European Union (EU) is a management course for public sector veterinarians which would bring together, and improve understanding between, what are effectively risk managers from the very varied cultural backgrounds of the EU. Involving stakeholders in any meaningful way in a process takes time, effort and money. In addition, it often requires a massive cultural shift in an organization to undertake the process effectively. Some of the advantages are also immediate, and include: • the fact that the process is democratic, and is seen to be; • the risk manager’s understanding of the issues involved is increased resulting in a better ability to gauge opinions on what are considered acceptable risks by different groups in society;

Risk assessment and management in policymaking

89

• it increases the manager’s overall knowledge base on the subject. However, many of the advantages are long term in nature and less intuitively obvious: • the overall time and expense of decision making is reduced, as unexpected challenges (both legal and practical) are also vastly reduced and have much less chance of popular support or legal success; • decisions reached with broad consensus generally require less regulatory enforcement or ‘policing’ and are therefore less costly to uphold; • when stakeholders have been involved at all stages, the necessity for huge public enquiries long after the event and at massive public cost is removed; • the transparency of the process improves the credibility of, and trust in, agencies that manage risk.

INTERNATIONAL RISK Risk Communication International risk communication is in essence no different from national and local risk management, and is vital if international agreements are to be effective, but there are some additional factors which may affect successful communication: • different native languages are an obvious source of potential misunderstanding, but it is probably the subtleties and nuances in all languages which are the true danger here, and these can be missed even by experienced translators; • countries may have different priorities and different political influences and levels of stability affecting their ability or willingness to communicate freely; • some cultures are more risk-averse than others, and whilst this can be true within any one country, it is likely to be accentuated internationally, as well as by any difference in economic importance of the product or industry involved; • commonly accepted standards of purity, safety and thus risk are also likely to depend heavily upon the wealth and regulatory infrastructure of the countries concerned.

90

Managing risk in policymaking

Finally on this topic, it must be remembered that although all viewpoints may be sought, all opinions listened to, and all weighed up in the final decision, if there are diametrically opposed requirements from different groups of stakeholders (as is very often the case), then it will be impossible to find a risk management strategy which fully satisfies all. This is often forgotten by those stakeholders who are least satisfied by the outcome, who may then complain bitterly that their opinions were sought only to be ignored. It is therefore essential to emphasize at an early stage the impossibility of pleasing everyone completely in such circumstances. Requirements of International Risk Management The general principles of international risk management are consistency, equivalence, fairness, transparency, feasibility and trust. Many of these principles are enshrined in WTO agreements, including the SPS agreement and a later paper will deal with this in considerable detail. However, some points require expansion here. The first is that of consistency which, in practice, means that for a given importing country a broadly similar level of import risk should be equally acceptable (or equally unacceptable) whatever the country of origin of that risk. However, the only way to know whether risks are of broadly similar level or not is to undertake some form of risk assessment. And, both from a theoretical and methodological viewpoint as well as a legalistic one, such a risk assessment should be based on up-to-date scientific knowledge. That a risk management policy should be feasible is obvious, and for this to be the case the risk mitigation measures required must be practicable. But common sense suggests these safeguards should also be proportional to the risk assessed. Demanding eight months quarantine when the longest known incubation period for the disease in question is two weeks is plainly out of proportion. However, what about a month – is this out of proportion? Here is possibly an area for negotiation. The Precautionary Principle At this point it is appropriate to mention the precautionary principle, about which there is currently a lot of interest, at least in the EU. Effectively what this says is that when the risk manager believes, on the basis of an initial examination of the evidence, that there is a risk of serious (and possibly irreversible) consequences contingent upon a given hazard, steps should be put in place to minimize that risk whilst a more thorough risk analysis (including a risk assessment) is performed. One difficulty is deciding what constitutes the cutoff point for the initial examination of the evidence – the precautionary principle

Risk assessment and management in policymaking

91

could potentially be invoked very widely to ban an import or process on flimsy evidence, perhaps for other reasons. A second difficulty is deciding what is a reasonable time limit for the performance of the more thorough risk analysis. Again there could be non-risk-related reasons for delaying the completion of such an analysis, used as an excuse to keep precautionary measures in place. That said, there is clearly a reasonable case for the existence of the precautionary principle for use in appropriate circumstances. Trust is not something which any agreement can by itself provide, but it is nevertheless the basis of successful international trade. Trust is built on communication and understanding, underpinned by monitoring and experience. Acceptable Risk The concept of acceptable risk has been mentioned, but what does actually constitute an acceptable risk? It should by now be clear that there is no single value which can be quoted, no simple number or rule of thumb. In any given case, the risk manager must decide, based on all the information to hand. This includes information on the risks, including assessed level of risk from the hazard in question and the context in which this risk exists. This means looking at other sources of the same risk (for example are low levels of medical radiation significant in the overall risk of radiation exposure?), other risks from the same source (for example do these chickens carry only Campylobacter, or do they carry Salmonellae as well?). It means checking for risks which develop as a result of mitigations taken to reduce the initial risk. It also includes information on the costs, benefits, local conditions, culture and community values, including political and ethical values. In short, the decision as to what constitutes an acceptable risk is a balancing act, with no correct answer, merely a functional one.

RISK MANAGEMENT The major phases in risk management are summarized in Figure 6.1. Good risk management comprises: • • • • •

using the best information using multisource information is feasible – and benefits are related to costs is effective uses alternative to ‘command and control’, if possible; for example by making recommendations, which may then have an effect on market forces

92

Managing risk in policymaking

• is sensitive to political and social situations • includes incentives.

The problem and its context Evaluation of decisions and actions

Bring the stakeholders into the process

Actions

Hazards and risks

Options for management Decisions

An iterative process Figure 6.1

The major phases in risk management

RISK ASSESSMENT IN POLICYMAKING Risk assessment has been mentioned on many occasions. What exactly is it, and what can it do? The main features of a risk assessment pertinent to risk management and policymaking need to be reviewed. What Risk Assessments Do (and Do Not) Provide • Risk assessment provides information for risk managers. A risk assessment may be qualitative or quantitative, and answers the question; ‘What is the risk of ...?’ Qualitatively, the probability is given in words, for example the probability may be high, low, negligible and so on. Quantitatively, point values of probability may be given, or if stochastic methods have been used a result such as ‘We can be 99 per cent certain that P is less than value y’ can be obtained. • Risk assessment does not give a totally accurate prediction of the future!

Risk assessment and management in policymaking

93

This is obvious when confidence limits are given, but less so with point value estimates, and this is sometimes forgotten. • Risk assessment does not give information clearly accessible to the public. Uncertainty is present, whether implicit or explicit, and most people prefer certainty in sources of information and decision making. It is an iterative process, and new information leads to an updating of the risk estimates. Such changes can be difficult to accept and handle. Risk results refer to populations, and most people are interested in their own personal safety; a risk assessment will tell them nothing about this. These factors may be barriers to understanding the results from a risk assessment unless carefully presented. What Risk Assessment Does (and Does Not) Do • Risk assessment is a useful decision making aid. • Risk assessment is used in conjunction with other information for risk management. It is not necessarily the most important information in reaching a final decision on management strategy. • Risk assessment does not answer the question ‘Is this risk acceptable?’ • Assessed risk must be clearly differentiated from acceptable risk in risk management and policymaking. • Risk assessment can help select effective safeguards. Once an acceptable risk level is decided, if the assessed risk is above that level, then a quantitative model can be used to test the effect of each considered safeguard on the level of assessed risk. Risk Assessment in International Policy • Risk assessment is necessary to underpin an international position in a trade dispute based on risk. If safeguards more stringent than accepted international safeguards are required in a particular situation, and result in a barrier to trade or a trading dispute, then a risk assessment is necessary to underpin its legality. That risk assessment must meet established international standards; in the SPS agreement the standards applied to animals and their products are set by the Office International des Epizooties (OIE).

94

Managing risk in policymaking

A SUMMARY OF RISK ASSESSMENT AND RISK MANAGEMENT IN POLICYMAKING Key points are: • • • • • •

Agreed policies reduce conflict Risk management is at the heart of policymaking Effective communication is vital Risk management balances assessed risk, acceptable risk and other factors Risk assessment provides necessary information Risk assessments underpin risk management and policymaking.

BIBLIOGRAPHY AND REFERENCES Selected general references: Risk Analysis The Commission (1997), The Presidential/Congressional Commission on Risk Assessment and Risk Management: Final Report. Volume 1: Framework for Environmental Health Risk Management. Volume 2: Risk Assessment and Risk Management in Regulatory Decision Making, Washington, DC: The Commission. Wooldridge, M., T. Bernardo, L. Audige, W. Hueston, D. Pfeiffer and D. Pritchard (1997), Risk Analysis and Animal Health: A Course Manual, Collated by R. Ruppaner for an International Training Course in Bubendorf, Switzerland: 13–18 July 1997.

Quantitative Assessment Methodology Vose, D. (1996), Quantitative Risk Analysis: A Guide to Monte Carlo Simulation Modelling, Chichester, England: John Wiley & Sons Ltd.

Specific References Fischhoff, B., P. Slovic, S. Lichtenstein and B. Combs (1978), ‘How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits’, Policy Studies, 9, 127–52. Frewer, L.J., M. Raats and R. Shepherd (1994), ‘Modelling the media: the transmission of risk information in the British quality press’, Journal of the Institute of Mathematics and its Applications to Industry, 5, 235–47. Frewer, L.J., C. Howard, D. Hedderley and R. Shepherd (1996), ‘What determines trust in information about food-related risks? Underlying psychological constructs’, Risk Analysis, 16(4), 473–86. Frewer, L.J., C. Howard, D. Hedderley and R. Shepherd (1997), ‘The elaboration likelihood model and communication about food risks’, Risk Analysis, 17(6), 759–70. Hunt, S. and L.J. Frewer (1997), ‘Public perception of naturally occurring radiation risks’, Proceedings of the Society for Risk Analysis Europe, Stockholm.

7.

Allowing for risk in setting standards Michael J. Nunn

Risk analysis has been recognized only recently as a formal study in its own right. Several attempts have been made to develop a standardized nomenclature in, for example, disciplines such as animal health (Hathaway, 1991; Ahl et al., 1993; Kellar, 1993; MacDiarmid, 1993; OIE, 1994; North, 1995) and food safety (Codex, 1993; ANZFA, 1996; Notermans and Mead, 1996). However, there is still some confusion about the precise definition of the elements in risk analysis (Krewski and Birkwood, 1987; Covello and Merkhofer, 1993). For example, some authorities use ‘risk management’ instead of ‘risk analysis’ for the overall term (for example SA/SNZ, 1995) while others use ‘risk analysis’ more narrowly as including elements such as risk identification, assessment and evaluation but excluding risk management and communication. Despite such variations in terminology, the basic principles are the same across all disciplines (Nunn, 1997). Risk analysis is used here as the overall term to encompass elements of risk assessment, risk management and risk communication. Risk assessment is the process of identifying and estimating the risk associated with an option, including evaluation of the likelihood of an event and of the consequences if that event were to occur. Risk management is the process of identifying, documenting and implementing measures to reduce risk (either the likelihood of occurrence or the consequences). Risk communication is the process of interactive exchange of information and opinions concerning risk between risk analysts and stakeholders. This terminology helps avoid the mistake of treating risk communication as the process of risk analysts advising stakeholders (such as policymakers or the general public) of the result of their risk assessment and their proposed risk management strategies. Such a limited approach implies that communication is primarily one-way and occurs only after the risk assessment and management steps have been completed, ignoring the need for two-way communication and consultation throughout the whole process, which is fundamentally iterative in nature. This chapter examines how standards relating to trade allow for risk. Its primary focus is on international obligations and standards that apply to import risk analysis used by quarantine authorities in the development of sanitary and 95

96

Managing risk in policymaking

phytosanitary measures against the spread of disease. However, it also considers how risk might be allowed for in other standards relating to trade, in areas such as genetically modified organisms.

INTERNATIONAL OBLIGATIONS The WTO agreement has two annexes with particular relevance to quarantine and trade – the SPS agreement and the TBT agreement (see Wilson and Gascoine, Chapter 11, this volume). The SPS agreement defines the basic rights and obligations of member countries with respect to ‘sanitary and phytosanitary measures’ to protect human, animal or plant life or health, and the TBT agreement covers food standards such as labelling and nutritional requirements. WTO members are obliged to ensure their quarantine measures are based on an assessment of the risks to human, animal or plant life or health, ‘taking into account’ risk assessment techniques developed by the relevant international organizations. For many years, the Office International des Epizooties (OIE) and the International Plant Protection Convention (IPPC) and the Codex Alimentarius Commission (Codex) have advocated the use of risk analysis principles in animal quarantine, plant quarantine and food safety, respectively. Through annex A of the SPS agreement, WTO recognizes these organizations as the custodians of the international standards, guidelines and recommendations for sanitary and phytosanitary aspects of international trade. Risk assessment and risk management are included in the SPS principle of ‘control, inspection and approval procedures’ as fundamental to the application of sanitary and phytosanitary measures to international trade. Risk communication is implicit in the SPS agreement, particularly in relation to its principle of transparency, which obliges members to notify changes of their quarantine measures, and in the consultative processes used by OIE, IPPC and Codex to develop their international standards, guidelines and recommendations. Thus risk analysis – including risk assessment, risk management and risk communication – is integral to international trade rules in the WTO.

PRINCIPLES The SPS agreement defines a number of principles governing sanitary and phytosanitary measures that affect international trade – basic rights and obligations; harmonization; equivalence; risk assessment; regionalization; national treatment; transparency; control, inspection and approval procedures;

Allowing for risk in setting standards

97

technical assistance; and special and differential treatment (see appendix). Similar principles apply to risk analysis in other disciplines, as reflected in approaches adopted to food safety (ANZFA, 1996) and environmental sciences (Beer, 1996). With respect to import risk analysis, a major review of Australian quarantine – the Nairn review (Nairn et al.,1996) examined the process used in Australia (AQIS, 1991) and recommended changes designed to increase consultation while continuing to meet international obligations. Its recommendations were largely adopted (DPIE, 1997) and are now in operation (AQIS, 1998). The Nairn review advocated early consultation with stakeholders to help obtain consensus on priorities, the scope and type of risk assessment, and the risk management strategies required to ensure any proposed import does not jeopardize the importing country’s animal and plant health status or have a negative effect on its natural environment. It argued that import risk analysis should be based on consultation, science, transparency, consistency and harmonization, and be subject to external review (Nairn et al., 1996). Risk analysis should be based on science. In particular, risk assessment should be ‘essentially a scientific endeavour based on experimentation and observation’ (ANZFA, 1996, p. 2), independent of policy or political considerations. In contrast, risk management ‘involves policy decisions based on a balance of scientific, social and economic considerations’ (ANZFA, 1996, p. 2). In recognition of this, some countries (including the United States) separate the regulatory application of risk analysis by assigning official responsibility to different agencies for risk assessment (its scientific or technical component) and for risk management (its policy or political component). From a regulatory perspective, risk analysis should be consistent with both government policy and international obligations. It should take account of international standards, guidelines and recommendations so that it is harmonized as much as possible with international practice. However, a government may in some circumstances elect to use criteria that are more rigorous than international practice, and international agreements may provide for such a decision. For example, SPS article 3.5 acknowledges that to meet its appropriate level of sanitary or phytosanitary protection, a country may impose risk management strategies or measures that are more stringent than international standards, guidelines and recommendations that are scientifically justifiable and otherwise consistent with international obligations. The need for appeal through both administrative and judicial approaches was confirmed in a recent report on the use of risk analysis in a wide range of regulatory agencies in the United States (CRARM, 1997). In international trade, the WTO’s Dispute Settlement Body oversees a formal system for resolving disputes related to among other things, quarantine and trade.

98

Managing risk in policymaking

DETERMINING THE TYPE OF RISK ANALYSIS The complexity of the issue is a primary determinant of the type of risk analysis that needs to be undertaken – simple cases can often be assessed quite quickly. For example, the vast majority of import access requests are routine and can be readily approved or rejected on sound scientific grounds. Requests that might be readily approved include those for importing a commodity from a source with a similar health status to that of a source already approved for imports of the same commodity. Alternatively, an import access request involving possible or likely contamination with an agent of a disease of concern that it is known cannot be removed or inactivated by the application of current risk management strategies can be readily rejected on sound scientific grounds. In more complex cases, detailed risk assessment is required before a scientifically sound decision can be made. For example, complex import access requests in Australia are assessed by formal risk analysis panels, which appoint or contract expert working parties to complete specific components of a detailed risk analysis. Risk analysis panels comprise appropriate experts – particularly in science for conducting detailed risk assessments and considering risk management options, and in economics for examining the potential economic loss of the introduction or establishment of any diseases of concern (AQIS, 1998).

METHODS FOR RISK ASSESSMENT The fundamental steps in risk assessment are the identification of the risks or hazards of concern, the assignment of a probability of occurrence of each risk and an estimation of the consequences resulting from the occurrence of each risk. There are several published reviews of methods used in import risk assessment (for example, Kellar, 1993; MacDiarmid, 1993; OIE, 1994 and Vose, 1997 for animal health and quarantine) and such papers and individual assessments should be examined for discussion of the details of specific approaches. In summary, the initial step in quarantine import risk assessment is to determine which diseases in the country of origin of a proposed import do not occur in the importing country and are of sufficient concern to warrant measures to exclude them. The assessment establishes a scenario tree or outline of the pathway or pathways of entry and establishment of unwanted diseases that might be associated with a proposed import. In qualitative approaches, emphasis focuses on the key points in the pathway where risk management factors can be applied to eliminate (for example, by heat treatment of a product) or reduce (for example, by vaccinating or testing live animals) the risk of

Allowing for risk in setting standards

99

importing diseases of concern. In semi-quantitative approaches, numerical values (for example, the prevalence of the disease of concern) are applied at each point for which data are available. In fully quantitative approaches, such data are applied at all points of the pathway of entry and establishment. Data used in the assessment are extrapolated from information on the biology and pathology of the disease of concern in countries where it occurs. Such information may be readily available in published scientific literature, may require specific investigation with the assistance of countries where the disease occurs, or in some cases may not be available and must be obtained by additional field or laboratory studies. In the case of genetically modified organisms, there is a relative lack of such information because of the difficulty of obtaining data on the effect of releases while at the same time ensuring the protection of the environment from any potential harmful effects (Wyndham and Evans, 1998).

QUANTIFICATION In many disciplines, there has been a marked trend towards the use of more quantitative methods of risk assessment over the past decade. For most import risk assessments, data gaps ensure that fully quantitative risk analyses are the exception rather than the norm. It is only in relatively simple cases that reliable quantitative data are available for all steps in an import risk assessment (that is, for all points in the potential pathway or pathways of entry and establishment of a potential disease of concern). In addition, in complex situations with multiple possible scenarios that each have only an extremely small probability of occurrence (as is often the case in import risk assessments), the mathematics of fully quantitative assessment is problematic and not yet well defined. Such situations are assessable only by qualitative or semi-quantitative approaches even if good data are available for all points in a scenario tree. From a practical perspective, it should also be appreciated that even when they are possible, more quantitative approaches are extremely resource-intensive, requiring skilled staff, large amounts of data, sophisticated computing resources and a large investment of time. Thus although quantitative approaches to risk analysis have some application in evaluating selected import access requests, semi-quantitative and qualitative approaches are more appropriate for the vast majority of import risk analyses. With the trend to increasing use of quantitative approaches to risk assessment, there has been a tendency to consider that more quantitative approaches are necessarily ‘better’ or ‘more scientific’ than less quantitative approaches. However, the ‘perception held in some quarters that quantitative approaches are inherently “better” or “more scientific” than qualitative approaches is

100

Managing risk in policymaking

misguided – a poor quantitative risk assessment (e.g. one using poor data or using inappropriate quantitative techniques) can be quite misleading and far less scientific than a good semi-quantitative or qualitative assessment’ (Nairn et al., 1996, p. 106). Quantitative approaches to risk assessment have been criticized for a perceived lack of objectivity resulting from the use of expert judgements that allegedly reflect not only scientific knowledge but also factors such as ‘policy values and cultural values’ (Breyer, 1993; Pollack, 1996). Some commentators have expressed concern that scientific judgement involved in risk assessment is not as objective as it may seem because assessments can have a large variability and uncertainty, particularly when applied to environmental problems. For example, one study of various estimates of the potential carcinogenicity of a particular chemical showed that quantitative assessments could vary by as much as eight orders of magnitude, a variation characterized as ‘clearly a dubious basis for issuing permits, setting clean-up levels, and setting standards’ (Ginsburg, 1993). However, explicit and clear acknowledgement and discussion of assumptions and data used in risk assessment should minimize any concerns regarding lack of objectivity, and permit more careful consideration of the need for obtaining better data where necessary. The use of stochastic rather than deterministic methods can also help to describe the uncertainty in risk estimates. Deterministic assessments assign a single number (for example, an amount or a probability) to each point in a scenario tree so that assessment leads to a single value, ignoring the fact that variation is an integral component in all natural systems. In contrast, stochastic methods assign a probability distribution to each point and determine a net probability density distribution by using techniques such as Monte Carlo simulation (Cohen et al., 1996; Vose, 1996).

SENSITIVITY ANALYSIS Simple scenario trees can be analysed in a semi-quantitative or quantitative manner even where there are some gaps in data. For example, an extreme value may be assumed for data missing at a particular point (for example, that the prevalence of a pathogen in the population of origin is 100 per cent) and the simulation run. Alternatively, expert opinion can be used to provide a ‘best guess’ of the value for a particular data point (for example using the Delphi technique). Such approaches enable the analyst to conduct sensitivity analyses to determine whether or not the particular parameter for which data are not available has a major effect on the overall risk. Sensitivity analysis often demonstrates that only a few critical points have a significant effect on the overall probability, so that if good data are available for these points the analyst can be confident that the assessment is robust.

Allowing for risk in setting standards

101

Conversely, if good data are not available for these critical points, robust risk analysis is not possible until further information is available to fill these gaps. Risk analysts reaching this conclusion might encourage research providers to commission or conduct appropriate research to fill the gaps identified, use a less quantitative approach, or focus on appropriate risk management options to reduce the risk.

THE PRECAUTIONARY PRINCIPLE In some cases, a risk analysis may determine that there are significant gaps in information that need to be filled by further research before a scientifically based decision can be made on a particular issue. Analysis might also lead to recommendations that specify the gaps and define the research needed to fill them. For example, a number of submissions to the Nairn review argued that where there is significant uncertainty or where there are significant gaps in knowledge needed to conduct a risk assessment, quarantine authorities should take a conservative approach. Some submissions went further and advocated adoption of the precautionary principle (or a variant of it) in cases they deemed involved significant uncertainty, probable delayed identification or reporting of incursions, or inadequate or no means of containing, controlling or eradicating incursions. The precautionary principle has been defined in various ways but may be simply seen as the principle of adopting a conservative approach when the relevant information needed to make an informed decision is limited – the greater the uncertainty, the more conservative should be the decision. Import risk analysis provides a good example of the valid application of the precautionary principle. Although the SPS agreement does not use the term precautionary principle, article 5.7 states that ‘in cases where relevant scientific information is insufficient’ member countries may provisionally adopt ‘sanitary or phytosanitary measures on the basis of available pertinent information’. However, the SPS agreement also states that in such cases, member countries ‘shall seek to obtain the additional information necessary for a more objective assessment of risk and review the sanitary or phytosanitary measure accordingly within a reasonable period of time’. Thus member countries’ international obligations preclude the ongoing or indefinite use of the precautionary principle as grounds for not taking a decision on any import access request. (For an EC view on the precautionary principle see the Appendix.)

DECIDING THE APPROPRIATE LEVEL OF PROTECTION Different stakeholders can have widely divergent perceptions of the risk associated with a decision. This divergence is often the result of disagreement

102

Managing risk in policymaking

about the level of risk associated with a particular option, in which case consensus can be approached through improving the underlying risk assessment. However, the disagreement is often more fundamental and rests on different perceptions of what level of risk a decision maker, stakeholder or society as a whole is prepared to accept. For example, despite statements by major reviews of quarantine (Australian Senate, 1979; DPIE, 1998) and the Australian Quarantine and Inspection Service (AQIS) itself (AQIS, 1991; Wilson and Banks, 1993), a significant number of individuals and organizations believe Australia has (or should have) a ‘no risk’ quarantine policy. The expression of this view was sufficiently frequent for the 1996 Senate Report on AQIS to state that it was ‘concerned about the persistence of the view that “no risk” is a viable option for quarantine policy, despite consistent and unequivocal dismissal of this approach by previous reviews’ (Australian Senate, 1996, p. xi). Similarly, the Nairn review concluded that ‘the continued perception in some quarters that there ever has been or ever can be a “no risk” quarantine policy for any country – let alone a major agricultural trading nation such as Australia – reflects a fundamental misconception that needs to be corrected in an ongoing awareness campaign’ (Nairn et al., 1996, p. 83). Improved risk assessment can improve the accuracy of the estimate of the risk associated with a decision, but does not help in deciding what constitutes an ‘acceptable risk’ (or in SPS terms, the ‘appropriate level of protection’). Improved risk communication and agreed guidelines or standards can help in reaching consensus, but ultimately what constitutes an acceptable risk is based not on science but on judgement – whether of an individual person, stakeholder group, community or country (see Deane, Chapter 8, this volume). Some authorities attempt to determine an appropriate level of protection by comparing the risk of a particular decision option with risks in other areas (for example, comparing the potential risk of death from ingesting a contaminant with the average risk of death from other causes such as aeroplane or automobile accidents). Others have argued that with appropriate consultation, a team undertaking a risk assessment should assess risks associated with a particular risk analysis and examine appropriate risk management strategies that might be used to reduce the level of risk to one that is ‘manageable’. For example, the Nairn review concluded that for quarantine risk analysis the pertinent concept is one of ‘manageable risk’ – not ‘no risk’ (which is unachievable). It acknowledged that a certain amount of judgement is implicit in this concept, but argued that consistency of application can be achieved by reference to existing national policies and procedures, by reference to relevant international standards, guidelines and recommendations, and through the contribution of experienced risk analysts (Nairn et al., 1996).

Allowing for risk in setting standards

103

BENEFITS OF A MULTIDISCIPLINARY APPROACH Risk assessment (particularly using more quantitative approaches) is an extremely demanding, complex and resource-intensive process. It involves consideration of scientific and economic factors, often requiring the use of multidisciplinary teams. In complex risk analyses of areas such as quarantine or genetically modified organisms, risk assessment teams may need to include specialists with skills in disciplines such as communications, mathematics, statistics, computer modelling, ecology and environmental science in addition to those in risk analysis, animal or plant health, and economics. There is also a need to include specialists in law, especially in international and trade law, to ensure that the risk analysis process meets statutory obligations domestically and that it takes account of all relevant international obligations. It can be expected that future consideration of issues related to risk and trade will require detailed risk analysis using a consultative approach similar to that now used for import access requests in Australia following the Nairn review. Some of these issues will require detailed scientific risk assessments, which will tend to use more quantitative approaches if only to provide a basic sensitivity analysis and comparison of the effect of different risk management options. Many will also require detailed economic assessment of the potential effects. For example, in import access requests economic assessment may be required of both the specific cost of the potential establishment of an exotic disease (for inclusion in the import risk analysis itself) and of more general economic effects of the trade on prices and markets (for consideration in possible industry adjustment measures or other policy options), and some issues will also require detailed environmental risk assessment. Similar needs for multidisciplinary input will also occur in other areas of risk analysis related to trade (for example, trade in genetically modified organisms). The standards, guidelines and recommendations provided for sanitary and phytosanitary measures (by WTO, the SPS agreement and TBT agreement, and the use of OIE, IPPC and Codex) provide a sound framework for the application of risk analysis to international trade in animals, plants and their products. This framework, although still evolving and imperfect, offers a practical and tested approach to the complexities of evaluating risks related to international trade. It is certainly worth careful consideration in the development of similar standards, guidelines and recommendations in other areas of risk analysis related to trade. There is thus an opportunity for greater interdisciplinary collaboration to develop better improved methods for risk assessment and better standards, guidelines and recommendations for risk analysis across all areas of potential risk related to international trade.

104

Managing risk in policymaking

REFERENCES Ahl, A.S., J.A. Acree, P.S. Gipson, R.M. McDowell, L. Miller and M.D. McElvaine (1993), ‘Standardization of nomenclature for animal health risk analysis’, Revue scientifique et technique de 1’Office International des Epizooties, 12, 1045–53. ANZFA (1996), Framework for the Assessment and Management of Food-related Risks, Canberra: Australian and New Zealand Food Authority. AQIS (1991), The Application of Risk Management in Agricultural Quarantine Import Assessment: A Discussion Paper, Canberra: Australian Government Publishing Service. AQIS (1998), Handbook on the AQIS Risk Analysis Process, Canberra: Department of Primary Industries and Energy. Australian Senate (1979), The Adequacy of Quarantine, Canberra: Australian Government Publishing Service. Australian Senate (1996), Australian Quarantine and Inspection Service. Report of the Senate Rural and Regional Affairs and Transport Legislation Committee, Canberra: Senate Printing Unit, Parliament House. Beer, T. (1996), ‘Risk assessment for environmental management’, in T.W. Norton, T. Beer and S.R. Dovers (eds), Risk and Uncertainty in Environmental Management: Proceedings of the 1995 Australian Academy of’ Science, Fenner Conference on the Environment, Canberra: Centre for Resource and Environmental Studies, Australian National University, pp. 74–83. Breyer, S. (1993), Breaking the Vicious Circle: Toward Effective Risk Regulation, Cambridge, MA: Harvard University Press. Codex (1993), Risk Assessment Procedures used by the Codex Alimentarius Commission and its Subsidiary and Advisory Bodies, Geneva: Codex Alimentarius Commission, Food and Agriculture Organization. Cohen, J.T., M.A. Lampson and T.S. Bowers (1996), ‘The use of two-stage Monte Carlo simulation techniques to characterize variability and uncertainty in risk analysis’, Human and Ecological Risk Assessment, 2, 939–71. Covello, V.T. and M.W. Merkhofer (1993), Risk Assessment Methods: Approaches for Assessing Health and Environmental Risks, New York: Plenum Press. CRARM (1997), Risk Assessment and Risk Management in Regulatory DecisionMaking, Final report (Volume 2) of the Presidential/Congressional Commission on Risk Assessment and Risk Management, Washington. DPIE (1997), Australian Quarantine: A Shared Responsibility – The Government Response, Canberra: Department of Primary Industries and Energy. DPIE (1998), Australian Quarantine Requirements for the Future: Report of the Quarantine Review Committee, Canberra: Australian Government Publishing Service. Ginsburg, R. (1993), ‘Quantitative risk assessment and the illusion of safety’, New Solutions (Winter), 8–15. Hathaway, S.C. (1991), ‘Risk assessment and its application to veterinary regulatory decisions’, Surveillance, 18, 24–8. Kellar, J.A. (1993), ‘The application of risk analysis to international trade in animals and animal products’, Revue Scientifique et Technique de 1’Office International des Epizooties, 12, 1023–44. Krewski, D. and P.L. Birkwood (1987), ‘Risk assessment and risk management: a survey of recent models’, in L.B. Lave (ed.), Risk Assessment and Management, New York: Plenum Press, pp. 399–406.

Allowing for risk in setting standards

105

MacDiarmid, S.C. (1993), ‘Risk analysis and the importation of animals and animal products’, Revue Scientifique et Technique de l’Office International des Epizooties, 12, 1093–107. Nairn, M.E., P.G. Allen, A.R. Inglis and C. Tanner (1996), Australian Quarantine: A Shared Responsibility, Canberra: Department of Primary Industries and Energy. North, D.W. (1995), ‘Limitations, definitions, principles and methods of risk analysis’, Revue Scientifique et Technique de 1’Office International des Epizooties, 14, 913–23. Notermans, S. and G.C. Mead (1996), ‘Incorporation of elements of quantitative risk analysis in the HACCP system’, International Journal of Food Microbiology, 30, 157–73. Nunn, M.J. (1997), ‘Quarantine risk analysis’, Australian Journal of Agricultural and Resource Economics, 41, 541–78. OIE (1994), ‘Import risk analysis’, Update to the International Animal Health Code, Mammals, Birds and Bees, Paris: Office International des Epizooties, pp. 27–28/1 0. Pollack, R.A. (1996), ‘Regulating risks’, Journal of Economic Literature, 33, 179–91. SA/SNZ (1995), Australian/New Zealand Standard. Risk Management, AS/NZS 4360:1995, Homebush and Wellington: Standards Australia/Standards New Zealand. Vose, D.J. (1996), Quantitative Risk Analysis: a Guide to Monte Carlo Modelling, Chichester: John Wiley and Sons. Vose, D.J. (1997), ‘Risk analysis in relation to the importation and exportation of animal products’, Revue Scientifique et Technique de 1’Office International des Epizooties, 16, 17–29. Wilson, D.B. and D.J.D. Banks (1993), ‘The application of risk assessment in animal quarantine in Australia’, Revue Scientifique et Technique de 1’Office International des Epizooties, 12, 1121–33. Wyndham, A. and G. Evans (1998), ‘National biosafety legislation and trade in agricultural commodities’, BINAS News, 4, 2–6.

8.

Public perceptions, risk communication and biotechnology Christine R. Deane

INTRODUCTION Modern biotechnology is a powerful, rapidly advancing technology, with the capacity to alter our way of life. It is viewed as an important tool, with the potential to improve living standards, but also as potentially hazardous and open to abuse. Individual attitudes to biotechnology are formed by complex interactions of socioeconomic factors, personal values, education and the attitudes of the society to which each individual belongs. Advances in biotechnology are accompanied by apprehension about safety, legal, ethical, social and economic concerns. This leads to debate among the ‘stakeholders’, and gives rise to emotive campaigning. As with any technology, biotechnology is neither inherently good nor bad, it simply provides a set of tools for the genetic manipulation of organisms. It is the application of this technology that determines whether it is beneficial or harmful. How biotechnology is applied to agriculture and medicine is a question for society as a whole. Biotechnology can help to solve many problems. For example, farmers encounter increasing demands from consumers for improved quality and to reduce chemical residues. There is also a demand to achieve more sustainable agriculture, and biotechnology can make a major contribution to this. The impact of the first genetically modified crops has already been observed. The US agrichemical industry has recorded reductions in herbicide sales following the introduction of transgenic herbicide tolerant crops. Medical applications of biotechnology have paved the way for treating many serious ailments. Diagnostic tests are being developed that can identify carriers of genes linked to serious heritable diseases such as cystic fibrosis, and advances in gene therapy and tissue engineering have opened new avenues of treatment and brought hope to many sufferers. Public opinion surveys indicate a shift towards acceptance of biotechnology in many developed countries. An extensive study carried out in Canada in 1993 highlighted three distinct attitudes: 25 per cent of respondents expected more benefit than danger from biotechnology; 24 per cent expected the reverse; while 39 per cent believed it may be equally beneficial and dangerous (Espey, 1993). 106

Public perceptions, risk communication and biotechnology

107

A more recent survey of Canadian attitudes (Environics Research Group, 1997) found a high level of respondents (87 per cent) were positive about biotechnology and specifically that they agreed ‘[Canada] must pursue opportunities presented by new technologies to maintain current quality of life’. Similarly, two surveys carried out in Japan indicate a general increase in support for agricultural biotechnology between 1995 and 1998 (Hoban, 1998). Surveys of public attitudes towards biotechnology in the US throughout the 1990s found consistently positive attitudes toward agricultural biotechnology. More than half of US respondents to surveys in 1994 and 1997 believed that biotechnology already provided benefits for them, while almost three-quarters of respondents believed that they would benefit from biotechnology in the future (Hoban and Katic, 1998). Conversely, comparisons of the 1996 ‘Eurobarometer’ survey with those of 1991 and 1993 indicate that optimism in Europe about the contribution of biotechnology to improving the