sign in sign up
<<
Home , European Social Movement, New Social Movements, Social movement, Social Movement Studies

Transnational in the New Urban World

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State

By

Matthew Schoene, M.A.

Graduate Program in

The Ohio State University

2015

Dissertation Committee:

Dr. Rachel Dwyer, Advisor

Dr. Andrew Martin

Dr. J. Craig Jenkins

Copyrighted by

Matthew Schoene

2015

Abstract

The world currently stands at a crossroads. has raised living standards all over the world, but globalization is also defined by rising inequality and extreme polarization (Sassen 2013). This disparity between rich and poor is most prominent in cities, now the primary lived experience for the majority of people. Grassroots social movement (SMOs) represent an important avenue for advancing in the globalized urban world, but there are significant gaps in the current understanding of this process. Most importantly, there is reason to believe that cities act as “movement spaces” (Nicholls 2009) that offer a better environment for SMOs as compared to less urban areas. However, it is not clear precisely how cities encourage social movement activism at both the organizational and individual level, how or if this process differs across national borders, or how much activism depends on contextual effects as opposed to individual factors.

This dissertation advances previous research by studying urban social movement activism in a cross-national, multilevel framework. First, I examine whether the global economic competitiveness of a city encourages SMO persistence. Using a sample of 672

SMOs in 67 global cities drawn from the Transnational Social Movement

Dataset 1953-2003 (Smith & Wiest 2012), I test the organizational, urban and national determinants of two organizational outcomes: viability and age. Results indicate that global urban competitiveness, professionalization, alliances with intergovernmental

ii

organization actors and urban resources best predict SMO persistence in global cities.

Next, I explore how urban status, and resource levels influence the likelihood of individual participation in six different forms of social and political activism in ’s recent wave. Using the 5th wave of the European Social , a series of multilevel logistic regressions indicate that European cities encourage citizen activism by connecting SMOs to the kinds of people most likely to participate. Finally, I further examine Europe’s protest wave and study how successful grassroots actors were at achieving media standing as compared to institutionalized, bureaucratic actors during

Europe’s anti-austerity movement. A content analysis of 4,486 quotes published in the

New York Times from January 2010-March 2013 indicates that institutionalized elites were better able to achieve media standing, but the overall sentiment was decidedly anti- austerity. Nevertheless, austerity became European policy, which indicates that the international media as measured here did not have a strong effect.

The findings of this dissertation indicate that cities act as movement spaces by promoting a better match between the supply of and demand for human and material resources, creating international and local networks of activism and connecting organizations to important allies like nongovernmental organizations and media actors.

Results also indicate that resources play a strong role in SMO formation and persistence.

Taken together, these findings point to a type of social movement convergence, in which organizational form and activity looks somewhat homogenous across national borders. I conclude with a discussion of the implications for future transnational social movements.

iii

Acknowledgements

I am forever indebted to several people, without whom this dissertation would not have been possible. I first thank my advisors and mentors, Dr. Rachel Dwyer and Dr. Andrew

Martin. Their contributions to this dissertation and my development as a scholar and person over the last six years are far too many to list, so I’ll just say they are everything I hope to someday be – as scholars, teachers, mentors, parents and friends. I also thank Dr.

Craig Jenkins, for his critical eye, unique insight and good humor as I have worked through this process. Craig has been extraordinarily generous with his time and energy, and I am a better scholar for his involvement in my career. I would also like to acknowledge Dr. Kazimierz Slomczynski, Dr. Irina Tomescu-Dubrow and Dr. Joshua

Dubrow. They introduced me to new areas of research, taught me new methodological skills and give me opportunities to see new parts of the world. I am also extremely appreciative to my gifted research assistants, whose hard work and enthusiasm helped make this dissertation possible: Ashley Shears, Kelly Isabella, Emily Norris, Peri

Strassner and Alli Hammond. This dissertation would not exist without their help.

Graduate school can be a grueling process, and I am thankful for the wonderful friends I’ve made here at Ohio State, especially Martin Kosla: my roommate for three years, officemate for six, and friend for life. I also acknowledge the G09 cohort, as I can’t imagine having a better group of people to take this journey with. The Department

iv of Sociology truly is a family, and I will always be proud to say I was a part of it. I also thank the hundreds of students I have been fortunate to teach at Ohio State.

I must also express my heartfelt gratitude to my family for their unwavering support of my academic and personal goals. Thanks to my parents, Ralph Schoene &

Eileen Corbett, for helping me understand the world, encouraging me to ask questions, and demonstrating a strong work ethic that I hope has become a part of me. Thank you especially for always encouraging me to the education I was fortunate to be getting, even when it seemed like I would never find my passion. Thanks to Kathy & Ray

McCarthy, for their constant encouragement and support. Thanks to Kenny, for distracting me with much needed sports discussions when I needed it the most. Thanks to

Jessica, for being on this journey with me. It hasn’t always been easy, but I couldn’t have done it without you. Last, but certainly not least, thank you to God, for everything and every day.

v

Vita

2009...... B.A. Sociology, Villanova University

2010...... University Fellowship, The Ohio State University

2011...... Graduate Research Associate, The Ohio State University

2011...... M.A. Sociology, The Ohio State University

2011-2015 ...... Graduate Teaching Associate, The Ohio State University

Publications

Schoene, Matthew. Forthcoming (2015). “Friend or Foe? The Media Coverage of Chicago’s Plan for Transformation.” Sociological Focus.

Kołczyńska, Marta & Matthew Schoene. Forthcoming (2017). “Survey as Context: Incorporating Metadata in Analyses of Substantive Issues in with Harmonized Survey Data.” Advances in Survey Methodology, eds. Timothy Johnson, Beth-Ellen Pennell, Ineke Stoop, and Brita Dorer.

Fields of Study

Major Field: Sociology

vi

Table of Contents

Abstract ...... ii

Acknowledgements ...... iv

Vita ...... vi

List of Tables ...... viii

List of Figures ...... x

Chapter 1: Introduction ...... 1

Chapter 2: Transnational Social Movement Organizations, Global Cities and Global Competitiveness ...... 18

Chapter 3: A Multilevel Study of Social and Political Activism in the European Union...... 49

Chapter 4: The International Media’s Coverage of the European Anti-Austerity Movement ...... 78

Chapter 5: Discussion and Conclusions ...... 104

References ...... 113

Appendix A: Tables and Figures from Chapter 2 ...... 128

Appendix B: Tables and Figures from Chapter 3 ...... 134

Appendix C: Tables and Figures from Chapter 4 ...... 140

vii

List of Tables

Table A.1: List of Social Movement Industries in Sample of Transnational Social Movement Organization Dataset, 1993-2003 ...... 128

Table A.2: Frequency and Percent of Global Cities Housing Social Movement Organizations in Sample of TSMOD, 1993-2003...... 129

Table A.3: Descriptive Statistics of Dependent and Independent Variables ...... 130

Table A.4: One-Way ANOVA Regression & Logistic Regression with Random Effects Models on Dependent Variables ...... 131

Table A.5: One-Level Logistic Regression Results of Organizational, Urban and National Variables on Organizational Viability ...... 132

Table A.6: Multilevel Mixed Effects Regression Results of Independent & Control Variables on Organizational Age ...... 133

Table B.1: Descriptive Statistics by Country ...... 134

Table B.2: Descriptive Statistics & Sources of All Dependent, Independent and Control Variables ...... 135

Table B.3: One Way ANOVA with Random Effects Models for All Dependent Variables...... 136

Table B.4: Percent of ESS Respondents at National Level Participating in Each Type of Social Movement Activism, 2010...... 137

Table B.5: Multilevel Mixed-Effects Logistic Regression Results of All Variables on Contacting Politician, Working for Party & Wearing a Protest Badge ...... 138

Table B.6: Multilevel Mixed-Effects Logistic Regression Results of All Variables on Signing a Petition, Boycotting a Product and Participating in a Demonstration ...... 139

viii

Table C.1: Number of Quotes by Institutional Affiliation in The New York Times, January 2010-March 2013 ...... 140

Table C.2: Sentiment of Quotes by Institutional Affiliation in New York Times, January 2010-March 2013 ...... 141

ix

List of Figures

Figure C.1: Weekly Count of Total Quotes from Sample of New York Times Articles, January 2010- March 2013 ...... 142

Figure C.2: Weekly Count of Pro-Austerity and Anti-Austerity Quotes from Sample of New York Times Articles, January 2010-March 2013 ...... 143

x

CHAPTER 1: INTRODUCTION

In the mid-1990s, Meyer & Tarrow declared that the world had become a “movement society” (1994). Modern capitalism is a system that produces winners and losers, and many of the classical theorists spend their lives thinking about the causes & consequences of capitalism, industrialization and , and suggesting alternatives that might alleviate human suffering. This style of thinking has long been a part of sociology, but capitalism has proven to be very flexible and able to change in response to external forces. When existing social structures benefit some more than others, conflicts inevitably arise. In addition, while is typically slow, it is far from impossible. Society changes in response to a multitude of factors, including population dynamics, economic shifts, global conflicts, and many, many others. It also sometimes changes due to the efforts of groups of people who organize collectively to press for social change, something called a social movement. However, their efforts are often bitterly resisted by those who do not wish to see society change. This conflict between those who fight for social change and those who resist it forms the very basis of modern society.

Diani defines a social movement as a network of informal interactions between a plurality of individuals, groups and/or organizations, engaged in political or cultural conflict, on the basis of shared collective identities (1992). When Meyer & Tarrow

1 deemed our world to be a movement society, they meant that movement participation is a normal and rational part of life (1994). What gave rise to the movement society, though?

Why do people participate in grassroots activism when the costs may be high and the rewards low or non-existent? How do social movements overcome Olsen’s classic free- rider problem, where because any potential gains won by a movement will be enjoyed by everyone, but the costs limited to only active participants, the rational choice is actually abstention (1965)? While students of social movements have greatly advanced our understanding of all these questions in the past several decades, recent trends demand increased attention on a few important factors that have the power to reshape the social movement landscape. Specifically, as the movement society emerged, it did so in an increasingly urban, globalized world. At some point in 2007, our world officially became an urban one. For the first time in human history, city-dwellers outnumbered the rural.

Cities now house the majority of the world’s people, a disproportionate share of global economic activity and typically have a highly stratified, unequal, polarized . Sassen has speculated that this is a recipe for potentially explosive social conflict within the world’s cities (2006), but the true role of urban space as an incubator of social movement activity is not yet fully known. Secondly, as Sassen also points out, the cities of the world are more connected than ever before (1990; 2006). This process, known as globalization, allows social relationships take on a distanceless and borderless quality, so that life is increasingly experienced in a global rather than a local sense

(Scholte 1997). While technology certainly reduces the importance of physical proximity in the contemporary world, and one can connect with others in faraway lands rather easily, most social movement activism still occurs locally, albeit with new transnational 2 elements (Smith 1998; 2001; 2008). How do globalization and urbanization influence social movement activity?

As these forces continue at a breakneck pace, the above question is timelier than ever. It is imperative to understand how the urban context influences in the globalized world because social movements represent a crucial pathway to advancing social justice. Global capitalism has made people richer everywhere it has ever appeared, but it has also generated wealth unevenly and left many people struggling to keep up.

While the global economy has grown exponentially, inequality has also risen in nearly every industrialized country on earth, even in traditionally low-inequality such as those of Northern Europe. Women still lag far behind men on many indicators of gender equality. The planet’s climate is at a crisis point, and global society needs to take immediate and comprehensive action to avert a catastrophe. Millions upon millions of people live in overcrowded, dangerous and polluted slums in the cities of the developing world, where 90% of future population growth is projected to be. These social problems are everywhere we look, and we know social movements can be an important part of the solution to many of these long-standing problems. Therefore, it is imperative to understand what motivates social movement activity, how cities influence these relationships, what factors produce strong SMOs and how this process plays out in a cross-national context. The next half-century will be a crucial point in human history.

This dissertation takes a step towards answering these questions, but before I introduce the specific research questions addressed here, I first must discuss what is known about , , globalization and the methods commonly used to study collective action. 3

Social Movement Theory

Social movements can transform society, but there is no guarantee any one movement will be successful. Because of their role as potential agents of change, social scientists have long been interested in social movements, uncovering why people participate, how informal movements coalesce into formal social movement organizations (SMOs), what factors encourage healthy SMO development and what potential allies are available to grassroots SMOs. Several broad, macro-style theories have emerged to explain social movement activism, which I briefly describe here. Classical theories of activism viewed participation as an irrational phenomenon that emerged because certain members of society were poorly integrated into social institutions (Blumer 1951). Therefore, people would turn to protest or even destructive behavior, such as rioting, as a last resort (King

1956). Relative deprivation theory, an offshoot of classical strain theory, follows a similar logic. Essentially, social movements arise from a combination of grievances, dissatisfaction and relative deprivation in comparison to both reasonable expectations and the living standards of others around them (Gurr 1970).

Relative deprivation and grievances add important insights to our understanding of collective action, but a shift occurred in the 1970s. Scholars who themselves participated in the of the 1960s realized that existing theory did not match up with their experiences (Morris & Herring 1984). In response to this mismatch between theoretical expectations and experiences, McCarthy & Zald’s landmark theory linked movement formation and outcomes to the level of human, material, organizational, moral and cultural resources that SMOs could marshal (1977). 4

SMOs exist as a formal manifestation of the social movement sector, or the totality of social movements within a society. While some social movements eschew formal organization to avoid hierarchal structure and the iron of oligarchy (Michels 1962), like anarchist movements or the recent 99% , most movements eventually produce formal organizations (McCarthy & Zald 1977) and are usually better off for it

(Freeman 1979). The social movement sector diverges into various movement industries, which are groups of similar social movements that often compete with one another for the same resources and participants. Using this framework, resource mobilization theory makes four key assumptions: that movement participation is normal and rational, that grievances are relatively constant in society but mobilization is not (Jenkins & Perrow

1977), that most resources can be mobilized from within a group’s constituency, and that the media acts as a crucial middleman in social movement conflicts. Social movements emerge and succeed or fail based on the amount of resources they have and how wisely they allocate those resources (Martin 2008).

However, movements exist within broader society, and are unquestionably influenced by the of the time and place they inhabit. Meyer, paraphrasing one of

Marx’s more famous statements, points out that “movements make history, just not in the circumstances they choose” (2004). What about the political context? theory attempts to account for how the can either constrain or encourage social movements (Meyer & Minkoff 2004). Within the political process model, there are four key conditions for movement activity (McAdam 1982): an insurgent consciousness, organizational readiness, an open political system and a low level of repression and social control. Movement activity will peak and be most 5 successful when conditions are right (Amenta & Zylan 1991; Amenta et al. 1992;

Brockett 1992).

While political opportunity theory produces some unique insights about the role of the state, it does not address how both influences and is influenced by social movements. The framing perspective posits that movement outcomes hinge upon how successfully movements can classify problems, identify who or what is responsible and suggest a course of action to rectify the situation (Snow et al. 1986). Master frames, derived from the core values of society, help groups create action frames, or the specific diagnostic framework used by SMOs to gain legitimacy (Snow & Benford 1992).

Groups must construct frames that resonate with a large segment of the population, or they risk irrelevancy (Babb 1996). Related to the framing perspective is collective identity, which can help overcome Olsen’s free-rider problem and mobilizes people on the basis of solidarity (Polletta & Jasper 2001; Kriesi et al. 1992). In the era of “new” social movements, movements tend to coalesce around identity, not social class. This differs from the classic Labor Movement, which drew upon the identity of workers (Fantasia 1998). This allows a social movement to mobilize more resources than they otherwise could, both human and material.

While the above theories have produced important insights into , much of this information speaks to social movement in a micro framework.

Too often, we learn a great deal about social movement emergence, operations and outcomes without knowing if this knowledge is generalizable to other movements or outside of the historical and national context in which the evidence was gathered.

Secondly, the vast majority of the current knowledge about social movements comes 6 from American or, slightly less commonly, Western European research. While some scholars have attempted to explain cross-national differences in movement emergence and activity, comparative movements have received much less scholarly attention.

Finally, while it is not known precisely how urban areas incubate social movements, there are many reasons to believe that cities provide the conditions necessary for the formation of strong social movements. I now turn my attention to this issue.

Urban Sociology

Meyer & Tarrow’s movement society has emerged during a time of unprecedented population shifts which have made this an urban world, and the city is now the primary lived experience for the majority of the planet’s inhabitants. The city can be broadly defined as a densely populated built environment that reflects the needs of capital, mediated by state policy, to promote economic activity and consumption. While ancient cities first formed when humans shifted from a hunter-gatherer lifestyle to an agrarian society, cities as we currently know them arose during the industrial (Hawley

1981). This trend shows no sign of slowing down. The world is just over half urbanized at the present date, but demographers project that over 70% of humanity will be city- dwellers by 2050, with this trend primarily driven by rural-urban migration and high fertility in the cities of the developing world (Chou et al. 2012). This mass movement of population, economic and social activity from the hinterlands to the cities is one of the most important trends in modern global society. Over time, urbanization has proven to be highly beneficial. Cities expand the reach of markets, raise living standards, promote a better match between the supply and demand of labor and diffuse technological 7 advancements more efficiently (Barioch 1988). The workers who first moved from the countryside of Britain, Germany, , the United States and others industrializing countries experienced a much higher standard of living, something that even urban disamenities like pollution and disease couldn’t erode (Williamson 1981).

In much the same way as social movement theory has undergone several transformations in the , so too has urban sociology. Classical urban ecology is concerned with urban form, development and land use patterns. It assumes that cities develop to facilitate economic production in the most efficient way possible. For example, the classic concentric zone theory used Chicago as its model, arguing that development occurs in circles that form around a central business district, with each zone reflecting a market need (Burgess 1925). Sector theory alters this argument and posits that cities expand outwards on major thoroughfares that begin in the central business district and extend out towards the suburbs (Hoyt 1939). This produces zones of development, not concentric circles. Multiple nuclei theory argues that uneven development occurs because certain sectors need proximity, while others demand distance, so urban form reflects individual sector needs (Harris & Ullman 1945). Modern urban ecology has pushed these ideas in new directions, but the unifying theme of urban ecology is that cities exist as a response to market needs (Berry & Kasarda 1977; Storper

& South 2009). However, perhaps more important than the question of why cities look the way they do, is the question of how urban form and features influence the people living there.

Urban ecology also fails to account for the very simple concept of power. Cities unquestionably serve market needs, but who is benefiting from the decisions that get 8 made in local urban politics? As many industrial cities of the American Northeast and

Midwest emptied out and decayed, who makes the decisions that redevelop the city towards a new function or purpose? Urban political economy argues that local and state policies create cities that serve the needs and desires of those with the power to make the city look and operate a certain way (Harvey 1982; Castells 1977). David Harvey, adapting Leferbve’s concept of circuits of capital, argues that cities exist to facilitate primary circuits of capital. That is, the processes of production and consumption.

Productive capitalism requires bodies to be in certain places, so that they can produce and consume. This production is known as a primary circuit of capital. The built environment, then, serves to facilitate this process, so things like houses, neighborhoods, factories and offices are secondary circuits. They exist as a support system for the process of production and consumption (Harvey 1982; 1985). Logan & Molotch advance

Harvey’s ideas by conceptualizing urban development and redevelopment as a process in which institutionalized actors, all of whom share a vested interest in growth, form a growth coalition that transforms the city into a growth machine to serve their needs

(1987). This includes business leaders with strong ties to a particular city, local officials, government officials, real estate developers and locally bounded institutions such as and hospitals (Logan et al. 1997; Cox & Mair 1988).

The political economy perspective advances urban sociology by putting the focus on the actor, precisely what classical urban Marxism is lacking. While cities are conceptualized as a reflection of economic restructuring of wider society, urban Marxism conceptualizes neighborhoods as little more than locations where the working class can be reproduced. This style of thinking lead some to ask whether there was even such a 9 thing as urban sociology, because the issues of the city were theorized as inseparable from the issues of macro-level society (Castells 1976). But one is left to wonder, is that all the bright lights of Paris are? Are the dynamic streets of New York City only a place of social reproduction? Really, cities are far more than just the buildings and people that exist within a physical space. They have their own unique culture, networks and ways of life. They can encourage or discourage collective action.

In fact, urban Marxism encourages us to start thinking about how actors work to shape the city, and scholars have long speculated that cities may produce better conditions for social movement activity (Castells 1977; Sassen 2006). Much of this attention is paid to how elites operate within local urban conflicts, but the concerns of neighborhood actors cannot be and are not ignored (Zukin 2011). Most scholars point to post-industrial cities as sites of great polarization, inequality and conflict (Sassen 2006).

Cities lay the groundwork for social movement activity in response to class conflict.

Essentially, cities can act as “social movement spaces,” physical areas where activism can thrive (Nicholls 20009; McCarthy & McPhail 2006). Furthermore, while conflict is experienced locally, the causes and consequences of urban social movement activity can stem from trends far outside the city limits (Peterson 1981). It is important to think about urban social movement activism in a transnational context, an issue I discuss next.

Globalization

As mentioned earlier, much of the current knowledge about social movements comes from American or Western European research, and this limits generalizability to a single movement or single national context. This is a major shortcoming, as many social 10 movements are starting to operate in a transnational framework (Smith 2001; 2008).

Globalization changes the operations of social movement organizations in some crucial ways. I previously noted that Scholte defines globalization as distanceless and borderlessness (1997). A more expansive way to describe globalization, however, might be that it is the “cluster of technological, economic and political innovations that have drastically reduced the barriers to economic, political and cultural exchange” (Drezner

2001). Originally, this term described purely economic linkages, but putting previously isolated countries and areas into contact with one another changes more than just the economy. It changes everything about society, and social movements are no exception

(Smith 2001). In fact, globalization has even produced movements expressly aligned against globalization, which speaks to the power of this concept (Castells 1997; Smith

2001).

How might globalization influence the social movement society? In recent years, scholars have pointed out that globalization opens new avenues for activism (Byrsk 1993;

1994), facilitates the growth of new institutions (Bartley 2007), diffuses resources, tactics and frames (Giguni 1998), diminishes repression (Maney 2002) and connects important potential allies like IGOs to SMOs (Lewis 2002). World Polity Theory, specifically, theorizes that globalization allows a world culture to emerge that countries and institutions must align themselves with to reap the benefits of globalization. This leads to a degree of “sameness” in institutional form (Kerr 1983; DiMaggio & Powell 1983) and policy (Drezner 2001). It is theorized that the emergence of a world polity will create movements that mimic one another and generally look similar across national borders

(Giugni 1998b). 11

However, while scholars have looked for convergence and divergence in places like the policy arena and institutional form, it is difficult to know exactly how the social movement sector is influenced by globalization, as researchers in this discipline have struggled to study movements in a comparative, cross-national setting. In this dissertation, I address this lack of comparative evidence by performing large-scale cross national studies of social movement organizations and activity, trying to find out what is similar and what is different in the various regions of the world. Of course, cross- national research of social and political behavior has certainly not been ignored.

Important examples of global social movement research include studies of protest waves in non-democratic countries (Almeida 2003), the emergence of private regulation of environmental and labor conditions in various industries (Bartley 2007), indigenous rights movements in South America (Brysk 1994), global civil society (Smith 1998), and political opportunity structure in different countries (Kitchelt 1986; Lewis 2002), among other questions. However, our knowledge of social movement activity and globalization is still rather limited, in large part due to the methodologies used to study comparative movements.

Methodology

Much of the success of the study of social movements can be attributed to the methodological diversity and uniqueness found in the field (Klandermans & Staggenborg

2002). In addition to the major theoretical breakthroughs sketched out above, the field has long been known for methodological innovation. Resource mobilization focused not just what a resource was, but how to measure it. Political opportunity was concerned with 12 not just the political system, but how exactly to study and compare structure. The same can be said for the other subfields in social movement research. Compared to the flurry of social movement research in the 1970s, we are much better informed about how movements emerge, act and end. Survey research, event data analysis, network analysis, organizational studies, in-depth interviews and participant observation have all been instrumental to these theoretical advances (Klandermans & Staggenborg 2002).

The major gap this dissertation seeks to fill regards the intersection of social movements, urbanization and globalization. Quite simply, studies of urban social movements rarely account for the cross-national context, while comparative movement research rarely examines what is special about the urban context. Essentially, I ask how different levels of context influence social movements: the global context and the urban context. Activism is obviously influenced by multiple factors. When people make the decision to join a social movement, or to protest, they do this based on their individual traits and the characteristics of the area in which they reside. Similarly, when SMOs form, their success or failure depends on both organizational characteristics and on contextual factors surrounding the organization. Social movement research must account for factors at both the individual and contextual level. However, the methods used to study social movement activity does not fully account for variation at both levels of analysis. Beyond the obvious nested structure of the data commonly used to study comparative movements, research must also account for the fact that people are not randomly distributed to locations, and those residing in one country are more similar to each other than they are to the residents of another land. The best way to account for this context is by using multi-level modeling, which is not common in in the study of social 13 movements. Therefore, I use two-level regression models in this dissertation to account for the nested structure of most social movement data.

Contributions of the Current Study

Flowing from this preceding discussion about social movements, urbanization, globalization and research methods, the following dissertation makes several contributions to the literature. First, I make a point to study how cities influence both social movement organizations and individual social movement activity, something that is not yet well understood. Second, I study social movements in a cross-national, comparative framework to help make the current knowledge more generalizable and comparable across national contexts. Third, while it is problematic that most of the current knowledge about social movements is domestic in scope, it is perhaps even more problematic that most research in this area focuses on one or a small handful of movements. This further limits generalizability. Therefore, I use quantitative, large- scale data to study social movements, and I do this at both the organizational and individual level of analysis. Fourth, from a methodological standpoint, the methods we use to study social movements also limit our generalizability. In response, I use multilevel modeling. This is crucial for not just understanding how movement activity differs across national context, but also for understanding how cities act as movement spaces independent of just the people who happen to be living there.

This dissertation is organized in the following way: chapter two examines transnational SMOs in global cities, or the most important economically important cities in the world. As centers of global society, there is reason to believe that global cities, the 14 most economically competitive, internationally-oriented places, may offer a better environment for the development of successful SMOs, and that the more globally competitive a city is, the better it will act as a movement space. So, does the global competitiveness of a city help produce organizations with attributes known to be important to success? Using the Transnational Social Movement Organization Dataset

1953-2003 (Smith & Wiest 2012), I fit a series of mixed-effects multilevel models to examine the effect of organizational, urban and national variables on two outcomes of

SMO health: organizational persistence (survival) and organizational age. Regression results of 672 SMOs in 67 cities from 1993-2003 indicate that urban status, professionalization and the level of resources best predict SMO persistence.

In the third chapter of this dissertation, I turn my attention to individual-level social and political activity. The contemporary European Union (EU) is currently experiencing a protest wave in response to recent demographic shifts and economic crisis. Protest waves offer the opportunity to study what motivates explosions in social and political activity. Using a sample of over 37,000 residents of 20 EU countries in the

5th wave of the European Social Survey, I study how relative deprivation, resources and urbanization influence an individual’s likelihood of participating in six different forms of social and political activity: contacting a politician, working for a , wearing a protest badge, signing a petition, boycotting a product and participating in a lawful demonstration. A series of multilevel logistic regressions indicate that urban status and the level of resources, both individually and regionally, best predict the likelihood of participating in the above activities.

15

In chapter four, I take a closer look at Europe’s protest wave by studying one especially controversial issue: the ongoing austerity policies implemented all over the region. There is a long history of identifying important allies and resources that social movements need to mobilize in order to be successful, and one such ally is the media.

While the media’s role in social movement disputes is well-studied, most studies focus on the local media and how they cover social movements. However, the local media is no longer the only outlet SMOs may try to engage. How do SMOs attain media standing in a major media outlet based in another country? Globalization and diffusion makes it easier for local movements to make transnational claims, and international media sources are some of the most accessible and respected outlets in the world. These international media outlets may help SMOs make international claims. So, I ask how successful are grassroots actors in gaining media standing when compared to institutionalized, bureaucratic elites? I perform a content analysis of New York Times articles about the anti-austerity movement in Europe from January 2010 through March

2013, the heart of the austerity crisis. Results indicate that institutionalized, bureaucratic actors indeed had a much easier time achieving media standing in the Times then their grassroots counterparts, but the aggregate coverage was slanted in the anti-austerity direction. Despite this decidedly anti-austerity sentiment, austerity still rocked many

European counties, a finding which calls into question the power of the international media to influence local politics.

Finally, chapter five closes with my attempt to synthesize this dissertation’s findings and what they mean for the future of social movements in the urban, globalized world. I draw from the insights of World Polity Theory to help explain the major 16 findings of these studies and how social movements organizations are starting to resemble each other even across national borders. I then discuss what this means for social movement tactics, the future of social movement activism in cities, and suggest areas for future research. The concepts explored here, urbanization, globalization and austerity, are not going to suddenly disappear or decline. People will not start a mass- migration back to the countryside, nor will the global economy suddenly take a backseat to domestic concerns (no matter how badly some Europeans want this to be true). These are trends that are only expected to increase. We know social movements have the power to improve lives, but the simple fact is that most movements fail (at least in the short term). This dissertation, then, is an attempt to uncover what these movements have in common, how they differ, and what this tells us about the world.

17

CHAPTER 2: TRANSNATIONAL SOCIAL MOVEMENT ORGANIZATIONS,

GLOBAL CITIES AND GLOBAL COMPETITIVENES

Does the global competitiveness of a city contribute to the organizational persistence of grassroots, transnational social movement organizations? By global competitiveness, I speak of the city’s role and integration in the global economy. Global competitiveness reflects two of the biggest trends in recent history: globalization and urbanization, both of which have occurred hand-in-hand since the mid-1970s. Globalization has brought untold changes to both domestic and international economies. Concurrently, cities have also emerged as the primary place for social, cultural and economic activity. These simultaneous forces have naturally created winners and losers inside the global system, both at the individual and national level. Due to the restructuring of the global economy

(especially in low or middle income countries), free trade agreements, deregulation and open borders that facilitate the movement of people and goods, the economy has grown exponential over the past 30 years. However, substantial chunks of humanity are either not seeing their living standards rise or are actually falling behind. This is often framed as a problem for developing countries, but citizens in many high income countries are also seeing their wages stagnate and their living standards threatened. In response, many

SMOs seeking to address these injustices have sprung up all over the world, many of which are expressly transnational in focus, operation and orientation (Smith 2001; 2008).

18

Furthermore, many of these SMOs are located in the world’s biggest, most influential and most economically competitive cities. How are SMOs influenced by their location?

This question is important because for the first time in human history, the urban population outnumbers the rural. While it would seemingly be easier than ever for humanity (and economic activity) to spread out and decentralize due to technology that allows people to communicate and interact instantly across great distances, cities instead have assumed a place of supreme importance in global society (Sassen 2006). The most important urban locations are now known at global cities (Taylor et al. 2011; Freidmann

1986; Sassen 2001; 2006). Global cities are important nodes in the world economy and are responsible for large chunks of national and regional economic activity. They serve as more than just command and control centers for global economic activity, however.

Places like New York, London and Tokyo are also spaces of dynamic social and cultural activity. Some scholars, in particular Sassen, have speculated that global cities may provide a favorable environment for the formation and success of social movements.

Global cities typically have high levels of stratification and inequality, so they are places where the world’s inequalities are laid bare (Sassen 2006; Castells 1983). That being said, global cities have high levels of both wealth and investment when compared to both rural areas and less-globally oriented cities, and the higher level of material and economic resources may help SMOs. Furthermore, they also tend to have large populations in dense areas, which may provide networks that encourage activism (Walton 1998). These networks include not just citizens, but also important IGO and NGO contacts, who can share information and resources with grassroots SMOs (Keck & Sikkink 1999; 2014).

However, the evidence regarding cities as movement spaces is relatively sparse, and it is 19 even less clear how this process works in a cross-national context. It is crucial that we better understand how global cities influence SMOs, as these transnational organizations represent an important pathway to advancing social justice in the areas where humans increasingly live (Smith 2001). Cities stand at the forefront of rising inequality in the global system, both within countries and between countries. Can transnational SMOs be an agent that helps alleviate this inequality, and if so, does global competitiveness (a source of this very same inequality) bestow organizational advantages upon these SMOs?

Despite this question’s obvious importance, however, there is nothing even resembling consensus in the literature on how exactly to define success for any one movement, let alone several movements across many different countries (Giugni 1998b).

A standardized measure of SMO outcomes that is applicable across time and national borders is necessary. Rather than try to define SMO success, this study focuses on factors known to contribute to strong, healthy, stable SMOs: organizational viability and organizational age. In other words, I focus on organizational persistence. Obviously, no

SMO can ever hope to achieve their goals if they disband, making viability perhaps the most rudimentary metric of success. Secondly, concerning age, I assume that older organizations tend to be more institutionalized, stable, respected and ultimately in a better position to achieve their goals, whatever those goals may be (Minkoff 1999). While these outcomes are admittedly an imperfect measure of SMO success, I accept these limitations because my outcomes allow me to address a second and crucial second gap in the literature: the lack of generalizable, cross-national findings. Most studies of social movements focus on one or a small sample of movements or SMOs in one or a limited number of countries. This is a significant limitation in a world that includes broad 20 diversity of -states, systems of government, cultural conditions and cities.

Additionally, it is crucial to understand these questions in a cross-national context, which can then be used as a starting point for deeper analyses into the social movement sectors of specific cities. Finally, a third shortcoming in the literature regards the methods commonly used to study social movements. Of course, SMOs are influenced by both organizational and contextual factors. This necessitates multilevel modeling, and there are very few examples of a hierarchical analytical strategy in the study of social movements, mostly because so few cases are studied in each analysis. In this chapter, I begin to address these shortcomings in using the Transnational Social Movement

Organizations Dataset (Smith & Wiest 2012) Using a sample of 672 transnational

SMOs in 67 global cities located in 47 countries, I test the effect of organizational, urban and national predictors on both the viability and age of SMOs. Results indicate that organizational factors best predict organizational viability, but it is urban resources that best predict organizational age. I then conclude with the implications of these findings for social movement theory and future research.

Theoretical Background

As technology increasingly made physical place less important to business transactions, urban planners feared that cities would become obsolete. You can send money from New

York to Tokyo with the push of a button and speak to a colleague in London with the push of a button. Would cities become meaningless? No, and in fact, precisely the opposite occurred. Social and economic activity has actually recentralized into what

Friedmann called “world cities” (1986; 2001). Sassen later refines this concept and 21 identifies three core requirements a city must meet to be considered a “global city.” First, global cities are command and control centers for the world economy. They house the major firms that manage and oversee business operations all over the world. Global cities also provide marketplaces for specialized financial services, which make up a disproportionate share of global economic activity. Finally, they are key site of production, although this also includes the production of innovation, not just material goods (e.g. major research universities). Within the world system, urban areas mesh into an interconnected hierarchy of global dominance, including high-income and low-income cities, as well as anything in between (Taylor et al. 2011). This is a crucial point. A city’s global status is linked to, but not entirely dependent on, the level of development within the region. These cities attract a disproportionate amount of foreign direct investment, headquarters and branch offices of major multinational corporations, and immigrants from both within and outside the country. This development allows cities to become far wealthier than their surrounding areas would indicate (Alderson & Beckfield

2004). This leads to much higher polarization within the city limits, a primary driver of current urban inequality (Alderson et al. 2006).

In a review of globalization and its effect on urbanization, King declared that due to the increased importance of the global economy to urban well-being, all cities were subsequently world cities (1990). Nevertheless, some cities are clearly more important to the world economy than others (Knox 1995), and a great deal of recent research has empirically examined the organization of global cities and their connections with other urban areas (Smith & Timberlake 2001; Taylor et al. 2011, for example). Global cities are dynamic places. Their position in the global hierarchy may rise or fall over time 22

(Smith & Timberlake 2001), such as Toronto outpacing Montreal as Canada’s most globally competitive city in the 1980s or Berlin challenging Frankfurt in the years following German reunification (Sassen 2006). They may even become so economically powerful that these cities eclipse the system of nations that contain them, something that is increasingly becoming the norm in developing countries prone to urban primacy

(Taylor et al. 2011). As urban scholars have greatly expanded our understanding of global cities and their importance to the world system, students of social movements have concurrently wondered if globalization may help the SMOs they also house (Castells

1977; Sassen 2006).

Before that question can be addressed, however, an even bigger theoretical question remains. How exactly does one define and measure SMO success? Giugni argues that it is extremely difficult, and perhaps impossible, to designate social movements as successes or failures (1998a). This is especially true when considering ongoing movements. Furthermore, Giugni also argues that it is inappropriate to measure the success of a social movement by looking at whether or not they achieved their stated goals (1998a). For one, goals can differ radically across SMOs working within the same social movement (McCarthy & Zald 1977). Two, the consequences of activism are long lasting for those involved both organizationally and individually, so what appears to be a failed social movement may ultimately produce conditions for successful activism in the future (McAdam 1989). Movement “success” can only be examined across time

(Andrews 2001). Nevertheless, Giugni also argues that even when SMOs fail to achieve their stated goals, just staying alive and active is a victory unto itself (1998a). This is a crucial point. No SMO, big or small, can ever hope to achieve their goals if they disband. 23

Thus, rather than study social movement success, I study two outcomes known to influence success: organizational viability (survival) and age. Viability is an obviously positive outcome for SMOs, but I also study age because I assume that older organizations will both be more likely to continue their activities and in a better position to eventually achieve their goals. They tend to be more institutionalized, more well- known, and are taken seriously by other important actors.

What factors encourage SMO viability? Much of the current knowledge of this question comes from studies of the American or Western European social movement sector. In a study of homeless advocacy SMOs in the US, Cress & Snow find that organizational viability was encouraged by access to important resources, like members, technical and logistical support and meeting space (1996; 2000). Similarly, Edwards &

McCarthy find that the healthiest SMOs are the most active ones, the groups who actually hold events that can generate specific resources necessary for the organization to continue existing (2003). In addition to mobilizing resources, social movements must also effectively manage and distribute resources in order for them to matter (Martin 2008).

Social capital (Edwards & McCarthy 2004) and networks of activists (Gould 1991) also encourage the development of strong SMOs. Additionally, evidence indicates that formal organization bestows a myriad of advantages upon social movement groups.

Professionalized SMOs are better able to ride the highs and lows of public support, which can be very unpredictable (Staggenborg 1988). Taylor and her colleagues show that the women’s movement entered a period of “abeyance” where the movement struggled to accomplish anything of tangible value, but kept their networks and professional structure in place (1989). The political opportunity structure waxes and wanes, and 24 professionalized SMOs with a committed membership were better able to survive these lean times. They then came roaring back when society was more receptive. Similarly, building institutions and organizations provided an important baseline from which to mobilize from at a later date during the desegregation of public schools in Mississippi

(Andrews 2002). This finding applies to in addition to traditional social movement groups (Meyer & Staggenborg 1996). So, in sum, the literature points to professionalization, network of activists, resources and well-connected political allies as crucial predictors of SMO viability and health.

In recent years, scholars have also begun to explore how globalization influences

SMOs all over the world. Smith uses the 1999 IMF riots in Seattle as a crucial crossroads for social movement activism (Smith 1998; 2001). Specifically, Smith argues that while activism used to be a local or national phenomenon, the Seattle incident shows how globalization opens up new networks, sources of funding and sources of information to social movement groups (2001). Smith observed this in Seattle, but it is also true in resource-poor areas, where SMOs can tap into the global flow of resources they previously had no way to access. These countries are often trying to create a social movement industry with little prior experience and fewer resources to draw upon. Maney argues that globalization encourages protest activity because it promotes , decreases political exclusion, disrupts traditional political alignments and diminishes repressive activity by the state (2002). All this combines to lower the cost of activism, and increases political opportunity. A more globally oriented society also opens up new avenues of information to activist groups (Rothman & Oliver 2002).

25

Globalization can also link SMOs with potential allies, like individuals in wealthier countries (Nepstad 2002) or other non-grassroots organizations like intergovernmental and nongovernmental organizations (IGOs & NGOs) (Caniglia 2002).

These actors and organizations make up a significant part of the networks of activism

Keck & Sikkink theorize about, in which international linkages form to the benefit of the social movement sector (1999; 2014). IGOs and NGOs are especially important pieces of the puzzle, because they represent the space and medium through which new opportunities, transnational diffusion of ideas, and transformation of state actors takes place (Reimann 2002). While IGO and NGO contacts were crucial in very diverse situations, from anti-dam movements in Brazil (Rothman & Oliver 2002) to the IMF protest in Seattle (Ayres 2002), the effect of contact with IGOs is not the same as NGOs.

IGOs tend to operate through grassroots organizations by linking them to government actors and acting as middlemen to transfer funding, grants and information. Contact with them should be beneficial for SMOs. However, NGOs act in a similar way to other grassroots SMOs, and are often in competition for the same resources. They cannot really be described as grassroots as they are often founded at the behest of bodies like the

UN or large foundations. NGOs then compete with grassroots SMOs for the same resources (Jad 2007). NGOs are not necessarily a bad thing for the social movement sector to develop, but they may damage SMO viability and age by providing competition for the very same things that SMOs need to persist (Bendaña 2006).

There is a clear connection here between more recent, globally oriented findings and the traditional studies of activism in the US & Europe. Resources, politically important allies, formal organization and all seem to encourage a strong social 26 movement sector, even in areas that should otherwise struggle to produce activism.

These are crucially important insights into globalized collective action, but the current understanding of this process does not expressly integrate how geographical factors influence SMOs. Often, macrosociological theories of SMO formation and activity take the geographical concentration of people, resources and economic activity for granted.

This is reflected in the way in which previous work theorizes about global cities, originally identifying this concept as a dichotomous indicator. A city is either global, or it is not (Timberlake 1985). However, more recent formulations of the concept view global city status as more of a continuum. While Timberlake identifies three kinds of cities (world, semi-periphery and periphery) which all serve different functions in the global hierarchy, it is useful to reconsider the factors that make a city “global” and how this influences the people living inside them. In fact, nearly all of the characteristics identified as making a city global are best conceptualized as continuous variables, so it is worthwhile to study variation inside global cities rather than just compare global to non- global cities.

In this paper, I hypothesize that the more global a city becomes, the stronger of a social movement sector that city will produce. Global cities do this because they increase relative deprivation in the area and provide resources that encourage social movement mobilization. Global cities house massive populations (Ades & Glaeser 1995) that live in very dense environments, something that can increase communication networks among people likely to participate in social movement activity (Hedstrom 1994). Furthermore, they provide transnational networks of activism (Keck & Sikkink 1998; 2014) that connect local SMO to international resources and allies like IGOs, NGOs, who can be 27 important sources of information to grassroots organizations. These contacts are also typically clustered in global cities, with the most global cities housing the greatest number of IGO and NGO groups (Bouteligier 2008; Taylor 2004; Su et al. 2014). The emergence of global civil society (Smith 1998) tends to be centered in urban areas.

Within this dense environment, people of very different ethnic and class backgrounds usually live side by side, and the inequality and polarization of the world system is laid bare for all to see. Relative deprivation has always been a more powerful motivator of social and political activism than absolute deprivation (Blau & Blau 1982; Walton 1998), and the most global cities are often places of the most evident polarization and inequality

(Sassen 2006). Global cities may be places of immense wealth and growth (Kasarda &

Crenshaw 1991), but it is impossible to hide the polarization that often accompanies growth.

In addition to these population dynamics, global cities are also much wealthier and provide much better access to material resources to SMOs. The more globally competitive a city is, the more material resources tend to be concentrated there (Ades &

Glaeser 1995). These resources include personal and regional wealth and also infrastructural advantages like telecommunications that allow SMOs to operate more efficiently and effectively. Money and material resources are consistently one of the most powerful indicators of a strong social movement sector, and the more global a city becomes, the more resources it tends to hold. Finally, global cities become integrated into a world polity (Meyer et al. 1997) that attracts a disproportionate amount of global media attention and encourages transparency and democratic development.

28

To summarize, all of the above factors are not dichotomous but rather continuous.

Population, population density, inequality, polarization, networks of activism and wealth are all things that run along a continuum, which means within global cities, some are more globally oriented and competitive than others. I argue that within global cities, the more globally oriented cities offer the best spaces for SMOs to succeed. Regional wealth is a characteristic of global cities, and some cities are wealthier than others. This same idea holds for population density, foreign direct investment, polarization, network formation and democratic values. Therefore, global cities cannot be considered in a dichotomous fashion, but should instead be studied continuously.

Furthermore, while we have access to knowledge about what factors influence

SMO persistence in various countries, we need more knowledge about whether or not this evidence is only applicable to the country in question or to the world system. In this paper, I employ a cross-national sample of organizations drawn from every corner of the globe, using multilevel modeling to account for cross-national differences. Ultimately, I ask whether or not we can find any commonalities in the factors that drive SMO persistence across many different contexts. Essentially, while I accept there are differences in social movement sectors across national borders, I seek to isolate the factors that produce comparable results within the world system. Furthermore, since I seek to examine how globalization has influenced the social movement sector, I study social movements in the world system in a theoretically limited timeframe of data: 1993-

2003. The 1990s were a time of intense globalization, both economic and political.

President Clinton took office in 1993 and NAFTA became official policy in 1994. The

Soviet Union also collapsed and brought many new nation states into the world system 29 during the 90s, exposing them to free-market conditions for the first time in generations.

This 10 year period is ideal to study how global forces affect the social movement sector.

I next turn my attention to the data and modeling strategy I use to accomplish my goals.

Data and Methodology

Data

For this project, I drew data from the Transnational Social Movement Organization

Dataset 1953-2003 (TSMOD) (Smith & Wiest 2012). The TSMOD contains information about 10,092 transnational SMOs, collected every other year from the Yearbook of

International Organizations. All organizational variables contained in the TSMOD are taken from this yearbook, or are derived from yearbook information. Not every organization qualifies as an SMO, however, and the authors explain how they decide whether to include or exclude an organization: “The main criterion for inclusion is that the organization's primary purpose is the pursuit of social change goals (broadly defined).

The group should be autonomous (i.e., not directed or led by a government or international agency) and not-for-profit“ (Smith & Wiest 2012: p. 4). The authors also exclude: IGOs, NGOs, foundations, trusts, funds, religious organizations and sections or committees of other international organizations.

Of course, the TSMOD comes with a few limitations. First, some important regions of the world are not represented, chiefly China. No one would seriously argue that because the TSMOD does not list any Chinese organizations that no such organizations existed, but this is unavoidable in areas where SMOs are better served not being public. A second shortcoming of the TSMOD is that it ends in 2003, so almost an 30 entire decade has passed at the time of this analysis. That being said, this is still the most comprehensive listing of transnational social movement organizations available to researchers, and for the purposes of this study, the 1990s are the key timeframe of interest. In addition to the TSMOD, I supplemented these organizational variables with urban and national indicators.

My unit of observation is the organization itself, and the TSMOD contains information on 10,092 yearbook entries in numerous countries, social movement industries and specific years. However, I reduced the sample to 672 organizations, dropping much of the sample for several reasons. First, since this paper examines SMOs in global cities, I dropped any organization that did not specify the city where that organization was headquartered. I then dropped SMOs not headquartered in a global city, using The Economist magazine’s global city metric. To be clear: this means that the following study examines the effect of global competitiveness within global cities, not the difference between global and non-global cities. Next, I limited my analysis to yearbook entries from 1993-2003, creating a ten year span of data to analyze. I do this for methodological and theoretical reasons. From a methodological standpoint, the older entries in the dataset are often riddled with missing data, and as the entries get more distant it is often impossible to find accurate contextual variables to supplement the dataset with. From a theoretical standpoint, the 1990s represent a crucial point in history where many new cities came into the world cities and many economic trends worked to increase both inequality and polarization.

My final sample includes 672 organizations in 67 global cities. The primary advantage, of this data structure is that it allows me to explore SMOs in a cross-national 31 context across a variety of social movement industries, in a multilevel model

(organizations nested within cities). In table A.1, I display the count and percentage of

SMOs in each social movement industry accounted for in my sample.

[Insert Table A.1 Here]

As Table A.1 shows, my sample includes a large number of SMOs working towards a wide variety of goals, from the modal industry of “human rights” to less common industries like animal rights, public health and democracy. Ensuring a broad diversity of social movement industries is crucial to uncovering generalizable results about contemporary SMOs, and my sample includes 29 different social movement industries.

My second concern is ensuring comparability across national borders. Table A.2 displays the of global cities in my sample and the number of organizations housed per city.

[Insert Table A.2 Here]

Of the 672 organizations, 134 are in North America, 24 are in Latin America, 416 are in

Europe, 65 in Asia and 33 in Africa1. Of special note is the modal city of Brussels,

Belgium, where almost 20% of all SMOs in my sample are located. As the headquarters for the European Union, it makes sense that many European-based SMOs would try to headquarter in the city where so much economic and political policy is debated and made. Nevertheless, my sample includes cities in every corner of the globe, making it ideal for studying organizational health in a cross-national perspective. I next turn my attention to the variables I use in my models. I display the full list of variables, sources and descriptive statistics in Table A.3, and I describe each variable in detail below. I

1 Again, while this is a cross-national dataset, it is heavily skewed towards European activism. Nevertheless, this is a constant problem when studying comparative movements, so while I acknowledge this is a weakness of the data I maintain this is the best such dataset for my research question. 32 differentiate my discussion of variables into four categories: dependent variables, organizational predictors, city-level predictors and national controls.

[Insert Table A.3 Here]

Dependent Variables

Organizational Viability: While it remains extraordinarily difficult to quantitatively measure SMO success, the simplest and most comparable metric of success is survival.

This dichotomous variable is taken directly from the TSMOD and assumes a value of 1 if the organization is still alive, and 0 if the organization has disbanded. All organizations were either alive in 1993 or came into being during this 10 year span. A value of 0 denotes that the organization disbanded at any point between 1993 and 2003.

Organizational Age: I next operationalize organizational success as age. Older organizations tend to be better staffed, better funded, better respected and altogether in a better position to accomplish their goals. They become accepted in the public debate.

This variable is taken from the TSMOD and ranges from 1-164, with a mean age of 25.5 years.

Organizational Variables

IGO Count: SMOs, especially transnational ones, often need to make alliances with more institutionalized groups and actors, both for information and resources. IGOs operate by funding other organizations, so they become a crucial source of material and informational resources. I include the TSMOD’s count of IGO contacts for each organization. This variable ranges from 0-22, with a mean count of 2.33. 33

NGO Count: NGOs also operate as an important player in the social movement sector, but they do so in a different way. Often, NGOs try to do similar things as SMOs, and thus compete for resources and funding. But, they can also be important allies when working on similar issues. Just like the IGO variable, this indicator is taken from the

TSMOD and ranges from 0-55, with a mean count of 4.18.

Professionalization: The literature from American and Western European SMOs identifies professional, formal organization as a key predictor of organizational strength and the ability to achieve stated goals. I control for this with a dummy variable derived from the TSMOD’s membership structure variable, 1 if professionalized and 0 if not.

Urban Predictors

Global Competitiveness Index: To measure how global cities influence the age and viability of the SMOs in this study, I utilize The Economist’s global competitiveness index. This variable is city specific and is measures how successful cities are in attracting foreign direct investment, capital, businesses, talent and visitors. In my sample, this variable ranges from 27.6 (Lagos, Nigeria) to 71.4 (New York City). It is important to note that this index was created in 2012, about a decade after the TSMOD stops collecting data. While most cities designated as global in 2012 were also incredibly important in the 1993-2003 timeframe. Future work will address alternative measures of competitiveness that match up with the years of analysis.

Urban Gross Domestic Product: A great deal of evidence points to economic and material resources as important predictors of social movement success. Organizations with better access to resources and better strategy in distributing these resources are 34 typically in a better position to succeed. Therefore, I account for the level of wealth within each city by using city-specific gross domestic product from the year that corresponds to the yearbook entry. I then log this variable both for ease of interpretation and to normalize a skewed distribution, which then ranges from 7.4 to 11.59.

Population Density: Urban theory speculates that global cities, by virtue of packing so many people of various social classes into a close proximity with one another should encourage both relative deprivation and networks of activism that help organizational recruitment. Because this proximity is crucial to network formation, I use density rather than just population. A key resource for SMOs is people, and urbanization creates the conditions that may help networks of activism form. I use population per square kilometer in 2003 or the closest year in which data was available, which I then log for ease of interpretation and to normalize the data.

National Controls

Inequality:2 I also measure the effect of inequality on SMO age. Walton notes that stratification and inequality tend to be more intense in global cities (1998). I use the

World Bank’s Gini coefficient measuring national inequality in this analysis. This is limitation, however, because inequality is primarily driven by rural-urban disparities, especially in developing countries. So, using a national variable does not really measure the polarization hypothesis. I choose to include it as a control, though, because the sample of organizations includes many whose goals include rural development, so national inequality may still be important.

2 For the timeframe of this study, I was unable to find city-level inequality measures. 35

Poverty:3 This variable is the percent of each nation’s population that lives in poverty, drawn from the CIA factbook. Similar to inequality, this is a country-level indicator and is not a test of the global city polarization hypothesis. I control for national poverty rates for the same reason as inequality: it is important to some of the SMOs in my sample, and can still motivate activism.

Human Rights: The organizations in my sample are headquartered in a wide variety of countries, some of which are more democratic than others. An oppressive regime may work to limit citizen activism, or ultimately encourage activism by pushing grievances to a breaking point. To model this, I include the UN’s human rights violations score, a scale that measures how many and what type of Human Rights abuses occurred within that country. As with above variables, I match up a country’s human rights score for the year that SMO was entered in the TSMOD.

Corruption Perceptions Index: Many studies of social movements identify public sector corruption as a key motivation for social movement activity. When people in a country lose faith in the public sector, they may resort to grassroots activism to pursue their goals.

So, I use Transparency International’s Corruption Perception Index to control for public sector corruption. In my sample, this variable ranges from 21-93, with a mean corruption score of 67. In this scale, the least corrupt countries score closer to 100 (e.g. Norway) and the most corrupt take on values closer to 0 (such as Egypt). Again, I use the value for the year that corresponds to the TSMODs entry.

Analytical Strategy

3 For poverty rates, the same problem applies. 36

Before presenting the full results of my analysis, I first fit two one-way ANOVA with random effects models, one for each outcome: viability and age. I do this to test the data’s suitability for multilevel modeling. These models contain no predictor variables and are used primarily to determine if there is enough variation at the second level (cities) to make multilevel modeling an improvement over standard models. The results of these tests on both outcomes are displayed in table A.4.

[Insert Table A.4 Here]

Of particular interest are the likelihood-ratio tests, which calculate the level of improvement (or lack thereof) of a two-level model over a standard, one-level model. As the results show, the likelihood ratio test for organizational viability is not significant, but the test for the second outcome of interest (organizational age) is highly significant.

These results tell me that the values for organizational viability do not vary significantly across cities, but the values of organizational age do. Therefore, multilevel modeling is only appropriate for the second outcome: age. In my analysis, I fit a series of one-level logistic regressions with observations clustered by city when regressing predictor variables on organizational viability. Then, I construct a series of multilevel, mixed effects models when regressing predictor variables at both the organizational and city level on organizational age. 4

With this in mind, models 1-3 are logistic regressions of organizational characteristics (model 1), urban traits (model 2) and a full model including national controls (model 3) on organizational viability. Models 4-6 follow the same pattern but are

4 All analyses were performed in Stata, using the “xtmelogit” command for models 1-3 and the “xtmixed” command for models 4-6. 37 instead multilevel mixed-effects models, with organizations nested within cities. Model

4 includes only organizational characteristics, model 5 adds urban variables, and model 6 is the full regression model including all previous variables and national controls. When interpreting coefficients for these logistic regressions, I refer to the odds ratio, which I present underneath each coefficient in the full table of results. These odds ratios should be interpreted as the increase or decrease in the odds of an organization being alive based on a one unit increase in the predictor.

Secondly, why use a mixed effects equation for models 4-6 rather than fixed or random effects regressions? I do this because the effect of each variable must be allowed to vary within each group (in this case, cities). Fixed effects models assume there is one true “effect” of each variable that remains constant across all units of analysis (in this case, SMOs). However, I assume that the effect of a variable may be very different in a high-income city like New York vs. a medium-income city like Buenos Aires vs. a low- income city like Lagos, so there is no one true “effect.” Random effects models will help account for the variability based on the city each SMO is housed in, and the differences in the effects of each variable relate directly to the hierarchy observed in where the data is gathered. However, a random effects model is limiting because it assumes the sample is drawn at random from the population (and random effects will estimate this variability), which is an assumption I cannot make with my data. This sample is drawn from a handbook of organizations, and thus is not fully random. Essentially, fixed effects are constant across individuals and random effects can vary. In contrast to the assumptions of fixed or random models, I assume that this hierarchical sample will include both types of effects, so I use a mixed-effects model to estimate coefficients. The fixed effects in 38 this sample are estimating the population-level coefficients, while the random effects account for individual units to react differently in response to an effect. To put this in the language of my study, a mixed model will include fixed effects to estimate the influence of each variable on an organization across cities, while also allowing each organization to respond to the variable differently. This captures the most variation.

Results

[Insert Table A.5 Here]

I first interpret the results of Table A.5, or predictors of organizational viability. In sum,

I find strong support for the idea that professionalization and IGO contacts are associated with organizational survival, but not much evidence that urban competitiveness is associated with the same outcome. IGO contacts and professionalization are strong predictors of organizational survival throughout all three models. I will interpret the full regression model for each outcome. In model 3, with all variables included, each additional IGO contact an organization makes is associated with a 1.18 times greater likelihood of survival (p<0.05). Professional organizations, in contrast to non- professional groups, are also 1.92 times more likely to be alive (p<0.01). I find no significant effect of NGO contacts at the organizational level. These effects refer to organizational characteristics independent of the city the SMO operates in. Overall, models 1-3 do a poor job of predicting organizational viability at the urban and national level. Global competitiveness is only significant (p<0.05) in model 2, and all other variables have no significant effect. Overall, it seems that when examining social movement groups in a cross-national sample, the strongest predictor of viability are the 39 characteristics of the organization itself. This is something that is largely (although not entirely) under the control of the organization.

City-level variables are much more meaningful in my next series of models, however, which predict SMO age. I find that organizational age is heavily influenced by

IGO contacts, the wealth of the headquarters city and global competitiveness. Table A.6 displays the results of models 4-6, which limits the sample to only organizations that survived the entire 10 year stretch.

[Insert Table A.6 Here]

Once again, the count of IGO contacts proves to be a strong predictor of SMO health. In model 6, the full regression model, I find that each additional IGO contact is associated with organizations that are 2.7 years older, net of other effects (p<0.001). Among viable organizations, I also find moderately significant and negative effects for the count of

NGO contacts, with each extra NGO contact being associated with SMOs about 0.27 years younger. So, while IGOs are associated with older organizations, NGOs are associated with younger organizations. While professionalization is associated with organizational survival, it is not associated with age once the sample has been reduced to only viable SMOs. Professionalization helps organization survive, which is a precondition of SMO health (age), but is not a major determinant of age itself.

When examining variables at the city and national level, I find strong evidence that resources are associated with older SMOs. The global competitiveness index is significant and positive throughout, independent of the characteristics of each SMO. In model 6, each unit increase in global competitiveness is associated with organizations about half a year older (p<0.05). Additionally, wealthier cities tend to house older 40 organizations, with each unit increase in logged urban GDP being associated with SMOs about 5 years older (p<0.05). Of course, there may be some reverse-causality at work here, as SMOs with a choice may decide to locate in wealthier cities. However, the finding clearly indicates that wealthier cities ultimately house older organizations, which is an important piece of knowledge for understanding the international social movement sector. Taken together, these findings indicate that the resources of a surrounding area, both material and human, are powerful indicators of older SMOs in a global context. At the national level, only the corruption index displays a significant coefficient, and the effect is rather weak. A one unit increase in the corruption score of a country (meaning, the less corrupt countries) is associated with organizations about 0.19 years younger.

Perhaps this means that in countries with the least corrupt public sectors, people do not feel the need to resort to grassroots activism, but more research is ultimately needed to determine this claim.

Discussion

A glaring hole in the social movements literature has been how the emergence of global cities, home to millions upon millions of people and a disproportionate share of economic activity, influences social movement organizations. The findings of this paper provide cross-national evidence for three ideas. First, I find that making contacts with IGOs strongly encourages both organizational viability and organizational age, and there is reason to believe these contacts are facilitated by global cities. Numerous scholars have either speculated or offered limited evidence that the twin emergence of both global cities and transnational activism offer crucial new resources SMOs previously lacked (Smith 41

2001; 2008; Sassen 2006; Caniglia 2002; Reimann 2002). I offer cross-national evidence that this is indeed the case. IGOs are a crucial medium between grassroots organizations and political elites through which opportunities and ideas can be diffused and state actors can possibly be transformed. Previously, SMOs were political outsiders, largely excluded from the decision making process even within industrialized .

IGOs, however, offer SMOs a semi-formal seat at the table, and give organizations access to information and money they did not previously have. My results show that organizations that can make large numbers of IGO contacts are both much more likely to survive, and likely to be older, indicating that IGO contacts serve as a buffer that protects

SMOs even as bystander interest in a group’s cause can be quite unpredictable. Global cities, then, offer a unique opportunity for SMOs to make these connections with IGOs, which are often located within global cities, too. These benefits extend beyond just , however. IGOs often work to fund and transfer resources to the SMOs they like, so having them as contacts is also likely picking up the effect of resources. Better funded organizations are more likely to survive and grow older, and IGOs are a well established source of this funding. From an urban perspective, most IGOs operate within capital or global cities, and the close proximity between grassroots actors and institutionalized actors helps produce healthier organizations. This relationship does not apply to NGOs, however, who seem to operate differently in the social movement sector. I do not find strong impacts of NGO contacts on SMO persistence, probably because NGOs compete with SMOs for the very same resources, information and legitimacy.

Secondly, and related to my first point, professionalized and formally organized

SMOs are much more likely to survive than those who lack formal organizations. While 42 some scholars have theorized that formally organized and structured groups will be more vulnerable to cooptation by elites (Piven & Cloward 1977), many other case studies of individual movements demonstrate the opposite (Freeman 1996; Jenkins 1983; McAdam

1982; Edwards & McCarthy 2003; Davis et al. 2005). My findings strongly show that in a cross-national context, professional organizations have more staying power, as they are almost twice as likely to be alive than non-professional groups. Professional organizations seem to have many advantages that non-professional groups lack, and this can helps sustain them through lean times (Taylor et al. 1989). This study confirms that not only is this true in the United States as many studies have shown, but in a cross- national context as well. Based on the multilevel nature of my data analysis, this finding can be applied across national borders, as the significance of the above coefficients holds independent of the city and national context. For activists, then, the lesson is clear: formal organization, professionalization, and IGO contacts serve to help solidify the group.

Third, I find that a major advantage of a globally competitive city to SMOs is the greater access to resources, predominantly economic and informational resources. If the findings at the organizational level are to be understood as independent of urban contextual effects, than urban variables should be understood as independent of the SMO.

When the sample is reduced to only viable organizations, I find that the count of IGO contacts is strongly associated with older organizations, providing these older groups with the resources needed to make an impact. While I cannot evaluate each group in my

43 sample on their win-loss record in the policy arena5, I can make the case that these older organizations are in a much better position to achieve their goals. The most globally competitive cities tend to also be the wealthiest cities, and as such, both global competitiveness and a city’s GDP per capita are associated with older organizations.

While neither of these variables influence organizational viability, once an organization is already stable and part of the social movement landscape, these economic resources bestow strong benefits upon an SMO independent of the organizational characteristics of the group. The most globally competitive cities seem to be the best location for activism going forward.

These findings also have implications for contemporary social movement tactics.

Cities are not going to suddenly deconcentrate – in fact, the pace of urbanization is likely to increase. More and more people are likely to find themselves within city limits. It appears to be far easier for organizations to form and survive in the most economically competitive cities. Transnational social movement organizations offer a pathway to increasing social justice throughout the world, but these groups need to be wise about their tactical choices, such as where to locate and focus their activities. These findings show that more globally competitive cities bestow benefits upon both the viability and health of SMOs. Much of the discussion about global competitiveness has been largely negative, highlighting the role of global cities in rising inequality and polarization. While my results do not speak to the differences between global and non-global cities, I show that there is more to these places that sometimes believed.

5 Which may or may not have even been their goal. 44

These findings also lead into a discussion of how SMOs can be not just viable, but successful going forward, especially in lower income areas. The urban poor tend to be very conservative in their protest tactics (Walton 1998). Many social movements in the

US and Europe are labor based, and tactics often include strikes and work stoppages, but the urban poor of the third world instead take action in the form of rent strikes or , the easiest but least effective form of collective action (Drakakis-Smith 1987).

This may be due to the culture in which they operate, the constraints placed upon the organization by the political opportunity structure, or the relative lack of resources available to the urban poor. Nevertheless, the first step to a strong protest movement is a strong organization or group of organizations. Moving forward, more research is needed into how cities influence the tactics of SMOs, but these findings help illuminate how

SMOs operate in global cities.

Conclusions

The results of this study also raise additional questions, many of which I intend to explore with future research in this same dataset. First, and most importantly, this study finds that global competitiveness in global cities has a positive impact on SMO persistence and age, but this is only an examination of cities already designated as global. This, in addition to isolating a ten year span, is the primary reason my sample dropped from over

10,000 SMOs down to 672. I plan to expand this analysis to include cities not designated as global cities by the global competitiveness index, with a dummy variable used to differentiate between global and non-global cities. This will provide more targeted insight into the role global cities play as movement spaces. 45

Next, I find that more globally competitive cities house older organizations. This is correlated with, but not identical to, national development. This may be because the most globally competitive cities, like New York, London, Singapore or Frankfurt, tend to be located in high-income countries. There has been more time for organizations to develop in these places, due to the political system and higher standard of living.

However, even if SMOs are healthiest and most effective in high-income countries, this may also be due in part to the rising inequality and polarization endemic to global cities, trends which have also had more time to take hold. So, the clustering of older, more viable organizations in these countries may partially reflect the persistence of the very factors known to motivate social movement activity in the first place.

This study also demonstrates that certain factors encourage organizational health and viability across national borders, but it does not explore how these factors vary over time. An extremely fruitful avenue of future research would be to study organizational age, viability and emergence by using event history analysis controlling for time, to see how effects change over the years. Furthermore, I acknowledge that these two outcomes do not fully speak to the numerous aspects of organizational success. This is another issue I will seek to address in future research. I plan to construct a new outcome that will better identify which organizations are the strongest and in the best position to succeed, one that accounts for organizational staff and budget as well as age. I have presented evidence that the viability and age of SMOs is influenced by several variables, but well- staffed, well-funded organizations are also in a better position to succeed for the same theoretical reasons. My future research will examine more indicators of SMO health.

46

Finally, I am not aware of any measure of city-specific inequality for global cities.

Such indicators exist for some cities, particularly American cities, but I cannot measure inequality and polarization within cities for all the locales examined in this study. This is problematic, because the primary engine of inequality at the national level is rural-urban disparities. Measuring inequality at the national level, then, is likely capturing rural- urban disparities and not really operationalizing the lived experience for people within global cities, particularly those in low or middle-income countries. The same goes for poverty rates, where the problem is similar to inequality. Poverty is more driven by rural residents, and within-city poverty is not being captured by a national metric. Moving forward, it is crucial that data collection efforts focus on providing more indicators than currently exist for global cities. Nevertheless, I still argue that measuring things like poverty and inequality at the national level matters for SMOs, because many of the organizations in this dataset are working on issues beyond their home. This paper argues that global competitiveness increases the health of SMOs and puts them in a better position to achieve their goals. These goals are varied, and often focus on rural issues like poverty and indigenous people’s rights. By, including national metrics of inequality and poverty, I account for issues that may be very important to the SMOs. That being said, more specificity is ultimately needed to strengthen the findings of this study.

Despite these limitations, this study both confirms previous research and advances our understanding of how SMOs can be successful in a cross-national context. I asked whether global cities present a better environment for social movement organizations to form, develop and untimely succeed. My results show that global cities tend to be home to older organizations, but it is the characteristics of the organization itself that determine 47 its viability. Although a great deal more work is needed to fully explore the relationships touched upon in this paper, this preliminary work goes a long way to establishing a more cross-national research agenda into the determinants of social movement formation and success.

48

CHAPTER 3: A MULTILEVEL STUDY OF SOCIAL AND POLITICAL

ACTIVISM IN THE EUROPEAN UNION

The European Union is currently facing a wide swath of economic, social and political problems. In response to these issues, various social movement organizations mobilized citizens in numerous European capital cities in protest of, among other concerns, the sovereign debt crisis and the harsh austerity measures designed to rectify it. While anti- austerity protests were the most visible of social movement activity in the EU, economic problems are by no means the only grievances Europeans hold. Citizens seem increasingly dissatisfied with their leadership in the European Parliament, sending

Euroskeptic representatives to Brussels in droves. Additionally, anti-corruption protests emerged in , and the Czech Republic. This is indeed a time of great activism in Europe, with protest movements aimed at both continental and domestic policy. While the protests in cities like Madrid, Athens, London, Dublin, Lisbon and may have capture the world’s attention, the factors responsible for spikes in social movement activity remain unclear. This is an important question, as social movements represent a haven for those who lack the ability to work for change through institutional channels and are an important pathway for advancing social justice.

In spite of the large body of work on social movement participation (Zald &

McCarthy 2002; Armstrong & Bernstein 2008; Fantasia 1998; Roscigno & Danahar

49

2001; Gamson 1990), scholars have been slow to address three gaps in the literature.

First, much of our knowledge about social movement activity is based on narrow studies of organizations within one social movement sector in one or a small handful of countries, limiting our understanding of cross-national activism (della Porta et al. 1998).

Second, individual choices clearly depend on both the characteristics of the individual and the context in which they live their lives. Socioeconomic and political conditions matter for the choices people make. Therefore, studying cross-national social and political behavior requires multi-level modeling that can separate individual from area effects and also control for the factors that make those who live around each other more similar to each other than the inhabitants of other areas. In other words, the context experienced by a Southern German and a small-town Pole are extremely different.

Controlling for these contextual differences demands multi-level modeling, which is very common in urban sociology but less common in the study of comparative social and political behavior. Third, there are many types of social movement activity, and the factors that influence one kind of activism may be very different from other forms.

Nevertheless, most studies examine only one type of outcome. In sum, social movement research needs to examine multiple kinds of activity in a cross-national framework using multilevel methods.

Additionally, there is a very under-theorized element of social movement theory in need of closer scholarly examination: urban status. Scholars have long noted that a hallmark of European cities, compared to those of other regions, is the link between economic growth and social cohesion (Ranci et al. 2014; Haussermann & Haila 2005).

This social cohesion results from multiple factors, chief among them national welfare 50 states that may discourage social movement activity (Le Gales 2002). As globalization took hold, however, inequality began to rise all over the continent, even in traditionally low-inequality countries like Sweden and Denmark (Alderson 2002; Immervoll &

Richardson 2011) and immigration mixed with polarization has created competition for jobs, resources and services that are most strongly felt in the cities of the European

Union. At the same time, the of social cohesion has been replaced with a new, individualized, market-based approach to urban life (Buck et al. 2005). The recent austerity crisis has actually caused most of the government to double down on this new ideology, which further increases the risks to Europe’s well-being (Crouch 2011). This polarization leads to new kinds of risk for Europeans, risks that are most acutely felt in cities (Taylor-Gooby 2004). Research indicates that the new risks to employment, worker’s rights, affordable housing and welfare programs are not distributed in a homogenous way, but are a primarily urban problem (Ranci 2010; Power 2000).

If cities are the places where these new social risks are most often seen, then cities can be an important incubator of social movement activity that rises in response to new problems. In this paper, I argue that cities are both the present and future of European social movement activity. Indeed, recent protests in Europe seem to take place in the downtowns of major cities, but is this purely a matter of convenience or do cities offer a more fertile ground for social movement activity? There are many reasons to believe that they do (Sassen 2006). First, there are more people to draw from (Walton 1998). Among this greater population pool, urban dwellers tend to be better educated and wealthier than rural counterparts. Grievances and relative deprivation are more acute in the city (Ranci

2010). Cities also offer more complex networks of like-minded individuals, which help 51 recruitment into SMOs (Gould 1991). Finally, cities are often the sites of severe and growing inequality, which high density makes impossible to hide (Sassen 2006).

I seek to address these shortcomings with a multilevel study of individual political and social movement activism in 20 European countries. I test the effect of urban status on social movement activity in Europe, while also accounting for grievance and resource variables on six different forms of activism: contacting a politician, working for a political party, wearing a protest badge, signing a petition, boycotting products and participating in a lawful demonstration. Using the 2010 multilevel file of the European

Social Survey’s 5th wave, I construct a series of two-level mixed effects logistic regression models to test the determinants of the above outcomes. Results indicate that the availability of resources and urban status best explain European social movement activity, while the grievances perspective offers some important insight but is best understood in the presence of urbanization and resources. I then conclude with a discussion about the implications of these findings for Europe’s social movement sector.

Theoretical Background

The European Union is currently at the peak of a protest cycle. Protest cycles tend to ebb and flow in response to a multitude of other factors (Brockett 1992; Jenkins et al. 2003)

What causes protest cycles? First, I discuss the grievances perspective, or relative deprivation. The role of dissatisfaction, grievances and relative deprivation in social movement activity has long been debated. Most would agree that some level of dissatisfaction is necessary to motivate social movement activity (Tarrow 1994). Without some perceived injustice, social movements could not exist. The question, then, is how 52 we move from general dissatisfaction to a movement society (Meyer & Tarrow 1998)?

Originally, the grievance perspective theorized that non-rational actors were excluded from mass society, so collective behavior was their only recourse for social change.

However, studies of grievances and relative deprivation have produced mixed results when studied over time (Morris & Herring 1984).

Nevertheless, those that take part in protests are obviously upset about something.

The American , for example, did not start with a dispute over a bus seat. It reflected decades (perhaps a century) of relative deprivation within the

African American community. Relative deprivation occurs when there is a mismatch between what people think they deserve and how their lives actually are. When a critical mass of people in a community begin to feel this way, they become dissatisfied with the way social institutions serve them, and this can motivate challenges to the system. Even if grievances alone cannot explain social movement cycles, this perspective can usefully augment other theories like resource mobilization and political opportunity (Amenta &

Zylan 1991; Jenkins et al. 2003). Material conditions in society do indeed have the ability to motivate collective action (Castells 1977; 1983). Offe finds that urban unemployment can drive protests (1985), and Smith demonstrates that undemocratic societies encourage challengers, especially in the modern era (2008). Additionally, the global economy can encourage challenges to the system, especially in a time when shifts in the economic policy of one nation can have major implications for another nation

(Nepstad 2002). To illustrate: anti-austerity protests in Greece targeted not just domestic leadership in Athens, but also European Union policies laid down by Brussels. Indeed, waves of protests in the early 2000s (Ayres 2002) and more recently in the European 53

Union seem to be economically rather than socially motivated, but it remains unclear exactly how large of a role material conditions and people’s perceptions of these conditions translate into social movement activity. Giugni argues that people have more sources of information, and information may mean greater dissatisfaction (1998a). To what extent is Europe’s ongoing protest wave motivated by relative deprivation and grievances? I test this by studying people’s satisfaction with some of society’s core institutions in my models: democracy, the economy, the government. I also include a measure of structural grievances, the regional unemployment rate.

Resource Mobilization

Resource mobilization theory posits that the fate of SMOs, and even entire social movements, is dependent upon their ability to marshal a high level of resources to their cause (McCarthy & Zald 1977). In other words, countries with greater resources should demonstrate higher levels of social movement activism and produce stronger social movement sectors. So what does a social movement need to be successful? The first is obviously people. SMOs with large membership rolls tend to be more successful than others (Cress & Snow 1996). More specifically, though, movements that have a paid, professional staff are usually healthier than groups that rely on volunteers (McCarthy &

Wolfson 1996). Money is a significant predictor of social movement activity for two reasons. First, better funded SMOs can hold more public events and be more active than those with fewer economic resources to draw from (Cress & Snow 1996). Second, wealthier areas will have more people with disposable income to give to SMOs

(McCarthy & Zald 1977; Stern et al. 1999). Education is another important resource, as educated individuals tend to be better informed and more socially active (Holford 1995; 54

Holst 2002). So, income at the individual and regional level stands out as an important resource.

However, SMOs are not just dependent on the number of people they can recruit and the amount of money they can raise. The type of people most likely to participate in social movements need to be in the right place to be recruited. Among personal attributes, education stands out as an attractive characteristic. The more educated an individual is, the more interest he or she tends to have in politics and

(Klandermans & Oegema 1987). This is especially true in urban areas (Anyon 2014).

Education is not evenly distributed, however, and some areas have a higher cluster of educated residents than others. Since educated people are more involved in politics, having them nearby is a powerful resource for social movements.

Beyond the individual resources, however, certain other factors also qualify as structural resources. A powerful indicator of a strong social movement sector is generalized trust, or the level of faith people have in those around them (McAdam,

Tarrow & Tilly 2003). The extent to which community life has declined or not declined in various countries is a hotly debated topic. What is clear, however, is that generalized trust in others is often a prerequisite for collective behavior (Moore 2006). Social trust helps build collective identities that form the basis of many . That being said, some areas have much higher generalized trust than others. This is particularly evident in Europe, where the people of former communist states in the East display much lower trust in those around them and social institutions (Wallace &

Latcheva 2006). Northern Europeans are much more trusting. Therefore, generalized trust in others can be conceptualized as a structural resource that will encourage social 55 movement activity. All this evidence suggests that personal & regional wealth, trust and education are important resources for SMOs.

Urbanity

Urbanization is one of the most important trends in global society, and yet, the role of cities in social movement activism is surprisingly understudied. This is not to say urban social movements (particularly protests) are ignored, but it is usually conceptualized as a just regular, garden-variety social movement activity that just happens to take place in cities. Little attention paid to what makes urban social movement activity unique. In contrast, I argue that cities are simultaneously places where the traditional motivators of activism (grievances, resources, political opportunities) are most prevalent, but also movement spaces in their own right that nurture social movement activity. This style of thinking traces back to David Harvey, who theorizes that cities are a reflection of global economic shifts (1973; 1982). These shifts produce conflict between the capitalist class and the working class, and unresolved conflict mixed with power differentials between populations in close proximity with one another often leads people to social movements (Castells 1977; Logan & Molotch 1987).

Cities are unique places that create many of the preconditions necessary for social movement activity.

From a structural perspective, how do cities encourage social movement activity?

Almost all social movement activity has to take place somewhere, and cities work to create movement spaces in which this activity can play out (Nicholls 2009). Cities, by definition, hold more people than rural areas (Walton 1998). Beyond just population size, however, population density is also a significant factor in creating a movement 56 space. Densely packed urban environments facilitate both recruitment (Gould 1991) and diffusion of ideas throughout a population (Hedstrom 1998). The networks of activism are most evident in cities.

Beyond just the structure of cities, I also argue that the factors known to encourage strong social movements are more prominently found in cities. The threats to people’s livelihoods and standard of livings are most prominently felt in urban areas, especially in affluent, industrialized democracies (Ranci et al. 2014). I therefore expect grievances to be higher in cities as compared to less urban areas. I already argued in a previous section that education and personal income are important resources to SMOs.

Urban residents tend to be both wealthier and better educated than their rural counterparts. The type of person likely to inhabit a city, however, makes a difference to local SMOs. From a resource mobilization perspective, educated people with disposable income are very important as a resource base to an SMO. Essentially, I argue that cities provide preferential access to the resources important to SMOs. These resources include an educated, politically active population, disposable income and wealth. While we can think of these as resource variables, I argue that urbanization makes the effect of these factors much more apparent and more powerful.

Other Issues

For the purposes of this paper, I focus most closely on grievances, resources and urbanization in the European Union. There are a few other issues worth highlighting and modeling, however. I must especially discuss the role of political opportunity structures.

In contrast to resource mobilization theory, political opportunity structure identifies certain moments in which the political system becomes more open to challengers (Meyer 57

& Minkoff 2004). Movements that mobilize during these times should achieve better outcomes than they would in unfavorable political conditions (Amenta & Zylan 1991;

Amenta et al. 1992; Brockett 1991). Of course, the components of political opportunity are highly variable across international and even local contexts, so this framework is best understood as a set of variables rather than a concrete framework (Meyer & Minkoff

2004). Political opportunity variables shown to be important to social movements include alliances with powerful elites (Jenkins & Perrow 1977), divisions among elites

(Jenkins et al. 2003), effective framing strategies utilized by activists (Gamson & Meyer

1996); level of state repression (Brockett 1991), the responsiveness and openness of the state (Kitchelt 1986; Krisei et al. 1992) and the presence or absence of a (Meyer & Staggenborg 1996).

However, for the purposes of this study, Europe does not display the variety of political systems and nation states observed when studying the world system. In a cross- national study, one would ideally want to study political opportunity with a range of political systems to draw from. The European Union is comprised of industrialized democracies. Still, though, there is a deep divide between the older democracies of

Western Europe and the new democracies of the East. The collapse of the Soviet Union and her satellite states brought many new democracies into the European system, and while all members of the EU qualify as industrial democracies some are more democratic than others. Therefore, it makes sense to classify countries as either old or new6 democracies when studying Europe, and I expect to observe differences between the two.

It seems intuitive that the most democratic states (or the oldest democracies) should

6 That is, countries which became democratic in the 1990s. 58 encourage social movement activity by lifting restrictions on political expression

(Goldstone 2004). However, there are two counterpoints to this idea. First, repressive regimes are never totally successful at suppressing social movement activity, and social movements exist in all societies, even less democratic ones (Pickvance 1999). The effect of democracy has long been theorized to be curvilinear – that is, in the least democratic societies, protest and other forms of activism are almost entirely absent, and it rises as the state becomes less repressive. At some point, however, the relationship reverses and the most democratic societies do not motivate protest either. Therefore, the alternate argument must be considered: that democratization hinders social movement activity. In less democratic societies, people cannot work for change through institutional channels, making grassroots activism their only avenue for social change. When societies democratize, those in preexisting social movements are brought into the political system.7

In other words, democratization removes one of the greatest reasons to join a social movement. Once absorbed into a more open, less repressive, less corrupt polity, they are vulnerable to cooption and struggle to make meaningful change (Katzenstein 1998).

Considering the East-West divide in European political history and the competing evidence, I seek to test whether the social movement sector is more or less prevalent in the new or old democracies.

Data and Methodology

7 It should be noted that while many scholars see social movements as inherently extra-institutional, others define incorporation into institutional politics as a desirable outcome, even a victory. See Gamson 1975, 1994 for a further discussion on this. 59

In this study, I use the 5th wave of the European Social Survey (ESS), which corresponds to the year 2010. The ESS is a cross-national survey program in the European Union and candidate countries carried out every other year since 2001. In each wave of the ESS, fieldworkers perform face-to-face interviews within a predetermined sampling frame to collect data on attitudes, beliefs, and patterns of behavior across the diverse populations of Europe. For the purposes of my research, the fifth wave has a unique advantage over other European datasets or other waves of the ESS: its multilevel structure. In addition to individual level variables, the multilevel ESS file also provides regional data for ESS

Nomenclature of Territorial Units (NUTS) at levels 1, 2 and 3.

While the ESS contains 26 countries, I dropped Russia & Israel outright. Russia and Israel are too politically and geographically separate from the rest of the EU to merit inclusion. Additionally, I also was forced to drop Cyprus, Bulgaria, Ukraine and Croatia, as these countries did not have regional level variables. So, my final sample includes 20 countries, the full list of which I include in Table B.1, along with the number of respondents and geographical units per country.

[Insert Table B.1 Here]

To construct my regression models, I make use of variables from the ESS and supplement these with a few external sources. Table B.2 displays the descriptive statistics and sources of all variables used in my models, which I discuss in detail below.

[Insert Table B.2 Here]

Dependent Variables

60

The dependent variables for models 1-6 are, in order, contact a politician, work for a political party, wore a protest badge, sign a petition, boycott a product and participate in a lawful demonstration. Each of the above variables are dichotomous outcomes, where 1

= participated in the activity in the last 12 months, and 0 = did not. Most of the above outcomes represent political participation and behavior, and is not concrete evidence of social movement participation (at least not formal participation). However, political interest and activity is an important precursor of social movement participation, especially in times of crisis such as the current European protest wave.

Independent Variables

Based on my review of the literature, I separate my independent variables into four general categories: grievances and relative deprivation, resources, urbanization and controls.

Grievances & Relative Deprivation Variables

Satisfaction (Economy). To address the relative deprivation perspective, I include a measure of satisfaction with the economy within the country in question. A poor economy can lead to several negative social outcomes and dissatisfaction with the system, which may drive individuals to activism. This is especially important when studying Europe post 2008, when many countries entered recessions. This variable is an

11 point scale, with 0 being total dissatisfaction with the economy and 10 representing complete satisfaction.

61

Satisfaction (government). Next, I supplement my economic satisfaction measure with a variable gauging contentment with the government. If people are happy with their government, they are less likely to participate in any higher-cost form of social movement activity. This is also an 11 point scale, measured 0-10 with 10 being most satisfied.

Satisfaction (democracy). However, if dissatisfaction with government might positively influence social movement action, then dissatisfaction with democracy might negatively influence activity. If people feel that citizen activity is meaningless, then dissatisfaction with how democracy works may serve to negatively influence social movement activity.

Like the other satisfaction variables, democratic satisfaction is measured from 0-10 with

10 being most satisfied.

Unemployment. Similar to satisfaction with the economy at the individual level, high unemployment is structural grievance, and is thus measured at the regional level.

Pervasive unemployment can have an extremely adverse effect upon a society. Even someone currently working will notice high levels of unemployment, and may be pushed towards social movement activity in response to worsening economic conditions.8

Essentially, they may fear that they are next. This is a percentage measured at the ESS regional level, with a mean of 8.72%.

Resources

8 To be clear, economic social movements are NOT always left-leaning. In recent years, far right movements have become quite prominent, especially in France, Spain, Greece and the Netherlands. So, high unemployment may be a recruitment tool for both left and right leaning groups. 62

Trust in Others. I include a measure of generalized trust to measure collective efficacy within each country. Social movements need people working together to be successful, so general trust is a structural resource. This variable can be conceptualized as both a resource indicator and an indicator of relative deprivation. This variable runs on an 11 point scale, with 0 being the lowest level of trust and 10 being the highest.

Income. Measuring income across 20 different countries, only some of whom use a common currency, is especially challenging. Rather than include a continuous measure of income, I construct two dummy variables that classify respondents as low, middle or high income. The ESS provides respondents with 10 categories and asks them to categorize themselves. I classify deciles 1-3 as low income, deciles 4-7 as middle income and deciles 8-10 as high income. This is an admittedly simplistic and imperfect measure of how wealthy people are, but when analyzing data from 20 countries, some depth must be sacrificed. I then omit low income as a reference category, including the other two in the models.

GDP. I expect wealthier areas to be more socially and politically active and offer more resources to social movements. The ESS measures GDP per capita at the regional level, and I transform this variable so that the coefficients should be interpreted as an increase of 1,000 dollars per capita is correlated with a change of X.9

Education. More educated people tend to be more socially and politically informed and active, so educated people are a resource for SMOs. To account for this, I measure years

9 I originally used the natural logarithm of regional GDP, but this variable was so highly collinear with numerous other independent variables that I could not include it in the final models. 63 of full-time education completed. This variable ranges from 0-50, with a mean of 12.26 years or, roughly a high-school diploma.

Urbanization Variables

Urban. To measure the effect of urban status, I include a dichotomous measure I constructed from the ESS place of residence variable. Respondents described their communities from five options, and I define urban as a “big city” and “suburbs or outskirts of a big city.” I define “not urban” as those residing in a “town or small city,”

“country village,” or “farm or home in countryside.”

Population Density. To further explore urbanity and its effect on social movement activity, I also include the region’s population density. Urbanization is a continuum.

This variable is included in the ESS and logged for ease of interpretation. This transformed variable then ranges from 0.69 to 8.79.

Control Variables

Sex. Sex is an important control variable in any kind of behavioral analysis. So, I include a dummy variable measuring sex, 1 if male, 0 if female.

Age. Social movement participation changes over the life course, with some forms of activity more likely amount the youth and others more likely among older residents. As individuals age, their interest in politics increases and social movement activity becomes more likely, but eventually this trend is likely to reverse. I include a simple continuous measure of age in my models, with a minimum of 16 and a maximum of 101.

64

Immigrant Status. I also include a control for immigrant status, 1 if foreign born, 0 if native. Immigrants are a very controversial issue in the contemporary European Union, and immigrants experience social life far differently than natives. I expect that immigrants should not demonstrate levels of social movement activity nearly as high as native born citizens.

New Democracy. Political opportunity theory, at least in a cross-national perspective, is usually applied to weak or authoritarian states. While the various countries of Europe are not all equally democratic, Europe is largely without a true autocracy.10 However, there remains a severe East-West divide in Europe that closely correlates with the old Soviet

Bloc and the democratic states of Western Europe. I control for new democracies but including a dummy variable, 1 if country became democratic in the 1990s and 0 if the country was democratic prior to the 1990s.

Analytical Strategy

Similar to the first chapter of this dissertation, I use multi-level modeling techniques.

Specifically, I fit a series of two-level logistic regressions, with individuals nested within regions of the European Union. Before proceeding, however, I must first establish the statistical basis for using a multilevel structure before I actually start analyzing the data.

As in chapter two, I fit a series of one-way ANOVA with random effects models which include only outcome variables and test the degree to which each outcome varies across the level-two group, which in this case is ESS regions. The results of these tests are displayed in table B.3.

10 Depending on how you feel about Belarus and Russia, neither of which are in this sample anyway. 65

[Insert Table B.3 Here]

The coefficients for all six outcomes are significant, but what I really care about is the result of the likelihood ratio test for each variable. Table B.3 shows that the result of this test is highly significant at the 0.001 alpha, which means that the values of each variable vary significantly across the countries in this sample. Since this variation is not random, the likelihood-ratio test shows that a multi-level regression model would be in improvement over a standard, one-level model. Therefore, I make use of multilevel modeling to analyze these variables.

Now that I have demonstrated the superiority of a hierarchical model over standard models, I then construct a series of two-level, mixed effects logistic regressions to measure the effect of these individual and regional variables upon social movement activity. In chapter 2, I addressed the relative strengths of a mixed-effects model over both fixed effects and random effects. The same reasoning for using a mixed-effects strategy in chapter 2 also applies here. Essentially, a mixed effects model allows for the most variation to be captured, and the error terms at the second level are not random. In these models, individuals are nested within regions as defined by the ESS (NUTS 2).

Models 1-6 measure the effect of all individual and regional variables on the propensity of respondents to contact a politician, work for a political party, wear a protest badge, sign a petition, boycott a product or participate in a lawful demonstration, respectfully.

All of these outcome variables refer to activity that took place within the previous 12 months from the time of interview. When interpreting effects, I also present the odds ratio associated with each unstandardized coefficient. The odds ratio should be understood as the change in the likelihood of an individual participating in that form of 66 activity based on a change of one unit in the predictor variable. An odds ratio of 1.00 implies no change at all, while any values below or above 1 indicate change.

Results

My first task is establishing the extent of social movement activity in the countries of my sample. Table B.4 displays the percent of people in each country who report participating in each of the six outcomes of interest. The last category is just a simple aggregation of all social movement activity, which I created by assigning a value of 1 to anyone who participated in any of the above six outcomes within the past year. While I do not model this outcome, I use it here for descriptive purposes.

[Insert Table B.4 Here]

As the table demonstrates, social movement activity varies immensely across the 20 countries of my sample. When examining all forms of activism, the ESS reports a low of about 12% in Portugal to a high of about 63% in Norway. What explains these broad variations? Table B-5 displays the results for outcomes 1-3, while Table B.6 displays the results for outcomes 4-6.

[Insert Table B.5 Here]

[Insert Table B.6 Here]

I will discuss my results in the order of each general theoretical perspective I seek to test. First, I focus on the idea that grievances, both individual and structural, can motivate social & political activity. For my other three grievances variables, I find mixed but generally not strong results. A one unit increase in economic satisfaction correlates with a 1.04 times greater likelihood of working for a political party, but the same unit 67 increase is associated with a 0.98 and 0.96 times lower likelihood of boycotting a product and participating in a demonstration, respectively. These are extremely small effects and may be due to the large N of my sample. For satisfaction in the government, I find significant and negative effects only on outcomes 4-6. Each one unit increase in this predictor correlates with a likelihood that is 0.95, 0.93 and 0.93 lower for signing a petition, boycotting a product or participation in a demonstration, respectfully. While these are relatively small effects, the consistently negative direction indicates the happier people are with their government, the less likely they are to participate in these forms of political behavior. Satisfaction in democracy is significant for only two outcome: protest badges and boycotts. A one unit increase in the predictor is associated with a likelihood of boycotts that is 0.97 as high and a likelihood of wearing a protest badge that is 1.03 times higher. Again, these are significant but not strong results. Regional unemployment, a structural grievance, is only a significant predictor of two outcomes: each one unit increase in the unemployment rate is correlated with a 1.04 times greater likelihood of working for a party and a 1.06 times greater likelihood of participating in a protest. When these four variables are considered as a group, I cannot make a strong case that grievances and relative deprivation are primary motivators of social movement activity.

I next discuss my resource variables, some of which are measured at the individual level and some at the regional level. I find consistently positive and significant effect of generalized trust on social movement activity. While trust is not correlated with contacting a politician, a one unit increase in generalized trust is correlated with a 1.05 times greater likelihood of working for a party, 1.08 times greater 68 likelihood of wearing a protest badge, a 1.06 times greater likelihood of signing a petition, 1.05 times greater likelihood of boycotting a product, and a 1.11 times greater likelihood of participating in a protest event. It seems that trust in those around you is a necessary prerequisite for the development of a strong social movement society. Next, I find that income is a consistently significant and positive predictor of social movement activism. When compared to lower income individuals, the middle income respondents are 1.13 times more likely to contact a politician, 1.34 times more likely to sign a petition and 1.14 times more likely to boycott a product. When compared to the same reference group, the upper income respondents is 1.25 times more likely to contact a politician,

1.18 times more likely to work for a party, 1.22 times more likely to sign a petition and

1.22 times more likely to boycott a product. I find no significant effect for the other outcomes. I also find that education is significantly and positively correlated with all six outcomes, indicating that greater education is one of the strongest predictors of social movement activism. Finally, I consider regional GDP, a variable that I normalized for ease of interpretation to correspond to units of 1,000 euros. The effect of this variable is strongly positive and significant across all six models, although the magnitude of the effects are somewhat weak. When you consider large gaps between rich and poor areas of Europe, however, and the effects can really add up. Each increase of 1,000 euros increases the likelihood of contacting a politician by 1.01 times, wearing a protest badge by 1.03 times, signing a petition by 1.04 times, and boycotting products by 1.03 times.

Wealthier areas produce stronger social movement societies, both at the individual and regional level.

69

I previously argued, in this paper and in the other chapters of this dissertation, that cities offer a better environment for social movements to take root and blossom. I test this proposition with two variables: urban status at the individual level and population density at the regional level. Testing both in the same models helps isolate individual from regional effects. I hypothesized that urban status will significantly increase social movement activity, but the results are more complex than that. Urban residents are 0.79 times as likely to contact a politician as their rural counterparts, and I find no effect on working for a party. However, urban residents are 1.15, 1.15, 1.22 and 1.37 times more likely to wear a protest badge, sign a petition, boycott a product and participate in a demonstration. While urbanity does not predict all forms of social movement activity, the higher-profile (and potentially higher costs) activities, like boycotts and protests, are much more likely among city dwellers. At the regional level, I find that population density is correlated with a lower likelihood of contacting a politician (0.94), wearing a badge (0.87) and boycotting a product, but a higher likelihood of signing a petition (1.06) and taking part in a protest event (1.09). Taken together, I conclude that social movement activism is indeed more likely in cities and denser areas.

Finally, I discuss my control variables. I find that men are more likely than women to contact a politician (1.42), work for a party (1.85) and participate in a protest

(1.30) but less likely to wear a protest badge (0.90), sign a petition (0.88) or boycott a product (0.93). Certain forms of activism skew female, and others skew male, indicating that there is clearly a role for everyone in a social movement campaign. Immigrants, reflecting their subordinate role in European society, are significantly less likely to participate in all six forms of activism than their native-born counterparts. The 70 magnitude of the effect of this variable is massive. Finally, age is positively correlated with contacting a politician or working for a party, but negatively correlated with wearing a protest badge, signing a petition and taking part in a protest. The milder forms of activism are usually performed by older individuals, while the more visible activities that require people to get out in the streets are still a young person’s domain. People age into and out of different forms of activism as they progress through the life-course. Finally, I controlled for the role of new democracies in European society. There is no effect on the first four outcomes, but citizens are far less likely to boycott products (0.52) and protest

(0.37) in the newer democracies of primarily Eastern Europe. These countries have a very different protest culture than the established democracies of Western Europe, but it is also interesting that this variable has no effect on the other types of activity. The new democracies of Europe clearly have some kind of social movement sector, but very likely lack the resources and trust necessary to motivate large scale protest and boycott behavior.

Discussion

In sum, I find mixed support for the various ideas about what motivates social movement behavior. Generalized trust increases the likelihood of working for a party, wearing a badge, signing a petition, boycotting a product and protesting. This demonstrates that collective action does indeed require a preexisting base of trust to encourage people to work together. This may be a reason why recent protests have been centered in Western

European cities rather than Eastern Europe, where collective efficacy is generally much

71 lower.11 More specifically, the countries in the sample with the highest levels of generalized trust are Northern European: Norway, Denmark, Sweden and Germany.

Conversely, former members of the Communist Bloc like Slovakia, Poland and the Czech

Republic all have levels of trust significantly below the mean. A strong social movement sector seems to be the consequence of a strong society where people trust one another, rather than the cause of it. This finding is slightly sobering, as it calls into question the ability of collective action to change society in exactly the conditions that need changing the most. Without the requisite levels of trust, social movement organizations will struggle to motivate citizens to be actively involved.

Furthermore, I find that satisfaction with the economy is positively associated with working for a political party and negatively associated with boycotts and protests.

Satisfaction with democracy positively correlates with wearing a badge and negatively correlates with boycotting a product. The magnitude of all these coefficients is generally weak, however. Satisfaction with government performs a bit better, as higher levels of satisfaction decrease the likelihood of petitions, boycotts and demonstrations. The performance of these three satisfaction variables demonstrate that while grievances certainly matter for social movement activity, satisfaction with material conditions is best understood in a supporting role to other theories rather than a primary explanation. It certainly makes intuitive sense that satisfaction with the government would help defuse potentially explosive issues that may lead to protests elsewhere. Grievances are not

11 Ukraine deserves a massive asterisk here, where grassroots protest (among other factors) was a crucial factor in overthrowing the country’s government in 2013-2014. 72 totally irrelevant, but they do not appear to be a strong motivator of social movement activity in modern Europe.

Resource mobilization suggests that areas with the greatest amount of resources should produce the healthiest social movement sectors, an idea for which I find strong support. While higher incomes are generally associated with higher levels of social movement activity, the trends differ by outcome. I find that higher incomes increase the likelihood of contacting a politician, working for a party, signing a petition and boycotting a product, but there is no effect on demonstrations. Perhaps this is because there is a clear pathway of escalating severity for social movement activity, and while higher incomes encourage milder forms of activism like signing a petition and boycotting a product, by the time protests break out, then income matters much less. What about resources at the regional level? I find that regional GDP is indeed associated with a higher likelihood of social movement activism, positively influencing working for a party, wearing a badge, signing a petition, boycotting a product and protesting. Richer areas produce more political and socially active citizens. Additionally, I find a strong effect of education, as years of school is positively associated with all six outcomes.

Educated people are an important resource for social movement groups, as they tend to be better informed and earn higher incomes. Regions with this crucial resource of educated people indeed produce higher levels of social movement activism. Once again, these findings lend support the idea that social movement activity is the result of an already strong, healthy society, one that is wealthier.

Finally, I did not test many political opportunity variables because the EU is largely filled with industrialized democracies. Testing political opportunity cross- 73 nationally requires a larger diversity of political systems that what is currently found in

Europe. Of course, Kitchelt’s seminal research shows that even within Western democracies, political opportunity factors like the openness and receptiveness of the state influence social movement activity (1986), and police tactics can also influence protest behavior in democratic societies (della Porta & Tarrow 2012). Political opportunity is not meaningless in Europe, but is certainly less meaningful.12 However, I do find huge disparity between the new and old democracies of Europe in how likely people are to boycott products and protest, with Eastern Europeans being much less likely to participate in these activities. I believe the answer can be found in the other variables known to motivate protests, like collective trust and resources. All of this is much less apparent in Eastern Europe, so as a result, their protest capacity is much more limited.

Finally, my findings advance previous research by examining the effect of urban status on social movement activity. Interestingly, the effect of urban status on social movement activity can be positive or negative depending on the outcome in question.

Urban residents are more likely to wear protest badges, sign a petition, boycott products and participate in demonstrations. In the EU, social movement activity has indeed centralized in cities. This has numerous implications for both SMOs and social movement theory. From the perspective of social movement groups, their efforts and precious resources seem best focused on cities, and this is not just because of their larger population pools. Cities offer much greater resources: more educated people, more wealth and disposable income, and presumably a more informed populace. Essentially,

12 In fact, a previous version of this study included several other national-level political opportunity variables. They never produced significant results. 74 the building blocks of social movement organizations: people, money and time, can all be found in greater abundance within cities. Some SMOs are local in nature and are tied to their location, but for those groups with geographical freedom, cities offer the better starting point. Furthermore, while resources play an important part in determining an area’s social movement capacity, cities are an important part of this equation. Cities modify the relationship between other social variables and their propensity to engage in social movement activity. Thus, even though social movement activity seems to be the product of strong, healthy societies, cities produce a better social movement sector than then many countries should theoretically have. For SMOs struggling to create change or even just stay active, the advice of this paper applies double. Cities seem to offer the best place for social movement work.

Conclusions

In conclusion, I find that when comparing the explanatory power a few different theoretical perspectives, resources and urban status seem to explain social movement activity in the contemporary European Union best. Of course, this study is limited in a few important ways. First and foremost, comparative studies of a broad geographical area are always limited to the areas of study. My sample of twenty countries includes nations from Western, Southern and Eastern Europe, but some theoretically interesting countries are not included. Ukraine, specifically, recently witnessed a large protest movement that toppled the government and preceded armed conflict in the eastern part of the country. It would be extraordinarily interesting to see how these findings apply to a country like Ukraine. This also speaks to a broader issue – specifically, that East Europe 75 is underrepresented in my sample. While I analyze data from places like Czech Republic,

Poland and Slovenia, this region has a much different protest and social movement culture than Western Europe. Further research should examine this region in more detail.

This study advances previous research in two crucial ways, one theoretical and one methodological. While students of social movements do study urban protests, it is rare to see urban status used as a variable in this research. Conversely, urban sociology rarely uses what we know about cities to contribute to the field of social movements. My research demonstrates that cities act as enablers for social movement activity and house citizens more likely to be socially and politically active. Therefore, urban status is a variable which can no longer be ignored in this kind of research. With the world rapidly urbanizing, cities will increasingly be the lived experience of the world’s inhabitants, and we need to better understand how this changes what we know about social movement activity. Secondly, from a methodological standpoint, multilevel modeling should be used more frequently in social movement research, especially in cross-national comparative work. Individual and regional characteristics are both important parts of the puzzle here, and this necessitates multi-level modeling.

However, this research also demonstrates that social movement activity is increasingly a consequence of an already stable society, not something that emerges as a strong causal factor in strengthening societies. This raises serious question about the future of citizen activism, not just in Europe but all over the world. However, it also suggests an interesting relationship between political opportunity and resource mobilization. Scholars always acknowledge that both perspectives have something to add to our understanding of the social movement society. I find that high levels of 76 activism arise in societies that can mobilize the greatest amount of resources and that in

Europe, political opportunity is less important. But for developing societies like those found in Latin America, the increasing amount of wealth and resources may combine to create a type of opportunity not seen in these countries before. This is a testable hypothesis, and my future research will examine this question outside of Europe. As it stands, however, these findings add a great deal to the current understanding of social movement activity in contemporary Europe.

77

CHAPTER 4: THE INTERNATIONAL MEDIA’S COVERAGE OF THE

EUROPEAN ANTI-AUSTERITY MOVEMENT

Can grassroots social movement actors achieve media standing in a transnational media outlet based in another country? If so, does this help their efforts? Students of social movements have long been interested in how the media covers social movement conflicts. Some scholars have highlighted the ability of the media to harm grassroots activism with negative coverage (Gans 1979; Gitlin 1980), but others instead point to the many benefits media coverage bestows upon SMOs (Tarrow 1994; Zald & McCarthy

2002; Andrews & Caren 2010). The media allows for social movement groups to advertise their goals, claims and grievances, facilitates the recruitment of new members and the transformation of the bystander public into constituents and helps draw public attention to the actions of SMOs. However, if the media is an important resource to

SMOs, then it can also be used in the same way by institutionalized elites, oftentimes much more effectively and efficiently (Schoene forthcoming). Despite this knowledge, existing research is overwhelming focused on local media outlets, but globalization opens new doors for SMOs. International media sources based outside the country of conflict are now more accessible than ever before, and their attention may be a powerful but largely untapped resource for grassroots SMOs.

78

This chapter builds upon the previous chapter’s work, which explored the determinants of individual level social and political activity in the recent European protest wave. The European Union, by its very nature, is a primary example of globalization. It links many different nation states into a complicated system of political, economic and social alliances. It binds the fates of many European nations to continental policy determined in Brussels. The recent economic crisis within the Eurozone, and subsequent austerity policies designed to reduce sovereign debt, has shaken the faith of many of Europe’s citizens in the common currency and even in the EU itself. While countries like France and Germany have largely escaped the crisis unscathed, many of their neighbors, like Greece, Spain, Italy, Portugal, Ireland and the United Kingdom implemented painful austerity measures that crippled their economies and stunted their growth. Austerity, an economic philosophy that cuts government spending while simultaneously raising taxes to generate the surplus cash necessary to service debt obligations, tends to be deeply unpopular among average citizens. Europe was no exception, and a series of loosely affiliated anti-austerity movements have emerged all across the EU in response. The media’s role in mediating this kind of conflict has been well studied in a variety of settings, but Europe’s anti-austerity movement differs in that it is expressly transnational in nature, and thus attracted the attention of international media outlets. When the media from one country covers a social movement conflict happening in another country, does this help the movement in any tangible way?

In this chapter, I apply Gamson’s concept of media standing to the European anti- austerity movement (1998; 2001). Media standing refers to the people and groups deemed worthy of participation in the public discourse surrounding a social conflict. 79

Everyone has the right to speak, but not everyone’s voice carries the same weight. The media, then, acts as a gatekeeper and bestows legitimacy upon the groups they cover and give voice to. Of course, the media also grants the same legitimacy to local and national elites, making the media a battlefield for standing between grassroots actors and elites.

This battle is not fought on an equal playing field, however, and media standing is typically much easier for elites to achieve (Gamson 2001). While it is well established that the media can be a valuable resource for social movement groups despite the other disadvantages they face, a crucial question remains unexplored. Can local SMOs achieve media standing in international media sources, and does this help their local efforts?

This study examines the New York Times’ coverage of Europe’s anti-austerity movement. I use the sovereign debt crisis to answer two questions. First, how successful were both established elites and grassroots actors in achieving media standing compared to bureaucratic, institutionalized elites? Secondly, was the New York Times coverage skewed in either the pro-austerity or anti-austerity direction? If so, how severe was this skew? To answer these questions, I perform a content analysis of 4,496 quotes taken from a sample of 1,197 New York Times articles from January 2010 through March 2013.

Following the example of Tuchmann (1972), Roth & Vander Haar (2006) and Schoene

(forthcoming), I use quotes as my unit of analysis and study how successful each side of this debate was at getting their rhetoric into the pages of one of the world’s most respected . The results show that institutionalized, bureaucratic elites were much better at achieving media standing compared to their grassroots counterparts, but that the aggregate coverage of the crisis was decidedly anti-austerity, something which calls into question the power of the international media in social movement conflict. I 80 then conclude with the implications for both social movement theory and organizational tactics.

Case Background

In 1992, long before the European Union ever created the common currency, EU member states signed the Maastricht Treaty, in which they agreed to limit deficit spending. Over the next two decades, however, it became clear that many EU states, such as Greece, had been hiding their true level of indebtedness for years.13 Once this mask came off, long term confidence among investors dropped, sending interest rates skyrocketing and the ratings of governmental bonds plummeting. While the details of each specific country’s crisis were unique and varied, the common theme that unified each struggling nation was debt. Banks were overextended and European politicians began to fear a catastrophic run on assets. In many countries, what originated as private bank exposure became sovereign debt through a series of bank bailouts orchestrated by the “Troika,” or the European

Central Bank (ECB), European Commission (EC) and the International Monetary Fund

(IMF). To complicate matters, the Eurozone has long shared a common currency without sharing fiscal policy. In other words, each nation is no longer in control of their currency, and cannot resort to a national stimulus akin to what the United States did during the recent Great Recession. They were at the mercy of creditors and the economically stable EU states, specifically Germany. To address the crisis, European

13 In reality, it is likely that the more stable nations of the Eurozone knew all along about high levels of deficit spending in Greece, Spain, Portugal and Italy. However, it would be silly to have a common currency if only Germany, France and Belgium could qualify, so these concerns were largely ignored. 81 policymakers implemented a series of austerity measures designed to reduce debts and stabilize the euro.

The effects of austerity were immediate and painful. Austerity is often felt most acutely by the average citizen who has been paying into a pension system or of drawing government wages. Suddenly, pensions are cut, wages are slashed, workers retire without replacements and taxes are raised on basic necessities like groceries. Meanwhile, wealthy citizens and larger businesses do not suffer nearly as much, with many able to ride out the tough times with their savings or park their money in tax havens and dodge the increasing tax rates.14 In other words, austerity tends to be regressive in nature. This reality is not lost on the average citizen, and austerity measures brought Europeans into the streets in droves. Protests (and sometimes riots) rocked Athens, Lisbon, London and

Madrid, as Europeans made their dissatisfaction with austerity known. While austerity measures were loosened somewhat over time, countries like Greece, the UK and Spain had to double down on their conservative economic course and Greece cycled through a revolving door of prime ministers, each one feeling the wrath of the voters more severely than the last.

From a social movement perspective, it is clear why the average citizen would oppose austerity measures. They lose their pensions, have their wages frozen or cut, lose important social services and welfare programs15 and struggle to find full time employment. Who, then, is benefiting from austerity policies? The simple answer is

14 Ironically, one of the most popular tax havens in Europe, the small island nation of Cyprus, needed a bailout of their banks after years of collecting foreign deposits and taxing them at artificially low rates. 15 It bears mentioning that these social programs are always more expansive in Europe than in the United States. 82 business and political elites, but even this is a reductionist take, as the owners of smaller businesses fear the impending recession and worry about declining demand. From an economic standpoint, the many social support programs that have been slashed across

Europe are not eliminated, they are instead privatized. This generates a profitable investment for the corporate class, who buy up the program for pennies on the dollar.

These savings are also used to pay for income tax cuts, which also primarily benefit wealthier residents. So, when the level of debt (which originated in the financial sector) reached a crisis point in Europe, the ECB and EC arrived with bailouts that carried austerity measures designed to encourage “financial discipline.” However, this bailout money is used to pay off the creditors owed by each country’s central bank and rarely stays to help the nation recover. This fact is not lost on citizens, who came to think of austerity as benefiting only the German and French bankers who encouraged deficit spending in the first place.

Theoretical Background

When viewed through the lens of social movement theory, the anti-austerity movement in

Europe can be seen as a clear example of conflict between grassroots citizens and international elites. There is a clear disparity in the institutional power held by each side of the fight in Europe. While citizens have the power to vote for parties that reflect their policy wishes, there is no guarantee the government will listen, especially in a political system in the EU where citizens are influenced by both European and national policy.

Several Greek governments fought against austerity measures, but none were ultimately able to ameliorate the situation. How are movements to achieve their goals in 83 circumstances such as these, where a clear power and resource differential exists? Social movement theory has long identified the media as a key resource for grassroots movements, sometimes calling it the single greatest ally SMOs can recruit (Lipsky 1968).

However, the relationship among movements, elites and the media is complex, highly variable by case, and likely changing in response to new global trends.

Why is media coverage important to grassroots movements? Quite simply, many

SMO activities require money or other resources, but media coverage is free. It is one of the most cost-effective ways for SMOs to raise awareness. This is crucial, because many grassroots SMOs are resource poor and must make careful decisions regarding resource allocation (Martin 2008). The media bestows untold benefits upon resource-deprived

SMOs, and can even be conceptualized as a resource in and of itself (McCarthy & Zald

1977). First, media coverage allows for SMOs to spread their message to the bystander public, who may or may not even be aware of the group’s existence or activities (Jenkins

& Perrow 1977; McAdam 1983). When the media pays attention to a group’s activities, that group is elevated in the public eye and deemed worthy of participation in the public debate (Gamson 1998; 2001; Roth & Vander Haar 2006). In this way, the media legitimizes groups who, by definition, exist outside mainstream political institutions.

SMOs can parlay this legitimacy into alliances with other actors (both political and non- political) and increased visibility. They can use their newfound standing to attract new human and economic resources (Andrews & Caren 2010).

It is sometimes argued that all media coverage is positive, but it must be noted that sometimes the media can harm movements with negative coverage (or no coverage at all). The media sets agendas and establishes talking points by deciding what stories to 84 cover and how to specifically cover them, and the way in which they use this power tends to ignore grassroots SMOs in favor of institutionalized elites (Kiousis 2004). The resulting coverage is skewed in favor of elite interests, both in the topics they cover and the content of that coverage (Ryan 1991; McCarthy et al. 1999). Gitlin attributes the decline of the leftist movement on American college campuses in part to the mass media, who originally gave the group an outlet for their message but eventually started to cover the group in a hostile and less-than-serious manner (1980). Without media allies, the movement eventually lost momentum. Even in circumstances where the stories appear to be free of moral judgment, research indicates that the news tends to be skewed in favor of one viewpoint over others (Rohlinger 2002; Budner & Kraus 1995). In circumstances where there is a conflict between grassroots actors and institutionalized elites, the media tends to favor elite voices. This has been observed in urban redevelopment (Kaniss 1991;

Fulton 2001), publicly subsidized sports stadiums (Delaney & Eckstein 2003; 2008) and public housing reform (Schoene forthcoming). The media is clearly not a staunch ally of social movement arguments.

That being said, it would be inappropriate to claim that the media is necessarily biased against SMOs. There are many examples of the media contributing to SMO success, however that success is defined (see chapter 2, for instance). When groups do not attract enough media attention, they often struggle to accomplish their goals

(Koopmans 2004). More important than just the amount of coverage, however, is the type of coverage received. Social movement groups appear to have some level of control over the stories published about them, and they control this by the process of framing

(Snow et al. 1986; Snow & Benford 1992). For example, during the anti-nuclear 85 movement, media coverage allowed SMOs to construct and distribute interpretative packages that framed the issue in a way that garnered support for their concerns.

Additional research shows that SMOs who can frame issues around the concerns of economic growth and development tend to receive more favorable treatment from media actors (Andrews & Caren 2010). Organizations that put more effort into their media message tend to be more successful at pushing their preferred frames into the public discourse (Rohlinger 2002).

The ultimate goal, as argued by previous work, should be media standing (Roth &

Vander Haar 2006; Schoene forthcoming). Media standing, a concept first developed by

Gamson (2001), is defined as an individual or group receiving the legitimacy necessary to participate in the public debate. Those who the media cover and interact with are elevated to a position of authority in the public debate. Institutional, bureaucratic elites tend to achieve media standing quite easily – their position grants it to them. For grassroots SMOs, however, the process is much more complicated. They have to prove themselves worthy of attention before they receive media standing. The literature cited above offers many different examples, some where the media encouraged SMO activism and some where the media hindered it. So, with that in mind, what factors attract the ability of SMOs to achieve media standing? I previously wrote that the media acts as a gatekeeper, and the way in which certain actors are either included or excluded strongly indicates that there are many factors involved beyond just newsworthiness.

The news represents not just an account of events, but reflects the practices of those who do the reporting (Molotch & Lester 1974). Journalistic practices can marginalize grassroots SMOs who lack any previously established media standing, but 86 who typically need the coverage the most (McCarthy et al. 1999; Tuchman 1972).

Protest events are still one of the key factors that attract the media’s attention (Schoene forthcoming). Within this type of event, however, larger events tend to be more strongly covered than smaller events (Andrews & Caren 2010; McCarthy et al. 1996). The old adage of “if it bleeds, it leads,” also holds true: violent confrontations receive more attention than peaceful demonstrations (Oliver & Meyer 1999). Formal organization is also positively correlated with media attention (Oliver & Maney 2000). Events deemed to be politically important (Myers & Caniglia 2000) or those that tap into current events

(Andrews & Caren 2010) receive disproportionate attention from reporters. Location seems to matter, too. Events in large cities or downtowns also are easier to cover, and thus attract more attention than events in sparsely populated areas or hard to reach locations (Ferree et al. 2002). The media is also motivated by publication concerns, and

SMOs that can tap into an issue the public is already concerned about will receive additional coverage, which Downs shows using ecology and groups (1972).

However, this is a double-edged sword, and Gamson & Modigliani demonstrate that when the public loses interest in a certain topic (which can happen quite suddenly and unpredictably) the SMOs working in this area will be hurt (1989). So, the amount of media coverage can fluctuate quite wildly, and this variation does not appear to be reducible to elite bias.

With all this in mind, how does globalization change this equation? Globalization opens up new avenues of resources, new sources of information and new potential allies

(Smith 2001; 2008). One potential new resource may be international media actors. It is not yet known how transnational media sources cover social movement conflicts, but 87 there are many reasons to believe they can be helpful to grassroots movements. The international media, such as the BBC, the New York Times and Al Jazeera, holds a certain level of international legitimacy that local news outlets can never hope to achieve.

Attracting their attention is a powerful statement that a grassroots SMO is worthy of public attention and deserving of a voice. The international media also reaches exponentially more people than the local outlets, and may help connect potential allies in other countries to SMO actors. Coverage by the international media is shown to be an important part of the strategy used by such diverse sources as ethnic insurgencies

(Clifford 2002), terrorist groups (Weimann & Winn 1994), environmental groups

(Bartley 2007), anti-free trade protests (Smith 2001), the South African anti- movement (Thörn 2007) and the Argentine (Byrsk 1993). These media outlets are part of the international system, which allows social change to be pursued both from above and below. Much of the social movement literature typical focuses on one of those two directions, but Byrsk shows how Argentina’s human rights movement benefited from international media coverage at the local level. In other words, change was possible due to efforts at the local and international level.

The international media thus represents a new arena in which a battle for media standing is waged. This arena is not “new” in the sense that it did not exist previously. It is new in that the ability of grassroots actors to have their conflicts covered and their concerns given voice has greatly expanded. Furthermore, thanks to new technologies, the resulting stories are easily accessed by a wider readership. While the battle for media standing is not fought with equal resources, Roth & Vander Haar point out in their study of Los Angeles parkland redevelopment that “elites do not have a monopoly on media 88 standing” (2006: p. 135). However, even though community opposition to this development was ultimately successful, the aggregate coverage was indeed skewed in favor of elite, pro-development voices. Where does this disparity come from? Many studies argue that bias in coverage of social conflicts is due to the practices and preferences of reporters rather than any kind of elite-level conspiracy. While journalists work in a field with high standards of objectivity (especially for the New York Times), they are also human beings. Reporters are under heavy pressure to meet tight deadlines, and they lean towards reliable sources (Tuchmann 1972). These sources tend to be bureaucratic, institutionalized and elite - exactly what grassroots SMOs are not. As a result, the aggregate coverage is often skewed in favor of the rhetoric preferred by these institutionalized elites. Therefore, since journalists need quotes to add both substance and legitimacy to their stories, they indirectly confer greater media standing upon institutionalized voices at the expense of the groups who need media standing the most.

This can hurt SMOs in their efforts for social change (Gamson 2001; Kaniss 1991; Ryan

1999; Roth & Vander Haar 2006). When considering the case of Europe’s anti-austerity movement, media standing can be a valuable tool for grassroots SMOs, but they stand at a natural disadvantage to voices that may be expected to support austerity policies: politicians, government officials, financial analysts and bankers. Of course, many of these elites still have to live in their home countries, so it is unclear whether they will highlight the positive or negative aspects of austerity in their public statements. In contrast, citizens and their leaders are less accessible to journalists and therefore have a harder time gaining media standing.

89

To test this, I conceptualize direct quotes as media standing and use them as my unit of analysis (Roth & Vander Haar; Tuchmann 1972; Schoene forthcoming). Most international media sources can be expected to cover the Europe’s austerity crisis, but coverage is not the same as standing. In contrast, a direct quote serves as a sign that this person or group deserves to be taken seriously in a social debate. Ferree and colleagues expand on this argument:

Commentary on the issue is an attempt to convey a preferred way of framing it and to increase the relative prominence of the preferred frames in the mass media arena. Those who are quoted are overwhelmingly spokespersons for collective actors – government ministries, political parties, or organizations that claim to represent the interests or values of some constituency, speaking on behalf of them. (Ferree et al. 2002: 11-12).

In this study, I use quotes published in The New York Times to measure both the aggregate coverage dedicated to both sides of the European austerity debate and to determine the sentiment, or general leaning of the statement, based on the affiliation of the individual quoted. Quotes serve as a better unit of analysis than entire articles because many times, a full article of several hundred words (or more) cannot be coded as holding any noticeable bias. Quotes are a smaller unit of analysis, so the sentiment can be better isolated.

I choose The New York Times for a few theoretical reasons. First, to study if a

European social movement can achieve media standing in a media outlet based in a non-

European country, then it makes sense to select a media outlet based in a country whose opinion and reporting matters most. In the austerity crisis, that country is the United

States, a place where bankers, financial actors, politicians and even the President were expected to comment about European . Secondly, the New York Times has

90 global reach, both in print and on the web. The Times circulates 110,747 copies every day in the European area, and can also be accessed in alternative methods, too (New

York Times Media Kit 2014). Third, among American competitors, this is by far the most respected in the country (Tietz 1999). Fourth, among other American options, The New York Times has a stronger focus on international news than many of their competitors. In sum, this source possesses the right mix of accessibility, respectability and international focus. Was the anti-austerity movement in Europe able to achieve media standing, as measured by quotes, in an American media outlet? If so, to what extent did this affect the movement?

Data and Methodology

To acquire data for this project, I used the Lexis-Nexus Academic search engine to access the New York Times archives. The Times has long been considered the gold standard in social movement media research, and is a prime example of an internationally-oriented newspaper. I choose the Times as my data source because it offers the best combination of broad circulation, consumer respect and international focus of other options.

Secondly, most previous studies of media standing use local or national newspapers as data sources, but the European austerity crisis goes beyond national borders. To my knowledge, there is no such study that examines how an internationally-oriented media source based in one country covers a social movement dispute in other countries, and what effect, if any, that has on the outcomes.

91

I then conducted searches using a broad variety of terms16 from January 1, 2010 to March 31, 2013. I began in 2010 because this is the start of the sovereign debt crisis. I chose to stop collecting data at the end of the first quarter of 2013 because bailouts and austerity policies (and their aftermath) were already in effect throughout most of Europe by this time, so it is clear that the anti-austerity push had failed. This search yielded

1,238 articles. Then, with the help of three research assistants, we carefully read and coded every quote published within this selection of articles for a total of 4,496 quotes.

For each quote, my team of research assistants extracted two pieces of relevant information: institutional affiliation and sentiment. Theory expects that bureaucratic, institutionalized sources will have an easier time achieving media standing. Therefore, we have to determine the source of each quote. I constructed ten affiliation codes. First,

I discuss the institutionalized sources: (1) politicians, (2) government officials, (3) bankers/financial analysts, (4) European Central Bank officials and (5) private businesses. On the community side, I code for (6) citizens and (7) citizen leaders.

Finally, I also control for other theoretically important voices: (8) authors,17 (9) experts18 and (10) other. After coding for affiliation, I coded each the sentiment of each quote: (0) pro-austerity, (1) anti-austerity or (2) neutral or no stand taken. Of course, as with any news story, the coverage of this conflict will not be uniformly distributed, and there is

16 When searching electronic archives, the results are totally dependent on what search terms are used. To ensure I found all related articles, I performed multiple searches of this timeframe with different search terms. My main search terms include “Austerity,” “European Austerity,” “European Central Bank,” “Protest” and “IMF.” 17 This category typically refers to columnists or individuals who write editorials for the New York Times. Sometimes, a guess columnist will identify themselves as having a different affiliation, in which case they are coded as their self-identified affiliation and not as an author. Inside this category, one particular voice dominated all others: Paul Krugman, the New York Times chief economics writer. 18 Those that fall into the category of expert are usually professors or researchers who are not affiliated with a bank or government (as both of those get their own category). 92 significant week-to-week variation in the amount of coverage the austerity crisis receives.

Figure C.1 displays the weekly count of total quotes published in the New York Times over the timeframe of the sample.

[Insert Figure C.1 Here]

Furthermore, there are times in which coverage of the austerity crisis skews more positive and times where it skews more negative (as measured by the sentiment of these quotes).

To provide an idea of the pro/anti-austerity split in each week, figure C.2 graphs the weekly count of both pro and anti-austerity quotes published in the New York Times over the timeframe of this study.

[Insert Figure C.2 Here]

By comparing how often each individual affiliation is quoted to the sentiment of their statement, I can evaluate how successful the anti-austerity movement was at achieving media standing, when compared to their opposition. It is important to note that “neutral or no stand taken” does not imply apathy or a lack of knowledge about the case. Instead, most of these are descriptive, and simply do not pass any moral judgement about

European austerity.

This approach has several advantages but is also limiting in a few important ways.

First, my singular focus on the New York Times leaves other newspapers and many other media platforms unexplored. While a study of how the Times’ coverage compares to other media outlets would be interesting, my research question concerns how groups achieve media standing, and the Times is consistently the most respected newspaper in the world. Therefore, I conceptualize quotes in the Times as the highest level of media standing. Secondly, studying the international media leaves the local media, which is 93 most likely to be read by the individuals most affected by austerity, largely unexplored.

However, the local media is not the focus of this paper.

In my analysis, I first present data describing the newspaper’s coverage of

Europe’s austerity crisis. Who was quoted more? Then, I examine the sentiment of each quote while controlling for institutional affiliation. Who supported austerity, who opposed it, and how often did these voices appear in the Times? I answer these questions using cross-tabulations of sentiment and affiliation, reporting what percentage of each affiliation’s total quotes fall in the three sentiment categories.

Results

Table C.1 displays the total number of quotes from each of the 10 institutional affiliations coded in my sample to appear in the Times’ coverage of the European austerity crisis.

Theory expects that institutionalized, bureaucratic sources will be better able to achieve media standing than grassroots social movement actors. Table C.1 demonstrates that this is indeed true.

[Insert Table C.1 Here]

The most quoted individuals in the sample are politicians, accounting for about 21% of the total sample of quotes. Bankers, financial analysts and IMF actors are the second most quoted group, at about 19%. Government officials are also commonly quoted, accounting for 15.5% of all quotes in the sample. Interestingly, the most powerful group of people in the sample, the European Central Bank, receives relatively little attention.

ECB officials only account for about 2.5% of total quotes. Next, private businesses are also not a major player in this case, as their quotes only make up 3.5% of the total 94 sample. Taken together, all these actors have varying degrees of institutionalized legitimacy, but they would certainly not be described as grassroots actors.

Institutionalized voices make up a whopping 60% of all quotes in my sample.

In contrast, the grassroots actors had a much harder time achieving media standing. These people are expected to be the most common and most vocal opposition to austerity measures, but when taken together, citizens and their leaders only account for about 13% of all quotations. Citizens make up about 11% of the total sample, while their leaders constitute an additional 2.5% of all quotes. These are the very people at the heart of the anti-austerity efforts, and yet they have a much harder time achieving media standing. This has implications for the grassroots fight against austerity, as the voices of the social movement are struggling to achieve media standing. However, it is inappropriate to say that they were ignored. Even an international media source like the

Times sought out community voices and incorporated their rhetoric into articles about the case.

The third broad category of affiliations, which I have named “other actors,” is primarily made up of experts and authors who write for the newspaper. For many of the authors, who account for 6% of the total sample, it is not entirely fair to categorize this as media standing, as they are part of the media. I simply use this category to point out how many articles about austerity came directly from the New York Times themselves. What is more interesting is that experts, financial or otherwise, account for a significant portion of quotes: about 17%. These voices are certainly seen as bureaucratic and legitimate, and their media standing reflects this. When in need of a quote, the newspaper is likely to

95 seek out the opinion of an expert, giving this group a leg up in the race for media standing.

However, table C.1 only displays the number of quotes per affiliation. It does not discuss the content of each quote. As discussed in the literature review, the point of offering a quote is to influence on an issue, or to take some kind of a stand. Table C.2 expands on the information provided in table C.1 by breaking down each affiliation’s quotes into three categories: pro-austerity, anti-austerity and neutral.

[Insert Table C.2 Here]

When analyzing the sentiment of quotes by institutional affiliation, an interesting picture emerges. First of all, the sentiment of the Times’s aggregate coverage is decidedly anti- austerity. 32% of total quotes over this timeframe were anti-austerity, while only 20% qualify as pro-austerity. Even with the vast majority of the quotes coming from bureaucratic, institutionalized sources, it is a mistake to assume that these voices felt the same way about austerity policies as the EU elites did. As many other scholars have shown, these results demonstrate the rhetoric of actors in a social movement dispute is more diverse that is sometimes thought, and the pages of the newspaper represent a contested battleground where actors fight for standing and influence.

Politicians offer a mixed opinion on austerity. They support it in about 39% of their quotes while criticizing it 24% of the time. Politicians are often in a very tough position. They are required by ECB and European Parliament to implement austerity measures that are deeply unpopular with their voters (as evidenced by the pattern seen in grassroots quotes, which I discuss momentarily). Politicians therefore must try to sell austerity as something that is indeed painful, but necessary for the future of the Eurozone 96 and the national economy19. Government officials also show a slightly pro-austerity pattern, but it is by no means overwhelming support. They speak positively about austerity in about 23% of their quotes and criticize it 15% of the time. Again, politics is a balancing act, where the government must try to appease their voters while toeing the line laid down by more powerful actors in the European Union. Many government officials are appointed by politicians in power, so if the politician is voted out, they are often out of a job. When examining the other institutionalized actors, the only affiliation with a decidedly pro-austerity position is the European Central Bank. As the chief driver of austerity measures, the ECB obviously tries to sell the positive, long-term impacts of their policies, but their voices do not appear very frequently. When they are quoted, however, they support austerity in about 52% of their quotes, compared to expressing critical rhetoric in only about 6% of their quotes. Bankers, financial analysts and IMF officials display a slightly anti-austerity sentiment. This group critiqued austerity in about 30% of their quotes, in comparison to praising it in about 21% of their quotes. Just like with politicians, bankers and financial analysts are a diverse group of people. They understand the complex nature of the European sovereign debt crisis and this is reflected in their statements. It is interesting, though, that their aggregate voice skews anti- austerity. Businesses are strongly opposed to austerity, expressing misgivings about these policies in about 51% of their quotes vs. supporting austerity 21.2% of the time.

They likely understand that demand drops during austerity, and they stand to be severely

19 Additionally, German politicians tend to be very supportive of austerity, as they are often driving the terms of national bailouts in countries like Spain, Greece and Italy. 97 hurt when people have less money to spend. All in all, institutionalized actors are not nearly as strong in their support of austerity as previously thought.

Next, I examine the grassroots actors. While citizens and their leaders did not achieve the same level of media standing as their institutionalized counterparts, their voices are much more unified. Citizens were opposed to austerity in 50% of their quotes and only in favor of it in 9% of their statements. While citizen leaders only account for about 3% of the total sample, they used what limited opportunities they had to fiercely criticize austerity measures, supporting the policy only about 5% of the time while criticizing it 57% of the time. Social movements need to stay on message to stay successful. The citizens active in the anti-austerity movement are obviously deeply opposed to it. This is why they protest in the first place. That being said, the average citizen has slightly more leeway to express diverse opinions. The leaders of the anti- austerity movement are much more constrained. They cannot risk saying anything that be remotely construed as pro-austerity.

When looking at the final three affiliations, it seems that the editorial board of the

New York Times was decidedly anti-austerity. Many of these articles come from the pen of Paul Krugman, one of the most consistently negative voices about European austerity policies. It seems that many of his colleagues who write for the Times followed his example. Experts, on the other hand, often try to balance their pro-austerity statements with an anti-austerity caveat. These people have a great freedom to offer nuanced takes on the situation, so it is hard to find a patter in their quotations, but they tend to identify the negative aspects of austerity more often then they highlight the positives.

98

Discussion

The results of this content analysis of the New York Times’s European austerity crisis produce three interesting findings. First, theory expects that institutionalized, bureaucratic actors will be better positioned to achieve media standing in a social movement conflict such as the one studied here. My content analysis suggests that this is indeed what happened when considering a media source based outside the country of conflict. Among the affiliations I identify as being major players in the austerity conflict, politicians, government officials, experts and financial analysts are the most quoted of the total sample. Grassroots actors had a much harder time achieving media standing, as citizens and their leaders account for only about 13% of total quotes. Grassroots actors are by no means ignored by the media, but their voices are in danger of being drowned out by their institutionalized counterparts. For social movement groups seeking to make a difference, they should not ignore media strategy but need to be aware of the limits of the media. This finding corroborates previous work that identifies the media as a resource to grassroots movements, but notes the limits of such a strategy (Schoene forthcoming; Soberij 2012; Rohlinger 2002). Additionally, it is interesting that arguably the most powerful player in the austerity debate, the ECB, was quoted so rarely. This may indicate that the more powerful an actor is, the less they need the media to publicize their claims. They have actual power, not just power granted to them by media actors.

The ECB operates in a non-democratic manner, often requiring austerity as a condition of any potential bailout. They do not need the public’s support and are not accountable to the voters. Future research should examine who institutionalized elites with varying degrees of power interact with the media. 99

Secondly, this finding advances previous research by showing how this process plays out in the pages of a highly respected international outlet. Previous research typically focuses on the local media, and one may expect that the international media is too far removed from the conditions on the ground to interact with social movement actors in any significant way. While these grassroots actors are not quoted nearly as much as their institutionalized counterparts, getting your voice in print within the pages of the Times has to be considered a major victory for social movement groups. 13% is not anywhere close to a majority of quotes, but this is a minority worth taking seriously.

But perhaps most importantly, it is easy to wonder if any of this even mattered.

The aggregate coverage of this case was actually skewed against austerity policies. Far more anti-austerity statements appeared in the Times than pro-austerity statements, about

31% vs. 21%, with the remaining quotes being neutral in slant. This is important because austerity measures were largely implemented at will by the European Central Bank and

European Commission in the nations suffering from sovereign debt issues. Anti-austerity voices had their say, but ultimately could not achieve their goals. Some countries put up more resistance to austerity than others, and at times continental elites allowed for austerity measures to be loosed in response to citizen outrage. Nevertheless, austerity policies became the law of the land throughout Europe. So, the question becomes: if the coverage of the gold standard for international media was categorically anti-austerity, and yet austerity became continental policy, then did media coverage matter at all to the movement?

It is likely that in the moment, the anti-austerity coverage of the Times did not make much of a difference in decisions made by the ECB, the EC or the European 100

Parliament. That being said, movements need to be studied over their life course. If the goal of the anti-austerity movement was to scale back unpopular economic policy, then the movement was a failure. However, the movement changed how the discussion was framed by economic elites. It is not possible to know exactly how much media coverage played into this transformation, but anti-austerity protests and sustained citizen pressure helped loosen austerity measures that crippled growth in Greece, Spain, Ireland and Italy.

This may have happened independent of a grassroots movement, as this content analysis shows that politicians and several other elite actors were not fond of austerity either.

Nevertheless, the fact that newspaper coverage trended anti-austerity and austerity became the law of the land anyway should serve as a warning to social movement groups fighting a similar battle. The media can be a resource, but it is far from the only thing necessary for a successful grassroots movement.

Conclusions

This study adds an important dimension to social movement theory, and our knowledge of the important relationship between the media, social movement actors and institutionalized elites. As activism becomes increasingly transnational, the international media may be an important resource for grassroots SMOs. I find that while social movement actors are not nearly as successful at attaining media standing than their bureaucratic, institutionalized counterparts, the media is available as a resource to them, even an international outlet like the New York Times. However, this study is limited in a few important ways. First, I study only the sentiment of each quote, not the particular way in which an argument was framed. One of the most valuable things direct quotes 101 give to a social movement is the ability to frame the argument according to their preferred rhetoric (Snow et al. 1986; Benford & Snow 1988; Rohlinger 2002). Frames allow social movements to diagnose the problem, suggest solutions and deliver this all to the public in neat packages. Applying this to the case at hand, and anti-austerity movement sometimes highlighted the unfairness of a system that unfairly targeted the working class at the expense of the upper class, while other times they stressed national pride and not letting policies laid down by Brussels dictate the direction of their country.

These frames appeal to different people, but I am not able to discern with my data if certain frames were more common than others or more effective than others. Future research should examine the framing strategies used by each side of the austerity debate.

Secondly, as I discuss in my methodology section, I focus on The New York

Times because it is one of the most respected media outlets in the world, but this is a limiting analytical strategy. First, future research should consider how the coverage of the Times compares to the local media coverage within a sample of European countries.

This is a challenging research project that demands diverse language skills and access to several foreign archives, but doing so would illuminate if social movement actors have an easier time achieving media standing in local or international outlets (I suspect local).

Second, and related to this point, it would be very interesting to study how the New York

Times coverage compares to other kinds of non-traditional media. While the Times is a well-respected newspaper, it is exactly that: a newspaper. Much has been made about the decline of newspapers and the rise of other forms of technology for recruitment, protest and activism. High quality newspapers tend to be seen as very trustworthy, but technology and non-traditional media is undoubtedly changing the game for activists. 102

Future research needs to examine how standing in these other forms of media compares to the newspaper.

Despite these limitations, this study illuminates an important gap in our understanding of the social movement – media relationship. Future research should continue to take the transnational element of social movements seriously, even in cases where activism is local. Globalization opens up many new possibilities in a constantly changing society and social movement sector. Those who advocate for the disadvantaged all over the world need to know about each and every resource available to them, and the media still represents a forum for their interest, even in this ever changing world.

103

CHAPTER 5: DISCUSSION & CONCLUSIONS

In a 1998 annual review piece about social movement outcomes, Marco Giugni provocatively asked “Was it worth the effort?” Surprisingly, students of social movements have spent very little time trying to answer this crucial question, and instead presuppose that of course social movements matter. They have to. This speaks to a methodological problem within the field, where scholars so often sample on the dependent variable and only study social movements that clearly influenced society. This arises partly because the successful movements are also the most interesting, partly because scholars must sample on the outcome when studying something like movement emergence, and partly because many scholars desperately want to believe in the power of grassroots activism. To sum up the findings of this dissertation, I find that when individual and organizational social movement activity is studied in a cross-national context, the level of resources both individually and regionally does the best job of predicting positive outcomes for social movements. I also find that professionalization helps keep SMOs viable and healthy, and ideally in a better place to pursue their goals. I find that cities offer a much better place for a social movement sector to take hold and thrive. Finally, I find that transnational actors can be important allies to social movements, which includes IGOs and the international media. This is highly dependent on context, however, as even when social movement actors are able to gain media

104 standing in a well-respected outlet, they can still ultimately struggle to achieve their goals.

The findings of this dissertation paint a mixed picture for grassroots social movements. Scholars have long recognized that the major theories of resource mobilization and political opportunity structure are not mutually exclusive. Reality is complicated, and both theories add important variables to our understanding of movements. While these perspectives are common within social movement research, it is less common to see competing theories tested against each other. When they are compared, it is rarely in a cross-national, quantitative framework. My first two chapters incorporate variables derived from both resource mobilization and political opportunity, and I find much stronger support for resources. Chapter two’s study of European activism finds strong effects of social class, education and regional GDP on the likelihood of personal activism. Chapter one’s study of organizational outcomes finds that urban GDP and global competitiveness are positively associated with organizational viability and age. Outside of these two main perspectives, I also find small support for the long-dismissed grievances perspective, which may have simply been framed incorrectly. I find that in Europe, dissatisfaction with institutions like the economy and the government is not enough to produce wide-scale activism, but it certainly matters.

The more satisfied people are with the institutions they rely on, the less likely they are to participate in social movement activism. What does this mean for social movement theory? Very simply, that while resources are the primary determinant of social movement activism (at least in the cases studied here), it is necessary to understand why people are so upset before understanding protest movements. Ultimately, though, I find 105 that resources best predict social movement organizational health and individual social movement activism. Most importantly, I find evidence that this process is comparable across national borders, which are indeed becoming less important in the era of globalization.

While I pay less attention to political opportunity variables, I do incorporate context into my analysis. My overall findings point to the convergence of social movement activity in the world system, although more cross-national research is necessary to truly confirm these claims. While world polity theory was not a major element of my literature reviews, when the findings of all three studies contained here are considered together, I see support for some of world polity’s ideas. Specifically, this perspective speaks of the emergence of a world culture that incorporates actors at the transnational, national and local level. These actors include nation-states, international bodies (such as the European Union discussed at length in this dissertation), international organizations (such as the IGOs discussed in chapter two) and the individuals influenced by these groups (McNeely 2012). As a cultural framework, World Polity Theory has its own set of norms and values that dictate how organizations and institutions approach problems like the European austerity crisis discussed in chapter four. Since problem solving is approached from a global, rather than local, framework, the solutions tend to be less creative and more homogenous (Boli et al. 2010).

The major contribution of this dissertation from a social movement perspective is that things are indeed looking similar rather than different across national borders. While chapter three examines a set of fairly similar countries in the contemporary European

Union, chapter one compares such diverse locations as Brussels, Belgium and Buenos 106

Aires, Argentina. Across a sample of 67 global cities, or the cities on the forefront of global society, resources, IGO allies and professionalization are strongly linked to positive outcomes for SMOs. This happens regardless of the level of development, although it must be noted that the majority of SMOs in the sample are located in high income countries. Even still, I test urban GDP and find that the same patters hold regardless of income levels. The factors that contribute to SMO success in the United

States and Western Europe also appear to be good predictors of strong SMOs in lower- income nations. This convergence, called institutional isomorphism, is a key element of

World Polity Theory. Isomorphism states that as institutions develop and grow, they become more like each other rather than innovating and diverging (DiMaggio & Powell

1983; Powell & DiMaggio 1991). Isomorphism has been observed in other institutions, such as governance and education, both of which have increasingly taken on a Western form. When specifically studying movements, Berkovitch finds that the women’s movements in multiple countries had so many similarities, they form a coherent, singular international movement (1999). In the era of globalization, institutions are taking on more similarities than differences, which may make political opportunity less important if opportunity structures do not differ greatly across national borders. Resources, however, are much unevenly distributed, and it appears the resources that encourage organizational health and viability in high income countries have similar effects in low-income nations.

In chapter two, I ran supplemental models with regional controls that I ultimately did not include. I omitted them because the regional control variables did not add anything extra to the model. It appears the social movement organizations have more similarities than differences, even in diverse national contexts. 107

Of course, this finding should be interpreted with some caution. Specifically, homogeneity does not happen overnight and the findings of this dissertation do not illuminate how SMOs are beginning to look alike. Additionally, homogeneity is often fiercely resisted by local communities. World culture and local often interact, which can produce not isomorphism but something entirely different. Before I truly apply my results to world polity theory, the existence of a world polity needs to be shown quantitatively, not just taken for granted. Even still, the fact that I find similar predictors of organizational health and individual level social movement activity across national borders adds an important dimension to our understanding of social movements in the globalized world.

Additionally, this dissertation makes the case for studying social movements in a multi-level framework. In chapters two and three, I find strong effects of contextual variables independent of individual or organizational traits, both defined as level-one variables. These contextual variables would not be clear without a hierarchical data structure. Future research needs to take this design more seriously. Cross-national surveys offer a wealth of individual and regional data that can be used to study movement behavior across a wide variety of national circumstances. Harmonization also allows researchers to link individual to regional data for a hierarchical structure. If the field is to be advanced, we need to take the small-scale studies of individual movements and apply these findings to cross-national research. If future activism will indeed be global, then more cross-national evidence must be added to the body of literature.

This dissertation also adds to the field of urban sociology, and specifically to the overlap between urban studies and social movements. This paper finds strong effects of 108 the urban context on social movement participation and organizational health. Cities, as has been previously theorized, act as movement spaces where social movement activity can play out. In chapter two, I find that the most globally competitive cities offer a better environment for social movement organizations, as measured by age and viability. In chapter two, I find that various forms of individual activism are more common in the cities of Europe. Social movement participation is encouraged by urban status and population density. While total population does not seem to influence participation, the networks that density encourages do. Social movement theory has been slow to examine expressly urban activism, usually focusing on major urban protests when they do.

However, these findings demonstrate that cities are also well suited for the more mundane things SMOs do that can ultimately make or break a movement.

Cities likely act as a mechanism that connects social movement organizations to other factors known to encourage activism. In other words, the things SMOs need are already most prevalent in cities. I find that resources exhibit a strong, positive effect on both individual activism in Europe and on SMO viability all over the world. Urban areas tend to be wealthier. I find that wealthier Europeans are more likely to participate in protests and political activism. Urban dwellers tend to be wealthier than rural inhabitants. I find that education is positively correlated with activism. Urban residents tend to be better educated than their rural counterparts. I find that IGO contacts strongly predict organizational viability and organizational age. These IGOs are usually located in major cities. Even chapter four, which departs from the style of research established in chapters two and three, shows that grassroots actors can attract the attention of powerful

109 actors. The international media pays more attention to cities than to rural areas. Cities pack all of the things SMOs need to be viable and successful into a very dense area.

For SMOs struggling and fighting to make a difference in an indifferent world, the lesson is clear. If you have limited resources but geographical mobility, your operations are best served in urban areas. Of course, SMOs have real constraints on where they can locate. This depends on their goals, constituency, membership structure and many other factors. However, cities offer SMOs the best value for their limited resources. Moving forward, social movement operations, activities and expansion should be primarily urban in nature, as this maximizes their opportunities for success. This does not mean rural movements cannot have success, nor does this mean all urban social movements will achieve their goals, but the results of this dissertation point towards the urban context as providing a better environment for movements to form.

However, it is the concept of “success” with which I would like to end this dissertation. In the study of social movements, it is too often taken for granted that movements must matter, and our research reflects this bias. However, a quick glance at the major findings presented here demonstrates that movements are not as powerful as some may want to believe. In chapter two, I find that global cities offer a better environment for transnational SMOs, but most of these SMOs are in high-income cities like Brussels, New York and London. In chapter three, I find that urban status and resources encourage social and political activism, but the countries with the most grassroots activity are the high-income, well-functioning democracies of Northern and

Western Europe. Countries like Norway, Sweden and Germany have the highest proportion of their citizens reporting being involved in some kind of political activity, be 110 it as moderate as signing a petition or as strong as taking part in a protest. The more corrupt, lower-income parts of Southern and Eastern Europe do not have nearly the same levels of citizen activity. This is especially when you consider high-cost activism like boycotts and protests. In chapter four, I find that while The New York Times allowed grassroots actors to gain media standing and produced coverage that was anti-austerity in sentiment, but austerity still became the official policy in many European countries. It is fair to ask, as Giugni did in 1998, was it worth the effort? Do social movements even matter?

When we study famous movements like Civil Rights or European anti-communist movements, it is easy to be impressed with the power of grassroots activism. However, it is also interesting that the countries least in need of a strong social movement sector produce the greatest amount of activism. The countries or areas with the most problems, be they economic, social, political, environmental or otherwise, do not produce levels of activism nearly as high as their more stable counterparts. This implies that social movements are the result of an already strong, stable society, and not as powerful of an agent for social change as many of us would like to believe. These findings paint a slightly sobering picture of grassroots activism going forward, but to say movements are less meaningful than sometimes thought does not imply they are irrelevant. For students of social movements, then, the task moving forward is to identify how social movements can be nurtured, encouraged and ultimately successful in environments like Eastern

Europe where they should otherwise fail. But, even with all this being said, it is well established that social movements are an important part of global society, and this

111 dissertation goes a long way to explaining how they can operate successfully in the globalized, urban world.

112

REFERENCES

Abrams, C. 1965. “Squatter Settlements: the Problem and the Opportunity.” Report to the US Agency for International Development, Department of Housing and Urban Development, Washington, D.C.

Ades, Alberto F. and Edward L. Glaeser. 1995. “Trade and Circuses: Explaining Urban Giants.”

Alderson, Arthur S. 2002. “Globalization and the Great U-Turn: Income Inequality Trends in 16 OECD Coutnries.” American Journal of Sociology 107(5): 1244- 1299.

Alderson, Arthur S. & Jason Beckfield. 2004. “Power and Position in the World City System.” American Journal of Sociology 109(4): 811-851.

Alderson, Arthur S. & Jason Beckfield. 2006. “Globalization and the World City System: Preliminary Results from a Longitudinal Dataset.” Ch. 2 in Cities in Globalization: Practices, Policies and Theories. Eds. Peter Taylor, Ben Derudder, Pieter Saey & Frank Witlox. New York, NY: Routledge.

Almeida, Paul. 2003. “Opportunity Organizations and Threat-Induced Contention: Protest Waves in Authoritarian Settings.” American Journal of Sociology 109:345-400. Amenta, Edwin and Yvonne Zylan. 1991. “It Happened Here: Political Opportunity, the New Institutionalism, and the Townsend Movement.” American Sociological Review 56: 250-265. Amenta, Edwin Bruce G. Carruthers, and Yvonne Zylan, 1992. “A Hero For the Aged? The Townsend Movement, The Political Mediation Model, and U.S. Old-Age Policy, 1934-1950.” American Journal of Sociology. 98(2): 308-39. Andrews, Kenneth T. 2001. “Social Movements and Policy Implementations: The Mississippi Civil Rights Movement and the War on Poverty, 1965 to 1971.” American Sociological Review 66: 71-95.

113

Andrews, Kenneth T. 2002. “Movement-Countermovement Dynamics and the Emergence of New Institutions: The Case of “White Flight” Schools in Mississippi.” Social Forces 80: 911-936.

Andrews, Kenneth T. and Neal Caren. 2010. “Making the News: Movement Organizations, Media Attention, and the Public Agenda” American Sociological Review 75:841-866.

Anyon, Jean. 2014. Radical Possibilities: Public Policy, Urban Education and a New Social Movement. New York. NY: Routledge.

Armstrong, Elizabeth A. and Mary Bernstein. 2008. “Culture, Power, and Institutions: A Multi-institutional Politics Approach to Social Movements.” . 26:74-99.

Ayres, Jeffrey M. 2002. “Transnational Political Processes and Contention against the Global Economy. Ch 11 in CFh.3 in Globalization and Resistance, eds. Jackie Smith & Hank Johnston. Lanham, MD. Rowman & Littlfield. Babb, Sarah. 1996. “’A True American System of Finance’: Frame Resonance in the U.S. Labor Movement, 1866-1886.” American Sociological Review. 61: 1033- 1052.

Bartley, Tim. 2007. “Institutional Emergence in an Era of Globalization: The Rise of Transnational Private Regulation of Labor and Environmental Conditions.” American Journal of Sociology 113: 297–351.

Bendaña, Alejandro. 2006. “NGOs and Social Movements: A North/South Divide?” Civil Society and Social Movements Programme Paper 22.

Bairoch, Paul. 1988. "Urbanization and Economic Development in Countries Affected by the Industrial Revolution." Ch. 21 in Cities and Economic Development. Chicago: Univ. of Chicago Press.

Berkovitch, Nitza. 1999. From Motherhood to Citizenship Women’s Rights and International Organizations. Baltimore, MD: Johns Hopkins University Press.

Berry, Brian J. and John D. Kasarda. 1977. "The Ecological Approach." Chap.1 in Contemporary Urban Ecology. New York: MacMillan.

Blau, Judith R. & Peter M. Blau. 1982. “The Cost of Inequality: Metropolitan Structure and Violent Crime.” American Sociological Reviewi 47: 114-129.

114

Blumer, Herbert. 1951. “Collective Behavior.” Pp. 166-222 in New Outline of the Principles of Sociology, Ed. A.M. Lee. New York, NY: Barnes & Noble.

Boli, John & George M. Thomas. 1997. “World Culture in the World Polity: A Century of International Non-Governmental Organization.” American Sociological Review 62(2): 172-188.

Boli, John. 2001. “Sovereignty from a World-Polity Perspective.” Pp. 53-82 in Problematic Sovereignty: Contexted Rules and Political Possibilities. Ed. Stephen D. Krasner. New York, NY: Columbia University Press.

Bouteligier, Sofie. 2008. “Global Cities and Networks of Global environmental Governance.” Presented at the 49th annual Convention of the International Studies Association, March 2008. San Francisco, California.

Brockett, Charles D. 1991. “The Structure of Political Opportunities and Peasant Mobilization in Central America.” Comparative Politics 23: 253-274. Budner, Stanley J. & Ellis S. Krauss. 1995. “Newspaper Coverage of U.S.–Japan Frictions: Balance and Objectivity.” Asian Survey 35(4): 336-356.

Burgess, Ernest W.1925. “The Growth of the City: An Introduction to a Research Project” in The City, Robert E. Park et al. Chicago: University of Chicago Press.

Brysk, Alison. 1993. “From Above and Below: Social Movements, the International System, and Human Rights in Argentina.” Comparative Political Studies 26(3): 259-285.

Brysk, Alison. 1994. The Politics of Human Rights in Argentina: Protest, Change and Democraticzation. Stanford, California: Stanford University Press.

Caniglia, Beth Schaefer. 2002. “Elite Alliances and Transnational Organizations.” Ch. 9 in Ch.3 in Globalization and Resistance, eds. Jackie Smith & Hank Johnston. Lanham, MD. Rowman & Littlfield.

Castells, Manuel. 1976. “Theoretical Propositions for an experimental Study of Urban Social Movements.” Urban Sociology: Critical Essays.

Castells, Manuel. 1977. The Urban Question: A Marxist Approach. Cambridge, MA: MIT Press.

Castells, Manuel. 1983. The City and the Grassroots: A Cross-cultural Theory of Urban Social Movements. Berkeley: University of California Press.

115

Castells, Manuel. 1997. “The Other Face of the Earth: Social Movements Against the New Global Order.” Pp. 68-109 in The Power of Identity. , ed. Malden, MA: Blackwell.

Cox, Kevin R. & Andrew Mair. 1988. “Locality and Community in the Politics of Local Economic Development.” Annals of the Association of American Geographers 78(2): 307-325.

Cress, Daniel M. and David A. Snow. 1996. “Mobilization at the Margins: Resources, Benefactors, and the Viability of Homeless Social Movement Organizations.” American Sociological Review 61(6):1089-1109. Cress, Daniel M. and David A. Snow. 2000. “The Outcomes of Homeless Mobilization: The Influence of Organization, Disruption, Political Mediation, and Framing.” American Journal of Sociology 105: 1063-1104.

Clarke, Susan E. & Gary L. Gaile. 1997. “Local Politics in a Global Era: Thinking Locally, Acting Globally.” The Annals of the American Academy of Political and 551(1): 28-43.

Clifford, Bob. 2002. “Political Process Theory and Transnational Movements: Dialectics of Protest among Nigeria’s Ogoni Minority.” Social Problems 49(3): 295-415.

Crouch, Colin. 2011. The Strange Death of Neoliberalism. Cambridge, UK: Polity Press.

Davis, Gerald F., Doug McAdam, W. Richard Scott, and Mayer N. Zald, eds. 2005. Social Movements and Organization Theory. Cambridge University Press.

Delaney, Kevin J. & Eckstein, Rick. 2003. Public Dollars, Private Stadiums. New Brunswick, NJ: Rutgers University Press.

Delaney, Kevin J. & Rick Eckstein. 2008. “Local Media Coverage of Sports Stadium Issues.” Journal of Sport and Social Issues 32(1):72-93.

Drakakis-Smith, David. 1987. Urban and Regional Development in Zimbabwe. New York, NY: Basil Blackwell. della Porta, Donatella, Hanspeter Kreisi, and Dieter Rucht. 1998. Social Movements in a Globalizing World. London: McMillan. della Porta, Donatella. 2011. “Communication in Movement: Social Movements as Agents of Participatory Democracy.” Information, Communication & Society 14(6): 800-819. 116 della Porta, Donatella & . 2012. “Interactive Diffusion: The Coevolution of Police and Protest Behavior with an Application to Transnational Contention.” Comparative Political Studies 45(1): 119-152. Diani, Mario. 1992. “The Concept of Social Movement.” The Sociological Review. 40: 1-25. DiMaggio, Paul J. & Walter W. Powell. 1983. “The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review 48(2): 147-160.

Downs, Anthony. 1972. “Up and Down with Ecology: The ‘Issue Attention Cycle.’” Public Interest 28: 38-50.

Drezner, Daniel W. 2001. “Globaization and Policy Convergence” International Studies Review 3(1): 53-78.

ESS Round 5: European Social Survey Round 5 Data (2010). Data file edition 3.2. Norwegian Social Science Data Services, Norway – Data Archive and distributor of ESS data.

Edwards, Bob and John D. McCarthy. 2003. “Resources and Social Movement Mobilization.” Pp. 116-154 in The Blackwell Companion to Social Movements, edited by David A. Snow, Sarah A. Soule, and Hanspeter Kreise. London: Blackwell Publishers.

Edwards, Bob & John D. McCarthy. 2004. “Strategy Matters: The Contingent Value of Social Capital in the Survival of Local Social Movement Organizations.” Social Forces 83: 621-653.

Fantasia, Rick. 1988. Cultures of Solidarity. Berkeley, CA: University of California Press.

Ferree, Myra Max, William A. Gamson, Jurgen Gerhards & Dieter Rucht. 2002. Shaping Discourse. Cambridge, MA: Cambridge University Press.

Freeman, Jo. 1996. “The Tyranny of Structurelessness.” http://www.hartford- hwp.com/archives/45/112.html

Friedmann, John. 1986. “The World City Hypothesis.” Development and Change 17: 69-84

Friedmann, John. 2001. “World Cities Revisited: A Comment.” Urban Studies 38:401- 424

117

Fulton, William. 2001. The Reluctant Metropolis: The Politics of Urban Growth in Los Angeles. Baltimore, Maryland: Johns Hopkins University Press.

Gamson, William. 1990. The Strategy of Social Protest. Homewood, IL: The Dorsey Press.

Gamson, William A. 1998. “Social Movements and Cultural Change.” Pp. 57-77 in From Contention to Democracy. Eds. M.G. Giugni, D. McAdam & C. Tilly. Lanham, Maryland: Rowman & Littlefield.

Gamson, William A. 2001. “Media and Social Movements.” Pp. 9468-9472 in International Encyclopedia of the Social & Behavioral Sciences. Eds. N. J. Smelser and P. B. Baltes. Amsterdam, Netherlands: Science.

Gamson, William & Antonio Modigliani. 1989. “Media Discourse and Public Opinion on Nuclear Power.” American Journal of Sociology 95: 1-37

Gamson, William A. and David S. Meyer. 1996. “Framing Political Opportunity.” Pp. 275–290 in Comparative Perspectives on Social Movements, edited by Doug McAdam, John D. McCarthy, and Mayer N. Zald. Cambridge, England: Cambridge University Press.

Gans, Herbert. 1979. Deciding What’s News: A Study of CBS Evening News, NBC Nightly News, Newsweek and Time. New York, NY: Vintage

Gitlin, Todd. 1980. The Whole World is Watching. Berkeley, CA: University of California Press. Giugni, Marco G. 1998a. “Was it Worth the Effort? The Outcomes and Consequences of Social Movements.” Annual Review of Sociology 98: 371-393. Giugni, Marco G. 1998b. “The Other Side of the Coin: Explaining Cross-National Similarities between Social Movements.” Mobilization 31:89-105 Goldstone, Jack A. 2004. “More Social Movements or Fewer? Beyond Political Opportunity Structures to Relational Fields.” Theory and Society 33: 333-364 Gould, Roger V. 1991. “Multiple Networks and Mobilization in the Paris Commune, 1871.” American Sociological Review 56: 716-729. Gurr, Ted. 1970. Why Men Rebel. Princeton, New Jersey: Princeton University Press. Haussermann, H. & A. Halia. 2005. “The European City: A Conceptual Framework and Normative Project.” Pp. 43-64 in Cities of Europe: Changing Contexts, Local Arrangements and the Challenge to Urban Cohesion. Oxford, UK: Blackwell.

118

Harvey, David. 1973. Social Justice and the City. Athens, Georgia: University of Georgia Press. Harvey, David. 1982. The Limits to Capital Chicago, IL: University of Chicago Press. Harvey, David. 1985. The Urbanization of Capital. Baltimore, MD: Johns Hopkins University Press. Harris, Chauncy D. and Edward L. Ullman. 1945. “The Nature of Cities.” Annals of the American Academy of Political and Social Science (242): 7-17.

Hawley, Amos. 1981. "The Beginnings of Modern Urbanization." Chapter 4 in Urban Society: An Ecological Approach, 2nd Ed. New York: Wiley.

Hedstrom, Peter. 1994. “Contagious Collectivities: On the Spatial Diffusion of Swedish Trade Unions, 1890-1940. American Journal of Sociology 99: 1157-1179. Holford, John. 1995. “Why Social Movements Matter: Adult Education Theory, Cognitive Praxis and the Creation of Knowledge.” Adult Quarterly Education 45(2): 95-111 Holst, John D. 2002. Social Movements, Civil Society, and Radical Adult Education. Critical Studies in Education and Culture Series. Westport, CT: Bergin & Garvey. Hoyt, Homer. 1939. “The Structure and Growth of Residential Neighborhoods in American Cities.” US Government Printing Office, Washington DC.

Immervoll, Herwig & Linda Richardson. “Redistribution Policy and Inequality Reduction in OECD Countries: What Has Changed in Two Decades?” IZA Discussion Paper No. 6030.

Jad, Islah. 2007. “NGOs: Between Buzzwords and Social Movements.” Development in Practice 17(4-5): 622-629. Jenkins, J. Craig. 1983. “Resource Mobilization Theory and the Study of Social Movements.” Annual Review of Sociology 9:527-553.

Jenkins, J. Craig and Charles Perrow. 1977. “Insurgency of the Powerless: Farm Worker Movements (1946-1972).” American Sociological Review. 42: 249-268.

Jenkins, J. Craig, David Jacobs, and John Agnone. 2003. “Political Opportunities and African American Protest, 1948-1997.” American Journal of Sociology 109: 277- 303.

Kasarda, John D. & Edward M. Crenshaw. 1991. “Third World Urbanization: Dimensions, Theories and Determinants.” Annual Review of Sociology 17: 467-501.

119

Kaniss, Phyllis. 1991. Making Local News. Chicago, IL: The University of Chicago Press.

Katzenstein, M.F. 1998. “Stepsisters: Activism in Different Institutional Spaces.” Pp. 195-216 in The Social Movement Society: Contentious Politics for a New Century, eds. D. Meyer & S. Tarrow. Totowa, NJ: Rowan & Littlefield Press.

Keck, Margret E. & Katherine Sikkink. 1999. “Transnational Advocacy Networks in International and Regional Politics.” International Social Science Journal 51: 89- 101.

Keck, Margret E. & Katherine Sikkink. 2014. Activists Beyond Borders: Advocacy Networks in International Politics. Ithica, NY: Cornell University Press.

King, C.W. 1956. Social Movements in the United States. New York, New York: Random House.

King, Anthony. 1990. Urbanism, Colonialism, and the World-Economy: Cultural and Spatial Foundations of the World Urban System. London, United Kingdom: Routledge Kegan & Paul.

Kiousis, Spiro. 2004. “Explicating Media Salience: A Factor Analysis of New York Times Issue Coverage during the 2000 U.S. Presidential Election.” Journal of Communication 54(1): 71-87.

Kitchelt, Herbert. 1986. “Political Opportunity Structure and Political Protest: Anti- Nuclear Movements in Four Democracies.” British Journal of . 16: 57-85. Klandermans, Bert & Dirk Oegema. 1987. “Potentials, Networks, Motivations, and Barriers: Steps Towards Participation in Social Movements.” American Sociological Review 52(4):519-531. Klandermans, Bert & Suzanne Staggenborg. 2002. Methods of Social Movement: Social Movements, Protest & Contention. Minneapolis, MN: University of Minnesota Press. Knox, P. 1995. “World Cities in a World System.” In World Cities in a World-System, eds. P. Knox & P. Taylor. London: Routledge.

Koopmans, Ruud. 2004. “Movements and Media: Selection Process and Evolutionary Dynamics in the Public Sphere.” Theory and Society 33: 367-91.

Kornhauser, W. 1959. The Politics of Mass Society. New York: .

120

Kriesi, Hanspeter, Ruud Koopmans, Jan Wilem Duyvendak, and Marco G. Giugni. 1992. “New Social Movements and Political Opportunities in Western Europe.” European Journal of Political Research 22: 219-244. Le Gales, Patrick. 2002. European Cities: Social Conflicts and Governance. Oxford, UK: . Lewis, Tammy. 2002. “Conservation TSMOs: Shaping the Protested Area Systems of Less Developed Countries. Ch.3 in Globalization and Resistance, eds. Jackie Smith & Hank Johnston. Lanham, MD. Rowman & Littlfield.

Lipsky, Michael. 1968. “Protest as a Political Resource.” American Political Science Review 62:1144-1158

Logan, John R. and Harvey Molotch. 1987. Urban Fortunes: The Political Economy of Place. Berkeley: University of California Press.

Logan, John R., Rachel Bridges Whaley & Kyle Crowder. 1997. “The Character and Consequences of Growth Regimes: An Assessment of 20 Years of Research.” Urban Affairs Review 32(5): 603-630.

Maney, Gregory M. 2002. “Transnational Structures and Protest: Linking Theories and Assessing Evidence.” Ch. 2 in Globalization and Resistance, eds. Jackie Smith & Hank Johnston. Lanham, MD. Rowman & Littlfield.

Martin, Andrew W. 2008. “Resources for Success: Social Movements, Strategic Resource Allocation, and Union Organizing Outcomes.” Social Problems 55(4) 501-524. McAdam, Doug 1982. Political Process and the Development of Black Insurgency, 1930- 1970. Chicago, IL: University of Chicago Press.

McAdam, Douglas. 1983. “Tactical Innovation and the Pace of Insurgency.” American Sociological Review 48(6): 735-754.

McAdam, Doug. 1989. “The Biographical Consequences of Activism.” American Sociological Review 54: 744-760.

McAdam, Doug, Sidney Tarrow & . 2003. “Dynamics of Contention.” : Journal of Social, Cultural and Political Protest 2(1): 99-102

121

McCammon, Holly, J. 2003. “Out of the Parlors and Into the Streets: The Changing Tactical Repertoire of the U.S. Women’s Suffrage Movements.” Social Forces, 81: 787-818.

McCarthy, John D. and Mayer N. Zald. 1977. “Resource Mobilization Theory and Social Movements: A Partial Theory.” American Journal of Sociology 82(6):1212- 1241. McCarthy, John D. and Mark Wolfson. 1996. “Resource Mobilization by Local Social Movement Organizations: Agency, Strategy, and Organization in the Movement against Drinking and Driving.” American Sociological Review 61: 1070-1088. McCarthy, John D. & Clark McPhail. 2006. “Places of Protest: The Public Forum in Principle and Practice.” Mobilization 11(2): 229-247. McCarthy, John D., Clark McPhail & Jackie Smith. 1999. “Images of Protest: Dimensions of Selection Bias in Media Coverage of Washington Demonstrations, 1982 and 1991.” American Sociological Review 61: 478-499.

McNeely, Connie. L. 2012. “World Polity Theory.” Wiley-Blackwell Encyclopedia of Globalization. Meyer, David S. 2004. “Protest and Political Opportunity.” Annual Review of Sociology 30: 125-145.

Meyer, David S. and Suzanne Staggenborg. 1996. “Movements, Countermovements, and the Structure of Political Opportunity.” The American Journal of Sociology 101: 1600 – 1628.

Meyer, David S. and Sidney Tarrow. 1998. “A Movement Society: Contentious Politics for a New Century.” Pp. 1-28 in The Social Movement Society, edited by David S. Meyer and Sidney Tarrow.

Meyer, David S. and Debra C. Minkoff. 2004. “Conceptualizing Political Opportunity.” Social Forces 82: 1457-1492.

Meyer, John W, John Boli, George M. Thomas & Francisco O. Ramirez. 1997. “World Society and the Nation State.” American Journal of Sociology 103: 144-81

Michels, Robert. 1962. Political Parties. New York, NY: Collier.

Minkoff, Debra. 1999. “Bending with the Wind: Strategic Change and Adoption by Women’s and Racial Minority Organizations.” American Journal of Sociology 104: 1666-1703.

122

Molotch, Harvey & Marilyn Lester. 1974. “News as Purposive Behavior: On the Strategic Use of Routine Events, Accidents and Scandals.” American Sociological Review 39(1): 101-112.

Moore, Oliver. “Understanding postorganic fresh fruit and vegetable consumers at participatory farmers’ markets in Ireland: reflexivity, trust and social movements.” International Journal of Consumer Studies 30(5): 416-426.

Morris, Aldon D. 1981. “Black Southern Student Sit-In Movement: An Analysis of Internal Organization.” American Sociological Review 46:744-767.

Morris, Aldon D. and Cedric Herring. 1984. “Theory and Research in Social Movements: A Critical Review.” Annual Review of Political Science. 2: 137-98.

Myers, Daniel J. & Beth S. Caniglia. 2004. “All the Rioting that’s Fit to Print: Selection Effects in National Newspaper Coverage of Civil Disorders, 1968-1969.” American Sociological Review 69(4): 519-543.

Nepstad, Sharon Erickson. 2002. “Creating Transnational Solidarity: The Use of Narrative in the US-Central America .” Ch 8. In Ch.3 in Globalization and Resistance, eds. Jackie Smith & Hank Johnston. Lanham, MD. Rowman & Littlfield. Nicholls, Walter. “Place, Networks, Space: Theorizing the Geographies of Social Movements.” Transactions of the Institute of British Geographies 34(1): 78-93. Offe, Claus. 1985. “New Social Movements: Challenging the Boundaries of Institutional Politics.” Social Research 52(4): 817-868.

Olsen Jr., Mancur. 1965. The Logic of Collective Action. Cambridge, Massachusetts: Harvard University Press.

Olivery, Pamela E. & Daniel J. Meyer. 1999. “How Events Enter the Public Sphere: Conflict, Location, and Sponsorship in Local Newspaper Coverage of Public Events.” American Journal of Sociology 105(1): 38-87.

Oliver, Pamela E. & Gregory Maney. 2000. “Political Processes and Local Newspaper Coverage of Protest Events: From Selection Bias to Triadic Interactions.” American Journal of Sociology 106(2): 463-505

Perlman, J.E. 1976. The Myth of Marginality. Berkeley: University of California Press.

Peterson, Paul E. 1981. City Limits. Chicago, Il: University of Chicago Press.

123

Pickvance, Christopher G. 1999. Democratization and the Decline of Social Movements: The Effects of Regime Change on Collective Action in Eastern Europe, Southern Europe and Latin America.” Sociology 33(2): 353-372

Piven Frances Fox and Richard A. Cloward. 1977. Poor People’s Movements: Why They Succeed, How They Fail. New York: Pantheon Books.

Polletta, Francesca and James M. Jasper. 2001. “Collective Identity and Social Movements.” Annual Review of Sociology 27: 283-305. Powell, Walter W., and Paul J. DiMaggio. 1991. The New Institutionalism in Organizational Analysis. Chicago, IL: The University of Chicago Press.

Raina, Peter. 1982. Independent Social Movements in Poland. London, United Kingdom: London School of Economics.

Ranci, Costanzo. Social Vulnerability in Europe: The New Configuration of Social Risks. New York, NY: Palgrave Macmillian.

Ranci, Costanzo, Taco Brandsen & Stefania Sabatinelli. 2014. “New Social Risks and the Role of Local Welfare: An Introduction.” Ch. 1 in Social Vulnerability in European Cities: The Role of Local Welfare in Times of Crisis. Eds. Constanzo Ranci, Taco Brandsen & Stefania Sabatinelli. New York, NY: Palgrave Macmillian.

Reimann, Kim D. 2002. “Creating networks from the Outside In.” Ch 10 in Ch.3 in Globalization and Resistance, eds. Jackie Smith & Hank Johnston. Lanham, MD. Rowman & Littlfield.

Rohlinger, Deana A. 2002. “Framing the Abortion Debate: Organizational Resources, Media Strategies, and Movement-Countermovement Dynamics.” Sociological Quarterly 43(4): 479-507

Roscigno, Vincent J. and William F. Danaher. 2001. “Media and Mobilization: The Case of Radio and Southern Textile Worker Insurgency, 1929 to 1934.” American Sociological Review 66: 21-48. Roth, Andrew L. & Emilie L. Vander Haar. 2006. “Media Standing of Urban Parkland Movements: The Case of Los Angeles’ Taylor Yard, 1985-2001.” City & Community 5(2): 129-151.

Rothman, Franklin Daniel & Pamela E. Oliver. 2002. “From Local to Global: The Anti- Dam Movement in Southern Brazil, 1979-1992.” Ch 7 in Ch.3 in Globalization

124

and Resistance, eds. Jackie Smith & Hank Johnston. Lanham, MD. Rowman & Littlfield.

Ryan, Charlotte. 1991. Prime Time Activism. Boston: South End Press. Sassen, Saskia. 1990. "Economic Restructuring and the American City." Annual Review of Sociology

Sassen, Saskia. 2001. The Global City: London, New York, Tokyo. 2nd ed. Princeton: Princeton University Press

Sassen, Saskia. 2006. Cities in a World Economy. Thousand Oaks, California: Sage.

Sassen, Saskia. 2008. “Two Stops in Today’s New Global Geographies: Shaping Novel Labor Supplies and Employment Regimes.” American Behaviorial Scientist 52:457-96.

Schoene, Matthew. Forthcoming. “Friend or Foe? Media Coverage of Chicago’s Plan for Transformation.” Sociological Focus.

Scholte, Jan Aart. 1997. The Globalization of World Politics. Oxford, UK: Oxford University Press.

Smith, Jackie. 1998. “Global Civil Society: Transnational Social Movement Organizations and Social Capital.” American Behavioral Scientist 42(1): 93-107.

Smith, Jackie. 2001. “Globalizing Resistance: The Battle of Seattle and the Future of Social Movements.” Mobilization 6(1): 1-20.

Smith, Jackie. 2008. Social Movements for Global Democracy. Baltimore, MD: Johns Hopkins University Press. Smith, Jackie & Dawn Wiest. 2012. “Transnational Social Movement Organization Dataset, 1953-2003.” ICPSR33863-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2012-07-16. http://doi.org/10.3886/ICPSR33863.v1 Smith, D. & M. Timberlake. 1995. “Cities in Global Matrices: Towards Mapping the World-System’s city system.” In World Cities in a World-System, eds. P. Knox & P. Taylor.

Smith, D. 2001. Transnational Urbanism. Oxford: Blackwell.

125

Snow, David A., E. Burke Rochford, Jr., Steven K. Worden, and Robert D. Benford. 1986. “Frame Alignment Processes, Micromobilization, and Movement Participation.” American Sociological Review 51: 464-481.

Snow, David A. and Robert D. Benford. 1992. “Master Frames and Cycles of Protest.” Pp. 133-155 in Frontiers of Social Movement Theory, Aldon D. Morris and Carol M. Mueller, eds. New Haven: Yale University.

Staggenborg, Suzanne. 1988. “The Consequences of Professionalization and Formalization in the Pro-Choice Movement.” American Sociological Review. 53: 585-605.

Stern, Paul C., Thomas Dietz, Troy Abel, Gregory A. Guagnano & Linda Kalof. “A value-Belief-Norm Theory of Support for Social Movements: The Case of .” Research in Human Ecology 6(2): 81-97.

Storper, Michael and Scott, Allen. 2009. “Rethinking human capital, and urban growth” Journal of Economic Geography, Vol. 9, pp. 147-167 Tarrow, Sidney G. 1994. Power in Movement: Social Movements, Collective Action, and Politics. Cambridge: Cambridge University Press.

Taylor, Verta, Katrina Kimport, Nella Van Dyke and Ellen Andersen. 2009. "Culture and Mobilization: Tactical Repertoires, Same-Sex Weddings, and the Impact on Gay Activism." American Sociological Review 74: 865-890. Taylor, Peter J. 2004. “The New Geography of Global Civil Society: NGOs in the World City Network.” Globalizations: 1(2): 265-277 Taylor, Peter J., Pengfei Ni, Ben Derudder, Michael Hoyler, Jin Huang & Frank Witlox. 2011. Global Urban Analysis: A Survey of Cities in Globalization. London: Earthscan.

Taylor-Gooby, Peter. 2004. New Risks, New Welfare: The Transformation of the European Welfare State. Oxford, UK: Oxford University Press.

Timberlake, Michael. 1985. Urbanization in the World Economy. New York: Academic Press. Thörn, Hakan. 2007. “Social Movements, the Media and the Emergence of a Global Public Sphere.” Current Sociology 55(6): 896-918.

Teitz, S. 1999. “America’s Best Newspapers.” Columbia Journalism Review 38(4):14.

Tuchman, Gaye. 1972. “Objectivity as Strategic Ritual.” American Journal of Sociology 77: 660-79 126

Chou, D, M. Inoue, C. Mathers, A.B. Moller, M. Oestergaard, L. Say, S. Mills, E. Suzuki & J. Wilmoth. 2012. “Trends in Maternal Mortality: 1990 to 2010: WHO, UNICEF, UNFPA, and The World Bank Estimates.” United Nations Fund for Population Activities.

Walton, John. 1998. “Urban Conflict and Social Movements in Poor Countries: Theory and Evidence of Collective Action.” International Journal of Urban and Regional Research 22(3): 460-81

Weimann, G. & G. Winn. 1994. Theater of Terror: Mass Media and International . White Plains, NY: Longman.

Williamson, Jeffrey G. 1981. “Urban Disamenitites, Dark Satanic Mills, and the British Standard of Living Debate.” Journal of Economic History 41: 75-83.

Zald, Mayer N. and John D. McCarthy. 2002. “The Resource Mobilization Research Program: Progress, Challenge, and Transformation.” (Eds.) Joseph Berger and Morris Zelditch, Jr. New Directions in Contemporary Sociological Theory. Lanham, MD: Rowman & Littlefield.

Zukin, Sharon. 2011. “Reconstructing the Authenticity of Place.” Theory & Society 40: 161-165.

127

APPENDIX A: TABLES AND FIGURES FOR CHAPTER 2

Table A.1: List of Social Movement Industries in Sample of Transnational Social Movement Organization Dataset, 1993-2003 SOCIAL MOVEMENT INDUSTRY Freq. Percent Various Human Rights 114 16.96 Environment 91 13.54 Women's Rights 68 10.12 Global Justice 67 9.97 Rights and Peace 49 7.29 Peace 46 6.85 Freedom of Expression 30 4.46 Development 27 4.02 Esperanto 17 2.53 16 2.38 Women and Development 16 2.38 Children's Rights 14 2.08 Rights and Democracy 14 2.08 International Integration 13 1.93 Peace and Democracy 11 1.64 Democracy 10 1.49 Other 9 1.34 Rights and Environment 9 1.34 Rights, Democracy, Peace 8 1.19 Peace and Development 7 1.04 Animal Rights 6 0.89 Rights, Peace, Development 6 0.89 Solidarity 5 0.74 Misc. Conservative 4 0.6 Alternatives to Capitalism 4 0.6 Public Health 4 0.6 Rights and Development 4 0.6 Ethnic Unity 2 0.3 Rights, Development, Democracy 1 0.15 Total 672 100

128

Table A.2: Frequency and Percent of Global Cities Housing Social Movement Organizations in Sample of TSMOD, 1993-2003. Global City Frequency Percent Global City Frequency Percent Brussels 135 20.09 Copenhagen 3 0.45 London 74 11.01 Frankfurt 3 0.45 Paris 60 8.93 Lagos 3 0.45 New York City 52 7.74 Lima 3 0.45 Geneva 48 7.14 Madrid 3 0.45 Washington D.C. 35 5.21 São Paulo 3 0.45 Rome 22 3.27 Seattle 3 0.45 Nairobi 20 2.98 Atlanta 2 0.3 Amsterdam 18 2.68 Bangalore 2 0.3 San Francisco 14 2.08 Dublin 2 0.3 Bangkok 12 1.79 Hamburg 2 0.3 Delhi 11 1.64 Miami 2 0.3 Stockholm 11 1.64 Philadelphia 2 0.3 Berlin 8 1.19 Seoul 2 0.3 Athens 7 1.04 Singapore 2 0.3 Barcelona 7 1.04 Sydney 2 0.3 Cairo 7 1.04 Taipei 2 0.3 Toronto 7 1.04 Alexandria 1 0.15 Santiago 6 0.89 Ankara 1 0.15 Tel Aviv 6 0.89 Boston 1 0.15 Manila 5 0.74 Bucharest 1 0.15 Moscow 5 0.74 Cape Town 1 0.15 Tokyo 5 0.74 Hong Kong 1 0.15 Budapest 4 0.6 Istanbul 1 0.15 Buenos Aires 4 0.6 Johannesburg 1 0.15 Dhaka 4 0.6 Krakow 1 0.15 Jakarta 4 0.6 Melbourne 1 0.15 Kuala Lumpur 4 0.6 Milan 1 0.15 Mexico City 4 0.6 Mumbai 1 0.15 Montreal 4 0.6 Panama City 1 0.15 Oslo 4 0.6 Riyadh 1 0.15 Prague 4 0.6 Vienna 1 0.15 Bogotá 3 0.45 Warsaw 1 0.15 Zurich 1 0.15 Total 672 100

129

Table A.3: Descriptive Statistics of Dependent and Independent Variables Variable Obs Mean s.d. Min Max Source Organizational Age 672 25.5 23.48 1 164 TSMOD Organizational Viability 672 0.77 0.42 0 1 TSMOD IGO Count 672 2.33 3.13 0 22 TSMOD NGO Count 672 4.18 6.54 0 55 TSMOD Professionalization 672 0.56 0.5 0 1 TSMOD Global Competitiveness 672 59.56 10.01 27.6 71.4 The Economist Ln(City GDP) 671 10.6 0.63 7.5 11.59 World Bank Ln(Population Density) 671 8.53 0.72 5.92 10.71 World Bank Inequality 672 35.38 7.35 23 65 World Bank Poverty 672 15.76 9.86 0 70 CIA Human Rights Index 672 1.56 1.38 0 5.42 UN Corruption Index 672 67.75 17.23 21 93 Transparency Int.

130

Table A.4: One-Way ANOVA Regression & Logistic Regression with Random Effects Models on Dependent Variables Model 1: Model 2: Organizational Organizational Viability Age Coef. 1.218*** 25.700*** Std. Error 0.101 1.614

RE Parameter 0.120 5.798 Std. Error 0.416 1.459

LR Test 0.02 15.29*** N 672 519 Groups 67 64

131

Table A.5: One-Level Logistic Regression Results of Organizational, Urban and National Variables on Organizational Viability Model 1 Model 2 Model 3 b b b Odds (s.e.) Odds (s.e.) Odds (s.e.) Organizational Variables IGO Count 0.176* (0.091) 0.168* (0.091) 0.167* (0.092) 1.193 1.184 1.182 NGO Count 0.022 (0.023 ) 0.022 (0.024 ) 0.021 (0.023 ) 1.022 1.022 1.021 Professionalization (a) 0.601** (0.19 ) 0.642** (0.189 ) 0.656** (0.19 ) 1.834 1.901 1.928 Urban Variables Global Competitiveness -- 0.020* (0.01) 0.019 (0.018) 1.02 1.02 Ln(City GDP) -- -0.136 (0.159 ) -0.058 (0.242 ) 0.873 0.944 Ln(Population Density) -- 0.153 (0.124 ) 0.206 (0.148 ) 1.166 1.229 National Variables Inequality -- -- -0.023 (0.023) 0.977 Poverty -- -- -0.014 (0.014 ) 0.987 Human Rights Index -- -- 0.183 (0.144 ) 1.201 Corruption Index -- -- 0.002 (0.009 ) 1.001 Constant 0.5** (0.16) -0.542 (2.336) -1.119 (3.1) N 672 671 671 Clusters 67 66 66 Standard errors presented in parentheses. Standard errors adjusted for clustering. Odds Ratios presented underneath coefficients. †p<0.10 *p<0.05 **p<0.01 ***p<0.001 a: Reference group is Non-professional structure

132

Table A.6: Multilevel Mixed Effects Regression Results of Independent & Control Variables on Organizational Age Model 4 Model 5 Model 6 b b B (s.e) (s.e) (s.e) Organizational Variables IGO Count 2.898*** 2.758*** 2.702*** (0.328) (0.326) (0.327) NGO Count -0.245† -0.278† -0.275† (0.152) (0.151) (0.150) Professionalization 2.752 2.878 2.568 (2.068) (2.048) (2.054) Urban Variables Global Competitiveness -- 0.281 * 0.454 * (0.130) (0.197) Ln(City GDP) -- 1.694 5.072* (1.759) (2.647) Ln(Population Density) -- -0.612 -1.102 (1.702) (1.783) National Variables Inequality -- -- -0.296 (0.246) Poverty -- -- 0.049 (0.172) Human Rights Index -- -- 1.330 (1.488) Corruption Index -- -- -0.191* (0.101) Constant 18.844 -6.811 -29.683 Variation(_cons) 17.697 5.61 0.594 Variation(Residual) 512.989 513.671 514.578 N 519 519 519 Groups 64 64 64

Standard errors presented in parentheses. †p<0.10 *p<0.05 **p<0.01 ***p<0.001 a: Reference group is Non-professional structure

133

APPENDIX B: TABLES AND FIGURES FOR CHAPTER THREE

Table B.1: Descriptive Statistics by Country N Groups Belgium 1,629 11 Switzerland 1,412 7 Czech Rep. 2,181 14 Germany 2,829 16 Denmark 1,497 5 Estonia 1,632 5 Spain 1,771 18 Finland 1,782 18 France 1,690 21 Great Britain 2,106 12 Greece 2,624 13 Hungary 1,351 20 Ireland 2,364 8 Netherlands 1,707 12 Norway 1,376 6 Poland 1,590 16 Portugal 1,896 5 Sweden 1,368 21 Slovenia 1,287 12 Slovakia 1,657 8 Total 35,749 248

134

Table B.2: Descriptive Statistics & Sources of All Dependent, Independent and Control Variables. Variable Obs Mean S.d. Min Max Source Contact Politician 35749 0.14 0.35 0 1 ESS Work for Party 35749 0.04 0.19 0 1 ESS Wore Badge 35749 0.07 0.26 0 1 ESS Sign Petition 35749 0.21 0.41 0 1 ESS Boycott Product 35749 0.16 0.37 0 1 ESS Demonstration 35749 0.06 0.24 0 1 ESS Trust in Others 35749 5.07 2.40 0 10 ESS Satisfaction (econ.) 35749 4.12 2.55 0 10 ESS Satisfaction (gov.) 35749 3.91 2.45 0 10 ESS Satisfaction (dem.) 35749 5.02 2.51 0 10 ESS Urban 35749 0.32 0.47 0 1 ESS Middle Income 35749 0.33 0.47 0 1 ESS Upper Income 35749 0.43 0.49 0 1 ESS Sex 35749 0.47 0.50 0 1 ESS Age 35749 48.08 18.33 15 101 ESS Immigrant Status 35749 0.08 0.28 0 1 ESS Education (years) 35749 12.50 4.14 0 50 ESS Unemployment 35749 8.83 4.20 2.1 26.2 ESS Population Density (log) 35749 4.98 1.26 0.69 8.79 ESS Regional GDP (log) 35749 35.86 17.50 12.26 84.59 ESS New Democracy 35749 0.27 0.44 0 1 World Polity

135

Table B.3: One Way ANOVA with Random Effects Models for All Dependent Variables.

Model 1: Model 2: Model 3: Model 4: Model 5: Model 6: Contact Work for Wore Sign Boycott Politician Party Badge Petition Product Demonstration Coef. -1.875*** -3.387*** -2.766*** -1.550*** -1.889*** -3.100*** Std. Error 0.033 0.043 0.061 0.065 0.066 0.074

RE Parameter 0.426 0.423 0.827 0.958 0.976 1.000 Std. Error 0.028 0.044 0.050 0.054 0.052 0.061

LR Test 507.52*** 77.06*** 1346.96*** 2724.20*** 2787.36*** 1167.38*** N 35749 35749 35749 35749 35749 35749 Groups 248 248 248 248 248 248

136

Table B.4: Percent of ESS Respondents at National Level Participating in Each Type of Social Movement Activism, 2010. Contact Work Wore Boycott All Country Politician Party Badge Petition Product Protest Activity Belgium 11.85 4.79 7.12 20.87 9.27 6.51 35.42 Switzerland 15.93 6.09 4.82 32.72 28.19 3.9 50.64 Czech Rep. 14.72 2.52 3.67 16.6 10.59 4.54 31.96 Germany 16.22 4.21 5.02 30.72 28.42 8.7 52.67 Denmark 18.24 4.21 8.48 30.13 22.51 7.75 52.51 Estonia 14.46 3.74 4.96 8.27 9.56 2.02 27.94 Spain 14 7.34 10.45 26.76 11.74 18.69 42.07 Finland 21.16 3.54 17.34 28.4 33.89 1.4 60.83 France 14.08 3.43 10.12 29.29 29.23 17.46 53.79 Great Britain 15.67 1.9 6.55 29.82 21.42 2.61 44.49 Greece 8.69 2.9 3.13 5.11 11.93 10.44 23.06 Hungary 12.21 2.59 2.74 3.03 6.44 2.66 18.21 Ireland 14.51 3.13 5.03 15.14 9.26 6.77 28.89 Netherlands 17.81 3.75 3.69 26.42 10.02 2.4 41.42 Norway 22.6 5.74 28.27 36.34 19.91 9.81 63.81 Poland 9.25 2.45 5.79 11.95 5.53 2.2 21.32 Portugal 5.01 1.85 2.06 5.91 2.11 3.11 11.87 Sweden 16.74 3.8 19.88 38.45 36.48 5.26 63.23 Slovenia 9.25 2.8 2.87 8.39 5.75 2.18 20.59 Slovakia 9.72 2.05 4.16 21.24 6.82 1.93 29.03 Total 13.99 3.57 7.32 20.96 15.98 6.25 38.29

137

Table B.5: Multilevel Mixed-Effects Logistic Regression Results of All Variables on Contacting Politician, Working for Party & Wearing a Protest Badge Contact Pol. Work Party Badge b (s.e.) b (s.e.) b (s.e.) Grievances Satisfaction (Econ.) 0.003 (0.010) 0.037* (0.017) 0.009 (0.013) 1.003 1.038 1.009 Satisfaction (Govt.) 0.001 (0.009) -0.028 (0.017) -0.021 (0.013) 1.001 0.972 0.979 Satisfaction (Dem.) 0.004 (0.009) 0.018 (0.016) 0.028* (0.012) 1.004 1.018 1.028 Unemployment -0.001 (0.008) 0.035** (0.011) 0.019 (0.013) 0.999 1.036 1.019 Resources General Trust 0.014 (0.008) 0.051*** (0.014) 0.079*** (0.011) 1.014 1.052 1.083 Middle Income (b) 0.123** (0.051) 0.060 (0.084) -0.006 (0.059) 1.131 1.061 0.994 High Income (b) 0.219*** (0.056) 0.169* (0.083) -0.010 (0.059) 1.245 1.184 0.990 Regional GDP 0.010** (0.003) 0.007 (0.004) 0.020*** (0.005) 1.010 1.007 1.020 Education 0.103*** (0.005) 0.100*** (0.007) 0.071*** (0.006) 1.108 1.106 1.073 Urbanization Urban (a) -0.234*** (0.031) -0.091 (0.069) 0.143** (0.051) 0.791 0.913 1.153 Population Density -0.069** (0.021) -0.047 (0.033) -0.174*** (0.037) 0.934 0.954 0.840 Controls Sex (c) 0.354*** (0.045) 0.616*** (0.060) -0.101** (0.043) 1.424 1.851 0.904 Age 0.008*** (0.001) 0.010*** (0.002) -0.011*** (0.001) 1.008 1.010 0.989 Immigrant Status (d) -0.454*** (0.042) -0.378** (0.119) -0.481*** (0.088) 0.635 0.685 0.618 New Democracy (e) -0.065 (0.090) -0.169 (0.141) -0.255 (0.166) 0.937 0.844 0.775 Constant -3.88*** (0.026) -6.26*** (0.361) -3.58*** (0.395) RE Parameter 0.301 (0.015) 0.12 (0.032) 0.317 (0.045) N 35,749 35,749 35,749 Groups 248 248 248 *p<.05 **(p<.01) ***(p<.001). Standard Errors presented in parentheses. a: reference is rural. b: reference is lower income. c: reference is female. d: reference group is native born citizen. e: reference group is older democracy.

138

Table B.6: Multilevel Mixed-Effects Logistic Regression Results of All Variables on Signing a Petition, Boycotting a Product and Participating in a Demonstration Petition Boycott Demonstration b (s.e.) b (s.e.) b (s.e.) Grievances Satisfaction (Econ.) 0.000 (0.009) -0.019* (0.010) -0.043** (0.014) 1.000 0.981 0.958 Satisfaction (Govt.) -0.054*** (0.009) -0.071*** (0.010) -0.071*** (0.014) 0.948 0.932 0.931 Satisfaction (Dem.) 0.002 (0.008) -0.029** (0.009) -0.020 (0.012) 1.002 0.972 0.980 Unemployment 0.016 (0.014) -0.008 (0.016) 0.060*** (0.017) 1.016 0.992 1.062 Resources General Trust 0.066*** (0.007) 0.050*** (0.008) 0.100*** (0.012) 1.068 1.051 1.106 Middle Income (b) 0.293*** (0.040) 0.132** (0.044) 0.049 (0.065) 1.341 1.142 1.050 Upper Income (b) 0.196*** (0.040) 0.200*** (0.044) 0.046 (0.065) 1.217 1.222 1.047 Regional GDP 0.034*** (0.005) 0.024*** (0.006) 0.006 (0.006) 1.035 1.025 1.006 Education 0.098*** (0.004) 0.117*** (0.004) 0.087*** (0.006) 1.103 1.125 1.091 Urbanization Urban (a) 0.139*** (0.034) 0.201*** (0.038) 0.320*** (0.055) 1.149 1.222 1.377 Population Density 0.061 (0.039) -0.104* (0.044) 0.086*** (0.049) 1.062 0.901 1.090 Controls Sex (c) -0.132*** (0.028) -0.069* (0.031) 0.265*** (0.046) 0.877 0.934 1.304 Age -0.008*** (0.001) 0.002 (0.001) -0.016*** (0.001) 0.992 1.002 0.984 Immigrant Status (d) -0.588*** (0.056) -0.334*** (0.059) -0.309*** (0.087) 0.555 0.716 0.734 New Democracy (e) -0.273 (0.175) -0.669** (0.201) -0.991*** (0.217) 0.761 0.512 0.371 Constant -4.192*** (0.407) -3.458*** (0.465) -4.584*** (0.503) RE Parameters 0.459 (0.055) 0.615 (0.069) 0.615 (0.084) N 35,749 35,749 35,749 Groups 248 248 248 *p<.05 **(p<.01) ***(p<.001). Standard Errors presented in parentheses. a: reference is rural. b: reference is lower class. c: reference is female. d: reference group is native born citizens. e: reference group is older democracy

139

APPENDIX C: TABLES AND FIGURES FOR CHAPTER 4

Table C.1: Number of Quotes by Institutional Affiliation in The New York Times, January 2010- March 2013 Affiliation Freq. Percent Cumulative Institutionalized Actors Politician 935 20.8% 20.8% Government Official 698 15.5% 36.3% Banker/Finance/IMF 870 19.4% 55.7% ECB Official 106 2.4% 58% Private Business 157 3.5% 61.5% Grassroots Actors Citizen 487 10.8% 72.4% Citizen Leader 118 2.6% 75% Other Actors Author 272 6% 81% Expert 761 16.9% 100% Other 92 2.1% 83.1% Total 4,496 100%

140

Table C.2: Sentiment of Quotes by Institutional Affiliation in New York Times, January 2010-March 2013. Pro- Anti- Neutral/No Affiliation Austerity Austerity Stand Total Institutionalized Actors Politician 362 225 348 935 38.7% 24.1% 37.2% 100% Government Official 159 106 433 698 22.8% 15.2% 62% 100% Banker/Finance/IMF 187 259 424 870 21.4% 29.8% 48.7% 100% ECB Official 55 6 45 106 51.9% 5.7% 42.5% 100% Private Business 33 81 43 157 21% 51.6% 27.4% 100% Grassroots Actors Citizen 45 245 197 487 9.2% 50.3% 40.5% 100% Citizen Leader 6 67 45 118 5.1% 56.8% 38.1% 100% Other Actors Author 18 128 126 272 6.6% 47% 46.4% 100% Expert 77 218 466 92 10.1% 28.7% 61.2% 100% Other 11 44 37 761 12% 47.8% 40.2% 100% Total 952 1,376 2,160 4,496 100% 100% 100% 100%

141

Figure C.1: Weekly Count of Total Quotes from Sample of New York Times Articles, January 2010-March 2013 Total Weekly Quotes 120

100

80

60

40

20

0

Jan '10 Jan Jan'12 Jan'13 Jan'11

142

Figure C.2: Weekly Count of Pro-Austerity and Anti-Austerity Quotes from Sample of New York Times Articles, January 2010-March 2013. 60 Pro-Austerity & Anti-Austerity Quotes 50

40

30

20 NumberofQuotes

10

0

Jan'10 Jan'11 Jan'12 Jan'13

Pro Anti

143