University of Southern California Dissertations and Theses

The murky risk of trade protectionism in an interconnected and uncertain global economy: a state, industrial, and regional analysis

(USC Thesis Other)

The murky risk of trade protectionism in an interconnected and uncertain global economy: a state, industrial, and regional analysis

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content The Murky Risk of Trade Protectionism in an Interconnected and Uncertain Global Economy A State, Industrial, and Regional Analysis By Nicolás Domingo Albertoni Gomez A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (POLITICAL SCIENCE AND INTERNATIONAL RELATIONS) August 2021 Copyright 2021 Albertoni Gomez, Nicolás Domingo ii Epigraph Between stimulus and response there is a space. In that space is our power to choose our response. In our response lies our growth and our freedom. VIKTOR E. FRANKL, MAN’S SEARCH FOR MEANING, 1946 iii Dedication For mamá, papá, Josefina, Florencia, and Alfonsina iv ACKNOWLEDGEMENTS Over the last five years as a doctoral student, I have benefited from the advice and support of many mentors, friends, colleagues, and students. I cannot imagine having written this dissertation and completing my PhD without their enormous support. I would like to start by thanking my family: Carlos, Sylvia, Josefina, Florencia and Alfonsina. They are my biggest supporter and I owe all of my successes to them. I owe many thanks to my dissertation committee: my doctoral advisor, Professor Carol Wise, and Professors Jeff Nugent and Joshua Aizenman. They not only taught me how to improve my research and better contribute to the discipline; they also taught me how to become a better scholar. From all of them I received excellent comments and suggestions that constituted an important part of the ideas and arguments I am presenting in this dissertation. I would especially like to thank Carol for her immense support in these five years that I have been working with her. It was due to her advice and enthusiasm that I entered the doctoral program. As a graduate student at Georgetown University, I read and followed her academic work, and I would never imagine at that time that I would work with her as one of her doctoral students at USC. Now, I can say that I am extremely grateful she invited me, for her guidance, patience, encouragement, and intellectual and personal support: sometimes teacher; sometimes like a mother; sometimes a senior scholar who supported my research. Always a mentor. Her many suggestions and careful reading improved every aspect of my doctoral works. In thanking her I would also like to extend my thanks to Roger, as he and Carol always let my wife and me feel like we had a family in L.A. v I would also like to thank Joshua Aizenman for his continuous encouragement to pursue new ideas. His support allowed me to reshape my research project. From Joshua I learned that our scholarly research has to be oriented to a better interpretation of the world. Jeff Nugent I would especially like to thank for his enormous generosity. Having him as a professor in my economics classes and then as part of a doctoral committee was a gift. He was always attentive of my research progress and supportive throughout my graduate studies. From Jeff I learned that the details are worth exploring. I am also extremely grateful to Ben Graham for his generous support over the last years. Ben gave me invaluable advice and comments on earlier drafts of my research project. It was due to his enthusiasm that I entered the USC’s Security and Political Economy (SPEC) Lab where I created the Trade Policy Project. Here, I was given the opportunity to create a fabulous team of students to work on my research projects. In that regard, I wish to thank the many students that have been part of the Trade Policy Project and from whom I received excellent research assistance: Madeline Zheng, Emily Heuring, John “Jack” Schwartz, Shir Attias, Sonum Patel, Amanda Joseph, and Kartik Mathur. I would also like to thank the USC Center for International Studies that awarded me the student research fund in 2020 and allowed me to work with Claire Schreder and Claire Post, from whom I received very helpful comments and editions of this dissertations. Several faculty members at the University of Southern California have been tremendously generous with their time and support: Gerardo Munck, Wayne Sandholtz, Saori Katada, Pablo Barberá, Patrick James, Iva Bozovic, and James Lo from my Department of Political Science and International Relations; Ratika Narag, Geert Ridder, and Hyungsik Roger Moon from the Department of Economics; and lastly, George Gerald Vega Yon, Kayla de la Haye, and Tom Valente from USC’s Center for Applied Network Analysis (CANA) for their suggestions on earlier versions of this paper. George specifically has been vi an immense support in this journey. I will always be extremely grateful for his support with the data analysis and the dataset in which we worked together to improve the quantitative analysis of this dissertation. Many scholars around the world contributed with insightful comments in different steps of this research project, including Marcelo Olarreaga (University of Geneva), Barbara Kotschwar (Georgetown University), Ricardo Ernst (Georgetown University), Diana Kapiszewski (Georgetown University), Andrew Bennett (Georgetown University), Andrés Rebolledo (Universidad SEK, Chile), Roberto Horta (Universidad Católica del Uruguay), and Bryn Rosenfeld (Cornell University). I also thank the more than 20 individuals from various backgrounds and professions around Latin America that accepted my invitation to be interviewed for this dissertation. For their invaluable advice and comments on earlier drafts I am grateful to many colleagues: Victoria Chonn Ching, Timo Dahler, Paul Orner, Miriam Barnum, Gabrielle Cheung, Kyle Rapp, Eboni Nola Haynes, Kelebogile Zvobgo, Brian Knafou, Joey Huddleston, Daniela Maag, and Manuel Martinez. I am also grateful for the constant support of Veri Chavarin and the entire staff of POIR. I would also like to thank the Global Trade Alert Dataset team, especially Piotr Lukaszuk, for his generosity in explaining to me many key details about the dataset. This dissertation has benefited from many grants and scholarships that allowed me to improve my study. I thank the National Science Foundation (NSF) for the Fellowship (1558713) I received to support my participation in the 11th Annual Political Networks Conference, where I received very useful comments on this project; the National Council on International Trade and Development (NCITD) for the 2018-2019 Award to this research project, which provided important support; and the USC Center For International Studies vii (CIS) fund for graduate research (2020) and the USC Korean Studies Institute Grad Student Affiliates’ grant, which also helped me in my fieldwork. Thanks to this dissertation project, I also obtained the American Political Science Association’s 2020 Fund for Latino Scholarship, and I was also recognized as the New Public Intellectuals in Latin America in 2018 by the Global American Journal. Additionally, I was awarded a USC research enhancement grant to participate at the Institute for Qualitative and Multi-Method Research, where I received many comments and suggestions that positively impacted my work. Naturally, any errors that remain are my own responsibility. viii TABLE OF CONTENTS EPIGRAPH ............................................................................................................................................................................... ii DEDICATION ......................................................................................................................................................................... iii ACKNOWLEDGEMENTS .................................................................................................................................................... iv LIST OF TABLES .................................................................................................................................................................... x LIST OF FIGURES ................................................................................................................................................................ xi ABSTRACT ........................................................................................................................................................................... xiv CHAPTER 1 – INTRODUCTION, RESEARCH QUESTIONS, THE LITERATURE, GOALS AND SIGNIFICANCE OF THIS DISSERTATION ....................................................................................................................... 1 1.1 Introduction .......................................................................................................................................................... 1 1.2 Definition and Conceptualization of the Main Variables ............................................................................... 14 1.2.1 Trade Policy and the Different Ways to Measure it .................................................................................. 15 1.2.1.1 What does trade policy really mean? ................................................................................................... 15 1.2.1.2 Tariffs, non-tariff measures and “murky” protectionism .................................................................... 16 1.2.1.3 The Complex Task of Defining Policy Responses ................................................................................ 20 1.2.1.4 Preferential Trade Agreements (PTAs) ............................................................................................... 22 1.2.1.5 Global Value Chains (GVCs) ............................................................................................................... 25 1.3 Theoretical Approaches in the Literature and Initial Hypotheses ................................................................ 28 1.3.1 State Level Dynamics in Trade Protectionism ........................................................................................... 29 1.3.2 Sectoral Dynamics ........................................................................................................................................ 33 1.3.3 Contribution of this Dissertation to the Trade Policy Literature ............................................................ 37 1.4 A Two-Step Research Design ............................................................................................................................ 39 1.4.1 Large-N Analysis .......................................................................................................................................... 39 1.4.2 Small-N Analysis and Case Studies ............................................................................................................. 40 1.5 Historical Context and Significance of this Research ..................................................................................... 42 1.6 Outline of the Dissertation ................................................................................................................................ 46 CHAPTER 2 – A HISTORICAL OVERVIEW OF 21st-CENTURY PROTECTIONISM: HOW DID WE ARRIVE AT THIS POINT? ................................................................................................................................................................... 47 2.1 Introduction ........................................................................................................................................................ 47 2.2 The Winding Road of the World Trade System Since the World War II .................................................... 48 2.3 China’s Accession to the World Trade Organization in 2001 ....................................................................... 54 2.4 Global Trade After the 2008-09 Global Financial Crisis ............................................................................... 58 2.5 Bilateral Tensions Between the U.S. and China in the Present and the Past ......................................................... 59 2.5.1 U.S.-China Trade Tensions Step by Step ................................................................................................... 63 2.6 Final Comments on 21st-Century Protectionism from a Historical Perspective ................................................... 66 CHAPTER 3 – UNDERSTANDING TRADE PROTECTIONISM PIECE BY PIECE. A DESCRIPTIVE ANALYSIS OF THE DATA .................................................................................................................................................. 69 3.1 Introduction ........................................................................................................................................................ 69 3.2 The Data .............................................................................................................................................................. 70 3.2.1 Global Trade Alert (GTA) ........................................................................................................................... 70 3.2.2 BACI-CEPII Trade Data ............................................................................................................................. 75 3.2.3 The IPE Data Resource ................................................................................................................................ 76 3.2.4 Data sources for other variables of interest: PTAs and GVCs ................................................................. 77 3.2.4.1 Measuring Preferential Trade Agreements (PTAs) ........................................................................... 77 3.2.4.2 Measuring Global Value Chains (GVCs) .......................................................................................... 78 3.3 A Descriptive Analysis at the State and Industrial Levels of Trade Policy Dynamics ................................ 79 3.3.1 Twenty-First Century Protectionism Piece by Piece: States, Sectors, and Measures ............................ 80 ix 3.3.1.1 State-Level Dynamics .......................................................................................................................... 80 3.3.1.2 Industry and Sectoral Dynamics ........................................................................................................ 86 3.3.1.3 Types of Measures Implemented ........................................................................................................ 89 A Descriptive Text Analysis of the Types of Measures Implemented ................................................................... 92 3.4 A Descriptive Analysis of Preferential Trade Agreements and Global Value Chains ................................ 95 3.3.1 Preferential Trade Agreements ................................................................................................................... 95 3.3.2 Global Value Chains ..................................................................................................................................... 97 3.4 Final Thoughts on Rising Protectionism vis-à-vis the COVID-19 Pandemic ............................................. 101 CHAPTER 4 – COUNTRY AND SECTORAL ANALYSIS ON PROTECTIONISM AND TRADE POLICY RESPONSES: A QUANTITATIVE APPROACH ............................................................................................................ 106 4.1 Introduction ...................................................................................................................................................... 106 4.2 The Models ....................................................................................................................................................... 109 4.3 The Data ............................................................................................................................................................ 111 4.3.1 Measurement of the Dependent Variables ............................................................................................... 112 4.3.2 Measurement of Regressors of Interest .................................................................................................... 114 4.3.3 Control Variables ....................................................................................................................................... 116 4.4 State-Level Analysis Results ........................................................................................................................... 119 4.5 Political and Economic Magnitude of the Empirical Findings at the State-Level Analysis ..................... 124 4.6 Sectoral-Level Analysis Results ...................................................................................................................... 130 4.7 Conclusions on the Quantitative Analysis ..................................................................................................... 138 CHAPTER 5 – A COMPARATIVE APPROACH: PROTECTIONISM AND TRADE POLICY RESPONSES IN LATIN AMERICA ................................................................................................................................................................ 142 5.1 Introduction ...................................................................................................................................................... 142 5.2 Case Selection: Why Latin America and Why These Three Countries? .................................................... 145 5.3 Fieldwork: Interviews with Key Trade Policy Actors in Latin American ................................................. 158 5.3.1 Interview Design and Methodology .......................................................................................................... 158 5.4 A Brief Background of Latin America’s Trade Policy Evolution ............................................................... 161 5.4.1 Latin America Regional Integration ........................................................................................................... 168 5.5 Country Case Studies ...................................................................................................................................... 177 5.5.1 The Case of Mexico ..................................................................................................................................... 178 5.5.1.1 Mexico’s Trade Policy Evolution and Key Actors Involved ............................................................ 178 5.5.1.2 Impacts of Mexico’s Trade Protectionism and Economic Responses ............................................ 185 5.5.1.3 Political Economy Insights Based on the Case of Mexico .............................................................. 194 5.6.1 The Case of Argentina ................................................................................................................................ 203 5.6.1.1 Argentina’s Trade Protectionism: Economic Impacts and Responses ........................................... 207 5.6.2 The Case of Brazil ....................................................................................................................................... 216 5.6.2.1 Brazil’s Trade Protectionism: Economic Impact and Responses ................................................... 221 5.6.2.2 Political Economy Insights Based on the Case of Mercosur .......................................................... 228 5.7 Conclusions and Country Cases’ Summary ............................................................................................. 235 CHAPTER 6 – CONCLUSIONS ......................................................................................................................................... 244 6.1 Main Research Contribution and Its Relevance to the Field ....................................................................... 244 6.2 Implications for International Political Economy Theory and Practice ..................................................... 249 6.3 Implications for Policy ..................................................................................................................................... 255 6.4 Insights for Further Research ......................................................................................................................... 259 BIBLIOGRAPHY ................................................................................................................................................................. 263 APPENDICES ....................................................................................................................................................................... 278 x LIST OF TABLES Table 1.1 – Relationship between the Literature and the Main Argument of this Dissertation ............ 9 Table 1.2 – General and Specific Research Questions of this Dissertation ........................................ 10 Table 1.3 – Mapping Trade Policy Dynamics by Region ................................................................... 14 Table 1.4 – Examples of NTMs Contemplated Under the UNCTAD’s Classification and GTA ....... 17 Table 1.5 – Interventions and their Level of Transparency ................................................................. 19 Table 1.6 – Capital, Intermediate and Consumption Goods (in BEC Classification) ......................... 28 Table 1.7 – General Hypotheses (Hypotheses 1, 2, and 3) .................................................................. 33 Table 1.8 – The Relationship Between Tariffs and NTMs with PTAs and GVCs (Hyp. 4 and 5) ..... 33 Table 1.9 – Sectoral Dynamics (Hypotheses 6, 7, and 8) .................................................................... 37 Table 3.1 – Name and Description of GTA Variables ........................................................................ 71 Table 3.2 – A Real Example of Data Collection: Step-by-Step .......................................................... 74 Table 4.1 – Summary Statistics, State-Level Analysis ...................................................................... 118 Table 4.2 – Summary Statistics, Sectoral-Level Analysis (% of Trade Affected by Sector) ............ 119 Table 4.3 – DV1a: All Protectionist Measures by A on B in Time t ................................................. 121 Table 4.4 – The Relationship Between Tariff and Non-tariff Measures with PTAs and GVCs (DV2a and DV2b) .......................................................................................................................................... 123 Table 4.5 – The Relationship Between Tariff and NTMs with PTAs and GVCs (DV3a and DV3b) 124 Table 4.6 – Responses by Sector Using BEC Classification (DV4) ................................................. 132 Table 4.7 – Responses by Sector Using HS classification (Sector 0 through Sector 9) (DV4) ........ 133 Table 4.8 – Summary of Hypotheses and Evidence from Results of Quantitative Analysis ............ 137 Table 5.1 – Main Economic and Trade Characteristics of Argentina, Brazil, and Mexico .............. 150 Table 5.2 – List of Interviewees ........................................................................................................ 160 Table 5.3 – Weighted Mean Tariff Rate Applied to All Products (%) .............................................. 164 Table 5.4 – Main Trade Characteristics of Latin America (2019) .................................................... 166 Table 5.5 – List of the Regional Agreements in Latin America ....................................................... 170 Table 5.6 – Latin America Intermediate Goods Destinations and Types of Measures Implemented Toward Those Destinations (2008-19) ............................................................................................... 175 Table 5.7 – Most Affected Sectors in and by Latin America (2009-19) ........................................... 176 Table 5.8 – Mexico Trade Agreements, Blocs, and International Forums (199-2019) ..................... 180 Table 5.9 – Main Actors Involved in the Development of Mexico’s Trade Policy .......................... 182 Table 5.10 – Top Destinations of Intermediate, Consumption, and Capital Goods (Average Share in Mexico’s Total Exports Between 2008 and 2018) ............................................................................. 184 Table 5.11 – Sectors Most Affected in and by Mexico (2009-2019) ................................................ 193 Table 5.12 – Mercosur Trade Agreements ........................................................................................ 202 Table 5.13 – Top Destinations of Intermediate, Consumption, and Capital Goods (Average Share in Argentina’s Total Exports, 2008-2018) ............................................................................................. 207 Table 5.14 – Most Affected Sectors in and by Argentina (2009-2019) ............................................ 214 Table 5.15 – Top Destinations of Intermediate, Consumption, and Capital Goods (Average Share in Brazil’s Total Exports Between 2008 and 2018) ............................................................................... 220 Table 5.16 – Most Affected Sectors in and by Brazil (2009-2019) .................................................. 227 Table 5.17 – Table Summary of Regional and Country Cases ......................................................... 243 Table 6.1 – Summary of Theory Background, Argument, Hypotheses, Methods, and Results ........ 248 xi LIST OF FIGURES Figure 1.1 – Comparing Protectionist and Liberalizing Measures Between 2009-2019 ..................... 2 Figure 1.2 – Between 2009 and 2018, the effect of G20 protectionism on LDCs has been concentrated in non-tariff measures. ................................................................................................... 12 Figure 1.3 – Sectors Within LDCs Most Affected by G20 Protectionism (2008-19) ........................ 13 Figure 1.4 – A Retaliatory Dynamic .................................................................................................. 21 Figure 1.5 – How do Global Value Chains Work? ............................................................................ 26 Figure 1.6 – Share of World Exports to G20 Countries Affected by Protectionist Measures (Average 2009-2018, by Country) ...................................................................................................................... 41 Figure 1.7 – Google Searches on “Trade Policy” and “Protectionism” (2008-19) ............................ 44 Figure 1.8 – World Trade Uncertainty index (1996 to 2019) ............................................................ 45 Figure 2.1 – The Evolution of Tariff Rates for Primary Products and Manufactured Goods Imposed by China, the U.S., and the EU. .......................................................................................................... 56 Figure 2.2 – Putting 2019 Global Trade Policy Dynamics in Perspective ......................................... 67 Figure 3.1 – Key Steps in GTA’s Data Collection ............................................................................. 72 Figure 3.2 – Example of an Official Trade Policy Implementation Document ................................. 74 Figure 3.3 – Summary of Data Sources Compiled in the Final Database .......................................... 76 Figure 3.4 – New Trade Policy Interventions Per Year ..................................................................... 80 Figure 3.5 – Share of World Imports Distorted by World Protectionist Restrictions ........................ 81 Figure 3.6 – Number of Protectionist and Liberalizing Measures by Region .................................... 82 Figure 3.7 – Share of World Exports to Each Region Affected by Each Region Protectionism Measures (Average 2009-2018) .......................................................................................................... 83 Figure 3.8 – Number of Protectionist Measures of G20 Countries Toward the Rest of the World’s Imports (Average 2009-18, by Country) ............................................................................................. 84 Figure 3.9 - Share of World Exports to G20 Countries Affected by Protectionism Measures implemented from the Rest of the World (Average 2009-18, by Country) ........................................ 85 Figure 3.10 – How Responsive Were Key Emerging Economies to Trade Protectionism? .............. 86 Figure 3.11 – Distribution of Protectionist Measures by Year and Sectors Affected ........................ 87 Figure 3.12 – Most Affected Sectors by Region (2008-19) ............................................................... 88 Figure 3.13 – Protectionist Policy Instruments Most Used Globally Pre- (2000-07) and Post-GFC (2008-19) by types of measure ............................................................................................................ 90 Figure 3.14 – Protectionist Policy Instruments Most Used Globally, Pre- (2000-07) and Post-GFC (2008-19) by the level of transparency ............................................................................................... 90 Figure 3.15 – Protectionist Instruments by Region Pre- (2000-07) and Post-GFC (2008-19) by the Level of Transparency ......................................................................................................................... 91 Figure 3.16 – Protectionist Instruments by Country Pre- (2000-07) and Post-GFC (2008-19) by the Level of Transparency ......................................................................................................................... 92 Figure 3.17 – Protectionist Policy Instruments with Name of Targeted Country (2009-19) ............. 95 Figure 3.18 – Preferential Trade Agreements Signed Until 2019 ...................................................... 96 Figure 3.19 – Number of PTAs of G20 Countries Until 1970-2019 ................................................. 97 Figure 3.20 –Example of the Global Value Chain for a Bicycle ....................................................... 98 Figure 3.21– GVC Share of Global Trade (%) .................................................................................. 99 Figure 3.22 – All Countries Participate in GVCs—But Not in the Same Way ............................... 100 Figure 3.23 – Global Production Networks and the Share of Value Added in Exports of Each Region by Source Region (2018) .................................................................................................................. 101 Figure 3.24 – WTO forecast: The Greater Trade Collapse of 2020 After COVID-19 .................... 103 Figure 3.25 – Unprecedented Trade Policy Response to COVID-19 .............................................. 104 Figure 4.1 –Modeling Retaliatory Dynamics by Sector ................................................................... 114 xii Figure 4.2 – Sectoral Retaliatory Dynamics Using BEC Classification .......................................... 134 Figure 4.3 – Sectoral Retaliatory Dynamics Using HS Classification ............................................. 134 Figure 5.1– Latin America as a Median Case in the Non-Developed World in Terms of Protectionism Measures ........................................................................................................................................... 146 Figure 5.2 – Distribution of the Three Case Countries in the G20 and Latin America (Average 2009- 18) ..................................................................................................................................................... 148 Figure 5.3 – Institutional Measures for Argentina, Brazil, and Mexico (Higher Indicates Better Performance) ..................................................................................................................................... 151 Figure 5.4 – How are the Three Selected Countries Distributed in the G20 and in Latin America on Preferential Trade Agreements ......................................................................................................... 152 Figure 5.5 – The GVC Participation Index (% share in total gross exports, in 2015) ..................... 154 Figure 5.6 – Trade in Merchandise Intermediates Products (2017) ................................................. 155 Figure 5.7 – Latin America Exports and Imports Trends (1960-2019) in trillions .......................... 165 Figure 5.8 – Latin American Exports of Goods and Services (% of GDP) ..................................... 165 Figure 5.9 – How Responsive is Latin America to Trade Protectionism? ....................................... 171 Figure 5.10 – Latin America Trade Policy Measures Within the Region and Toward G20 (2008-19) ........................................................................................................................................................... 173 Figure 5.11 – Latin America Exports of Intermediate Goods by Region ........................................ 175 Figure 5.12 – Mexico’s Export of Intermediate, Consumer, and Capital Goods Between 2000 and 2018 (in thousands of dollars) ........................................................................................................... 184 Figure 5.13 – Share of Mexico’s Total Exports Affected (t-1) vs. World Exports to Mexico Affected (time t) by Protectionist Measures (Average 2009-19) ..................................................................... 186 Figure 5.14 – Share of Mexico’s Exports Impacted by Close and Not Close Trading Partners’ Protectionism ..................................................................................................................................... 187 Figure 5.15 – Share of Close and Not Close Trading Partners’ Exports to Mexico Impacted by Mexico’s Protectionism .................................................................................................................... 187 Figure 5.16 – Share of Mexico’s Exports Impacted by GVC Ties vs. No GVC Ties ..................... 188 Figure 5.17 – Mexico’s Number of Protectionist Measures Implemented Toward Close and Not Close Trading Partners ...................................................................................................................... 189 Figure 5.18 – Number of Protectionist Measures Toward Mexico by Close and Not Close Trading Partners .............................................................................................................................................. 190 Figure 5.19 – Share of Non-Transparent Protectionist Interventions by Mexico toward Close and not Close Trading Partners (average 2009-2019) ................................................................................... 191 Figure 5.20 – Share of Non-Transparent Protectionist Interventions toward Mexico by Close and not Close Trading Partners (average 2009-2019) ................................................................................... 191 Figure 5.21 – Mexico’s Employment in Industry, Agriculture, and Services (% of total employment) ........................................................................................................................................................... 194 Figure 5.22 – Total Number of Measures Affecting Mexico by Main Economic Sectors (2009-2019) ........................................................................................................................................................... 194 Figure 5.23 – Trade Policy During Cristina Fernandez vs. Mauricio Macri ................................... 204 Figure 5.24 – Argentina’s Export of Intermediate, Consumption, and Capital Goods Between 2000 and 2018 (in U.S. dollars) ................................................................................................................. 206 Figure 5.25 – Share of Argentina’s Total Exports Affected (t-1) vs. World Export to Argentina Affected (time t) by Protectionist Measures (Average 2009-2019) .................................................. 209 Figure 5.26 – Share of Argentina’s Exports Impacted by Protectionism from Close and not Close Trading Partners ................................................................................................................................ 210 Figure 5.27 – Share of Close and not Close Trading Partners Exports to Argentina Impacted by Argentina’s Protectionism ................................................................................................................. 211 Figure 5.28 – Argentina’s Number of Protectionist Measures Implemented toward Close and Not Close Trading Partners ...................................................................................................................... 212 xiii Figure 5.29 – Number of Protectionist Measures toward Argentina by Close and Not Close Trading Partners .............................................................................................................................................. 212 Figure 5.30 – Share of Non-Transparent Protectionist Measures by Argentina toward Close and Not Close Trading Partners (average 2009-2019) ................................................................................... 213 Figure 5.31 – Share of Non-Transparent Protectionist Measures toward Argentina by Close and Not Close Trading Partners (average 2009-2019) ................................................................................... 214 Figure 5.32 – Argentina’s Employment in Industry, Agriculture, and Services (% of total employment) ..................................................................................................................................... 215 Figure 5.33 – Number of Protectionist Measures Affecting Argentina by Main Economic Sectors (2009-2019) ....................................................................................................................................... 216 Figure 5.34 – Brazilian Exports of Intermediate, Consumer, and Capital Goods Between 2000 and 2018 (in US$) .................................................................................................................................... 220 Figure 5.35 – Share of Brazil’s Total Exports Affected (t-1) vs. World Export to Brazil Affected (time t) by Protectionist Measures (Average 2009-2019) ................................................................. 222 Figure 5.36 – Share of Brazil’s Exports Impacted by Close and not Close Trading Partners’ Protectionism ..................................................................................................................................... 222 Figure 5.37 – Share of Close and not Close Trading Partners’ Exports to Brazil Impacted by Brazil’s Protectionism ..................................................................................................................................... 223 Figure 5.38 – Brazil’s Number of Protectionist Measures Implemented Toward Close and not Close Trading Partners ................................................................................................................................ 224 Figure 5.39 – Number of Protectionist Measures toward Brazil by Close and not Close Trading Partners .............................................................................................................................................. 224 Figure 5.40 – Share of Protectionist Interventions that Use Non-Transparent Measures by Brazil toward Close and not Close Trading Partners (average 2009-2019) ................................................ 225 Figure 5.41 – Share of Protectionist Interventions that Use Non-Transparent Measures toward Brazil by Close and not Close Trading Partners (average 2009-2019) ....................................................... 226 Figure 5.42 – Brazil’s Employment in Industry, Agriculture, and Services (% of total employment) ........................................................................................................................................................... 228 Figure 5.43 – Number of Total Measures Affecting Brazil by Main Economic Sectors (2009-2019) ........................................................................................................................................................... 228 Figure 6.1 – A Logical Diagram Summarizing Theory and Main Research Questions .................. 247 xiv ABSTRACT This dissertation is neither about free trade nor protectionism. It is about the point where the two may converge. One distinctive aspect of rising protectionism over the last decade is that it is occurring against a backdrop of increased global interconnectivity. Recently, countries around the world have become more integrated through preferential trade agreements (PTAs) and global value chains (GVCs), even though they are erecting trade barriers within these same trade venues. What explains these contradictory trajectories? The recent literature tells us how PTAs and GVCs promote trade interdependences, and some of the earlier work on the political economy of trade protectionism offers compelling arguments about the circumstances under which governments decide to protect their economies. However, we have little evidence on how these two trajectories may simultaneously interact. This dissertation concentrates on how states––especially in the developing world––respond to the protectionist trends that have arisen in context of high interconnectivity and uncertainty. This question is vital to debates about international political economy and trade policy from a theoretical and policy perspective. We have now entered an era in which major trade policy tensions are growing faster than the generation of theories to explain them. Adopting a mixed methods approach based on a large-N statistical analysis combined with a detailed country case analysis, this research offers new evidence that institutional mechanisms like PTAs, which for decades have fueled the diffusion of trade liberalization, can also become channels for the spread of less-transparent protectionism (e.g., Non–Tariffs Measures) in an interconnected and uncertain global economy. I also conduct an initial empirical analysis of how GVCs have exacerbated bilateral protectionist dynamics. My time frame is the period from the 2008-09 Global Financial Crisis (GFC) to 2019. I examine: (1) xv the role of PTAs and Non–Tariffs Measures (NTMs) in the escalation of trade protectionism, (2) how emerging economies (with a focus on Latin America) have been affected by and responded to rising protectionist trends especially in the realm of NTMs, and (3) the specification of how industry/sectoral dynamics have shaped protectionist policies and responses. The key contribution of this research is my analysis of how three main variables have interacted with ongoing trade policy dynamics since the GFC. By this, I refer to the political economy “trilemma” between trade protectionism, trade interconnectivity (e.g., PTAs and GVCs), and uncertainty (mostly generated by economic crises). This dissertation is a cautionary tale about trade protectionism in an era of trade interdependence and high levels of uncertainty. This project bridges academic analysis with trade policy-making and offers a roadmap for the kinds of commercial policy reforms that will be essential for the successful revival of world markets post-COVID-19. 1 CHAPTER 1 – INTRODUCTION, RESEARCH QUESTIONS, THE LITERATURE, GOALS AND SIGNIFICANCE OF THIS DISSERTATION We live in a difficult world. If everyone is becoming more protectionist, then we have to protect ourselves too. We know, though, that we have to be careful with our close trading partners, as not to escalate the problem. That’s why our union asked the government to work “in silence” and implement policies that could help us become more competitive without being as obvious to the rest of the world. We are not trying to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context. ANONYMOUS INDUSTRIAL UNION LEADER IN MEXICO 1 1.1 Introduction It has been more than a decade since the 2008-09 global financial crisis (GFC) and the accompanying collapse of trade (Baldwin, 2009a). 2 In the last decade, many countries have erected new trade barriers that contrast sharply with the thrust of global trends over the past seventy years, which saw a sustained opening of national markets under the Bretton Woods 1 Author’s interview with a union industry leader in Mexico, December 2019. See appendix 5 for the full interview 2019 (interviewee requested no attribution). 2 As different studies have shown, from 2008 to 2009 global trade fell by 15%, surpassing the fall in real world GDP (Baldwin, 2009; Bems et al., 2013). 2 system. 3 The World Trade Organization has been urging its members “to resist protectionism and get trade moving again” (WTO, 2016 para. 1). 4 However, the recent literature on trade policy suggests this is not occurring. Indeed, recent research has focused on bilateral trade tensions that are occurring around the world, such as the “U.S.-China Trade War,” which formally began in March 2018 when the U.S. imposed safeguard tariffs on a long list of Chinese products (Bown & Kolb, 2019). Recent academic work on trade protectionism has concentrated on these sorts of bilateral trade conflicts, rather than explaining what has really happened in the decade since the GFC. My main starting point concerns the rise of trade protectionism between the outbreak of the 2008-09 GFC and the current U.S.-China trade war. Figure 1.1 shows unilateral measures implemented by 170 countries of different regions across the world, representing 98% of global trade flows. Figure 1.1 – Comparing Protectionist and Liberalizing Measures Between 2009-2019 Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. 3 As a WTO (2016, para. 11) report shows, in 2016 “of the 2,835 trade-restrictive measures, including trade remedies, recorded for WTO Members since 2008 by this exercise, only 708, or 25%, had been removed by mid-May 2016.” 4 WTO (2016, July 4) Report to the TPRB from the Director-General on trade-related developments, <https://www.wto.org/english/news_e/news16_e/trdev_22jul16_e.htm>. 3 This dissertation analyzes the rise of a new pattern of trade protectionism beginning with the GFC up to the present. My emphasis is on the developing world and my research question concerns how countries have responded to rising protectionism in a context of high trade interdependence and economic uncertainty. This question is vital to debates about international political economy and trade policy from a theoretical and policy perspective. We have now entered an era in which major trade policy tensions are erupting faster than the generation of theories to explain them. This study of trade policy trends from the GFC to the present promises to put the Sino-American tariff war in perspective and situate it as part of a deeper and longer pattern of 21st-century protectionism (Evenett, 2019). One distinctive aspect of this rising protectionism is that it is occurring against a backdrop of intense globalization. Paradoxically, through PTAs and GVCs, countries around the globe are becoming more interconnected even as they erect new trade barriers. This trend starkly contrasts with past patterns of protectionist escalation. The present literature argues that the more interconnected countries are, the less protectionist they are likely to become (Baldwin, 2012; Lamy, 2013; Gawande et al., 2015; Jensen et al., 2015). However, it has barely spoken to this newer contradictory phenomenon where trade protectionism occurs under conditions of high interdependence and economic uncertainty (Ahir et al., 2018). My dissertation is motivated by this contradiction. I hypothesize that protectionist trends over the last decade reveal a possible downside of PTAs and GVCs. That is, protectionism is rising in the form of less observable, non–tariff measures despite more open trade and production schemes. In other words, economic interdependence in the context of high trade uncertainty is devolving into a spiral of protectionism because “many 4 governments simultaneously face pressure to reflate national economies and defend national commercial interests” (Evenett, 2019, p. 26). This research project offers preliminary evidence that institutional mechanisms like PTAs, which for decades have fueled the diffusion of trade liberalization, can also become channels for the spread of protectionism in the context of high economic uncertainty. 5 In the case of GVCs, the goal of this dissertation is to introduce a preliminary analysis of their potential effects in bilateral protectionism dynamics, since it is still very challenging to have a perfect measure of them (Gaulier et al., 2019). As I will explain later, GVCs refer to international production sharing, a phenomenon where production is broken into activities and tasks carried out in different countries. Hence, given its complex structure, there is considerable debate about how to properly measure GVCs (Gaulier et al., 2019; Hummels et al., 2001). The share of intermediate goods traded between two countries is normally used as a proxy measure for the GVC between them. Gaulier et al. (2019) shows that in the last decade, “the share of intermediate goods in world trade in nominal terms is fairly well correlated to various Global Value Chain (GVC) indicators based on international input- output matrices” (p. 1). Hence, I use the exchange of intermediate goods as the best approximation to measure the potential “GVC interconnection” between two countries (Escaith et al., 2010; De Backer & Miroudot, 2014). No data is perfect, but in later chapters I endeavor to acknowledge the flaws to uncover potential biases in econometric work in chapter 4. 5 Different to PTAs, which undoubtedly have fueled the diffusion of trade liberalization, GVC can be seen as a result of trade liberalization and not a cause of it. 5 At its heart, this dissertation is an analysis of how countries’ bilateral protectionist dynamics have evolved over the last decade. My focus will be on: (1) the role of Preferential Trade Agreements (PTAs), Global Value Chains (GVCs) and Non–Tariffs Measures (NTMs) in the escalation of trade protectionism; (2) how emerging economies (with a focus on Latin America) have been affected by and responded to rising protectionist trends; and (3) how industry/sectoral dynamics affect protectionist responses. In sum, my research highlights the risk of protectionism in an interconnected world. As China’s President Xi Jinping has remarked, “in a world of deepening economic globalization, practices of the law of the jungle and winner-takes-all only represent a dead end.” 6 The COVID-19 pandemic has stopped globalization in its tracks, an extraordinary circumstance that makes for a natural experiment on how drastically slowed trade affects the global economy and developing countries in particular. 7 Keeping this in mind, I compile a novel dataset that captures how global trade has been hampered by the rise of protectionist policy interventions over the past decade. The significance of my dissertation is the empirical documentation of the damage that closed markets and trade nationalism can inflict on developing countries. In essence, my dissertation is a cautionary tale about economic nationalism in an era of trade interdependence and high levels of uncertainty. It also 6 Keynote speech by President Xi Jinping at opening ceremony of 1st China International Import Expo, <http://www.xinhuanet.com/english/2018-11/05/c_137583815.htm>; also see BBC, Xi Jinping pledges to cut Chinese import tariffs (November 5, 2018), <https://www.bbc.com/news/business- 46093478>. 7 In a June 2020 report about the effects of COVID-19 on trade, the World Trade Organization suggested that the collapse in trade now could be far bigger than in response to the 2008 financial crisis. See WTO (2020a) Trade falls steeply in first half of 2020, https://www.wto.org/english/news_e/pres20_e/pr858_e.htm; and Wolf, Martin (2020) The dangerous war on supply chains. Financial Times (June 23, 2020), <https://www.ft.com/content/e27b0c0c-1893-479b-9ea3-27a81c2506c9?>. 6 contributes to the debate among trade policy scholars on whether or not the rise of protectionism following the GFC is on par with other major trade collapses in history, such as that which occurred during the Great Depression of the 1930s. Some scholars prematurely argued that there was a limited increase in protectionism after the GFC (Rodrik, 2009; Boffa & Olarreaga, 2012; Kee et al., 2013) and stated that this “should be seen as a victory for the political consensus and economic institutions that underpin the global economy” (Bems et al., 2013, p. 376). However, hindsight now shows that increases in protectionism were much larger and more pervasive than these scholars originally thought (Evenett, 2019). In this dissertation, I argue that there is a need for improved measurement and careful analysis of the various types and forms of protectionism that have evolved since the GFC. As Evenett (2019, p. 10) has highlighted, trade policy theories and practices are now in a state of flux: The global financial crisis and its aftermath is an excellent laboratory to test the robustness of key research findings in extreme circumstances. Abrupt shifts in government preferences may result in big shifts in public policy (such as attitudes towards more interventionist industrial policy) opening the door for fresh testing of theories of the impact of business-government relations on policy choice. I contend that the standing research on post-GFC protectionism has not caught up with the array of responses (especially in the global south) in the form of less observable NTMs. This rise of less visible protectionist measures has important policy implications. First, by showing that the developing world has indeed responded to protectionism by relying on NTMs and less transparent tools, I can capture a pattern of backsliding on trade policy reform in many countries. As Baldwin (1970) argued long ago, a decrease in tariff measures 7 over the last decades is most likely offset by NTMs. My point is that this practice has accelerated at an alarming pace since the GFC. Second, I will show that although the proliferation of PTAs may have changed the way that countries trade among themselves (Baldwin, 2012; Gawande et al., 2015), these institutions have not prevented countries’ increased reliance on NTMs. I also study whether the same mechanisms may occur where potential GVCs linkages exist. In other words, greater trade interconnections between countries through PTAs has not limited the escalation of protectionism (Baldwin, 2012; Boffa & Olarreaga, 2012; Gawande et al., 2015; Rodrik, 2009; Jensen et al., 2015; Lamy, 2013). Instead, the current context of international trade requires a broader conception of trade restrictions. As Evenett (2019, p. 6) states, “an up-to- date definition of protectionism would refer to all government acts that actually discriminate in favor of local commercial interests over one or more foreign rivals whatever the form of international commerce or the form of policy instrument used.” Following Evenett’s (2019) approach, this dissertation is based on a broader conceptualization of trade protectionism and requires a new method of measurement, which will be explained below. As the literature and recent events have shown, in a context of crisis and uncertainty, governments face social and business pressures to cushion their domestic economies from harsh external blows (see, e.g., Baldwin, 1989). Under such circumstances, “the likelihood of ‘copycat’ behavior arises, especially with respect to discriminatory measures that go against the spirit, if maybe not the law, of international trade norms” (Evenett, 2019, p. 13). My theoretical intuition is that PTAs––and to some extent GVCs, too––create trade dependencies, which under conditions of high economic uncertainty can increase defensive trade mechanisms between partner countries. 8 GVCs––which require further analysis in the future once data become more available and accessible––have in particular eroded the influence of large firms over national-level trade policy making. By definition, when there is a shift in employment abroad, domestic workers receive less support, which affects the “political influence” of large firms to combat trade protectionism (Gereffi, 2014). Hence, the effect of GVCs can swing both ways (Ghodsi & Stehrer, 2016); they can have an important effect on trade liberalization but also give rise to protectionism (Gereffi, 2018). In an interview I conducted with Victor Stolzenburg, research economist with the Economic Research and Statistics Division (ERSD) of the WTO, 8 he stated: The network of trade agreements and the rise of global value chains have undoubtedly impacted the way in which countries design trade policy since they have changed the underlying parameters of trade policy in numerous ways. Most of the effects have been in favor of a more liberal trading system but not all. Trade agreements (the most important one in this context being the WTO) have limited countries’ policy space. In order to obtain greater market access abroad, countries have committed to keep their tariffs below certain bound levels, move away from quantitative restrictions, improve transparency on non-tariff barriers. … This, by default, implies that countries’ responses to events like the financial crisis, that traditionally would have led to higher policy-related trade costs, are more muted since they operate in a more confined environment. … The more countries have constrained their tariffs, the more likely it is that they switch to alternative measures that can reach the same objectives. As true as this may be, the types of protectionism occurring now in a more interconnected world are murkier and less transparent. New, evasive NTMs can make it difficult for policy makers to quickly identify when and how a trading partner is negatively affecting their country (Grundke & Moser, 2019; Baldwin & Evenett, 2009). In sum, this dissertation takes the rise of protectionism following the GFC as a fait accompli. As such, my focus is not to prove the existence of this phenomenon but to identify 8 Author’s interview with Victor Stolzenburg, July 2020. See appendix 5 for full interview. 9 the factors that drive and perpetuate it. In 2017 (before the start of the U.S.-China trade war), the WTO secretariat lamented that “G20 economies may have opted in favor of implementing less traditional and transparent measures to curtail trade, the secretariat may have had more difficulties in gaining access to the relevant information and/or G20 economies implemented fewer such measures during this particular review period” (WTOa, 2017, p. 6). The following chart depicts my argument. Table 1.1 – Relationship between the Literature and the Main Argument of this Dissertation Source: Author’s creation. My methodological approach is geared toward theory expansion. It is both deductive and inductive. It is deductive in the sense that I revisit previous theories in the trade policy literature, first identifying a pattern and then testing whether data fit within a new international context. It is inductive in the sense that I analyze data in order to identify a pattern that can contribute to theory building. The following table summarizes the connection between research questions and research design. 10 Table 1.2 – General and Specific Research Questions of this Dissertation General research question Specific research questions Research Design How do countries respond to trade protectionism in a context of high economic uncertainty and trade interconnectivity? STATE LEVEL DYNAMICS Large-N quantitative analysis using primary data - How does economic interconnectivity (PTAs and GVCs) affect countries’ propensity to engage in protectionist retaliation? - How does economic interconnectivity (PTAs and GVCs) affect the type of measures countries deploy to respond to protectionism? INDUSTRIAL (SECTORAL) DYNAMICS - How do industrial/sectoral factors affect the way countries respond to protectionism? REGIONAL DYNAMICS Qualitative analysis. Case studies: Mexico, Brazil and Argentina; secondary sources and process-tracing interviews. - How have Latin American emerging economies been affected by and responded to protectionism in the decade since the GFC? Source: Author’s creation. Why is this topic relevant from a global development perspective? I will show in the following chapters that sectors, workers, and consumers are all hurt by protectionist measures, including opaque NTMs. For example, over the past 10 years, G20 members 9 have implemented around 1,200 protectionist measures that have affected the trade flows of the 47 nations constituting the Less Developed Countries (LDC) 10 group (Evenett & Fritz, 2018). 9 G20 members are Argentina, Australia, Brazil, Canada, China, European Union, France, Germany, India, Indonesia, Italy, Japan, the Republic of Korea, Mexico, the Russian Federation, the Kingdom of Saudi Arabia, South Africa, Turkey, the United Kingdom, and the United States. 10 For the list of the 47 LDCs, see UNCTAD, <https://unctad.org/topic/least-developed-countries/list>. 11 The United Nations (2020) estimates that three-quarters of the 880 million individuals in the LDC group are living in poverty. 11 As Evenett and Fritz (2018, p. 49) highlight: Without suggesting that they have specifically targeted the LDCs, three G20 members have put in place over 100 policy interventions since November 2008 that harm exporters, their employees, and migrant workers from the LDCs. The three G20 countries and the number of times they have hit LDC commercial interests are, respectively, China (101 times), India (235 times), and the United States (344 times). When we disaggregate the protectionist instruments that G20 countries have used most often against the LDCs, as I have done in Figure 1.2, we see that that a majority of them are NTMs (e.g., non-automatic licensing, quotas, price-control measures, subsidies). 12 The figure shows the differences between tariff and tariff measures in terms of their impact on exports from LDCs to G20 countries. While in 2018 the G20’s tariff measures only affected 9% of LDC exports to G20 countries, its NTMs affected 53% of LDC exports to G20 countries. This figure offers just one insight into how protectionism in general and NTMs in particular are affecting the developing world. 11 UN (2020) UN recognition of the least developed countries, <https://unctad.org/topic/least-developed- countries/recognition>. 12 In this section I present a detailed explanation for each type of measure. 12 Figure 1.2 – Between 2009 and 2018, the effect of G20 protectionism on LDCs has been concentrated in non-tariff measures. Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 1.3 shows the five sectors within the LDCs that have been most adversely affected by G20 protectionism (both tariff and non-tariff measures) between 2009-2018. 13 Metal products (Central Product Classification (“CPC”) 429) 14 were the target of 228 protectionist interventions; wearing apparel, except fur apparel (CPC 282) were affected by 186 interventions; products of iron or steel (CPC 412) by 166 interventions; vegetables (CPC 012) by 147 interventions; and electric motors, generators & transformers and parts (CPC 461) by 125 protectionist interventions. 13 For more details about how G20 economies measures have affected LDCs exports, see Global Trade Alert, <https://bit.ly/3959qCa>. 14 United Nations Statistics Division, Central Product Classification (CPC), <https://unstats.un.org/unsd/classifications/unsdclassifications/cpcv21.pdf>. 13 Figure 1.3 – Sectors Within LDCs Most Affected by G20 Protectionism (2008-19) Source: Global Trade Alert, 2019 When we expand the analysis globally, we can discern important variations in the level of impact from and responses to protectionism according to region. Table 1.3 summarizes key information about trade protectionism in sub-Saharan Africa, South Asia, North America, the Middle East and North Africa, Latin America and the Caribbean, the European Union, and East Asia Pacific. For each region, the table shows: 1) the number of protectionist and liberalizing measures each region implemented (2009-2019); 2) the percentage of exports in a region affected by protectionism emanating from the rest of world (RoW) and vice versa (percentage of RoW exports to the region affected by a given region’s protectionism); and 3) which sectors have been most affected by RoW protectionism and vice versa (which sectors have been more protected by each region). 14 Table 1.3 – Mapping Trade Policy Dynamics by Region TRADE POLICY INFORMATION Sub- Saharan Africa North America Middle East & North Africa Latin America & the Caribbean European Union East Asia Pacific Number of protectionist measures implemented (2009- 18) 613 2619 1015 2151 2701 4842 Number of liberalizing measures implemented (2009-18) 398 348 494 1366 423 2657 Share of protectionist interventions that use nontransparent (“murky”) instruments 40% 77% 51% 44% 87% 83% Region’s percentage of exports affected by RoW’s protectionism (average 2009-18) 29% 30% 27% 36% 22% 39% RoW’s exports to the region affected by the region’s protectionism (average 2009-18) 25% 40% 10% 19% 23% 21% Region’s most used protectionist measures affecting RoW’s exports Finance measures that affect trade Subsidies Tariff measures Tariff measures and non- automatic licensing Subsidies Subsidies and non-automatic licensing Region’s most affected sector by RoW’s protectionism - Motor vehicles, trailers (CPC 491) - Machinery for mining (CPC 444) - Products of iron or steel (CPC 412) - Fabricated metal products (CPC 429) - Products of iron or steel (CPC 412) - Basic organic chemicals (CPC 341) - Products of iron or steel (CPC 412) - Fabricated metal products (CPC 429) - Motor vehicles, trailers (CPC 491) - Products of iron or steel (CPC 412) - Products of iron or steel (CPC 412) - Basic organic chemicals (CPC 341) Source: Author’s creation based on Global Trade Alert, 2019 Notes: RoW is Rest of the World; sub-Saharan Africa consists of 47 countries, North America of 3 countries, the Middle East and North Africa of 20 countries, Latin America and the Caribbean of 39 countries, the European Union of 28 countries, and East Asia Pacific of 37 countries (including China). 1.2 Definition and Conceptualization of the Main Variables There are at least four key concepts related to the main variables of interest in this dissertation: 1) trade policy and the different types of measures used, 2) trade policy retaliation, 3) Preferential Trade Agreements (PTAs), and 4) Global Value Chains (GVCs). 15 1.2.1 Trade Policy and the Different Ways to Measure it 1.2.1.1 What does trade policy really mean? Trade policy is policy that regulates the flow of goods and services across borders and has a direct or indirect impact on the domestic prices of those imports and exports (Rodrik, 1995; Milner, 1999). 15 For example, the implementation of an import tariff on a specific product or group of products constitutes trade policy. The key conceptual element of trade policy measurement is that it concerns a specific measure implemented by a given country. This measure could affect one specific country or a group of countries. Protectionist trade policies are defensive measures implemented by one country that negatively affect market access and the commercial interests of other countries. This includes both tariffs and NTMs such as quotas, which limit market access to foreign commercial interests. Trade policy covers the flows of goods, services, and capital (portfolio and foreign direct investment (FDI)) (Evenett, 2019). Trade policy-making involves the dynamics of supply and demand, both of which are embedded in politics (Milner, 1999). The demand side of trade policy is shaped by domestic interest groups (e.g., unions) and individuals, who in turn will be affected by trade policy outcomes. These actors demand that political institutions implement a specific type of trade policy. The supply side is shaped by government—that is, the executive, legislature, and administrative courts, all of which guide the implementation of trade policy decisions (Rodrik, 1995). These two dimensions will be relevant to other discussions in this dissertation. 15 In this dissertation, I use the terms “trade policy,” “commercial policy,” and “international trade policy” interchangeably. 16 1.2.1.2 Tariffs, non-tariff measures and “murky” protectionism As previously described, trade policy outcomes are the results of various countries’ specific trade measures, which can be divided into tariff and non-tariff instruments. Import tariff measures are taxes imposed, typically in percentage terms, on the value of a good. For example, a 10% tariff would require importers to pay 10% of the appraised value of a good to the government of the importing country before selling the product in that government’s domain. 16 While tariff measures are easy to identify and define, non-tariff measures (NTMs) are just the opposite. NTMs require more detailed definition and differentiation. The United Nations Conference on Trade and Development (UNCTAD) Multi-Agency Support Team (MAST) 17 says, “NTMs are policy measures other than ordinary customs tariffs that can potentially have an economic effect on international trade in goods, changing quantities traded, or prices or both. A detailed classification is therefore critical in order to clearly identify and distinguish among the various forms of NTMs.” 18 This definition, which relates to the taxonomy of NTMs, will be used throughout the dissertation. I will further complement this definition of NTMs with details found in the Global Trade Alert (GTA) dataset, as measures such as import tariff increases, factors affecting the flow of workers across borders, and factors affecting capital flows are not included in the MAST classification. The International Monetary Fund (IMF) has noted that the GTA database “has the most comprehensive coverage of all types of trade-discriminatory 16 OAS SICE, Foreign Trade System Dictionary of Trade Terms, <http://www.sice.oas.org/Dictionary/TNTM_e.asp#TNTM>. 17 More information on the MAST Group’s classification of non-tariff measures can be found at: <https://unctad.org/en/Pages/DITC/Trade-Analysis/Non-Tariff-Measures/MAST-Group-on- NTMs.aspx>. 18 UNCTAD, Classification of NTMs, <https://unctad.org/en/Pages/DITC/Trade-Analysis/Non- Tariff-Measures/NTMs-Classification.aspx>. 17 and trade-liberalizing measures, although it begins only in 2008’’ (IMF, 2016, p. 76). Unlike tariff measures, NTMs are categorized into many different types. All of these in some way affect trade flows from one country to another. The following table presents general examples of NTMs under the MAST classification. Table 1.4 – Examples of NTMs Contemplated Under the UNCTAD’s Classification and GTA Name Description and examples Sanitary and phytosanitary measures Measures that are applied to protect human or animal life from risks arising from additives, contaminants, toxins, or disease-causing organisms in food, such as a requirement limiting the use of hormones and antibiotics in the production of meat or a sample test on imported oranges to check for the residue level of pesticides. Technical barriers to trade Measures referring to technical regulations and procedures for assessment of conformity with technical regulations and standards, such as restrictions on toxins in children’s toys or a requirement that refrigerators carry a label indicating their size, weight, and electricity consumption level. Contingent trade-protective measures Measures implemented to counteract particular adverse effects of imports in the market of the importing country contingent upon the fulfilment of certain procedural and substantive requirements, such as when Country A imposes an anti-dumping duty on imports of biodiesel products from country B to offset an injurious dumping by country B found to exist via an investigation, or when Country A imposes a countervailing duty on imports of semiconductors from country B to offset the subsidies granted by country B on the production of semiconductors found to exist via an investigation. Non-automatic licensing, quotas, prohibitions, and quantity control Control measures generally aimed at restraining the quantity of goods that can be imported, such as when only hotels and restaurants are allowed to import alcoholic drinks or a quota of 100 tons of tuna fish that can be imported during any one year. Pre-shipment inspection and other formalities Examples include live animals that need to be cleared at a designated customs office for inspection, or a requirement that goods must be shipped directly from the country of origin, without stopping at a third country. Price-control measures Measures implemented to control or affect the prices of imported goods, such as a minimum import price for fabric and apparel, or imports of fresh blueberries may enter free of duty between January 1 and May 31, while in other months seasonal duties apply. Export-related measures Export-related measures are measures applied by the government of the exporting country on exported goods, such as when exports of processed sea food products must be inspected for sanitary conditions or exports of cultural heritage objects— sculptures or other precious works—are prohibited. Source: UNCTAD, Classification of NTMs, <https://unctad.org/en/Pages/DITC/Trade-Analysis/Non-Tariff- Measures/NTMs-Classification.aspx>. 18 Overall, it is clear that non-tariff measures are less transparent than tariffs, which are more objective and explicit when countries implement them. As the Transparency in Trade (TNT) initiative states: As tariffs have been gradually reduced, increasingly prevailing trade restrictions take the form of so-called non-tariff measures (NTMs). These are inherently less visible and transparent than simple tariffs. … The situation is particularly murky for firms when it comes to trade in services, where it is often necessary for companies to hire specialists to help decipher which policies determine whether they will be able to provide their services to buyers in a foreign market. 19 To define these new measures that go beyond the traditional import tariff, Baldwin and Evenett (2009, p. 4) define “murky protectionism” as policies that “are not direct violations of WTO obligations; they are abuses of legitimate discretion which are used to discriminate against foreign goods, companies, workers and investors. Examples include abuses of health and safety regulations, and clauses in stimulus packages that confine spending to domestic producers.” Considering these definitions of tariff and NTMs, the Global Trade Alert database, a freely available dataset that I use for the quantitative analysis in this dissertation, includes numerous measures that “discriminate against foreign commercial interests” 20 in the list of trade interventions and categorizes them as transparent or non-transparent. 21 Table 1.5 shows the list of all intervention types considered by the Global Trade Alert database and their level of transparency. As mentioned, this list is based on United Nations’ MAST classification of 19 Transparency in Trade Initiative (TNT), Improving transparency in trade, <http://www.tntdata.org/about_tnt.html>. 20 Global Trade Alert, <https://www.globaltradealert.org/>. 21 In my quantitative analysis this categorization is used as a dummy variable (1 when it is nontransparent, and 0 when it is transparent). 19 non-tariff measures and supplemented by additional categories not found in MAST, such as import tariff increases and measures affecting the flow of foreign direct investment. 22 Table 1.5 – Interventions and their Level of Transparency Transparent Non-Transparent Import tariff Consumption subsidy Sanitary and phytosanitary measure Capital injection and equity stakes Technical barrier to trade Financial assistance in foreign market Anti-circumvention Financial grant Anti-dumping Import incentive Anti-subsidy In-kind grant Import monitoring Interest payment subsidy Safeguard Loan guarantee Special safeguard Price stabilization Import quota Production subsidy Import tariff quota State aid Instrument unclear State loan Export quota Tax or social insurance relief Export tariff quota Import ban and licensing requirement Export tax Internal taxation of imports Local labor, operations and sourcing Localization incentive Trade balancing measure Intellectual property protection Control on personal and commercial transactions, and investments Controls on credit operations Repatriation and surrender requirements Entry and ownership rule Financial incentive Treatment and operations Competitive devaluation Trade payment measure Import-related non-tariff measure Public procurement access Public procurement localization and preference margin Public procurement Labor market access Post-migration treatment Export ban, licensing requirement, subsidy Export-related non-tariff measure Foreign customer limit Tax-based export incentive Trade finance Source: Global Trade Alert, 2019 22 For a more detailed analysis of each instrument, see Evenett & Fritz (2017), pp. 37-41. 20 1.2.1.3 The Complex Task of Defining Policy Responses It is not an easy task to define a “trade policy response.” 23 The reason is twofold. First, many of the policies a country implements constitute unilateral protectionist dynamics, rather than retaliatory measures. Second, countries do not necessarily convey that a policy is a direct response to another country’s policy. In some cases, the term “policy response” is used in a more general way throughout the literature (e.g., Aizenman & Marion, 1996; Gawande et al., 2011, 2015; Bussière et al., 2011; Aizenman & Jinjarak, 2010; Wise & Quiliconi, 2007). These types of studies analyze what happens before and after a given political and/or economic event (e.g., the China shock or an international crisis) to see how countries, companies, or people responded to it. For instance, Bussière et al. (2011) studied how countries responded overall to the GFC and found that until 2011, there was no significant increase in protectionist measures. Aizenman and Jinjarak (2010) study the role of fiscal policy in response to the financial crisis. Other trade policy studies use the term “response” in a more specific way in the context of examining trade policy-making at the bilateral level. For instance, Boffa and Olarreaga (2012) study instances of bilateral retaliation after the GFC and find that, until 2012, there was little evidence of retaliation. Their definition of a protectionist response is as follows (p. 746): Arguably, a share of the overall protectionist response may be associated with retaliation, and anecdotal evidence suggests that this was the case during the recent global crisis. For example, China’s response to the 35% tariff on imports of Chinese tires imposed by the U.S., was to impose countervailing duties of 31% on imports of chicken from the United States, later increased to 105%. Another example is the 23 In this dissertation, I use the terms “response” and “retaliation” interchangeably. 21 response by some Canadian towns to the “Buy America” provisions in the U.S. fiscal package. In May 2009 some Canadian towns adopted legislation that barred U.S. companies from municipal contracts. … To move beyond anecdotal evidence, we … examine whether countries that have been affected by a protectionist measure adopted by a trading partner, systematically retaliate by adopting measures that hurt exports of that trading partner. In taking the conceptual framework of retaliation suggested by Boffa and Olarreaga (2012), this paper defines a potential retaliation policy as any protectionist measure that country A implements against country B at time t, after country B implemented an original protectionist measure against country A at time t-1. We can see this depicted in Figure 1.3. It is important to highlight the idea of a systematic pattern of protectionism. If we consider the notion of policy response within a window of just two years, we may only observe a bilateral measure implemented by coincidence between two countries. However, if this pattern occurs in a systematic and consistent way over an extended period of time, and we include controls for other bilateral factors and countries’ fixed effects, we can then argue that this mechanism is more than just anecdotal evidence. That is why my window of time for the quantitative analysis extends from 2009 to 2018. Figure 1.4 – A Retaliatory Dynamic Source: Author’s creation. For the quantitative analysis that undergirds this dissertation, I used a linear model with lagged variables of analysis to avoid making erroneous assumptions. The independent 22 variable is lagged so that the influence a trading partner has on a focal country (and vice versa) can be more carefully studied. The lagged influence model accounts for the time that the country of interest may take before it decides to retaliate. I will discuss this model in more detail in Chapter 4. 1.2.1.4 Preferential Trade Agreements (PTAs) Preferential Trade Agreements (PTAs) are probably one of the most important transformative changes of international trade policy since the early 1990s (Baccini, 2019). They have expanded in number and diversity. In the 1990s, there were 50 PTAs; as of January 2020, there were 303. 24 The World Bank provides a more general description of international trade agreements as “all the agreements that have the potential to liberalize trade” (Dür et al. 2014, p. 1). 25 I will use this definition in this dissertation as it is the one most used in the literature (Hofmann et al., 2019; Ruta, 2017). Hence, when I refer to a PTA, I am including free trade agreements, customs unions, and partial free trade agreements, which can be bilateral (two countries), plurilateral (many countries), plurilateral with a third country (e.g., the EU-Australia Trade Agreement), or region-region (e.g., the EU-Mercosur Trade Agreement). These distinctions are important because, as I mentioned above, the WTO categorizes PTAs as non-reciprocal preferential schemes. Examples include the Generalized System of Preferences (GSP), under which developed countries grant other developed countries preferential tariffs on their imports. RTAs, on the other hand, are reciprocal preferential trade 24 WTO, Regional Trade Agreements, <https://www.wto.org/english/tratop_e/region_e/region_e.htm>. 25 Other studies also explain why the term PTA is preferred to the term RTA (Hofmann et al., 2019; Ruta, 2017). 23 agreements between two or more partners. Under the WTO’s framework, these two terms are distinct. Thus, the WTO keeps two separate databases for PTAs and RTAs. 26 RTAs are allowed under WTO rules as exemptions and are subject to the rules stated in GATT Article XXIV, Article V, and the Enabling Clause. 27 The main function of the WTO––which succeeded the GATT in 1995––is to ensure that international trade is conducted as smoothly, predictably, and freely as possible. 28 But how can this be achieved in practice? In seeking to fulfill this key objective, the WTO created the idea of the “most favored nation” (MFN) treatment, a central norm within the global trading system, which holds that any benefit one member country grants to another must also be granted to each other member of the organization. 29 Specifically, MFN requires that “any advantage, favor, privilege or immunity granted by a contracting party to a product originating in or destined for another country, shall be granted immediately and unconditionally to any similar product originating in the territories of all other contracting parties or destined to them.” 30 It is conceivable that no country would prefer to dismantle trade barriers as it would then be obligated to grant the same benefits to all WTO members. That is why Article XXIV of the General Agreement on Tariffs and Trade (GATT) establishes exceptions to MFN treatment. It stipulates that when a free-trade zone or a customs union is created, signatory countries are not required to extend the same benefits to all WTO member countries. In other 26 WTO, Database on Preferential Trade Arrangements, <http://ptadb.wto.org/?lang=1>, and Regional Trade Agreements’ Dataset, <http://rtais.wto.org/UI/PublicMaintainRTAHome.aspx>. 27 WTO, Article XXIV: Territorial Application, Frontier Traffic, Customs Unions, and Free-trade Areas, <https://www.wto.org/english/tratop_e/region_e/region_art24_e.htm>. 28 WTO, What is the WTO?, <https://www.wto.org/english/thewto_e/whatis_e/whatis_e.htm>. 29 For an extended analysis of the main functions of the WTO, see Albertoni (2019). 30 GATT (1947), Article I, <https://www.wto.org/english/docs_e/legal_e/gatt47_01_e.htm>. 24 words, a given country is not in violation of the MFN rule when it enters into a Free Trade Agreement (FTA) or Customs Union (CU). As for FTAs, the WTO states “the trade within the group is duty free but members set their own tariffs on imports from non-members (e.g. NAFTA).” 31 Article XXIV of the GATT is more specific and says, “A free-trade area shall be understood to mean a group of two or more customs territories in which the duties and other restrictive regulations of commerce are eliminated on substantially all the trade between the constituent territories in products originating in such territories.” 32 The phrase “substantially all trade” is not clearly defined, but is usually understood to mean 80% of the volume of trade between the countries entering into the agreement. 33 When two or more countries decide to deepen their trade integration, they can go from a Free Trade Agreement (FTA) to a Customs Union (CU), which is another type of Preferential Trade Agreement. The WTO defines a CU as an integration scheme in which the members apply a common external tariff (CET) (e.g., the African Union). Hence, in a CU, members not only remove regulations on “substantially all” bilateral trade but also create a common external tariff. The CET is meant to prevent tariff-jumping, or the ability of outsiders to enter the integration scheme through lower tariff member countries. In the absence of a CET, most PTAs (and RTAs) set rules of origin or content requirements. 31 See, WTO’s Glossary, Free Trade Area, <https://www.wto.org/english/thewto_e/glossary_e/glossary_e.html>. 32 GATT (1947), Article XXIV, <https://www.wto.org/english/tratop_e/region_e/regatt_e.htm>. 33 See Departamento de Relaciones Exteriores y Comercio Exterior Austalia (2004) Negociando Acuerdos de Libre Comercio: Una Guía, <http://www.oas.org/dsd/Tool- kit/Documentosspa/ModuloV/Goode%20Chapter%203%20Reading.pdf>. 25 Finally, when countries decide to further advance the depth of integration within a PTA, they create a common market (CM). A CM establishes not only a common external tariff, but also the free movement of factors of production among the signatory countries. For example, any citizen within the CM can work freely in any of the countries that are part of the agreement, without restrictions (e.g., the European Union). All of the agreement types mentioned above are normally referred to as preferential (or regional) trade agreements (PTAs). Of the 303 PTAs currently in force, approximately 90% of these agreements are free trade areas, while the remaining 10% represent deeper integration projects, such as a CU. 34 The number of PTAs has been growing consistently since the 1990s, a trend that was defined in the literature as “New Wave of Regionalism” (Mansfield & Milner, 1999, p. 589). 35 As I will explain in greater detail later in this dissertation, to measure PTAs between countries, I will rely on data from the Design of Trade Agreements (DESTA) project, which “systematically collects data on various types of preferential trade agreements (PTAs). These may be customs unions or free trade agreements.” 36 DESTA is today “the most comprehensive [dataset] in terms of both items coded and number of agreements included” (Dür et al., 2014, p. 353). 1.2.1.5 Global Value Chains (GVCs) GVCs are also among the most important transformative changes of international trade policy in the last decade. However, they are still very difficult to measure, and thus it is difficult to understand their actual impact on the world economy. As Ruta (2017, p. 175) 34 WTO, Regional trade agreements, <https://www.wto.org/english/tratop_e/region_e/region_e.htm>. 35 To see how this “Wave of Regionalism” evolved in the Middle East and North Africa, see Nugent and Hakimian (2003). 36 Design of Trade Agreements Database, <https://www.designoftradeagreements.org/project- description/>. 26 states, the “GVC has changed international trade, with trade in parts and components increasing almost six times between 1990 and 2015, faster than the 4.5 times for other forms of trade.” A recent report from the World Bank (2020) entitled The Age of Global Value Chains defines GVCs as follows: “a global value chain breaks up the production process across countries. Firms specialize in a specific task and do not produce the whole product” (p. 5). The following figure shows how GVCs works. Figure 1.5 – How do Global Value Chains Work? Source: World Bank (2020) World Development Report 2020: Trading for Development in the Age of Global Value Chains. Overview booklet. World Bank, Washington DC, p. 5. Given the complex structure of GVCs, there is considerable debate about how to properly measure them (Gaulier et al., 2019; Hummels et al., 2001). The share of intermediate goods traded between two countries is normally used as a proxy measure for the GVC between them. Gaulier et al. (2019) shows that in the last decade, “the share of intermediate goods in world trade in nominal terms is fairly well correlated to various Global Value Chain (GVC) indicators based on international input-output matrices” (p. 1). Hence, I use the exchange of intermediate goods as the best approximation to measure the potential “GVC interconnection” between two countries (Escaith et al., 2010; Altomonte et al., 2012; De Backer & Miroudot, 2014). 27 To specifically measure the share of trade of intermediate goods, I rely on the classification groups created by Broad Economic Categories (BEC), which provides “a means for international trade statistics to be analyzed by broad economic categories such as food, industrial supplies, capital equipment, consumer durables and consumer non- durables.” 37 Table 1.6 shows the three basic classes of goods categorized in the BEC. Intermediate good are those in the middle of the table. 38 When speaking on the intermediate goods’ linkages with trade protectionism, Bems et al. (2009, para. 10-13) state that: Imported intermediate goods linkages have several distinct implications for the response of trade to changes in final demand. These effects operate independently of, and in addition to, standard trade transmission channels that work through the endogenous response of final demand to shocks. First, imported intermediate goods linkages imply that a country’s exports and imports tend to move in the same direction in response to changes in either domestic or foreign demand. For example, a decline in U.S. demand for cars will typically imply decreased demand for cars imported from Canada. Since Canadian cars are produced using imported inputs from the U.S., a decline in the production of Canadian cars will mean fewer U.S. exports of car parts to Canada, both U.S. imports and exports fall. This does not happen if the imported intermediates channel is absent. Second, imported intermediate goods linkages influence each country’s exposure to foreign shocks. 39 37 Classification by Broad Economic Categories (Rev.4), https://unstats.un.org/unsd/tradekb/knowledgebase/50089/classification-by-broad-economic- categories-rev4. 38 A concept normally related to intermediate good is “intra-industry trade” (IIT), which is a more general way of defining trade within similar industries. The OECD (2005) defines IIT as “the trade in similar products (“horizontal trade”) with differentiated varieties (e.g. cars of a similar class and price range), and trade in “vertically differentiated” products distinguished by quality and price (e.g. exports of high-quality clothing and imports of lower-quality clothing).” There could be Intra- Industry Trade in both intermediate and final goods. While IIT can applied to different stages of a good, intermediate goods are normally considered to measure GVCs. See Lüthje (2001) and Peterson and Thies (2015) for studies on intra-industry trade (IIT) and intermediate goods. 39 Bems, Rudolfs, Robert Johnson, Kei-Mu Yi (2009, 27 November) The collapse of global trade: Update on the role of vertical linkages, VoxEU CERP Policy Portal. https://voxeu.org/article/collapse-global-trade-update-role-vertical-linkages. 28 Table 1.6 – Capital, Intermediate and Consumption Goods (in BEC Classification) Capital goods Intermediate goods Consumption goods 41 – Capital goods (except transport equipment) 521 – Transport equipment, industrial 111 – Food and beverages, primary, mainly for industry 121 – Food and beverages, processed, mainly for industry 21 – Industrial supplies not elsewhere specified, primary 22 – Industrial supplies not elsewhere specified, processed 31 – Fuels and lubricants, primary 322 – Fuels and lubricants, processed (other than motor spirit) 42 – Parts and accessories of capital goods (except transport equipment) 53 – Parts and accessories of transport equipment 112 – Food and beverages, primary, mainly for household consumption 122 – Food and beverages, processed, mainly for household consumption 321 – Fuels and lubricants, processed (motor spirit) 522 – Transport equipment, non- industrial 61 – Consumer goods not elsewhere specified, durable 62 – Consumer goods not elsewhere specified, semi-durable 63 – Consumer goods not elsewhere specified, non-durable Source: Author’s creation based on UN Trade Statistics, Intermediate Goods in Trade Statistics, <https://unstats.un.org/unsd/tradekb/Knowledgebase/50090/Intermediate-Goods-in-Trade-Statistics>. In light of my literature review, GVCs are another example of countries’ interconnectedness, through their trade relations (something that PTAs mostly show) and also through their trade production. This is why GVCs are another relevant variable for understanding trade interconnectivity in the current international political economy. 1.3 Theoretical Approaches in the Literature and Initial Hypotheses How systemic level dynamics shape international trade policy-making processes remains an open question in the literature on international trade. As Lake (2009, p. 237) stresses, if international institutions “really matter, they will alter the interest and possibly institutions within states as well.” With this quote in mind, I aim to provide preliminary evidence that reciprocity can also have a negative side. That is, when trading partners apply protectionist measures, locally affected industries can lobby for domestic tariff increases as a defensive mechanism. This could press a partner into removing tariff or non-tariff barriers that they implemented previously, or it could spiral into higher trade barriers. I analyze two different 29 categories to understand the patterns of retaliatory dynamics: state level and industrial (sectoral) factors. 1.3.1 State Level Dynamics in Trade Protectionism Broadly, the characteristics of a given state can affect the likelihood that policy makers will resort to retaliatory protectionism (Gould & Woodbridge, 1998; Miyagiwa et al., 2016). In particular, the degree of bilateral trade integration (Kono, 2007), global value chains (Gereffi, 2018; Gawande et al., 2015; Tanaka, 2009; Yi, 2009), and the relative economic size of a country (Miyagiwa et al., 2016; Evenett, 2019) can shape its trade policy dynamics. Concerning relative economic size, Evenett (2019, pp. 24–25) argues that: Larger economies tend to resort less to traditional trade restrictions. Perhaps governments in larger economies feel they are under more scrutiny and so resort to less transparent forms of discrimination. However, economic size might matter in different ways. For instance, nations with larger domestic markets may be more inclined to implement industrial policies that seek to nurture domestic firms, doing so not with tariffs but with state largesse that a larger tax base may support. Rather than attribute agency to government, MNEs [multinational enterprises] may lobby for different types of assistance in economies of different sizes; again, the size of a nation’s tax base may influence how deep its government’s pockets are or the scale of lending by state- controlled development banks. Governments invariably pursue policies that maximize their support from influential political groups. Studies have shown that uncertainty about an economic outcome will lead one country in a group of two or more trading nations to impose a gradually rising tariff on the others (Baldwin & Evenett, 2009). Eventually, this added cost on imported goods induces retaliation from the other nations. This retaliation can encourage either liberalization or a protectionist war, with market characteristics acting as the deciding factor. At the same time, the very threat of retaliation may play a key role in triggering trade liberalization (Gould & Woodbridge, 1998). 30 The extent of protectionism triggered by the GFC remains open to debate. As mentioned at the beginning of this chapter, earlier studies conducted in the wake of the GFC found no evidence of trade retaliation between 2008 and 2010; rather, this research found just the opposite. This was the case for Boffa and Olarreaga (2012), who analyzed protectionist measures from November 2009 to December 2010. Similarly, in their study on the use of antidumping duties and tariffs after the GFC, Kee et al. (2013) showed that bilateral responses along these lines were minimal. In this dissertation project, I extend this analysis through 2018 and include the effects of GVCs and PTAs on trade dynamics. I found that with a longer timeline and additional variables, a more protectionist scenario takes shape. Most studies on the role of GVCs and PTAs suggest that trade interdependence fosters more trade liberalization. This argument goes back to the international political economy (IPE) and peace research agenda, both of which state that increased economic and political interdependence reduces the likelihood of conflict between countries (Copeland, 1996). However, realist authors, mostly after the Cold War, challenged this argument and questioned the relationship between economic interdependence and international peace (Barbieri, 1996). In fact, Waltz (2000, p. 14) highlights that interdependence “multiplies the occasions for conflicts”: Close interdependence is a condition in which one party can scarcely move without jostling others, a small push ripples through society. The closer the social bonds, the more extreme the effect becomes, and one cannot sensibly pursue an interest without taking others’ interests into account. One country is then inclined to treat another country’s acts as events within its own polity and to attempt to control them. As mentioned before, the literature on the political economy of PTAs has been concentrated mostly on why countries negotiate and sign PTAs (Baccini & Dür, 2012; R. 31 Baldwin, 1993; Chase, 2003; Manger, 2009) and more recently, on the effect PTAs have in signatories’ countries in terms of trade flows (Baier & Bergstrand, 2007) and other areas of political concern, such as their ability to reduce bilateral conflict between PTA members (Hafner-Burton, 2005; Mansfield et al., 2008; Mansfield & Reinhardt, 2008). Overall, academic studies have concentrated on the positives of interdependence (through PTAs or GVCs) and have not sufficiently considered how interdependence can create and encourage channels of protectionism. Only a few authors to date have probed in the context of interdependence protectionism (Blanchard et al., 2016; Bems et al., 2009; Yi, 2009, Manger, 2009), with Yi (2009)––one of the few––arguing that “massive reorientation of trade flows towards multiple-step supply chains” has played an important role in current protectionist dynamics (p. 2). In an interview I conducted with Dorotea López, Director of the Institute of International Studies of the University of Chile, she said, “The effects of free trade agreements, probably in the deformation of economists, are rather more a legal instrument than an instrument of real economic connection. Sometimes I think free trade agreements have been overvalued, in terms of what they generate.” 40 The main goal of this dissertation project is to increase understanding of this interdependence protectionism. In Tables 1.7 and 1.8, I present my hypotheses concerning state-level dynamics in trade protectionism. The first set of hypotheses concentrates on a general perspective of how countries respond to protectionism and the effect of PTAs and GVCs on such responses. 40 Author’s interview with Dorotea López, April 2020. See appendix 5 for the full interview. 32 First, I hypothesize that, in general terms, when a country is adversely affected by another country’s protectionist measure, it will systematically respond to that country with retaliatory protectionist measures. Hence, on Hypothesis 1, I suggest that an increase in import restrictions by country B on country A at time t-1 has a positive effect on import restrictions by A on B at time t. However, the existence of PTAs between two countries reduces the possibilities of protectionist escalation. In other words, a PTA––ceteris paribus–– lowers the probability and severity of protectionist bilateral dynamics. In line with the literature on trade interdependence, we would expect that the more interconnected countries are, the less protectionist they are likely to become (e.g., Baldwin, 2012; Gawande et al., 2015; Rodrik, 2009; Jensen et al., 2015; Lamy, 2013). Hence, in Hypothesis 2, I suggest that PTAs and GVCs are negatively correlated with protectionist responses. In Hypothesis 3, I also test to what extent GVCs decrease the likelihood of protectionist response. A second group of hypotheses concerning state-level dynamics in trade protectionism tests how responses vary when they are divided between tariff and non-tariff measures. In Hypotheses 4 and 5, I suggest that countries with high trade interdependence (PTAs and GVCs) will retaliate more with less transparent non-tariff measures than they will with tariff measures. Thus, PTAs and GVCs are negatively correlated with protectionist tariff responses (Hypothesis 4), while positively correlated with non-tariff responses (Hypothesis 5). 33 Table 1.7 – General Hypotheses (Hypotheses 1, 2, and 3) RESEARCH PROPOSITION A: In general terms, when a country is adversely affected by a protectionist measure adopted by another country, it will systematically respond to that country by adopting protectionist measures in return. However, the existence of PTAs and GVCs decreases the likelihood of protectionist response. HYPOTHESIS 1: Countries will retaliate against those nations that initially target them with a protectionist measure (Direct Retaliation). HYPOTHESIS 2: Preferential Trade Agreements (PTAs) are negatively correlated with protectionist responses (Conditional Effect of PTAs). HYPOTHESIS 3: Global Value Chains (GVCs) are negatively correlated with protectionist responses (Conditional Effect of GVCs). Table 1.8 – The Relationship Between Tariffs and NTMs with PTAs and GVCs (Hyp. 4 and 5) RESEARCH PROPOSITION B: Countries with high trade interdependence (PTAs and GVCs) will retaliate more with less transparent non-tariff measures than tariff measures. HYPOTHESIS 4: Preferential Trade Agreements (PTAs) are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses (Conditional Effect of PTAs). HYPOTHESIS 5: Global Value Chains (GVCs) are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses (Conditional Effect of GVCs). 1.3.2 Sectoral Dynamics Another focus of this dissertation is on the sectoral dynamics that underpin trade protectionism. As several studies show, trade policy dynamics are highly differentiated for industry-level data. First, key findings in the literature on industrial level protection show that a given sector is more likely to receive trade protection when, among other factors, it is labor intensive and closely integrated into a GVC or intra-industry trade (Ray, 1991; Ray & Marvel, 1984). 41 For example, Lee and Swagel (1997), who focus on the relationship between non-tariff barriers (NTBs) and industry level protection, show that NTBs are more concentrated in low-wage industries. However, one caveat is that industry-level studies are 41 I use the terms “within-industry” and “intra-industry” interchangeably. 34 still highly concentrated on industrialized economies (Milner & Kubota, 2005); thus, I focus on developing economies in this dissertation. Second, studies on retaliation under the WTO’s dispute settlement mechanism (DSM) largely explored patterns of response under that formal dispute body while less attention has been paid to retaliation when the DSM cannot reach a decision. In this event, the affected member country can formally request permission to retaliate against the other country in the dispute. 42 During 2017, WTO members filed 17 requests for consultation (the first step of the dispute process), compared with 16 in 2016 and 13 in 2015. 43 In general, it is not much less than the number of cases filed during the 1990s (an average of 37.8 cases per year between 1995 and 1999), and not far from the average number of cases filed between 2007 and 2016 (around 19 cases per year). Between 1995, when the WTO’s DSM was established, and 2017, the DSM dealt with 590 requests for consultation and issued over 355 dispute settlement decisions (Reich, 2017). Countries in the WTO may request to earn the right for cross-sector retaliation, which is the suspension of concessions in an industry different than the sector in which the trade dispute originated (Abbott, 2009). 44 Although this mechanism has been an option since 2017, it has only been requested 14 times (Chisik & Fan, 2020). While a large body of research addresses patterns of retaliation under the WTO’s DSM, few studies have considered how these dynamics have played out beyond formal 42 Examples are New Zealand and the U.S. against Canada in a dispute over milk and dairy products, and Brazil, Canada, Chile, the EU, India, Japan, Korea, and Mexico against the U.S. and its Byrd Amendment. For more information, see Bonomo (2014). 43 Canada, Qatar and the United States filed the most disputes in 2017 (three each), followed by Russia and Ukraine (two each). For more details, see WTO (2018) A Year in Review, <https://www.wto.org/english/res_e/booksp_e/06_anrep18_ayearinreview_e.pdf>. 44 A well-known case of cross-retaliation was the US-Mexico cross-border trucking dispute under NAFTA, for an in depth analysis of that case, see Alexander and Soukup (2010). 35 institutional boundaries, i.e., when the DSM adjudication panel deadlocked. It is in this realm where an important number of retaliatory dynamics can occur in “murkier” forms. Even within the WTO, there have been new types of requests that go beyond the traditional framework of tariff or NTB cross-retaliation. For example, in Canada’s dispute against Brazil’s export financing program for its Embraer Aircraft, Canada requested the suspension of such concessions under parallel WTO agreements, such as the Agreement on Textiles and Clothing and on Import Licensing. In this specific case, Bonomo (2014, p. 3) highlights that “prolific Canadian trade diplomats also created another invention in the same dispute against Brazil on regional aircrafts, they included non-WTO retaliation in their mix by informing the membership that ‘Canada may also suspend Brazil from the list of countries eligible for General Preferential Tariff treatment, a unilateral tariff preference granted by Canada for most imports from developing countries.’” Recent research on trade policy seeks to explain variations within industries by studying product differentiation, firm size, and the location of a given firm’s global operations. For instance, Jensen et al. (2015) find that the location points of a given firm’s global supply chain operations can influence the trade policy objectives of that firm. More recently, Kim (2017) showed that firm-level determinants can help to explain trade liberalization: high levels of product differentiation and cross-border production enable exporting firms to overcome collective action problems and lobby for tariff reductions to facilitate production across the entire value chain. Kim’s (2017) empirical study finds that vertically integrated global firms lobby more actively when they produce differentiated products, and that goods with low substitutability have lower applied tariffs on average. 36 A related subset of this literature concerns the trade preferences of societal groups and their influence on trade policymaking. More specifically, how do domestic actors overcome the so-called coordination problem and formulate policy demands? Why are some groups better than others in the provision of public-collective goods (Grossman & Helpman, 1994; Olson, 1965)? Considering within-industry dynamics, studies have shown that countries with strong intra-industry trade are less likely to engage in conflict than cross- industry trading partners (Gowa & Mansfield, 2004). Moreover, intra-industry trade promotes the emergence of similar foreign policy preferences among trading partners. However, this does not happen in reverse; states with similar foreign policy preferences do not necessarily engage in intra-industry trade (Peterson & Thies, 2015). As mentioned above, recent studies on the relationship between global supply chains and trade policy responses to the 2008 GFC show that participation in global value chains (within-sector trade) is “a powerful economic factor [in] determining trade policy responses” (Gawande et al., 2015, p. 102). There are lower levels of trade friction in contexts of within- sector trade because the demand for cheap inputs by domestic suppliers and exporters decrease the pressure from domestic lobbies for market protection. In Table 1.9, I depict my hypotheses concerning sectoral dynamics. For instance, due the existence of GVCs, we would expect that sectoral bilateral dynamics affect the way countries respond to protection. I test this proposition using three different factors. First, I test whether protectionist retaliation is more likely to happen cross-sector rather than within-sector (Hypothesis 6). Second, I test whether the level of industrialization of the sector impacts within-sector retaliation (Hypothesis 7). Finally, I test whether protectionist responses are less likely with trade in intermediate goods, due to a possible GVC effect (Hypothesis 8). 37 Table 1.9 – Sectoral Dynamics (Hypotheses 6, 7, and 8) RESEARCH PROPOSITIONS C: Sectoral bilateral dynamics affect the way countries respond to protection. HYPOTHESIS 6: Protectionist retaliation is more likely to happen cross-sector rather than within-sector. (Cross-Sector Retaliation). HYPOTHESIS 7: The higher the level of industrialization of the sector, the lower the within-sector retaliation. (Within-Sector Retaliation – industrialization) HYPOTHESIS 8: Protectionist responses are less likely with trade in intermediate goods. (Within-Sector Retaliation – intermediate goods) 1.3.3 Contribution of this Dissertation to the Trade Policy Literature There are at least three major challenges related to the current literature on protectionism and policy responses in the context of high trade interdependence. The first is related to the data used in current research on protectionist responses. To date, the literature has focused on the magnitude of retaliation with regard to protectionist responses, calculating this magnitude by counting the total number of measures trading partners adopted, not the share of domestic trade affected (Boffa & Olarreaga, 2012; Evenett, 2019). In this dissertation, I go a step further and also analyze how these policy interventions affect the actual share of trade in bilateral goods. In doing so, I can better specify how countries harm one another when implementing protectionist measures. The new measures presented in this study have important implications for global trade moving forward, as will become evident in the case study chapters of this dissertation. 45 45 For instance, if we consider the total number of measures implemented by the U.S. and China toward Argentina, Brazil, and Mexico, the U.S. seems to be (disproportionately) more protectionist than China. However, when we consider the actual effect of those measures, the story changes substantially. Indeed, when considering the average share of exports between 2009 and 2017 that protectionist measures of China and the U.S. affected Argentina, Brazil, and Mexico, China is more protectionist than the U.S. Apart from being an interesting descriptive finding, this also highlights the relevance of focusing on the effect of protectionist measures rather than just the aggregate number of measures implemented. 38 Finally, most of the current literature is still centered on the developed economies (Evenett & Fritz, 2018). Studies that analyze developing countries focus primarily on tariffs, at the exclusion of other types of discriminatory measures (Gawande et al., 2011). Most of the work to date has been Large-N cross-country analyses. Few scholars have focused on the developing economies, and few have incorporated sectoral analyses to assess how different sectors have been more impacted than others (Gawande et al., 2011; Evenett & Fritz, 2018). The broader research agenda on post-GFC trade policies is still limited due to the emergence of new types and forms of trade protectionism that have yet to be measured. This dissertation will also contribute to a broader debate on trade policy substitution between NTMs and tariffs, a topic that has been under-researched (Beverelli et al., 2019; Niu et al., 2018). We know the more tariffs are lowered via binding rulings created by WTO rules as more countries rely on NTMs, but this raises bigger questions about the relationship between trade policy and transparency. Speaking about this relationship, Kono (2006, p. 369) writes that “democracy has contradictory effects on different types of trade policies because electoral competition generates more information about some than about others. It generates considerable information about policies whose effects on consumer welfare are easy to explain to voters, but less information about policies whose effects are more complex.” In the same vein, Garrett (2000) argues that democracies are relatively more protectionist because they are more responsive to the demands of citizens than autocracies; more specifically, he indicates that “democracy makes leaders more accountable to their citizens, promoting trade liberalization to the extent that this is good for society as a whole” (p. 973). 39 The research agenda to which this dissertation seeks to contribute is still limited due to the limited ways in which countries can enact protectionist measures, but new types and forms of trade protectionism are still being discovered and will be measured. Evenett (2019) discusses the consequences of limited data in measuring 21st-century protectionism. This creates an “inadequate scrutiny bias,” which means that once an idea has been established it become unquestionable, such as the argument that the GFC has not substantially increased protectionism (p. 12). Data limitations also create a pattern of “omitted variable bias” in regression studies, which is “particularly important when analyzing the impact of commercial policy, as governments can substitute between transparent and murkier forms of protectionism” (p. 13). Still, as mentioned before, Evenett (2019, p. 13) highlights that the GFC and its aftermath can be “an excellent laboratory to test the robustness of key research findings in extreme circumstances.” Taking all of this into account, my dissertation will leverage the empirical findings in the literature since the GFC and glean new insights to gain a better understanding of the sheer volume and nature of 21st-century protectionism. Here, I will present a unified database that expands and combines data from multiple datasets to draw a clearer picture of the unfavorable impacts of protectionism. 46 1.4 A Two-Step Research Design 1.4.1 Large-N Analysis The first portion of this dissertation is based on a Large-N quantitative analysis, which will be constructed using directed dyad-year data at the state and industry level. Ultimately, two 46 As will be explained in following chapters, the database presented in this dissertation combines data from Global Trade Alerts (GTA), the BACI-CEPII database, the Classification Harmonized System (HS), and Broad Economic Categories (BEC). The development of this new dataset at the monthly and trimester level is part of a joint project with George Gerald Vega, from the USC Center for Applied Network Analysis (CANA). 40 datasets will be created by generating new data and merging it with existing data, as discussed below. To test the state- and industry-level hypotheses, I will use a lagged model, as it does not assume contemporaneous influence; rather, influence is lagged so that a focal country is influenced by its trading partners’ network after those trading partners have implemented protectionist policies that affect the focal country. Hence, what the lagged influence model considers is the time that the focal country may take (e.g., quarters within one year) before it decides to retaliate. 1.4.2 Small-N Analysis and Case Studies A central chapter of this dissertation focuses on a more in-depth analysis of three emerging economies in Latin America, specifically Mexico, Brazil, and Argentina. For each, I will compile a country narrative of trade policy performance amid current protectionist trends. With the large-N quantitative analysis completed, this case study section will offer a robust examination of these country cases based on the different levels of analysis offered above. Although my focus is on Latin American emerging economies, the three cases that I will study offer rich theoretical insights. First, all three states are G20 members, and the G20 represents 80% of gross world product (GWP) and three-quarters of global trade (UNCTAD, 2020). Second, considering all of G20’s protectionist measures from 2009 to 2018 (see Figure 1.6), Argentina, Brazil, and Mexico are well distributed in the share of exports that have been affected by protectionist measures. Argentina, for example, was the most protectionist G20 country considering the share of exports from the rest of the world to that country. As I will explain in chapter 5, although these three countries are resource dominated and only Mexico has a significant manufacturing sector, I use a regional focus in Latin 41 America for reasons that I will explain further, and I will concentrate on differences between (1) Mexico and (2) Argentina and Brazil. A detailed justification of each case is explained in chapter 5. Figure 1.6 – Share of World Exports to G20 Countries Affected by Protectionist Measures (Average 2009-2018, by Country) Source: Author’s creation based on GTA, 2019. Third, while Argentina, Brazil, and Mexico effectively weathered the GFC, assessments to date completely overlook the role of trade protectionism. For example, Wise et al. (2015, p.1) argue that emerging regions such as Latin America and East Asia showed remarkable resilience when faced with the GFC: One of the more surprising features of the 2008-09 global financial crisis was the comparative ease with which emerging economies in Asia and Latin America rebounded. That rebound was a radical departure from the effects of previous crises on these regions, be it the decade-long recession wreaked on Latin America by the 1982 debt shocks or the financial crisis that dramatically slowed Asian economies in the late 1990s. The quick recovery of emerging economies in 2010-12 was, moreover, instrumental in deterring a full-blown global depression. This section of my dissertation aims to bring trade protectionism into the framework of Wise et al. (2015). Closer inspection of the data on these three countries’ trade measures 42 reveals highly diverse policy responses. Argentina and Brazil, for instance, need to be better integrated into the global economy, while Mexico is perhaps too open for its own good. Recent studies have shown the important direct and indirect effects of current trade disputes on emerging economies and especially those in Latin America. 47 More to the point, when we analyze less observable trade measures, we find that Latin America has been implementing protectionist measures during the last decade. This small-N analysis will offer more detail as to whether countries with strong trade ties to the rest of the world via PTAs and GVCs have been differentially affected by protectionist dynamics than those with fewer trade ties. For instance, a recent study developed by the Chilean General Directorate of International Economic Relations (2018) shows that 63% of Chile’s exports go to countries in the core and periphery of the Sino- American trade conflict (India, Turkey, EU, Canada, Mexico, and Russia). However, given Chile’s strong web of free trade agreements, few of the many tariff increases have been enacted against Chile. 1.5 Historical Context and Significance of this Research As the historical chapter of this dissertation will further explain, this is a timely project in the field of International Political Economics (IPE). Although trade protectionism has grabbed the attention of the media and has been at the center of political debates (Beyers & 47 Moody’s Analytics (2018) estimates the impact of the “Trump Trade War” from 2018 to 2020 and offers different scenarios. Interestingly, under the scenario of 25% tariffs on all Sino-U.S. trade and other “qualitative” actions taken by China against American firms operating there, the results show that protectionism reduces GDP growth from second quarter 2018 to second quarter 2020 in Argentina, Brazil, and Mexico in very different ways: in Argentina, GDP growth is reduced by more than 0.6%; in Brazil, 0.3 to 0.6%; and in Mexico, less than 0.3%. See Moody’s Analytics (2018), Trump Trade War, <https://www.moodysanalytics.com/-/media/article/2018/trump-trade-war.pdf>. For a review of 17 recent studies of the form and consequences of contemporary trade conflict, see Evenett and Fritz (2018). 43 Kerremans, 2007), there have been few articles on trade policy that explore post-GFC protectionism. Figure 1.6 reflects data on Google searches for words related to “trade policy” and “protectionism” since 2008. The figure shows “search interest relative to the highest point on the chart for the given region and time. A value of 100 is the peak popularity for the term. A value of 50 means that the term is half as popular. A score of 0 means there was not enough data for this term.” 48 As the figure shows, the “interest” worldwide on news related to “trade policy” and “protectionism” has been significant since 2008. 49 However, since 2008 only 9 articles have been published on questions related to “trade protectionism” in the prestigious American Political Science Review, and an average of 2.3 articles per year on this topic have been published in the journal International Organization (a total of 26 between 2008 and 2019). 50 48 Google Trends Data, <https://trends.google.com/trends/?geo=US>. 49 Altman & Bastian (2019) shows that since March 2018, when President Trump announced new steel and aluminum tariffs against China, Google searches for “trade war” spiked to 100, which is the peak popularity in Google searches. 50 In main academic journals of economics, for instance, we see a similar scenario: between 2008 and 2019, there were 11 articles about “trade protectionism” in The American Economic Review, and 27 articles in The Economic Journal. 44 Figure 1.7 – Google Searches on “Trade Policy” and “Protectionism” (2008-19) Note: A value of 100 is the peak popularity for the term. A value of 50 means that the term is half as popular. A score of 0 means there was not enough data for this term. Source: Author’s creation based on Google trend data. Another reason this research is timely relates to the high degree of uncertainty that surrounds trade policy, especially since the U.S. declared a trade war against China in 2018. Before the outbreak of COVID-19 in 2020, this trade war was one of the most pressing challenges to the global economy. The World Trade Uncertainty (WTU) index is a measure of trade uncertainty introduced by Ahir et al. (2018) to address the impact of erratic changes in trade policy (see Figure 1.8). 51 It shows how trade uncertainty is rising sharply, after 20 years of relative stability. 51 The index is constructed by using “Economist Intelligence Unit (EIU) Country Reports” to measure uncertainty related to trade for each of 143 countries, on a quarterly basis, from 1996 onwards. Data presented is weight averaged. The font in blue indicates the tariff measure taken, and the font in black indicates the narrative of the World Trade Uncertainty index. The WTU index is computed by counting the frequency of uncertain words (or the variant) that are near the following words: protectionism, North American Free Trade Agreement (NAFTA), tariff, trade, United Nations Conference on Trade and Development (UNCTAD) and World Trade Organization (WTO). The 45 Figure 1.8 – World Trade Uncertainty index (1996 to 2019) Source: Bown and Kolb (2019), <https://www.weforum.org/agenda/2019/07/how-trade-uncertainty-is- impacting-the-global-economy>. This uncertainty does not only impact diplomatic relationships among governments. It also affects international actors at every level, including political institutions, corporations, small businesses, and consumers; the effects are omnipresent. An uncertain economic environment is not conducive to long-term business decisions. Trade policy impacts consumers especially since they are most likely to bear the ultimate burden of higher-cost processes. When political uncertainty reaches typical businessowners and consumers, it shapes public opinion about trade policy and its future. WUI is then normalized by the total number of words and rescaled by multiplying by 100,000. A higher number means higher trade uncertainty and vice versa. For more information about this index, see <https://voxeu.org/article/trade-uncertainty-rising-and-can-harm-global-economy>. 46 1.6 Outline of the Dissertation This dissertation proceeds in five more chapters. Chapter 2 offers a historical overview of 21st-century protectionism, with an emphasis on how we arrived at this new protectionist juncture. Chapter 3 is a descriptive analysis of the dataset used for the quantitative and qualitative analysis. Chapter 4 conducts a quantitative analysis that captures the “big picture” on current protectionist trends and tests the main research hypotheses of this dissertation. Chapter 5 concentrates on Latin America as a regional case and studies three emerging economies in greater detail (Argentina, Brazil, and Mexico). It examines a small number of country-industry cases in depth to qualitatively test some of the main hypotheses discussed in the quantitative chapter. Chapter 6 concludes with a summary of the main findings and the intellectual merit of this dissertation project. 47 CHAPTER 2 – A HISTORICAL OVERVIEW OF 21st- CENTURY PROTECTIONISM: HOW DID WE ARRIVE AT THIS POINT? “More than at any point in the past seven decades, We’re to be in danger of forgetting the lessons from Bretton Woods” ROBERTO AZEVÊDO, WTO DIRECTOR-GENERAL, 2019 52 2.1 Introduction This chapter does not attempt to offer an extensive historical overview of trade policy, which has been done masterfully by other scholars (e.g., Irwin, 1998; Bordo et al. 1999; Wilkinson, 2006; Canto, 1983; Bhagwati, 2003; Baldwin, 1989). Instead, my main goal is to situate this dissertation in its proper historical context by analyzing the main trade policy milestones of the 21st-century. This section begins with the winding road of trade liberalization since World War II, briefly tracing how we arrived from the early stages of the Bretton Wood System to the current moment of stagnation of the multilateral system and rising protectionism. I then turn to three key events to understand the current new reality: China’s 52 WTO, DG Azevêdo: We’re in danger of forgetting the lessons from Bretton Woods, October 17, 2019, <https://www.wto.org/english/news_e/spra_e/spra287_e.htm>. 48 accession to the World Trade Organization (WTO) in 2001, the Global Financial Crisis (GFC) of 2008 to 2009, and the current trade war between the United States and China. I conclude with some final comments on the relevance of my research from a historical perspective. 2.2 The Winding Road of the World Trade System Since the World War II The rise of protectionism in the 1930s was believed to be one of the primary causes of World War II (WWII), and global leaders were anxious to prevent this from happening again. As a result, they convened in Bretton Woods, New Hampshire, in 1944 to discuss the need for international institutions to lend order to the world economy. A year later, the majority of these countries reconvened to discuss the establishment of a new International Trade Organization (ITO), but their plan never came to fruition due to isolationist politics in the US, with the ITO bill never sent by President Truman to Congress for final approval. Thus, the lack of support from the U.S. was the final nail in the ITO coffin. In its place, the General Agreement on Trade and Tariffs (GATT) became the de facto Bretton Woods institution under which international trade was liberalized in the post-WWII era. Designed solely to govern trade in goods, GATT was limited in the scope and control it could exert over trade policy. The GATT’s first meeting, held in Geneva in 1947, resulted in the reduction of tariffs and created steps towards the liberalization of global trade. The Cold War incentivized the United States to push its democratic agenda, including liberal trade policies, on a global scale as the liberalization of trade continued through the economic growth of the 1950s and 1960s. In the 1970s, however, economic decline and the emergence of the East Asian newly 49 industrializing economies (NICs) triggered a return to protectionism. Since GATT restricted the use of tariffs, many countries resorted to non-tariff barriers (e.g., quotas, subsidies). The 1982 ministerial meetings of the GATT reached a nadir in its history. It was the first time since WWII that there was a decrease in international trade, which highlighted the shortcomings of the trading system as the number of GATT members increased and the global economy grew more complex. Global leaders began realizing the need for a more effective trade organization, one that would be broader in scope and administrative capability (Grant & Kelly, 2005). The U.S. led the way in pushing for free trade ideology, heavily influenced by the leadership and economic values of the Reagan and H.W. Bush administrations. The eighth GATT Round in 1986, also known as the Uruguay Round, underscored the need to organize and promote liberalization efforts for an economy moving into the 21st century. The inclusion of agri-business, services, intellectual property, and trade-related investment was discussed, and it became clear that an agreement made solely for goods had become obsolete. The Uruguay Round didn’t conclude until 1995. 53 While this lengthy session took place, leaders of the United States, Canada, Japan, and the EU discussed creating an international trade organization to more effectively cover global trade flows not included in the original GATT. The World Trade Organization (WTO), founded in 1995, differed from the GATT in fundamental ways to address some of the latter’s weaknesses. Most importantly, new and binding procedures for settling disputes were implemented, as this was a major downfall of 53 For more information about the Uruguay Round, visit < https://www.wto.org/english/thewto_e/minist_e/min98_e/slide_e/ur.htm>. 50 the GATT. Especially contentious issues in the Uruguay Round––such as agriculture, intellectual property (IP), and services––were also incorporated into the WTO. In contrast with the provisional nature of the GATT, everything decided upon within the WTO was mandatory for participating nations. By 2003, 146 countries were members of the WTO, and it had dealt with 298 cases. While the WTO’s structure increased efficiency in settling disputes and upholding rules over trade policy, its creation also raised concerns about national sovereignty. Many anti-globalists feared the implications of the WTO and wanted to see its collapse. The first ministerial conference of the WTO was held in Singapore in 1996. At this round, working groups were established around four main issues, which came to be known as the “Singapore Issues.” These included transparency in government, trade facilitation, trade-related investment, and competition policy. Although most of these issues were opposed by developing countries, the more powerful bloc of developed nations insisted that any new negotiations must include the “Singapore Issues.” After the Singapore meeting in 1996, the 1999 Seattle ministerial conference, which took place amid massive anti-globalization protests, marked an important milestone for the WTO. Over 40,000 protesters demonstrated, angered by environmental degradation and labor rights; the WTO’s poor incorporation of developing nations only exacerbated this ire. The Seattle meeting was a patent failure. Up to this point, negotiations had followed a “green room” setup, where a few key G-7 decision makers would be in the room negotiating, rather than all WTO member countries. This exclusion of developing nations made decisions less effective and many saw this as a violation of the multilateral norms upon which the WTO 51 had been founded (Grant & Kelly, 2005; Hopewell, 2016; Jones, 2015). Massive anti-WTO social movements and the developing countries themselves began demanding more transparent and inclusive styles of negotiation (Casey-Sawicki, 2018). After the Seattle debacle, and without sufficiently addressing those earlier demands, a WTO ministerial meeting was held in Doha, Qatar, in November 2001 and the ninth negotiation Round of the GATT/WTO was officially launched. The main objective was to involve developing countries more authentically in global trade negotiations. Doha presented a much less feasible destination for 40,000 protesters than Seattle, especially in the wake of the 9/11 terrorist attacks on the U.S. This meeting birthed the Doha Development Round, which sought to address issues on the “old” (agriculture, NTMs, and the market access for manufactured exports from South to North) and “new” (IPRs, services, and investment) trade agendas. Developing countries also had the opportunity to discuss the difficulties they had in implementing earlier commitments made under the Uruguay Round (Grant & Kelly, 2005). Targeted completion date of the Doha Round was 2005; however, by 2006, the negotiations were suspended, and the round has not ended as of publication date. Many developing countries had been under the impression that the Doha Round would right previous wrongs around transparency and inclusiveness. In particular, they hoped to see more favorable decisions made regarding market access for agriculture and industrial goods. Yet the United States and the EU refused to offer significant concessions on both (Bhagwati et al., 2016). The stagnation of the Doha Round was accompanied by (and in part a consequence of) important events in the multilateral system. One was the decreasing U.S. support of multilateralism and especially of the international trading system over time, something that Allee (2012) attributed to three factors: “the decline in U.S. hegemony; the role of ideas, and, 52 particularly, the change from the free trade mentality to one that focuses more on the unfair practices of other nations; and the role of domestic interest groups, and, particularly, the increasing effectiveness of import-competing interests” (p. 235). The decreasing support of developed nations for the international trading system led developing countries to have a more active role in the defense of the system. In particular, countries such as Brazil, Russia, India, and China (BRICs) have become important supporters of a more balanced agenda between the developing and developed world. However, as Vickers (2012) said, while these emerging economies “show greater activism in the organization, activism does not equate with leadership” (p. 254). In that sense, it is key to understand the role of the G20 developing nations’ group (different from the G20 of developed economies) in the Doha Round and the high-profile role played by Brazil and India especially. 54 The other two emerging economies with an increasing role in the international trading system, Russia and China, were in different circumstances than that of Brazil and India. Russia did not join the WTO until 2012, and China was concentrating on implementing its 2001 accession commitments. The increasing role of these non-central economies is not only explained by their economic growth, but also their very active trade and investment relations—especially with East Asia countries—which challenge the centralized management of the world trading system by central economies. As Vickers (2012, p. 256) has cautioned, this “shift of systematic influence” signaled a drastic restructuring of the international economic order, and it reflected a “balance of power [that is] more multipolar, even multicultural.” 54 G-20 is a coalition of developing countries “pressing for ambitious reforms of agriculture in developed countries with some flexibility for developing countries.” See Groups in the WTO, < https://www.wto.org/english/tratop_e/dda_e/negotiating_groups_e.pdf>. 53 A third significant ministerial conference, albeit short-lived, met in Cancun, Mexico, in 2013, and erupted in violence. 55 The ministerial meeting held in Hong Kong in 2005 and Geneva in 2008 ended in similar stalemates. The developing nations tried to be involved in the negotiations and advocated for their interests, but gridlocks between them and developed countries prevented much from getting done. Within the context of these aforementioned events, there are at least two relevant examples that show how the global trading system is currently in the midst of profound changes: first, the “mega-regional agreements” that facilitate easier consensus by omitting some countries, and second, the “regionalization and bilateralization of trade,” which has led to more RTAs. Moreover, the 2008-09 GFC has likely contributed to the “second wave” of protectionism, or the “protectionism resurrection.” For example, the WTO has increased restrictive measure by 11 percent between 2008 and 2016 (Albertoni, 2018, pp. 156-157). 56 Perhaps most absent in the Doha Round was sound U.S. leadership. In fact, the early 2000s marked the beginning of mini-trade wars by the United States. In 2002, for example, President George W. Bush instituted temporary tariffs on steel imports. The EU retaliated by placing tariffs on U.S. goods such as Florida oranges and American-made cars. The WTO found that the actions of the U.S. were in violation with its rules, and they ended 18 months after implementation. 55 To learn more about this Ministerial Conference in Cancún, visit <https://www.wto.org/english/thewto_e/minist_e/min03_e/min03_e.htm>. 56 As of January 17, 2020, 303 RTAs were in force. See World Trade Organization, ‘Regional Trade Agreements’, <https://www.wto.org/english/tratop_e/region_e/region_e.htm#facts>. 54 2.3 China’s Accession to the World Trade Organization in 2001 The singularly important event in terms of trade policy dynamics during the 21st century was China’s accession to the World Trade Organization (WTO) in December 2001. Trade policy experts had downplayed the effect of China’s WTO accession because the U.S. had granted China the most-favored nation (MFN) status as far back as the 1980s (Autor et al., 2016). 57 Yet studies on the impacts conducted a few years after China’s entrance demonstrate that China has been the main beneficiary, including “US$31 billion a year from trade reforms in preparation for accession and additional gains of $10 billion a year from reforms after accession” (Ianchovichina & Martin, 2004, p. 3). In addition, among China’s commitments in its accession to the WTO were reforming its system of tariffs and quotas to ease the circulation of commodities; changing its practice of state trading to encourage volume control; and introducing critical service sectors, including “telecommunications, distribution, banking, insurance, asset management, and securities to foreign direct investment” (Lardy, 2001). More specifically, Lardy (2001, p. 11) argues that “the protocol governing its accession sets forth China’s commitment to abide by international standards in the protection of intellectual property and to accept the use by its trading partners of a number of unusual mechanisms that could be used to reduce the flow of Chinese goods into foreign markets.” 58 57 Studies have shown that the possibility of a return to non-MFN tariffs, which averaged 37.0% in the late 1990s and compared to average MFN tariffs of only 3.4% in those years (Pierce & Schott, 2016), “dissuaded Chinese firms from investing in exporting to the U.S. WTO accession removed this uncertainty and encouraged China-U.S. trade” (Autor et al., 2016, p. 11). Also see Salam, R. (2018), Normalizing trade relations with China was a mistake, The Atlantic, < https://www.theatlantic.com/ideas/archive/2018/06/normalizing-trade-relations-with-china-was-a- mistake/562403/>. 58 Lardy, N. R. (2001). Issues in China’s WTO Accession. Brookings Institute. (May 9, 2001). Retrieved from: <https://www.brookings.edu/testimonies/issues-in-chinas-wto-accession/>. 55 China’s accession to the global trading system was also an institutional signal to the rest of the world about its intention to compete under the same rules. In this regard, Lardy (2001) notes that being part of the WTO “impel[s] China to be accountable to an internationally agreed set of rules and bind[s] them to wide-ranging economic and systemic changes in order to meet the commitments they have agreed to undertake as a part of WTO accession.” 59 However, China has evaded full compliance with its commitments and its current trade tensions with the U.S. has triggered a debate about whether letting China into the WTO was a mistake. Levy (2018, para. 3) reminds us that during the 15 years of negotiation leading up to China’s accession, WTO country members had set several conditions for China’s admission, which “involved concessions such as dropping tariffs on many categories of goods, opening up agricultural trade, and allowing in foreign service providers. In contrast, the United States did not need to make any new market-opening concessions.” 60 Autor et al. (2016) see China’s accession to the WTO as one of the three ‘China Shocks’ to the global economy. The first shock occurred in the 1980s and 1990s with the economic opening of China and the take-off of India’s growth, which expanded production based on low-skilled labor. According to Reisen and Stemmer (2018), the second ‘China shock’ spans the time of its accession to the WTO in 2001 up to the 2008-09 GFC, and is based on the “pervasive convergence of poor countries largely due to increasingly China- 59 Idem. 60 Levy, P. (2018). Was Letting China into the WTO a Mistake? Foreign Affairs. Retrieved from: <https://www.foreignaffairs.com/articles/china/2018-04-02/was-letting-china-wto-mistake>. 56 centric growth and higher raw material prices.” 61 The third shock is still underway and started with the 2008-09 GFC. This period has seen a reversal of previous trends “as China is transforming its production and trade patterns toward consumption, away from investment and intermediate trade.” 62 According to the World Bank (2020), since its accession to the WTO, China has reduced tariff rates considerably for both primary and manufactured products thereafter. While China’s average tariff rate for primary products five years before its accession to the WTO (1996 to 2000) was 19%, the average was 8% between 2001 and 2006. 63 For manufactured products the averages were 15% and 7%, respectively, before and after the accession. Since the GFC, China’s tariff rate reductions have approximated average tariff rates in the U.S. and the EU (see Figure 2.1). Figure 2.1 – The Evolution of Tariff Rates for Primary Products and Manufactured Goods Imposed by China, the U.S., and the EU. A. Tariff rate, applied, weighted mean, primary products (%) 61 Reisen, Helmut and Michael Stemmer (2018). The Three Acts of the ‘China Shock.’ ShiftingWealth Blog. Retrieved from: <http://shiftingwealth.blogspot.com/2018/04/the-three-acts-of-china-shock.html>. 62 Idem. 63 World Bank Open Data, <https://data.worldbank.org/> 57 B. Tariff rate, applied, weighted mean, manufactured products (%) Source: Author’s creation based on World Development Indicators. Finally, based on UNCTAD’s Non-Tariff Measures (NTMs) dataset, 64 China increased NTMs after its accession to the WTO. While China had an average of 244 NTMs five years before its accession to the WTO (1996 to 2000), this average jumped to 1,582 between 2001 and 2006. 65 These barriers have made China the “country notifying the second largest number of technical barriers to trade (TBTs),” just after the United States (Ghodsi 2018). This is in line with Evenett and Fritz (2018), who have shown that now, many countries are increasingly using nontransparent policy instruments (non-tariff or “murky” protectionism) as a main trade policy instrument. While China agreed to limit tariffs when it joined the WTO, its trade policy tools simply took on a new form. Garred (2018) found that China has also influenced international trade through export restrictions and value-added tax rebates. This international phenomenon of utilizing various trade policy tools suggests that 64 Examples of NTMs are sanitary and phytosanitary measures, technical barriers, quantity control guidelines, and price controls. See United Nations Conference on Trade and Development (UNCTAD), Non-Tariff Measures data, < https://trains.unctad.org/Forms/TableView.aspx?mode=modify&action=search>. 65 U.S. average NTMs initiated by the U.S. five years before China’s accession to the WTO (1996 to 2000) were 539. Between 2001 and 2006, this same figure for the U.S. was 328. See UNCTAD, Non- Tariff Measures data, op. cit. 58 today’s trade war is nothing new to global trade, but rather “the latest example of an ongoing battle whose skirmishes have taken many forms” (Garred, 2019). 66 Recently, much of Donald Trump’s rhetoric throughout his 2016 candidacy for president framed his protectionist trade policy as a direct response to other global players who were “stealing” American jobs. However, research has found that support for protectionism was not due to globalization or trade shocks. There wasn’t a significant correlation between households in economic distress and support for Trump. Instead, those voters who supported his policies felt the threat of a changing social dynamic at home, and a loss of U.S. dominance abroad (Noland, 2020). 2.4 Global Trade After the 2008-09 Global Financial Crisis Since the 2008-09 global financial crisis (GFC), many countries have erected new trade barriers. However, traditional protectionist measures (e.g. tariffs) did not rise as dramatically as expected. 67 New trade policy tools that are less transparent have taken their place. Niu et al. (2018) show that although average tariff rates have fallen since the GFC, there has been a sharp increase in the number of non-tariff measures. These tend to be more restrictive and are clustered in the form of technical barriers to trade as well as sanitary and phytosanitary standards, especially in Central Asia, North America, South Asia, and Europe. 68 Again, this contrasts sharply with global trends over the past 50 years, which saw a sustained opening of 66 Garred, J. (2019). The persistence of trade policy beyond import tariffs. VoxEU. Org (9 July), <https://voxeu.org/article/persistence-trade-policy-beyond-import-tariffs>. 67 See Desilver, D. (2018). U.S. Tariffs Are Among the Lowest in the World – and in the Nation’s History.” March 22, Pew Research. Retrieved from: < https://www.pewresearch.org/fact- tank/2018/03/22/u-s-tariffs-are-among-the-lowest-in-the-world-and-in-the-nations-history/> . 68 An example of a technical barrier to trade is a safety standard for manufactured goods. Sanitary and phytosanitary measures concern safety standards for food or animals. 59 national markets. Since 2009, in contrast, G20 governments implemented restrictive measures on 9,041 occasions (Evenett & Fritz, 2018, p. 6). 69 These are all examples of the rise of protectionism during the last decade. Recent studies have analyzed the impact that the 2008 financial crisis had on international trade policy instruments. Countries have overwhelmingly moved towards protectionist policies, with 70 percent of world trade impacted by these interventions (Evenett, 2019). While global crises induce more restrictive measures, nearly all of the policies implemented post-2008 have fallen within the boundaries of WTO rules. Non-tariff measures (NTMs) are more complex, harder to detect, and have “taken over the center stage of trade policy instruments” (UNCTAD, 2010, p. xiii). They also have greater power in influencing modern problems, such as environmental guidelines for the protection of firms and consumers. The counter-productive use of NTMs has gripped the attention of international agencies, which are now working to define, collect information on, and analyze this misappropriation of modern trade policy instruments. These new dynamics of trade protectionism after the GFC will be elaborated upon in chapter 3. 2.5 Bilateral Tensions Between the U.S. and China in the Present and the Past Since 2016, the U.S. has been leading trade wars with the world on multiple fronts. Yet signs of protectionist tension can be found as far back as 2009 during the Obama administration. The U.S., for example, limited the number of Mexican trucking firms that could operate within specific areas of the country. At the same time, Obama “softened his tough rhetoric on 69 When we refer to restrictive measures, we are talking about any kind of protectionist policy instrument that can take the form of a tariff or a non-tariff measure under the classification of the UN MAST classification, which can be found at <https://unctad.org/en/Pages/DITC/Trade-Ana lysis/Non-Tariff-Measures/NTMs-Classification.aspx> 60 free trade, warning repeatedly against tit-for-tat protectionism in the midst of an economic crisis” (Alexander & Soukup, 2010, p. 324). 70 Concrete demands for protectionist policies under Trump began with requests filed in 2017 by the solar panel and washing machine industries. The U.S. International Trade Commission quickly discovered that foreign imports were harming domestic businesses, and the Trump administration imposed tariffs on solar panels and washing machines in January 2018. This affected approximately $8.5 billion in solar panel imports and $1.9 billion worth of washing machines, largely from China. China countered with tariffs on soybeans imports from the United States shortly thereafter, while Korea and China filed WTO disputes against the U.S. The question here is how current trade tensions might differ from previous ones. What explains the length and depth of the current trade slump experienced in the wake of the 2008-09 GFC? Trade tensions or protectionist trends caused by economic shocks that spur the adoption of beggar-thy-neighbor measures are not new in economic history. But why might this time be different from previous trade disputes and protectionist spirals in history? Let’s first revisit some similar situations in economic history, which may prove to be illuminating on the state of protectionism today. In 1929, when the U.S. passed the Smoot-Hawley tariff act, raising tariffs by 60 percent for more than 3,000 products, at least 60 other nations implemented retaliatory measures against the U.S. This almost doubled the world’s average level of trade protection, and world trade contracted (Boffa & Olarreaga, 2012). Between 1929 and 1933, total global trade decreased by 25 percent (Canto, 1983), and according to Irwin (1998), 70 percent of 70 See The Economist (2009) Obama and Trade: Low Expectations Exceeded, <https://www.economist.com/united-states/2009/04/30/low-expectations-exceeded>. 61 that global trade reduction could be attributed to the Smoot-Hawley tariff and the subsequent retaliatory measures it invoked. However, in 1934, the U.S. sought to promote economic recovery by reducing tariffs; the U.S. Congress passed the Reciprocal Trade Agreement Act, which reduced these tariffs (Irwin, 1998). Under the 1934 Trade Act, the executive took command of trade negotiations, and the U.S. signed bilateral trade agreements with 20 countries. Tariffs were eventually lowered to 50 percent of the rates established by the Smoot-Hawley act in 1929 (Canto, 1983). Under section 301 of the Trade Act of 1974 and the Omnibus Trade and Competitiveness Act of 1988, the U.S. government demonstrated “increased willingness to threaten retaliation against protected foreign markets” (Gould & Woodbridge, 1998, p. 116). During this time (the late 1980s and early 1990s), a main objective was to induce macroeconomic reforms and trade opening in Japan. In 1993, U.S. President Bill Clinton said, “I am particularly concerned about Japan’s growing global current account and trade surpluses and deeply concerned about the inadequate market access for American firms, products, and investors in Japan.” 71 This is not too far from what we hear currently about China’s current account surplus and trade practices. During the 1990s, through the use of various trade and non-trade mechanisms (e.g., antidumping duties and the use of countervailing sanctions), the U.S. threatened to implement protectionist measures against Korea, Japan, and European countries, seeking to improve U.S. access to these markets (Gould and Woodbridge, 1998). Evenett and Fritz (2018) identify at least three main differences between the 1980-90s trade tensions (mostly between the U.S. and Japan) and 71 Public Papers of the President of the United States, William J. Clinton, 1993 (January 20 to July 31). 62 current Sino-U.S. trade disputes: Japan is not only a “military ally” of the U.S., but the country poses less of a threat both demographically and economically than China does. Moreover, unlike Japan, China has acted as “the host to the quantum of American foreign investment,” which consequently led to “sustained criticism of discriminating against foreign multinationals inside its borders” (Evenett and Fritz, 2018, p. 13). It is important to distinguish between the type of protectionism originated by the GFC and that originated by the Trump administration. In an interview I conducted with Alejandro Jara, former WTO Deputy Director General: The surge in protectionism is a reality; first as a result of the 2008-09 crisis and later in Trumpian world. But perhaps such protectionism is reduced to certain sectors (steel and aluminum are examples) and not a widespread phenomenon. Until the Trumpian world, a great deal of protectionism took place as a result of subsidization and trade remedies (particularly antidumping), which may indicate that the appeals for protection were presented as either “I’m too big / strategic to fail” or “it’s about fairness, not about protectionism.” 72 Beyond the bilateral differences between previous and current trade tensions, there are also major global factors that may be impacting the long-term consequences of current trade disputes and protectionist dynamics. Unlike what happened in the 1980s and 1990s, the high level of trade interdependence through global value chains (GVC) and preferential trade agreements (PTA) render seemingly bilateral trade conflicts far from solely bilateral; possible systemic consequences can be key explanatory variables in explaining why the current trade recovery is still relatively anemic compared to recovery from previous global 72 Author’s interview with Alejandro Jara, June 2020. See appendix 5 for full interview. 63 trade collapses. 73 As Lamy (2013) states, one of the major changes we see nowadays is the level of interdependency in trade: Almost 60 percent of trade in goods is now in intermediates. ... An important consequence of the integration of production networks is that imports matter as much as exports when it comes to contributing to job creation and to economic growth. In 1990, the import content of exports was 20 percent; in 2010, it was 40 percent, and it is expected to be around 60 percent in 2030. This is why enacting protectionist measures in the modern world to protect jobs, such as raising import barriers, can have an inverse reaction in economies that are increasingly reliant on imports to complete their exports. 74 2.5.1 U.S.-China Trade Tensions Step by Step The U.S.-China trade war has been mounting since Donald Trump hit the campaign trail. Trump spoke frequently of the U.S. trade deficit, which has been the world’s largest since 1975. His move towards protectionism, and specifically policies pointed towards China, are supposed to improve national well-being; yet experts, economic theory, and history all prove that trade wars will only cause trouble (Thoms, 2019; WEF, 2019). Import tariffs bring economic losses to countries, producers, and consumers (Crowley, 2019) and it is estimated that the U.S. trade war has already lowered GDP by 0.6% (Amadeo, 2019). As studies have shown, most of these price increases are passed on to American consumers and not offset by production benefits. At the start of 2018, U.S. tariffs on Chinese imports were 3.1%, while Chinese tariffs on U.S. exports were 8%. In January of 2020, those rates stood at around 23.8% and 25.1%, respectively (Bown & Kolb, 2019). However, in December 2019, China and the U.S. agreed to pause the tensions as part of a compromise that “requires structural 73 As Bussière et al. (2011) show, one of the distinctive characteristics of the post-2008-09 period was that trade decreased much more than output. In 2009, real world output contracted by 0.7 percent, whereas real trade flows fell by 11 percent. These features are surprising because they stand in sharp contrast with past experiences. That is why the dynamics of trade in 2009 became widely known as the “Great Trade Collapse” (Baldwin, 2009). 74 Lamy (2013), Global value chains, interdependence, and the future of trade <https://voxeu.org/article/global-value-chains-interdependence-and-future-trade>. 64 reforms and other changes to China’s economic and trade regime in the areas of intellectual property, technology transfer, agriculture, financial services, and currency and foreign exchange.” 75 Trump’s legal authority to impose the first round of tariffs came from Section 301 of the U.S. Trade Act of 1974, which states that the president can impose tariffs if the U.S. International Trade Commission finds that imports are causing harm to an industry. However, Section 301 has rarely been used in recent history. The last use was in 2001 when George W. Bush imposed steel tariffs. Trump once again used a rarely used section of trade policy, this time, Section 232 of the Trade Expansion Act of 1962, to impose further tariffs. In April 2017, he instructed the Commerce Secretary to investigate the steel industry, and by March 2018 he imposed steel tariffs. These tariffs temporarily exempted several countries, including Mexico, Canada, Brazil, Australia, and Korea, as well as the European Union. China retaliated shortly after by imposing tariffs on $2.4 billion worth of U.S. goods, which closely matches the $2.8 billion that was affected by the steel tariffs. The steel tariffs continued to follow a tit-for-tat strategy, with the U.S. raising them for certain countries and those countries then retaliating. Although the tariffs were successful in creating U.S. jobs, they came at a high cost. Each of the 8,700 jobs costs about $650,000 to create. Poor, developing countries were hit the hardest by these tariffs, experiencing a 12% decline in steel exports to the U.S. and a 15.5% decrease in revenue (Bown & Kolb, 2019). Under Trump’s leadership, the United States continued to impose new tariffs throughout 2018 and 2019 to protect industries such as automotive, consumer goods, intermediate goods, and technology. 75 USTR, United States and China Reach Phase One Trade Agreement, <https://ustr.gov/about- us/policy-offices/press-office/press-releases/2019/december/united-states-and-china-reach>. 65 Unsurprisingly, these actions have sparked retaliation from countries on the receiving end, and this explains the current trade war. The mounting tariffs have also done more to deepen the U.S. trade deficit, rather than closing it. The deficit reached a “10-year high of $621 billion in 2018” (Curran, 2019), and economists believe that these tariffs have contributed to this gap by hindering economic growth rates for China and Europe. A trade war between the two largest countries in the global economy also creates a level of uncertainty that has made international leaders uneasy. The role of the dollar-based payment system as the backbone of global commerce also gives America additional ways of influencing trade, which “has been weaponized” (The Economist, 2019, p. 13). For instance, a Chinese technology company was banned from doing business with America, which effectively isolated it from the global financial system. As foreign trust in the Federal Reserve declines, global business leaders are looking for alternatives in a “post-American era” (Ibid, p. 14). While analyzing the U.S.-China trade war can provide valuable insights, it is important to view this in the context of the global trading environment. Evenett (2019) argues that the energy spent on bilateral trade fights may be disproportionate to their actual effect on the global economy. For example, just 2.6% of traded goods have been affected by these bilateral disputes, and U.S.-China trade accounts for only 4.4% of traded goods. Meanwhile, 16.5% of goods are impacted by tariffs of any nature, and 27.2% of world trade is affected by foreign firms trying to compete with subsidized domestic producers. 76 Most of 76 Evenett, S. (2019), “For trade war junkies: Why not put the same energy into analyzing & fixing two trade policy problems that implicate much more global trade than the Sino-US trade dispute? ...” 66 the goods traded globally are influenced by more than a bilateral trade dispute, and this fact shouldn’t be overstated by the attention that something like the U.S.-China trade war garners. 2.6 Final Comments on 21st-Century Protectionism from a Historical Perspective The question raised by this historical review is to what extent protectionism will be the new norm. Blaming foreign influences for domestic distress is common practice for American politicians, which makes free trade an easy target during election season (Irwin, 2016). Even Democratic politicians, such as Hillary Clinton and Barack Obama, voiced concerns about free trade agreements. Donald Trump claims his protectionist policies will help Americans, but instead these reckless U.S. actions risk triggering a global trade war that will have adverse effects on all countries (Irwin, 2017). The trade war has already brought an annual loss of $68.6 billion to U.S. producers and consumers, and an aggregate loss of $7.8 billion to the U.S. economy (Fajgelbaum et al., 2019). In the past decade, “murky” trade policies have further complicated this issue. As countries continue to move toward less transparent instruments, the ways in which we’ve historically measured their impact are becoming outdated. The opaque nature of those policies implemented over the past decade is a key element to be considered. As Evenett (2019) has said (and the following figures show), “for every tariff hike this year, more than 3 trade-distorting subsidies have been imposed. Perspective needed in judging 2019 global [Twitter post, Sep, 2019] Retrieved from @SimonEvenett: <https://twitter.com/SimonEvenett/status/1169553731060297728>. 67 trade policy dynamics—US-China is not the only game in town.” 77 Interestingly, in 2019, 618 new harmful trade measures were implemented worldwide, with 168 of them corresponding to China and the U.S. (GTA, 2019). Figure 2.2 – Putting 2019 Global Trade Policy Dynamics in Perspective A. Harmful measures found by 2019 B. Types of measures used by governments Source: Global Trade Alert, 2019. Yet these non-transparent trade policies aren’t getting the media or academic attention needed to properly understand their effects (Evenett, 2019). Most research focuses on tariffs, as we have the most data on these measures, rather than looking at ways in which global trade policy is evolving. This lack of appropriate data and research leads to “inadequate scrutiny bias” among leading scholars. They maintain that global trade was unaltered by the financial crisis, yet more comprehensive research on trade distortions could disprove this viewpoint (Ibid). 77 Evenett, Simon (August, 2019). “If it bleeds it leads” I get that. But for every tariff hike this year, more than 3 trade-distorting subsidies have been imposed….” [Twitter post] Retrieved from @SimonEvenett: < https://twitter.com/SimonEvenett/status/1165645814267559940/photo/1> 68 This increasingly murky context explains why there is uncertainty about international trade across the globe. As the new Index of Trade Uncertainty shows, it is “rising sharply, having been stable at low levels for about 20 years.” 78 This quickly increased in quarter three of 2018, the same time when the U.S.-China tariff war launched from Washington, D.C. This index movement has been tracking U.S.-China tensions closely. When officials announced that they were halting tariff escalations at the G-20 summit in the fourth quarter of 2018, the index went back down. It then spiked again when China implemented several tariffs on U.S. goods in March 2019. As the world’s largest economies, the U.S.-China disputes set trends and frame how foreign officials view the global trade environment. Analysts have cautioned that the trade war is harming the global economy, and several have come together to urge China and the U.S. to find common ground. 79 Both governments have promised trade talks, yet coming to an agreement will prove to be difficult, especially since the U.S. has been unclear about what it wants out of these negotiations (Evenett, 2019). 78 Hites, Bloom, and Furceri (2019), New Index Tracks Trade Uncertainty Across the Globe, <https://blogs.imf.org/2019/09/09/new-index-tracks-trade-uncertainty-across-the- globe/#.XXdyqn18_Zk.twitter>. 79 Donnan, Shawn, Economists Call for Alternative Path to U.S.-China Trade Wars (October 26, 2019), Bloomberg, <https://www.bloomberg.com/news/articles/2019-10-27/economists-call-for-alternative- path-to-u-s-china-trade-wars?utm_source=google&utm_medium=bd&cmpId=google>. 69 CHAPTER 3 – UNDERSTANDING TRADE PROTECTIONISM PIECE BY PIECE. A DESCRIPTIVE ANALYSIS OF THE DATA “If I had to identify a theme at the outset of the new decade, it would be increasing uncertainty.” KRISTALINA GEORGIEVA MANAGING DIRECTOR OF THE IMF, 2020 80 3.1 Introduction This chapter is divided into two main sections. First, I describe the different data sources used for this dissertation. Second, I present a more detailed analysis at the state and industry/sectoral levels and discuss the types of protectionist measures that have been implemented over the last decade. For clarity, most of the descriptive analysis at the state level is concentrated in G20 countries, which represent 80% of gross world product (GWP) and three-quarters of global trade. 81 This group of countries also includes the three Latin American cases (Argentina, Brazil, and Mexico) that will be analyzed in detail in subsequent chapters. 80 IMF Managing Director Kristalina Georgieva Speech at the Peterson Institute for International Economics, in Washington, D.C., Jan. 17, 2020, < https://www.imf.org/en/News/Articles/2020/01/17/sp01172019-the-financial-sector-in-the-2020s>. 81 The G20 is “the premier forum for international economic cooperation. The G20 brings together the leaders of both developed and developing countries from every continent.” (See: G20, What is the G20? < https://g20.org/en/about/Pages/whatis.aspx>.) It is different from the developing nations’ G20 group in the WTO, which is a coalition of developing countries formed in 2003 during the Doha Round Negotiations, “pressing for ambitious reforms of agriculture in developed countries with some flexibility for developing countries” (See: Groups in the WTO, <https://www.wto.org/english/tratop_e/dda_e/negotiating_groups_e.pdf>. 70 3.2 The Data This research is based on several data sources that were retrieved from public repositories. 82 The four main datasets used for this project are: 1) Global Trade Alert (GTA), which provides information on those government-led trade interventions undertaken since 2008, which are most likely to affect foreign trade, 83 2) BACI–CEPII trade data, which disaggregates data on bilateral trade flows for more than 5,000 products and 200 countries, 84 3) the International Political Economy Data Resource (IPE), 85 which compiles data from different IPE data sources into a single dataset (Graham & Tucker, 2017), and 4) data from the Harmonized System (HS) classification of products, which classifies “traded goods on a common basis for customs purposes.” 86 3.2.1 Global Trade Alert (GTA) Since its inception in 2009, the Global Trade Alert (GTA) has tracked state-to-state interventions that may influence foreign trade, including “state interventions affecting trade in goods and services, foreign investment and labor force migration.” 87 It was formed to help prevent concerns that “the global financial crisis would lead governments to adopt widespread 1930s-style beggar-thy-neighbor policies” (Evenett & Fritz, 2020, p. 1). Since it was launched, many studies have relied on the GTA database (examples are Evenett et al., 2011; Boffa & Olarreaga, 2012; Henn & McDonald, 2014; Georgiadis & Gräb, 2016; 82 See the appendix for a detailed description of the raw data used to generate the entire project. 83 For more details about the Global Trade Alert (GTA), visit <https://www.globaltradealert.org/>. 84 For more detail about BACI – CEPII trade data, visit < http://www.cepii.fr/CEPII/en/bdd_modele/bdd_modele.asp>. 85 For more details about the IPE Data Resource, visit <https://doi.org/10.7910/DVN/X093TV>. 86 For more details about the Harmonized System (HS), visit <https://unstats.un.org/unsd/tradekb/Knowledgebase/Harmonized-Commodity-Description-and- Coding-Systems-HS>. 87 Global Trade Alert, <https://www.globaltradealert.org/>. 71 Evenett, 2019). Although the GTA is relatively new, covering the past 12 years, this project has been praised for its extensive scope. For instance, in 2016, the International Monetary Fund (IMF) noted in its World Economic Outlook that GTA “has the most comprehensive coverage of all types of trade-discriminatory and trade-liberalizing measures, although it begins only in 2008’’ (IMF, 2016, p. 76). As explained by Evenett & Fritz (2020, p. 1), authors of this dataset: Each GTA database entry documents a government statement … which included a credible announcement of a meaningful and unilateral change in the relative treatment of foreign versus domestic commercial interests … includes various forms of government action from national legislation to the contract terms of individual state agencies. All documented changes reflect unilateral government action and thus exclude changes coordinated within bilateral trade agreements or the multilateral trading system. The foreign commercial interests considered by the GTA are trade in goods and services, investment as well as labor force migration. For the purpose of this project, it is relevant to note that the GTA records both trade protectionist and liberalization measures. It thus measures both sides of the trade policy mechanism. This means that the GTA has its own built-in control variable. This enables me to test the real weight of both protectionist and liberalizing measures when we consider them as explanatory variables. The following table shows those variables included in the GTA database, which will be used for this dissertation. Table 3.1 – Name and Description of GTA Variables Name of the variable Description Implementing country The country that implements the trade policy Affected country The country that is affected by the trade policy Implementing date The day the policy was implemented Intervention type The type of intervention implemented (e.g., Import tariff, subsidy, etc.) Affected sectors The three digits sector (or group of sectors) most affected by the measure Affected product The six-digit product (or group of products) affected by the measure 72 For each entry, the GTA provides the official document issued by the acting institution within a given country. As described in the GTA’s codebook (Evenett & Fritz, 2020, p. 1): For cases where an official statement cannot be located, press clippings from multiple original sources are analyzed for their consistency. All database entries undergo a two-stage review process before publication. The monitoring period starts in November 2008, when the G20 heads of state pledged to refrain from protectionist action in the aftermath of the Global Financial Crisis. The GTA database only includes relative treatment changes announced since this pledge. The database is updated continuously as new information arrives. Before each database submission, the GTA team checks that the following criteria have been met: first, a possible measure is identified and investigated. Then, a report is compiled and reviewed separately by two senior members of the GTA research team. An entry is included in the dataset when these two review checks have been satisfied. Once the data is published, third parties also check for errors. The following table summarizes this process of data collection from start to finish. Figure 3.1 – Key Steps in GTA’s Data Collection Source: GTA Codebook, 2017 73 In recent years, the GTA has incorporated new strategies to increase the efficiency of its data collection and coverage. For example, as Evenett (2019, p. 10) highlights: The GTA team has built up a library of websites of government agencies, ministries, and gazettes (official journals) and of international organizations that are consulted on a regular basis. Whenever possible, official sources are used to document an entry, and this has been accomplished with over 93% of entries in the GTA database. However, the GTA team scours newspapers, reports by industry associations and law firms for leads of government policy intervention. Once a lead is identified, a team member investigates it and the goal is to find an official source to support any write-up. Increasingly, website scraping tools are being used to identify potential changes in government policy. Automation offers the potential to cover even more government websites and expand the GTA’s coverage. In 2018, the GTA included in its data the percentage of goods traded bilaterally that have been affected by policy interventions implemented between 2009 and 2018. This newly included data offers an enormous opportunity to analyze not only the number of measures implemented, but also the extent to which protectionist measures affect trade flows. Hence, at the empirical level, one of the main contributions of this new data included by GTA is to concentrate not only on the number of measures implemented by countries, but also on how these measures affected the share of exports and imports; the latter is still an open question in debates related to current trade disputes. Following the model that the GTA team uses to disaggregate the share of trade affected annually, in this dissertation, I use data that I have collected using the Harmonized System Classification (HS) and the Classification by Broad Economic Categories (BEC) to run the sectoral models presented in the following chapter. 88 For a clear understanding of how the data is collected and structured, I present a step-by-step example in Table 3.2. Figure 3.2 presents the original document from which the information was collected. 88 My development of this new dataset is part of a joint project with George Gerald Vega, from the USC Center for Applied Network Analysis (CANA). 74 Table 3.2 – A Real Example of Data Collection: Step-by-Step 1. On August 17, 2018, the Government of Mexico temporarily increased the import duty on certain steel products. 2. The document was published by the Secretariat of Economy (DOF: 17/08/2018) on its webpage: <http://dof.gob.mx/nota_detalle.php?codigo=5535213&fecha=17/08/2018>. 3. The official document states that it is implemented on November 30, 2018. The affected country will be the United States and the sectors affected will be those with the following HS codes: 7207.12.01, 7207.12.99, 7224.90.02, 7216.32.04, 7216.33.02, 7216.33.01. 4. Based on the official document, the GTA coders report the following information: implementing country, affected country, implementing date, intervention type, affected sector, and affected product. Source: Global Trade Alert, <www.globaltradealert.org>. Figure 3.2 – Example of an Official Trade Policy Implementation Document Source: Secretary of Economy of Mexico, <http://dof.gob.mx/nota_detalle.php?codigo=5535213&fecha=17/08/2018>. 75 3.2.2 BACI-CEPII Trade Data The quantitative analysis of this dissertation draws mainly on BACI-CEPII trade data. This dataset contains estimates of the composition of trade between countries. In particular, for each country pair (exporter A, importer B), it has an entry of a product (HS 2012 code) with the estimate of how much that product accounts for exports from A to B. 89 The version of BACI-CEPII that I am using is HS 2012. 90 Following the GTA dataset structure, I use three years of data (2005, 2006, and 2007) and take the annual averages, which allows me to avoid endogeneity issues. In some cases, there is only information about one or two years, so when computing the annual averages, I use less than three years. 91 Hence, the main variable that I consider from the BACI-CEPII Trade Data is the percentage (pcent suffix in the dataset) that estimates the proportion of exports from A to B of a given product in the total exports from A to B. Then, by combining that percentage by product from BACI-CEPII with the data provided by the GTA, or the product affected by each protectionist measure implemented from A to B, I can obtain the percentage of trade between 89 The raw data is publicly available at < http://www.cepii.fr>. 90 The Harmonized System (HS) is an international nomenclature for the classification of products. It allows participating countries to classify traded goods on a common basis for customs purposes. At the international level, the Harmonized System (HS) for classifying goods is a six-digit code system. It is composed of approximately 5,300 article/product descriptions that appear as headings and subheadings, arranged in 99 chapters, grouped in 21 sections. The six digits can be broken down into three parts. The first two digits (HS-2) identify the chapter in which the goods are classified, e.g., 09 = Coffee, Tea, Maté and Spices. The next two digits (HS-4) identify groupings within that chapter, e.g., 09.02 = Tea, whether or not flavored. The next two digits (HS-6) are even more specific, e.g., 09.02.10 Green tea (not fermented). Up to the HS 6-digit level, all countries classify products in the same way (a few exceptions exist where some countries apply old versions of the HS). For this project, I use HS2012. For more details about the Harmonized System (HS), visit: <https://unstats.un.org/unsd/tradekb/ Knowledgebase/Harmonized-Commodity-Description-and-Coding-Systems-HS>. 91 Because of this, I created a new variable (with _by_3) in which instead of taking the average with whatever number of observations we observe, I simple divide the total export flow (sum through the 2005-2007 range) by three. 76 A and B that is affected by each protectionist measure. Figure 3.3 shows how the data of the GTA and BACI-CEPII are combined to develop a unique dataset that illustrates the percentage of trade affected by protectionist measures from one country to another. In sum, for this research, I will use two types of variables to measure protectionism. One is the number of protectionist measures one country implemented against another. The second measurement I will consider, which has not been used extensively within the literature, is the share of trade that is affected by these protectionist measures. Given the specificity of this data, the time frame of some factors will vary (i.e., for some measures I have data until 2018, while I have data up to 2019 for number of measures implemented). Figure 3.3 – Summary of Data Sources Compiled in the Final Database Source: Author’s creation. 3.2.3 The IPE Data Resource The IPE Data Resource is a standardized dataset that compiles variables from 89 IPE data sources into a single dataset. Its main goal is “increasing efficiency and reducing the risk of 77 data management errors” (Graham & Tucker, 2017, p. 149). The IPE Data Resource attempts to draw all data from its original source (i.e., downloading the data directly from the scholar or the institution that produced the data). For both country-year and dyad-year, the unit of analysis in the dataset is Gleditsch-Ward number (gwno) and year (Gleditsch & Ward, 1999). Alternative country identifiers, i.e., Correlates of War (COW) codes and International Financial Statistics (IFS) codes, are provided for convenience, but the Gleditsch-Ward numbers are the basis upon which component datasets are merged together. For the purpose of this dissertation, from the IPE Data Resource, I consider the relative economic difference bilaterally (B’s GDPpc / A’s GDPpc), 92 bilateral investment treaty between A and B, common official language, and Geographic Data CEPII: population weighted distance (typically used in a gravity model). 93 3.2.4 Data sources for other variables of interest: PTAs and GVCs 3.2.4.1 Measuring Preferential Trade Agreements (PTAs) The main dataset used in this dissertation to measure PTAs derives from Baccini & Urpelainen (2014). It compiles bilateral data on Preferential Trade Agreements from 1948 to 2018 for 203 countries. As Ruta (2017, p. 175) states, the “GVC has changed international trade, with trade in parts and components increasing almost six times between 1990 and 2015, faster than the 4.5 times for other forms of trade.” PTAs have not only expanded in 92 The International Political Economy Data Resource (IPE), which compiles data from 89 IPE data sources into a single dataset (Graham & Tucker, 2017), is available at <https://doi.org/10.7910/DVN/X093TV>. 93 See Mayer & Zignago (2011) Notes on CEPII’s distances measures: the GeoDist Database. CEPII Working Paper 2011-25, <http://www.cepii.fr/CEPII/en/bdd_modele/presentation.asp?id=6>. 78 number, but they are also diversifying in content, beginning from the negotiation of 50 PTAs in the 1990s to 303 as of January 2020. 94 Following the World Bank’s definition, Dür et al. (2014) provide a more general description of international trade agreements as “all the agreements that have the potential to liberalize trade” (Dür et al. 2014, p. 1). I will use this definition in this paper, as it is the one most used in the literature (Hofmann et al., 2019; Ruta, 2017). Hence, when I refer to a PTA, this includes free trade agreements, customs unions, or partial free trade agreements, which can take ––as previously said–– bilateral (two countries), plurilateral (many countries), plurilateral with a third country (e.g., EU-Australia Trade Agreement), or region-region (e.g., EU-Mercosur Trade Agreement) forms. 3.2.4.2 Measuring Global Value Chains (GVCs) To measure GVCs, I combine the Global Trade Alert data, the Classification by Broad Economic Categories (BEC), and BACI-CEPII trade data. As mentioned, BACI-CEPII in particular collects data on trade flows (by product) for each country pair (A, B) (exporter, importer). The data is collected using the HS 2012 classification and with an estimate of how much a given product accounts for exporter A to importer B of the total exports from A to B. Once I have the bilateral data (country A, country B, year, and percentage in total flows of each product), I merge this information with Broad Economic Categories (BEC). BEC is the data normally used to measure the share of trade in intermediate goods (a proxy of GVC). The Broad Economic Categories (BEC, para. 1) provides “a means for international trade 94 WTO, Regional Trade Agreements, <https://www.wto.org/english/tratop_e/region_e/region_e.htm>. 79 statistics to be analyzed by broad economic categories such as food, industrial supplies, capital equipment, consumer durables and consumer non-durables.” 95 3.3 A Descriptive Analysis at the State and Industrial Levels of Trade Policy Dynamics In this section, I present a detailed descriptive analysis of state and industry- and sectoral- level protectionism over the last decade (between 2008 and 2018). As mentioned, for clarity, I will concentrate on global and regional dynamics. For individual countries, I will mostly focus on G20 economies, which represent 80% of the gross world product (GWP) and three- quarters of global trade. As I previously discussed, trade protectionism has not only increased over the last decade but also changed its form: historical defenders of trade liberalization are protecting their domestic economies more by using less observable trade policy measures (Evenett, 2019; WTO, 2016). These findings raise a number of questions: 1) Which countries are the most protectionist today, determined not only by the number of measures implemented, but––most importantly––by the share of trade (from the rest of the world) that is affected? 2) Which countries have been most affected by protectionism in the past decade? 3) What measures have most countries used? and 4) What sectors have been most affected trade protectionism? These questions are answered in the remaining sections of this chapter. 95 UNSTATS, Classification by Broad Economic Categories (Rev.4) <https://unstats.un.org/unsd/tradekb/knowledgebase/50089/classification-by-broad-economic- categories-rev4>. 80 3.3.1 Twenty-First Century Protectionism Piece by Piece: States, Sectors, and Measures 3.3.1.1 State-Level Dynamics From a global perspective, we can see that there has been a clear deficit of trade liberalization compared to protectionism measures since 2008. More specifically, between 2008 and 2019, there were 2,646 newly implemented protectionist interventions, while there were just 781 liberalizing measures, creating a “liberalizing deficit” of 1,865 measures. 96 As of the writing of this chapter (May 2020), the Global Trade Alert database reports 390 new protectionist interventions and 197 liberalizing measures between January and May 2020. Most of these trade policy measures have been attributed to the COVID-19 pandemic, which I elaborate on at the end of this chapter. Figure 3.4 – New Trade Policy Interventions Per Year Note: GTA tracks policies by 225 countries and jurisdictions, this data is based on reporting-lag adjusted statistics. Source: Author’s creation based on the Global Trade Alert. 96 Global Trade Alert tracks policies by 225 countries and jurisdictions. It includes state interventions affecting trade in goods and services, foreign investment, and labor force migration: <https://www.globaltradealert.org/global_dynamics/area_all/year-to_2019/day-to_0531>. 81 Figure 3.5 shows the share of global trade that has been affected by protectionist measures around the world. In other words, it demonstrates the extent to which world exports have been affected by protectionist measures. In general, we can see an important growth of global trade distortions thanks to these protectionist restrictions. The only year in which we see a decrease in import distortions is 2015, which is probably explained by the weakness of trade during that year. This was due to a number of factors, including “an economic slowdown of China, a severe recession in Brazil, falling prices for oil and other commodities, and exchange rate volatility” (WTO, 2017, p. 16). Figure 3.5 – Share of World Imports Distorted by World Protectionist Restrictions Source: Author’s creation based on Global Trade Alert. Upon further examination of particular regions and countries, we can see that protectionism has been more concentrated in Asia and North America. Figure 3.6 shows the total number of protectionist and liberalizing measures by region. As we can see, in terms of number of measures implemented, Asian countries have been much more protectionist than the rest of the regions. However, it is interesting to note (see Figure 3.6) that when 82 considering the share of world exports affected by a region’s protectionism (the average from 2009 to 2018), we see a very different story. Examining the share of world exports to individual regions, North America has affected almost twice as many as Asia has (Figure 3.7). We learn at least two things from the following graphs: 1) the number of measures implemented is not necessarily correlated with the share of bilateral trade affected; and 2) many protectionist dynamics now can be related more to signals to defend rather than measures that substantially affect trade. Take Asia as an example. Although Asia would appear to be more protectionist based on the number of measures adopted, Asian countries do not affect world exports to the region in the same way that North American countries do. This raises the question of the extent to which interdependence factors such as global value chains (GVCs) and preferential trade agreements (PTAs) play a role here. For instance, countries more dependent on intermediate goods cannot implement as many protectionist measures as those less dependent. Figure 3.6 – Number of Protectionist and Liberalizing Measures by Region Source: Author’s creation based on the Global Trade Alert. 83 Figure 3.7 – Share of World Exports to Each Region Affected by Each Region Protectionism Measures (Average 2009-2018) Source: Author’s creation based on the Global Trade Alert Trade policy dynamics at the state level also reveal interesting patterns. China, for instance, ranks first in the G20 in terms of the number of protectionist measures implemented and second in the share of world exports to China that are affected. In comparison, the U.S. sits in fourth place for the number of protectionist measures adopted and is in sixth place on the share of bilateral trade (from the rest of the world) that is affected. In other words, we may say that the U.S. has emitted more protectionist signals than actual measures that affect bilateral trade flows. An extreme case is Argentina, which is one of the “least” protectionist country in terms of the number of measures implemented and one of the two “most” protectionist in terms of the share of bilateral trade (from the rest of the world) affected. This suggests that the “few” measures implemented have been very powerful. 84 Figure 3.8 – Number of Protectionist Measures of G20 Countries Toward the Rest of the World’s Imports (Average 2009-18, by Country) Source: Author’s creation based on the Global Trade Alert. In terms of the share of world exports affected by protectionism, Figure 3.9 shows that there is an important difference among the Latin American countries within the G20 (Argentina, Brazi, and Mexico): While Argentina is the most protectionist economy in the G20, Brazil is almost exactly in the middle of the list, and Mexico is among the least protectionist countries. 85 Figure 3.9 - Share of World Exports to G20 Countries Affected by Protectionism Measures implemented from the Rest of the World (Average 2009-18, by Country) Source: Author’s creation based on the Global Trade Alert. Finally, Figure 3.10 shows six different country cases with regard the share of a country’s exports to the world that is affected in t-1 (from the rest of the world’s protectionist measures), with the rest of the world (RoW)’s exports to the country affected in time t. As the different figures show, in most cases, there is a positive relationship between protectionist dynamics in t-1 (from the RoW toward specific country) and what happened in time t. Especially in the cases of China and the U.S., what happens in time t seems to be closely related with what happens in time t-1. Although these are just six examples, this is a trend that could portend some potential retaliatory dynamics, which is something that will be quantitively tested in the following chapter using a larger number of cases and years. 86 Figure 3.10 – How Responsive Were Key Emerging Economies to Trade Protectionism? CHINA U.S. ARGENTINA BRAZIL MEXICO INDIA Note: Share of Country’s Total Exports Affected (t-1) vs. Share of World Exports to the Country Affected (time t) by Protectionism Measures (Average 2009-2019). Source: Author’s creation based on the Global Trade Alert. 3.3.1.2 Industry and Sectoral Dynamics Globally, the sectors most affected by protectionist measures implemented between 2008 and 2018 are those within HS3 (Other transportable goods, except metal products, machinery and equipment); HS4 (Metal products, machinery and equipment); HS2 (Food products, beverages and tobacco); and HS0 (Agriculture, forestry and fishery products). Figure 3.11 shows the weight of each sector in the total measures implemented globally between 2008 87 and 2019 (using the sum of all measures implemented). These same sectors were also highly affected in terms of the share of trade flows. Figure 3.11 – Distribution of Protectionist Measures by Year and Sectors Affected Source: Author’s creation based on the Global Trade Alert. By region, we can identify a pattern similar to the global one, in which sectors 3 (Other transportable goods, except metal products, machinery and equipment); 4 (Metal products, machinery and equipment); and 2 (Food products, beverages and tobacco) were those most affected by global protectionism. Interestingly, when we analyze which sectors were most affected by protectionism within a given region, we can discern a response that relates to these same products. This pattern may change when testing for statistical patterns by year, which is analyzed in depth in the following chapter of this dissertation. A detailed sectoral analysis of the three Latin American countries of interest (Argentina, Brazil, Mexico) is presented in chapter 5. 88 Figure 3.12 – Most Affected Sectors by Region (2008-19) Most affected sectors in the region (by RoW’s protectionism) Most affected sectors by the region’s protectionism (to RoW’s exports to the region) LATIN AMERICA NORTH AMERICA ASIA EUROPEAN UNION Source: Author’s creation based on Global Trade Alert. 89 3.3.1.3 Types of Measures Implemented For a deeper understanding of the rise of protectionism post-GFC, I now turn to the type of protectionist measures that have been implemented globally, by region, and by country. As Figure 3.13 shows, the protectionist measures most used during the pre-GFC years (2000 to 2007) are technical barriers to trade and sanitary and phytosanitary measures, both of them considered (under the GTA classification) as transparent measures (see Table 1.5 in Chapter 1). 97 During and after the crisis (2008 to 2019), we can observe more diversification of measures implemented, where the protectionist measures most used are contingent trade- protective measures, subsidies, export-related measures, tariff measures, and non-automatic licensing. In terms of the level of transparency, we can see that pre-GFC (2000 to 2007), 78% of the measures globally have been transparent (e.g., tariff and sanitary and phytosanitary measures), while the other 22% consist of nontransparent measures (e.g., price controls). 98 These figures vary significantly after the GFC, when 58% of the measures globally have been transparent and the other 42% consist of nontransparent measures. Although the non- transparent measures are not the majority, they represent a significant portion of protectionist 97 Technical barriers to trade are “measures referring to technical regulations, and procedures for assessment of conformity with technical regulations and standard: Restrictions on toxins in children’s toys; Refrigerators need to carry a label indicating their size, weight and electricity consumption level.” For a detailed description of each type, see Table 1.4 in chapter 1. Also, for more information about each classification, see: <https://unctad.org/en/Pages/DITC/Trade-Analysis/Non-Tariff- Measures/NTMs-Classification.aspx>. 98 Data from UNCTAD’s Trade Analysis Information System (TRAINS), <http://i- tip.unctad.org/Default.aspx>. Given that the structure of the data of TRAINS differs from that in GTA, it is very challenging to conduct a detailed quantitative analysis of pre- and post-GFC years. In any case, this dissertation’s main focus is to understand protectionist patterns after the GFC and by no means does it pretend to claim that the findings of this analysis are generalizable to other economic periods. I discuss this in chapter 4. 90 measures between 2009 and 2019. Moreover, the significant increase of nontransparent measures after the crisis also shows an important trade policy substitution between NTMs and tariffs, a topic that, as recent studies have shown, has been under-researched (Beverelli et al., 2019; Niu et al., 2018). We know the more tariffs are lowered via binding rulings created by WTO rules, the more countries rely on NTMs, but this raises bigger questions about the relationship between trade policy and transparency (Kono, 2006). Figure 3.13 – Protectionist Policy Instruments Most Used Globally Pre- (2000-07) and Post- GFC (2008-19) by types of measure Pre-GFC (2000-07) Post-GFC (2008-19) Source: Author’s creation based on Global Trade Alert. Figure 3.14 – Protectionist Policy Instruments Most Used Globally, Pre- (2000-07) and Post- GFC (2008-19) by the level of transparency Pre-GFC (2000-07) Post-GFC (2008-19) Source: Author’s creation based UNCTAD database on Non-Tariff Measures, <http://trains.unctad.org/Forms/MeasureView.aspx?mode=modify&action=search>. 91 When we disaggregate by region, between 2008 and 2019, we can identify a pattern similar to the global one. The protectionist measures most used in Latin America were tariff measures and non-automatic licensing; 99 in North America, subsidies; in Asia, subsidies and non-automatic licensing; and in the European Union, subsidies (see Figure 1.3, in chapter 1). Figure 3.15 shows the protectionist instruments used by region and by level of transparency in -pre- (2000-07) and post-GFC (2008-19). Interestingly, for all the regions, we can see a notable increase of non-transparent measures after the GFC. In particular, the cases of North America, Asia, and the European Union show a notable variation in the share of measures consisting of non-transparent barriers before and after the GFC. Figure 3.15 – Protectionist Instruments by Region Pre- (2000-07) and Post-GFC (2008-19) by the Level of Transparency Source: Author’s creation based on Global Trade Alert and UNCTAD NTM data. 99 Non-automatic licensing is “control measures generally aimed at restraining the quantity of goods that be imported. Examples are only hotels and restaurants are allowed to import alcoholic drink; A quota of 100 tons of tuna fish can be imported any time of the year.” See <https://unctad.org/en/Pages/DITC/Trade-Analysis/Non-Tariff-Measures/NTMs- Classification.aspx>. 92 When we consider the level of transparency for the six G20 economies depicted in Figure 3.16, we can also see important changes between pre- (2000-07) and post-GFC (2008- 19) periods. In all cases, the share of non-transparency trade policy instruments increases substantially after the GFC. Between 2008 and 2019 (post-crisis period), we can see that Brazil, China, India, and the U.S. have relied disproportionally on non-transparent trade policy instruments. In the case of Argentina and Mexico, the majority of policy instruments were transparent (e.g., tariff). Nevertheless, in both cases, a notable proportion of the measures were non-transparent––38% in Argentina and 34% in Mexico. Figure 3.16 – Protectionist Instruments by Country Pre- (2000-07) and Post-GFC (2008-19) by the Level of Transparency Source: Author’s creation based on UNCTAD Global Database on Non-Tariff Measures (for pr-GFC, <https://trains.unctad.org/>) and Global Trade Alert (for post-GFC, <https://globaltradealert.org>). A Descriptive Text Analysis of the Types of Measures Implemented Another way to understand the level of transparency of a given trade policy instrument is to assess just how explicit the initiating country is when formally implementing it. As 93 mentioned above, when WTO countries implement a specific trade measure, they are legally obligated to publicize it on a government website or through the press (see Figure 3.2 for an example). Technically, the WTO requires that countries publicize a new trade measure before it can be enforced by customs. In any case, governments may want to inform citizens of a new policy measure that protects domestic goods from foreign competition (Hollyer et al., 2014). Although governments may want to send economic signals to domestic lobbies, they may also want to keep their policies below the radar of foreign countries. If a foreign trade partner is implementing a protectionist measure against a country, a government may implement punitive policies in retaliation. As a result, the way countries communicate their trade policy to each other is relevant and should be considered. One way to do this is to see whether a country mentions a specific trade partner that will be affected by the measure. It could be that instead of implementing a measure against a specific trade partner, countries implement more general measures by only identifying which goods will be protected, rather than which countries will be affected. Note the difference between a measure that states there are “tariff increase on some articles of leather” 100 (which affects all leather exporters to the implementing country) and a measure that states there is “an increase in the fee for registering vehicles imported from the United States.” 101 The former is a general trade policy measure that does not specify a specific country; the latter is specific measure toward a specific trade partner. 100 For instance, in January 2010, Mercosur members announced a change in import duties that especially affect articles of leather. , See Mercosur, CMC (Consejo del Mercado Común) Decision no. 27/09 of 7 Dec. 2009: <http://200.40.51.218/SAM%5CGestDoc%5CPubWeb.nsf/B6BA5D474C5B065A032585BC000B9F 98/$File/DEC_027-2009_ES_Confecciones.pdf>. 101 For instance, in 2016, the Canadian government raised the fees of vehicles imported specifically from the U.S. See <https://www.cbsa-asfc.gc.ca/publications/cn-ad/cn16-30-eng.html>. 94 These dynamics are illustrated in Figure 3.17, wherein I take all the official documents in which countries announce the implementation of a specific trade measure and download them from the GTA’s repository of State Acts (https://www.globaltradealert.org/latest/state-acts). I then use the stringr package in R to study different patterns of the text of those official trade policy documents. I pay special attention to documents that mention a country (or group of countries) that will be affected by the measure officials are publishing. As we can see, in Figure 3.17, an average of 8% of measures implemented between 2009 and 2019 actually name the country affected in the official document. This suggests that most of the measures have centered on the product they are protecting and not the country that is affected. The fact that most of the measures implemented between 2009 and 2019 were “general trade policy” doesn’t mean that many of them were retaliatory measures. It could be that countries implement a retaliatory measure without specifying the name of the trading partner; instead, they specify a specific product that could be imported from the country against which they wish to retaliate. This is something I discussed in Chapter 1 when talking about the complex task of defining policy responses. 102 102 In appendix 3 of this dissertation, I create a world cloud of the most used words in official documents announcing trade policy instruments between 2009 and 2019. The word cloud keeps track of words that are mentioned frequently (more than 10 times) in official documents. 95 Figure 3.17 – Protectionist Policy Instruments with Name of Targeted Country (2009-19) Source: Author’s creation based on the Global Trade Alert. 3.4 A Descriptive Analysis of Preferential Trade Agreements and Global Value Chains Two other variables of interest here are GVCs and PTAs. While I briefly referred to the relevance of these two phenomena in the global economy in Chapter 1, I will now analyze them in more detail. 3.3.1 Preferential Trade Agreements As of June 2020, there were 303 PTAs that have been formally notified to the WTO. Of these, 94% are flexible integration agreements (e.g., Free Trade Agreements, Partial Scope Agreement), and the remaining 6% are considered too deep to be integration agreements (e.g., Customs Union). 103 Figure 3.19 offers a global view of those PTAs signed up until 103 By flexible integration agreements, I mean those cases in which there is a bilateral compromise between countries, which means that the members of such an agreement are free to maintain an independent trade policy with third countries. This is not the case with a Customs Union, under which 96 2019, as well as the number of agreements signed by each country. As the figure shows, Europe and Latin America are the two regions with the most trade agreements signed, both within these respective regions and with the rest of the world. Figure 3.18 – Preferential Trade Agreements Signed Until 2019 Source: Design of Trade Agreements (DESTA) Database, <https://www.designoftradeagreements.org/>. Turning to those trade agreements (goods and services) signed by the G20 countries, we can see that, apart from the EU bloc, Mexico and Turkey are the two countries with the most trade agreements in force. China, Canada, and the U.S. are in the middle of the list, while Brazil and Argentina are among those countries with fewer than ten trade agreements. members are obligated to negotiate together with third countries that are not part of the Union. For more information, see WTO, Regional Trade Agreements Database, <http://rtais.wto.org/UI/PublicMaintainRTAHome.aspx>. 97 Figure 3.19 – Number of PTAs of G20 Countries Until 1970-2019 Source: Author’s creation based on the WTO´s Regional Trade Agreements Database, <http://rtais.wto.org/>. 3.3.2 Global Value Chains As mentioned before, GVCs are production networks that gather products using parts (semi- finished goods) from around the globe. Today, around 80 percent of world trade is driven by these production networks (Kolb, 2019). There is probably no better way to describe a GVC than by drawing upon real-life examples. A recent World Bank (2020b) report on global value chains highlights that the bicycle industry offers one of the best examples for understanding the nature of GVCs. Global bicycle production grew from about 10 million units in 1950 to more than 130 million units today. Bicycles are assembled using semi- finished components from all over the world. Take the example of Bianchi bicycle model: design, prototyping, and conception work happen in Italy, and assembly takes place in Taiwan, using parts and components from China, Italy, Japan, and Malaysia. Shimano of Japan, for example, makes brakes for the Bianchi, while the handlebars are made in China (World Bank, 2020b, pp. 16-17). Figure 3.21 depicts this sophisticated GVC for a bicycle. 98 Figure 3.20 –Example of the Global Value Chain for a Bicycle Source: World Bank. 2020. World Development Report 2020 (p. 16), using data from UN Comtrade database. Using the latest data, it has been found that “the overall share of GVC trade in total world trade … grew significantly in the 1990s and early 2000s, but it appears to have stagnated or even declined in the last 10 years. Still, about half of world trade appears to be related to GVCs” (World Bank, 2020b, p. 40). As Figure 3.21 shows, the GFC had an important impact in that it reduced the share of total world trade conducted within GVCs from 52% in 2007 to 43% in 2016 (which is the last year of aggregated data available on GVCs). 99 Figure 3.21– GVC Share of Global Trade (%) Source: World Bank. 2020. World Development Report 2020 (p. 19), using data from Eora26 database; Borin and Mancini (2015, 2019); and Johnson and Noguera (2017). It is important to understand that although GVCs span the globe, countries’ participation in them is uneven when we consider the types of exports that each country contributes to different GVCs (Figure 3.22). According to the World Bank (2020b, p. 21): Some countries export raw materials for further processing; others import inputs for assembly and exports; and still others produce complex goods and services. In addition, some are heavily reliant on GVCs for trade, whereas others export largely domestic goods for consumption. To capture these distinct features of participation, countries are classified into four main types—commodities, limited manufacturing, advanced manufacturing and services, and innovative activities— based on the products they export and their participation in GVCs. 104 104 This taxonomy is based on the following criteria by the World Bank (2020b, pp. 22-23): “Countries are classified based on (1) the goods and services exported, (2) the extent of GVC participation, and (3) measures of innovation. A country’s sectoral specialization of exports is based on the domestic value added in gross exports of primary goods, manufacturing, and business services. A country’s extent of GVC participation is measured as backward integration of the manufacturing sector as a share of the country’s total exports. Higher backward integration in manufacturing is an important characteristic of countries entering or specialized in non-commodity GVCs. Two measures are used to capture a country’s innovative activities: (1) intellectual property (IP) receipts as a percentage of GDP and (2) research and development (R&D) intensity, defined as its expenditure of public and private R&D as a percentage of GDP.” 100 Figure 3.22 – All Countries Participate in GVCs—But Not in the Same Way Source: World Bank. 2020. World Development Report 2020 (p. 21), based on the GVC taxonomy for 2015. When we study GVC participation in more detail, the World Bank (2020b) has identified a variety of scenarios according to different regions. While Europe is seeing an increase in regional fragmentation, Asia is becoming internationally fragmented but regionally integrated, and Latin America is becoming more integrated across the board. More specifically, the World Bank (2020b, p. 24) explains that: In Europe, regional fragmentation of value chains increased through successive rounds of enlargement in which Eastern European countries, including Bulgaria, Hungary, and Poland, progressively joined older members’ production networks … global fragmentation was equally important, driven mostly by the larger European economies such as France, Germany, and the United Kingdom, whose linkages with countries in Asia such as China or India expanded. In East Asia, linkages are more regional than global and GVCs became more internationally fragmented after 1990 because of both regional and global fragmentation in the 1990s and 2000s, although regional integration dominated. … The NAFTA GVCs depend somewhat more on global partners than regional partners, and integration has been increasing on both fronts. … In Latin America and the Caribbean, value chains are more globally linked, but they have increased both regionally and globally. 101 Figure 3.24 shows how global production networks are organized around three major regions (East Asia and Pacific, Europe, and North America). Figure 3.23 – Global Production Networks and the Share of Value Added in Exports of Each Region by Source Region (2018) Source: World Bank. 2020. World Development Report 2020 (p. 25) based on data from full Eora database (latest year for which data are available is 2018). 3.4 Final Thoughts on Rising Protectionism vis-à-vis the COVID-19 Pandemic Although this dissertation concentrates primarily on understanding trade protectionism between 2008 and 2019 (in part, due to the amount of data available on the variables of interest), at the moment that I am writing this dissertation, we are in the midst of a global health crisis. The COVID-19 pandemic, which will have an unprecedented impact on the global economy and multilateralism, is causing even more uncertainty in international markets (Baker et al., 2020; Albertoni & Wise, 2020; Pinna & Lodi, 2021). We still have limited data on the trade and economic impact of this pandemic; however, it is important to 102 consider this shock at least as part of this descriptive analysis. After all, the goals of this dissertation are to capture and explain global trade policy dynamics within a highly uncertain context in the world economy. Thus, it is relevant to consider how countries are responding to the COVID-19 pandemic. As the World Bank (2020) states in its first Global Economic Prospects after the start of the pandemic: The COVID-19 pandemic is expected to result in a 5.2 percent contraction in global GDP in 2020—the deepest global recession in eight decades, despite unprecedented policy support. Per capita incomes in the vast majority of emerging market and developing economies (EMDEs) are expected to shrink this year, tipping many millions back into poverty. The pandemic highlights the urgent need for health and economic policy action, including global cooperation, to cushion its consequences, protect vulnerable populations, and improve countries’ capacity to prevent and cope with similar events in the future. Once the health crisis abates, structural reforms that enable strong and sustainable growth will be needed to attenuate the lasting effect of the pandemic on potential output. From the side of the WTO (2020, p. 1) there has been a special concern on the inevitable negative impact the COVID-19 pandemic has already on global trade, which is expected to fall between 13% and 32% in 2020 because of the pandemic’s disruption of “normal economic activity and life around the world” (Figure 3.24). 103 Figure 3.24 – WTO forecast: The Greater Trade Collapse of 2020 After COVID-19 Source: WTO (2020). Notes: Trade volumes are an average of exports and imports; Figures for 2020 and 2021 are projections. (2015 = 100) The WTO (2020) highlights that one of the main challenges of this pandemic is that it happens in a world interconnected by “complex value chains,” which can be severely affected in this context of high economic uncertainty. More specifically, the WTO (2020, pp. 1-4) predicts: Trade will likely fall steeper in sectors with complex value chains, particularly electronics and automotive products. … According to the OECD Trade in Value Added (TIVA) database, the share of foreign value added in electronics exports was around 10% for the United States, 25% for China, more than 30% for Korea, greater than 40% for Singapore and more than 50% for Mexico, Malaysia and Vietnam. Imports of key production inputs are likely to be interrupted by social distancing, which caused factories to temporarily close in China and which is now happening in Europe and North America. However, it is also useful to recall that complex supply chain disruption can occur as a result of localized disasters such hurricanes, tsunamis, and other economic disruptions. Managing supply chain disruption is a challenge for both global and local enterprises and requires a risk-versus-economic efficiency calculation on the part of every company. 104 Indeed, the impact that the current pandemic could have on global trade that might even surpass the damages inflicted by the 2008-09 GFC. This is why current studies already suggest that we could be looking at a “Greater Trade Collapse” (the GFC on trade was called the “Great Trade Collapse”) (Baldwin & Evenett, 2020, p. 3). Recent studies have shown that as the pandemic has spread across the globe, governments all around the world responded “with a chain reaction of unprecedented trade policy measures” (Joller & Kniahin, 2020, p. 1). Around the world both import and export measures were quickly implemented to fend off a total collapse in domestic economies. Figure 3.26 shows the proliferation of such measures between April and June 2020. 105 Figure 3.25 – Unprecedented Trade Policy Response to COVID-19 Source: ITC Market Access Map COVID-19 page (as of 15 June 2020), <https://www.macmap.org/covid19>. The reason why the collapse would be more catastrophic than during the GFC is twofold. For starters, the COVID-19 pandemic impacts every country, whereas the GFC 105 For more information about trade policy measures generated in the context of the pandemic, see the daily tracker on the ITC Market Access Map: <https://www.macmap.org/covid19>. 105 primarily affected the U.S. and the UK (Baldwin, 2020). In addition, the present crisis is hitting both demand and supply, severely limiting the options for remedial economic policies. As Baldwin and Evenett (2020, p. 4) explain: While the point-of-impact of the 2008 financial crisis was the U.S. and the UK, today’s crisis hit all the world’s largest trading nations within a few months. The US, China, Japan, Germany, Britain, France, and Italy ––all of which were hit hard by the virus in the first quarter–– account for 60% of world supply and demand (GDP), 65% of world manufacturing, and almost as much of world manufacturing exports. While the Great Trade Collapse was primarily caused by a collapse in demand, today’s “Great Lockdown,” as the IMF is calling it, is a serious supply-side disruption that is affecting all sectors in all of the largest economies in the world. As in 2008, today’s trade shock has been accompanied by rising concerns about a return to protectionism. 106 The COVID-19 pandemic has stopped globalization in its tracks. Understanding what happened in the decade after the GFC will help us forecast what could happen in the post- pandemic era. These extraordinary circumstances offer a natural experiment on the effect of drastically slowed trade on the global economy. As this descriptive chapter shows, one of the contributions of this dissertation is the empirical documentation of the damage that closed markets and trade nationalism can inflict upon the world economy. In essence, this dissertation is a cautionary tale about trade nationalism and the merits of a more open global trade regime. My empirical documentation of the damage currently at play is meant to alert us to what lies ahead in the near future and to better prepare us to cope with the challenges now faced by the global trade regime. 106 Baldwin (2020) “The Greater Trade Collapse of 2020”, VoxEU.org, 07 April 2020. Retrieved from: <https://voxeu.org/article/greater-trade-collapse-2020>. 106 CHAPTER 4 – COUNTRY AND SECTORAL ANALYSIS ON PROTECTIONISM AND TRADE POLICY RESPONSES: A QUANTITATIVE APPROACH “Not everything that counts can be counted, and not everything that can be counted counts.” WILLIAM BRUCE CAMERON, INFORMAL SOCIOLOGY, 1963 4.1 Introduction To test the hypotheses of this dissertation, a mixed-method approach is used––blending both quantitative and qualitative techniques of analysis. 107 This chapter focuses on the quantitative portion of the study, where large-N statistical analysis is used to present the reader with a “big picture” idea of what the main patterns of protectionism have looked like over the last 107 For this chapter, I received excellent comments and suggestions from Joshua Aizenman and Jeff Nugent, and Ratika Narag, Geert Ridder, and Hyungsik Roger Moon from the USC Economics Department. My colleague and friend Timo Dahler gave me very helpful comments too. Earlier versions were presented at courses and seminars in the USC Economics Department where I received useful comments on the models and analysis presented here. I also want to thank George Gerald Vega Yon from USC’s Center for Applied Network Analysis (CANA) for their suggestions on earlier versions of this chapter. George has been an immense support in this journey. I will always be extremely grateful for his support with the data analysis and the dataset in which worked together that allowed to improve this chapter a lot. I am also grateful for the excellent research assistance from USC’s Security and Political Economy (SPEC) Lab’s trade policy team. I also thank the National Science Foundation (NSF) for the fellowship (1558713) to support my participation in the 11th APSA Networks Conference, where I also received useful comments for the improvement of this chapter. Lastly, I want to thank the Global Trade Alert Dataset team, especially Piotr Lukaszuk, for his generosity to explain to me many key details about the dataset. 107 decade. In this analysis, I include more than 170 countries from regions across the world, which represents, on average, 98% of global trade flows (Evenett & Fritz, 2020). These quantitative insights will serve as a baseline for Chapter 5, where an in-depth qualitative analysis of three country cases––Argentina, Brazil, and Mexico––is conducted. These case studies serve two purposes: first, they allow for the assessment of the validity of my findings from the large-N analysis undertaken in this chapter; and second, they give rise to potentially new and more detailed questions about protectionist responses that can best be identified at the country level. Before delving into the details, I reiterate the main goals of this dissertation. In analyzing the explosion of trade protectionism since the 2008-09 global financial crisis, I have uncovered the role that preferential trade agreements (PTAs) and global value chains (GVCs) have come to play in the expansion of non-tariff barriers in particular. Ultimately, I argue that PTAs and GVCs, in the current context of high economic uncertainty, have had counterintuitive and contradictory effects on global trade policies. It is true that PTAs and GVCs have fostered healthy patterns of trade interdependences. Both––especially PTAs–– have been a force for trade liberalization the world over, and common wisdom has held that the more countries depend on one another, the less likely that protectionist dynamics will arise between them (Baldwin, 2012; Gawande et al., 2015; Jensen et al., 2015; Lamy, 2013). However, a newer, contradictory phenomenon has emerged since the GFC, one in which trade protectionism occurs within the very institutions that have been designed to mitigate it. This phenomenon has barely been discussed or studied (Ahir et al., 2018). 108 For the remainder of this chapter, various hypotheses will be tested relating to the insertion of non-tariff barriers within PTAs and GVCs, and they will seek to explain how economic uncertainty exacerbates this phenomenon. This is approached through a large-N statistical analysis constructed with directed dyad-year data at the state and industry level. This chapter is divided into three main sections. First, I analyze my general hypotheses at the state level and probe how countries respond when they are adversely affected by a protectionist measure adopted by another country. At the same time, I ask whether the existence of PTAs and GVCs decreases the likelihood of a protectionist response. In other words, the existence of PTAs and GVCs–– ceteris paribus––lower the probability or severity of bilateral protectionist dynamics: HYPOTHESIS 1 – Direct Retaliation: Countries will retaliate against those nations that initially target them with a protectionist measure. HYPOTHESIS 2 – Conditional Effect of PTAs: PTAs are negatively correlated with protectionist responses. HYPOTHESIS 3 – Conditional Effect of GVCs: GVCs are negatively correlated with protectionist responses. Second, I focus on the relationship between tariff and non-tariff measures (NTMs) within PTAs and GVCs. Specifically, I look at whether countries with high trade interconnectivity (i.e., the existence of PTAs and GVCs) will retaliate more with less transparent (NT) measures than with transparent (T) measures. HYPOTHESIS 4 – Conditional Effect of PTAs: PTAs are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses. HYPOTHESIS 5 – Conditional Effect of GVCs: GVCs are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses. 109 Third, I concentrate on sectoral dynamics and how these affect the way countries respond to protection: HYPOTHESIS 6 – Cross-Sector Retaliation: Protectionist retaliation is more likely to be cross-sectoral, rather than within-sector. HYPOTHESIS 7 – Within-Sector Retaliation – industrialization: The higher the level of industrialization of the sector, the lower the within-sector retaliation. HYPOTHESIS 8 – Within-Sector Retaliation – intermediate goods: Protectionist responses are less likely with trade in intermediate goods. Again, the different levels of quantitative analysis will be conducted using directed dyad- year data at the state, industry, and sectoral levels. The unit of analysis is dyad countries by year. For both the state- and industry-level hypotheses, a lagged model is used because it does not assume contemporaneous influence. Rather, it considers the time that the focal country may take—e.g., one year—before it decides to retaliate with protectionist policies that its trading partner might levy against it. 108 4.2 The Models First, to test the hypotheses concerning the likelihood of responses (Hypothesis 1), the conditional effects of PTAs and GVCs in that response (Hypothesis 2, Hypothesis 3), and the difference between the likelihood of tariff and NTMs responses (Hypothesis 4, Hypothesis 5), the models can be stated as: 108 Data from Global Trade Alert shows that, in general, it takes time (sometimes many months) for governments to formally implement trade policy measures. Moreover, it shows how this time has quickened since the last GFC. However, we can see that to collect an important threshold of data (around 1000 policy interventions), they have taken a minimum of 187 days in the past four years. This may suggest that most of the recent protectionist dynamics are more political threats than real measures. All this is to say that a period of one year is the time in which retaliation dynamics move beyond mere anecdotal evidence (see Figure 4.1 in the Appendix of this dissertation). 110 The main difference between these is the dependent variable. YABall,t represents all protectionist restrictions by country A on country B at time t, which is measured by the share of bilateral trade affected by protectionism measures in a given year. YABtariff,t represents protectionist tariff restrictions by country A on country B at time t, and YABNTM,t represents protectionist non-tariff restrictions by country A on country B at time t. For the three equations, YBAall,t-1 represents the protectionist restrictions by country B on country A at time t-1; PTAAB,t represents the number of trade agreements in place between A and B; and GVCAB,t represents whether trade between A and B is interconnected through a global value chain. XAB,t are control variables consisting of political and economic data regarding A and B such as population weighted distance, normally used for gravity models. The share of trade positively affected is shown by liberalizing measures by B on A in time t-1, bilateral investment treaty between A and B, relative economic difference (B’s GDPpc/A’s GDPpc), common colonizer, common official language, among others control measures; whereby α is the intercept and ε is an error term. Second, for the sectoral-level analysis, I use a similar model to that used at the state level, but now the policy itself becomes the main unit of analysis. Given that we need to test cross-sector retaliation and within-sector retaliatory dynamics, the main variables of analysis 111 now take the form of binary measures. The analysis at the sectoral level will be developed as follows: 1) I consider different scenarios (one for each industrial section described in the UN’s Central Product Classification or BEC classification). 2) I consider each of these sectors as a separate independent variable to see how it correlates with retaliatory dynamics toward the same and other sectors. 3) The following variables show an example for sector 0 (Agriculture, forestry and fishery products): Dependent Variable: Import restrictions over sector 0 from B to A in time t; Independent Variable: Import restrictions over sectors 0 through 9 from A to B in time t-1. Hence, to test the hypotheses concerning the likelihood of retaliation at the cross- sector versus within-sector level (Hypothesis 6), the effect of different levels of industrialization of the sector in within-sector retaliation (Hypothesis 7), and the effect of intermediate goods in retaliation (Hypothesis 8), the models can be stated as: Where YABsec,t represents all protectionist restrictions by country A on country B on a specific sector at time t; YBAsec,t-1 represents all the protectionist restrictions by country B on country A on a specific sector at time t-1; XAB,t are control variables consisting of political and economic data regarding A and B mentioned above, whereby α is the intercept and ε is an error term. 4.3 The Data In Chapter 3, I introduced the structure of different datasets considered in this dissertation and described the different patterns these reflect at the state and sectoral levels of analysis. In this section, I concentrate on how the different datasets used for this dissertation are related to the specific variables of interest. 112 Keep in mind that this dissertation is based on several datasets that have been retrieved from public repositories. The four main data sources were 1) Global Trade Alert (GTA), which provides information on state-level trade policy interventions taken since 2008 that are likely to affect foreign trade; 109 2) BACI – CEPII trade data, which disaggregate data on bilateral trade flows for more than 5,000 products and 200 countries; 110 3) the International Political Economy Data Resource (IPE), which compiles data from 89 IPE data sources into a single dataset (Graham & Tucker, 2017); 111 and 4) data from the Harmonized System (HS) classification of products, which classifies “traded goods on a common basis for customs purposes.” 112 4.3.1 Measurement of the Dependent Variables As mentioned above, the main dataset used for my models derives from GTA, which provides information on state-to-state—dyad-year—interventions undertaken since 2008 and are most likely to affect foreign trade. This includes “state interventions affecting trade in goods and services, foreign investment and labor force migration.” 113 GTA was launched in 2009 when it was “feared that the global financial crisis would lead governments to adopt widespread 1930s-style beggar-thy-neighbor policies” (Evenett & Fritz, 2020, p. 1). Since it was launched, many studies have relied on GTA data (Evenett et al., 2011; Evenett et al., 2011; Boffa & Olarreaga, 2012; Henn & McDonald, 2014; Georgiadis & Gräb, 2016; 109 For more details about GTA, visit <https://www.globaltradealert.org/>. 110 For more details about BACI – CEPII trade data, visit <http://www.cepii.fr/CEPII/en/bdd_modele/bdd_modele.asp>. 111 For more details about the IPE Data Resource, visit <https://doi.org/10.7910/DVN/X093TV>. 112 For more details about GTA, visit <https://www.globaltradealert.org/>; For the Harmonized System (HS), visit <https://unstats.un.org/unsd/tradekb/Knowledgebase/Harmonized-Commodity- Description-and-Coding-Systems-HS>. 113 Global Trade Alert, <https://www.globaltradealert.org/>. 113 Evenett, 2019). It is relevant to note that GTA documents both trade liberalization and protectionist measures, which lets us measure both sides of trade policy mechanisms and offers control variables to test the real weight of both types of measures when we consider them as explanatory variables. Based on GTA, in order to measure the dependent variable for the state-level analysis, the following possible outcomes are used: DV1: All protectionist measures by A on B in time t—the share of bilateral trade affected (Hypothesis 1, Hypothesis 2, and Hypothesis 3); DV2a & DV2b: Number of Transparent and Non-Transparent protectionist measures by A on B in time t—continuous variable (Hypothesis 4 and Hypothesis 5); DV3a & DV3b: Proportion of Transparent and Non-Transparent protectionist measures in total measures by A on B in time t—percentage (Hypothesis 4 and Hypothesis 5). To measure the dependent variable at the sector-level analysis, the following outcome is used: DV4: Share of bilateral trade affecting capital goods, intermediate goods and consumption goods—for BEC classification; and sec0-sec9, for HS classification. More specifically, for the sector analysis, two different variables are used to test for cross sectoral retaliation. First, the BEC classification is relied on considering the share of trade between A and B affecting capital, consumption, and intermediate goods by measures implemented by A toward B in time t, as the dependent variable; and the share of trade between A and B affecting capital, consumption, and intermediate goods by measures implemented by B toward A in time t-1 as the independent variable. Second, the HS classification is relied on and a dummy variable is created for every time country A 114 implemented a measure toward B in each sector (sec0 – sec9) in time t, as the dependent variable; and every time country B implemented a measure toward A in each sector (sec0 – sec9) in time t-1, as the independent variable. In Figure 4.1, Figure 1.3—at the state level—is replicated to show how the main model looks at the sectoral level of analysis. Figure 4.1 –Modeling Retaliatory Dynamics by Sector Source: Author’s creation. 4.3.2 Measurement of Regressors of Interest The measure of PTAs is based on Dür et al. (2014), which contains bilateral data on PTAs from 1948 to 2018 for 203 countries. PTAs have expanded in number and have diversified in content. Around PTAs had been negotiated by the 1990s; 303 had been negotiated as of January 2020. 114 Following the World Bank’s definition, Dür et al. (2014) provides a more general description of international trade agreements as “all the agreements that have the 114 WTO, Regional Trade Agreements, <https://www.wto.org/english/tratop_e/region_e/region_e.htm>. 115 potential to liberalize trade” (p. 1). 115 This definition will be used in this dissertation, as it is one of the most widely used definitions in the literature (Hofmann et al., 2019; Ruta, 2017). As said, when referring to a PTA, this includes free trade agreements, customs unions, or partial free trade agreements, which can take bilateral (two countries), plurilateral (many countries), plurilateral with a third country (e.g., EU-Australia Trade Agreement), or region- region (e.g., EU-Mercosur Trade Agreement), forms. The diversity of PTA classifications complicates the inherent meaning of PTA, since a PTA could be one of many trade agreements. As mentioned above, as of June 2020, there were 303 PTAs that had been formally notified to the WTO. Of these, 94% are flexible integration agreements (e.g., Free Trade Agreements or Partial Scope Agreements), and the remaining 6% are considered too be deep integration agreements (e.g., Customs Union). This shows that when talking about PTAs, in reality we are talking about FTAs as well. The measure of GVCs is based on the combination of Global Trade Alert data and BACI-CEPII trade data, which contains estimates of the composition of trade between countries. As Ruta (2017, p. 175) says, “GVCs [have] radically changed international trade.” However, as said before, it is still very challenging to capture bilateral GVC dynamics. One of the better––although imperfect—proxies we have today is the share of intermediate goods between two countries. Both GTA and BACI-CEPII let us capture these dynamics. For each pair of countries A and B—exporter and importer—the data includes the product, HS 2012 code, with an estimate of how much that product accounts for total exports from exporter A to importer 115 For other studies on why the term PTA is preferred to the term RTA see, for instance, Hofmann et al. (2019); and Ruta (2017). 116 B. 116 As mentioned, the classification by Broad Economic Categories (BEC, para. 1) is also used, which provides “a means for international trade statistics to be analyzed by broad economic categories such as food, industrial supplies, capital equipment, consumer durables and consumer non-durables.” 117 Hence, the main variable obtained from BACI-CEPII trade data is a percentage that estimates the proportion that a certain product contributes to total exports from country A to country B. Then, combining this percentage with the data provided by GTA, whose product is affected by each protectionist measure implemented from A to B, we can obtain the percentage of trade affected between A and B. 118 Finally, to avoid endogeneity issues, these shares of intermediate goods are calculated using pre-crisis, 2005- 2007, trade values. 4.3.3 Control Variables The control variables include trade liberalizing measures which are the share of exports positively affected by liberalizing measures and the Bilateral Investment Treaties (BITs) 116 The raw data is publicly available at <http://www.cepii.fr>. BACI-CEPII uses HS 2002 version, so I merged it with the equivalent code in HS 2012 version. See: UN Statistics Division, Correlation and Conversion Tables Used in UN Comtrade, <https://unstats.un.org/unsd/trade/classifications/correspondence-tables.asp>. 117 UN Statistics Division, Classification by Broad Economic Categories, < https://unstats.un.org/unsd/tradekb/knowledgebase/50089/classification-by-broad-economic- categories-rev4>. See Table 1.6 in Chapter 1. It shows the three basic classes of goods categorized in the BEC. As explained above, a concept normally related to intermediate goods is “intra-industry trade” (IIT), which is a more general way of defining trade within similar industries. The OECD (2005) defines IIT as “the trade in similar products—horizontal trade” with differentiated varieties, e.g. cars of a similar class and price range, and trade in “vertically differentiated” products distinguished by quality and price, e.g. exports of high-quality clothing and imports of lower-quality clothing. There could be Intra-Industry Trade in both intermediate and final goods. While IIT can applied to different stages of a good, intermediate goods are normally considered to measure GVCs. See Lüthje (2001); and Peterson and Thies (2015) for studies on intra-industry trade (IIT) and intermediate goods. 118 Based on these new measures, further studies can apply this to the share of trade between A and B of total trade of A, to get a “back-of-the-envelope” calculation for the income effects of retaliation. This will be further discussed later. 117 between A and B, 119 from the IPE Data Resource, the Relative Economic Difference—B’s log GDPpc / A’s log GDPpc. 120 They additionally draw from the GeoDist Database CEPII, which uses the population-weighted distance that is typically used in gravity models, common colonizer, and common official language. 121 The following tables show the summary statistics of all the variables considered for the analysis. 122 119 UNCTAD (2018) International Investment Agreements Database, <http://investmentpolicyhub.unctad.org/IIA/IiasByCountry#iiaInnerMenu>. 120 The International Political Economy Data Resource (IPE) compiles data from 89 IPE data sources into a single dataset (Graham & Tucker, 2017). See <https://doi.org/10.7910/DVN/X093TV>. 121 Mayer, T. & Zignago, S. (2011) Notes on CEPII’s distances measures: the GeoDist Database CEPII Working Paper 2011-25, <http://www.cepii.fr/CEPII/en/bdd_modele/presentation.asp?id=6>. 122 Given the high concentration in zero values in the main variables of interest, I use log(x+1) transformation, which is widely used among data scientists (Bellégo & Pape, 2019). 118 Table 4.1 – Summary Statistics, State-Level Analysis VARIABLE NAME MIN. MEAN MAX. SHARE OF TRADE AFFECTED BY MEASURES All protectionist measures by A on B in time t (share of bilateral trade affected, log) 0 0.192 0.693 All protectionist measures by B on A in time t-1 (share of bilateral trade affected, log) 0 0.106 0.693 Non-transparent (non–tariff) protectionist measures by A on B in time t (share of bilateral trade affected) 0 0.145 0.693 Transparent (tariff) protectionist measures by A on B in time t (share of bilateral trade affected) 0 0.085 0.693 N O OF ALL PROTECTIONIST MEASURES N o of all protectionist measures by A on B in time t 0 5.43 1661 N o of all protectionist measures by B on A in time t-1 0 4.87 1661 N o of Non-transparent (non–tariff) protectionist measures by A on B in time t 0 5.15 1660 N o of Transparent (tariff) protectionist measures by A on B in time t 0 0.47 74 REGRESSORS OF INTEREST Preferential trade agreement (PTA) A and B 0 0.006 2 GVC proxy variable: Percentage of intermediate goods in total trade between A and B 0 0.007 0.57 CONTROLS Liberalizing measures by B on A in time t-1 0 0.99 79 Bilateral Investment Treaty between A and B 0 0.004 1 Population-weighted distance (gravity model) 35 6449 19649 Relative Economic Difference (B’s GDPpc / A’s GDPpc) 0.51 0.70 0.94 Common colonizer 0 0.05 1 Common official language 0 0.13 1 Source: Author’s creation. 119 Table 4.2 – Summary Statistics, Sectoral-Level Analysis (% of Trade Affected by Sector) VARIABLE NAME MIN MEAN MAX. BEC CLASSIFICATION (1-Capital goods; 2-Consumption goods; 3- Intermediate goods) Continuous Variable 1-Capital goods, A to B, time t 0 0.0003 0.083 2-Consumption goods, A to B, time t 0 0.0008 0.080 3-Intermediate goods, A to B, time t 0 0.0026 0.083 Capital goods, B to A, time t-1 0 0.0006 0.002 Consumption goods, B to A, time t-1 0 0.0002 0.007 Intermediate goods, B to A, time t-1 0 0.0006 0.008 HS CLASSIFICATION (SEC 0 – SEC 9) Dichotomous Variable Sec. 0 - Agriculture, A to B, time t 0 0.0196 1 Sec. 1 - Ores and minerals, A to B, time t 0 0.0865 1 Sec. 2 - Food products, A to B, time t 0 0.0942 1 Sec. 3 - Other transportable goods, A to B, time t 0 0.0931 1 Sec. 4 - Metal products, A to B, time t 0 0.0564 1 Sec. 5 - Constructions services, A to B, time t 0 0.0369 1 Sec. 6 - Distributive trade services, A to B, time t 0 0.0417 1 Sec. 7 - Financial and related services, A to B, time t 0 0.0749 1 Sec. 8 - Business and product. services, A to B, time t 0 0.1049 1 Sec. 9 - Community, A to B, time t 0 0.0570 1 Sec. 0 - B to A, time t-1 0 0.0048 1 Sec. 1 - B to A, time t-1 0 0.0252 1 Sec. 2 - B to A, time t-1 0 0.0273 1 Sec. 3 - B to A, time t-1 0 0.0332 1 Sec. 4 - B to A, time t-1 0 0.0203 1 Sec. 5 - B to A, time t-1 0 0.0112 1 Sec. 6 - B to A, time t-1 0 0.0138 1 Sec. 7 - B to A, time t-1 0 0.0215 1 Sec. 8 - B to A, time t-1 0 0.0273 1 Sec. 9 - B to A, time t-1 0 0.0225 1 Source: Author’s creation. 4.4 State-Level Analysis Results The first set of hypotheses presented in this dissertation predict that, in general terms, when a country is adversely affected by another country’s protectionist measure, it will systematically respond to that country by adopting its own protectionist measures in return (Hypothesis 1). However, the existence of PTAs (Hypothesis 2) and GVCs (Hypothesis 3) decrease the likelihood of protectionist responses. Table 4.3 shows that, holding all other variables constant, a one-decimal-point increase in the share of bilateral trade affected by import restrictions by B on A in time t-1 120 results in a 0.043 (4.3%) import restriction increase by A on B in time t (Model 1). This positive correlation is statistically significant even when controlling for other bilateral characteristics such as the liberalizing measures by B on A in time t-1, bilateral investment treaty between A and B, population-weighted distance—the gravity model—the relative economic difference, B’s log GDPpc / A’s log GDPpc, common colonizer, and common official language (Model 2), and when including countries A and B fixed effect and time fixed effect (Model 3). These results support Hypothesis 1, which states that countries will retaliate against nations that initially target them with a protectionist measure, which is known as Direct Retaliation. Second, the following table also shows that the existence of a PTA between A and B is negatively correlated with protectionist responses. In other words, if a PTA exists, retaliation is less strong than would otherwise be the case. Something similar happens when a GVC exists between A and B. This negative correlation for both variables is statistically significant even when controlling for other bilateral characteristics. Countries A and B and year fixed effects were controlled for because some countries may implement more import restrictions than the average of countries simply to trade more than others, as do the U.S. and China. Other countries may be targeted with additional protectionist measures related to political concerns and pressures, like the economic sanctions imposed on countries such as Iran and Cuba. When controlling for fixed effects, GVCs do not seem to be as strong as PTAs in restraining protectionist dynamics. This may be signaling that GVCs are highly concentrated in a few central countries. Many countries do not play a relevant role in GVCs. Some of them do not have much potential at having GVC ties, as is the case with Argentina. Others, though, have important ties, such as Mexico. Both of these cases will be further 121 explained in Chapter 5. This second set of results supports Hypotheses 2, which suggests that PTAs are negatively correlated with protectionist responses, known as the Conditional Effect of PTAs. Table 4.3 – DV1a: All Protectionist Measures by A on B in Time t DV1a: Share of bilateral trade affected (log) Model 1 Model 2 Model 3 A&B fixed effect and time fixed effect All protectionist measures by B on A in time t-1 (share of bilateral trade affected) (log) 0.043*** 0.070*** 0.020** (0.011) (0.010) (0.010) Trade agreements between A and B -0.094*** -0.110*** -0.050*** (0.012) (0.010) (0.010) % of intermediate goods in total trade between A and B (GVC proxy variable) -0.604*** -0.090* -0.070 (0.040) (0.040) (0.040) CONTROL VARIABLES NO YES YES Constant 0.200*** -0.390*** 9.670*** (0.001) (0.01) (0.344) R2 0.01 0.06 0.39 Adj. R2 0.01 0.06 0.39 Num. obs. 53831 50077 50077 RMSE 0.24 0.23 0.18 ***p < 0,001, **p < 0,01, *p < 0,05 The second set of hypotheses concentrates on the relationship between tariffs and NTMs within the institutional context of PTAs and GVCs. Hypotheses 4 and 5 predict that countries with high trade interdependence—with PTAs and GVCs—retaliate more with non- tariff measures than with tariff measures. For this set of hypotheses, two different dependent variables were considered to verify the robustness of the results. First, there was the number of transparent (DV2a) and non-transparent (DV2b) protectionist measures levied by A on B in time t, the continuous variable. Second, was the proportion of transparent (DV3a) and Non-Transparent (DV3b) protectionist measures out of total measures created by A and used on B in time t as a percentage. 122 Table 4.4 shows that, holding all other variables constant, one unit of increase in the number of import restrictions by B on A in time t-1 increases the number of tariff restrictions by A on B in time t by 0.230 (Model 1a). The number of non-tariff restrictions in time t also increases by 0.490 (Model 1b). Second, it shows that when we use an interaction term between having a PTA and being affected by a protectionist measure in t-1, there is a negative relationship with tariff retaliation (Model 2a). This means that when a PTA exists between A and B and B implements a protectionist measure toward A, the latter responds, not with tariff measures (Models 2a and 4a), but rather with non-tariff protectionist measures (Models 2b and 4b). This same pattern holds with GVCs (Models 3a, 4a, 3b, and 4b). In other words, having a PTA or GVC is negatively correlated with tariff responses, while positively correlated with non-tariff responses (Models 3b and 4b). These results are consistent with the expectations of Hypotheses 4 and 5: both PTAs and GVCs are negative, statistically significant predictors of tariff responses, but positive, statistically significant predictors of nontariff responses. 123 Table 4.4 – The Relationship Between Tariff and Non-tariff Measures with PTAs and GVCs (DV2a and DV2b) DV2a: Tariff protectionist measures by A on B in time t (number of measures) (log) DV2b: Non-tariff protectionist measures by A on B in time t (number of measures) (log) Model 1a Model 2a Model 3a Model 4a Model 1b Model 2b Model 3b Model 4b Protectionist measures by B on A in time t-1 (num/ of measures) (log) 0.230*** 0.230*** 0.230*** 0.181*** 0.490*** 0.486* 0.520*** 0.500*** (0.000) (0.000) (0.000) (0.000) (0.004) (0.004) (0.010) (0.010) Interaction term: Trade agreements * Protect. measures in t-1 -0.041* -0.040* 0.130** 0.141** (0.020) (0.020) (0.048) (0.050) Interaction term: % of intermediate goods (GVC) * Protect. measures in t-1 -0.224*** -0.360*** 0.010*** 0.010*** (0.030) (0.031) (0.001) (0.001) CONTROL VARIABLES NO NO NO YES NO NO NO YES Constant 0.110*** 0.110*** 0.113*** -0.482*** 0.352*** 0.352*** 0.360*** 3.01*** (0.000) (0.000) (0.002) (0.024) (0.005) (0.005) (0.005) (0.060) R2 0.23 0.23 0.24 0.28 0.20 0.20 0.20 0.26 Adj. R2 0.23 0.23 0.24 0.28 0.20 0.20 0.20 0.26 Num. obs. 51872 51872 51872 50441 51872 51872 51872 50441 RMSE 0.45 0.45 0.45 0.44 1.07 1.06 1.06 1.03 ***p < 0,001, **p < 0,01, *p < 0,05 Finally, Table 4.5 supplements the previous results by considering the proportion of transparent and non-transparent protectionist measures in total measures by A on B in time t as the dependent variable. I find very similar results regarding the effect of GVCs in both transparent (Model 3a) and Non-Transparent (Model 3b) protectionist measures, even when considering different control variables (Models 4a and 4b). Additionally, I find similar results regarding the effect of PTAs in both Transparent (Model 2a) and Non-Transparent (Model 2b). However, this statistical significance diminishes when the control variables are included (Models 4a and 4b). These results are mostly consistent with the expectations of Hypothesis 5: GVCs are negative, statistically significant predictors of tariff responses, while they are positive, statistically significant predictors of nontariff responses. 124 Table 4.5 – The Relationship Between Tariff and NTMs with PTAs and GVCs (DV3a and DV3b) DV3a: Proportion of Tariff protectionist measures by A on B in time t (%) DV3b: Proportion of Non-Tariff protectionist measures by A on B in time t (%) Model 1a Model 2a Model 3a Model 4a Model 1b Model 2b Model 3b Model 4b Protectionist measures by B on A in time t-1 (num/ of measures) 0.053*** 0.054*** 0.230*** 0.100*** 0.130*** 0.146*** 0.154*** 0.150*** (0.001) (0.001) (0.000) (0.000) (0.001) (0.001) (0.002) (0.010) Interaction term: Trade Agreements (PTAs) * Protect. measures in t-1 -0.041*** -0.002 0.0005* 0.0002 (0.012) (0.002) (0.0003) (0.002) Interaction term: % of intermediate goods (GVC) * Protectionist measures in t-1 -0.074*** -0.001*** 0.003*** 0.003*** (0.020) (0.000) (0.000) (0.001) CONTROL VARIABLES NO NO NO YES NO NO NO YES Constant 0.080*** 0.078*** 0.080*** -0.710*** 0.146*** 0.144*** 0.148*** -0.970*** (0.001) (0.001) (0.001) (0.020) (0.001) (0.001) (0.001) (0.020) R2 0.05 0.05 0.05 0.05 0.14 0.14 0.14 0.20 Adj. R2 0.05 0.05 0.05 0.05 0.14 0.14 0.14 0.20 Num. obs. 51872 51872 51872 50077 51872 51872 51872 50441 RMSE 0.27 0.27 0.27 0.30 0.36 0.36 0.36 0.36 ***p < 0,001, **p < 0,01, *p < 0,05 4.5 Political and Economic Magnitude of the Empirical Findings at the State-Level Analysis The empirical results presented above are merely statistical relationships between the variables of interest. As such, the political and economic magnitude of the findings cannot be inferred directly. However, careful interpretation of the econometric findings can shed light on how interconnectedness through a PTA or GVC may affect the dynamics of bilateral protectionism. From a quantitative perspective, the main goal here is to first establish the existence of the relationship between the main variables of interest and then to explore them in detail through country case studies presented in Chapter 5. However, as a political scientist, I’m also particularly interested in the political economy magnitudes of these results, 125 which is one of the areas in which further research can build upon the results presented in this dissertation. I will talk more about this in Chapters 5 and 6. Let us return to the results presented in Table 4.3, in which Hypotheses 1, 2, and 3 are tested. This first set of results shows interesting stories from an economic perspective. As discussed earlier, both the DV and the main IV present the share of bilateral trade affected by protectionist measures by one country against another. This characteristic of the variables of interest—which is a percentage between 0 and 100 represented by the share of value traded between two countries affected by protectionist measures by year—already gives a rough estimate of how much of the value between two trading partners is affected through retaliation. However, it is still unclear how much this trade retaliation actually hurts the partner and what is the economic magnitude of the effect of an existing PTA. To understand this, I take a closer look at the coefficients and apply them to the maximum, minimum, and average of the variables of analysis presented in Table 4.1. In doing so, the following results were found: • If the share of trade affected by protectionist measures from B toward A in time t-1 increases 100 percentage points, from 0% to 100%, retaliation from A toward B increases 2%, on average, based on Model 3 of Table 4.3. • If the share of trade affected by protectionist measures from B toward A in time t-1 increases by 1 percentage point, such as from 1% to 2%, retaliation from A toward B increases 0.020%, on average. Even though this is not nearly a 1-to-1 relationship, it still indicates that retaliation dynamics exist. From Table 4.1, we know that the average of all protectionist measures by B on A in time t-1, which is the share of bilateral trade affected in terms of log, is 0.106 (10.6%). With this data in mind, the average protectionist measure response can be presented, which is 0.21% (0.106*0.020*100). The conclusion is if country B has no protectionist measures in 126 place, and it simply introduces the average protection measures across the same, which is 10.6%, then, ceteris paribus, country A will certainly retaliate with 0.21%. The interesting part of the analysis presented above is the effect of a PTA on bilateral protectionist dynamics. What Table 4.3 shows is that having a PTA in place, ceteris paribus, means that the retaliation is 5% smaller than not having it. The size of this effect is considerable when compared to the 2% effect of the protectionist measure. In other words, if country B protectionism toward A (as a share of trade) increases from 0% to 100%, A will increase protectionism toward B by 2%. If this result was reversed such that B goes from 100% to 0% protectionism toward A, then A will only reduce its protectionist measure toward B, on average, by 2%. This is 3 percentage points less than the reduction effect in the protectionism of PTAs—a decrease of 60%. Let’s think about these results from a political perspective. The main political outcome here is that PTAs seem to be effective in deterring a spiral of protectionism and help prevent a tit-for-tat dynamic in exactly the way that they are designed to do. Hence, instead of dismantling them, countries should work to deepen trade-related constraints through trade agreements. However, the results presented in Tables 4.4 and 4.5 introduce two key elements related to the real effect of PTAs in the context of murkier protectionism. If there is a trade agreement in place, when B introduces one new protectionist measure against A, retaliation from A to B of tariff measures is negative, as demonstrated by Models 2a in both Tables 4.4 and 4.5. However, if there is a trade agreement in place when B introduces one new protectionist measure toward A, retaliation from A to B of nontariff measures is positive. This has very important policy implications: given that nontariff measures are not easily 127 observable, countries with PTAs tend to respond much more with nontariff measures than with tariff measures. Here a step by step of what may be the political dynamics that explain the findings: 1. Country B implements a protectionist measure toward A. There is a PTA in place between the two countries. 2. If the measure implemented by B is nontransparent, the affected sector normally notices that measure before the government. For example, if it is a consumption subsidy for domestic products in country B the affected sector in country A will notice that by a decrease in its exports to B. 3. Once affected sector in country A notices the protectionism measure, it will pressure its government to ask country B to do something in response. 4. Country A’s government, in order to not escalate the tensions with B but give a response to its domestic industry, will have much more incentive to implement a nontariff measure in response. This protectionist measure would probably be to defend the sector affected domestically. That is why this dynamic can also be explained from the sectoral perspective, as shown in a following section of this dissertation. Finally, as for the effect of GVCs, no dependable results were discerned based on the economic magnitude of GVCs. This is likely due to the limitations of GVC measures. As different studies have shown, it is still a very complex task to measure them (World Bank, 2020b). This is due to not only the measurements itself but also to the lack of consensus about the right measure to use. This is something that policymakers note also. For example, in the interview I conducted with Beatriz Leycegui Gardoqui, former undersecretary of foreign trade at the Secretariat of the Economy of Mexico (from 2006 to 2011), she said that: “For example, in Latin America the big problem is that it depends a lot on commodities and not on products with more added value. Then each reality depends on the explanation of why 128 they have not also fared in a globalized world. That is why it is very complex to generalize.” 123 To address this complexity that Gardoqui discusses, in this dissertation, I rely on intermediate goods as my main variable of analysis, following a similar logic of studies of the OECD (2014); World Bank (2020b); and WTO (2020). Using intermediate goods as a GVC proxy is still imperfect because as soon as countries trade in intermediate products and add some value along the chain, they are included in a GVC. As such, it is quite difficult to say which countries are partners in a GVC and which ones are not. In general, GVC data quality is still limited, as is the data reported by multinationals. Take for example the case of the value chain of an iPod from Apple. Studies have shown that in exports from China to the U.S., the value added in China constituted only 2 percent of the final sales price of an iPod (Dedrick et al., 2010). 124 More generally, when analyzing GCV, we should be skeptical about the reports of multinationals due to transfer pricing and limited information about the actual value added; there are plenty of tax and accounting issues, making it a work of art and science to find the true value added at each station along the GVC. Thus, the results on GVCs have to be taken as conclusive rather than just preliminary evidence until more detailed and publicly available data available becomes available. If we take the case of the U.S.-China trade war as an example, we can see that the results above are in line with the bilateral tension between the two countries. Chong and Li (2019) evaluate the domino effect of the U.S. setting higher tariffs and China setting higher tariffs in response and take the worst-case scenario of bilateral trade between China and the 123 Author’s interview with Beatriz Leycegui Gardoqui, January 2020. See Appendix 5 for the full interview. 124 For an extended discussion about this Dedrick et al.’s findings, see Hal R.Varian, An iPod Has Global Value. Ask the (Many) Countries That Make It, The New York Times (June 28, 2007), <https://www.nytimes.com/2007/06/28/business/worldbusiness/28scene.html>. 129 U.S. dropping by 27%. Additionally, they theorize that “China will lose 1.1% of job positions and about 1% of total output. Due to capital control policies and massive reserves in China, shocks to foreign reserve assets should be manageable” (Chong and Li, 2019, p. 199). For all these reasons, it is important to put the quantitative results in context. The fact that a coefficient is highly statistically significant still does not tell too much about the real economic and political significance. This is why in Chapter 5, an in-depth analysis of three specific country cases is conducted to elucidate the political and economic magnitude of the empirical findings. This analysis will be centered in Latin America on three countries in particular— Argentina, Brazil, and Mexico—for a more detailed analysis that will answer questions about how much an increase of protectionism has an effect on the economy, taking into account country-specific variables. Examples of this include: How many jobs are at stake with protections? What share of GDP is really affected by recent protectionist dynamics? Likewise, I study the profile of the trade between specific countries and their main trading partners, from the perspective of what underlying factors are traded. For example, if it is mainly manufacturing being targeted, such as in the case of Mexico, then jobs get lost through retaliation. If it is resources or services, it may well be that only capital incomes fall through the retaliation by trading partners, but not a significant number of jobs are directly affected. 130 4.6 Sectoral-Level Analysis Results The last set of hypotheses focuses on how sectoral bilateral dynamics affect the way countries respond to protectionism. At this level of analysis, it is argued that protectionist retaliation is more likely to happen cross-sector rather than within-sector (Hypothesis 6). The higher a sector’s level of industrialization, the less likely within-sector retaliation will occur (Hypothesis 7). Finally, I argue that protectionist responses are less likely to occur with trade involving intermediate goods (Hypothesis 8). In this dissertation, country-level hypotheses are more inductive in nature, but also amenable to findings that can contribute to theory building. Sectoral-level hypotheses are more deductive, having been revisited in previous theories in the trade policy literature to identify certain patterns and then test whether the data fits within a new international context. Taken together, this methodological approach brings us closer to theory expansion. Turning to the sectoral analysis, Table 4.6 tells two stories. It supports Hypothesis 8 by showing that protectionist responses are less likely for trade involving intermediate goods. As Model 5 shows, when B implements measures toward A that impact intermediate goods, it has a negative impact on retaliation, from A toward B, in this same sector. This negative correlation is significant at the bilateral level of variables, as seen in Model 5. However, it retains the negative sign but does not remain statistically significant when we include controls, as seen in Model 6. Indeed, when B implements measures toward A that impact intermediate goods, we see a positive cross-sectoral retaliation from A to B impacting capital goods, like transport equipment, and industrial, which also supports Hypothesis 6––that protectionist retaliation is more likely to happen across sectors rather than within them. 131 In the BEC classification, one of the sectors with high levels of industrialization is Capital Goods such as with transport and industrial equipment; in the HS classification, the sectors that contain more industrialized products are Sector 3, other transportable goods, except metal products, etc.; Sector 4, metal products, machinery and equipment; and Sector 5, constructions and construction services. With these sectors in mind, we can see that results for both types of classifications validate Hypothesis 7, which suggests the higher a sector’s level of industrialization, the lower the within-sector retaliation. As we can see in Table 4.6, there is not a strong positive correlation in the realms of capital goods retaliation (Model 1 and 2). Using HS classifications, it can be seen that, although there is within-sector retaliation in Sector 3 (Model 3 in Table 4.7), the same pattern for Sectors 4 and 5 is not seen (Model 4 and 5 in Table 4.7). Figures 4.2 and 4.3 summarize the results presented in Tables 4.6 and 4.7 and describe how sectoral dynamics are playing out within the BEC and HS classification schemes. For both type of measures, the data show that most of the retaliation is cross- sectoral (Hypothesis 6), which is in line with a growing consensus within the literature on the political economy of trade policy. 132 Table 4.6 – Responses by Sector Using BEC Classification (DV4) DV: share of bilateral trade affected of Capital, Intermediate, and Consumption Goods (BEC classification) by A on B in time t (% of bilateral trade affected by sector) (log) Capital Goods, t Consumption Goods, t Intermediate Goods, t Model 1 Model 2 Model 3 Mode 4 Mode 5 Mode 6 Capital, t-1 0.011 0.001 0.100*** 0.090*** 0.130** 0.120** (0.001) (0.010) (0.020) (0.020) (0.040) (0.040) Cons. Goods, t-1 0.020*** 0.010** -0.040*** -0.041*** 0.030 0.050* (0.001) (0.005) (0.010) (0.010) (0.020) (0.020) Intermediate, t-1 0.010*** 0.00 -0.009 -0.002 -0.040*** -0.020 (0.001) (0.00) (0.000) (0.004) (0.010) (0.01) CONTROL VARIABLES NO YES NO YES NO YES Constant 0.0006*** 0.0030*** 0.001*** -0.009* 0.010*** -0.010*** (0.000) (0.000) (0.000) (0.004) (0.007) (0.001) R2 0.00 0.01 0.00 0.01 0.00 0.01 Adj. R2 0.00 0.01 0.00 0.01 0.00 0.01 Num. obs. 32643 25847 32643 25847 32643 25847 RMSE 0.01 0.01 0.01 0.01 0.01 0.01 ***p < 0,001, **p < 0,01, *p < 0,05 133 Table 4.7 – Responses by Sector Using HS classification (Sector 0 through Sector 9) (DV4) DV: share of bilateral trade affected of Capital, Intermediate, and Consumption Goods (BEC classification) by A on B in time t (% of bilateral trade affected by sector) (log) Model 0 Sec 0, t Model 1 Sec 1, t Model 2 Sec 2, t Mode 3 Sec 3, t Mode 4 Sec 4, t Mode 5 Sec 5, t Mode 6 Sec 6, t Mode 7 Sec 7, t Mode 8 Sec 8, t Mode 9 Sec 9, t Sec 0, t-1 -0.010 -0.030 0.010 0.050** 0.063*** 0.018 0.009 -0.005 -0.023 0.064*** (0.011) (0.02) (0.020) (0.02) (0.020) (0.016) (0.016) (0.021) (0.024) (0.019) Sec 1, t-1 -0.010 0.023** 0.030*** 0.030** 0.012 0.001 0.016* -0.013 0.000 0.031*** (0.010) (0.010) (0.010) (0.01) (0.008) (0.010) (0.007) (0.009) (0.011) (0.009) Sec 2, t-1 -0.003 0.019* 0.020* 0.020** 0.030*** 0.014** 0.011 0.009 -0.016 0.027** (0.004) (0.010) (0.010) (0.01) (0.084) (0.006) (0.007) (0.009) (0.010) (0.008) Sec 3, t-1 0.006 0.020** 0.010 0.05*** 0.003 0.004 0.002 0.015* -0.008 0.055*** (0.00) (0.010) (0.010) (0.01) (0.007) (0.010) (0.010) (0.008) (0.009) (0.008) Sec 4, t-1 0.001 0.011 -0.009 0.020** 0.004 0.020*** 0.024** -0.012 0.005 0.015 (0.005) (0.011) (0.010) (0.01) (0.010) (0.010) (0.008) (0.010) (0.012) (0.009) Sec 5, t-1 0.003 0.020 -0.004 0.020 0.05*** 0.018* 0.050*** 0.016 -0.038** 0.047*** (0.007) (0.015) (0.020) (0.020) (0.01) (0.010) (0.011) (0.014) (0.016) (0.013) Sec 6, t-1 -0.020** 0.006 0.010 0.030** 0.046*** 0.012 0.030** 0.032*** -0.013 0.063*** (0.010) (0.010) (0.010) (0.010) (0.012) (0.01) (0.009) (0.012) (0.014) (0.011) Sec 7, t-1 -0.076 -0.018* -0.010 0.042*** 0.023*** 0.003 0.014 0.006 -0.008 0.050*** (0.053) (0.010) (0.010) (0.010) (0.010) (0.010) (0.008) (0.010) (0.011) (0.009) Sec 8, t-1 -0.090* -0.026*** 0.010 0.017* 0.004 0.010 0.001 0.023*** -0.006 0.002 (0.048) (0.010) (0.010) (0.009) (0.010) (0.010) (0.007) (0.009) (0.010) (0.008) Sec 9, t-1 0.092* 0.023** 0.040*** 0.007 0.031*** 0.011 0.011 -0.001 -0.006 0.055*** (0.052) (0.010) (0.01) (0.010) (0.010) (0.007) (0.008) (0.010) (0.011) (0.009) CONTROL VARIABLES YES YES YES YES YES YES YES YES YES YES Constant 0.070*** 0.270*** 0.360*** 0.600*** 0.200*** 0.020 0.098*** 0.284*** 0.752*** 0.340*** (0.012) (0.020) (0.020) (0.019) (0.020) (0.021) (0.017) (0.022) (0.025) (0.020) R2 0.01 0.01 0.01 0.02 0.01 0.01 0.00 0.006 0.023 0.022 Adj. R2 0.01 0.01 0.01 0.02 0.01 0.01 0.00 0.005 0.022 0.021 Num. obs. 28002 28002 28002 28002 28002 28002 28002 28002 28002 28002 RMSE 0.16 0.32 0.32 0.32 0.27 0.22 0.23 0.288 0.332 0.265 ***p < 0.01, **p < 0.05, *p < 0.1 134 Figure 4.2 – Sectoral Retaliatory Dynamics Using BEC Classification Source: Author’s creation. Figure 4.3 – Sectoral Retaliatory Dynamics Using HS Classification Source: Author’s creation. 135 Given the deductive and descriptive nature of this sector analysis, it is important to contrast the empirical findings with evidence from previous studies that explore sectoral retaliation dynamics––especially under the WTO’s dispute settlement mechanism (DSM). As introduced in Chapter 1, while a large body of research addresses patterns of retaliation under the WTO’s DSM, few studies have considered how these dynamics have played out beyond formal institutional boundaries, i.e., when the DSM adjudication panel becomes deadlocked. It is in this realm that an important number of retaliatory dynamics can occur in “murkier” and less identifiable forms. 125 First, low within-industry retaliatory dynamics is in line with Gowa and Mansfield (2004), who show that countries with strong intra-industry trade are less likely to engage in conflict than cross-industry trading partners. Also, studies on the relationship between global supply chains and trade policy responses to the 2008 crisis show that participation in global value chains within-sectors is “a powerful economic factor [in] determining trade policy responses” (Gawande et al., 2015, p. 102). As we saw in Chapter 1, intra-industry trade promotes the emergence of similar foreign policy preferences among trading partners. However, this does not happen in reverse; states with similar foreign policy preferences do not necessarily engage in intra-industry trade (Peterson & Thies, 2015). As mentioned above, there are lower levels of trade friction 125 As noted earlier, even within the WTO, there have been new types of requests that go beyond the traditional framework of tariff or NTB cross-retaliation. For example, in Canada’s dispute against Brazil’s export financing program for its Embraer Aircraft, Canada requested the suspension of such concessions under parallel WTO agreements, such as the Agreement on Textiles and Clothing and on Import Licensing (Bonomo, 2014). 136 in contexts of within-sector trade because the demand for cheap inputs by domestic suppliers and exporters decreases the pressure from domestic lobbies for market protection. My results can be used to explain and understand why there are lower levels of retaliation when protectionist policies target intermediate goods (Hypothesis 8). There seem to be lower levels of trade friction in contexts of within-sector trade due the demand for cheap inputs by domestic suppliers and exporters, which decreases domestic lobbies’ pressure on governments for market protection. In other words, governments will not levy protectionist trade policies against another government that is integral to the production, an input, of the first government’s products, an output. And if a government has to due to internal pressures, it will do so in the least visible way possible (Hypothesis 5, evidence in Models 3b and 4b of Table 4.5). 137 Table 4.8 – Summary of Hypotheses and Evidence from Results of Quantitative Analysis Hypotheses Evidence from quantitative analysis HYPOTHESIS 1: Countries will retaliate against those nations that initially targeted them with a protectionist measure (Direct Retaliation). From Table 4.3: One unit of increase in the share of bilateral trade affected by import restrictions by B on A in time t-1, increases 0.043 (around 4.3%) import restrictions by A on B in time t. This positive correlation is statistically significant, even when controlling for other bilateral characteristics and A and B fixed effect and time fixed effect (Model 3). HYPOTHESIS 2: Preferential Trade Agreements (PTAs) are negatively correlated with protectionist responses (Conditional Effect of PTAs). HYPOTHESIS 3: Global Value Chains (GVCs) are negatively correlated with protectionist responses (Conditional Effect of GVCs). From Table 4.3: The existence of a PTA between A and B is negatively correlated with protectionist responses. This negative correlation is statistically significant, even when controlling for other bilateral characteristics and fixed effects. However, for the effect of GVC in protectionist dynamic, we see less consistent evidence when controlling for fixed effects. In other words, GVCs do not seem to restrain protectionism as PTAs do. HYPOTHESIS 4: Preferential Trade Agreements (PTAs) are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses (Conditional Effect of PTAs). HYPOTHESIS 5: Global Value Chains (GVCs) are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses (Conditional Effect of GVCs). From Tables 4.4 and 4.5: One unit of increase in the number of import restrictions by B on A in time t-1 increases the number of tariff restrictions by A on B in time t by 0.230 (Model 1a), and the number of non-tariff restrictions in time t also increases by 0.490 (Model 1b). When we use an interaction terms between having a PTA and being affected by a protectionist measure in t-1, there is a negative relationship with protectionist tariff measures (Model 2a). The same happens with GVCs (Model 3a). Table 4.5 shows very similar results for the effect of GVCs in both Transparent (Model 3a) and Non-Transparent (Model 3b). Similar results are also found for the effect of PTAs in both Transparent (Model 2a) and Non-Transparent (Model 2b). However, these lose statistical significance when we include control variables (Models 4a and 4b). HYPOTHESIS 6: Protectionist retaliation is more likely to happen cross-sector rather than within-sector (Cross-Sector Retaliation). HYPOTHESIS 7: The higher the level of industrialization of the sector, the lower the within-sector retaliation (Within-Sector Retaliation). HYPOTHESIS 8: Protectionist responses are less likely with trade in intermediate goods (Within-Sector Retaliation). From Tables 4.6 and 4.7 For both types of measures (HS and BEC classification) we can see that most of the retaliation is cross-sectoral (Hypothesis 6) (See Figures 4.2 and 4.3). In general, there does not appear to be a strong positive correlation within capital goods retaliation (Models 1 and 2). From HS classification we see that although there is within- sector retaliation in Sector 3 (Model 3 in Table 4.7), we do not see the same pattern for Sectors 4 and 5 (Models 4 and 5, Table 4.7). Protectionist responses are less likely with trade in intermediate goods (in Table 4.6, Models 5 and 6). Source: Author’s creation. 138 4.7 Conclusions on the Quantitative Analysis As mentioned in the introduction of this dissertation and through the review of the existing literature, in contexts of high economic uncertainty––like the one experienced since the 2008-09 GFC, which has worsened in the wake of the COVID-19 pandemic––countries are more likely to increase protectionism (Ahir et al., 2020; Baldwin, 2009; Baldwin & Evenett, 2009; Evenett, 2019). 126 Through this quantitative analysis, I seek to answer the following questions: How have countries responded to trade protectionism since the shock of the GFCs? How does economic interdependence—of PTAs and GVCs—affect countries’ propensity to engage in protectionist retaliation? How does economic independence—of PTAs and GVCs—affect the type of measures countries deploy to respond to protectionism? How do industrial and sectoral factors impact the way countries respond to protectionism? In this chapter, the answers to these questions are sought through a Large-N quantitative analysis using primary data. In contexts of high uncertainty, protectionism has been used not as an economic tool—e.g., to protect domestic sectors, or mitigate trade deficit—but rather as a policy tool used to do things such as respond and send signals to special interest and business lobbies. During periods of high economic uncertainty, special-interest groups have gained the leverage to pressure governments into protecting domestic markets. Inevitably, countries tend to protect themselves against those countries with which they have the strongest trading relationship. 127 It is less plausible that trade policymakers would guard against those 126 See Appendix 5 of this dissertation for a list of different measures of uncertainty for the COVID- 19 Crisis. Also see Altig et al. (2020); Baker et al. (2020). 127 Trade interconnectivity normally increases bilateral trade flows between countries with such interconnectivity (PTAs and GVCs) (Urata & Okabe, 2010). 139 countries with lower trade flows. However, when governments implement restrictive trade measures against strong trade partners, this escalates protectionism. While bilateral trade flows are most negatively affected, there can be multiple effects in the sense that this signals a general mood of protectionism. In an interview I conducted with Eduardo Egas, former Vice Minister of Foreign Trade of Ecuador, he said: When the pressure from local groups is present … it only depends on the authorities to yield or maintain respect for the commercial agreements signed with their foreign partners. Influence groups seek restrictions on the side of sanitary measures, technical measures, or environmental measures, justified by an apparent consumer defense. These types of restraining weapons are very common in recent times and often exceed the real response capacities of the supplying countries. 128 What I argue here is that, in this current context of high uncertainty, governments need to find the optimal balance between giving in to protectionist domestic pressures and mitigating against an upward protectionist spiral with its main trading partners. While striving to maintain some balance many countries appear to have embraced murkier and less transparent trade policy measures. To recapitulate, my quantitative results at the state level show that trade protectionism in a context of high uncertainty over the past decade has assumed the followings features: • In general terms, when a country is adversely affected by a protectionist measure adopted by another country, it will systematically respond to that country by adopting protectionist measures in return. However, the existence of PTAs and GVCs decreases the likelihood of a protectionist response (Table 4.3) • Countries with high trade interdependence—PTAs—will retaliate more with non-tariff measures which are less transparent rather than tariff measures (Table 4.3). At the sectoral bilateral level of analysis, we see that the sectoral characteristics of bilateral trade affect the way countries respond to protectionism. More specifically: 128 Author’s interview with Eduardo Egas, July 2020. See appendix 5 for the full interview. 140 • Protectionist retaliation is more likely to happen cross-sector rather than within- sector. • Protectionist responses are less likely with trade in intermediate goods. • There is no strong evidence that the higher the level of industrialization of the sector, the lower the within-sector retaliation. This is known as within-sector retaliation industrialization. It is important to note that this quantitative analysis constitutes a tentative first step, meaning that there are important limitations to the analysis in its current form. First, although this study goes one step forward by not considering bilateral data rather than monadic (which looks at one state as the unit of analysis), this is still not enough to have a clear picture of the current global trade dynamic, which is essentially a network in which many factors and countries interact at the same time. In other words, although this dissertation makes an important contribution by introducing a large-N bilateral analysis to potential retaliatory responses, this may not only be a story of countries A and B, but also country C. One example is when in December 2017, during the WTO Ministerial Conference in Buenos Aires, the U.S. and Japan joined forces to develop a common position vis-à-vis Chinese trade practice. In September 2018, they issued a statement that was also signed by the EU making specific reference to concerns with non-market-oriented policies and practices of third countries, concerns with forced technology transfer policies and practices of third countries, and on industrial subsidies and state-owned enterprise (Evenett & Fritz, 2018). Further research should build on these results to gain a greater understanding of how trading partners affect each other’s trade policy-making decisions. In particular, the main focus of this analysis is on dyadic dynamics rather than triadic ones, though triadic dynamics would be an interesting area of research extension. Further research could also employ relational event approaches, in which systemic effects can be better captured to gain a better grasp of how the system is faring as a whole (Butts & Marcum, 2017). The whole system, in other words, may 141 offer a better explanation than the sum of its parts. A second limitation to this analysis, due to data restrictions, is the relative short period taken into consideration, i.e., the aftermath of the 2008-09 GFC. There is no data available before 2008 at the different levels of analysis used in this dissertation. This is discussed more extensively in the conclusions of this dissertation. In the context of economic crisis and post-crisis, governments tend to implement more protectionist measures than they normally do in the hopes of redirecting demand toward domestically produced goods. Although the period studied in this analysis goes until 2018, studies have shown that the recovery of global trade after the GFC is still weak compared to the marked acceleration in global trade in the previous two decades before the crisis (Wozniak & Galar, 2018). It is thus possible that much of the retaliation captured here is specific to the GFC and inapplicable to other less turbulent economic contexts. As such, it cannot be claimed that the findings of this analysis are generalizable to other economic periods. Finally, although this large-N analysis helps us understand big-picture trends, there may be specific or unique country characteristics that affect the ways that governments are responding to protectionism, as well as the precise forms of their retaliation. In other words, the quantitative analysis in this chapter must be complemented by a more in-depth analysis of individual country cases. This is exactly what the following chapter attempts to accomplish by analyzing three emerging economies in Latin America: Mexico, Brazil, and Argentina. Hence, Chapter 5 will offer a robust examination of these country cases based on the analysis offered in this chapter. 142 CHAPTER 5 – A COMPARATIVE APPROACH: PROTECTIONISM AND TRADE POLICY RESPONSES IN LATIN AMERICA “No tool is omnicompetent. There is no such thing as a master-key that will unlock all doors.” ARNOLD TOYNBEE, A STUDY OF HISTORY, 1936 5.1 Introduction This chapter moves away from the large-N analysis undertaken in Chapter 4 and concentrates on three country case studies in the Latin American region: Argentina, Brazil, and Mexico. 129 As mentioned throughout this dissertation, I have used a mixed-method approach to test the hypotheses and causal claims presented in earlier chapters. Chapter 4 consisted of a quantitative analysis based on large-N data sets. In this chapter, I undertake comparative case-study research, which combines qualitative and quantitative information derived from within-country data. As different studies on policy-making have shown, the case-study approach helps to “directly analyze and reconstruct the process of political decision making,” in which actors’ interests and political and economic factors play a key role (Mukherjee, 2016, p. 23). Case analysis helps us ascertain causal mechanisms in more 129 Earlier versions of this chapter were presented at the USC’s Comparative Politics Group, where I received excellent comments and suggestions. I especially want to thank Bryn Rosenfeld, Gerardo Munck, and Saori Katada for their suggestions. My colleague and friend Victoria Chonn Ching also gave me very helpful comments on earlier versions of this chapter. Additionally, this chapter was presented at the Institute for Qualitative and Multi-Method Research, where I received many comments and suggestions that positively impacted my work. 143 detail than can be gleaned from large-N analyses (Collier et al., 2004; King et al., 1994). As Mukherjee (2016, p. 23) states, “a contextual comparison of individual cases also ensures that the generalizations made from the large-N analysis are not too sweeping.” Hence, these small-n case studies not only complement my quantitative findings, but also offer a robustness test of the large-N statistical analysis presented earlier. As Seawright (2016, p. 42) notes, “an integrative multimethod design provides a family of compelling strategies for linking qualitative and quantitative components of an overall design, while also enhancing the quality of causal inference.” I chose Mexico, Brazil, and Argentina as my country cases because they represent 72 percent of total regional exports and 64 percent of total imports (UN Comtrade, 2019). My task is to better understand their trade policy performance and illuminate how they were affected by and responded to protectionism since the GFC. As a main departure point, I build upon Wise et al. (2015) Unexpected Outcomes, which argues that emerging regions such as Latin America and East Asia showed remarkable resilience when faced with the GFC. This phenomenon differed from previous historical cases in the 1980s and 1990s and may have played a role in “deterring a full-blown global depression” (p.1). I take this framework further and explore how the industrial characteristics of my three country cases shape the likelihood of a response to rising global protectionism. Within each country, I analyze specific industry-level cases that have been directly or indirectly affected by current protectionist trends and I examine the political economy dynamics that are playing out in each case. Recent studies have shown important direct and indirect effects of post-GFC trade disputes in emerging economies, especially in Latin 144 America. 130 In light of this, I probe whether countries that have strong trade ties with the rest of the world via preferential trade agreements have been more affected by less transparent protectionist dynamics than those with fewer formal trade ties. In other words, one of the essential research questions will be the extent to which trade interconnectivity increases the risk of being harmed by non-transparent protectionism. A study developed by the Chilean General Directorate of International Economic Relations (2018) shows that 63 percent of Chile’s exports went to countries in the core and periphery of the U.S.-Sino trade conflict—India, Turkey, EU, Canada, Mexico, and Russia. Given Chile’s strong web of free trade agreements (FTAs), many tariff increases will not be implemented against Chile by its various trading partners. 131 This does not mean that the country has escaped the effects of non-tariff protectionist measures. Indeed, Chile has been affected by 2,124 measures between 2009 and 2019, 79 percent of which have been “murky” instruments implemented by countries that comprise the rest of the world—that is, countries with which Chile trades in the absence of an FTA. 132 What happened in Argentina, Brazil, and Mexico? Have they been affected by protectionism over the last decade? And if so, how did they respond to it? 130 Zandi et al. (2018) estimate the impact of the “Trump trade war” for the years 2018-2020. Interestingly, under the scenario of 25% tariffs on all Sino-U.S. trade and other “qualitative” actions taken by China against American firms operating there, their results show that protectionism reduces GDP growth from second quarter 2018 to second quarter 2020 in Argentina, Brazil, and Mexico in very different ways: in Argentina by more than 0.6%; in Brazil, 0.3 to 0.6%; and in Mexico, less than 0.3%. For more information, see Evenett & Fritz (2018). 131 As part of its follow-up process concerning the U.S.-Sino trade conflict, Chile has created three sectoral committees (agriculture, agroindustry, and fisheries and aquaculture). See Direcon (2018) “Riesgos y oportunidades de la guerra comercial”. Departamento de estudios, <https://www.subrei.gob.cl/wp-content/uploads/2018/07/riesgos_y_oportunidades_gc_final.pdf>. 132 Global Trade Alert (GTA), <https://www.globaltradealert.org/>. 145 This chapter is organized as follows: I begin with an explanation of my case selection strategy and Latin America’s relevance. Second, I describe my fieldwork method, including my interviews with key social, economic, and political actors involved in the making of Latin American trade policy. Third, I present an in-depth analysis of the three cases of interest. I conclude with some finals thoughts on this comparative analysis of protectionism and trade policy responses in Latin America. 5.2 Case Selection: Why Latin America and Why These Three Countries? Latin America, in general, and my specific three countries of interest are relevant for both theoretical and methodological reasons. First, from a theoretical point of view, the latest studies and political debates on recent trade dynamics have focused on the U.S., China, the EU, Canada, and Japan. Less has been written on the effects of protectionist global dynamics in emerging regions since the GFC. This fixation on major economies, though pertinent, overlooks the fact that emerging countries such as Brazil, Mexico, and Argentina have also been impacted by the rise of trade protectionism over the past decade (Zandi et al., 2018; Evenett & Fritz, 2018). Moreover, the literature has paid little attention to how these Latin American countries have reacted to this increasingly adverse global trade environment. Second, I selected these three Latin American emerging economies because they are also part of the G20 group of countries, which is the premier forum for international financial and economic cooperation. From a regional perspective, Latin America has emerged as an interesting “median” case in the non-developed world with regard to protectionist and liberalizing trade policy measures (Seawright & Gerring, 2008). As Figure 5.1 shows, Latin America stands between 146 Asia and Africa, two non-developed regions, in both total numbers of protectionist and liberalizing measures between 2008 and 2018 (left axis) and the share of RoW’s exports to the region affected by protectionist measures (right axis). As a regional “median” case in the non-developed world, the current literature has paid scant attention to the impact and responses to protectionist dynamics in Latin America (Seawright & Gerring, 2008). Figure 5.1– Latin America as a Median Case in the Non-Developed World in Terms of Protectionism Measures Source: Global Trade Alert, 2019, <https://www.globaltradealert.org/>. Also relevant is the fact that Argentina and Brazil, two of my three case countries, are members of the Southern Cone Common Market (Mercosur), which is a regional integration project established in 1991 by Argentina, Brazil, Paraguay, and Uruguay. Mercosur comprises more than 270 million people, and its combined GDP is US$5.8 trillion, making this Southern bloc equivalent to the fifth-largest economy in the world. 133 According to the Treaty of Asuncion––the founding document of the bloc—Mercosur’s main goal was to establish a common market by December 31, 1994. This involved “the free movement of 133 See Mercosur, ¿Qué es el MERCOSUR? <https://www.mercosur.int/quienes-somos/en- pocas-palabras/>. 147 goods, services and factors of production between countries through … the elimination of customs duties and non-tariff restrictions on the movement of goods, and any other equivalent measures.” 134 Yet thirty years since its founding, the formation of a common market has eluded bloc members. That is why Mercosur is normally defined as an “imperfect customs union” or simply a “free trade agreement” (Felter et al., 2019; Albertoni, 2016, 2019). To date, the members have agreed to eliminate some customs duties, to implement a common external tariff on certain products originating outside of Mercosur, and to adopt a common trade policy toward outside countries and blocs (Felter et al., 2019). Mexico, too, conducts most of its trade policy strategy through bilateral and unilateral policies. Having joined the North American Free Trade Agreement (NAFTA) with Canada and the U.S. in 1994, Mexico maintains an independent trade policy toward non-NAFTA trade partners. These differences are relevant in terms of my comparative case analysis because they offer sub-regional (Argentina and Brazil) and bilateral (Mexico) perspectives regarding the respective trade policy dynamics of these countries. The selection of these three cases also shows “diversity” within both the G20 and the Latin American region. Figure 5.2 shows how the three countries at hand are distributed in terms of share of world exports to the countries affected by protectionist measures in the G20 and Latin American countries. For both subgroups, Argentina has been the most protectionist, Brazil is in second place, and Mexico is the least protectionist of the three countries. 134 OAS-SICE, Treaty Establishing a Common Market between the Argentine Republic, the Federal Republic of Brazil, the Republic of Paraguay and the Eastern Republic of Uruguay, <http://www.sice.oas.org/trade/mrcsr/treatyasun_e.asp>. 148 When we consider trade measures implemented by the G20, which represent 80 percent of gross world product (GWP) and three-quarters of global trade from 2009 to 2019, 135 Argentina, Brazil, and Mexico are well distributed in terms of both the number of protectionist measures that each country has implemented as a G20 member and the share of world’s exports to each country that has been affected by protectionist measures within the Latin American region. Argentina has been the most protectionist country within the G20, Brazil falls in the middle of the list of G20 countries, and Mexico was the second least protectionist of the G20 economies. Figure 5.2 – Distribution of the Three Case Countries in the G20 and Latin America (Average 2009-18) Share of World Exports to G20 Countries Affected by Protectionist Measures Share of World Exports to Latin American Countries Affected by Protectionist Measures Source: Author’s creation based on GTA, 2019. 135 G20, What is the G20? <https://g20.org/en/about/pages/whatis.aspx>. 149 The countries at hand also represent an interesting mix in terms of their main economic and trade indicators. Table 5.1 shows the economic differences and commonalities between the three countries. First, when considering average growth of GDP between 2008 and 2018, Argentina registered the highest growth rate during those ten years (43 percent), ranking it as number 91 out of 196 countries in terms of GDP growth. 136 Both Mexico and Brazil’s GDP growth rates were 10 percent, ranking them 143 and 144 in terms of GDP growth, respectively, over this same time period. We see a very different scenario in terms of growth of GDP per capita, which is a more comparative figure that compensates for population differences. Argentina, again, is the country that has grown more in terms of per capita GDP growth (29 percent) from 2008 to 2018, ranking at 83 of 196 countries on this variable. Brazil’s per capita GDP growth was 1 percent over this same period, ranking it 144th; and Mexico had a negative per capita GDP growth rate of -3.43 percent, ranking it 159th in the 196-country sample. Second, the three countries have different markets as their main export destinations. Argentina’s main destination is Brazil, which is 18 percent of its total exports; for Brazil, China is the main destination (26 percent); and for Mexico, the U.S. is the main export destination (70 percent). While China is in the top three destinations for Argentina and Brazil, it is not one of Mexico’s main destinations. Mexico demonstrates an important concentration of exports to the U.S., while Argentina and Brazil have better diversification in terms of destinations. Significant variations also arise when considering the origin of imports for the three countries. Argentina’s main origin is Brazil, which is 24 percent of its total 136 Closer to 0 means a better performance. For more information visit, The Observatory of Economic Complexity, <https://oec.world/>. 150 imports; for Brazil, China is the main origin (20 percent); and for Mexico, the U.S. is the main import origin (57 percent). Third, in terms of sectoral representation in international trade, we can see a very similar pattern between Argentina and Brazil. In both cases, the main export sector is related to soybeans. Mexico is a more industrialized economy in which the three main exporting sectors are cars (11 percent), vehicle parts (7 percent), and delivery trucks (6 percent). In the main import sectors, we can see that Argentina is more dependent on industrial goods, with cars registering 8 percent and vehicles parts 5 percent as its two main importing sectors. Mexico’s import sectors are refined petroleum (8 percent) and vehicle parts (7 percent). Table 5.1 – Main Economic and Trade Characteristics of Argentina, Brazil, and Mexico ARGENTINA BRAZIL MEXICO ECONOMIC FIGURES GDP GROWTH (Avg. 2008 - 2018) 44% (RANKING 91 OF 196) 10% (RANKING 143 OF 196) 10% (RANKING 144 OF 196) GDP PC $11,684 (RANKING 66 OF 196) $8,921 (RANKING 82 OF 196) $9,673 (RANKING 79 OF 196) GDP PC GROWTH (Avg. 2008 - 2018) 29.5% (RANKING 83 OF 196) 1.02% (RANKING 141 OF 196) -3.43% (RANKING 159 OF 196) TRADE FIGURES Main Export Destinations Brazil: 18% China: 7% U.S.: 7% China: 26% U.S.:12% Argentina: 6% U.S.: 70% Canada: 4% Germany: 2% Main Import Origins Brazil: 24% China: 17% U.S.: 13% China: 20% U.S.:17% Germany: 6% U.S.: 57% China: 13% Germany: 4% Main Exporting Sectors Soybean Meal: 15% Corn: 6% Trucks: 6% Soybeans: 14% Crude Petroleum: 10% Iron Ore: 8% Cars: 11% Vehicles Parts: 7% Trucks: 6% Main Importing Sectors Cars: 8% Vehicles parts: 5% Soybeans: 4% Refined Petroleum: 7% Vehicles Parts: 4% Crude Petroleum: 3% Refined Petroleum: 8% Vehicles Parts: 7% Office Machine Pts: 4% Source: Author’s calculation of country rankings based on The Observatory of Economic Complexity, 2018, <https://oec.world/en/profile/country/>. Regarding institutional factors such as the rule of law, effectiveness of government, and regulatory qualities, we can see in Figure 5.3 that there are no major differences between 151 the three countries of interest. For this analysis, we are comparing cases that deal with similar structural challenges. Figure 5.3 shows that, in the three periods of analysis, Mexico had a better performance in terms of the effectiveness of governments and regulatory quality. During the three periods of analysis, Brazil had a better performance in terms of rule of law. Argentina has the worst performance in all three periods for the three variables. Figure 5.3 – Institutional Measures for Argentina, Brazil, and Mexico (Higher Indicates Better Performance) Source: World Bank’s Worldwide Governance Indicators (WGI) Project (2017, Total Countries, 215): WGI is produced by the Revenue Watch, Brookings Institution, World Bank Development Research Group, and the World Bank Institute. It reports aggregate and individual governance indicators for 215 countries. The indicators are based on information from 32 individual data sources provided by organizations such as survey institutes, think tanks, NGO’s, international organizations, and private-sector firms. They track six main dimensions of governance: Voice and Accountability, Political Stability and Absence of Violence, Government Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption, <http://info.worldbank.org/governance/wgi/index.aspx#home>. Finally, the region in general and the three countries specifically have their differences and commonalities in terms of PTAs and GVCs, two variables central to this analysis. In terms of PTAs, Mexico with 22 PTAs is second only to the European Union countries in the number of PTAs it has signed up to 2019. Brazil and Argentina (Mercosur) 152 are among the least integrated countries in terms of PTAs, as Mercosur’s common trade policy mandates that the four countries negotiate outside PTAs as an entire bloc. Figure 5.4 – How are the Three Selected Countries Distributed in the G20 and in Latin America on Preferential Trade Agreements Number of PTAs of G20 Countries Until 2019 Number of PTAs of Latin American Countries Until 2019 Source: Author’s creation based on WTO’s Regional Trade Agreements Database, <http://rtais.wto.org/>. To show how my three selected countries stand in comparison with other countries in terms of GVCs, I used WTO data on “Trade in Value-Added (VA) and Global Value Chains (GVCs),” which provides “an overview of the key indicators that can be derived from trade in value-added statistics.” 137 Some of the measures presented in the WTO’s profiles in VA and GVCs are the interconnection of national economies within GVCs and the level of trade 137 WTO, Trade in Value-Added and Global Value Chains: Statistical Profiles, <https://www.wto.org/english/res_e/statis_e/miwi_e/countryprofiles_e.htm>. 153 in merchandise and services in intermediate products. 138 The GVC participation index specifically provides “an estimation of how much an economy is connected to global value chains for its foreign trade.” There are two methods of GVC participation, including “upstream and downstream links in international production chains”: Individual economies participate in global value chains by importing foreign inputs to produce the goods and services they export (backward GVC participation) and also by exporting domestically produced inputs to partners in charge of downstream production stages (forward GVC participation). 139 The measure of trade in intermediate products and services shows the share in total trade of “tangible and intangible products utilized as inputs in production, excluding fixed assets. Trade statistics on intermediate products reflect the exchanges of parts, components, accessories, and intermediary services taking place within international production chains.” 140 As explained in earlier chapters, the definition of intermediate goods relies upon the United Nations’ Broad Economic Categories (BEC) classification, which groups commodities by main end-use, distinguishing between consumption, capital, and intermediate goods (see Figure 1.6 in Chapter 1). Figure 5.5 shows how the three selected countries are distributed in terms of the GVC Participation Index, which combines backward and forward GVC participation: Backward GVC participation refers to the ratio of the ‘foreign value-added content of exports’ … to the economy’s total gross exports. This is the ‘buyer’ perspective or sourcing side in GVCs, where an economy imports intermediates to produce its exports. Forward GVC participation corresponds to the ratio of the ‘domestic value added sent to third economies’ … to the economy’s total gross exports. It 138 These WTO profiles on value-added and GVCs rely on the OECD trade in value added database (TIVA) stemming from a joint OECD-WTO TIVA initiative. The TIVA online database can be accessed at <http://oe.cd/tiva>. TIVA indicators “cover goods and services and are available for a set of 64 economies and 36 industries from the international standard industrial classification, (ISIC revision 4). Hence, the number of profiles available as well as the main partners and industries shown in the tables reflect the current coverage of the TIVA database” (OECD-WTO TIVA). 139 WTO, Trade in Value-Added and Global Value Chains: Statistical Profiles, op. cit. 140 Idem. 154 captures the domestic value added contained in inputs sent to third economies for further processing and export through value chains. This is the ‘seller’ perspective or supply side in GVC participation. 141 As we can see, in terms of GVC participation, the three countries present interesting variation, both among themselves and between the average of developing and developed economies. Mexico appears to be in a better position in terms of GVC interconnectivity, while Brazil and Argentina fall below average compared to the rest of the world. In Figure 5.6, when we consider trade in merchandise intermediate products, we see that Brazil and Mexico are the two countries in which imports of intermediate goods have a higher representation in total imports. In comparison, Argentina and Brazil have a higher representation of intermediates in total exports. Figure 5.5 – The GVC Participation Index (% share in total gross exports, in 2015) Source: WTO, Trade in Value Added and Global Value Chains, <https://www.wto.org/english/res_e/statis_e/miwi_e/countryprofiles_e.htm>. Note: The last publicly available data by country is for 2015. 141 Idem. 155 Figure 5.6 – Trade in Merchandise Intermediates Products (2017) Exports of intermediates (% in total exports) Imports of intermediates (% in total imports) Source: Author’s creation based on WTO, Trade in Value Added and GVC, <https://www.wto.org/english/res_e/statis_e/miwi_e/countryprofiles_e.htm>. Considering the differences at hand, I used a “diversity” approach as the main case selection technique. Seawright and Gerring (2008) state that in case selection and analysis, the “diverse” method serves to “exemplify diverse values” of the dependent and independent variables, and it is necessary when conducting an “expiratory or confirmatory” analysis in which the goal is to illuminate the full range of variation on the dependent and independent variables (p. 297). As shown, diversity as the case selection technique fits the three cases mentioned above from both a regional and global perspective. Taking three different Latin 156 American country cases, which are “medians” in the non-developed world (see Figure 5.1), helps us understand regional patterns, as well as trade policy dynamics in other non- developed regions. Simultaneously, the three countries selected shed light on G20 dynamism. As different figures above show, the three countries selected are well distributed within the G20, in key variables of interest. These countries are situated in diverse positions at the regional and global levels, thus justifying the case analysis presented in this chapter (Gerring & Cojocaru, 2016; Seawright & Gerring, 2008). As mentioned, another key element regarding case selection is that it aims to complement the large-N statistical analysis presented in Chapter 4. In this regard, considering the six types of case study suggested by Lijphart (1971)––theoretical, interpretative, hypothesis-generating, theory-confirming, theory-infirming, and deviant––this case study has theory-confirming goals. Hence, having selected these cases, it is important to note that this is not simply country case analysis but rather country-sector analysis—meaning that country-sector’s reflection of the impact of protectionist measures emanating from the rest of the world will be the focus here. Particular attention will be paid to these research questions: 1) Who are the domestic economic actors involved in developing a trade policy response? 2) When they are adversely affected by a trade partner, how might they lobby for protectionist responses? 3) Do they promote diplomatic approaches before retaliation, such as bilateral working groups, to pressure the implementing country to lower trade restrictions? 4) Do representatives from the affected export sector coordinate with other traditional protectionist sectors, like importers, to demand retaliation? My intention in posing these questions is to add important insights to the trade policy literature by demonstrating that both export and import sectors may coordinate to demand 157 protectionism when faced with adverse policies from trade partners. As the trade literature has extensively demonstrated, the exporting sector typically lobbies for trade openness while import-competing industries tend to lobby for protection (Roosevelt, 2016). Can we say that both sectors find an equilibrium and together lobby for protection when faced with a protectionist policy response? The primary groups that were interviewed for this dissertation are the main industry lobbies—such as, in Brazil, the National Confederation of Industry—as well as other actors, including economic interest groups such as trade chambers, unions, and exporter associations. Before delving into the details of each country case, let me mention what I expect to see in each country in light of my quantitative results reported in the previous chapter. First, if my theoretical intuitions are correct, I expect that when they are adversely affected, all three countries will systematically respond to protectionist measures imposed by another country (H1). However, the existence of PTAs and GVCs can decrease the likelihood of a protectionist response (H2 and H3). Thus––considering Figures 5.4, 5.5, and 5.6––I would expect that Mexico, which is more interconnected to the world through PTAs and GVCs, would be less responsive to protectionism than Argentina and Brazil, which are less interconnected to the rest of the world. Second, I expect that countries with high trade interdependencies, such as Mexico, will retaliate more with non-tariff measures, which are less transparent, than with tariff measures. With regard to sectoral dynamics, I expect that protectionist retaliation is more likely to happen in a cross-sectoral fashion, rather than a within-sector one (H6); the higher the 158 level of industrialization of the sector, the lower the within-sector retaliation (H7); and protectionist responses are less likely with trade in intermediate goods (H8). 5.3 Fieldwork: Interviews with Key Trade Policy Actors in Latin American Between fall 2019 and summer 2020, I developed a list of relevant actors to interview in each country of interest. I then designed a semi-structured questionnaire to guide my fieldwork interviews with key social, economic, and political actors relevant to trade policy. Apart from gathering these insights on the topic of interest, another goal was to collect data on specific sectors that have been affected by protectionist measures. In particular, the interviews helped me pinpoint some of the factors that prompted policy makers to adopt more restrictive trade policies in the wake of the 2008-09 GFC. This fieldwork also let me understand the policy-making processes that undergird my large-sample analysis. 142 5.3.1 Interview Design and Methodology As mentioned before, my research design has two steps. First is the large-N quantitative analysis presented in Chapter 4 based on dyad-year data at the state and industry level, which encompassed different countries from different regions across the world. 143 The second part of my research design, which this chapter focuses on, is based on a more in-depth analysis of three emerging economies in Latin America: Argentina, Brazil, and Mexico. The following sections consist of a country narrative of trade policy performance amid current protectionist trends for each of the three cases. In doing so, this section offers a robust examination of these country cases according to the different levels of analysis offered above. Again, a 142 Due to the COVID-19 pandemic, I conducted most of these interviews via Zoom or Skype. 143 This interview design was reviewed and approved by the University of Southern California Institutional Review Board (IRB) (ID: UP-20-00392). 159 qualitative interview method is used to flesh out those factors that prompted policy makers to adopt more restrictive trade policies in the wake of the 2008-09 GFC. Interviewees primarily consist of industry leaders, “elites” from government, academia, the media, and research institutions. The sample size consists of at least one to two individuals from each category in each country. Contact information for the participants came through the standard snowball method, via contacts known by me or my faculty advisor. Additional contacts came from other known contacts or subjects in or familiar with the region. Twenty-four individuals were selected for interviews who came from various backgrounds and professions as listed in Table 5.2. Once an email address or other forms of contact information were obtained, an email was sent to potential subjects briefly describing the research and what would be required of the meeting, including dates and locations, the list of talking points based on the general themes of the research, and the informed consent disclosure. 144 If subjects agreed, they were asked to speak for approximately 30 to 60 minutes in an individual, one-time interview. 145 Talking points for these meetings were based on my extensive literature review on those factors that prompted policy makers to adopt restrictive trade policies. Participants chose the location, day, and time for the conversation via Skype or Zoom. Handwritten notes were taken for all meetings, and further impressions and notes were recorded with typewritten notes. 146 The list of interviewees was divided into five main groups: 1) global trade perspectives; 2) Latin American perspectives; and experts and key actors from 3) 144 I would especially like to thank the contacts related to strategic trade policy in Latin America provided by Prof. Carol Wise, which enabled me to reach high-level experts and policy makers. 145 For an example of the email sent to potential subjects, see Appendix 5 of this dissertation. 146 The interviewees were expected to speak Spanish, Portuguese, or English. 160 Argentina, 4) Brazil, and 5) Mexico. Table 5.2 shows a full list of those interviewed and includes a short biographical description. Some of them, due to the “no attribution” request, are listed with a generic description. 147 Table 5.2 – List of Interviewees Interview ID Name Short Bio GLOBAL TRADE PERSPECTIVES 001 Victor Stolzenburg A research economist with the Economic Research and Statistics Division (ERSD) of the World Trade Organization. Coordinates the WTO’s Global Value Chains Report. 002 Alejandro Jara Chilean lawyer, trade negotiator, former Director General of International Economic Relations, former Ambassador to WTO, and former WTO Deputy Director General. Presently practices international trade law in Santiago, Chile, and teaches trade policy in Chile, China, and Mexico. LATIN AMERICAN PERSPECTIVES 101 Jorge Sahd Director of the Center for International Studies at the Catholic University of Chile. 102 Andrés Rebolledo Former Vice Minister of Trade in Chile and former Minister of Energy. 103 Eduardo Egas Eduardo Egas, Executive President of the CORPEI Export and Investment Corporation; former Minister of Industries; former Vice Minister of Foreign Trade in Ecuador. 104 Dorotea López Director of the Institute of International Studies at the University of Chile. 105 Olga Lucia Lozena Former Vice Minister of Foreign Trade of Colombia and Alternate Representative of Colombia to the WTO, where she chaired the Subsidies Committee. 106 Magali Silva Peruvian economist, former Minister of Foreign Trade and Tourism of Peru (2013-2016), and Vice Minister of SMEs and Industry. MEXICO 201 Juan Moreno Director of International Relations of COPARMEX, the Mexican Employers’ Association. 202 Jesús Flores Undersecretary for Administrative Affairs in the Ministry of Commerce and Industrial Development from 1994 to 2000, General Director for Latin American Negotiations and Chief Negotiator between Mexico and Latin America. Currently, Director for Trade and Institutional Affairs at Tenaris (Mexico). 203 Beatriz Leycegui Served as Undersecretary of Foreign Trade at the Secretariat of the Economy of Mexico from December 2006 to August 2011. She has served as legal director of the Trade and Industrial Development Secretariat and was at the Secretariat of the Economy during the NAFTA negotiations. 204 Not for attribution Anonymous industrial union leader in Mexico. 205 Daniel Vázquez Mexican trade policy expert. He has advised companies and countries on 147 Following good practices of qualitative analysis, the full text of the interviews is presented in Appendix 5 so the reader of this dissertation can review the contents of the full answers and the specific portions I relied on to support my arguments in this dissertation. My main goal is to convert this information into replicable research data. For more information on good practices of qualitative data management, see Qualitative Data Repository, <https://qdr.syr.edu/guidance/managing>. 161 various issues, including market access, investment, and technical barriers to trade. ARGENTINA 301 Luis Miguel Etchevehere Minister of Agriculture, Livestock, and Fisheries of Argentina in 2019, Minister of Agribusiness of Argentina from 2017 to 2018, and President of the Rural Association of Argentina from 2012 to 2017. 302 Miguel Braun Secretary of Commerce of Argentina from 2015 to 2018 and Secretary of Economic Policy of the Ministry of the Treasury. 303 Ricardo Rozemberg Argentine economist, Specialist in International Economic Relations at Center iDeAS/UNSAM. 304 Not for attribution An anonymous trade policy maker of Argentina during the Macri Administration (2015-2019). 305 Not for attribution An anonymous trade policy maker of Argentina during the Macri Administration (2015-2019). 306 Felix Peña Argentine Specialist in International Trade Relations and Economic Integration. He currently works at the National University of Tres de Febrero (UNTREF) and the Industrial and Commercial Bank of China Foundation (ICBC). BRAZIL 401 Renato Flores Director of the International Intelligence Unit at the Foudaçao Getulio Vargas, Rio de Janeiro, where he is also a Professor at the Graduate School of Economics (FGV EPGE) and acts as Special Aide to the President of FGV. He has an extensive career in Brazilian and European institutions as well as in the Brazilian public sector at the Ministry of Planning. 402 Tatiana L. Palermo International Trade and Public Policies expert, former Brazilian Vice-Minister of Agriculture and Chief Agricultural Trade Negotiator, and chair of Palermo Strategic Consulting LLC. 403 Welber Barral Former Secretary of International Trade of Brazil from 2007 to 2011. Partner at BMJ Consultants. 404 Tatiana Lacerda Senior Fellow at the University of International Business and Economics in Beijing, China. Previously, Senior Advisor to the Director-General of the WTO from 2013 to 2018 and Secretary of Foreign Trade of Brazil from 2011 to 2013. Source: Author’s creation. 5.4 A Brief Background of Latin America’s Trade Policy Evolution The modern economic history of Latin America has seen numerous reforms since the mid- 1980s. External shocks have periodically hampered the economic success of the region. The widespread and prolonged reliance on import-substitution-industrialization (ISI), which originated in the 1960s, was a main drag on these economies. ISI “represents a shift away from the outward orientation to export promotion, to an inward-looking orientation. ISI was designated to replace imports with domestic production under the guiding hand of the state. Governments used activist industrial, fiscal, and monetary policy to achieve growth” 162 (Franko, 2018, p. 615). ISI measures wreaked havoc on regional fiscal accounts and induced periodic balance-of-payments crises. In the 1970s, international bank lending was directed toward Latin America under the auspices of sovereign borrowing. However, high oil prices in 1973 and 1974 and again in 1978 and 1979 fueled global inflation. The U.S Federal Reserve responded with high interest rates, sparking a debt crisis in Latin America that caused commercial banks to halt new lending to the region. The 1980s saw the collapse of the prevailing ISI model, as countries could no longer afford the subsidies and tax breaks. Nor could they generate the foreign exchange needed to pay for increasingly costly imports (Gereffi, 2014). After almost three decades of an ISI strategy based on high tariffs to protect national industries, in the 1980s many Latin American countries launched ambitious reforms based on trade liberalization, privatization, and deregulation. These were in line with the Washington Consensus, a set of policy instruments that Washington institutions such as the World Bank and the IMF prescribed for developing countries after the lost decade that followed the 1982 debt crisis. The Washington Consensus advocated a package of policies to achieve economic recovery (fiscal discipline, public expenditure priorities, tax reforms, financial liberalization, exchange rates, FDI, privatization, deregulation, property rights, and trade liberalization). On trade liberalization specifically, the policy recommendations were (Williamson, 1990, pp. 14-15): Quantitative trade restrictions should be rapidly replaced by tariffs, and these should be progressively reduced until a uniform low tariff in the range of 10 per cent (or at most around 20 per cent) is achieved. There is, however, some disagreement about the speed with which tariffs should be reduced (with recommendations falling in a band between three and ten years), and about whether it is advisable to slow down the process of liberalization when macroeconomic conditions are adverse (recession and payments deficit). 163 In the mid-1980s, no Latin American country had a positive current account balance, and with regional capital “dried up due to the debt crisis, current account deficits become unsustainable without external financing” (Idem, p. 234). Conditionality tied to emergency loans from the International Monetary Fund throughout Latin America promoted this trade liberalization. Mexico was the first of my three case countries to move from trade protectionism to trade liberalization, which it did in the context of acceding to the GATT in 1986. However, these reforms were not uniform throughout Latin America, and many countries still protected certain industries. Some trade liberalization efforts were quickly counteracted with protectionist policies in the form of non-tariff measures meant to reduce imports (IDB, 2018). Such are frequently designed to protect domestic losers from liberalized trade, as they demand compensation from losses when trade is liberalized. As such, domestic actors that represent these losing commercial interests also have an influential role in the trade policy-making process. These trade reforms in Latin America also increased exposure to global volatility, especially in terms of commodities. For commodities, terms of trade are contingent on constant and unpredictable price fluctuations. Ultimately, in many countries throughout the region, trade policy choices were not made at the national level but instead were rooted in the rules of membership in the WTO (IDB, 2018). In part due to the numerous trade reforms of the 1980s and 1990s, the average tariff rate in Latin America decreased significantly. For instance, the regional average tariff was nearly 50 percent in 1985, with rates as high as 80 percent in Brazil and Colombia. By the mid-1990s, the average tariff had come down to 12 percent and to 10 percent in 2002. Interestingly, as Table 5.3 shows, tariffs in Latin America have been consistently decreasing 164 since the mid-1980s. Andrés Rebolledo, former Vice Minister of Trade of Chile from 2014 to 2016, is worth quoting on this topic: Tariffs in the region are less and less relevant as protection measures. In fact, except Venezuela, Mercosur remains the set of countries that have the highest applied tariff in the region at a level of around 10%. Effective protection and tariff escalation with a high peak are where a relevant level of protection is found. It is precisely formal non-tariff barriers and discriminatory administrative practices that really aim to manage trade. 148 While tariffs have decreased in the last several decades, exports and imports of goods and services have increased consistently since the 1960s (Figure 5.7). In addition, exports as a share of GDP have also witnessed significant growth since the 1980s (Figure 5.8). Table 5.3 – Weighted Mean Tariff Rate Applied to All Products (%) Country Mid-1980s Late-1980s Mid-1990s 2000-2007 2008-2019 Argentina 28 44 14 8 7 Bolivia 20 19 10 7 5 Brazil 80 51 13 9 8 Chile 36 15 11 5 3 Colombia 83 48 11 10 6 Costa Rica 53 n/a 12 4 2 Ecuador 50 40 11 9 6 El Salvador 23 n/a 9 6 3 Guatemala 50 n/a 11 6 2 Honduras n/a 42 18 7 6 Mexico 34 11 14 8 3 Nicaragua 54 n/a 17 4 2 Panama n/a 33 n/a 7 8 Paraguay 71 19 9.4 7 4 Peru 64 68 16 9 2 Uruguay 32 27 10 5 5 Venezuela 30 33 12 13 10 Avg. Latin America 50 34 12 8 5 Source: Data from IADB and World Bank, <https://data.iadb.org/> and <https://data.iadb.org/>. At the same time, as Moreira et al. (2019, p. 118) note, “since trade is liberalized, there is no guarantee that it will remain liberalized.” Indeed, these authors distinguish 148 Author’s interview with Andrés Rebolledo, November 2019. See Appendix 5 for the full interview. 165 between two types of policy trajectories since the liberalization of trade in the 1980s and 1990s: “Chile, Colombia, Costa Rica, and Mexico, have continued to reduce trade barriers, signed preferential trade agreements with an increasing number of regional and extra regional partners, and persisted in maintaining and deepening open trade. … Argentina, Brazil, and Venezuela have resurrected some trade barriers or come up with new ones” (p. 118). Figure 5.7 – Latin America Exports and Imports Trends (1960-2019) in trillions Note: Exports of goods and services (constant 2010 US$), in trillion dollars Source: World Bank national accounts data, and OECD National Accounts data files. Figure 5.8 – Latin American Exports of Goods and Services (% of GDP) 166 Note: Exports of goods and services (constant 2010 US$), in trillion dollars. Source: World Bank national accounts data, and OECD National Accounts data files. Another important factor for understanding Latin American trade policy dynamics is the composition of trade in terms of sectors and destinations. As Table 5.4 shows, Latin America’s main trading partners for both imports and exports are the U.S., China, Canada, and Brazil. As well, the structure of exports and imports by product groups shows that the region is still a raw materials provider and a capital goods consumer. One pressing challenge is for the region to add value to its exports. Table 5.4 – Main Trade Characteristics of Latin America (2019) Trade Policy Information Data Top 3 export destinations U.S. (43%) China (12%) Canada (3%) Top 3 import origins U.S. (31%) China (18%) Brazil (5%) Exports by product groups Raw materials (30%) Intermediate goods (19%) Consumer goods (22%) Capital goods (23%) Imports by product groups Raw materials (7%) Intermediate goods (22%) Consumer goods (31%) Capital goods (37%) Main products exported Vehicles (13%) Mineral fuels (11%) Machinery (9%) Main products imported Electrical machinery (14%) Machinery (13%) Mineral fuels (12%) Source: World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. To fully understand trade policy in Latin America, specific attention must be given to the political economic context that periodically invokes backlash against free trade. Here, we must consider the mechanisms behind policy-making in the region. Within the executive branch of government, trade negotiations generally fall under the responsibility of ministries 167 of foreign trade or external relations. Occasionally, when a country has a ministry assigned to both of these roles, there can be tension between the roles in the policy-making process. Examples are found in the cases of Costa Rica (Monge-González & Rivera, 2018) and Mexico (Córdova Bojórquez et al., 2018). Further, when other ministries have constituents affected by trade policy, officials may seek out an administrative role in the implementation of trade policy (Moreira et al., 2019). In terms of the legislative branch, most countries require approval to pass trade legislation. In some countries, the legislature’s ability to introduce amendments before passing negotiations can slow this process (Moreira et al., 2019). Further, the level of the legislature’s involvement in trade policy-making can lead to different policy outcomes. For instance, legislatures that are strongly influenced by domestic business interests can produce more protectionist policies if this meets the demands of their constituents. Moreover, incumbent legislators seeking reelection may respond to popular sentiment on trade policy among those they represent (Conconi et al., 2014). The legislature can also serve an important role in providing checks and balances on executive power in the trade policy- making process. This role ultimately provides countries with a more consistent and predictable policy that is less susceptible to rapid, partisan changes. The private sector in Latin America plays a pivotal role in the formulation of trade policy. Consistent with economic theory, more efficient firms tend to enjoy open trade policies due to their comparative advantage, while less productive firms push to maintain trade barriers (IDB, 2018). Latin American producers, not consumers, have the largest stake in these policy negotiations and are the most politically engaged. Therefore, political 168 mobilization and organization mostly occurs at the level of producers. Domestic producers primarily seek representation through conventional lobbying efforts directed toward the executive and legislative branches at both the national and local levels. With the regional context in mind, the following subsections briefly analyze variables of interest in this dissertation from a regional angle. I analyze regional interconnectedness through the lens of PTAs and the kinds of protectionist tools that policy makers are deploying within this context. 5.4.1 Latin America Regional Integration Latin America is very well interconnected in terms of regional agreements. Recent studies have shown that “excluding those agreements that cover less than 80 of products, there are currently 33 LAC-to-LAC PTAs in force, including … CACM, Mercosur, CAN, and CARICOM.” This means that 80 percent of regional trade flows today are under some type of preferential agreements (Mesquita Moreira, 2018, p. 15). In general terms, Latin America is much more connected regionally than with the rest of the world. However, there is a lack of harmonization of trade rules. There are many subregional agreements, but there is a lack of coordination among them. That is why some studies have suggested that for better integration one of Latin America’s major policy alternatives is “to build bridges among the existing RTAs—strive to achieve some form of convergence or gradual harmonization of the various RTAs in the Americas and to implement cumulation of production among them” (Estevadeordal et al., 2009, p. 10). In the interview I conducted with Jorge Sahd, member of Chile’s Foreign Policy Council, he says: The restrictions in the region have a relevant component of non-tariff barriers, where the lack of regulatory convergence and excess administrative requirements end up making the integration process difficult. Undoubtedly, 169 internal structural reforms are necessary to maximize the benefits derived from trade. 149 Table 5.5 lists the most important regional agreements since the 1960s. Most of them have trade and economic objectives, such as creating a free trade zone or a common market. Others focus on political and social issues, for example, by creating regional intergovernmental organizations such as the Bolivarian Alliance for the Peoples of Our America (ALBA) and the Union of South American Nations (UNASUR). All of these have been crucial for regional integration in Latin America. 150 As the table shows, one of the first steps toward Latin American integration was taken when Argentina, Brazil, Chile, Mexico, Paraguay, Peru, and Uruguay signed the Treaty of Montevideo in 1960 (TM60), which established the Latin American Free Trade Association (LAFTA). Between 1961 and 1967, Colombia, Ecuador, Venezuela, and Bolivia also joined. The main objective of LAFTA was to create a free trade zone among the countries of Latin America over 12 years. Note that during this time period, the region’s economic model was based on ISI, as defined earlier. Hence, the main idea behind integration at that time was to encourage the economies of scale needed for industrialization (Tussie, 2009). During the 1980s, its members transformed LAFTA into the Latin American Integration Association (LAIA) through a revised 1980 Treaty of Montevideo (TM80), the main objective of which was to create a Latin American Common Market over an extended period. In the 1990s, Latin America started a new phase in its economic integration history defined by a proliferation of regional agreements (Albertoni & Rebolledo, 2018). As Table 5.5 shows, more than ten regional initiatives were developed between the 1970s and the beginning of the 21st century. 149 Author’s interview with Jorge Sahd, June 2020. See Appendix 5 for the full interview. 150 For an extended analysis of Latin America regional integration, see Albertoni & Rebolledo (2018). 170 Table 5.5 – List of the Regional Agreements in Latin America Regional bloc and original treaty Members and type of agreement Latin American Free Trade Association (LAFTA)/Latin American Integration Association (LAIA), Treaty of Montevideo 1960/1980. Members: 13 members a Type/objective: FTA (TM60), common market (TM80) Caribbean Community and Common Market (CARICOM), Treaty of Chaguaramas 1973. Members: 15 Caribbean nations and dependencies b Type/objective: trade and economic integration, common market Latin American and Caribbean Economic System (SELA), Panama Convention 1975. Members: 28 countries of Latin America and the Caribbean Type/objective: creation of a regional intergovernmental organization Rio Group, Declaration of Rio de Janeiro 1986. Members: signed by Argentina, Brazil, Colombia, Mexico, Panama, Peru, Uruguay, and Venezuela Central American Integration System (SICA), 1991, which succeeded the Central American Common Market (CACM), 1960. Members: created by Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama. Subsequently, Belize and the Dominican Republic joined as a full members Type/objective: creation of an institutional framework for regional integration in Central America Southern Cone Common Market (Mercosur), Treaty of Asuncion 1991. Members: Argentina, Brazil, Paraguay, Uruguay and Venezuela (joined in 2012, suspended in 2017) Type/objective: trade and economic integration, customs union North American Free Trade Agreement (NAFTA), 1994. Members: Canada, the United States, and Mexico Type/objective: trade and economic integration, FTA Andean Community of Nations (CAN), 1994, which succeeded the Andean Pact, Cartagena Agreement, 1969 Members: Bolivia, Colombia, Ecuador and Peru Type/objective: trade and economic integration, customs union. Bolivarian Alternative for Latin America and the Caribbean (ALBA), ALBA Establishment Agreement, 2004. Members: 11 countries d Type/objective: intergovernmental organization Dominican Republic-Central America FTA (CAFTA– DR), 2004 Members: United States, Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Dominican Republic Type/objective: trade and economic integration, FTA PetroCaribe, Puerto La Cruz, Venezuela, 2005. Members: 18 members e Type/objective: regional project to promote energy sector cooperation Union of the South American Nations (UNASUR), 2008, which succeeded the South American Community of Nations, 2004 Members: 12 South American nations f Type/objective: political/intergovernmental organization CELAC (Comunidad de Estados Latinoamericanos y Caribeños), 2011. Members: 33 countries in the Americas excluding Canada and the United States g Type/objective: political integration with some features of economic integration. Pacific Alliance (PA), Lima Declaration, 2011. Chile, Colombia, Mexico and Peru Type/objective: trade and economic integration, FTA. Notes: a LAIA: Argentina, Bolivia, Brazil, Chile, Colombia, Cuba, Ecuador, Mexico, Paraguay, Panama, Peru, Uruguay, and Venezuela. b CARICOM: Antigua and Barbuda, Bahamas, Barbados, Belize, Dominica, Grenada, Guyana, Haiti, Jamaica, Montserrat, Saint Lucia, St. Kitts and Nevis, St. Vincent and the Grenadines, Suriname, and Trinidad and Tobago. c SELA: Argentina, Bahamas, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominican Republic, Ecuador, El Salvador, Grenada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Suriname, Trinidad and Tobago, Uruguay, and Venezuela. d ALBA: Antigua and Barbuda, Bolivia, Cuba, Dominica, Ecuador, Grenada, Nicaragua, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines and Venezuela. e PetroCaribe: CARICOM (excluding Barbados, Montserrat, and Trinidad and Tobago), Venezuela, Cuba, and Dominican Republic. Haiti and Nicaragua joined the union at the third summit. Guatemala joined in 2008 but left the organization in November 2013. f UNASUR: Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guyana, Paraguay, Peru, Suriname, Uruguay, and Venezuela. g CELAC: Mercosur, UNASUR, SELA, ALBA. CELAC is the successor of the Rio Group and the Latin American and Caribbean Summit on Integration and Development (CALC). Source: Albertoni and Rebolledo (2018) based on the original treaties and the information provided on their webpage. 171 5.4.2 Trade Protectionism from a Regional Perspective Overall, Latin America has been very responsive to global protectionism in the last decade. Figure 5.9 shows that there is a positive relationship between protectionist dynamics in t-1 (from the RoW toward Latin America) and what happened in time t (Latin America toward the RoW). In conjunction with the regional integration that has increased over the past decades, the Latin American region has also experienced some policy reversals in the liberalization process. That is why the IDB recently has alerted that “the pace towards open markets has not been steady” (IDB, 2018, p. 1). More specifically, these policy reversals have “led to tariff increases or the establishment of non-tariff barriers including the pervasive use of instruments like antidumping, import license requirements, and quantitative restrictions on imports” (idem). Figure 5.9 – How Responsive is Latin America to Trade Protectionism? Note: Share of Country’s Total Exports Affected (t-1) vs. Share of World Export to the Country Affected (time t) by Protectionist Measures (Average 2009-2019). This is an average of share of trade affected in Argentina, Brazil, and Mexico. Source: Author’s creation based on the Global Trade Alert, <www.globaltradealert.org>. 172 In an interview with Olga Lucia Lozano, former Vice Minister of Foreign Trade of Colombia, 151 while discussing Latin American protectionism, she noted: Surely trade interconnection has generated less transparency in trade policy. Likewise, Latin America has been less successful than Europe, for example, in making its trade policy complex in such a way that it continues to protect without making it very obvious. The commercial policy officials today, seeing that in Europe they have applied increasingly transparent measures, feel that if we from Latin America apply some, it is more to level the playing field. 152 Given the region’s high interconnectivity though regional agreements, we would expect that Latin America as a region would not protect intra-regionally because most countries have at least some trade agreements with one another. Remarkably, the data show that Latin American has actually been consistently more protectionist. Moreover, 47 percent of the measures implemented within the region have been “murky instruments,” while the share of “murky instruments” leveraged by Latin America toward the G20 countries is 33 percent. Figure 5.10 shows Latin America trade policy measures within the region and toward the G20 countries between 2008 and 2019. Although the region has been more protectionist against G20 countries overall—with 1,578 measures versus 1,022 measures in the region—the proportion of non-transparent murky measures is almost the same: 530 murky measures against G20 and 490 murky measures within the region. Latin America has implemented considerably more liberalizing measures toward the G20 than it has intra- regionally. In the interview I conducted with Beatriz Leycegui Gardoqui, former Undersecretary of Foreign Trade of Mexico from 2006 to 2011, she observed that “the difference of Latin America with other blocs, such as the European Union and ASEAN, is that, within these 151 She was also former chair of the Subsidies Committee at the WTO. 152 Author’s interview with Olga Lucia Lozano, April 2020. See Appendix 5 for the full interview. 173 blocs, they have managed to liberalize trade. Unfortunately, the region has not achieved that. Intra-regional trade is very small.” 153 Figure 5.10 – Latin America Trade Policy Measures Within the Region and Toward G20 (2008-19) Note: When considering the G20, I removed Argentina, Brazil, and Mexico given that they are Latin America countries. Source: Authors’ creation based on GTA. Figure 5.10 shows that Latin America’s interconnectivity through regional trade agreements does not seem to be a decisive factor in deterring protectionism or murky instruments, i.e., non-transparent measures. On the prevalence of NTM in Latin America, Cadestin et al. (2016, p. 39) states the following: Measures arising from national regulation, such as sanitary and phytosanitary (SPS) measures and technical barriers to trade (TBT), have grown in importance globally in recent years while the usage of more traditional NTMs such as, for example, quantitative restrictions and non-automatic licensing has declined. The incidence of NTMs in Latin American countries is no higher than, for example, in Europe or Asia but there is important heterogeneity across Latin American countries, ranging from more than 40% coverage ratios in Argentina, Brazil, and Chile to less than 20% in some countries in Central America. This means that traders in different countries in the region face very different challenges in dealing with NTMs. 153 Author’s interview with Beatriz Leycegui Gardoqui, January 2020. See Appendix 5 for the full interview. 174 Additionally, during my fieldwork in Latin America, trade policy makers also highlighted that PTAs do not deter less transparent trade measures. For instance, in an interview with Magali Silva, former Minister of Foreign Trade and Tourism of Peru, she stated that: It did happen in the Peruvian case of applying protectionist measures in countries with which we have free trade agreements. … Indeed, when protection is not given at the tariff level but rather at the non-tariff level, through subsidies to local companies, the defense process requires an investigation that generally takes some time. Governments must be well-informed enough not to yield to power groups under political pressure. They must make decisions based on proven evidence for the good of the entire country and not based on private interests. During my three years as a minister, I fought for transparent management. Every time I met with the different actors in the sector, I informed the press about the agreements made. I had to negotiate and sign the Pacific Alliance and the TPP, and I had very positive experiences and fortunately few of the others. 154 Another perspective on the relevance of regional trade protectionism is whether there exists a pattern between the destinations of exports of intermediate goods by region and the types of measures implemented. As depicted in Figure 5.11, since 2008, Asia and the U.S. have consistently been the leading destinations for Latin American intermediate goods. Between 2008 and 2018, we can say that in terms of possible GVC linkages—measured by intermediate goods exports—the order of regional importance has been North America, Asia, Europe, Latin America, the Middle East, and Africa. 155 Interestingly, Table 5.6 shows a similar order regarding the total number of murky protectionist instruments implemented from Latin America toward other regions. In other words, regions where Latin America is exporting more intermediate goods are also those where Latin America has also implemented murkier instruments. 154 Author’s interview with Magali Silva, April 2020. See Appendix 5 for the full interview. 155 For an in-depth analysis of participation in GVCs in Latin America, see Cadestin et al. (2016) and Hernández et al. (2014). 175 Figure 5.11 – Latin America Exports of Intermediate Goods by Region Note: Exports in U.S. dollars by thousands. Source: World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. Table 5.6 – Latin America Intermediate Goods Destinations and Types of Measures Implemented Toward Those Destinations (2008-19) North America Asia Europe Latin America Middle East Africa Total number of protectionist measures implemented between 2008 and 2019 850 1054 1221 1022 353 202 Total number of murky protectionist instruments implemented between 2008 and 2019 450 306 342 480 166 117 Share of protectionist measures that use murky instruments 53% 29% 28% 47% 56% 58% Source: Author’s creation based on GTA, <https://www.globaltradealert.org/>. Table 5.7 shows the most affected sectors in Latin America by protectionism from RoW and the most affected sectors by Latin America’s protectionism toward RoW exports. In terms of the impact of protectionism on sectors and potential retaliatory dynamics, the following table tells two stories. First, in general terms, Latin America has been less protectionist toward the RoW than vice versa. At the same time, there were some sectoral retaliatory dynamics within industrialized values chains—such as motor vehicles, trailers and semi-trailers, and products of iron and steel—which is in line with Hypothesis 7. 176 Table 5.7 – Most Affected Sectors in and by Latin America (2009-19) Source: Author’s creation based on GTA, <https://www.globaltradealert.org/> 177 5.5 Country Case Studies This section concentrates on my three case studies. The methodology used to analyze each case study draws on Frieden (2000), who argues that the following components must be considered when conducting a political economic analysis of a given country case: knowledge of the recent history, identification of key actors and their goals, specification of actors’ policy preferences, and the aggregation of actors into groups and social institutions. The underlying goal of this analytical framework is to better understand policy outcomes, how government policies––for instance, trade policy retaliation––are adopted, and their associated results (whether, how, and when actors decide to reject, reform, or build political institutions). I begin with the case of Mexico. I will then turn to Brazil and Argentina. Again, as members of Mercosur (which aims to be a custom union), Brazil and Argentina are not entirely free to make unilateral trade policy decisions. As explained earlier, Mexico offers the opportunity to study trade policy dynamics in a country where unilateralism has been the main trade policy approach. Argentina and Brazil shed light on cases that operate regionally and need to coordinate with other regional partners to define and carry out trade policy. Moreover, despite this regional characteristic in the trade policy of Mercosur countries, my analysis shows that Argentina and Brazil have engaged in some trade policies that have their own unique dynamics. For the three cases, I contextualize each country’s trade policy evolution and identify the key actors involved. I then focus on specific protectionist policies and their impacts and responses (H1) from the perspective of the types of trade policy instruments used (H4 and H5), the effects of PTAs (H2 and H4), and how GVCs (H3 and H5) shape trade policy responses and sectoral dynamics (H6, H7, and H8). 178 5.5.1 The Case of Mexico In the last several decades, Mexico’s trade policies have undergone significant change. After almost three decades of import substitution industrialization (ISI) policies, Mexico initiated a trade liberalization process that resulted in its 1986 accession to the GATT and its entry into the North America Free Trade Agreement (NAFTA) with Canada and the U.S. in 1994. This section recounts Mexico’s trade policy evolution and the key actors at play. Following this contextualization, I focus on the specific impact of and responses to rising trade protectionism since the 2008-09 GFC. 5.5.1.1 Mexico’s Trade Policy Evolution and Key Actors Involved After 30 years of high growth under ISI, which was embedded in the country’s stabilizing development model, Mexico was one of the first countries in Latin America to embrace trade liberalization. As Table 5.3 shows, the weighted mean tariff rate applied to all products was 34 percent in the mid-1980s in Mexico, while the average in Latin America was 50 percent (Brazil reached 80 percent at the same time). In the 1980s, Mexico lowered its tariffs to 11 percent, while the regional average was 34 percent. Upon Mexico’s accession to NAFTA, its average tariff was 14 percent. Over the past two decades, Mexico has been substantially more open to trade, with an average tariff of 8 percent between 2000 and 2007, and 3 percent between 2008 and 2018; the regional average has been 5 percent over the same period (see Table 5.3). This rapid reduction of Mexico’s tariff explains, in part, why this country is one of the most open economies in the developing world today (Zabludovsky & Pasquel, 2010). Pastor & Wise (1994, pp. 462-463) identify three phases in Mexico’s trade policy from the 1980s to the 1990s: first, a “loosened restrictive import regime” under the administration of President Miguel de la Madrid (1982-88) implemented during the “debt crisis and continuing 179 balance-of-payments shortfall”; second, a “new four-step tariff reduction schedule” instated to “eliminate all official prices for imports and exports” and “promote external competition as a way to restrain domestic prices”; third, a process of “fine tuning” these tariff initiatives. Zabludovsky and Pasquel (2010) define the 1980s in Mexico as the decade of “unilateral” trade policies, in which there were two main events: the reforms implemented for the recovery from the 1982 debt shocks and the GATT accession. In the late 1980s, the administration of President Carlos Salinas de Gortari (1988-1994) intensified the reform process, with a main goal of harnessing “the unilateral liberalization of the 1980s … taking advantage of the accession to GATT in order to expand Mexico’s export markets in North America, Europe and East Asia” (Garciadiego & Hernández, 1994, p. 16). However, this approach shifted in the early 1990s in response to the fall of the Berlin Wall in 1989. In response to this new world order, President Salinas concluded that Mexico had “to take advantage of its geographical proximity to the largest economy in the world and become part of a trading bloc. Such thinking prompted Mexico’s proposal to the United States to negotiate a free trade agreement in the early 1990s” (Salinas De Gortari, 2002, pp. 47-48). After four years of negotiations, on January 1, 1994, the North America Free Trade Agreement (NAFTA) entered into force, making a radical turning point in Mexico’s trade policy. As Zabludovsky and Pasquel (2010, p. 92) state, “NAFTA creates the opportunities for negotiating similar agreements with other Latin American nations, the European Union, the European Free Trade Agreement (EFTA) countries, Israel and Japan. … [It] also strengthen[s] Mexico’s presence in other multilateral and regional forums.” 156 Thus, between 156 For an extensive analysis of Mexico’s trade policy with Latin America, see Pastor and Wise (1994) and De Mateo Veiga (2003). 180 1990 and 2019, Mexico negotiated 14 FTAs with over 50 countries (constituting 60 percent of gross world product) and joined various multinational organizations such as the Asia- Pacific Economic Cooperation (APEC) forum in 1993, the Organization for Economic Cooperation and Development (OECD) in 1994, and the Pacific Alliance in 2014. Mexico also joined the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP), signed in 2018. 157 Table 3.8 shows the main trade policy agreements signed by Mexico since the 1990s. Now, 82 percent of Mexico’s total exports are sent to countries with which it has standing trade agreements. 158 Table 5.8 – Mexico Trade Agreements, Blocs, and International Forums (199-2019) Date of Entry into Force Agreement 1992 FTA with Chile 1993 APEC a (21 members) 1994 NAFTA (Mexico, Canada, and the U.S.) OECD b (37 members) 1995 FTA with Costa Rica FTA with Bolivia FTA with Venezuela 1998 FTA with Nicaragua FTA with Uruguay 2000 FTA with Israel FTA with EU c 2001 FTA with EFTA FTA with Guatemala, Honduras, and El Salvador 2005 FTA with Japan 2012 FTA with Peru 2015 FTA with Panama 2016 Pacific Alliance (Mexico, Peru, Colombia, and Chile) 2020 USMCA (United States-Mexico-Canada) Notes: a APEC, 21 members, < https://www.apec.org/About-Us/About-APEC>; b OECD, 37 member, <https://www.oecd.org/about/members-and-partners/>; c EU, 27 members, <https://europa.eu/european- union/about-eu/countries_en>. Source: Zabludovsky & Pasquel (2010) and OAS Foreign Trade Information System, <www.sice.oas.org>. 157 OAS Foreign Trade Information System, Mexico’s Trade Agreement, <http://www.sice.oas.org/ctyindex/MEX/MEXAgreements_e.asp>. 158 ITC calculations based on UN COMTRADE, <https://www.trademap.org/>. 181 Another way to understand Mexico’s trade policy-making is to explore the actors involved. There are at least four key actors involved in Mexico’s trade policy-making process: the executive, the Senate, state actors, and private/nonstate actors. As Pastor and Wise (1994, p. 182) highlight: “an organizational overhaul, particularly at the level of the state, played a strong role in facilitating the successful implementation of Mexico’s trade reforms.” In an interview I conducted with Daniel Vázquez, a Mexican trade policy expert, he stated that in Mexico, very “influential sectors have lobbied the government, not only to respond but also to protect and safeguard their interests.” More specifically, Vázquez continued: Depending on the government and the technical capacity of its officials, these supports have been aligned with international commitments or have taken unorthodox forms, such as making the entry of merchandise more difficult by modifying customs requirements. For example, in the case of sugar in Mexico, the sector asked the government to protect access to the American market. At the same time, products competing with sugar—such as fructose—were proposed to be subject to retaliatory measures. When conflicts escalated to the USMCA negotiation, the same sectors demanded that access to the American market be protected, while protecting the local market. 159 The following table shows a summary of the key actors involved in Mexico’s trade policy-making. 159 Author’s interview with Daniel Vázquez, April 2020. See appendix 5 for the full interview. 182 Table 5.9 – Main Actors Involved in the Development of Mexico’s Trade Policy State actors The participation of state actors in trade negotiations is regulated by the Constitution. The Mexican Constitution states that any legislation that emanates from an international agreement entered into by the president and approved by the Senate becomes supreme law of the land and is, therefore, part of Mexican domestic law (Mexican Constitution:1475/98) (Zabludovsky & Pasquel, 2010, p. 99) The executive There are a number of agencies the executive is involved in with regard to the negotiation and implementation of international trade agreements. The Ministry of the Economy is in charge of the negotiation process and is aided by the Ministry of Agriculture, Ministry of Foreign Affairs, Ministry of Finance, Ministry of Interior, and officials from the Central Bank. The Senate Once a PTA is negotiated, it has to be signed by the executive and ratified by the Senate to become Mexican legislation. Private and nonstate actors The following institutions are normally consulted during trade negotiations: Confederation of Industrial Chambers of Mexico (CONACIM); the Employer’s Confederation of the Mexican Republic (COPARMEX); the Coordinator for Foreign Trade Business Organizations (COECE); and the Confederation of National Chamber of Commerce (CONCANACO). Note: For an extensive list of Mexico’s trade policy bodies and other trade-related institutions, see Appendix 5 of this dissertation. Source: Zabludovsky & Pasquel (2010, pp. 99-108) and OAS Foreign Trade Information System, <www.sice.oas.org>. I interviewed Juan Rodrigo Moreno, the Director of the Employer’s Confederation of the Mexican Republic (COPARMEX) for International Trade Negotiations, about the group’s role in the negotiation process: COPARMEX actively participates in the ‘Attachment Room,’ which is a group that has been created to favor the participation of the business sector in the negotiation processes of Free Trade Agreements in which Mexico is involved. In this group, we work in coordination with the federal government to know and jointly propose the terms of negotiation of any free trade agreement to be concluded by our country. This group has representatives of accredited companies in Mexico, from all productive sectors, as well as those responsible for the area and/or foreign trade issues of the various organizations and business chambers that exist in Mexico. In addition, we work closely with the Ministry of Economy, specifically with the Undersecretary of Foreign Trade, to learn about the plans, programs, and projects that are promoted by the federal government to boost the foreign trade of Mexican companies. We maintain a representation chair in these spaces, and we participate in the various events that are held by the Federal Government in this area. 160 Regarding the group’s position on trade protectionism, he said: We advocate fair trade, based on the guidelines of multilateralism and dispute settlement mechanisms established in each FTA, as well as those promoted by the WTO. However, we must also recognize that Mexico has a lot of experience in 160 Author’s interview with Juan Moreno, November 2019. See appendix 5 for the full interview. 183 tariff and non-tariff measures (depending on the circumstances), carousel strategies that have been used in the past to guarantee fair conditions, as well as to defend free market principles, commercial reciprocity, most favored nation treatment, etc. We have noticed that applying these types of measures often results in issues of illegality and lack of ethics. For this reason, COPARMEX is in favor of tariff policies which benefit national economies by allowing the strengthening of their industries by becoming much more competitive within the international market, which in the end brings greater benefits to their entrepreneurs, their workers and, therefore, to the economy of each country. 161 Finally, when I asked Moreno about how they react as an institution when they are affected by a protectionist measure from a close partner, he said that, in this situation: Tariff policies are synonymous with legality and ethics, values that COPARMEX defends and demands, so the position we take on this issue is the search for negotiations through the corresponding instances and mechanisms for dispute resolution, in order to ensure that our country can participate freely within the international market, demonstrating in this way the capacity and competitiveness of our industries, products, and services offered inside and outside the country. 162 Finally, as mentioned, Mexico has been very active in GVC participation, not necessarily as an intermediate goods exporter but rather as a capital goods exporter. Many products are thus at the end of the “factory line.” A recent study by The Economist (2020, p. 1) explores how well Latin America is prepared for post-COVID trade dynamics in regard to GVCs: Mexico seems best placed to increase its position in the U.S. supply chain, but there are issues here too. In the very near term, the problem will center around the country’s slow emergence from the health and economic effects of Covid-19. More fundamentally, Mexico will struggle without government policy to actively promote supply chain shifts, and without policy that attracts investment more broadly. 163 Figure 5.12 shows that Mexican exports have been growing consistently in terms of intermediate, consumer, and capital goods. The top ten destinations of intermediate, 161 Idem. 162 Idem. 163 In appendix 5 of this dissertation, see an extensive list of countries of Latin America and how they are prepared to be part of GVCs. 184 consumption, and capital goods (average share in total exports between 2008 and 2018) represent, on average, 83 percent of total exports of Mexico’s intermediate goods between 2008 and 2018. Table 5.10 shows five key players in Mexico’s exports and potential GVC dynamics. The U.S. is the major destination by far for all three categories. However, Canada, China, Colombia, and Brazil also appear as consistent destinations in the three categories. Figure 5.12 – Mexico’s Export of Intermediate, Consumer, and Capital Goods Between 2000 and 2018 (in thousands of dollars) Source: Author’s creation based on World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. Table 5.10 – Top Destinations of Intermediate, Consumption, and Capital Goods (Average Share in Mexico’s Total Exports Between 2008 and 2018) Consumption goods % in total exports Intermediate goods % in total exports Capital goods % in total exports U.S. 79% U.S. 65% U.S. 87% Canada 3% Canada 3% Canada 3% Germany 3% China 2% China 1% Colombia 2% UK 2% Brazil 1% Brazil 2% Colombia 3% Germany 1% China 1% Belgium 2% Netherlands 1% Chile 1% Brazil 3% Colombia 1% Argentina 1% Guatemala 1% France 1% Guatemala 1% Switzerland 2% Japan 0.5% Panama 1% Spain 1% UK 0.4% Source: Author’s creation based on World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. 185 5.5.1.2 Impacts of Mexico’s Trade Protectionism and Economic Responses Here, I concentrate on three main issues. First, I focus on how protectionist dynamics from Mexico’s main trading partners have affected Mexico in the last decade. For this task, I concentrate on those countries with which Mexico has close trade relationships. Second, I analyze how Mexico has affected these close partners with its own protectionist dynamics. For both cases, I analyze the types of measures countries have used and the specific sectors that have been affected. Finally, I test the differences between “cross” and “within” sectoral dynamics in Mexico’s retaliation. In light of the descriptive evidence presented, I predict that when Mexico is adversely affected by a protectionist measure adopted by another country, it will systematically respond to that country by adopting protectionist measures in return (H1). However, the existence of PTAs (e.g., NAFTA or the Pacific Alliance) decreases the likelihood of a protectionist response (H2). As I stated before, I expect that Mexico has been generally less affected by those countries with which it has a PTA. Finally, I believe that Mexico will retaliate more with non-tariff measures (less transparent) than tariff measures. With regard to the sectoral dynamics, I argue that Mexico’s protectionist dynamics are more likely to occur cross-sector rather than within-sector (H6); the higher the level of industrialization of a given sector, the lower the within-sector retaliation (H7); and protectionist responses are less likely with trade in intermediate goods (H8). Figure 5.13 reflects Mexico’s total exports affected (in t-1) versus world exports to Mexico affected (time t) by protectionist measures (average 2009-2019). As we can see, Mexico has been very responsive to protectionist measures. Overall, this confirms H1, which 186 states that, generally, when a country is adversely affected by a protectionist measure adopted by another country, it will systematically respond to that country by adopting protectionist measures in return. Figure 5.13 – Share of Mexico’s Total Exports Affected (t-1) vs. World Exports to Mexico Affected (time t) by Protectionist Measures (Average 2009-19) Source: Author’s creation based on the Global Trade Alert. Figure 5.14 shows that, in the last decade, Mexico’s share of exports has been consistently more impacted by those countries with which Mexico has no FTA. On average, between 2009 and 2019, 14 percent of Mexico’s exports were affected by close trading partners and 21 percent by those countries with no trade agreement with Mexico. In other words, having a PTA leads to a 7 percentage points reduction in the share of Mexico’s exports impacted by protectionism. In Figure 5.15, we can see the share of close and not close trading partners’ exports to Mexico impacted by Mexico’s protectionism. Since 2014, countries with no trade agreements with Mexico have been consistently more impacted by Mexico’s protectionism 187 than those with an FTA with Mexico. This is exactly in line with the results presented in Table 4.3, where we can see from a large-N analysis that the existence of a PTA between A and B is negatively correlated with protectionist dynamics. Figure 5.14 – Share of Mexico’s Exports Impacted by Close and Not Close Trading Partners’ Protectionism Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.15 – Share of Close and Not Close Trading Partners’ Exports to Mexico Impacted by Mexico’s Protectionism Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. 188 At the same time, my earlier results in Chapter 4 showed that GVCs don’t seem to restrain protectionism along the same line as do PTAs. In Figure 5.16, I apply the same exercise to those close trading partners through potential GVC ties with Mexico (e.g., the U.S., Canada, China, Colombia, and Brazil, as shown in Table 5.9). Interestingly, we see a very different scenario. On average, between 2009 and 2019, 36 percent of Mexico’s exports have been affected by close trading partners and 21 percent by those countries that have no PTA with Mexico. These results show that GVCs don’t seem to restrain protectionism as PTAs do. The case of Mexico thus confirms H2 on the negative relationship between protectionist dynamics and PTAs. Moreover, it doesn’t confirm H3, in line with the quantitative results presented earlier (model 3, Table 4.3). Figure 5.16 – Share of Mexico’s Exports Impacted by GVC Ties vs. No GVC Ties Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Relying now on the number of protectionist measures implemented, Figure 5.17 shows that since 2011, Mexico has consistently implemented more protectionist measures against countries with which it has no trade agreements than against close trading partners 189 (e.g., with a PTA in force). The same occurred with those measures that affected Mexico (Figure 5.18). It has been consistently more affected by countries with which it has no trade agreements. The only exception is in 2019, when the trade war between the U.S. and China spread to Mexico and then-U.S. President Trump threatened to impose an extraordinary escalating tariff on Mexico if it didn’t try harder to address the problem of migration flows from Central America. 164 Although much of this undocumented bilateral tension was at the rhetorical level, there were numerous non-transparent measures implemented by the U.S. toward Mexico. 165 Figure 5.17 – Mexico’s Number of Protectionist Measures Implemented Toward Close and Not Close Trading Partners Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. 164 For more information about this bilateral tension between the U.S. and Mexico, see Pérez Arguello & Albertoni, “Tariffs on Mexico? That Could Result in Trump’s Nightmare,” The New York Times, June 20, 2019, <https://www.nytimes.com/2019/06/20/opinion/tariffs-on-mexico-would-mean-more- immigrants.html>. 165 Examples are the reclassification of surfboard transport systems, resulting in the imposition of a tariff (implemented in November 2019), <https://www.cbp.gov/document/bulletins/customs-bulletin- weekly-vol-53-september-25-2019-no-34>; and $10 million in Agricultural Trade Promotion Program support to the U.S. Meat Export Federation (implemented in July 2019), <https://www.fas.usda.gov/atp-funding-allocations>. For other measures implemented from the U.S. toward Mexico in 2019, see <https://www.globaltradealert.org/country/222/affected- jurisdictions_132/period-from_20190101/period-to_20191231>. 190 Figure 5.18 – Number of Protectionist Measures Toward Mexico by Close and Not Close Trading Partners Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Focusing now on the level of transparency of trade measures, in Figure 5.19 we can see that the share of protectionist interventions that rely on non-transparent instruments used by Mexico toward close trading partners is much higher (on average, 47 percent between 2009 and 2019) than with countries that have no trade agreement with Mexico (32 percent on average between 2009 and 2019). Furthermore, the share of non-transparent protectionist interventions toward Mexico by close trading partners (82 percent) is higher than by countries with which it has no trade agreements (80 percent). This is in line with research proposition B (H4), which states that we would expect to see more non-transparent protectionist dynamics between close trading partners (e.g., through PTAs). 191 Figure 5.19 – Share of Non-Transparent Protectionist Interventions by Mexico toward Close and not Close Trading Partners (average 2009-2019) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.20 – Share of Non-Transparent Protectionist Interventions toward Mexico by Close and not Close Trading Partners (average 2009-2019) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Table 5.11 concentrates on sectoral dynamics in Mexico, which shows a cross- sectoral retaliatory dynamic. This can be explained by the fact that most of the sectors affected are embedded in GVCs, such as fabricated metal products, basic iron and steel, and 192 products made of iron or steel. 166 Hence, a retaliatory dynamic from within would severely impact consumers of Mexico’s key products. As the table shows, one of the sectors in Mexico most impacted by recent protectionism has been the steel sector. During my fieldwork, I interviewed Jesús Flores, the Director for Trade and Institutional Affairs at Tenaris Mexico, one of the most important global manufacturers and suppliers of steel pipes and related services. 167 Flores notes that the current protectionism against the steel sector can be explained by China’s oversupply of steel and its impact on global trade: The problem of excess capacity has been under discussion in the OECD Steel Committee for the last 10 years. In the year 2001, China became part of the WTO and negotiated a transition period of 15 years to complete its reforms and become a market economy. This did not happen because China did not modify its model. The U.S. reacted with different measures, but by the end of the Obama administration did not apply drastic actions. The [Mexican] steel sector has been affected by [section] 232 [Investigation on the Effect of Imports of Steel on U.S. National Security] and section 301 [of the Trade Act of 1974]. Now we live under administrated trade (quotas, tariffs, tariff-rate quota). 168 166 However, recent studies have shown that Mexico’s participation in GVCs has failed to capture enough value added, in part due to its low production interconnectivity with China (Chiquiar & Tobal, 2019). 167 Tenaris, <https://www.tenaris.com/en>. 168 Author’s interview with Jesús Flores, November 2019. See Appendix 5 for the full interview. Section 232 authorizes the President to “impose imports restrictions on products, imported into the United States in such quantities or under such circumstances as to threaten to impair the national security.” (See <https://ustr.gov/about-us/policy-offices/press-office/press- releases/2020/august/statement-presidential-proclamation>.) Under Section 301 of the Trade Act of 1974, the USTR can initiate “an investigation to determine whether China’s acts, policies, and practices related to technology transfer, intellectual property, and innovation are unreasonable or discriminatory, and burden or restrict U.S. commerce” (See <https://ustr.gov/issue- areas/enforcement/section-301-investigations/tariff-actions>.) 193 Table 5.11 – Sectors Most Affected in and by Mexico (2009-2019) Sectors Most Affected in Mexico (by Main Trading Partners’ Protectionism) Sectors Most Affected by Mexico’s Protectionism (Toward Main Trading Partners) By Close Trading Partners Products of iron or steel Other fabricated metal products Basic iron & steel Motor vehicles, trailers & semi-trailers Prepared & preserved fruits & nuts Products of iron or steel Cereals Sugar & molasses Food products Wearing apparel, except fur apparel By Countries with no Trade Agreements with Mexico Motor vehicles, trailers & semi-trailers Computing machinery & parts Electric motors, generators & transformers Electricity distribution & control apparatus Pharmaceutical products Products of iron or steel Other fabricated metal products Cereals Basic iron & steel Wearing apparel, except fur apparel Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.21 illustrates the percentage of employed persons within each key sector out of the total workforce in Mexico. In Figure 5.22, we can see that the industry sector was by far the most affected. When I link these two figures by sector and the total number of employed persons in Mexico, we can approximate the number of employed people who are affected by protectionism in Mexico. More specifically, an average of 12.3 million people working in the industrial sector were exposed to 12,406 protectionist measures from the rest of the world between 2009 and 2019 (90 percent of total measures that impacted Mexico); 27.3 million worked in the service sector, which were exposed to 862 measures (6 percent of total measures that impacted Mexico); and 5.9 million people worked in the agricultural sector, which was exposed to 500 measures (4 percent). In sum, 90 percent of the protectionist measures that affected Mexico were against a sector that represents, on average, 25 percent of employed persons in Mexico. 194 Figure 5.21 – Mexico’s Employment in Industry, Agriculture, and Services (% of total employment) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.22 – Total Number of Measures Affecting Mexico by Main Economic Sectors (2009-2019) Source: Author’s creation combining data from International Labour Organization, ILOSTAT database, and Global Trade Alert. 5.5.1.3 Political Economy Insights Based on the Case of Mexico The case of Mexico raises some interesting questions. For example, how have these protectionist measures interacted with the ongoing development strategy? Contrary to the case of Mercosur, NAFTA has integrated Mexico into powerful GVCs. However, because of 195 Mexico’s role as an assembler rather than an innovator, and its over-dependence on the U.S. market, NAFTA has hampered Mexico’s exports diversification and become a trap that prevents the country for retaining high value added. What are the implications for policy reform? During the 1970s, Mexico’s ISI development strategy had gone into crisis. The scarcity of fiscal resources, rising inequality, a dual agricultural sector, and an external debt crisis all stymied economic growth. These realities reflected the disconnect between Mexico’s socioeconomic and political systems (Riguzzi & de los Ríos, 2012). In the late 1980s, Mexico adopted WC policy recommendations from Washington, with mixed results. Since then, Mexico went through two additional economic crises in the past decades. First was the 1994 peso meltdown, which can be explained by a reliance on short-term flows of portfolio capital to achieve internal macroeconomic stability. This new financial context also reflected the perils of a rapid opening of the capital account, as advocated by the WC. Investors were incentivized to look for high returns in Mexico, but this created an asset bubble that exploded into a joint currency and banking collapse. The second major crisis was the 2008-09 GFC, when Mexican policy makers misread the shock from Wall Street and raised interest rates in 2009 due to a fear of inflation. This delayed the recovery and drove Mexican GDP to decrease, percentagewise, more than any other emerging economy in 2009 (Wise et al., 2015). The current development trajectory of Mexico is defined by its position in the United States–Mexico–Canada Agreement (USMCA). In short, Mexico’s unilateral trade opening and trilateral FTA with the U.S. and Canada have been a disincentive for its policymakers to foster solid forward and backward linkages for small- and medium-sized firms and the domestic economy overall. Mexican linkages to GVCs must expand beyond 196 North America. For instance, Japan, Germany, and Korea have incorporated Mexican contract assemblers into sophisticated GVCs. However, these account for low levels of trade and FDI. The country’s average GDP growth since 2000 is anemic and its per capita GDP growth is strongly negative. As Wise (2020, p. 195) states: “[A]s warning signs went unheeded and more promising routes were not taken, Mexico lost the economic dynamism that had carried it through the postwar era up to through the 1970s. However, the low growth and paltry per capita gains that have persisted since the early 1980s were far from predetermined. These policy failures have been self-inflicted.” Mexico’s resort to protectionist measures is testimony to these policy failures. Although the three main pillars of the Washington Consensus (WC)—liberalization, privatization, and deregulation—were quickly and rigorously adopted post-1988 in Mexico (Wise, 2020), Mexico is facing a major reversal on its previous progress and gains with trade liberalization. In other words, just as a wave of liberalization occurred as a result of Mexico’s strong political and economic ties with the U.S. and its eagerness to join NAFTA, current protectionist responses are a reaction to rising U.S. protectionism and a more “anti-trade” narrative that has emerged since the GFC. As a Dorotea López observes, “today [in Mexico] there is much less consensus on the advantages of trade openness that seemed very evident in all countries.” 169 In the same vein, Beatriz Leycegui Gardoqui)—who served as undersecretary of foreign trade at the Secretariat of the Economy of Mexico (2006-2011)— has observed: [After the liberalization reforms,] something that was not well calculated is that trade policy should have been accompanied by internal policies that addressed certain structural problems that were specific to each country ... . I think that the 169 Author’s interview with Dorotea López, April 2020. See Appendix 5 for the full interview. 197 impulse towards the outside was very strong but unfortunately it was not accompanied by other types of policies that also explain why so much inequality has been generated. So, in a context where this commercial opening and neoliberal policies did not seem to be the model to follow, there were no policies that addressed productive development to achieve more education, innovation, greater competition in the markets to have more accessible prices for the entire population. 170 The data I have reviewed thus far suggests that Mexico “has been a laggard in terms of institutional reform and aggregate and per capita growth” (Wise, 2020, p. 191). The lack of an industrial competition policy to complete trade policy reforms is a primary cause of Mexico’s underperformance. Moreover, policymakers must strive to secure better access to markets beyond North America. In 2013, Mexico started this process by introducing an ambitious program of reform in 11 different sectors, including (WTO, 2017b): The implementation of reforms around competition policy, tax policy, energy, financial services, and telecommunications, which walled for the amendment of the Constitution. A review of taxation was provided in order to improve tax collection. The measures adopted included an expansion of the base for income tax collection, fewer exemptions on payment of value added tax (VAT), revised taxation for the mining sector, and the introduction of “green” taxes. These led to a growth in tax revenue, from 9.7% of GDP in 2013 to 13.1% in 2015, while government finances became significantly less dependent on oil revenue. Despite these efforts, the public sector deficit rose from 2.3% of GDP in 2014 to 3.2% in 2015 and is forecast to be 3.5% in 2016. The efforts at reform have, nonetheless, managed to install a general climate of macroeconomic stability, although Mexico is still facing major challenges, particularly to the wide disparity in income, insecurity, and shortcomings in governance, as well as high dependence on a single market (p. 2). What this dissertation suggests is that Mexico’s dependence on the U.S. market is not just about trade. That is, NAFTA was more than just trade for Mexican producers and decision makers; it also was a tool to “stabilize the policy environment” (Pastor & Wise, 1994, p. 489). However, more than a quarter century later, this stabilizing tool has become 170 Author’s interview with Beatriz Leycegui Gardoqui, January 2020. See Appendix 5 for the full interview. 198 increasingly counter-productive for both domestic politics and U.S.-Mexico political and economic relations (Puyana, 2018). In this context, it is crucial for Mexico to craft a new and more proactive trade agenda, especially toward China. As Wise (2020, p. 191) states, “at the turn of the new millennium it was China that had become more problematic for Mexico, as reflected in the acerbic comments made by Xi Jinping 171 while visiting the country prior to assuming the Chinese presidency in March 2013. Until the mid-1990s, the two countries had enjoyed a cordial bilateral relationship on issues ranging from academic exchanges to science and technology collaboration.” At the same time, modernization of trade agreements signed already (beyond NAFTA) may be required. Interestingly, in April 2020, Mexico concluded negotiations with the EU to modernize and update the 1997 EU-Mexico Economic Partnership, Political Coordination and Co-operation Agreement (Rodríguez-Piñero, 2020). The modernization of this agreement with the EU can be a good roadmap to be continued by Mexico with other countries with which it has PTAs (Japan, Israel, Chile, Peru, Colombia, Bolivia, and countries in Central America). Mexico has attracted less Chinese FDI than smaller states like Chile and Ecuador, and its trade deficit with China accounts for 80% of Latin America’s total trade deficit. Rather than leverage the Sino-Mexico relationship to balance against U.S. trade and investment dominance, the country’s leaders have spurned Chinese overtures; the “renegotiation” of NAFTA to the USMCA in late 2018 was imposed upon Mexico and Canada by the Trump administration. The USMCA is historic in that this is the first time the U.S. has actually raised trade barriers in the context of a PTA. With the advent of the Biden administration in 2021, Mexican policy makers should now push to 171 “First, China does not export revolution; second, it does not export famine and poverty; and third, it does not mess around with you. So, what else is there to say?” See Xi Jinping, Wall Street Journal, October 19, 2010, <https://www.wsj.com/articles/SB10001424052702304410504575559623101416204>. 199 modernize USMCA along the lines of the updated EU-Mexico PTA. Among other things, the modernization of the EU-Mexico trade agreement includes (Rodríguez-Piñero, 2020): 1. In the area of agricultural goods, more than 85% of tariff lines would be fully liberalised while sectors such as dairy and meat would remain subject to restrictions such as quotas and tariff rate quotas. 2. In addition, it includes chapters on rules of origin, trade facilitation, trade remedies, technical barriers to trade, and sanitary and phytosanitary rules. 3. On services, the modernization would make it easier for EU firms to do business in Mexico (including in the maritime transport, telecommunications, and financial sectors), while protecting both parties’ right to regulate. 4. The modernized agreement would also include chapters on dispute settlement, anti-corruption measures, trade and sustainable development (TSD), transparency, energy and raw materials, small- and medium-sized enterprises, subsidies, competition, good regulatory practices, animal welfare and antimicrobial resistance, as well as annexes on motor vehicles and wine and spirits (p. 1). 5.6 Mercosur and a Brief Contextualization of Trade Policy in Argentina and Brazil In contrast to Mexico, Argentina and Brazil have been part of the same trade bloc (with Uruguay and Paraguay)—the Southern Common Market (Mercosur)—since the early 1990s. This means that these two neighbours have shared a trade policymaking institutional framework for thirty years (see section 5.2 for more detail on Mercosur’s decision-making institutions). 172 Again, Mercosur was founded in 1991 with the signing of the Treaty of Asuncion. 173 Venezuela became a full member in July 2012 (but was suspended in 2017 when an authoritarian regime came to power); Bolivia has been in the process of becoming a full member since 2012. Chile, Colombia, Ecuador, and Peru have remained associate members. Today, Mercosur is the most important bloc in South America, with more than 275 172 This subsection draws on Albertoni and Rebolledo (2018, p. 104-105). 173 See Treaty of Asuncion, <http://www.sice.oas.org/trade/MRCSR/treatyasun_e.asp>, accessed February 2016. 200 million inhabitants and a gross product of US$ 3.5 trillion (making it equivalent to the fifth- largest economy in the world). By 1986, after years of rivalry, Argentina and Brazil simultaneously transitioned to democratic rule. Newly elected presidents sought to enter into a cooperative relationship by creating the Argentine–Brazilian Economic Integration Program (ABEIP, PICAB in Spanish), aimed at economic and political cooperation. The political side of ABEIP tried to strengthen these infant democratic regimes, which had emerged after long periods of military rule in both countries. From an economic perspective, ABEIP sought to expand and diversify trade between the two countries with a special focus on agribusiness and the automotive sector (Manzetti, 1993). When President Carlos Menem in Argentina and Fernando Collor de Mello in Brazil took office (in 1990 and mid-1989, respectively), free market ideas about trade liberalization and integration were now becoming accepted in the region. As Manzetti (1993) points out, during those years “both countries re-affirmed their commitment to the integration effort. Indeed, economic integration was a key component of both their foreign policies” (p. 104). In 1990, after several meetings, Collor de Mello and Menem signed the Buenos Aires Act, which established the objective of creating a common market by the end of 1994. When Argentina and Brazil started to formalize this agreement, Paraguay and Uruguay joined the proposal. Finally, on March 26, 1991, the Foreign Ministers of the four countries signed the Treaty of Asuncion, which defined the common market as: The free movement of goods, services and factors of production between countries through, inter alia, the elimination of customs duties and non-tariff restrictions on the movement of goods, and any other equivalent measures; the establishment of a common external tariff and the adoption of a common trade policy in relation to 201 third States or groups of States, and the co-ordination of positions in regional and international economic and commercial forums; The co-ordination of macroeconomic and sectorial policies between the States Parties in the areas of foreign trade, agriculture, industry, fiscal and monetary matters, foreign exchange and capital, services, customs, transport and communications and any other areas that may be agreed upon, in order to ensure proper competition between the States Parties; the commitment by States Parties to harmonize their legislation in the relevant areas in order to strengthen the integration process. 174 Based on the conceptual underpinnings of a common market, the main objectives of the Treaty of Asunción were as follows (Manzetti, 1993): 1. An across-the-board-tariff reduction would replace the sector-by-sector approach used by the ABEIP. 2. The coordination of macroeconomic policies in accordance with the tariff reduction schedule, and the elimination of non-quantitative restrictions. 3. The establishment of a common external tariff for trade partners outside Mercosur, with the objective of increasing the competitiveness of the member countries. 4. The development of accords for specific sectors of the economy in order to optimize the use and mobility of production factors and achieve efficient economies of scale. 5. The implementation of an institutional framework to solve trade litigation (p. 105). Mercosur’s main goal is to “create a Customs Union in which goods, independently of where they are produced, can circulate free from import tariffs and certificates of origin within the integrated zone. In a complete Customs Union, the requirement of a certificate of origin for trade within the Union is considered to be equivalent to a non-tariff barrier” (Vaillant, 2005, p. 1). After sluggish progress in the 1990s, the Common Market Council (CMC) passed Decision 32/00 in 2000. This was meant to be a new impetus to deepen economic integration between Mercosur members. During the summit of Buenos Aires in June 2000, the members “re-launched” the process of integration in order to strengthen the bloc both internally and externally. Known as the “Re-launching Agenda of Mercosur” (CMC decisions 22/00 to 174 Treaty of Asuncion, <http://www.sice.oas.org/trade/MRCSR/treatyasun_e.asp>, accessed February 2016. 202 32/00), the strategy sought to identify Mercosur’s bottlenecks and to generate proposals to resolve them. In June 2001, a CMC meeting took place in Asuncion and emphasized the need to make progress in the following key areas (European Commission 2007): a) reformulation of the system of dispute settlements; b) identification and elimination of intra-regional barriers to trade (internal); c) elaboration of common trade disciplines to prevent the imposition of trade- distorting measures (internal); d) creation of a Free Trade Area of the Americas (external); e) creation of an Inter-Regional Association Agreement with the EU (external). Despite the many challenges Mercosur still faces—most of them derived from asymmetries among the four main partners—after the EU, it is nevertheless one of the longest lasting, most sophisticated, and deepest regional integration initiatives. Mercosur as a bloc currently has trade agreements with the countries listed in Table 5.12. Around 60 percent of Mercosur’s total exports go to countries with which the bloc has negotiated these agreements (UN COMTRADE, 2019). 175 Table 5.12 – Mercosur Trade Agreements Date of Entry into Force Mercosur with 1996 Chile 1997 Bolivia 2004 Colombia – Ecuador – Venezuela 2005 Peru 2006 Mexico 2007 Israel 2008 Cuba 2009 India 2010 Morocco 2016 Southern African Customs Union (SACU) 1 2017 Colombia Egypt 2019 2 EU Notes: 1 Botswana, Eswatini, Lesotho, Namibia and South Africa. 2 Reached an agreement in 2019, still not signed. Source: OAS Foreign Trade Information System, <www.sice.oas.org>; and Mercosur < https://www.mercosur.int/relacionamiento-externo/red-de-acuerdos/>. 175 In Brazil, 17 percent of total exports go to countries that have a trade agreement with Mercosur and in Argentina, 29 percent do so (ITC calculations based on UN COMTRADE, <https://www.trademap.org/>). 203 5.6.1 The Case of Argentina Keeping in mind the context of Mercosur, which is the basis upon which Argentina develops its trade policy, this section concentrates on the economic impact and policy responses of Argentine trade protectionism. First, I will briefly trace Argentina’s trade policy evolution beyond its 1991 accession to Mercosur. Argentina’s trade policy has been a roller coaster since the Peronist era in the 1950s. The Menem administration (1989-1999) quickly adopted the Washington Consensus policy reforms and moved to reduce and streamline trade barriers with the new Mercosur bloc. After an unprecedented economic depression between 1998 and 2002 (the economy shrank 28 percent during that period), due to the haphazard and reckless implementation of market reforms, Argentina swung back toward extremely protectionist trade policies during the administration of Nestor Kirchner (2003-2007) and Cristina Fernandez (2007-2015). 176 This, too, was a failure and led to a “return to the world” (volver al mundo) strategy during the Mauricio Macri administration (2015-2019) that followed. 177 Figure 5.23 shows trade policy trends during the administrations of Cristina Fernandez and Mauricio Macri, which covers the time period of my dissertation. As we can see, the number of protectionist measures under Macri decreases and the number of liberalizing measures increases in comparison with trade policies implemented under the Cristina Fernandez administration. 176 See Saxton (2003) for a detailed description of Argentina’s economic crisis and causes. 177 See O’Farrell et al. (2018) for an analysis of the political economy of trade policy in Argentina under Macri’s administration. 204 Figure 5.23 – Trade Policy During Cristina Fernandez vs. Mauricio Macri Source: Author’s creation based on Global Trade Alert. Pita (2017, pp. 35-39) distinguishes between two main periods of trade policymaking during the Kirchner and Fernandez administrations: Internal and external turbulence (2007-2011), in which Kirchner’s response to the growing inflationary problem was simply to sweep the problem under the rug, postponing any economic solution. In January 2007, he decided to understate the true inflation rate in the official statistics [given] a brief, yet intense, worldwide recession lasting approximately from spring of 2008 to winter of 2009… there were problems, but there were also successes: economic growth, high salaries, cheap tariffs, exchange rate appreciation, expansion of public spending, fiscal deficit, and the financing of this deficit through the Central Bank. 178 In an interview I had with Miguel Braun, Argentina’s Secretary of Commerce during the Macri administration, he recalls: Argentina was already in 2008 one of the most closed economies in the world, both by level of tariffs and by volume of trade. This closure was deepened in 2011 to 2015 with exchange controls, para-tariff barriers and the DJAI system, prior import control, which was punished in a WTO panel initiated by the United States, the EU and Japan. The result was a 30% drop in exports between 2011 and 2015, and the loss of about 5,000 of 15,000 exporting companies. This greater protectionism was not a response to the 2008 crisis, but a strategy to sustain an artificially low (appreciated) exchange rate that led to the loss of reserves. In this ultra-closed context, any request for protection from the private sector was generally met, generating higher costs for 178 Interestingly, between 2008 and 2009, around 85 percent of trade policies implemented by countries consisted of protectionist measures. However, only a third of these policies consisted of tariffs, while the remainder were non-traditional protections measures (Kosacoff et al., 2008, p. 36). 205 consumers (prices up to 3 times higher than in the U.S. for toys, clothing, electronics, and appliances) and loss of competitiveness for industries dependent on imported inputs. The government that assumed power in 2015 sought to implement a gradual opening of the economy, based on free trade agreements, initially with partners that did not represent a risk to Argentine production, and ensuring long terms of tariff reduction (EU-Mercosur, deepening of Mercosur and regional agreements such as Colombia, Chile, and Mexico). 179 Very different from his two immediate predecessors, Macri “considered the inward- looking orientation of the previous government a limitation to economic development. During the presidential campaign, Macri and his political coalition advanced an idea of a more open economy, publicly presented with expressions such as an ‘intelligent integration to the world’ (integración inteligente al mundo)” (O’Farrell et al., 2018, p. 12). In my interview with Luis Miguel Etchevehere, the Minister of Agriculture during the Macri administration, he highlighted this idea of more trade openness: Contrary to what is observed in some countries of the world, with protectionist advances, our government has decided to integrate itself into globalization, multilateralism, and commerce … . In these four years we have opened more than 250 markets for products from agriculture, livestock, and fisheries … . We find that there are many tools to help … . International agreements such as EU-Mercosur, multilateral agencies such as the WTO, and global forums such as the G20 are some examples. They are platforms from which a country can generate many instances of negotiation and interaction with others, to promote dialogue and trade. Undoubtedly, these institutional tools make the effects of a protectionist trend soften, and for our country they are essential. 180 Despite these noble intentions, recent studies have identified some protectionist tools that were adopted during the Macri administration. O’Farrell et al. (2018, p. 12) summarize these trade policies as follows: 1) lower tariff barriers, within the limits imposed by Mercosur CET; 2) replacement of the system of non-automatic import licenses (NALs) implemented by the previous administration for a new scheme of licenses that respected the rules of the WTO (and that would be progressively dismantled); 3) changes in the 179 Author’s interview with Miguel Braun, October 2019. See Appendix 5 for the full interview. 180 Author’s interview with Luis Miguel Etchevehere, November 2019. See Appendix 5 for the full interview. 206 methodology of anti-dumping (AD) measures; and 4) increased use of technical regulations in replacement of non-tariff barriers to trade, such as non-automatic import licenses. One useful way to understand Argentina’s protectionism during the time period of my study (2008-2019) is to analyze the different types of products exported. Figure 5.24 shows that Argentina’s export of intermediate, consumption, and capital goods has been declining since 2008, which reflects the impact of the blanket protectionism embraced by the Kirchner and Fernandez administrations. In contrast with Mexico, Argentina is weakly connected to GVCs as an exporter of both capital and intermediate goods. Table 5.13 shows the five key destinations for Argentina’s exports and potential GVC dynamics. Brazil is by far the major destination for Argentina’s consumption and capital goods. Chile, the U.S., Uruguay, and Paraguay also appear as consistent destinations in all three product categories. Interestingly, China is not among its main trading partners in any of the three categories. Figure 5.24 – Argentina’s Export of Intermediate, Consumption, and Capital Goods Between 2000 and 2018 (in U.S. dollars) Source: Author’s creation based on World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. 207 Table 5.13 – Top Destinations of Intermediate, Consumption, and Capital Goods (Average Share in Argentina’s Total Exports, 2008-2018) Consumption goods % in total Consumption goods’ exports Intermediate goods % in total Intermediate goods’ exports Capital Goods % in total Capital goods’ exports Brazil 40% Brazil 10% Brazil 50% Chile 12% U.S. 6% Chile 5% U.S. 8% India 5% Mexico 5% Uruguay 5% Canada 4% U.S. 5% Paraguay 4% Netherlands 4% Germany 3% Netherlands 2% Spain 4% Venezuela 3% Venezuela 2% China 3% Uruguay 3% Bolivia 1% Vietnam 3% Colombia 2% Mexico 1% Indonesia 3% Paraguay 2% Colombia 1% Switzerland 3% Peru 2% Source: Author’s creation based on World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. 5.6.1.1 Argentina’s Trade Protectionism: Economic Impacts and Responses In this section, I first concentrate on how protectionist measures imposed by Argentina’s main trading partners have affected Argentina over the past decade. In this regard, I concentrate on those countries with which Argentina has close trade relationships through trade agreements and potential GVC ties (Table 5.12). Second, I analyze how Argentina has responded to these close partners and the effects of its own protectionist measures. Here, I study the types of measures countries have used and the specific sectors that have been affected. Finally, I explore how trade protectionism has impacted Argentine employment in industry, agriculture, and services. In light of the descriptive evidence I have in hand, I expect to find the following. First, if my theorical intuitions are correct, I expect that Argentine policy makers will systematically respond to adverse protectionist measures adopted by another country and adopt protectionist measures in return (H1). Second, the existence of PTAs (e.g., Mercosur) decreases the likelihood of a protectionist response overall (H2). Again, I expect that, as a 208 member of Mercosur, Argentina has been less affected by those countries with which the bloc has PTAs. Third, I expect that Argentina will retaliate more with (less transparent) non- tariff measures than tariff measures. With regard to sectoral dynamics, I expect that Argentina will respond with cross-sectoral measures rather than within-sector ones (H6); the higher the level of industrialization of a given sector, the lower will be the within-sector retaliation (H7); and protectionist responses will be less likely with trade in intermediate goods (H8). On my first assumption, Figure 5.25 shows the share of Argentina’s total exports affected (in t-1) versus how world exports to Argentina are affected (in t) by protectionist measures (on average by year between 2009 and 2019). As we can see, Argentina has been especially responsive to protectionist measures up until 2015. Since 2016, the first full year of Macri’s administration, the level of protectionism targeted toward Argentine exports remains the same, but its protectionist responses decreased. This is in line with Figure 5.23, where we can see that during the Fernandez Administration, Argentina was more protectionist (and responsive in turn). Overall, until 2015, the case of Argentina confirms H1, which states that when a country is adversely affected by a protectionist measure adopted by another country, it will systematically respond to that country by adopting protectionist measures in return. 209 Figure 5.25 – Share of Argentina’s Total Exports Affected (t-1) vs. World Export to Argentina Affected (time t) by Protectionist Measures (Average 2009-2019) Source: Author’s creation based on the Global Trade Alert. In Table 5.12, I depicted Argentina’s main trading partners through trade agreements. The following two figures discriminate between close (e.g., PTA) and not close trading partners. Table 5.26 shows that up until 2014, Argentina’s share of exports was more impacted by protectionism on the part of not close trading partners. Since 2014, Argentina has been similarly impacted by protectionist measures of close and not close trading partners. This can be explained by the increasing protectionism of Brazil, as this is the key destination of Argentine exports. On average, between 2009 and 2019, 50 percent of Argentina’s exports have been affected by close trading partners and 47 percent by those countries with which it has no trade agreements. In other words, for the case of Argentina, PTA membership has not reduced protectionist measures, but rather increased them. Figure 5.27 shows the share of exports from close and not close trading partners impacted by Argentine protectionism. Remarkably, the pattern of protectionism between close and not close trading partners moves 210 in tandem. This evidence for Argentina does not support the results from the large-N analysis presented in Table 4.3, where we saw that the existence of a PTA between A and B is negatively correlated with protectionist dynamics. This can be explained by the fact that Argentina does not have PTAs with Chile or the U.S., two other key market destinations for its exports. Moreover, Brazil, the top trading partner of Argentina, has become increasingly protectionist, despite its Mercosur membership. From a policy perspective, this result shows that countries like Argentina would do well to sign more trade agreements, especially strategic ones. That means signing agreements with those countries with which Argentina has trade complementarities. 181 However, Argentina is hampered from doing this because of its Mercosur membership. In the final section of this chapter, I concentrate on how modernizing reforms could allow greater flexibility for this customs union. Figure 5.26 – Share of Argentina’s Exports Impacted by Protectionism from Close and not Close Trading Partners Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. 181 On this debate about signing PTAs with countries where there are strong trade complementarities, see Albertoni et al. (2020). 211 Figure 5.27 – Share of Close and not Close Trading Partners Exports to Argentina Impacted by Argentina’s Protectionism Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Considering the number of measures implemented (and not the share of trade affected), Figure 5.28 shows that Argentina has consistently implemented more protectionist measures against countries with no trade agreements than toward close trading partners (e.g., with a PTA in force). Even more consistent is the effect of protectionist measures by countries with no trade agreements on Argentina (see Figure 5.29). This evidence for Argentina supports the results from the large-N analysis, where the existence of a PTA between A and B is negatively correlated with protectionist dynamics. 212 Figure 5.28 – Argentina’s Number of Protectionist Measures Implemented toward Close and Not Close Trading Partners Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.29 – Number of Protectionist Measures toward Argentina by Close and Not Close Trading Partners Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. As for the level of transparency in protectionist interventions, in Figure 5.30 we can see that the use of non-transparent instruments by Argentina toward close trading partners is much higher (33 percent on average between 2009 and 2019) than for countries with which it 213 has no trade agreements (24 percent). This is in line with research proposition B (H4), which states that we would expect to see more non-transparent protectionist dynamics between close trading partners (e.g., through PTAs). Ricardo Rozemberg, an Argentine economist and specialist in international economic relations at Center iDeAS/UNSAM, spoke with me about this phenomenon: In Argentina these discussions are less subtle, and the logic is “in a protectionist world, we have to be protectionist.” While in Mexico, for example, they tell you “although NAFTA was not all that good, we increased exports and jobs.” And while the previous government (of Macri) did not share this protectionist vision, the government now [of Alberto Fernandez] agrees more in this regard. 182 Figure 5.30 – Share of Non-Transparent Protectionist Measures by Argentina toward Close and Not Close Trading Partners (average 2009-2019) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. 182 Author’s interview with Ricardo Rozemberg, October 2019. See Appendix 5 for the full interview. 214 Figure 5.31 – Share of Non-Transparent Protectionist Measures toward Argentina by Close and Not Close Trading Partners (average 2009-2019) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Finally, Table 5.14 concentrates on sectoral dynamics for the case of Argentina. Here, there are important within-sector retaliation dynamics in the industrial sector, including motor vehicles and trailers & semi-trailers. Recall that H7 says we would not expect to find much within-sector retaliation in those sectors with high levels of industrialization. The reverse finding for Argentina can be explained by the fact that it is poorly integrated into GVCs. With regard to these sectoral dynamics, Ricardo Rozemberg continued: The sectors where Argentina has problems are the same sectors where everyone has a problem: clothing (textiles), footwear, which are more automated sectors, that generate employment, and today are most exposed to competition internationally. The largest sectors, automotives, iron and steel, are the ones with more problems abroad, where Argentina exports. But above all the most problematic is that of iron and steel. In this sector there are two important companies: Techin and Acindar. Techin (mostly Argentine capital), which is the largest, does two types of things: flat pipes and tubes that are not traded goods. These companies have private and public reactions (to protectionism). On the private ones, since they are transnational, they invest in other countries to break through the barriers. For example, they put a plant in the U.S., in Asia. On the other hand, in the domestic market they ask to be protected against Chinese and Russian competition, and against global overproduction and low prices. 183 Table 5.14 – Most Affected Sectors in and by Argentina (2009-2019) 183 Author’s interview with Ricardo Rozemberg, October 2019. See Appendix 5 for the full interview. 215 Most affected sectors in Argentina Most affected sectors by Argentina Close Trading Partners Cereals Motor vehicles, trailers & semi-trailers Sugar & molasses Pharmaceutical products Pumps, compressors, hydraulic & engines Motor vehicles, trailers & semi-trailers Other fabricated metal products Other general-purpose machinery & parts Prepared & preserved fish, crustaceans Wearing apparel, except fur apparel Not Close Trading Partners Products of iron or steel Pharmaceutical products Fruits & nuts Prepared & preserved fruits & nuts Motor vehicles, trailers & semi-trailers Other fabricated metal products Motor vehicles, trailers & semi-trailers Domestic appliances & parts Other general-purpose machinery & parts Wearing apparel, except fur apparel Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.32 illustrates the percentage of workers in each sector out of the total number of employed persons in Argentina. In Figure 5.33, I divide Argentine employment into the three main sectors of the economy (industry, services, and agriculture), and we can see that the industry sector has been most affected by protectionist measures. More specifically, between 2009 and 2019, an average of 2.5 million people worked in the industrial sector, which was exposed to approximately 5,800 protectionist measures from the rest of the world (80 percent of total measures that impacted Argentina). Over 33 million worked in the services sector, which was exposed to 850 measures (5 percent of total measures that impacted Argentina). And 0.2 million people worked in the agriculture sector, which was exposed to 862 measures (6 percent). In sum, 80 percent of the protectionist measures that affected Argentina were against the industrial sector. Interestingly, although protectionism in the industrial sector is highest, it has less impact on the workforce. Moreover, if we consider that Brazil is the top destination of Argentine exports, and that, on average, 43% of its exports to Brazil were impacted by protectionist measures (Global Trade Alert, 2020), this alerts us to some serious bottlenecks within Mercosur. Figure 5.32 – Argentina’s Employment in Industry, Agriculture, and Services (% of total 216 employment) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.33 – Number of Protectionist Measures Affecting Argentina by Main Economic Sectors (2009- 2019) Source: Author’s creation combining data from INDEC, International Labour Organization, ILOSTAT database, and Global Trade Alert. 5.6.2 The Case of Brazil 217 This section investigates the economic impacts and responses of trade protectionism in the Brazilian case. As in the case of Argentina, I analyze Brazil’s trade policy as a country, beyond its membership in Mercosur. Brazil has maintained a relatively closed economy compared to other countries in Latin America. From the 1930s to the late 1960s, the Brazilian economy was dominated by an import substitution industrialization model. Under this model, Brazil established successful automobile and steel production industries, marked by import barriers, subsidies, and state sponsorship. However, after fifty years of these protectionist policies, Brazil’s role in global trade was disproportionately low given its population. To be sure, Brazil was not immune to the Latin American debt crisis of the 1980s. The neoliberal Washington Consensus (WC) policies that emerged in the 1990s led to privatization and trade liberalization throughout the region. However, foreign trade strategies in Brazil maintained a protectionist industrialization agenda compared to those of neighboring countries. Further, these WC policy changes were not followed up with further tariff reductions in Brazil, and non-tariff barriers to trade have increased since then. This agenda, coupled with passive involvement in preferential trade agreements, has hindered exports, economic growth, and allowed protectionism to flourish. Brazil’s historic lack of exposure to global trade fostered the economic success of certain industries that remained protected even after the trade reforms of the 1990s. Overall, manufacturing in Brazil represents a smaller portion of GDP than it did under import substitution industrialization. Industrial sectors like automobiles and capital goods that thrived under Brazil’s import substitution strategy receive a large share of foreign direct 218 investment, mainly because they have retained preferential treatment. Analysis of the prominent actors and sectors in Brazilian trade policymaking reveals the roots of protectionism in the country. Special interest groups can assert political influence because of their ample resources for lobbying and because of the substantial number of workers in the industrial sector. Other more competitive sectors that emerged under export-oriented strategies, predominantly agriculture, have failed to effectively challenge this historic industrial protectionism. In the time period of this study, exchange rate over-valuation has also favored industry over agriculture. The institutional framework in which trade policy is made sheds further light on how domestic actors have managed to maintain protectionist policies in Brazil. Structural changes have occurred within civil society and state institutions due to two main forces: the unilateral nature of post-debt-crisis liberalization, and Mercosur. The liberalization that followed the 1980s debt crisis led to the mobilization of civil society, especially in the business sector. This was later influenced by the launching of hemispheric negotiations for a Free Trade Area of the Americas (FTAA), to be completed by 2005. Ultimately, Brazilian internal institutional restructuring, which concentrated power in the Ministry of Development and Foreign Trade, gave rise to greater influence and political power within the industrial sector as a force in Brazilian trade policymaking. Another important Brazilian entity is the Foreign Trade Chamber (CAMEX), an inter- ministerial body that coordinates different institutions involved in the trade policymaking process. Originally formed in 1995 under the office of the president, the Foreign Trade Chamber was moved to the Ministry of Development and Foreign Trade. Specialized bodies 219 in the Foreign Trade Chamber advise their local business councils on specific policy issues, which cover all areas of trade, including tariff policy, trade defense, trade negotiations, and trade facilitation. Meanwhile, in the 1990s, consultation and coordination mechanisms were developed to account for the new relevance of trade policy negotiation even though these changes ran counter to Mercosur and the ongoing FTAA negotiation. On tariff policy, existing “exception mechanisms” enabled the shift of product-specific tariffs between levels, which provided lobbyists a way to exert influence. Additionally, special import regimes have the ability to exempt or suspend certain tariffs. This process is costly and bureaucratic, favoring larger, richer companies that can bear the burdensome costs. These two mechanisms in particular render Brazilian trade policymaking vulnerable to the demands of special domestic interests seeking protection. In 2001, the Foreign Trade Chamber was given the discretion to apply trade defense measures on investigations conducted by the Department of Trade Defense. The conclusions from the Department of Trade Defense are legally binding and the Foreign Trade Chamber must adopt them. Exceptions to these legally binding recommendations exist in cases of public interest arising in the private sector or with another CAMEX member. Within the government, many public entities that are not directly involved with trade issues were brought into the discussion and formulation process. These mechanisms incorporated representatives from civil society as well. Indeed, the mobilization of civil society in trade policy incentivized stronger connections between public and private actors. In the public sector, mechanisms to 220 strengthen these connections were implemented. In the executive branch, the National FTAA Secretariat was created under the Ministry of Foreign Affairs to monitor hemispheric negotiation groups. These groups sometimes allowed participation by the private sector. In the legislative branch, Congress must ratify negotiated trade agreements. This can give the private sector a voice in the policymaking process. For instance, the Congress rejected a commitment made to the WTO to eliminate import duties under pressure from the Textile Parliamentary Front. Finally, the following graph and table show Brazil’s export of different types of products (intermediate, consumer, and capital goods). Figure 5.34 shows that, after a sizeable decrease in 2008, Brazil’s exports of intermediate, consumer, and capital goods have been growing in the last decade. Table 5.15 shows that Mercosur countries, the U.S., and the Netherlands are important destinations for intermediate, consumption, and capital goods (as is China, but only for intermediate goods). Figure 5.34 – Brazilian Exports of Intermediate, Consumer, and Capital Goods Between 2000 and 2018 (in US$) Source: Author’s creation based on World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. Table 5.15 – Top Destinations of Intermediate, Consumption, and Capital Goods (Average Share in Brazil’s Total Exports Between 2008 and 2018) Consumption goods % in total Consumption Intermediate goods % in total Intermediate Capital Goods % in total Capital goods’ 221 goods’ exports goods’ exports exports Argentina 20% United States 15% United States 23% United States 14% Netherlands 9% Argentina 13% Netherlands 7% China 9% Netherlands 12% Paraguay 4% Argentina 7% Mexico 6% Mexico 3% Germany 3% Singapore 4% Belgium 3% Italy 3% Chile 4% Singapore 3% UK 3% Panama 4% Uruguay 3% Canada 3% Germany 4% Chile 3% India 2% Colombia 2% Colombia 2% Korea 2% Paraguay 2% Source: Author’s creation based on World Bank-World Integrated Trade Solutions (WITS), <https://wits.worldbank.org/countrysnapshot/en/LCN>. 5.6.2.1 Brazil’s Trade Protectionism: Economic Impact and Responses Figure 5.35 shows the share of Brazil’s total exports affected (in t-1) versus world exports to Brazil affected (time t) by protectionist measures (average by year between 2009 and 2019). As we can see, Brazil has been very responsive overall to protectionist measures since 2012, which marked the first year of Dilma Rousseff’s (of the Worker’s Party) Presidency. Since then, the level of impact of protectionism toward Brazilian exports has remained the same, but the responses have increased. Overall, the case of Brazil confirms H1, which states that, in general terms, when a country is adversely affected by a protectionist measure adopted by another country, it will systematically respond to that country with protectionist measures in return. 222 Figure 5.35 – Share of Brazil’s Total Exports Affected (t-1) vs. World Export to Brazil Affected (time t) by Protectionist Measures (Average 2009-2019) Source: Author’s creation based on the Global Trade Alert. Considering the information above about Brazil’s main trading partners (Table 5.13) through trade agreements, the following figures distinguish between close (e.g., PTA) and not close trading partners. First, Table 5.36 shows that during the period of analysis, Brazil has been similarly impacted by protectionism on the part of close and not close trading partners. This can be explained by the increasing protectionism of key trade partners (China, U.S., and Mercosur), which are the main destinations for Brazilian exports. Figure 5.37 shows the share of Brazil’s protectionism and its effect on close and not close trading partners. We can see that there is little difference in the impact of Brazil’s protectionism between close and not close trading partners. Figure 5.36 – Share of Brazil’s Exports Impacted by Close and not Close Trading Partners’ Protectionism 223 Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.37 – Share of Close and not Close Trading Partners’ Exports to Brazil Impacted by Brazil’s Protectionism Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Considering now the number of measures implemented (and not the share of trade affected), Figure 5.38 shows that Brazil has consistently implemented more protectionist measures against countries with which it has no trade agreements than toward close trading partners (e.g., with an FTA in force). Even more consistent is the case that those protectionist 224 measures that most affected Brazil are from countries with which it has no trade agreements (Figure 5.39). This evidence for Brazil supports the results presented in Table 4.3, where we can see from a large-N analysis that the existence of a PTA between A and B is negatively correlated with protectionist dynamics. Figure 5.38 – Brazil’s Number of Protectionist Measures Implemented Toward Close and not Close Trading Partners Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.39 – Number of Protectionist Measures toward Brazil by Close and not Close Trading Partners Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. 225 Focusing now on the level of transparency in trade measures, Figure 5.40 shows that the share of protectionist non-transparent instruments used by Brazil toward close trading partners is much higher (74 percent, on average, between 2009 and 2019) than with countries with which it has no trade agreements (56 percent). This is in line with research proposition B (H4), which states that we would expect to see more non-transparent protectionist dynamics between close trading partners. I quote from an interview I had with Tatiana Palermo, former Brazilian Vice-Minister of Agriculture and chief agricultural trade negotiator: Mercosur is a good example … . Although the countries do not apply tariffs among them (with some exceptions), there are numerous non-tariff and other barriers (like, for example, extremely complex rules of origin, and different sanitary and phytosanitary requirements that make agricultural trade within the bloc very difficult). I agree with your statement that governments are influenced by interest groups that advocate for more protectionism and import substitution. But I would not say that, in the case of existing trade agreements, this happens in a less transparent and perceptible way. Non-tariff barriers in Brazil are usually not directed toward specific countries. At the same time, it is usually more difficult for the Brazilian government to justify such barriers within Mercosur. 184 Figure 5.40 – Share of Protectionist Interventions that Use Non-Transparent Measures by Brazil toward Close and not Close Trading Partners (average 2009-2019) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. 184 Author’s interview with Tatiana L. Palermo, May 2020. See Appendix 5 for the full interview. 226 At the same time, Brazil has been consistently more affected by non-tariff barriers from not close trading partners (Figure 5.41). This is probably because Brazil (and Mercosur as a bloc) is a global player and many countries implement barriers against it. In an interview I had with Welber Barral, former Secretary of International Trade in Brazil, he said: If we look at all the countries to which Brazil exports, we can see that Brazil suffers a lot from non-tariff protectionism, such as health barriers, especially with meat. It is with these types of measures that Brazil has had the most problems. If you look at the cases that Brazil has brought to the WTO, you can see that many are cases of anti-dumping and some cases of subsidies, as was the case with Canada on the matter of Embraer, or the case against the United States on the subject of cotton. For Brazilian exports, the vast majority of cases are derived from subsidy measures that were used by importing countries of our products. If we take the case of Brazilian exports to Europe, there are many sanitary barriers and also very high tariffs. What I can say is that, in the case of Brazil, the 2008 crisis did not greatly affect Brazilian exports, but rather followed the practices that already existed. 185 Figure 5.41 – Share of Protectionist Interventions that Use Non-Transparent Measures toward Brazil by Close and not Close Trading Partners (average 2009-2019) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Table 5.16 concentrates on sectoral dynamics. As we can see, for the case of Brazil, there is an important “within” sector retaliation in motor vehicles, trailers & semi-trailers, and fabricated metal products, which are highly industrialized sectors. This is not in line with 185 Author’s interview with Welber Barral, June 2020. See Appendix 5 for the full interview. 227 H7, which predicts that we would not expect to find much within-sector retaliation in sectors with high levels of industrialization. This can be explained by the fact that Brazil (and Mercosur in general) is producing outside of GVCs. I discussed these sectoral dynamics with Renato Flores, Professor and Special Aide to the President of Fondação Getulio Vargas in Rio de Janeiro: Much like the dynamic in most Western economies, threatened sectors lobby for protection or retaliation before the government. A whole governmental apparatus exists––again not much different from those existing in many countries––for channeling such demands. These usually end up in a conciliatory agreement or with a standard WTO panel. Other actions may involve specifically designed non-tariff barriers, usually imposed by regulatory or supervisory agencies, that may cause effective damage to those which started the measure. Finally, with political support, taxation may also be used for certain retaliations. 186 Table 5.16 – Most Affected Sectors in and by Brazil (2009-2019) Most affected sectors in Brazil (by main trading partners’ protectionism) Most affected sectors by Brazil’s protectionism (toward main trading partners) Close Trading Partners Sugar & molasses Motor vehicles, trailers & semi-trailers Products of iron or steel Pumps, compressors, pneumatic & engines Other general-purpose machinery & parts Electric motors, generators & transformers Basic organic chemicals Motor vehicles, trailers & semi-trailers Electrical energy Other fabricated metal products Not Close Trading Partners Products of iron or steel Basic iron & steel Other fabricated metal products Motor vehicles, trailers & semi-trailers Aircraft & spacecraft Basic organic chemicals Machinery for mining & construction Electric motors, generators & transformers Other fabricated metal products Motor vehicles, trailers & semi-trailers Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.42 illustrates the proportion of employment that each sector represents out of the total number of employed persons in Brazil. Then, in Figure 5.43, I divide the share of employment into the three main sectors of the economy (industry, services, and agriculture). As we can see, employment in industry was the most affected by far. When I link these 186 Author’s interview with Renato Flores, January 2020. See Appendix 5 for the full interview. 228 figures with the proportion of employed persons in each sector to the total number of workers in Brazil (Figure 5.42), we can reach an approximate number of employed people who are exposed to protectionism in Brazil. More specifically, we can see that between 2009 and 2019, an average of 19.2 million people worked in the industrial sector, which was exposed to approximately 10,800 protectionist measures from the rest of the world. In addition, 4.7 million worked in the service sector, which was exposed to 950 measures. And 29.6 million people worked in the agricultural sector, which was exposed to 800 measures. Figure 5.42 – Brazil’s Employment in Industry, Agriculture, and Services (% of total employment) Source: Author’s creation based on Global Trade Alert, <https://www.globaltradealert.org/>. Figure 5.43 – Number of Total Measures Affecting Brazil by Main Economic Sectors (2009-2019) Source: Author’s creation combining data from International Labour Organization, ILOSTAT database, and Global Trade Alert. 5.6.2.2 Political Economy Insights Based on the Case of Mercosur 229 Many questions arise surrounding the political economic impact of protectionism in Argentina’s and Brazil’s respective developmental trajectories. First, the lost opportunities for trade-led development in these two cases is shocking. The data show that these two countries are not only highly protectionist, but they haven’t even cultivated a GVC with each other as part of Mercosur. Moreover, Brazil––a main players in the G20 and the world’s ninth-largest economy in 2020—has forsaken GVC integration with other trade partners in favor of protectionism. The third decade of the 21st century is very late in the game for this rampant protectionism and lack of trade interconnectivity for both countries. It also shows how far away these countries are from having a strategic regional integration model that can drive productive economic growth. Both countries have been in a severe recession since 2015 and now face further economic stress from the impacts of the COVID-19 pandemic. Mercosur is an example of an economic integration project gone wrong. The combination of deep integration with high levels of murky protectionism, even between members, has not served these member countries well. Indeed, their participation in GVCs should be much higher, as we saw in the Mexican case. It seems clear that both Argentina and Brazil need to seek PTAs beyond Mercosur, but to do so necessitates that the bloc modernizes and that each country simultaneously undertakes serious domestic reforms. Let’s start with the latter. Brazil, for example, has implemented several trade-related reform initiatives aimed at buttressing growth (WTO, 2017a): including in trade facilitation, anti-dumping, production and trade incentives (e.g., SME support), state-owned enterprises, energy, manufacturing, transport infrastructure, and more are under consideration. Monetary policy interventions were calibrated to contain inflation, which remained at the upper 230 limit of the tolerance range established by the Central Bank of Brazil (BCB) during most of the period under review. The easing of inflationary pressures since 2016 has enabled the BCB to progressively cut its policy rate from a peak of 14.15%, which should help rekindle growth. The BCB has not sought to influence the exchange rate, limiting its currency market interventions to containing excessive short-term volatility. During 2013-15, the real depreciated by around 20% in real effective terms, moving towards levels more consistent with fundamentals, but appreciated by around 6% in 2016 (p. 10). Beyond these reforms, unilateral trade liberalization in Brazil as one of its main domestic reforms that could create a more favorable context for domestic industrial production (Oliveira et al., 2018). As for Argentina, O’Farrell et al. (2018, p. 69) state that there were at least three main reform mechanisms in the last decade: compensation (“to potential losers from liberalization”), transformation (“to support firms and workers to transit from a ‘restructuring process’ as a consequence of trade liberalization to dynamic firms”), and concertation (“of jobs, through which government took a gradual stance in the liberalization of cell phones and the special promotion regime”). However, there was “low incidence in minimizing resistance to liberalization from potential losers or gaining support from potential winners. The reasons for that are multifold and specific to each of them.” For both countries, it is key to take their relations with China into account as an explanation for some current trade policy dynamics. Wise (2020, p. 169) talks about a “Southern Cone Style of Developmentalism” with China: The onset of the China boom in 2003 saw both countries with newly elected leftist executives, Lula in Brazil and Nestor Kirchner in Argentina. Although the political economy literature is laced with references to this period as “post- neoliberal,” there are important nuances between the market reforms embraced by the two countries in the 1990s. These differences, in turn, shaped the varying content and emphases of the developmentalist programs that both administrations rolled out in the early 2000s. 231 However, there are distinct differences in the economic strategies employed by Brazil and Argentina during the China boom (Wise, 2020, p. 177): In Brazil, Lula’s new developmentalism had a strong institutional base in BNDES, and the effort to promote company innovation and an outward orientation was transparent and explicit. The macro prudent economic policy framework worked at cross-purposes and ultimately hampered the quest to promote value-added exports with greater technological intensity. In Argentina, the strategy was based on the Kirchners’ coalition of the urban unemployed and underemployed, blue-collar workers, and the powerful ABCD trading companies, which have passed soybean export taxes on to weakly organized domestic producers. Given the discretionary nature of this spending, developmentalism was whatever the Kirchners said it was. If there was an industrial policy, it was difficult to discern. As for China, Argentina had naively expected south–south solidarity and a lender-of-last-resort relationship that never gelled; with Brazil, China has been an active trade and investment partner since the onset of the China boom. Both countries have had trade conflicts with China, although the China–Argentina relationship has been more acrimonious. Regarding Argentina specifically, Wise (2020) says that: Policy makers in the post-GFC period embraced some measures, such as nationalization, capital controls, and multiple exchange rates, that were simply not appealing to any rational investor…. Brazil ranks third behind Peru as a destination for outflows of Chinese FDI into the LAC region…. Argentina’s total Chinese FDI from 2000 to 2017 was around US$11.87 billion, whereas this indicator for Peru in the same period was roughly US$19.34 billion. Argentina … ranks fourth, behind Venezuela, Brazil, and Ecuador, as a recipient of Chinese lending (p. 163). On Mercosur as a bloc, as I have written elsewhere, it is first relevant to say that it is more than just a commercial agreement (Albertoni, 2019, p. 9). This is demonstrated by its normative framework, which reflects a complex shared history of its members with broader ambitions of forming a seamless customs union (similar to the EU), the defence of democracy, and broader policy coordination. The challenge facing the organization, which is not limited to Mercosur but is plaguing many of the region’s multilateral bodies, is to use this framework as an opportunity for driving development rather than the obstacle to genuinely free trade among its members and becoming a formidable market. We also need to be 232 realistic. A 37-year-old institution is not easily dismantled or reformed, but to stay relevant Mercosur must evolve with the times. The bloc’s leaders need to identify and attempt to implement a clear, predictable plan of action to promote the export sectors of member countries and maintain the confidence of international investors. Incremental changes without considering the ramifications will accomplish nothing. Mercosur members need to look to the United Kingdom’s ongoing Brexit debacle for an example of how not to proceed. To avoid a Brexit effect during Mercosur reform, the foreign ministers of Argentina, Brazil, Paraguay, and Uruguay must cooperate to maintain transparency among member countries and for the world at large. Given some of Mercosur’s downsides mentioned above, the best initial move would be to allow freedom of action for individual member states to negotiate trade agreements bilaterally without losing the advantages already provided by the bloc. This reform would allow small sub-groups within Mercosur to negotiate at faster speeds than the bloc as a whole. But here the small economies (Uruguay and Paraguay) must tread carefully. For them, reform efforts could easily turn current bloc partners into competitors (Albertoni, 2019, p. 9). Yes to modernization, but the question that arises is how. I suggest four key steps for reform of this regional bloc (Albertoni, 2019, p. 9): 1. The first step in the modernization process should be the signing of an agreement in which the members ratify the consolidation and deepening of the existing free trade area. This would lend certainty to all the economic actors of the member countries that the regional benefits of Mercosur will not be in doubt. This consolidation must be on paper, and with a commitment by member states to address all of the exceptions that have been used up until now to limit market access. 233 2. Once this crucial first step has been accomplished, members will need to amend the bloc’s founding language and processes to allow greater flexibility for the customs union and its members to negotiate international agreements with other countries or blocs. In conducting individual trade policy members will be transitioning to a PTA rather than a customs union (about this transition I talk in the following point). In recent years, bloc-wide negotiations have been plagued by delays in the process between the signing of agreements and their ratification. It’s become typical for agreements to be signed by governments, only for parliaments to spend years debating their ratification. If Mercosur is going to modernize, this tedious process needs to come to an end. 3. It is important to consider ‘flexible clauses’ in order cases such as the text of the renegotiated Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP, formerly TPP). The revised agreement, signed by eleven countries, established that the CPTPP would come into force “60 days after the date on which at least six or at least 50 per cent of the number of signatories to this Agreement, whichever is smaller, have notified the Depositary in writing of the completion of their applicable legal procedures.” 187 A similar mechanism for Mercosur reform would require that the agreement be ratified by two-member countries (one of which must be either Uruguay or Paraguay, to protect smaller members). 4. In terms of bilateral negotiations and negotiations between subgroups, each country should be free to negotiate bilaterally with countries outside of Mercosur. For this to be possible within a balanced and transparent framework, Chapter 32 of the new trade agreement between the United States, Mexico and Canada (USMCA) serves as a useful guideline. The chapter established that if one of the three partners wishes to start a commercial agreement with a country outside the bloc, it must notify the other two USMCA members three months before beginning negotiations. To properly modernize Mercosur, I contend that it should concentrate on mechanisms under a transparent process rather than binding consultation or veto. To avoid indirect harm to the bloc’s domestic economies, it is essential for all member countries to know if another member is planning on undergoing bilateral negotiations with a non-Mercosur country. 187 Comprehensive and Progressive Agreement for Trans-Pacific Partnership, <https://www.international.gc.ca/trade-commerce/trade-agreements-accords-commerciaux/agr- acc/cptpp-ptpgp/text-texte/cptpp-ptpgp.aspx?lang=eng>. 234 A final aspect regarding Mercosur modernization, which links this analysis with the Mexico case study, is to consider working more closely with the Pacific Alliance countries (Chile, Colombia, Mexico, and Peru). As previous studies suggest, “there are enough reasons to conclude that the Pacific Alliance and Mercosur could attempt convergence in the future” (Albertoni & Rebolledo, 2018, p. 108). Of relevance here is that there are numerous plurilateral agreements currently in the pipeline around the world. Hence, the convergence of two blocs such as Mercosur and the PA makes sense from an economic perspective. For example, convergence could provide the region with stronger negotiating power with the rest of the world, and it could foster integration into global supply-chain trade. Baldwin et al. (2014, p. 12) note that “traditional trade means selling into one nation goods that were made in another nation … . Supply-chain trade arises when high-tech firms combine their know- how with low-wage labor in developing nations; supply-chain is thus mostly about making things internationally, although international selling is also important.” In June 2014, during the summit meeting of the presidents of PA countries, Chile announced the proposal towards convergence between the two blocs. At that meeting, Chile also conceded that in the event of a convergence between the two blocs, a “two-speed” approach should not be disregarded. The main conclusion here is that, considering the divergent patterns of integration with Latin America, regional leaders need to rethink trade, economic, and political ties in ways that would let the region become a more active and dynamic player in the global economy. Given the relevance of PA and Mercosur for the region, convergence between these initiatives could help considerably in achieving this aim. 235 5.7 Conclusions and Country Cases’ Summary As demonstrated in previous chapters, one of the most challenging issues facing the global economy is the rapid proliferation of opaque or murky trade protectionism. Latin America is no exception to these global trends. Indeed, it is still one of the most protectionist regions in the world. Much has been written about how different nations have been affected by and responded to protectionism (Amadeo, 2019; R. Baldwin & Evenett, 2020; Evenett & Fritz, 2018) and, more generally, about the political economy of protectionism (Milner & Yoffie, 1989; Fajgelbaum et al., 2019; Hillman, 1989; Baldwin, 1989, 1982). However, this literature pertains almost exclusively to developed countries. We know little about how developing countries have been affected by and responded to protectionist trends since the GFC. This chapter has sought to expand our knowledge of the political economy of protectionism between 2009 and 2019 by exploring these dynamics in three key Latin American emerging economies. I conclude here with some political economy insights based on the main results presented above. Before I dig into each case, recall that (as shown in detail in previous sections of this chapter) Latin America has recently experienced a policy reversal in the liberalization process. A slew of studies on this topic have cautioned that “the pace towards open markets has not been steady” (IDB, 2018, p. 1). Most countries in this region have at least some type of trade agreement with one or more countries. My results show that interconnectivity through regional trade agreements does not seem to be a decisive factor in deterring protectionism and non-transparent protectionism in particular. Some 47 percent of the measures implemented within the region have been “murky instruments,” while the share of “murky instruments” deployed by Latin American countries toward the G20 countries is 33 236 percent (see Figure 5.9). Additionally, during my fieldwork interview in Latin America, trade policymakers confirmed that PTAs do not deter less transparent trade measures. Another perspective on Latin American trade protectionism concerns whether there is a pattern between the destinations of intermediate goods exported by region and the types of measures implemented. As shown, since 2008, Asia and the U.S. have consistently been the leading destination of Latin American intermediate goods (possible GVC linkages). Interestingly, the regions where Latin America is exporting more intermediate goods are also those with which it has implemented murkier instruments (see Figure 5.10 and Table 5.6). This not only confirms one of the main research propositions of this dissertation (countries with high trade interdependence will retaliate more with less transparent measures), but it also shows how far away Latin America is from being integrated into more extra-regional GVCs. In terms of the impact of protectionism on sectors and potential retaliatory dynamics, the results show that the sectors most affected in Latin America by protectionism from the RoW are also those most affected by Latin America’s protectionism toward RoW exports. Still, in general terms, Latin America has been less protectionist toward RoW than RoW has been toward Latin America. For example, while the maximum number of measures by sector affected in Latin America is around 70—for such products as iron or steel in 2019—the maximum number of protectionist measures deployed by Latin American toward RoW in terms of sector was 26—for such products as cereals in 2019. At the same time, there are some potential “within” sector retaliatory dynamics in sectors that normally have high involvement in industrialized global value chains—goods such as motor vehicles, trailers, 237 semi-trailers, and products of iron or steel. This is not in line with H7, which states that the higher the level of industrialization of the sector, the lower the within-sector retaliation. Again, this within-sector retaliation in Mercosur has been a big obstacle to productive growth in Argentina and Brazil. In the case of Mexico, we can see that it has been very responsive to protectionist measures (see Figure 5.13). Overall, this confirms H1: When a country is adversely affected by a protectionist measure adopted by another country, it will systematically respond to that country by adopting protectionist measures in return. Also, Mexico’s share of exports has been consistently more impacted by those countries with which Mexico has no FTA. On average, between 2009 and 2019, 14 percent of Mexico’s exports have been affected by close trading partners (e.g., the U.S.), and 21 percent by those countries with which it has no trade agreement. When we consider close trading partners through potential GVC ties (e.g., the U.S., Canada, China, Colombia, and Brazil, as shown in Table 5.9), there is a contrasting scenario. Indeed, between 2009 and 2019, on average, 36 percent of Mexico’s exports have been affected by close trading partners and 21 percent by those countries with which it has no trade agreements. This suggests that GVCs do not seem to restrain protectionism as do PTAs. In the same way, the case of Mexico confirms H2 on the negative relationship between protectionist dynamics and PTAs. Moreover, it does not confirm H3 (GVCs are negatively correlated with protectionist responses), in line with the quantitative results presented above. My results also show that, since 2011, Mexico has consistently implemented more protectionist measures against countries with which it has no trade agreements than toward 238 close trading partners (e.g., with an FTA in force). We can also see that the same occurred with those measures that affected Mexico. The only exception is in 2019, when Mexico found itself wedged between the U.S.–China trade war. Mexico’s exports to China between 2018 and 2019 decreased in value by 4 percent (UN COMTRADE, 2019). Moreover, President Trump threatened to impose an extraordinary escalating tariff on Mexico if it did not address the problem of undocumented migration flows from Central America. 188 Although much of this bilateral tension was at the rhetorical level, there were many non- transparent measures implemented by the U.S. toward Mexico. 189 Focusing on the level of transparency of trade measures, the share of non-transparent protectionist interventions used by Mexico toward close trading partners is much higher (on average, 47 percent between 2009 and 2019) than for countries with which it has no trade agreements (32 percent on average between 2009 and 2019). The share of non-transparent protectionist interventions toward Mexico by close trading partners (82 percent) is also higher than such interventions by countries with which it has trade agreements (80 percent). This is in line with research proposition B (H4), which states that we would expect to see more non-transparent protectionist dynamics between close trading partners (e.g., through PTAs). 188 For more information about this bilateral tension between the U.S. and Mexico, see Pérez Arguello & Albertoni, “Tariffs on Mexico? That Could Result in Trump’s Nightmare,” The New York Times, June 20, 2019, <https://www.nytimes.com/2019/06/20/opinion/tariffs-on-mexico-would-mean-more- immigrants.html>. 189 Examples are reclassification of surfboard transport systems, resulting in the imposition of a tariff (implemented in November 2019), <https://www.cbp.gov/document/bulletins/customs-bulletin- weekly-vol-53-september-25-2019-no-34>; and $10 million in Agricultural Trade Promotion Program support to the U.S. Meat Export Federation (implemented in July 2019), <https://www.fas.usda.gov/atp-funding-allocations>. For other measures implemented by the U.S. toward Mexico in 2019, see <https://www.globaltradealert.org/country/222/affected- jurisdictions_132/period-from_20190101/period-to_20191231>. 239 When we concentrate on sectoral dynamics, we can see that there are more cross- sectoral retaliatory dynamics for the case of Mexico, rather than from within sectors. This can be explained by the fact that most of the sectors affected are embedded in GVCs, such as fabricated metal products, basic iron and steel, and products of iron or steel. Hence, a retaliatory dynamic from within sectors would severely impact the domestic consumption of Mexico’s key products. With regard to Brazil and Argentina, we can see interesting protectionist dynamics from a political economy perspective. Start with Argentina. It was very responsive to protectionist measures until 2015. Since 2016, the first year of Macri’s presidency, the impact of protectionism toward Argentine exports remained the same, but its protectionist responses decreased. Second, we can see that, until 2014, Argentina’s share of exports had been consistently more impacted by protectionism from not close trading partners. However, since 2014, Argentina has been impacted by protectionism from close and not close trading partners alike. This can be explained by increasing protectionism in Brazil, the key destination of Argentine exports. On average, between 2009 and 2019, 50 percent of Argentina’s exports have been affected by close trading partners, and 47 percent by those countries with which it has no trade agreements. For the case of Argentina, its only PTA is Mercosur, and the data clearly show that rather than reducing the protectionist impact, Mercosur has increased it. This evidence for Argentina does not support my hypothesis that the existence of a PTA between A and B is negatively correlated with protectionist dynamics. As mentioned above, from a policy perspective, this result suggests that Argentina needs to negotiate more 240 strategic trade agreements with countries that it exports to, such as Chile and the U.S. In other words, the country must pursue PTAs with those countries with which trade complementarities exist (Albertoni et al., 2020). When considering the current number of measures implemented (and not the share of trade affected), Argentina has consistently implemented more protectionist measures against countries with which it has no trade agreements than toward its Mercosur trading partners. Further, Argentina has been consistently more affected by protectionist measures from countries with which it has no trade agreements. As for the level of transparency, we can see that the share of murky protectionist instruments used by Argentina toward close trading partners is much higher (33 percent on average between 2009 and 2019) than with countries with which it has no trade agreements (24 percent). This is in line with H4, which states that we would expect to see more non- transparent protectionist dynamics between close trading partners (e.g., through PTAs). When we concentrate on sectoral dynamics, we can see that there exists an important within-sector retaliatory dynamic between Argentina. The same is present in Brazil in motor vehicles and trailers & semi-trailers, all in the industrialized sector. This contradicts H7, which predicts that we would not expect to find much within-sector retaliation in those sectors with high levels of industrialization. This can be explained by Argentina’s very low access to GVCs, as well as its high protectionism. Regarding Brazil, my results demonstrate that it has been very responsive to protectionist measures overall since 2012, which marked the first year of Dilma Rousseff’s presidency. Since then, the level of protectionism toward Brazilian exports has remained the 241 same, but Brazil’s protectionist responses have increased (see Figure 5.35). Moreover, we can see that Brazil has been impacted (and responded) almost in the same way by protectionism from close and not close trading partners. This can be explained by the increasing protectionism of key trading partners such as China, the U.S., and its own Mercosur partners (see Figures 5.36 and 5.37). Considering the number of measures implemented (and not the share of trade affected), we can see that Brazil has steadily implemented more protectionist measures against countries with which it has no trade agreements than toward its Mercosur partners (e.g., with an PTA in force). Even more consistently, I find that Brazil repeatedly has been more affected by countries with which it has no trade agreement (see Figures 5.38 and 5.39). This is in line with proposition A (H2), which states that we expect that PTAs are negatively correlated with protectionist responses. Regarding the levels of transparency of measures, we can see that the share of protectionist interventions that use non-transparent instruments by Brazil toward close trading partners is much higher (74 percent, in average between 2009 and 2019) than with countries with which it has no trade agreements (56 percent). This is in line with research proposition B (H4), which states that we expect to see more non-transparent protectionist dynamics between close trading partners (e.g., through PTAs). At the same time, Brazil has been consistently more affected by non-tariff barriers used by not close trading partners. This is most likely because Brazil (and Mercosur as a bloc) is a global player that has retained highly protectionist policies and hence many countries implement barriers against it in return. Finally, regarding sectoral dynamics, we can see that for the case of Brazil there are important “within” sector retaliatory dynamics in motor vehicles, trailers & semi-trailers, and 242 fabricated metal products, which are highly industrialized sectors. This is not in line with H7, which predicts that we would not expect to find much within-sector retaliation in those sectors with high levels of industrialization. This can be explained by the fact that Brazil (and Mercosur in general) does not participate in significant GVCs. (Table 5.17 shows a summary of the results obtained in each case for each research proposition). In this respect, Brazil has been its own worst enemy. Rather than playing the role of industrial anchor and innovator within Mercosur, it has thwarted GVC development with the bloc by coddling its own motor vehicle sector—not to mention Mercosur’s prospects for deeper industrialization overall. 243 Table 5.17 – Table Summary of Regional and Country Cases 244 CHAPTER 6 – CONCLUSIONS “Not everything is as it seems, and not everything that seems is. Between being and seeming there is always a point of agreement, as if being and seeming were two inclined planes that converge and become one. There is a slope and the possibility of sliding down that slope, and when that happens, one reaches a point at which being and seeming meet.” JOSE SARAMAGO, SKYLIGHT, 2011 6.1 Main Research Contribution and Its Relevance to the Field This dissertation is neither about free trade nor protectionism. It is about the point where the two may converge. One distinctive aspect of rising protectionism over the last decade is that it is occurring against a backdrop of increased global interconnectivity. As I have argued throughout, countries around the world have become more integrated through preferential trade agreements (PTAs) and global value chains (GVCs), even though they are erecting trade barriers within these same trade venues. What explains these contradictory trajectories? Earlier work on the political economy of trade protectionism offers compelling arguments about the circumstances under which governments decide to protect their economies (Milner, 1999), and the recent literature tells us how PTAs and GVCs promote trade interdependences and openness. However, the question of now how these two trajectories––openness and protectionism––may simultaneously interact has been largely ignored. Thus, my overriding 245 research question concerns why and how increased global trade interconnectivity has also become a channel for new forms of trade protectionism. These counterintuitive dynamics constitute the newest wave in the literature on trade interdependence. The literature on trade policy has heretofore concentrated on one side of the coin with respect to the effects of an interconnected global economy: the more political and economic linkages countries build amongst themselves, the fewer tensions they will generate across borders (Baldwin, 2012; Gawande et al., 2015; Jensen et al., 2015; Lamy, 2013). From a trade policy perspective, this causal claim has held steady for many decades. Under Bretton Woods, the world saw sustained trade liberalization for half a century. However, since the completion of the Uruguay Round in 1994 and the incorporation of the General Agreement on Tariffs and Trade (GATT) into a newly created World Trade Organization (WTO) in 1995, the world has experienced a rapid proliferation of preferential trade agreements (PTAs), which have radically changed the internal logic of the global trading system. PTAs, for instance, have transformed the ways in which countries trade among themselves (Egger et al., 2011) and how they implemented liberalizing policies with closer trading partners (Cieślik & Hagemejer, 2011). In a similar way, GVCs have overhauled the ways in which countries and their large multinational corporations produce internationally. The ways these institutional trade and production mechanisms have also become channels for the spread of protectionism are subtle and enshrouded in significant economic uncertainty. This dissertation is motivated precisely by this contradiction. I hypothesize that protectionist trends over the last decade reveal a possible downside to the proliferation of PTAs and GVCs, as these have become the institutional locus for less observable, non-tariff measures. In other words, economic interdependence in the context of 246 high economic uncertainty can devolve into a spiral of protectionism because “many governments simultaneously face pressure to reflate national economies and defend national commercial interests” (Evenett, 2019, p. 26). The broader research agenda on post-GFC trade policies is still limited: Just one decade has elapsed after the GFC, and the emergence of new types and forms of trade protectionism are difficult to measure. Nevertheless, these phenomena have been under- researched. This dissertation contributes to this research agenda and the broader theoretical debate on trade policy substitution between NTMs and tariffs (Beverelli et al., 2019; Niu et al., 2018). The data limitations mentioned in earlier chapters have given rise to a pattern of “omitted variable bias” in regression studies, which is “particularly important when analyzing the impact of commercial policy, as governments can substitute between transparent and murkier forms of protectionism” (Evenett, 2019, p. 13). What we do know is that the more tariffs are lowered via binding rulings created by WTO rules, the more countries are coming to rely on NTMs. This raises a bigger challenge on the relationship between trade policy and transparency. It is precisely these opaque NTMs and their effects that this dissertation has sought to specify and quantify. As mentioned at the outset, the combined effects of the GFC, the U.S.-China trade war, and COVID-19 have rendered the international political economy a virtual social science laboratory in which the variables are still at play (Baker et al., 2020; Albertoni & Wise, 2020; Pinna & Lodi, 2021). Taking these extenuating circumstances into account, this dissertation leverages the empirical findings in the literature since the GFC and gleans new insights to better understand the sheer volume and nature of 21st-century protectionism. 247 Figure 6.1 maps out my theory and the main research questions of this dissertation. Based on empirical evidence, three key dynamics are taken into account: (1) a highly uncertain global economy; (2) given this uncertain context, governments face more domestic pressure to protect their products and markets; and (3) this current uncertainty is nested in an interconnected global economy that has no precedent (e.g., PTAs and GVCs). Table 6.1 offers a summary of my argument, hypotheses, methods, and results. Figure 6.1 – A Logical Diagram Summarizing Theory and Main Research Questions Source: Author’s creation. 248 Table 6.1 – Summary of Theory Background, Argument, Hypotheses, Methods, and Results Background: PTAs and GVCs have increased countries’ interconnectivity and bilateral trade flows (Gereffi, 2014; Baccini & Dür, 2012). At the same time, the last decade has been marked by countries’ economic and trade uncertainty and resort to trade protectionism (Ahir et al., 2018). The WTO has mostly concentrated on the regulation of tariff measures while domestic trade policy has increasingly emerged in less observable non-tariff measures (“trade policy substitution”) (Beverelli et al., 2019; Grundke & Moser, 2019; Evenett, 2019). Argument: For decades, the mechanisms that have fueled the diffusion of trade liberalization under the WTO (PTAs and GVCs), have also become channels for protectionism (based on less observable non-tariff or murkier measures). I argue that countries have changed the way they respond to protectionism, which impacts both countries’ bilateral relations. Hypotheses: Country-Level Analysis: General Hypotheses: H1: Countries will retaliate against those nations that initially target them with a protectionist measure (Direct Retaliation). H2: Preferential Trade Agreements (PTAs) are negatively correlated with protectionist responses (Cond. Effect of PTAs). H3: Global Value Chains (GVCs) are negatively correlated with protectionist responses (Cond. Effect of GVCs). The Relationship Between Tariff and Non-tariff Measures with PTAs and GVCs: H4: Preferential Trade Agreements (PTAs) are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses (Non-tariff & Conditional Effect of PTAs). H5: Global Value Chains (GVCs) are positively correlated with protectionist non-tariff responses and negatively correlated with tariff responses (Non-tariff & Conditional Effect of GVCs). Sectoral-Level Analysis: H6: Protectionist retaliation is more likely to happen cross-sector rather than within-sector. (Cross-Sector Retaliation). H7: The higher the level of industrialization of the sector, the lower the within-sector retaliation. (Within-Sector Retaliation–Industrialization). H8: Protectionist responses are less likely with trade in intermediate goods (Within-Sector Retaliation–Intermediate Goods) Methods: To test the hypotheses and causal claims presented in this dissertation, a mixed-method approach is used to blend quantitative and qualitative analytical techniques. Chapter 4 presents a quantitative analysis based on large-N data sets. Chapter 5 undertakes comparative case-study research based on three Latin American countries—Argentina, Brazil, and Mexico—which combines qualitative and quantitative information derived from within-country data. Results: Country-Level Analysis: H1 (Direct Retaliation): Confirmed – When a country is adversely affected by another country’s protectionist measures, it will systematically respond to that country by adopting its own protectionist measures in return. H2 (Conditional Effect of PTAs): Confirmed – The existence of a PTA between A and B is negatively correlated with protectionist responses. In other words, if a PTA exists, then retaliation is lower than would otherwise be the case. H3 (Conditional Effect of GVCs): Partially confirmed – Combining quantitative and qualitative results, there is no clear evidence of the effect of GVCs in protectionist retaliation. GVCs do not seem to be as strong as PTAs in restraining protectionist dynamics. H4 (Non-tariff & Conditional Effect of PTAs): Confirmed – Having a PTA is negatively correlated with tariff responses while positively correlated with non-tariff responses. H5 (Non-tariff & Conditional Effect of GVCs): Partially confirmed – Combining quantitative and qualitative results, there is no clear evidence about the real effect of GVCs in non-tariff protectionism. Sectoral-Level Analysis: H6 (Cross-Sector Retaliation): Confirmed – For both types of measures (HS and BEC classification), we can see that most of the retaliation is cross-sectoral. H7 (Within-Sector Retaliation–Industrialization): Partially confirmed – It does not appear that there is strong evidence that the higher the level of industrialization of the sector, the lower the within-sector retaliation. H8 (Within-Sector Retaliation–Intermediate Goods): Confirmed – Protectionist responses are less likely with trade in intermediate goods Source: Author’s creation. 249 6.2 Implications for International Political Economy Theory and Practice Interestingly, since the launching of the GATT treaty in 1947, the global economy has experienced four major recessions, in 1975, 1982, 1991, and 2009. However, as Kose et al. (2020, p. 3) note, “the 2009 GFC global recession, set off by the global financial crisis, was by far the deepest and most synchronized of the four recessions.” 190 Despite the failure to create an International Trade Organization in 1948, the GATT sufficed as the multilateral trade forum in 1995 when the WTO was founded. Ironically, it was at this same time that countries formally started to negotiate trade agreements among themselves. For example, by 1972, just 12 regional trade agreements had been publicized to the GATT and by the end of the 1990s, the number of PTAs had increased to 81. 191 Since then, the number of PTAs has grown consistently, a trend defined in the literature as a “New Wave of Regionalism” (Mansfield & Milner, 1999, p. 589). In sum, the 2009 recession was the first to occur after countries had created a solid network of trade agreements and global value chains, which had become a main locus of production. 192 The Great Recession triggered high levels of economic uncertainty and motivated governments to seek out PTAs to hedge against further risks in a more interconnected world. From a trade policy perspective, the different debates about the real 190 For a deep understanding of how the 2009 GFC and subsequent “Great Recession” was different to previous ones, see Reinhart and Rogoff (2009). Also see Harvard Behavioral Finance & Financial Stability, < https://www.hbs.edu/behavioral-finance-and-financial-stability/data/Pages/global.aspx>. 191 WTO, Regional Trade Agreements, <https://www.wto.org/english/tratop_e/region_e/region_e.htm>. 192 The International Monetary Fund defines a global recession as “a decline in annual per-capita real World GDP (PPP weighted), backed up by a decline or worsening for one or more of the following macroeconomic indicators: Industrial production, trade, capital flows, oil consumption, unemployment rate, per-capita investment, and per-capita consumption” (See Bob Davis, “What's a Global Recession?"”. The Wall Street Journal. 22 April 2009. Retrieved 21 Dec 2020, <https://blogs.wsj.com/economics/2009/04/22/whats-a-global-recession/>. 250 consequences of the 2009 recession are still underway. As Reinhart and Rogoff (2009, p. 1) claim, it was a different kind of recession because “the essence of the this-time-is-different syndrome is simple. It is rooted in the firmly held belief that financial crises are things that happen to other people in other countries at other times; crises do not happen to us, here and now. We are doing things better, we are smarter, we have learned from past mistakes. The old rules of valuation no longer apply.” Following Reinhart and Rogoff (2009), we would also expect that the ways in which countries were affected by and responded to protectionism and recession also shifted. Could it be that countries are protecting their markets in “smarter ways” because they may “have learned from past mistakes”? During my fieldwork interviews, I probed the possible effect of PTAs on trade policy and commercial flows. I quote trade policy expert, Dorotea López, Director of the Institute of International Studies at the Universidad de Chile, on this subject: The effects of free trade agreements … are more legal, rather than instruments of real economic connection. Sometimes I think free trade agreements have been overvalued in terms of what they generate. When one asks students about those channels that connect us commercially, free trade agreements is always an answer, when in reality this is not always the case. Just look at Mexico, which has free trade agreements and only remains connected to the United States. I do not know if free trade agreements have been overestimated. So, I’m not really sure if today the world is more connected through PTAs or not. 193 At the same time, different studies challenge the widespread perception among analysts and practitioners that traditional trade barriers have become irrelevant. For instance, Goldberg and Pavcnik (2016, p. 1) provocatively ask in one of their studies, “does trade policy still matter?” They note that over the last two decades, international trade research has shifted its focus from trade policy to other forms of trade friction. They also challenge this idea by examining a large body of evidence and conclude that: 193 Author’s interview with Dorotea López, April 2020. See Appendix 5 for the full interview. 251 The perception that trade policy is no longer relevant arises to a large extent from the inability to precisely measure the various forms of non-tariff barriers that have replaced tariffs as the primary tools of trade policy. Better measurement is thus an essential prerequisite of policy-relevant research in the future. Despite measurement challenges and scant evidence on the impact of actual policy changes, existing evidence when properly interpreted points to large effects of trade policy on economically relevant outcomes, especially when trade policy interacts with other developments, e.g., technological change (p. 10). As I have shown in previous chapters, trade policy is still very relevant, but we need to consider and measure new factors such as murky trade policies that impact trade relations between countries. In doing so, this dissertation seeks to better understand the past, but also provide a roadmap of sorts for the kinds of commercial policy reforms that will be essential for the successful revival of world markets following the pandemic downturn. This dissertation was finished during a period of intense flux. In terms of theory expansion, the main contribution of this research is my analysis of how three main variables have interacted with ongoing trade policy dynamics since the GFC. By this, I refer to the political economy “trilemma” between trade protectionism, trade interconnectivity (e.g., PTAs and GVCs), and uncertainty (mostly generated by economic crises). More specifically, I have explored how rising trade interconnectivity via PTAs and GVCs can also generate an unexpected increase in non-transparent protectionism, given the context of exceedingly high economic uncertainty. In such times, countries protect themselves from trade partners with which they have strong commercial ties through less observable, non-tariff protectionist measures. These patterns have important implications for the relationship between PTAs and GVCs, as well as for political economy and policy trends overall. 252 I do not dispute the robustness of existing research indicating a correlation between PTAs, GVCs, and trade openness measured by tariff levels and, to some extent, by the more traditional non-tariff barriers such as antidumping measures. Rather, what is demonstrated in this dissertation is that trade interconnectivity, generated by PTAs and GVCs in the context of high economic uncertainty, induces governments to substitute transparent trade barriers with less transparent ones. These results are in line with the “optimal obfuscation” that Kono (2006, p. 369) talks about in a seminal work in which he finds that “democracy leads to lower tariffs, higher core NTBs, and even higher quality NTBs …. [D]emocracy promotes ‘optimal obfuscation’ that allows politicians to protect their markets while maintaining a veneer of liberalization.” Keeping Kono’s framework in mind, I suggest that PTAs and GVCs also promote the “optimal obfuscation” that he talks about. More generally, my research contributes to the fields of international political economy, development studies, and trade policy. Another main point of my research has been to measure how the proliferation of protectionist policies since the 2008-09 GFC has affected key economic sectors. Relying on a mixed methods approach, I undertook a quantitative analysis of the spread of trade protectionism based on a large sample of countries. For the qualitative part of this dissertation research, case studies were compiled based on three Latin American emerging economies—Argentina, Brazil, and Mexico. To elaborate on this last point, Yi (2009, p. 2) argues that a “massive reorientation of trade flows towards multiple-step supply chains” has played an important role in spurring protectionist dynamics. Bems et al. (2009, para. 4) are even more specific: Vertical specialization transmission mechanism is subtle, with several ways in which it could help generate a large and widespread collapse in trade …. There 253 could be re-nationalization of international production chains (triggered perhaps by an increase in protectionism). Second, growing vertical specialization implies that more cross-border transactions occur between separate stages of the production process. If the elasticity of substitution across stages is very low, then shocks to production in one country could be transmitted forcefully to other stages undertaken elsewhere. Third, if demand shocks are concentrated on goods that are vertically specialized, then trade is highly sensitive to changes in demand. At the same time, Bems et al. (2009, para. 2) assert that “while all these channels seem plausible and many analysts have asserted that they have played an important role in the trade collapse, there has been, to date, little evidence supporting the notion.” Gawande et al. (2015), using trade policy data for seven large emerging market countries (Argentina, Brazil, China, India, Mexico, South Africa, and Turkey), find that participation in global value chains is “a powerful economic factor determining trade policy responses” (p. 102). In the same vein, Blanchard et al. (2016) show that GVCs have increased the share of firms that rely on international supply and lobby for international trade openness. In the interview I had with Victor Stolzenburg, research economist with the Economic Research and Statistics Division (ERSD) of the WTO, about the potential negative side of GVCs, he says: Not all effects of trade agreements and GVCs work in favor of an ever-freer trade regime. As trade agreements have increasingly become deeper, they touch upon a wider array of subjects, including sensitive subjects like food or health regulations. This has caused a strong backlash in some countries against trade agreements from CETA to TTIP or TPP. Similarly, the offshoring of jobs that is associated with GVCs could have weakened the political sway of trading firms as they have reduced their domestic workforce. 194 194 Author’s interview with Victor Stolzenburg, February 7, 2020. See Appendix 5 for the full interview. The CETA (Comprehensive Economic and Trade Agreement) is a free trade agreement between Canada and the European Union concluded in 2014. The TTIP (Transatlantic Trade and Investment Partnership) was a trade agreement proposed between the EU and the U.S., in 2010. The TPP (Trans-Pacific Partnership Agreement), was a proposed trade agreement between 12 countries signed in February 2016. In January 2017, then-President Donald Trump withdrew the U.S. from the TPP and the remaining countries created a new agreement called the CPTPP (Comprehensive and Progressive Agreement for Trans-Pacific Partnership), which entered into force on December 30, 2018. 254 On the effect that PTAs can have with regard to trade policy, the literature on the political economy of PTAs has been concentrated mostly on why countries negotiate and sign PTAs (Baccini & Dür, 2012; R. Baldwin, 1993; Chase, 2003; Manger, 2009). More recently, the focus has been on the effect PTAs have on signatory countries in terms of trade flows (Baier & Bergstrand, 2007) and other areas of political concern, such as their ability to reduce bilateral conflict between PTA members (Hafner-Burton, 2005; Mansfield et al., 2008; Mansfield & Reinhardt, 2008). Some more recent studies report contradictory findings. For instance, Kono (2007), in investigating the effect of PTAs in 30 countries from 1988 to 1998, found that they have “important but contradictory conditional effects: they promote multilateral liberalization when members’ intra- and extra-FTA comparative advantages are similar but impede such liberalization when these comparative advantages are different. FTAs can thus, depending on the circumstances, either help or hinder broader trade liberalization” (p. 165). Nevertheless, most of the studies on this topic have concentrated more on the positive aspects of PTAs (Limao, 2016). PTAs can cause various levels of integration, including “deep” integration (Young, 2017). It is relevant to analyze within-sector dynamics because these may play a role in a state’s protectionist dynamics and improve our understanding of how the composition of trade influences protectionist retaliation. As Peterson and Thies (2015, p. 177) state, “higher proportions of bilateral intra-industry trade within bilateral relationships may promote the formation of preferential trade agreements, foster increasing political affinity, and reduce 255 states’ propensity to engage in military conflict.” As the World Bank’s 2020 Global Economic Perspective states: COVID-19 has sharply worsened economic conditions in Latin America and the Caribbean (LAC). The regional economy is projected to contract by 7.2 percent in 2020, a much steeper decline than during the global financial crisis, reflecting the impact of the measures necessary to slow the spread of the pandemic, significant deterioration in financing conditions and commodity prices, and spillovers from a global recession. As mitigation measures are scaled back and financing, commodity price, and external demand conditions become more supportive, regional growth is projected to recover to 2.8 percent in 2021. However, the near-term outlook is subject to significant downside risks. These include a resurgence of last year’s wave of social unrest, increasingly adverse market reactions to rising public debt, weaker-than-expected commodity prices, and persistent pandemic-related uncertainty slowing the recovery of the services sector (p. 81). In sum, the contribution of this academic work is threefold: first, it fills a gap in our knowledge about the rise of post-GFC protectionism in developing countries. Second, it identifies a new murky protectionism in the form of opaque and pernicious non-tariff barriers, a topic that has been under-researched and demands attention. Third, I have built a series of datasets that better enable us to measure the damage done to sectors, producers, workers, and consumers since the rapid escalation of non-tariff barriers over the past decade. A central element of this project is that it bridges academics and trade policymaking and contributes to the search for commercial policy reforms that will be essential for the successful revival of world markets post-COVID-19. 6.3 Implications for Policy Regarding policy prescriptions, a key finding here is that a reliable set of trade agreements can be a buffer against straight tariff protectionism but not much use in combatting non-tariff protectionism. Moreover, the number of PTAs a given country negotiates is less important than the type of trade agreements that are signed. Based on the positive effects of trade 256 interdependence highlighted by previous studies, many countries have sought to strengthen their trade ties with new foreign partners. However, my research suggests that trade interconnectivity can be “double edged”: it limits tariff protectionism overall but can also prompt non-transparent protectionism. In other words, PTAs without clear enforcement mechanisms can actually be a negative long-term tool to improve bilateral trade. Evidence of this is that some original PTAs are now being renegotiated in a process called modernization. Most of these renegotiations include chapters related to non-tariff measures. Examples of this are the recent modernization of such PTAs as the EU-Chile, EU-Mexico, EU-Canada, China- Chile, Canada-Ukraine, and European Free Trade Area (EFTA) with Turkey and Canada. 195 Many of these agreements that are—or were—in the process of modernization concentrate on topics that go well beyond tariff issues and concentrate on “murkier” regulations that were not fully considered in “first-generation” PTAs (Lechner, 2016). Politically, this dissertation suggests that PTAs and GVCs are neither promoters nor deterrents of protectionism. That is, although PTAs and GVCs seem to have important effects in limiting tariff protectionism, they can be strong channels for non-transparent protectionism (e.g., consumption subsidies, loan guarantees, import bans, and licensing requirements, among others presented in Table 1.5). From a policy perspective, trade interconnectivity via PTAs and GVCs can lure policymakers into lowering their guards as tariffs are reduced drastically and in a transparent manner. However, non-tariff barriers can discreetly increase. This means two things from a policy point of view. First, apart from rules to limit traditional NTMs, which are included in many agreements, trade negotiators must 195 For more information on these modernization trade agreements, see <http://www.sice.oas.org/news_e.asp>. 257 also seek ways to address other types of murky measures. Second, policymakers must be proactive in the enforcement of these updated rules. New PTAs, in short, have to be as creative as the sponsoring governments have been in implementing new types of NTMs. In an era of “murky” protectionism, we need “bright” rules. Andres Rebolledo, former Vice Minister of Trade of Chile and chief negotiator of the Chile-China FTA, elaborated on this point during our interview: It is very difficult to find a specific chapter that talks about NTMs in a PTA. What we do find is specific articles that talk about specific types of measures such as sanitary and phytosanitary measures, and technical barriers to trade, which are examples of NTMs. 196 However, in all cases these chapters tend to be very general, in which countries commit to transparency in the implementation of these norms. From my experience, what PTAs do with regard to NTMs is to discipline their implementation but not necessarily to remove them for the simple reason that enforcement here is very difficult. Once a trade agreement is signed, I would say that 80% of the conflicts are related to non-tariff measures. And the main problem is that we rarely find specific and clear chapters on how NTM restrictions will be enforced. I think the problems is not the PTA itself but how countries use these less transparent norms within a PTA. In sum, all this discussion is a good example that PTAs are neither heaven nor hell, they are just one instrument that by itself doesn’t do too much if there is no commitment to transparency by trading partners. 197 Countries normally rely on the WTO general framework text for NTMs when they negotiate bilateral PTAs. Using text analysis techniques, recent studies show that the presence of WTO text in PTAs is substantial: “at least one hundred PTAs take 80 percent or more of their contents directly from a single, existing treaty [the WTO]—with many copying and pasting 95 percent or more” (Allee & Elsig, 2019, p. 603). When specifically talking about NTMs (e.g., SPS and TBT), the percentage of text in PTAs that is copied and pasted 196 Examples of this in recent agreements are Chapter 9 and 11 in the USMCA and Chapters 5 and 6 in the Mercosur and EU trade agreement. See Agreement between the United States of America, the United Mexican States, and Canada 12/13/19 Text, <https://ustr.gov/trade-agreements/free-trade- agreements/united-states-mexico-canada-agreement/agreement-between>; and Mercosur and EU agreement, <https://publications.iadb.org/pt/acordo-mercosul-uniao-europeia-impactos- normativosregulatorios-no-mercosul> 197 Author’s interview with Andrés Rebolledo, November 2019. See Appendix 5 for the full interview. 258 from WTO agreements reaches 60 percent. At the same time, in the context of the WTO, the handling of NTMs is still an open debate. For instance, in July 2019, the WTO Goods Council meeting suggested the importance of enhanced transparency and strengthening notification requirements under WTO agreements. 198 In sum, the current structure of PTAs’ rules with regard to NTMs is still not clearly defined at the multilateral level (Allee & Elsig, 2019, p. 603). Many institutions have reiterated concerns about the challenge of non- transparent protectionism. This is clear from a recent European Commission (2019, p. 9) report: Analysis of non-tariff barriers and their impact remains challenging. The main reason is that non-tariff barriers are characterized by different degrees of restriction. Other than outright bans, most trade-restrictive measures do not fully eliminate trade but reduce it. Moreover, restrictions regarding the same products or services may overlap. As a result, additional barriers may not necessarily mean additional impact, nor does the removal of one barrier imply automatic improvement in market access. Finally, the world community faces important unknowns related to the economic fallout of the COVID-19 pandemic. As of January 2021, there is an emerging consensus that the virus is still far from being under control, although the vaccine is now being distributed around the world. 199 Here, commitments and resources have been uneven, and U.S. infection rates continue to surge. This delay on the part of the world’s largest economy will surely prolong the pandemic and wreak further havoc on global markets. What will the international political economy look like once the pandemic is finally under control? The lack of answers to these questions has wrought a level of uncertainty 198 WTO (2019), WTO members consider transparency reforms at Goods Council meeting, < https://www.wto.org/english/news_e/news19_e/good_10jul19_e.htm>. 199 Coronavirus trend: The pandemic is far from over, Deutsche Welle, January 29, 2021, Retrieved from: <https://www.dw.com/en/coronavirus-global-pandemic-trend/a-53954594>. 259 and insecurity on par with the Great Depression, which lasted a full decade (Albertoni & Wise, 2020). 6.4 Insights for Further Research This dissertation constitutes a first step in a research agenda with plenty of room to expand. Future studies can build on the results presented here to gain further understanding of the risk of trade protectionism in an interconnected and uncertain global economy. For example, the main quantitative focus of this dissertation is on dyadic dynamics rather than triadic ones. Although this study goes one step forward by considering bilateral data rather than monadic—one state as the unit of analysis—we still need a clearer picture of current global trade dynamics. In essence, a network analysis in which many factors and countries interact at the same time would tell us more. My large-N bilateral analysis sheds light on retaliatory responses between country A and B, but not country C. A state’s position in an international trading network also affects its retaliatory dynamics. That is, given the high levels of trade interconnectivity within the global economy, it is likely that dyadic mechanisms may give way to more triadic dynamics. The application of network theory to the study of international trade is becoming more relevant in the literature. As Chaney (2016, p. 754) states, “the structure of international trade intrinsically resembles that of a network: countries are connected to each other by trade linkages.” According to this systemic view of trade protectionism, countries are categorized not just by how much they apply protectionist trade policies but also “against” whom. Recent literature suggests that the field of international relations can benefit from the incorporation of network analysis into methodological approaches (Lupu & Traag, 260 2013). 200 For example, relational event approaches (Butts & Marcum, 2017), in which the systemic effect can be better specified, could capture the whole system and perhaps render a better explanation than one based on the sum of its parts. In Appendix 6, I offer specific ideas on how to extend this dissertation by using network analysis. Another limitation of this dissertation relates to data restrictions, including the short period of analysis, i.e., the 2008-09 GFC to 2019. In a context of economic crisis—and post- crisis—governments tend to implement more protectionist measures than they normally would so as to redirect demand toward domestically produced goods. Although this study goes until 2019, other studies have shown that the recovery of global trade after the GFC has been weak compared with the marked acceleration in global trade in the two decades before the GFC (Aslam et al., 2017; Wozniak & Galar, 2018). It is thus possible that much of the retaliation captured here is more explicit to the GFC but not to other less turbulent economic contexts. As such, it cannot be claimed that the findings of this study are generalizable to other periods of economic history. On the country case analyses, further research could be expanded to other regions such as Africa and Asia as a way to validate my domestic level of analysis. A new causal explanation for why governments may be willing to retaliate against a country that previously targeted them may originate in the exporting sector of the affected country. Before lobbying for retaliation, ex ante, the exporting sector may pressure the government to initiate a diplomatic conversation to persuade the trading partner to remove a protectionist measure. If “trade diplomacy” does not work, the exporting sector could then put pressure on 200 For a critical review of the literature that has recently applied network theory to the study of international trade, see Fagiolo (2017) and Chaney (2016). 261 the government to retaliate. The effect of media coverage on the implementation of a protectionist measure by a trading partner could also be relevant in further studies. 201 Further studies should build on this research at the country level and explore different casual mechanisms as to why countries are embracing new forms of protectionism. For instance, one reason countries may or may not be willing to escalate trade protectionist dynamics, especially in developing economies in Asia and Latin America, can be found in lessons learned from past protectionist experiences during the 1970s and 1980s. There could, in other words, be a “path dependence” mechanism at work in the way countries decide to respond to protectionism. Indeed, as Wise et al. (2015) document, emerging economies in both Latin America and East Asia showed remarkable resilience and restraint from protectionism when faced with the global financial crisis from 2008 to 2009. As they describe it, “one of the more surprising features of the 2008-09 global financial crisis was the comparative ease with which emerging economies in Asia and Latin America rebounded. That rebound was a radical departure from the effects of previous crises on these regions, be it the decade-long recession wreaked on Latin America by the 1982 debt shocks or the financial crisis that dramatically slowed Asian economies in the late 1990s” (p. 1). In general, at the country level, further studies could identify those emerging economy cases in which the exporting sector—which is traditionally part of the lobby for trade openness—is pushing for trade protection. This approach may help in specifying trade preferences at the firm level. As of now, there is little evidence of exporters lobbying for protection. This type of exporter protection lobby may be transitory and does not necessarily 201 For more information about the public diplomacy of interactional trade see Kostecki and Naray (2007) and Albertoni (2018). 262 mean that exporters have changed their overall trade preferences. However, it does show that in the context of a trade conflict, exporters may be willing to set aside their traditional trade preferences in order to solve the conflict and return the trade relationship to status quo. Exporters could lobby for protection, for instance, if a “trade war” happens between the U.S., China, and the EU. This could invoke retaliatory responses that result in a never-ending spiral of protectionism with few incentives or intentions to negotiate a resolution to the conflict. On a final note, this study offers a cautionary tale about trade protectionism in an era of high interconnectivity and economic uncertainty. From a policy perspective, it shows how relevant it is to direct our attention toward the increased need for trade transparency, something that multilateral institutions have acknowledged without promoting consistently in ongoing policy debates. The escalation of trade protectionism during and after a COVID-19 economic recession may be the (unfortunate) catalyst for pushing this topic higher up on the multilateral policy agenda. 263 BIBLIOGRAPHY Abbott, F. M. (2009). Cross-Retaliation in TRIPS: Options for Developing Countries. ICTSD Issue Paper, 8. https://www.files.ethz.ch/isn/102241/2009-06_cross-retaliation-in-trips.pdf Ahir H., Bloom, N., & Furceri, D. (2018). The World Uncertainty Index. Retrieved from https://nbloom.people.stanford.edu/research Ahir, H., Bloom, N., & Furceri, D. (2020). 60 Years of Uncertainty. Finance & Development, 57(1), 58–60. Aizenman, J., & Jinjarak, Y. (2010). The Role of Fiscal Policy in Response to the Financial Crisis. United Nations’ Development Policy and Analysis Division. http://csf.rrojasdatabank.info/wesp2011paper_aizenman.pdf Aizenman, J., & Marion, N. P. (1996). Volatility and the Investment Response. NBER Working Paper Series, No. 5841. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=225627 Albertoni, N. (2016, May 2). Los 25 años del Mercosur. El País. https://elpais.com/internacional/2016/05/02/america/1462201971_247462.html ______ (2018). The New Dynamics of the International Trading System. Global Policy, 9(1), 156- 158. ______ (2019a, January 1). Can Mercosur be updated and reformed or is it a relic of the past? Global Americans. https://theglobalamericans.org/2019/01/can-mercosur-be-updated-and-reformed- or-is-it-a-relic-of-the-past/ ______ (2019b). Uruguay como solución. Su inserción internacional cuando lo importante se transforma en urgente. Penguin Random House. Albertoni, N., & Wise, C. (2020). International Trade Norms in the Age of Covid-19 Nationalism on the Rise? Fudan Journal of the Humanities and Social Sciences, 14(1), 41–66. Albertoni, N., & Rebolledo, A. (2018). Trade, Economic and Political Integration in Latin America: The Cases of the Southern Common Market (Mercosur) and the Pacific Alliance. In P. Sauvé, R. P. Lazo, J-M. Alvarez Zarate, (Eds.), The Pacific Alliance in a World of Preferential Trade Agreements (pp. 99-112). Springer. Albertoni, N., Iturralde, A., & Correa, R. (2020). Índice de Vulnerabilidad Comercial. Centro de Estudios para el Desarrollo. Centro de Estudios para el Desarrollo. Alexander, K. W., & Soukup, B. J. (2010). Obama’s first trade war: The US-Mexico cross-border trucking dispute and the implications of strategic cross-sector retaliation on US compliance under NAFTA. Berkeley Journal of International Law, 28(2), 313–342. 264 Allee, T. (2012). The role of the United States: A multilevel explanation for decreased support over time. In M. Daunton, A. Narlikar, and R. M. Stern (Eds.), The Oxford Handbook on the World Trade Organization (pp. 235–253). Oxford University Press. Altig, D., Baker, S., Barrero, J., Bloom, N., Bunn, P., Chen, S., Davis, S., Leather, J., Meyer, B., Mihaylov, E., Mizen, P., Parker, N., Renault, T., Smietanka, P., & Thwaites, G. (2020, July 24). Economic uncertainty in the wake of the COVID-19 pandemic. VoxEU.Org https://voxeu.org/article/economic-uncertainty-wake-covid-19-pandemic Altman, S., & Bastian, P. (2019). DHL Global Connectedness Index. (2019, February 25). https://www.dpdhl.com/content/dam/dpdhl/en/media-center/media relations/documents/2019/dhl-gci-2019-update-full-study.pdf Amadeo, K. (2019). Why Trade Wars Are Bad and Nobody Wins. The Balance (2019, May 2). https://www.thebalance.com/trade-wars-definition-how-it-affects-you-4159973 Autor, D., Dorn, D., & Hanson, G. (2016). The China Shock: Learning from Labor-Market Adjustment to Large Changes in Trade. Annual Review of Economics, 8(1), 205–240. Baccini, L., & Dür, A. (2012). The New Regionalism and Policy Interdependence. British Journal of Political Science, 42(1), 57–79. Baccini, L., & Urpelainen, J. (2014). International Institutions and Domestic Politics: Can Preferential Trading Agreements Help Leaders Promote Economic Reform? The Journal of Politics, 76(01), 195–214. Baier, S. L., & Bergstrand, J. H. (2007). Do free trade agreements actually increase members’ international trade? Journal of International Economics, 71(1), 72–95. Baker, S., Bloom, N., Davis, S., & Terry, S. (2020, April 13). COVID-Induced Economic Uncertainty and its Consequences. VoxEU.Org https://voxeu.org/article/covid-induced- economic-uncertainty-and-its-consequences Baldwin, R.E. (1970). Non-Tariff Distortions of International Trade. The Brookings Institution. ______ (1989). Trade Policy in a Changing World Economy. University of Chicago Press. Baldwin, R. (1993). A Domino Theory of Regionalism. NBER Working Paper Series, No. 4465. (1993, January 18). https://www.nber.org/papers/w4465 ______ (2009a, November 27). The great trade collapse: What caused it and what does it mean? VoxEU.Org https://voxeu.org/article/great-trade-collapse-what-caused-it-and-what-does-it- mean ______ (2009b). The Greater Trade Collapse of 2020. VoxEU.Org (2009, April 7). https://voxeu.org/article/greater-trade-collapse-2020 265 ______ (2012). WTO 2.0: Global Governance of Supply-Chain Trade. Centre for Economic Policy Research, Policy Insight No. 64. https://voxeu.org/article/greater-trade-collapse-2020 Baldwin, R., Kawai, M., & Wignaraja, G. (2014). A world trade organization for the 21st century: The Asian perspective. Edward Elgar Publishing. Baldwin, R. & Evenett, S. (2009). The collapse of global trade, murky protectionism, and the crisis: Recommendations for the G20. Centre for Economic Policy Research (CEPR). ______ (2020). COVID-19 and Trade Policy: Why Turning Inward Won’t Work. Centre for Economic Policy Research (CEPR). Barbieri, K. (1996). Economic Interdependence: A Path to Peace or a Source of Interstate Conflict? Journal of Peace Research, 33(1), 29–49. Bellégo, C., & Pape, L.-D. (2019). Dealing with Logs and Zeros in Regression Models. Political Methods: Quantitative Methods EJournal, 1(2), 1–16. Bems, R., Johnson, R., & Yi, K. (2013). The Great Trade Collapse. Annual Review of Economics, 5(1), 375–400. ______ (2009, November 27). The Collapse of Global Trade: Update on the Role of Vertical Linkages. VoxEU CERP Policy Portal. https://voxeu.org/article/collapse-global-trade-update- role-vertical-linkages Beverelli, C. & Boffa, M. & Keck, A. (2014) Trade policy substitution: Theory and evidence from Specific Trade Concerns. WTO Staff Working Papers ERSD-2014-18. https://ideas.repec.org/p/zbw/wtowps/ersd201418.html ______ (2019). Trade Policy Substitution: Theory and Evidence. Review of World Economics, 155(4), 755–783. Beyers, J., & Kerremans, B. (2007). The Press Coverage of Trade Issues: A Comparative Analysis of Public Agenda-Setting and Trade Politics. Journal of European Public Policy, 14(2), 269– 292. Bhagwati, J. (1991). The World Trading System at Risk. Princeton University Press. ______ (2003). Free Trade Today. Princeton University Press. Bhagwati, J., Krishna, P., & Panagariya, A. (2016). The World Trade System: Trends and Challenges. MIT Press. Blanchard, E., Bown, C., & Johnson, R. (2016). Global Supply Chains and Trade Policy. NBER Working Paper No 21883. https://doi.org/10.3386/w21883 Boffa, M., & Olarreaga, M. (2012). Protectionism During the Crisis: Tit-for-tat or Chicken-Games? Economics Letters, 117(3), 746–749. 266 Bonomo, D. (2014). Hitting Where it Hurts: Retaliation Requests in the WTO. CEPR Policy Paper (2014, March 1). https://voxeu.org/article/retaliation-wto Bordo, M. D., Eichengreen, B., & Irwin, D. A. (1999). Is Globalization Today Really Different than Globalization a Hundred Years Ago? National Bureau of Economic Research. Bown, C., & Kolb, M. (2019). Trump’s Trade War Timeline: An Up-to-Date Guide. PIIE, 2–15. Bussière, M., Pérez-Barreiro, E., Straub, R., & Taglioni, D. (2011). Protectionist Responses to the Crisis: Global Trends and Implications. The World Economy, 34(5), 826–852. Butts, C. & Marcum, C. (2017). A Relational Event Approach to Modeling Behavioral Dynamics. In A. Pilny & M. Scott (Eds.), Group Processes: Data-Driven Computational Approaches (pp. 51–92). Springer. Cadestin, C., Gourdon, J., & Kowalski, P. (2016). Participation in Global Value Chains in Latin America: Implications for Trade and Trade-Related Policy, OECD Trade Policy Papers, No. 192. OECD Publishing. https://doi.org/10.1787/5jlpq80ts8f2-en Canto, V. A. (1983). U.S. Trade Policy: History and Evidence. Cato Journal, 3(3), 679-703. Casey-Sawicki, Katherine. (2018). Seattle WTO Protests of 1999. Encyclopedia Britannica. Chase, K. A. (2003). Economic Interests and Regional Trading Arrangements: The Case of NAFTA. International Organization, 57(1), 137–174. Chiquiar, D. & Tobal, M. (2019). Global Value Chains in Mexico: A Historical Perspective. Working Papers 2019-06 Banco de México. https://ideas.repec.org/p/bdm/wpaper/2019-06.html Chisik, R., & Fan, C. (2020). Cross-retaliation and International Dispute Settlement. Working Papers 078, Ryerson University, Department of Economics. https://ideas.repec.org/p/rye/wpaper/wp078.html Chong, T. T. L., & Li, X. (2019). Understanding the China–US Trade War: Causes, Economic Impact, and the Worst-Case Scenario. Economic and Political Studies, 7(2), 185-202. Collier, D., Brady, H. E., & Seawright, J. (2004). Sources of leverage in causal inference: Toward an Alternative View of Methodology. In H. E. Brady & D. Collier (Eds.), In Rethinking Social Inquiry: Diverse Tools, Shared Standards (pp. 229-266). Rowman & Littlefield Publishers. Conconi, P., Facchini, G., & Zanardi, M. (2014). Policymakers’ Horizon and Trade Reforms: The Protectionist Effect of Elections. Journal of International Economics, 94(1), 102–118. Copeland, D. C. (1996). Economic Interdependence and War: A Theory of Trade Expectations. International Security, 20(4), 5–41. Córdova Bojórquez, G., Meardon, S., Nava Aguirre, K. M., & Schaffler-González, F. (2018). The Political Economy of Mexico-U.S. Trade Policy. IDB - Working Paper. 267 Crowley, M. A. (2019). Trade War: The Clash of Economic Systems Threatening Global Prosperity. CEPR Press. De Backer, K., & Miroudot, S. (2014). Mapping Global Value Chains. OECD Trade Policy Papers, 15(1), 1–24. De Mateo Veiga, F. (2003). La política comercial de México con América Latina. In R. F. De Castro (Ed.), En la frontera del imperio (pp. 10-30). Planeta. Dedrick, J., Kraemer, K., & Linden, G. (2010). Who Profits from Innovation in Global Value Chains? A Study of the iPod and Notebook PCs. Industrial and Corporate Change, 19(1), 81–116. DIRECON. (2018). Riesgos y Oportunidades de la Guerra Comercial. Departamento de Estudios. https://www.subrei.gob.cl/sala-de-prensa/noticias/detalle-noticias/2018/07/20/en-primera- mesa-publico-privada-sobre-el-tema-direcon-da-a-conocer-informe-sobre-riesgos-y- oportunidades-de-la-guerra-comercial. Dür, A., Baccini, L., & Elsig, M. (2014). The Design of International Trade Agreements: Introducing a New Dataset. The Review of International Organizations, 9(3), 353–375. Escaith, H., Lindenberg, N., & Miroudot, S. (2010). Global Value Chains and the Crisis: Reshaping International Trade Elasticity. Global Value Chains in a Postcrisis World: A Development Perspective. The World Bank. https://openknowledge.worldbank.org/handle/10986/2509 Estevadeordal, A., Suominen, K., Harris, J., & Shearer, M. (2009). Bridging Regional Trade Agreements in the Americas. IDB Special Report on Integration and Trade. Evenett, S. (2019). Protectionism, State Discrimination, and International Business Since the Onset of the Global Financial Crisis. Journal of International Business Policy, 2(1), 1–28. Evenett, S., & Fritz, J. (2017). Will Awe Trump Rules? Global Trade Alert 21st Report CEPR. https://www.globaltradealert.org/reports/download/42. ______ (2018). Brazen Unilateralism: The US-China Tariff War in Perspective. Global Trade Alert Report 23. CEPR Press. https://www.globaltradealert.org/reports/45. ______ (2020). The Global Trade Alert database handbook. Manuscript. https://www.globaltradealert.org/data_extraction. Evenett, S., Fritz, J., Gerasimenko, D., Nowakowska, M., & Wermelinger, M. (2011). The resort to protectionism during the great recession: Which Factors Mattered? [Working Paper] University of St Gallen. https://www.alexandria.unisg.ch/publications/89753 Fajgelbaum, P., Goldberg, P., Kennedy, P., & Khandelwal, A. (2019). The Return to Protectionism. Working Paper No. 25638 National Bureau of Economic Research. https://doi.org/10.3386/w25638 268 Felbermayr, G., Teti, F., & Yalcin, E. (2019). Rules of Origin and the Profitability of Trade Deflection. Journal of International Economics, 121(1), 103-248. Felter, C., Renwick, D., & Chatzky, A. (2019, July 10). Mercosur: South America’s Fractious Trade Bloc. Council on Foreign Relations. https://www.cfr.org/backgrounder/mercosur-south- americas-fractious-trade-bloc Franko, P. (2018). The Puzzle of Latin American Economic Development. Rowman & Littlefield Publishers. Frieden, J. (2000). The Method of Analysis of Modern Political Economy. In J. A. Frieden, M. Pastor, & M. Tomz (Eds.), Modern Political Economy and Latin America: Theory and Policy (pp. 35–37). Westview Press. Garciadiego, J., & Hernández, B. (1994). El TLC día a día: Crónica de una negociación. Miguel Angel Porrúa Editorial. Garred, J. (2018). The Persistence of Trade Policy in China After WTO Accession. Journal of International Economics, 114(1), 130–142. ______ (2019, July 9). The Persistence of Trade Policy Beyond Import Tariffs. VoxEU.Org. https://voxeu.org/article/persistence-trade-policy-beyond-import-tariffs Garrett, G. (2000). Capital Mobility, Exchange Rates and Fiscal Policy in the Global Economy. Review of International Political Economy, 7(1), 154–171. Gaulier, G., Sztulman, A., & Ünal, D. (2019). Are Global Value Chains Receding? The Jury is Still Out. Key Findings from the Analysis of Deflated World Trade in Parts and Components. International Economics, 161(1), 219-236. Gawande, K., Hoekman, B., & Cui, Y. (2011, November 10). Determinants of Trade-Policy Responses to the 2008 Financial Crisis. VOX CERP Policy Portal. https://voxeu.org/article/determinants-trade-policy-responses-2008-financial-crisis Gawande, K., Hoekman, B., & Cui, Y. (2015). Global Supply Chains and Trade Policy Responses to the 2008 Crisis. The World Bank Economic Review, 29(1), 102–128. Georgiadis, G., & Gräb, J. (2016). Growth, Real Exchange Rates and Trade Protectionism Since the Financial Crisis. Review of International Economics, 24(5), 1050-1080. Gereffi, G. (2014). Global Value Chains in a Post-Washington Consensus World. Review of International Political Economy, 21(1), 9–37. ______ (2018). Global value chains and development: Redefining the contours of 21st century capitalism. Cambridge University Press. Gerring, J., & Cojocaru, L. (2016). Selecting Cases for Intensive Analysis: A Diversity of Goals and Methods. Sociological Methods & Research, 45(3), 392–423. 269 Gleditsch, K. S., & Ward, M. D. (1999). Interstate System Membership: A Revised List of the Independent States Since 1816. International Interactions, 25(4), 393–413. Ghodsi, M., & Stehrer, R. (2016). Non-tariff measures trickling through global value chains. Productivity, non-tariff measures and openness. [Working paper]. Vienna Institute for International Economic Studies, Vienna. https://www.etsg.org/ETSG2016/Papers/281.pdf Gould, D. M., & Woodbridge, G. L. (1998). The Political Economy of Retaliation, Liberalization and Trade Wars. European Journal of Political Economy, 14(1), 115–137. Gowa, J., & Mansfield, E. (2004). Alliances, Imperfect Markets, and Major Power Trade. International Organization, 58(4), 775–805. Graham, B., & Tucker, J. (2017). The International Political Economy Data Resource. The Review of International Organizations, 12(1), 1–13. Grant, W., & Kelly, D. (2005). The Politics of International Trade in the Twenty-First Century. Palgrave Macmillan. Grossman, G., & Helpman, E. (1994). Protection for Sale. American Economic Review, 84(4), 833– 850. Grundke, R., & Moser, C. (2019). Hidden Protectionism? Evidence from Non-Tariff Barriers to Trade in the United States. Journal of International Economics, 117, 143–157. Hafner-Burton, E. M. (2005). Trading Human Rights: How Preferential Trade Agreements Influence Government Repression. International Organization, 59(3), 593–629. Henn, C., & McDonald, B. (2014). Crisis Protectionism: The Observed Trade Impact. IMF Economic Review, 62(1), 77–118. Hernández, R., Martínez Piva, J. M., & Mulder, N. (2014). Global Value Chains and World Trade: Prospects and Challenges for Latin America. ECLAC. Hites, A., Bloom, N., & Furceri, D. (2019, September 9). New Index Tracks Trade Uncertainty Across the Globe. IMF Blog. https://blogs.imf.org/2019/09/09/new-index-tracks-trade- uncertainty-across-the-globe/ Hofmann, C., Osnago, A., & Ruta, M. (2019). The Content of Preferential Trade Agreements. World Trade Review, 18(3), 365–398. Hollyer, J., Rosendorff, P., & Vreeland, J. (2014). Measuring Transparency. Political Analysis, 22(4), 413–434. Hopewell, K. (2016). Breaking the WTO: How Emerging Powers Disrupted the Neoliberal Project. Stanford University Press. Hummels, D., Ishii, J., & Yi, K.-M. (2001). The Nature and Growth of Vertical Specialization in World Trade. Journal of International Economics, 54(1), 75–96. 270 Ianchovichina, E., & Martin, W. (2004). Impacts of China’s Accession to the World Trade Organization. The World Bank Economic Review, 18(1), 3–27. IDB. (2018). The Political Economy of Trade Policy in Latin America and the Caribbean. Research Department Integration and Trade Sector RG-K1198. https://research- proposals.iadb.org/system/files/RES/documents/EZSHARE-1728116555-3751.pdf IMF. (2016). World Economic Outlook. Subdued Demand: Symptoms and Remedies. IMF. https://www.imf.org/en/Publications/WEO/Issues/2016/12/31/Subdued-Demand-Symptoms- and-Remedies Irwin, D. (1998). Against the Tide: An Intellectual History of Free Trade. Princeton University Press. ______ (2016, July/August). The Truth About Trade. What Critics Get Wrong About the Global Economy. Foreign Affairs. https://www.foreignaffairs.com/articles/2016-06-13/truth-about- trade ______ (2017, May/June). The False Promise of Protectionism. Foreign Affairs. https://www.foreignaffairs.com/articles/united-states/2017-04-17/false-promise- protectionism Jensen, J., Quinn, D. P., & Weymouth, S. (2015). The Influence of Firm Global Supply Chains and Foreign Currency Undervaluations on US Trade Disputes. International Organization, 69(4), 913–947. Joller, Y., & Kniahin, D. (2020, July 8). Global Chain Reaction: Unprecedented Trade Measures to Tackle COVID-19. ITC Blog. https://www.intracen.org/covid19/Blog/Global-chain-reaction- Unprecedented-trade-measures-to-tackle-COVID-19/ Jones, K. (2015). Reconstructing the World Trade Organization for the 21st Century: An Institutional Approach. Oxford University Press. Kee, H. L., Neagu, C., & Nicita, A. (2013). Is Protectionism on the Rise? Assessing National Trade Policies during the Crisis of 2008. Review of Economics and Statistics, 95(1), 342–346. Kim, I. S. (2017). Political Cleavages within Industry: Firm-level Lobbying for Trade Liberalization. American Political Science Review, 111(01), 1–20. King, G., Keohane, R. O., & Verba, S. (1994). Design Social Inquiry. Scientific Inference in Qualitative Research. Princeton University Press. Kolb, M. (2019, February 4). What Is Globalization? And How Has the Global Economy Shaped the United States? The Peterson Institute for International Economics. https://www.piie.com/microsites/globalization/what-is-globalization Kono, D. (2006). Optimal Obfuscation: Democracy and Trade Policy Transparency. The American Political Science Review, 100(3), 369–384. 271 ______ (2007). When Do Trade Blocs Block Trade? International Studies Quarterly, 51(1), 165–181. Kosacoff, B., López, A., & Pedrazzoli, M. (2008, January). Trade, Investment and Fragmentation of the Global Market: Is Latin America Lagging Behind? CEPAL Studies and Perspectives. https://www.cepal.org/en/publications/4867-trade-investment-and-fragmentation-global- market-latin-america-lagging-behind Lake, D. (2009). Open Economy Politics: A Critical Review. The Review of International Organizations, 4(3), 219–244. Lamy, P. (2013, December 18). Global Value Chains, Interdependence, and the Future of Trade. Vox CEPR Policy Portal. https://voxeu.org/article/global-value-chains-interdependence-and- future-trade Lardy, N. R. (2001, May 9). Issues in China’s WTO Accession. Brookings Institution. https://www.brookings.edu/testimonies/issues-in-chinas-wto-accession/ Lee, J., & Swagel, P. (1997). Trade Barriers and Trade Flows across Countries and Industries. Review of Economics and Statistics, 79(3), 372–382. Levy, P. (2018, February 4). Was Letting China into the WTO a Mistake? Foreign Affairs. https://www.foreignaffairs.com/articles/china/2018-04-02/was-letting-china-wto-mistake Lijphart, A. (1971). Comparative Politics and the Comparative Method. The American Political Science Review, 65(3), 682–693. Lüthje, T. (2001). Intra Industry Trade in Intermediate Goods. International Advances in Economic Research, 7(4), 393–408. Manger, M. (2009). Investing in Protection: The Politics of Preferential Trade Agreements Between North and South. Cambridge University Press. Mansfield, E., & Milner, H. (1999). The New Wave of Regionalism. International Organization, 53(3), 589–627. Mansfield, E., & Reinhardt, E. (2008). International Institutions and the Volatility of International Trade. International Organization, 62(4), 621–652. Mansfield, E., Milner, H., & Jon, P. (2008). Democracy, Veto Players and the Depth of Regional Integration. The World Economy, 31(1), 67–96. Manzetti, L. (1993). The Political Economy of Mercosur. Journal of Interamerican Studies and World Affairs, 35(4), 101–141. Mercosur. (2021) ¿Qué es el MERCOSUR? Retrieved from https://www.mercosur.int/quienes- somos/en-pocas-palabras/ 272 Mesquita Moreira, M. (2018, May). Connecting the Dots: A Road Map for a Better Integration of Latin America and the Caribbean. Inter-American Development Bank. https://doi.org/10.18235/0001132 Milner, H. (1999). The Political Economy of International Trade. Annual Review of Political Science, 2(1), 91–114. Milner, H., & Kubota, K. (2005). Why the Move to Free Trade? Democracy and Trade Policy in the Developing Countries. International Organization, 59(1), 107–143. Miyagiwa, K., Song, H., & Vandenbussche, H. (2016). Size Matters! Who is Bashing Whom in Trade War? International Review of Economics & Finance, 45(1), 33–45. Monge-González, R., & Rivera, L. (2018). The Political Economy of Trade Openness in Costa Rica. IDB Working Paper (March 2018. https://publications.iadb.org/publications/english/document/Productive-Development- Policies-in-Costa-Rica-Market-Failures-Government-Failures-and-Policy-Outcomes.pdf Moreira, M., Stein, E., Li, K., Merchán, F., Martincus, C., Blyde, J., Trachtenberg, D., Cornick, J., Frieden, J., & Chatruc, M. R. (2019). Trading Promises for Results: What Global Integration Can Do for Latin America and the Caribbean. Inter-American Development Bank. Mukherjee, B. (2016). Democracy and Trade Policy in Developing Countries. University of Chicago Press. Niu, Z., Liu, C., Gunessee, S., & Milner, C. (2018). Non-tariff and Overall Protection: Evidence Across Countries and Overtime. Review of World Economics, 154(4), 675–703. Noland, M. (2020). Protectionism Under Trump: The China Shock, Deplorables, and the First White President. Asian Economic Policy Review, 15(1), 31-50. Nugent, J. B., Hakimian, H. (2003). Trade Policy and Economic Integration in the Middle East and North Africa. Routledge. O’Farrell, J., Obaya, M., & Marín, A. (2018). The Political Economy of Trade Policy in Argentina under Macri’s Administration (2015-2018). CENIT. OECD. (2005, November). Measuring Globalisation: OECD Handbook on Economic Globalisation Indicators. OECD. https://www.oecd.org/sti/ind/ oecdhandbookoneconomicglobalisationindicators.htm Oliveira, I., da Motta Veiga, P., Polónia Rios, S., & Ribeiro, F. (2018). The Political Economy of Trade Policy in Brazil. Working Paper CINDES/IPEA. https://www.ipea.gov.br/portal/index.php?option=com_content&view=article&id=35017&Ite mid=432 273 Olson, M. (1965). Logic of Collective Action: Public Goods and the Theory of Groups. Harvard University Press. Pastor, M., & Wise, C. (1994). The Origins and Sustainability of Mexico’s Free Trade Policy. International Organization, 48(3), 459–489. Peluffo, A. (2020). China US Trade War: Will it impact on Latin American Countries? Serie Documentos de Trabajo - FCEA, 08(20). Pérez Arguello, M., & Albertoni, N. (2019, June 20). Tariffs on Mexico? That Could Result in Trump’s Nightmare. The New York Times. https://www.nytimes.com/2019/06/20/opinion/tariffs-on-mexico-would-mean-more- immigrants.html Peterson, T., & Thies, C. (2015). Intra-Industry Trade and Policy Outcomes. In Lisa L. Martin (Ed.), The Oxford Handbook of the Political Economy of International Trade (pp. 177-196). Oxford University Press. http://www.oxfordhandbooks.com.libproxy1.usc.edu/view/10.1093/oxfordhb/978019998175 5.001.0001/oxfordhb-9780199981755-e-17 Pierce, J., & Schott, P. (2016). The Surprisingly Swift Decline of US Manufacturing Employment. American Economic Review, 106(7), 1632–1662. Pinna, A., & Lodi, L. (2021). Trade and Global Value Chains at the Time of Covid-19. The International Spectator, 1(1), 1–19. Pita, N. (2017). Is Populism Against Trade? Argentina’s Trade Policy in the Context of Latin American Populism. [Senior Thesis]. University of South Carolina. https://scholarcommons.sc.edu/senior_theses/2017 Puyana, A. (2018). Trump Politics Towards Mexico. Renegotiating NAFTA while Invocating the Monroe Doctrine. Real-World Economics Review, 85(1), 123–141. Ray, E. (1991). Protection of Manufactures in the United States. In D. Greenaway (Ed.), Global Protectionism: Is the U.S. Playing on a Level Field? (pp. 12–36). Macmillan. Ray, E., & Marvel, H. (1984). The Pattern of Protection in the Industrialized World. Review of Economics and Statistics, 66(3), 452-458. Reich, A. (2017). The Effectiveness of the WTO Dispute Settlement System: A Statistical Analysis. EUI Working Paper Department of Law 2017/11. https://cadmus.eui.eu/bitstream/handle/1814/47045/LAW_2017_11.pdf?sequence=1 Reisen, H., & Stemmer, M. (2018, April 17). Shifting Wealth: The Three Acts of the ‘China Shock.’ Shifting Wealth. http://shiftingwealth.blogspot.com/2018/04/the-three-acts-of-china- shock.html 274 Riguzzi, P., & de los Ríos, P. (2012). Las relaciones México-Estados Unidos 1756–2010. Universidad Nacional Autónoma de México. Rodríguez-Piñero, I. (2020). Modernization of the Trade Pillar of the EU-Mexico Global Agreement. EPRS, European Parliamentary Research Service, Members’ Research Service. Rodrik, D. (1995). Political Economy of Trade Policy. In R. Jones, P. Kenen, G. Grossman, & K. Rogoff, (Eds.), Handbook of International Economics (Vol. 3) (pp. 1457–1494). Elsevier. Rodrik, D. (2009, October 12). The Myth of Rising Protectionism. Project Syndicate Blog. http://www.project-syndicate.org/commentary/the-myth-of-rising- protectionism?barrier=accesspaylog Roosevelt, M. (2016, November 20). Disparate Distribution of Exporting vs Import-competing Firms and their Implications for Trade Policy Openness [Working paper presentation]. Annual meeting of the International Political Economy Society, Durham, NC. https://ncgg.princeton.edu/new_ipes/2016/papers/F945_rm3.pdf Ruta, M. (2017). Preferential Trade Agreements and Global Value Chains: Theory, Evidence, and Open Questions. The World Bank Working Paper S8190. http://documents.worldbank.org/curated/en/991871505132824059/Preferential-trade- agreements-and-global-value-chains-theory-evidence-and-open-questions Salinas De Gortari, C. (2002). Mexico: The policy and politics of modernization. Plaza y Janes. Saxton, J. (2003, June). Argentina’s Economic Crisis: Causes and Cures [Testimony]. Joint Economic Committee of the United States Congress. https://www.jec.senate.gov/public/_cache/files/5fbf2f91-6cdf-4e70-8ff2- 620ba901fc4c/argentina-s-economic-crisis---06-13-03.pdf Seawright, J. (2016). Better Multimethod Design: The Promise of Integrative Multimethod Research. Security Studies, 25(1), 42–49. Seawright, J., & Gerring, J. (2008). Case Selection Techniques in Case Study Research: A Menu of Qualitative and Quantitative Options. Political Research Quarterly, 61(2), 294–308. Tanaka, K. (2009). Trade Collapse and International Supply Chains: Japanese Evidence. In R. Baldwin & S. Evenett, The Great Trade Collapse: Causes, Consequences and Prospects (pp. 199-206). CEPR Press. The Economist. (2009, April 30). Low expectations exceeded. https://www.economist.com/united- states/2009/04/30/low-expectations-exceeded ______ (2019, January 24). Globalisation has Faltered: It is Now Being Reshaped. https://www.economist.com/briefing/2019/01/24/globalisation-has-faltered 275 ______ (2020). Will Latin America take advantage of supply chain shifts? The Economist Intelligence Unit. https://www.eiu.com/n/campaigns/will-latin-america-take-advantage-of- supply-chain-shifts/ Thoms, A. (2019, November 1). Why trade wars have no winners. World Economic Forum. https://www.weforum.org/agenda/2019/11/who-benefits-from-a-trade-war/ Tussie, D. (2009). Latin America: Contrasting Motivations for Regional Projects. Review of International Studies, 35(1), 169–188. UN COMTRADE. (2019) The International Trade Statistics Yearbook. United Nations Press. UNCTAD. (2010). International Trade After the Economic Crisis: Challenges and New Opportunities. United Nations Publication. UNCTAD. (2020). Trade and Development Report 2020: From Global Pandemic to Prosperity for All: Avoiding Another Lost Decade. United Nations Publication. Urata, S., & Okabe, M. (2010). The Impacts of Free Trade Agreements on Trade Flows: An application of the Gravity Model Approach. Free Trade Agreements in the Asia Pacific, 11(1), 195–240. Varian, H. (2007, June 28). An iPod Has Global Value. Ask the (Many) Countries That Make It. The New York Times. https://www.nytimes.com/2007/06/28/business/worldbusiness/28scene.html Vaillant, Marcel (2005): Mercosur: Southern Integration under Construction. In Internationale Politik Gesellschaft No. 2 (2005): pp. 52-71. https://mpra.ub.uni- muenchen.de/id/eprint/47251 Vickers, B. (2012). The Role of the BRICS in the WTO: System-Supporters or Change Agents in Multilateral Trade. In A. Narlikar, M. Daunton, & R. Stern (Eds.), The Oxford Handbook on the World Trade Organization (pp. 254–273). Oxford University Press. Waltz, K. (2000). Structural Realism after the Cold War. International Security, 25(1), 5–41. WEF. (2019). Global Future Council on International Trade and Investment. World Economic Forum. https://www.weforum.org/communities/gfc-on-trade-and-investment Wilkinson, R. (2006). Multilateralism and the World Trade Organisation: The Architecture and Extension of International Trade Regulation. Routledge. Williamson, J. (1990). What Washington Means by Policy Reform. In J. Williamson (Ed.), Latin American Adjustment: How Much Has Happened (pp. 353–420). Institute for International Economics. Wise, C. (2020). Dragonomics. How Latin America is Maximizing (or Missing Out on) China’s International Development Strategy. Yale University Press. 276 Wise, C., & Quiliconi, C. (2007). China’s surge in Latin American Markets: Policy Challenges and Responses. Politics & Policy, 35(3), 410–438. Wise, C., Elliott Armijo, L., & Katada, S. (2015). Unexpected Outcomes. How Emerging Economies Survived the Global Financial Crisis. Brookings Institution Press. Wolf, M. (2020, June 23). The dangerous war on supply chains. Financial Times. https://www.ft.com/content/e27b0c0c-1893-479b-9ea3-27a81c2506c9? World Bank. (2020a). World Development Report 2020: Trading for Development in the Age of Global Value Chains. World Bank. ______ (2020b). Global Economic Prospects. A World Bank Group Flagship Report. http://pubdocs.worldbank.org/en/267761588788282656/Global-Economic-Prospects-June- 2020-Highlights-Chapter-1.pdf Wozniak, P., & Galar, M. (2018, January). Understanding the Weakness in Global Trade. European Commission, Economic Brief 033. https://ec.europa.eu/info/sites/info/files/economy- finance/eb033_en_0.pdf WTO. (2016, July 4). Report to the TPRB from the Director-General on Trade-Related Developments. World Trade Organization. https://www.wto.org/english/news_e/news16_e/trdev_22jul16_e.htm ______ (2017a, November 9). World Trade Organization. Report on G20 Trade Measures. World Trade Organization. https://www.wto.org/english/news_e/news17_e/g20_wto_report_november17_e.pdf ______ (2017b). World Trade Statistical Review 2017. World Trade Organization. https://www.wto.org/english/res_e/statis_e/wts2017_e/wts17_toc_e.htm ______ (2017c). Trade Policy Review Brazil. WTO Secretariat WT/TPR/S/358. https://www.wto.org/english/tratop_e/tpr_e/tp458_e.htm ______ (2017d). Trade Policy Review Mexico. WTO Secretariat WT/TPR/S/352. https://www.wto.org/english/tratop_e/tpr_e/tp452_e.htm ______ (2020a). Trade falls steeply in first half of 2020. WTO Secretariat. https://www.wto.org/english/news_e/pres20_e/pr858_e.htm ______ (2020b). Trade Set to Plunge as COVID-19 Pandemic Upends Global Economy. WTO Secretariat 20-2749. https://www.wto.org/english/news_e/pres20_e/pr855_e.pdf Xi, J. (2018). Keynote speech by President Xi Jinping at opening ceremony of 1st China International Import Expo. http://www.xinhuanet.com/english/2018-11/05/c_137583815.htm 277 Yi, K.-M. (2009). The Collapse of Global Trade: The Role of Vertical Specialisation. In R. Baldwin & S. Evenett (Eds.), The Collapse of Global Trade, Murky Protectionism, and the Crisis: Recommendations for the G20 (pp. 45-48). CEPR Press. Yu, Z. (2000). A Model of Substitution of Non-Tariff Barriers for Tariffs. The Canadian Journal of Economics, 33(4), 1069–1090. Zabludovsky, J., & Pasquel, L. (2010). The Case of Mexico. In A. Capling & P. Low (Eds.), Governments, Non-State Actors and Trade Policy Making (pp. 89–124). Cambridge University Press. Zandi, M., Kamins, A., & Cohn, J. (2018). Trump Trade War. Moody’s Analytics. https://www.economy.com/getlocal?q=d3a731c7-0018-450d-90f5- 20e2efa8c6e0&app=eccafile 278 APPENDICES Appendix – Chapter 3 In the following figure, I use the wordcloud2 package in R to create a diagram of words that appear with high frequency in the official documents announcing trade policy instruments, and then I divide them according to transparent versus nontransparent measures. The word cloud keeps track of words that are mentioned frequently (more than 10 times) in official documents. For a better visualization, I limit the number of words in the cloud to 2,500. As we can see in Appendix Figure 3.1, in the section addressing transparent measures there are special references to countries, something we don’t see in the non-transparent section. In sum, this descriptive text analysis shows that the transparency of a measure is not only associated with the type of policy but also how it is officially publicized by the implementing country. Figure A3.1 – World Cloud of Most Used Words in Official Documents Announcing Trade Policy Instruments Between 2009 and 2019 Source: Author’s creation based on the Global Trade Alert. Appendix – Chapter 4 279 Figure A4.1– Global Commercial Flights Carrying Air Cargo Have Started to Rise Note: The number of flights fell 74% between January 5, and April 18, 2020, but have since risen. While these increases are from historic lows, container-port throughput also looks like it is recovering, and other indicators of exports show some improvement (WEF, 2020, par. 9). Source: WTO, 2020, <https://www.wto.org/english/news_e/pres20_e/pr858_e.htm>. Figure A4.2 –Container-Port Throughput, 2007-2020 Note: The WTO highlights that, while container-port throughput showed a partial recovery, that followed near-historic declines, and the WTO says it’s too early to draw any definitive conclusions (WEF, 2020, par. 10). Source: WTO, 2020, <https://www.wto.org/english/news_e/pres20_e/pr858_e.htm>. Figure A4.3 – Indices of new export orders started to recover in May 280 Note: After record drops, purchasing managers’ indices showed export orders started to recover in May 2020 (WEF, 2020, par. 11). Source: WTO, 2020, <https://www.wto.org/english/news_e/pres20_e/pr858_e.htm>. Figure A4.4 – Signs of Better Auto Sales Note: Automobile sales also show signs of recovery from lows. Sales of cars in China fell almost 80% in February 2020 compared with a year earlier but rose 5% year-on-year in May 2020. Declines in car sales in Western Europe and the U.S. slowed in May 2020. The WTO says these tentative signs of recovery will be monitored closely as lockdowns are lifted around the world, and economies start to open up. Source: WTO, 2020, <https://www.wto.org/english/news_e/pres20_e/pr858_e.htm>. 281 Appendix Chapter 5 Table A5.1 – Measures of Uncertainty for the COVID-19 Crisis Note: The VIX is the implied volatility (over the next month and over the next 24 months) on the S&P 500 index from the Chicago Board of Options Exchange, expressed in annualized units. Values downloaded from Federal Reserve Bank of St. Louis, <https://fred.stlouisfed.org/series/VIXCLS>. The daily Economic Policy Uncertainty index values are from <http://www.policyuncertainty.com/media/All_Daily_Policy_Data.csv> and constructed as described in Baker, Bloom and Davis (2016). Subjective sales growth uncertainty is computed as the activity- weighted average of firm-level subjective uncertainty values, which are computed as the standard deviation of each firm’s subjective forecast distribution over its own future sales growth rate from the current quarter to four quarters hence (Altig et al., 2020b). U.S. data are from the Survey of Business Uncertainty conducted by the Federal Reserve Bank of Atlanta, Stanford University, and the University of Chicago Booth School of Business. UK data are from the Decision Maker Panel Survey conducted by the Bank of England, Nottingham University, and Stanford University. Forecast disagreement is measured as the standard deviation across forecasters of one-year- ahead annual real GDP growth rate forecasts. U.S. data are from the Survey of Professional Forecasters conducted by the Philadelphia Fed. UK data are from the Survey of External Forecasters conducted by the Bank of England. Source: Altig et al. (2020), Economic Uncertainty in the Wake of the COVID-19 Pandemic, < https://voxeu.org/article/economic-uncertainty-wake-covid-19-pandemic> 282 Table A5.2 – Example of an Initial E-mail to Potential Interviewees (Interview Questions and Informed Consent Disclosure will be Attached) Dear [INSERT NAME], I am very pleased to be in touch with you through [INSERT NAME OF CONTACT]. I am currently a first year Ph.D. candidate in Political Science and International Relations at the University Southern California in Los Angeles, California, where I study the intersection of international political economy and comparative politics. I am currently working on my dissertation research project to investigate the impact and responses of various Latin American countries to protectionism originated after the global financial crisis of 2008. As part of this research I am conducting interviews regarding this topic. I am hoping I can have your opinion given your experiences and understanding about recent trade policy dynamics in Latin America. I would only ask for about a half hour to an hour of your time, by Zoom or Skype. I would be available anytime. I’ve attached a list of talking points to this e-mail for the conversation. If you decide you are interested, our discussion does not have to focus on these questions, but I thought they might be a good starting point. I would be very pleased to address other issues you think are important along the way. In the event that you wish to keep some parts of the interview, or even the entire interview, as a “non- attributable” version, we appreciate that you indicate it from the beginning and in those cases your will not be cited by your name. Additionally, if you know of any other experts who may be interested in giving me his opinion about this topics, please feel free to pass along my contact information. Finally, a bit of background about me: I am a Ph.D. Candidate in Political Science and International Relations at the University of Southern California. My research broadly engages questions in international political economy and comparative politics with an emphasis on Latin America trade policy-making and economic integration. My dissertation project, The Risk of Trade Protectionism in an Era of Trade Interdependences and Economic Uncertainty, analyzes how states ––especially in the developing world–– have been affected by and responded to rising protectionist trends that occurred in context of high trade interdependence (through preferential trade agreements and global value chains) and economic uncertainty. I hope this gives you a picture of my research project and some information about me. I am happy to answer any questions you may have either by e-mail, Zoom or Skype. Thank you very much for your time, and I look forward to hearing from you soon. Best regards, Nicolás Albertoni 283 Table A5.3 – Example of a Full Interview Sent to Interviewed EXPERTS INTERVIEW ENGLISH VERSION INTRODUCTION This research work is for academic purposes only and is coordinated by Nicolás Albertoni, PhD candidate of the University of Southern California. The research is titled: The Risk of Trade Protectionism in an Era of Trade Interdependences and Economic Uncertainty. The study analyzes the responses of various Latin American countries to protectionism originated after the global financial crisis of 2008. CONSENT: When responding to this interview, the interviewee authorizes the researcher to include the answers with the corresponding appointment, as part of the aforementioned academic study. In the event that the interviewee wishes to keep some parts of the interview, or even the entire interview, as a “non- attributable” version, we appreciate that you indicate it from the beginning and in those cases the specific name of the interviewee will not be mentioned. Thank you very much for the time spent answering this interview. Your answers will be of great value for this academic work. The answers have no word limit. Once this interview has been answered, please send it to nalberto@usc.edu A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? 284 ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader Latin America, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work “in silence” and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What’s is your opinion about this thought? 4. Finally, a question about the political economy of trade policy (responses): As a trade expert of Latin America, what you can say about how interest groups reacted to these current protectionism trends (you can concentrate in the case of your country)? Did they pressure for retaliation or domestic measures (e.g., tax reduction) to compensate for the losses of protectionism measures implemented from other countries? Finally, can you give a short BIO that we can use for the introduction when we cite your answers? 285 Interview A5.1 – Full Interview with Victor Stolzenburg (Interview ID: 001) Interview ID: 001 Name: Victor Stolzenburg BIO: Research economist with the Economic Research and Statistics Division (ERSD) of the World Trade Organization. He coordinates the WTO’s Global Value Chains Report. Disclaimer: The opinions expressed in this interview are those of the interviewee acting in a personal capacity. They are not intended to represent the positions or opinions of the WTO or its members and are without prejudice to members’ rights and obligations under the WTO. Any errors are attributable to the interviewee. 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: The network of trade agreements and the rise of global value chains have undoubtedly impacted the way in which countries design trade policy since they have changed the underlying parameters of trade policy in numerous ways. Most of the effects have been in favor of a more liberal trading system but not all. First and foremost, trade agreements (the most important one in this context being the WTO) have limited countries’ policy space. In order to obtain greater market access abroad, countries have committed to • keep their tariffs below certain bound levels, • move away from quantitative restrictions, • improve transparency on non-tariff barriers, • provide equal treatment to foreign and domestic goods, • do not differentiate among goods from WTO members, • etc. This, by default, implies that countries’ responses to events like the financial crisis, that traditionally would have led to higher policy-related trade costs, are more muted since they operate in a more confined environment. Of course, there are exceptions within trade agreements. Prominent examples are trade remedy provisions or GATT ARTICLE XX. This still leaves considerable policy space in times of unusual economic distress. Nevertheless, policy space has increasingly become smaller with the spread of trade agreements. Secondly, trade agreements and global value chains have led to a dramatic increase of trade flows both in absolute numbers and relative to GDP. This, in turn, has altered the political economy of trade policy. The political economy of trade policy evolves around the conflicting interests and lobbying power of import- competing firms vis-à-vis exporting and import-dependent firms. Third, the impact of GVCs operates through the raised economic costs caused by higher trade costs. GVCs describe the fact that domestic firms increasingly source inputs from foreign countries which themselves have sourced their inputs from abroad. Similarly, domestic firms sell their products increasingly to foreign countries for further processing and export. This has led to a complex network of cross-country dependencies and exposures. As a result, raising trade costs has strong ripple effects that get magnified through the supply chain and are harder to predict and manage. This makes the case for protectionist policies less appealing because the damage from it is likely to be much larger than in a non-GVC world. However, not all effects of trade agreements and GVCs work in favor of an ever-freer trade regime. As trade agreements have increasingly become deeper, they touch upon a wider array of subjects, including sensitive subject like food or health regulations. This has caused a strong backlash in some countries against trade agreements from CETA to TTIP 286 or TPP. Similarly, the offshoring of jobs that is associated with GVCs could have weakened the political sway of trading firms as they have reduced their domestic workforce. Nevertheless, I believe that the net effect of trade agreements and GVCs on trade policy is positive in the sense that they have constrained higher tariffs and NTMs. This is supported by data looking at the response of countries to the financial crisis. While trade costs went up to some extent with countries using sometimes opaque trade-restricting measures, the response was much smaller than in previous major recessions (see, for instance, https://voxeu.org/article/wto-success-no-trade-agreement-no-trade-war and https://www.sciencedirect.com/science/article/pii/S1877042812036749/pdf?md5=af88c5589ed188bc1f3411e9b c084e75&pid=1-s2.0-S1877042812036749-main.pdf). 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from the influence groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: There is certainly something to this argument. The more countries have constrained their tariffs, the more likely it is that they switch to alternative measures that can reach the same objectives. In line with this, several studies have observed such trade policy substitution and how it relates to the tariff water—i.e., the difference between bound and applied duties – that is available to countries (see, for instance, https://ideas.repec.org/a/spr/weltar/v155y2019i4d10.1007_s10290-018-00338-7.html; https://link.springer.com/article/10.1007/s10290-018-0317-5, https://unctad.org/en/PublicationsLibrary/ditctab2018d3_en.pdf). That said, this does not render trade agreements meaningless. They make non-tariff measures more transparent by imposing notification requirements and by limiting the use and design of these measures. They also provide countries with temporary tools to respond to protectionist pressures such as antidumping or countervailing duties which have been successful in keeping countries from using either tariffs or NTMs (see, for instance, https://www.wto.org/english/res_e/reser_e/ersd201906_e.htm). Finally, the rise of NTMs observable in the data is also likely to be driven by changes in societies' preferences related to sensible policy objectives. With a rising demand for sustainable and fair production, there is for example a heightened need for labelling and testing measures. These might be trade-restrictive, but protectionism is not the driving force. 287 Interview A5.2 – Full Interview with Alejandro Jara (Interview ID: 002) Interview ID: 002 Name: Alejandro Jara BIO: Chilean lawyer, trade negotiator, former Director General of International Economic Relations, former Ambassador to WTO and former WTO Deputy Director General. Presently practices international trade law in Santiago, Chile, and teaches trade policy in Chile, China, and Mexico. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: First, the proliferation of PTAs is due mainly to the fact that exporters hate being discriminated against. Thus, if the U.S. and South Korea announce they will hold negotiations for an FTA, the EU cannot tolerate being crowded out of the attractive market of South Korea and pushes the latter to also negotiate an FTA with them. This I’ve seen happen many times. When MERCOSUR was formed, or when Mexico and three Central American countries concluded an FTA, the Chilean exporters reacted by demanding that the government should also negotiate FTAs with MERCOSUR and Central America in order not to lose market. Once concluded, and provided these agreements are successful in creating new trade flows, they are difficult to undo or revert because there will be a constituency to protect their interests. That said, this observation is valid as long as the PTAs are truly free trade agreements that the parties fully implement. This excludes agreements that portray as FTAs (or Customs Union) but are honored in the breach. GVC is a different story because they have a solid regional basis (Asia, North America and Europe) but the trade, capital and technology of some countries participate globally (China, U.S. and possibly Germany and Japan). Vast areas of the world do not really participate in GVC for reasons that may vary depending on the jurisdiction. For example, with the exceptions of Mexico and Costa Rica, the rest of Latin America is not part of GVC (there are exceptions of course such as EMBRAER of Brazil). Be that as it may, GVC, and consequently the need to measure trade on the basis of value added to have an accurate picture, shows that over time the origin of a good is more and more difficult to determine. Increasingly it is not possible to differentiate between us and them, making protectionism irrelevant and counterproductive. If GVCs are the most efficient way to organize production, then the costs of a reversal may be such that the process cannot be easily undone. The surge in protectionism is a reality; first as a result of the 2008-9 crisis and later in Trumpian world. But perhaps such protectionism is reduced to certain sectors (steel, aluminum, are examples) and not a widespread phenomenon. Until the Trumpian world, a great deal of protectionism took place as a result of subsidization and trade remedies (particularly Antidumping) which may indicate that the appeals for protection were presented as either “I’m too big/strategic to fail” or “It’s about fairness not about protectionism.” 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: This is partly answered in the previous reply. Exception made of the Trumpian world of tariff wars, I agree that trade restrictions are less about tariffs and more about behind-the-border measures. It is more effective and politically acceptable to gain protection through regulations made for the purpose of promoting a 288 public policy such as protecting the environment or health, etc. The process leading to drafting standards (or sanitary or phytosanitary measures) is central to ensure transparency and prevent it from falling captive to specific interests that reduce welfare. It is imperative that at the multilateral, regional and bilateral level, governments commit to open and transparent regulatory processes. ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader Latin America, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could make us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: First, in today’s world there is no such thing as being able to “silently” implement policies that go unnoticed to the rest of the world. The union leader probably has a pre-internet mindset. Argentina’s informal licensing system under Kirchner/C. Fernandez and the ensuing litigation in WTO illustrates this very well. That said, some get away with some murder because of the opaqueness of government (Russia and China) or are closed (North Korea). Even then anecdotal and indirect evidence are bound to be able to pinpoint to protectionist or discriminatory policies and measures. 4. Finally, a question about the political economy of trade policy (responses): As a trade expert of Latin America, what you can say about how interest groups reacted to these current protectionism trends (you can concentrate in the case of your country)? Did they pressure for retaliation or domestic measures (e.g., tax reduction) to compensate for the losses of protectionism measures implemented from other countries? ANSWER: I believe the answer to be complex because the realities are very different in each country. For example, Argentina and Brazil show a flurry of demands for the application of trade remedies, especially anti- dumping duties, and particularly with respect to China. The same is true for Chile, but with more limited results. Steel, dairy and wheat are “frequent flyers” of trade remedies. The steel industry has sought different kinds of protection against products from China, Mexico and Turkey. It has been successful with downstream consequences but contained because of the limits the Chilean law imposes on the duration of the measures. With regards to dairy, part of the producers regularly seek protection, but their problem is insecurity (conflict with indigenous people in Southern Chile and drought). The petitioners have had very limited success. On wheat and wheat flour, the problem has been mainly Argentina. Measures have been applied and litigation in WTO has taken place with results unfavorable to Chile. Over the last years higher international prices have quieted down claims in this sector. 289 Interview A5.3 – Full Interview with Jorge Sahd (Interview ID: 101) Interview ID: 101 Name: Jorge Sahd BIO: Director of the Center for International Studies at the Catholic University of Chile. Member of the Foreign Policy Council. Assistant professor of Economic Law and Government and Business at the School of Law. Additionally, he is the academic coordinator of the Probidad, Transparencia y Buen Gobierno (Integrity, Transparency, and Governance) academic program, and member of the United Nations Convention against Corruption (UNCAC). A LATIN AMERICAN PERSPECTIVE 1. Ten years have passed since the global financial crisis (CFG) of 2008-2009. This has generated new dynamics in the global economy, including what’s been called “the collapse of global trade” (Baldwin 2009). In recent years both developed and developing countries have increased some barriers to international trade. How do you think Latin America in general has responded to these new protectionist tendencies that have occurred in the last decade? Do you think there are different types of responses depending on the country analyzed? ANSWER: I think that Latin America has been “resting on their laurels” regarding advances in integration. The Pacific Alliance, whose start was very auspicious, has had a slow progress in recent years. Mercosur, although it reached an agreement with the European Union, remains a block highly dependent on political power, which ends up conditioning its progress. In addition to low intra-regional trade, the region has a space to advance in trade in services and greater integration of its value chains. 2. Recent studies have shown that, although there have been no significant increases in trade tariffs in Latin America, non-tariff mechanisms such as subsidies to local companies and quotas, among others, have been used. Do you think that the type of response that Latin America has had to the new reality of the world economy has been more through non-tariff than tariff? If so, do you think this shows that the region still has necessary structural trade reforms? ANSWER: Indeed, the restrictions in the region have a relevant component of non-tariff barriers, where the lack of regulatory convergence and excess administrative requirements end up making the integration process difficult. Undoubtedly, internal structural reforms are necessary to maximize the benefits derived from trade. A GLOBAL PERSPECTIVE 3. A differential aspect of this last decade, with respect to other events of the past in which there have been commercial tensions, is that today the world is more interconnected through a network of. commercial agreements and global value chains. Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: Without a doubt, this wide network of agreements has been fundamental. In the Asia-Pacific area, the most dynamic economic region in the world, there is an important development of the value chains of Asian countries and China, as well as the recent integrative efforts of continents such as Africa. In addition, this network of treaties establishes mechanisms to resolve commercial disputes or investor-State treatment that have contributed to give greater stability and certainty to economic relations between countries. The current challenge is whether we will witness a phenomenon of deglobalization or globalization without the United States. 4. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that 290 integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from the influence groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: Agree. The risk today is called “protectionism 2.0,” which is behind the borders. Indeed, that is the way some countries impose protectionist measures on countries, even if they have signed free trade agreements. A GENERAL CONCLUSION 5. Finally, several studies have highlighted that a special feature of the last CFG, and the subsequent “collapse of global trade” (Baldwin 2009), has been the non-existence of a massive tendency to protect markets as if it happened in similar events throughout history (for example, in the crisis of 1929). Many of these studies have focused specifically on the measurement of tariff measures. From a perfect generation, what role do you attribute to the non-tariff and less transparent measures applied by various developed and developing countries in the last decade? Does the increasing use of non-tariff and less transparent measures for trade protection not reveal the lack of reforms that exist in the multilateral trading system to regulate these problems? ANSWER: Agree. One of the criticisms of the WTO is its lack of updating the rules and the inability to respond to these new forms of protectionism. The challenge, no doubt, is to point to these new restrictions, ranging from state subsidies, intellectual property infringements, etc. 291 Interview A5.4 – Full Interview with Andrés Rebolledo (Interview ID: 102) Interview ID: 102 Name: Andrés Rebolledo BIO: Vice Minister of Trade of Chile (2014-2016) and Minister of Energy of Chile (2016-2018). A LATIN AMERICAN PERSPECTIVE 1. Ten years have passed since the global financial crisis (CFG) of 2008-2009. This has generated new dynamics in the global economy, including what’s been called “the collapse of global trade” (Baldwin 2009). In recent years both developed and developing countries have increased some barriers to international trade. How do you think Latin America in general has responded to these new protectionist tendencies that have occurred in the last decade? Do you think there are different types of responses depending on the country analyzed? ANSWER: There is no response to this crisis in Latin America, there have been several models in the region for years, reflecting fragmented integration processes and almost no coordination of the region in global issues. This crisis occurs at a time when some Latin American countries embraced with determination the commercial opening and the negotiation of commercial agreements. On the other hand, especially the Atlantic countries, a more heterodox model that combines an international projection effort with the protection of their domestic and subregional markets. Trade barriers that proliferate in the world are likely to have consolidated and deepened the region's trade patterns with the rest of the world, particularly in the northern hemisphere (they should be ratified with data). A post-crisis scenario and weakened global rules is the most complex for our region. A challenge for the region will be to find its role and comparative advantages in the context of an international trade that incorporates the effects of the digital revolution. 2. Recent studies have shown that, although there have been no significant increases in trade tariffs in Latin America, non-tariff mechanisms such as subsidies to local companies, quotas, among others, have been used. Do you think that the type of response that Latin America has had to the new reality of the world economy has been rather through non-tariff than tariff measures? If so, do you think this shows that in the region there are still necessary structural reforms related to trade? ANSWER: Tariffs in the region are less and less relevant as protection measures. In fact, except Venezuela, Mercosur remains the set of countries that have the highest applied tariff in the region at a level of around 10%. Effective protection, and tariff escalation with high peak is where a relevant level of protection is found. It is precisely formal non-tariff barriers and discriminatory administrative practices that really aim to manage trade. In any case, the above is not recent, but a traditional practice in our region. The structural reforms that our countries require have to do with the challenge of productivity and additional international competitiveness. They are unavoidable tasks for the governments of the region, beyond foreign trade. A GLOBAL PERSPECTIVE 3. A differential aspect of this last decade, with respect to other events of the past in which there have been commercial tensions, is that today the world is more interconnected through a network of commercial agreements and global value chains. Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: The paradox is that the normative effort of protectionism tends to be overcome by the technological disruption that jumps borders and brings markets closer. That is, the challenge, how to govern this phenomenon, but not try to oppose. At least, that is not an option for medium, small and peripheral countries. Indeed, the second best to integrate into international markets has been the negotiation of many bilateral and plurilateral trade agreements. In general, these Agreements have been beneficial for their partners, but 292 systemically they impose the great challenge of segmenting markets and increasing transaction costs in international trade. 4. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressures from influence groups that, following a global protectionist trend, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: The first this is regarding the incorporation into global value chains, the countries of the region still do not find the appropriate formula and policies to benefit from this phenomenon. Perhaps it is one of the most important challenges for the region. It is very difficult to find a specific chapter that talks about NTMs in a PTA. What we do find is specific articles that talk about specific types of measures such as sanitary and phytosanitary measures, and technical barriers to trade, which are examples of NTMs. 202 However, in all cases these chapters tend to be very general, in which countries commit to transparency in the implementation of these norms. From my experience, what PTAs do with regard to NTMs, is to discipline their implementation but not necessarily to remove them for the simple reason that enforcement here is very difficult. Once a trade agreement is signed, I would say that 80% of the conflicts are related to non-tariff measures. And the main problem is that we rarely find specific and clear chapters on how NTM restrictions will be enforced. I think the problems is not the PTA itself but how countries use these less transparent norms within a PTA. In sum, all this discussion is a good example that PTAs are neither heaven nor hell, they are just one instrument that by itself doesn’t do too much if there is no commitment to transparency by trading partners. A GENERAL CONCLUSION 5. Finally, several studies have highlighted that a special feature of the last CFG, and the subsequent “collapse of global trade” (Baldwin 2009), has been the non-existence of a massive tendency to protect markets as if it happened in similar events throughout history (for example, in the crisis of 1929). Many of these studies have focused specifically on the measurement of tariff measures. From a perfect generation, what role do you attribute to the non-tariff and less transparent measures applied by various developed and developing countries in the last decade? Does the increasing use of non-tariff and less transparent measures for trade protection, does it not reveal the lack of reforms that exist in the multilateral trading system to regulate these problems? ANSWER: Developing countries tend to apply less transparent measures to protect their markets compared to developed countries. However, I believe that the last few years are measures of this type that restrict trade and less tariffs. The future evolution of trade barriers is likely not even to be applied at borders, but in domestic markets that raise standards and requirements for product marketing. Undoubtedly, the weakening of multilateralism and the global governance of trade issues has to fragment the markets, and it is a step backward towards the traditional logic of the great and powerful in front of its trading partners. 202 Examples of this in recent agreements are Chapter 9 and 11 in the USMCA and Chapters 5 and 6 in the Mercosur and EU trade agreement. See Agreement between the United States of America, the United Mexican States, and Canada 12/13/19 Text, <https://ustr.gov/trade-agreements/free-trade-agreements/united-states-mexico-canada-agreement/agreement- between>; and Mercosur and EU agreement: <https://publications.iadb.org/pt/acordo-mercosul-uniao-europeia-impactos- normativosregulatorios-no-mercosul> 293 Interview A5.5– Full Interview with Eduardo Egas (Interview ID: 103) Interview ID: 103 Name: Eduardo Egas BIO: Executive President of the CORPEI Export and Investment Corporation; former Minister of Industries; former Vice Minister of Foreign Trade, Ecuador. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: Definitely yes. In recent times, many trade agreements have emerged, but they’re more bilateral than multilateral; which causes more distortions in trade for those countries that are outside of these agreements. For example, Colombia and Peru independently have trade agreements with Mexico. Ecuador only has a partial agreement, so when entering that market, we must pay a much higher tariff for our cocoa, for example, than the one paid by our competitors in Colombia and Peru. Mexico has strong protection for its fishing industry, making it very expensive for Ecuador to enter its shrimp and tuna. Faced with both previous cases, and others similar, Ecuador is determined to negotiate a comprehensive trade agreement with Mexico, yielding to Mexico's claims, to defeat its protectionist measures. On the other hand, the integration of global value chains in certain sectors or products provokes a mixture of interests among the countries that participate in it. Here protectionism is not an option but collaboration. An example of this is given to the tuna industry in Ecuador, which is an important supplier to the European canning industry. This complementarity means that Ecuadorian tuna (raw material) is not impeded from entering the European market, but instead, Europe places great protectionism on its canning industry (manufactured product). Without doubt, trade agreements and the establishment of global value chains have diminished protectionist actions at the global level. 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: This pressure from local groups is always present and it only depends on the authorities to yield or maintain respect for the commercial agreements signed with their foreign partners. Influence groups seek restrictions on the side of sanitary measures, technical measures or environmental measures, justified by an apparent consumer defense. These types of restraining weapons are very common in recent times and often exceed the real response capacities of the supplying countries. ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader Latin America, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could make us more competitive without being as obvious to the rest of the world. We are not pretending to hurt 294 our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: In Latin America we face the permanent fight to have a product in conditions that can compete with those offered from other parts of the world, mainly because the innovation applied by our competitors is causing better and more desired products. Some businessmen confuse what must be understood by competitiveness and believe that it is to receive internal covert subsidies, tax exemptions, permanent devaluations, among other privileges that, in reality, do not make us more competitive. Our challenge is to be more competitive because we give more value to our resources; because we produce more efficiently; because we have good strategies that allow us the sustainability of our product in international markets for a long time. 4. Finally, a question about the political economy of trade policy (responses): As a trade expert of Latin America, what you can say about how interest groups reacted to these current protectionism trends (you can concentrate in the case of your country)? Did they pressure for retaliation or domestic measures (e.g., tax reduction) to compensate for the losses of protectionism measures implemented from other countries? ANSWER: These are very common reactions from interest groups in our countries. When they find protectionist positions in their markets of interest and, mainly, when the protectionism used is “gross” or not logical, they lobby with the government to request similar response measures. Ecuadorian shrimp has been making efforts, for several years, to unblock sanitary measures that prevent its entry to the Brazilian market. Brazilian pressure groups, made up of local shrimp farmers, have managed to keep protectionist measures from their government for several years. Neither diplomatic efforts have served to unlock these restrictions. This position has been so absurd and in violation of the rules that the shrimp farmers and Ecuador's public opinion requested equivalent sanctions against products from Brazil. 295 Interview A5.6 – Full Interview with Dorotea López (Interview ID: 104) Interview ID: 104 Name: Dorotea López BIO: Director of the Institute of International Studies, Universidad de Chile. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? Today there is much less consensus on the advantages of trade openness that seemed very evident in all countries. The effects of free trade agreements are more legal, rather than instruments of real economic connection. Sometimes I think free trade agreements have been overvalued in terms of what they generate. When one asks students about channels that connect us commercially, free trade agreements is always an answer, when in reality this is not always the case. Just look at Mexico, which has free trade agreements and only remains connected to the United States. I do not know if free trade agreements have been overestimated. So, I’m not really sure if today the world is more connected through PTAs or not. What is certain is that the world is more interconnected by global value chains. Of that there are no doubts. Because production, which is not a new concept either, but of all life in economic theory, makes integration much closer and much more fragmented. And I believe that they undoubtedly have an effect on how countries have responded to protectionist tendencies. They have done so in terms of the interests they have and to what degree of global value chains they are. 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? First, it is worth saying that today we see a returned need for state economic protection. Protectionism today is equal to creativity: whoever manages to be as discreet as possible while being more competitive without openly being protectionist, is the great winner in these processes. So, tariff barriers become less relevant. Today it becomes more difficult to follow up on existing barriers. It is difficult to understand how one jumps from the 2008 crisis to the trade war without an in-depth study of what happened in between. Those who really want to understand the commercial world today should not continue worrying about tariffs but about all the non-tariff measures that are actually the new way of doing trade policy. Therefore, all this protectionism that has arisen, such as subsidies to local companies and promotion to consume products from the country have resurfaced significantly. Furthermore, they have been very difficult to follow up on and very difficult to identify. All of this poses challenges to trade policy. This also happens to financial regulation, which always lags a little behind the creativity of the real market movement. For example, today the great problem in the world of services is the difficulty of identifying the true barrier that can often be complex and leave competitors out. That is why the WTO and all economic theory positions the tariff as everyone's favorite measure. Because if we could get everyone to implement tariffs (instead of non-tariff measures) we would know where protectionism is. 296 ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader Latin America, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could make us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? It would be good to know first, what this trade unionist understands by a trading partner. Is he thinking about Free Trade Agreement partners or those partners that buy and sell to the country? I see that there is this tendency towards silent protectionism. I even think that crises like those that COVID is opening today are going to sharpen this trend. What is certain is that, in general, we have seen a demand for the presence of the State. This is because the market has been super aggressive and also the competition policy laws have been very weak. So, when you enter this type of competition so predatory that it can arise when a free trade agreement eliminates tariffs and you enter competition without sufficiently strong competition policies, the level of depredation is complex. For this reason, there is a return of demand from the State. Without going any further, today in Chile one of the great discussions is the famous subsidiarity of the state. Today Chile is on a return to rethink the role of the State. More in terms of intervention than protectionism. Chile is well aware of its size and its (scarce) pricing power in the world and its dependence on the external sector. Clearly, having an opening coefficient close to 70% maintained in recent years is not the same as being a country like the United States that today has an opening coefficient of 20%. In other words, one is aware that autarky is not the way but that these interventions at this time have to be quieter. 297 Interview A5.7 – Full Interview with Olga Lucia Lozano (Interview ID: 105) Interview ID: 105 Name: Olga Lucia Lozano BIO: Former Vice Minister of Foreign Trade of Colombia; Alternate Representative of Colombia to the WTO, where she chaired the Subsidies Committee. Business consultant for over 15 years. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: The PTAs and the GVCs have specifically influenced the issue of anti-dumping investigations, in which more and more, at least in the case of Colombia, policymakers take into account who they are buying from before making an eventual decision of anti-dumping duty and compensatory measure. The country’s productive system has a reality that is interconnection, since it is difficult to find companies that only base their production on national inputs. So, when a policymaker seeks to put an anti-dumping measure, the voices of those who are within that industry clearly have a role. However, in economies such as Colombia where there is a significant integration of inputs from third countries that we have to incorporate in order to export, although there is an interconnected awareness that is specifically reflected in examples of the one that I have just given you, there is no doubt that the mindset of public officials today in the face of protectionist measures is a totally different one from the one that existed in the 1990s. Today he won the battle of “if we can protect, let’s do it.” I see this in Colombia and in several other countries. For example, I also see it in even Chile where I hear voices defending the application of anti-dumping measures, which previously would consider this as heresy five years ago. So, I think the world has changed and public officials have changed their mental model: they have gone from the religion of the nineties that trade will save the world, to the disappointment of the positive things of trade that began to originate in the 2000s. It began to be seen that the recipe needed to be changed. They go on to take the positive things out of commerce, but with the notion that local industry is important. The important thing is that when an anti-dumping measure is applied, for example, consideration is given to whether the party with which the process is initiated has a trade agreement. In the case of Colombia, for example, there is an interesting element, and that is that Congress approved a law for an analysis of the impact of free trade agreements and it is presented annually. For example, I was in the government in the 1990s, and to think of a countervailing duty against the United States was egregious. No one would have raised that. Today everything changed. For example, Colombia is investigating the U.S. for ethanol subsidies. Today trade relations have migrated towards an understanding that, regardless of the political relationship or even that of the ties between the two economies, it is possible to use measures that commercially defend the national industry. When I refer to interconnection, I mean that, when Colombia is going to take a measure, it asks the industry that could be affected since they consume a certain product and their production. We are interested in knowing if this measure affects someone in the internal market. That is why there is always an analysis of political economy that will always be present and more so for a country like Colombia. 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? 298 ANSWER: Without … doubt trade interconnection has generated less transparency in trade policy. Likewise, Latin America has been less successful than Europeans, for example, in making its trade policy complex in such a way that it continues to protect without making it very obvious. The commercial policy officials today, seeing that in Europe they have applied increasingly transparent measures, feel that if we from Latin America apply some it is more to level the playing field. I believe that all of this changed in the first place because the relative weight of trade policy is no longer the same as it was in the 1990s. Before, any Latin American country had its strategy of internationalization of the economy. It was a fundamental part of his speech and those of us who worked in the commerce sector saw each other periodically. The consensus of those years was to lower tariffs, privatize, etc. That was defended as a religion. Over the years, the world showed that this path was not indicated. They began to realize that perhaps the recipe was misunderstood and that just by opening the markets it was not lacking, and that it was also necessary to complete those policies with others such as infrastructure, education. That is why the result of that recipe was not what was expected. And that is what has been happening in Chile and the social revolts, which led us to ask ourselves more seriously: is free trade the one that will solve Latin America’s problems? Well, no. If I solved them, what happened in Chile would not be happening. 299 Interview A5.8 – Full Interview with Magali Silva (Interview ID: 106) Interview ID: 106 Name: Magali Silva BIO: Former Minister of Foreign Trade and Tourism of Peru and Vice Minister of MYPE and Industry. She is an economist who works at the Central Reserve Bank of Peru as Advisor to the General Management. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: Indeed, the proliferation of PTAs and the possibility of participating in the GVC has been one of the valuable ways in which different countries have responded to current protectionist trends. PTAs allow emerging economies access to technology and knowledge as well as expand markets and respond to more demanding consumer tastes. Participation in GVCs allows emerging economies to integrate into the world economy without having to be competitive throughout the entire production chain. What has definitely not happened in most emerging economies, with the exception of China and the countries of Southeast Asia, are the expected ascents in the value chain towards more sophisticated activities and productions. This would partly explain why in some cases emerging economies have failed to overcome protectionist policies in the developed world. In many cases, developed countries have used sanitary authorizations or permits as para-tariff measures to delay or allow the entry of a new product. 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: From my experience, countries are less protected if they are integrated via global value chains and sell together in third markets, but it did happen in the Peruvian case of applying protectionist measures in countries with which we have free trade agreements. This question raises a hypothesis that may indeed exist. Indeed, when protection is not given at the tariff level but rather at the non-tariff level, through subsidies to local companies, the defense process requires an investigation that generally takes some time. Governments must be well-informed enough not to yield to power groups under political pressure. They must make decisions based on proven evidence for the good of the entire country and not based on private interests. During my three years as a minister, I fought for transparent management. Every time I met with the different actors in the sector, I informed the press about the agreements made. I had to negotiate and sign the Pacific Alliance and the TPP and I had very positive experiences and fortunately few of the others. My strategy was always to find more markets for the smallest companies in the country and to promote the integration of small entrepreneurs. ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader Latin America, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s 300 why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: I firmly believe in the ability of trade to generate new investment and employment opportunities in the poorest countries. However, free trade agreements are not enough to achieve this. It is necessary to work on an internal agenda. It is necessary to fight against informality and corruption. It is necessary to build basic infrastructure that facilitates trade and take advantage of technology to improve access to health and education. Having said this, I am of the opinion that the first thing a country that faces restrictions on the entry of its products, increased controls (para-tariff measures) or increased tariffs, even while in the framework of a treaty, must do is act at the technical level, and apply all the legal mechanisms allowed by the same treaty in the chapter “Dispute Settlement.” The second is to raise the case to the political level, to the ministerial level, and then to the highest level of the heads of state. In both cases, internal communication is extremely important because the affected sectors must be informed. It is necessary to meet with unions and their workers to inform strategies in this situation. You have to take the time and measure the impacts and if you don’t receive positive responses, you need to start reviewing case by case to see the country’s imports that are tightening their conditions. I believe that one must be very respectful of the current legal framework but be very firm in asserting the rights of our countries. In this case, you can be very strict in verifying compliance with the rules of origin or other obligations of the country that applied protectionist practices. In the case mentioned about the union’s position of “asking the government to help them be more competitive in the difficult international context” it seems to me that this is a request that should be more permanent. Governments are strengthened when they receive requests from employers, unions and other workers because they allow them to identify problems and work in coordination to solve them. And it seems to me a very mature position of the union. 4. Finally, a question about the political economy of trade policy (responses): As a trade expert of Latin America, what you can say about how interest groups reacted to these current protectionism trends (you can concentrate in the case of your country)? Did they pressure for retaliation or domestic measures (e.g., tax reduction) to compensate for the losses of protectionism measures implemented from other countries? ANSWER: It is completely natural that when a sector is affected by protectionist practices it reacts quickly. And therefore, there must be an open channel between the government and the private sector to resolve the impasse in the best way and in the shortest possible time. The private sector will present its solutions to the problem from its perspective, and it will correspond to each government, depending on the particularities of each case (number of workers affected, type of industry affected—health, food, energy, or others—importance in generation currency, impact on the supply chain, etc.) find a solution. I am of the opinion that businessmen and government should not be confronted. You need an empathetic government, promoter of free enterprise, that encourages job creation and that listens to the private sector. A new type of entrepreneur is also needed: solidarity, protector of decent employment, who is committed to the health and education of his workers, innovative and respectful of the environment, who pays his taxes. Those are the challenges we have. 301 Interview A5.9 – Full Interview with Juan Rodrigo Moreno (Interview ID: 201) Interview ID: 201 Name: Juan Rodrigo Moreno BIO: Director of International Trade Negotiations of the Employers’ Confederation of Mexico (COPARMEX). About COPARMEX 1. First, we would be interested to know the approximate distribution of companies by sector that form COPARMEX. That is, approximately, how many of the members are agricultural, industrial, etc.? ANSWER: The Employers’ Confederation of the Mexican Republic (COPARMEX, in Spanish) currently has 36,000 members of which 215 correspond to COPARMEX Nacional, these in turn are classified into Large Companies and Associations, the corresponding distribution is as follows: Distribution of COPARMEX by Sector The rest of the members companies are part of the partners attached to one of the 65 COPARMEX Business Centers located throughout the Mexican Republic. 85% of these companies are SMEs. 2. How does COPARMEX see the current international context in which there seems to be a resurgence of protectionist currents in international trade? What is the general vision of the institution regarding this issue? ANSWER: COPARMEX is an institution that is in favor and defends the principles and guidelines of the international multilateral trading system, based on the policies of the World Trade Organization (WTO), and therefore advocates an open free trade system. Mexico seeks to strengthen its commercial relations taking as a reference the various free trade agreements that it has signed, and from the business sector we represent, we defend the importance of strengthening commercial diversification, open and with clear, transparent and equal rules for all. Mexico has achieved a favorable balance in its trade balance and productivity growth thanks to the opening of international flows; This could be strongly affected if we return to the protectionist currents that some countries have been promoting in recent years. From COPARMEX, we are certain that the best way to grow in international markets is with measures of commercial openness that allow and propitiate long-term growth in the economy and an equitable participation of all the actors involved. 3. What concrete measures have been taken by COPARMEX as an institution to deal with possible measures that affect Mexican trade? Have you had meetings with the government to apply measures that help Mexican companies to be better positioned in case protectionism affects them? ANSWER: COPARMEX actively participates in the “Attachment Room,” which is a group that has been created to favor the participation of the business sector in the negotiation processes of free trade agreements in which Mexico is involved. In this group, we work in coordination with the federal government to know and jointly propose the terms of negotiation of any free trade agreement to be concluded by our country. This group has representatives of accredited companies in Mexico, from all productive sectors, as well as those responsible 302 for the area and/or foreign trade issues of the various organizations and business chambers that exist in Mexico. In addition, we work closely with the Ministry of Economy, specifically with the Undersecretary of Foreign Trade, to learn about the plans, programs and projects that are promoted by the federal government to boost the foreign trade of Mexican companies. We maintain a representation chair in these spaces, and we participate in the various events that are held by the Federal Government in this area. We are also the only business body in Mexico that is part of the Business at OECD, the business arm of the Organization for Economic Cooperation and Development (OECD), in which there is a Committee on Trade. Here trends and main challenges of international trade worldwide are discussed, which allows us to stay informed about the main trends, risks and challenges facing international trade in the current context. This allows us to provide Mexican companies with useful information for decision-making so that, as a business sector, we are prepared for any situation that may arise in the aforementioned subject. In 2019, during a working tour of Geneva, our National President, Gustavo de Hoyos Walther, met with Roberto Azevedo, Director General of the WTO, to discuss the concerns of the Mexican business sector regarding international trade and to directly about the opportunities that Mexican companies have to improve their participation and insertion schemes in international trade and the support that this organization can provide to our country to strengthen our trade standards. In addition, at various times we have expressed our disagreement that international trade protectionist measures are taken, and we have observed with concern the actions that various countries have promoted to unleash trade wars or tariff impositions in an arbitrary manner that go against the rules marked by the multilateral international trading system. 4. Finally, from COPARMEX which are believed to have been the sectors within Mexican trade that have been most affected by the latest protectionist currents? If there have been conversations with the government to deal with the current protectionist context, have COPARMEX taken strategies by sectors or general measures? ANSWER: In general terms we observe that there are some sectors that are always more vulnerable in this type of negotiations, but in a particular way there is always a latent concern for the automotive, agri-food and agro- industrial, textile, footwear, agribusiness, digital commerce, and steel industries. 5. Could you provide us with any specific contact from any COPARMEX member company that you think is relevant that we should know your opinion about how commercial protectionism has affected you? We are particularly interested in companies related to the automotive, iron and steel sector, and computer and computer assembly. ANSWER: Before that, it is necessary to carry out a consultation and request the authorization of potential companies to participate in this exercise, since we cannot share your information due to data protection. With pleasure, we can analyze possibilities with a specific proposal for these effects. A GENERAL PERSPECTIVE ON MEXICO 6. Current data related to the new protectionist currents show that, in 2018, approximately 50% of Mexican trade has been exposed to some type of protectionist measure implemented by countries to which Mexico exports. Among the most affected sectors are the automotive sector and iron and steel sectors. When there are direct effects on Mexican industries, do you think Mexico should respond with protectionist measures as well? ANSWER: At COPARMEX we are convinced that protectionist measures to international trade are not a sustainable tool to promote greater or better economic growth. We believe that you should always look for measures and/or solutions that guarantee and favor free and open trade, under general, clear rules and equal opportunities for all actors. We advocate fair trade, based on the guidelines of multilateralism and dispute settlement mechanisms established in each FTA, as well as those promoted by the WTO. However, we must also recognize that Mexico has a lot of experience in tariff and non-tariff measures (depending on the 303 circumstances), carousel strategies that have been used in the past to guarantee mirror and fair conditions, as well as to defend free market principles, commercial reciprocity, most favored nation treatment, etc. 7. When Mexico responds with tariff measures, an escalation of protectionism could be generated. Likewise, some governments in the region, instead of responding with tariff measures, have implemented non-tariff measures (subsidies, for example) that are not easily discernible to other countries and, in turn, cause the government to show itself in response to Local businesses. Do you agree with this type of non-tariff measures? ANSWER: No, these types of schemes only disguise protectionism and prevent other countries that do not have this type of support from competing freely in the market. We have noticed that applying these types of measures often results in issues of illegality and lack of ethics. For this reason, COPARMEX is in favor of tariff policies which benefit national economies by allowing the strengthening of their industries by becoming much more competitive within the international market, which in the end brings greater benefits to their entrepreneurs, their workers and, therefore, to the economy of each country. Some tariff measures have sometimes been justified to guarantee principles of legality and competitiveness of the markets. For example, on some occasion, quotas have been applied to Chinese footwear to protect and safeguard the competitiveness of the production of national footwear. In other cases, there are countries that have government subsidies, important supports that impact the cost of manufacturing products or implement other practices that require an institutional compensation mechanism. The important thing in these cases is to ensure that these measures are within the framework of the legality and of the minimum standards established by the multilateral international trade standards. 8. Finally, we want to ask you a very important question for this study: in recent years, the number of trade agreements signed by various countries has grown, as well as global value chains. Mexico integrates various trade agreements and is also part of some global value chains. Considering the above, how do you think Mexico should respond when a business partner (with whom there is an agreement) or a country with which a global value chain is shared, implements a measure that affects Mexican trade? Should Mexico respond? And to answer, is it better to do it with tariff or non- tariff measures? ANSWER: Tariff policies are synonymous with legality and ethics, values that COPARMEX defends and demands, so the position we take on this issue is the search for negotiations through the corresponding instances and mechanisms for dispute resolution, in order to ensure that our country can participate freely within the international market, demonstrating in this way the capacity and competitiveness of our industries, products, and services offered inside and outside the country. 304 Interview A5.10 – Full Interview with Jesús Flores (Interview ID: 202) Interview ID: 202 Name: Jesús Flores BIO: Director for Trade and Institutional Affairs; Tenaris Mexico. Undersecretary for Administrative Affairs in the Ministry of Commerce and Industrial Development (1994–2000). General Director for Latin America Negotiations and Chief Negotiator of Mexico’s negotiations with Latin America. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: Protectionism is the U.S. reaction to the frustration to negotiate with China adjustment to its economic model dominated by government intervention, SOE’s, subsidies, tax rebates and lack of reform commitments in the WTO. This created excess capacity, market distortions, and monopolistic actions. All this had an important impact in GVC, wealth generation with a relevant impact in social and skill training. Now there are two economic models. Capitalism and state capitalism. This in part explains the NAFTA renegotiation in order to modify the rules of origin; mainly those related to the steel value chain. Also explains the WTO crisis situation. ABOUT MEXICO 1. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: Mexico is not in the condition to apply a protectionism policy. Mexico needs to play a coordinated strategy with the United States. This explains USMCA. In other words, it looks we are moving to blocs competition. 2. Some of the sectors that has been most affected by protectionist measures from the rest of the world was “Products of iron or steel” (525 interventions between 2009-2019) and “Basic iron & steel” (415 interventions). Many of these interventions from the rest of the world have been implemented before the trade war and there were not necessarily tariff measures, most of them were nontariff mechanisms that are less transparent and complex to identify (murky protectionism): 2.1 How much you think your sector has been affected by current protectionist mechanism? ANSWER: Correct. The problem of excess capacity has been under discussion in the OECD Steel Committee for the last 10 years. In the year 2001 China became part of the WTO and negotiated a transition period of 15 years to end its reforms and become a market economy. This did not happen because China did not accept to transparent its information and neither modify its model. The U.S. reacted attempting with different measures, but at the end the Obama administration did not apply drastic actions. The steel sector has been affected by 232 and 301. Now we live in an administrated trade (Quotas, tariffs, TRQs.). 305 2.2 Do you think this protectionist mechanism are coming more from countries that you have strong trade ties (FTAs, or GVCs)? If so, why is that the case? ANSWER: Not necessarily. For example, China conditioned trade and investment subject to share your knowledge and intellectual property. 2.3 When you realized that your sector is affected by a specific protectionist measures imposed by a foreign country, do you talk with the government to impose even retaliatory measures or domestic measures (e.g., tax reduction) to compensate the losses? ANSWER: 232 has been an important change. Now countries like Mexico can export but they need to control subject to historical information. Mexico retaliated, but in other sectors. 306 Interview A 5.11 – Full Interview with Beatriz Leycegui Gardoqui (Interview ID: 203) Interview ID: 203 Name: Beatriz Leycegui Gardoqui BIO: Served as undersecretary of foreign trade at the Secretariat of the Economy of Mexico from December 2006 to August 2011. She has served as legal director of the Trade and Industrial Development Secretariat (now the Secretariat of the Economy) during the NAFTA negotiations. ABOUT LATIN AMERICAN 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: I considered that, in general, for decades there have been different types of Latin America, it is difficult to speak in Latin America integral or with a common vision. That explains the inability we have had to achieve liberalization through a mega regional agreement that includes all countries. Liberalization has occurred in a very fragmented way and there are different times also depending on who is in power at this time in each country, but there are countries that have certainly followed a line of greater protection and not trade liberalization as we organize them internationally. It is nothing else in the wake of the 2009 crisis that certain countries have adopted protectionist policies in particular. Brazil and Argentina are two very important economies in the region who have tended to have closed economies and have bet almost everything on their regional integration from Mercosur. And, in general, they have had partial agreements with nearby countries. So, we have Argentina and Brazil, which after Mexico is the largest economy, has not followed a position of liberalization. There have been two Latin Americas. On the one hand, the entire block where Brazil, Argentina, Venezuela, Bolivia, Ecuador, Nicaragua has been. When Venezuela was strong that time it was strong, it also took several countries to ALBA that had very different. On the other hand, there are the countries that grew in globalization, in the commercial opening, which are Chile, Colombia, Peru, Costa Rica, and Mexico…. These countries have clearly followed another way of thinking. For all this, I do not believe that protectionism began in 2009, perhaps what he did was to accentuate and deepen in certain countries by the arrival of governments that were also antagonistic to globalization, but in general it has been in some countries something common and in the long term, and not only through the use of commercial measures in their own right, but they have also used domestic or national public policies that have also been an important factor in curbing trade. For example, with Brazil, much of the problem with Brazil has been the fiscal issue. It makes it very difficult for countries to enter to make investments, in addition to their tariffs. Because in the country they face other very high taxes and at least heard of Mexican businessmen, discriminatory deals because they have a very complex tax system also at the State level. It is not a general tax system that applies to the entire country, but there is a very burdensome series of local taxes applied. This makes trading with Brazil very complicated. It was in my turn as undersecretary to try to make a free trade agreement with Brazil. I was like a year and a half almost totally dedicated to this effort and I realized that, in addition to the position that there could be in Brazil at this time, it seemed that it was not so much or that the government had more control over that, in Mexico a very strong negative reaction towards something with Brazil because the companies that had invested in Brazil had already taken a lot of work to position and did not want more competition. And the companies that had tried to trade with Brazil had said that it had been practically impossible because of the tax issues and also because of the customs system, where there are many non-tariff barriers that make it very difficult for a company to introduce its products through customs. 307 2. Recent studies 203 have shown that, although there have been no significant increases in trade tariffs in Latin America, non-tariff mechanisms such as subsidies to local companies, quotas, among others have been used. Do you think that the type of response that Latin America has had to the new reality of the world economy has been mostly through non-tariff measures (and less transparent from a commercial point of view)? If so, do you think this shows that in the region there are still necessary structural reforms related to trade? ANSWER: In any case, I would say that this issue of non-tariff measures is a subject not typical of Latin America, but of the world in general. When tariffs were liberalized under the World Trade Organization (except for agriculture that has had a much stronger delay), in general, the only thing left for countries to use and to make protectionists are the non-tariff barriers worldwide. Then, there may be zero tariffs, but if there are no significant non-tariff barriers in the countries, the products do not enter. Latin America has not been the exception and needs structural reforms. The difference of Latin America with other blocks, such as the European Union and ASEAN, is that, within these blocks, they have managed to liberalize trade. Latin America unfortunately has not achieved that. Intra-regional trade is very small. ABOUT MEXICO 3. What do you think have been the sectors most affected by the protectionist tendencies of these last decades? ANSWER: I would say that more than sector is the problem of our commercial opening has been that it has not been a democratic opening that unfortunately has been concentrated in five industrial sectors entering certain regions of the country and elevated to a limited number of companies by sectors. The main sectors of Mexico are three: the automotive sector, the electric-electronic (which go together) and the machinery sector. Others have been developed such as aeronautics, an agricultural sector that has products that have been successful, but the three stars are those mentioned. In Mexico, 75-80% of exports are attributable to 500 companies, approximately 9 of the 32 states concentrate 75-80% (of exports), and 5 sectors of 17 industrial sectors concentrate much of the trade. So, this is a problem. Those not benefited by trade have been SMEs that are companies that have not been able to compete worldwide because of their size and have faced more competition in the domestic market. So, more than in sectors, the impact has been on the type of company. And SMEs have been the most affected. 4. As a Mexican expert, how do you think governments in Mexico have responded in the last decade to current protectionist tendencies? Have there been pressure from local industrialists for the government to protect the national economy? And if so, were these measures tariff or mostly non- tariff (for example, subsidies and credits to local companies)? ANSWER: The Mexican business sector has been organized since the North American free trade agreement are sectors that participate in the negotiations in the room next door. The agricultural sector has always had a very strong weight in the negotiations. And in general, the strongest sectors have been very well organized: the steel, automotive, pharmaceutical sector. Above all, the goods products sectors. The service sectors have not been as general presentations. Depending on how much power the executive branch has or if it has the majority in the Senate, who approves the treaties, there is more room to negotiate or not. When you don't have that majority (as I was the secretary), the strength of the business sector is much stronger because they knew they could stop the treaties in the Senate. The president of the PRI (Peña Nieto), had control over the Senate and so he could do in a very ambitious trade policy because he had that majority in the Senate. Although, in general, Mexico has followed an opening policy and although there was a transition to a leftist government and very different from what Mexico had had in the past, in the commercial policy part with López Obrador currently, it has resolved to continue the policy of trade opening and respect for free trade agreements. In short, when the government does 203 Evenett, S. J. (2019). Protectionism, state discrimination, and international business since the onset of the Global Financial Crisis. Journal of International Business Policy, 2 (1), 9-36. 308 not have a majority, the lobbies are much stronger because they know they can stop the agreements. In Mexico, the private sector has had a strong presence to set its conditions and, as a result, the government negotiates. ABOUT THE GLOBAL CONTEXT 5. On the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressures from influence groups that, following a global protectionist trend, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: For me it has been very surprising that, having created these sophisticated global value chains, added to the commercial opening and technological revolution that we had with the internet in the nineties where companies moved to produce where it was cheaper because there was better communication and control over each stage of the production process, it seemed that this whole process was going to be almost irreversible. Even consumer products had been cheaper, trade grew above world GDP growth. It seemed that everything was going towards a path of greater openness and that this trend towards the first bilateral integration, then regional and then already more broadly global through mega regional agreements, would also promote multilateral liberalization that had not been possible give. Everything was going in the same direction. I believe that what was not well calculated is that trade policy should have been accompanied by internal policies that addressed certain structural problems that were specific to each country and that were going to be the positive multiplier effects of trade were felt more. I think that the impulse towards the outside was very strong but unfortunately it was not accompanied by other types of policies that also explain why so much inequality has been generated. So, in a context where this commercial opening and neoliberal policies did not seem to be the model to follow, there were no policies that addressed productive development to achieve more education, innovation, greater competition in the markets to have more accessible prices for the entire population. A series of voices that many of our countries have. I explain much of the questioning of globalization and trade openness to the emergence of these populist governments where it is much easier to attribute the causes of inequality and economic crises to trade and other countries, than to attribute them to a number of multidimensional factors They make it very difficult to win an election. It seems easier to say, as Trump has done, that “our big problems are Mexico and Canada and the disastrous free trade agreement we had or China” than to say, “here we needed to practice a series of measures that we don't take care of.” So, I think this explains Brexit, Trump, and many countries that, if they were already protectionists, they will be even more so. Now they will say, “I continue with my protectionist policies” because the opposite has explained why other countries are so bad. I believe that there have been more explanations for why inequality. There was also a serious analysis of how the opening could have been done better. There are certain things that we have tried for free trade that could have been better taken care of. But that alone would not have been enough. Much has to do with internal policies, having taken a little more care of the transformation of the productive sectors towards the sectors where there was more competitiveness and comparative advantage. Another issue that also reached us is the Technological Revolution 4.0 which has meant many layoffs or a transformation in the production plants. So, I think that many actors got together many factors that explain why there is so much bad being in the world. But the issue of GVCs must be seen which countries you speak of. For example, in Latin America the big problem is that it depends a lot on commodities and not on products with more added value. Then each reality depends on the explanation of why they have not also fared in a globalized world. That is why it is very complex to generalize. There are many factors. For example, the issue that China was not adequately addressed at the multilateral level and was not organized for the world to see what they were integrating. While there are things that could have been done in the international sphere, things should also have been done in the internal sphere of each country. And all that is now clearly bursting what is sad in Latin America, is that the model country for all was Chile that had achieved good levels of growth, today is with so many problems. And the great fear that I have is to see that they attribute the problem of Chile to its commercial opening when I think it is the other way around: if Chile had not gone out into the world to sell as such a small country and a small economy in terms of consumption, they 309 would have grown and achieved the levels of development he has achieved. Free trade agreements and trade policy is one of many policies that must be put into play in order for a country to achieve economic growth. You cannot expect that nothing more than openness alone will take you to levels of productive development. It would not be adequate nor if you achieved economic foundation and many benefits and positive things, you cannot say that it is seen only because it is a separate country opening. That is the debate that has always been and right now because I believe that we must have the humility of all the analysts of trade and the economic real to be able to attribute to each factor its specific non-real weight in order to also define the new public policies require because the world in search of a new model. Interview A 5.12 – Full Interview with an anonymous industrial union leader in Mexico (Interview ID: 204) Interview ID: 204 Name: Not for attribution BIO: Anonymous industrial union leader in Mexico. 1. In order to understand how it looks to Mexico in very general terms, as well as from a political point of view, the implications of the trends in protectionist tendencies over the last decade have impacted Mexico significantly, an economy that turns out to be one of the most open in America Latina. What does the new norm of offensive protectionism look like from Mexico? ANSWER: First, it is good to highlight that Mexico is surely the most open economy in Latin America and the most integrated into the processes of international trade. Take into account that Mexico represents at least 70% of all exports of manufactured goods from Latin America. It is an economy that is not based on raw materials, but on manufacturing production and export. From its profile, Mexico has been one of the most proactive countries in the WTO, and in the G20, especially since the 2008 crisis in supporting political decisions aimed at eliminating protectionist measures. Its political position has been very aligned in the need to counter non-tariff measures. In the report that the G20 makes each year, on what new protectionist measures have been adopted, Mexico has a very consistent position against protectionism. At the multilateral level in the WTO and at national and bilateral level, especially with the United States. 2. However, Mexico has been affected by some international protectionist measures, as has happened to other countries, such as Brazil, for example. In these scenarios, how has Mexico's response been? How protectionism has been affected to your industry? ANSWER: We live in a difficult world. If everyone is becoming more protectionist, then we have to protect ourselves too. We know, though, that we have to be careful with our close trading partners, as not to escalate the problem. That’s why our union asked the government to work “in silence” and implement policies that could help us become more competitive without being as obvious to the rest of the world. We are not trying to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context. However, I have to say that Mexico has been one of the countries that has most resorted to dispute resolution mechanisms under NAFTA (or now T-MEC) and those of the WTO. The most emblematic cases were the tuna labeling that the U.S. implemented several years ago and in which the WTO finally proved it right to Mexico. The second emblematic case was the breach of the only substantive provision that had not been implemented in NAFTA, which was the one that freed all cross-border freight transport between the three countries. From the unions’ pressure on all of the transport in the United States, it was that Mexico did not have the freedom of transit in cross-border transport. Then, a pilot program was developed by the United States Department of Transportation, which is then died in the House of Representatives if I remember correctly. In that case, Mexico exhausted the multilateral consultation instances and once it had the green light of the agencies, a series of reprisals equivalent to the economic damage represented by the measure implemented against Mexico. Mexico had the possibility to implement approximately 2 billion dollars in reprisals to the United States. Then an inquisition began district by congressional district to identify which legislators had voted against the program, as well as its main export product to Mexico; and Mexico retaliated accordingly. I 310 don't remember the specific cases, but it was, for example, in Idaho it was toothpaste, apples from Washington State. And from those targeted shots, we managed to reverse the measure. Another recurring theme has been about tomato exports to the U.S. Mexico is a tremendously complementary agrifood market because we complement the station in the U.S. and Canada, where Mexican exports can fill gaps in the national market. Then we have a success story of NAFTA. The political lobby in Florida, which is a major tomato product, has repeatedly sought to accuse tomato exporters of unfair competition or dumping. This has been another case in which we have suffered from protectionism and in which Mexico has been very blunt with unequal successes, because it is an issue that is undefined, so that the provisions of the free trade agreement are enforced. 3. Although for these cases they have decided to go to the route of the WTO or NAFTA, there have also been cases in the world of automatic measures, non-tariff measures that affect Mexican producers and in which Mexico could have been forced to act faster to protect their local producers. Has Mexico been faced with cases such as these in which it had to act more appropriately and outside the parameters the WTO or NAFTA? ANSWER: Rarely has Mexico resorted to such measures. According to my point of view, Mexico has resorted to such measures due to the sheer density of our network of trade agreements. It is a network of commercial forums, that I think is the largest in the world, with more than 50 countries. Yes, we have sometimes resorted to emerging national protection measures for vulnerable sectors such as textiles and footwear, for example, when China entered the WTO, and with Vietnam. but now Vietnam is in the transpacific trade agreement. Then a series of transitions and safeguards are negotiated that have prevented us from resorting to unilateral measures. We have not done much to my knowledge, but I do not know the subject matter. If we are concerned about the application of the measures, we fight against them and we strongly state our position. Japan is an emblematic case, since it is a very difficult market to access. But we already have a trade agreement with them for 15 years. It was only the second that Japan negotiated with another country. 311 Interview A5.13– Full Interview with Daniel Vazquez (Interview ID: 205) Interview ID: 205 Name: Daniel Vazquez BIO: Mexican trade policy expert. He has advised companies and countries on various issues, including market access, investment, and technical barriers to trade. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: No. I think those two factors are tangential. Countries respond to perceived threats. For example, China’s abuse of the current trading system, taking predatory stances in certain sectors, is what generates the protectionist reaction. That position of China is exacerbated by the facilities created by the PTAs and the GVCs. 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: Governments will act using the most effective tools that cause the least political pain. At the USMCA, one of the big challenges was how to leave the reduction schedules almost intact, because once they are opened, a cascade effect can be created that leads to the total collapse of the scheme. Now, there will be cases in which it would be easier to do this by means of tariffs—like China vs. U.S.—if the political cost, although initially large, has much greater political returns ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader Latin America, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: That they are not clear about what they want… 4. Finally, a question about the political economy of trade policy (responses): As a trade expert of Latin America, what you can say about how interest groups reacted to these current protectionism trends (you can concentrate in the case of your country)? Did they pressure for retaliation or domestic measures (e.g. tax reduction) to compensate for the losses of protectionism measures implemented from other countries? 312 ANSWER: Yes, I know. Influential sectors have lobbied their governments, not only to respond but also to protect and safeguard their interests. Depending on the government and the technical capacity of its officials, these supports have been aligned with international commitments or have taken unorthodox forms, such as making the entry of merchandise more difficult by modifying customs requirements. For example, in the case of sugar in Mexico, the sector asked the government to protect access to the American market. At the same time, products competing with sugar—such as fructose—were proposed to be subject to retaliatory measures. When conflicts escalated to USMCA negotiation, the same sectors demanded that access to the American market be protected, while protecting the local market. 313 Interview A5.14 – Full Interview with Luis Miguel Etchevehere (Interview ID: 301) Interview ID: 301 Name: Luis Miguel Etchevehere BIO: Minister of Agriculture, Livestock and Fisheries of Argentina (2019), Minister of Agribusiness of Argentina (2017-2018) and President of the Rural Association of Argentina from 2012 to 2017. A LATIN AMERICAN PERSPECTIVE 1. Ten years have passed since the global financial crisis (CFG) of 2008-2009. This has generated new dynamics in the global economy, including what’s been called “the collapse of global trade” (Baldwin 2009). In recent years both developed and developing countries have increased some barriers to international trade. How do you think Latin America in general has responded to these new protectionist tendencies that have occurred in the last decade? Do you specifically believe that Agriculture, Livestock and Fisheries in the region have been affected by the new protectionist currents? ANSWER: Contrary to what is observed in some countries of the world, with protectionist advances, our government has decided to integrate itself into globalization, multilateralism, and commerce. We believe in the trade agreement, the opposite of protectionism. In these four years we have opened more than 250 markets for products from agriculture, livestock, and fisheries. With respect to Latin America, this continent does not represent a homogeneous analytical unit. There are countries that are very integrated into the global commercial circuit, such as Chile, and others that have closed completely, such as Venezuela. The four countries of the Mercosur bloc signed a very important free trade agreement, especially in agro-industrial matters, with the European Union, at the beginning of the year. In that sense, we consider that these new protectionist tendencies may be a reality for some developed countries, but it is not a reality that we have lived in our region in the last four years. 2. Given this new international context, it has coordinated with Mercosur countries to implement support policies for Agriculture, Livestock and Fisheries in the region (for example, soft credit support program, etc.) ANSWER: In Mercosur and the CAS (Agricultural Council of the South) we have coordinated efforts towards greater health and commercial integration. We believe that the greatest incentive that producers need is to open markets and opportunities. For that, the health we can guarantee in our food is key. Our health systems have worked to standardize standards, assessments, and controls at the borders and within our countries. On the other hand, Argentina has made huge investments in infrastructure and capital to support this activity. The improvements in competitiveness due to the investment in routes, highways, trains, ports and airports, with the most ambitious infrastructure program of the last 50 years, add 70 million dollars specifically destined to small family farmers throughout the country. ABOUT THE ARGENTINE PERSPECTIVE 3. Recent data on the protectionist currents of recent years has shown that, in Argentina, sectors such as cereals and fruits have been affected by the new protectionist measures that various countries in the world have implemented. In these cases where there are direct effects on certain sectors, have pressures arisen from sector organizations for the Argentine government to implement response measures or at least sectoral internal support measures to address international uncertainty? Specifically, what types of measures have been used to respond to the new protectionist currents that often directly or indirectly affect productive sectors of Argentina? (For example, raise trade tariffs so that local businesses have less competition, or internal soft subsidies or credits for businesses) 314 ANSWER: While it is true that protectionist measures are observed in the world, we cannot say that these have been applied to Argentina with as much damage as with other economies. The big problem of the trade war between China and the United States for Argentina was the drop in international food prices, especially oilseeds and their derivatives, which represent the main product exported by our country (more than 60 million dollars in 2018). However, thanks to the effort of our government to integrate intelligently into the world, since countries where some protectionist measures such as China, the United States, and the European Union have increased their commercial openness towards us, we see very positive indicators that reflect a greater integration of our country with the world. To mention just a few examples of concrete measures, since 2018 we have opened the market for cherries, honey, bovine genetics, soybean meal, blueberries, beef, sheep, goats and pigs, and mandarins to China. In addition, lemons and meats (a market that had closed to us 17 years ago) to the United States. We’ve opened the pecan, fractional honey, and shrimp markets to Brazil. The export of lemons and seeds from China to India was enabled. And we signed the free trade agreement with the European Union that had 20 years of negotiations, lowering almost all entry tariffs for agro-industrial products of Mercosur to 0%. All this allowed us to increase exports of many products of agriculture, livestock, and fisheries in the last year. Comparing the first nine months of 2019 against those of 2018, we observe that exports of orange juice (285%), strawberries (230%), cane sugar (156%), cherries (78%), butter (67%) and potato (65%) all increased. Finally, we can say that in the last 4 years, beef exports quadrupled, and pig meat exports increased almost eight times. We believe that trade is one of the keys to the development of economies. And we always try to increase our business relationships with the world. That is why we affirm that the protectionist tendency observed in some parts of the world has harmed us by the generation of uncertainty and the drop in international prices of the goods we export, but not by the increase in tariffs or tariff barriers. In that sense, we note that thanks to the trust generated by our government, the trend has been rather the opposite. 4. In recent years, Argentina has begun to access some global value chains. A differential aspect of this last decade, when compared with other events of the past in which there have been commercial tensions, is that today the world is more interconnected through networks of commercial agreements and global value chains. Do you think that these two factors have impacted the way in which countries (and Argentina specifically) have responded to current protectionist trends? What have you noticed? ANSWER: As I said before, contrary to the protectionist trend, our country experienced a process of great commercial opening in these last four years. And when we start it, we find that there are many tools to help this way. International agreements such as EU-Mercosur, multilateral agencies such as the WTO, and global forums such as the G20, are some examples. They are platforms from which a country can generate many instances of negotiation and interaction with others, to promote dialogue and trade. Undoubtedly, these institutional tools make the effects of a protectionist trend soften, and for our country they are essential. 5. It seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from the influence groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: We believe in commerce as a fundamental tool for the development of the economy. And in particular, for food production, we see trade and market access as essential mechanisms to guarantee food security, given the deep interrelation between countries, when producing them. Argentina is an actor of great relevance in global food value chains, because it has unique characteristics such as climate, soils and the talent of producers. In that, our country has a very powerful comparative advantage that makes it a strategic ally for any world power. The world demands food in quantity, quality and safety, and not everyone can provide it. That is why, beyond what other countries do, our position is to allow trade to flow, and lower tariff and para- tariff barriers as much as possible. 315 Interview A5.15 – Full Interview with Miguel Braun (Interview ID: 302) Interview ID: 302 Name: Miguel Braun BIO: Argentinian Trade Policy Expert. Secretary of Commerce of Argentina (2015-2018). 1. Data on the new protectionist currents show that, in 2018, approximately 40% of Argentine exports were exposed to some type of protectionist measure implemented by countries to which Argentina exports. Among the most affected sectors are the automotive sector and iron and steel sectors. From your experience, how do you think Argentina has responded to all these protectionist currents that originated after the global financial crisis of 2008? While it is clear that the outgoing government has been a defender of free trade, in the face of such international contexts, it is expected that there will be pressures to implement response or domestic measures (subsidies, credit support, etc.) that help to the affected sectors to face an uncertain international scenario. ANSWER: Argentina was already in 2008 one of the most closed economies in the world, both by level of tariffs and by volume of trade. This closure was deepened in 2011-2015 with exchange controls, para-tariff barriers and the DJAI system (prior import control, which was punished in a WTO panel initiated by the United States, the EU and Japan). The result was a 30% drop in exports between 2011 and 2015, and the loss of about 5,000 of the 15,000 exporting companies. This greater protectionism was not a response to the 2008 crisis, but a strategy to sustain an artificially low (appreciated) exchange rate that led to the loss of reserves. In this ultra-closed context, any request for protection from the private sector was generally met, generating higher costs for consumers (prices up to three times higher than in the U.S. for toys, clothing, electronics, and appliances) and less competitiveness for industries dependent on imported inputs. The government that assumed in 2015 sought to implement a gradual opening of the economy, based on free trade agreements, initially with partners that did not represent a risk to Argentine production, and ensuring long terms of tariff reduction (EU-Mercosur, deepening of Mercosur and regional agreements such as Colombia, Chile, and Mexico). At the same time, it normalized foreign trade according to parameters acceptable by the WTO, replacing the DJAI with a system of automatic and non-automatic import licenses (the latter for sensitive products such as textiles, toys, etc.) and sought to simplify bureaucratic procedures and Customs to reduce export costs. In this context, there were cases where the protectionism of other countries affected Argentine production. In particular, the case of dumping and subsidies in the United States against Argentine biodiesel, and the imposition of tariffs on steel and aluminum by the United States. In the second case, a direct and specific negotiation was initiated, coordinated with the affected local private sector, and Argentina was exempted from tariffs. In the case of biodiesel, it was also sought to negotiate, and all legal and institutional tools were applied to fight the case in the U.S., and it was also used to negotiate concessions in other markets, especially opening agricultural markets. There were no subsidies or other domestic support for these industries, beyond what exists. In the case of steel and aluminium because a successful negotiation was achieved, and in the case of biodiesel because exporting companies could easily export soybean oil without suffering significant economic damage. 1.1. But no domestic aid measures were applied to local industry that could be seen as protectionists during the Macri government? Because many times measures are applied that, while not perceived as commercial, end up having a significant commercial impact in foreign countries. ANSWER: There was rather an approach to improve the lowering of export costs and lowering the autocracy to open markets and so on, but the truth is that I do not remember a case where we clearly needed a conversation about import competition. Surely there was something, but a very concrete case does not come to mind. 316 2. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: Argentina undoubtedly suffered in 2015 a very disadvantageous situation for its exports, but more for domestic policies than for external protectionism. High tax burden, artificially low exchange rate, logistic costs higher than our competitors (especially due to union pressure), high financial costs, and excessive bureaucratic burden implied high costs to export. The government worked to lower these costs, and in fact it was possible to reverse the 2011-2015 export decline and have export growth between 2015-2019. We worked with each export sector in sector tables seeking to unlock problems that generated costs for exporters. That is, it was explicitly sought, both in macro and micro policies, to enhance export competitiveness. Of course, much remains to be done, but I share the spirit of the phrase of the trade unionist quoted in the question. 2.1. In what sense do you share this quote? ANSWER: I share it in the sense that it is a claim ... It is not a phrase one would hear from an Argentine unionist because protection is much more explicit. And because Argentina starts from a much more closed base than a country like Mexico, for example, which has many more commercial agreements, lower average tariffs and obviously in NAFTA, so today it is much more integrated with the U.S. In Argentina it would be weird to hear such a phrase. In general, the request for protection is much more explicit and at the same time it is less necessary because it is already very protected. But this phrase in Argentina is not entirely applicable. 3. What are the mechanisms and spaces in Argentina in which the sectors affected by protectionism coordinate with the government for the design of trade policy that helps them cope with certain adverse contexts? (For example, does the dialogue take place through business chambers, unions?). Could you tell us a specific example in which you have been? ANSWER: The dialogue took place in 2015-2019 1) through chambers, in meetings convened by the Ministry of Commerce on a regular basis and in meetings requested by the chambers, 2) in meetings with companies, requested by companies to address specific issues, 3) in meetings with trade unions, cited by the Ministries of Production and Labor (and then unified), 4) at the sectoral tables coordinated by the Ministry of Production, and to which the President attended in the final instances, 5) in meetings of the Production Council coordinated by the Ministry of Production, 6) in formal meetings of trade policy processes (e.g., dumping cases). A concrete example: several meetings with steel and aluminum export companies when the problem of tariffs imposed by the U.S. arose. 3.1. What was the specific claim of these companies? Did they encircle the government or vice versa? ANSWER: At that time the request was that we negotiate, that we look for the way for Argentina to be excluded (from the list), that we negotiate at least quotas. A request to minimize the impact for companies because the American market was very important especially in the case of aluminum more so than in the case of steel. 3.2. But was it rather a request with the intention of dialogue? Or was there also a request to do something if that dialogue did not happen and the affectation continued? ANSWER: The truth is that at least with me (I don’t know if in other places they will have raised it differently) but with me it was always a positive claim. As saying: “Help us avoid this problem ... never raised the possibility, for example, of threatening some kind of commercial reprisal.” And the truth is that we, as a government, do not consider it because what we most import from the United States are inputs, machinery, 317 which would impose a punishment. And since Argentina is a small country at the level of global demand, the possibility of generating a real and credible threat to the United States is void. 3.3. And do you think that this same logic (of a positive claim), does it occur with a closer partner with whom there are commercial ropes such as Brazil, when making a decision that is not favorable to Argentina? ANSWER: In those cases, there is a more equal to equal dialogue, more possibility to negotiate between markets. That is, “if you complicate me, then I’m going to do something.” There is more comprehensive negotiation in what would be the Mercosur countries. With more distant partners like the U.S. and China it is more complicated. In the regional logics there is one more logic of retaliation precisely because of what I said before, as the negotiation is more integral and the dialogue is more continuous, there is more than possibility of negotiating between markets. And thus, appear explicit and implicit threats of retaliation. That is why retaliation between partners who have greater integration and more equal-to-equal trade is more normal … I remember many bilateral meetings with Brazil. In these meetings the positive but negative logic is also latent. 3.4 So, the closer commercially the greater the chances of retaliation threats? ANSWER: Yes. Absolutely. 4. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). It seems possible to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from the influence groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. What do you think about these factors? ANSWER: As the evolution of the measurement of non-tariff barriers by the WTO shows, it is correct that there is a growth of barriers to international trade, which generates greater discretion when applying trade policy. However, global trade continued to grow in the last decade, although its growth rate slowed. This implies that it is more important than ever to have good trade agreements in order to continue promoting exports. It is also necessary to strengthen the geopolitical relevance of a country to obtain greater commercial benefits. This was the path that began in 2015-2019 in Argentina. 318 Interview A5.16 – Full Interview with Ricardo Rozemberg (Interview ID: 303) Interview ID: 303 Name: Ricardo Rozemberg BIO: Argentinian economist, specialist in international economic relations at Center iDeAS/UNSAM 1. Question: Data on the new protectionist currents show that, in 2018, approximately 40% of Argentine exports were exposed to some type of protectionist measure implemented by countries to which Argentina exports. Among the most affected sectors are the automotive sector (314 protectionist measures in destinations to which ARG exports) and iron and steel sectors. From your experience, how do you think Argentina has responded to all these protectionist currents that originated after the global financial crisis of 2008? While it is clear that the outgoing government has been a defender of free trade, in the face of such international contexts, it is expected that there will be pressures to implement response or domestic measures (subsidies, credit support, etc.) that help to the affected sectors to face an uncertain international scenario. ANSWER: First, I would be interested in giving a little context to this topic. The literature that is looking at this topic on what we can call the reduction of globalization, tends to say that there are transient and more permanent factors. Trade grows much less than before: before it believed two to three times more than what the product grew and now it grows at most the same. A world in which global value chains have shrunk a bit. Where foreign direct investment has been reduced. It is a different world from the one we knew before the crisis in which globalization and productive and financial integration came with great force. The media and analysts woke up when Trump, Brexit, and the commercial war between China and the U.S. appeared, but this is a process that comes after the crisis. What we would say in economic terms is that there is an elasticity of exports or imports in relation to the product, lower than before. Many times, I hear that the great era of globalization had to do with the insertion of China and Eastern Europe into the global economy, and now that these two are already inserted in the world you cannot expect trade growth as strong as before. Others say that the crisis revealed that there are large trade imbalances in countries like the U.S., which cannot happen again ... It is a phenomenon of transitory factors. And all this also explains the reaction of countries in Latin America and the world and lets you understand what the strategy is ahead. In Argentina and Latin America in general, everyone says, we have to export more, but they remember to export more just when the world is in a very different scenario than before. The commercial war comes today from a scenario that had already changed. The problem in Latin America is different because the inconvenience has always been the exchange rate appreciation. As a friend in Argentina says, “the exchange rate is late or late.” The process of increasing the prices of commodities and capital inflows has generated pressure on the important exchange rate that they have not done otherwise in the last 10 years than to delay the exchange rate. And those who look at trade policy do not always look at the exchange rate. And it seems to me that in Latin America and in Argentina in particular you have to look at both. Because when you have such volatility, trade policy generally or often compensates or tends to offset macro policies. This is different from what happens in developed countries where volatility is much lower. But in Latin America we must see exchange rate policy with the head of political economy. Because when economic actors ask for obstacles, protection and subsidies, what they are really saying is “I can't ask you for changes to macro policy” (because you can't ask for a better exchange rate), instead go up a fee or give me a subsidy if you can do it. It is easier to ask for that. That is why we tend to demonize trade instead of demonizing the true factors of imbalances that are macro factors. The reasoning of the domestic actor is “why I am going to ask that what I need is stability and a good high exchange rate” when I know that it is an obvious thing. The policymaker is going to tell you, “Yes, I want that too, but I don't get it for different reasons.” So, the domestic actor ends up asking for things related to the commercial. So, against that, added that they are economies that have lived for many decades in international dependence and that have companies and entrepreneurs who think of the world more as threatening than of opportunities, there is a group of sectors that have a very protectionist look. And the protectionist look is tempting for governments because something can always be done to make them happy. 319 About Argentina, I believe that what the Kirchnernist government brought was that when it became difficult the appreciation would change there the protection measures began. But they have had to do with the vision that the world became more difficult, and the exchange rate appreciation began to make a strong effect. The sectors where Argentina has problems are the same sectors where everyone has a problem: clothing (textiles), footwear, which are more automated sectors, that generate employment, and today are most exposed to competition internationally. The largest sectors, automotives, iron and steel, are the ones those more problems abroad, where Argentina exports. But above all the most problematic is that of iron and steel. In this sector there are two important companies: Techin and Acindar. Techin (mostly Argentine capital), which is the largest, does two types of things: flat pipes and tubes that are not traded goods. These companies have private and public reactions (to protectionism). On the private ones, since they are transnational, they invest in other countries to break through the barriers. For example, they put a plant in the U.S., in Asia. On the other hand, in the domestic market they ask to be protected against Chinese and Russian competition, and against global overproduction and low prices. In some cases, your requests are that you go out to respond. Worse, the first thing they ask for is “give me protection.” So, if you look at the products that have had high tariffs or anti-swing in Argentina, you will see that the steel sector is at the forefront. This is bad because that protection generates cost overruns that complicate other sectors that use the company as input. And the automotive sector is another very different case because it is a well-regulated sector in which Argentina basically sells to Brazil, beyond selling a little in Mexico and Europe. But it is a sector that mainly exports to the region where trade is highly regulated. For example, there is an agreement between Argentina and Brazil for the automotive sector so that trade is not too unbalanced. In the automotive sector there are two major actors, the automakers and the auto parts. The automakers are on both sides (Brazil and Argentina). Then it is more an arbitration: if Brazil gives an incentive to domestic consumption or production, there may be a problem. But the important thing to see is that automakers ask you to have here the same things they have in Brazil. Companies want to produce on both sides because they have capital invested in both countries, but they are not the most worried about this. But 80% of automotive trade is in Brazil, the rest must be Mexico and some utility that is sold to European countries. But the auto parts sector has shrunk a lot. There are some success stories such as a company makes gearboxes in Cordoba and sells to Germany. Then there is a valve manufacturer that supplies the European automotive industry that even sells it to Ferrari. But in general, the auto parts industry has shrunk a lot due to competition from Brazil and China. 2. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: In Argentina these discussions are less subtle, and the logic is “in a protectionist world, we have to be protectionist.” While in Mexico, for example, they tell you “although NAFTA was not all that good, we increased exports and jobs” And while the previous government (of Macri) did not share this protectionist vision, the government now [of Alberto Fernandez] agrees more in this regard. 3. What are the mechanisms and spaces in Argentina in which the sectors affected by protectionism coordinate with the government for the design of trade policy that helps them cope with certain adverse contexts? (For example, does the dialogue take place through business chambers, unions?). Could you tell us a specific example in which you have been? 320 ANSWER: In Argentina, the private-public thing is mentioned a lot because it is good marketing, but in reality, there is little interaction. The first thing to know is that, in Argentina, the legislative branch has very little influence on trade policy because it is delegated to the executive branch. Tariffs can be raised without a law. So, the dialogue is basically between the private sector and the executive. And the type of dialogue will depend on the times. The previous government (of Macri) listened little to businessmen in this regard and the current government listens to them too much. Argentina is like that, bipolar. In the Macri government, those who made commercial and productive policy believed that Argentina (and I agree with that) had to go towards a more open economy. I also add a macro element and it is the exchange rate. It is not the same to have a low tariff with a high exchange rate, then a high tariff with a low exchange rate. Just look at the cases of Germany or China to see how a foreign exchange policy goes hand in hand with trade policy. So, it seems to me that the Macri government had a naive version of all this that wanted the opening to anything, regardless of the macro. And this new government, at least the signs it has given so far, is that it matters more to preserve the sources of employment, or to help lobbies. Note that the number one measure on foreign trade was to put non-automatic licenses to a significant number of products. A vision on this measure is that it was taken because exports fall 25% a year, the level of internal recession that exists is high, therefore, the chances of imports growing are low. So, even so, they put a lot of non-automatic licenses. The truth is that with this government the lobbies have a more important role and politics wants to give a signal that the government is no longer the same as before. Although later with time the looks are balanced. For example, the previous government began with a very opening speech and after that it was realizing that in practice it was not possible since private parties play an important role as well. What does happen in Argentina, unlike Brazil, is that the winners of the economic opening play a very silent role. For example, analyzing the case of Mercosur, in 2011, Argentina put non-automatic licenses in response to the crisis, and Brazil also put them. At that time the logic was “if the world is complicated, we must protect too.” But the important point to highlight is that, in the face of the same phenomenon, in Brazil, the National Confederation of Industrialists of São Paulo, took a proposal the next day to the government saying, “look we are very grateful that they want to take care of us, with what we have to manage the trade in this difficult world, but we are very much inserted in the value chains, we have imported inputs everywhere, so please ask us to take out the licenses.” This is very different from what happened historically in Brazil. In Argentina that did not happen, nobody said “take that out.” Neither those who favor nor those affected. Because there are many sectors that these measures make inputs, intermediate goods, more expensive. However, those sectors do not go out so much to demand openness. It is something I never quite understand why it happens from a political economy point of view. That is, the openings do not have much voice. The protectionists do have a voice and publicly ask and thank. In short, the openings are quieter. In part, because, for example, closed and open sectors coexist in the Argentine Industrial Union (UIA), so it is best not to say too much. 321 Interview A5.17 – Full Interview with Not for attribution (Interview ID: 304) Interview ID: 304 Name: Not for attribution BIO: Anonymous trade policy maker of Argentina. 1. Question: Data on the new protectionist episodes how that, in 2018, approximately 40% of Argentine exports were exposed to some type of protectionist measure implemented by countries to which Argentina exports. Among the most affected sectors are the automotive sector and iron and steel sectors. From your experience, how do you think Argentina has responded these protectionist trends that have been originated directly after the global financial crisis of 2008? While it is clear that the outgoing government has been a defender of free trade, in the face of such international contexts, it is expected that there will be pressures to implement some sort of response or domestic measures (subsidies, credit support, etc.) that will provide aid to the affected sectors who face uncertainty in the face in the face of an uncertain international climate. ANSWER: My answer will focus essentially on two specific situations. On the one hand, the tariffs applied by the United States to the import of Argentine biodiesel after an investigation was conducted to find evidence of dumping, the use of subsidies, and the application of tariffs by the United States on the import of steel and aluminum, from which Argentina obtained an exemption based on accepting a maximum export quota for those products. In both cases, industry pressured the government to support both sectors, both through the use of bilateral responses to the United States and through domestic measures. However, for different reasons, the government was reluctant to adopt and support concrete measures and limited itself to its efforts to diplomatic dialogue with the United States government. In the case of [the investigation regarded] biodiesel, a particular phenomenon occurred because the situation that gave rise to the research was the differential in the rate of export duties charged to soybeans and soybean oil compared to biodiesel - with the understanding of the United States government that this difference constituted a subsidy to the biodiesel input and also required rebuilding the price since the market price could not be taken into account. This situation posed a particular challenge for the government and for the industry. For the government, export rights to soybeans and biodiesel are fundamental sources of public income, which made their reduction impossible– the only options were to either eliminate the differential or to offer some compensation to produces. On the other hand, for the producers themselves, the reduction of export duties—thought to compensate for the loss of market access or to improve their competitiveness—would have aggravated the problem since it would have increased the differential that led to the investigation. In addition, government analyzes suggested that in macroeconomic terms the situation was not a serious problem. The additional added value of the export of biodiesel compared to the export of soybean oil was not significant in relation to the loss of revenue resulting from the differential in export duties. In addition, most of the companies involved have participated in both steps of the value chain and therefore could easily convert their biodiesel exports into soybean oil exports. The employment of the sector is not significant and therefore was not a serious challenge either. These factors discouraged the adoption of additional measures to compensate producers. Ultimately, the government opted to increase the aliquot of biodiesel export rights and establish a gradual schedule to reduce the aliquot on the export of soybeans and soybean oil that would have made both aliquots converge—these projects were later abandoned with the deepening of the economic crisis and the increase of export duties. On the other hand, the choice to continue the litigation in the United States—in a coordinated manner with the industry and indicates a change of circumstances. Finally, the reopening of other markets that had also applied anti-dumping and countervailing tariffs that Argentina questioned in the WTO, allowed the industry to reorient its biodiesel exports and alleviate the situation. 322 In the case of steel and aluminum, there were two main factors that led to the government not adopting measures to compensate producers. On the one hand, the immediate achievement of an exception while the negotiations were taking place and the successful conclusion of these allowed companies to maintain an export level similar to what they had and therefore not require compensation. On the other hand, these are two sectors that have historically had protection by trade defense measures and tariffs and that market their products in the domestic market at import prices. The administration of President Macri saw this situation with concern about its impact on the competitiveness of the sectors that use these inputs in production and therefore was reluctant to adopt additional protection measures and rather sought to promote a price reduction of those products in the domestic market. 2. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: I have two observations about this reflection. The most limited is that in the case of the Republic of Argentina, trade union movements have a very scarce—almost nil—relationship with the government areas linked to productive issues and focus their attention on the Ministry of Labor and the President. That attention is even less for international or foreign trade issues, where they not only do not act proactively but also do not respond in detail when they are consulted. More generally, in my experience the business sectors, I have seen that they have made such requests to the government—although perhaps without making the connection so explicit. In the cases mentioned, the modalities of support that were suggested had to do with the elimination of certain measures that affect the competitiveness of exports—in particular, export duties—or with the protection of the internal market in the case of producers of diffused inputs such as steel and aluminum, especially seeking to influence the adoption of trade defense measures both at a general level—the position of the Argentine government on the classification of China as a “market economy” for the purpose of anti-dumping investigations—as a particular—promoting the adoption of compensatory or anti-dumping tariffs. 3. What are the mechanisms and spaces in Argentina in which the sectors affected by protectionism coordinate with the government for the design of trade policy that helps them cope with certain adverse contexts? (For example, does the dialogue take place through business chambers, unions?) Could you tell us a specific example in which you have been? ANSWER: In the two cases on which I can comment concretely, the mechanisms and spaces occurred mostly through direct dialogue between the affected companies and the government—although in the case of biodiesel exporters the representative chamber of the sector also participated. I understand that this is primarily due to the characteristics of the sectors impacted by protectionist measures. In the case of aluminum, there is only one exporting company. In steel, although there is a small group of exporting companies, there is virtually absolute control of the export market at the head of a single company. In the case of biodiesel, although the number of exporting companies is slightly higher, it is a small group and large companies. It is possible that, in the case of other affected sectors, the dynamics were different, and the business chambers had a more important role—extrapolating the role they play in the dialogue with the government on international negotiations. 323 4. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). It seems possible to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from the influence groups that, following global protectionist tendencies, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. What do you think about these factors? ANSWER: In principle, it is a difficult response from the Argentine perspective given the poor integration of its productive activity in the global economy and the practically null network of preferential trade agreements. That said, the alternative described was contemplated and developed by technical officials as a policy tool not in relation to protectionist measures of other countries but as a more sophisticated way of exercising control over imports once the decision to disarm the non-automatic import licensing system that had been implemented by the previous administration and had been subject to questions in the WTO. In fact, technical regulations were used as an alternative mechanism to offer some level of protection to the domestic market against imports that were suspected of potentially unfair competition—especially from China. 324 Interview A5.18 – Full Interview with Not for attribution (Interview ID: 305) Interview ID: 305 Name: Not for attribution BIO: an anonymous Argentine policy maker during the Macri administration. 1. Context: Data on the new protectionist events show that, in 2018, approximately 40% of Argentine exports were exposed to some type of protectionist measure implemented by countries to which Argentina exports. Among the most affected sectors are the automotive sector (314 protectionist measures in destinations to which ARG exports) and iron and steel sectors. From your experience, how do you think Argentina has responded to all these protectionist currents that originated after the global financial crisis of 2008? While it is clear that the outgoing government has been a defender of free trade, in the face of such international contexts, it is expected that there will be pressures to implement response or domestic measures (subsidies, credit support, etc.) that help to the affected sectors to face an uncertain international scenario. ANSWER: In recent years, Argentina has been observing with caution the trade war between the United States and China. For Argentina it is a delicate situation because they are both very important political as well as commercial partners. With the U.S. there is a strong historical relationship, we have achieved a very fluid dialogue beyond the existing agenda with the IMF. We have had permanent contact with the White House; with the USTR and the FDA, among several others. On the other hand, with China we have also maintained a highly fluid dialogue for everything related to the opening of agricultural markets. We managed to open 12 markets in four years, a historical result, which has today allowed us to become involved (the current situation of the coronavirus can have a very negative impact on trade) with our first trading partner (destination of agriculture) to ever displace Brazil. Beyond that, in general terms, Brazil remains our primary trading partner. We have achieved a balance in the face of recent commercial trends, maintaining a permanent link with the two (China and the U.S.) without having to choose one over the other. But clearly this protectionist trends impact us all. Although some celebrated how China raises tariffs for the U.S. and lowers them for the rest of the world seeing it as an opportunity for Argentina, we also knew that they were momentary opportunities creating even more uncertainty. Under the Macri government we support a multilateral integration agenda because we are part of Mercosur, we believe in the block, in productive integration and together we have become the developed world´s food exporter. We have much more to gain as a team, than individually. The U.S. challenges the world in the face of a bilateral relationship strategy. It is reviewing its bilateral agenda with each market. Trump did it with Mexico, then Canada with China. A few days ago, we left the list of beneficiaries of the S.G.P. (General System of Preferences) which fully impacts our country and other developing countries. For example, we had 3,600 (out of a total of 5,500) tariff positions with a 0% benefit with the U.S. under that system. And that we achieved as part of renegotiation efforts in 2017. This has had very negative impacts in our productive sectors, mainly in the agricultural sector. Renegotiation is urgently required. 2. In these contextual situations, many countries apply domestic economic measures, without realizing that they can and often do function as methods of undercover protectionism (non-tariff measures). What is your opinion about this type of reflection? Have you seen yourself in a similar situation given your experience as a policy maker? ANSWER: In Argentina there is still a great fear of openness. In the less informed segments of society, the word import represents a threat to the local industry. Personally, I do not agree with that. I think it is possible to open markets having a constant dialogue with the private sector. Letting them know the economic impact. We are the third most closed country in the world. We adopted a policy of opening dialogue with the world. Our exports increased significantly to Southeast Asia and it is because we are also opening up to a region where its 325 middle class is growing, the one that is willing to pay the export values that our country can offer today. The truth is that when you start this opening process growth is achieved, in fact it can be seen that between 2016- 2019 through the rate at which exports increased, exports of this magnitude have not been seen since 2008. The markets opened to Argentina, but they also need to sell us more. While our imports fell (due to the economic situation, the volatility of the exchange rate, the contraction of Brazil, among others). Our permanent dialogue is with the intermediate entities (chambers). All the chambers have been part of the trade agenda both for the signing of the agreement with the European Union, and with the EFTA countries, both agreements culminated in 2019. We renegotiated the GSP in 2017, which also went hand in hand of all regional economies, provincial governments and business domes. And, in that dialogue, we listen to sensitivities and always try to protect highly sensitive sectors or products where competitiveness is low, leaving them exempt or under a prolonged opening process. For example, the MCS EU agreement has openings of sectors in “baskets” of 15 years. I have not made the decision to ban or stop the importation of a sector or product. The National Foreign Trade Commission (CNCE) was established during the implementation of the management process so that it would be independent and potential damage or dumping claims could be directed there. On the other hand, we systematized the import control, setting aside the discretion highly sought after by the WTO while eliminating the bureaucracy that generated cost and time for firms. We managed to make some export companies more competitive by importing inputs that were not produced here, additionally are engaged and committed to a permanent dialogue with other agencies such as the Industry Secretary or National Trade Secretary. Everyone was heard. For example, sensitive sectors such as the automotive sector were able to take historic steps such as the renegotiation of the ACE 54 economic complementation agreement, achieving a total open in a few years, meanwhile increasing the export quota from Mexico to Argentina, the agreement with Brazil was renegotiated, Paraguay among others. In all cases attending to the sensibilities of our industry and opening other markets where we are competitive, e.g., Colombia with the export of pickups (shipping/distribution services). 3. A defining aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (internationally) have responded to current protectionist trends? What have you noticed? ANSWER: The dialogue was the basis of everything. For example, when the U.S. announced the possible increase in tariffs on steel and aluminum at the end of 2019, we established an immediate dialogue with the authorities through the Minister of Production Dante Sica and the Secretary of Commerce Wilbur Ross. We have taken all the measures that were required in each case to avoid any circumstance that put us at risk. The dialogue is the basis of trust. In the case of Brazil, when a tension arose from the application of non-tariff measures the contact was immediate. Both from the Ministry of Foreign Trade and from the Secretariat of Agroindustrial Markets of which I have been in charge, we have taken measures that stress relations, but on the contrary, we have worked to expand and strengthen them. For four years the dialogue with Brazil was fluid and it was appeasing measures for historical tariffs (especially from a sanitary point of view). It is important to remember that Argentina has historically had very closed organizations (those that have control over the entry or exit of merchandise). It is a work agenda that had to be modernized and systematized. As with other countries one maintains a closer link with (through a trade agreement). Such as that with Brazil, when a tension arises there is no doubt the application of non-tariff measures is among the options considered. But we have not yet implemented any. To me, as secretary of agribusiness in the Ministry of Agriculture and then as Secretary of Foreign Trade, there was the option, but it was never the decision. We immediately implemented the dialogue with Brazil because historically we were two countries where tariff 0 began with a list of non-tariff averages. For four years in dialogue with Brazil, what we did was try of sweeping, all those 326 non-tariff barriers. There is still a great deal between the two countries. Because we are two countries that, we play a competition who is more bureaucratic. What happened was that the retaliation came from the Brazilian side when we realized that in 24 hours olive oil did not enter Brazil, when the grapes stopped entering Brazil, or in the fishing sector that were more ph in the Hake then slowed the containers. We understood this was a message and immediately we contacted our counterpart in Brazil to see what was happening. And on the other side the response was very cordial but many times it came with another request that we had to attend to. It is important to remember that Argentina had historically very closed organisms, mainly the SENASA that is our sanitary organism, that historically put some conditions of access to Brazil. So that had to be cleaned and modernized in these last four years. What happened with the measures that Brazil sent us when something did not access its market was as a claim for us to also lower some other measure that existed before the Macri government. 327 Interview A5.19 – Full Interview with Felix Peña (Interview ID: 306) Interview ID: 306 Name: Felix Peña BIO: Argentine specialist in international trade relations and economic integration. He currently works at the National University of Tres de Febrero (UNTREF) and the Industrial and Commercial Bank of China Foundation (ICBC). 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: Both preferential trade agreements and global value chains have an impact on current trends in protectionism that may eventually emerge in a country that participates in a PTA, or whose companies—and therefore their Comex—are inserted in a GVC. It is difficult, however, to give a precise answer to this question without taking into account what country it may be, the impact that a PTA may have on its strategy of international insertion, and the real impact that the GVC have or may have. for that country, for example in its foreign trade or in its investments of international origin. 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: Difficult to give an opinion when asked too broad a question. What is clear is that today the tendency of many countries (eventually all), is not to make evident their measures that have or may have protectionist objectives, less when it is applied to other countries belonging to the same PTA. Hence, in such cases, it is natural to resort to non-tariff measures, making them as less transparent as possible, and avoid that they may have an impact related to certain countries in particular. In real life, it is often the case that a country's politics may make its protectionist objectives explicit—even for reasons of internal politics. But this is increasingly unobservable. ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: I think it is an argument that corresponds to what is observed in reality in many countries and at different times. This does not imply a value judgment. I do not say that it is right or wrong. I say that is what is frequently observed in many countries and at different times. 328 4. Finally, a question about the political economy of trade policy (responses): As a trade expert of Latin America, what you can say about how interest groups reacted to these current protectionism trends (you can concentrate in the case of your country)? Did they pressure for retaliation or domestic measures (e.g., tax reduction) to compensate for the losses of protectionism measures implemented from other countries? ANSWER: What is relatively normal is that in every country—including mine—it tends to propose measures that take into account the need to counteract negative effects that may result from protectionism that other countries apply and that affect their specific interests. Of course, it is known that such measures of replication of possible protectionisms will depend on their effectiveness on the relative value that one country and another have among themselves. Hence, it is difficult to comment on a subject if it is presented in a very abstract way. In practice, if protectionism comes from a country with greater relative power, the answer may be very careful and not very transparent. 5. What do you think about trade policy dynamics in the post pandemic era? ANSWER: It is very difficult to give a solid opinion on an issue that will depend a lot on the relative power relations between the respective countries in the “post-pandemic” international trading system. The only thing that is clear is that everything will be more complicated if the trends observed in recent years to devalue a rule-oriented international trading system persist. It is yet another reason to worry about the fate of the WTO if its current crisis is not overcome, and to appoint, as Director or Director-General, someone with the capacity to coordinate countries with increasingly different interests and relative power. 329 Interview A5.20 – Full Interview with Renato Flores (Interview ID: 401) Interview ID: 401 Name: Renato Flores BIO: Professor, EPGE, and Special Aide to the President Fondação Getulio Vargas Rio de Janeiro. Professor at the Graduate School of Economics—EPGE/FGV, Rio de Janeiro; President of the Management Committee of Program PEP (Poverty and Economic Policy), based in Canada; President of the Brazilian chapter of the European Community Studies Association; Member of the Advisory Board of the Chair Mercosur, at Sciences Po, Paris, and Member of the Fiscal Board of CEBRI—Brazilian Centre for International Relations. He is a former Member of the Permanent Group of Experts of the Committee on Subsidies and Countervailing Measures, WTO, Geneva, where he also acted as negotiator for the Fourth Protocol of the Agreement on Telecommunications, after the Uruguay Round. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: From one side yes, GVCs in particular create a mesh of connections that may very easily turn a protectionist measure into a shot in one’s own foot. Politicians and decision makers are becoming aware of this. From the other side, protectionism is still broadly strong in agriculture—a sector where GVCs have a smaller presence, and oftentimes left outside, at least partially, in many PTAs—and in specific goods that have an intrinsically strategic value, as found in many telecoms intermediates. Moreover, established constituencies in advanced economies, related to medium to low technology goods, are still very active. ABOUT BRAZIL 2. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: If I understood well the above saying, it refers to the old dilemma of opening up now versus taking more time, during which the sector—thanks to the “polices led in silence” —would be in better conditions to face international competition via trade openness. This is a classic issue, considered in the literature to have been well equated in Japan, South Korea and most Asian tigers, for instance, which set time limits to their domestic manufacturers become able to face foreign competition. In Brazil one finds a mixed picture, with sectors over-protected and without incentives to modernise—many under the umbrella of the powerful São Paulo industries’ federation (FIESP) —as well as other ones with internationally competitive performances, as in the agribusiness, some machinery subsectors, metallurgy & steel, and the flagship example of small and mid- sized aircraft. Liberalisation continues to be a goal but, at present, no consistent and coordinated policy is under discussion. 3. Some of the sectors in Brazil that have been most affected by protectionist measures from the rest of the world were “Products of iron or steel” (592 interventions between 2009-2019) and “Motor 330 vehicles, trailers & semi-trailers; parts” (415 interventions). Many of these interventions from the rest of the world have been implemented before the trade war and there were not necessarily tariff measures, most of them were nontariff mechanisms that are less transparent and complex to identify (murky protectionism): Do you think this protectionist mechanisms are coming more from countries that you have strong trade ties (FTAs, or GVCs) with? If so, why is that the case? ANSWER: To begin with, FTAs, PTAs and mega trade agreements are at a watershed nowadays. This is not only because the incumbent U.S. President has discarded all forms of trade integration and openly resorted to managed trade—even towards allies, but also because the whole design and narrative of trade agreements needs a thorough overhaul, given the new realities and modalities of trade itself—both in goods and services—and the slowly growing fatigue of negotiators with the uncontrolled complexity of themes, areas and non-specifically trade-related requirements of the present negotiations. Add to all this a considerable lack of transparency that has progressively let the people ignorant and unaware of “what has been going on,” and consequently both indifferent and against them, and the present impasse is quite understandable. Within this context, and in a situation of low growth and high unemployment in many traditional Brazilian partners, protectionist measures have been increasing, oftentimes not under WTO principles and rules. The ‘iron and steel’ (or ‘metallurgy and special steels,’ as mentioned above) is a good example, the sector being relatively young in Brazil (relatively to the standard industrial countries) and generally using leaner, cleaner, better and less expensive technologies. Reaction is nearly unavoidable from all those economies—even advanced ones—with more traditional domestic technologies. 4. As an expert in Brazil trade policy, when industrial leaders realized that they are being affected by a specific protectionist measures imposed by a foreign country, do they talk with the government to impose retaliatory measures or domestic measures (e.g., tax reduction) to compensate the losses? In other words, what do you think is the political economy of trade policy retaliation in Brazil? ANSWER: Not much different from that in most Western economies. Menaced sectors lobby for protection or retaliation before the government. A whole governmental apparatus—again not much different from those existing in many countries—exists, for channelling such demands. These, usually end up in a conciliatory agreement or with a standard WTO panel. Other ways may involve specifically designed non-tariff barriers, usually imposed by regulatory or supervisory agencies, that may cause effective damage to those which started the measure. Finally, with political support, the tax galaxy may also be used for certain retaliations. 331 Interview A5.21 – Full Interview with Tatiana L. Palermo (Interview ID: 402) Interview ID: 402 Name: Tatiana L. Palermo BIO: International Trade and Public Policies expert; former Brazilian Vice-Minister of Agriculture and chief agricultural trade negotiator; chair, Palermo Strategic Consulting LLC. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: Trade was an essential driver of the economic recovery of most countries since the 2008 financial crisis. Economies like the European Union, the U.S. (until the current administration), Canada, Australia, New Zealand, China, ASEAN countries and, here in Latin America, the Pacific Alliance countries negotiated many FTAs and PTAs to reduce protectionism, to consolidate GVCs and to speed up the recovery. At the same time, Brazil and Mercosur countries managed to sign only a few PTAs and FTAs, the most significant one, with the EU, was concluded in 2019, but still has not been signed. The average import duty applied by Brazil and Argentina is still among the highest in the world and these countries are not very integrated in the global value chains. So, I would say that the Mercosur countries did not significantly changed their trade policy during the last decade. In 2019, Brazil started to design a different trade policy. President Bolsonaro’s government declared that it wanted to increase Brazilian foreign trade as a share of GDP from the current 23% to 30% in four years. According to the government, this objective would be fulfilled through new trade agreements and through a unilateral reduction of tariffs on inputs and capital goods. After the conclusion of FTAs with the EU and EFTA, Brazil, together with the other Mercosur countries, is currently negotiating FTAs with Canada, Singapore and South Korea. It also plans to start free-trade negotiations with Japan, the U.S., Mexico, and Vietnam. 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: Mercosur is a good example for your argument. Although the countries do not apply tariffs among them (with some exceptions), there are numerous non-tariff and other barriers (like, for example, extremely complex rules of origin, and different sanitary and phytosanitary requirements that make agricultural trade within the bloc very difficult). I agree with your statement that governments are influenced by interest groups that advocate for more protectionism and import substitution. But I would not say that, in the case of existing trade agreements, this happens in a less transparent and perceptible way. Non-tariff barriers in Brazil are usually not directed toward specific countries. At the same time, it is usually more difficult for the Brazilian government to justify such barriers within Mercosur. ABOUT LATIN AMERICA 332 3. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: When we talk about Brazil, we must recognize that Brazil is becoming more and more transparent. There is a pressure from specific interest groups, but the government is trying to balance the interests and to promote a trade openness. As a part of the process of Brazil’s accession to the OECD, the government is undertaking policy reforms to strengthen Brazil’s integration into global trade, including reduction of non-tariff barriers. The current coronavirus crisis is causing more protectionism around the world and will redesign GVCs, especially in critical sectors like medical supplies. The on-going technological decupling between the U.S. and China is also influencing the supply chains. These factors will definitely increase the pressure on governments to adopt more protectionist measures. 4. As a trade expert of Latin America, what you can say about how interest groups reacted to these current protectionism trends (you can concentrate in the case of your country)? Did they pressure for retaliation or domestic measures (e.g., tax reduction) to compensate for the losses of protectionism measures implemented from other countries? ANSWER: Existing tariffs and non-tariff barriers and the lack of trade agreements to reduce such barriers resulted in a high concentration of Brazilian exports in a small group of commodities. Just 10 commodities were responsible for 50% of the total Brazilian exports in 2019: soybeans, crude oil, iron ore, cellulose, beef, chicken meat, soybean meal, coffee beans, raw sugar, and intermediate products from iron or steel. There is a common understanding within the private sector that Brazil needs to diversify its exports and, in order to do this, it needs to negotiate more trade agreements to reduce tariff and non-tariff barriers in export markets. At the same time, during the last decade, there was a significant increase in the use of trade remedies, especially on imports from China (Brazil currently imposes trade remedies on imports of as many as 53 different Chinese products). The Brazilian government is also making important steps to adopt domestic policies to increase Brazil’s competitiveness in foreign markets (the government is currently preparing a proposal of a comprehensive tax reform to be submitted to the Brazilian Congress). In general, Brazil resorts to negotiations with its partner countries to try to reduce their protectionist measures, instead of retaliation. 333 Interview A5.22 – Full Interview with Welber Barral (Interview ID: 403) Interview ID: 403 Name: Welber Barral BIO: Former Secretary of International Trade of Brazil (2007-2011). Partner at BMJ Consultores. A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: The case of Brazil is quite particular. For example, from Brazil, in 2008, when I was in government, we made the G20 speak out against protectionism. There were two statements seeking to curb protectionism. But if we look at all the countries to which Brazil exports, we can see that the country suffers a lot from non-tariff protectionism, such as health barriers, especially meat. It is in this type of measures that Brazil has had the most problems. If you look at the cases that Brazil has brought to the WTO, you can see that many are cases of anti-dumping and some cases of subsidies, as was the case with Canada on the matter of Embraer, or the case against the United States on the subject of cotton. For Brazilian exports, the vast majority of cases are derived from subsidy measures that were used by importing countries of our products. If we take the case of Brazilian exports to Europe, there are many sanitary barriers and also very high tariffs. What I can say that, in the case of Brazil, the 2008 crisis did not mean a greater effect on Brazilian exports, but rather followed the practices that already existed. In terms of the protectionism that Brazil applies, from the Lula government, there was a discourse of commercial opening and reduction of Brazilian protectionism. But in reality, some measures continued to be imposed. For example, in 2012 and 2013 Brazil was one of the countries that applied antidumping the most. And Brazil has also used Mercosur’s list of exceptions to impose higher tariffs on imports. What happens with Brazil is that they have always been very specific topics. That is, a specific production and always in response to the demand of a specific sector of the national economy. In summary, in the case of protectionism against Brazil, our country has always been more affected by non-tariff protectionism. In the case of Brazil’s protectionist responses, they have been of both tariff and non-tariff types. There has been a lot of anti-dumping. When we speak of protectionist measures as responses, it is important to consider three factors in the case of Brazil. The first is to know which country we have in front of us, for example, if the U.S. is taking the measure, Brazil will always respond because it is a very important market. But when there are minor economies, we do not make that response. For example, a few years ago, Ecuador imposed barriers against Brazilian products and Brazil did not bother to respond. It did not make sense. What has happened a lot in 2008 have been measures to implement response policies against Argentina. This was a time when Argentina began to put up many non-tariff barriers. And in that case, since it was an important market for Brazil, we always responded. In many cases not to the same products, but to others. In general, we responded to products that were not so important from an economic point of view but that had a significant political impact. For example, once Argentina put barriers to Brazilian exports of household appliances, or barriers to Brazilian shoes, we also imposed non-automatic licenses on Argentine exports of garlic and wine, which are not such important products from the point of view from a macro point of view, they did have a significant political impact if you looked at the affected provinces. So, to the question of whether PTAs and GVCs are important factors when making a trade policy decision, my answer is yes and no. Take for example the case of Argentina as an example. That country began to implement protectionist measures against Brazil when integration was very strong. In other words, there were good political ties. If Argentina made a decision (which affected Brazil) there were and are rapid coordination mechanisms between the Secretary of Commerce of Brazil and Argentina. And we all sat at the table and sought to discuss each of the measures that affected each country. And when these dialogues were opened, there were complaints from both sides, because Brazil also applied many measures. Sometimes there would be a list of 40 topics to discuss in the meetings, for example, technical work, health, etc. So, there were mechanisms of dialogue in the first instance. In addition to reaching the dispute settlement system in the framework of Mercosur. So, since there was a deep integration, these dialogue mechanisms existed, things that we do not have with commercial partners in Asia or Africa. 334 2. Continuing with the impact of trade agreements and global value chains, it seems logical to imagine that countries could protect themselves less against those partners in a trade agreement or that integrate the same global value chain. At the same time, it could happen that the protection was reduced in tariff terms but that the same did not happen with non-tariff and less transparent barriers (for example, subsidies to local companies). That is to say, it could be considered that governments have access to pressure from interest groups that, following a global protectionist tendency, promote protection through less transparent and perceptible measures by those countries with which there are important commercial ties, for example, through commercial agreements. Faced with this argument and scenario, what is your opinion? ANSWER: The truth is that whenever we had these coordination mechanisms, it was to remove non-tariff measures that existed in both countries (licenses, certifications, etc.). It is true that there were cases that went beyond dialogue. For example, Argentina claimed that its tractor manufacturers were disadvantaged compared to Brazilians by loans from the National Development Bank of Brazil (it is the main public bank in Brazil and today has twice the World Bank budget). These loans were only granted to tractors produced in Brazil. This created a distortion within a market in which Argentina was very competitive. This claim Argentina always put on the table, but it did not have an easy solution, in fact, until today it could not be solved. ABOUT LATIN AMERICA 3. In an interview with an anonymous union industry union leader of Mexico, he told me: “We live in a difficult world. If everyone is getting more protectionist, we have to protect ourselves too. We know we have to be careful with our trading partners, though, as not to escalate the problem. That’s why our unions asked the government to work ‘in silence’ and implement policies that could become us more competitive without being as obvious to the rest of the world. We are not pretending to hurt our partners; we are just asking the government to help us be more competitive in this difficult international context.” What is your opinion about this thought? ANSWER: For example, we had three meetings a year on average with our main trading partners: three with the U.S., three with Europe, two meetings a year with China, three with Uruguay. But with Argentina we had meetings every month because there was always a list of bilateral claims. In fact, it was a very close relationship, but in which there were always commercial obstacles to be solved. And this bilateral dialogue was due in part to the failures of Mercosur. In the 1990s, the National Development Bank of Brazil was a very important actor in granting financing to the national industry, and this Argentina always saw as an obstacle. 335 Interview A5.23 – Full Interview with Tatiana Lacerda (Interview ID: 404) Interview ID: 404 Name: Tatiana Lacerda BIO: Tatiana Prazeres, PhD, is Senior Fellow at the University of International Business and Economics in Beijing, China. Previously, she was Senior Advisor to the Director-General of the WTO (2013-2018) and Secretary of Foreign Trade of Brazil (2011-2013). A GLOBAL PERSPECTIVE 1. A differential aspect of this last decade, with respect to other events of the past in which there have been protectionist escalation, is that today the world is more interconnected through a network of trade agreements (PTAs) and global value chains (GVCs). Do you think that these two factors have impacted the way in which countries (globally) have responded to current protectionist trends? What have you noticed? ANSWER: I would argue that it is natural that countries will be reluctant to impose measures on others to which they are particularly dependent. PTAs and GVCs have in common that they normally create deeper trade relations and because of that I would agree that countries will be more careful with not to damage a trade relation that they also depend on and for potential responses that can be damaging for the country too. I can tell the experience of Brazil. I was Brazil foreign trade secretary from January 2011 to May 2013 and my view is that regional trade agreements in particular have created some constraints on Brazil’s protectionist policies. For example, there was a moment in Brazil that there was an interest of some groups to increase tariffs. And we have some space to do that because the allowed tariff under the WTO is much higher than the applied tariff in Brazil. And the fact that was a Mercosur common external tariff served as a deterrent to those pressures. So, it was because Mercosur would be willing to increase tariff that somehow contain tariff those impositions …. So, for Brazil, RTAs and the legal framework of Mercosur somehow helps to contain protectionist trends. So, I would agree with the point that RTA impact in some way the way in countries implement trade policy measures to both against the agreement members and the rest of the world. Also, in the case of Brazil, I don’t think it is just the regional framework but the multilateral system that helps to contains some protectionist pressures. Given that we know (from the bureaucracy) that know the value of multilateral rules, we were not in full support of going all the way in terms of the protection that the industry was calling for in some moments in times. The WTO legal framework is particularly important and is normally underestimated as a deterrent of protectionism. For example, with the application of antidumping measures. At the time I was the Trade Secretary Brazil was WTO member that opens the higher number of antidumping investigations and also the country with higher number of antidumping measures’ implementation. We knew that we were in the spot so playing under the rules is something that we valued very much. At that time, there was pressure from our industry sector to apply especial safeguards against China on a broad range of textiles and clothing and it didn´t happen and there also a petition to open a general (globally) safeguard on wines and that also didn’t go ahead. So, despite being considered and labeled as a protectionist country, we were very mindful of legal framework, we value the legal framework, and we use it to protect the policies that we thought should be protected. For example, the textile sector was very vocal in making their case for more protection. And we somehow use that argument to explain why it's not something that could be done under the WTO. 336 Table. A5.4 – National Trade Policy Bodies and Other Trade-Related Institutions Institution Web link National Trade Policy Bodies Presidencia de la República https://www.gob.mx/presidencia Secretaría de Agricultura, Ganadería, Desarrollo Rural, Pesca y Alimentación (SAGARPA) http://www.sagarpa.gob.mx/ Secretaría de Comunicaciones y Transportes (SCT) http://www.sct.gob.mx/ Secretaría de Economía (SE) https://www.gob.mx/se/ - Información Estadística y Arancelaria - Instituto Nacional del Emprendedor (INADEM) - Sistema de Información Arancelaria vía Internet - Sistema de Información de Tratados Comerciales - Solución de Controversias - Representaciones Comerciales en el extranjero - Unidad de Prácticas Comerciales Internacionales (UPCI) Secretaría de Energía (SENER) http://www.energia.gob.mx/ Secretaría de Gobernación (SEGOB) http://portales.segob.gob.mx/es/PoliticaMigratoria/home Secretaría de Hacienda y Crédito Público (SHCP) https://www.gob.mx/hacienda Secretaría de Relaciones Exteriores (SRE) https://www.gob.mx/sre Cámara de Diputados http://www.diputados.gob.mx/ Senado de la República https://www.senado.gob.mx/ Other Trade-Related Institutions Banco Nacional de Comercio Exterior, S.N.C. (BANCOMEXT) http://www.bancomext.com/ Comisión Federal de Competencia Económica (COFECE) https://www.cofece.mx/cofece/ Comisión Federal de Mejora Regulatoria (COFEMER) http://www.cofemer.gob.mx/ Compras de Gobierno https://www.comprasdegobierno.gob.mx/ Instituto Mexicano de la Propiedad Industrial (IMPI) https://www.gob.mx/impi/ Sistema Integral de Información de Comercio Exterior (SIICEX) http://www.siicex.gob.mx/portalSiicex/ Sistema Nacional de Información e Integración de Mercados http://www.economia-sniim.gob.mx Trade and Investment Promotion ProMéxico http://www.promexico.gob.mx/ Directorio de Exportadores Mexicanos (DIEX) http://diex.promexico.gob.mx/ Hecho en México B2B http://www.hechoenmexicob2b.com/ Sistema de Información Empresarial Mexicano (SIEM) https://siem.gob.mx/ Sistema Nacional de Información e Integración de Mercados http://www.economia-sniim.gob.mx Ventanilla Única https://www.ventanillaunica.gob.mx Support to SMEs Fondo Nacional Emprendedor (FNE) https://www.sistemaemprendedor.gob.mx/ Fondo de Apoyo para la Micro, Pequeña y Mediana Empresa (FONDO PYME) http://www.fondopyme.gob.mx/ Instituto Nacional del Emprendedor (INADEM) https://www.inadem.gob.mx/ Mujer PYME Internacional https://mujerpyme.com/ Programa Nacional de Financiamiento al Microempresario (PRONAFIM) https://www.gob.mx/pronafim Red de Apoyo al Emprendedor https://www.inadem.gob.mx/a/ Private Sector Associations American Chamber - Mexico https://www.amcham.com.mx/ Canadian Chamber of Commerce in Mexico http://www.canchammx.com/ Consejo Empresarial Mexicano de Comercio Exterior, Inversión y Tecnología (COMCE) http://www.comce.org.mx/ Entidad Mexicana de Acreditación, A.C. http://www.ema.org.mx/ United States-Mexico Chamber of Commerce http://www.usmcoc.org/ Source: OAS, SICE, http://www.sice.oas.org/trl/MEX_e.asp 337 Figure A5.1 – Who’s ready to enter supply chains? A consideration of key variables (score out of 10 in the Business Environment Rankings) Source: Economic Intelligence Unit, Business Environment Rankings. 338 Appendix – Chapter 6 Table. A6.1 – Specific Ideas on How to Elaborate More on this Dissertation by Using Network Analysis Possible Network Dynamics: For example, explore possible “common enemies dynamics,” which happen when two (or more) countries with strong trade ties cooperate against a third economy (e.g., EU+U.S. versus China). We would expect, for instance, that the existence of a PTA between two countries increases the likelihood of these two countries implementing protectionist measures against the same third country. In December 2017, during the WTO Ministerial Conference in Buenos Aires, the U.S. and Japan joined forces to develop a common position vis-á-vis Chinese trade practices. In September 2018, they issued a statement (also signed by the EU), making specific reference to “Concerns with Non-Market-Oriented Policies and Practices of Third Countries” (Evenett & Fritz, 2018, p. 11). Fear of Trade Deflection: U.S., Canada and the EU against China Common Enemy: U.S. and Japan against China Methodology: Such dynamics can be better explained by using Exponential Random Graph Models (ERGMs), which is the lingua franca of social network analysis. The structure of the trade system can be understood as a signed network with positive ties (e.g., PTAs), and negative ties (e.g., tariffs), that can be explained using exponential random graph models. ERGMs are designed specifically for this type of data, i.e., an endogenous model, which cannot be estimated directly with other methods such as GLM, which depends on ID data; or spatial regressions, which in turn depend on having independent spatial weights. Some of the key papers that describe the ERGMs and specific parameters, on which furthers studies on trade network dynamics are Robins et al. (2007); Robins and Morris (2007); and Snijders et al. (2006). Design: One possible way to test the “common enemy” hypothesis, which analyzes the extent to which a positive link between two nodes (e.g., PTA) increases the likelihood that they both share a protectionist tie with a third country, would be studies that consider the dataset presented in this dissertation (all links from 2009- 2018) and consider different estimates of “protectionist edges.” For instance, consider ties as a binary variable: a first perspective of the tie will be based on a binary variable that describes whether or not there is a protectionist measure between A and B. A tie exists if at any time between 2008 and 2018, A implements a protectionist measure against B, which is in essence the network to model. Also, studies can rely on a weighted tie as a dyad attribute, which measures how much these ties matter (what share of exports are affected by the measure). This alternative of a weighted tie as a dyad attribute can be used to test different alternatives for the model to construct a binary variable from this weighted tie. I can show different levels of protectionism between two countries. For each link of the network, the following options based on the weighted tie can be considered: A new dyad attribute can take the form of 1 when the main weighted tie fulfills the following scenarios: Country A sends a tie to B if and only if the size of the policies sent is above the median (above the 25th, 33rd, etc.) of what A sends to B in all the years considered in the analysis (from 2009 to 2017). A sensitivity analysis on these different measures can be conducted. Moreover, it can be worthy to do the analysis in different blocks of time, saying, groups of 3 years (2009-2011, 2012-2014, 2015-2017), for example. Source: Author’s creation.

Abstract (if available)

Abstract This dissertation is neither about free trade nor protectionism. It is about the point where the two may converge. One distinctive aspect of rising protectionism over the last decade is that it is occurring against a backdrop of increased global interconnectivity. Recently, countries around the world have become more integrated through preferential trade agreements (PTAs) and global value chains (GVCs), even though they are erecting trade barriers within these same trade venues. What explains these contradictory trajectories? The recent literature tells us how PTAs and GVCs promote trade interdependences, and some of the earlier work on the political economy of trade protectionism offers compelling arguments about the circumstances under which governments decide to protect their economies. However, we have little evidence on how these two trajectories may simultaneously interact. This dissertation concentrates on how states?especially in the developing world?respond to the protectionist trends that have arisen in context of high interconnectivity and uncertainty. ? This question is vital to debates about international political economy and trade policy from a theoretical and policy perspective. We have now entered an era in which major trade policy tensions are growing faster than the generation of theories to explain them. Adopting a mixed methods approach based on a large-N statistical analysis combined with a detailed country case analysis, this research offers new evidence that institutional mechanisms like PTAs, which for decades have fueled the diffusion of trade liberalization, can also become channels for the spread of less-transparent protectionism (e.g., Non–Tariffs Measures) in an interconnected and uncertain global economy. I also conduct an initial empirical analysis of how GVCs have exacerbated bilateral protectionist dynamics. My time frame is the period from the 2008-09 Global Financial Crisis (GFC) to 2019. I examine: (1) the role of PTAs and Non–Tariffs Measures (NTMs) in the escalation of trade protectionism, (2) how emerging economies (with a focus on Latin America) have been affected by and responded to rising protectionist trends especially in the realm of NTMs, and (3) the specification of how industry/sectoral dynamics have shaped protectionist policies and responses. ? The key contribution of this research is my analysis of how three main variables have interacted with ongoing trade policy dynamics since the GFC. By this, I refer to the political economy “trilemma” between trade protectionism, trade interconnectivity (e.g., PTAs and GVCs), and uncertainty (mostly generated by economic crises). This dissertation is a cautionary tale about trade protectionism in an era of trade interdependence and high levels of uncertainty. This project bridges academic analysis with trade policy-making and offers a roadmap for the kinds of commercial policy reforms that will be essential for the successful revival of world markets post-COVID-19.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The political economy of Brazilian trade policy: domestic determinants, world and regional strategies

PDF

Financial crises and trade policy in developing countries

PDF

Developing countries in the digital era: state and business interactions for industrial upgrading

PDF

Beauty has a price: the global political economy of beauty among youth in Guadalajara, Jalisco, Mexico

PDF

Trade and legalization in East Asia: government-business collaboration in trade dispute settlement

PDF

An empirical characterisation of the Bombay Stock Exchange

PDF

Economic policy-making in a changing environment: the politics of success in Peru vis-à-vis China

PDF

Essays on sovereign debt

PDF

The impact of trade liberlaization on firm performance in developing countries -- new evidence from Pakistani manufacturing sector

PDF

Environmental policy reforms in a global and regional world: mimetic, diffusive, and coercive policy adoptions in France and Korea

PDF

Supply and demand of economic hierarchy: the Northeast Asia case

PDF

Global governance of sport in a digital age: the political economy of sport integrity regulation

PDF

Essays on political economy and corruption

PDF

The role of local actors in international development: how incentive types affect foreign aid effectiveness

PDF

Social policy and presidential ideology in Latin America: the political economy of social spending and anti-poverty programs

PDF

Essays in political economy and mechanism design

PDF

Three essays in international macroeconomics and finance

PDF

Essay on monetary policy, macroprudential policy, and financial integration