non-state actors in international climate change negotiations · 2015-01-17 · diss. eth no. 21193...

diss. eth no. 21193

Non-State Actors inInternational Climate Change Negotiations

A dissertation submitted to

ETH ZURICH

for the degree of

Doctor of Sciences

presented by

CAROLA BETZOLD

Master of Philosophy, University of Cambridge

born February 19, 1984

citizen of Germany

accepted on the recommendation of

Prof. Dr. Thomas Bernauer

Prof. Dr. Vally Koubi

Prof. Dr. Jens Steffek

2013

iii

Acknowledgements

A doctoral dissertation is not the work of an individual alone. I would therefore like tothank everyone who helped me write this dissertation for their intellectual, moral, andfinancial support.

I am especially thankful to my main advisor Thomas Bernauer without whom thisresearch project would not have been feasible. I would also like to thank Vally Koubifor her support and advice over the past four years. I thank both of them for theenergy and time they have invested in me and this dissertation. Their inputs as well asencouragement and support in attending different courses, workshops, and conferenceshelped creating a stimulating academic environment and are much appreciated.

I would also like to thank Jens Steffek for his support and help, in particular towardthe end of my doctoral studies, and his willingness to serve as an examiner for thisdissertation. I am similarly grateful to David Victor for his hospitality during my researchstay at the University of California at San Diego; although short, the research staycertainly taught me a lot in many respects.

A lot of support and help came from my colleagues in Zurich and elsewhere. Specialthanks goes to all members of the International Relations research group at eth Zurich.Gabriele Spilker and Michael Bechtel provided important methodological input duringvarious stages of my research project, while Axel Michaelowa’s experience with the cli-mate change negotiations helped me design and implement my questionnaire. Manyothers helped, too; in particular, I would like to thank Anaïd Flesken for her invaluableadvice and feedback, Cornelius Klöck for the many discussions, and Theresa Tribaldosfor sharing our office.

The elements of this dissertation were presented at various workshops and conferences,including the Annual Conference of the Swiss Political Science Association in Lucerne(2012) and Zurich (2013); the Annual Meeting of the International Studies Associationin San Diego (2012) and San Francisco (2013); the Annual Conference of the MidwestPolitical Science Association in Chicago (2012 and 2013); and the Open Meeting of theGerman Political Science Association in Munich (2011). I also had the opportunity topresent my work at the PhD workshops on International Climate Policy in Freiburg(2012) and Paris (2013) as well as at the research colloquia of the nccr Democracy in

iv

Zurich. The discussants and participants of these workshops and conferences providedhelpful comments and feedback, for which I would like to thank them.

Empirically, I relied on the help of those who filled in my survey and responded to myquestions. I am grateful to all respondents for their willingness to share their knowledgeand experience with me.

Finally, this research would not have been possible without the generous funding ofthe Swiss National Science Foundation. This dissertation was written in the contextof the National Center for Competence in Research on Challenges to Democracy inthe 21st Century (nccr Democracy). The support in the framework of this specialresearch programme is greatly appreciated, as it allowed me to present my work atseveral conferences, and to attend the climate summit in Durban in 2011.

v

Contents

List of Figures vii

List of Tables ix

List of Abbreviations xi

Summary xiii

Zusammenfassung xv

1 Introduction 1

1.1 ngos in Global Environmental Governance . . . . . . . . . . . . . . . . . 31.2 When do ngos Participate in International Environmental Negotiations? . 41.3 How Do ngos Participate in International Climate Change Negotiations? 61.4 Whom Do ngos Lobby in International Climate Change Negotiations? . . 71.5 Press Briefings in International Climate Change Negotiations . . . . . . . 91.6 Concluding Remarks and Outlook . . . . . . . . . . . . . . . . . . . . . . . 11

2 Civil Society in Global Environmental Governance 15

3 Indigenous Peoples in International Environmental Negotiations:Evidence from Biodiversity and Climate Change 19

3.1 Indigenous Peoples and the Environment . . . . . . . . . . . . . . . . . . . 203.2 Indigenous Peoples and International Environmental Politics . . . . . . . . 213.3 Explaining the Difference: Analytical Framework . . . . . . . . . . . . . . 253.4 Biodiversity and Climate Change Compared . . . . . . . . . . . . . . . . . 27

3.4.1 Goals and Grievances . . . . . . . . . . . . . . . . . . . . . . . . . 273.4.2 Opportunities and Constraints . . . . . . . . . . . . . . . . . . . . 30

3.5 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

4 Business Insiders and Environmental Outsider? Advocacy Strategiesin the Climate Change Negotiations 39

4.1 Explaining Advocacy Strategies . . . . . . . . . . . . . . . . . . . . . . . . 41

vi

4.2 Data Collection and Measures of Variables . . . . . . . . . . . . . . . . . . 454.3 Business Insiders and Environmental Outsiders? . . . . . . . . . . . . . . . 47

4.3.1 Bivariate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 494.3.2 Multivariate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 51

4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

5 Responsiveness or Influence?Whom to Lobby in International Climate Change Negotiations 575.1 Whom to Lobby? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

5.1.1 Friends or Foes? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595.1.2 Strong or Weak? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

5.2 Data and Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 635.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

5.3.1 Clustered Logistic Regression . . . . . . . . . . . . . . . . . . . . . 685.3.2 International vs. National Groups . . . . . . . . . . . . . . . . . . 715.3.3 Survey Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

5.4 Discussion and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

6 The Media in International Negotiations: The Case of Climate Change 776.1 Mediatisation and press briefings in climate cops . . . . . . . . . . . . . . 806.2 Two Logics of Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 846.3 Research Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

6.3.1 Dependent Variable . . . . . . . . . . . . . . . . . . . . . . . . . . 866.3.2 Independent Variables . . . . . . . . . . . . . . . . . . . . . . . . . 866.3.3 Control Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . 876.3.4 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

6.4 Explaining Press Briefings Initiated By Governments . . . . . . . . . . . . 896.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 926.6 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Bibliography 99

vii

List of Figures

3.1 Number of ipos attending cbd and unfccc negotiations . . . . . . . . . 223.2 Geographic origin of ipos active in cbd and unfccc negotiations . . . . . 233.3 Framework for analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

4.1 Use of Advocacy Activities For All ngos, Percentages . . . . . . . . . . . 484.2 Mean Use of Advocacy by Group Types . . . . . . . . . . . . . . . . . . . 494.3 Mean Use of Advocacy by Experience . . . . . . . . . . . . . . . . . . . . 504.4 Mean Use of Advocacy by Membership . . . . . . . . . . . . . . . . . . . . 504.5 Histogram for Mean Use of Inside and Outside Advocacy . . . . . . . . . . 56

5.1 Countries contacted most frequently . . . . . . . . . . . . . . . . . . . . . 685.2 Importance of different factors for ngos’ choice of lobbying targets . . . . 74

6.1 Media attention for climate change . . . . . . . . . . . . . . . . . . . . . . 816.2 Number of press briefings by cop and actor type . . . . . . . . . . . . . . 826.3 Aggregate number of press briefings by actor type . . . . . . . . . . . . . . 826.4 Number of press briefings by most active governments . . . . . . . . . . . 83

ix

List of Tables

4.1 Descriptive Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474.2 Regression Analysis of Inside and Outside Advocacy . . . . . . . . . . . . 514.3 Regression Analysis of Degree of ‘Insiderness’ . . . . . . . . . . . . . . . . 534.4 Correlation of Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

5.1 Number of reported contacts with government delegations . . . . . . . . . 645.2 Distribution of delegations and contacts according to country-level variables 675.3 Logistic regression analysis of ngos’ choice of lobbying targets . . . . . . 695.4 Logistic regression analysis of ngos’ choice of lobbying targets for inter-

national and national groups . . . . . . . . . . . . . . . . . . . . . . . . . 73

6.1 Descriptive statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 886.2 Zero-inflated negative binomial analysis of governments’ decision to hold

press briefings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 916.3 Variables and data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . 946.4 Correlation coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 956.5 Correlation coefficients using political rights . . . . . . . . . . . . . . . . . 956.6 Estimation results using political rights. . . . . . . . . . . . . . . . . . . . 966.7 Estimation results using total CO2 emissions. . . . . . . . . . . . . . . . . 97

xi

List of Abbreviations

CO2 carbon dioxide

alba Bolivarian Alliance of the Peoples of our Americas

cacam Central Asia, Caucasus, Albania and Moldova group

cocarae Confederation of Indigenous Organisations of the EcuadorianAmazonian Region

coica Indigenous Organisations of the Amazon Basin

g77 Group of 77 and China

gfbv Society for Threatened Peoples

iwgia International Work Group for Indigenous Affairs

opec Organization of the Petroleum Exporting Countries

unccd United Nations Convention to Combat Desertification

uniproba Unissons-nous pour la Promotion des Batwa

aosis Alliance of Small Island States

cait Climate Analysis Indicators Tool

cbd United Nations Convention on Biological Diversity

cop Conference of the Parties

eu European Union

field Foundation for International Environmental Law and Develop-ment

gdp gross domestic product

igo intergovernmental organisation

iifb International Indigenous Forum on Biodiversity

ipcc Intergovernmental Panel on Climate Change

ipos indigenous peoples organisations

ldcs Least Developed Countries

xii

ngo non-governmental organisation

oecd Organisation for Economic Cooperation and Development

redd Reducing Emissions from Deforestation and Forest Degradationand the Role of Conservation, Sustainable Management of Forestsand Enhancement of Forest Carbon Stocks in Developing Coun-tries

unfccc United Nations Framework Convention on Climate Change

un United Nations

us United States

xiii

Summary

A diverse set of non-governmental organisations (ngos)—from environmental groups tobusiness association, trade unions, or indigenous peoples organisations (ipos)—participatein international climate change negotiations. Why do these diverse groups decide to en-gage in the negotiations, and how do they do so? After an introductory chapter (chapter1) and a brief overview over civil society in environmental governance (chapter 2, co-authored with Thomas Bernauer), I address these questions in four substantive chapters,based on interviews, a survey, and archival data.

Chapter 3, co-authored with Anaïd Flesken, starts with developing a framework foranalysing ngos’ decision to get involved in international decision-making, and appliesthis framework to ipos. ipos are active participants in the biodiversity regime, yetare only marginally present in climate change negotiations. How can this imbalance beexplained? A systematic comparison of the two sets of negotiations points to importantdifferences in their institutional environment, in particular in the timing and degree ofrecognition of indigenous peoples as stakeholders and thus legitimate political actors.We relate these to differences in how climate change and biodiversity affect indigenousrights, especially land rights: it is when indigenous peoples feel their right to land beingaffected by an issue—or related policies—that they become active on that issue.

I then turn to the strategic choices of ngos when attending international negotiations(chapter 4, single-authored). ngos can participate in the negotiations in different ways:they may seek to directly influence the negotiations, for instance by lobbying governmentnegotiators; or they may opt for more indirect paths, for example media contacts. Whichof these strategies do ngos pursue, and why? Using data from a survey I conductedamong ngos that are active in international climate change negotiations, I find that, whilemost groups pursue both inside and outside advocacy, they do display a preference towardone or the other strategy. This difference can most importantly be explained with theirorganisational type—environmental organisations preferring outside advocacy, businessgroups opting for an inside strategy—as well as their experience with the negotiationprocess—the more experienced an organisation, the more it will use direct advocacy.

In chapter 5 (single-authored), I focus on the interactions between ngo representativesand government negotiators. Which of the almost 200 government delegations present

xiv

at a typical climate summit do ngos lobby, and why? Do they approach influentialdelegations whose voice matters at the negotiation table? Or do they rather turn to(potentially) responsive to delegations, for instance those advancing similar positions?Using survey data, I find that most ngos contact relatively few government delegations;however, the odds of being lobbied increase significantly with both, a delegation’s degreeof responsiveness toward ngo, as well as its influence in the negotiations. Most impor-tantly, however, ngos seek out their home government: they use the annual climatesummits primarily to extend their domestic lobbying activities.

In the final chapter (chapter 6), my co-authors Thomas Bernauer, Vally Koubi and Iexamine the role of the media in international climate change negotiations. The annualclimate summits have become highly mediatised events, with hundreds of journalists at-tending. ngos and governments alike reach out to these journalists and invite them topress conferences. While it would be interesting to assess which ngos talk to the press,we have little systematic information on which ngos organise press briefings. We there-fore concentrate our analysis on press briefings organised by government representatives,and ask which governments speak to the press, and why? Do governments address themedia, and through them, the general public, because of strong normative expectationsof transparency? Or do governments rather see these press events as opportunities toincrease their leverage in the negotiations? We find moderate support for these two mo-tivations in our analysis, with other factors, such as capacity, seemingly playing a moreimportant role.

My dissertation thus links the study of interest groups with research on (international)environmental politics. Using new methodological approaches and novel data, I couldestablish important links between issue, ngo, and government characteristics and ngo

behaviour in the area of climate change—an area in which ngos play a particularly largerole.

xv

Zusammenfassung

Unterschiedlichste Nichtregierungsorganisationen (ngos)—darunter Umweltgruppen, In-dustrievereinigungen, aber auch Gewerkschaften oder Indigenenorganisationen—nehmenan den jährlich stattfindenden internationalen Klimaverhandlungen. Was sind die Beweg-gründe dieser Organisationen für ihre Teilnahme an internationalen Verhandlungen, undwelche Form hat diese? Nach einem einleitenden Kapitel (Kapitel 1) und einer kurzenÜbersicht über Zivilgesellschaft in Umweltpolitik (Kapitel 2, zusammen mit ThomasBernauer geschrieben) gehe ich auf diese zentralen Fragen in vier Kapiteln ein, wobei ichmich auf Interviews, eine Umfrage sowie Archivdaten stütze.

In Kapitel 3 entwickle ich zusammen mit meiner Ko-Autorin Anaïd Flesken zunächsteine analytische Struktur, mit deren Hilfe wir die Entscheidung von Nichtregierungsor-ganisationen, an internationalen Verhandlungen teilzunehmen, am Beispiel von Indige-nenorganisationen untersuchen. Indigenenorganisationen sind bei den Verhandlungen zuBiodiversität sehr aktiv, treten jedoch nur marginal bei den Klimaverhandlungen auf.Wie ist dieses Ungleichgewicht zu erklären? Ein systematischer Vergleich der beidenVerhandlungen zeigt wichtige institutionelle Unterschiede auf, insbesondere in Bezug aufwann und in welchem Ausmaß Indigene als Stakeholder und damit als legitime politischeAkteure anerkannt wurden. Wir erklären diese institutionellen Unterschieden mit denunterschiedlichen Folgen von Klimawandel und dem Verlust von Biodiversität für indi-genes Landrecht: Indigenenorganisationen werden dann aktiv, wenn ein Problem—oderdessen politische Lösung—Folgen für das indigene Recht auf Land hat.

In Kapitel 4 (selbstständig verfasst) wende ich mich dann dem strategischen Verhaltenvon Nichtregierungsorganisationen bei internationalen Verhandlungen zu. Nichtregierung-sorganisationen haben verschiedene Handlungsoptionen während der Verhandlungen; sokönnen sie direkt Einfluss zu nehmen suchen, indem sie beispielsweise Regierungsvertreterlobbyieren. Alternativ können sie einen indirekteren Weg wählen und zum Beispiel an diePresse treten, um von außen auf die Verhandlungen einzuwirken. Welche dieser Optio-nen wählen Nichtregierungsorganisationen, und warum? Basierend auf einer Befragungvon bei den Klimaverhandlungen aktiven Organisationen stelle ich fest, dass die meistenGruppen sowohl direkten wie auch indirekten Aktivitäten nachgehen. Nichtsdestotrotzkann man eine Präferenz für die eine oder andere Strategie feststellen. Diese Präferenz

xvi

ist hauptsächlich die Folge von Organisationstyp: Umweltorganisationen stützen sichverstärkt auf indirekte Methoden, während Industrievereinigungen mehrheitlich eine di-rekte Strategie verfolgen. Aber auch Erfahrung mit dem Verhandlungsprozess spielt eineRolle: je erfahrener eine Gruppe, desto mehr nutzt sie direkte Methoden.

Im folgenden, selbstständig verfassten, Kapitel (Kapitel 5) konzentriere ich mich aufdie Interaktion von Nichtregierungsorganisationen und Regierungsvertretern. Welche derfast 200 Regierungen, die an einem typischen Klimagipfel vertreten sind, werden von ngo-Repräsentanten lobbyiert, und warum? Wenden sich Nichtregierungsorganisationen eheran einflussreiche Delegationen, deren Stimme am Verhandlungstisch Gewicht hat? Oderkontaktieren sie vielmehr (potenziell) responsive Delegationen, zum Beispiel Delegatio-nen mit ähnlichen Positionen in den Verhandlungen? Laut meiner Umfrage gibt es nurrelativ wenige Kontakte zwischen Regierungen und Nichtregierungsorganisationen, dochdie Wahrscheinlichkeit, lobbyiert zu werden, steigt sowohl mit der Responsivität wie auchdem Einfluss einer Delegation signifikant. In erster Linie kontaktieren Nichtregierung-sorganisationen jedoch ihre eigene Regierung; auf den jährlichen Klimagipfeln setzen siesomit vorrangig ihre heimischen Lobbyingaktivitäten fort.

Im letzten Kapitel (Kapitel 6) untersuche ich mit meinen Ko-Autoren Thomas Bernauerund Vally Koubi die Rolle der Medien in den internationalen Klimaverhandlungen. Diejährlichen Klimagipfel haben sich zu stark mediatisierten Ereignissen entwickelt, an denenHunderte von Journalisten teilnehmen. Nichtregierungsorganisationen wie Regierungs-vertreter suchen den Kontakt zu diesen Journalisten und laden sie zu Pressekonferenzenein. Obwohl die Frage, welche Nichtregierungsorganisationen Pressekonferenzen organ-isieren, von Interesse ist, fehlen hierzu systematische Informationen, weswegen wir unsauf Pressekonferenzen von Regierungsvertretern konzentrieren. Insbesondere sind wir ander Frage interessiert, welche der vertretenen Regierungen tatsächlich mit der Presse inKontakt treten, und welche Beweggründe dies haben mag. Treten Regierungsvertreteraufgrund starker normativer Erwartungen durch die Medien an die Öffentlichkeit? Odersehen die Organisatoren in Pressekonferenzen vielmehr eine Möglichkeit, ihre Verhand-lungsposition zu stärken? Unsere Analyse stützt diese beiden Motivationen nur moderat;andere Faktoren wie beispielsweise Kapazität spielen offenbar eine größere Rolle.

Die vorliegende Arbeit verbindet so die Analyse von Interessengruppen mit Arbeit zu(internationaler) Umweltpolitik. Mit neuen methodischen Ansätzen konnte ich wichtigeVerbindungen zwischen Problem-, Organisations- und Regierungsmerkmalen und demVerhalten von Nichtregierungsorganisationen im Bereich internationaler Klimapolitikherstellen—ein Bereich, in dem Nichtregierungsorganisationen eine besonders große Rollespielen.

1

1 | Introduction

Over the past decades, non-governmental organisations (ngos)1 have become an integralelement of global environmental governance. In particular since the 1992 Rio Summit,multilateral negotiations on the environment have seen an unprecedented growth in ngo

participation. The climate change negotiations are a prime example of this trend: hun-dreds of different ngos—from environmental groups to business associations, from tradeunions to women groups, faith-based organisations, or farmers’ associations—attend theannual Conference of the Parties (cop) of the United Nations Framework Conventionon Climate Change (unfccc). So many ngo representatives in fact participate in these‘mega-conferences’ (Gaventa 2010) that they sometimes even outnumber government ne-gotiators.2

Unsurprisingly, the soaring ngo engagement in global environmental governance hasattracted academic interest. Numerous studies analyse why governments open up in-tergovernmental policy making to non-state actors (e.g. Bernauer and Betzold 2012;Raustiala 1997; Yamin 2001), and whether these non-state actors actually matter forpolicy making beyond the state (e.g. Albin 1999; Arts 1998; Betsill 2006; Betsill andCorell 2001; 2008; Corell and Betsill 2001) In contrast, the actual behaviour and deci-sions of ngos before and during negotiations has received little attention. Yet, if we areto understand the role and influence of ngos in international negotiations on the envi-ronment, we also need to look at their motivations to participate in these negotiations,and their activities and strategies when attending intergovernmental summits. My thesisthus contributes to the growing literature on non-state actors in global environmentalgovernance by opening the ‘black box’ of the intergovernmental policy-making processand analysing more closely the behaviour of ngos at the international level.

In the five papers that make up this cumulative dissertation, I first, jointly withThomas Bernauer, provide a brief overview over the participation of civil society inglobal environmental governance (chapter 2). In chapter 3, my co-author Anaïd Flesken

1ngos in global environmental governance are variously referred to as ‘non-state actors’, ‘civil so-ciety organisations’, etc. For simplicity, I will use the term non-governmental organisations or groupsthroughout the text.

2Compare participant data available at the unfccc website: http://unfccc.int/parties_and_observers/ngo/items/3667.php.

2

and I then look at when ngos decide to become involved in international politics, bycomparing the participation of indigenous peoples organisations (ipos) in the climatechange and biodiversity negotiations. While ipos are active and vocal actors in thelatter negotiations, they are underrepresented in climate change politics, despite beingparticularly vulnerable to climatic changes. How can this difference be explained? Moregenerally, what are factors that help or hinder ngo participation in international policymaking?

I then turn to the behaviour of ngos during the climate summits. I start withaddressing the action repertoires of ngos in chapter 4 (single-authored). Although ngos,as observers, have no official voice in the negotiations, they can pursue a broad rangeof activities to make their voice heard, including direct contacts with negotiators, sideevents and exhibits, media interviews, or demonstrations. These activities are usuallyclassified as insider or outsider advocacy, depending on whether they aim at impactingdecision making directly, or seek to create pressure from the outside, through the mediaand the public. What explains variation in the use of these two strategies? Do actionrepertoires vary with group type, that is, are environmental groups different from businessassociations, as common wisdom has it? What about other group characteristics, suchas type of membership or experience?

One important element of ngo insider advocacy is lobbying, that is, direct contactswith negotiators. However, with almost 200 parties represented at a typical climatesummit, ngos need to lobby selectively. Whom, then, do ngos lobby? Are lobbyingefforts focused on responsive delegations who are likely to take up ngo input and bringit to the negotiation table? Or, alternatively, are influential delegations whose voice isheard at the negotiation table attractive lobbying targets? I tackle these questions inchapter 5 (single-author).

Finally, jointly with Vally Koubi and Thomas Bernauer, I focus on the media inchapter 6—an important, but often overlooked, actor in international environmentalpolitics. A large number of journalists attend climate change negotiations, and ngosand governments alike actively reach out to these in press conferences. Press conferenceshave grown in frequency over time, but who uses this instrument, and why? Due to dataavailability, we focus on government-organised events, and ask whether governmentssee communication with the public through press conferences as appropriate, or whetherpress conferences rather serve as an instrument for increasing leverage in the negotiations.

In this introductory chapter, I first address the existing literature on non-state par-ticipation in environmental governance, an issue that is dealt with in chapter 2. I thenbriefly provide an overview of the four longer chapters that make up the core of mydissertation.I conclude with some caveats, as well as options for further research. Notethat the dissertation was written in the context of the project ‘Civil society–governmentinteractions in global governance’ of the Swiss National Centre for Competence in Re-

3

search Challenges to Democracy in the 21st Century,3 and focuses on the ngo side of theinteraction between civil society and governments; others in the project have looked moreat the government side of this interaction (see Bernauer and Böhmelt 2013; Bernauer andGampfer 2013; Böhmelt 2012; Böhmelt et al. 2013).

1.1 ngos in Global Environmental Governance

At least since the 1972 United Nations Conference on the Human Environment, theenvironment has made it to the top of the international agenda, as the increasing numberof multilateral agreements on diverse environmental issues, including ozone, biologicaldiversity, climate change, or desertification, show.4 While these agreements are theresults of intergovernmental decision making, non-governmental actors play a key role inthe negotiation of multilateral environmental agreements. Think of the climate changenegotiations: When representatives of the almost 200 parties convene each year, theirdiscussions are closely observed and actively shaped by thousands of ngo representatives.

Why do governments open up multilateral negotiations? And to what extent doesngo participation make a difference, both for the negotiation process and its outcome?Academic attention paralleled the unprecedented growth in the size and scope of ngo

participation in global environmental governance (e.g. Brühl 2003; Charnovitz 1997; 2003;Lipschutz 1996; Wapner 1996; Weiss and Gordenker 1996). Scholars identified differentfunctions and roles of ngos in international environmental politics, from agenda settingto implementation of policies (e.g. Albin 1999; Betsill 2006; Gemmill and Bamidele-Izu 2002; Newell 2000; Raustiala 1997; Yamin 2001). From these diverse contributions,two main arguments for why governments allow ngo participation in environmentalnegotiations arise (cf. chapter 2): On the one hand, ngos provide governments withinformation, advice, and expertise, at little or even no cost for the governments. ngosthus help governments improve the quality of decisions in a cost-efficient way. On theother hand, ngos increase the transparency and legitimacy of intergovernmental decision-making, which is often criticised as undemocratic because of the long chain of delegation.ngos hence also help to improve the quality of the decision-making process beyond thestate (e.g. Betsill and Corell 2008; Biermann and Gupta 2011; Dryzek 2012; Steffek 2008;Steffek and Ferretti 2009).

But is the inclusion of ngos in multilateral negotiations mere window dressing, or doesngo participation actually improve the negotiation process and its outcome? Althoughit is notoriously difficult to analyse influence and causally relate outcomes to specificactors and activities, many scholars have addressed the question of ngo influence (Albin1999; Arts 1998; 2001; Betsill 2002; Betsill and Corell 2001; 2008; Corell and Betsill 2001;

3See http://www.nccr-democracy.uzh.ch for further information.4On environmental agreements, see the International Environmental Agreements (iea) Database

Project at http://iea.uoregon.edu/.

4

Gulbrandsen and Andresen 2004; Michaelowa 1998; Skodvin and Andresen 2003).

The literature largely agrees that ngos matter: that they can, and do, influence inter-national environmental politics, if to differing degrees. Yet, most of the existing researchis based on qualitative case studies that examine the roles and effects of one type of organ-isations, mostly environmental groups (Duwe 2001; McMahon 1993; Pulver 2005; Rietig2011; Tjernshaugen and Lee 2004; Willetts 1996); to a lesser extent, business groups(Falkner 2009; Kolk 2001; Vormedal 2008), or these two groups comparatively (Giorgetti1998; 1999; Gough and Shackley 2001; Lisowski 2005).5 In contrast, more systematicstudies, based on large-N analyses, remain rare (but see Böhmelt 2012; Orr 2006). Atthe same time, we lack information on the incentives and perspectives of ngos: When dothey engage in international decision making? What strategies do they adopt to influencenegotiations, and why? For a comprehensive understanding of ngo influence, we alsoneed to know who is participating, as well as how these groups participate (cf. Betsill2006: 186). I address these questions in my dissertation, thus unpacking the ‘black box’of international negotiations themselves.

1.2 When do ngos Participate in International Environmental

Negotiations?

Since the first climate change summit, cop1 in Berlin in 1995, the number of accreditedngos that may attend the summits as observers has increased tremendously. As of Jan-uary 2013, over 1500 organisations are accredited, clustered in so-called ‘constituencies’,loose groups of broadly like-minded organisations.6 The environmental and business andindustry constituencies are the oldest such groups, and with the research and independentngos, the largest.7 Other constituencies are much smaller; indigenous peoples organisa-tions (ipos), for instance, account for only about two percent of all organisations.8

Why are there so few ipos active in climate change politics? Indigenous peoples aredisproportionately affected by environmental degradation, as they typically live in sensi-tive ecosystems, are closely related to, and directly depend on, the natural environmentand its resources (e.g. Feldt 2009; Toledo 2001). As a response to environmental pressure,indigenous peoples have organised and gained a voice in environmental policy makingon issues such as persistent organic pollutants (Selin and Selin 2008; Tennberg 2010) orbiological diversity (Colchester 2003; Oldham 2002). Why, then, the low engagementwith climate change politics? When do ipos, or ngos more generally, become active in

5There are some exceptions that take into account a broader variety of groups, see e.g. Kruse (2011).6See the unfccc website at http://unfccc.int/files/parties_and_observers/ngo/

application/pdf/constituency_2011_english.pdf.7These three constituencies account for 80 percent of all ngos. Note that while membership in a

constituency is not mandatory, the large majority (almost 90 percent) of ngos are affiliated with aconstituency.

8Cf. unfccc website at http://maindb.unfccc.int/public/ngo.pl, as of January 2013.

5

international policy making?

Questions of why and how groups organise have been dealt with extensively in socialmovement theory. My co-author Anaïd Flesken and I borrow from this literature indeveloping a framework for analysis. This framework has two dimensions: the ‘politicalopportunity structure’, that is, ‘consistent—but not necessarily formal or permanent—dimensions of the political environment that provide incentives for people to undertakecollective action by affecting their expectations for success or failure’ (Tarrow 1994: 85;see also Meyer and Staggenborg 1996: 1633). The main elements of the political oppor-tunity structure are organisational capacities, institutional openness, and other actors.The second dimension of our framework refers to ‘goals’ or ‘grievances’, that is, the de-gree to which indigenous communities are affected by an issue; or, more concretely, thedegree to which indigenous peoples perceive to be affected by an issue, both directly orindirectly by the political responses (cf. Kriesi 2004; Meyer 2004).

Building on interviews with representatives of ipos and experts, as well as primaryand secondary documents, we use this framework to systematically compare indigenousparticipation in the climate change and biodiversity negotiations. The comparison high-lights important differences, both in how the two environmental problems affect indige-nous communities, as well as in the institutional environment.

Both Conventions, the United Nations Convention on Biological Diversity (cbd) andthe unfccc were negotiated and signed at the Rio Summit in 1992. At this time, itwas clear that the loss of biological diversity, and the political response to slow thatloss, would directly bear on indigenous land, which contains many so-called ‘biodiver-sity hotspots’ (Sobrevila 2008; Toledo 2001). Indigenous communities therefore wereconcerned about their territorial rights and self-determination—and in response becameactive in cbd negotiations. Climate change, in contrast, has more indirect effects onindigenous livelihoods, which were recognised relatively late (Green and Raygorodetsky2010; Salick and Ross 2009). Again, it was the link to indigenous land and land rightsthat spurred ipos into action. It was only when forests were put on the agenda as apossible mitigation strategy that indigenous peoples became active, concerned about po-tential negative impacts on their rights to land and self-determination (e.g. Degawan2008; Feldt 2009).

These differing effects of biodiversity climate change policies on indigenous peoplesalso explain differences in the openness of these regimes toward indigenous participa-tion, with the biodiversity regime offering more support and opportunities for indigenousparticipation. Finally, because of the earlier mobilisation of ipos with respect to biodi-versity, they are more aware of, and experienced with, these negotiations, and, in light ofresource constraints, reluctant to extend their activities to climate change negotiations.9

9This may also explain why there is little overlap between the two negotiations; as of June 2011, therewere only six indigenous organisations that were accredited with both, the cbd and unfccc secretariats.

6

Although climate change is a pressing issue for indigenous communities, and its ef-fects closely related to loss of biological diversity, organisations representing indigenousinterests are much less active in the unfccc negotiations compared to the cbd pro-cess. While this may be surprising, given the greater political relevance of the former(cf. field 2008), a systematic comparison of the two negotiations highlights importantdifferences. In particular, the earlier recognition of biodiversity as a problem with di-rect implications for indigenous communities and specifically indigenous land rights, andthe relative institutional openness of the cbd explain why ipos focus their internationalactivities on biodiversity.

1.3 How Do ngos Participate in International Climate Change

Negotiations?

While much smaller than other constituencies, ipos have formally the same status asother ngos, that of observer organisations. As such, ngos have no official voice in thenegotiating process, yet ngos have a broad range of activities they can pursue to maketheir voice heard. Among other things, ngos lobby and contact negotiators directly,intervene in the formal debate, talk to media representatives, or participate in protestsand demonstrations.

These activities are usually classified as insider or outsider advocacy, depending onthe channel of influence: While insider activities aim at influencing the negotiations fromwithin, through direct interactions with negotiators, outsider activities seek to createpressure from the outside, by mobilising public opinion (e.g. Binderkrantz 2005; Grant2001). The question of which groups pursue which type of activity has long been anissue in research on interest groups at the domestic level (e.g. Grant 1989; Walker 1991),or more recently, the European level (e.g. Dür and Mateo 2012; 2013), yet has receivedlittle attention at the level of global negotiations.

In international environmental politics, studies adhere to the insider/outsider distinc-tion, and differentiate between insider advisory organisations and activist outsiders (e.g.Betsill 2006; Breitmeier and Rittberger 1998; Gulbrandsen and Andresen 2004). The-oretically, three explanations for such tactical specialisation are put forward: A firstexplanation concerns the potentially conflicting goals of ngo participation in multilat-eral negotiations, namely, to serve as ‘watchdogs’ providing accountability and trans-parency; and to provide input to the negotiation process as ‘deliberators’ (Steffek andFerretti 2009). Secondly, groups differ in the degree to which they have access to deci-sion makers. While it is assumed that insider advocacy is more efficient, and hence thepreferred option, lack of access may be compensated by outsider advocacy (Beyers 2004;Binderkrantz and Krøyer 2012; Kriesi et al. 2007). Finally, ngos do not only seek toinfluence policy, but also need to secure their organisational survival. How these goals

7

are balanced has effects on the advocacy strategies they adopt (Dür and Mateo 2012;2013).

Because of their more concentrated membership and economic resources, businessgroups tend to enjoy privileged access to policy makers and rely to a larger extent on aninsider strategy; in contrast, environmental organisations depend on the greater publicvisibility of outside advocacy to communicate with their far-flung membership. Yet, theenvironmental and business constituencies in the climate change negotiations are hetero-geneous. Further, there are many more groups active in the negotiations, such as theindigenous peoples’ organisations analysed in chapter 3. Beyond group type, I thereforetest explicitly whether group characteristics, notably experience and membership type,also affects ngos’ advocacy strategies.

In order to obtain data on the advocacy behaviour of ngos active in internationalclimate politics, I conducted a survey among all accredited ngos10 in 2011. I askedrespondents how often their organisation engages in a set of insider and outsider activ-ities, and then test whether responses vary systematically with group type and groupcharacteristics.

The results support a distinction between business insiders and environmental out-siders. Business groups engage in significantly more inside and significantly less outsideadvocacy than environmental and other ngos. Environmental organisations, on the otherhand, are more likely to adopt an outsider strategy. Over and above group type, I findthat experience also has a significant effect on the choice of advocacy strategy. Across alltypes, the more experienced an ngo, the more it turns to inside advocacy, at the expenseof outside advocacy. Membership, in contrast to my expectations, does not affect advo-cacy behaviour. I do not find that organisations whose members are mainly individuals,engage in more outside advocacy, as I would have expected based on the literature.

ngos are admitted to intergovernmental negotiations because they are thought toimprove the quality of decision-making by providing information and expertise, as wellas ideas and views from a wide range of societal actors. If, however, some of these viewsand ideas are heard more easily than others, the effectiveness of non-state participationin the negotiations may be questioned.

1.4 Whom Do ngos Lobby in International Climate Change

Negotiations?

Some group differences notwithstanding, certain activities are widely used by all ngos.The survey showed that direct contacts with negotiators play an important role for ngosacross the board: over 80 percent of the organisations in the sample indicated that

10Universities were excluded from the sample because they arguably attend climate change meetingsmainly for research purposes.

8

they contact negotiators directly. However, almost 200 governments are present at atypical climate change summit—ngos thus can only selectively establish contacts withgovernment representatives. Whom, then, do ngos lobby?

While scant attention has, to date, been paid to the question of whom ngos contact ininternational negotiations, it has attracted considerable interest at the domestic level (e.g.Baumgartner et al. 2009; Hall and Deardorff 2006; Hojnacki and Kimball 1998). Buildingon this literature, I argue that, from a theoretical viewpoint, two factors play a role for anngo to have an impact on the negotiation process and its outcome. With the negotiationsbeing intergovernmental in nature, ngos have no formal place at the negotiating table,but rely on government delegations to bring their views into the negotiations. Yet forthe process and outcome to reflect the ngo’s views and positions, it is not enough tohave a few delegations bring ideas and information into the debate; other, non-targeted,delegations, also need to take up these ideas and information. Lobbying hence has twodimensions: ngos need to ensure that their preferences are brought into the debate, butalso need to ensure that their input is supported by other delegations. In light of thisprocess, ngos can focus their lobbying on responsive delegations that can be expectedto insert ngo input into the negotiations; or they may address influential delegationswhose voice is heard in the negotiations.

Who, then, among the government delegations, is responsive, and who is influential?Responsiveness, I argue, hinges on two elements, relative position and regime type. Onthe substantive side, if an ngo advances positions similar to those of a certain governmentdelegation, this delegation should be relatively open to arguments and information by thengo as it can then use that input to make its case stronger in the negotiations (see Halland Deardorff 2006). On the procedural side, responsiveness is related to regime type,whereby delegations from democratic countries should be more willing to engage withngos. This is because ngos communicate voter preferences—on which policy makers indemocratic systems depend (see e.g. Keck and Sikkink 1998; Yamin 2001).

Influence, in turn, depends on (issue-specific) power resources: financial capacityand—in the case of climate context—greenhouse gas emissions. Any climate agreementwill only address the problem effectively if large greenhouse gas emitters cooperate, whilefinancial resources provide means of pressure. Influence, however, can also be obtainedfrom the negotiation structure. Heads of coalitions or negotiation bodies have more op-portunities to structure and impact discussions, as they coordinate positions internallyand are often included in last-minute bargaining (Depledge 2005; Lang 1989; Tallberg2010).

Whom, then, do ngos lobby in practice? I use the survey I conducted in 2011 toempirically assess the importance of the above factors in ngo-government interactionsat the climate change summits. The questionnaire asks respondents which government(s),if any, their organisation contacted. I pair each respondent with the almost 200 parties

9

to the unfccc; the unit of analysis is thus the ngo-government pair, and the dependentvariable a dichotomous variable that takes a value of 1 if ngo i indicated it contactedgovernment j at the last climate change summit or the previous year. I can then useregression analysis to identify whether a delegation’s degree of influence or responsivenessaffects its likelihood of being lobbied.

The empirical analysis shows clearly that ngos selectively contact delegations. Theodds of being lobbied increases significantly with a delegation’s degree of responsiveness,and even more so, its influence. In particular, delegations from democratic countries—but not delegations advancing similar positions—are contacted more often. ngos alsostrategically approach delegations from poor countries; possibly, this indicates that poorcountries, due to their lower negotiation capacity, depend more on external support andadvice, and are hence more responsive to ngo input. Similarly, ngos seek out largegreenhouse gas emitters, while shying away from small emitters. Finally, the presidentof the climate summit is also a very attractive lobbying target. First and foremost, how-ever, ngos simply extend their domestic activities to the international level by primarilylobbying their home government. Although the climate summits gather all the world’sgovernments, ngos do not seem to make full use of the opportunities such a gather-ing affords them, but rather continue to focus their lobbying on their respective homegovernment.

ngos have only limited resources, and therefore need to contact government dele-gations selectively. Theoretically, ngos can focus their lobbying efforts at responsivedelegations that can be expected to bring ngo input to the negotiation table; or theymay address influential delegations whose voice is heard at the negotiation table. Em-pirically, my analysis indicates that both factors, responsiveness and influence, matter.Most importantly, however, ngos simply extend their domestic activities to the cop

by contacting their home delegation, that is, the delegation(s) representing the country(countries) in which the ngo has its headquarters or subsidiary offices.

1.5 Press Briefings in International Climate Change Negotiations

Another important, if often overlooked, actor in global environmental negotiations is themedia. The media provide a central link between intergovernmental policy making andthe general public, and a key venue for media representatives to gather information ispress briefings that take place throughout the two weeks of climate change negotiations.At these press conferences, or press briefings, the organiser(s)—a government delegation,an ngo, an intergovernmental organisation (igo), or a combination of these actors—communicate information about their positions, developments in the negotiations, orother issues, and respond to questions by journalists.

Press conferences have become increasingly frequent; as the number of journalists that

10

attend the annual climate summits is growing, so is the effort of non-governmental andgovernmental delegates to communicate with them. To some extent, this developmentis part of a general growth of the climate change negotiations, which have turned into‘environmental mega-conferences’ (Gaventa 2010). Yet, there is clear variation in theoccurrence of press conferences, both over time, and, more importantly, across actors.

While an analysis of press briefings organised by ngos would be interesting, thereis little systematic information about the ngos that organise press briefings. Due tothe lack of comparable data on ngos, such as their size, membership, or resource base,we concentrate our analysis on press briefings initiated by governments. Who amongthe almost 200 parties talks to the press in press conferences, and why? Theoretically,we argue that government delegations have two primary motivations to invite the pressto a briefing: a logic of appropriateness, and a logic of consequence (cf. March andOlsen 1996; 2004). According to the former logic, governments should hold such eventsbecause of normative expectations of transparency and democratic legitimacy, wherebypress conferences are seen as appropriate and ‘the right thing’ to do; alternatively, underthe latter logic, press conferences serve to signal resolve and positions toward othergovernments and thus to increase leverage.

If the former logic is driving government behaviour, we should see democracies organ-ising more press events, since democratic polities arguably care more about questions oftransparency and legitimacy. If, in contrast, governments are motivated by a logic ofconsequence, we should see governments with extreme positions hold press conferences,both pushers and laggards, as these stand to gain the most from moving the negotiationoutcome toward their ideal point. We use a climate risk indicator and greenhouse gasemissions as proxies for pusher and laggard positions, respectively, where countries moreat risk from the negative effects of a changing climate should have a more ambitiousstance in the negotiations, while countries with high levels of greenhouse gas emissionsstand to loose economically from strict policies and should therefore hold more laggardpositions.

We use a zero-inflated negative binomial regression analysis to empirically relate cer-tain country characteristics to governments’ inclination to organise press briefings. Ourfindings provide only moderate support for the two logics of action, as we find no robustand significant relationship between democracy, CO2 emissions, or climate risk and thelikelihood of organising press conferences. It seems that other factors, such as capacity,the relevance of the climate summit, or a country’s function in the negotiations can betterexplain which governments inititate press briefings, and which do not.

11

1.6 Concluding Remarks and Outlook

Climate change is one of the key challenges of the 21st century. Since 1995, the yearin which the United Framework Convention on Climate Change entered into force, theannual climate summits have become the prime venue to address that challenge. Thesesummits have turned into real mega-conferences that are attended not only by the al-most 200 governments party to the Convention, but also by hundreds of ngos—fromenvironmental groups to business associations to indigenous peoples organisations toyouth groups, farmers, or trade unions.

The very strong participation of non-state actors in the climate change negotiationshas unsurprisingly attracted considerable academic interest, and produced a considerablebody of literature. So far, however, research has focused on the government side of thecivil society–government nexus, examining the reasons for allowing ngo participation,and assessing its effects. The perspective of ngos, in contrast, has received less atten-tion. In this dissertation, I hence look more closely at the strategic decisions of thesediverse groups: when and how do they participate in multilateral environmental negoti-ations? I thereby bring insights and approaches from related, but thus far unconnectedresearch strands to the study of civil society in international environmental politics: so-cial movement research and, more importantly, research on interest groups in domesticpolitics.

The contribution of this dissertation is thus threefold: On a theoretical level, it appliesinsights from research on domestic interest groups to ngos in climate change negotia-tions, linking their behaviour to characteristics of the ngo itself, the governments withwhich they interact, and the institutional environment in which they work. On a method-ological level, it complements the—largely case study-based—literature on civil societyin environmental governance with a quantitative approach, based on a large-N survey ofa broad variety of ngos. On an empirical level, it goes beyond the focus on environ-mental versus business groups, and takes into account the wide range of ngos that isactive in international climate change negotiations, while also shifting attention towardthe role of the media, an actor that has so far received very little attention in studies ofenvironmental governance.

Some caveats, however, apply, which at the same time open avenues for further re-search. First, research on domestic interest groups has shown that, beyond factors atthe group and the institutional levels, issue-level factors, such as salience or degree ofpoliticisation, play a role in explaining interest group behaviour (e.g. Mahoney 2008).Accordingly, the observations made in this dissertation may not easily apply to otherissue areas, as climate change differs, in many respects, from other (environmental) issueareas. Future research should thus take this project forward and analyse ngo behaviourcomparatively across different issues, both environmental and non-environmental. A

12

questionnaire similar to the one used here could be administered to ngos active in,for instance, the international negotiations on biodiversity; desertification; or trade inendangered species, but also to organisations active in areas such as human rights ordevelopment. A comparative analysis of ngos across issue areas can thus shed lighton the role of group, institutional, and issue-level characteristics in explaining advocacybehaviour.

At the same time, the climate agenda by now is quite encompassing. In this disserta-tion, I analyse ngo participation in climate change broadly, and do not look at potentialdifferences in behaviour under specific agenda items; comparisons across agenda items,however, if challenging methodologically, promise to improve our understanding of ngo

behaviour. Case studies may here be a helpful complement to the survey method; de-tailed analyses of ngo campaigns and activities on specific issues, such as forestry orblue carbon, could similarly shed light on the role of issue characteristics in explainingngo advocacy.

Second, in my analysis of advocacy strategies, I distinguish between environmental,business and other groups. While the size of some constituencies, both in the sample andin the population, make a remainder category necessary, more detailed work on specificconstituencies, such as the women and gender, trade union, or farmer constituencies, aswell as on groups that have so far no official constituency, for instance faith-based groups,would improve our understanding of ngo advocacy in climate politics (cf. Muñoz Cabré2011). Since I find, in line with other studies, that much variation in advocacy behaviourcan be explained by a simple distinction between environmental and business groups,a shift in focus toward ‘non-traditional’ types of organisations could shed light on whygroup type plays such an important role: what is it that makes business associationsbehave so differently?

Third, while the survey method allows me to cover a variety of groups, the amount ofinformation that a brief online survey, carried out anonymously, can elicit is limited. Itherefore lack information on the exact nature of contacts with governments: how oftenwere negotiators contacted? What was the content and nature of that contact? Whatdo ngos expect from their interactions with governments? Similarly, I have no dataon the conditions under which groups pursue certain advocacy activities, such as whenthey address the media, or on which occasions they organise a press conference. Similarcaveats apply to our analysis of press briefings, where our data is exclusively drawn fromthe official programmes. We hence lack information on ngos’ and governments’ expec-tations vis-à-vis the media; the content of these briefings; or the journalists that attendthese events. Again, case studies based on participant observation, interviews, docu-ment analysis and/or an analysis of webcasts of these press conferences, where available,can remedy, at least to a certain extent, the shortcomings of the data used here. Casestudies are also better able to uncover causal mechanisms underlying the correlational

13

relationships that my quantitative approach has identified.

While the PhD project thus adds to our understanding of civil society–governmentinteractions in international climate politics, it raises further questions and opens avenuesfor further research.

15

2 | Civil Society in Global Environ-mental Governance

Thomas Bernauer and Carola Betzold1

Non-governmental organisations play an increasingly important role in theformation and implementation of environmental policies and institutions.The growing involvement of non-state actors in environmental governanceis generally welcomed for two reasons: civil society presumably helps govern-ments reach more effective and democratic agreements by providing informa-tion and legitimacy. Yet, there are reasons to doubt civil society’s capacity tofulfill these two functions. Many non-governmental organisations themselveslack democratic legitimacy; and weak international agreements are often theresult not of lack of information, but of lack of political will. While non-stateactors clearly have an important role to play in global environmental gover-nance, this article calls for more research to identify when and how non-stateactors indeed contribute to more effective and democratically more legitimategovernance.

Global environmental diplomacy has intensified remarkably since the 1972 UnitedNations Conference on the Human Environment in Stockholm. Both the impressiveincrease in the number of new international agreements2 and the number of meetings andparticipant numbers per meeting are testimony to this trend. Even though governmentshave remained firmly in the ‘driver’s seat,’ the number and diversity of ‘passengers’has increased, to keep with the metaphor. Global governance has become less statecentric, as civil society3 seems to be playing an increasingly important role, both in theimplementation of global environmental policy and in the formation of new policies andinstitutions.

1This chapter was published in 2012 as an invited editorial in The Journal of Environment & Devel-opment, 21(1): 62-66.

2See http://iea.uoregon.edu/page.php?file=home.htm&query=static.3We subscribe to a broad definition of civil society in this article. This definition follows the accred-

itation criteria for observers under the United Nations Framework Convention on Climate Change. Forexample, it includes not only ngos but also business associations (but not private firms) and researchinstitutions.

16

From the perspective of governments, it may be more cost-efficient to outsource theimplementation of some policies to nongovernmental organisations (ngos). The lattermay, arguably, sometimes be in a better position to deliver public goods, because theyare present on the ground and have considerable experience and specialised expertise thatmay be more costly to obtain from within the government apparatus. In line with theneoliberal ideal of a lean state, governments have thus funneled increasing amounts offunding to civil society groups and the private sector. These actors have in fact becomepublic service providers in some areas of environmental policy implementation, first inbiodiversity conservation and more recently for generating carbon-emission reductions inreturn for certified carbon credits.

States also seem to turn toward ngos during the policy formation stage. The climatechange negotiations are, arguably, the most spectacular example, with currently morethan 1,500 accredited nongovernmental observer organisations of various types (e.g., en-vironmental ngos, business associations, academic institutions).

Moreover, civil society representatives are, at an increasing rate, also formally in-cluded as members of national delegations to climate change negotiations. Their de factoinfluence with respect to national bargaining positions and international negotiations dif-fers strongly from delegation to delegation (or country to country). However, the merefact that civil society actors are, in large numbers, formally granted a ‘seat at the table’and thus become country and nongovernmental representatives in one person is aston-ishing, because it is almost unthinkable in other important global policy arenas, such asinstitutions governing global trade, monetary and financial relations, or security.

The increasing role of civil society in global environmental diplomacy is often explainedwith two arguments:

– Civil society representatives provide valuable information and expertise to govern-ments and thus help them reach ‘better,’ that is, more effective, agreements. Thisinformation provision role becomes particularly important when governments facebudgetary constraints.

– They provide legitimacy to intergovernmental negotiations and thus mitigate the‘democratic deficit’ in global policy making, which takes place far away from do-mestic political arenas and the national demos (see, for example Betsill and Corell2008; Biermann and Gupta 2011; Dryzek 2012; Steffek and Ferretti 2009).

If these are indeed the reasons why governments seek civil society involvement and civilsociety agrees to become involved in global environmental policy making, we should wel-come the trend toward more civil society participation. More civil society participationwill then, presumably, result in more effective and more legitimate global environmentalagreements.

Let’s take a step back, however, and ask whether there is sufficient empirical evidence

17

for these assumptions or claims. To what extent can civil society fulfill the two functionsof enhancing effectiveness in problem solving and enhancing democratic legitimacy? Arethere reasons to question the capacity of civil society to live up to these expectations?The extensive scientific literature on interest (lobby) groups at the domestic level suggeststhat we should remain cautious. This literature, which is based on a long tradition oftheoretical and empirical research in the social sciences, evaluates interest group partic-ipation in politics rather critically, notably in light of potential capture of public policyby private interests at the expense of the public good (e.g. Baumgartner et al. 2009;Baumgartner and Leech 2001; Olson 1982).

Two types of criticism of civil society involvement in politics are often voiced: Onerelates to problem solving and the other to legitimacy. Such criticism has been directedprimarily at interest groups within national political systems. However, at least someof it is likely to apply also to the presumably beneficial contributions of civil society toglobal environmental governance efforts.

As to the problem solving part, existing research offers episodic, but not systematicand strong empirical evidence that more civil society participation has contributed tomore effective agreements. Although it is plausible to argue that civil society can helpreduce information deficits and thus facilitate agreement, the problem could also be toomuch information rather than too little. In other words, slow progress in solving many ofthe pressing environmental problems may not stem from a lack of information and know-how, or from insufficient involvement of civil society, but rather from (well-informed)government preferences that stand in the way of effective collective action.

As to the legitimacy part, civil society is often said to lack legitimacy and accountabil-ity. Many civil society representatives claim to speak for the public at large, but neitherthe positions nor the actions of these actors emanate from that public. Decisions aretaken by a small number of persons who may or may not be elected based on standardsthat are benchmarks for liberal, representative democracies. The general public—andoften not even the membership of civil society organisations—usually has no possibilityto hold the leadership of civil society groups accountable for their action or inaction.Also, it might not care either. Empirically, there is no robust and generalizable evidencefor whether the public (voters) views negotiations and their outcomes as more legitimateand thus acceptable when civil society is involved to a greater extent. In other words, wesimply do not (yet) know whether civil society participation can promote public supportfor global environmental policies.

What does this mean for global environmental governance? Should we adopt thecautious stance of the literature on domestic interest group politics and plead for areturn to more state centric or perhaps corporatist models of governance? We would notgo that far. Civil society clearly has a role to play in global environmental governance,and probably an important role for that matter. However, more is not automatically

18

better, even though civil society participation may have evolved into a necessary, but nota sufficient condition for effective and legitimate global environmental governance.

Social sciences research on global environmental governance has, thus far, concentratedon selectively describing civil society involvement and, in doing so, has often used anarguing-by-example approach. Not surprisingly then, the general tone in this literatureis very sympathetic to civil society. However, the conclusions are likely to suffer fromselection bias, in the sense of being based on empirical analysis of cases where civil societyappears to have contributed to problem solving.

Future research in this area needs to move in a direction that generates larger-scalecomparisons that pay equal attention to potentially positive, negative, or irrelevant im-plications of civil society involvement. Such research could also have policy implications.When interviewing national delegations to the climate summit in Durban in December2011, the authors of this essay noted that many governments appear to be reconsideringthe role of civil society in global climate negotiations. Notably, they are reconsidering thepros and cons of formally including civil society representatives in national delegations, orinvolving them in other forms. Improved empirical evidence on informational/know-howand legitimacy effects of civil society involvement could be helpful in this respect.

19

3 | Indigenous Peoples in InternationalEnvironmental Negotiations:Evidence from Biodiversity and Cli-mate Change

Carola Betzold and Anaïd Flesken1

Indigenous peoples are among the populations most vulnerable to the ad-verse effects of environmental problems like the loss of biodiversity or climatechange. Although indigenous peoples organisations (ipos ) participate in in-ternational environmental negotiations, participation differs starkly betweenissues: ipos are well represented in the negotiations under the Conventionon Biodiversity (cbd), yet remain sidelined in the negotiations under theUnited Nations Framework Convention on Climate Change (unfccc), evenif indigenous peoples have voiced concerns about climate change and climatepolitics in international fora. How can this difference be explained?

Drawing on social movement theory, we build a framework that allows usto compare indigenous mobilisation along two dimensions, the perceptionof being affected by an issue and/or related policies as well as institutionalopportunities and constraints. Our analysis points to important differencesbetween the cbd and unfccc regimes in both respects. In the cbd regime,indigenous peoples were recognised as legitimate stakeholders from the verybeginning, as measures to protect biological diversity directly impacts indige-nous land and land rights. In contrast, climate change was only relativelyrecently discovered as an indigenous problem, when forests were discussed aspart of mitigation.

1This chapter has been accepted for publication in an edited volume that is forthcoming in Septem-ber 2013: Thoko Kaime (ed.), International Climate Change Law and Policy: Cultural Legitimacy inAdaptation and Mitigation. Abingdon: Routledge.

20

3.1 Indigenous Peoples and the Environment

About 300 million people belong to one of the roughly 5,000 indigenous peoples world-wide. While their ecological footprint is very small, indigenous peoples are disproportion-ately affected by environmental problems, for they tend to live in sensitive ecosystems, areclosely related to, and directly depend on, the natural environment and its resources (e.g.Toledo 2001). Environmental degradation hence compounds the already difficult situa-tion of indigenous peoples, who typically suffer from problems such as poverty, marginali-sation, discrimination, and exclusion from policy-making processes (Macchi 2008; Nilsson2008).

In response to these pressures, indigenous peoples have over the past decades organisedthemselves at the domestic, regional, and international levels, with considerable success.2

Indigenous peoples organisations (ipos)3 now routinely participate in international pol-icymaking processes, in particular with regard to environmental issues (Rossbach deOlmos 2004; Selin and Selin 2008; Tennberg 2010; see also Morin and Saladin d’Anglure1995). Yet participation varies greatly between issues. Whereas ipos have a strong andvocal presence in the negotiations under the Convention on Biological Diversity (cbd),they are virtually absent from the international debate under the cbd’s sister agree-ment, the United Nations Framework Convention on Climate Change (unfccc).4 Howcan this variation in indigenous involvement be explained? Why do ipos focus theiractivities on the cbd negotiations, seemingly reluctant to get involved in the unfccc

negotiations—even if climate change has been recognised as affecting indigenous peoplesdisproportionately?5

Indigenous peoples as actors in international environmental politics have receivedscant attention;6 if mentioned at all, they are merely described as ‘passive and helplessvictims’ (Salick and Ross 2009: 137). We seek to remedy this unbalanced perceptionby focusing on the agency of indigenous organisations. In this paper, we concentrate onthe institutional framework in which indigenous agents act. First, we give an overviewof indigenous peoples’ participation in international environmental politics. Drawingon social movement theory, we then develop a theoretical framework that allows us to

2Key achievements are, among other things, the two International Decades of the World’s Indige-nous People (1995-2004 and 2005-2015), the establishment of the United Nations Permanent Forum onIndigenous Issues in 2000, and the 2007 United Nations Declaration on the Rights of Indigenous Peoples.

3Note that we use ‘indigenous peoples organisations’ or ‘indigenous organisations’ to refer to allorganisations that declare to defend indigenous interests, regardless of whether their members are in-digenous or not.

4Besides the cbd and the unfccc, the United Nations Convention to Combat Desertification(unccd) was agreed on at the Rio Earth Summit in 1992. While a comparison of indigenous par-ticipation of all three so-called Rio Conventions would be very interesting, it is beyond the scope of thisarticle to also consider the unccd.

5For a discussion of indigenous peoples and climate change, see e.g. Salick and Ross (2009).6For notable exceptions, see Koivovura and Heinämäki (2006); Rossbach de Olmos (2004); Schroeder

(2010); Selin and Selin (2008); Tennberg (2010).

21

systematically compare environmental negotiations on biodiversity and climate change,and thus to highlight the conditions under which indigenous organisations become activeon particular issues and in specific policy making venues. Lastly, we offer some concludingremarks.

3.2 Indigenous Peoples and International Environmental Politics

While there exists no universally accepted definition of indigenous peoples (see e.g. Colch-ester 2003; Sanders 1999), United Nations (un) special rapporteur José Martínez Cobo’sworking definition is widely used. It emphasises historical continuity with pre-colonialsocieties, as well as the right to self-identification (Cobo 1986). The un builds on thisworking definition and confirms the right of indigenous peoples to determine their iden-tity (unfccc Secretariat 2004). We subscribe to the notion of self-identification andconsider those organisations to be indigenous that declare themselves to be indigenous.In addition, we also consider organisations that explicitly state a concern with indige-nous peoples and indigenous affairs but whose members are not necessarily identifyingthemselves as indigenous.

The lack of a comprehensive definition points to the heterogeneity of indigenous peo-ples. Indigenous communities inhabit all biomes of the world, from Arctic ice to tropicalrainforests, from high-altitude mountains to low-lying islands. When measured by spo-ken language, indigenous peoples account for 80 to 90 percent of the world’s culturaldiversity (Toledo 2001). In spite of this diversity, indigenous peoples worldwide sharea close relationship with, and direct dependence on, their physical environment. Theytherefore have a natural interest in environmental policy making, and indeed gaineda voice in negotiations on issues such as persistent organic pollutants (Selin and Selin2008; Tennberg 2010). In particular, Arctic indigenous organisations have been at theforefront of political activism, both at the national and international level (Koivovuraand Heinämäki 2006; Lindroth 2006; Selin and Selin 2008).

Since international policy-making is of an intergovernmental nature, indigenous peo-ples can only participate in these negotiations as observers.7 They hence need to formnon-governmental organisations (ngos) and then have the same status as environmen-tal ngos or business ngos, although ipos are arguably of a different kind, given theterritorial identity of indigenous peoples (see Yamin and Depledge 2004: 55). Indeed,indigenous peoples have gained increasing recognition of their collective rights in inter-national law. The United Nations Declaration on the Rights of Indigenous Peoples, forexample, recognises their right to self-determination, ‘by virtue of which they [...] freelypursue their economic, social and cultural development’ (United Nations General Assem-

7Note that our discussion only considers organisations that send their own delegations to thesenegotiations. We do not take into account that representatives of indigenous peoples may also beincluded in governmental delegations.

22

bly 2007). However, this has, so far, only allowed indigenous peoples self-representationat the national, not the international, level (Koivovura and Heinämäki 2006; Lindroth2006).

ipos have sought observer status with both, the negotiations on biological diversityand those on climate change. Indigenous involvement in the cbd, however, is starklymore pronounced, as a comparison of indigenous participation at the Conferences of theParties (cops), the main decision making bodies of these two conventions, shows (seefigure 3.1).

Figure 3.1: Number of ipos attending cbd and unfccc negotiations

In the unfccc negotiations, indigenous organisations make up only a tiny proportionof all admitted observer organisations. As of November 2011, 30 organisations coordinatetheir work as indigenous peoples organisations. Seven further organisations have a strongindigenous focus, such as the Climate Alliance, but do not categorise themselves as ipos.Out of a total of 1508 accredited ngos, then, only 37 can be counted as indigenous—lessthan 2.5 percent.8 In the biodiversity negotiations, in contrast, a total of 328 indigenousorganisations have attended at least one cop.9 On average, over 50 ipos attend a cbd

cop, whereas the average number of ipo delegations at a unfccc cop is only six.Interestingly, there is very little overlap between the two sets of negotiations: only sixindigenous organisations are accredited with both processes.10

In order to obtain more information on the organisations, we sorted all indigenousorganisations by their geographic origin, as shown in figure 3.2. The graph indicates that

8See list of admitted ngos, http://maindb.unfccc.int/public/ngo.pl, as of November 1st, 2011.9Unfortunately, the cbd secretariat does not publish a list of all admitted ngos as the unfccc

secretariat does. Figures are taken from lists of participants and official cop reports, which are allavailable at http://www.cbd.int/ and upon request from the cbd secretariat.

10These organisations are: the Coordinator of Indigenous Organisations of the Amazon Basin (coica),the Forest Peoples Programme, the International Alliance of the Indigenous-Tribal People of the TropicalForest, the International Work Group for Indigenous Affairs (iwgia), the Society for Threatened Peoples(gfbv) as well as the Inuit Circumpolar Conference.

23

American organisations dominate in both negotiations. In particular, Latin Americangroups represent a large majority in the cbd negotiations, where almost half of all ipos(48 percent, 157 organisations) are from Latin America. The remaining groups are rela-tively evenly spread over the other continents. Note that the 5 percent of organisationsfrom Europe are mainly international ngos like the International Work Group for In-digenous Affairs (iwgia)11 that have their headquarters in Europe. For 11 organisations(3 percent), it was impossible to obtain information about their origin.12 The picture forthe unfccc process is relatively similar. Although Latin American groups do not repre-sent such a large majority as in the cbd negotiations, organisations from the Americastogether make up almost two thirds of all organisations (24 organisations, 60 percent),of which 40 percent (16 organisations) are North American. The remaining groups arespread relatively evenly among Asia, Africa, and Europe, with one group from Oceania.

Figure 3.2: Geographic origin of ipos active in cbd and unfccc negotiations

Small numbers, however, need not be equated with little influence, and vice versa. Anassessment of indigenous participation in the climate and biodiversity negotiations alsoneeds to take into account the degree of activism and vigour within the negotiations.An overview over the few studies that analyse indigenous participation in internationalenvironmental negotiations confirms the quantitative picture laid out above: Indigenouspeoples are an integral and fairly visible part of the biodiversity negotiations, yet tendto be overlooked in the climate change process.

The unfccc Secretariat, while admitting small delegation sizes for indigenous organ-isations, describes their participation as ‘consistent and vigorous’ (unfccc Secretariat2004). For indigenous peoples of the Arctic and Pacific regions, Crump (2008: 11) simi-larly argues for vigorous participation ‘in a number of international negotiating processes,

11See the organisation’s website at http://www.iwgia.org/sw617.asp.12The cbd does not publish detailed lists of participants for all cops, but provides a list of observer

group names as part of the cop report. Some organisations could not be identified further from thename as it appeared in the report.

24

including the unfccc’. In a 2008 report, the Foundation for International Environmen-tal Law and Development (field) confirms this view by highlighting that ‘IndigenousPeoples play an active role at unfccc sessions’, although the report at the same time de-nounces the exclusion of indigenous peoples ‘from decision-making processes that directlyaffect their communities’ (field 2008: 12; 3).

Others share this more cautious evaluation of indigenous participation in the climatedebate. Schroeder (2010: 328) remarks that ‘[i]ndigenous peoples have been moderatelyactive relative to business, ngo or local government constituencies’, their participationbeing ‘constrained by weaker legal standing, lack of support for their participation bytheir national governments and lack of resources and capacity of engagement in thesetypes of events’. Degawan (2008: 59) even argues that indigenous peoples are purpose-fully excluded: ‘Despite years of trying to influence the discussions, indigenous peoplesremain at the sidelines of the official meetings. Discussions on the issue seem to be de-liberately designed to exclude indigenous peoples’, being overly technical and scientific(see also Green and Raygorodetsky 2010; Mihlar 2008; Nilsson 2008).

In the biodiversity negotiations, indigenous organisations also face severe challenges.Mauro and Hardison (2000: 1265) write that ‘indigenous representatives have been ableto make interventions, and participate in some negotiating sessions, but this has beenlimited and the final decisions rest with the Parties’. Further, ‘[m]any Parties have actedto deny or circumscribe these rights [rights of indigenous peoples to lands, languages,religions, and cultures] in the cbd.’

Nonetheless, indigenous organisations are much more visible participants in the biodi-versity negotiations, in particular since the establishment of the International IndigenousForum on Biodiversity (iifb) at cop3 in 1996. The above-mentioned field report,for instance, argues that indigenous and local communities, despite their lack of formaldecision-making powers, ‘play a very visible and often influential role at cbd meetings’(field 2008). Oldham (2002: 39) similarly ascribes ‘increasing prominence’ to the iifb

on a number of levels, at least since cop5, which, according to him, reflects ‘a growingopenness on the part of both the Parties and the Secretariat to engage with indigenouspeoples’.

How can these pronounced differences in ipo participation between the unfccc andcbd negotiation be explained? Drawing on social movement theory and its concept ofpolitical opportunity structure, we build a theoretical framework that helps us to eluci-date the institutional framework in which ipos act and thus to compare systematicallyindigenous participation in the cbd and unfccc negotiations.

25

3.3 Explaining the Difference: Analytical Framework

The questions of why and how groups organise have been dealt with extensively in socialmovement theory. Since the work of Eisinger (1973), this literature uses the ‘politicalopportunity structure’ to explain political mobilisation. The concept is not well-defined,but generally thought to refer to ‘consistent—but not necessarily formal or permanent—dimensions of the political environment that provide incentives for people to undertakecollective action by affecting their expectations for success or failure’ (Tarrow 1994: 85).In other words, the political opportunity structure consists of ‘processes that alter thecosts and likely outcomes of collective action’, that is, decision maker’s cost-benefit cal-culations (Meyer and Staggenborg 1996: 1633). Note that, as Elster (1989: 20) pointsout, although opportunities are objective and exogenous, action is explained by actors’desires together with their ‘beliefs about the opportunities’.

We can apply the concept of political opportunity structure to group’s decision ofwhether to become active in particular institutional environments. In our case, then,the political opportunity structure helps ipos assess their prospects for success, in otherwords, whether investing in advocacy in a particular institution seems worthwhile.

The opportunity structure has several dimensions, which we call opportunities andconstraints. First of all, the organisational capacities of a group have consequences forthat group’s cost-benefit calculation. Expertise in an issue area and funding opportu-nities, for instance, will make an organisation’s participation in that issue area morelikely. Second, venues may be more or less open to the involvement of indigenous organ-isations. Openness comprises both formal rules of access and informal recognition as astakeholder (Kriesi 2004: 68).13 Other actors also play an important role. The presenceof actors with similar interest may present opportunities for building coalitions, sharinginformation and pooling resources, all of which reduce the costs of lobbying. This should,ceteris paribus, positively influence an organisation’s decision to also become active inthat venue, in particular for resource-poor organisations. Similarly, groups with opposinginterests—as long as they are not too strong so as to deter participation—may in factincrease incentives for participation, since other groups will want to counteract the effortsof their opponents (Gormley and Cymrot 2006; Hojnacki 1997; Holyoke 2003; Mahoney2007).

However, the structure of the political environment is insufficient to explain whenand where indigenous peoples organisations take action. Different issues may affectindigenous peoples to varying degrees; if a particular issue has only little bearing onindigenous communities, it should come as no surprise that few organisations becomeactive on that issue. Mobilisation thus requires what the social movement literature

13See also Meyer and Minkoff (2004) and references therein on openness as a dimension of the politicalopportunity structure.

26

calls ‘goals’ or ‘grievances’, which is a function of the perception of being affected. Inorder to become active on a particular issue, a group needs to recognise an issue asa problem, and needs to desire its solution. Again, it is important to stress that theobjective impacts of an issue play a lesser role than the subjective perception of beingaffected by these impacts (Kriesi 2004; Meyer 2004). We distinguish two dimensions tothe perception of being affected: A group can be affected directly by the issue itself;or indirectly because policy responses to that issue have (positive as well as negative)impacts on local stakeholders.

goals/grievances- perception of being affected directlyby an issue ↘- perception of being affected indirectlyby political responses to an issue

↑↓ decision to become activein a particular venue

opportunities/constraints↗- institutional openness

- organisational capacities- behaviour of allies and adversaries

Figure 3.3: Framework for analysis

Our analytical framework combines these two sets of factors, as figure 3.3 lays out.Note that the factors—goals and grievances on the one hand, and opportunities and con-straints on the other—are not independent of each other. Recognition as a stakeholder,which falls into the opportunities/constraint category of our framework, for example, hasan effect of how a problem is framed and perceived, that is, on goals and grievances.

Since we are primarily interested in the perceptions of grievances and opportunitiesand constraints, we sought to supplement existing information with first-hand informa-tion from indigenous groups active in environmental negotiations. We therefore con-ducted semi-structured interviews with representatives from indigenous organisationsthat are involved in the unfccc and/or cbd negotiations, as well as with experts on thetopic.14 The information obtained through interviews complemented information gath-ered from the organisations’ websites where available, official documentation from thetwo agreements, as well as secondary literature. The material presented here is intendedas an indicative analysis, rather than a comprehensive and conclusive assessment, of in-digenous involvement in the cbd and unfccc negotiations. We present a snapshot ofindigenous participation in two dynamic regimes (see Schroeder 2010: 323).

14To some extent, our data collection efforts hence suffer from bias, as we only look at organisationsthat are active in either or both fora of interest. While we admit that it would be extremely interestingto also investigate indigenous organisations that have chosen not to participate in international environ-mental negotiations, lack of data makes it impossible to identify the appropriate population from whichto sample.

27

3.4 Biodiversity and Climate Change Compared

3.4.1 Goals and Grievances

Direct Impacts

A fundamental characteristic of indigenous peoples is their close relationship with anddirect dependence on their physical environment. This environment is typically both,rich in biological diversity as well as fragile and vulnerable to change.

Traditional indigenous territories, according to a report by the World Bank, make upabout one fifth of global land surface, yet hold four fifths of the world’s biodiversity—indigenous land contains many so-called ‘biodiversity hotspots’ (Sobrevila 2008; Toledo2001). Scientists recognised early that this is no coincidence. Indigenous peoples areknown to not only conserve and nurture natural diversity, but also to augment it throughlandscape manipulation (Gadgil et al. 1993).15 For instance, the majority of biologicaldiversity in domesticated plants and animals is due to interventions by indigenous peoples(see Mauro and Hardison 2000).

Indigenous organisations highlight this relationship between indigenous peoples andbiodiversity, stressing their role in both cultural and biological diversity. ‘Over thousandsof years, indigenous lifestyles have contributed to not only conserve, but create, biologicaldiversity’, says one of our interviewees.16 Similar remarks abound, and indicate thatindigenous peoples perceive themselves to be ‘guardians of biodiversity’, and as suchclaim a role in the cbd negotiations as well as its implementation:

The Biodiversity Convention should enhance the pre-existing relationship be-tween indigenous peoples and biodiversity, [...] and unless indigenous peoplesare placed at the centre of protecting and managing the biodiversity on theirterritories, a major opportunity of facilitating collaborative conservation willbe lost (International Alliance of Indigenous-Tribal Peoples of the TropicalForests 1996).

It should be noted that although indigenous organisations promote themselves ascustodians of diversity, the concept of biological diversity as taken up in the Conventionon Biological Diversity is a Western concept, rooted in conservation biology (Redfordand Stearman 1993). It may not be readily understandable for indigenous peoples, asPotvin et al. (2002) point out:

Biodiversity is a western scientific and political construct that is influential atthe international and national levels. At the local level, the concept might beeither not understood or considered irrelevant, since ecosystems are viewed

15See also interview, Ahmed Djoghlaf, cbd Secretariat, 08 June 2011; interview, Thomas Brose,Climate Alliance, 29 April 2011.

16Interview, Thomas Brose, Climate Alliance, 29 April 2011.

28

in terms of the resources they provide for direct and in some cases indirectuse (see also Colchester 2003; Oldham 2002).

This perception of biodiversity in terms of resources, however, is useful to under-stand the immediate impact that the extinction of species has on indigenous peoples.Indigenous peoples require high biodiversity for food security and social reproduction, so‘loss of biological diversity means loss of livelihood’,17 and indigenous peoples directlyexperience biodiversity loss as they notice a diminishing availability of resources.18

Climate change, in contrast, has more mediated effects for indigenous peoples, whichwere recognised comparatively late. Climate research only picked up the issue of in-digenous peoples in the mid-2000s. It was not until the fourth assessment report ofthe Intergovernmental Panel on Climate Change (ipcc) that climate change impactswere explicitly linked to indigenous peoples (Parry et al. 2007). The connection mainlystems from the fact that some ecosystems, including low-lying shores or mountainousareas, are more vulnerable to climatic changes, and these fragile ecosystems are ofteninhabited by indigenous peoples. Indigenous peoples, being themselves weak emitters ofgreenhouse gases, hence are disproportionately affected by the negative consequences ofclimate change (Green and Raygorodetsky 2010; Salick and Ross 2009).

Indigenous peoples themselves began to report changes in their natural environmentthat they connect to climate change, including changes in precipitation and seasonalpatterns as well as increased frequency of extreme weather events such as droughts andfloods (see e.g. testimonies in Mihlar 2008; Permanent Forum on Indigenous Issues2008). While indigenous peoples admit their particular vulnerability to climate change,the relationship between indigenous communities and climate change is not as readilyvisible, as one interviewee explains:

Although the consequences of climate change will be felt, and indeed partiallyare already felt, strongly by indigenous peoples, they have not yet made thelink between climate change and their territories. Indigenous organisationsare trying to change this, but for organisations at the local level, climatechange is still at a distance. Climate change is not yet as much on theindigenous agenda.19

It is important here to stress that for indigenous peoples, the changes that theyexperience in their environment cannot be related either to climate change or to lossof biodiversity; for indigenous peoples, these are not two separate issues, but pertainto one single environment.20 Indeed, climate change and biological diversity are closely

17Interview, Heidi Feldt, 18 May 2011; see also interview, Charles Masabo, Unissons-nous pour laPromotion des Batwa (uniproba), 13 June 2011.

18Interview, Maurizio Ferrari, Forest Peoples Programme, 12 May 2011.19Interview, Grace Balawag, Tebtebba, 03 June 201120Interview, Grace Balawag, Tebtebba, 03 June 2011; interview, Charles Masabo, uniproba, 13 June

2011; interview, anonymous representative, indigenous peoples organisation, 15 June 2011.

29

interrelated, and climatic changes are known to affect biological diversity. Differencesin direct impacts are thus unlikely to explain differences in indigenous involvement inpolicy making, even if these impacts pose a—sometimes existential—threat to indigenouscommunities.

Indirect Impacts

As opposed to direct effect, indirect consequences seem to be a stronger explanatoryfactor for the mobilisation of indigenous peoples organisations. In both, the cbd andthe unfccc, indigenous organisations seem to be mainly concerned with the impacts ofresponse measures on their rights, in particular their right to and ownership over land.

In light of the close link between indigenous land and biodiversity, indigenous territorywas likely to be impacted by conservation efforts under the cbd, for governments seemedto think that the easiest way to implement the cbd is through the establishment of nat-ural parks and protected areas, which in many cases end up overlapping with indigenousterritories, as one interviewee explains.21

These efforts are based on the Western conception of conservation as the preserva-tion of ‘untouched wilderness’, ideally without any human interference (Colchester 2003;Tauli-Corpuz 2010). Hence, indigenous peoples were concerned about their rights to landand self-determination. As Colchester (2003: 22) writes, protected areas, to indigenouspeoples, are no different to conventional development programmes such as dams or roads:‘Both are experienced as topdown impositions on indigenous communities whereby landsare taken away from the control of local communities and allocated to uses determinedby outsiders.’22 Since the Amazon Basin here plays an important role, the focus onprotected areas explains why Latin American ipos make up such a large majority in thecbd process (see figure 3.2 above). Indeed, as Potvin et al. (2002) write, the cbd forumis used by many Latin American ipos to advance territorial rights.23

In a similar way, the cbd’s objective of the ‘fair and equitable sharing of the bene-fits arising out of the utilisation of genetic resources’ (United Nations 1992a: Art. 1)raises concerns about the economic use of traditional knowledge and thus about indige-nous rights and indigenous identity. Indigenous peoples may experience the privatisationof their resources and their knowledge as a form of modern colonialism. From an in-digenous, holistic perspective, one cannot ‘own’ knowledge, only ‘guard’ it (gfbv 2008).Accordingly, knowledge is an integral and indivisible part of indigenous identity, and itscommercial exploitation a threat to this identity, as the iifb emphasises:

21Interview, Maurizio Ferrari, Forest Peoples Programme, 12 May 2011.22On this point, see also Coordinator of Indigenous Organizations of the Amazon Basin 1989; Feldt

2009; interview, Maurizio Ferrari, Forest Peoples Programme, 12 May 2011.23African ipos, on the other hand, are more concerned with traditional knowledge, see Potvin et al.

(2002).

30

Our knowledge of biodiversity is indivisible from our identities and our laws,institutions, value systems and cosmovisions as Indigenous Peoples. [...] Theprivatisation and commodification of our knowledge and natural resourceswill undermine the political, social, economic, and cultural integrity of ourpeoples (iifb 2011).

In the context of climate change, it is again the link to indigenous land and in-digenous rights that seems necessary to mobilise indigenous organisations. Indigenousorganisations mainly took interest in the issue of climate change when the internationalnegotiations began to include forests as a mitigation strategy in what is referred to asredd, Reducing emissions from deforestation and forest degradation and the role of con-servation, sustainable management of forests and enhancement of forest carbon stocksin developing countries. Just like protected areas, redd projects are prone to be im-plemented on indigenous territory, especially in the tropical forests, and as with theprotected areas, indigenous peoples fear negative impacts on their rights to land andauto-determination. Under redd, governments receive financial support to protect theirforests. If indigenous peoples are excluded from the negotiation and implementation ofsuch forest protection schemes, this may mean a shift in control over forests away fromindigenous communities to governments, and in the worst case result in eviction of in-digenous communities (Degawan 2008; Feldt 2009; Lemaitre 2011).24 Other mitigationefforts, such as the construction of hydro-electrical dams or increased investments intobio-fuel, affect indigenous land rights in a similar way. Nilsson (2008: 9) hence assessesthat ‘[i]ncreasingly, international and national climate change mitigation strategies posean additional threat to indigenous peoples’ territories and coping strategies’ (see alsoMihlar 2008).

3.4.2 Opportunities and Constraints

Institutional Openness

Institutional openness also plays an important role in indigenous peoples organisations’assessment of costs and benefits of advocacy in a venue, and hence in their decision tobecome active in that venue. Institutional openness includes formal rules of access aswell as recognition as stakeholder.

The key obstacle to indigenous participation in environmental negotiations is theintergovernmental nature of the negotiations: Unless states include indigenous membersin their official delegations, indigenous peoples can only participate as observing ngo,and thus through relatively informal channels.25 While this is the case for both the

24Interview, Heidi Feldt, 18 May 2011.25Interview, Thomas Brose, Climate Alliance, 29 April 2011; interview, Segundo Bueno Quichimbo,

Confederation of Indigenous Organisations of the Ecuadorian Amazonian Region (cocarae), 12 May2011; interview, Heidi Feldt, 18 May 2011.

31

unfccc and cbd negotiations, the cbd opens up channels for indigenous participationwithin the implementation of individual articles. Thus, working groups that concernArticle 8(j) and Related Provisions, on the use of traditional knowledge, give indigenousparticipants the opportunity to co-chair meetings, intervene in the debate, or participatein the bureau: Workshops on the article were characterised by ipo participation asearly as 1997 (Oldham 2002: 26, 35). In addition, indigenous delegates may participatein expert panels with the endorsement of government delegations (Oldham 2002: 26;Schroeder 2010: 326). Several interviewees emphasise that in contrast to the relativelyclosed unfccc process, indigenous peoples are included in the cbd decision-makingprocess from an early stage onwards, and that concerns and statements by indigenouspeoples will be listened to and acted upon.26 In climate change, on the other hand, ‘ourconcerns, our ideas, our initiatives are simply left out’, as one respondent complains.27

Another interviewee elaborates,

in the biodiversity negotiations, the indigenous peoples observers get includeddirectly during the negotiations, while in the climate change negotiations, wehave to talk with friendly government delegations to gain their support forour proposals in the negotiating texts. [...] We only have a few minutes todeliver our statements, so we need to have indigenous peoples representativeson government delegations to be able to negotiate directly.28

These differences in formal access result from differences in the recognition of indige-nous and local communities as stakeholders in the respective negotiations. In the cbd,indigenous peoples were recognised as stakeholders from the very beginning: Scientistsemphasised a direct link between biodiversity and indigenous peoples early on, attributingan important role to traditional knowledge and practices for conservation efforts (Mauroand Hardison 2000; Rossbach de Olmos 2004).29 In addition, the inclusion of stipula-tions toward the commercial use of traditional knowledge recognised indigenous peoplesas the legitimate holders of this knowledge and as trading partners to be consulted in thenegotiations (Oldham 2002: 30-33). The Convention text of 1992 hence contains severalreferences to ‘indigenous and local communities embodying traditional lifestyles’, mostimportantly in art. 8(j) (United Nations 1992a).

Beyond article 8(j), the Convention on Biological Diversity confirmed the importanceof indigenous peoples and local communities at cop5 in Nairobi in 2000, where it es-tablished clear rules for the participation of ipos in the negotiations; previously, it haddepended on the chair of the respective session whether contributions would be heard.

26Interview, Ahmed Djoghlaf, cbd Secretariat, 08 June 2011; interview, Grace Balawag, Tebtebba,03 June 2011; interview, anonymous representative, indigenous peoples organisation, 15 June 2011. Seealso field 2008: 13.

27Interview, Segundo Bueno Quichimbo, cocarae, 12 May 2011.28Interview, Grace Balawag, Tebtebba, 03 June 2011.29See also interview, Thomas Brose, Climate Alliance, 29 April 2011; interview, Heidi Feldt, 18 May

2011.

32

Thus, cop5 established an Ad Hoc Open-Ended Working Group on Article 8(j) and re-lated provisions, and reached out to indigenous peoples. For instance, time allocated foripo contributions increased and the negotiators departed from the exclusive use of En-glish during the contact group stage (Oldham 2002; United Nations 1992a). Since then,the social dimension of biodiversity, and in particular the role of indigenous peoples insustainable resource management, has been further recognised.30 The cbd decided toestablish a Voluntary Fund for Indigenous and Local Communities to financially sup-port indigenous and local communities’ attendance at cbd related meetings at cop8 inCuritiba, Brazil, in 2006.31 At the last cop, cop10 in Nagoya, Japan, the Japanesegovernment and the un University Institute of Advanced Studies together launched theSatoyama Initiative, aimed at the conservation of sustainable socio-ecological productionlandscapes, thus further emphasising the need for the inclusion of indigenous and localcommunities on issues of biological diversity.32

In unfccc negotiations, on the other hand, no explicit link was made between indige-nous peoples and climate change until ipos themselves claimed a link between indigenousforest management and greenhouse gas emissions in the mid-2000s (Macchi 2008; Ross-bach de Olmos 2004). Accordingly, official documents first mention indigenous peoplesin 2005, much later than in cbd decisions and reports. Since 2007, however, there hasbeen growing recognition of the social dimension of climate change as well; cop13 inBali, Indonesia, has been described as a ‘turning point’ in this regard.33

While indigenous concerns are gaining recognition in the climate change negotiations,there remain considerable obstacles. Similar to the inclusion of indigenous peoples forconservation efforts in biodiversity, the recognition of indigenous peoples in the climatechange negotiation is mainly due to the realisation that redd efforts will only work effec-tively with indigenous peoples. In addition, there is growing awareness that indigenousand traditional approaches to dealing with environmental changes could provide usefulfor adaptation measures for climate change.34 Cynically speaking, then, the inclusionof indigenous peoples does not result from their recognition as legitimate partners indecision making on issues of biodiversity and climate change; rather, they are seen asnecessary to implement the decisions made on their behalf.

30Interview, Maurizio Ferrari, Forest Peoples Programme, 12 May 2011.31The full title is Voluntary Trust Fund to Facilitate the Participation of Indigenous and Local

Communities in the Work of the Convention on Biological Diversity (Convention on Biological Di-versity 2006). The Voluntary Fund remains the only un fund for this purpose, yet funding is limited.See http://www.cbd.int/doc/programmes/socio-eco/traditional/tk-vf-en.pdf See also interview,Ahmed Djoghlaf, cbd Secretariat, 08 June 2011.

32In the same year, the decision was taken to integrate the implementation of article 10(c) aboutthe customary sustainable use of biodiversity with the work on article 8(j) and related provisions, seehttp://satoyama-initiative.org/en/; see also interview, Maurizio Ferrari, Forest Peoples Programme,12 May 2011.

33Interview, Thomas Brose, Climate Alliance, 29 April 2011. See also Feldt (2009); Macchi (2008);Mihlar (2008).

34Interview, Heidi Feldt, 18 May 2011.

33

Organisational Capacities

A second important element in ipos’ evaluation of the institutional environments relatesto their organisational capacities. Participation in international environmental negotia-tions does not only require physical presence at the cops; just as important for effectiveparticipation is thorough preparation, which requires considerable financial resources,knowledge, and expertise. Indeed, the problem with participation is not so much theformal or informal access granted by the forum, but with how ipos can make use of theopportunities for participation.35 Several respondents in this context comment on thedifference between participation and meaningful participation.36

Financial resources seem to be a key factor, as one interviewee explains: ‘If we had themoney, we would send delegations to all negotiations on environmental protection.’37 TheVoluntary Fund here is an important source of funding participation, even if the largestshare of funds may still be coming from ipos themselves.38 Funding may also comefrom other ngos, foundations, or other un agencies. Here, one interviewee raises thequestion of who obtains financial support, and in what way this affects the independenceof recipients. Not only does the structure of the accreditation process favour Westernorganisations; it is much easier to obtain funding if you agree with donor positions.39

Beyond funding, expertise plays a central role. International environmental negoti-ations are very complex and multilayered processes. Organisations need to follow theentire process closely in order to maintain contacts and obtain information on negoti-ation statuses and strategic opportunities to bring in their concerns.40 This is difficulteven for some states that, due to their size, often do not have their own expert staff(Chasek 2001). For indigenous peoples, it is even more difficult to organise permanently,in order to prepare and lobby national governments beforehand. The cops themselvesare just ‘the tip of the iceberg’.41

Both, climate change and biological diversity, are complex processes, so complexityper se cannot explain differences in indigenous participation.42 Yet complexity of thenegotiations and the costs involved in informed and active participation point towards the

35One respondent notes that there is a limited number of ipos in the cbd that are capable of effectiveand sustained participation, see interview, Maurizio Ferrari, Forest Peoples Programme, 12 May 2011;see also e.g. interview, Thomas Brose, Climate Alliance, 29 April 2011.

36Interview, Ahmed Djoghlaf, cbd Secretariat, 08 June 2011, interview, anonymous representative,indigenous peoples organisation, 15 June 2011. See also Fisher and Green (2004).

37Interview, Segundo Bueno Quichimbo, cocarae, 12 May 2011, emphasis added38Interview, Maurizio Ferrari, Forest Peoples Programme, 12 May 2011; interview, Charles Masabo,

uniproba, 13 June 2011; interview, representative, anonymous indigenous peoples organisation, 15 June2011.

39Interview, Rebecca Sommers, 29 June 2011.40Interview, Thomas Brose, Climate Alliance, 29 April 2011.41Interview, Thomas Brose, Climate Alliance, 29 April 2011, interview, Heidi Feldt, 18 May 2011; see

also Feldt (2009).42Note, however, that respondents evaluated the complexities of the negotiations on these issues

differently. While some argued that biodiversity is more complex, others believed the opposite to betrue.

34

importance of familiarity with the issue, institution, and other actors. The cbd is a mucholder theme among indigenous peoples, and ipos have become aware of the negotiations atthe international level and the possibility of participating in them. Information about theclimate change negotiations, in contrast, is more recent, and while indigenous peoples areaware of the issue itself, they know less about the international negotiations and how theseaffect their communities.43 Thus, the continuing difference between ipo participation incbd and unfccc negotiations—despite growing recognition of indigenous peoples asstakeholders in climate change processes—may be a result of path dependence. In otherwords, ipos that are acquainted with the cbd negotiations may be reluctant to extendtheir activities to the unfccc process.

Allies and Adversaries

Finally, the behaviour of allies and adversaries may also play a role in the decision toparticipate in negotiations on a particular issue or in a particular venue. The presence ofgroups with similar interests thereby provide opportunities for coalition building, resourcepooling, and information sharing, making advocacy at the same time less costly and moreeffective (Hojnacki 1997; Holyoke 2003; Mahoney 2007).

Several interviewees emphasise their cooperation with like-minded groups, in par-ticular, and stressed cooperation among indigenous organisations.44 In the cbd, thiscooperation was formalised as early as 1996, when ipos formed the iifb with the aim tohelp coordinate indigenous strategies at the meetings, provide advice to the governmentparties, and influence the interpretations of government obligations to recognise and re-spect indigenous rights to the knowledge and resources within the cbd as well as ‘otherimportant international environmental meetings’.45 The iifb was officially recognisedas advisory body by the cbd cop in 2000 and now receives logistical support from theSecretariat of the Convention (Convention on Biological Diversity 2000).46

In climate change negotiations, ngos form ‘constituencies’, lose networks that roughlycoordinate their positions and actions. Indigenous peoples have been recognised as aconstituency in 2001, as the fourth constituency (after environmental ngos, businessngos, and local and municipal groups). The recognition as constituency has given them‘slightly more recognition and capacity and they can make statements at preliminaries ofcop meetings’ (Mihlar 2008: 2-3). Yet, and in spite of fewer numbers, coordination andcooperation in the unfccc negotiations has not been as close as in the cbd negotiations,mainly because of disagreement on how to go about achieving their goals.47

In both fora, environmental ngos present natural partners of ipos as they share43Interview, Rebecca Sommers, 29 June 2011.44E. g. interview, Charles Masabo, uniproba, 13 June 2011.45See the iifb website at http://www.iifb.net/.46For further information, see the cbd website at http://www.cbd.int/traditional/general.shtml.47Interview, Ahmed Djoghlaf, cbd Secretariat, 08 June 2011.

35

roughly the same interests. Indeed, there has been coordination and exchange betweenthe two types of organisations.48 However, interests do not always correspond. In thecbd, for example, ipos stress the linkages between biodiversity and rights to land, terri-tories and resources, whereas environmental ngos focus more on the conservation aspectitself and may even consider the indigenous inhabitants to be a disruptive factor.49 Over-all, however, there seem to be no differences in how ipos cooperate with environmentalgroups between the cbd and unfccc processes.

The presence of opposing interests also affects participation, as they may either deterparticipation or provide incentives for counteractive lobbying (Holyoke 2003). Interest-ingly, our respondents did not identify clear opponents, although some governments wererepeatedly mentioned as blocking indigenous concerns, especially when the issue undernegotiation is related to indigenous rights. Thus, one interviewee reports that withinthe cbd negotiations, most countries do not oppose stipulations on traditional knowl-edge; the issue becomes a problem for some Parties only when it is linked to territorialrights as traditional knowledge has to be practiced on the land and with resources.50

This resonates with another interviewee’s comments on the difference between acceptingindigenous rights formally, and including ipos in the negotiations to bring in their ownpositions.51 Again, however, these comments did not point to differences between thetwo regimes.

3.5 Concluding Remarks

Although climate change seems to be just as pressing an issue for indigenous communitiesas the loss of biodiversity, ipos are much more active in the international negotiations onbiodiversity. This seems surprising, for the unfccc process is the more visible process,as a field report argues: ‘In addition to being the most relevant agreements when itcomes to adaptation, the unfccc and Kyoto Protocol carry significantly more politicalweight than the cbd. Even small steps in the unfccc may be of greater value thanbig steps in the cbd’ (field 2008: 13). As opposed to the climate change negotiations,however, the cbd negotiations are relatively more open toward indigenous participation.Together with the earlier recognition of biodiversity as a problem with direct implicationsfor indigenous communities, this institutional openness explains why ipos focus theirinternational activities on the cbd process.

Our analysis thus points to two important findings: Firstly, indigenous peoples world-wide feel affected by both, the loss of biological diversity and climate change, and report-

48Interview, Heidi Feldt, 18 May 2011.49Interview, Thomas Brose, Climate Alliance, 29 April 2011.50Interview, Maurizio Ferrari, Forest Peoples Programme, 12 May 2011; see also interview, Grace

Balawag, Tebtebba, 03 June 2011.51Interview, Heidi Feldt, 18 May 2011; interview, anonymous representative, indigenous peoples or-

ganisation, 15 June 2011.

36

edly are already experiencing negative consequences of both phenomena. Compared toclimate change, however, biodiversity is not only perceived to be the more direct threat,it was also recognised as a problem earlier; already in the early 1990s, indigenous organ-isations had taken interest in the issue of biodiversity. Climate change, in contrast, onlyreached the indigenous agenda in the late 1990s when it became clear that mitigationprojects would likely be implemented on indigenous land.52 The emergence of forestsas a topic in the climate change debate in the mid-2000s sparked further interest, butparticipation is only slowly increasing. For ipos to mobilise and engage in internationalpolicy making, a direct connection to rights, in particular indigenous rights to land andself-determination, seem key.

This leads to our second key finding: Indigenous participation in international nego-tiations necessitates external recognition as a stakeholder. In particular, scientific ac-knowledgement that indigenous peoples are affected by an environmental problem seemsto open up negotiations on that problem, which in turn facilitates political activismby indigenous groups in these negotiations. Scholars had recognised the importance ofindigenous peoples and traditional knowledge in the conservation and management ofbiological diversity by the time the Convention on Biological Diversity was negotiated,so that the Convention explicitly refers to indigenous peoples and encourages indigenouspeoples to get involved in the implementation and further negotiation of the Convention.In contrast, indigenous peoples were almost completely missing in the climate debateuntil the introduction of redd in the negotiations in the mid-2000s, that is, about fif-teen years after the start of the unfccc negotiations. Since then, it has proved difficultfor indigenous organisations to get their foot into the door of a highly politicised andtechnical process.

With the increasing recognition of indigenous peoples as stakeholders, and as impor-tant actors in finding solutions to the problem of climate change mitigation and adap-tation, however, the unfccc process may open further toward indigenous participation.The cbd process, while far from perfect, may serve as an example in that respect.

Let us conclude with a note of caution: Our analysis does not allow us to distinguishclearly between cause and effect. We cannot assert with confidence that institutionalopenness, in the case of biodiversity, led to an increased participation of indigenous or-ganisation. It might also be that indigenous involvement led to institutional opennessin the first place, which then helped other indigenous organisations to become activein the cbd negotiations. Future work needs to take up the issue of causality, and lookat a broader range of indigenous organisations involved in international environmentalnegotiations. Further interviews, but also large-N surveys of stakeholders in these nego-tiations, both indigenous and non-indigenous, can help clarify the direction of causality,that is, whether indigenous agency led to institutional openness, or whether institutional

52Interview, Heidi Feldt, 18 May 2011.

37

openness drove indigenous agency.

List of Interviews

Thomas Brose Climate Alliance, 29 April 2011.Segundo Bueno Quichimbo Confederation of Indigenous Organisations of the Ecuado-

rian Amazonian Region (cocarae), 12 May 2011.Maurizio Ferrari Forest Peoples Programme, 12 May 2011.Heidi Feldt expert, 18 May 2011.Grace Balawag Tebtebba Foundation, 03 June 2011.Ahmed Djoghlaf Secretariat of the United Nations Convention on Biological Diversity,

08 June 2011.Charles Masabo Unissons-nous pour la Promotion des Batwa (uniproba), 13 June

2011.Representative indigenous peoples organisation, 15 June 2011. (Wished to remain

anonymous.)Rebecca Sommers expert, 29 June 2011.

39

4 | Business Insiders and Environmen-tal Outsider? Advocacy Strate-gies in the Climate Change Ne-gotiations

Carola Betzold1

Interest groups in international climate change negotiations have broad ac-tion repertoires to influence negotiations, ranging from inside advocacy suchas direct contacts with negotiators to outside advocacy, including media con-tacts or demonstrations. These diverse activities are typically classified asbelonging to either an inside strategy, or an outside strategy, depending onwhether they aim at influencing politics directly, or whether they seek to ex-ert influence from the outside, via the media and the general public. Whatexplains variation in the use of these two advocacy strategies?

Based on research on national interest groups, I put forward three reasonsfor specialisation in an insde or outside strategy: First, the goals of non-governmental participation—namely, to provide accountability and to im-prove the negotiation process—are potentially conflicting. Second, groupsdiffer in the degree of access to policy makers; if they lack access, they areunable to engage in inside advocacy. Third, groups do not only seek to influ-ence policy, for which inside advocacy is presumably more suitable; they arealso interested in their survival as organisations, for which outside advocacyseems more appropriate.

I test these reasons with novel data from a survey conducted among organisa-tions active in international climate politics. The statistical analysis providesevidence for differences across group types. Business groups use more insideadvocacy, environmental organisations tend toward an outside strategy. Overand above group type, however, the data also indicate that some group char-acteristics affect the choice of advocacy strategy; notably, experience increasesthe use of inside advocacy.

1This chapter has been accepted for publication in Interest Groups and Advocacy.

40

There are many ways in which interest groups can and do influence the policy-makingprocess at the national level. The action repertoires of interest groups in internationalnegotiations, such as those on climate change, are similarly broad. Among other things,non-governmental organisations (ngos)2 contact negotiators directly; contribute to draft-ing legal text; present their organisation in side events and exhibitions; organise pressconference or prepare press releases; and mobilise supporters in protest marches. Thesedifferent activities tend to be categorised as belonging to an inside or outside strategy,depending on whether they aim at influencing decision making directly, or seek to createpressure from the outside, by influencing public opinion and mobilising the public in adirection desired by the interest group (e.g. Binderkrantz 2005; Kollman 1998; Walker1991).

How do ngos choose among these many activities? What advocacy strategies3 dothey pursue, and why? This paper starts from the assumption that ngos have twomain objectives: influencing policy and securing their survival as organisations (Dürand Mateo 2012; 2013), or what Schmitter and Streeck (1999) refer to as the logic ofinfluence and the logic of membership. How groups balance these two goals, it argues,affects their choice of strategy: when a group needs to secure the continuous support ofmembers, it turns to outside advocacy, whereas inside advocacy is the preferred optionfor influencing policy. Additionally, differences in access to policy makers help explainvariation in advocacy, since access is a precondition for inside advocacy. The paper thusanalyses differences across group types, and in particular compares environmental withbusiness organisations, but also looks at the effect of group characteristics, specifically,expertise and membership type, on advocacy behaviour.

Empirically, the paper uses original data from a survey of ngos active in the inter-national negotiations on climate change. 1026 organisations accredited with the UnitedNations Framework Convention on Climate Change (unfccc) were invited to partici-pate, and 216 fully completed questionnaires obtained, which corresponds to a responserate of just over 21 percent. The questionnaire asked how often participating ngos en-gage in different activities. Responses to this item serve to measure groups’ use of insideand outside advocacy, as well as the relative importance of inside advocacy in a group’saction repertoire.

This analysis adds to a growing literature on the strategies of interest groups (Beyers2004; Binderkrantz 2008; Binderkrantz and Krøyer 2012; Dür and Mateo 2012; 2013; Gaisand Walker 1991; Mahoney 2008; Maloney et al. 1994; Schlozman and Tierney 1986). By

2The term interest group is mainly used for groups that seek to influence policy making at thenational level, while groups that are active in international policy making are usually referred to asnon-state actors or non-governmental organisations (ngos) that are admitted as observer organisations.I use these terms interchangeably to refer to groups that seek to influence policy at the national and/orinternational level.

3I use advocacy, being a broader term than lobbying, to denote all ngo activities that aim at reachingthe ngo’s goals. A strategy refers to a combination of activities (cf. Dür and Mateo 2013).

41

looking at advocacy in international climate change negotiations, it applies insights fromresearch on national interest groups to global policy-making processes. It thus also addsto the literature on civil society in global environmental governance (e.g. Lisowski 2005;Newell 2000; Orr 2006; Rietig 2011; Vormedal 2008), where large-N surveys of differenttypes of ngos remain rare. The results support previous research that finds variationin strategies according to group type; in particular, it confirms that business groups areinsiders with privileged access to policy makers. However, other factors also play a role.Notably, the paper finds a positive and significant association between experience andthe use of inside advocacy. In contrast, membership type does not impact the choiceof advocacy strategy. This suggests that differences in access are better able to explainvariation in the use of advocacy strategies, as opposed to how groups balance the goalsof influencing policy and securing organisational survival.

If we assume that advocacy activities differ in their effectiveness, it is important totake into account variation in the use of advocacy strategies. Since the value of non-state participation in intergovernmental negotiations stems from the diversity of viewsand ideas that different stakeholders can bring to the process, we need to ask whetherall views and ideas have the same chance of being heard. The findings here are hencerelevant for debates on the legitimacy of the negotiations, as well as, more generally, onthe role of civil society in environmental governance.

4.1 Explaining Advocacy Strategies

The participation of non-state actors in global environmental governance has increasedtremendously over the past decades, both in terms of size and scope. This surge innon-governmental activity at the international level has attracted considerable academicinterest (see Betsill 2006), including on how ngos seek to influence negotiations (Arts1998; Betsill and Corell 2008; Giorgetti 1999; Lisowski 2005; Newell 2000; Orr 2006;Raustiala 2001; Rietig 2011; Vormedal 2008). However, the strategic behaviour of in-terest groups has been analysed more extensively at the national level. In particular inthe United States, there exists a long tradition of studies on interest group activitiesin Washington (e.g. Bauer et al. 1963; Baumgartner and Leech 1998; Berry 1977; Mil-brath 1963; Schlozman and Tierney 1983; Walker 1991); more recent work also looksat advocacy behaviour in US states (Nownes and Freeman 1998), the European Union(Beyers 2004; Dür and Mateo 2012; 2013; Mahoney 2008; Marshall 2010; Woll 2006), orindividual European countries (Binderkrantz 2005; 2012; Binderkrantz and Krøyer 2012;Dür and Mateo 2010).

This literature distinguishes broadly between two types of advocacy, inside and out-side advocacy. The former includes all activities that rely on the direct transmissionof information to decision makers; the latter, in contrast, seeks to create pressure more

42

indirectly, by influencing public opinion and mobilising the public in a direction desiredby the interest group (Grant 2001; Gulbrandsen and Andresen 2004; Kollman 1998; Ma-honey 2008; Richards and Heard 2005). Some authors refer to the distinction as onebetween direct and indirect advocacy (Binderkrantz 2005; 2008), between access andvoice (Beyers 2004), between engagement and confrontation, or between the politics ofpartnership and the politics of blame (Alcock 2008).

Much of the environmental governance literature adopts this distinction and suggeststhat non-governmental groups in international politics specialise in one or the other ad-vocacy technique; activist outsiders are thus differentiated from insider advisory organ-isations (e.g. Betsill 2006; Breitmeier and Rittberger 1998; Gulbrandsen and Andresen2004). What are the reasons for such a tactical specialisation?

Steffek and Ferretti (2009) propose that such a specialisation results from the two po-tentially conflicting goals of non-governmental participation in environmental governance,namely, to act as ‘watchdogs’ and provide accountability; and to act as ‘deliberators’ andprovide input into the negotiation process. Because the first role may prevent civil so-ciety actors from effectively fulfilling their deliberative role, and vice-versa, the authorsargue for a division of labour between these two types of groups (see also Steffek 2008).

A second explanation for tactical specialisation that is commonly put forward in theinterest group literature centers on differences in access to policy makers. Generally,inside advocacy—for which access is a precondition—is regarded as preferable, as ‘thedefault strategy for any political actor’ (Kriesi et al. 2007: 53). In contrast, only groupswithout access will opt for outside advocacy so as to compensate their lack of access.From this perspective, outside advocacy is thus a mere fallback option for those groupswho do not have, or do not want, access (e.g. Beyers 2004; Binderkrantz 2012; Gais andWalker 1991; Grant 1989; Maloney et al. 1994).

Finally, a third explanation points to trade-offs in interest groups’ objectives. Interestgroups are typically assumed to have two main objectives: they seek to influence policy,but at the same time are also interested in their survival as organisations (Dür and Mateo2012; 2013; Schmitter and Streeck 1999). Membership type affects how groups addressthat trade-off and balance these two objectives. The more fragmented the membership,the more important it will be to signal to members that the organisation is activelydefending the group’s cause, that it is ‘worthy of continued support’ (Gais and Walker1991: 105). This applies in particular to organisations whose members are united onlyby their support for the group’s cause; and to organisations that depend on members’financial contributions, specifically if these contributions are spontaneous and based onapproval of the organisation’s activities (cf. Schmitter and Streeck 1999). Outside advo-cacy is better suited to signal an organisation’s worthiness of support to fragmented andfar-flung members, since outside activities such as media interviews or demonstrationsare much more visible than inside advocacy (Binderkrantz 2005; 2008; Gais and Walker

43

1991; Maloney et al. 1994).

In international environmental politics, two types of ngos are typically distinguished,environmental ngos and business associations. In line with the two goals of ngo

participation (Steffek and Ferretti 2009), environmental ngos are described as ‘watch-dogs’, while business groups tend to act as ‘deliberators’ (Giorgetti 1999; Lisowski 2005;Vormedal 2008). This distinction parallels studies on domestic lobbying, where citizengroups are differentiated from business groups. The assumption is that the former relymore on outside advocacy, the latter using more inside advocacy (Binderkrantz 2008; Dürand Mateo 2013; Schlozman and Tierney 1986). Group type is indeed likely to affect agroup’s advocacy behaviour, with environmental organisations engaging in more outsideand business organisations, in more inside advocacy, for both of the above reasons: ac-cess and membership. First, environmental groups typically are citizen or cause groupswhose membership is united only by their diffuse interest in protecting the environment.These groups need the public visibility of outside advocacy to signal to their membershiptheir active engagement and to maintain member support. In contrast, business groupsdefend the specific interests of their more concentrated members; they therefore rely toa lesser extent on public visibility to communicate with their members. Second, businessorganisations tend to enjoy privileged access to policy makers because of the economicresources they possess (Binderkrantz 2005); they also tend to possess specialised infor-mation that can be exchanged with policy makers for access (Dür and Mateo 2012; 2013;Maloney et al. 1994). For these reasons, I expect a greater reliance on outside activitiesamong environmental ngos, whereas business groups should engage in relatively moreinside advocacy:

H1a Environmental organisations use relatively more outside advocacy than businessgroups.

H1b Business and industry groups pursue more inside advocacy compared to environ-mental groups.

It should be noted that the above explanations—access and membership—rest ondifferent assumptions. The explanation based on differences in access presumes thatorganisations opt for outside advocacy for lack of access, and thus because they cannotengage in inside advocacy. In contrast, the argument that organisations opt for a certainadvocacy strategy as a result of how they balance the objectives of influencing policyand organisational survival assumes that organisations have a choice: Groups may optfor outside advocacy because they want to. The above expectations do not address thequestion of choice versus need, as both factors should lead to a tendency for environmentalorganisations to pursue outside advocacy, and for business associations to pursue insideadvocacy. However, it is—to some extent—possible to address that question directly.Access and membership type is not necessarily a characteristic of group type per se, andtheir influence on advocacy strategies can be tested empirically.

44

If advocacy behaviour is a result of differences in access, and thus of external con-straints, we should find a relationship between access to policy makers and use of insideadvocacy, independent of other factors. How is access gained? Inside advocacy is oftendescribed as an exchange: In return for certain (valuable) goods, policy makers grant ac-cess to interest groups (DeGregorio 1998; Dür and Mateo 2013; Hall and Deardorff 2006;Maloney et al. 1994; Milbrath 1963; Victor 2007). Access thus depends on the goodsthat interest groups can offer to policy makers. In particular in the context of complex,technical issues such as climate change, a valuable good to policy makers is knowledge:specialised information or technical expertise (Maloney et al. 1994; Walker 1991). It fol-lows, then, that interest groups that can provide relevant technical advice and scientificexpertise should have easier access to policy makers, and thus be more inclined to turnto inside advocacy—in particular if inside advocacy is the generally preferred method forinterest groups and outside advocacy a mere fallback option:

H2 The more expertise an interest group has, the more it will pursue inside advocacy.

If, in contrast, advocacy behaviour is a result of choice, we should see a relationshipbetween membership type and advocacy, again, independent of other factors. When theirmembers are diffuse and far-flung, groups—environmental, business, and others—needto put greater weight on securing these members’ continued support. A disparate mem-bership will rely on public rather than group internal channels to learn about the interestgroup’s actions. To communicate to their members that their interests are actively pro-moted, the interest group therefore needs to engage in publicly visible activities, in otherwords, in outside advocacy (Binderkrantz 2005; Gais and Walker 1991). This applies inparticular to groups who have individuals as members, since individuals primarily relyon public information to learn about group activities, much more so than other types ofmembers, such as companies or other institutions. As Binderkrantz (2008: 192) writes,‘individuals must be regarded as the type of members that are most likely to be reachedonly through publicly visible strategies’. Accordingly, outside advocacy, which is by def-inition more visible to the general public, should be used more often by groups withindividuals as members:

H3 Groups whose membership consists of individuals use more outside advocacy.

These explanations and expectations are not mutually exclusively; it is plausible, andindeed likely, to expect organisations to combine elements of inside and outside advocacy.Since the effect of any single activity is uncertain, an organisation will want to pursuea range of activities to reduce uncertainty and increase its likelihood of success (Beyers2004). Accordingly, research has shown again and again that the majority of groups preferto pursue a wide range of tactics rather than relying on a single strategy (Baumgartnerand Leech 1998; Kriesi et al. 2007). Nonetheless, research also points to variation in therelative importance of inside and outside advocacy, and it is this relative importance thatis expected to vary with group type, access, and membership.

45

4.2 Data Collection and Measures of Variables

A survey among groups registered with the unfccc was conducted in 2011 to collectdata on interest group activities in the international climate change negotiations. All1026 accredited ngos were invited to participate in the survey;4 216 fully completedquestionnaires were returned; the return rate is thus just above 21 percent.

In order to obtain information on the advocacy behaviour of these organisations, thequestionnaire asked respondents to indicate for a total of 13 advocacy activities (seefigure 4.1) how often they engage in these activities in the context of climate changenegotiations, with a four-point answering scale ranging from never to very often.5

The dependent variables are then the mean frequencies with which groups pursueinside and outside activities; a second indicator to measure the extent to which an interestgroup pursues inside advocacy in proportion to its overall advocacy was obtained bydividing the mean frequency of inside advocacy by the sum of the mean frequencies ofboth types of advocacy.6 Accordingly, if an organisation uses inside activities on averagemore often than outside activities, the indicator takes on a value larger than .5, andthe corresponding organisation can be portrayed as an insider. In contrast, an indicatorsmaller than .5 indicates a tendency toward outside advocacy, since the organisationpursues, on average, outside activities more frequently. Finally, a value of around .5points to an evenly balanced action repertoire.

The questionnaire also asks for a number of organisational characteristics, includingthe organisation’s type of membership and observer constituency; in which country orcountries the organisation has its headquarters and subsidiary offices, if any; since whenthey are active in the climate change negotiations; and how many delegates they sent tothe last climate meeting the organisation attended.

The climate change secretariat recognises currently nine so-called observer constituen-cies, loose groupings of broadly like-minded ngos; the two oldest such groups are theenvironmental and the business and industry ngos, newer ones include youth groups,farmers, women organisations, or indigenous peoples organisations. According to theconstituency to which ngos belong, I classify them as environmental (engo), businessand industry (bingo), or other ngo.7

As a proxy for expertise, I include experience with the negotiation process, that is, thenumber of years since the first climate meeting that a group attended. The longer a group

4Only universities were excluded from the sample, since they arguably attend climate change meetingsmainly for research purposes and do not aim at influencing policy.

5The item wording is: ‘At the last cop that your organisation attended, how often did your organi-sation engage in the following activities?’

6The indicator is thus computed as insiderness = meaninsidemeaninside+meanoutside

.7Note that while membership in a constituency is not mandatory, about 90 percent of all registered

ngos do identify with one of the constituencies. See http://maindb.unfccc.int/public/ngo.pl.

46

has been involved in the negotiations, the more it will be familiar with the issues undernegotiations, but also with the negotiation process—and the people negotiating. Thus,experience to some extent also measures ‘connectedness’, the degree to which groups areintegrated in the negotiations, yet arguably, this connectedness is part of expertise, sinceknowledge of people makes it easier to acquire substantial information.

The questionnaire asks what types of members the organisation mainly has. Possibleanswers were individuals, other groups or associations, firms or corporate groups, andothers. Based on the responses to this item, I create a binary variably that takes a valueof one if a group indicates that its members are mainly individuals, and zero otherwise.

Finally, I control for material resources. Advocacy is costly, so the level of resourcesshould influence a group’s level of activism. The literature is unclear which type of advo-cacy is more costly; while many authors argue that inside advocacy is positively associ-ated with material resource endowment, others emphasise the costs of outside advocacy(Kollman 1998; Thrall 2006). I argue that certain activities require more resources thanothers, but that inside advocacy is not systematically more costly than outside advocacy,and vice versa. Hence, material resources should increase the overall use of advocacy,but not favour one type of advocacy over the other (cf. Dür and Mateo 2013). I haveseveral measures to gauge a group’s level of resources. First, I use its delegation size,that is, the number of delegates at the last climate summit that the group attended. Asecond measure of organisational resources is a binary variable that captures whethera group is international, that is, whether it has offices in more than one country, theassumption being that groups with more offices tend to be better resourced, with morestaff and larger budgets. Lastly, I include a binary variable for whether the group isfrom a country member of the Organisation for Economic Cooperation and Development(oecd), since again, groups from industrialised countries tend to be better resourced.

As table 4.1 shows, most respondents belong to the environmental constituency. Theenvironmental constituency is the largest constituency in the climate change negotia-tions, with about 40 percent of all registered ngos; in my sample, they are somewhatoverrepresented, with 127 environmental ngos, or 55 percent of all respondents. Busi-ness and industry groups, in contrast, are proportionately represented, with 31 groups,or 14 percent in the sample, as compared to 15 percent of all registered ngos.8 Theremaining category comprises about one third of the respondents, and covers all otherconstituencies, as well as seven groups that are not member of any constituency (notreported).

Experience with the negotiations varies. The average group has been involved in in-ternational climate change negotiations for about seven years, although the sample com-prises a number of new organisations that have only started to attend climate summitsin the past years, as well as experienced organisations who have followed the negotiations

8See the unfccc website at http://maindb.unfccc.int/public/ngo.pl.

47

frequency/ percent/ min max Nmean std. dev.

dependent variablesinside advocacy 1.00 0.69 0 2.83 215outside advocacy 1.43 0.64 0 3 216‘insiderness’ 0.38 0.18 0 1 215

group typeengo 127 55.46 0 1 229bingo 31 13.54 0 1 229other groups 71 31.00 0 1 229

group characteristicsexperience 6.66 5.41 0 16 229individual members 126 55.02 0 1 229

resourcesdelegation size 8.19 20.73 0 200 229oecd 141 61.57 0 1 229int. presence 90 40.72 0 1 221

Variables in italics are continuous variables.

Table 4.1: Descriptive Statistics

since the very first meeting in Berlin in 1995.

About 55 percent (126 organisations) of the organisations in the sample have individ-uals as members; the remainder has other types of members, including corporate actorsand other associations, but also universities, government agencies, or subnational gov-ernments. Note that group type and membership are weakly correlated (see table 4.4in the appendix); there are comparatively more environmental groups with individualsas members (correlation coefficient of .14), and fewer business associations (correlationcoefficient of -.31) that often are professional or umbrella associations.

The sample is fairly heterogeneous with respect to material resources. On average,organisations send about nine delegates to a climate summit, yet delegation sizes vary alot. Some respondents indicated that they had no delegation of their own at the summit,and most organisations sent five representatives or fewer to the summit, while threeorganisations had over 100 representatives each (not reported). Roughly two thirds of allrespondents (141 organisations) have their headquarters in a oecd country, while about40 percent (90 organisations) are present in more than one country.

4.3 Business Insiders and Environmental Outsiders?

Table 4.1 also includes summary statistics for the dependent variables; figure 4.1 furtherreports the responses to the question of how often organisations use different activities,across all organisations. (Figure 4.5 in the appendix provides additional information on

48

Figure 4.1: Use of Advocacy Activities For All ngos, Percentages

the distribution of the main dependent variables). The table and figures indicate thatinside activities are, for the most part, pursued less frequently than outside activities.Distributing information and contacting negotiators directly, however, are quite common,being used regularly or often by over half of the respondents. The remaining four insideactivities—preparing submissions to the secretariat, participating in the debate, draftingtext, and sitting on a government delegation—are rather rare; only around 20 percentof the respondents engage in the activities on a regular or frequent basis. This may alsopoint to the nature of these activities: they are all subject to government restrictions,since the latter decide for instance who to admit to their national delegations, or whenobservers may speak in the negotiations.

The most common outside activities include side and parallel events as well as in-forming the public about the negotiations: between roughly 60 and 70 of all respondentsindicate that they resort to these three activities regularly or often. Media activities suchas interviews with journalists or press releases follow, with about half the respondentsusing them regularly or often. Demonstrations are comparably rare, only 50 percentof the organisations indicate that they participate in demonstrations, and less than onethird demonstrate regularly or often. Even rarer are press conferences, which are usedon a regular or frequent basis by just over 20 percent of the respondents.

49

4.3.1 Bivariate Analysis

Does the picture change when looking at different groups separately? Hypotheses 1a and1b expect business associations to rely to a greater extent on inside advocacy; environ-mental groups, in contrast, should use more outside advocacy. Figure 4.2 compares themean frequencies of inside and outside advocacy by group type. As expected, environ-mental organisations pursue outside advocacy more often than other groups, and muchmore so than business groups. Although even for the latter, outside advocacy is slightlymore common than inside advocacy, the difference in mean frequency is relatively small,indicating the relatively strong emphasis on inside advocacy. Other groups are in themiddle, with relatively high frequencies for both inside and outside advocacy. Are thedifferences in means across group types statistically significant? According to a t-test,environmental, business, and other groups do not differ significantly in their reliance oninside advocacy (t = −.79, p = .43 for environmental/business; t = −.88, p = .38 forenvironmental/other; and t = .15, p = .88 for business/other). In contrast, and in linewith hypothesis 1a, environmental groups differ significantly from business groups, asdo other groups, in their much greater reliance on outside advocacy (t = 2.87, p < .01

for environmental/business and t = −2.05, p < .05 for business/other); the differencebetween environmental and other groups is insignificant (t = 1.14, p = .26).

never rarely regularly

other NGOs

BINGOs

ENGOs

inside advocacy outside advocacy

Figure 4.2: Mean Use of Advocacy by Group Types

According to hypothesis 2, the use of inside advocacy should increase with a group’slevel of expertise, which is measured by experience. Figure 4.3 depicts the mean fre-quency of inside and outside advocacy as a function of how long groups have participatedin climate change negotiations. Against my expectations, the figure does not suggest anincrease in the use of inside advocacy over time. Neither the use of inside nor of out-side advocacy seems to correlate to experience: despite slight variations, both lines arerelatively flat.

50

neve

rra

rely

regu

larly

very

ofte

n

0 5 10 15experience in years


Figure 4.3: Mean Use of Advocacy by Experience

Finally, figure 4.4 responds to the expectation that type of membership affects advo-cacy behaviour, with organisations with individuals as members pursuing more outsideadvocacy than other organisations. The figure suggests that this is not the case. Or-ganisations with individual memberships do not pursue more outside advocacy; to thecontrary, organisations with other types of members are more active in outside advocacy,but also engage in more inside advocacy than individual-based organisations. The differ-ences between these two categories are, however, very small, and insignificant (t = −.72,p = .47 for inside advocacy; t = −1.14, p = .26 for outside advocacy).

never rarely regularly

other members

individuals as members


Figure 4.4: Mean Use of Advocacy by Membership

51

4.3.2 Multivariate Analysis

In order to isolate the effect of group type and of other characteristics, ordinary leastsquares regression analyses are used, with the dependent variable being the use of insideadvocacy (mean; models 1-3) or the use of outside advocacy (mean; models 4-6). Ina second step, the paper looks at the proportion of inside advocacy in relation to theoverall action repertoire, or degree of ‘insiderness’, as dependent variable.

Models 1 and 4, respectively, only include the group type as predictor for the use of in-side and outside advocacy, respectively. Models 2 and 5, in turn, use group characteristics—expertise and membership type—as predictors, while models 3 and 6 include both typesof predictors, and also controls for material resources. The two last models also includefor the mean frequency of outside (inside) advocacy, as a predictor for the respectiveother strategy, since the two advocacy strategies are correlated (correlation coefficient of.56; for a full list of correlations, see table 4.4).


(1) (2) (3) (4) (5) (6)

engos −0.114 −0.451*** 0.380*** 0.400***(0.139) (0.118) (0.128) (0.112)

other groups −0.020 −0.211* 0.270* 0.247**(0.149) (0.124) (0.138) (0.117)

experience 0.035*** 0.038*** 0.005 −0.018**(0.009) (0.007) (0.008) (0.007)

individual members 0.095 0.107 0.104 −0.030(0.092) (0.077) (0.089) (0.073)

delegation size 0.001 0.005***(0.002) (0.002)

oecd −0.200** −0.025(0.082) (0.078)

int. office 0.123 0.051(0.076) (0.072)

avg. outside advocacy 0.597***(0.062)

avg. inside advocacy 0.531***(0.055)

constant 1.066*** 0.711*** 0.223 1.134*** 1.336*** 0.687***(0.124) (0.092) (0.150) (0.114) (0.089) (0.134)

R2 0.005 0.073 0.444 0.040 0.007 0.413adjusted R2 −0.004 0.065 0.421 0.031 −0.002 0.389N 215 215 208 216 216 208

Coefficients are estimated with ordinary least squares regression. Standard errors in parentheses.*p < .10, **p < .05, ***p < .01

Table 4.2: Regression Analysis of Inside and Outside Advocacy

Group type alone is unable to explain variation in the use of inside advocacy. Al-though the coefficients in model 1 have the correct sign, neither coefficient is significant.However, when group characteristics and control variables are included, the differences

52

between group types become significant. Business organisations, as expected, engagein significantly more advocacy than environmental (p < .01) or other (p < .1) groups(model 3). In contrast, group type explains differences in the use of outside advocacy,regardless of other predictors. According to both, model 4 and model 6, environmentalgroups engage in significantly more outside advocacy than business groups (p < .01); sodo other types of groups, although here the difference is not quite as stark (p < .1). Theeffects are quite large; the difference between environmental and business groups, whenother variables are included, is 40 percentage points, that between other and businessgroups about 20 percentage points (models 3 and 6). The multivariate analysis thusprovides support for hypotheses 1a and 1b: there is systematic variation in advocacystrategies between different group types.

This finding, however, does not tell us whether environmental groups turn to outsideadvocacy because of lack of access, or because they accord higher priority to securing or-ganisational survival. A look at the effects of group characteristics other than group typecan, however, shed some light on this question. As hypothesis 2 expected, experiencedgroups indeed engage in significantly more inside advocacy, regardless of whether it is anenvironmental, business, or other organisation (p < .01, models 2 and 3). An additionalyear of experience with the climate change negotiations increases the mean frequency ofinside advocacy by about 4 percentage points. At the same time, at least according tomodel 6, experience significantly decreases the use of outside advocacy (p < .05).

Membership, in contrast, does not affects groups’ choice of advocacy strategy. Con-trary to hypothesis 3, organisations whose membership consists of individuals do notpursue more outside advocacy. The coefficients have the wrong sign in three out of fourmodels (models 2, 3 and 6), and are not significant in any of the specifications. Theregression results thus do not offer support for my expectation that an individual mem-bership results in more outside advocacy. This is in line with recent research that findsonly slight variation in the behaviour of organisations with different types of members(Kotzian and Steffek 2013). The regression analysis thus supports the argument thattactical specialisation results from differences in access, rather than from differences inhow the goals of influencing policy and securing organisational survival are balanced.Apparently, organisations prefer inside advocacy, and will turn to it when they can, thatis, when they have access to policy makers—and business organisations tend to enjoybetter access, in the climate change negotiations as well as in domestic policy making.

With regard to material resources, the analysis does not find consistent support forthe expectation that resource-rich organisations generally pursue more advocacy. Whiledelegation size positively and significantly affects the use of outside advocacy (p < .01,model 6), it does not significantly increase the use of inside advocacy. Interestingly,organisation with headquarters in an oecd country engage in significantly less insideadvocacy (p < .05, model 3); the coefficient for outside advocacy is also negative, yet

53

insignificant (model 6). These findings supports scholars like Kollman (1998) and Dür andMateo (2013) who argue that resources are also important for outside advocacy. Whetheran organisation has offices in more than one country has no significant effect on advocacystrategies, although both coefficients have the expected sign. Finally, organisations whoare quite active in one type of advocacy also are likely to be quite active in the other typeof advocacy: an increase in inside activity also significantly increase the use of outsideadvocacy, and vice-versa (p < .01, models 3 and 4).

‘insiderness’

(1) (2) (3)

engos −0.148*** −0.176***(0.035) (0.038)

other groups −0.075* −0.080**(0.038) (0.040)

experience 0.008*** 0.010***(0.002) (0.002)

individual members −0.002 0.023(0.025) (0.025)

delegation size 0.000(0.001)

oecd −0.069***(0.026)

int. office 0.042*(0.024)

constant 0.485*** 0.329*** 0.447***(0.032) (0.025) (0.042)

R2 0.086 0.053 0.189adjusted R2 0.078 0.045 0.161N 215 215 208Coefficients are estimated with ordinary least squares regression.Standard errors in parentheses. *p < .10, **p < .05, ***p < .01

Table 4.3: Regression Analysis of Degree of ‘Insiderness’

As a robustness check, the explanatory variables are regressed on the second dependentvariable, the relation of inside advocacy to the entire action repertoire, or degree of‘insiderness’ (table 4.3). In model 1, only group type is used as predictor; model 2includes other group characteristics, while model 3 combines models 1 and 2 and addscontrol variables. As expected, business groups put significantly more emphasis on insideadvocacy compared to outside advocacy than both environmental (p < .01) and othergroups (p < .1 in model 1 and p < .05 in model 3). Business groups hence can bedescribed as insiders with privileged access in the climate change negotiations.

Access, however, can also be gained through experience. Beyond group type, experi-ence is significantly and positively associated with a higher degree of ‘insiderness’: thelonger any group has been involved in the climate change negotiations, the more it willrely on inside advocacy, at the expense of outside advocacy (p < .01). Membership,

54

in contrast, has no significant effect on the degree of ‘insiderness’; organisations whosemembers are mainly individuals do not use in less inside advocacy, as hypothesis 3 wouldhave expected.

Finally, organisations from oecd countries are more likely to adopt an outside strat-egy, while those that are present in more than one country are more likely to adopt aninside strategy. Delegation size, in contrast, has no effect on the relative importance ofinside advocacy, which suggests that resources in terms of staff are important for both,inside and outside advocacy.

4.4 Conclusion

A variety of interest groups seeks to influence the international climate change negotia-tions, and does so using different advocacy strategies that include a range of activities, in-cluding direct contacts with negotiators, media interviews, press conferences, side events,and demonstrations. Drawing on research on domestic interest groups, and using noveldata from a survey conducted among different types of ngos active in the climate changenegotiations, this paper demonstrates that advocacy at the international level resemblesdomestic advocacy. As at the domestic level, there is systematic variation in advocacybehaviour between groups, with environmental groups using relatively more outside ad-vocacy, and business organisations rather engaging in inside advocacy. More noteworthyis that the paper also finds an effect of experience on the choice of advocacy strategy,over and above group type: the more experienced a group, the more it will rely on insideadvocacy, and the less on outside advocacy. Contrary to the theoretical expectationsand other studies, the analysis finds no effect of membership on advocacy behaviour.Groups with individuals as members are not more likely to adopt an outsider strategy.The findings of this paper therefore support arguments centered on access: Organisationsseemingly prefer inside advocacy, but are not always able to gain the necessary access topolicy makers. How they balance the logics of influence and membership, in contrasts,seems to be play a lesser role.

Nonetheless, most organisations do not exclusively rely on one advocacy strategy, andengage in both, inside and outside advocacy. The latter is in fact more widespread thaninside advocacy, across all groups. The greater reliance on outside advocacy may bedue to two factors. On the one hand, outside activities such as side or parallel eventsalso provide opportunities to network and to increase the group’s visibility, which mayserve group goals other than influencing policy and is popular among a diverse range ofngos. On the other hand, an outside strategy may be less a choice of interest groupsthan a response to government restrictions on the use of inside advocacy. Removing suchrestrictions, as is currently discussed in the negotiations, is thus a welcome development.Nonetheless, attention needs to be paid to the differing degrees with which inside advo-

55

cacy is used among different groups, so as not to favour certain groups that are already(better) able to circumvent barriers to influencing the process from within. In 1960,Schattschneider (1960: 35) famously pointed to the ‘heavy upper-class accent’ of interestrepresentation in the United States; research has since confirmed the privileged positionof business groups in domestic policy making (e.g. Dür and Mateo 2010; 2012). Similarobservations can be made for international policy making: business groups enjoy betteraccess to policy makers in the climate change negotiations, even if experience improvesaccess for all types of organisations.

To what extent this better access of business groups also translates into more influ-ence, however, is not a question that can be answered with the data used in this analysis.Nonetheless, if we assume that advocacy activities differ in their effectiveness, it is im-portant to take into account variation in the use of advocacy strategies. Since the valueof non-state participation in intergovernmental negotiations stems from the diversity ofviews and ideas that different stakeholders can bring to the process, we need to askwhether all views and ideas have the same chance of being heard. The findings hereare hence relevant for debates on the legitimacy of the negotiations, as well as, moregenerally, on the role of civil society in environmental governance.

56

Appendix

Figure 4.5: Histogram for Mean Use of Inside and Outside Advocacy

avg.

insider

avg.

outsider

‘insiderness’

engos

bin

gos

other

grou

ps

experience

indiv.mem

bers

deleg.size

oec

d

int.

office

avg. insider 1avg. outsider 0.561 1‘insiderness’ 0.672 -0.050 1engos -0.097 0.108 -0.256 1bingos 0.062 -0.156 0.225 -0.443 1other groups 0.058 0.000 0.107 -0.743 -0.271 1experience 0.248 -0.003 0.245 0.104 0.066 -0.161 1indiv. members 0.050 0.071 -0.026 0.135 -0.308 0.085 -0.078 1delegation size 0.188 0.277 0.014 -0.006 0.036 -0.021 -0.021 -0.037 1oecd -0.111 -0.130 -0.053 -0.180 0.192 0.050 0.137 -0.070 0.096 1 .111int. office 0.184 0.181 0.097 0.004 -0.068 0.046 0.022 0.035 0.183 0.085 1

Table 4.4: Correlation of Variables

57

5 | Responsiveness or Influence?Whom to Lobby in InternationalClimate Change Negotiations

Carola Betzold

This paper analyses the lobbying behaviour of non-governmental organisa-tions (ngos) during international climate change negotiations. With limitedresources, ngos need to strategically invest these so as to maximise theirimpact. A key decision here concerns the target of lobbying efforts: Whomdo ngos lobby, and why? I contrast two possible explanations for choosingwhom to lobby: influence and responsiveness. Accordingly, ngos can focuson responsive targets that are likely to bring ngo input to the table; or theycan focus on influential targets whose voice is heard at the negotiation table.

I test these two explanations with data from a survey of ngos active ininternational climate change negotiations, and find that ngos strategicallytarget their lobbying efforts, approaching both responsive and, more impor-tantly, influential delegations. However, ngos primarily contact their homedelegation. This is in particular true for national groups, while internationalorganisations take into account more factors.

58

Non-governmental organisations (ngos)1 have become an integral part of global envi-ronmental governance. In particular since the 1992 Rio Summit, multilateral negotiationson the environment have seen an unprecedented growth in ngo actor participation. ngosspend considerable time, effort, and money on attending these negotiations with a viewto incorporate their perspectives and positions into the negotiating process as well as itsoutcome.

While the enormous growth of ngo participation has attracted much academic inter-est, research to date almost exclusively focuses on assessing whether ngos matter, and ifso, to what extent (see e.g. Albin 1999; Arts 1998; Betsill 2006; Betsill and Corell 2008;Böhmelt and Betzold 2012). This ex post evaluation of ngo influence, however, neglectsimportant questions on ngo behavior at the negotiations themselves. ngos, as rationalactors, seek to maximise their influence, and hence need to make strategic decisions onhow to best spend their limited resources. A key choice here concerns the target of theiradvocacy efforts: Among the many governments represented in negotiations on globalenvironmental problems, who should an ngo try to influence? Is time and money bestspent on persuading decision-makers with opposing interests to take positions more inline with the ngo’s preferences? Or may it be a better choice to strengthen governmentswith similar positions? Power may also matter: Intuitively, one would expect ngos toseek out powerful countries whose voice matter in the negotiations, either because theirhigh greenhouse gas emissions and economic weight, or because their position in thenegotiation structure, as chairs of negotiation bodies or country coalitions.

With this paper, I seek to contribute to the literature on ngos in global environmentalgovernance by opening the ‘black box’ of the intergovernmental negotiation process itselfand analysing more closely the behavior of different ngo at the international level. I thuscomplement the existing ex post evaluation of ngo influence with an ex ante analysisof the strategic considerations of these groups in their endeavour to exert influence inglobal environmental governance. Based on a survey of almost 150 organisations activein negotiations under the United Nations Framework Convention on Climate Change(unfccc), my analysis indicates that these groups strategically lobby delegations that areresponsive and/or influential. In particular, ngos target democratic delegations as wellas delegations from low-income countries; they also seek out contacts with the delegationrepresenting the host country, as well as delegations representing large greenhouse gasemitters, while shying away from delegations representing small emitters. First andforemost, however, ngos contact their home delegation, the delegation from countries inwhich the group has its headquarters or subsidiary offices. In other words, ngos use theinternational climate summit to extend and strengthen their domestic advocacy. This isespecially valid for groups that work in one country only, while international groups who

1ngos in global environmental governance are variously referred to as ‘non-state actors’, ‘civil so-ciety organisations’, ‘non-governmental organisations’ etc. For simplicity, I will use the terms ‘non-governmental organisations’ or groups throughout the text.

59

have offices in more than one country seem to put more weight on other factors.

5.1 Whom to Lobby?

Although the question of whom ngo lobby has, as yet, received scant attention in in-ternational environmental politics,2 many authors have addressed this question at thedomestic level. In particular, there exists a large body of literature on the choice oflobbying targets in the context of the United States Congress (see for example Hall andDeardorff 2006; Hojnacki and Kimball 1998; 1999; Smith 1995). Even if internationalnegotiations differ from policy-making processes in national legislatures, mainly in thatdecisions are taken by consensus rather than by vote, this literature provides a goodstarting point for exploring patterns of ngo activity in global environmental politics.

5.1.1 Friends or Foes?

Within this literature, policy-makers3 are typically distinguished according to their posi-tion relative to that of the interest group. Policy-makers who a priori share the interestgroup’s position can be said to be ‘friendly’ or ‘allied’ to the ngo, whereas those policy-makers whose positions are a priori in opposition to those of the ngo are termed ‘foes’or ‘opponents’. Intuitively, it seems wiser to spend resources on persuading one’s op-ponents, rather than ‘waste’ resources on policy-makers who anyway support the ngo’sperspective. Yet, arguments can be made for both types of targets, friends and foes.

Why would an ngo lobby an ally whose positions are already in line with those of thengo? At first glance, it makes little sense to invest precious time and effort to obtainsomeone’s support when this support can be expected without any costly intervention.However, a precondition for lobbying is access. Access is more easily gained in the caseof decision-makers with similar preferences, which makes these decision-makers ‘natural’targets (for example Bauer et al. 1963; Baumgartner et al. 2009; Denzau and Munger1986; Mahoney 2008; Mahoney and Baumgartner 2008; Milbrath 1963). Another, closelyrelated, perspective is to conceive lobbying as a tool for mobilisation, or ‘legislativesubsidy’ (Hall and Deardorff 2006; see also Hall and Wayman 1990). If the main purposeof lobbying is to enroll policy-makers as agents on behalf of the ngo, legislative allies areagain the most appropriate target. By providing them with information and resources,friendly policy-makers become (more) active on issues of importance to the ngo, andhence have greater chances to influence legislation in line with the ngo’s preferences.From this point of view, ngos act as ‘service bureaus’ to their policy-making friends

2Gullberg (2008a;b), however, addresses lobbying on climate politics in the European Union.3For readability, I will refer to the potential lobbying targets as ‘policy-makers’ in the national

context, and use the term ‘delegation’ for the international level. Similarly, ngos in national politicstend to be referred to as interest groups. I will thus use the term ‘interest groups’ for the domestic, andthe term ‘ngo’ for the international level.

60

(Bauer et al. 1963: 353; see also for example Baumgartner et al. 2009; Bolton andNash 2010; Denzau and Munger 1986; Hall and Deardorff 2006; Hall and Wayman 1990;Richards and Heard 2005).

Austen-Smith and Wright (1992; 1994) take yet another approach. Although theauthors argue that interest groups in principle lobby their opponents, organised interestsmay be forced to lobby their allies so as to counteract the lobbying activities of competinginterest groups. All else equal, however, interest groups should focus on lobbying theiropponents. In particular in cases of binary decisions such as simple yes/no votes, the aimof lobbying is to increase the size of the supportive coalition, for which it is necessaryto address opponents rather than friends who already are in the supportive coalition(Austen-Smith and Wright 1994; Gullberg 2008a; Hojnacki and Kimball 1998). On theother hand, it is also most difficult and costly to persuade an opponent, and successmuch less certain, which is why it may be unwise to spend limited resources on opponents(Hojnacki and Kimball 1999: 1001).

The argument for lobbying opposing policy-makers holds especially for votes, whereit is important to have a large number of policy-makers cast their votes for the optionthe lobby group prefers. In international negotiations, however, decisions are reached byconsensus, not by vote. This procedure hence resembles early stages in national policy-making, where the discussion aims at defining and framing issues, setting the agenda,and changing the content of proposed legislation. Under these circumstances, it is moreimportant to influence that discussion, rather than expand the size of the supportivecoalition. If that is an ngo’s policy objective, it seems more advisable to target friendlylegislators when providing arguments, issue interpretations, and information that makethe strongest case for the interest group’s position (see Baumgartner and Leech 1998;Hojnacki and Kimball 1998).

Empirical evidence supports arguments for lobbying allies rather than opponents.The literature focused on the United States Congress mainly agrees that interest groupsaccord higher priority to lobbying allies than opponents (e.g. Hojnacki and Kimball1998). These results were confirmed in the context of environmental governance. At theEuropean level, Gullberg (2008a) finds that interest groups lobby mostly their friends,although especially business groups seem to extend their efforts to foes as well (see alsoGullberg 2008b). In the same vein, Richards and Heard (2005: 32) note the importanceof identifying ‘friendly’ politicians and decision-makers in their study of environmentalngos active in European marine protection.

5.1.2 Strong or Weak?

Beyond classifying policy-makers according to their policy position relative to those ofthe ngo, they may also be rated as strong or weak according to their power position in

61

the decision-making process. Again, arguments can be made for both types of decision-makers, the strong and the weak.

Theoretically, each policy-maker, both at the domestic and international level, isequally important, nationally because of the one-man-one-vote principle, internation-ally due to the principle of consensus. In practice, however, some policy-makers mattermore than others, and these influential policy-makers are attractive targets for lobbygroups. Several studies confirm that interest groups are more likely to lobby influentialpolicy-makers. Hojnacki and Kimball (1998; 1999) find that lobby groups in Washingtonprioritise policy-makers that lead committees or are member of sub-committees. Accord-ing to Marshall (2010), interest groups in the European Union similarly prefer to lobbyinfluential members of the European Parliament, mainly those with leadership positionsin party groups or committees.

In international negotiations, in contrast, some authors have argued for lobbying del-egations from small and middle powers, as these are supposedly more susceptible tolobbying. In their review of international negotiations to ban cluster munitions, Boltonand Nash (2010) find evidence to support the argument that middle powers—relativelywealthy but small to medium-sized states—lack the capacity to coerce, and hence tradi-tionally rely on ‘soft power’ that is, persuasion and attraction, as well as internationallaw and international institutions to advance their interests. These characteristics makedelegations from such middle powers attractive for civil society actors, which is why theyhave formed ‘high-impact partnerships’ (Bolton and Nash 2010: 174; see also Risse 2007)

Theory: The Two Goals of Lobbying

What drives lobbying success at the international level from a theoretical perspective? AsHojnacki and Kimball (1998; 1999) write, interest group lobby with two goals in mind:securing agents and gaining broader support, or, in the words of Hall and Deardorff(2006), subsidy and persuasion (see also Hall and Wayman 1990). According to the firstgoal, the targeted delegation needs to respond to the ngo input, such that it brings thegroup’s ideas and interests into the negotiations. Yet, in multilateral decision-making, itis not enough to have one or a few targeted delegations bring ideas and information intothe debate. It is also the remaining, non-targeted, delegations that need subsequentlyto support the position of the targeted delegation(s) for the final negotiation outcome toreflect the ngo’s preferences, in line with the second goal. There are thus two dimensionsto successful lobbying: ngos need to ensure that their preferences are brought to thenegotiation table through individual delegations, but for the final outcome to reflect theirpreferences, they also need to ensure that non-targeted delegations support the targeteddelegation(s).

Given this process, ngos have two pathways for deciding whom to lobby. They can

62

focus on responsive targets that are likely to bring ngo input to the table; or they canfocus on influential targets whose voice is heard at the negotiation table.

According to the first explanation—responsiveness—the key determinant for choosingan appropriate lobbying target is whether that target is a priori susceptible to lobbyingby the ngo. From this perspective, lobbying is targeted at delegations contingent ontheir likelihood to respond (positively) to ngo input.

A delegation’s responsiveness to lobbying by a specific group can be divided intotwo parts, a procedural and a substantive part. On the substantive side, responsivenesshinges on the relative proximity between the positions of the delegation and the ngo athand. If a delegation holds positions similar to those of the ngo, it should be more opento information by that group, since this delegation can then use the information to playa more active role in the negotiations. This is the argumentation brought forward byHall and Deardorff (2006), according to which delegations depend on information andsupport from ngos to (more) actively participate in the policy-making process. Clearly,if the information provided is in line with the delegation’s preferences and helps themmake their case stronger, they will be more likely to listen to the ngo.

On the procedural side, I expect responsiveness to be related to the regime type ofthe country that the delegation represents. Ceteris paribus, delegations from democraticcountries should be more willing to engage with ngos. ngos are important links betweenpolicy-makers and voters in democratic systems, communicating voter preferences topolicy-makers. Since democratic policy makers depend on voter approval, they shouldbe interested in the positions of the various ngos (and thus voter groups) attending theclimate negotiations. Furthermore, ngos help delegations increase the transparency andthus accountability of international negotiations—a service that is more important todelegations from democratic countries (see Keck and Sikkink 1998; see also for exampleRaustiala 1997; Yamin 2001).

H1a An ngo is more likely to lobby delegations that represent positions similar to thoseof the ngo.

H1b An ngo is more likely to lobby delegations that represent democratic countries.

In contrast, the influence explanation suggests that ngos choose targets whose voicematters. When an influential delegation prefers a certain policy position, it is likely thatthe negotiation outcome reflects their preference. Hence, if the ngo’s preferences aretaken up by a targeted influential delegation, the likelihood that these preferences makeit into the final outcome is high. But which delegations are influential?

First, negotiation theory suggests that delegations vary in general and issue-specificpower resources. General bargaining power refers to differences in political and economicweight at the international level in general, which should also be reflected in the climatechange negotiations. Issue-specific power, in contrast, is bargaining power that is specific

63

to the climate change context. More powerful delegations have more opportunities toinfluence the content and fate of proposed text. Delegations from countries with con-siderable political and economic resources tend to be included in consultations and finaldeal-making (see Yamin and Depledge 2004: 457; Depledge 2005: 124f), and furtherhave the means to pressure other delegations into supporting an agreement. In addition,issue-specific resources may translate into leverage over the successful implementationof an agreement. In the climate change negotiations, for example, large greenhouse gasemitters can contribute most to solving the problem, whereas without their support, anagreement is unlikely to be effective.

Second, delegations can also use their positions within the negotiation structure toexert influence. Substantive negotiations typically take place in smaller bodies suchas committees or working groups. Delegations presiding over these negotiations areresponsible for drawing up the agenda of the respective body, for organising the debate,as well as for presenting substantive proposals and forging agreement on these proposals(Depledge 2005: 41ff). Presiding delegations thus have some leeway when writing jointtext and may use their role as a basis for exercising influence (see Depledge 2005; Lang1989; Tallberg 2010). Similarly, leaders of negotiation coalitions such as the Allianceof Small Island States (aosis) or the group of Least Developed Countries (ldcs) havemore opportunities to shape decision-making. Delegations leading such sub-groups are,just as chairs of working bodies, responsible for structuring debates within the sub-group and crafting joint positions. Furthermore, coalition chairs are more likely to beinvolved in case of consultations in small groups such as the so-called ‘Friends of theChair’ (see Depledge 2005; Yamin and Depledge 2004; see also Hojnacki and Kimball1998; 1999). Delegations that hold such positions should thus be particularly attractivelobbying targets for ngos.

H2a An ngo is more likely to lobby a delegation the stronger the (issue-specific and/orgeneral) bargaining power of the country the delegation represents.

H2b An ngo is more likely to lobby delegations that chair negotiation bodies or coali-tions.

5.2 Data and Method

In order to test the above expectations, I use data from a survey among accredited non-governmental observer organisations in the international climate change negotiations.The questionnaire asks participants about their activities at the yearly climate summits,the Conference of the Parties (cop) of the United Nations Framework Convention onClimate Change (unfccc), and was sent to all accredited non-governmental observer

64

organisations in May and November 2011;4 147 fully completed questionnaires were ob-tained and are used for the analysis in this paper. Despite this relatively small samplesize, respondents vary in important aspects: all ngo constituencies are represented;5 theorganisations are geographically diverse, and also differ with regard to size and resources.

The questionnaire asks respondents to specify with which governments, if any, theirorganisation had contacts during the last unfccc cop their organisation attended, aswell as during the year preceding that summit.6 Based on this question, I can construct adataset with the respondent-country pair as the unit of analysis, similar to the design usedin Hojnacki and Kimball (1998; 1999). The dependent variable is then a dummy variablethat takes a value of one if country i was contacted by organisation j at the last cop

or the preceding year, and zero otherwise. Since contacts may also have been initiatedby governments, a third questionnaire item asks respondents which governments, if any,contacted their organisation, either during the cop or the year before. If an organisationreports contacts with country i, and at the same time reports being contacted by countryi, this contact is excluded and the dependent variable takes a value of zero.

It should be noted that, overall, ngos report only very few contacts with governmentdelegations in the survey. As table 5.1 shows, the number of contacts ranged between 0and 28, and between 0 and 21 when excluding contacts that may have been initiated bygovernments. The average respondent contacted 3.2 government delegations, where 3.3delegations were, on average, contacted at the last unfccc cop they attended, while2.7 contacts occurred during the preceding year, mostly with the same countries. Withalmost 200 parties and observer states, then, the probability for any delegation to belobbied by one of the responding organisations is just below 2%.

last cop preceding year overall*

mean 3.75 3.31 3.18median 3 2 0mode 1 1 2

minimum 0 0 0maximum 27 28 21

*This excludes contacts that may have been initiated by governments.

Table 5.1: Number of reported contacts with government delegations, by ngo (N=147)

I have two measures of responsiveness, one for its procedural, the other for its substan-tive aspect. For the substantive aspect of responsiveness, I use the similarity of positionsbetween ngos and countries. To capture the similarity of positions, I ask respondents

4Universities were excluded from the sample based on the assumption that they mainly attendnegotiation sessions for research purposes and not for lobbying. The survey is the same as the one usedin chapter 4.

5The environmental and youth constituencies are overrepresented, while organisations with no con-stituency are underrepresented. All other constituencies are roughly proportionately represented.

6Note that over 85 percent of respondents indicated that they last attended a cop in 2009 or later.

65

which of the major negotiation groups7 have positions similar to those of their organi-sation. The variable ‘similarity of positions’ accordingly is a binary variable that takesthe value of one if country i is in any coalition mentioned as having similar positions tothose of organisation j.

The procedural aspect of responsiveness is measured by countries’ democratic status.Because of its broad temporal and geographic scope, I use the classification of countries asnot free, partly free, and free by Freedom House as reported in the Quality of GovernmentDataset (Teorell et al. 2011) and on the Freedom House website.8

To measure delegations’ influence in the negotiations, I use three indicators: twofor bargaining power and one for delegations’ positions in the negotiation structure. Idistinguish between general and issue-specific bargaining power; for the former, countries’per capita gross domestic product (gdp) is used as proxy, for the latter, greenhousegas emissions. To avoid correlated predictors, I use per capita gdp, which captures acountry’s level of wealth, and overall carbon dioxide (CO2) emissions, which captures acountry’s contribution to anthropogenic climate change and thus its role in mitigation.9 Isplit countries into three groups for both income and CO2 emissions, whereby countries inthe lowest quartile are coded as poor or low emitter, respectively; countries in the highestquartile as rich or high emitter; and countries in the remaining quartiles as mid-incomeor medium emitter. This increases the comparability of the explanatory variables, aswell as accounts for the large range and skewed distribution of the two variables.

A delegation’s position in the negotiation structure is proxied by a binary variable,chairmanship. A delegation is classified as chair if it is either member of the cop Bureau,including the president of the cop,10 or chair of one of the main coalitions.11 Because of

7The negotiation groups in the questionnaire include the African Group; aosis; the BolivarianAlliance of the Peoples of our Americas (alba); the Central Asia, Caucasus, Albania and Moldovagroup(cacam); the Coalition of Rainforest Nations; the Environmental Integrity Group; the EuropeanUnion; the Least Developed Countries; the Organization of the Petroleum Exporting Countries (opec);and the Umbrella Group. Respondents could also specify other groups or single countries.

8See http://www.freedomhouse.org.9Economic data is from the World Bank’s World Development Indicators and the International Mone-

tary Fund’s World Economic Outlook Database (see http://databank.worldbank.org/Data/Home.aspxand http://www.imf.org/external/pubs/ft/weo/2011/02/weodata/index.aspx.); data on CO2 emis-sions is from the World Bank and the United States Energy Information Administration (see http://205.254.135.7/cfapps/ipdbproject/IEDIndex3.cfm?tid=90&pid=44&aid=8). For all yearly data, Iuse the year in which the cop that the respondent indicated as the last that their organisation tookplace. In case no data is available for that year, I use data from the closest available year.

10The Bureau of the cop consists of the President of the cop; seven vice-presidents; the chairs ofthe two subsidiary bodies of the Convention; and a rapporteur. The Bureau assists the President withmanaging the negotiation process. See http://unfccc.int/bodies/body/6430.php.

11Data on Bureau membership was retrieved from the official unfccc cop reports, available from theunfccc website at http://unfccc.int/documentation/documents/items/3595.php. While the Groupof 77 and China (g77), the European Union (eu), and aosis provide a list of all coalition chairs ontheir website (http://www.g77.org/doc/presiding.html for the g77, http://www.consilium.europa.eu/council/presidency-websites.aspx for the European Union, and http://aosis.info/ for aosis.),the cop reports were used to obtain that information for the group of Least Developed Countries andthe African Group. The report lists the countries delivering group statements, which are usually thecoalition chairs; Note that no information on presidencies could be found for some coalitions, notably

66

the special importance of the cop presidency, usually the country hosting the summit, Ihave a separate variable for the cop president or host.12

ngos may simply extend contacts and lobbying efforts from the national to the in-ternational level, and lobby primarily the delegation from their respective home country,irrespective of the expected responsiveness or influence of that delegation. I hence con-trol for geographical presence, that is, whether organisation j has its headquarters or asubsidiary office in country i. Under the unfccc’s Kyoto Protocol, only the countrieslisted in Annex I of the Protocol—mostly rich Western countries—have legally bindingcommitments to reduce their greenhouse gas emissions, which may make these countries’delegations particularly interesting for ngos. I therefore control for being an Annex Icountry. Finally, I need to control for certain ngo characteristics, since some groups maybe more active in lobbying delegations, resulting in an overall larger probability of beinglobbied by that group. I expect the delegation size of an ngo to increase the level of lob-bying for that group, as more representatives should translate into more interactions withgovernment delegations. Similarly, the longer an ngo has been attending climate changenegotiations, and the more experience with the climate change cops it has, the easier itshould become to contact government delegates. More experienced groups should thusalso be able to interact with more government delegations.

As table 5.2 shows, ngos mostly contact delegations from countries that advanceddifferent positions, which also made up the largest proportion of all delegations (73percent). Only one third of all reported contacts are with delegations promoting positionssimilar to those of the respective organisations, which is roughly proportionate to thenumber of delegations that are perceived as advancing similar positions (27 percent ofthe delegations). About 25 percent of the delegations represent countries rated as notfree; 30 percent, countries rated as partly free; and 44 percent, countries rated as free.Yet, only 15 percent of the reported government contacts occur with delegations fromcountries rated as not free, which compares to 20 percent of contacts with delegationsfrom partly free countries, and 65 percent of contacts with delegations from countriesrated as free.

Similarly, a disproportionate number of contacts are with delegations that chair coali-tions or negotiation bodies: 14 percent of all contacts, compared to 8 percent of alldelegations. The difference is even more striking when looking at the delegation rep-resenting the cop presidency. There is only one cop president out of the almost 200parties per cop, that is, the presidency makes up only .5 percent of all delegations, yet4 percent of all contacts are with the cop president.

Since income level and CO2 emissions are measured in quartiles, the distribution across

for the Umbrella Group and opec.12Rarely, the country serving as president of the cop is not hosting the summit. For these cases, the

variable takes the value of one for both, the cop president and the host country.

67

% of delegations % of contacts

Relative position(Different position) 72.89 66.46Similar position 27.11 33.54

Democracy(Not free) 25.39 14.58Partly free 30.89 20.00Free 43.72 65.42

Negotiation positionChair 8.00 13.54Presidency 0.54 4.17

Income levelPoor 26.19 18.96(Medium) 49.71 35.83Rich 24.10 45.21

CO2 emissionsLow 24.39 5.21(Medium) 50.24 39.38High 25.37 55.42

ngo presenceYes 1.50 18.96(No) 98.50 81.04

Annex IYes 22.06 48.75(No) 77.94 51.25

Table 5.2: Distribution of delegations and contacts according to country-level variables. Refer-ence levels in parentheses. (N= 27’232).

these categories is roughly 25, 50, and 25 percent, with slight variations due to missingvalues. The contacts in these respective categories, in contrast, are unequally distributed.Delegations from middle-income and rich countries are very roughly contacted to thesame extent (36 percent and 45 percent, respectively), while contacts with delegationsfrom poor countries account for less than one fifth of all contacts (19 percent). Thedifference is even more pronounced for CO2 emissions: Only 5 percent of the contactedgovernment delegations are from low-emitting countries, compared to 39 percent and 55percent, respectively, from medium and high-emitting countries, respectively.

Only in 1.5 percent of all ngo-country dyads have the ngo headquarters or a sub-sidiary office in the respective country, yet 19 percent of all contacts are with delegationsfrom countries in which the ngo is present. Finally, 22 percent of all delegations repre-sent Annex I countries, while almost 50 percent of the contacts are with these Annex Idelegations.

5.3 Results

Before presenting the results from the statistical analysis, I show in figure 5.1 whichcountries are contacted most often (by at least five ngos). At first glance, it seems

68

likely that bargaining power and being cop president play a role. The European Unionand the United States are named most frequently, by 28 and 25 ngos, respectively,out of 147 groups in the sample. These two countries are followed by several largeand influential countries like the United Kingdom, Germany, Canada and Australia, aswell as the presidents of the past climate summits, South Africa, Mexico and Denmark.Further down the list, however, there are also a number of more unexpected countries,such as Indonesia, Congo (Kinshasa), or Niger, each of which was reportedly contactedby between five and nine organisations in the sample.

0

5

10

15

20

25

30

EU

USA UK

South A

frica

Mexico

Germany

Canada

Australia

Denm

ark

Brazil

Kenya

Norway

China

France

Indonesia

Netherlan

ds

Japan

India

Sweden

Banglad

esh

Switzerlan

d

Congo (Kin

shasa)

Ghana

Russia

Finlan

d Philippin

es

Bolivia

Ecuador

Niger

Papua N

ew Guin

ea

numbe

r of con

tacts

Figure 5.1: Countries contacted most frequently (over five contacts)

5.3.1 Clustered Logistic Regression

To test more systematically which factors explain ngos’ choice of lobbying targets, I uselogistic regression with robust standard errors, clustered by countries. This seems theappropriate modeling approach, given the binary dependent variable and the setup of thedata: as each responding ngo is paired with 189 countries,13 the observations are notindependent. Although the baseline probability of being lobbied is very low (see table5.1 above), the statistical analysis (see table 5.3) indicates that the odds of being lobbiedincrease significantly with the degree of responsiveness and influence.

Contrary to what H1a posited, ngos do not focus their lobbying on delegations thatadvance similar positions. Although the odds of being contacted by an organisation areabout one third higher for delegations that advance positions similar to those of theorganisation, the difference is not statistically significant. ngos do, however, lobby dele-gations that come from democratic countries, as H1b expected. Compared to delegationsrated as not free, delegations from countries rated as partly free or free have higher oddsof being lobbied, yet the difference is much larger between delegations from free andunfree countries, the odds of the former being 84 to 106 percentage points higher than

13The exact number of observations per ngo varies because of missing values. Missing values alsoexplain why ngos are paired with 189 countries, rather than the currently 195 Parties to the Convention.For instance, there is no democracy data available for Niue and the Cook Islands.

69

Model 1 Model 2

odds (robust odds (robustratio std. err.) ratio std. err.)

Country-level variablesRelative position(Different position)Similar position 1.325 (0.269) 1.325 (0.269)

Democracy(Not free)Partly free 1.223 (0.219) 1.217 (0.219)Free 2.063** (0.493) 1.844* (0.451)

Negotiation positionChair 1.228 (0.163)Presidency 4.922*** (1.397)

Income levelPoor 2.252*** (0.504) 2.294*** (0.518)(Medium)Rich 1.052 (0.124) 1.120 (0.132)CO2 emissionsLow 0.250*** (0.052) 0.262*** (0.054)(Medium)High 2.203*** (0.321) 2.142*** (0.306)

ngo presenceYes 10.300*** (1.947) 10.270*** (1.947)(No)

Annex IYes 1.585* (0.342) 1.715* (0.365)(No)

ngo-level variablesngo experience 1.024 (0.023) 1.024 (0.023)ngo delegation size 1.004 (0.005) 1.004 (0.005)

Constant 0.004*** (0.001) 0.004*** (0.001)

+ p < .10, * p < .05, ** p < .01, *** p < .001

Table 5.3: Logistic regression analysis of ngos’ choice of lobbying targets (robust standarderrors adjusted for clustering of observations on ngos; N= 27’232).

the odds of the latter. This difference is statistically significant (p < .01 in model 1, andp < .05 in model 2).

ngos also strategically target their lobbying conditional on delegation’s function inthe negotiation structure, income level, and greenhouse gas emissions. The analysis doesnot find evidence that delegations chairing coalitions or negotiation groups are lobbiedsignificantly more often. Figure 5.1 above, however, suggested that ngos seek to interactwith representatives from the country that presides over the negotiations, which is whyI substitute the variable for chairmanship with a binary variable for cop presidency inmodel 2. According to this specification, the odds of being lobbied increase about fivefoldfor the cop president, which is statistically highly significant (p < .001). This highlightsthe importance of the cop presidency and the leverage the presiding country has overthe negotiations. The findings thus offer partial support for hypothesis H2b: While

70

ngos do seek contacts with the president of the climate summit, they do not generallyapproach coalition chairs or Bureau members other than the cop president significantlymore often.

The odds of being lobbied also depend significantly on a country’s income level; inter-estingly, however, the effect is the opposite of what I hypothesised in H2a: ngos targetdelegations from poor countries significantly more often than delegations from medium-income countries. The odds of being lobbied for the former increase more than twofoldcompared to those of the latter (p < .001). While the odds are also slightly higher fordelegations from rich countries, the difference between medium-income and rich countriesis not significant. Possibly, this finding suggests that per capita gdp does not measurea delegation’s level of influence in the negotiations, but rather proxies its negotiationcapacity. Delegations from low-income countries tend to be small and thus have only fewexperts; accordingly, their need for external support is relatively high, which may makethem susceptible to ngo input (cf. Chasek 2001). In other words, negotiation capacity,as measured by per capita gdp, is a proxy for responsiveness rather than for influence.

Greenhouse gas emissions, in contrast, seem to provide a better measure of influence.As expected in H2a, the odds of being contacted for delegations from countries with highCO2 emissions are more than twice the odds of their colleagues from medium-emittingcountries. At the same time, delegations from countries with low emissions are signifi-cantly less likely to be contacted by ngos, the odds being about 75 percentage pointslower than those of delegations from the medium category. Both these differences—between high and medium emitters, and between low and medium emitters—are highlysignificant (p < .001). Overall, then, H2a seems to be correct: the likelihood of beinglobbied increases with a delegation’s bargaining power, yet bargaining power, in the cli-mate change negotiations, appears to hinge on CO2 emissions rather than per capitagdp.

I also find significant effects for my control variables. While an organisation’s ex-perience and number of delegates at the climate summit does not explain variation inngo-government contacts, whether a country is listed in Annex I of the Convention does.Delegations from Annex I countries are contacted significantly more often, the odds beingabout 59 to 72 percentage points higher (p < .05). More importantly, however, ngoscontact their home delegations. If an organisation has its headquarters or a subsidiaryoffice in country i, the odds of that organisation contacting country i increase abouttenfold compared to a country in which the organisation is not present. This effect isby far the largest effect in the regression analysis. It indicates very clearly that ngosfirst and foremost extend their lobbying activities from the domestic level to the inter-national level. While the climate summits bring together representatives from roughlyevery country in the world, and thus afford many lobbying opportunities for the ngosattending these summits, contacts still are mainly focused on group’s home countries.

71

5.3.2 International vs. National Groups

Given that domestic ties play such an important role in explaining contacts betweenngos and government delegations, we may wonder to what extent national groups thatwork exclusively in one country behave differently from international groups that arepresent in more than one country. Since the latter have no single home delegation, theymay generally be less focused on lobbying delegations from countries in which they haveoffices. I therefore split the sample into international groups (45 percent of the groups inthe sample) and national groups (55 percent), and re-run the above analysis separatelyfor international (models 1 and 2) and national (models 3 and 4) groups, as presented intable 5.4.

The results indicate that indeed, international groups differ from national groups; thelatter are to a much larger extent focused on the delegation from their respective homecountry, whereas the former tend to condition their government contacts on a wider rangeof factors—although contacts with the delegations from those countries in which theywork remain very important.

For ngos with an international presence (models 1 and 2), the results resemble theabove results for the entire sample: They report significantly more contacts with delega-tions from democratic countries; the delegation presiding over the cop; with delegationsfrom poor countries; with delegations from Annex I countries; and with delegations fromcountries with high levels of greenhouse gas emissions, while avoiding delegations fromlow-emitting countries. Most importantly, again, they contact delegations from countriesin which they work. As above, the odds of being lobbied for those countries in whichthey are present are much larger—about eight times as large—compared to those ofcountries in which they are not present, a difference that is statistically highly significant(p < .001).

In contrast, national organisations appear to target their lobbying based on fewerfactors. The democratic nature, income level, or listing in Annex I of the Convention,of the country a delegation represents does not increase the odds of being lobbied bya national organisation. Influence, however, still matters. Just as international groups,national organisations approach high greenhouse gas emitters while shying away fromlow emitters (p < .001); and, even more so than international groups, they seek out thecop president: the odds of being lobbied by a national group are about six times thoseof other delegations (p < .001). Most importantly, however, they contact their homedelegation: the odds of being lobbied for the delegation from which the organisationcomes are more than 17 times those of other countries—the difference is about twicethat for international organisations, and is statistically highly significant (p < .001).

The results, both for the entire sample and the split sample, are robust across anumber of specifications. They do not change when excluding respondents who reported

72

the last cop they attended was before 2007; when only looking at contacts at the lastcop, that is, when excluding contacts that had occurred in the year prior to that cop;and when including in the dependent variable contacts that may have been initiated bygovernments. In the latter case, however, the effect of presence becomes much larger,which suggests that governments, if they contact ngos, primarily approach their domesticgroups.

73

International groups National groups

Model 1 Model 2 Model 3 Model 4

odds (robust odds (robust odds (robust odds (robustratio std. error) ratio std. error) ratio std. error) ratio std. error)

Relative position(Different position)Similar position 1.144 (0.335) 1.144 (0.334) 1.548 (0.426) 1.549 (0.429)

Democracy(Not free)Partly free 1.401 (0.295) 1.397 (0.295) 1.020 (0.296) 1.013 (0.295)Free 2.143** (0.526) 2.007** (0.517) 1.935 (0.799) 1.651 (0.690)

Negotiation positionChair 1.167 (0.200) 1.258 (0.259)Presidency 3.328** (1.519) 6.462*** (2.355)

Income levelPoor 3.051*** (0.794) 3.093*** (0.807) 1.516 (0.572) 1.542 (0.592)(Medium)Rich 1.117 (0.176) 1.161 (0.191) 0.988 (0.174) 1.076 (0.175)

CO2 emissionsLow 0.256*** (0.064) 0.263*** (0.066) 0.239*** (0.086) 0.257*** (0.090)(Medium)High 2.054*** (0.364) 2.025*** (0.355) 2.392*** (0.572) 2.286*** (0.527)

ngo presenceYes 7.870*** (1.847) 7.878*** (1.841) 17.200*** (5.505) 16.900*** (5.510)(No)

Annex IYes 1.830+ (0.567) 1.918* (0.588) 1.339 (0.410) 1.495 (0.447)(No)ngo experience 1.031 (0.030) 1.031 (0.030) 1.014 (0.036) 1.014 (0.036)ngo delegation size 1.004 (0.005) 1.004 (0.005) 0.973 (0.017) 0.973 (0.017)

Constant 0.004*** (0.001) 0.004*** (0.001) 0.006*** (0.003) 0.006*** (0.004)

N 12’231 12’231 15’001 15’001

Table 5.4: Logistic regression analysis of ngos’ choice of lobbying targets with robust clustered standard errors, for international (N= 12’231) and national(N=15’001) groups

74

5.3.3 Survey Results

As a further robustness check to validate the results obtained from regression results,the questionnaire asks respondents directly which factors play a role in the decision tocontact a specific delegation. I provide five different factors—the similarity of positions,the delegation’s influence, whether it chairs meetings or coalitions, personal acquaintancewith delegates, and whether the organisation is from the country in question—and askrespondents to rank the importance of each of these factors on a four-point scale rangingfrom not important to all to very important.

Figure 5.2: Importance of different factors for ngos’ choice of lobbying targets (relative fre-quency per category; N=122)

From the responses to this questionnaire item, a similar picture emerges as from theregression analysis. As can be seen from figure 5.2, ngos tend to contact delegates thatthey know personally. For more than 60 percent of respondents, personal acquaintanceand geographic presence is very important or important. It is thus not surprising thatngos interact primarily with delegations from the country in which their organisation isactive. About as important as presence is how influential a country is. ngos address del-egations that are thought to have a lot of influence—two thirds of the respondents rankinfluence as very important or important for their organisation’s advocacy. In contrast,whether a delegation is chairing negotiation meetings or coalitions, or whether a delega-tion represents positions similar to those of the organisation are not as important. Justunder (similar position) or over (chairmanship) half of the respondents perceive thesefactors to be very important or important. Respondents could also specify additionalfactors that they take into consideration when approaching delegations. Here, respon-dents referred to delegations’ ‘willingness to communicate with environmental ngos’;their having opposite positions; and their taking ‘a popular (negative or positive) standin the negotiations’.

75

5.4 Discussion and Conclusion

With limited resources at their disposal, ngos need to be selective in their advocacy.In order to maximise their impact, ngos should strategically seek out government del-egations among the almost 200 parties that participate in international climate changenegotiations. This paper puts forward two alternative explanations for whom ngos lobby.On the one hand, ngos may aim at responsive countries that are likely to take up thelobbyist’s arguments and information; alternatively, they may rather seek out influentialdelegations whose voice is heard at the negotiation table.

So whom do ngos lobby in international climate negotiations? When testing thetwo explanations empirically, the results suggest that both factors—responsiveness andinfluence—matter. The two goals of lobbying, mobilisation and persuasion, seem to becomplements rather than contrasts.

First, ngos lobby responsive delegations, that is, those from countries rated as demo-cratic and free, as hypothesis 1b expected. In contrast, and against my expectation inhypothesis 1a, the similarity of positions does not have an effect on the likelihood of be-ing lobbied. As opposed to previous studies, I do not find evidence that interest groupslobby their allies in the climate change negotiations. This non-finding may be due to twofactors: First, I ask for perceptions of similar positions at the level of the coalition, whilethe analysis is carried out at the level of the country. My empirical design may henceunderestimate the importance of positions—although respondents, when asked directly,do not assign high priority to positions (cf. figure 5.2). Second, and more importantly,the climate agenda has grown enormously and covers a variety of issues, from foreststo insurance against loss and damage to bunker fuels from air and maritime transport.While governments may promote interests that are in line with an ngo in one area, theymay have opposing interests in another area; who is an ally may thus depend on thespecific issue under negotiation. Even if ngos lobby their allies, but alliance patternsare issue-specific, I am unable to identify such patterns with my research design.

At the same time, ngos approach influential delegations, that is, those delegationsrepresenting large greenhouse gas emitters, while shying away from low emitters, in linewith the expectation that issue-specific power matters (hypothesis 2a). At the sametime, and as hypothesis 2b suggested, ngos seek out the delegation presiding over thecop, which illustrates the large degree of leverage that the cop president has over thenegotiation process and outcome. Delegations from low-income countries, in contrast towhat I expected with respect to generally bargaining power (hypothesis 2a), also seemto be attractive lobbying targets. Possibly, then, income measure negotiation capacityand thus indirectly responsiveness rather than influence.

Lastly, and most importantly, ngos appear to simply extend their domestic activitiesto the international level by primarily addressing the delegation from their home coun-

76

tries, countries in which they have their headquarters or a subsidiary office, regardlessof that delegation’s degree of responsiveness and influence. The effect is very large, andshows the crucial importance of domestic ties, including in international politics. ngosmay hence simply opt for the ‘easy’ way out, as suggested by Bauer et al. (1963), ap-proaching those delegates that they know personally, an interpretation that is supportedby respondents’ emphasis on personal acquaintance (see figure 5.2). Alternatively, thestrong role of domestic ties may indicate that ngos prioritise national legislation overinternational decisions. The climate change negotiations proceed only very slowly, andsubstantive commitments remain rare. At the last cop in Doha, only those countriesthat already have considerable domestic legislation in place, notably the European Union,agreed to a second commitment period under the Kyoto Protocol. Yet, if the interna-tional negotiations mainly serve to codify domestic legislation, it is but rational for ngosto focus lobbying on their home governments.14

This paper is a first attempt to scrutinise the relationships between ngos and gov-ernment delegations from the ngos’ point of view. While the question of why govern-ments take non-state actors aboard has recently attracted some interest (see for exampleBöhmelt 2012), the international lobbying activities of ngos have so far received less at-tention. It is notoriously difficult to obtain reliable empirical data on the many contactsbetween non-governmental representatives and government delegates that often occur athaphazard on the corridors of the climate change summits. The annual climate meetingsattract thousands of negotiators and ngo representatives, and it is all but impossible totrace who talked to whom, when, how, and why. Survey data can to some extent remedythe situation, although the survey method is, of course, not without fault: The almost 150organisations in the sample represent just 10 percent of all accredited non-governmentalobserver organisations, and respondents may not have comprehensively listed all govern-ment delegations their organisation contacted. Finally, we have no information on thenature of the contact, the intensity and frequency with which groups met with individ-ual government delegations, or the content of that contact. Further research, includingdetailed case studies of individual organisations, is thus needed to better understand theinteractions between governments and ngos in international environmental negotiations.This analysis provides a starting point for that endeavour.

14See also personal communication with David Suzuki Foundation, 09 February 2012.

77

6 | The Media in International Ne-gotiations: The Case of ClimateChange

Carola Betzold, Vally Koubi and Thomas Bernauer.

One manifestation of the larger phenomenon of mediatisation of global policy-making in recent decades is the very strong presence of mass media represen-tatives from around the world at global climate change conferences, and theextensive news media coverage of those events. The most prominent mech-anism for interaction between governments and the mass media on thoseoccasions is press briefings. We explore patterns in the use of press briefingsby governments in this context, arguing that governments use press briefingsboth for reasons of appropriateness and consequence. Using information onpress briefings collected from the United Nations Framework Convention onClimate Change (unfccc) archives, we find limited support for the argumentthat democracies, which are more likely to be motivated by a logic of appro-priateness relative to other political systems, are more likely to initiate pressbriefings. We also find only moderate support for the logic of consequenceargument, which holds that government choices to initiate press briefings aremotivated by physical vulnerability to climate change as well as economicvulnerability to strict climate policies. Other factors, such as capacity anda government’s function in the negotiation structure, seem to play a moreimportant role for explaining government decisions to hold press briefings.

78

Upward denationalisation refers to the shift of policy-making to the internationallevel: more and more issues are negotiated, regulated and decided at the internationallevel. Environmental politics is a prime example of this trend. Mitchell (2013; 2003), forinstance, counts almost 30 new multilateral environmental agreements, amendments, orprotocols per year for the past two decades, as opposed to less than ten in the 1950s.

While the international level is the appropriate level for decision-making for globalenvironmental problems, intergovernmental decision-making processes tend to have longchains of delegation and are often criticised for being non-transparent and undemocratic(Moravcsik 1997; Rieff 1999; Rootes 1999). In response to this ‘democratic deficit’, gov-ernments have opened up intergovernmental negotiations to civil society; accordingly,civil society participation in global environmental governance has grown tremendouslyover the past decades (Arts 2005; Charnovitz 1997; Raustiala 2001). At the same time,media attention to international environmental politics has increased. Yet while the roleof civil society in global environmental governance has become a prominent subject forresearch (Bernauer and Gampfer 2013; Betsill and Corell 2008; Böhmelt 2012; Gulbrand-sen and Andresen 2004; Hjerpe and Linnér 2010; Newell 2000; Orr 2005; 2006; Schroederand Lovell 2012), the strong presence of the mass media in global environmental gover-nance efforts has received only scant attention in the literature (but see Carpenter 2001;Newell 2000). Just as in national policy-making, however, the media play a key roleas intermediary between policy-makers and the public. By increasing transparency, themedia contribute to improving the democratic accountability and legitimacy of globalenvironmental governance.

Despite the lack of scholarly attention, policy-makers seem to be aware of the impor-tant role the media play, as the annual climate change summits illustrate. These Con-ferences of the Parties (cops) have become ‘environmental mega-conferences’ (Gaventa2010), attended not only by hundreds of negotiators and civil society observers, but alsoby hundreds of journalists. The 2011 climate summit in Durban, for instance, attracted1200 formally accredited journalists—in addition to around 5000 government delegatesand another 5000 civil society representatives.1 Governmental and non-governmentaldelegates alike actively reach out to these journalists by inviting them to press confer-ences, or press briefings (we use these terms interchangeably): 30-minute meetings withthe press where negotiators and observers can inform the media, and through them, thepublic, about developments in the negotiations, communicate positions and preferences,and answer to questions by journalists.

Why do governments organise press conferences? Are these briefings element of a stan-dard operating procedure that governments follow automatically, or do press conferencesserve other functions? In this paper, we explore the ‘landscape’ of press conferences in

1See list of participants, available at http://unfccc.int/documentation/documents/items/3595.php.

79

climate change negotiations, and analyse variation in the use of press briefings across gov-ernments. In particular, we contrast two possible motivations for holding press briefings,a logic of appropriateness and a logic of consequence. Under a logic of appropriateness,governments communicate with the media in press conferences because there is a strong(normative) expectation, in the sense described above: Press briefings are a mechanismto increase transparency and legitimacy and a standard element of policy-making in thedomestic context, at least in democratic settings. Initiating such briefings is hence ‘theright thing to do’—an argument that should in particular count for democratic govern-ments. In contrast, the logic of consequence is instrumental and extrinsic, rather thannormative and intrinsic. It presumes that governments use press briefings to achieveparticular purposes. Under this logic, press conferences are a means to an end, an instru-ment to increase leverage in the negotiations. We expect this logic to be more importantto frontrunner and laggard countries, since these are under most pressure to justify theirposition and to shift the negotiations toward their ideal point. We submit that thesetwo logics are not mutually exclusive, since most political action involves elements ofboth logics. As March and Olsen (1998: 952) note, ‘Political action generally cannotbe explained exclusively in terms of a logic of either consequences or appropriateness’.Nonetheless, the relative importance of the two logics should vary by government typeand position in the negotiations.

As opposed to studies of the media in domestic politics that are typically based oncontent analyses of news media reports, we take the perspective of political actors andcompare all countries globally in their decision to communicate with the media and thepublic via press conferences.2 Using information on press briefings collected from theunfccc archives for all cops in the period from 1995 to 2011, we find that democraticgovernments are less likely to never hold a press briefings, arguably because they believethat press briefings are appropriate, or ‘the right thing to do’. We also find a smallerlikelihood to never organise a press conference for countries more at risk from climaticchanges, as well as that countries with large total CO2 emissions are associated with alarger number of press briefings, presumably in order to promote their frontrunner andlaggard positions, respectively, in the negotiations. However, the effects are rather weak;it thus seems that other factors play a larger role when it comes to deciding whetherto hold a press conference or not. For instance, countries with large delegations andcountries hosting a climate summit communicate more with the media via press briefings.More research is hence needed to shed light on the motivations and expectations ofgovernments when speaking to the press.

2It would be interesting to explain not only which types of governments/countries are initiating pressbriefings, but also which types of media are attending particular types of press briefings organised byparticular types of governments on particular issues. Such research is virtually impossible, however,because there exist no data on participant characteristics and content of press briefings. The unfcccdocuments do not record such information, although webcasts of press briefings have been provided formore recent cops. Moreover, government press briefings are often only open to media representatives(and not to researchers).

80

6.1 Mediatisation and press briefings in climate cops

It is widely agreed that politics is increasingly mediated and mediatised (cf. Strömbackand Esser 2009). While often used interchangeably, the two concepts of mediation andmediatisation are closely related, but distinct. Mediatisation—in contrast to ‘mediation’,which is viewed as a neutral or transparent process—assumes that institutions not onlyconduct their activities through the media, but that in doing so, they become dependentupon and guided by the ‘institutional and technological modus operandi of the media’(Hjarvard 2008: 113; Marcinkowski 2005). Mediatisation is thus ‘an inherently process-oriented concept, focused on how media influence has increased in a number of differentrespects’ (Strömback and Esser 2009: 208), or, put differently, a process ‘in which thereis a development toward increasing media influence’ (Asp and Esaiasson 1996: 80f).Mazzoleni and Schulz (1999: 250) apply the concept of mediatisation to media influenceon politics. They argue that ‘mediatized politics is politics that has lost its autonomy, hasbecome dependent in its central functions on mass media, and is continuously shaped byinteractions with mass media.’ The authors also stress, however, that although politicalinstitutions increasingly are dependent on and shaped by mass media, they still remainin control of political processes and functions. The growing influence of the media indomestic politics has been extensively studied by communication scientists (e.g. Esser2013). Less attention, in contrast, has been paid to the media in international relations.Yet, under conditions of upward denationalisation, more and more decisions are taken atthe international level.

The media follow this upward trend, as the case of international climate change nego-tiations illustrates. The unfccc was signed at the Rio Summit in 1992, and entered intoforce in 1995. Since its entry into force, the Parties to the Convention meet annually atthe cop, typically in late November/early December, to negotiate the implementationof the Convention’s objective, namely, to avoid ‘dangerous’ interference with the climatesystem (United Nations 1992b: art. 2). While climate change has generally receivedconsiderable media attention over the past two decades, so have the climate change ne-gotiations. Existing studies clearly show that news-media reporting on climate changehas increased since 1995, and regularly peaks during the cops, as can also be seen inFigure 6.1 (Boykoff 2007; Dirikx and Gelders 2010; Liu et al. 2011; Schäfer et al. 2011a).

Political actors involved in climate change politics seem to be aware of this mediaattention and to adapt to the rules of the mass media system, in line with the con-cept of mediatisation. cop participants—governments, civil society, and internationalorganisations—actively seek media exposure during the two-week negotiation sessions,including through press briefings. The number of such events has increased considerablyover time (see Figure 6.2); in particular, the last negotiation meetings (cop15 to cop17)have seen a far higher number of press briefings than previous summits. Note, however,

81

Source: Schäfer et al. (2011b)

Figure 6.1: Media attention for climate change in France, Spain, Singapore and the USA

that throughout the history of the climate change negotiations, some cops attracted adisproportionately large number of briefings. In the early period of the negotiations,notably the Kyoto and The Hague cops (cop3 and cop6) each saw just over 100 events,compared to an average number that is clearly below 100, which testifies to the polit-ical importance of these two cops for the climate change regime: At cop3, the KyotoProtocol was adopted, while cop6 remains the only cop to not have reached agreementand to be continued in a second meeting, cop6bis.3 In the last round of negotiations,starting with cop13 in Bali, each cop routinely saw well over 100 press briefings. TheCopenhagen Summit (cop15) breaks the record with over 350 events, while the last twocops saw about 250 events each. Put differently, just under half of all press briefingsfrom 1995 to 2011 took place during the last three cops.

Does the increase in the number of press briefings thus simply correspond to the overallincrease in the cops? To account for the general trend toward ever larger climate con-ferences, we divide the number of press briefings by the number of attending journalists(in 100s). The black line in Figure 6.2 displays the proportion of briefings to journalists.It indicates that press briefings have become more frequent. Compared to only 2.5 press

3On cop6, see e.g. Grubb and Yamin (2001).

82

Figure 6.2: Number of press briefings by cop and actor type

briefings per 100 journalists at cop3 in Kyoto, there were almost 20 briefings per 100journalists at cop17 in Durban.

Who organises these events? Figure 6.3 shows the aggregate number of press briefingsduring all cops, according to actor type, that is, according to who initiated these events:government, ngo, or intergovernmental organisation (igo). Press briefings that were co-organised by different actors are shown accordingly. The figure indicates that less thanten percent of all press briefings were joint press briefings by different actor types. Mostevents—over 800 during the 17 cops from 1995 to 2011, or 43 percent—were initiatedby ngos. Governments follow closely, with roughly 700 events (36 percent). igos weresomewhat less active; they were responsible for about 300 events (16 percent), but tendto co-organise with governments and ngos. Joint press conferences of governments andngos are the most rare, with only nine events (less than one percent). The same holdsfor press briefings co-organised by ngos and igos.

Figure 6.3: Aggregate number of press briefings by actor type

83

Out of a total of 1920 recorded press briefings, 93 percent (1781 events) had a singleinitiator, compared to 137 events that were co-organised by between two (104 events, 5percent) and six (two events, 0.1 percent) actors. It should, however, be noted that thesefigures reflect only imperfectly cooperation since only the main organiser appears in theDaily Programme, even if other governments or ngos are then invited to speak at thepress conference.

We now turn to the 700 events organised by governments.4 Figure 6.4 lists the partiesto the unfccc that have initiated the most press briefings during climate cops. TheEuropean Union is by far the most active party; it invited the media to a total of 135press briefings. The United States ranks second; it held a total of 96 press briefings.Other particularly active governments are Japan (39 press briefings), China (34 briefings),Canada (32 briefings), and Brazil (31 briefings). Among the most active initiators,however, are also a number of smaller countries, such as Bolivia (16 briefings), Ecuador(11 briefings), and Bangladesh (ten briefings).5

0

20

40

60

80

100

120

140

EU

USA

Japan

China

Canada

Brazil

Germany

Indonesia

Bolivia

Mexico

South A

frica

Switzerlan

d Ecuador

France

Polan

d Banglad

esh

India

Denm

ark

Costa

Rica

Netherlan

ds

UK

numbe

r of p

ress briefi

ngs

Figure 6.4: Number of press briefings by most active governments

These numbers indicate that only few countries use the opportunity to hold pressbriefings. Of the 196 countries that are party or observer state to the unfccc, 123countries (63 percent) have never held a press briefing; 31 countries (16 percent) haveheld only one briefing, and 17 countries (nine percent) have held ten press briefings ormore.6

It is worth noting that the two weeks of climate negotiations are extremely busy,with countless meetings, formal negotiations, and informal consultations, and with many

4As we mentioned above we chose to focus on these events because of data availability. While itwould be just as interesting to analyse which ngos hold press conferences, the Daily Programmes justprovide a name for each ngo; further information is restricted to contact information on the unfcccwebsite at http://maindb.unfccc.int/public/ngo.pl?.

5When also including coalitions, the G77 (28 briefings), the Alliance of Small Island States (aosis,22 briefings) and the African Group (17 briefings) would also be on the list.

6Note that these figures include co-organised events.

84

of these activities taking place in parallel. Although the unfccc secretariat providesthe infrastructure and coordinates time slots, organising a press conference still requiresprecious time and effort. Why do governments then invest in organising press briefings?

6.2 Two Logics of Action

International relations scholar distinguish two different incentives, or two logics of action:a logic of appropriateness and a logic of consequence (March and Olsen 1996; 2004). Thesetwo logics put forth different explanations for whether a government decides (or not) toinitiate a press briefing.

The logic of appropriateness holds that political actors behave in accordance withprevailing norms of proper conduct—that is, they seek to ‘do the right thing’ in a givensocial context. What, then, is the prevailing norm of proper conduct in the context ofglobal intergovernmental negotiations? As mentioned above, global (environmental andother) negotiations tend to be criticised as being non-transparent and undemocratic (e.g.Moravcsik 1997; Rieff 1999; Rootes 1999). More transparency does not automaticallyresolve the democracy deficit problem presumably associated with global internationalpolicy-making, but transparency is, arguably, a precondition for accountability and le-gitimacy. Such criticism therefore points to a normative ideal of transparent negotiationprocesses, where the general public (electorates) is informed about the actions of officialsand institutions and can therefore hold decision-makers to account. Since the publicreceives this information typically through the media, there is an expectation towarddecision-makers to interact with the media and allow for more transparency in decision-making processes. Press briefings are element of the standard operating procedure of(democratic) policy-making, at least in the national context, and this element may sim-ply be transferred to the international context.

Although there may also be an element of the logic of consequence in this argument(see below)—governments initiate press briefings to increase transparency and counter-act claims of a democratic deficit—we consider hypothesis 1 to reflect primarily a logicof appropriateness. The public in democratically governed countries is more used tohigher levels of transparency and accountability, relative to the public in non-democraticcountries. We thus expect governments from such countries to have internalised thesecommon expectations and act accordingly at all political levels, including global settingssuch as the climate cops. Hypothesis 1 reflects this argument.

H1 Democratic governments are more likely to initiate press briefings.

The logic of ‘consequence’ or ‘effectiveness’ assumes that governments are rationalutility maximisers: they select strategies and behave in ways that are most likely toachieve a desired goal. In other words, governments do what they think best advances

85

their interests. According to this line of reasoning, press briefings are organised in orderto increase the government’s leverage at the bargaining table.

How can public press conferences help governments make their position in the nego-tiations stronger? Press conferences provide an opportunity for governments to publiclycommit to a certain position; when a government has publicly committed to a position,it is reluctant to move away from that position, since this is interpreted as weaknessand failure and thus met with public disapproval (e.g. Tomz 2007). In other words, bypublicly committing to a position, governments create audience costs and tie their hands(Fearon 1997). If, however, the cost of backing down is high, then the commitment toa particular position becomes more credible and can serve as an informative signal ofresolve toward other governments.

The cost of backing down depends entirely on the governmental win-sets. The conceptof win-sets can be defined as the sum of foreign policy actions at the international levelwhich a government can successfully ratify at the domestic level (Moravcsik 1993; Putnam1988). Hence the win-sets give clues to the content of foreign policy decisions that areto be expected from a government. As a result, the size of win-sets affects both theflexibility and the bargaining power of governments on the international level. Largedomestic win-sets ceteris paribus enhance the flexibility of governments in their dealingson the international stage as well as the feasibility of policy trade-offs and package deals inorder to reach compromises and achieve cooperation between governments. On the otherhand, small win-sets impose stricter limits on the governmental room for manoeuvre, andthus they paradoxically strengthen the international bargaining power of the respectivegovernment (‘paradox of weakness’, see Schelling 1960). The tight domestic constraintsassociated with a small win-set increase the credibility of a government’s threat to walkaway from international negotiations without agreement and to refuse to compromise onits preferences, which is a primary source of international bargaining power (Milner 1997;Moravcsik 1991; 1993; Schelling 1960).7

Consequently, we expect governments to hold more press briefings in order to justifyand indirectly garner support for their positions at the international level when their do-mestic win-set is small. This should be particularly the case when a country is physicallyvulnerable to climatic changes and when a stricter climate agreement is expected to hurta country’s economy.

7One could also argue that press conferences serve as a signaling device vis-à-vis other governmentsand there exist some anecdotal evidence to support such proposition. For instance, on the last dayof the Copenhagen Summit (cop15), as the formal negotiations stalled, the so-called ‘Friends of theChair’—a small group of countries met privately and crafted the brief declaration that was to becomethe Copenhagen Accord. As soon as the Accord was written, the involved government leaders, suchas us President Obama, presented the agreement in press conferences—even before the document wasintroduced in the plenary. Although this ‘going public’ is described as a ‘strategic mistake’ (Dimitrov2010), it meant that opponents to the Accord would publicly be seen as blockers to the agreement andhence their room for manoeuvre was restricted.

86

H2a Countries more vulnerable to climate change are more likely to initiate press brief-ings.

H2b Countries expecting to lose economically with stricter climate agreements are morelikely to initiate press briefings.

6.3 Research Design

6.3.1 Dependent Variable

The unfccc Secretariat publishes a Daily Programme for each day during cops. TheDaily Programme includes, among other items, a list of press briefings that are scheduledfor the day, as well as information on who organises the briefing. This data is freelyavailable from the unfccc website, but has, to our knowledge, not been subject tosystematic analysis. We built a dataset based on these Daily Programmes, focusing onevents organised by governments. To test the above hypotheses, we need informationon which governments held press briefings and on the number of press briefings theyheld. We hence count the number of press briefings organised or co-organised by eachgovernment for each of the cops; accordingly, our unit of analysis is the country-cop

dyad. The predictors are located both at the country level and the level of the cop.

6.3.2 Independent Variables

Our measure reflecting the logic of appropriateness (hypothesis 1) is an index capturingthe extent of democratic participation in government, Polity, from the Polity IV dataset.It is a composite index that includes the following elements: presence of competitive po-litical participation, guarantee of openness and competitiveness of executive recruitment,and existence of institutionalised constraints on the exercise of executive power. Polityranges from –10 (most autocratic) to 10 (most democratic) (Marshall and Jaggers 2012).With a view to the above-mentioned theoretical arguments, we expect the sign of thisvariable to be positive.

To check the robustness of our results, we also use political rights scores as reportedby Freedom House. The Freedom House organisation rates all countries of the world ondimensions of political and civil8 rights. The political rights dimension captures mainlythe fairness and freedom of elections; whether elections, if any, are free and fair; andwhether an opposition exists and has the opportunity to take power with the consent ofthe electorate. Freedom House rates countries on a 1 to 7 scale; we converted the valuessuch that higher values indicate more political rights, and accordingly expect a positiverelationship between this variable and our dependent variable.

8The civil liberties part of the index measures constraints among other things on: association andorganisational rights, the rule of law and human rights and personal autonomy and economic rights.

87

We use two different measures to operationalise the logic of consequence. First andwith regard to a country’s physical vulnerability to climate change (hypothesis 2a), weinclude the Global Climate Risk Index, provided by Germanwatch.9 The index captureslosses, both material and human, from weather-related events. Using data from NatCat-SERVICE of Munich Re (Munich Re 2013), it aggregates four indicators: total deathtoll, deaths per 100’000 inhabitants, total losses in us dollar (adjusted for inflation) andtotal losses per unit gross domestic product in percent, for weather-related event classes,namely storms, floods, and extreme temperature events (Anemüller et al. 2006). Werecoded the index such that higher scores mean higher vulnerability. The expectation isthat more vulnerable countries adopt a more ambitious stance in the negotiations.

Assuming that fossil fuel consuming and producing countries are facing higher oppor-tunity costs of stricter climate policies, we proxy opposition to stricter climate policies(hypothesis 2b) with a country’s per capita carbon dioxide (CO2) emissions; because ofthe skewed nature of the variable, we log it. To check for robustness, we also include(logged) total CO2 emissions. The data come from the Climate Analysis IndicatorsTool (cait, see World Resources Institute 2008). The expectation is again a positiverelationship between CO2 emissions and our dependent variable.

6.3.3 Control Variables

First, we control for the political relevance of climate cops. We include the total numberof participants for each cop as a control variable. Arguably, the more important ameeting is perceived to be, the more participants—and potentially also the more pressconferences—it will attract.

Second, we control for the capacity of a country to hold press briefings. Althoughthe unfccc Secretariat administers the time slots and provides the infrastructure, apress briefing still requires considerable time and effort on the part of the initiator. Forsmall countries, the opportunity costs of holding a press briefing can be considerablebecause national delegates present in press briefings are not available for negotiationsand consultations that usually take place in parallel. We use the size of a country’snational delegation to proxy for capacity. Data on delegation size is taken from the listof participants, as coded by Böhmelt (2012).

Third, the country hosting a climate cop has a special role in the negotiation pro-cess. Rather than merely pursue its national interest, the cop presidency is expected tomake sure that the conference runs smoothly and that compromises are reached. Thepresidency is also responsible for communicating developments in the negotiations tothe media and thus the general public. The country presiding over a cop is therefore

9See Anemüller et al. (2006); Harmeling (2008; 2009; 2010; 2011; 2012; 2013). All reports are alsoavailable online at http://germanwatch.org/en/cri. Note that no yearly data are available for theperiod before 2004. We use the decadal average for this period, as provided in Anemüller et al. (2006).

88

expected to organise regular press conferences. We use unfccc documents, in particularthe cop reports, to obtain information on the cop presidency and host.10

Fourth, countries often negotiate as part of coalitions, such as the Alliance of SmallIsland States, the Group of Less Developed Countries (ldcs), or the European Union.The country that chairs one of these coalitions usually speaks on behalf of the coalition inthe negotiations and, for that reason, possibly also communicates more with the media.Accordingly, countries chairing a coalition should be more likely to initiate a press brief-ing. Information for this variable was extracted from unfccc documents, in particularthe cop reports, and, where this information was not available, from the coalition web-sites. Note that not all coalitions publish information on chairmanships; in particular, noinformation on chairmanship is available for the Umbrella Group and the Organization ofthe Petroleum Exporting Countries (opec). Table 6.1 presents the descriptive statistics;in the Appendix, Table 6.3 summarises our variables and data sources, and Tables 6.4and 6.5 report the correlation coefficients.

N mean std. deviation minimum maximum

press conferences 3510 0.21 1.18 0 16polity IV 2669 3.23 6.57 −10 10climate risk index 3071 −82.58 38.91 1.83 168.67CO2 emissions (total, log) 3218 2.28 2.63 −5.75 9.03CO2 emissions (per capita, log) 3218 0.60 1.69 −3.88 4.34delegation size 3490 14.06 30.08 0 737cop host/president 3510 0.01 0.07 0 1coalition chair 3510 0.02 0.15 0 1cop participants (total) 3510 8004 5548.83 1575 27’294

Table 6.1: Descriptive statistics

6.3.4 Method

We use zero-inflated count models. Since our dependent variable is the number of pressconferences per cop for each government, we are bound to have a lot of zeros in ourdata. A zero-inflated count model thus helps in dealing with excess zeros in the data(Greene 1994). The first stage of zero-inflated models (the inflation equation) uses abinary specification to estimate whether a press briefing is possible (i.e., whether thereis a positive probability of a government holding a press briefing). The second stage(the press briefings equation or count estimation) accounts for variation in the numberof press briefings initiated by a country that has a positive probability of holding apress briefing. This approach has several advantages. The zero-inflated models split thepress briefing initiation process into two stages for statistical and theoretical reasons.

10Usually, the country hosting the cop acts as the cop president. However, for some meetings, thehost did not hold the cop presidency. For instance, Zimbabwe officially presided over cop2, yet themeeting was held in Geneva, Switzerland. In such cases, both host and president were coded.

89

It is unrealistic to assume that all governments have a strictly positive probability ofholding a press briefing, as the standard models assume. Since it is not obvious ex antewhen a country becomes ‘relevant’, the two-step procedure separates observations withzero probability of press briefing initiation from those with a positive probability. Tomodel temporal dependence, time as well as its squared and cubic term (t, t2 and t3)are included in the models. This approach acknowledges that a government’s decision tohold a press briefing today depends strongly on its behaviour in the years before and thuscontrols for time effects. The use of t, t2 and t3 (cubic time polynomial) instead of cubicsplines, recommended by Beck et al. (1998), has the advantage of a more straightforwardinterpretation of the baseline hazard, whereas the approximation of the baseline hazardis at least as good as that with cubic splines (Carter and Signorini 2010).

6.4 Explaining Press Briefings Initiated By Governments

Table 6.2 reports the results for the zero-inflated models. The first column (model 1)contains the model for the logic of appropriateness, with democracy as the only predictor;the second column (model 2) only includes the predictors for the logic of consequence,i.e., environmental and economic vulnerability. Model 3 combines these two sets ofpredictors; model 4 also includes our control variables. Finally, we include in model 5dummy variables for the United States (us) and the European Union (eu). Becausethese two parties are responsible for the majority of all press conferences ever held (seeFigure 6.4), they may drive our results, which is why we include dummies as controls.

The appropriateness model in the first column supports the argument that democra-cies are less likely to never hold a press briefing (the inflation equation). The coefficientis negative, as expected, and the effect is statistically highly significant (p < .01). Thecount estimation similarly suggests a positive and significant relationship between democ-racy and the number of press conferences a government initiates: the more democratic,the more press conferences a government will initiate. However, this relationship losesits significance as soon as other variables are included (models 3 to 5); the relationshipin the inflation equation is also insignificant in model 4, and significant at the 10 per-cent level only in model 5. It seems, thus, that there is no strong relationship betweendemocratic governments and the organisation of press briefings, against our expectationsin H1. Democracies, while less likely to never organise a press conference, do not sys-tematically organise more press conferences. Substituting the polity variable with thepolitical rights one from Freedom House produces the same results (see Table 6.6 in theAppendix).

How about the logic of consequences (model 2)? The inflation equation suggeststhat more vulnerable countries are significantly less likely to never hold a press briefing,although the significance level drops from the 1 percent to the 10 percent level when eu

90

and us dummies are introduced in model 5. In contrast, according to the count model,there is no relationship between vulnerability to the negative effects of climate change andthe number of press conferences organised: governments from more vulnerable countriesdo not speak to the press in briefings more often. Economically vulnerable countries,measured by logged per capita CO2 emissions, in contrast, are associated with a highernumber of press briefings (count estimation). The relationship is positive and significantin models 2 to 4 (p < .01), yet the result seems to be driven by the two parties mostactive in organising press conferences, the us and the eu. When these are controlledfor, the effect of per capita CO2 emissions loses considerably in size and in significance(model 5). Similar results are obtained when using logged total, rather than per capita,CO2 emissions (see Table 6.7 in the Appendix).

91

(1) (2) (3) (4) (5)logic of ap- logic of both logics full model full model

propriateness consequence with eu/us

number of press conferencespolity 2 0.081*** 0.023 0.025 −0.007

(0.040) (0.033) (0.022) (0.018)climate risk 0.003 0.002 −0.001 0.007

(0.008) (0.009) (0.008) (0.004)CO2 per capita (logged) 0.573*** 0.492*** 0.406*** 0.170

(0.119) (0.122) (0.153) (0.123)delegation size 0.007** 0.004***

(0.003) (0.001)cop host/president 1.149*** 1.534***

(0.353) (0.209)coalition chair −2.073** −1.890**

(0.887) (0.782)cop participants (total) 0.000 0.000**

(0.000) (0.000)eu dummy 2.018***

(0.289)us dummy 1.295***

(0.254)constant −0.276 −0.450 −0.449 −1.027* −0.288

(0.400) (0.546) (0.561) (0.608) (0.322)

inflatepolity 2 −0.124*** −0.093*** −0.056 −0.075*

(0.036) (0.034) (0.041) (0.039)climate risk −0.026*** −0.024*** −0.016** −0.010*

(0.007) (0.006) (0.007) (0.006)CO2 per capita (logged) −0.006 −0.023 0.324 0.116

(0.166) (0.171) (0.244) (0.182)delegation size −0.049*** −0.045***

(0.011) (0.010)cop host/president −22.544*** −22.716***

(0.720) (0.638)coalition chair −0.934 −1.709

(1.569) (1.854)cop participants (total ) −0.000** −0.000***

(0.000) (0.000)eu dummy −2.238***

(0.486)us dummy −1.140**

(0.536)t 0.978*** 0.954*** 0.898*** 0.713*** 0.611***

(0.176) (0.189) (0.185) (0.224) (0.203)t2 −0.096*** −0.089*** −0.088*** −0.055 −0.045

(0.027) (0.028) (0.029) (0.041) (0.037)t3 0.003** 0.002** 0.003** 0.002 0.001

(0.001) (0.001) (0.001) (0.002) (0.002)constant 0.422 −2.325*** −1.378** 0.884 1.786***

(0.412) (0.730) (0.639) (0.858) (0.587)natural log of alphaconstant 1.015*** 0.953*** 0.824*** −0.028 −1.049**

(0.234) (0.310) (0.290) (0.176) (0.491)

N 2669 2873 2501 2500 2500Notes: The inflation equation estimates the probability that a national government never holds a press briefing (belongsto the Always 0 group). The count equation estimates the expected number of press briefings, weighted by the probabilitythat a government belongs to the Not Always 0 group. Robust standard errors are clustered on the country and listedin parentheses. *p<0.10;**p<0.05;***p<0.01.

Table 6.2: Zero-inflated negative binomial analysis of governments’ decision to hold press brief-ings

92

Turning now to the control variables, we find that countries with large delegationsand countries that host the summits are more likely to hold many press briefings. Onthe other hand, and contrary to our expectation, we did not find any evidence thatcountries chairing coalitions initiate more press briefings, but rather are associated witha significantly lower number of press conferences. Unsurprisingly, the larger a cop,the more press briefings it witnesses. While the effect size is small, the relationship ishighly and robustly significant (p < .01). Finally, the time variables also have largeand significant effects; having organised a press conference in the past is thus a strongpredictor of organising a press conference in the future.

6.5 Conclusion

Press briefings are an important element of international climate change negotiations;it is here that government negotiators communicate with the media, and through them,with the general public, to relate positions and the status of the negotiations. Pressconferences are widespread, and their number has increased sharply over the past roundof negotiations, yet the increase is not uniform over time, with some cops attracting alot of media attention and accordingly, a particularly high number of press conferences.Focusing on press conferences organised by government representatives, we argue thatthere are two main motivations for governments to hold a press conference: a logic ofappropriateness, according to which press conferences are ‘the right thing to do’; anda logic of consequence, according to which press conferences are used as an instrumentto increase bargaining leverage. We test these motivations empirically, using data fromthe unfccc Daily Programmes for the climate summits from 1995 through 2011. Ourempirical analysis finds limited support for these two logics of action. While democraciesare less likely to never hold a press conference, the relationship between level of democracyand organisation of press conferences is rather weak. Our analysis also provides onlylimited support for the logic of consequence. We find that vulnerable countries—whichface severe material and human losses from changing climatic conditions—are similarlyless likely to never organise a press conference; and that large CO2 emitters—which likelysuffer economically from ambitious climate policies—are associated with a larger numberof press conferences. However, the effects of our main variables are less significant thanother factors, including capacity or hosting the climate summit.

The role of the media in international negotiations has so far received very little aca-demic attention, although journalists attend international negotiations in large numbers.While our analysis provides a first step to address this gap, some caveats are in order.First, we only look at the case of climate change. There may be important differencesacross issue areas that we cannot capture with our research design. Second, and moreimportantly, we have no information about the content of the press conferences: our

93

data lacks any information on what was said during the event. Similarly, we have noinformation on who attended these events, both on the side of the organiser(s) and on theside of the media. Finally, we do not know for sure what organisers and attendees expectfrom these press conferences: which events do journalists in fact attend, and why? Howis the information given in the press conferences then conveyed to the general public?Further research will be needed to better understand press conferences in internationalnegotiations.

94

6.6 Appendix

concept variable data source

dependent variablenumber of press conferences unfccc Daily Programmes

logic of appropriatenessdemocracy polity 2 Polity IV dataset

political rights* Freedom House/Teorell et al. (2011)

logic of appropriatenessphysical vulnerability climate risk index Germanwatcheconomic vulnerability CO2 emissions, per capita cait

CO2 emissions, total* cait

control variablescapacity delegation size Böhmelt (2012)cop presidency cop president or host unfccccoalition chair chairmanship of a coalition unfccccop relevance number of cop participants unfccc

*Variables marked with an asterisk are used for robustness checks but not in the main analysis.

Table 6.3: Variables and data sources

95

press

conferences

polity2

clim

aterisk

index

CO

2em

ission

s

delegationsize

coppresident/host

coalitionchair

copparticipan

ts

press conferences 1

polity 2 0.148 1

climate risk index 0.127 0.165 1

CO2 emissions 0.137 0.218 0.090 1

delegation size 0.590 0.228 0.200 0.226 1

cop president/host 0.404 0.050 0.057 0.045 0.306 1

coalition chair -0.021 0.021 0.016 -0.011 0.072 -0.012 1

cop participants 0.121 0.034 0.111 0.021 0.304 -0.002 0.001 1CO2 emissions are per capita and logged. 0.011

Table 6.4: Correlation coefficients

press

conferences

political

righ

ts

clim

aterisk

index

CO

2em

ission

s

delegationsize

coppresident/host

coalitionchair

copparticipan

ts

press conferences 1

political rights 0.148 1

climate risk index 0.127 0.165 1

CO2 emissions 0.137 0.355 0.070 1

delegation size 0.590 0.206 0.202 0.204 1

cop president/host 0.404 0.053 0.056 0.041 0.307 1

coalition chair -0.021 0.038 -0.009 -0.016 0.061 -0.012 1

cop participants 0.121 -0.004 0.114 0.022 0.290 -0.002 0.006 1CO2 emissions are per capita and logged. 0.011

Table 6.5: Correlation coefficients using political rights

96



number of press conferencespolitical rights 0.281* 0.110 0.079 −0.009

(0.149) (0.121) (0.070) (0.060)climate risk 0.003 0.004 −0.000 0.006

(0.008) (0.008) (0.007) (0.004)CO2 per capita (logged) 0.574*** 0.486*** 0.377** 0.158

(0.119) (0.135) (0.150) (0.121)delegation size 0.007** 0.004***


(0.346) (0.208)coalition chair −2.436** −2.046***

(1.231) (0.742)cop participants (total) 0.000 0.000*

(0.000) (0.000)eu dummy 2.022***

(0.277)us dummy 1.323***

(0.250)constant −0.759 −0.450 −0.014 −0.638 −0.363

(0.481) (0.546) (0.671) (0.531) (0.345)

inflatepolitical rights −0.208* −0.251** −0.176 −0.225**

(0.111) (0.119) (0.118) (0.108)climate risk −0.026*** −0.024*** −0.016** −0.010*

(0.007) (0.006) (0.007) (0.006)CO2 per capita (logged) −0.006 −0.109 0.385 0.176

(0.166) (0.170) (0.243) (0.179)delegation size −0.053*** −0.048***


(0.764) (0.687)coalition chair −3.155 −3.016

(5.202) (2.287)cop participants (total) −0.000*** −0.000***

(0.000) (0.000)eu dummy −2.149***

(0.480)us dummy −1.142**

(0.539)t 1.127*** 0.954*** 0.925*** 0.708*** 0.624***

(0.157) (0.189) (0.182) (0.219) (0.196)t2 −0.112*** −0.089*** −0.087*** −0.049 −0.042

(0.023) (0.028) (0.028) (0.040) (0.036)t3 0.003*** 0.002** 0.002** 0.001 0.001

(0.001) (0.001) (0.001) (0.002) (0.002)constant −1.128*** −2.325*** −2.828*** 0.172 0.792

(0.304) (0.730) (0.754) (0.853) (0.626)natural log of alphaconstant 1.131*** 0.953*** 0.876*** −0.011 −1.026**

(0.231) (0.310) (0.290) (0.179) (0.490)

N 3456 2873 2858 2857 2857

Notes: The inflation equation estimates the probability that a national government never holds a press briefing (belongsto the Always 0 group). The count equation estimates the expected number of press briefings, weighted by the probabilitythat a government belongs to the Not Always 0 group. Robust standard errors are clustered on the country and listedin parentheses. *p<0.10;**p<0.05;***p<0.01.

Table 6.6: Estimation results using political rights.

97



number of press conferencespolity 2 0.081*** 0.027 0.029 0.007

(0.040) (0.028) (0.019) (0.017)climate risk −0.006 −0.004 −0.004 0.003

(0.006) (0.006) (0.005) (0.005)CO2 (logged) 0.341*** 0.346*** 0.333*** 0.143*

(0.045) (0.048) (0.072) (0.079)delegation size 0.003*** 0.003***


(0.247) (0.208)coalition chair −1.968*** −1.892**

(0.758) (0.752)cop participants (total) 0.000* 0.000**

(0.000) (0.000)eu dummy 1.622***

(0.447)us dummy 1.116***

(0.315)constant −0.276 −1.803*** −1.888*** −2.415*** −1.057*

(0.400) (0.556) (0.557) (0.606) (0.634)

inflatepolity 2 −0.124*** −0.093*** −0.050 −0.065*

(0.036) (0.030) (0.035) (0.035)climate risk −0.021*** −0.018*** −0.017*** −0.012*

(0.006) (0.005) (0.006) (0.007)CO2 (logged) −0.156 −0.217** 0.131 0.070

(0.096) (0.098) (0.108) (0.114)delegation size −0.042*** −0.043***


(0.636) (0.624)coalition chair −1.554 −1.855

(1.564) (1.730)cop participants (total) −0.000*** −0.000***

(0.000) (0.000)eu dummy −2.453***

(0.618)us dummy −1.188**

(0.541)t 0.978*** 0.830*** 0.768*** 0.766*** 0.629***

(0.176) (0.167) (0.159) (0.214) (0.203)t2 −0.096*** −0.076*** −0.075*** −0.066* −0.049

(0.027) (0.025) (0.025) (0.038) (0.037)t3 0.003** 0.002* 0.002* 0.002 0.002

(0.001) (0.001) (0.001) (0.002) (0.002)constant 0.422 −0.985 0.342 0.401 1.334

(0.412) (1.075) (0.976) (0.940) (0.936)natural log of alphaconstant 1.015*** 0.475 0.372 −0.583* −1.142**

(0.234) (0.420) (0.400) (0.339) (0.495)

N 2669 2873 2501 2500 2500

Notes: The inflation equation estimates the probability that a national government never holds a press briefing (belongsto the Always 0 group). The count equation estimates the expected number of press briefings, weighted by the probabilitythat a government belongs to the Not Always 0 group. Robust standard errors are clustered on the country and listedin parentheses. *p<0.10;**p<0.05;***p<0.01.

Table 6.7: Estimation results using total CO2 emissions.

99

Bibliography

Albin, C. (1999). Can NGOs Enhance the Effectiveness of International Negotiations?International Negotiation, 4(3):371–387.

Alcock, F. (2008). Conflicts and Coalitions Within and Across the ENGO Community.Global Environmental Politics, 8(4):66–91.

Anemüller, S., Monreal, S., and Bals, C. (2006). Global Climate Risk Index 2006.Weather-related Loss Events and their Impacts on Countries in 2004 and in a Long-Term Comparison. Technical report, Germanwatch, Bonn.

Arts, B. (1998). The Political Influence of Global NGOs: Case Studies on the Climateand Biodiversity Conventions. International Books, Utrecht.

Arts, B. (2001). The Impact of Environmental NGOs on International Conventions. InArts, B., Noortmann, M., and Reinalda, B., editors, Non-State Actors in InternationalRelations, pages 195–210. Ashgate, Aldershot.

Arts, B. (2005). Non-State Actors in Global Environmental Governance: New Arrange-ments Beyond the State. In Koenig-Archibugi, M. and Zürn, M., editors, New Modes ofGovernance in the Global System: Exploring Publicness, Delegation and Inclusiveness,International Political Economy Series, pages 177–200. Palgrave Macmillan.

Asp, K. and Esaiasson, P. (1996). The Modernization of Swedish Campaigns: Individ-ualization, Professionalization, and Medialization. In Swanson, D. L. and Mancini,P., editors, Politics, Media, and Modern Democracy: An International Study of In-novations in Electoral Campaigning and Their Consequences, pages 73–90. Praeger,Westport, CT.

Austen-Smith, D. and Wright, J. R. (1992). Competitive Lobbying for Legislative Votes.Social Choice and Welfare, 9:229–257.

Austen-Smith, D. and Wright, J. R. (1994). Counteractive Lobbying. American Journalof Political Science, 38(1):25–44.

Bauer, R. A., de Sola Pool, I., and Dexter, L. A. (1963). American Business and PublicPolicy: The Politics of Foreign Trade. Atherton Press, New York.

Baumgartner, F. R., Berry, J. M., Hojnacki, M., Kimball, D. C., and Leech, B. L. (2009).Lobbying and Policy Change. Who Wins, Who Loses, and Why. University of ChicagoPress, Chicago and London.

Baumgartner, F. R. and Leech, B. L. (1998). Basic Interests: The Importance of Groupsin Politics and in Political Science. Princeton University Press, Princeton.

Baumgartner, F. R. and Leech, B. L. (2001). Interest Niches and Policy Bandwagons:Patterns of Interest Group Involvement in National Politics. The Journal of Politics,

100

63(4):1191–1213.

Beck, N., Katz, J., and Tucker, R. (1998). Taking Time Seriously: Time-Series-Cross-Section Analysis with a Binary Dependent Variable. American Journal of PoliticalScience, 42(2):1260–1288.

Bernauer, T. and Betzold, C. (2012). Civil Society in Global Environmental Governance.Journal of Environment and Development, 21:62–66.

Bernauer, T. and Böhmelt, T. (2013). National Climate Policies in International Com-parison: The Climate Change Cooperation Index. Environmental Science & Policy,25:196–206.

Bernauer, T. and Gampfer, R. (2013). Effects of Civil Society Involvement on Popu-lar Legitimacy of Global Environmental Governance. Global Environmental Change,23:439–449.

Berry, J. M. (1977). Lobbying for the People: The Political Behavior of Public InterestGroups. Princeton University Press, Princeton.

Betsill, M. M. (2002). Environmental NGOs Meet the Sovereign State: The KyotoProtocol Negotiations on Global Climate Change. Colorado Journal of InternationalEnvironmental Law and Policy, 13(1):49–64.

Betsill, M. M. (2006). Transnational Actors in International Environmental Politics. InBetsill, M. M., Hochstetler, K., and Stevis, D., editors, Palgrave Advances in Interna-tional Environmental Politics, pages 172–202. Palgrave Macmillan.

Betsill, M. M. and Corell, E. (2001). NGO Influence in International EnvironmentalNegotiations: A Framework for Analysis. Global Environmental Politics, 1(4):65 – 85.

Betsill, M. M. and Corell, E., editors (2008). NGO Diplomacy: The Influence of Non-governmental Organizations in International Environmental Negotiations. MIT Press,Cambridge, MA.

Beyers, J. (2004). Voice and Access: Political Practices of European Interest Associations.European Union Politics, 5(2):211–240.

Biermann, F. and Gupta, A. (2011). Accountability and Legitimacy in Earth SystemGovernance: A Research Framework. Ecological Economics, 70:1856–1864.

Binderkrantz, A. S. (2005). Interest Group Strategies: Navigating Between PrivilegedAccess and Strategies of Pressure. Political Studies, 53:694–715.

Binderkrantz, A. S. (2008). Different Groups, Different Strategies: How Interest GroupsPursue Their Political Ambitions. Scandinavian Political Science, 31(2):173–200.

Binderkrantz, A. S. (2012). Interest Groups in the Media: Bias and Diversity over Time.European Journal of Political Research, 51(1):117–139.

Binderkrantz, A. S. and Krøyer, S. (2012). Customizing Strategy: Policy Goals andInterest Group Strategies. Interest Groups & Advocacy, 1(1):1–24.

Böhmelt, T. (2012). A Closer Look at the Information Provision Rationale: Civil So-ciety Participation in States’ Delegations at the unfccc. Review of InternationalOrganizations.

Böhmelt, T. and Betzold, C. (2012). The Impact of Environmental Interest Groups in In-ternational Negotiations: Do ENGOs Induce Stronger Environmental Commitments?International Environmental Agreements, in print.

101

Böhmelt, T., Koubi, V., and Bernauer, T. (2013). Civil Society Participation in GlobalGovernance: Insights from Climate Politics. European Journal of Political Research.

Bolton, M. and Nash, T. (2010). The Role of Middle Power–NGO Coalitions in GlobalPolicy: The Case of the Cluster Munitions Ban. Global Policy, 1(2):172–184.

Boykoff, M. T. (2007). Flogging a Dead Norm? Newspaper Coverage of AnthropogenicClimate Change in the United States and United Kingdom from 2003 to 2006. Area,39(4):470–481.

Breitmeier, H. and Rittberger, V. (1998). Environmental NGOs in an Emerging GlobalCivil Society. Tübinger Arbeitspapiere zur Internationalen Politik und Friedens-forschung Nr. 32, Eberhard Karls Universität zu Tübingen, Germany.

Brühl, T. (2003). Nichtregierungsorganisationen als Akteure internationaler Umweltver-handlungen: Ein Erklärungsmodell auf der Basis der situationsspezifischenRessourcennachfrage. Studien der Hessischen Stiftung Friedens- und Konfliktforschung.Campus, Frankfurt and New York.

Carpenter, C. (2001). Business, Green Groups and the Media: The Role of Non-Governmental Organizations in the Climate Change Debate. International Affairs,77(2):313–328.

Carter, D. B. and Signorini, C. S. (2010). Back to the Future: Modeling Time Dependencewith Binary Data. Political Analysis, 18(2):271–292.

Charnovitz, S. (1997). Two Centuries of Participation: NGOs and International Gover-nance. Michigan Journal of International Law, 18:183 –286.

Charnovitz, S. (2003). The Emergence of Democratic Participation in Global Governance(Paris, 1919). Indiana Journal of Global Legal Studies, 10:45–77.

Chasek, P. S. (2001). NGOs and State Capacity in International Environmental Ne-gotiations: The Experience of the Earth Negotiations Bulletin. Review of EuropeanCommunity & International Environmental Law, 10(2):168 – 176.

Cobo, J. M. (1986). Study of the Problem of Discrimination Against Indigenous Popu-lations. UN Economic and Social Council E/CN.4/Sub.2/1982/2/Add.6, New York.

Colchester, M. (2003). Salvaging Nature: Indigenous Peoples, Protected Areas and Bio-diversity Conservation. Forest Peoples Programme, Moreton-in-Marsh.

Convention on Biological Diversity (2000). Decision V/16: Article 8(j) and Related Pro-visions. Published in the Report of the Fifth Meeting of the Parties to the Conventionon Biological Diversity, Nairobi, Kenya, 15-26 May 2000. UNEP/CBD/COP/5/23:139-146, Nairobi.

Convention on Biological Diversity (2006). Decision VIII/5: Article 8(j) and Related Pro-visions. Published in the Report of the Eighth Meeting of the Parties to the Conventionon Biological Diversity, Curitiba, Brazil, 20-31 March 2006. UNEP/CBD/COP/8/31:139-158, Curitiba.

Coordinator of Indigenous Organizations of the Amazon Basin (1989). Two Agendas onAmazon Development. Cultural Survival Quarterly, 13(4):75–87.

Corell, E. and Betsill, M. M. (2001). A Comparative Look at NGO Influence in In-ternational Environmental Negotiations: Desertification and Climate Change. GlobalEnvironmental Politics, 1(4):86 – 107.

102

Crump, J. (2008). Snow, Sand, Ice, and Sun: Climate Change and Equity in the Arcticand Small Island Developing States. Sustainable Development Law and Policy, 8(3):8–12, 59–60.

Degawan, M. (2008). Mitigating the Impacts of Climate Change: Solutions or AdditionalThreats? Indigenous Affairs, 1-2:52–59.

DeGregorio, C. (1998). Assets and Access: Linking Lobbyists and Lawmakers inCongress. In Hernson, P. S., Shaiko, R. G., and Wilcox, C., editors, The InterestGroup Connection. Electioneering, Lobbying and Policymaking in Washington, chap-ter 9, pages 137–153. Chatham House.

Denzau, A. T. and Munger, M. C. (1986). Legislators and Interest Groups: How Unor-ganized Interests Get Represented. American Political Science Review, 80(1):89–106.

Depledge, J. (2005). The Organization of Global Negotiations: Constructing the ClimateChange Regime. Earthscan, London.

Dimitrov, R. S. (2010). Inside UN Climate Change Negotiations: The CopenhagenConference. Review of Policy Research, 27(6):795–821.

Dirikx, A. and Gelders, D. (2010). To Frame is to Explain: A Deductive Frame-analysisof Dutch and French Climate Change Coverage During the Annual UN Conferences ofthe Parties. Public Understanding of Science, 19(6):732–742.

Dryzek, J. S. (2012). Global Civil Society: The Progress of Post-Westphalian Politics.Annual Review of Political Science, 15:101–119.

Dür, A. and Mateo, G. (2010). Irish Associations and Lobbying on EU Legislation:Resources, Access Points, and Strategies. Irish Political Studies, 25(1):107–122.

Dür, A. and Mateo, G. (2012). Who Lobbies the European Union? National InterestGroups in a Multilevel Polity. Journal of European Public Policy, 19(7):969–987.

Dür, A. and Mateo, G. (2013). Gaining Access or Going Public? Interest Group Strategiesin Five European Countries. European Journal of Political Research.

Duwe, M. (2001). The Climate Action Network: A Glance behind the Curtains of aTransnational NGO Network. Review of European Community & International Envi-ronmental Law, 10(2):177–189.

Eisinger, P. K. (1973). The Conditions of Protest Behavior in American Cities. AmericanPolitical Science Review, 67(1):11–28.

Elster, J. (1989). Nuts and Bolts for the Social Sciences. Cambridge University Press,Cambridge.

Esser, F. (2013). Mediatization as a Challenge: Media Logic Versus Political Logic.In Kriesi, H., Lavenex, S., Esser, F., Matthes, J., Bühlmann, M., and Bochsler, D.,editors, Democracy in the Age of Globalization and Mediatization. Palgrave Macmillan,Houndmills.

Falkner, R. (2009). Business and Global Climate Governance: A Neo-Pluralist Per-spective. In Ougaard, M. and Leander, A., editors, Business and Global Governance.Routledge, London.

Fearon, J. D. (1997). Signaling Foreign Policy Interests: Tying Hands versus SinkingCosts. Journal of Conflict Resolution, 41(1):68–90.

Feldt, H. (2009). Stärkung indigener Organisationen in Lateinamerika: Indigene Völker

103

und Klimapolitik. Deutsche Gesellschaft für Internationale Zusammenarbeit, Es-chborn.

Fisher, D. R. and Green, J. F. (2004). Understanding Disenfranchisement: Civil Soci-ety and Developing Countries’ Influence and Participation in Global Governance forSustainable Development. Global Environmental Politics, 4(3):65–84.

Foundation for International Environmental Law and Development (field) (2008).Ways for Indigenous Peoples’ Groups to Advance Adaptation Concerns and SolutionsThrough International Fora. London.

Gadgil, M., Berkes, F., and Folke, C. (1993). Indigenous Knowledge for BiodiversityConservation. Ambio, 2-3:151–156.

Gais, T. L. and Walker, J. L. J. (1991). Pathways to Influence in American Politics.In Walker, J. L. J., editor, Mobilizing Interest Groups in America, pages 103–121.University of Michigan Press, Ann Arbor.

Gaventa, J. (2010). Environmental Mega-Conferences and Climate Governance beyondthe Nation-State: A Bali Case Study. St Antony’s International Review, 5(2):29–45.

Gemmill, B. and Bamidele-Izu, A. (2002). The Role of NGOs and Civil Society inGlobal Enviromental Governance. In Esty, D. and Ivanova, M. H., editors, GlobalEnvironmental Governance: Options and Opportunities. Yale School of Forestry andEnvironmental Studies, New Haven, CT.

Giorgetti, C. (1998). The Role of Non-Governmental Organizations in the ClimateChange Negotiations. Colorado Journal of International Environmental Law and Pol-icy, 9:115–137.

Giorgetti, C. (1999). From Rio to Kyoto: A Study of the Involvement of Non-Governmental Organizations in the Negotiations on Climate Change. N.Y.U. En-vironmental Law Journal, 7:201–245.

Gormley, W. T. and Cymrot, H. (2006). The Strategic Choices of Child AdvocacyGroups. Nonprofit and Voluntary Sector Quarterly, 35(1):102–122.

Gough, C. and Shackley, S. (2001). The Respectable Politics of Climate Change: TheEpistemic Communities and NGOs. International Affairs, 77(2):329–345.

Grant, W. (1989). Pressure Groups Politics and Democracy in Britain. Philip Allan,Hemel Hempstead.

Grant, W. (2001). Pressure Politics: From ‘Insider’ Politics to Direct Action? Parlia-mentary Affairs, 54:337–348.

Green, D. and Raygorodetsky, G. (2010). Indigenous Knowledge of a Changing Climate.Climatic Change, 100:239–242.

Greene, W. H. (1994). Accounting for Excess Zeros and Sample Selection in Poisson andNegative Binomial Regression Models. Stern School of Business Working Paper 94-10.

Grubb, M. and Yamin, F. (2001). Climatic Collapse at The Hague: What Happened,Why, and Where Do We Go From Here? International Affairs, 77(1):261–276.

Gulbrandsen, L. H. and Andresen, S. (2004). NGO Influence in the Implementation ofthe Kyoto Protocol: Compliance, Flexibility Mechanisms, and Sinks. Global Environ-mental Politics, 4(4):54 – 75.

Gullberg, A. T. (2008a). Lobbying Friends and Foes in Climate Policy: The Case of

104

Business and Environmental Interest Groups in the European Union. Energy Policy,36:2964–2972.

Gullberg, A. T. (2008b). Rational Lobbying and EU Climate Policy. InternationalEnvironmental Agreements, 8:161–178.

Hall, R. L. and Deardorff, A. V. (2006). Lobbying as Legislative Subsidy. The AmericanPolitical Science Review, 100(1):69–84.

Hall, R. L. and Wayman, F. W. (1990). Buying Time: Moneyed Interests and theMobilization of Bias in Congressional Committees. The American Political ScienceReview, 84(3):797–820.

Harmeling, S. (2008). Global Climate Risk Index 2008. Weather-related Loss Events andtheir Impacts on Countries in 2006 and in a Long-Term Comparison. Technical report,Germanwatch, Bonn.

Harmeling, S. (2009). Global Climate Risk Index 2009. Weather-related Loss Events andtheir Impacts on Countries in 2007 and in a Long-Term Comparison. Technical report,Germanwatch, Bonn.

Harmeling, S. (2010). Global Climate Risk Index 2010. Who is Most Vulnerable?Weather-related Loss Events since 1990 and How Copenhagen Needs to Respond.Technical report, Germanwatch, Bonn.

Harmeling, S. (2011). Global Climate Risk Index 2011. Who Suffers Most From ExtremeWeather Events? Weather-related Loss Events in 2009 and 1990 to 2009. Technicalreport, Germanwatch, Bonn.



Hjarvard, S. (2008). The Mediatization of Society: A Theory of the Media as Agents ofSocial and Cultural Change. Nordicom Review, 29(2):105–134.

Hjerpe, M. and Linnér, B.-O. (2010). Functions of COP Side-Events in Climate-ChangeGovernance. Climate Policy, 10:167–180.

Hojnacki, M. (1997). Interest Groups’ Decision to Join Alliances or Work Alone. Amer-ican Journal of Political Science, 41(1):61–87.

Hojnacki, M. and Kimball, D. C. (1998). Organized Interests and the Decision Whomto Lobby in Congress. American Political Science Review, 92(4):775–790.

Hojnacki, M. and Kimball, D. C. (1999). The Who and How of Organizations’ LobbyingStrategies in Committee. The Journal of Politics, 61(4):999–1024.

Holyoke, T. T. (2003). Chosing Battlegrounds: Interest Group Lobbying Across MultipleVenues. Political Research Quarterly, 56(3):325–336.

International Alliance of Indigenous-Tribal Peoples of the Tropical Forests (1996). Sub-mission to the CBD Secretariat 1996. In International Alliance of Indigenous-TribalPeoples of the Tropical Forests and International Work Group for Indigenous Af-fairs, editor, Indigenous Peoples, Forests, and Biodiversity: Indigenous Peoples and

105

the Global Environmental Agenda. International Alliance, London.

International Indigenous Forum on Biodiversity (2011). Statement at the Ad Hoc Open-Ended Working Group on Access and Benefit Sharing. Convention on Biological Di-versity, 22–26 October 2001, Bonn, Germany.

Keck, M. E. and Sikkink, K. (1998). Activists Beyond Borders. Advocacy Networks inInternational Politics. Cornell University Press, Ithaca, NY.

Koivovura, T. and Heinämäki, L. (2006). The Participation of Indigenous Peoples inInternational Norm-Making in the Arctic. Polar Record, 42(221):101–109.

Kolk, A. (2001). Multinational Enterprises and International Climate Policy. In Arts,B., Noortmann, M., and Reinalda, B., editors, Non-State Actors in International Re-lations, pages 211–225. Ashgate, Aldershot.

Kollman, K. (1998). Outside Lobbying: Public Opinion and Interest Group Strategies.Princeton University Press, Princeton.

Kotzian, P. and Steffek, J. (2013). Do Members Make a Difference? A Study of Transna-tional Civil Society Organizations. European Political Science Review, 5(1):55–81.

Kriesi, H. (2004). Political Context and Opportunity. In Snow, D. A., Soule, S. A.,and Kriesi, H., editors, The Blackwell Companion to Social Movements, pages 67–90.Blackwell Publishing Ltd., Oxford.

Kriesi, H., Tresch, A., and Jochum, M. (2007). Going Public in the European Union:Action Repertoires of Western European Collective Political Actors. Comparative Po-litical Studies, 40:48–73.

Kruse, J. (2011). Civil Society Organizations’ Strategies in Global Environmental Gover-nance: Between Collaboration and Protest. Paper presented at the Dreiländertagung,January 13-14, Basel, Switzerland.

Lang, W. (1989). The Role of Presiding Officers in Multilateral Negotiations. In Mautner-Markhof, F., editor, Processes of International Negotiations, pages 23–42. WestviewPress, Boulder.

Lemaitre, S. (2011). Indigenous Peoples’ Land Rights and REDD: A Case Study. Reviewof European Community & International Environmental Law, 20(2):150–162.

Lindroth, M. (2006). Indigenous Relations in the UN: Establishing the Indigenous Forum.Polar Record, 42(222):239–248.

Lipschutz, R. D. (1996). Global Civil Society and Global Environmental Governance: ThePolitics of Nature from Place to Planet. State University of New York Press, Albany.

Lisowski, M. (2005). How NGOs Use Their Facilitative Negotiating Power and BargainingAssets To Affect International Environmental Negotiations. Diplomacy & Statecraft,16(2):361–383.

Liu, X., Lindquist, E., and Vedlitz, A. (2011). Explaining Media and CongressionalAttention to Global Climate Change, 1969-2005: An Empirical Test of Agenda-SettingTheory. Political Research Quarterly, 64(2):405–419.

Macchi, M. (2008). Indigenous and Traditional Peoples and Climate Change. IssuesPaper, Internationl Union for the Conservation of Nature.

Mahoney, C. (2007). Networking vs. Allying: The Decision of Interest Groups to JoinCoalitions in the US and the EU. Journal of European Public Policy, 14(3):366–383.

106

Mahoney, C. (2008). Brussels Versus the Beltway: Advocacy in the United States andthe European Union. Georgetown University Press, Washington D.C.

Mahoney, C. and Baumgartner, F. R. (2008). Converging Perspectives on Interest GroupResearch in Europe and America. West European Politics, 31(6):1253–1273.

Maloney, W. A., Jordan, G., and McLaughlin, A. M. (1994). Interest Groups and PublicPolicy: The Insider/Outsider Model Revisited. Journal of Public Policy, 14(1):17–38.

March, J. G. and Olsen, J. P. (1996). Institutional Perspectives on Political Institutions.Governance-an International Journal of Policy and Administration, 9(3):247–264.

March, J. G. and Olsen, J. P. (1998). The Institutional Dynamics of InternationalPolitical Order. International Organization, 52(4):943–969.

March, J. G. and Olsen, J. P. (2004). The Logic of Appropriateness. Technical report,Centre for European Studies, University of Oslo.

Marcinkowski, F. (2005). Die ‘Medialisierbarkeit’ politischer Institutionen. In Rössler,P. and Krotz, F., editors, Mythen der Mediengesellschaft - The Media Society and itsMyths, pages 341–369. UVK, Konstanz.

Marshall, D. (2010). Who To Lobby and When: Institutional Determinants of Inter-est Group Strategies in European Parliament Committees. European Union Politics,11(4):553–575.

Marshall, M. G. and Jaggers, K. (2012). Polity IV Project: Political Regime Charac-teristics and Transitions, 1800-2010. Available online at http://www.systemicpeace.org/polity/polity4.htm.

Mauro, F. and Hardison, P. D. (2000). Traditional Knowledge of Indigenous and Lo-cal Communities: International Debate and Policy Initiatives. Traditional EcologicalKnowledge, 10(5):366–383.

Mazzoleni, G. and Schulz, W. (1999). ‘Mediatization’ of Politics: A Challenge for Democ-racy? Political Communication, 16(3):247–261.

McMahon, V. M. (1993). Environmental Nongovernmental Organizations at Intergov-ernmental Negotiations. In Susskind, L. E., Moomaw, W. R., and Najam, A., editors,Papers on International Environmental Negotiations, volume III. PON Books, Cam-bridge, MA.

Meyer, D. S. (2004). Protest and Political Opportunities. Annual Review of Sociology,30:124–145.

Meyer, D. S. and Minkoff, D. C. (2004). Conceptualizing Political Opportunity. SocialForces, 82(4):1457–1491.

Meyer, D. S. and Staggenborg, S. (1996). Movements, Countermovements, and theStructure of Political Opportunity. American Journal of Sociology, 101(6):1628–1660.

Michaelowa, A. (1998). Climate Policy and Interest Groups - A Public Choice Analysis.Intereconomics: Review of European Economic Policy, 33(6):251–259.

Mihlar, F. (2008). Voices That Must Be Heard: Minorities and Indigenous PeoplesCombating Climate Change. Minority Rights Group International Briefing, London.

Milbrath, L. (1963). The Washington Lobbyists. Rand McNally, Chicago.

Milner, H. V. (1997). Interests, Institutions, and Information: Domestic Politics and

107

International Relations. Cambridge University Press, Cambridge.

Mitchell, R. (2013). International Environmental Agreements Database Project (2013.1edition). Available online at http://iea.uoregon.edu/.

Mitchell, R. B. (2003). International Environmental Agreements: A Survey of TheirFeatures, Formation, and Effects. Annual Review of Environmental Resources, 28:429–461.

Moravcsik, A. (1991). Negotiating the Single European Act: National Interests andConventional Statecraft in the European Community. International Organization,45(1):19–56.

Moravcsik, A. (1993). Introduction: Integrating International and Domestic Theoriesof International Bargaining. In Evans, P. B., Jacobson, H. K., and Putnam, R. D.,editors, Double Edged Diplomacy: International Politics and Domestic Bargaining.University of California Press, Berkeley.

Moravcsik, A. (1997). Taking Preferences Seriously: A Liberal Theory of InternationalPolitics. International Organization, 51(04):513–553.

Morin, F. and Saladin d’Anglure, B. (1995). L’éthnicité, un outil politique pour lesautochtones de l’arctique et de l’amazonie. Études/Inuit/Studies, 19(1):37–68.

Muñoz Cabré, M. (2011). Issue-linkages to Climate Change Measured through NGOParticipation in the UNFCCC. Global Environmental Politics, 11(3):10–22.

Munich Re (2013). Downloadcenter for Statistics on Natural Catastrophes (NatCatSER-VICE). Available online at http://www.munichre.com/en/reinsurance/business/non-life/georisks/natcatservice/default.aspx.

Newell, P. (2000). Climate for Change: Non-state Actors and the Global Politics of theGreenhouse. Cambridge University Press, Cambridge.

Nilsson, C. (2008). Climate Change from an Indigenous Perspective: Key Issues andChallenges. Indigenous Affairs, 1-2:8–15.

Nownes, A. J. and Freeman, P. (1998). Interest Group Activity in the States. The Journalof Politics, 60(1):86–112.

Oldham, P. (2002). Negotiating Diversity: A Field Guide to the Convention on BiologicalDiversity. CESAGEN and Lancaster University, Lancaster.

Olson, M. (1982). The Rise and Decline of Nations: Economic Growth, Stagflation, andSocial Rigidities. Yale University Press, New Haven, CT.

Orr, S. K. (2005). Organized Interests and International Climate Change Policy. PhDthesis, Wayne State University, Detroit, Michigan.

Orr, S. K. (2006). Policy Subsystems and Regimes: Organized Interests and ClimateChange Policy. The Policy Studies Journal, 34(2):147–169.

Parry, M. L., Canziani, O. F., Palutikof, J. P., van der Linden, P. J., and Hanson,C. E., editors (2007). Contribution of Working Group II to the Fourth AssessmentReport of the Intergovernmental Panel on Climate Change. Cambridge UniversityPress, Cambridge.

Permanent Forum on Indigenous Issues (2008). Report on the Seventh Session (21 April-2May 2008). E/C.19/2008/13, New York.

108

Potvin, C., Revéret, J.-P., Patenaude, G., and Hutton, J. (2002). The Role of IndigenousPeoples in Conservation Actions: A Case Study of Cultural Differences and Conser-vation Priorities. In Prestre, P. G. L., editor, Governing Global Biodiversity: TheEvolution and Implementation of the Convention on Biological Diversity, pages 159–175. Ashgate, Aldershot.

Pulver, S. (2005). A Public Sphere in International Environmental Politics: The Caseof the Kyoto Protocol Negotiations. Paper presented at the Berlin Conference on theHuman Dimensions of Global Environmental Change: International Organizations andGlobal Environmental Governance.

Putnam, R. D. (1988). Diplomacy and Domestic Politics: The Logic of Two-level Games.International Organization, 42(3):427–460.

Raustiala, K. (1997). States, NGOs, and International Environmental Institutions. In-ternational Studies Quarterly, 41:719–740.

Raustiala, K. (2001). Nonstate Actors in the Global Climate Regime. In Luterbacher, U.and Sprinz, D. F., editors, International Relations and Global Climate Change, pages95 – 117. MIT Press, Cambridge, MA.

Redford, K. H. and Stearman, A. M. (1993). Forest-Dwelling Native Amazonians andthe Conservation of Biodiversity: Interests In Common or In Collision? ConservationBiology, 7(2):248–255.

Richards, J. P. and Heard, J. (2005). European Environmental NGOs: Issues, Resourcesand Strategies in Marine Campaigns. Environmental Politics, 14(1):23–41.

Rieff, D. (1999). The False Dawn of Civil Society. The Nation.

Rietig, K. (2011). Public Pressure vs. Lobbying - How Do Environmental NGOs Matterthe Most in Climate Negotiations? Working paper no. 70, Grantham Research Instituteon Climate Change and the Environment, London School of Economics and PoliticalScicence, London.

Risse, T. (2007). Transnational Actors and World Politics. In Walther Ch. Zimmerli,M. H. and Richter, K., editors, Corporate Ethics and Corporate Governance, pages251–286. Springer, Berlin and Heidelberg.

Rootes, C. (1999). Acting Globally, Thinking Locally? The Prospects for a GlobalEnvironmental Movement. Environmental Politics, 8(1):290–310.

Rossbach de Olmos, L. (2004). Klimawandel, internationale Umweltpolitik und indigeneVölker. Anthropos, 99:551–564.

Salick, J. and Ross, N. (2009). Introduction: Traditional Peoples and Climate Change.Global Environmental Change, 19:137–139.

Sanders, D. E. (1999). Indigenous Peoples: Issues of Definition. International Journal ofCultural Property, 8(1):4–13.

Schäfer, M. S., Ivanova, A., and Schmidt, A. (2011a). Global Climate Change, GlobalPublic Sphere? Media Attention for Climate Change in 23 Countries. Studies inCommunication | Media, 0(1):131–148.

Schäfer, M. S., Ivanova, A., and Schmidt, A. (2011b). Media Attention for ClimateChange. KlimaCampus, University of Hamburg: http://www.klimacampus.de/macc.html.

109

Schattschneider, E. E. (1960). The Semisovereign People: A Realist’s View of Democracyin America. Holt, Rinehart and Winston, New York.

Schelling, T. C. (1960). The Strategy of Conflict. Harvard University Press, Cambridge,MA.

Schlozman, K. L. and Tierney, J. J. (1986). Organized Interest and American Democracy.Harper and Row, New York.

Schlozman, K. L. and Tierney, J. T. (1983). More of the Same: Washington PressureGroup Activity in a Decade of Change. The Journal of Politics, 45(2):351–377.

Schmitter, P. C. and Streeck, W. (1999). The Organization of Business Interests: Study-ing the Associative Action of Business in Advanced Industrial Societies. Max-Planck-Institut für Gesellschaftsforschung Discussion Paper 99/1.

Schroeder, H. (2010). Agency in International Climate Negotiations: The Case of In-digenous Peoples and Avoided Deforestation. International Environmental Agreements,10:317–332.

Schroeder, H. and Lovell, H. (2012). The Role of Non-nation-state Actors and SideEvents in the International Climate Negotiations. Climate Policy, 12(1):23–37.

Selin, H. and Selin, N. E. (2008). Indigenous Peoples in International Environmen-tal Cooperation: Arctic Management of Hazardous Substances. Review of EuropeanCommunity & International Environmental Law, 17(1):72–83.

Skodvin, T. and Andresen, S. (2003). Nonstate Influence in the International WhalingCommission, 1970-1990. Global Environmental Politics, 3(4):61 – 86.

Smith, R. A. (1995). Interest Group Influence in the U.S. Congress. Legislative StudiesQuarterly, XX(1):89–139.

Sobrevila, C. (2008). The Role of Indigenous Peoples in Biodiversity Conservation: TheNatural But Often Forgotten Partners. World Bank, Washington, DC.

Society for Threatened Peoples (2008). Biodiversität und indigene Völker: Ausverkaufvon biologischen Ressourcen und traditionellem Wissen. Society for Threatened Peo-ples Human Rights Report 50, Göttingen.

Steffek, J. (2008). Zähmt zivilgesellschaftliche Partizipation die internationale Politik?Vom exekutiven zum partizipativen Multilateralismus. Leviathan. Berliner Zeitschriftfür Sozialwissenschaft, 36(1):105–122.

Steffek, J. and Ferretti, M. P. (2009). Accountability or ‘Good Decisions’? The Compet-ing Goals of Civil Society Participation in International Governance. Global Society,23(1):37–57.

Strömback, J. and Esser, F. (2009). Shaping Politics: Mediatization and Media Inter-ventionism. In Lundby, K., editor, Mediatization. Concept, Changes, Consequences,pages 205–223. Peter Lang, New York, NY.

Tallberg, J. (2010). The Power of the Chair: Formal Leadership in International Coop-eration. International Studies Quarterly, 54:241–265.

Tarrow, S. (1994). Power in Movement. Cambridge University Press, Cambridge.

Tauli-Corpuz, V. (2010). Indigenous Peoples’ Actions Towards Solving the BiodiversityCrisis. Speech during the 2010 International Year of Biodiversity, UNESCO High LevelSegment, Paris, 21-22 January 2010.

110

Tennberg, M. (2010). Indigenous Peoples as International Political Actors: A Summary.Polar Record, 46(238):264–270.

Teorell, J., Charron, N., Samanni, M., Holmberg, S., and Rothstein, B. (2011). The Qual-ity of Government Dataset, version 6Apr11. University of Gothenburg: The Qualityof Government Institute, http://www.qog.pol.gu.se.

Thrall, A. T. (2006). The Myth of the Outside Strategy: Mass Media News Coverage ofInterest Groups. Political Communication, 23:407–420.

Tjernshaugen, A. and Lee, H.-C. (2004). Shaming and Framing: Norwegian Nongovern-mental Organizations in the Climate Change Negotiations. CICERO Working Paper2004:09.

Toledo, V. M. (2001). Biodiversity and Indigenous Peoples. In Levin, S. A., editor,Encyclopedia of Biodiversity, pages 451–463. Elsevier, New York, NY.

Tomz, M. (2007). Domestic Audience Costs in International Relations: An ExperimentalApproach. International Organization, 61(4):821–840.

unfccc Secretariat (2004). Promoting Effective Participation in the Convention Process.Note by the Secretariat. FCCC/SBI/2004/5, Bonn.

United Nations (1992a). United Nations Convention on Biological Diversity. TreatySeries, 1760(30619):79–307.

United Nations (1992b). United Nations Framework Convention on Climate Change.Treaty Series, 1771(30822):107–321.

United Nations General Assembly (2007). Declaration on the Rights of Indigenous Peo-ples. A/61/L.67, New York, NY.

Victor, J. N. (2007). Strategic Lobbying. Demonstrating How Legislative Context AffectsInterest Groups’ Lobbying Tactics. American Politics Research, 35(6):826–845.

Vormedal, I. (2008). The Influence of Business and Industry NGOs in the Negotiation ofthe Kyoto Mechanisms: The Case of Carbon Capture and Storage in the CDM. GlobalEnvironmental Politics, 8(4):36–65.

Walker, J. L. (1991). Mobilizing Interest Groups in America. The University of MichiganPress, Ann Arbor, MI.

Wapner, P. (1996). Environmental Activism and World Civic Politics. State Universityof New York Press, Albany.

Weiss, T. G. and Gordenker, L., editors (1996). NGOs, the UN, and Global Governance.Lynne Rienner, Boulder, CO.

Willetts, P. (1996). From Stockholm to Rio and Beyond: The Impact of the Environmen-tal Movement on the United Nations Consultative Arrangements for NGOs. Reviewof International Studies, 22(1):57 – 80.

Woll, C. (2006). Lobbying in the European Union: From sui generis to a ComparativePerspective. Journal of European Public Policy, 13(3):456–469.

World Resources Institute (2008). Climate Analysis Indicators Tool. Available online athttp://cait.wri.org/.

Yamin, F. (2001). NGOs and International Environmental Law: A Critical Evaluationof their Roles and Responsibilities. Review of European Community and International

111

Environmental Law, 10(2):149–162.

Yamin, F. and Depledge, J. (2004). The International Climate Change Regime: A Guideto Rules, Institutions and Procedures. Cambridge University Press, Cambridge.

non-state actors in international climate change negotiations · 2015-01-17 · diss. eth no. 21193...

Documents