State-building and tax regimes in Central America 9781107019096

Citation preview

State-Building and Tax Regimes in Central America

In Central America, dynamic economic actors have inserted themselves into global markets. Elites atop these sectors attempt to advance a state-building project that will allow them to expand their activities and access political power, but they differ in their internal cohesion and their dominance with respect to other groups, especially previously constituted elites and popular sectors. Differences in resulting state­ building patterns are expressed in the capacity to mobilize revenues from the most dynamic sectors in quantities sufficient to undertake public endeavors and in a relatively universal fashion across sectors. Historical, quantitative, and qualitative detail on the five countries of Central America are followed by a focus on El Salvador, Honduras, and Guatemala. The greatest changes have occurred in El Salvador, and Honduras has made some advances, although they are almost as quickly reversed by incentives, exemptions, and special arrangements for particular producers. Guatemala has raised revenues only mar­ ginally and failed to address problems of inequity across sectors and between rich and poor. Aaron Schneider is the Jill H. and Avram A. Glazer Professor of Social Entrepreneurship and Associate Professor of Political Science and Latin American Studies at Tulane University. He is especially interested in the sources of and uses for the wealth of states, including taxation and expenditure. By treating public finance as a window into the politics of economic development, his work characterizes the way in which state actors interact with the evolution of global capitalism. He has con­ ducted research in Central America, Brazil, Peru, India, sub-Saharan Africa, and closer to home in New Orleans. With a mix of quantitative and qualitative methods, Professor Schneider’s outputs have included journal articles, book chapters, working papers, and policy briefs. Prior to joining Tulane University, he was a Fellow at the University of Sussex in the Institute of Development Studies (IDS) and Advisor to the Chief Economist for Central America and the Caribbean at the Inter­ American Development Bank.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

“Aaron Schneider explains with great clarity the specific political strategies or paths followed in Central American countries in order to implement tax reforms and strengthen the state. Anyone interested in the comparative analysis of the interaction of fiscal policy, conflict, state building, and glob­ alization should read this book.” - Juan Alberto Fuentes Knight, former Minister of Finance of Guatemala “ This analytically ambitious and empirically thorough comparative study represents a dramatic and much needed advance in research on contem­ porary Central American political economy. Anchored in recognition of often overlooked transformations of the region’s economies and their impact on the confi guration of elites, Aaron Schneider’s book underscores how patterns of taxation are shaping the contours of states, and distribu­ tional outcomes, in highly unequal societies in which revenue constraints hinder efforts to consolidate stable democratic institutions. The insights he puts forth are invaluable for scholars and practitioners seeking to under­ stand the challenges to state-building that confront fledgling democracies in Central America and beyond.” - Eric Hershberg, American University “ How to raise taxes to fund public policy constitutes a major challenge for all developing countries. Professor Schneider offers an outstanding inter­ pretation of why some countries have been more successful than others in meeting it. Based on a comparative study of the recent experience in El Salvador, Guatemala, and Honduras, he shows how the power and cohe­ sion of the new transnational elite determines the nature of tax reforms. Countries like El Salvador where the elite is cohesive and dominates the process of policy making are more likely to expand taxes than others, even if through regressive means. By studying the tax system, Professor Schneider also offers a new understanding of the process of state building in small countries in the era of globalization. This book is political econ­ omy as its best.” - Diego Sánchez-Ancochea, University of Oxford “ Of the many questions about the history of Central America, the fiscal aspect is one of the most important. The construction of the national state and its operation cannot be understood without looking at institutions and fi scal policies, the tax system, and state resources, both historically and today. Professor Schneider’s ambitious work is of the utmost importance; it is probably the first work on this subject, with the additional merit of refer­ ring to Central America. It is both historical and current. This is a valuable book, which I enthusiastically recommend for anyone wishing to better understand this region.” - Edelberto Torres-Rivas, United Nations Development Programme

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

State-Building and Tax Regimes in Central America

A A R O N S C H N E ID E R

Tulane University

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi, Mexico City Cambridge University Press 32 Avenue of the Americas, New York,

ny

10 0 13 -2 4 7 3 , USA

www.cambridge.org Information on this title: w ww.cam bridge.org/9781107019096 © Aaron Schneider 2 0 12 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2 0 12 Printed in the United States of America A catalog record for this publication is available from the British Library. Library o f Congress Cataloging in Publication data Schneider, Aaron, 1 9 7 1 State-building and tax regimes in Central America / Aaron Schneider. p. cm. Includes bibliographical references and index. i s b n 978 -1-107-0 19 09 -6 (hardback) 1. Tax administration and procedure - Central America. 2. Nation-building - Central America. 3. Central America - 1979 - Politics and government. I. Title. H J2471.S36 2 0 12 336.2009728-dc23 2 0 110 4 7 8 3 7 ISBN 978 -1-107-0 19 09 -6 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of u r l s for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

State-Building and Tax Regimes in Central America

In Central America, dynamic economic actors have inserted themselves into global markets. Elites atop these sectors attempt to advance a state-building project that will allow them to expand their activities and access political power, but they differ in their internal cohesion and their dominance with respect to other groups, especially previously constituted elites and popular sectors. Differences in resulting state­ building patterns are expressed in the capacity to mobilize revenues from the most dynamic sectors in quantities sufficient to undertake public endeavors and in a relatively universal fashion across sectors. Historical, quantitative, and qualitative detail on the five countries of Central America are followed by a focus on El Salvador, Honduras, and Guatemala. The greatest changes have occurred in El Salvador, and Honduras has made some advances, although they are almost as quickly reversed by incentives, exemptions, and special arrangements for particular producers. Guatemala has raised revenues only mar­ ginally and failed to address problems of inequity across sectors and between rich and poor. Aaron Schneider is the Jill H. and Avram A. Glazer Professor of Social Entrepreneurship and Associate Professor of Political Science and Latin American Studies at Tulane University. He is especially interested in the sources of and uses for the wealth of states, including taxation and expenditure. By treating public finance as a window into the politics of economic development, his work characterizes the way in which state actors interact with the evolution of global capitalism. He has con­ ducted research in Central America, Brazil, Peru, India, sub-Saharan Africa, and closer to home in New Orleans. With a mix of quantitative and qualitative methods, Professor Schneider’s outputs have included journal articles, book chapters, working papers, and policy briefs. Prior to joining Tulane University, he was a Fellow at the University of Sussex in the Institute of Development Studies (IDS) and Advisor to the Chief Economist for Central America and the Caribbean at the Inter­ American Development Bank.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

“Aaron Schneider explains with great clarity the specific political strategies or paths followed in Central American countries in order to implement tax reforms and strengthen the state. Anyone interested in the comparative analysis of the interaction of fiscal policy, conflict, state building, and glob­ alization should read this book.” - Juan Alberto Fuentes Knight, former Minister of Finance of Guatemala “ This analytically ambitious and empirically thorough comparative study represents a dramatic and much needed advance in research on contem­ porary Central American political economy. Anchored in recognition of often overlooked transformations of the region’s economies and their impact on the confi guration of elites, Aaron Schneider’s book underscores how patterns of taxation are shaping the contours of states, and distribu­ tional outcomes, in highly unequal societies in which revenue constraints hinder efforts to consolidate stable democratic institutions. The insights he puts forth are invaluable for scholars and practitioners seeking to under­ stand the challenges to state-building that confront fledgling democracies in Central America and beyond.” - Eric Hershberg, American University “ How to raise taxes to fund public policy constitutes a major challenge for all developing countries. Professor Schneider offers an outstanding inter­ pretation of why some countries have been more successful than others in meeting it. Based on a comparative study of the recent experience in El Salvador, Guatemala, and Honduras, he shows how the power and cohe­ sion of the new transnational elite determines the nature of tax reforms. Countries like El Salvador where the elite is cohesive and dominates the process of policy making are more likely to expand taxes than others, even if through regressive means. By studying the tax system, Professor Schneider also offers a new understanding of the process of state building in small countries in the era of globalization. This book is political econ­ omy as its best.” - Diego Sánchez-Ancochea, University of Oxford “ Of the many questions about the history of Central America, the fiscal aspect is one of the most important. The construction of the national state and its operation cannot be understood without looking at institutions and fi scal policies, the tax system, and state resources, both historically and today. Professor Schneider’s ambitious work is of the utmost importance; it is probably the first work on this subject, with the additional merit of refer­ ring to Central America. It is both historical and current. This is a valuable book, which I enthusiastically recommend for anyone wishing to better understand this region.” - Edelberto Torres-Rivas, United Nations Development Programme

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

State-Building and Tax Regimes in Central America

A A R O N S C H N E ID E R

Tulane University

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi, Mexico City Cambridge University Press 32 Avenue of the Americas, New York,

ny

10 0 13 -2 4 7 3 , USA

www.cambridge.org Information on this title: w ww.cam bridge.org/9781107019096 © Aaron Schneider 2 0 12 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2 0 12 Printed in the United States of America A catalog record for this publication is available from the British Library. Library o f Congress Cataloging in Publication data Schneider, Aaron, 1 9 7 1 State-building and tax regimes in Central America / Aaron Schneider. p. cm. Includes bibliographical references and index. i s b n 978 -1-107-0 19 09 -6 (hardback) 1. Tax administration and procedure - Central America. 2. Nation-building - Central America. 3. Central America - 1979 - Politics and government. I. Title. H J2471.S36 2 0 12 336.2009728-dc23 2 0 110 4 7 8 3 7 ISBN 978 -1-107-0 19 09 -6 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of u r l s for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

State-Building and Tax Regimes in Central America

In Central America, dynamic economic actors have inserted themselves into global markets. Elites atop these sectors attempt to advance a state-building project that will allow them to expand their activities and access political power, but they differ in their internal cohesion and their dominance with respect to other groups, especially previously constituted elites and popular sectors. Differences in resulting state­ building patterns are expressed in the capacity to mobilize revenues from the most dynamic sectors in quantities sufficient to undertake public endeavors and in a relatively universal fashion across sectors. Historical, quantitative, and qualitative detail on the five countries of Central America are followed by a focus on El Salvador, Honduras, and Guatemala. The greatest changes have occurred in El Salvador, and Honduras has made some advances, although they are almost as quickly reversed by incentives, exemptions, and special arrangements for particular producers. Guatemala has raised revenues only mar­ ginally and failed to address problems of inequity across sectors and between rich and poor. Aaron Schneider is the Jill H. and Avram A. Glazer Professor of Social Entrepreneurship and Associate Professor of Political Science and Latin American Studies at Tulane University. He is especially interested in the sources of and uses for the wealth of states, including taxation and expenditure. By treating public finance as a window into the politics of economic development, his work characterizes the way in which state actors interact with the evolution of global capitalism. He has con­ ducted research in Central America, Brazil, Peru, India, sub-Saharan Africa, and closer to home in New Orleans. With a mix of quantitative and qualitative methods, Professor Schneider’s outputs have included journal articles, book chapters, working papers, and policy briefs. Prior to joining Tulane University, he was a Fellow at the University of Sussex in the Institute of Development Studies (IDS) and Advisor to the Chief Economist for Central America and the Caribbean at the Inter­ American Development Bank.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

“Aaron Schneider explains with great clarity the specific political strategies or paths followed in Central American countries in order to implement tax reforms and strengthen the state. Anyone interested in the comparative analysis of the interaction of fiscal policy, conflict, state building, and glob­ alization should read this book.” - Juan Alberto Fuentes Knight, former Minister of Finance of Guatemala “ This analytically ambitious and empirically thorough comparative study represents a dramatic and much needed advance in research on contem­ porary Central American political economy. Anchored in recognition of often overlooked transformations of the region’s economies and their impact on the confi guration of elites, Aaron Schneider’s book underscores how patterns of taxation are shaping the contours of states, and distribu­ tional outcomes, in highly unequal societies in which revenue constraints hinder efforts to consolidate stable democratic institutions. The insights he puts forth are invaluable for scholars and practitioners seeking to under­ stand the challenges to state-building that confront fledgling democracies in Central America and beyond.” - Eric Hershberg, American University “ How to raise taxes to fund public policy constitutes a major challenge for all developing countries. Professor Schneider offers an outstanding inter­ pretation of why some countries have been more successful than others in meeting it. Based on a comparative study of the recent experience in El Salvador, Guatemala, and Honduras, he shows how the power and cohe­ sion of the new transnational elite determines the nature of tax reforms. Countries like El Salvador where the elite is cohesive and dominates the process of policy making are more likely to expand taxes than others, even if through regressive means. By studying the tax system, Professor Schneider also offers a new understanding of the process of state building in small countries in the era of globalization. This book is political econ­ omy as its best.” - Diego Sánchez-Ancochea, University of Oxford “ Of the many questions about the history of Central America, the fiscal aspect is one of the most important. The construction of the national state and its operation cannot be understood without looking at institutions and fi scal policies, the tax system, and state resources, both historically and today. Professor Schneider’s ambitious work is of the utmost importance; it is probably the first work on this subject, with the additional merit of refer­ ring to Central America. It is both historical and current. This is a valuable book, which I enthusiastically recommend for anyone wishing to better understand this region.” - Edelberto Torres-Rivas, United Nations Development Programme

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

State-Building and Tax Regimes in Central America

A A R O N S C H N E ID E R

Tulane University

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi, Mexico City Cambridge University Press 32 Avenue of the Americas, New York,

ny

10 0 13 -2 4 7 3 , USA

www.cambridge.org Information on this title: w ww.cam bridge.org/9781107019096 © Aaron Schneider 2 0 12 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2 0 12 Printed in the United States of America A catalog record for this publication is available from the British Library. Library o f Congress Cataloging in Publication data Schneider, Aaron, 1 9 7 1 State-building and tax regimes in Central America / Aaron Schneider. p. cm. Includes bibliographical references and index. i s b n 978 -1-107-0 19 09 -6 (hardback) 1. Tax administration and procedure - Central America. 2. Nation-building - Central America. 3. Central America - 1979 - Politics and government. I. Title. H J2471.S36 2 0 12 336.2009728-dc23 2 0 110 4 7 8 3 7 ISBN 978 -1-107-0 19 09 -6 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of u r l s for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

Contents

List o f Figures and Tables

page vii

List o f Acronyms

ix

Preface

xiii

1.

Revenues, States, and Central America The Nature o f Tax The Argument Methodology o f Research and Comparison Structure o f the Book

i 12 16 19 23

2.

State-Building in a Globalized Political Economy Globalization State-Building Fiscal Sociology Framework o f State-Building and Tax Regimes Conclusion

25 25 29 34 40 50

3.

Historical Junctures in Central American State-Building and Tax Central American State-Building Over Time and Across Space Tax Regimes 1980s Crisis Conclusion

52 54 64 75 77

1990s Transnational Integration: Quantitative Evaluation of Socioeconomic Actors, Democratic Institutions, and Tax Regimes Introduction Socioeconomic Actors Political System

80 80 83 92

4.

v Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

Contents

vi

5.

6.

7.

8.

Fiscal Conditions Conclusion

99 106

Inside-Out State-Building in El Salvador: Dominant and Cohesive Transnational Elites Introduction Increase Taxes, but Not on the Wealthiest Dominant and Cohesive Transnational Elites Facilitating State-Building Reform Abundance Reforms by Fiat and Retreat Conclusion

109 109 110 115 124 128 13 2 13 5

Outside-In State-Building in Honduras: Dominant but Divided Transnational Elites Increasing Taxes and Tax Breaks Dominant but Divided Transnational Elites Parallel Institutions Reform Cycles Reform from Outside and Erosion Conclusion

13 7 138 14 2 15 5 159 16 2 169

Crisis in Guatemalan State-Building: Divided, Subordinate Transnational Elites Introduction Emergency Increases, Stagnating Tax Divided and Subordinate Transnational Elites Institutional Obstacles Many Reforms, Little Change Reform Collapse and by Stealth Conclusion

17 1 17 1 17 2 17 5 183 18 7 190 196

Conclusion: Globalization and State-Building in Developing Countries The Politics o f Tax Regimes Countervailing Power 2009

198 199 203 205

Appendix: Database on Tax Legislation

2 13

Bibliography

2 15

Index

239

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

Figures and Tables

Figures 1 .1 . 3 .1. 4 .1. 5.1. 5.2. 6.1. 7 .1.

Change in tax/GDP since 1990 Proportion of revenue from different tax sources Central American market share of U.S. apparel imports El Salvador, revenue structure El Salvador, economic change Honduras, revenue structure Guatemala, revenue structure

page 7 73 86 111 117 140 174

Tables 1 .1 . 3 .1. 3.2. 3.3. 3.4. 3.5. 3.6. 4 .1. 4.2. 4.3. 4.4. 4.5.

Emerging Elites, State-Building, and Tax Regimes Tax as Percent GDP Divergent State-Building Over Time and Space Revenues in Pesos Tax Regimes at the End of the Liberal Period, 1929 Revenue During the Depression Post-WWII Tax as Percent of GDP Indicators of Integration, Annual Averages (US$ million) Party Systems, Central America Tax Burdens Since 1990 Central America Average Tax, Percent of Total Free Trade Zone Incentives, Exemptions, and Durations, 2005 4.6. Average Tax Proportions as a Percentage GDP, 2003-2007 5.1. Productivity of Tax, Selected Sectors 5.2. Party System, El Salvador

19 54 63 66 68 69 72 87 94 10 1 102 104 106 113 123 vii

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

viii

5.3. 6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 7 .1. 7.2. 8.1.

Figures and Tables Tax Reforms, El Salvador Honduras, Tax Burden by Base and Income Decile Financial Sector and International Alliances Business Elites in Government, by Party and Sector Legislative Seats, Honduras Party System Polarization, Honduras Tax Reforms, Honduras Seats in Guatemalan Legislative Elections Tax Reforms, Guatemala Dominance and Cohesion, Patterns of Reform

13 1 14 1 145 14 7 15 3 154 16 1 18 1 188 204

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

Acronyms

AHIBA AHM ON AIFLD ALBA ANDI ANDI ANEP ARENA ASC ASESA ASIES BCIE CACIF CACM CAEM CBI CD CGS CGT CICIG CIEDEG CLE COPMAGUA

Honduran Association of Banking Institutions Honduran Association of Municipalities United States Institute for the Development of Free Labor Bolivarian Alliance for the Americas National Association of Industries National Industrial Development Association Salvadoran Private Enterprise Association Republican National Alliance Civil Society Assembly Agreement on Socioeconomic and Agrarian Issues Social Research and Study Association Central American Bank for Economic Integration Coordinating Committee of Agricultural, Commercial, Industrial and Financial Associations Central American Common Market Entrepreneurial Chamber Caribbean Basin Initiative Democratic Convergence General Union Central General Workers’ Central International Commission against Impunity Evangelical Church of Guatemala Conference Free Market Chamber Coordinator of Mayan People’s Organizations in Guatemala IX

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

x

COS CPPF CTH CTS CUTH DEES DEI FEDECAM ARA FENAGH FENASTRAS FIDE FLACSO FOSDEH FM LN FRG FUSADES GANA ICEFI IEMA IETAAP IPES LET MINUGUA NGO PAN PCN PDC PL PN PP RIT SAT SICA UASP UNE

Acronyms Collective of Social Organizations Preparatory Commission of the Fiscal Pact Honduran Workers’ Confederation Salvadoran Laborers Central Unitary Workers’ Central Department for Economic and Social Study Executive Tax Directorate Federation of Chambers of Commerce and Industry National Federation of Agriculturalists and Cattle National Federation of Salvadoran Workers Foundation for Investment and Export Development Faculty of Latin American Social Sciences Social Forum of External Debt and Honduran Development Farabundo Marti National Liberation Party Guatemalan Republican Front Salvadoran Foundation for Economic and Social Development Grand National Alliance Central American Institute for Fiscal Studies Mercantile and Agribusiness Tax Extraordinary and Temporary Tax in Support of the Peace Accords Institute of Political, Economic, and Social Studies Tax Equity Law UN Mission to Guatemala nongovernmental organization Party of National Advancement National Conciliation Party Christian Democratic Party Liberal Party Nationalist Party Patriotic Party Temporary Import Regime Tax Administration Superintendence Central American Integration System Public Sector Analysis Unit National Unity of Hope

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

xi

Acronyms UNOC UNTS U RN G U RN G USAID VAT ZIP ZOLI ZOLT

National Unity of Workers and Peasants National Salvadoran Workers United Guatemalan Revolutionary National Unity National Revolutionary Unity of Guatemala U.S. Agency for International Development Value-Added Tax Industrial Processing Zone Laws Tax Free Zone Tax Free Tourism Zone

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

Preface

In 1993, I arrived in San Antonio Los Ranchos, El Salvador, a commu­ nity of repatriated refugees and Farabundo Marti Front for National Liberation (FMLN) ex-guerrilla combatants not far from the Honduran border. I was assigned the task of teaching them accounting skills, the set of double-ledger calculations that could tell the men that they were los­ ing money on the sides of mountains they cultivated with back-breaking labor and indicate to the women that they were barely breaking even on the putting-out system of sewing they conducted in their homes. I lived among them for a year, sharing their experiences of the war and see­ ing their struggles to care for family and community. Everything I now understand about global political economy I first learned from them. This book is close to my heart because Central America is close to my heart. While I lived in El Salvador for a year, I toured the isthmus by motorcycle, and I was struck by the beauty and the horror of the region spectacular mansions and abject poverty. To help me make sense of these contrasts, I owe much to the community leaders I met in San Antonio Los Ranchos and associated movements, men like Father Jon Cortina. I also owe a great debt to José Roberto Gochez, his wife Ana Teresa, and children Diego, Antonio, and Rebeca. They were my Salvadoran family, welcoming me into their home on nights spent in San Salvador, making sure I did not crack my head open riding a motorcycle on still war-torn roads. Roberto was my boss for 19 9 3-19 9 4 , he is now a vice minister of the Salvadoran government, and he has been a friend and source of insights always. To begin to organize such insights for myself, I returned to graduate school and wrote a Master’s thesis about FM LN political strategy in xiii Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

xiv

Preface

its transition from guerrilla fighting force to electoral competitor. I was guided through graduate school by attentive and wonderful advisors and professors, especially David Collier, Ruth Berins Collier, David Leonard, and Henry Brady. They helped me put order to my thoughts on political economy, teaching me especially to choose questions that are interesting and important, and how to go about answering such questions. Of course, no graduate school experience is completed alone, and I could never have survived without the companionship and incredibly good ideas of my col­ leagues, all of whom are still good friends: Ben Goldfrank, Ken Greene, Marcus Kurtz, Emil Levendoglu, Evan Lieberman, Julia Lynch, Khalid Medani, David Moon, Marc Morje-Howard, Lise Morje-Howard, and Lauren Morris-Mclean. While learning from them in graduate school, I began an ongoing curiosity about Brazil, where I saw global economic change through the experience of a troubled, but emerging, developing world power. M y reintroduction to Central America came in 20 0 1, when I joined the Institute of Development Studies at the University of Sussex, and I met my wife, Rebeca Zuniga-Hamlin, a Nicaraguan who reminded me that the small, open, vulnerable countries of Central America have much to teach us about the highs and lows of the global economy. Rebeca and her family, brother Marcelo and mother Maria, along with their commu­ nity of friends in Nicaragua and Guatemala, have taught me much about the role of struggle and dignity and justice. At the Institute of Development Studies, I was encouraged to narrow my political economy focus to public finance issues, especially taxation. I was guided in this task by excellent colleagues: Arnab Acharya, Robert Chambers, Richard Crook, John Gaventa, Peter Houtzager, Richard Jolly, Anu Joshi, Raphie Kaplinsky, James Manor, Lyla Mehta, Mick Moore, Peter Newell, Rachel Sabates-Wheeler, and Hans Singer. Working in the U.K. taught me a different way to be an academic and provided opportu­ nities to explore methodologies, empirical cases, and policy engagement that would never have been available to me in the United States. One such opportunity was a U.K. Department for International Development placement in a temporary position at the Inter­ American Development Bank (IADB), also supported by the Swedish International Development Cooperation Agency and Norwegian Agency for Development Cooperation. At the IADB, I was placed in the Central America and Caribbean division, directed by Máximo Jeria, and assigned to Chief Economic Advisor Manuel Agosín. I joined his team of research­ ers, Roberto Machado, Claudia Martinez, Vibeke Oi, Humberto Petrei,

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

Preface

xv

and Alejandro Tamola, who had devised fi seal reforms for the govern­ ments of the region, but they were baffled as to why so few seemed to get off the ground. They wanted a political scientist to help them understand, and in the proeess they taught me to engage and understand eeonomist colleagues. Manuel was particularly helpful as a guide for how to com­ bine policy relevance and academic contribution, and he helped me navi­ gate the labyrinthine maze of large, international bureaucracies. That experience set up this project directly, as another colleague, Peter Solis, helped me to identify trust funds for IADB research set up by other donor agencies. They backed my plan to coordinate colleagues in the region in the analysis of tax reforms in each of their countries, follow­ ing a parallel and complementary methodology. Supervised through the application process by Juliana Pungiluppi, guided in the initial stages of research by Susana Sitja-Rubio, and coaxed to completion by Alberto Barreix, I produced methodological guides, selected researchers from each of the countries, and oversaw the completion of their work. I was assisted in these tasks by Juliana Arbalaez-Mungo. Jorge Vargas Cullel produced a background paper on Costa Rica; I assisted Leticia Gonzalez in the production of a background paper on Guatemala; and, Alex Segovia worked with me to produce the background paper on El Salvador. On the background paper I produced on Honduras, Hugo Noe Pino provided first-hand knowledge. These papers served as useful inputs, with primary research offered by colleagues who understand their own countries far better than I. The visits involved in this collaboration produced many additional contacts and an equal number of debts to individuals from the region. I am particularly indebted to Juan Alberto Fuentes Knight, who started the Central American Institute for Fiscal Studies (ICEFI) and invited me to become a member of its advisory board. Through that position, I made the acquaintance of regional experts, such as the next ICEFI direc­ tor, Fernando Carrera, and the rest of the ICEFI team, Adelma Bercian, Maynor Cabrera, Maria Castro, Vivian Guzman, Jonathan Menkos, Iliana Pena, and Martin Rodriguez. In Honduras, Nelson Avila has been a friend and useful source of contacts, and Armando Sarmiento provided insights to tax administration as well as visits to a local watering hole. Lourdes Aguilar welcomed me with an occasional place to stay and pro­ vided an indispensable double-check on my incipient understanding of Honduran society. None of my work would have been possible without the aid of Juan Carlos “ Bebe” Gómez Sabaini, an Argentine economist and perhaps the foremost expert on tax in the region, who provided data

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

xvi

Preface

and background materials. With Bebe and the other regional experts on fiscal affairs, “ fiscalistas” as they call their select club, I look forward to further exchanges, especially at the annual Regional Seminar on Fiscal Policy of the Economic Council of Latin America and the Caribbean, run by Ricardo Martner. Since 2008 I have been at Tulane University in New Orleans, where I have benefited from the guidance and support of excellent colleagues. In addition to dribbling circles around me on the soccer field (Tony Pereira) or leaving me in the dust on the bicycle (Tom Langston), colleagues like Gary Brooks, Brian Brox, Mary Clark, Chris Fettweis, Khaled Helmy, Casey Kane-Love, Celeste Lay, Nancy Maveety, Gary Remer, Ray Taras, Martyn Thompson, Raúl Sanchez Urribarri, Mark Vail, and Dana Zartner have smoothed my reentry to U.S. academia. They helped me make sense of teaching undergraduates again, and they supported my year sabbatical in 2010, making it possible for this project to be completed. Research sup­ port was provided by the Tulane Stone Center for Latin American Studies, where I am indebted to James Huck and Director Tom Reese and the research assistance provided by Mariana Alcoforado, Jennifer Boone, and Leslee Dean. Additional funding has been provided by the University of New Mexico Latin American and Iberian Institute, through a 2010 Richard Greenleaf Library Fellowship. In Albuquerque, I have been oriented by Andrew Schrank and longtime Central America observer, Bill Stanley. I am particularly grateful to the team at Cambridge University Press, especially Lew Bateman, who has shepherded this project to its comple­ tion. Anonymous reviews strengthened the book significantly, and the professional production staff ensured that the final product was polished and tight. This project would certainly not have been possible without the help of the people and institutions mentioned here, and I am sure there are many others I have missed. Utmost gratitude goes to my family, father M ark, mother Susan, and sister Miriam. This book bears your imprint, you believed before I did that I was capable of completing such a project. I promise not to miss any more Thanksgivings. The most important acknowledgment goes to my wife, Rebeca, who forgives my occasional absences and frequent distractions. Most of all, Rebeca inspires me to do work that is both professionally rigorous and morally just. This is the way she lives her life, and it is the way I hope to live mine. Rebeca, I fall in love with you every day, and because of you, I have fallen in love all over again with the people of Central America, for whom this book is written.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061

1

Revenues, States, and Central America

Building states is both more important and more difficult in an age in which resources, information, products, and (occasionally) people flow across borders with greater speed than ever before.1 Under such circum­ stances, the tasks of state-building must be undertaken anew, calling into question the responsibilities of states to citizens, the responsibilities of states to holders of capital, and the responsibilities of states to other states. No issue captures these potential conflicts more clearly than taxation, by which state authorities muster the resources needed to govern and in the process negotiate contributions from citizens. Citizens form groups and associations to negotiate their obligations and the benefits they expect in return and establish political institutions to formalize these state-society relationships. In time, this can have lasting impacts on political practice and regime outcomes (Lieberman, 2003; Gehlbach, 2008). There has been significant attention given to the role of revenues in building early Western European states, and even some attention given to formative moments of state-building in developing countries (Tilly, 1992; Centeno, 2002). However, we have limited insight into what happens when economies change significantly, with new leading sectors, new patterns of social orga­ nization, and new requirements of state authorities. How do governments sustain themselves in the face of more mobile sources of revenue, and how do new social actors fit into already established political institutions? 1 Whereas globalization renews debates on state-building, the topic has attracted repeated attention from social science observers over the years, including during the formation of the European interstate system and the expansion of industrial capitalism (Weber, 1968), the process of decolonization (Huntington, 1968), and the reform of state institutions in developing countries (World Bank, 1997). 1

1 Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

2

State-Building and Tax Regimes in Central America

Struggles over tax are a particularly good place to look for evidence that sectors emerging as a result of global integration are pursuing a state­ building agenda. Taxes express in fi scal terms the way groups organize and advance a project for their state, and it is in taxes that we can find evidence that state-building is occurring, or not, and in what ways. Taxes institutionalize who pays for what benefits enjoyed by whom. As recent scholarship on the topic has noted, “ Taxes formalize our obligations to each other. They define the inequalities we accept and those that we col­ lectively seek to redress. They signify who is a member of our political community, how wide we draw the circle of ‘we.’ They set the boundaries of what our governments can do. In the modern world, taxation is the social contract” (Martin, Mehrotra, and Prasad i 2009: 1). This book studies state-building under globalization by comparing the politics of revenue in five Central American countries: Guatemala, Nicaragua, Honduras, Costa Rica, and El Salvador, with particular emphasis on El Salvador, Honduras, and Guatemala. These countries have relatively similar economic and social structures, buffeted in compa­ rable fashion by global forces beyond their control. At first glance, their responses appear similar, with governments of the region largely adopt­ ing the set of policies denoted in the Washington Consensus, especially when it comes to trade integration, privatization, and market liberal­ ization (Williamson, 1990; Bird, 1992). In revenue terms, the countries have generally dropped trade taxes, raised domestic indirect taxes, and marginally increased revenue capacity (Agosin, Barreix, and Machado, 2005). Viewed in these terms, the state-building experiences of Central American countries are instructive for those curious about all developing countries. On closer examination, however, the countries diverge in the precise fiscal changes that have occurred. These divergences are revealed in the degree to which states increase their revenues, apply the rules and regu­ lations of tax regimes universally across sectors, and mobilize revenues from wealthy sectors, especially newly emerging transnational elites adapted to an integrated global economy.2 Together, these dimensions characterize the tax regimes taking shape in the region. 2 Victor Bulmer-Thomas, in observing Central American economic history over the last century, noted that “ although Central America has had no political unity ... it has had a certain ‘economic unity’ as a result of its subjection to common external influences. These have filtered through domestic institutions to affect each economy in slightly different ways. Thus, the region exhibits both conformity and diversity and the problem facing an author is to see the one without losing sight of the other” (Bulmer-Thomas, 1987).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

3

These dimensions can be summarized in the concepts of tax capacity, universality, and progressivity. The capacity to extract revenues is funda­ mental to government functioning; a state cannot be considered authorita­ tive over a territory if it cannot collect taxes.3 For this reason, the amount of taxes collected is a useful indicator of the legitimate capacity of the state, measured in the amount of national wealth mobilized in revenues. This is most directly measured in terms of revenues as a percentage of GDP, although revenues per capita, absolute or nominal revenues, or rev­ enues in relation to spending or debt can provide other impressions of the relative amount of resources available to state authorities (Peters, 1991). For some comparisons, wealth mobilized through other means, such as social security or natural resource royalties, are relevant, but taxes are distinctly useful indicators of state-society relations, as taxes are “ unrequited” payments to the state, requiring a basic degree of state legit­ imacy and embeddedness in society (World Bank, 1988: 79). Taxes are especially useful because there is no quid pro quo, no direct exchange. Other types of contributions, such as fees for services or the profits of public enterprises, may contribute to state coffers and thereby allow the state to provide public goods, but they do not have the character­ istic of an open-ended, individual commitment to the state. Margaret Levi (1989: 49) highlights this aspect of the tax relationship in terms of “ quasi-voluntary compliance,” as tax collection must rest on a basis of consent and not simply the threat of punishment or the promise of some immediate benefit. In a Latin American region where taxes are generally low, Central American governments have traditionally taxed at some of the lowest rates, making any advance in capacity a rich opportunity to explore exactly what has happened in political terms. The second dimension of tax regimes, tax universality, is the degree to which tax law is applied in equal fashion regardless of the way wealth is created, stored, or transferred. For example, universality sug­ gests that incomes generated in one sector are not treated differently than incomes from another, nor are the sales of one kind of goods ver­ sus another. The butcher and the baker who earn an equal sum do not pay different amounts of tax, regardless of to whom and where they sell their wares. Everyone pays their part. In this sense, the problem of universalistic taxation is very much a coordination problem, as groups must agree not to seek particular treatment and benefits within the tax 3 North (1982: 2 1) suggests state “ boundaries are determined by its power to tax constituents.”

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

4

State-Building and Tax Regimes in Central America

code (Bates, 1989; Steinmo, 1993). This aspect of tax regimes gets to the horizontal equity of taxes, which is the degree to which individuals or firms of similar resources carry similar burdens. In the absence of horizontal equity, the impression that similarly situated competitors are not paying the same amount creates a sense of injustice and a resistance to payment, complicating collection and suggesting that government is either too captured or too weak to balance the tax burden among dif­ ferent sectors. One way to approach universality of tax is through the bases and rates defined in tax legislation. Reforms that narrow or widen bases or apply special rates to specific bases have important horizontal equity implica­ tions, creating particular relationships with sectoral groups rather than a universal relationship with the citizenry as a whole. Not only is particu­ larism damaging to the legitimacy of taxes, it also complicates collection by forcing administrations to manage multiple rates and rules. The attention to fairness also implies concern with vertical equity, or the progressiveness of a tax system. This refers to the degree to which those with more wealth pay a greater share. In addition to an ethical con­ cern with distributive justice, progressivity also follows from the intuition that the marginal utility of income decreases with wealth; taking an extra dollar away from a wealthy person has a far lower impact on welfare than taking an extra dollar away from a poor person (Lieberman, 2001: 94). To measure vertical equity, several indicators are useful. Traditional divisions of tax bases into direct and indirect categories are a start, with income, property, and capital gains taxes in the direct category, assumed to fall more heavily on the wealthy. Although this may not perfectly mea­ sure the impact on inequality, there are both objective and subjective rea­ sons to consider this categorization useful. First, most evidence suggests that direct taxes are more progressive than indirect taxes (Musgrave , 1959). In addition, people perceive direct taxes in that way. Opposition to consumption taxes is often made on the basis of a perceived inequity in tax burdens, as poor people pay a greater portion of their income to consume their basic needs. Other measures of inequality capture the degree of progressivity more precisely, such as the distributional impact of taxes, with Gini coefficients before and after taxes among the most common (Goñi, López, Servén, 2008). Estimating Gini coefficients is a difficult task, as it depends on assumptions about behavior had a tax not occurred; and most studies that use Gini coefficients in this fashion model different possible taxpayer responses.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

5

An additional aspect of progressivity is the ability to tap into the most dynamic and lucrative sectors. In the current international political econ­ omy, these are the sectors adapted to globalized production and mar­ kets. Finding a way to engage dynamic sectors is especially important for governments in highly unequal societies. Dynamic sectors offer the large sums necessary for financing public needs, and these sectors tend to be concentrated, making it more convenient to find the resources here rather than scattered throughout the rest of the economy. Also, bargaining over tax with the actors producing in these sectors offers an opportunity to exchange information and identify the ways state power can be used to promote dynamic sectors. Less-developed countries have long faced the problem of export enclaves with few links to the rest of the economy. In addition, the primary products they tend to export face competition from other producers and offer limited returns, much lower than the f nished products they import. In a globally inte­ grated political economy, individual entrepreneurs may be able to iden­ tify market niches, but they need state power and resources if they are to survive, extend linkages to other sectors, and scale up to more complex production processes. There is much to be gained through state engagement with dynamic sectors over the issue of tax, but it is no easy task. Many of these firms and sectors barely existed two decades ago, forcing states to design new administrative tools and legal authority to tap into undef ned tax bases and previously unknown entities. Tax experts call these tax “ handles,” and they have to be invented anew. This is a delicate task, as clumsy or excessive attempts to collect can weigh heavily on the few prof table activities appearing in a poor econ­ omy. The state must be careful not to scare off incipient industries and firms, especially those that can quickly move to other jurisdictions. This is particularly the case under liberalized capital market and trade regimes, when the state has few tools to compel emerging elites to stay home. Tax regimes are characterized by capacity, universality, and progressivity; and in many ways it is progressivity that holds the key. Unfortunately, as will become clear, the state-building projects advanced by emerging elites in Central America have occasionally demonstrated the path to greater capacity and universality, but they have yet to stumble on the pol­ itics of progressivity. Central America is a particularly good place to study the difficult task of constructing tax regimes, as the governments of the region have never been particularly effective at mobilizing revenues. The region has been the

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

6

State-Building and Tax Regimes in Central America

source of lament for at least forty years for international observers who worry about the failure to tax enough to support basic state activities, and especially the failure to tax leading sectors (Joint Tax Program OAS/ IADB, 1966).4 Indeed, observers have long been convinced that there is a connection between the failure to tax the wealth of powerful elites and the fragility of democratic politics in the region (Best, 1976). Historically, Central America presents a perverse version of “ no taxation without rep­ resentation” - for long periods the region has enjoyed neither. Much of our understanding of democratic weakness and underdevel­ opment in the region comes from the rich comparative historical literature that traced the paths producing military authoritarianism, patrimonial dictatorship, and democratic transitions (Stone, 1992; Rueschemeyer, Stephens, and Stephens , 1992; Paige , 1997; Yashar, 1997; Mahoney, 2001). These studies share the conviction that weak revenues contributed repeat­ edly to regime breakdown and lackluster development, as fragile gov­ ernments pitched into crisis when economies slowed or revenues proved insufficient to meet popular demands, elite rivalry, or external challenges. By contrast, these studies do not locate revenues at the heart of their analysis of state construction. They neither explain moments at which rev­ enues increase, nor do they explore the way changes to revenue regimes express alternative efforts by emerging elites to enact state-building proj­ ects. This study, by focusing specifically on revenues, can identify those rare and isolated moments in which tax regimes modernize and describe the alternative state-building projects they reflect. By exploring the tax implications of state-building projects, the study might even offer some hope. Revolutionary upheavals of the 1970s and 1980s threatened states with breakdown; the debt crisis and economic stagnation of the 1980s cut off resources; and subsequent stabilization, peace agreements, economic transformation, and democratization gener­ ated new demands. At least since the 1990s, this has provoked recurrent fiscal crises of Central American states, which have been met with hun­ dreds of efforts to renegotiate tax regimes in terms of new laws, admin­ istrative bureaucracies, and changes to bases and rates. Outcomes have been mixed. Among the three cases studied most closely here, El Salvador has increased revenues and applied them more universally over the last two decades, although the tax structure has not particularly improved 4 A similar concern directed those exploring developing countries more generally, “ Will Underdeveloped Countries Learn to Tax?” (Kaldor, 1963).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

7

------ El Salvador ----- Honduras ------ Guatemala

f ig u r e

1 . 1 . Change in tax/GDP since 1990.

Source: Author calculations from United Nations data collected by the Economic Commission of Latin America and the Caribbean (ECLAC or CEPAL, in Spanish), www. cepal.org/ilpes.

distribution nor has it tapped into the most internationalized segments of leading sectors - the operations of Salvadoran financiers outside Salvadoran territory. Honduras has seen less significant and more infre­ quent increases in revenue capacity and has even witnessed some increases in progressivity, but both capacity and progressivity have been eroded through the steady extension of particularist benefits to narrow groups. Finally, in Guatemala the revenue regime has remained pocked full of particularist holes, with limited capacity and no gains in progressivity. This book seeks to explain the continued weakness of tax regimes in the region while also tracing the moments and processes by which steps for­ ward are taken. In prior work, colleagues and I described this process as “ two steps forward and one step back,” as all countries in the region took steps to improve tax capacity, but they have done little to fight particu­ larism and inequity and have reversed some of their hard-fought gains in capacity (Schneider, Agosín, and Machado, 2008). As an indication of the kind of changes occurring, Figure 1 .1 displays the amount of change in tax regimes in terms of tax capacity over the last fi fteen years. For each of the countries, tax as a percentage of GDP in 1992 is set to 100, with subsequent years appearing as increases or decreases on that base. With only a slight slowdown in 19 9 6 -19 9 7 , El Salvador increased its revenues fairly smoothly, reaching a total almost 40 percent higher than where it started. Honduras increased by almost 30 percent, but there were almost as many downturns as there were upticks. Guatemala increased by only 10 percent.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

8

State-Building and Tax Regimes in Central America

These changes raise some fundamental questions about what is hap­ pening in the region. How has El Salvador increased its revenues so markedly? Does this mark a new chapter in the kind of state governing that small, war-torn country; or have revenues increased merely to benefit the objectives of a narrow elite? What does the 2009 election of the former guerrilla, leftist opposition mean? What about Honduras, where revenues have advanced but infrequent attempts to raise them have been interspersed with deterioration and backsliding? Further, what does the 2009 coup tell us about the kind of state being built in Honduras and the implications for ongoing changes to the tax regime? Finally, what is going on in Guatemala? How can a country make significant occasional gains yet make only minimal headway overall? What does this tell us about the ability of newly emerging elites to get their act together and advance a state-building agenda? These questions attend to deep social and political processes. In partic­ ular, tax regimes are shaped by contemporary patterns of state-building.5 Those patterns are devised by elites empowered by processes of global integration and negotiated with other actors through political institu­ tions. Because emerging elites vary, as do the sectors they encounter and the institutions in which they operate, state-building looks different across the region. Tax regimes are the outcome of these state-building projects, and they reveal divergence across the region. It may seem strange to talk about Central American state-building in the new millennium, almost two hundred years after independence. Yet, state-building is precisely what is occurring in response to global integra­ tion, as governments adapt existing institutions and invent new ones.6 As in other moments in Central American history and other parts of the world, contemporary economic transformations produce new leading sectors, which pursue policies and institutional changes to secure their political access and their needs of the state. These needs include infra­ structure, policy frameworks, and human capital formation to expand their activities and accumulation. To accomplish these tasks, they devise state-building projects composed of the policy changes and institutional adaptations they need from the state. 5 “ Taxes underwrite the capacity of states to carry out their goals; they form one of the cen­ tral arenas for the conduct of state-society relations, and they shape the balance between accumulation and redistribution that gives states their social character” (Brautigam, Fjeldstad, and Moore, 2008: 1). 6 This echoes Fukuyama’s definition of state-building: “ the creation of new governmental institutions and the strengthening of existing ones” (Fukuyama, 2004: 1).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

9

The concept of state-building will be developed more fully in Chapter 2, but for now it is appropriate to raise questions about the leading sectors that design a state-building project, how they relate to other groups and negotiate a national agenda, and how those plans are filtered through and shape already existing, and newly created, political institutions. The core variable distinguished among the state-building projects in the region is the structure of emerging elites. Newly emerging elites share certain characteristics. Based in small countries with small domestic economies, they must export; this requires them to insert and reinsert themselves in changing international markets. Above all, current patterns of international insertion are transnational in the sense that Central American producers operate within produc­ tion processes that are themselves integrated across borders. With low labor costs and close geographical proximity, Central American coun­ tries have become processing zones, importing materials for assembly and reexport, often to the United States. Central American firms have identifi ed opportunities in transnational processes, including assembly manufacture for export; nontraditional agriculture, tourism, and finan­ cial and other services; as well as absorbing the remittances of Central Americans who now live outside the region but send money home. Whereas transnational sectors are largely similar across countries, they diverge in terms of the way elites from these sectors organize themselves internally and relate to other groups, concepts labeled here as cohesion and dominance . Cohesion refers to the patterns of internal organization and inter­ est articulation that characterize emerging elites. Cohesion is important because it indicates an ability to unify around a single, coherent, state­ building agenda. To pursue this task, emerging elites combine within social and economic organizations such as family networks, firms, and business associations. In political terms, it is here that they undertake collective action to coordinate what they need from the state. Cohesion is indi­ cated by relations among rival firms and factions of the economic elite, especially within business associations, and how these associations and their members carry state-building projects into public policy. Political vehicles include political parties, individual leaders taking on electoral and governmental roles, think tanks to articulate their interests and pol­ icies, sectoral lobbying, and more informalized pressure through family and personal connections. Across the region, clear differences emerge in the strategies of interest articulation and aggregation among emerging elites, characterized here by the degree of cohesion.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

10

State-Building and Tax Regimes in Central America

This book also explores the relationship of new elites to other actors. As new sectors emerge, sunrise firms engage with rivals and challengers, such as elites from declining firms in sunset sectors, popular actors, state bureaucrats, and politicians. To the degree that transnational elites can impose their interests over rivals, or convince them to share an interest in expanding transnational accumulation, transnational elites can be viewed as dominant actors within the domestic political economy. Dominance is necessarily a relational term, requiring analysis of the interests and capac­ ities of other actors, as well as the incentives and constraints placed on them by existing political institutions.7 In the highly unequal societies of Central America, where popular sectors are numerous and have pro­ vided support to social movements, antisystem parties, and revolution­ ary uprising, the relative strength of popular sectors deserves particular attention. Still, it is the relative dominance of emerging elites that is criti­ cal, as it captures the degree to which their state-building project is likely to advance. Together, the cohesion and dominance of emerging elites form the crux of the argument made here about the kind of state-building project under­ taken in each country. This echoes other studies of capitalist factions, emphasizing their internal organization and relationships to other sectors and the state (Ross Schneider and Maxfield and Schneider, 1997; Shadlen, 2005). In some ways, this study most closely resembles the approach to capitalist factions taken in Rueschemeyer et al. (1992).8 In a compar­ ative historical study of democratization across regions, Rueschemeyer et al. suggest that the emergence of a domestic bourgeoisie is propitious for democracy but only if balanced by other sectors, including state actors and popular forces. Their argument hinges on three variables: “ the balance of power among different classes and class coalitions ..., the autonomy of the state apparatus and its interrelations with civil soci­ ety, and the impact of transnational power relations” (Rueschemeyer et al., 1992: 5). Central America is included among their cases, focus­ ing on the period around the Great Depression and its aftermath, when democracy emerged in Costa Rica, but elsewhere domestic bourgeoisies were too weak with respect to oligarchic elites who were occasionally aided by U.S. intervention. 7 Mann (1986) takes such a relational approach to state-building processes, exploring the relationship between and among social groups and state actors in the construction of early Western European states. 8 Formal models suggest a balance of class power is propitious to democracy because it lowers elite fears of the democratic majority (Acemoglu and Robinson i 2006).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

11

The current analysis adds to theirs in at least three ways. First, their understanding of both Central America and Latin America tends to emphasize the inherent conflict between a weak domestic bourgeoisie and external capital. The moment at which Latin American bourgeoisies enjoyed the greatest autonomy from international capital was a period of relative international economic decline, the Great Depression, when national bourgeoisies were emboldened to confront declining oligarchic sectors. Rueschemeyer et al. emphasize the risk of external actors ally­ ing with traditional oligarchies to stunt the emergence of domestic bour­ geoisies, something that is unlikely at the current moment. In a period of global integration, factions of the domestic bourgeoisie are emerging precisely because they have been skilled at linking themselves to interna­ tional actors and transnational economic processes. The character of contemporary emerging elites draws attention to a second distinction from Rueschemeyeret al. As they make comparisons across regions, they differentiate among domestic bourgeoisies in terms of their relative strength. Within a rapidly integrating Central American region, where the domestic bourgeoisies are increasingly intertwined, the main source of difference is not only relative strength but also the pre­ cise strategy of international insertion. Transnational elites vary in terms of whether they prioritize attracting external capital or choose to pro­ vide capital to neighboring countries, domestically based versus offshore accumulation. These strategies distinguish among emerging elite sectors, characterize the process of securing internal cohesion, and add nuanced content to state-building projects. The current project retains the Rueschemeyer et al. focus on the bal­ ance of class forces, defined here in terms of the structure of transnational elites. In particular, the current argument concurs with Rueschemeyer et al. that cohesion among emerging elites may be necessary for coherent state-building projects to emerge, but the character of those projects is deeply fl awed without active input from popular sectors. This is partic­ ularly expressed in the progressivity of tax, an outcome largely absent from the changes occurring in the region over the last two decades. This calls for a reframing of the social democratic model that moti­ vates the Rueschemeyer et al. book. The book rests on a post-World War II social democratic capitalist bargain between large-scale labor and large-scale capital enforced by the state.9 These were the elite and 9 This is also the model of the regulation school of comparative political economy (Marglin and Schor, 1990).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

12

State-Building and Tax Regimes in Central America

popular actors in the ascendant sector of the economy, and social dem­ ocratic regimes varied according to the degree to which class balance could sustain dynamism while sharing out the benefits. In contemporary Central America in an age of globalization, neither large-scale labor nor large-scale capital exists. Instead, factions of capital and popular sectors have haphazardly integrated in transnational production processes. The challenge is to build states that share out benefits to popular sectors while sustaining transnational dynamism. The Nature of Tax This calls for a more explicit discussion of determinants of taxation, which has received ample attention in the economic and political litera­ ture. Most economic explanations for the characteristics of tax regimes refer to the nature of economic development and economic structures, highlighting characteristics of modernization such as wealth, urbaniza­ tion, literacy, and industrialization.101 Political theories tend to focus on pressure groups and institutions, emphasizing variables such as the power of different interest groups and regime type.11 The economics literature on taxation has focused on efficiency - how to extract revenues while creating the least economic distortion. Jean Batiste Colbert summarizes this consideration with the observation, “ The art of taxation consists in so plucking the goose as to obtain the largest possible amount of feathers with the smallest possible amount of hiss­ ing” (Hillman, 2009). Such approaches suppose that state efforts to tax are bounded by the amount of resources in society and the way in which wealth is produced. Certain economic activities are easier to tap into; others are more difficult.12 In poor countries, the main constraint, by def­ inition, is limited resources. Further, the economic activity that does exist is often irregular or informal, and when it is not, external actors control large portions, all of which make it diff cult for national authorities to capture any revenue. Poor countries are also plagued in that the few resources available to them are highly uncertain. Their export products are vulnerable to fluc­ tuations in price, and their f nances can be thrown completely off track 10 There are several reviews of economic theories of taxation in developing countries (Bird, 19 9 2; Thirsk, 19 9 7; Tanzi i 1992). 11 Useful reviews of political theories of tax can be found in Gould and Baker (2002) and Mahon (2004). 12 See Stotsky and Woldemariam (1997; 2002).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

13

by changes in climate, demography, and international market conditions (Newberry and Stern, 1988; Lledo, Schneider, and Moore , 2004).13 These factors are exacerbated by extensive domestic need, as resources that stay in the domestic economy are needed for consumption, leaving little to sustain government.14 The result is a vicious cycle in which poor coun­ tries are unable to meet their revenue needs, and because of this, they cannot escape poverty (Bird, 1992; Tanzi, 1992; Burgess and Stern, 1993; Tanzi, 2000).15 Current patterns of global integration in Central America add addi­ tional complications. As new economic activities appear, they increase the amount of resources available, but often in hard-to-reach sectors led by actors adapted to moving around the international economy.16 Two examples of emerging sectors are tourism and finance; tourism can locate wherever there is a modicum of infrastructure and available nat­ ural beauty, and finance is notoriously footloose. The region has further excelled as a platform for textile assembly and reexport, mostly to the United States, with firms locating labor-intensive stages of production within free trade zones in each country. These investments tend to be difficult to monitor, especially for weak public administrations, and are further complicated by incentives, special regimes, and exemptions used to attract investors in the first place. Even if governments wanted to raise revenues on these dynamic sec­ tors, there would be pressure to keep taxes on profit and income low.17 Attempts to raise taxes drive individuals and firms to vote with their feet (and their portfolios), moving activities to places in which tax rates are lower.18 This threat is particularly available in countries like those in 13 Caiden and Wildavsky (1980) call this “ complex functional redundancy.” Although their language is cumbersome, they simply refer to the practice of multiple fallback systems, quality checks, and other practices that can ensure that essential tasks are accomplished. 14 Caiden and Wildavsky (1980) and Nagel and Caiden (1996) stress the budgetary impli­ cations of these economic factors. Structural poverty simultaneously constrains revenues while increasing the needs of citizens. 15 It was for these reasons that Nicholas Kaldor famously asked whether developing coun­ tries would ever “ learn” to tax (Kaldor, 1963). 16 Much economic literature focuses on optimizing tax systems to different economic struc­ tures (Ramsey, 19 2 7; Diamond and Mirrlees, 19 7 1a ; Slemrod, 1990; Heady, 1993). 17 Whereas epistemic communities of tax advisors lie beyond the scope of the current anal­ ysis, there has been a notable shift in what was assumed to be acceptable tax policy (Goode, 1993). A comparison of Latin American tax reforms over time can be found in Rodriguez (1993). 18 On the issue of tax competition in developing countries, see Gillis (1989).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

i4

State-Building and Tax Regimes in Central America

Central America, as there is little to distinguish one country from another with respect to the comparative advantages they provide (Rennhack et al., 2004; Rodlauer and Schipke, 2005). Further, new technologies allow firms to organize production across multiple territories and transfer capital across these territories in increas­ ingly complex financial instruments. Such instruments are poorly under­ stood and constantly innovating, usually by highly paid bankers and investors whose job it is to come up with ever more complex ways to evade the revenue and regulatory efforts of much lower paid government officials (Reich, 2007). Even more slippery is the trade that occurs within firms, which now accounts for much of international trade. As f rms transfer goods and materials from one branch to another, they set transfer prices for account­ ing and budgeting purposes. These transfers have no equivalent market price, as they are inputs to a final product that is only given a price when sold outside the firm. As a result, enterprises can manage their accounting to show more income in territories where nominal tax rates are low and less income where taxes are high. The unregulated proliferation of tax havens makes this even easier, bedeviling both developing and developed country governments with bank secrecy and low regulations to assist tax avoiders and evaders (Gravelle and Library of Congress, 2009). In sum, globalized international capitalism exacerbates constraints on Central American governments when it comes to tax. Structural limita­ tions include shortages of revenues, expanding needs, and uncertainty. Sectoral complications include informalization, diff cult-to-tax sectors, and a new facility in tax avoidance as a result of f nancial sophistica­ tion. Vito Tanzi calls this combination of factors “ fiscal termites,” as they erode tax structures and force countries to depend on inadequate, regres­ sive, and distortionary tax regimes (2000). These economic constraints would seem to overdetermine weak tax regimes in Central America. Yet, all countries in the region have signif cantly changed their tax regime in recent years, in some cases even making gains that would seem surprising compared to the predictions of economic theory. Further, although the countries share a similar eco­ nomic outlook, differences have emerged, as national governments adopt varying political and policy responses. This suggests that perhaps the determinants of tax in the region are more than economic; perhaps they are rooted in the social and political context of revenue extraction. Prior attention to such factors emphasize societal, institutional, and regime determinants of tax capacity. A pluralist,

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

15

or societally oriented, approach suggests tax depends on the pressures of different interest groups. For ideological and material reasons, certain social sectors tend to favor or at least be amenable to higher levels of taxation, such as popular sectors, organized labor, and left-wing political parties (Garrett, 1998). Where such sectors are powerful, and economic elites are less so, one can expect governments to extract more tax and in a more progressive fashion. However, such explanations are not partic­ ularly useful in Central America, where popular sectors have never been particularly powerful. Even in Nicaragua and El Salvador, where cur­ rently governing parties evolved from leftist revolutionary movements, there has been only limited taxation of elites. Also, the country with the most active civil society and longest history of openness to popular sec­ tors, Costa Rica, has done perhaps the least to alter its revenue regime to respond to changes brought by global integration. O f course, even where pressure groups militate for more revenues, they do not always get what they want. They must have the institutional tools to implement their policy preferences. After all, it is through author­ itative and occasionally coercive tools that states enforce taxation, espe­ cially when it must be done over the objections of powerful opponents. Bird and Casanegra (1992) emphasize tax administration as the key component of tax policy, and innovations such as semi-autonomous tax administrations and large-payer collection divisions have been champi­ oned as tools by which recalcitrant taxpayers can be tapped (Taliercio, 2004). Yet institutional capacity seems a difficult variable to incorporate into Central American explanations, as coercive capacity was most often expanded to protect elites from popular demands for redistribution, not to raise new revenues. Other political explanations go beyond institutional capacity to make arguments about broader regime questions of democratic and authoritar­ ian regimes. Such arguments tend to find a positive relationship between democracy and tax that flows in both directions.19 Democratic elections and accountability provide legitimacy, including support for tax. And, tax revenues offer resources to the state that can be used to offer ser­ vices and benefits to legitimate democracy and make it possible to govern without excessive use of coercion (Cheibub, 1998; Boix, 2003; Moore 1 2007).20 This theory would predict an increase in revenue capacity for 19 Results are not uniform, as some studies find authoritarian regimes tax more; see Fauvelle-Aymar (1999). 20 As a corollary, where governments have other sources of revenue besides tax, such as natural resource rents, there is a greater propensity of dictatorship (Ross, 2004; 2006).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

1 6

State-Building and Tax Regimes in Central America

Costa Rica after the democratic transition of 1948 and for the rest of the countries since their transitions since the 1980s. In fact, democratic rule has coincided with only a spotty increase in revenues in the region, and least of all in Costa Rica, where one might expect it most. Moving down from the level of regimes, scholars have attempted to identify variations in the types of democracy that lead to taxation. One possibility is in the degree of engagement or legitimacy of the regime, as indicated for example in the level of participation in elections, which Boix (2003) finds associated with higher levels of taxation. Others frame tax as a collective action problem, in which economic actors would like to contribute toward needed public goods but are tempted to free ride on the contributions of others (Lieberman , 2003). In his comparison of Sweden, the United Kingdom, and the United States, Steinmo finds that proportional representation parliamentary systems do better than first-past-the-post and presidential systems to shape coalitions of orga­ nized labor, industrial elites, and politicians in support of tax capacity (Steinmo , 1993). In a later study of OECD (Organisation for Economic Co-operation and Development) countries, he found that a moder­ ate number of political parties facilitated collective action around tax (Steinmo and Tolbert, 1998).21 Yet, even these nuanced comparisons across democratic regimes offer little guidance to Central America. The countries of the region are more similar than different in the timing and types of democracies that have emerged; what differs more significantly are the internal dynamics of emerging sectors, their relationships with other groups, and the institu­ tions that guide their behavior within policy-making processes. The Argument To explain tax regimes and patterns of change in tax regimes, this book focuses on a concept of state-building that includes relationships among key social groups and their connection to and functioning within the institutions of the state. The key social group here is that dynamic fac­ tion of capital adapted to globally integrated production and markets, transnational elites. Where these elites are cohesive and dominant, they can design a coherent state-building project, including institutions to 21 Others have focused on more specific institutional arrangements, such as the relative power of the executive, the party system, parliamentary institutions, electoral rules, uni­ tary systems, and the origins of political institutions (Gerring, Thacker, and Moreno, 2005; Brautigam, Fjeldstad, and Moore, 2008).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

17

facilitate their political access while constraining opposition and policies that expand their strategies of accumulation. This requires attention, not only to the structure of rising and declining elites and popular sectors , but also to the institutions in which social forces interact with one other and with state authorities in the policy-making process. Tax regimes are the outcome of the state-building project advanced by leading capitalist sectors as they interact among themselves and with other actors while moving through and shaping policy-making institutions. This volume has provided a label for the kind of state-building project emerging in each country. These labels are intended to summarize the nature of the social coalition on which the state rests and the pattern of policy making that operates within political institutions. In El Salvador, transnational elites were led by financial sectors with investments spread throughout the Central American region. For twenty years they held together more domestically based factions of the transnational elite, such as those in assembly manufacturing for export, and advanced a strat­ egy of using state power to project their activities regionally, giving the Salvadoran state-building project the label “ inside-out.” Transnational elites were cohesive within business associations and financial networks, coordinated a state-building strategy through an autonomous think tank, and advanced that project through a business-oriented political party. The institutions in which they operated facilitated their dominance by providing privileged access to transnational elites while distracting or co-opting opposition. This book ends just as the external preference of financial elites eroded their links to other capitalist factions, weakening the very cohesiveness and dominance that had allowed them to hold onto power. It is too early to say whether their “ inside-out” state-building pro­ ject has ended; possible future directions under a newly elected opposi­ tion government will be addressed in the Conclusion to this book. In marked contrast, transnational elites in Honduras were dominant with respect to other social groups, but they remain divided among sev­ eral large conglomerates, family networks, and traditional political par­ ties. They agreed on a state-building project of attracting international firms to Honduras, with local transnational elites positioned as subordi­ nate partners to mediate access. This gave Honduran state-building the label “ outside-in,” and it was advanced by a coalition of domestic and international technocratic advisors. To coordinate them, new institutions invited international technocrats from donor organizations to participate in Honduran policy making, operating in parallel to existing institutions that concentrated on parceling out benefits to traditional political elites

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

18

State-Building and Tax Regimes in Central America

and their clientele. The implications of the 2009 coup will be explored in the Conclusion, but all indications are that the coup aborted an alterna­ tive state-building project oriented toward popular sectors and preserved the regnant “ outside-in” state-building project. In Guatemala, transnational elites are neither dominant nor cohe­ sive, accommodated to traditional sectors within a single peak business association and a fragmented and volatile party system. Between rival factions of transnational and traditional, urban and rural elites, no sin­ gle state-building project dominates, and competitors appear content to ensure that no faction can use the state to impose dominance over the rest. Institutions set high thresholds for change and encourage continued division among elites. The result is stalemate, giving Guatemalan state­ building the label “ crisis,” as no coherent state-building project has taken shape. The argument of this project is that these patterns of state-building can be understood in terms of the degree of dominance and cohesion among emerging elites and the institutions that channel actors into the policy­ making process. The project also argues that tax regimes and changes to tax regimes, in terms of capacity, universality, and progressivity, are traceable to these state-building constellations. Table 1 . 1 summarizes tax regime outputs. In El Salvador, cohesive and dominant emerging elites have advanced an “ inside-out” state-building project, and this has allowed them to achieve important gains in tax capacity and universality but only limited tax progressivity. In Honduras, dominant but divided emerging elites have structured institutions to facilitate the occasional entrance into domestic policy making of external actors, producing what can be called an “ outside-in” state-building project. Moments at which this alliance can advance a state-building agenda have produced some advances in tax capacity and even some progressivity, but these moments pass quickly and are replaced by the more normal Honduran policy-making process characterized by patronage and clientelism in the form of particularist and regressive tax exemptions, incentives, and special regimes. Finally, in Guatemala, emerging elites have been neither dominant nor cohesive, and they have found even their occasional attempts to advance a state­ building project blocked by institutions that complicate policy making. This is not to say that nothing has happened, however; the pattern of state-building that has emerged can be characterized as recurrent “ crisis” with neither a coherent state-building program, a leading social sector to advance it, nor the institutions that might make it possible to imple­ ment. This is expressed in the tax regime in perpetually insufficient funds,

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America t a b l e

19

1 .1 . Emerging Elites, State-Building, and Tax Regimes Structure of Emerging Elites

State-Building Project

Cohesion Dominance El Salvador Honduras Guatemala

High Low Low

High High Low

Change in Tax Regime Capacity

Inside-Out Outside-In Crisis

Universality Progressivity

High Moderate Moderate Low Low Low

Low Low Low

supported by temporary and stop-gap increases to avoid bankruptcy, but failing to advance either a more universal application of taxation or a more progressive collection. Methodology of Research and Comparison The evidence presented in this book was gathered through extensive field­ work in the region and interviews with leading figures from state bureau­ cracies, donor agencies, emerging elites, leaders of popular organizations, and academic observers. I spent the bulk of eighteen months during 2006 and 2007 working with the Inter-American Development Bank traveling throughout the region; I continue to return at least once a year for the board meetings of a local think tank; and I lived for an extended period in Honduras in 2007. During this period, I worked in the Secretariat of the Presidency, loaned by the Inter-American Development Bank using a trust fund provided by the UK Government and the Swedish Government. This was privileged access to the top of the policy-making world. In particular, the staff of the Inter-American Development Bank and other donor agencies provided what amounted to diplomatic reports on their view of economic associations, governments, and individual politi­ cians in each of the countries. These observations were relayed in con­ versations over coffee and meals and were often filled with a surprising amount of disdain for country counterparts. “ Why won’t they do what we tell them will work?” was the gist of donor community complaints. Rarely did donors acknowledge that advice given at one point, “ Close down the Secretary of Planning,” was reversed only a few years later, “ You really need a Secretary of Planning!” 22 From my vantage point in 22 The foibles of the international development community are reviewed in two books titled similarly, “ Globalizers,” one by Woods (2007) and the other, focused on Honduras, by Jackson (2005).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

20

State-Building and Tax Regimes in Central America

a government ministry, it was abundantly clear that Central American leaders were fully aware of the two-faced games donors play and were largely happy to play along with games of their own. The circles of donors and government officials also provided access to financiers and industrialists, who spoke directly and honestly about their approach to politics and views of taxation. Some of these views were particularly uncensored. One research associate reported that a Guatemalan industrialist described his view of taxation while conduct­ ing the entire interview with a pistol sitting on his desk. I led a heated group discussion among Salvadoran bankers in which a leading financier proclaimed, “ Income tax is a penalty on success; and, in my opinion, it should be eliminated.” In a private conversation in a lavish office, a Honduran beverage magnate explained that the problem with revenue was that Hondurans drank too much beer and not enough more highly taxed alcohol, or maybe it was the other way around. These conversa­ tions provided a unique view into the leading sectors of Central American society, warts and all. Whereas access to policy makers and economic elites was easily obtained through the diplomatic channels of donors and their counter­ parts, popular organizations were another story. The leaders of leftist parties, labor unions, and social movements were wary of research con­ ducted with the imprimatur of the international community. To secure interviews with popular actors, I mobilized a different set of contacts from acquaintances and counterparts cultivated in almost two decades of solidarity work in the region. In 1993 and 1994, I had lived among demobilized guerrillas in El Salvador, forming durable bonds of friend­ ship, information, and refl ection. When told of the nature of the project, most people in these networks insisted that their perspective was neces­ sary to correct the picture painted by government officials, donors, and economic elites. Additional assistance in each country was provided by local experts.23 They facilitated contacts, conducted some of the interviews, and gathered primary data that would have been difficult for a foreigner to obtain. Jorge Vargas Cullel opened a particularly useful avenue by collecting the universe of tax legislation changes that occurred over the last twenty years in Costa Rica. Leticia Gonzalez collected similar data for Guatemala and wrote a helpful background paper. I collected similar data for Honduras 23 Assistance was provided by Leticia Gonzalez (Guatemala), Alex Segovia (El Salvador), Francisco Baez (Nicaragua), and Jorge Vargas Cullel (Costa Rica).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

21

and El Salvador, and categorized the legislation in terms of the statesociety relationships indicated by different policy-making processes. This methodology is discussed in the Appendix. To make sense of these observations and test hypotheses about the determinants of tax regimes, this study adopts a series of comparisons. The first comparison is within countries, between two contrasting reform patterns. In each country, I selected one reform as a positive example, in the sense that it increased revenues, made the revenue regime more uni­ versally applied, and/or increased equity. I also selected a second reform that did the opposite; it decreased revenues, introduced exceptions to the revenue regime, and/or increased inequity. By selecting cases of positive and negative change within each country, the comparison holds constant the host of country-specific variables that might influence tax regime changes, and it focuses on the micro-level differences that explain success or failure of reforms within state-building projects. Of course, these specifically selected reforms were not necessarily rep­ resentative of all of the changes that occurred in the countries, and the project also explored the universe of tax legislation changes that occurred. This ranged from 70 laws in Honduras to 140 laws in El Salvador. Quantitative analysis of these legislative changes allows for another ana­ lytical strategy in which I categorize reforms in terms of what they tell us about state-society relations in the process of policy making and what that tells us about the link between state-building and tax. The Appendix discusses the methodology for constructing that database. These within-case comparisons draw attention to the micro-level dynamics driven by social and political actors as they tinker with tax regimes, the machinations of day-to-day politics. This is the stuff of pol­ icy making, and this book uses this information to ask questions about what gets done in tax reform and what does not. In this approach, the project operates with the microeconomic logic common to many stud­ ies of policy making, in which individuals and groups maximize their interests in response to the incentives and constraints presented by insti­ tutional rules and structural conditions.24 This micro-level approach identifies potential institutional tweaks and strategies by which desired reforms might be made more likely. This part of the enterprise addresses the question that has dogged reformers everywhere - what increases the likelihood of reform? 24 For a good example applied to public finance, see Hallerberg, Scartascini, and Stein (2009); and, for a variety of policy areas, see Nelson and Kaufmann (2004).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

22

State-Building and Tax Regimes in Central America

Yet, the micro-level approach alone would miss the character of state-building processes as more than the sum of their parts. It is not enough to study interest maximization by transnational elites as though group interests were simply the aggregation of individual inter­ ests; group entities interpret complex realities and frame the interests of members.25 This macro logic is important because it defies simple assumptions about interests and behaviors and suggests a need for close understanding of the way interests are formed, shaped by pat­ terns of organization and historical legacies, negotiated through social relationships, and adapted within institutional boundaries, and the w ay interests construct new institutions. This project is thus at once micro-level, in its concern for the strategic behavior of individuals and groups, and macro-level, in its concern for the structural forces that constitute social actors, form interests, and construct institutions to condition behavior. To pursue analysis at this macro level, comparison across country cases controls for regional similarities. Despite relatively similar social structures and emerging sectors, state-building projects vary significantly across countries. By comparing across countries, the project can explore the macro logic of state-building, as organizations, institutions, and inter­ action construct interests and create different state-building agendas. Comparison across cases allows the project to test the impact of variation in the nature of state-building on tax regimes. Additional cross-country comparison includes historical and statis­ tical techniques. Historical comparison makes use of archival materi­ als, secondary sources, and historical data compiled by Oxford and the Economic Commission of Latin America. These allow for a tracing of the evolution of tax regimes across the region, identifying moments of major change. The historical comparison tests whether state-building efforts by elites of prior eras were connected to moments of tax reform and why such reforms emerged in some places but not others. The second cross-country comparison is quantitative. Contemporary data on economic change, political representation, state institutions, and fiscal outcomes are available from cross-national databases. This project brings these databases together and situates the Central American cases in their regional and international context, with particular attention to

25 These approaches have received discussion in terms of structure versus agency (Mahoney and Snyder, 1999) and ontological and methodological individualism (Della Porta and Keating, 2008: 2).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

Revenues, States, and Central America

23

the most recent period when comparable data for large numbers of coun­ try cases have become available. Structure of the Book After this introduction, the chapters that follow address these compari­ sons. Chapter 2 explores the concept of state-building and places Central American political development in the context of changes to the world system of capitalist development. The latest manifestation of interna­ tional capitalism, patterns of economic integration labeled globalization, has called forth a new cast of Central American political and economic actors, constrained as always by their roles as peripheral participants in the global economy. The chapter explores the meaning of state-building as it applies to different parts of the globe and different moments of his­ torical time. This sets the stage for an exploration of state-building as it is occurring in Central America, with variations produced by the structure of emerging elites, their relationships with other sectors, and the institu­ tions of the state itself. Chapter 3 begins the empirical investigation, first addressing the his­ tory of tax regimes in the region, exploring moments of major historical change for evidence of emerging elites and state-building projects. Prior insertions into the international economy have produced socioeconomic changes as momentous as those occurring today, offering an opportunity to explore history for lessons about patterns of divergence. Chapter 4 brings together a series of different quantitative databases to compare recent socioeconomic, political, and f scal trends in Central America as well as how the region compares to other parts of the world. The chapters that follow examine case studies of three of the f ve countries of Central America. Each chapter is structured around the same sequence, with a section on actors, a section on institutions, and a sec­ tion on processes. These variables characterize the state-building project occurring in the countries and trace the connection between patterns of state-building and tax regime outcomes. Of particular interest is the orga­ nization of emerging elites and their relationship to other sectors, as well as the institutions in which they interact and the processes they undertake in producing policy outcomes. The concluding chapter takes the opportunity to examine patterns revealed in the within-country and cross-country comparisons as well as to outline additional avenues for research and potential future policy les­ sons. The three case studies are qualitatively different, shaped by unique

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

24

State-Building and Tax Regimes in Central America

national histories and trajectories, and they have generated differing tax regime outcomes. The one outcome none of these state-building patterns produced is a tax regime with more capacity, more universal applica­ tion, and more progressivity. In characterizing existing state-building patterns and their tax regime outcomes, this book identifies the miss­ ing state-building balance of class forces, institutional arrangements, and reform processes that could generate progressive, universal, and capable tax regimes. It is to this effort that this book hopes to contribute.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.003

2

State-Building in a Globalized Political Economy

Contemporary state-building has to be understood in the context of recent transformations of capitalism, in which economic integration has acceler­ ated rapidly to reach a global scale. Rapid economic change brings forth new economic actors, especially those adapted to transnational processes of accumulation, and they operate alongside and in competition with previously established actors and popular sectors. The task of accommo­ dating these groups and promoting further accumulation falls to state institutions. To manage these challenges, state institutions inherited from prior periods of Central American history have been inadequate, and this book explores patterns of state-building for their impact on taxation. This chapter proceeds in four sections. The first section addresses the concept of globalization as a stage in the evolution of capitalism, charac­ terized especially by the integration of economic activity across multiple territories and the emergence of transnational actors adapted to this type of accumulation. The next section addresses the concept of state-building as it has been viewed in the literature and what it means for the trans­ national elites of Central America. Next, the chapter considers the link between state-building and revenue, rooted in theories of fiscal sociology. The last section develops a framework to explore the dimensions of state­ building in the region. Globalization State-building challenges have to be understood in the context of a changing international economy that has expanded, integrated, and transformed over centuries (Wallerstein. 1974). This process has moved 25 Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

26

State-Building and Tax Regimes in Central America

in fits, starts, and occasional reversals, characterized best by “ punctuated equilibrium” in which long periods of gradual change are interrupted by moments of rapid transformation (Eldredge and Gould, 1977). These transformations alter patterns of production, distribution, and consump­ tion that are marked by shifting patterns of accumulation. The transformations that occurred in the 1970s shifted world capital­ ism toward a globalized pattern of liberalized trade and capital markets and integrated global production processes. This liberalization and glob­ alization were associated with a technological shift in the emergence of high-speed information and transportation flows; an international regime shift with the breakdown of Bretton Woods arrangements; and a pol­ icy shift in removing many national-level regulations (Marglin and Schor, 1990; Eichengreen, 1995; Harvey, 1999). The process of economic inte­ gration that this produced brought more people, from farther regions, into greater contact with one another in the process of capitalist production. As a concept, this process entails a more intensive application of capital­ ist social relations - liberalization - to a more extensive number of places and people - globalization (Arrighi and Silver, 1999; Arrighi, 20 10 ).1 The impact on production processes was a shift toward globalized chains. Stages of production cross borders as productive chains extract raw materials, transform them into intermediate goods, assemble them into finished products, and distribute them to final markets where they are consumed. These processes need not occur in close geographic prox­ imity, and firms reorganized themselves to locate production in the loca­ tions that provided the greatest returns. The result for firms is that they can adapt management techniques to respond more quickly to market demands. Firms set up production chains with tight turnarounds, minimal inventories, and quick outputs, so production can occur “ just in time” for each link in the production chain. Workers are trained and machines are designed to be specialized but flexible, allowing them to switch quickly from production of one out­ put to another to meet rapidly changing consumer tastes (Zysman, 1984; Piore and Sabel, 1986). The entrepreneurs and elites who adapt to these requirements are qual­ itatively different from those emerging in prior moments of capitalism.1 1 The process is of course partial and political. Liberalization has occurred for some coun­ tries but not others; for example, the United States and Europe retain agricultural subsidies and protections while requiring developing countries to remove theirs. In addition, there are large parts of the world that remain significantly outside of globalized production and exchange, such as parts of sub-Saharan Africa.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

27

As they operate across borders in integrated production chains, they are no longer bound by residence or productive activities located in one ter­ ritory. This differs from periods in which capitalists operated in several markets, but they organized their activities from start to finish in pro­ cesses limited to one jurisdiction at a time. For example, a Ford factory could be built from bottom to top to produce cars in several countries, making it a national or multinational concern. The greater integration of globalization now allows them to operate transnationally, making parts in one country to be assembled in another and sold finally in yet another (Gereffi, 1990; Dicken, 2007). Through the interpenetration of national capitalist sectors, cross-border mergers and acquisitions, interlocking of financial networks, and strategic alliances among business associations and networks, they have produced a transnational class of emerging elites. Liberal observers posited that the transnational character of these social actors would eclipse the nation-state, as they outgrew state institu­ tions as either unviable or unnecessary (Williamson, 1990a; Friedman, 2007). Shifts in the international regime, in the form of treaties and new institutions such as the WTO, appeared to confirm the increasing constraint on regulatory power and the role of states, while structural adjustment programs imposed limits on the policy space of individual governments (George, 1990). For critical globalization scholars, “ economic integration processes and neoliberal structural adjustment programs are driven by transna­ tional capital’s campaign to open up every country to its activities, to tear down all barriers to the movement of goods and capital, and to create a single unifed feld in which global capital can operate unhindered across all national borders” (Robinson, 2008: 19). This view holds that owners of capital had removed their obligations to the state or to other members of society and could simply pursue accumulation wherever and however they wanted (Amin, 1994). This stateless and apolitical version of globalization misses the contin­ ued relevance of national political patterns. In many national contexts, the legacy of prior institutional arrangements has proven durable. The character of nation-states has not been so easily brushed aside, and there is signif cant room for divergence, even as economies integrate. These national particularities remain evident in enduring institutions and politi­ cal practices, indicated by wide varieties in the size of states, the structure of welfare systems, and the regulation of work and environment (Evans, 1997; Hall and Soskice, 2001).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

28

State-Building and Tax Regimes in Central America

In addition, there remain unique historical functions necessary to manage globalization (Rodrik, 1996).2 The functions of states have not so much disappeared as transformed to deal with challenges of eco­ nomic integration. State leaders compete to attract stages of production to their borders, offering incentives that attract footloose capital. The difficult task is to ensure that the stages of production attracted are high value-added processes with linkages to other sectors and firms, such that they allow for decent livelihoods and sustainable development (Reich, 1992). Thus, the question is not whether the state has disap­ peared, but rather the nature of the states being built in the context of globalization. This calls into question the character of emerging, transnational elites in each country. They are the bearers of new state-building agendas, but they do not enjoy the same economic or political success in each place. They will be more dynamic in some places than others, and their strate­ gies for international integration can signif cantly differ. This introduces national particularities into the content of their state-building agendas as they attempt to hold together diverse factions of transnational elites, encounter rivals, remodel national institutions, and pursue distinct polit­ ical strategies. In Central America, the homogenizing external pressure of interna­ tional integration and the persistence of national particularities are both evident. The changes associated with globalization were accelerated by the 1980s debt crisis and revolutionary upheaval, ushering in govern­ ments favoring neoliberal reforms. With slight variations, the region followed the rest of Latin America in adopting the liberalization, deregu­ lation, and privatization policies that characterized structural adjustment (Lederman et al., 2006). These were locked into trade arrangements, such as the Caribbean Basin Initiative, begun during the 1980s, and the Central American Free Trade Agreement of 2006, which gave producers in Central America access to the U.S. market. This attracted Central American entrepreneurs into new export activities and attracted U.S. and thirdcountry investors to locate in the region, creating and empowering those emerging elites who could ally with them to advance transna­ tional processes of production. The governments of Central America competed to enhance their attractiveness to these sectors by creating 2 Even the World Bank acknowledged a new set of requirements of the “ State in a Changing World” in its 1997 World Development Report (World Bank, 1997).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

29

special tax and regulatory regimes that held down production costs while offering geographical proximity and trade access to developedcountry markets. This process has played out in Central America in specific ways. The sectors most attractive to U.S. markets have been tourism, final assembly manufacturing (called maquilas in Spanish), nontraditional agriculture , and remittances. In addition, financial sectors and other services have prospered by acting as intermediaries between local interests and trans­ national firms (Segovia , 2005). As these sectors expand, the emerging elites who lead them devise unique strategies of accumulation and objec­ tives for policy. To these ends, they seek access to the state and the insti­ tutional changes that can facilitate their state-building efforts. The question posed by this book is: What kind of state-building project have these transnational elites devised in the context of globalization? Is there a difference in the content of that project across countries, and to what degree is it being advanced in a coherent fashion? What factors in the political and institutional environment shape their success, and how are these expressed in differences in the tax regimes of the region? Before turning to these questions in more depth, it makes sense to consider the concept of state-building itself. State-Building This book considers state functions under globalization as an opportunity to reevaluate previous conceptualizations. Most traditional approaches to state-building rest on two assumptions. First, state-building is some­ thing that occurs with the establishment of national, sovereign authority over a territory, generally at the founding moment of a national unit, in which boundaries and national myths are set (Gellner, 1983; Mann, 1986; Poggi, 1990). Second, state-building includes those efforts to approxi­ mate or move toward a Weberian ideal (Evans and Rausch, 1999). Much of our understanding of states begins with M ax Weber, who defined the state by its unique power of coercion within a territory, “ [It is] a ‘state’ if and insofar as its administrative staff successfully upholds a claim on the monopoly of the legitimate use of violence in the enforcement of its order” (Weber, 1968: 154). For early modern states, this was a process of asserting power, especially military power, by central governments over principalities, fiefdoms, and warlords (Ziblatt, 2008).3 3 For an application to South Asia, see Slater (2010).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

3o

State-Building and Tax Regimes in Central America

Weber also emphasized the institutionalization of the state through the establishment of rational-legal authority, removing the crutch of tra­ ditional or charismatic legitimacy. Under rational-legal authority, pro­ fessional politicians establish the legal authority of a meritocratic and impersonal bureaucracy, resting on “ a belief in the ‘legality’ of patterns of normative rules and the right of those elevated to authority under such rules to issue commands (legal authority)” (Weber, 1947: 328). In the context of capitalist modernization, the bureaucracy manages the challenges of public administration through the application of rules, science, and rationality. This replaces the communal values, religious prin­ ciples, and traditional practices that might have guided decision making under premodern arrangements. As a more effective and efficient mecha­ nism of organizing authority, Weber believed that rational-legal bureau­ cracies would eventually outlive and outcompete alternative ones. Al Stepan specifies additional dimensions focused on the nature of state linkages to groups within society. These are important to carry interests into the state, such as through mechanisms of representation, and also to allow the state to shape the social groups and cleavages that exist in soci­ ety. The state, for Stepan, encompasses “ the continuous administrative, legal, bureaucratic and coercive systems that attempt not only to struc­ ture relationships between civil society and public authority in a polity but also to structure the many crucial relationships within civil society as well” (Stepan, 1978: xii). Mann refers to this as “ infrastructural power,” the capacity to penetrate society and make effective the intentions of the state (Mann, 1986). To these domestic efforts, state-builders must also look outward, as “ states are inherently Janus-faced, standing at the intersection of interna­ tional and domestic processes” (Evans, Reuschemeyer, and Skocpol, 1985: 350).4 As Skocpol notes, international “ contexts impinge on individual states through geopolitical relations of interstate domination and com­ petition, through the international communication of ideals and models of public policy, and through world economic patterns of trade, divi­ sion of productive activities, investment flows, and international finance” (Evans, Rueschemeyer, and Skocpol, 1985: 8). This suggests two distinct mechanisms of pressure from the external environment. The first is in terms of direct challenges or infl uences on national state authority, as in the form of war, economic competition, and policy diffusion. The second 4 This draws on Otto Hintze, who traced the nature of domestic regimes to external mili­ tary threats (Hintze, 1975).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

31

mechanism shifts the domestic social structures and social coalitions that provide support to national states, for example as a result of socioeco­ nomic changes produced by international relations of trade, production, investment, and finance. The internal and external dimensions of states have been incompletely incorporated into our conceptualization of state-building. In part, this is because contemporary state-building discussions are limited to formative moments of nationhood, directing attention to centuries-old processes in Europe (Tilly, 1992), and are only extended to limited contemporary contexts. These include apparently failed states in places like sub-Saharan Africa (Doornbos, 2002; Ghani and Lockhart, 2008) and attempts by external actors to rebuild states after conflict and invasion in places like Iraq and Afghanistan (Kalyvas, Shapiro, and Masoud, 2008). This book suggests that state-building tasks are not reserved to for­ mative moments of nationhood; rather, state-building occurs repeatedly, triggered by major shifts in international capitalism. Economic shifts stimulate domestic sectors, and they press for political access to advance a policy agenda that will help them sustain their dynamism, including administrative adaptations and the construction of new political institu­ tions, where necessary, as well as creating new state-society linkages to link themselves to the state and manage relations among rising, falling, and excluded groups. Successful contemporary state-building generates power and resources that can protect emerging sectors where they are vulnerable and help to insert them into transnational processes of accu­ mulation where they are not. A combined external-internal approach is especially important in the context of contemporary liberalized markets in Central America. Global integration has accelerated changes in domestic actors and placed new and more sophisticated requirements on state institutions, but the capac­ ity of political elites and state institutions to accomplish these tasks is by no means assured. National territories remain pocked with spaces and policy areas in which governance is incomplete, and the recent interna­ tional financial crisis has shaken the foundations of ongoing integration processes (O’Donnell, 2004).5 The crisis has exposed, once again, funda­ mental ideological and political struggles over the appropriate role for the state.

5 Further, the assumption that state-building occurs only at foundational moments creates the misguided impression that it occurs on a blank slate, wiped clean with each invasion or regime change.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

32

State-Building and Tax Regimes in Central America

These struggles are expressed in Central America in terms of competing state-building agendas. One such agenda draws on liberal traditions and seeks to capitalize on moments of crisis to roll back the state, removing it from economic functions, at once resolving crisis and preventing the state from intervening in the economy in the future (Klein, 2008). This view orients some potential state-builders in Central America, who are drawn to the liberal view out of both conviction and fear. They are convinced that the appropriate role for the state is a minimal one, and they are even more fearful of the populist alternative, in which short-term ameliora­ tion of crisis becomes a long-term threat to managing limited resources (Corrales, 2002).6 In the past, this fear of populism has motivated elites and middle classes to support authoritarian solutions (Avritzer, 2002). The room for authoritarian responses has been limited in recent years, and domestic elites have sought democratic alternatives, focusing on plu­ ralism as a manageable form of democratic opening. Under pluralism, politics is limited to a competition among interest groups, directed by political institutions and mechanisms of social control to protect minor­ ities from the whims of the majority (Dahl, 19 7 1).7 The kinds of insti­ tutional rules this has included in Central America are electoral rules to limit majorities and powerful constitutional courts, not to mention the implicit veto power held by militaries over legislative and executive decisions. As a complement to pluralism, technocratic insulation removes pop­ ular sectors and democratic process from core areas of decision making (Centeno, 1997). Under such arrangements, democratic competition is limited to noneconomic areas, with areas essential to capitalism, such as finance or industrialization, placed in the hands of a protected bureau­ cracy (Dominguez, 1997).8The residue of such practices litters the region with semi-autonomous institutions and bureaucracies granted indepen­ dence from party politics and executive discretion, only to find them­ selves abandoned or absorbed once policy makers lose interest or move on to other challenges (Geddes, 1996). 6 This echoes fears of public-choice theorists that expanding state mandates during crises creates self-interested bureaucrats who protect themselves and their lobbies even after the crisis has passed (Friedman, 1990; Brennan and Buchanan, 2006). 7 Mechanisms of social control include using the state to demobilize popular mobilization and prevent organization (Kurtz, 2006). 8 The neoliberal fear of self-interested bureaucrats seems at odds with the faith in techno­ crats. This has been resolved by Weberian assumptions about the role rational and merit­ ocratic bureaucrats play as shepherds to capitalist expansion, rather than inherent threats to it (Evans, 1995).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

33

Some conflicts over state-building are waged within the state itself (Hirsch, 1996). Conflicts within and among branches of the state and the executive bureaucracy express these internal struggles for power, dislocat­ ing ideological and political struggle from society into state institutions as competing factions within the political elite align with alternative state­ building agendas.9 Such moments of intraelite conflict have appeared in recent years in Central America in the 2009 coup in Honduras and the near collapse of the government in Guatemala . What has not yet occurred is a pattern of state-building that renegoti­ ates international patterns of economic insertion while also reorganizing domestic political and economic relationships. This is what happened in the developmental state models of East Asia, and to a lesser extent in the statist development models of post-World War II Brazil, Mexico, and Argentina (Evans, 1995).10 These efforts included institutional charac­ teristics of modernizing bureaucracies, especially in terms of meritocratic and rational-legal orientation (Amsden, 1992); changes in the relationship of the state to leading social sectors, especially dynamic, export-oriented ones (Evans, 1995); and increased state ability to trade on geopolitical position to renegotiate the terms of international engagement, for exam­ ple by securing markets for exports and upgrading technology (Wade, 2003b). Peter Evans describes this as a triple alliance between the state, domestic capital, and international capital (Evans, 1979). Such an alliance would require a domestic balance of class forces that has not appeared in Central America. Galbraith, in describing U.S. capi­ talism, coined a term to describe such a balance: “ countervailing power” (Galbraith , 1993). His focus was the postwar industrialization boom, in which large-scale business organizations confronted large-scale unions, and the balance of class power produced a state capable of reproducing accumulation while satisfying popular needs. Such an arrangement was appropriate to the nature of Fordist industrialization and even had its imperfect reflection in the national populist models of Latin American developmentalism. Yet, transnational Central American elites have been preoccupied with the policies that can stimulate their expansion and smooth their profits, 9 Fiscal crisis is taken by some to indicate fundamental conflicts over the goals of the state and the inability to accommodate competing visions within a limited fiscal envelope (Swedberg, 19 9 1). 10 He observed that even in the best of circumstances, Latin American states tended to be more “ predatory than developmental,” stifling dynamic sectors rather than “ shepherd­ ing” them toward sustainable patterns of development (Evans, 1995).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

34

State-Building and Tax Regimes in Central America

seeking state assistance to insert dynamic sectors into rapidly chang­ ing transnational processes. This includes state subsidy to research and development, promotion of dynamic sectors, and renegotiation of rela­ tionships with international investors, consumers, and creditors, among others. Less emphasis has been given to the reproduction of popular sec­ tors, which have long pent-up demands for decent livelihoods, obtainable only through more forceful state regulation of the markets in which they operate. The room to manage elite and popular demands in peripheral econ­ omies like those of Central America is more difficult and qualitatively different. The pattern of accumulation that has taken hold in the region produces too little wealth, in too few hands, through sectors with lim­ ited potential for sustained development (Agosín, Machado, and Nazal, 2004).11 The resources they generate are hardly sufficient to satisfy popu­ lar demands, the requirements of renewing accumulation, and sustaining the state. Nor have the social formations produced by the patterns of growth occurring in the region created the coalitions that might provide a countervailing balance of class forces. Instead, state-building projects have been defined by the nature of internal and external relations of emerging elites, with only limited input from popular sectors. This book explores the differences in state-building projects in the region and traces their impact on tax regimes. This places the study in that body of literature that moves fiscal affairs, tax in partic­ ular, to the center of social analysis. Fiscal Sociology That literature has been labeled fiscal sociology, and in 19 18 it was elo­ quently trumpeted by Josef Schumpeter, economic historian, philosopher, and one-time Austrian finance minister, who noted the complex link between the economics of tax and broader social processes: The spirit of a people, its cultural level, its social structure, the deeds its policy may prepare - all this and more is written in its fiscal history, stripped of all phrases. He who knows how to listen to its message here discerns the thunder of world history more clearly than anywhere else. Most important of all is the insight which the events of fi scal history provide into the laws of social being and becoming1 11 Writing within the confines of economic theory, Chu, Davoodi, and Gupta (2000) note the intersection of economic distribution and problems of state capture, referring implic­ itly to structuralist traditions associated with dependency and underdevelopment (Frank, 19 78 ; Amin, 1977).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

35

and into the driving forces of the fate of nations, as well as into the manner in which concrete conditions, and in particular organizational forms, grow and pass away. The public finances are one of the best starting points for an investigation of society, especially though not exclusively of its political life. The full fruitful­ ness of this approach is seen particularly at those turning points, or better epochs, during which existing forms begin to die off and to change into something new, and which always involve a crisis of the old fiscal methods. This is true both of the causal importance of fiscal policy (insofar as fiscal events are an important ele­ ment in the causation of all change) and of the symptomatic significance (insofar as everything that happens has its fiscal reflection). (Swedberg, 19 9 1)

Public finances are causal and symptomatic. Revenues are necessary to build states; the act of gathering revenues structures societies; produc­ tive societies and capable states generate revenues. To make sense of the specific causal mechanisms that link revenues, social structure, and state institutions, comparative political science directs attention to the political processes by which tax regimes are built. These political mechanisms follow two overarching logics, rooted in the concepts of capacity and consent (Levi, 1989: 1 2 1 ; Brautigam, Fjeldstadt, and Moore, 2008). Capacity arguments focus on the bureau­ cratic, technical, and authoritative capacity of state elites. Their capacity emerges over time and is defined in part by bureaucratic effectiveness, mil­ itary power, and the degree of autonomy from social pressures (Skocpol, Evans, and Rueschemeyer, 1985). Where states are sufficiently capable, they can extract funds from social actors, even wealthy ones whose polit­ ical leverage might be significant. Impulses to expand capacity and secure more revenues often come at moments of extreme pressure, such as when states face an external threat from a neighbor or an internal threat of insurrection (Tilly, 19 7 5; Ertman, 1997).12 This tradition places war at the heart of greater tax efforts. Drawing on the concern with external pressure that oriented state-building theorists, militarist theory “ argued that military competition and the development of taxation went hand in hand. ... In the classical version of this theory as expounded by Schumpeter, taxation was the key to the rise of the state, because taxation furnished the resources that allowed states to make war and eliminate their competitors” (Martin, Mehrotra, and Prasad, 2009: 9 -10). This functionalist logic explains revenues by the wars they are meant to finance, noting that “ those that did not would be crushed on the 12 For some, this occurs when states are backed by popular sectors, turning to confiscatory forms of raising revenue from the wealthy, although this is certainly not the case in any of the Central American cases.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

36

State-Building and Tax Regimes in Central America

battlefield and absorbed into others” (Mann, 1986: 195). As evidence for this link, Wiseman and Peacock (196 1) note the coincidence of wars and increases in expenditures and revenues in U.K. history, and Webber and Wildavsky (1986) do the same for the Western world in general.13 More cynically, some emphasize the inherent desire of political elites to expand resources and power (Levi , 1989). Public choice theories simply assume that rulers maximize revenues (Brennan and Buchanan, 2006). In some cases, this is a matter of duping taxpayers: “ Conceal the burden of taxation and exaggerate the benefits of public spending” (Buchanan and Tullock, 1962; Martin, Mehrotra, and Prasad, 2009: 8-10). This tradi­ tion spawned a generation of public-choice economists skeptical of taxes and expenditure, casting doubt on their value for money and pointing to the self-preservation instinct of bureaucrats as part of the explanation for different tax structures. Not all state efforts to expand revenues are successful, however; they depend on the consent of the governed. Consent arguments come in two forms, either suggesting that states must enter into direct bargaining rela­ tionships with contributors or that tax is a result of collective action among social groups. According to bargaining arguments, legislatures and other sites of state-society communication allow for the negotiation of a fiscal contract in which social groups negotiate with the state over obligations and benefits. Under this model, states will vary significantly in the services they provide, depending on the groups from which they mobilize revenue (Timmons, 2005). Tax is akin to an exchange - contri­ butions for benefits - in which the state “ sells” services to citizen “ con­ sumers,” who provide tax payments. Some see this exchange as a more diffuse transaction in which gov­ ernments provide universal public goods, such as democratic participa­ tion, transparent decision making, and effective implementation. Such goods are public because they are nonexclusive and nonrivalrous; once established, everyone benefits, whether or not they contributed to pay­ ing for them. Public goods are notoriously underprovided, as they must overcome the collective action problem of preventing free riders (Olson, 1965; Ostrom, 1990). According to this view, taxation is a collective action problem of getting individual taxpayers to pool resources to pur­ sue social ends (Lieberman , 2003). Over time, state-society relations that sustain tax contributions, pol­ icy benefits, and collective action generate deeper and longer-lasting 13 Mann (1986: 486) makes a similar observation for the U.K.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

37

legitimacy and attachment. This is encapsulated in a culture of tax, in which citizens identify tax payment with their membership in a society, their acquiescence and participation in representative government, and the pursuit of collective goods (Slemrod, 1990; Torgler, 2005). These connections provide the outline for fiscal sociology, in which revenues, social forces, and state institutions are understood as closely related phenomena. Early attempts to connect these phenomenon drew from cameral sciences, or chrematistics, which link the wealth of states and satisfaction of societies - “ the wealth and happiness of states and thereby the people living in the house of the state” - to more general pro­ cesses of modernization and development.14 Rudolf Goldscheid, a contemporary of Schumpeter, laid the outlines of fiscal sociology by placing fi scal history in the context of broader social processes, the “ gradual expropriation of the state” (Goldscheid, 1958: 203). Under feudal arrangements, medieval nobles owned most of the land, the main productive resource. As these nobles were part of the state, in its premodern form, the wealth of the state was the same as the wealth of the economy and society (Anderson, 1989: 195). Yet, mismanagement, crisis, and especially war strained the capacity of nobles. To generate additional revenues, they ceded to private actors opportunities for wealth creation in the form of land, feudal monopoly, and commercial opportunity. Eventually, it was the private sector that became the overwhelming source and location of wealth, forcing leaders to develop sophisticated bureaucracies and representative mechanisms to coax or coerce tax contributions to the state. Goldscheid traces this history to note that state efforts to generate rev­ enues do not necessarily conflict with the expansion of the private sector, as assumed by most liberal economic and political observers of tax who argue that state efforts to accumulate wealth necessarily “ crowd out” the private sector by withdrawing resources from private uses (Musgrave and Peacock, 1958; Barro, 1997). Rather, Goldscheid noted that expanded state capacity and increased private wealth go together, especially under modern capitalist relations. As commerce and industry expanded, privately generated wealth was both more difficult to extract and also more propitious to building state capacity. To begin, commercial and industrial capitalists generated great wealth, more easily identifi ed and tracked as it moved through major 14 The impetus to these studies came from King Frederick William of Prussia, in 17 2 3 , who sought the systematic analysis of public finance (Backhaus, 2002).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

38

State-Building and Tax Regimes in Central America

ports and concentrated in major cities. To get an equal amount of revenue from rural areas required either great extension or agricultural moderni­ zation, or both (Moore , 1966). “ On the whole, taxes on land were expen­ sive to collect as compared with taxes on trade, especially large flows of trade past easily controlled checkpoints” (Tilly, 1985: 18 2).15 On the other hand, emerging capitalists had the advantage of mobil­ ity, which meant that they could resist tax contributions by shifting their sites of accumulation. If rulers wished access to private wealth, it would have to come with certain guarantees. The first of these was some degree of oversight of the use of revenues, most clearly through elected legisla­ tures. Representation in such legislatures began with holders of property, who used the legislative power of the purse to restrain military adven­ tures, limit excessive spending, and force more effective governance by the nobility (Moore , 2004).16 A second guarantee frequently sought by holders of wealth was the repayment of their contributions, which they offered in loans rather than taxes. Loans at one period indicate taxes in future periods to repay the loans, making loans almost as good as taxes when it came to checking executive authority. Loans had another advantage to creditors, which is that they were repaid with interest, greatly enriching financiers. Fortunes could be made by bankrolling the military adventures of the state, and expanding military power turned out to be a great impetus both to state capacity and private finance capital. In the United States, it was precisely this insight that led Alexander Hamilton to encourage federal absorp­ tion of subnational government revolutionary war debts, thereby secur­ ing credibility for the fledgling republic while concentrating wealth in the hands of urban capitalists who could drive industrialization, “ These are the great and invaluable ends to be secured, by a proper and adequate provision, at the present period, for the support of public credit” (U.S. Department of the Treasury, 1790).17 15 Landed elites had more power to resist, and rural popular sectors either fled or used feudal institutions to prevent extraction beyond what was considered legitimate (Scott, 1976). 16 The most effective check on central power came from participative local institutions in which diverse local and sectoral interests were aggregated to resist absolutism, offering taxation and state-building only on the condition of “ constitutionalist power-sharing” through parliamentary democracy (Ertman, 1997). 17 According to Goldscheid, the state shifted from feudal dominance to capitalist penury, “ Thus, exploitation by the state ended up in exploitation of the tax state” (Goldscheid, 1958: 205). This was paradoxical, as it was the state that created financial wealth, “Always the capitalists have needed the State to establish profits on a really grand scale, never have

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

39

The lesson from fiscal sociology is that a more productive and dem­ ocratic society rests on public finances. “ Only a rich state can be a just state,” wrote Goldscheid, calling for close analysis of the interplay between revenues, social forces, and state institutions (Goldscheid , 1958: 210). This nexus offered the potential for interdisciplinary exploration of core social science processes and stimulated Schumpeter to proclaim in 19 18 , “ In some historical periods the immediate formative influence of the fi scal needs and policy of the state on the development of the econ­ omy and with it on all forms of life and all aspects of culture explains practically all the major features of events” (Swedberg, 19 9 1). For most of the last century, however, the study of fiscal sociology has been frag­ mented among distinct theoretical and disciplinary concerns (Martin, Mehrotra, and Prasad, 2009).18 To bring these traditions back together, the “ new” fiscal sociology “ abandoned blanket contrasts between tradition and modernity and the search for general covering laws of history. Instead, students of fiscal sociology today are more likely to puzzle over differences in tax policy across states or countries at similar levels of development, particularly because ... variations in tax structure seem resilient even in the face of putative pressures to converge brought about by globalization” (Martin, Mehrotra, and Prasad , 2009: 13). Some of the most significant work in this new fiscal sociology cat­ egory directly addresses questions of state-building, bridging militarist, elite, and modernization theories. Much of this research has focused on Western Europe, as in Charles Tilly ’s masterful work (1992), and extended to postindependence Latin American (Centeno , 1997), postco­ lonial Africa (Bates, 1 981 ) and post-Soviet independent states (Gehlbach, 2008), noting important differences in regional and historical contexts. These studies focus on the key social groups and ideologies present at the formative moments of nation-states and the process by which the state­ building projects of founding groups are consolidated for long periods in institutions such as tax regimes.19 they been able to consolidate their economically, socially, and politically pre-dominant position without the backing of the public finances. ... Capitalists have used the public household on the largest scale to enhance their profits and extend their power since capi­ talism has emerged triumphant in the form of finance capital” (Goldscheid, 1958: 2 11) . 18 Martin, Mehrotra, and Prasad (2009: 6 - 11) divide fiscal sociology traditions into three camps: modernization, elite theory, and militarist theory. 19 Such an approach guides Evan Lieberman’s (2003) exploration of “ national political community” in Brazil and South Africa, which traces initial decisions about racial and regional identity to current-day tax regimes.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

4o

State-Building and Tax Regimes in Central America

However, the fiscal sociology of state-building requires updating if it is to deal with the impact of state-building on tax under contemporary evolutions of capitalism. As new sectors emerge in Central America, they struggle to construct new social coalitions and linkages to policy mak­ ing, establishing a state-building project to advance their interests. This requires state support to high-value sectors and mechanisms to maintain peaceful domestic social relations. Fiscal sociology approaches tell us that the revenues to pursue these functions will be forthcoming only if the state can provide institutionalized linkages to key groups in society. To rising sectors, these linkages must offer guarantees that the revenues they contribute will be used to promote economic expansion; whereas declin­ ing sectors demand protection and popular sectors demand attention to social needs. The current project examines similar pressures of integration in global markets and divergent fiscal responses rooted in national patterns of state-building. State-building projects sit at the juncture of internal state-society linkages and external economic integration. Although the opportunities for external economic insertion are largely similar across Central America, national emerging elites have differed in their trans­ national accumulation strategies and their internal organization. They have further differed in their relationship to other domestic social actors and institutionalized state-society linkages. As a result, the state-building projects emerging in each country are unique, and they can be character­ ized according to the actors involved in shaping them, the institutions in which they are formed, and the processes by which they produce policies. These dimensions of state-building projects explain differences in the tax regimes evolving in the region. Framework of State-Building and Tax Regimes The question at the heart of this study is what impacts state-building dif­ ferences have on tax regimes. State-building differences are characterized by different internal structures among emerging elites (their cohesion) and their relation to other groups (their dominance). Additional vari­ ation is evident in the institutions in which these social forces interact, which provide incentives and constraints on the actors involved in pol­ icy making. As actors interact within institutions to produce policy, they engage one another and their institutional hosts, drawing attention to the dynamic processes by which actors and institutions interact over time. As a way to systematically explore these dimensions, the project defines

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

41

the concepts of actors, institutions, and processes such that they can be explored in each case. Political Actors

The project combines analysis of individual actors, such as legislators responding to the incentives of congressional procedure, with socio­ economic actors, such as those emerging elites associated with new patterns of production and exchange. In part, interests of these actors with respect to state-building are shaped by positions within urban, rural, elite, middle sector, and popular groupings, and most impor­ tant is the relationship to processes of global integration. Actors con­ nected to dynamic transnational processes of production generate great wealth, and as they gain economic presence, they also seek to mobi­ lize, organize, and secure political access. As such, newly emerging eco­ nomic elites play a leadership role as a “ transnational capitalist class,” shaping productive relations and governmental structures in their own countries and in the countries in which they operate (Robinson, 2008: 28-30). Yet, transnational elites do not enter an empty space; they enter a political fray in which previously constituted actors already occupy some political spaces and organizations. In Central America, power continues to be exercised by economic elites drawn from prior relationships to the international economy, such as industrial producers oriented to domestic markets and landed agricultural interests producing traditional exports. These actors retain influence and are often embedded within family net­ works and social ties that connect them to political elites and the leaders of emerging sectors. One of the key variables to consider in each case is the degree to which emerging elites have gained an upper hand with respect to rival elite factions within sectoral institutions and partisan organizations. In recent years, popular groups have also found opportunities for activism and organization, especially those whose own political and eco­ nomic role has expanded as a result of international integration.20 Such groups may face new forms of exploitation, but they also mobilize new identities, using newfound political resources, and they present their own priorities within debates about the appropriate role for the state (Sikkink 20 One such popular sector is the women’s movement, empowered by the processes of dem­ ocratic opening occurring in the region, their increasing role within the economically active population, and their ability to connect globally to international allies within offi­ cial donor organizations and nongovernmental organizations (Jaquette, 1989).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

42

State-Building and Tax Regimes in Central America

and Keck, 1998; Della Porta and Tarrow, 2004). Yet, there is a discon­ nect between the literature on social movements, which describes this newfound dynamism of popular sectors, and the literature on globaliza­ tion, which predicts an unproblematic ascendance of transnational elites. These processes remain uncertain, as the hegemony of newly emergent elites must be organized and articulated, terms of alliance and authority negotiated with rivals, and new forms of domination secured over subor­ dinate groups. The way this unfolds varies from country to country and depends in part on ideological and political struggles. The main ideological struggle of concern here regards the view of the state in the context of globalization. This is partly an issue of material interests, as social groups seek assistance and protection to integrate eco­ nomic processes internationally. However, interests are also nonmate­ rial, as actors can be convinced to look past short-term self-interest and cooperate around long-term efforts to pursue public goals. State-building projects are partly ideological struggles in which the challenge is to alter the norms by which the state is evaluated and understood, articulating a coalition of social groups, desirable policies, and institutional adapta­ tions (CEPAL, 1998). Whether a consensus around a state-building agenda will emerge and the content of that agenda depend critically on the technical and political capacity of different actors. Technical capacity refers to the skills, informa­ tion, and resources necessary to analyze complicated issues and develop alternative proposals for state reform. Few actors possess this technical capacity, and those who do often dictate much of the debate. In fact, transnational technocratic elites, ensconced in national and international institutions, have used privileged knowledge (as well as their leverage over developing country governments gained from aid and debt) to insert themselves into national policy processes (Goode, 1993; Centeno , 1994; Dominguez, 1997). With these technocratic elites, transnational elites infil­ trate and forge alliances, guided by an agreed preference for international integration as well as shared social networks (Jackson , 2005). Political capacity refers to the power to influence others and assert dominance within policy making. This can include efforts at persua­ sion, negotiation, force, or other means. One of the major ways in which social groups exert influence is through their sectoral and social organi­ zations. These organizations deliberate over interests, convert them into public policy preferences, and act strategically within the policy process. As such, the internal and external characteristics of sectoral and interest organizations will be a major focus of this book.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

43

Internally, sectoral organizations vary in terms of cohesion. This partly reflects internal political processes and resources dedicated to deliberat­ ing around state-building projects; not all organizations are able to artic­ ulate a unifi ed position. By definition, homogeneous organizations have less diverse membership, will have fewer internal conflicts, and will have greater political capacity speaking with a single voice in policy making processes. Yet, emerging elites often share their sectoral associations with rival elite factions and occasionally even divide among themselves. These divisions can be organizational, as they enter rival political par­ ties or networks of firms, or they can be ideological, as specific policy objectives diverge among factions of transnational elites. For example, domestically based exporters have to worry about infrastructure and human capital development, whereas offshore financiers have to worry more about macroeconomic frameworks and access to credit. Both favor the policies and institutions that can enhance transnational production processes, but there are potential disagreements on particular issues. The question to ask is whether they are cohesive, able to articulate a single, coherent agenda for state-building, and able to carry it into the state. Sectoral organizations can also be examined in terms of their external influence. Groups with significant resources, especially media and com­ munication resources, have an advantage in infl uencing others. Further, organizations whose members can call on other kinds of loyalties, such as business connections, family relationships, or patronage networks, add further elements of infl uence, projecting their interests onto other actors and into the political system. When it comes to public policy making, the most effective vehicles for emerging elites and other actors to communicate their interests into pol­ itics are political parties, present in all branches of government.21 Party organizations can be influenced by campaign finance, membership, media support, and direct representation of emerging elites in the cadre and leadership of party organizations. On the other hand, political parties and the politicians and activists that fill them are not driven solely by the interests of their social base. Occasionally, political elites distance themselves, driven by the require­ ments of political institutions, the demands of political competition, and the impossibility of responding to multiple and conflicting social demands. At these moments, political elites demonstrate a degree of autonomy from 21 This includes the highly politicized judicial branches in Central America, which are divided according to partisan appointments and loyalties.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

44

State-Building and Tax Regimes in Central America

sectoral interests, and their behavior is directed by affairs of state. These can be institutional interests in expanding state power or more narrow political interests such as reelection, expanding bureaucratic power, or remuneration. This book defines actors within state-building projects in terms of the individuals, factions, and groups who are able to articulate a state­ building agenda and who mobilize the technical and political capacity to influence policy outcomes. The book highlights transnational elites stimulated by global integration, and it emphasizes the way in which they relate to already existing elites and popular sectors, especially through sectoral associations and partisan organizations. The question asked of each case is whether the structure of emerging elites and their relations to other groups characterizes them as cohesive and dominant, or not. Institutions

There is more to actor dominance and cohesion than simply the pursuit of interests with different capacities. Actors operate within institutional boundaries that set the “ rules of the game” for interaction (Steinmoi 2008: 123). These rules define what options exist at any given moment and include or exclude certain actors from the policy process. Institutional design provides incentives to actors that make certain behaviors more appealing and constraints that rule out others. These incentives and con­ straints direct actors toward behaviors and policy outcomes that would be unpredictable or unlikely if we only considered their cohesion and dominance. The approach taken to institutions acknowledges both formal, written rules and the informal norms and customs that bound behavior (Hall and Taylor, 1996; Streeck and Thelen, 2005). Both are relevant in the incipi­ ent democratic states of the region, and to make sense of these rules, this study applies a combination of rational choice, sociological, and historical institutionalist approaches. Although they have signif cant differences, they share the intuition that institutional rules structure behavior (Hall and Taylor, 1996).22 22 Rational choice institutionalism places emphasis on individuals as self-interest m axi­ mizers, responding to institutional incentives and constraints that induce cooperation; and this cooperation is rational if it makes all better off than in its absence (Bates et al., 1998). Sociological institutionalism places individuals in their group context, responding not primarily to self-interest, but rather to habitual behaviors and norms of interaction. The resulting patterns of cooperation may produce collective benefits

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

45

Among the institutions that will receive attention in each of the cases examined here are legislatures, executives, and in certain instances, judi­ cial branches. Legislatures play a central role in the aggregation of inter­ ests at the site of debate over state-building efforts. Legislative bodies are organized procedurally, with committee structures, leadership structures, partisan blocs, and rules for the design, deliberation, and authorization of legislation (Cox and McCubbins, 2005). These rules are the product of formalized and written procedures as well as the gradual accretion of informal practices and custom, and they shape the way actors operate within the policy process. Within legislatures, attention will be paid to issues of access, agenda, and authorization. Which actors, either sectoral or partisan, have access to the legislature and gain representation and shape policy debates? In particular, do transnational elites, rival elites, and popular sectors enjoy equal access? Further, how are agendas set and controlled, and do agenda rules facilitate or inhibit the consideration of state reforms? Finally, what are the rules for authorization, in terms of number of votes, in which committees, and under what threats of executive veto? Do changes require supermajorities or constitutional amendments? In short, what aspects of institutional design impinge on the way state-building proceeds? The executive branch presents an alternative set of incentives and con­ straints, embodied in the complex set of institutional rules governing core ministries and decentralized bodies. In exploring executive institutions, this study is particularly interested in four factors: the degree of con­ centrated power in the presidency, relative power with respect to other branches of government, patterns of linkage to social sectors and power­ ful individuals, and external dependence. Executive branches vary in the degree to which they are dominated by the person of the president or power is shared among cabinet members. Further, executive-legislative relations differ in the degree to which the executive can control the agenda, design legislation, and veto congressio­ nal decisions. Linkages to social groups and powerful individuals vary in but more importantly generate consistency between collective worldviews and appro­ priate social interaction (March and Olsen, 19 89 ; DiM aggio and Powell, 19 83). Finally, historical institutionalism distinguishes itself, not by the motivations attrib­ uted to individuals and groups, but rather to the role placed on institutions as mark­ ers of historical process. Institutions are the congealed accumulation of choices by individuals and groups, which thereby leave legacies shaping the operation of future politics (Pierson, 1993).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

46

State-Building and Tax Regimes in Central America

terms of the way emerging elites have gained access to executives, either by entering positions of power directly or by gaining influence over exec­ utive branch actors. Finally, although executive branches are not the only entities in contact with external actors, it is the executive branch that makes most foreign policy decisions.23 International relationships achieve the status of institutions in and of themselves to the extent that they constrain domestic actors and invite external actors to operate within national policy processes. Aid relation­ ships often include mechanisms to incorporate the interest of donors into domestic processes, such as through conditions attached to loans and grants or, even more explicitly, with donor representatives exercising infl uence over executive or legislative actors. This may appear an intru­ sion into national sovereignty, yet it is prevalent and increasing, especially in countries that depend on external assistance for a significant portion of their revenue. Also of relevance here are tax administrations, formally part of the executive branch but deserving of independent analysis for their role in shaping the application of revenue rules. Revenue administrations are not independent, even where they are granted a degree of autonomy; chief administrators are usually named by sitting presidents, and the rules for tax administration are often passed in legislatures (Taliercio, 2004). Yet, revenue administrations are internally complex bureaucracies, and there are differences in the way international taxes are administered by customs agencies and domestic taxes are administered by revenue agents. Newly emerging sectors, such as owners of assembly factories, have often asked for, and secured, special revenue bureaucracies to ease their fiscal relationships to the state, such as customs agents housed in and assigned specifically to industrial parks. Finally, in some of these cases, the judicial branch is an important insti­ tutional source of influence. In places like Guatemala and Costa Rica, the constitution includes legal stipulations on revenues that the courts have expansively interpreted. The existence of these rules and the willingness of the Supreme Court to use its power to strike down legislative changes constrain state-building efforts. In cases where it is relevant, this book examines the power of the court, its willingness to exercise oversight, and the rigidity of constitutional stipulations. 23 Legislative branches are not without influence on international matters, as they must approve treaties and often have significant influence over loans, but it is the executive branch that generally initiates these efforts.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

47

In short, this study pays attention to the incentives and constraints pre­ sented by different institutions. This includes legislative, executive, and judicial branches, as well as the formal and informal rules governing their interaction. Further, in the context of small, dependent countries, external relationships take an institutional life of their own, creating additional incentives and constraints that shape who is involved with policy making and according to what rules. Processes

Actors operate within institutions, but they attempt to shape those insti­ tutions and are themselves influenced by the policy outcomes produced. This study takes seriously the notion of processes, by which are meant the dynamic interactions, sequences, and feedback effects among vari­ ables. One can think of actors and institutions as stochastic stock vari­ ables, in which process variables are the dynamic flow by which other stock variables come together. Processes refer to the ways in which actors and institutions produce outcomes over time in interactive, repeated, and nonlinear patterns. Three dimensions guide this study: timing and speed, number of political actors, and inclusiveness. The first characteristic to consider within reform processes is their tim­ ing and speed. The notion of timing refers to the moment within electoral cycles, and speed refers to the pace of reform itself. Many observers of reform argue that early introduction of reform and rapid implementation is useful to overcome a strong and concentrated opposition.24 They have described “ first generation” reforms in these terms, including macroeco­ nomic adjustment, privatization, liberalizing trade, and marketization (Nelson, 1989; Haggard and Kaufman, 1992; Haggard and Webb, 1994). The strategy promoted by many was both rapid speed and early implementation. Jeffrey Sachs, reflecting on his experiences in Poland, Russia, and Bolivia, promoted a “ Big Bang” approach: “ rapid, compre­ hensive, and extensive reforms to implement normal capitalist relations” (Sachs, 1994). He advised recently elected, new governments to take advantage of their political capital to undertake reform before opposi­ tion could organize. Similarly, governments that suffered a crisis, natural 24 These types of problems are particularly vexing under new democratic regimes (Lal, 1994). In the early 1990s, as countries faced the double transition of democratization and macroeconomic adjustment, it was particularly difficult to imagine how weak gov­ ernments could advance difficult reforms in the face of newly mobilized and invigorated social actors.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

48

State-Building and Tax Regimes in Central America

disaster, or other shock could take advantage of disruptions that demobi­ lized opposition and united supporters (Bates and Krueger, 1993).25 Others have observed a slower and more gradual process of reform, building only toward the end of any electoral cycle. The conditions under which this might apply are those in which there are significant transaction costs, as the very act of moving from one set of rules to another implies costs. The challenge, in these circumstances, is to distribute these costs by reaching prior agreement among interested actors.26 The analogy is to a contract in which the details can be specified and agreed before the trans­ action occurs (Coase, 1960; Alchian and Demsetz, 1972). One example of the use of contracts within reform processes is the use of social pacts, consultations, and consensus (CEPAL, 1998; Kaufman and Nelson, 2004). This type of reform implies a slow pace and drawn-out timing, so that all actors can negotiate and agree to the terms of shared burdens. Only after all the terms have been agreed can the reform proceed.27 Another useful strategy is linking difficult reforms with policies that have already achieved a degree of consensus. The process of reaching consensus, for example through a poverty reduction strategy or demo­ cratic transition, establishes a degree of trust among state representatives, society, and international actors. This trust can be mobilized to support additional reforms without having to establish entirely new social rela­ tionships or institutional frameworks.28 25 Sachs uses the example of Poland after the election of Solidarity to justify his support for a Big Bang approach. In Poland, Solidarity won all but one of the Senate seats in 1989. The Communist Party, which ran unopposed for thirty-fi ve seats in the lower house, actually lost two seats, as people turned in blank ballots rather than cast a vote to give the Communist Party candidate 50 percent (Sachs, 1994: 38). The result was a resound­ ing victory for Solidarity and an unprecedented legitimacy for the incoming government. It proceeded to pursue drastic reforms in rapid succession: “ liberalization, stabilization, privatization, creation of a social safety net” (Sachs, 1994: 79). The government reduced tariffs, freed prices, and opened trade to the West, especially the European Union, which it would eventually join. Stabilization focused first and foremost on the reduction of sub­ sidies, raising revenues, and balancing of the budget. 26 Literature on transaction costs was the subject of a Nobel Prize for Oliver Williamson, who demonstrated that markets can reach suboptimal equilibriums because of the diffi­ culty of entering transactions, which require transfers of value but imply costs that have to be distributed among parties (e.g., 1986; 1993). 27 One example is the replacement of trade taxes with value-added taxes. Over time, this may not imply a loss of revenue or welfare for society, but still implies costly switching of burdens. Some sectors will surely pay more, others will pay less, and the benefits of shifting the economy toward an outward orientation will be unequally distributed. 28 In Chile, for example, the consensus that was built around the transition to democracy served as glue to hold together a coalition that could also discuss the terms of expanding social spending and increased taxes to pay for that spending (Boylan, 1996).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

49

A second aspect of process that will receive attention here is the number of actors involved in reform. For political theorists applying bargaining models, the number of actors is a crucial determinant of political out­ comes. The main proposition is that increasing numbers of political actors complicate collective action by creating incentives to free ride.29 Many of the reforms characterized as “ second generation” have operated accord­ ing to this logic, such as civil service reforms, electoral reforms, or decen­ tralization (Weyland, 1996; Graham, 1998; Corrales, 1999). They impose costs but distribute benefits in a nonexclusive and nonrivalrous fashion. On the other hand, the difficulty of public policy is probably not strictly increasing with the number of actors. With too many actors, coordina­ tion can be difficult, as argued by bargaining theorists; but, on the other hand, with too few actors, the pressure to reform can be limited and the pattern of reform can be authoritarian and unaccountable.30 The focus on the number of actors also raises the issue of what kind of actors are included, characterized here as the inclusiveness of reform pro­ cesses. This reflects a concern with the degree to which reform processes incorporate different social interests. In the context of state-building in response to globalization, reform necessarily involves decisions about winners and losers, and including the representatives of different sectors makes it more likely that diverse interests will be represented in reform outcomes. In Central America, where the distribution of wealth is severely unequal and much of the population lives in poverty, it is especially important to pay attention to the inclusiveness of reform. Wealthy sectors have suc­ cessfully evaded and avoided taxes, leaving states dependent on revenues drawn from middle sectors, through consumption taxes and levies on wages (Tanzi , 2000). In part, this is because policy processes have system­ atically excluded popular actors, allowing upper classes to manipulate revenue regimes to limit their contributions.31 29 A corollary of the collective action problem is that the degree of coordination may vary between reformers and opponents. In the context of World Bank attempts to “ right-size” public employment, opponents of reform are seen as highly organized “ entrenched inter­ ests.” To assist reformers, they developed an entire website to reviews and strategies for uncoordinated reformers to overcome coordinated opponents (http://wwwi.worldbank. org/publicsector/civilservice/ epublishdocs/rightsizing/rightsizingchallenges.htm). 30 Those examining party systems have forcefully made this argument (Huntington 1968). In single-party dominant systems, there may be little incentive to make any changes, especially if radical change might upset dominance (Greene, 2007). 31 Michael Best noted that the main obstacle to revenue raising and state-building in Central America was the intransigence of “ oligarchic elites,” in which he was particularly con­ cerned with large landowners (Best, i976).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

5o

State-Building and Tax Regimes in Central America

The ability of governments to secure revenues from new sectors requires a shift in government priorities, as well as an enhanced capacity to mon­ itor and tap into globalized production. For two reasons, more inclusive policy processes are necessary to achieve these goals. First, the state needs a broad support base among popular sectors if it is to impose burdens on privileged elites, especially elites in hard-to-tax sectors. Second, a new fiscal regime entails a new set of state-society relationships and relation­ ships among groups in society, something that cannot be negotiated with­ out broad and open participation in policy making. Conclusion This chapter has developed a framework to compare the state-building projects occurring in each country, focusing on the concepts of actors, institutions, and processes. Actors, especially transnational elites, will be explored in terms of the degree to which they are cohesive in defining their internal structure and dominant in their capacity to advance a state­ building project. One of the determinants of their ability to advance a project is the set of institutions they must pass through if they are to make policy. Institutions that facilitate access for transnational elites, constrain opponents, and ease passage of legislation carry transnational elite state-building projects forward. Finally, because policy making has a dynamic element, state-building projects will be compared in terms of the process by which actors and institutions come together and act on one another in the process of making policy. These patterns can shed light on the precise fiscal sociology emerging in each country in terms of the way revenues refl ect deeper aspects of state capacityand state-society relations. In particular, fiscal sociology directs attention to the terms of the social contract that conditions the revenues that sustain states and preserve accumulation. These processes further shed light on state-building itself, which includes both inter­ nal management of social relations and external projection of national actors. For this study, concerned as it is with state-building in the pro­ cess of globalization, the main issue is the character of the economic integration currently taking place in the region. Globalization has made possible a new pattern of transnational accumulation, and in the process it has excited new domestic elites in each country adapted to transna­ tional processes. One of the insights of this framework is that state-building occurs with each transformation in international capitalism and the ensuing

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

State-Building in a Globalized Political Economy

51

emergence of new, domestic capitalist factions. It is to the historical rec­ ord of revenues in Central America that the next chapter turns, testing whether episodes of tax regime change across the region and over time can be linked to alternative state-building efforts by elites emerging under different patterns of international insertion.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:02, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.004

3 Historical Junctures in Central American State-Building and Tax

This chapter explores the historical trajectory of Central American tax regimes, highlighting moments of change over time, as well as divergent paths across countries. To make sense of these changes, the chapter builds on the already rich historiography of political regimes in the region, which are used here to describe state-building as it occurred over time and across countries. In bringing analysis of tax regimes to the secondary literature on the evolution of regimes, the chapter tests the underlying argument of this book: State-building patterns influence the nature of tax regimes and changes to tax regimes. The first section of this chapter reviews some of the main conclusions of the comparative historical literature to identify periods of conflict between rising and declining elites over their designs for the state. These designs moved largely in parallel for all five countries, driven by emerging elites seeking the policies and institutional adaptations that would allow them to expand their exports to the international economy.1 Changing export profiles triggered changing social formations, which comparative historians have divided into four distinct periods of Central American history. These include: the postindependence oligarchic regimes of the early nineteenth century (Stone, 1992); the rise of liberal coffee exporters toward the end of the nineteenth century (Mahoney, 2001); different responses to popular mobilization during the Great Depression and after (Rueschemeyer, Stephens, and Stephens, 1992; Paige, 1997); and the post-World War II rise of capitalist agricultural exporters and incipient industrialists (Yashar, 1997). 1 As Rodrik, Hausmann, and Hwang (2005) have argued, “ What you export matters.”

52 Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

53

Each period was defined by the nature of social cleavages between rising and declining elite sectors. As rising elites pressured for political access and government policies to advance their expansion, they encoun­ tered declining elites and popular sectors, with alternative and conflict­ ing designs for building states policies and institutions. The resolution of these conflicts established institutions and social relationships with last­ ing duration, traced forward to long-term processes of democratization and development.2 To this secondary literature on periodization and divergent paths, this chapter adds close attention to changes in tax regimes. Tax regimes change in terms of the amount of revenue raised and the sources of rev­ enue. More revenues raised suggest state-building projects were enjoying success, and the source of revenues offers an indication of which groups were willing to provide their wealth to support the state. Variations across time indicate changes in the character of dominant sectors and their state-building enterprise, whereas variations across countries indi­ cate divergent trajectories in the nature of state-building projects. To make this argument, the chapter explores historical data from the region. There are of course problems with these data, attributed by careful historians to poor record keeping, tax evasion, government cor­ ruption, and regime breakdown; and the older data are the most prob­ lematic (Gudmundson and Lindo-Fuentes i 1995: 69). To make the data as comparable as possible, cross-country comparisons concentrate on data collected in similar fashion during similar time periods, for example, comparing Guatemalan data and Honduran data from the nineteenth century drawn from a single source. Cross-time analysis of individual countries, such as changes over time to Guatemalan tax regimes, are less simple to limit to single sources, but changes in measurement and detail from different sources are smoothed with national sources as much as possible. Even with the limitations of imperfect data, tax regimes clearly shift according to the state-building efforts of emerging elites, evolving over time and diverging across countries. For example, the Table 3 .1 shows tax levels as a percentage of GDP in selected years from 1920 to 1980. The data is from the Oxford University Latin American Economic History Database, which gathers data in local 2 One of the major causes for dispute among historical studies, which will not be resolved here, is which moment of social conflict was decisive. This is important because these stud­ ies operate with a path-dependent logic, in which the paths chosen at precise moments lead to the unfolding of subsequent events for long periods (Mahoney and Rueschemeyer, 2003).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

State-Building and Tax Regimes in Central America

54 t a b l e

3 .1 . Tax as Percent G D P

Costa Rica El Salvador Guatemala Honduras Nicaragua

1920

i9 29

1 933

1939

9.39 5.10

11.6 5

8.46 11.0 0

8.80 14.29

5.85

7-73

8.26

10 .12

10.95

1 96 5

1 975

12.06 13.86

18 .0 1 z 3 v95

9 .0 9

8.9 7

12 .3 1 11.2 0

1 2 .59 1 3 .7 4

Source: Oxford Latin American Economic History Database, http://oxlad.qeh.ox.ac. uk/search.php, consulted 12/10 /10 .

currency units for both GDP and revenues. The years in the table were selected to cover years during the height of the liberal period (19 2 1 and 1929), the Depression (1933 and 1939), and the post-World War II period (1965 and 19 75).3*Across time, the periods would appear to mark shifts for each country. For example, Guatemala during the 1920s looks different from Guatemala during the 1930s and Guatemala during the 1960s and 1970s. Also, within periods there are clear differences across countries; Costa Rica is far ahead of the other countries by 19 75, with Guatemala notably lower. This chapter places historical data on tax in the context of state-building differences across periods and across countries. Central American State-Building Over Time and Across Space In the background of most historical work on Central America is the work of Barrington Moore. He focused on the role of agricultural trans­ formation in Western European paths to modernity, arguing that without a modernizing bourgeoisie there could be no democracy (Moore , 1966). This has led to a focus on the nature of class relations in the Central American rural sector, distinguishing between “ labor-repressive forms of export agriculture in the North and the greater importance of small holders and merchant capitalists in the South” (Paige , 1997: 6-7), help­ ing to distinguish between long-standing democracy in Costa Rica and dictatorships in the rest of the region. The Moore thesis is insufficient to fully explain Central American variation, however, as there were great differences between the military authoritarian regimes of Guatemala and El Salvador during the twentieth century and the relatively more patri­ monial authoritarian regimes of Honduras and Nicaragua. 3 Data for the oligarchic postindependence period is measured differently and will be pre­ sented later.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

55

To study these patterns more closely, historians have described four broad periods marked by the way Central American countries have fit into the world capitalist system. These included (1) oligarchic regimes exporting nonfood, agricultural exports after independence; (2) liberal regimes exporting coffee from the last quarter of the nineteenth century; (3) the Great Depression and popular mobilization; and (4) capitalist agricultural expansion and incipient industrialization after World War II (Barry, 1987; Robinson, 2001a; Brockett, 2005).4 Each period was asso­ ciated with the rise of new elites atop emerging export sectors, and they struggled politically to capture state power and introduce policies and institutions that would assist in their expansion. Even as they were drawn from relatively similar sectors and sought relatively similar state trans­ formations, the precise character of their state-building projects and the conflicts they encountered with other sectors varied across countries. The following paragraphs draw on the secondary literature to describe pat­ terns of state-building over time and across space. Period One

The collapse of Spain and independence struggles elsewhere in Latin America thrust independence on Central America, which had been first incorporated into Mexico but was declared an independent federation in 1824 by the restoration of Conservative Party elites linked to the Church. The immediate postindependence period preserved the colonial pattern of international insertion and social structure, with oligarchic elites con­ trolling land and exports; “ cacao, sugarcane, and cattle were produced for the national market, while indigo and cochineal, respectively, met the demands of the European textile industry. Under such conditions, the colonial status quo could maintain and consolidate itself” (Torres-Rivas, 1993: 1 0 - 1 1 ) . 5 Stone has argued that oligarchic landowners formed a durable dom­ inant elite during the period, resting authority on family ties and cul­ tural values of domination. Differences across the region emerged in part 4 “ Each epoch of world capitalism has produced a re-articulation of Central America to it, including the transformation of social forces and the restructuring of economies, classes, states, and power blocs” (Robinson, 2001b: 149). 5 There was some variation across the region in terms of the precise composition and destination of exports. “ Guatemala had become an important exporter of Salvadoran and Guatemalan indigo. Exports beyond the isthmus from the other provinces were not very great, but Honduras and Nicaragua, and to a lesser extent Costa Rica, were impor­ tant suppliers of livestock and agricultural foodstuffs to the indigo-producing regions” (Cardoso, 19 9 1: 8).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

56

State-Building and Tax Regimes in Central America

because of the structure of those elites and the economic conditions in which they operated. The relative abundance of land, capital, and man­ power in Guatemala and El Salvador allowed economic elites to dedicate themselves to accumulation and leave the military or political classes to manage politics. In Costa Rica, by contrast, the oligarchic elite had less abundance with which to work and were forced to enter direct control of the state (Stone, 1979). While Salvadoran and Guatemalan state-building efforts, such as they were, focused on amassing military power, the Costa Rican state used its power to promote exports, encouraging small-holder cultivation of coffee and eventually bananas.6 Nicaragua and Honduras represented a third pattern, characterized by a weak oligarchy, frag­ mented along regional lines atop relatively stagnant economies with lim­ ited exports to the rest of the region. Period Two

By the end of the nineteenth century, most exporters across the region turned to coffee.7 This marked the emergence of liberal state-building efforts dominated by newly emerging coffee exporters. They generally identified with the Liberal Party, through which they sought to concentrate their control over land, capital, and labor. To this end, they offered land grants and privatized state and Church holdings, created tax breaks to encourage property concentration and export, and offered financing, set­ ting up the first state banks. They also undertook infrastructure projects, especially railroads, ports, and roads, to bring products to international markets. Regulations, vagrancy laws, and civil codes created polarized rural class relations that were enforced by modernized militaries.8 Despite this common trend, differences emerged. In El Salvador and Guatemala, Mahoney (2001) describes the “ radical liberal” project of a dominant agrarian bourgeoisie who established a centralized state, com­ plete with a powerful military-coercive apparatus to sustain polarized 6 In a later work, Stone argues that Costa Rican elites used the state to resist the entrance of potential rivals, blocking immigrant populations; while in Honduras and El Salvador, Arab and Jewish immigrants played an important role in commerce, banking, and new agricultural products (Stone, 1992). 7 Coffee had come to dominate Costa Rican exports as early as the 1 830s and had replaced cochineal from Guatemala by 18 7 1 and indigo from El Salvador by 18 8 1 (Cardoso, 19 9 1: 4 0 -4 1). 8 “ The kind of dependent economic growth experienced in Central American countries had no use for workers with full labor rights and citizenship. On the contrary, it needed firm political and social control and low wages. Costa Rica was an exception, but only a partial one” (Cardoso, 19 9 1: 62-63).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

57

rural class relations. In Costa Rica, a more “ reformist liberal” pattern dominated, in part because coffee emerged earlier, grown on land par­ celed into small estates, and the conservative oligarchy had remained weak after independence. The new agrarian bourgeoisie, allied to urban merchants, could centralize the state apparatus to promote export with­ out having to rely on a repressive military to discipline rural labor.9 Both Nicaragua and Honduras saw their period of liberal state-building aborted by external intervention. Although a weak oligarchy appeared to allow Honduras to follow the Costa Rican path of reformist liberalism, the late start of liberal reforms and the relative weakness of domestic exporting classes led to conflict with foreign banana enclave owners.101To protect enclave interests, U.S. intervention deposed liberal reformers and reaccommodated traditional Conservative Party elites. Nicaragua suf­ fered a similar fate when Liberal Party reformers introduced nationalist measures that potentially threatened transportation routes.11 The United States invaded, restored oligarchic dominance, and occupied the country repeatedly until 19 33 (Mahoney, 2001: 33-4 1). Period Three

A third period was marked by the Depression of the 1930s when inter­ national trade collapsed, real incomes fell, and massive unemployment ensued.12 The countries pursued relatively similar responses, which is to say they did nothing to stimulate their economies, running fiscal surpluses at the height of the downturn (Torres-Rivas , 19 9 1: 75).13 In response, pop­ ular movements inspired by trade unionism and the Russian Revolution

9 “As the linkages towards industrial development were virtually nil due to the reduced size of the economies, only investment in human capital and transportation infrastruc­ ture would remain accumulated over the long term. In these terms, it was clear that Costa Rica achieved a significant advantage, already visible at the end of the 19th century” (Pérez-Brignoli, 2000: 1 1 5 - 1 1 6 ) . 10 The Conservative oligarchy was considered too weak by itself to resist liberal state­ building efforts (Torres-Rivas, 19 9 3: 7 0 -7 3, 90-95). 11 The United States had developed an interest in routes across the isthmus through Nicaragua, shortening significantly the time it took to bring gold from California to New York (Stone, 1992). 12 “ The levels of foreign trade, public spending, and the gross domestic product in general recovered only after 19 4 5, and in some cases, such as Honduras, even later” (TorresRivas , 19 9 1: 73). 13 Even had elites wanted to pursue alternative, or countercyclical, policies, there were few instruments available. Most infrastructure projects had been finished earlier, having been limited in scope to railroads, ports, and other transportation avenues (Pérez-Brignoli, 2000).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

58

State-Building and Tax Regimes in Central America

of 19 17 organized, pursued direct action for wages and access to land, and threw their support behind the electoral prospects of dissident Liberal Party elites. Once again, differences across the region were evident. In Costa Rica, El Salvador, and Guatemala, popular sectors organized more forcefully against the centralized states built by coffee exporters during the liberal period. Meanwhile, the lack of a centralized state and the weakness of export elites in Nicaragua and Honduras provided no clear target, and popular demands diverted toward subnational conflict or nationalist movements against outside influence (Mahoney, 2001). In Guatemala and El Salvador, the liberal elite had been prosperous and equipped a powerful military, they were “ an oligarchy in the true sense of the word, controlling the state directly or via military officers” (Torres-Rivas, 19 9 1: 232). Their response was immediate and harsh. In El Salvador, rural uprising was met with repression, leaving more than thirty thousand dead in an episode known simply as “ The Massacre.” 14 Guatemalan elites responded to popular mobilization by bringing the military to power in 1 9 3 1, further centralizing authority, purging Communists and radicals from unions, and repressing social organiza­ tion in a regime that lasted until 1944 (Mahoney, 2001: 45). In Honduras and Nicaragua, the liberal elite were neither hegemonic nor particularly prosperous, and foreign interests maintained a strong presence. In Honduras, the president elected in 19 3 2 ruled until 1949 with the approval of United Fruit and, instead of widespread repression, managed popular demands with a patronage-oriented political machine. In Nicaragua, occupying U.S. troops withdrew in 19 3 3 , leaving power to a military strongman whose family retained patrimonial control of the country until 1979. In Costa Rica, the liberal period had not polarized rural landhold­ ings or class relations to the same degree. The Depression was met by a coalition of middle sector and popular groups who pressured Liberal Party elites to enact social programs, including the eight-hour day, rights to unionize and strike, and the establishment of a corporatist system of labor courts (Rueschemeyer et al., 1992: 235; Yashar, 1997). 14 For Paige, the episode forged a particularly reactionary ideology among coffee elites, “ The liberal ideology of the nineteenth-century coffee elite was modified substantially by interaction with insurrectionary movements associated with counter-ideologies” (Paige, 1997: 9 ). Rueschemeyer, Stephens, and Stephens (1992: 233) also trace postwar author­ itarianism to this period, “ marking the beginning of the military-oligarchic pact under which the military dominated in political office holding and provided the oligarchy with the conditions for continued economic and social domination.”

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

59

Period F our

After World War II, export markets for coffee and bananas recovered, and prices on commodities rose, with surpluses invested in agricul­ tural modernization. This raised productivity in traditional exports and opened new exports in more capital-intensive products such as beef, sugar, and cotton.1516These opened the agricultural frontier, with sugar and cotton particularly suited to the Pacific lowlands, which had been too hot and wet for coffee cultivation but now achieved rapid expansion through more capital-intensive methods.16 The rise of capital-intensive agriculturists spurred a new state-building effort, as they sought to expand state credit and drive a new round of land concentration.17 They were complemented by emerging industrial elites, mostly targeted to domestic and then regional markets, with the greatest boost coming with the Central American Common Market (CACM) after the 1960 treaty, creating free intraregional trade and a common external tariff.18 Incipient industrialists sought import substitution industrialization poli­ cies such as those outlined by the United Nations Economic Commission of Latin America, including government investment and protection to cultivate industrial expansion.19 Despite industrial growth, the project was largely doomed by the small size of regional markets and political conflicts over distributing benefits

15 Investment was poured into expanding Central American opportunities, improving acre­ age planted and productivity. In the decade after 1944, values of coffee exports increased from almost twenty million to more than ninety million dollars in El Salvador, and the value of Honduran exports increased more than thirty-five times (Bulmer-Thomas 1 1987: 110 -111). 16 “ The new exports of cotton, beef, and sugar represented a transformation of the agricul­ tural sector and led ultimately to a significant degree of export diversification” (BulmerThomas, 1987: 159). 17 The tendency toward concentration of credit was less intense in Costa Rica and Honduras, where coffee production was in the hands of smallholders (Bulmer-Thomas, I 9 8 7 : I 5 5 ). 18 Costa Rica entered in 19 63. Tariffs were raised substantially in the San José Protocol of 1968, leaving out only agro-exports and products subject to excises. Other regional initiatives included infrastructure projects, such as a Regional Highway Program and Regional Telecommunications Commission. 19 Policies oriented toward the urban sector included “ the transformation of handicrafts and secondary trade, the expansion of services, the formalization of merchant activity, and state-sponsored education” (Robinsoni 2003: 150). States extended medium and long-term finance through Central Banks or newly formed development banks, and sought international assistance, especially from the United States. The main target for these funds was infrastructure, especially roads, dams, electricity, and airports, with the public sector responsible for between one-fifth and one-third of all investment.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

6o

State-Building and Tax Regimes in Central America

among the countries.20 Governments failed to coordinate their industrial­ ization efforts, launching competing enterprises and a race of tax incen­ tives to attract international investors. War eventually broke out between Honduras and El Salvador, leading to the departure of Honduras from the Common Market in 19 70 .21 The pattern of growth during the period increased the importance of middle sectors, which pressured rising industrial and capitalist agricul­ tural elites for access. “As the influence of the middle sectors increased, many elites came to share their commitment to economic development, viewing an expanding economic pie as the solution that would satisfy the mounting aspirations of those below without sacrificing their own mate­ rial advantages” (Brockett, 2005: 42). Once again, however, the countries diverged in how they managed social relations among rising elites, falling elites, middle sectors, and popular groups. Costa Rica stands out in Central America and in Latin America for the social democratic peace that took hold after World War II (Itzigsohn, 2000). The country was gripped by civil war in 1948, but by 19 52 the National Liberation Party had consolidated an alliance of industrial elites, middle peasants, and urban sectors behind a generous welfare state regime, taking the additional step of eliminating the military. The party would remain dominant until the 1990s, interrupted in power infre­ quently by one of several center-right competitors. Even as the rest of the region experienced authoritarianism and slipped into war, Costa Rica remained democratic and expanded social programs, including health, education, and social security, attaining a level of welfare that was among the highest in Latin America (Mesa-Lago, 2003). 20 Average rates of growth of value added in manufacturing were 1 1 8 percent in the 1960s, and they more than quintupled for the region from 19 50 to 1980. From 19 50 to 1980, industrial value added as a percent of GDP at factor costs expanded in Costa Rica from 1 1 to 16 percent; 12 to 17 percent in El Salvador; 1 1 to 15 percent in Guatemala; 9 to 15 percent in Honduras; and 12 to 24 percent in Nicaragua (author calculations from Bulmer-Thomas, 19 87: 308, 322). Still, industrial output remained focused on consumer goods, with the food sector accounting for around half of gross industrial outputs. 21 Robinson emphasizes political problems with the model. “ The agro-export oligarchy dominated the project ... given the political weakness of the elite and the inchoate response of the subordinate classes to the 19 30s crisis; authoritarian civilian-military regimes remained in place. . The project in Central America stood out for the very narrow autonomy exhibited by local states and dominant groups, the extremely firm insertion of the CACM into U.S.-sponsored accumulation processes in the Greater Caribbean Basin, and the lack of popular and working class incorporation into the pro­ ject” (Robinson, 2003: 15 1) .

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

61

In marked contrast, authoritarian governments in the rest of the region produced inferior economic and social outcomes, with especially brutal repression in El Salvador and Guatemala. In Guatemala, the two govern­ ments following 1944 were led by reformist military officers in a coali­ tion with middle-class sectors and workers. Governments pursued labor regulation, public investment, and developmentalist policies but faltered when they attempted land reform, leading United Fruit to ally with reac­ tionary rural elites and military officers to stage a coup in 19 5 3 .22 For the next two decades, the country would be led by the military as an institu­ tion or its reactionary civilian allies, and popular movements turned to revolutionary uprising ending only in 1996 (Rueschemeyer et al., 1992; Yashap 1997; Mahoneyi 2001). El Salvador also endured the post-World War II period under mil­ itary authoritarian regimes. After the Depression, General Martinez established what William Stanley refers to as a “ protection racket state” in which “ the military earned the concession to govern the country (and pillage the state) in exchange for its willingness to use violence against class enemies of the country’s relatively small but powerful economic elite” (Stanley, 1996: 6-7).23 During the 1950s, military capacity was expanded and its control of the state was consolidated with an oligarchic base of support, with only moderate attempts at developmentalism.24 Popular mobilization triggered a reactionary response, driving opposi­ tion into a revolutionary movement that only ended with the 1992 Peace Agreement. Like El Salvador and Guatemala, authoritarianism characterized the Nicaraguan and Honduran regimes, but theirs were traditional regimes with a less capable military, co-optation of popular opposition through patronage, and more infrequent repression. In Nicaragua, the dictator who emerged in the 1930s, Anastasio Somoza Garci a, stepped aside temporarily in 1948 but regained the presidency until his assassination 22 “ The ability to sustain democratizing coalitions, however, rested on their capacity both to redistribute elite property and to develop political control of the countryside” (Yashar, 1 9 9 7 : 4 ). 23 Martinez himself hoped for a personalist authoritarian regime of the type that took hold in Honduras, but he was chased from power in 1944. 24 “ The military, as the institutional heart of the state, intended to balance the conflicting demands in Salvadoran society ... [but was] unwilling or unable to chart a course that conflicted with the core interests of agrarian elites. The military effectively shifted the costs for maintaining military rule onto the population, preferring in the end, to kill large numbers of its own citizens rather than bear the costs involved in challenging the more conservative faction of the oligarchy” (Stanley, 1996: 89-90).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

62

State-Building and Tax Regimes in Central America

in 1956. Even with his death, patrimonial authoritarianism remained, as he was succeeded in the presidency by his son Luis (19 5 7 -19 6 3 ) and his son Anastasio Somoza Debayle (19 6 7 -19 79 ), until the Sandinista revolution of 1979. The Somoza dynasty made use of the Liberal Party to rebuff rival Conservative Party and dissident elites, distributing patronage to an extensive clientelist network in urban and rural popu­ lar sectors. When elections were insufficient, the government made use of the National Guard, “ a praetorian outfit that possessed little auton­ omy from Somoza himself,” sufficient to the task of repression but with­ out the institutional modernization of the military in Guatemala and El Salvador (Mahoney, 2001). The Depression-era leader of Honduras, Tiburcio Carias, was unable to establish similarly personalist control, but he left a Nationalist Party with extensive patronage networks, especially in rural areas. The Nationalists competed with the old Liberal Party, who experimented with developmentalist policies and attempts to incorporate popular sectors when they came to power in 19 5 7 (Euraque, 1996: 9 7 -10 1). These efforts were cut short by a coup in 19 6 3, and it was another mil­ itary coup in 19 7 2 that initiated the more coherent Honduran effort at developmentalist state-building of 19 7 2 -19 7 5 . The general who took power in 19 7 2 , General López Arellano, had been dismayed by the dismal performance of the Honduran army in the brief war with El Salvador in 19 6 9 25 and attempted to advance the developmentalist and welfarist state-building project that had begun to take shape among San Pedro Sula liberals.26 This was incorporation from above, imposed by the military, and it was extremely late, even after the CACM had fallen apart. As such, the project was doomed, with conservative military officers deposing López Arellano in 19 75 and restoring the patrimo­ nial regime managed by the patronage-oriented Liberal and Nationalist Parties in 1982. Table 3.2 traces the state-building trajectories just described. Along the top of the chart are the broad periods of change, marked along the bottom in terms of patterns of international insertion and leading domes­ tic elite sectors. As newly emerging elites attempted to introduce state­ building programs that would assist in their efforts at accumulation and

25 He had also led the 1963 coup. 26 “ López Arellano’s coup empowered the agenda, or the vision, espoused by a national industrial bourgeoisie and labor as these sectors developed on the North Coast” (Euraque, 1996: 156).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

t a b l e 3 .2

. Divergent State-Building Over Time and Space Costa Rica

Honduras

Nicaragua

Postindependence Oligarchy

Weak oligarchy switches early to coffee

Weak oligarchy exports limited products regionally

Consolidated oligarchy exports indigo and cochineal

1870s to 1920s Liberalism

Rising coffee exporters pursue reformist liberal state centralization and incorporate middle sectors

Rising coffee exporter reforms aborted by invasion and accommodate declining elites and foreign enclaves

Rising coffee exporters adopt radical liberal centralization of the state, and modernized military enforces polarized rural class relations

1930s Depression

Liberals meet popular mobilization with social programs

Liberals turn to patrimonial authoritarianism after limited popular mobilization

Liberals cede state control to military authoritarian regime after popular mobilization

Post-WWII Developmentalism

Rise of capitalist agricultural exporters and industrialists promotes developmentalist policies and social benefits under democratic welfarist regime

Capitalist agriculturalists and industrialists emerge at the end of the period, leaving patrimonial authoritarianism intact except for 1970s military, developmentalist, welfarist experiment

Capitalist agriculturalists and industrialists leave military authoritarian control of state to preserve polarized rural class relations while adopting some developmentalist policies

Limited capitalist agriculture and industrialization provokes no change in patrimonial authoritarianism and limits developmentalist state-building

El Salvador Guatemala

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

64

State-Building and Tax Regimes in Central America

access to power, outcomes diverged as a result of the resolution of con­ flicts between rising and falling elites and popular sectors . Tax Regimes These periods of state-building and the divergent state-building patterns across countries influenced changes to tax regimes. In the next sections, evidence from the secondary literature on revenues and primary data on revenues are compared across time and across countries. When compared across time, shifts in the policies and institutions of state-building proj­ ects are evident in changing amounts of revenue mobilized and the chief sources of revenue. In addition, when the data are compared across coun­ tries at similar points in time, differences in the relative success of rising elites and the accommodations they made to resolve social conflicts are also evident. In general, when rising elites have a unified and coherent program that they are dominant enough to assert, they are more will­ ing to support the state with their own resources and allow the state to gather more revenues overall. Period One

The oligarchic elites who inherited control of the state after independence had little interest in enhancing state capacity. The exports in which they specialized required little state support, and the main activity for which they might have sought to raise revenues was to f ght wars. They fought frequently, but they did so with support from foreign loans.27 Only Costa Rica, separated by geography and by the character of its oligarchy, was relatively isolated from the wars of the other states. In general, oligarchic elites mobilized few resources through tax. In part, this was because of their ideological commitment to a laissez-faire state, a residue of their independence from colonial limitations. They had bristled for so long under the mercantilist restrictions on colonial trade that liberalized trade and removal of government intervention guided their state-building preferences. In addition, the generally weak adminis­ trative bureaucracies they inherited were poorly equipped to collect the few taxes they imposed. The period was marked by many changes to tax laws but little increase in revenues.28 27 Gudmundson and Lindo-Fuentes (1995: 7 1) cite fifty-one wars among the five countries in the fifty-year period after independence. 28 “ The disorganization of the bureaucratic apparatus that followed independence encour­ aged tax evasion. Therefore, tax revenues fell regardless of the intentions of the new

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

65

During the period of the federation, from 1824 until 1838, Conservative Party elites abolished taxes to stimulate the economy.29The federation col­ lected import-export duties and revenues from tobacco and mail monop­ olies, but the individual states resisted sharing revenues, and expanding contraband trade evaded tax payments.30 Even though the federation raised import duties, little additional revenue was collected. After the breakup of the federation, individual state governments made few improvements in revenue. They continued to depend on mail services, tobacco, liquor, and other monopolies inherited from colonial times. Small sums were available from taxes on stamped paper for legal transactions and those on cattle slaughter and gunpowder, as well as inter­ nal transaction taxes, sales of government lands, and small items, such as the right to organize cock fights. The rates for trade duties continued to rise, but weak bureaucracies and contraband kept revenues insufficient to cover expenditures (Woodward and Bethell, 19 9 1: 17 ).31 The main site of collection, ports, were limited to collecting customs duties on imports, which were ultimately passed along to consumers, as exporters toppled governments that attempted to raise export duties. In Table 3.3, revenues per capita are shown from Gudmundson and Lindo-Fuentes (1995). In 18 5 5 , none of the countries mobilized signifi­ cant revenues, although Costa Rica was already far ahead of the other countries. Smallholders had already begun producing coffee, and they sup­ ported taxes that would facilitate infrastructure investments and human capital investments to expand their earnings.32 The table includes data from 1892, by which time exporters in all of the countries had switched

29 30 31 32

authorities. ... Inexpensive imports appealed to the general public but damaged arti­ sans; low taxes stimulated trade but lowered fi scal revenue. These contradictory forces caused frequent changes in customs laws and tariffs and in the mix of internal taxes” (Gudmundson and Lindo-Fuentes, 1995: 29-30). “ Without taking into account the needs to finance growing administrative expenses” (Gudmundson and Lindo-Fuentes, 1995: 69). “ When Manuel José Arce became president of the federation in 18 25, he found a total of six hundred pesos in the federation’s coffers” (Gudmundson and Lindo-Fuentes, 1995: 70). Short-lived Liberal Party governments interspersed during the period attempted head taxes, but these were quickly rolled back by dominant Conservative Party elites. Stone attributes this difference to a value-difference between the dominant elite of the period, “ In northern countries, when legislators establish tax policies, they are prone to giving greater consideration to encouraging production rather than seeking added rev­ enue for the treasury. They also seem to attribute greater importance to collections. In the south, taxes are levied with greater concern for the treasury than for production. In short, where abundance of land, labor, and capital have made entrepreneurial activities attractive, the thinking of legislators reflects a concern with production more than where there is scarcity of these factors” (Stone, 1992: 95-96).

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

State-Building and Tax Regimes in Central America

66

t a b l e

3.3. Revenues in Pesos Total 1 8 5 5

Costa Rica El Salvador Honduras Guatemala Nicaragua3

506,920 446,824 286,895 69 9 ,5 4 3 296,374

Per Capita 1 8 5 5

Total 1 8 9 2

Per Capita 1 8 9 2

4 .7

5 , 8 0 8 ,4 7 4

2 3 .9

1.1

6,896,000 2,416,620 12,099,220

9.8

2,847,729

7-9

!-3

0.8 1.2

7.3 8.5

a Nicaragua data from 18 5 2 and 18 9 1. Source: Gumundson and Lindo-Fuentes , 1995: 7 1.

to coffee, and liberal reforms started to emerge in other countries to pro­ mote exports. Even in 1892, however, Costa Rica far outpaced the rest. Period Two

The transition to regimes dominated by coffee exporters supporting lib­ eral state-building required more significant action by governments to pay for infrastructure, provide credit to expanding agricultural exporters, and especially to consolidate and centralize military power. Governments greatly increased taxes on trade, the fastest expanding sector and the source of wealth for the most productive agro-export sectors. Yet, coffee elites were wary of committing too many of their resources to public pur­ poses, preferring to direct government attention toward other tax han­ dles, such as imports, administrative transactions, or nontax revenues. Taxes on trade accounted for more than half of total taxes for most of the period, making them the key focus of state-building efforts. Governments focused revenue administration on customs houses and mostly targeted imports, as expanding revenues on exports required the acquiescence of coffee elites. They offered their resources where they enjoyed the greatest control over liberal regimes, and export taxes were greatest in Costa Rica, El Salvador, and Guatemala and more limited elsewhere. Ideologically, liberal elites resisted government intervention in trade, and they preferred to direct attention to domestic sources of revenues such as consumption, gathered through excises on tobacco and alcohol, as well as goods and services taxes that applied to commercial patents and insurance. In Guatemala and El Salvador, liberal regimes collected more than 20 percent of revenues through consumption taxes.33 33 Nicaragua also collected signifi cant revenues through consumption taxes until external invasion rolled back liberal reforms.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

67

Liberal governments also collected significant nontax revenues from public enterprises, such as mail, telegraph, and other leftover colonial monopolies. These monopolies accounted for up to 30 percent of rev­ enues in Costa Rica and Honduras, while nontax revenues hovered closer to between 5 and 10 percent in the other countries. In all countries, direct taxes on income or property were extremely limited. Only Costa Rica experimented with an income tax, enacting one in 19 16 only to suspend it only a year later as a result of the decline in world demand associated with World War I.34 The tax was reinstated in 19 22 but excluded coffee producers and other key groups, making it rel­ atively insignificant as a source of revenue until it was modernized in the 1940s. All of the countries also had a property tax collected by municipal governments, but in no country was it applied to large estates or used to collect significant amounts of revenue. It hovered around 2 to 4 percent of total taxes in El Salvador and Guatemala, and it produced less than 1 percent of the total in Honduras and did not even exist until the 1940s in Costa Rica. In no country did direct taxes surpass 5 percent of revenues. This mix of revenues depended on international demand for Central American exports, which increased up to World War I and during the 1920s but declined during the Great Depression and during the World Wars. Declining terms of trade and high year-to-year fluctuations made revenues extremely unpredictable, and the limited amount of revenue they generated remained insufficient to cover expenses. Countries ran chronic deficits that they were forced to finance with loans from internal and external creditors. As collateral for these loans, governments offered revenues from trade taxes, and creditor nations militarily intervened to take over customs houses and ensure repayment to their banks. Several times from 19 13 to 1928, for example, the United States occupied Nicaraguan ports and took over management of its customs houses (Walter, 1993: 10 - 12 ) .35 Table 3.4 presents per capita revenues in U.S. dollars, as well as the proportion of total revenues from different sources. Costa Rica main­ tained its advantage over the other states, even though El Salvador and 34 It was replaced temporarily by a tax on bank and business profits, although this too was repealed after only five years (ICEFI, 2007c). 35 Smedley Butler, U.S. marine, in a speech on his military service noted, “ During that period, I spent most of my time being a high class muscle-man for Big Business, for Wall Street and for the Bankers. In short, I was a racketeer, a gangster for capitalism. ... I helped in the raping of half a dozen Central American republics for the benefits of Wall Street. The record of racketeering is long. I helped purify Nicaragua for the international banking house of Brown Brothers in 1 9 0 9 - 1 9 1 2 ” (Butler, 2003: 10) .

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

State-Building and Tax Regimes in Central America

68

t a b l e

3.4. Tax Regimes at the End o f the Liberal Period, 1929 Revenue Revenue As Percentage of Total Revenues ($US Per Import Export Direct Consumption N ontax millions) Capita Duties Duties Taxes Tax Revenue

Costa Rica El Salvador Guatemala Honduras Nicaragua

8.9 z 3-5 15.4 6.9 6.6

18

56.8

7.9

2.8

7.8

5 0 .7

5.2

7 .2

4 7 .4

9.8

58.6 58.6

II.9 13.6 I .3 1.2

10 .1

I .3 0 0

5.8 22.3 I 7 .I I 0.8

28.7 9 .6

20.7 2 7 .7

Source: Bulmer-Thomas, 2003: 178 ; Pérez-Brignoli, 2006 .

Guatemala were by the end of the liberal period raising more total reve­ nues, if not more per capita. All countries continued to depend on import duties, although Costa Rica, El Salvador, and Guatemala were able to secure contributions from exporters to a significant degree. These tax regimes refl ect the nature of liberal state-building projects. Where emerging coffee elites were dominant in El Salvador, Guatemala , and Costa Rica, they were confident enough in their control of the state to allow it to collect more revenue from export sectors. In the other countries, liberal elites operated in the most dynamic sectors, but they were not ascendant politically, and they were less successful in advancing a state-building agenda. They were wary of handing over their wealth or allowing the state to gather significant revenues, even if these revenues would have been necessary to support economic expansion. Period Three

The Depression was most significant because it produced significant mobilization of popular sectors. As such, it marked the decisive point at which liberal coffee elites and declining oligarchic elites decided the terms of their alliance and the nature of the state-building project they would support. This included some surprising shifts in tax regimes . In Costa Rica, where popular pressure encouraged some demo­ cratic opening and welfare provision, the state expanded its collection of revenues on domestic consumption. These doubled as a proportion of total revenues, from 6.5 percent to 12.6 percent during the period. Meanwhile, income taxes fell from 7.4 percent to 3.3 percent of the total. This shift indicates an increased burden on popular sectors and middle classes, thereby pooling their resources to finance the welfare benefits they demanded of state authorities.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures t a b l e

69

3 .5 . R ev en u e D u rin g the D epressio n

Costa Rica El Salvador Guatemala Nicaragua

Revenue Per Capita

Revenue as % of GDP

1933

1939

1933

1939

45.28 I I .4 I 4.19

7 0 .4 9

8.80 I 4 .2 9

5 -7 1

6.67

8.46 11.0 0 I 0 .I 2

12.58 5.45

1 0 -95

Source: Oxford Latin American Economic History Database, http://oxlad.qeh. ox.ac.uk/search.php, consulted 12/10 /10 .

Meanwhile, in Guatemala and El Salvador, popular mobilization was met with harsh repression and the consolidation of an elite alliance between liberal and oligarchic export elites in support of military author­ ities. In both countries, income and property taxes increased as a propor­ tion of the total, from 4 percent to 10 .7 percent in Guatemala and from 6 percent to 7.4 percent in El Salvador. In Guatemala, total revenues even increased to their highest level as a percentage of GDP. Both trends sug­ gest that economic elites were willing to pay the costs of preserving their privilege in highly unequal and repressive societies. Nicaragua and Honduras, where popular mobilization resulted in the introduction of patrimonial authority, did not produce significant change in revenue capacity or structure. There was a slight increase in revenues per capita and as a percentage of GDP in Nicaragua, but nothing beyond normal fluctuations. Also, there was no change in the relative distribution of revenues; for example, in Honduras income and property taxes were 0.3 percent of the total in 19 30 and 0.6 percent of the total in 1939. The changes to the character of liberal regimes that occurred in the Depression were played out in tax. In Costa Rica, where economic elites reached out to popular sectors, middle classes and workers were asked to share part of the burden in taxes on consumption. In El Salvador and Guatemala, by contrast, economic elites chose to support military reac­ tion, and they paid for it with more direct taxes and more taxes over­ all. In Honduras and Nicaragua, patronage-oriented patrimonial regimes had little claim to shift burdens and mobilized few new revenues. Period F our

After World War II, with the rise of capitalist agriculture and incipient industries, tax systems bore the imprint of newly formed state structures and social coalitions. Taxes increased, accompanying economic expan­ sion, finding tax handles in an expanding market economy, and meeting

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

7o

State-Building and Tax Regimes in Central America

some of the necessities of states that were taking an increasingly active economic role (Bulmer-Thomas , 1987: 15 0 - 1 7 1) . Policy advice from the United States and international lending organizations, newly interested in promoting development in Latin America, encouraged tax reforms appropriate to peripheral industrialization; it is summarized in a study of tax reform by the U.S. Alliance for Progress: “ promoting savings and investment; maintaining reasonable price stability; and contributing more of their domestic resources to economic and social development proj­ ects” (Sommerfeld, 1966: 17). In El Salvador and Nicaragua, revenues tripled between 1947 and 19 54, while they expanded by 1 1 6 percent in Guatemala (CEPAL, 1956: 23). Still, from 19 50 to 1980, as a percent of gross income, the total tax collected by central governments surpassed 15 percent only in Costa Rica and actually fell during the 1960s in Nicaragua. Local revenues, mostly property taxes, were extremely low, accounting for more than 1 percent of gross income only in Guatemala. Part of the explanation for the lim­ ited expansion of tax capacity was the failure to sufficiently mobilize tax from emerging sectors. In a report on a mission to Guatemala, U.S. officials found the tax take too low to sustain development efforts and the government unwilling or unable to increase. “ The government has been remiss in increasing the level of taxation because it does not wish to reduce consumption among lower and middle income earners who support the largest portion of the tax burden. It has not achieved, on the other hand, access to the wealth of the relatively small group of privileged elites” (Adler et al., 1952: 35-36). The inability to tap into emerging sectors was exacerbated by the pol­ icies they pressed to advance their expansion. Industrialists demanded tax exemptions for investments, and industrial promotion laws prolifer­ ated in the 1950s. The first was enacted in 19 52 in El Salvador, followed in 19 55 in Nicaragua, 1958 in Honduras, and 1959 in Costa Rica and Guatemala (Bulmer-Thomas , 1987: 168). As industrialization acceler­ ated and attempts to coordinate incentives failed, countries competed by offering ever-increasing exemptions to potential investors. All countries lost revenues, with losses estimated at 265 percent of the amount actually collected in Guatemala (Bulmer-Thomas , 1987: 18 1) .36 Industrialists pressed at first for tariffs to protect them from for­ eign competitors, increasing the revenues earned on imports. Once they 36 These figures assume zero substitution effects and are therefore exaggerated, but they still give an indication of the scope of the problem.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

71

turned their attention to integration through the Central American Common Market, however, these trade revenues fell. Tariffs on intrare­ gional trade were rapidly dropped, and the most expensive imports, cap­ ital equipment and inputs for industry, were offered privileged rates. To protect local industry and recover some of the lost revenues, the coun­ tries agreed to the San José protocol in 1968, attaching a tax surcharge on extraregional imports. This had limited effect, however, as producers shifted tariff-free intraregional trade even more severely toward con­ sumer goods, while external trade patterns shifted toward those goods with the lowest duties (ICEFI, 2007c). The failure to mobilize revenues from industrialists suggests that they were too weak in economic terms or too uncertain about their access to the state in political terms, or both. Quite simply, they were unwilling to offer their resources to support the state. A review of the tax structures that emerged over the period reveals differences across the states. Governments still depended on trade taxes, which rebounded after World War II, increasing by half in El Salvador to reach 63 percent of all taxes in 19 5 1, increasing to 64 percent of total taxes in Costa Rica by 19 5 5 , and to 50 percent of the total in Honduras by 1954. Within trade taxes, however, Costa Rica and Honduras were able to increase the value of revenues from exports, while the Salvadoran and Guatemalan governments dropped their export taxes in the 1950s. A similar pattern emerged in income taxes. Income taxes arrived lat­ est to Guatemala in 1963, replacing the existing tax on business profits that had existed since 1938. An income tax law was initiated in 1949 in Honduras, 19 5 1 in El Salvador, in 19 53 in Nicaragua, and was updated in 1946 in Costa Rica. In each country, the income tax fell mostly on sal­ aried, formal sector employees; and in El Salvador and Guatemala, coffee exporters were explicitly exempted. For most of the 1950s, income taxes contributed approximately 10 percent of total taxes, rising to around 20 percent during the industrial expansion of the 1960s and 1970s in all countries except Guatemala, where income taxes did not rise to 20 per­ cent of the total until the 1990s.37 Other direct taxes, such as property or wealth taxes, fared even worse, falling or staying at the same levels virtu­ ally everywhere except Nicaragua and El Salvador. 37 Best (1976) explains the distorted structure of Central American taxes, heavily reliant on indirect taxes, with the dominance of agricultural elites, whose aversion to income and property tax was reflected directly in the ordering of tax structures.

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

State-Building and Tax Regimes in Central America

72

3.6. Post-WWII Tax as Percent o f GDP 1960

197°

1980

9 -5 4

12 .2 7 10.73 12.04 8.24 9 '9 2

z 3 '5 3 12 .3 1 11.0 5 8.67 10 .52

14 .8 1 11.8 8 11.2 3 12.6 2

8.19 9 .5 7

6.67 6.72

OO H

Costa Rica Honduras El Salvador Guatemala Nicaragua3

1 950

H 's i

t a b l e

a Nicaragua data from 1949 and 1978. Source: Tax as Percentage of GDP from Oxford Latin American Economic History Database; Honduras 19 50 and 1960 from Mitchell (2003)

Countries also turned to new indirect taxes, beginning with retail taxes, but El Salvador and Guatemala failed to introduce them. Instead, they relied on administratively difficult, regressive, and unpredictable stamp duties. The general weakness of revenues, even where they expanded, placed fiscal pressure on all governments of the region during the 1960s and 1970s. Bulmer-Thomas observes, “ The fiscal crisis exposed many of the weaknesses of the Central American development model in the 1960s. Powerful pressure groups were able to cripple efforts at fiscal reform, while leaving traditional exports virtually untouched; this left the “ old model of export-led growth based on primary products firmly in place, despite the appearance alongside it of a ‘new’ model based on industrialization and balanced regional development” (Bulmer-Thomas, 1987: 184). To meet fiscal crisis, governments in Guatemala, El Salvador, and Nicaragua chose to cut severely into public investment and social spend­ ing while Costa Rica sustained its spending with expansions in tax capac­ ity and an even more impressive expansion in social security. By the end of the period, social security revenues in Costa Rica were equal to 20 percent of total revenues, almost twice the amount in the other coun­ tries. Honduras also avoided severe cuts in spending and undertook some expansion of its tax capacity in the 1960s and 1970s, although it neither expanded as much nor offered as many benefits as in Costa Rica. Still, Costa Rica and Honduras ended the period with the highest revenues, at 17 .8 1 percent and 14 .8 1 percent, with Guatemala and El Salvador below 12 percent, and Nicaragua barely above. Data is also available on the proportions of revenue provided by differ­ ent sources. In Figure 3 .1, these proportions are displayed for the period

Downloaded from https://www.cambridge.org/core. YBP Library Services, on 27 Aug 2018 at 20:49:03, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139094061.005

Historical Junctures

73

1950 to 19 7 5 .38 The change in proportions shows the general trend that occurred across the region. Revenues from taxes on trade increased in the early part of the period, as incipient industrialists sought protection from imports, but these fell over time as trade barriers were removed in the process of regional integration. Domestically levied revenues on con­ sumption and income taxes increased their share. Despite these similar overall trends, a few observations are evident. The main one is that Costa Rica would appear to have started already with a higher proportion of revenues from domestic taxes such as income and Costa Rica 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

■ Non-tax ■ Consumption ■ Trade ■ Property ■ Income N