Cities, Business, and the Politics of Urban Violence in Latin America

By Eduardo Moncada

This book analyzes and explains the ways in which major developing world cities respond to the challenge of urban violence. The study shows how the political projects that cities launch to confront urban violence are shaped by the interaction between urban political economies and patterns of armed territorial control. It introduces business as a pivotal actor in the politics of urban violence, and argues that how business is organized within cities and its linkages to local governments impacts whether or not business supports or subverts state efforts to stem and prevent urban violence. A focus on city mayors finds that the degree to which politicians rely upon clientelism to secure and maintain power influences whether they favor responses to violence that perpetuate or weaken local political exclusion. The book builds a new typology of patterns of armed territorial control within cities, and shows that each poses unique challenges and opportunities for confronting urban violence. The study develops sub-national comparative analyses of puzzling variation in the institutional outcomes of the politics of urban violence across Colombia's three principal cities—Medellin, Cali, and Bogota—and over time within each. The book's main findings contribute to research on violence, crime, citizen security, urban development, and comparative political economy. The analysis demonstrates that the politics of urban violence is a powerful new lens on the broader question of who governs in major developing world cities.

• Language: English
• Publisher: Stanford University Press
• Release date: Jan 6, 2016
• ISBN: 9780804796903

Book preview

Stanford University Press

Stanford, California

© 2016 by the Board of Trustees of the Leland Stanford Junior University.

All rights reserved.

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press.

Printed in the United States of America on acid-free, archival-quality paper

Library of Congress Cataloging-in-Publication Data

Moncada, Eduardo, author.

Cities, business, and the politics of urban violence in Latin America / Eduardo Moncada.

pages  cm.

Includes bibliographical references and index.

ISBN 978-0-8047-9417-6 (cloth: alk. paper)

1. Urban violence—Colombia—Case studies.   2. Municipal government—Colombia—Case studies.   3. Business and politics—Colombia—Case studies.   4. Patron and client—Colombia—Case studies.   I. Title.

HN310.Z9V5523    2016

303.609861—dc23

2015010586

ISBN 978-0-8047-9690-3 (electronic)

Typeset at Stanford University Press in 10/14 Minion

Cities, Business, and the Politics of Urban Violence in Latin America

Eduardo Moncada

Stanford University Press

Stanford, California

CONTENTS

Acknowledgments

Abbreviations

1. Rethinking the Politics of Urban Violence

2. Parties, Clientelism, and Violence: Exclusionary Political Order in Colombia

3. Medellin: Reshaping Political Order and Criminal Coexistence

4. Cali: The Derailment of a Pioneering Participatory Project

5. Bogota: Building and Branding a Global City

6. The Politics of Urban Violence: Comparisons and Next Steps

Notes

References

Index

ACKNOWLEDGMENTS

This project benefited from the feedback and support that I was fortunate to receive from many individuals and institutions. First and foremost, I hope that my parents view this book as a small symbol of my immeasurable gratitude for the everyday sacrifices they made as part of raising me in a country separated by geography, language, and culture from the countries they knew as their homes. My wife, Angie, provided encouragement, patience, humor, and not so gentle reminders of what matters. And along the way Elisio and Nacine arrived, surprised and amazed me, and they have not stopped ever since.

At Brown University, I am indebted to Richard Snyder for providing me with an abundance of his time, guidance, and critical but always insightful commentary. His enthusiasm for big important questions is contagious. Melani Cammett, Pauline Jones-Luong, and Patrick Heller each contributed to the research and arguments presented here. All four individuals showed me that what scholars do matters in many different ways.

Many people provided feedback at critical moments in the project’s evolution. Enrique Desmond Arias, John Bailey, Kent Eaton, and Benjamin Goldfrank each received an email from me with draft chapters attached, and before we even had a chance to meet in person, each wrote back with detailed and helpful comments and critiques. They did so again several years later. Responding to the questions and comments posed by the manuscript’s anonymous reviewers was both challenging and rewarding, and made this a stronger book.

Other individuals that offered insightful comments and questions on the project at different points in its development include Jorge I. Domínguez, John Gershman, Matthew Ingram, Stathis Kalyvas, Rogan Kersh, Peter Kingstone, James Mahoney, Barbara Stallings, Judith Tendler, Mark Ungar, and Elisabeth Jean Wood. A number of colleagues provided much needed feedback and friendship, including Jorge Antonio Alves, Kelly Bay, Erin Beck, Mila Dragojevic, Angélica Durán-Martínez, Ayo Jegede, Jeremy Johnson, Cecilia Perla, and Ravi Perry. I was fortunate to present parts of the project during workshops and conferences in front of thoughtful and tough audiences at Brown University, City University of New York, Columbia University, Harvard University, New York University, the Universidad de Chile, the Woodrow Wilson International Center for Scholars, and Yale University.

Funding and support from several institutions was crucial for carrying out this project. I am grateful for the generous funding for field research provided by the American Society of Criminology, the Smith Richardson Foundation, the Ford Foundation, the Fulbright-Hays Doctoral Dissertation Research Abroad Program, and the Drugs, Security, and Democracy Program, which is funded by the Open Society Foundation (OSF) and is a partnership between the OSF, the Social Science Research Council, the Universidad de los Andes, and the Centro de Investigación y Docencia Económica. Support from the Harry Frank Guggenheim Foundation enabled me to synthesize my findings while spending a productive year at Yale University’s Order, Conflict, and Violence Program. New York University’s Postdoctoral and Transition Program for Academic Diversity supported my immensely enriching stay at the Wagner Graduate School of Public Service. The Department of Political Science at Rutgers University, Newark, provided a supportive environment within which to finalize the manuscript. And a grant from the Research Council at Rutgers University made it possible for me to visit Mexico and start the process of extending the argument. I also thank Geoffrey Burn at Stanford University Press for believing in this project, and James Holt for guiding the manuscript through the production process. Excerpts of the book were first published as articles in Studies in Comparative International Development and the Journal of Comparative Politics. I thank the editors of those journals and the anonymous reviewers for the feedback they provided. And of course, I must acknowledge my immense gratitude to the many individuals in Colombia and Mexico that agreed to be interviewed as part of my research—they provided me with their time and confided in me with their memories, insights, fears, and hopes.

ABBREVIATIONS

1

Rethinking the Politics of Urban Violence

Medellin is a new city. The King of Spain came to Medellin, the Organization of American States held its annual meeting here, people are back on the streets because they feel safe, new businesses have emerged and the city is a magnet for international business, four and five star hotels are under construction, foreign capital is back, and the homicide rate has dropped significantly, which is how safety is measured abroad.

—Fabio Orlando Acevedo, aka Don Efe (former paramilitary leader in the Bloque Cacique Nutibara of Medellin, Colombia, January 9, 2009, author interview)

This statement was how Don Efe responded when I asked him how it was that the city of Medellin, which in the early 1990s was considered the most violent in the world, had become a global model of urban governance. The interview took place six years after Don Efe and other members of the urban paramilitary unit known as Bloque Cacique Nutibara (BCN) demobilized in Medellin, and six months before he was arrested and jailed in the same city on charges of coordinating extortion rackets, drug trafficking, forced displacement, illegal capture of private lands, and several homicides. And although I was not completely aware of it while I was interviewing Don Efe, his statement alluded to some of the basic elements of the main argument that I develop in this book to explain what shapes the nature and trajectory of the political projects that cities launch in response to urban violence.

During the late twentieth century Medellin was the epicenter of a global illicit narcotics trade coordinated by the Medellin drug cartel headed by the infamous Pablo Escobar. Efforts by the Colombian government to dismantle the drug trafficking organization (DTO) and extradite its leaders to the United States in the late 1980s prompted the DTO to launch a war against the state that unfolded partly on Medellin’s streets. The city’s homicide rate peaked in 1991 at an astonishing 381 per 100,000 people. During this conflict a novel political project was launched to stem urban violence in Medellin by strengthening violence control measures and increasing socioeconomic investment to peripheral parts of the city where the majority of the victims and perpetrators of the violence resided.¹ But more broadly, the project explicitly sought to increase the participation of socioeconomically marginalized communities in local governance via new participatory institutions and greater inclusion in the formal political spaces long controlled by a local ruling class of political and economic elites. The architects of this ambitious participatory political project argued that deconstructing the historically exclusionary nature of local politics would reduce violence in the long term and, more broadly, deepen local democratic rule. The participatory project, however, was derailed. The city soon reverted back to its historical struggle to maintain order through a mix of coercive force by state and nonstate actors, fleeting and fragile peace pacts with armed actors, and robust networks of clientelist exchange coordinated by local traditional parties—what I call reactionary political projects.

A decade later violence in Medellin again escalated when leftist insurgents and right-wing paramilitaries from the country’s decades-long civil war entered peripheral neighborhoods where lucrative illicit markets and hundreds of small but powerful armed actors were based. The city once more attempted a participatory political project that focused in part on the peripheries that had been the target of the earlier failed participatory project. This time, however, the participatory project was sustained, and soon became the foundation for Medellin’s dramatic rebranding from the world’s most violent city to a dynamic emerging global city. Spaces for political participation increased at multiple scales, including community involvement in the development of major urban infrastructural projects and neighborhood-level participatory budgeting (Brand and Dávila 2011; Castro and Echeverri 2011). Between 2002 and 2009, foreign direct investment (FDI) into Medellin increased tenfold,² and the city soon joined the ranks of the top Latin American metropolises for doing business.³ Medellin is today widely touted as a model of how urban governance can be reshaped to address violence in major developing world cities in ways that increase local economic competitiveness and foster inclusionary local governance. And though this book shows that the construction and nature of the Medellin miracle is far more complex than either its political architects or international donors concede, as in the other cases analyzed in this book, it is emblematic of the significant ways in which the politics of urban violence can restructure urban governance in complex and far-reaching ways. Hence it is important that we develop our understanding of the factors that shape how different cities respond to violence. This is the central objective of this book.

PREVIEWING THE ARGUMENT

Multiple scales of power overlap, intersect, and interact within cities (Davis 2011; Sellers 2002). This confluence imparts both substance and political form to the political projects that cities deploy in response to urban violence. And while power within cities is deployed to advance the interests of formal political authorities, it is also exercised to generate gains among the many informal and alternative forces that operate within and outside of cities. Because cities are entangled in complex crosscurrents of political power rooted in both jurisdictional and nonjurisdictional authority, explaining the nature and trajectory of political projects in response to urban violence requires a multilevel analytic framework. Such a framework must unpack political dynamics within the formal political spaces that govern cities while also theorizing the conditions under which the efforts of actors operating at other territorial and institutional scales are able to influence the institutional outcomes of the politics of urban violence. Thus to account for why cities pursue distinct political projects in response to urban violence, and why these projects can sometimes assume forms quite different from those originally envisioned by the political forces that develop them, I build a novel analytic framework that focuses on interaction between urban political economies and patterns of armed territorial control.

By urban political economy I am referring to the nature of the relationship between city governments headed by popularly elected mayors and urban business,⁴ two actors that have long constituted the local ruling class within major developing world cities. These relationships are shaped by the interaction between the individual preferences of the public and private sectors and the formal and informal linkages that join the two. The resulting relations can be collaborative, conflictive, or mired in a state of disengagement. The nature of these relations weighs heavily on the degree to which other actors within or outside of a city are able to influence the nature and trajectory of political projects in response to violence, or what I refer to as the scope of the politics of urban violence.

Patterns of armed territorial control vary in their coordination of criminal leadership and the level of lethal violence. Criminal coordination refers to the degree to which the regulation of illicit markets within peripheral urban territories is concentrated in a single or small group of armed actors (high coordination) or is dispersed and disconnected across an array of competing armed actors (low coordination). Violence in this book refers to the use or threat of physical force to inflict bodily or psychological harm against an individual, community, or social group (see WHO 2002). By level of lethal violence, I am referring to the most visible form of violence: homicide. The intersection between these two dimensions represents distinct types of territorial control that pose different challenges and opportunities for actors seeking to influence the trajectory of political projects in response to urban violence. Taken together, a focus on the nature of urban political economies and patterns of armed territorial control strengthens our understanding of the politics of urban violence, which scholars and policy-makers alike consider a major challenge to the construction and deepening of democracy in Latin America (Koonings and Kruijt 2007; Moncada 2013a; World Bank 2011) and other parts of the developing world, including Africa (LeBas 2013), the Caribbean (Jaffe 2013), and South Asia (Varshney 2002; Weinstein 2013).

WHY STUDY THE POLITICS OF URBAN VIOLENCE?

Urban violence disproportionally impacts the most vulnerable and marginalized segments of society that are the focus of international development efforts (Moser and McIlwaine 2006; Winton 2004). Violence exacerbates poverty and worsens inequality, which is positively associated with violence (Fajnzylber et al. 2002). The result is a vicious and deadly circle that inhibits both social mobility and productivity. And while cities are increasingly seen as the engines of economic growth in the developing world, urban violence constrains growth through multiple direct and indirect channels, ranging from the redirection of scarce public resources into efforts to stem and address the consequences of violence to the negative effects of violence on investor confidence (Muggah 2012, 36). Frustration and fear in the face of intense violence and heightened feelings of insecurity can contribute to the emergence of vigilantism and strategies of popular justice as ways to establish order (Godoy 2006; Goldstein 2004).⁵ These same sentiments can reshape the physical geography of a city as wealthier segments of society retreat to privately guarded compounds that foster self-segregation along class lines (Caldeira 2000). Finally, violence can erode the legitimacy of state institutions, including democratic political regimes.⁶

More than half of the world’s population now resides in cities. Latin America is today the world’s most urbanized region, with over 80 percent of its population residing in cities. It is also the world’s most violent region, accounting for more than one-third of the world’s homicides in 2012 (UNODC 2013). And it is in the region’s large cities where citizens are most vulnerable to becoming the victims of criminal violence (Gaviria and Pagés 2002). Both emerging and growing cities thus confront the challenge of intense violence. A key focus in this book is unpacking the political conditions that shape and result from the varying ways in which cities respond to this violence after inheriting a range of formal political powers and resources delivered by the wave of decentralization that washed over the developing world in the late twentieth century (Dickovick 2011). These institutional outcomes in the politics of urban violence provide a powerful new lens on the broader question of who governs in major developing world cities.

Finally, it is important to note that whereas urban politics is an important vein of research within the discipline of political science (see, for example, Dahl 1961; Stone 1989), the study of urban politics within political science has long been the domain of scholars focused on cities located in the Global North. Many of our theories and concepts of urban politics are therefore rooted in studies of city politics in Atlanta, Chicago, and Detroit, among others. The institutional and socioeconomic contexts of these cities differ in important ways from those of Bogota, Johannesburg, Rio de Janeiro, and other cities in the Global South where the pace of urbanization has far exceeded that of North America and Europe (Cohen 2004). Thus an overarching aim of this book is to underscore the need for greater dialogue among scholars of urban politics whose work center on distinct regional settings, but who may be analyzing similar issues, including microlevel political relations, the forces that mediate access to public goods and services, dynamics of local democracy, and challenges to development.

The remainder of this chapter is divided into three sections. In the next section I discuss how existing scholarly approaches to the study of the politics of criminal violence provide limited analytic traction for explaining the institutional outcomes of the politics of urban violence. In the third section I build the new analytic framework to explain the institutional outcomes of the politics of urban violence using a focus on urban political economies and patterns of armed territorial control. I conclude in the fourth section by previewing the subsequent empirical chapters in which I use this analytic framework to develop controlled-comparative analyses of the politics of urban violence in Colombia.

RECONCILING MACRO- AND MICROLEVEL APPROACHES TO THE POLITICS OF CRIMINAL VIOLENCE

Existing scholarly treatments of the politics of criminal violence fall under one of two overarching analytic approaches. The first is a macrolevel approach that looks to the national level to explain patterns of violence and state efforts to establish order. In contrast, scholars using a microlevel approach develop nuanced accounts of the dynamics and consequences of violence at the ground level most proximate to where violence unfolds. Both macro- and microlevel approaches provide important insights into patterns of criminal violence, the actors that carry it out, and the ways in which states respond. Yet, neither approach can satisfactorily account for the nature and trajectory of the political projects that cities launch in response to violence. Doing so requires an approach that takes the city as the primary unit of analysis but that also theorizes how and when forces situated above or beneath the city level influence the institutional outcomes of the politics of urban violence.⁷

Scholars working from a macrolevel approach explain cross-national patterns of criminal violence by analyzing where countries reside on the production chains of global illicit economies. Changes in cross-national patterns of violence are attributed to geographic shifts in the production sites and trafficking routes associated with illicit markets, often prompted by state efforts to disrupt these processes in one set of countries that, in turn, displaces the processes to another set of countries.⁸ Violence increases as new and competing criminal organizations struggle to establish control over reconfigured illicit supply chains (Bagley 2012, 11–13; Shifter 2007, 59). Both criminal organizations and states are conceptualized as locked in perpetual battle, which entails a combination of violence and corruption that challenges national security by destabilizing state institutions, fraying state relations between producer and consumer countries, and weakening the perceived legitimacy of democratic regimes as they struggle to stem violence and impose order (Bailey and Chabat 2002; Rabasa and Chalk 2001; Shifter 1999; Tokatlian 1988). Other scholars unpack the consequences of the overwhelmingly militarized nature of state responses to these challenges (MacCoun and Reuter 2001). The anti-narcotics efforts of the United States in Latin America, for example, have long emphasized disrupting and dismantling the production, processing, and transport of illicit narcotics while largely neglecting domestic consumer demand. This has generated counterproductive effects in target countries, including undermining human rights, generating new opportunities for the corruption of state security forces, and weakening incentives for national politicians to uphold democratic practices (Youngers and Rosin 2005).

A macrolevel approach provides important insights into the national and global dimensions of criminal violence, but offers weak leverage for unpacking the politics of urban violence. The focus on the national level makes it difficult to detect subnational variation in the frequency and forms of criminal violence. The focus on violence wielded by large criminal organizations overlooks how these institutions and illicit markets are reliant upon diverse arrays of smaller armed actors based in subnational locales and who exhibit varying levels of autonomy from larger criminal groups. But most importantly, the national-level emphasis limits our ability to analytically disaggregate the state and criminal organizations and, in turn, theorize how distinct sets of relations between different levels of both actors may generate varied patterns of violence and (dis) order.⁹

Macrolevel approaches provide particularly limited traction for studying the role of business in the politics of urban violence, which has largely been overlooked in extant research, and which I place at the center of my analysis alongside the political actors and institutions traditionally found within studies of crime and violence.¹⁰ Analyses of the financial costs that violence inflicts on private sector interests rely on aggregate indicators, such as fluctuations in economic growth and levels of foreign investment (Gaviria 2002; Pshisva and Suarez 2006). But these approaches disembed the private sector from politics, making it difficult to analyze the linkages between business interests and political forces and, in turn, how these linkages impact the private sector’s preferences and mobilization in efforts to confront violence. This constrains our ability to theorize the differentiated effects that violence poses for distinct configurations of the private sector, and how local political conditions impact business responses to violence.¹¹

More broadly, in recent years one of the most transformative institutional changes to take place in political systems across the developing world is the decentralization of a range of policy domains, including security, to subnational governments.¹² The widespread establishment of direct elections for city mayors has reshaped the territorial distribution of incentives that politicians face to harness security policy-making in order to build political power.¹³ Mayors increasingly wield varying levels of authority over municipal police forces, and realization that their political fortunes hinge on local security conditions have led some to resist recentralization of such powers (Eaton 2008, 14–18).¹⁴ But decentralization has also provided subnational governments with a range of powers beyond violence control that can be used to craft political projects in response to violence. Indeed, the very concept of security is changing in ways that encourage local political leaders to become protagonists in confronting urban violence. Citizen security, which entails measures to reduce criminal violence and the fear of victimization (Casas-Zamora 2013, 2), is both a prerequisite and an outcome of human development (UNDP 2013, 3).¹⁵ Violence prevention efforts can thus include a range of targeted local-level measures, including socioeconomic investment in marginalized neighborhoods and the reshaping of the built environment, among others. A national-level focus makes it difficult to capture and theorize this rich variation in types of political projects that cities launch to address violence.

In contrast, a microlevel approach scales down from the national to the ground level, where units of analysis can vary from jurisdictionally defined territories, such as neighborhoods, to nonjurisdictional units, including individuals, armed groups, and informal settlements. The overarching feature and strength of the microlevel approach, however, is its ability to generate insights into the micro-dynamics of criminal violence through its pinpoint focus.¹⁶ A key finding from microlevel studies is that the relationship between illicit markets and violence is mediated by a range of variables, including the nature of relations between criminal and political actors (Arias 2013; Leeds 1996; Rios 2012; Snyder and Durán-Martínez 2009). Microlevel approaches show how social networks can either facilitate or inhibit criminal violence (Arias 2006). Other scholars examine how the institutions forged between criminal and political actors explain variation in types of localized political orders, which in turn have consequences for associational life, economic markets, and political mobilization (Moncada 2013a).

Yet extant microlevel studies underemphasize the role of several important city-level actors whose interests may include the microlevel, but also extend well beyond it, such as business. When business does appear in microlevel studies, it is often in the form of idiographic description with an emphasis on the survival strategies of small businesses saddled with tight budgets, including the turn toward alternative forms of security by bribing local police, paying extortion fees to armed actors, or organizing social cleansings targeting impoverished youth and others considered socially undesirable.¹⁷ The large-scale business interests that are protagonists in urban political economies are absent in microlevel studies. Certainly small firms are disproportionally targeted for varied acts of