An East Asian Model for Latin American Success (The International Political Economy of New Regionalisms Series)

  • 39 79 8
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

An East Asian Model for Latin American Success (The International Political Economy of New Regionalisms Series)

The International Political Economy of An East Asian Model for New Regionalisms Series Latin American Success The New

994 20 16MB

Pages 100 Page size 843.12 x 595.68 pts (A4)

Report DMCA / Copyright


Recommend Papers

File loading please wait...
Citation preview

The International Political Economy of

An East Asian Model for

New Regionalisms Series

Latin American Success The New Path

The International PolWcal Economy o/New Regionalisms Series presents innovative analyses of a range of novel regional relations and institutions. Going beyond established, formal, interstate economic organizations, this essential series provides informed interdisciplinary and international research and debate about myriad heterogeneous intermediate level interactions. Ref ective of its cosmopolitan and creative orientation, this series is developed by an international editorial team of established and emerging scholars in both the South and North. I t reinforces ongoing networks of analysts in both academia and think-tanks as well as international agencies concerned with micro-, meso- and macro-level regionalisms. Editorial Board

ANILHIRA Simon Fraser University, Canada

Timothy M. Shaw, Royal Roads University, Victoria, BC, Canada Isidro Morales, Universidad de las Americas, Puebla, Mexico Maria Nzomo, Embassy of Kenya, Zimbabwe Nicola Phillips, University of Manchester, U K Johan Saravanamuttu, Science University of Malaysia, Malaysia Fredrik SOderbaum, Goteborg Universitet, Sweden Recent titles in the series

Regional Integration and Poverty Edited by Dirk Willem te Velde and the Overseas Development Institute

Redefining the Pacif c? Regionalism Past, Present and Future Edited by Jenny Bryant-Tokalau and Ian Frazer

The Limits of Regionalism NAFTA's Labour Accord Robert G. Finbow

European Union and New Regionalism Regional Actors and Global Governance in a Post-Hegemonic Era Second Edition Edited by Mario TeM



Ani! Hira 2007 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior pennission of the publisher. Anil Hira has asserted his moral right under the Copyright, Designs and Patents Act, 1 988, to be identified as the author of this work. Published by Ashgate Publishing Limited Gower House Croft Road Aldershot Hampshire GU l l 3HR England

Ashgate Publishing Company Suite 420 1 0 1 Cherry Street Burlington, VT 0540 1 -4405 USA

Hira, Ani! An East Asian model for Latin American success: the new path. (the international political economy of new regionalisms series) I. Industrial promotion Latin America 2. Industrial promotion - East Asia 3. Industrial policy Latin America 4. Industrial policy - East Asia 5. Latin America Economic policy 6. East Asia Economic policy


Why an Active Industrial Policy is Needed for Development



Industrial Policy Lessons for Latin America from East Asia



Lessons from the Rise of China: The Continuing Relevance of the East Asian Model for Latin American Industrial Policy


Sectoral and Political Foundations for a New I ndustrial Policy Path for Latin America

1 03

Putting Institutions and Sectoral Policy to Work: Lessons from Success Stories in East Asia and Latin America

13 1



I. Title

338.9'8 Library of Congress Cataloging-in-Publication Data

Hira, Ani\' An East Asian model for Latin American success: the new path / by Ani! Hira. p. cm. -- (International political economy of new regionalisms series) Includes bibliographical references and index. ISBN 978-0-7546-7 1 08-4 I . Industrial promotion--Latin America. 2. Industrial policy--Latin America. 3 . Latin America--Economic policy. 4 . Industrial promotion--East Asia. 5. Industrial policy--East Asia. 6. East Asia--Economic policy. I. Title. HC 1 30.I53H57 2007 338.98--dc22 200701 4527


vi ix xxi x.xiii

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy

British Library Cataloguing in Publication Data

Printed and bound in Great Britain by MPO Books Ltd. Bodmin. Cornwal l

List ofFigures and Tables Preface Acknowledgements List ofAcronyms

Bibliography Index

1 59 181

List Figures and Tables



Key Relationships for a true IP

1 19


3.5 3.6 3.7 3.8 3.9 3.10 3. 1 1 3.12 3.13

U S economic and military assistance and defense expenditures abroad, 1 95 0-73 GDP per capita, totals and growth, 1 9 1 3-98 Employment in LA and EA, % of population, 1 950-98 Percent growth in value of merchandise exports at constant prices 1 870- 1 998 Openness, % of GDP per capita, 1 950-2000 Gross national savings (% of GNI) inflation Debt, 1 990-200 1 FDI in LA vs. EA Countries Overall central government deficit/surplus Tax revenue source as a % o ftotal revenue Government expenditures by function Interest rates and business environment Comparing standards of living and inequality i n E A and L A

4.1 4.2 4.3 4.4

Comparing long-run economic performance: China's amazing spurt Comparison of FDI inf ows for various developing countries Share of state-owned banks, 1 998 Average annual growth in tax revenue in China, 1 994-1 999

88 89 93 96

5.1 5.2 5.3a 5.3b 5 .4 5.5 5.6 5.7 5.8

Structure of output (value added % of GDP, 200 1 ) Comparing manufacturing in LA vs. EA, 1 995 Comparing sectoral production mix in EA and LA by output, 1 995 Comparing sectoral production mix in EA and LA by output, 1 995 Sectoral production mix by number of employees, 1 995 Key exports and i mports, EA vs. LA Innovation in EA vs. LA Rank order of leading US manufacturing sectors, 1 997 The most important frms in the world

1 05 1 06 1 07 1 08 1 09 1 10 1 12 1 16 118

3.1 3 .2 3.3 3.4

60 72 74 74 75 75 76 78 81 81 82 83 84


An East Asian Modelfor Latin American Success

nationalization, or relying solely on the newer proposals of regional autonomy and self-sufficient agriculture. Nor does it require the cross the fingers hope that macroeconomic stability will lead eventually to some benefits for the impoverished majority. I offer a plan for developing both growth and equity simultaneously and expect that the idea itself can help us to overcome political divisions. History shows the power of ideas to move political mountains and to inspire generations to move forward. While my focus is on industrialization policies that are closer to the BRIC (Brazil, Russia, India, and China) model of using large domestic markets to spur competitive nations, I also present a variant on the strategy that is closer to smaller developing economies. In the former case, a direct approach towards fomenting sectors is helpful, while in the latter a flexible niche strategy is more appropriate. We review both strategies in detail in our discussions of China as the l arge market case and Taiwan, Singapore and Costa Rica as the small market cases. We turn then to a serious discussion on how to win the political battle for pushing LA to adopt an EA model. We close with policy recommendations for both the state and the private sector, using success stories from both regions.


Academics, contrary to graduate school perceptions, is strongly geared against innovative thinking. My former professor, David Arase, helped to introduce me to the East Asian l iterature, opening up a door to new ideas for which I am forever grateful. Though I had thought about this project for a long time, and worked through it for a number of years, it was an uphill battle to bring it to fruition. One prospective publisher even told me at one point to give up. But I suppose it is dogged faith in the possibilities of ideas, more than anything else, that allowed me to continue the effort. Several anonymous reviewers, some encouraging, some not, were very helpfu l in pushing the arguments forward through its multiple versions. However, the proj ect would not have come to fruition without the support of the managing editor, Kirstin Howgate at Ashgate, and the series editor, Tim Shaw, who "had a little faith in me." The latter suggested adding the f nal ehapter and gave strong encouragement throughout the intensive editing process. I am very fortunate to have his guidance. The stable and encouraging environment that Simon Fraser University and my colleagues in the Political Science Department create has allowed me to move forward on bold propositional projects such as this one. I would l ike to thank my colleagues who read all or parts of the manuscript, Tsuyoshi Kawasaki, Mike Howlett, Paul Haslam, and Gerardo Otero. Three reviewers deserve special recognition for their painstaking review that allowed small changes to lead to l arge improvements: Ron Hira, Richard Lipsey, and Kathryn Lavelle. My partner, Patricia, was dedicated as al ways in the fumes of the marathon, helping not only with moral support but also in preparing the final drafts.

List of Acronyms


Brazil, Russia, India, and China East Asia Export Oriented I ndustrialization Foreign Direct Investment I nter-American Development Bank industrial policy I mport Substituting Industrialization l S I of capital and economies of scale intensive goods, such as durables Latin America Ministry of I nternational Trade and I ndustry (Japan) Multinational Corporations State Owned Enterprise

Chapter 1

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy This book is dedicated to Patricia, a true angel who taught me everything I need to know. Introduction

In the preface, we discussed the problems with the implicit foundations of the predominant approaches to LA development. Building upon the oldest story in the world, that of conspiratorial fate, the Left focuses on emotion-based critique, pointing out that misery must have some blame. As we have seen, the problem with this approach is not only is it non-productive, but it also glosses over many of the complications inherent in the LA landscape. For example, as I have pointed out elsewhere (Hira 1 998), a class - and/or imperialist-based analysis not only begs the question of the haphazard nature of rebellion (leading to bringing in exogenous variables such as control through ideology), but also misses the multi-faceted aspects of individual and group identity, such as religion, race, gender, ethnicity and geography as well as the wide variety of possible reaction-interactions over time and space. On the Right, the glib faith that the only thing that matters is individual maximization of economic goods also not only misses the rich social context of development, but the fact that other goals such as domestic equity and a sense of place may also affect behavior. This leaves us with the last predominant approach of the literature, namely that of policy analysts. This group, including primarily economists in state institutions, multilateral development banks, and think tanks, is by far the most influential in setting policy, matching resources with supposed dominance by way of "scientif c" expertise and privileged access to policymakers. This group creates barriers from conversing with others through its insistence that mathematically-based economic models are the only sensible way to discuss policy issues. The end result is incredibly sophisticated ways of refecting the problems that exist, such as fscal imbalances' effects on growth, but no attempt to answer why or how such problems can be fxed, and certainly no sense of evaluating or assigning responsibility to agency, that is the decisions people choose to make. I The problem of exclusion by this group hit home to me when I had the good fortune to be invited by a multilateral development bank to give a paper I had 1 For examples, see Norman Loayza, Pablo Fajnzyl ber, and Cesar Calderon, Economic Growth in Latin America and the Caribbean: Stylized Facts, Explanations, and Forecasts, Washington: World 8ank, 2005; Andres Solimano, ed., Vanil'hlng Growth in Latin America: The Late Twentieth Century Experience, Northampton, MA: Edward Elgar, 2006.

An East Asian Modelfor Latin American Success


prepared on why the FTAA (Free Trade Agreement of the Americas) was going to fail on the basis of international relations bargaining theory. My discussant, an economist from the region, stated that no commentary could be given since "there was no regression model" in the paper. As I sat through paper after paper at the conference of economic modeling I could not help but ask one of the experts what the public policy implications of his work were. There seemed to be no clear answer to this question, nor could 1 find a "recommendations" section at the end of most papers. To be sure, the mainstream economists who are the dominant policy advisors have had to respond to the problem of contextualizing economic policies. The initial blindness of the effects of structural adjustment policies on the poor was just one of a series of criticisms (again without suggestions for f xing fiscal problems and other vulnerabilities that led to the imposition of such policies) that have pointed this weakness out. While the World Bank and its regional counterparts have now discovered that good governance and institutions might indeed matter, at least insofar as recognizing the importance of transactions costs,2 they have largely ignored the work of a rich emerging vein of political economy scholars who have been dealing with these problems in a way that pro-actively considers policy and political context in making recommendations (see Tendler 1 997 for example). This work builds upon emerging work in that vein by cross-fertilizing the literature from two regions, EA and LA, with historical, policy-oriented, and statistical analysis.3 The problem again seems to center around the fact that economists are uncomfortable with the lack of a unifying theory of political and social change. More importantly, by defnition as either multilateral institutions or as "neutral" policy advisors their mandate forces them to shy away from any direct political stance. Thus the policy advice of the supposed optimality of a theoretical free market remains at the most general level, progressing to an attempt to "count" the economic costs of red tape and corruption on the otherwise theoretically well-functioning market-based economy. The end result of both class-based analysis and the optimality of markets as the ultimate basis of analysis is that they squeeze out the wide variation and richness of historical experience, thus creating blinders to the huge variety of possibilities. My contention here is that understanding the historical context of the problem is the essential starting point for a solution-oriented empirical analysis. Therefore, 2 Most of the new work espouses a 2nd generation of reforms for neoliberalism, downplaying the shortcomings and conflicts unresolved by such policies. See, for example, Inter-American Development Bank, The Politics of Policies: Economic and Social Progress in Latin A merica 2006 Report, Washington: Inter-American Development Bank, which takes a pure rules-based view of politics, and Pedro-Pablo Kuczynski and John Williamson, After the Washington Consensus: Restarting Growth andReform in Latin America, Washington: Institute for Intemational Economics, 2003, which takes the usual recipe approach of recommending what policy reforms need to happen for the market to function efficiently. As Chang points out, institutions in the North have evolved historically, through a mix of necessity, confict, . opportunities, and policy decision-making. See Ha-Joon Chang, "Institutional Development in Historical Perspective," 499-521 in Chang, ed., rethinking development economics, London: Anthem, 2003. 3 Recent work by Sanjaya Lall and Ha Joon Chang argue along somewhat similar lines to my own efforts for industrial promotion, though they are not applied specifica ll y to Latin America See bibliography for relevant citations from these authors. .

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy


before getting to a direct argument of how a "new path" to equity and growth could be found in Latin America, we must f rst demonstrate the overall feasibility of change in historical context. In this chapter, I build upon the ideas approach to Latin American economic history that I have suggested elsewhere in order to demonstrate two key planks of this foundational argument.4 First, I wish to show that ideas have had the ability to influence economic decision-making throughout Latin American history.5 Second, I endeavor to demonstrate that the Latin American state has always been the key actor in economic decision-making (for background discussion see Vellinga 1 998). Third, I begin the argument, which I f nish in the fnal chapter, that the timing and politics currently in LA lend themselves to a paradigm shift in the EA direction I am proposing. This means that a move towards an EA-style development model is not only time ly, but eminently possible, once the battle of ideas is won. Hopeful ly, this book will be a f rst salvo in that battle. The foundation of my own approach to explaining Latin American political economy lies in building my policy-oriented decision-making framework upon the "historical-structural" approach laid out 30 some years ago by Fernando Henrique Cardoso and Enzo Faletto (Cardoso and Faletto 1 979). The basic idea behind this approach is to look at the importance of historical relationships and circumstances in constraining and promoting certain types of economic decisi0ns. By structures, I refer here to sets of relationships among powerful pol itical actors, as well as norms and ideas about acceptable behavior and shared goals that are embedded in a culture. "Structures" guide but do not determine behavior. Structures are multi-layered, and include actors, ideas, and norms at the international, regional, domestic, and! or local level. In short, structures are relationships between important entities that seem to have some lasting value and staying power. Practical ly speaking, structures are always changing, both in evolutionary and revolutionary terms. In any given structure, moreover, there are multiple possibil ities for achieving an equilibrium (stable situation) among the different powerful actors. Similarly, while the Cardoso and Faletto approach implies "path dependency," in the sense that historical patterns tend to have a staying power, by no means does history determine decisions. Ifanything, I would argue that the array ofchoices always available is vastly underestimated because ofthe cognitive "semi-blindness" that the sway of a particularly strong set of ideas has in different historical periods. One example is the academic literature that arose in the 1 960s surrounding "bureaucratic authoritarianism" in Latin America. This literature was created in the wake of a series of military takeovers of democratic governments throughout Latin America. In general, the literature linked the military takeovers to an economic growth crisis in 4 Parts of this chapter are adapted from my unpublished dissertation, "Ideas and Political Economy in Latin America," PhD. Political Science, Claremont Graduate University, Califomia, 1997. The best general economic history is found in Rosemary Thorp, Progress, Poverty, and Exclusion: An Economic History of Latin America in the 20'h Century,

Washington: Johns Hopkins University Press, 1998. 5 I avoid conducting an extensive literature review here in order to save space. However, I have included a number ofthe most important books on the topic in the bibliography. A broad review of ideas and ideologies in LA is found in Leslie Bethe ll, ed., Ideas and ideologies in twentieth century Latin America, New York: Cambridge Universi ty Press, 1996.


An East Asian Modelfor Latin A merican Success

the region, partly by relating the need to suppress skilled wages in order to improve trade and savings balances. Both this and the related literature on authoritarianism that emerged at the time were strongly pessimistic about the chances for restoring democracy in the short-run. In fact, the literature underplayed the political factors of anti-Communism, which was a key motivator behind the coups (the military seeking to restore power relationships). The revival of democracy in the region throughout the 1 9808 and 1 9908 was as surprising to many Latin American specialists and policymakers as the "fall" ofthe Soviet Union was to international relations theorists. The point is that the operating ideas about the region, which worked in synergy with the policies adopted and political developments, were much more controlling of the possibilities for action than was realized at the time. Ideas and cultural values are perhaps the most underestimated factors in explaining economic decision making. They do not f t neatly into any economics framework, yet they are transcendent in explaining not only why some policy choices are made at certain times, but also the general traj ectory of policy choices over long periods oftime. Moreover, as we explained in the preface, neoliberalism is currently confronting a major crisis so the timing i s right for a paradigm change in the region. Nonetheless, academics tend to be very skeptical about the power of ideas much of social science work is based on more easily followed interests and constraints related to the use of coercion and/or allocation of resources. There is therefore a general under-appreciation for how ideas can shape coercive and allocation-based decisions. It is absolutely vital, therefore, to make the case that ideas can make a di lerence in order to argue that the set of ideas the follow in this book can change Latin American economic policies. Structure, Actors, and Discourse: Three Frameworksfor Explaining Latin American Economic Ideas

In order to understand how ideas affect economic policy, we need to move beyond merely stating the content of the ideas and explain exactly how and why certain ideas became i mportant at certain times, and the exact ways in which they inf uence economic policy at the time. I n order to accomplish this task, I will buil d upon two frameworks that I developed in my previous work,6 adding a third and new framework for this project (rhetoric and interests). We can then apply these to Latin American history to set up the context for our enquiry as to how to move forward. How the Frameworks Work Together

An oversight of most social scientists, and perhaps the weakest point of theorizing in either the social or natural sciences, is the inability to incorporate multiple levels of analysis simultaneously. By multiple levels of analysis, I refer not only to the .spatial interaction of the international, domestic, and local levels, but also to the 6 Anil H ira, Ideas and Economic Policy in Latin America: Regional, National, and Organizational Perspectives. Westport: Greenwood, 1998. I n the book, I not only present the frameworks for how ideas atTect political economy. but apply them in detailed case studies to Latin America. Indonesia. and Egypt.

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy


interweaving of one temporal period with another. To state it simply, it is easy to see that the past is involved in the present, if nothing else through shaping perceptions, and the future (possibilities) also figures into present decisions. I would add to this a third step of sophistication that we need to move forward in our understanding, that is, the interaction of ideas, culture, and political action. These three aspects are the source of the frameworks I propose here. Now, let's look at how each framework operates separately, and then we shal l review how they all fit together in more concrete terms in Latin American economic history. Framework 1:

Moving Beyond Economic Knowledge Networks

It goes without saying that ideas are important. Whil e we would not say that ideas are everything, since action also matters, without ideas, no actions would be purposeful or meaningful. Ideas, whether we are talking about general frameworks, such as monetarism, or specific questions, such as the value of a progressive tax system to social harmony, are the definers of the context, tool s and substance withOn) which pol itical action can take place. However, we also nced to concretize ideas into political action. In order to understand how ideas become policy, we need to iso late empirical referents, including actors and institutions that can act either for or against certain economic ideas. Ideas can exist as a casual conversation, an intellectual exercise, or a work of art, but when we are examining the nexus between ideas and pol icy, we wi l l look to the actors that create, promote, dissemi nate, and fight for or against certain sets or ideas in the policy arena.7 Ideas exist in every stratum of society, but the main conversations that matter to economic policy tend to be concentrated among an el ite set of actors. Economic policies are in and of themselves intricate, requiring an advaneed knowledge of macro and microeconomics. Moreover, unl ike social policy issues, such as abortion rights, economic policy decisions, particularly on the macroeconomic level, may seem more remote to the general popUlation. Lastly, when we think of the main institutions of econom ic policy in most countries these days, we think of large businesses or coalitions of businesses or unions that can lobby the govcrnment, and, more importantly, top-level government agencies. In general, the decision makers in these agencies, such as in central banks, tend to be relatively �

7 My perspective contrasts with approaches that focus on institutional arrangements, interest groups and/or technocrats (bureaucrats who are economic experts as the key sources of economic policy change in LA. For examples, see Kathryn Sikkink, Ideas and institutions: developmentalism in Brazil and Argentina, Ithaca: Cornell University Press, 1 99 1 ; Miguel Angel Centeno, Democracy Within Reason: Technocratic Revolution in Mexico, 2nd ed., University Park, PA: The Pennsylvania State University Press, 1 994; Eduardo Silva, The state and capital in Chile: business elites, technocrats, and market economics, Boulder: Westview, 1 996; Jorge I. Domingeuz, ed., Technopols: free politics and markets in Latin America in the 1990s, University Park, PA: Pennsylvania State University Press, 1 997; Veronica M ontecinos, Economists, Politics and the State: Chile 1958-1994, Amsterdam: CEDLA, 1 998; Miguel A. Centeno and Patricio Silva, The Politics of Expertise in Latin America, New York: St. Martin's Press, 1998; Glen Biglaiser. Guardians of the Nation? Economists, Generals, and Economic Refbrm in Latin America, N otre Dame: University of Notre Dame Press, 2002.


An East Asian Modelfor Latin American Success

insulated from direct reclamations from the public, absent crisis conditions.s Thus we can see that economic policy decisions are largely made by a network of public, private, and academic, and domestic and foreign actors who advise and influence political decision-makers. This is not to say that the public has nothing to say about economic ideas. We know from our own experience that in economic hard times, such as hyperinflation, people often mobilize for demonstrations in reaction against a perceived failure of government policy. Much rarer, however, i s a pro-active and well-organized message from the public at large. In collective action terms, the public as a whole is too diverse a group and the stakes of everyday economic policy action are perceived (mistakenly) as too remote from daily lives to warrant active political stances. The possible counter-argument that there are numerous citizens' advocacy groups lobbying for specific issues, such as consumer R ights, ignores the elitist nature of those efforts. That is, the efforts oflobbying groups are quite often one step removed from their general constituency, and carried on by staff members who live in a capital city, have upper-level educational credentials, and generally come from an above average class background. In Latin America, most well-organized non-governmental organizations (NGOs) usually depend upon funding from outside, primarily US or European funding. Perhaps a good way to differentiate this is to say that the public at large has "notions," or much more basic understandings and ideas about economic policy, than those close to decision makers. We therefore might find that our framework should be one that explains how elite groups affect economic policy action. I have set up a preliminary framework elsewhere (Hira 1 998) that makes a suggestion in this direction. The framework basically posits that "experts" are often well-organized into groups that share common principles and beliefs about economic policy. They help to justify and take responsibility for politicians' economic policy decisions, and so receive the support of political groups in return. They also help to articulate and formulate the recommendations of interest groups, such as large businesses, that have strong opinions on economic policy in terms of the national interest. In return for this service, they often receive financial support (directly or indirectly) from lobbying groups. I feel now that this framework, which I sketched out i n 1 998, may be a bit too limited to explain the variety of ways in which experts can influence economic policy. I think the framework works very well in many situations, such as Chile under General P inochet, or on certain issues in which the level of policy sophistication required is very high, such as the US Government effort to bail out the semiconductor industry in the 1 980s. However, as I will explain in Framework 3 below, the way that an economic issue is portrayed more widely, not just the sophistication of the ideas themselves, may be crucial in opening up the dialogue about policy to a wider audience.9 Thus, as Ravi Roy has pointed out, economic notions can be carried not only in elite 8 Interestingly, this very separation from everyday politics seems to have given economic experts a kind of "wizard-like" quality in which their unusually high-level expertise can guide the country out of its problems relatively easily. 9 Though I di sc ussed this level ofthe discourse in my pre v ious work, I w as unable for lack oftime and resources to fully explore it in the case studies. ,

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy


economic knowledge networks, but also in situations of active political parties or public interest groups, by advocacy coalitions.lO These coalitions have a different type of conversation than the high-level sophisticated conversations that I have so far explored, but one that is i mportant to include in order to increase the scope of our analytical framework. We can broadly call these "ideas coalitions," for simplicity, though they are really ideas-interests coalitions. We will include this level of analysis in Framework 3, which I present below. Framework 2:

Historical-Ideological Periods

Experts in the guise of economic knowledge networks do not act in a vacuum. Rather, they work within an existing palette of widely accepted perspectives and "facts" that guide economic behavior. These perspectives inform the economic policy conversation or discourse about what a country's economic policy ought to be. The notion of a discourse in development policy, marked by distinct historical­ ideological periods, and Thomas Kuhn's framework of paradigmatic change can be used to understand how the ideological frameworks behind development policies change overtime. I I By using a discursive framework, we can explain the evolutionary trajectory as well as the ongoing dynamic of ideas over time and space. In some historical periods there may not be a government capable of setting an economic policy agenda, or there may such divisions among the population so that no one side can "win" the argument about which economic policy paradigm is best. By paradigm, I mean a certain perspective on economic policy that has a 1 0 This is the approach that one would have to take in devising a strategy for changing ideas on the national level - mapping out interest group coalitions and points of equilibria that mirror major changes in policy ideas, and then f nding the Right types of catalysts (for example media coverage, mass appeals, elite-level policy discussions, education, and so on) to move the ideas in this new direction. See Ravi K. Roy and Arthur Denzau, Fiscal Policy

Convergence from Reagan to Blair: the Left veers Right. New York: Routledge, 2004 for discussion of the advocacy coalition approach to ideas, which develops the approach outlined in Paul Sabatier's

Policy Change and Learning: An Advocacy Coalition Approach,


i n The Politics of Freeing Markets in Latin America: Chile, A rgentina, and Mexico, Chapel Hill, NC: University of North Carolina Press, 200 1 . See also Marcus 1. Kurtz, 'The boundaries Westview Press, 1993. Judith A. Teichman takes a similar policy networks approach

of Antipoverty Policy: Economic Ideas, Political Coalitions, and the Structure of Social

pp. 139-166 in Peter p. Houtzager and Mick Moore, cds., Changing Paths: International Development and the New Politics ofInclUSion, Ann Arbor: U. of Michigan Press, 2003. Promotion of neoliberal ideas, with strong value-laden assumptions,

Provision in Chile and Mexico,"

is not in some ways unlike emerging networks of social advocates who push a certain view of the world. For more on this, see Margaret E. Keck and Kathryn Sikkink, Activists beyond borders: A dvocacy networks in international politics, Ithaca: Comell University Press, 1998. 11 Thomas Kuhn, The Structure ofScientific RevolutiOns, Chicago: University of Chicago Press, 1962. The modem foundations for the application of ideas to economic policy are found in Peter Hall, Policy Paradigms, Social Learning, and the State: The Case of Economic Policymaking in Britain," Comparative Politics, 25,3 (Apr 1993): 275-96, and The Political Power of Economic Ideas: Keyneslanism A cross Nations. Princeton: Princeton University Press, 1989. "


An East Asian Model for Latin American Success

meta-level diagnosis of causes, effects, and opportunities that explain the current situation and future possibilities. For example, one school of Latin American thought, dependency, sees attempts by the developing world at cooperation with the First World as not only futile but counter-productive. Thus, the dependency paradigm set a context of inevitable exploitation that guided economic policy prescription by those in that camp and in strong opposition to the Latin American policies of Import Substituting Industrialization (lSI) during the l ate 1 960s and early 1 970s, which they correctly saw as simply changing the nature of the dependency. The problem, being, of course, that they offered no alternative. Development paradigms, whether neoliberalism or dependency, have tended to paint the world in stark conditions, perhaps exaggerating and over-simplifYing in order to push forward new action. We also find certain historical periods, on the other hand, that are marked by a stability among the interest groups, the politicians, and the expert groups. While the personnel may change in each of these groups, the policy outcomes remain the same because the essential relationships among the actors and the paradigms they share are stable. These hypothesized periods are called historical-ideological periods. Historical-ideological periods need not be limited to particular nations. Since development is an object of international study, it follows that it is an i nternational discourse, so that we f nd certain regional paradigms during given time periods, such as the period of the early 1 960s, when many African countries followed a type of state socialism while Latin American countries followed an import substituting industrialization paradigm. The notion of discourse implies constant change, though not necessarily progression. Kuhn's framework provides one important source of change, namely crisi s-driven change. In our examination of historical-ideological periods, we will see that Kuhn's idea that there is a paradigmatic cycle in which certain ideas become dominant and gain acceptance and, upon entering a crisis period, subsequently become susceptible to challenge from new paradigms, seems to fit many of the major changes in regional economic policy ideas for Latin America. However, unlike Kuhn, we are dealing here with the real world of political and economic i nterests and seemingly diverging perspectives. Therefore, we must link Framework 1 with Framework 2 by stating that the existence of a dominant economic paradigm in a particular time period speaks to the existence of a strong and stable alliance of an economic knowledge network, private interests, and politicians. During these periods of time, the more polarized aspects of the national-level d iscourse seem much less i mportant since a general consensus among those who control economic policy has been achieved. Still, there is always room for diverging interpretations of the paradigm itself and in the reaction to the results of policy decisions. Thus we can speak of a "mini-discourse" within these periods in which people who share the same fundamental paradigm argue about how best to implement it in the real world. Framework 3:

How Paradigms Move from Elite to Mass Levels During

Historical-Ideological Periods: The Relationship of Rhetoric and Interests

Thus far. we have seen that there is a symbioti c, multi-level rel ationship between the ideas coalition of experts-politicians-interests and the historical-ideological

Why Ideas and Slate Leadership are the Key Variables in Latin American Political Economy


period. However, we should not neglect the very important idea of legitimacy. By legitimacy, we mean whether the economic policies selected are considered to be the appropriate ones at the time by the public at large. The ideas coalition interacts with the public discourse in three fundamental ways. The f rst is that they seek both concrete and tacit support of their policy decisions. By support, we mean that important groups in society as well as the public-at-Iarge must be willing to make contributions in terms of paying taxes, complying with the new policies, and rallying their organizations' (for example, unions, citizens' groups) active efforts behind the economic plans. In terms of tacit support, we refer to the sense of the public, most prominently filtered through the media, that the economic plan as widely understood is the best one for the country at the time. The sense of the public is clearly reflected in public opinion polls, but there is also a general sense of the tone of the public discourse that is refected not only in the media, but in the daily conversations of the citizens. Economic policy affects every aspect of citizens' lives, from daily working conditions to the prices of groceries, so that every person really talks about economic policy on a daily basis without realizing it. In general, the public discourse does not play a pro-active role in the actual economic policy decisions, unless economic policies seem to prejudice long-held values or expectations, or fundamentally seem to fail. Thus, the second way that the public interacts with the ideas coalition is through manifestations, strikes, protests, and other ad hoc spontaneous communications of displeasure. Throughout Latin American history, there have been frequent strikes and work stoppages, with relative degrees of organization, that have disrupted the intentions of the predominant ideas coalition. These manifestations can happen at times of unexpected or unpreventable economic crisis that may have little to do with the economic policies at the time, such as when the US raised worldwide interest rates and commodity prices sank, forcing a huge downward adjustment by Venezuela in the early 1980s. They might also be directly linked to the economic policies being pursued, such as the recent spate of strikes against the privatization of water in Cochabamba, Bolivia in the early 1 990s. The point remains that the protestors generally do not understand the technical level of the issues - their world view, what they are reacting to and what they want are much more basic than those of policy planners. Thus, there is a natural conflict between those who react to the results of economic policy and those technicians who are trying to figure out how to solve the multiple trade-off.. of incredible complexity, high unpredictability and limited control in terms of the national economy and its relationship with the world economy. It i s no surprise, i n short, that conf ict persists i n LA both because marginal groups are momentaril y upsetting the power equilibrium and because these groups live i n a different world than those deciding policy. For such ongoing conflict to become a manifestation that forces a policy shift, the intensity of the adjustment or perception of injustice must be acute, and may take time to boil over. For example, in the recent Bolivian case, the privatization of the water system had already taken place. It was only when the price increases in the water, in combination with a host of unrelated pain-creating economic conditions occurred, did the momentum against privatization start to build in Bol ivia. We should remember that, in general, most people do not have the


An East Asian Modelfor Latin American Success

realization or the resources to become involved in economic policy making, and do not find readily available channels to do so, thus, only when they perceive a crisis on a personal level do they become involved. The perhaps most important, if subtle, way in which the ideas coalition interacts with the public is through what we might call rhetoric or propaganda. While these terms seem to portray a negative content, whether "spin control" is negative or positive depends on your view of the economic policies in question. Every ideas coalition spins its policies along the lines of the dominant values of the society to whom they are selling them. Thus, economic policies are usually portrayed as benefcial to the "national interest," fair to all groups, especially those who are more marginalized, and creating a system that is rational and "divorced from special interests." As our Frameworks 1 and 2 tell us, nothing could be farther from the truth in regard to economic policy! Every economic policy involves winners and losers, change and stasis, and ideas and interests. Thus, the trick of the "spin doctor," is not only a negative mission. It is always the case that a number of people in the ruling coalition believe sincerely in these ideas, and thus they try to convince both organized interests and the public of their conviction. Whether these people are "true believers" or "pragmatists," they play a very important role in passing on the nature, costs and

Why ldeas and State Leadership are the Key Variables in Latin A merican Political Economy I I Latin American Economy History Through the Three Frameworks

Introduction I make the argument in this chapter that the LA state has always been the predominant actor in development decision-making. The private sector has been too parasitic and too weak; labor unions even more so. Foreign actors have been dominant at times, yet the numerous acts of strong nationalism through LA economic history show that the strength of external forces is highly exaggerated, rather used as a scapegoat by LA elites who do not want to identify their interests with those of foreigners. Building upon my previous work, I synthesize the ideas-interest group configuration of each period of Latin American economic history using the following categories: general ideology of how an economy should run; the policy prescriptions which follow from that ideology; the perceived costs and benefits of different policy actions, such as whether welfare subsidies are beneficial in building up human capital or whether they throw off the ability to reach balanced fiscal budgets and reduce the incentives to work; the dominant interest groups, and the degree to which economic and political "leaders"

benefits of complex economic policies in a way that is generally understandable.

are insulated from them to be able to pursue national interests;

However, the way that policies are "sold" to the public often overshoots the truth

the crisis events leading to the breakup of the dominant ideas interest configuration;

in the sense that unrealistic expectations are set up for political reasons. Thus, the

how the overall economic paradigm is legitimized to members of the public.

pattern of distinct historical-ideological periods is marked at the boundaries not only by changes in concrete economic conditions, but also by a general disappointment

The literature on LA economic history generally identifies six major periods of Latin

at the lack of delivery of some promises of the framework. Conditions change

American economic history, which we call historical-ideological periods. While we

unexpectedly, or in a way that the existing ideas-interest coalition is unable to react

must always remember the important national variations within these periods, ideas

to in the short-run, creating the conflict over relative expectations. In the developing

tend to spread and create imitators. Thus, we can add to our frameworks by saying that

world, the fragility of power coalitions leads naturally to a huge gap between the

there is a demonstration, or "learning" effect, when one country seems to adopt a set of

hopes and aspirations of the economic policy and the actual perfornlance within the

economic policies that seems to analyze and solve similar economic policy problems

important constraints of the world economy, opening the way for economic policy

quite successfully. This can lead to the rise of similar coalitions in other countries, which

"entrepreneurs" to undermine any economic policy framework over time and to sell

might not have happened without the spread of ideas across borders. The neoliberal

their own magical solutions. It is important to point out that the relevant "public" for the selling of ideas has

revolution spread from Chile not only to the rest of Latin America, but also eventually back to the developed world where, for example, pension fund privatization, pioneered

expanded over time historically. Whereas in colonial times, the public was members

in Chile, is a current topic amid social security crises. Therefore, we see that the history

of the elite who had to be convinced of the divine right of the King to accept

of ideas is quite regionalized in spatio-temporal terms.

decisions, from around the 1920s, with changes in literacy, voting, industrialization,

Finally, it is important to emphasize that a crisis alone is insufficient to explain

mass media, and related changes, we begin to see a wider public to whom economic

change. There are a number of historical crises in LA history that have not led to

ideas increasingly become legitimized. Through the three frameworks, we have now

ideological change, such as the financial crisis in Chile in 1982; the peso crisis in

discussed the ways in which ideas matter in economic policy. In order to demonstrate our frameworks' validity, we test them out through a brief illustration of changes in Latin American economic history.

Mexico a decade later (1994); and the more recent fnancial crises in Brazil (1999) and Argentina (2000). Rather, what we see in LA economic history is that in each case of major change, the three gears of ideas change

coalitions, new widely-accepted

ideas, and a crisis of legitimacy for the reigning paradigm, must all work together to usher in a new era of economic policy. The latter is what distinguishes "reform" from an ideas "revolution." A new set of ideas is crucial to change policy. Its absence may lead instead to a strong political reaction, including repression, or chaos amidst a lack of consensus; this seems to describe well recent events in Venezuela, Bolivia, Colombia, and Peru, where no consensus on economic policy exists. other than that the

A n East Asian Modelfor Latin A merican Success


present one is not working. In the rest of this Chapter, we demonstrate the utility of our framework in explaining historical shifts in Latin American economic policy. 1.

Colonial Period (150ff-1800)

The colonial period may be thought of as a period of economic imposition for Latin America, rather than one of national development. Nonetheless, the centrality of the state to Latin American history begins with its origins. The colonial economic system in Latin America was typical of European colonial arrangements throughout the world, which followed the precepts of mercantilism. 12 Latin America suffered from the legal requirement of having to trade through the home country, whether Spain or Portugal, resulting in a slow and halting industrialization process (R. Shafer 1 978). One result of this unidirectional trade is that intra-regional trade within Latin America was negligible (Crow 1 992). Moreover, the economies of these countries developed an outward trade orientation, which has been maintained. Of course, this outward orientation was also a result of the formidable geographic barriers between Latin Ameriean colonies. Latin American government was hierarchically arranged by viceroys in Spanish America. Much of the aristocracy was created by the monarchs from whom they received large land grants. The result was a conf guration of large­ scale estates, which often included allotments of Indian workers (Donghi 1 993). Thus, clearly power relations and the dynamic tools of power centered around land and control of the military. Another important characteristic which contrasted the Latin American with the English colonies was the importance of the Catholic Church, which worked hand-in­ hand with the colonial state. In Latin America, the Church played an important moral role in underpinning the legitimacy of the colonial enterprise, with a few exceptions (R. Shafer 1 978, 68). I ntellectual activity, often based on the Church, tended to emphasize humanistic studies and was characterized by a fear of innovation, which was in keeping with its association with the staid institutions of Church and monarchy (R. Shafer 1 978, 1 20). Our three frameworks show that the ideas-interest group framework during colonial times was based upon the mercantilist paradigm that a colony was an appendage of the home country, and so should provide resources for the interests of the empire as a whole. The coalition consisted of the primary agricultural and mining interests in the colonies, the colonial administration and the King, and, to a lesser extent, the Church. The nature of the ideas of the Spanish Empire (based on the experience of the recapture of the Iberian peninsula), particularly the role of the Catholic Church, the formation of huge landholdings, and the stifling of industry and internal commerce set out a distinct trajectory for LA from the English colonies. 12 The intentions of the colonizing country were to procure raw materials and primary products for home consumption, wealth accumulation, and trade with other developed countries. In return, the home country provided government, trading facilities, and fnished products for consumption in the colonies. For a review of mercantilism, liberalism, and other international political economy terms, see George T. Crane and Abla Amawi, editors, The Theoretical Evolution (�f International Political Economy: A Reader, New York: Oxford University Press, 1 99 1


Why Ideas and State Leadership are the Key Variables in Latin American Political Economy 1 3

The real crisis of legitimacy started to arise with the b irth of new generations of elites in the colonies, who resented the long-distance decision making of the Crown from overseas, and the meddling and favoritism of the "peninuslares," or Spanish! Portuguese at home, thus undermining the legitimacy of the foremost mission being to enrich the mother country. 2.

Latin America in the Age ofLiberalism and Independence (l 80ff-1870):

Liberals versus Conservatives

Latin America's development, from the beginning, then, was shaped by ideas and the state. The development of l iberal ideas in Europe and the liberal revolutions in the wake of the Napoleonic Wars were directly responsible for changes in the political and economic traj ectory of Latin America in the turbulent nineteenth century. This period is a great example of what happens when there is no consensus around a new paradigm to order society after major changes in social relations and power take place. The creoles, elites who were born in the New World, were an instrumental group for liberal ideas because of their dissatisfaction with the colonial status quo. Creoles desired political equality with Spaniards, freer economic trade with Europe, to institute the new liberal ideas of political Rights, voting, and representation for the (upper) middle class, and for a separation of Church and State. Members of the upper and privileged class as well as some middle class creoles fought as conservatives against the l iberal movement. This schism between liberals and conservatives continues to be a prominent feature of the Latin American political landscape, albeit in much mollified form. The loosening of commercial and trade restrictions for trade within the Empire began in 1 764, but occurred only gradually. In fact, the opening of free trade within the Spanish Empire was only slowly implemented, having been fully extended only by 1 789 (R. Shafer 1 978, 1 88-9). By the latter part of the eighteenth century, the ideas of the Enlightenment had begun to seep into Spanish America, yet it took more than a century for a consensus to be achieved around these ideas, and they were never fully legitimized. These ideas wcre spread by the constant interchange of people between Latin America and Europe. The Enlightenment essentially held reason as a guiding force for showing man how best to make decisions and form society. These ideas gradually worked their way into the elite discourse. Most importantly, valuing reason as a means to truth has a way of implicitly "leveling the playing f eld." Thus, once adopted, the idea that the best mode of logic should carry the day can start to creep in on superiority on the basis of birthright, class, connections, swordsmanship, morality, and other forms of nobility. The net effect on societies has been to slowly increase the legitimacy ofthe power of those who are decently educated, namely, the middle class. However, there was a major mismatch between these new ideas and the much slower changes in power relations. The early proponents of liberalism were harshly dealt with by authorities, with little public outcry.13 In fact, most creoles in the eighteenth century were more 1 3 Many were inspired by the ideals of the French Revolution, such as Antonio Narino in Bogota and Francisco Miranda in Venezuela. See R. Shafer, pp. 292-3. The American


An East Asian Modelfor Latin American Success

concerned with the need for greater economic freedoms than political independence (R. Shafer 1 978, 293-5). The French Revolution and the subsequent Napoleonic Wars served as a catalyst for change for the crisis of legitimacy introduced among the LA elite by the new ideas. Napoleon forced Spain to ally with France, thus placing an enduring burden on Spain for resources in the ensuing wars. This burden, for example, led directly to the seizure of Church property (R. Shafer 1 978, 299-300). More importantly, Napoleonic rule led to the default of local decision-making by creoles in Latin America while Spain was occupied and the Portuguese Crown fed to Brazil. By the early 1 800s, republican governments were set up everywhere, and nobility was abolished. Though economic power continued to be concentrated in the hands of a small wealthy class allied with the military, with economic growth, a middle class, albeit one imitative and subservient to the upper landowning classes, began to appear (R. Shafer 1 978, 357-8). As the new governments faltered under open political challenges, an era of caudillismo [charismatic military based leadership] spread across Spanish America. Military caudillos in the capital regions slowly consolidated power from regional counterparts backed by large landowners. Exports were generally concentrated in agricultural products, and led to the development of related industries, such as financial houses for export loans, railways, and shipping. Other sectors of the economy, including manufacturing, were very slow to develop, with export industries dominating economic policies, such as maintaining low tariffs as well as continuing capital fight. The result was a lack of protection for infant industries and an inadequately developed local financial system (R. Shafer 1 978, 373-6). In sum, the system of liberal economic policies was set up to benefit the large exporting landowners (Donghi 1 993, 1 1 8). Thus, we see the early years of independence as characterized by the gradual formation of two distinct elite-based ideas coalitions seeking to create their own view of what the political economy should entail- and the catalyst for change being the delegitimization of the Spanish and Portuguese Crowns. The l iberal coalition, backing federalist decentralization, minimalist government, and a primary export policy, was dominant immediately after independence. After a series of civil conf icts throughout Spanish America in the latter 1 8th century, however, the conservatives gradually won the battle for political centralization. In the end, Church inf uence was reduced (an aim ofthe liberals), but the central role of the state was solidified (an aim of the conservatives). As in the previous historical period, outside actors wielded important power within the region. The European powers as well as the United States protected and expanded their economic interests through military intervention on numerous occasions during this period. By the end of the 1 9th century, new ideas of positivism would help to usher in a growing role for economic advisors to the LA state, and a gradual consensus was created around the compromise of greater secularism but also a centralized state. Revolution was equally important for inspiring Latin American creoles, such as Bolivar. While the United States failed to provide material, as wel l as ideological, support, both the United States and Britain served as havens for creole exiles. The British were somewhat more forthcoming in supporting the early efTorts by creole activists. See Crow. pp. 4 1 4-42 1 .

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy 1 5 3. Export Boom and Early National Development, 1 870-1930: Latin America s First Age of Liberal Economics

The beginnings of the 20th century were marked by the steady development of a politically active middle class and labor unions that would became important factors by the 1 930s and work hand-in-hand with an increasingly active LA state that began to openly discuss development. While centralization had won out over liberal-supported federalism, and the military still played a political role; suffrage increased, and states moved increasingly toward rejection of a role for the Church (R. Shafer 1 978, 503-4). Except for the Mexican Revolution, this period of LA economic history includes a relative consensus among elites on the role of the state as primary products exporting facilitator. The elite consensus behind this idea of the state facilitating development was reinforced by periods of economic boom for many of the countries, who invested to increase their production and reduce the transportation costs of their exports. The world economy, along with technological developments such as refrigeration (leading to beef booms in the Southern Cone) supported strong demand for LA commodities. Foreign investors were the major suppliers of capital for this period, and therefore owners of much ofthe important economic stock, such as the railways. The period was the first one in which population increased rapidly, along with increasing urbanization, and su bsequent waves of European immigration. The population increase, along with the increasing earnings from exports, led to the creation of a viable middle class for the first time. '4 The middle class tended, in typical fashion, to identify with the upper class, rather than with the other emerging social actor - the urban lower class, thus legitimacy around this paradigm was preserved for a time. The first significant middle-class parties, such as the Radical party in Argentina, appeared at this time. The elite-dominated system continued to prevail despite changes in formal political rights, especial ly in rural areas, where large landholdings continued to be the rule. Only in Mexico, after its seven year revolution, was there any significant land reform. The role of the government in society increased dramatically in this era, as a middleman-facilitator between agricultural interests and foreign imperial ones, and partly to deliver new social and economic goods to the emerging middle class, gains made possible by surplus generated from the export boom. For example, literacy rates in most countries shot up as the state took on the role of primary educator. The beginnings of a role for the bureaucracy also developed. The public role in the economy was limited, of course, by the continuing predominance of primary product exports in the economies, and dependence on unreliable and insufficient trade taxes as the primary sources of revenue (R. Shafer 1 978, 502-6). Therefore, foreign imperial companies continued to play crucial roles in the development of infrastructure to service the primary export-based economies. Yet, despite all of these apparent gains in economic growth and standards of living, the consensus around this model remained in legitimacy crisis. Economic 1 4 In Argentina, for example, the middle class was considered to be one third of the population by 1 930.

A n East Asian Modelfor Latin American Success


policy maintained a laissez-faire orientation, despite growing resentment at foreign investors. Quickly growing income was accompanied by more obvious income and wealth inequality. Though economic and demographic changes began to foster an emerging middle class, access to political structures remained highly restricted. The crisis of legitimacy was a crisis of a lack of adaptation and adj ustment. The lack of development in institutions led directly to the Mexican Revolution, and subsequently to the populist dictatorships that began to attempt to address the pressing ideas of nationalism and redistribution. The final catalyst for change was the world economic market collapse in 1 929, setting off the Great Depression. This shock from outside would completely and forever change Latin American political economy. 4.

The Age ofNational Populism amid a Changing World Economy: The

Beginnings ofNational Industry Through Import Substitution, I930 �55

The export-led era was forcibly ended when world trade abruptly shut down with the commencement of the Great Depression and, subsequently, World War I I . Besi des losing export markets, and their main source of wealth, Latin American economies also lost their sources of finished goods, necessitating domestic industrial production. Latin American governments became increasingly active in the economy in promoting the cause of industrialization, as well as attempting to take advantage of the huge world demand in commodities created during the World Wars. This period of early industrialization probably accelerated the shift, in Latin American economies due to the changing nature of the world economy 1 5 and in response to the legitimacy crisis of the primary product exporting (PPE) model. Following Engel 's law, as European incomes increased, and families became smaller, European demand for primary goods declined (relative to demand for manufactures as a proportion of their incomes). At the same time, supply m arkets became more competitive, which was heightened by the high tariff barriers that agricultural exports faced in developed countries. These trends would continue, generally, over the course of the century. The parallel trend, of course, was skyrocketing demand for manufactured goods (E. Williamson 1 992, 3 1 9-2 1 ). The Great Depression alone would be insufficient to explain the nature of the change that occurred, which has been described as the first phase of import substituting industrialization ( I S Il ), during which consumer goods begin to be manufactured domestically (Haggard \ 990). The vacuum of foreign interests during this period gave the LA state space to expand its political role and room for nationalists to push for industrialization as a way of increasing autonomy. Industrialization went hand-in-hand with the aforementioned demographic and socioeconomic changes,


Clearly, industrialization began long before the Great Depression period. Stimulation

from the export boom at the tum of the century had already led, notes Bulmer-Thomas, to the beginnings of i mport substitution. With the First World War, however, primary commodities

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy 1 7

cementing a skilled working and nascent white collar middle class that now became a strong constituency for the new consensus behind lSI. I S I I typically began with consumer products, such as processed food and textile factories, with few technological requirements, and extended next to light electrical industry, relying in the latter phase on foreign capital which had been received before the Crash of 1 929. World War II brought the beginnings ofthe second phase of import substitution - the move towards production of heavy industry, capital goods, and consumer durables. The eventual acceleration to ISI2 did not just happen naturally or "inevitably" as some economic historians suggest. Governments throughout the region set up industrial planning boards, such as CORFO (Corporation for the Development of Production) in Chile, that sought to diversifY the economy away from primary products. The change in economic policy was refected by a shift in the relative national power of pol itical interest groups from a dominant rural oligarchy toward the state, urban industrializers, and, to a lesser extent, organized labor. More importantly, it was the new consensus of this coalition that created l S I and explains its timing. Government involvement in economic activity would increase steadily until the debt crisis ofthe 1 9805. The goal of national economic development, latent in Latin American history, was consciously adopted for the first time by the state.16 Nor is adding interests to crisis sufficient for explaining the change. As with the new ideas of liberalism from the period of Independence, we should remember that the milieu of the age centered around the new ideas of national development through dictatorships in Germany and Italy, as well as the radical ideas of the Soviet Union and labor, and around the development of mass political parties which, along with industrialization, mobi lized new elements of the population into political discourse. The shifts in international economic structure created by the Great Depression and the World Wars catalyzed a parallel adj ustment of power relations in LA, but the new ideas of populist national development determined their direction. Politically, with few exceptions, the "solution" for these new challenges of ideas and interests, for LA, was one of creating populist dictatorships, inspired by European examples. The dictatorships attempted to legitimize themselves to a wider constituency for the first time, using mass communication. The new nationalist populism correspondingly justified government actions which asserted economic independence. National popUlism tended to be personalistic as in previous eras, however they were now married to strong state bureaucracies. The new popular dictators took pains to cultivate a broad apparent following, in contrast to the back room oligarchic politics of the past. Leaders such as Vargas in Brazil, Peron in Argentina, and Cardenas in Mexico not only organized the labor sectors of their economies, but began the first serious programs which promoted domestic industry. The first steps in their plans generally included nationalization of foreign-owned industries, such as petroleum in Mexico. Ultimately, these dictators were able to create two parts of Peter Evans' triple alliance the state and organized industrial capitalists and workers (Evans 1 979). Unfortunately, the lack of available capital

became much more valuable to the warring industrial powers. The Great Depression led to the first conscious efforts at industrial ization and self-sufficiency in manufactures. Victor Bulmer­ Thomas. "The Latin American economies, 1 929-39," in Bethell, v.VI,


Latin America since

Economy, Society and Polltlcw: Part I Economy and Society, 1 994 pp. 65-1 1 5 . ,


While economic goals were a part of political groups' ambitions throughout Latin

American hi story, an active state role. even in Mexico. which had undergone a socioeconomic revolution. came in fact onl y after the tirst World War. See Donghi. pp. 274-n.

A n Easi Asian Modelfor Latin A merican Success


equipment and investment on the world market meant that LA had to rely on a third alliance partner, foreign capital. LA industries therefore emerged from the war period as mere infants. Moreover, the sudden move to industrialization created a strong anti-productive bias against the old regional centers of export production who were also the source by which to finance lSI. Thus, the i nherent contradiction of taxing a source (primary production exports) which you are creating a bias against led eventually to reliance on foreign partners' financial and technical aid. This was also the period in which economists began to dominate the discourse of development in the region. We have to remember that the concept of development originates only after World War II. Thus, our historical view of "ideology" must be tempered with the fact that economic knowledge and theory were still quite underdeveloped, for example the consensus about fighting inf ation through monetary policy was not solidified until the 1 980s. Positivism had made a deep impression on the Latin American militariesl? who then turned to using their supposed organizational and planning superiority on the economy. The period was one in which not only the mass public, but also Leftist parties and bureaucracies, which generally had taken shape in the beginning of the century, began to wield real influence over economic policy. I n Chile, for example, as in other Latin American countries, an urban working class, urban capitalist, and state coalition began its consolidation of power, squeezing agriculture through its economic policies. The primary product surplus from exports, especially in agriculture, was thereby gradually eliminated until incentives became so reversed that Chile became a net food importer by the 1 970s. The strategy of using agricultural surplus was not inherently wrong, so long as it was not bled towards diminishing returns, but the capital invested in industry did not pay off in many cases. The industries which developed were inefficient and dependent upon state protection for the most part, and much of the capital surplus actually went abroad in capital flight (Mamlakis 1 976). The consensus around lSI is reflected by the fact that many of the countries in the region enjoyed regular elections, with the m i litary serving only as a guarantor of peace for civilian rule. This is the period when framework 3, the importance of "public" opinion, begins to make a clear impact on decision-making. The dictators captured the continual and growing public sentiment of anti-imperialism and the desire for wresting national control from foreign interests, culminating in seizures of assets such as Mexican nationalization of the oil industry under Cardenas and Peron's nationalization of the British railways in Argentina. Ironically, it was the very success of lSI in industrialization and in access to political structures by the middle and upper working class that led ultimately to its legitimacy crisis, catalyzed by the Cuban Revolution in 1 959. The desire for greater redistribution was insufficiently addressed by lSI. Though growth and industrialization proceeded, relative inequality 17 For example, the military in Brazil was, in fact, led by an intellectual, Benjamin Constant, to overturn the monarchy in 1 889. Constant was inspired by the Federalists of the United States, such


Hamilton. Jefferson. and Madison. and was a member of one of the

numerous national Republican clubs 553-58,

which began to appear around 1 870 in Brazi I. Crow. pp.

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy


worsened. The increasing popular mobilizations around the region for more drastic redistribution and the specter of Communism backing such popularly supported changes pushed the m i litary to step in. 5.

The Communist Menace, ISI2, and the Rise ofEconomic A dvisors, 1 955-80

From about the middle of the 1 950s, a new age of Leftist political operations set in motion a cycle of military rule repressing recently activated l abor forces, which seemed integral to national development, and centrist rule attempting to compromise with it. A new battle of ideas on the Left, center, and Right was waged, leading to a new cycle of economic instabi lity reflecting a lack of consensus. In hindsight, as in the 1 91h century, it is the lack of ideological-interest consensus that caused LA to miss thc advantages of a unique period of world economic boom, as much as the policy errors made. The cycl e ended, in general, by the end of the 1 960s with harsh military dictatorsh i ps constructed across LA for the express purpose of preventing Commun ist revolutions, such as the Cuban Revolution of 1 959. The Cuban Revolution had brought to the t(lrefront anti-U nited States feelings which had been present in LA ttH some time and which reflected continuing foreign dependence through the triple a l l iance. At the same time, LA regimes were strongly influenced by American advice and resources for economic medication to the radical threat. The resu lt was an attempt, and fai lure, in most cases, at moderate economic reform, including land red istribution, gradual national ization of key industries and exports, and regional integration in the 1 95 0s and early 1 960s, which pleased neither of the polarized pol itical extremes and so were i n feasible. Examples abound, including Frci in Chi le, Quadros in Brazil, and Frondizi in Argentina. Both moderation and polarization were ended by crushing repression from conservative clements and the military throughout Latin America by the mid- 1 970s. Given the common enemy of Communism, United States policy, and that of international institutions, such as the United Nations, turned increasingly to development planning a la the Marshall Plan. The prevailing wisdom i n the early 1 9605 was that economic development would undercut the lower class bases for Communism. In some cases, such as Guatemala in 1 954 and the Dominican Republic in 1 965, the US directly intervened to prevent popularly elected Leftist governments from taking power, but in most cases, the infl uence was indirect through economically-administered incentives and punishments and aid to military regimes. The period was understood by some Latin American analysts as bureaucratic­ authoritarianism, in contrast to the more personalistic dictatorships of previous years. While the reaction to the Leftist threat was the primary military justification for the initial wave of dictatorships, the LA military moved towards an economic paradigm of greater autonomy from the world economy through ISI2 as its main source of legitimacy. However, though industrialization did deepen throughout the region, with the major exceptions of Brazil and Mexico, most countries in this period failed to sustain the economic growth rates which were thought necessary to reinforce middle class support. The period was one of frequent macroeconomic instabil ity, which carried over into a good part to the 1 970s. when the solution of greater government


An East Asian Modelfor Latin A merican Success

intervention became ever more frequent. The turn to greaterc intervention belies the dependency idea that LA states are incapable of responding to internal crises and the neoliberal idea that the state is incapable of fostering development. The responses to infationary and exchange rate volatility during the period, in fact, were sound, including efforts to increase exports, speed intra-regional integration, and increase agricultural productivity, but there was inadequate political consensus around them preventing adequate implementation. Some analysts, with good reason, believe that the move towards a second stage of import-substituting industrialization was inherently doomed to failure by several (in hindsight, anyway) obvious economic factors (Skidmore and Smith 1 989, 567). These include the continuation of balance of trade problems, since the need to import f nished manufactures was now replaced by the need to import the capital goods to make them, particularly given continuing technological advances by developed countries' industries. In line with this problem was the need to have heavy foreign involvement in the initial stages to f nance, set up, and run the capital goods producing factories. Secondly, the limited domestic markets meant inefficient scales of production. The result was the typical developing country dualistic economy of a small pocket of highly developed industries, few local owners, heavy state and foreign participation, surrounded by a sea of poverty and restiveness. Why did LA countries continue with ISI2 when most neoliberal economists claim that such efforts of state intervention are doomed to failure? In good part because they had achieved industrialization and product diversification through such policies, despite the heavy costs. Perhaps the most inspiring economic development was the Brazilian military's early economic success. After taking over in 1 964, the military adopted a monetarist reform program, with inflation indexing of wages, rents, interest, taxes, and other economic transactions, encouraging foreign capital, and procuring more aid. Wages were indexed, however, so that real wages actually declined. The monetarist correction was popular with the United States, the World Bank, and the International Monetary Fund, as was a new found anti-Leftist vigor, which certainly helped Brazil to obtain more aid from them. At the same time, the Brazilian state increased its fiscal policy interventions, undertaking large-scale and high-risk projects in several areas, leading to state domination of the oil, steel, energy, and some parts of the transportation industries. The state mollified private capitalists' displeasure with the policies by spreading around contracts. The policies worked spectacularly in terms of gross national product growth, and both Mexico and Brazil were deemed "miracle" economies. However, I S I2 began to fal l apart in 1 974, in the wake of the world oil price hikes. At that point, inflation soared, government deficits went out of control, and the state, responding by borrowing on world capital markets, began to accumulate a huge foreign debt (Skidmore and Smith 1 989, 689-9 1 ). Nevertheless, the Brazilian military, which continued to justity its stay in power by the economic miracle during the 1 9605, did succeed in diversitying exports somewhat for the long-run and in developing several manufacturing industries for export, such as the highly successful aircraft manufacturer, Embraer (see Chapter 6), and world class steel, mining and petrochemical industries. The Brazilian military's economic activism, more

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy 2 1

importantly, inspired other Latin American militaries to adopt similarly economic agendas, which they also used to legitimize their rule, such as that which took power in Peru in 1 968. Leftist i deas were inspiring but could not develop political consensus. Cuba's Communist experiment based on Castro's unexpected turn amidst government collapse was all but politically impossible to repeat in most places. Not only did domestic conservative forces, including the military and the United States, rally to defeat such ventures, such as Allende's ill-fated presidency in 1 970, but Castro's success in stabilizing the political economy relied on several unique factors. These included the surprise ofCastro 's conversion to Communism; the wholesale emigration, both voluntary and forced, of opponents ofthe regime to the United States, including large portions of the upper and middle classes; the United States' reluctance to move overtly against Castro after the Bay of Pigs fiasco and its tacit agreement with the Soviets; the massive subsidization of the economy by the Soviet Union; and the reluctance of the Soviet Union to overextend its economic and military resources, especially after the brinksmanship of the Cuban missile crisis. While a number of M arxist guerrilla movements spread throughout Latin America, not until 1 979 did one succeed. The Sandinista regime in Nicaragua, however, suffered economic collapse under the continuing civil war and international embargoes, which were financed and directed in good part by the United States. With the collapse of Soviet financing in the late 1 980s, the ever pragmatic Sandinistas settled for peace, giving up most overt control, while maintaining control of the army and organized labor for continuing political involvement. The Cuban model was thus never really a viable economic paradigm for LA because it was never convincing to the middle and upper classes who have power. Its widespread inspirational appeal stems back to the Creole aspirations for greater autonomy from external interference during the colonial period- in sum, anti-imperialism, not Communism, is the favorite card for LA. By the 1 970s, military governments in much of LA, using heavy repression of Leftists, had won the civil wars but had not reached consensus or legitimacy for their rule. This led them to move towards economics as the basis of their continuation in power. The boom in oil prices during the 1 970s helped Latin American countries which produced petroleum, such as Venezuela and Mexico, to directly finance government spending, while others relied more heavily on the vastly increased private capital markets to borrow at low but fexible interest rates for the same purposes. By the 1 970s, most countries in the region, in attempting to gain the foreign exchange needed for the new sky-high oil prices and in adopting policies of state-led growth, borrowed heavily from the emerging world capital market. The Mexican government, which actually enjoyed the benefits of higher oil prices, still borrowed heavily while engaging in high cost and low efficiency state development projects and social spending. In a sense, we can see that the industrialization progress of the 1 960s was impossible to sustain given the oil price increases of the 1 970s and the resulting contraction of the world economy. Instead of retrenching their industrial expansion plans, governments followed a political, rather than economic logic, in terms of borrowing heavily to continue large projects with no accountability for results. In other words. the consensus around the ideas of ISI2 prevailed within the


An East Asian Modelfor Latin American Success

military regimes over a myriad of necessary possible adj ustments in state-market relations, including moving straight into E xport-Oriented Industri alization (EOI) as had East Asia, developing more backward linkages, or neoliberalism, the path of least resistance ultimately taken. Therefore the pressing problems exposed by l S I were never addressed. In effect, Latin American states continued to rely upon imports of Northern technologies, capital goods imports, and intermediate goods, only temporarily, until their import substitution phase was complete. Secondary import substitution policies meant protection of those same industries which the military governments had decided to champion, through a continuing effective taxation on agriculture. Overvalued exchange rates, borrowing heavily on world capital markets, and price subsidies which benefted urban workers were typical of this scheme. The costs to primary product producers was by now very heavy, and continuing nationalizations within the region led to a decline in activity and export earnings in these sectors. Pressures for social reform were held in check by nominal attempts at reform; co-optation of agitators through selective social spending; and. above all , repression of radical movements by the military. The m i litary governments pushed to new heights the LA state's claim of economic expertise and thus its vulnerability to legitimacy crisis at any sudden downturn. The change in the world economic conditions and the economic crises of the 1 980s would lead to middle-class rejection of these regimes, ironically, with devastating results for moving towards addressing the ongoing underlying weaknesses of the economy, such as inequality in land redistribution, export concentration and lack of export prowess in industrialized goods, lack of investment in human capital, and weak fnancial structures. 6.

The Debt Crisis and the Move to Neoliberalism, 1 98{}'-?

The golden age of government spending for social stability and rapid industrialization came to a screeching halt with the sudden increases in interest rates, partly refecting the new United States' monetary policy, in the early 1 980s. Given the huge debt burdens incurred, and a steady deterioration of commodity prices, most Latin American countries, and the military regimes in charge ofthem, found themselves in severe economic and political crisis, yet as we have seen the consensus around I S I2 had begun to break down by the late 1 960s. The solution for the m i litary regimes was simple: to abdicate responsibility for the economic mess by handing over the reins of government to democratic regimes. More importantly, the threat of Communism had faded from the societies, and the Left moderated its political position, opening the way for a more peaceful and acceptable centrist rule. In a way it is ironic that we more recently returned in a sense to the status quo ante of the early 1 960s centrism, however, this time the center seems to enjoy considerably more legitimacy and economic knowledge has increased. The Chilean "miracle" provided a new model of development that spread throughout the region. The m i litary took over in September 1 973, ending the widespread chaos of the Allende years. The military erected a highly repressive political regi me. whi le maintaining a few of the economic gains of the Allende administration. including the nationalization of copper. and. to a lesser degree.

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy 2 3

land reforms. I S These policies began a new regional development paradigm for Latin America, which involved the shrinking of the state from an active role in industriali �ation, a return to emphasizing primary exports, the opening of the economy to foreign investment and trade, and tight monetary and fiscal controls. The Pinochet regime also marked the rise of an overt group of technocrats who made economic decisions and openly vied for political support of their economic programs, for the first time since Diaz's cientificos formed their own political party. The " Chicago Boys" in Chil e soon inspired counterparts in other countries of Latin America. Ironically, the neoliberal revolution requires a stronger state, one which can not only cut off special interests who enjoyed lSI subsidies, but also one which can enforce property rights, set up a stable regulatory environment, provide insulation for monetary policy, and act strongly to cut spending and raise and collect new taxes as part of fscal policy. Even well-ensconced Northern governments would have a hard time making such adjustments while dealing with huge external debts, high unemployment, and the myriad of political, social, and economic problems. The 1 980s marked the rise and acceptance of technocrats throughout Latin America to make economic decisions, particularly in instituting the monetarist policies needed to survive the debt-induced austerity of the period. Many of the new pol iticians since this period had distinctively technical degrees in Business or Econom ics, for example, this includes every Mexican President: de la Madrid, Salinas, Zedillo, and Fox (H ira 2007). By the 1 990s, with the inability to utilize fiscal policy given debt burdens, newly democratic Latin American states shopped for a new economic policy framework which would a l low them to deal with the external debt and create some growth base for legitimacy among the upper and middle classes. By this time, the "East Asian miracle" of fantastic and sustained economic growth rates could no longer be ignored. The miracle was widely ascribed to foUowing sound macroeconomic policies and openness, particul arly by international organ izations, a classic case of the infl uence of misrepresented ideas, as we discuss in Chapters 3 and 4. The rise of Latin American economists trained at United States' universities to political power and the success of the monetarist policies of Chile were additional factors in laying out the basis for a new political economy. Given the failure of Communism, the Left was also by now largely persuaded of the need for a capitalistic system, although there was great objection as to the degree to which government spending on social services should decline. In Argentina, for example, a leader from the Leftist Peronist Party, Menem, was the first to succeed in cementing the new monetarism. From Mexico to Brazil, businessmen and economists rose to power as leaders on expressly economic platforms. Carlos Salinas set the precedent for courting international capital by setting tight monetary policies for Mexico and liberalizing the banking and investment sectors. In each case, strong economists came to the forefront on monetary policy, such as Foxley in Chile, Aspe in Mexico, and Cavallo in Argentina. In Peru, the economic formula was first applied by the popular d ictator, Fuj imori, and then by Toledo, a former World Bank economist. The economic policymakers follow the same recipe. In many cases, the 18

See H ira.

) 998

for a full-fledged discussion of the Chilean case.


An East Asian Modelfor Latin A merican Success

leaders who first applied the policies, such as Salinas and Alfonsin, have become, in part, scapegoats for the heavy costs of attempting to transform fiscal policy and drastically reducing protection. However, with their succession, the policies have nonetheless not fundamentally changed. Certainly, Latin America had limited economic choices given the harsh world l iquidity situation of the early 1 980s. Still, given the history recounted herewith, the return to an embrace of foreign investment, including ownership of key industries and resources, a return to a reliance on primary product exports, and the retreat of the state from economic and social activism, similar to the primary product export period of the turn of the century, must remain a surprising twist of circumstances indeed. The attitude towards the North is one of renewed invitation on the surface. The policies concentrated on monetary and fiscal restraint, which, of course, have added to the slowdown in the economies, although restoring some semblance of stability to the investment climate. The slowdown in economic growth and the accompanying cuts in fiscal subsidization were and are being felt particularly by the middle and lower classes, while expectations have continued to increase, following the trajectory of the century. However, it is also important to note that the LA state had to create some insulation in order to push forward these changes it had to reduce subsidies and ownership which were deeply entrenched in the interest groups of the society. The fact that the state was able to precipitously reduce the role of labor unions and state-owned or reliant domestic private interests (unexpected by the "rent-seeking" theories of economists) underscores more than ever the strength of the LA state to change direction. As we outlined in the preface, neoliberalism seems to have run its course amidst the current crisis of equity and growth, the brief consensus around these policies refected in the popUlarity of Fujimori, Salinas, and Menem, has gone the way of those erstwhile champions. The region is now ready for another shift. Two of the three elements of change are present: c hanges in interests, and de-legitimization. What we hope to provide in this book is a third element - a new set of ideas that can create a new consensus. Conclusion: Patterns of Growth and Equity in Latin American Economic Hist01Y

We have now accomplished the goals of this chapter by examining LA economic historical change through the prism of ideas, consensus, and legitimacy. We have shown that the fortunes of LA's economy are intimately tied to the world economy. As the world economy has shifted, this has created a forced change in the outer wheel national policies to respond to these conditions. The change in the outer wheel, as well as evolutionary and other changes, including im- and emigration, have meant continual shifts in domestic political economy coalitions in LA. We have seen that from colonial times inclusive to the present neoliberal reduction of state involvement that the LA state has been at the center of all development decision­ making. It is the state, not a revolutionary movement by "the masses" nor a deus ex machina [God in the machine] market, that is at the center of development decisions. In the ultimate crises. even a successful revolution as in the case of Cuba follows

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy 25

state leadership. In the case of the debt crisis, the movement toward neoliberalism and now towards reforming it, is led, ironically, by the state. At any given point, LA could have adjusted in a myriad of ways, yet it was the ideas in fashion that determined the direction and type of reaction. Otherwise, we would never have seen winning coalitions geared to make the state work for them, give up power. LA could have continued with a monarchy, as was tried in Mexico by Maximilian, in Brazil by Pedro II and in Paraguay by Solano Lopez. However, Enlightenment ideas of republicanism and liberalism shaped the Independence period in a different direction. The Independence period could have led to a continuation of agricultural elites, in the pseudo-liberal forms adopted by the end of the 1 9th century. But the ideas of positivism and state responsibil ities for development activities, such as mass literacy, went hand-in-hand with the world economy's support of the primary product export boom to yield a state for the first time actively discussing national development. This period of elite alliance with neo-imperial economic interests continues in some form today, but the cozy restiveness of the beginning of the century changed as new waves of ideas came to Latin America's shores. Changes in the world economy beginning with the Great Depression went hand­ in-hand with changes in ideas. The ideas of anarchism, socialism, communism, and fascism all had major impacts upon the region's development trajectory, as did liberalism beginning earlier. National populism was the version of these new ideas that created temporary consensus in LA. Conscious policies of suffrage and middle class wel fare (a la Battle in U ruguay) to increase middle-class inclusion demonstrate the power of ideas to affect power relations. l S I became a national development strategy that moved the region forward in industrialization, despite the ravages ofthe Cold War and the Cuban Revolution. lS1 could have been modified in response to the oil crisis and the debt crisis ofthe 1 9808. It could have been adjusted downwards into more modest goals, including paying more attention to agricultural productivity and an EA-style export orientation. Unfortunately, the ideas of neoliberalism prevailed instead, leading the state to cut down and change the nature of the patron-client relationships built up over a century of state-building (though these have been merely transformed into patronage networks with far less social or national benefit). In some senses, the LA state has now decided to return to the late 19th century situation of primary commodity exports, except now including labor-intensive exporting, and abandon independent national industrialization projects. In sum, we have shown that ideas have been central to the choices made in any given age. They are central to the overall evolutionary direction of national development and to the adjustments that need to be made to changes in the world economy. They provide not only a focal paradigm for development, but act as a cementing force for decision-making elite interest groups, and a bulwark of legitimacy to the marginalized elements of large portions of the population. Their "stickiness" can be explained by both the achievement of some consensus and occasionally legitimacy and by the lack of other ideas that have compellingly met the needs ofboth power interests and national interests of development. By showing that ideas matter, we have proven that a new alternative path, based on a convincing set of ideas, and given adequate political support by elites, is a clear possibility for LA now. as it was in the past. We have also shown. however. that the history of LA


An East Asian Modelfor Latin American Success

economy has been one of only temporary consensus. The most primary motivations: anti-imperialism/independence/recognition; raising standards of living to First World levels, and improving equity and reducing poverty, have never found a sustainable winning formula. The frustration at this inability has led to a grinding fatalism as reflected in dependency on the one hand, and a rejection of the basic reality of the hurtful and ineffective market policies under neoliberalism on the other. It is this frustration that feeds the l imited but increasingly popular solutions of grassroots autonomy from the state and the world capitalist system. On the other hand, while politicians continue to look for short-term solutions to immediate political problems, they still are operating within a certain ideological road map, one that allows them to configure national interest among the special domestic and foreign interests that constitute power in ways that make sense for the nation. This is why historical-ideological periods can be identified across the highly differentiated region. Framework 3, the general legitimacy of economic ideas, has increased in importance over time. Recentl y, public opinion has led to signifcant changes in policy throughout the region. In Chile, to human rights accountability for Pinochet. In Nicaragua, Costa Rica, and elsewhere, to corruption trials. In Bolivia, to the deposition of President Gonzalo Sanchez de Lozada. In Argentina, to a stronger negotiating stance by the government with the I MF and private foreign banks. In Brazil, to a strong rejection of an FTAA on the terms set out by the US. Neoliberalism has become associated in the region with inequity, volatile and unreliable growth, and a new form of global imperialism. All of these changes suggest that the neoliberal period may be in crisis - thus we are in the node of another potential historical shift in Latin American policy. We have seen in our survey that change is both possible and necessary. There is an evolutionary trajectory, ongoing policy experiments over time and space within a historical-ideological paradigm, and mini-crises that raise doubts and lead to further adjustments. If the neoliberal paradigm continues its tailspin, when the system is in crisis, a new set of ideas can push forward the shift to a new type of economic development policies. The problem so far is that, unlike the immediate aftermath of the Cuban Revolution, when socialism was seen as a viable alternative paradigm, no such alternative exists. Cuban-style socialism is rejected by the large majority of Latin Americans who want equity, but also growth and democracy. They recognize that neither the US nor the military would allow for a Cuban Revolution even if opinion somehow miraculously moved in that direction. What is needed therefore is an alternative framework, one that can inspire the current generation of policymakers just as the ideas of industrialization inspired lSI and market efficiency, neoliberalism. In this chapter, we have sought to extinguish the meta-paradigm of dependency and the popular version of Marxism that historical and imperial forces have and continue to doom LA to helpless poverty and that a revolution overturning everything is the only solution. We have seen that rather than helplessness, Latin American history has been f lled with a purposeful, if haphazard, search for locally-appropriate, economically effective, and pol itical ly supported answers to its unique versions of development problems. Within constraints, LA has been able to engage in a wide variety of policy experiments, ranging from nationalization to labor mobilization

Why Ideas and State Leadership are the Key Variables in Latin American Political Economy


to marketization. Fatalism is simply false and must be discarded. Ideas work within the constraints of interests in the short-run but the array of possible stable idea­ interest conf gurations is far wider than perceived at any given time. Thus, ideas, pragmatical ly applied, can and have changed the perceptions of interests and the direction of policy, as we have seen throughout LA's economic history. As many analysts have pointed out on both the Left and the Right, the state in LA has been the primary vehicle for oppression, inefficiency and lack of opportunitie s (Vargas L l osa 2005). Nonetheless, it is only the state, ultimately, that can resolve these problems. In the next chapter we need to point our lance at a different type of fatalism a faith among mainstream economists in practice, if not in quiet conversation, that markets solve all problems. What we see is that there is no such thing as a free market - state intervention is ubiquitous. Once we have extinguished the myth of a free market along with fatalism, on the "ash heap of history," we can begin to lay out an alternative blueprint of economic policies between the state (failed revolutionary central ism) and the market (failed neoliberalism ) for Latin American development that will achieve the best promised features of lSI, neoliberalism , and dependency theorists, that is simultaneous achievement of greater and more autonomous national development, efficiency of market outcomes, and greater equity. The main prerequisite for this change in fortunes is constructing a consensus around this idea, which is precisely what this book hopes to help to catalyze.

Chapter 2

Why an Active Industrial Policy is Needed for Development

N ow that we have demonstrated the central importance of the state as an actor in LA history and the abi lity of ideas to shape state policies in a variety of ways in response to the same constraints, dismissing the fatalism, helplessness and hopelessness of the dependency view, we must address the next red herring before getting to which ideas make the most sense for LA economic policy. Much ofthe debate among mainstream economists and dependistas [those who believe in dependency] concerns the optimal ity of the market; often this discussion takes on moral overtones. Mainstream economics is based on theories that show that markets are almost always more efficient and preferred to any state intervention, while many ofthe anti-globalization protests against free trade assert that markets reinforce inequality and unsustainable consumption. My contention in this chapter is that this question, like the question of the inevitabil ity of either "failure" or "progress" of L A , is misplaced. This chapter demonstrates that there is no such thing as a free market in the sense economists would have us believe. When economists debate whether it was markets or institutions that led to East Asian growth, they are missing the fact that not only EA but every Northern state interferes heavily in its markets. As Karl Polanyi pointed out in The Great Transformation, markets arc embedded in historical contexts and social understandings. However, this is not a book about sociology but economics. My objective in this chapter is therefore to show that state interference in market outcomes is ubiquitous, and that LA must therefore adopt an active industry policy (IP) as the center of its development efforts. There is no real situation in which the state does not have a hand in market outcomes. The question then will become what type of state intervention in markets makes the most sense for LA? Below we begin to catalogue the different ways in which states intervene in markets through industrial policy. We concentrate in this book on industry policy towards manufacturing here because it has been the main source of high-paying jobs and social mobi lity since the 1ndustrial Revolution. While services are acquiring increasing importance, there are very little statistical data or study of their role in growth and development. Still, the arguments we make here about promotion of manufacturing could just as easily be made for growing services, such as fnancial, software, and other professional services.l As I have shown e lsewhere in some detail, Northern countries understand very well how important key industries are for their I document the growing importance of services for developing countries particularly those in Asia. in the book Ron and Anil Hira. Outsourcing America, Amacom, 2005.

An East Asian Modelfor Latin A merican Success


economies (Hira 2003b, 4 1 -58). Make no mistake, no matter the consequences for economic theory, the US is not going to let its key manufacturing sectors simply disappear. The US and European governments, the main movers behind the push for neoliberal policies in the developing world, are supreme hypocrites. They protect, subsidize, and coddle industries. They play favorites and interfere in the market in a myriad of ways beyond the atrocious subsidies in agriculture, the purported comparative advantage of most developing countries.2 In a study of foreign direct investment on developing countries, mainstream economist Theodore Moran points out in detail that developing countries have to compete not only in market conditions with the built-in advantages of the North, but also with a wide range of government interferences, including subsidies, tax breaks, export financing, abuse of countervailing duties and anti-dumping procedures, and so on (Moran 2002, 7). In lieu of the trite claims of the sanctity of instant redistribution or macroeconom ic stability as the supreme priorities, a more complex, balanced, and helpful dialogue on development found in the industrial policy literature has been largel y forgotten. The industrial policy l iterature, l ike its sister literature on competitiveness, seems to have reached its peak in the l ate 1 980s in tune with the American recession, the rise of East Asia, and the general decline in the US's dominant postwar international position. The US boom in the 1 990s, centered around expectations based on the information technology revolution, extinguished this discussion. As we begin the new millennium, with the increasing failures of neoliberal policies to de l iver either stable growth or equity, this chapter demonstrates that it is high time to revisit the industrial policy arguments. The continuing increases in relative inequality, the bursting of the dot com bubble and American prosperity, and the stagnation of large parts ofthe developing world all call for an exploration of new economic paradigms, and this chapter shows one important avenue for making that conversation fruitful­ namely selective, limited, and purposefu l state regulation and shaping of markets for national outcomes. Brief Literature Review of Discussions about Industrial Policy and Development3

Political economists have long tried to argue with mainstream economists about the benefts of industrialization for development, and, more acutely, about the benefts of state-led industrialization. Unfortunately, these arguments have often been dismissed by economists because they go against the grain of mainstream theory and formal


See Oxfam's brilliant report, RiggedRules andDouble Standardl': trade, globalisation, and thefight against poverty, 2002, available at 3 A good general overview of the topic is found in Edward J. Malecki, Technology and Economic Development: The Dynamics of Local, Regional and National Competitiveness,

2nd ed., Essex, England: Longman, 1 997. I have tried to avoid reviewing the whole of the immense literature on industrial, innovation, and technology policy, and instead tried to cull out the important pol icy lessons for developing countries, with Latin America specifically in mind.

Why an Active Industrial Policy is Neededfor Development


methods which calls for free markets as the only method of achieving growth. This of course ignores basic history. As Change and Grabel state,4 Britain, for example, prospered at a time in history when it could (and did) colonize and/or dominate weaker nations, engage in slave trade, openly sell opium to China, and force young children to work 1 2-hour days under miserable working conditions. During its development, Britain also routinely violated IPRs (intellectual property rights) and marinated a law from 1 750 to 1 842 that banned exports of machinery to competitor economies. The economy of the USA benefited from very similar circumstances. Additionally. the USA benefited from its vast geographic scope (as the govemment was able to exterminate and/or forcibly relocate Native Americans), a large population of immigrant labour, and its exeeptionally rich endowment of natural resources.

Several seminal works will be highlighted here to introduce readers unfamiliar with the literature to some of the key ideas in regard to industrial policy and development. Perhaps the primary work on the need for state leadership in industrialization is that of Aleksander Gerschenkron, whose book, Economic Backwardness in Historical Perspective, has had a profound infl uence on development thinking (Gootenberg 200 I ). Gerschenkron pointed out through comparative historical case studies that over time, nations that successfully industrial ized, from Britain to the US to Germany to Japan and Russia, progressively used increasing levels of state intervention. This led to an important concept in the political economy literature, namely "late industrialization," implying the importance of the timing of when industrial ization (or even entry into a certain type of production) begins. In the 1 960s, Raul Prebisch'5 work on commodity prices vs. manufactures, as embodied by the Prebisch-Singer hypothesis, added the idea that industrialization was desirable because the prices of industrial products were both less volatile and increasing as compared to commodity prices over time.s Part of the reason is that as incomes rise ( Enge l 's law), a relatively smaller proportion is spent upon basic food products; another part is that over time the technological and skilled labor value of the production process increases while the raw material portion of production costs tend to decrease as a proportion of the final good's value. However, a great deal of skepticism continues to surround the relative prices of commodities, with institutions l ike the World Bank pointing out that commodity producing nations such as New Zealand can be prosperous. In more recent reports, the World Economic Forum and the Inter-American Development Bank, have acknowledged through econometric studies that the evidence points clearly to the conclusion that commodity-based economies grow more slowly than technology-based export ones.6 Moreover, the 4 Ha-Joon Chang and Ilene Grabel, Reclaiming Development: An Alternative Economic Policy Manual, New York: Zed, 2004, p. 44. See Chang's more impressive and eomprehensive argument in Kicking Away the Ladder: Development Strategy in Historical Perspective, London: Anthem Press, 2002. 5 Prebiseh's key work i s The Economic Development ofLatin America andIts Principal Problems. New York: UN, 1 950. 6 World Economic Forum, Global Competitiveness Report 2001�2002, New York: Oxford Uni versity Press, 2002, 46, and Inter-Ameriean Development Bank, Economic and social progress In Latin America. Competitivene,�.\·; The Business of Growth. Washi ngton :


A n Ea5t Asian Modeljor Latin American Success

rise ofthe OPEC petroleum cartel led to increasing hopes that commodity producers could organize themselves. Unfortunately, thesc hopes have never been realized into sustainable cartels in any other product, and the grandiose plans of for a New International Economic Order to improve returns to commodity producers through price setting, buffer stocks, and other deliberate mechanisms, seem to be a thing of the past. In line with the context of the Marshall Plan following World War II, in Latin America the argument was not over whetherthere should be industrial promotion, but what type should occur (Hira 1 998). Discussions included whether a "big push" of an entire production process, in a centralized location, should take place, or whether decentralized, regional, and piece-meal industrial development made more sense. In general, Latin American states opted for the big push idea, in part because of the inf uential ideas of "spillovers," or, as Albert Hirschman called them, "linkages." (Hirschman 1 96 1 ) The idea is that industries are linked a viable steel industry facil itates a viable auto industry, for example. This is related to the idea of the "value-added" chain, by which one seeks to capture more and more ofthe processes that add value to a product. For example, an auto industry facilitates the viability of national tire and parts sub-contractors. Spillover effects, similarly, are processes that are enabled from the development or big push in one industrial area. Developing engineers for oil ref ning could create a pool of talented (or human) capital that could lead the charge into petrochemicals. Radical extensions of Prebisch's insights led to a general skepticism not only towards commodities production, but also towards all kinds of dependent links with the developed world. Oftentimes, these led to nationalizations of foreign property andlor stringent conditions on multinational investment. In general, Prebisch 's later ideas that import substitution had to be replaced by the production of capital goods, regional integration and export-oriented industrialization were all but forgotten. Therefore, in practice the Latin American industrialization process was one of substituting only for fnished products, while intermediate input products continued to be imported and, by the 1 970s, fnanced by borrowing on the international capital markets. Thus, Latin America's external dependence was not signifcantly altered. Raymond Vernon's seminal article positing a product cycle added to political economists' arguments that small nations need state help to industrialize, and that high technology and industrial goods can lead to improved living standards (Vernon 1 966). Vernon's product cycle is a theory of product innovation, during which monopoly or oligopoly prof ts can be gained, followed by a stage of mass production, in which the labor-intensive and now standardized parts of the process are moved out to international sites with lower labor costs, followed by a general decline in demand for the product, at which point production sites are determined largely on the basis of labor costs and likely to be completed therefore wholly in developing countries. Vernon's product cycle suggests that production of most items should move over time via comparative advantage to lower cost countries, but the Johns Hopkins University Press for the IADB, 200 1 , 49. See also Sanjaya Lall, "The technological structure and performance of developing country exports, 1 985- 1 998, Oxford Development Studies. 2000 (28): 337-368, p. 3S0.

Why an Active Industrial Policy is Needed/or Development


there are two implications that must be considered before considering this bedrock theory of international economics. The first is that technological products, such as computers and autos, are constantly being re-invented into what are, in essence, new products, thereby renewing the product cycle of innovation. The second is that the productivity of developed country workers more than compensates for the lower labor costs ofworkers in the South, and that constant innovation leads to increasingly sophisticated production processes, requiring more highly skilled workers. That productivity advantage came initially from a combination of greater skills and capital equipment. As we see in the outsourcing phenomenon, the acquisition of skills is reducing such advantages to the point where countries such as India and China have targeted industrial policies to close the gaps in certain industries. A simple example of this is the computer industry. Begun in the US, prices have slowly come down as production has moved overseas to lower cost labor markets. Some labor-intensive programming has moved overseas to India as diaspora skills and an active Indian industrial policy have attracted those production components. Yet, new Windows programs, new semiconductor chips, and new products like the iPod still allow US manufacturers to retain control over market profts. Developing countries need to do more than just close the remaining gaps in terms of capital investment and skills. It is only when developing countries can begin to innovate and market their own latest versions of these products, as only Japan has done so far, that they will be able to gain the lion's share of the profits involved (and lead a new cycle of innovation for the next line). Moving into such a position of competing in international manufacturing markets never happened in LA. With the inflationary, balance of payments, and commodity price crises in LA from the late 1 9605, and spurred, ironically, by the OPEC oil price shocks, industrialization through import substitution was considered a failure, and, in l ine again with international trends, was gradually replaced by the neoliberal paradigm of liberalization. The late 1 980s-early 1 990s wave of East Asian l iterature revived the earlier interest in industrialization by attempting to make the case that East Asian states, following Gerschenkron's historical pattern, were successful in terms of growth and equity precisely because they intervened so deeply into industrial development. Stephan Haggard, in his famous Pathways from the Periphery, pointed out the important contrast between East Asian industrialization's export-orientation VS. Latin America's import substitution strategy (Haggard 1 990). The World Bank's rejoinder came in 1 993 with the publication, The East Asian Miracle, in which the institution attempted to refute much of the EA industrial policy literature.7 Regardless of the merits of the argument, which we examine in more 7 World Bank, The East Asian Miracle: Economic Growth and Public Policy, New York: Cambridge Universi ty Press, 1993. A solid rejoinder, among others and along with basic logic and knowledge ofthe region, is found in K.S. Jomo, ed., Manufacturing Competitiveness in Asia: How internationally competitive nationalfirms and industries developed in East Asia,

New York: RoutledgeCurzon, 2003. Marcel Timmer stakes out a middle ground in which he suggests that improvement in various factors, particularly labor productivity was key, but so were gove rn ment policies that directed improvements within selected industries. See The Dynamic.� ofAsian Manufacturing: A Comparative Per,�pective in the Late Twentieth Century, Northampton. MA: Edward E lg ar 2000, .

detail in Chapters 3 and 4, by the 1 980s, the argument was a moot point in much of the developing world, with the onset of the debt crisis, followed by the series of financial shocks and the ongoing recession in much of East Asia. Industrialization through protection or promotion is l argely considered passe and, indeed impossible given the constraints on dcveloping country budgets and the pressures to join free trade agreements in order to secure access to markets. Despite this overall veneer of consensus, behind thc scenes, states remain universally active in sectoral promotion policies, albeit in much more subtle and hidden ways than during the 1 960s. I n order t o finally extinguish the Wizard o f Oz-type illusion of free markets, w e tum in the rest of the chapter to the reasons behind having an industrial policy (or more generally government intervention in markets) and then tum to cataloguing the ways in which governments actually intervene. This effectively makes the mainstream economist view of "the two paradigm debate" that EA succeeded because of markets and macroeconomic stability alone irrelevant in the face of the fact of ubiquitous state intervention everywhere. We close the chapter with a brief discussion on a secondary challenge - refuting the idea that globalization renders industrial policy ineffective. Industrial Policy: Reasons Given fo r the Need for Government Intervention in Markets

Industrial policy has been understood in various ways. Michael Dietrich defines industrial policy "as long-run supply-side initiatives aimed at restructuring or promoting the activities of particular firms or sectors." (Dietrich 1 992, 1 7) Here are the most common reasons given for the need for government intervention in markets: 1.

Why an Active Industrial Policy is Neededfor Development

An East Asian Model for Latin American Success


Market Failure

The market failure school of thought posits that industrial policy is necessary because of key faults that occur in markets. These problems could include very high economies of scale that limit the level of competition; lack of incentives to create new technology; externalities created in the process of production; and high barriers to entry and exit for production. Government intervention must occur, in this sense, for the optimal outcomes to be obtained, that is the outcomes that would have been reached had the market been functioning adequately. For example, it may require some concerted effort for a government to promote research and development in an industry that does not yet exist, but for which there is a potential comparative advantage. Thus, the Brazilian Government's promotion of its pharmaceutical industry has, after many decades, led to viable generic drugs that are more affordable for the local population. A case can also be made that some production produces positive or negative externalities the common example being pollution (a negative one), whose full cost is not borne by the producer. In the case of industrial policy, we expect strong externalities for the region as a result of employment and income providing a tax base and a fiscal base for the quality of life. Similarly, a parallel


literature traces the lack of information o r information asymmetries a s part of the transactions cost l iterature. From this point of view, as Joseph Stiglitz pointed out, there may be inefficiencies in markets, such as the market for used cars, because of a lack of information by one or both parties, the buyer in this case. Risk can be considered a related type of category. Risk may not be a market fai lure in the sense that some worthwhile endeavors may take on the characteristics of public goods. Research on malaria, for example, could be a vital social task, but one in which the payoff may be too long-term and too uncertain for Northern pharmaceutical companies to adequately invest in it. If we consider the highly volatile and "shallow" nature of developing country financial markets, as well as the l imited size of the domestic markets and the lack of technological and marketing sophistication of some national companies, even more ordinary industrial endeavors will likely be out of the reach of Latin American entrepreneurs, particularly in smaller countries. The key mainstream rejoinder to the market failure school of thought would be foremost that the government is unable to overcome the market failure. Since there is no profit-related market discipline by government enterprise, the government will not create the atmosphere in which a viable, competitive, and efficient market producer can be created. Instead, the government should focus on creating the conditions, e.g. ensuring macroeconomic balance and viability of financial institutions that will allow for entrepreneurs to take on the risk (and the accompanying gains/losses) of any endeavor. If an enabling market atmosphere is created, and national businessmen still do not enter the industry, the industry may not be viable within the country itself. The government may become active in improving the human capital and infrastructure of a country (though even those are potentially better run by private firms or through government awards of sub-contracts) to attempt to attract foreign investors who will then bring in their expertise, capital, and technology to the task, and take on the risk. I f the government creates a viable investment climate, the population will gain not only from the employment and income created, but also from the spillover effects of the learning process. The government may also look at certain items as public or "club" goods, in which, similar to externalities, the costs and benefits are not equally shared. For example, research and development with long-term payoffs may be viewed in some cases as a public good. A breakthrough i n biotech could benefit an entire nation's pharmaceutical industry and the medical well-being of its population, but may be too costly and/or too risky for a private firm (particularly in a small developing economy) to undertake on its own. Thus, as Chang points out, the question is the balance between the socialization of risk for public benefit and the moral hazard problem by which risks are unnecessarily taken for private, rather than social gain (Chang 1 994, 78-9).



Late Industrialization/Infant Industry and Industrial/Sectoral Targeting

This argument, in line with the previous discussion, is that there are high start-up and learning curve features which serve as "barriers to entry" to viable production. These could include technology. managerial and other expertise, and even inadequate


regulation of markets. The question here again is why the government is better poised than the private sector to lead this in the form of state-owned enterprises, which in Latin America, generally have a very poor record of performance. This issue was partly answered in the East Asian literature by noting that the governments of Japan, South Korea, and Taiwan all worked with private domestic frms. In the case of protection, the question is which industries the government should choose to protect, for how long, and under what conditions. Again, the East Asian literature points out that stringent performance requirements, including export and local content requirements, were placed upon domestic companies, and that disciplinary measures, including phase-out or cut-off of protection and fnancial support were used to attempt to ensure performance. 3.

Why an Active Industrial Policy is Neededfor Development

An East Asian Modelfor Latin American Success

Technology/Product Cycle Management/Market Dynamics

As noted above in terms of the Vernon product cycle and the dependency view of the world, technology has long been considered part of the key difference in the economic structure of the North versus the South. H owever, the same types of market failure arguments as well as the overal l viability of a technological sector for a developing country have been put into question. When one considers the distance between the educational and research and development infrastructure of the US and Europe compared to the developing world, the task seems daunting. However, larger countries, including China, India, and Brazil have recently created viable leading edge products, such as software development and aerospace products that are internationally competitive. Even a smaller country such as Costa Rica has created a technology pact with large international software companies in an attempt to enter the information technology market. The real question, then, is which countries, under what conditions, in which markets is technology development possible, and, more importantly, how can it be accomplished? Peter Katzenstein's seminal work, Small States in World Markets, adds an interesting wrinkle to this issue (Katzenstein 1 985). Katzenstein points out that Western and Central European corporatist states also have little infuence on demand and supply conditions in international markets, yet have relatively stable growth and equity performance. He suggests that part of the reason for this performance is a highly adaptable and f exible industrial structure that is premised upon an extremely skilled workforce, and state cooperation with industry and labor to facilitate adjustment to improvement of products and new product lines. The problem with this approach is that Latin American states have not shown anything near the capacity required for such leadership. With high levels of corruption and class and group conf ict, establishing a European-style corporatism in Latin America must be viewed with great skepticism. On the other hand, market facilitation, and adjustment policies, such as retraining, are often paid lip service by most Northern countries. The perennial question here is how retraining can be accomplished- the track record of retraining long-time workers in downsizing industries (such as coal mining or steel production) is not very good, reflected by the high levels of protection for those industries internationalIy, including Northern economies (H ira and Hira 2005).



Spillovers/Military-Industrial Complex

An issue that is an important part of the industrial policy literature is the relative importance ofdeveloping military technology that can then be sold on the international weapons markets and have spillover effects in civilian production. Certainly the development of the internet by the Defense Advanced Research Project Agency (DARPA) of the US, and the close ties between US aerospace and defense contracts points to strong evidence that military spending can have very important effects on industrial structure, and has had particularly profound impacts on leading edge services, including information technology. The question in this area beyond those stated above would be whether in fact it makes sense to attempt to create military spillovers rather than simply funding/protecting an industrialization project directly, as in the case of Europe's Airbus. Brazil's successful development of Embraer, an international ly competitive aerospace industry, falls directly into this category. 5.

Employment, Declining Industries, and Education

Besides the fact that human capital, or a capable, well-fed, healthy, and wel I ­ educated population, i s needed t o create the viable conditions for development, as most mainstream economists acknowledge at the macro-level, there are important implications of industrial policy for employment as well. One ofthe most contentious sectors of multilateral discussion is certainly automobiles, precisely because production in that sector involves so many high paying jobs. It is no coincidence that the supposedly free-market US Government bailed out Chrysler in the 1 980s. In declining industries, the question of promotion and/or retraining becomes especially acute, as the country may view a particular sector as vital not only in economic, but also in political and cultural terms. The European Union has effectively stated this about its unviable (in many sub-sectors) agricultural sector. As noted elsewhere, the role the government can effectively play in economic adjustment either in creating or downsizing remains especially murky. For a developing country experiencing ongoing urbanization, population growth, and pressures on public services, as well as, in the case of Latin America, major internal factionalization, employment creation is both an economic and political necessity. The questions here are similar to those stated earlier- how can this be done, and when and under what conditions is protection worthwhile (rather than allowing the private market to work). The creation of human capital could not conceivably raise any opposition, however, the real question is what type of educational promotion is best for a particular country? The question is about degrees, institutional structures, and opportunity costs among different systems, rather than a simple prof con argument for human capital development. An additional complication is the probable complementarities among different services like education, health, and population growth. There are countries such as Argentina, with high literacy rates, but a poor economy, so the precise relationship between human capital and economic growth is not at a l l clear. On the other hand, as in the case of the military-industrial complex, there is strong evidence that there is an education-industrial complex. The US. Canada. and Europe all heavily subsidize higher education and university-based


An East Asian Model/or Latin American Success

research with the idea that it will benefit private industry. Besides large research grants, Northern countries also identify particular sectors for promoting re-training, such as having established or increased the funding of technical institutes and research centers once infonnation technology became considered a key industry. 6.

Creating Larger Firms (Economies ofScale) and Facilitating Exports

Given the relative size of international finns in key industrial and even service sectors, an argument can be made that domestic finns should be allowed a defacto oligopoly or monopoly within the domestic market so that it can reach the physical and marketing economies of scale to compete in world markets. The EA literature even suggests that Japan, among others, protects its domestic markets so that domestic consumers can effectively subsidize national companies' exports. Similarly, through JETRO (Japan External Trade Organization), the Japanese Government facilitates the targeting of possible export markets and the entry of national companies into them. While all developed countries have well-honed export facilitation and subsidization institutions, including trade infonnation services and export financing, the extent and types of promotion vary considerably by country, sector, and time period. Moreover, it is quite unclear the extent to which the agencies are really effective and efficient at finding market opportunities that private sector companies could have found themselves. The efficiency of public expenditures argument also comes into play here. In a developing country context, one can make the argument that the domestic economy is probably too small in most of the countries to create a national champion on the basis of the domestic market. More importantly, there is the question of why domestic consumers should pay a collective price for the benefit of the few workers and managers lucky enough to work for a subsidized company. Hence part of the key to an IP argument in a developing setting, as we shall see in our examination of EA IP, is that the end goal must be to create an internationally competitive industry, that is, one that can provide the same or better goods and services to domestic consumers at the same or better price as foreign fnns. For taxpayers, the benefit of this investment is that the creation of employment, incomes, and export earnings from the industry provides a strong fiscal base, lower crime, etc. On the other hand, this also requires the conscious national development of champions, embracing a domestic oligopoly, so that national finns can reach the adequate production, fnancial, and technological economies of scale in order to be able to compete in international markets. The key question, then, is how to provide protection domestically while pushing discipline for competition internationally. 7.

Regional Development

A constant feature of industrial policy, discussed by a vast literature, is the question about using promotion and protection to develop poorer regions within a country. Undoubtedly, the government can develop infrastructure in rural regions to facilitate market access, but economists will raise the question ofthe viability of such schemes. As in the case of adjusting workers and regions in declining industries, schemes to diversify backwards areas, such as Canada's promotion ofnew industries and services

Why an Active Industrial Policy is Needed/or Development


in over-fished Newfoundland, generally have a poor track record. Economists point out that migration and urbanization takes place because the labor force reacts to market signals- both in terms of economic opportunity and quality of life. The question of the many subsidizing the few is also brought up in this context, and the sustainability of supporting losing enterprises rather than simply bearing the shorter­ tenn costs of adjustment is a serious issue. On the other hand, poverty and related social problems are often concentrated in particular regions. Moreover, especially in the developing world, living conditions have become almost unbearable in the large cities, both in tenns of pollution and ability to carry out nonnal business. The strength and concentration of the urban population always raises the question about whether developing governments actually subsidize those urban centers, which was certainly the case in Latin America during the Import Substituting Industrialization (lSI) period. The responses to such arguments do not have to be made purely in terms ofequity considerations. The fact is that oftentimes poorer regions lack in both physical and social infrastructure. Hence strategic investments in these areas can be viewed as a type of facilitation, allowing the market to work as industry moves to areas with lower labor costs. The Arguments Against IP: Is the Medicine Worse than the Symptoms?

The arguments against industrial policy are in line with those against all fonns of government intervention in the economy (most publications from the World Bank and IMF echo this line, see bibliography for examples). They include the attack on the idea that the government has any ability to "pick winners," in tenns of emerging/viable industries and sectors; the likelihood that politically, rather than economical ly, optimal decisions will be made, leading to corruption, cronyism, inefficiency, and poor service, affecting other economic sectors and leading to a net drain (or suboptimal return) on economic resources; the probability that recipients of government aid will not only be tied in politically, but use political avenues as means to sustain such support, leading to a parasitic, rather than independent, viable, and competitive private sector, i.e. "infants that never grow up;" the stifling of foreign direct investment, technology transfer, managerial expertise, and global marketing and retail/distribution chains; the stifling of entrepreneurship and efficient domestic finns; and, most fundamentally, the loss to domestic consumers of the benefits of market competition. M oreover, critics caution that industrial policies would simply invite reciprocal retaliation by other countries. Like most economic theories, these arguments may hold some water in a perfect world of no intervention by any party, but that is not the world we live in. States always intervene in markets and always will, for the reasons noted above, particularly that perfect competition does not exist in most markets that provide employment and high incomes. Since states by definition are political and notj ust economic creatures, they will and should attempt to shape market outcomes for their national interests. The real question, then, is given the myriad of possibilities, some of which may be incompatible, which types of intervention make the most sense for a developing country? In the rest of this chapter, we lay out the possible types of intervention and

An East Asian Modelfor Latin A merican Success


then in the next chapter argue why an EA arrangement makes the most sense for LA.

What Tools are Used in Industrial Policy?

Based on the factors above that make almost all markets imperfect in some way, "Strategic Trade and Investment Policies" is the term often used to describe the many different avenues by which countries seek to promote national competitiveness.s There are many definitions and measurements of competitiveness. The most well­ known is Michael Porter's famous diamond model, with the four interrelated factors of firm strategy, structure, and rivalry; demand conditions; factor conditions; and related and supporting industries, all of which are indirectly affected by government policies and chance (Porter 1 990). Our focus here will be on this indirect link . between government and national competitiveness. We can summanze the ways industrial policy is carried out to achieve national competitiveness according to the following categories: Anti-trust and Labor and Environmental Codes

An often overlooked aspect ofi ndustrial policy deeply engrained in Western tradition is the use of anti-trust and labor protection codes to change markets. A nti-trust is used to break up dominant companies, as with the recent cases ofAT&T and M icrosoft in : order to introduce competition. It is interesting to note that perhaps the most stunnmg omission of economists, in their promotion of liberalization in order to increase domestic competition through imports, is that they ignore that dominance can also occur on an international level (thus there is little foreign competition to create a competitive market). Thus we have long-standing companies controlling significant market shares in key global markets over long periods of time. Similarly, we now take for granted basic labor protections, such as worker safety and chil� labor, �s wel l as environmental curbs on market behavior, though they do not effectively eXIst in much of the world on whom we rely for part of our goods. While not the focus of this book, it bears repeating again, the fact that in a global economy, there is little coordination in these areas, by contrast with international macroeconomic issues, such as exchange rates.

Why an Active Industrial Policy is Neededfor Development


company of Brazil. I n a developing country scenario, SOEs were often set u p to lead the national plan for industrialization. Defining exactly what an SOE is can be quite tricky, as there are many examples of mixed public-private ownership as well. A country entering a new market, with all the problems we discussed in the frst section, may wel l consider an SOE to be a more reliable vehicle than a private sector firm, especially where foreign competition i s already well-established. Protection/rom Competition

In terms of protection, again a wide variety of instruments are available, including quotas, tariffs, differential treatment of imports through regulation, differential treatment offoreign investment, ownership, and personnel, government procurement rules, and a wide variety of non-tariff barriers. Protection can be viewed in some ways as promotion of domestic firms. Sectoral Promotion, including Subs idies

Sectoral promotion is arguably much more key than state-owned enterprises because of the myriad of methods by which it takes place. Van Beers and de Moor suggest five different kinds of public subsidies. The first is on-budget, and includes direct payments or grants to a party. Off-budget subsidies account for the other types. The first category of these is tax policies, including credits, exemptions, deductions, rate relief, holidays, and preferential treatment. The second is public provision below cost. I n this category, we could include infrastructure provision and complementary services, and research and development grants and purchases. The third category would be capital cost subsidies, incl uding all fonns of fnancial incentives. Examples of financial subsidies are preferential loans, liability guarantees, debt forgiveness, and other forn1s of providing capital at below market cost and above market access. Lastly, they cite subsidies through the market mechanism. These include domestic-oriented policies, such as price regulation, quotas, and procurement policies; and trade-oriented measures, such as those listed above under protection (van Beers and de Moor 200 1 , 5). To these suggestions we can add government efforts to aid in marketing or distribution; and even subsidization offactors of production, including infrastructure. I fwe consider the latter avenue of sectoral promotion, the advantages built in by the North over the South are egregious and outside of the scope of trade agreements.

State-Owned Enterprises (SOE8)

Regional Growth A reas/Clusters Based on Flexible Specialization Strategy

There is a strong consensus among economists that state-owned enterprises tend to be less efficient than their private counterparts, based in part upon the difference in incentives and s upposedly levels of competition. Despite this, numerous examples of successful SOEs exist throughout the world, such as Petrobras, the petroleum

An ongoing discussion i n the industrial policy literature questions whether IP should be focused on the national or regional level, similar to the question of whether the government should follow a broad-based TP for growth (for example facilitating


See Jeffrey A. Hart and Aseem Prakash for an overview of the strategic trade policy

l iterature, "Globalization, governance, and strategic trade and investment policies," pp.

243-6 in Prakash and Hart. Globalization and Governance. New York: Taylor and Francis, 1 999.


An East Asian Modelfor Latin American Success

markets to operate efficiently, such as providing infrastructure and macroeconomic stability) or targeting specif c sectors.9 On a regional level, government procurement and other methods of favoritism can help to develop the "research triangles," or clusters of similar activities which are highlighted in P iore and Sabel's famous book The Second Industrial Divide (Pi ore and Sabel 1 984). Piore and Sabel point out that in some cases, such as Silicon Valley or Milan, there seem to be "horizontal" economies of scale in terms of companies in the same industry benefiting from regional concentration, such as the sharing of ideas and a pool of human capital. "Flexible specialization" is the term used to describe the supposed change from mass production, based on lowest cost economies of scale and transactions costs to just-in-time, specialized and flexible production based on quality and niche marketing to customers' specifications. Flexibility and customization are the key words for the supposed new paradigm. Regional clusters are always supported by local and national government support, such as local universities that produce a steady stream of labor, or government subsidization of infrastructure for research and development.1O Some authors in the I iterature prefer to see these as network, rather than geographically-based clusters (see De la Mothe and Link 2002). Regulatory Policy as a Source ofIndustrial Policy

In my book, Political Economy ofEnergy in the Southern Cone, and in my chapter "Regulatory Games States Play," I argued that regulation could be used as a form of industrial policy. Regulation sets up the rules for how a market operates. 1 t can set the level of competition, the types of standards, the level of entry of foreign companies, and the general red tape, among a myriad of other factors, all of which affect how a market operates. However, most analysts would think of industrial policy as working through more traditional instruments of protection and promotion. Investment in Factors ofProduction and Ways to Reduce Transactions Costs

As noted in the various sections above, the levels of military and direct educational subsidization that feed into research and development of products on the private market are vastly superior in the North, leading to a vicious cycle of increasing gaps in competitiveness. These would include the vastly superior investments in education and health care that give human capital, on the whole, a decided edge in the North. We could extend this line of thinking to include investments that reduce transactions costs, such as Singapore's setting up of broadband transmission throughout its territory. Again, while mainstream economists, such as Theodore Schultz and Robert Barro, along with mainstream institutions such as the World 9 For an overview of these debates, see Malecki. Some authors claim that there are now supra-national innovations systems. See Robert Anderson, Theodore Cohn, Chad Day, Michael Howlett, and Catherine Murray, eds., Innovation Systems in a Global Context: The North A merican Context, Montreal: McGill-Queen's University Press, 1 998. 10 For an overview of this literature. see Rod B. McNaughton and Milford B. Green, Global Competition and Local Networks, Burlington, VT: Ashgate 2002. .

Why an Active industrial Policy is Neededfor Development


Bank, often call for investments in human capital to create growth, they fall far short of human capital for an industrial policy. In fact, one could argue that mainstream economics effectively prevents human capital investments through the clampdown on fiscal budgets created by structural adjustment policies designed above to pay back external debt. The fact that countries with high levels of human capital, such as Argentina, India, Sri Lanka, Egypt, and Russia have struggled for many years with both growth and equity, show that investments in human capital alone are inadequate. How does IP Need to be Adjusted for a Global Economy?

Thus far, we have explained the reasons for and the tools of IP, with emphasis on developing country settings, demonstrating en route that intervention in markets is ubiquitous and the wide variety of forms it takes. There is an important and growing vast l iterature on how globalization affects late industrialization in the developing world, effectively limiting the possibilities and outcomes of an active J P. H owever this literature contains all kinds of explanations, some of which seem to have empirical backing and some of which are more speculative. In general. we can see that there are two different views, one focusing on constraints and the other on possibilities for learning, for the possible outcomes of developing country I P in a globalizing economy. Global Commodi�y Chains

One slice of the literature on globalization looks at the emergence of global commodity chains which limit independent industrialization in the developing world. In line with much of the writing about the development of global enterprises, the commodity chain literature focuses on the human networks and localized economic effects created around the international production of a particular good. Most of the commodity chain writers take a critical perspective on such operations; in a dependency-type fashion they see a system of transformation of raw materials and unskilled labor in the periphery as important inputs into Western consumer products. The global commodity chain l iterature also seeks to explain the development of the semi-periphery countries such as South Korea, where semi-skilled labor yields higher local returns, but an overall dependency upon foreign capital, technology, and export markets continues. Gary Gereffi points out that commodity chains can be producer or buyer driven. Producer-driven chains, such as automobiles, aircraft, and machinery, are organized around a multinational corporation's production system. The key decisions are to minimize the costs of production of the different components and various stages of production. By contrast, buyer-driven chains are labor-intensive goods such as c lothing, footwear, toys, and electronics, in which larger retailers use international contracting to minimize the costs of production (without actually producing the goods themselves) (Gereffi 1 994, 95-1 22). In producer-driven chains, companies such as Ford not only control the retail and distribution network, but also the technology, managerial expertise, and capital to produce the good. In buyer-driven chains, firms such as Nike more often rely upon


An East Asian Modelfor Latin A merican Success

branding as well as their access to retail chains owned by others. Besides brand, their main power comes from their design and marketing know-how. The global commodity chains l iterature is quite valuable not only in terms of its sectoral studies,l I but also because it looks at sectors to answer the question- to whom do the benefts of value-added steps of the production process accumulate? This l iterature also does well in pointing out the complex and subtle differences among national strategies for industrialization. Thus the literature points out that upward mobility is possible, but the individual characteristics of industrialization process for example the nature of the global commodity chains of which national production is part - is vital to understanding the overall well-being of the economy. In this sense, through this l iterature, an overall critical stance of globalization i s favorably watered down by the complexities o f real situations that acknowledge that mobility in the industrial ladder and world economy can still take place. The Role of Multinational Corporations

Multinational corporations (MNCs) have been at the center of controversy about development from at least the 1 960s.12 The dependency argument sees MNCs as agents of exploitation. MNCs use their strength in capital, technology, expertise, and access to outside markets to push out local competition, purchase local raw materials and labor at exploitative prices, engage in destructive environmental practices, and avoid taxes and support corrupt e lites, including governments, with no concern for the welfare of their own population. On the other side, MNCs are supposed to open the door for developing eountries to acquire access to markets, to current technology, to world-class training, and to capital and managerial know-how that are unavailable to local entrepreneurs. They are supposed to spur the development, technology, and competitiveness of local firms. One of the most important aspects of MNCs revolves around technology transfer. In practice, it i s quite difficult to separate out MNCs from local developments of technology. A common example occurs when technology has been adapted to the needs of local markets, or when an MNC uses its dominant position to buy out or otherwise usurp local entrepreneurs or subcontractors' breakthroughs. The question about where MNCs place resources has led to a growing literature on decision-making within MNCS.13 Most of this literature is concerned with how 1 1 Teresa Shuk-Ching Poon, Competition and Cooperation in Taiwan s Information Technology Industry: Inter-Firm Netlvorks and Industrial Upgrading, Westport: Quorum, 2002, is an interesting study that combines GCC with the leaming framework discussed

Why an A ctive Industrial Policy is Neededfor Development


firms decide to make foreign investment decisions. The most common analytical framework in the literature is Dunning's Ownership, Location, and Internalization (OLl), also known as Dunning's "eclectic" paradigm. Basically, ownership refers to some specific advantages that the frm possesses, which allow it to overcome the costs of competing in a foreign market. These are often referred to as "core competencies," such as proprietary technology or managerial, production, or retail know-how. Location refers to the advantages of the particular host country, such as cheap labor or tax or other benefits. Internalization, which is viewed as a subsequent consideration, refers to the fact that a l arge multinational enterprise can internalize transactions costs of operating in the market, that is, money is saved by not having to contract out services and bargain a price with an external agent, who has their own overhead, etc. Institutional instability, exchange rate risks, and unreliability of sub-contractors could all be factors in frm decisions on whether to purchase a local firm, set up contractual relationships with local companies, develop its own subsidiary and production facilities, or otherwise develop an optimal strategy for its invcstment. These factors are considered along with strategic interaction with other firms and time lags from strategic changes in the market. So, which view about MNCs is correct? Whi le the literature does contain some insights into investment decision-making, there are still major empirical gaps that prevent us from answering this question. For example, most economists acknowledge that the theory and data behind intra-industry and intra-firm trade is quite lacking.14 We simply do not know much about the magnitUde or nature of intra-MNC or intra­ commodity chain transfers. There are a growing number of case studies, but these tend to be concentrated in particular areas of interest, such as automobiles and texti les. The bottom line is that much of this information is private to the finn and so inaccessible to researchers. Moreover, there are few well-developed empirical studies of technology transfer, and even fewer by particular industry (Cantwell and Molero 2003). In this sense, foreign investment should be seen as a bargain between states and M NCs. The nature and terms of the bargain will vary greatly by sector, time, and the resources that each side promises to bring to the table. There is an additional type of bargaining that takes place- that between government, business, and consumers. Dunning calls "alliance capitalism" the need for cooperation among these different partners (Dunning 1 997). For example, governments need tax revenues; corporations need skilled workers; and consumers want good jobs. All three have a stake in creating a situation where a healthy corporate sector can grow on the national level, and adequate attention is paid to investments in human capital and research and development, leading to a virtuous circle of growth. As we have

below onto Taiwan's information technology industry.

1 2 More recent critical treatments along these lines include: the seminal David C. Korten, When Corporations Rule the World, Bloomf eld, CT: Kumarian Press, 200 1 ; and specifically in regard to LA: Michael Edwards and John Gaventa, eds., Global Citizen Action, Boulder: Lynne Rienner, 200 1 , and Frederick Stirton Weaver, Latin America in the World Economy: Mercantile Colonialism to Global Capitalism, Boulder: Westview, 2000. 1 3 For examples, see Thomas L. Brewer, Stephen Young, and S tephen Guisinger, The New Economic Analysis ofMullinational.v: An Agendafor Management. Policy and Research, Northampton. MA: Edward Elgar. 2003; Yadong Luo. Multinational Enterprises in Emerging

Markets, Copenhagen: Copenhagen Business School Press, 2002; William N. Cooke, ed., Multinational Companies and Global Human Resource Strategies, Westport, CT: Quorum, 2003. 14 See Sven W. Amdt and Henryk Kierzkowski, Fragmentation: New Production Patterns in the World Economy, New York: Oxford University Press, 200 I ; and p. J. Lloyd and Hyun-H oon Lee, eds., Frontiers ofResearch In Intra-Industry Trade, New York: Palgrave Macmi llan. 2002.


An East Asian Modelfor Latin American Success

seen throughout the book, dichotomous views of good and bad or right and wrong simply don't match the complex reality, which is always mixed. Learning and Governing the Market

Alice Amsden's Asia s Next Giant, on the rapid industrialization of South Korea, typifes the l iterature on learning and governing markets for development. 15 Amsden asserts that late industrializing states face extremely difficult barriers of entry due to timing (well-established f rms already in the market who are already in the product cycle). The costs of creating an infrastructure for the continual project innovation that is needed in world markets, is an extremely expensive and difficult undertaking. A developing country will have a hard time competing with the education and research and development institutions as well as the well-developed and large private sector organizations working in a particular sector. As a result, Amsden suggests, late industrializers are more likely to compete on innovations in process rather than in product. That is, she examines the continual efforts to improve the productivity of South Korean manufacturing as the essential component for arriving closer to world market competitiveness. Amsden's studies of South Korea pointed out that late industrializing nations could enter new industrial markets in part on the basis of competing through constant efforts to reduce the costs of production, rather than simply through product innovation. She also created the term "dynamic comparative advantage," a direct affront to mainstream economists in stating that comparative advantage in some products could actually be created. Amsden's later study of Taiwanese high technology industries builds upon her earlier work (Amsden and Chu 2003). She points to the importance of rapid adjustment; the necessity of national ownership to capture the benefits of research and development and to encourage entrepreneurship; and the goal of creating outward foreign direct investment which can take advantage of specializing the different segments of production. Like Robert Wade, whose seminal book, Governing the Market, demonstrated that the dichotomy between markets and state-led economies is a false one, Amsden believes that one of the keys to success is the dense network of personal relationships between the state and the private sector (Wade 1 990). For Wade, governing the market means that a developing government should engage in "market-friendly" management of the domestic economy. The upshot is that attention must be paid to macroeconomic stability at the same time as creating selective interference on the microeconomic level. This interference is part of long-term planning to move up the value-added chain and capture market shares for higher revenue and wage-producing goods. Selected tools might include conditionalities on foreign investment, such as export and training requirements; subsidized credit to leading f rms; government-led marketing intelligence efforts; and heavy investment in human capital to create a 1 5 Alice H. Amsden, Asia s Next Giant: South Korea and Late Industrialization, New York: Oxford University Press, 1 989. See also Amsden, The Rise of "the Rest ": Challenges to the Westfrom Late-Industrializing Economies, New York: Oxford University Press, 200 1 , and Michael Storper. Stavros B. Thomadakis and Lena J . Tsipourit, Latecomers in the Global Economy. New York: Routledge. 1 998.

Why an Active Industrial Policy is Neededfor Development


skilled and ready domestic labor pool. Such interventions should be tempered with performance requirements, which are provided in part through selective and time­ limited competition for government aid, and through the discipline of having to compete on the international market through exports. Factor Markets as Parts o/National innovation Systems

Clearly, factor markets play a key role in terms of national competitiveness. Lall suggests that there are three basic factor markets in regard to internationally competitive industries: human capital; financial capital; and information, specif cally access to foreign technology for transfer (La11 200 1 , 23, 3 1 -33). Obviously these are factors that can be created, rather than either the resource "destiny" or "comparative advantage" that both the dependency school and neoliberals claim locks developing countries into a certain economic profle. However, countries with ready access to capital and high literacy rates, such as oil-rich Venezuela, have also had inconsistent economic growth. Countries with heavy foreign investment and multinational activity, such as Mexico, have not clearly increased their growth rate as a result. More importantly, in none of these cases has the country in question been able to create a nationally-based competitiveness with Northern countries. Therefore, the question is not simply investment in factor markets, but setting up institutions by which those markets can be applied fruitfully to actual innovation and technology problems. The literature on innovation reveals that there is no clear theory on how innovation occurs, or what exactly stimulates it. Some of the literature suggests networks of entrepreneurs; strong intel lectual and institutional property rights; evolutionary product cycles; and Schumpeterian values for entrepreneurship. 16 However, much of the literature suggests that technological development can be stimulated through national innovation systems - systems of higher education, research and development funding for inventors and entrepreneurs, and public institutions dedicated to this task (Nelson 1 993). These national innovation systems should recognize that it is applied science, "design" that leads to product innovation (Kline and Rosenberg 1 986, 275-305). Trust and social capital are words used to describe the necessary social pre-requisites for healthy cooperation both within a national innovation system and among international partners, generally firms, cooperating on research and development. 17 1 6 For examples, see HariolfGrupp, Foundations ofthe Economics ofInnovation: Theory, Measurement and Practice, Northampton, MA: Edward Elgar, 1 998; Andrew Hargadon, How Breakthroughs Happen: The Surprising Truth About How Companies Innovate Boston: Harvard Business School Press, 2003; Andrew H. Van de Ven, Douglas E. Polley, Raghu Garud, and Sankaran Venkataraman, The Innovation Journey, New York: Oxford University Press, 1 999; and Jon Sundbo and Lars Fuglsang, eds., Innovation as Strategic RefleXivity, New York: Routledge, 2002. 1 7 John H. Dunning and Gavin Boyd, eds., Alliance Capitalism and Corporate Management: Entrepreneurial Cooperation in Knowledge Baved Economies, Northampton, MA: Edward Elgar, 2003; David O. Faulkner and Mark de Rond, eds., Cooperative Strategy: Economic. Business. and Organlzatlonal l.lwue.l', New York: Oxford University Press, 2000.

An East Asian Modelfor Latin American Success


National innovation systems are often thought of as triangular- involving a partnership among government, industry and education. Technology is considered to be of particular interest as it is presumed to have multiple spill-over effects. I S I n other words, technology development is considered akin to a public good, in the sense that the costs of development may be high up front, but the benefits are spread across a wide range of companies and consumers. Clearly, there is a strong connection between defense spending and technological development, for example the aerospace industry developed around military needs and continues to rely upon military contracts for viability. Corporate research and development, concentrated in fields such as biotechnology, relies heavily upon intel lectual property rights for patent protection and developed through public subsidies to university researchers. Moreover, for national innovation systems to be successful, one needs a strong partnership between the public and private sector. Technological universities and institutes must be funded to develop researchers, and for long-term and public-goods type basic research. The educational infrastructure must have social network links to the private sector in order to develop viable applications. Given these infrastructure requirements, as well as the head start and the larger market size that developed economies have, it is no surprise that until recently developing countries have struggled to develop their own technology. The literature points increasingly to "strategic alliances" by different MNCs, and suggests that the wave of merger and acquisition (M&A) activity of the late 1 980s and early 1 990s was a reflection of Northern governments' recognition that the economies of scale for innovation had increased, though this probably underplays many of the financial motivations of the wave of M&A (for example see Narula 2002). Given the levels of expenditures required for b iotechnology, aerospace, and information technology innovation, and the rapid change of technological level in leading sectors, it is not surprising that national innovation systems seem to be melding with strategic MNC alliances. Therefore, MNCs are seen as a key steppingstone to international competitiveness in terms of technology transfer. Geographical Clusters or "Growth Poles "

I n terms of regional development, the literature also points out the importance of regional c lustering and inter-firm networks for sectoral success stories. 1 9 One need only think about Detroit, Silicon Valley, and Milan, to recognize how common the concentration of rival firms in the same area seems to be. The literature i s not really clear on what gives rise to such clusters - much of it seems to be historical accident 18

An overview of the economic literature on technology and growth is found in Albert

N. Link and Donald S. Siegel, Routledge,


Technological Change and Economic Performance, New York:


Reviews of this literature are found in Peter Brown and Rod B. McNaughton,

3-37 in 2002; Anna Grandori, ed., Interfirm Networks: Organization and Industrial Competitiveness, New York: Routledge, 1 999; and John H. Dunning, ed., Regions. Globalization. and the Knowledge-Based Economy, New York: Oxford University Press, 2000. "Global Competitiveness and Local Networks: A Review of the Literature," pp. McNaughton and Green,

Why an A ctive Industrial Policy is Neededfor Development


- or how or when clusters could be pro-actively created. Paniccia offers various suggestions. The first is that clusters could be "autopoietic," or chaotic systems that are self-organizing. The second are push factors, including l ocal expertise, entrepreneurship, or extended-family kinship networks who have an interest in local development of a particular production process. M ore recently, we see deli berate government policies designed to create such clusters through pro-active policies, as suggested here. Pull factors can also lead to geographical clusters, where partnerships between firms and sub-contractors; between firms and universities; and between firms and resources andlor customers can take advantage of proximity and inter-personal cooperation. The nature of the product may require distinct craftsman-like (highly specialized) interrelated stages of development which lends itself to a clustering of small skilled enterprises (Paniccia 2002, especially 33, 1 49-5 1 ). Once a large firm or set of firms is created, the human capital and infrastructure create a spill-over effect which lowers the costs of entry to rival firms, as was obviously the case in Silicon Valley with a huge number of start-ups in the 1 9905. Clustering could also lead to negative externalities, such as congestion; increases in factor costs, particularly labor and real estate; and a tendency towards groupthink. These factors also seem to be present in Sil icon Valley, where costs for land and programmers skyrocketed during the boom. The literature points out the important "enabling" effects that governance at a variety of levels can have in terms of aiding regional clusters, such as predictable fiscal policies; investment in human capital ; and creating top-notch and tailored infrastructure. The literature emphasizes the importance of shared values, trust, and informal ties that allow for the successful operation of networks. It points out that there are successful historical cases of government promotion leading to industrial clusters (Paniccia 2002, 1 52). Thus, clustering brings into question the premise of globalization of highly mobile supply chains. The truth seems to be that both modularization of production chains and clusters explain aspects of the global economy. Conclusion: IP in a Context a/ Globalization

The question of how globalization affects I P mirrors the pessimism and optimism of the dependency/neoliberal duality. Some authors see severe constraints which limit the possibilities for independent industrialization, while others see the possibilities for learning. Like all changes in structure, we can resolve this debate in the same way as we addressed the other in the preface both are simultaneously correct. Changes in structure from globalization create both new constraints and opportunities for developing countries. Having shown in the beginning of this chapter that IP is inevitable, we came to the question of how to create the best possible constellation of policies that will raise the living standards of the largest number of people as high and quickly as possible. We did not find any clear answers in the existing l iterature, however a number of suggestions worth investigating were reviewed. For example, our review of the l iterature suggests that some sectors are more "sticky" than others in the sense that they are not so easily moved from one place to another. I feel the best way to answer the question of "what works" is to look at how EA has succeeded in entering these difficult to move sectors.


Why an Active Industrial Policy is Neededfor Development

A n East Asian Modelfor Latin American Success

Why therefore the East Asian Miracle "Debate" is a False One

The departure point of the rest of this book is that there was indeed an East Asian miracle of economic growth though much of the post World War I I period, which we demonstrate in the statistical section of the next chapter. We do not deem it worthwhile, given our arguments above to devote exhaustive space in this book to "prove" that IP was truly responsible for the East Asian miracle. On one side, the World Bank launched the first counter-attack to the foundational literature of the developmentalist state with its influential study, The East Asian Miracle (World Bank 1 993). The World Bank's contribution seems to contain mixed messages, in some parts acknowledging the effectiveness of IP (in selected cases), but overall arguing that IP were "largely ineffective (p. 3 1 2)." The World Bank's volume has become the backbone of the "second generation" of reform espoused now by mainstream economists and international financial institutions, which diagnoses the problem of development as centered on problems of making markets work efficiently through institutional reform, which for the most part has completely failed in LA. Much of this literature actually sees IP as an impediment to growth in EA, which according to this view, has grown primarily by getting macroeconomic fundamentals "right."20 Since the end of the Asian financial crisis at the end of the 1 9908, this view seems to have acquired dominance. Not only are there questions about East Asia's ability to continue its IP policies, but also whether the financial crisis is forcing a fundamental restructuring of EA economies towards a US model (Hira 2006, Tan 1 999). East Asia i s not seen as an I P model for these mainstream institutions, but rather as a reinforcement that macroeconomic fundamentals, and perhaps some secondary attention to human capital investment and institutions for markets are the true lessons for Latin America (Birdsall and Jasperson 1 997). Interestingly, this view of the optimality of moving towards liberalization is sometimes echoed by those on the left who see globalization as offering no other choice.21 We leave such debates, like whether free trade is truly optimal, to economists to argue over.22 From the point of view of many political economists, such debates are not worthy of serious investment of time and effort for several reasons. As we have seen, states always intervene in markets. More importantly, there is no control case of macroeconomic stability without heavy IP in East Asia. We do have the case of macroeconomic star Chile, but it does not have a serious manufacturing 20 For examples, see Joseph E. Stiglitz and Shahid Yusuf, eds., Rethinking the East Asian Miracle, Washington: World Bank and New York: Oxford University Press, 200 1 , and M. G. Quibria, Growth and Poverty: Lessons from the East Asian Miracle Revisited, ADB Institute Research Paper 33, Tokyo: Asian Development Bank Institute, Feb. 2002. 2 1 These views are often expressed from a world systems view. For example, see Satoski Ikeda, The TrifUrcating Miracle: Corporations, Workers, Bureaucrats, and the Erosion of Japan :S National Economy, New York: Routledge, 2002. 22 An exciting contribution in this vein will be found in Richard G. Lipsey, Kenneth I Carlaw and Clifford Bekar, Economic Transformations: General Purpose Technologies and Long Term Economic Growth, Oxford University Press forth com ing 2005. See also Dani Rodrik's work, such as "Rethinking Growth Policies in the Developing World," Luca D' Agliano Lecture. Torino. Italy, Oct. 11. 2004.


base. However, Chile did have a strong and pro-active I P promotion of new export industries. Still, as we detail in the next chapter, Chile's success has not been on the same level as that of EA, particularly in terms of equity. Nor does "free trade" exist anywhere - it is an abstract principle, just as perfectly competitive markets are. All trade agreements are subject to negotiations and concessions on a sectoral level in practice. Indeed most trade agreements are set up around the mechanism of allowing joining parties to set up exceptions to the principles laid out, and trade negotiations are ful l of conditionalities and opt out and safeguard clauses. In essence, free trade agreements have the effect of "locking in" winners who had a head start in entering global industries.


We have laid out in this chapter the reasons for IP and the ways in which IP takes place every day everywhere in order to promote the idea that an active IP for LA is a must. We have also demonstrated that the real argument must be what type of IP makes sense in a developing country setting. This situation works under the radar of a disingenuous discussion about the failure of I P, when as we have demonstrated the reality is that, by their actions, no states rely solely upon the international institutions' prescription to developing countries of macroeconomic discipline and free markets as the avenue to prosperity. This prescription is, in a sense, a manipulation of these institutions, reflecting the dom inance of the North, to preserve the advantages of early timing and its own I P interventions. Regardless of theoretical discussions of optimality, we see across the board that sectors from automobiles to computers are all subject to huge and ongoing types of government meddling. It i s hard to understand the psychology of the mainstream economist who cannot see these basic facts. Perhaps he/she is so enamored with the beauty of economic theory that the economist is unable to see what those who simply open their eyes can. Or perhaps there is a strong level of groupthink going on, with the immersion in formal models preventing them from seeing the clear fai lure of neoliberalism throughout the region. The bottom line is that no honest discussion about how development can take place until the facts about IP are brought to the forefront. One ()f the foremost facts that is completely ignored by economists is that macroeconomic stability and an active IP are completely compatible. In the next two chapters we sketch out the particular

institutional arrangements behind these harmonious relationships in EA. We now turn to explaining this difference in outcomes as a difference in their brand of IP.

Chapter 3

Industrial Policy Lessons for Latin America from East Asia


While we have shown that state intervention occurs in every economy, in LA, any pretense of coherent and deliberate state leadership or vision towards economic development has been dropped under the current paradigm of neoliberalism. Neoliberalism, including economic l iberalization and privatization is under heavy fre in Latin America. The recent elections of Chavez in Venezuela, Lula in Brazil, Correa in Ecuador, Kirchner in Argentina, and Morales in Bolivia indicate a strong level of unhappiness with the low rates of growth and growing inequality in the region. While a return to the heavy protectionism ofearlier decades seems impossible, particularly given the pressures of the external debt load of many Latin American countries, the role of the state remains a subject of implicit debate in the region. It is important to point out that every East and Southeast Asian country has had a unique historical traj ectory. There have been differences in ethnicity and historical development, colonizers, in the types of security threats, the timing of different economic policies, the existence and development of natural and human resources, and in national politics among many other things. This chapter argues that despite all of these huge differences, we can discern a common approach to development that has led across-the-board to superior results in terms of both growth and equity. The problem with the formal or large number of cases approach such as those taken by many economists is that it does not make sense in the context of a discussion about general development strategy, where there are considerably more factors at play than the number ofcases. Following the last chapter, we find that we simply cannot separate the macro from the micro-level variables. In this sense, we cannot simply isolate investments in education, as a number of growth models do, and conclude from that a positive correlation explains growth. A particular correlation can contribute to an explanation, but cannot be separated from the myriad of other variables that overlap (are endogenous and interactive) in the cases. Nor can we can simply increase the number of cases - East and Southeast Asian countries began their industrial expansions in unique country - and international-specific historical contexts. There are only a few East Asian countries, and only a few Latin American countries who are constantly changing as the context changes, with generally incomparable and limited datasets truncated by major political and economic shocks. What we can accomplish here is to use the comparative method to show that, despite major diferences among East Asian economies, including China, a succes::.ful formula of IP relationships can be identified over time. EA, facing the


Industrial Policy Lessons for Latin America from East Asia

An East Asian Modelfor Latin American Success


The questions we now tum to are what are the dynamic and pro-active elements of EA industrial policy that could give guidance to future LA efforts to change their position in the value-added chain of global production? The literature provides no clear answers to what seems to be a relatively straightforward question. The frst set of literature on the topic, which is quite limited, explicitly compares the economic development of the two regions. Most of the literature simply provides snapshots of the development of different countries in each region, with no real cross-analytical synthesizing comparison; thus both conclusions and recommendations on what LA could learn are largely absent (for example, see Lin 1 989). Stephan Haggard's Pathwaysfrom the Periphery and Gary Gereffi and Donald Wyman's Manufacturing

Miracles paved the way in the early 1 990s for a serious look at industrial policy in Latin America from a comparative point of view. Michael Shafer's 1 994 Winners and Losers: How Sectors Shape the Developmental Prospects ofStates is a relatively overlooked but pioneering contribution in pointing out the vital importance of different economic sectors to development outcomes. Shafer's book was rather sketchy, though, in terms of accounting for important nuances, such as intra-sectoral differences. Jeffry Frieden's 1 99 1 overview of Latin American political economy also contained some important analysis of sectoral attributes, following in the footsteps of Peter Evans' important conceptual work and domestic political economy traditions of the work of Helen M ilner, Robert Rogowski, and Peter Gourevitch. Sylvia Maxf eld and Ben Ross Schneider's edited 1 997 volume has some important discussions about state-business relationships from a comparative perspective.2 The main problem with these books is that they tend to be heavy on general political economy comparisons, but quite sketchy on the detaiIs of sectors. Haggard has some later work on the political economy of adjustment that does look at domestic actors, but this is more conceptual than empirical. In fact, only Shafer looks at specif c countries and specif c sectors in them, but his rubric is highly limited to just two general aspects ofthe sector. As he himself admits, his is an important but preliminary first step in sectoral analysis. A second but more limited area of the literature is specific case studies of sectors. Terry Karl's 1 997 book, The Paradox ofPlenty: Oil Booms and Petro-States, may be the best known of these, with its focus on Venezuelan petroleum, but there are other important and more complex treatments as well. Peter Kingstone, Sarah Schoonmaker, and Roy Nelson, among others, have conducted important case studies of particular Latin American sectors. Unfortunately, these books reveal the problems of sectors in adequate complexity, but more from a post hoc point of view. They uncover in sometimes great detail the myriad of problems from industry lobbying, foreign investors' dominance, and lack of state capacity, but do not suggest how these aspects could be aligned for more favorable results. A third set of literature would include the more technical literature by sectoral experts at institutions such as UNIDO (United Nations Industrial Development Organization) and the World Bank (eg. Sercovich). This literature includes the seminal World Bank report The East Asian Miracle and the 1 997 Asian Development Bank Report, Emerging Asia. This report suggests that macroeconomic stability, rather than state intervention, was responsible for high growth rates in EA. Furthermore, it reaches the unsupported conclusion that government intervention actually had more negative than positive effects. As we saw in the previous chapter, claiming that free markets were responsible for EA growth simply does not square with the fact of state intervention everywhere.3 Chang points out several fundamental flaws in the report's analysis. F irst, it does not adequately consider the role of protective industrial

F or examples, see Sung Gul Hong, The Political Economy ofIndustrial Policy in East Asia: The Semiconductor Industry in Taiwan and South Korea, Northampton, MA: Edward Elgar, 1 997; John A. Mathews and Dong-Sung Cho, Tiger Technology: The Creation of a Semiconductor Industry in East Asia, New York: Cambridge University Press, 2000; and Jomo K.S., et aI., eds, Southeast Asia �' Misunderstood Miracle: Industrial Policy and Economic Development in Thai/and, Malaysia and Indonesia, Boulder, co: Westview, 1 997.

See bibliography for full citations. Sanjaya Lall points to a variety of substantive and methodological errors. See "'The East Asian Miracle' Study: Does the Bell Toll for Industrial Strategy," pp. 1 07-123 in Lall, Learningfrom the Asian Tigers: Studies in Technology and Industrial Policy, New York: St. Martin's Press, 1 996.

same constraints of dependency on finance, technology, and a limited number and type of export markets, provides the most comparable (and only) successful case for other developing countries. The case of China is particularly important to show that the EA blueprint still works in a globalized economy, which is why we address it in a separate chapter. The similarities in the way that the Asian economies are set up provides a basic blueprint for rearranging Latin American economies towards innovation, growth, equity, and competitiveness in world markets. In this sense, we do not see EA as a "miracle," rather as a general set of guidelines for development strategy that can be followed elsewhere. A small and largely-ignored literature (at least by mainstream economists) is growing to more closely study the IP components that have succeeded in EA, sometimes with indirect applications to LA.l This book seeks to reinforce the foundation of these types of studies as a more productive direction for Latin American political economy research in the coming years. To continue our thread from the last chapter, which concluded that there must be an active industrial policy, the conclusion of this chapter will be to begin to address what type of industrial policy makes sense (Rodrik 2004). There are a number of reasons why a move to an activist state a la EA is usually dismissed out of hand in LA, for reasons that we address in depth in this chapter. We must begin the discussion with a more detailed look at the history of industrial policy in LA, which helps us to understand the context in which an EA model is consistently dismissed as a possibility for the region. Then we can begin to address these objections, an exercise that we continue throughout the book. We then tum to an objective comparison of whether and why the nature and quality of the raising of standards ofliving in EA have changed. Building upon the basis of our conclusions in the last chapter, that state intervention is universal, we see in this chapter that it was not state intervention that failed in LA, but the particular type of state intervention. Literature Comparing EA and LA Comes Up Short on Proposals

2 3


An East Asian Modelfor Latin American Success

policies both in giving some greater insulation to frms struggling with new products and a shortage of f nancing, and in promoting a firm's ability to engage in structural change, that is to undergo wholesale changes in the human, physical, and relational capitaL Second, the report does not adequately examine spill-over effects on linked industries, and cross-sectoral learning effects. Third, the report does not consider dynamic and incremental learning effects of change and i mprovement over time, including improvement in institutions (Chang 2003a, 1 8-40). We can add, as we demonstrate in the fnal chapter, that it is the long-term learning process ofthe private sector that must be protected on a graduated basis in order for late industrializers to have a chance to enter into new sectors, even ones where they theoretically have a comparative advantage. As Malaysian economist K.S. Jomo puts it, in calling the World Bank report "erroneous," (Jomo 200 1 , 1 6- 1 7): the role and contribution of industrial policy instruments in the development of the three second-tier Southeast Asian NICS, especially since 1 970, is undeniable. The role of governments in promoting industrialization beyond what would have been possible and likely without intervention is suggested by the contrasts between the late colonial economies of Malaysia and Indonesia and these national economies today. Despite all the f aws and abuses involved, there is now little doubt that the structural transformation and industrialization of these economies would not have been achieved by exclusive reliance on market forces and private sector initiatives . . . Though the consequences of state intervention . . . have been mixed, this is largely because much of the responsible state intervention has been motivated by considerations other than accelerating late industrialization.

Industrial Policy Lessonsfhr Latin America/j'om East Asia


and have a more neoliberal-compatible nature to them (Peres 2002, 8 1 -20 1 ). We can point out severe limitations in regard to an active IP role of the Latin American state. First, the clear priority is macroeconomic stability above all, including equity and microeconomic policies, such as labor and IP. Second, industrial promotion takes current comparative advantage more seriously, and tends to focus on existing industries/resources in terms of their international export potential. There is a sense of growing competition for access to developed markets (dismissing internal development as a possible avenue), reflected in the growing interest in free trade agreements and integration, as well as courting foreign investment. Third, current IP shies strongly away from starting new state-owned enterprises (SOEs), favoring instead creating stable investment c limates and indirect aid to industries. Indeed, many large SOEs have been privatized in the region, with heavy foreign ownership taking over. Fourth, in line with this fact, the role of foreign companies is not necessarily viewed as clearly separable from the development of national industry. As in the case of Mexican maquiladoras, Latin American governments often see foreign companies as necessary partners, bringing capital, technical expertise, and international marketing and distribution networks. Does this new approach towards I P make sense? A review of EA IP indicates otherwise in terms of long-term growth and equity. Before we get to a detailed assessment of the two strategies, we must continue to climb the mountain of implausibility that an active I P a la EA could be adopted in LA. Why did I P Succeed in E A vs. LA?: Dispensing with the Easy Objections

Since Miracle, a series of technical studies have emerged from international institutions. These studies usually exhibit a bias against industrial policy on the one hand, and an inability to discuss politics on the other. They are important fountains of analysis and information, but they do not provide any comprehensive vision or planning. Nor is there any explicit analysis ofEA to understand why industrial policy succeeded there but not in LA. Conclusion

Because of the external debt load and the change in ideas, a ful l blown East Asian industrial policy never caught on in Latin America. Indeed, a wave ofprivatization has swept through Latin America, changing the whole context for economic policy.4 As I have discussed elsewhere, the claims that emerging policies labeled as " neostructuralism" represent a new and strong deviation from neoliberalism, in emphasizing a combination of stable macroeconomic policies, more active social welfare policies, and export promotion (Hira 1 998), are highly dubitable. There is some evidence that Latin American countries have not completely abandoned IP. As Wilson Peres points out, most IP efforts now fall under the guise of"competitiveness," Luigi Manzetti, Privatization: South American Slyle. New York : Oxford University also Judith A. Teichman, The Politic.v ofFreeing Markets In Latin America: Chile. Argentina. and Mexico. Chapel H i l l : University af N orth Carolina Press. 200 1 . 4

Press, 1 999. See

There are a number of reasons why an EA model is dismissed as impossible for LA. We have so far addressed the two most general ones the first being that LA is helplessly constrained in its dependency condition and lack of relative power vis it vis external powers and internal elites (Chapter 1 ), and the second that free markets are really the only answer for LA's problems (Chapter 2). In this section, and through the rest of the book, we begin to systematically examine additional objections and demonstrate why, while not without substance, they are not true obstacles for such a paradigm shift in the region. The Democracy Debate

A central and frequent objection of LA observers i s that any replication ofEA would come with the same authoritarianism, particularly with the case of China. This i s another of many possible debates, including class, ethnicity, effects of different colonizers, and so on, which this book does not consider integral to the argument made here. As in the previous discussion of the EA Miracle, r do not see any clear causality between IP and authoritarian structures. The examples of the previous chapter, such as Airbus and the US's support of the semiconductor and aerospace industries, are from democratic states. Japan has a long-standing democracy in the region, as well as providing the blueprint for EA intervention. Moreover, democracy has developed considerably in EA during the period of lP, with South Korea, Taiwan,


A n East Asian Model/or Latin A merican Success

Malaysia, and Indonesia all making important strides towards stable democratic rule, despite the Communist threat and civil strife we discuss below. On the other hand, China and Vietnam have developed their own version ofthe EA models, while continuing authoritarian structures. My conclusion from this differentiation is that there is no clear relationship between political regime type and IP. This conclusion is echoed in much ofthe political economy of development literature about economic growth and regime type (for example, see Haggard and Kaufman 1 992). To be sure, there must be certain aspects of state leadership and certain types of institutional arrangements, as we discuss throughout the book. Yet, there is no reason whatsoever why a state cannot receive support for, and be ultimately accountable to, a population for the results of its IP through democratic institutions. Western European countries such as Switzerland and the Netherlands do so all the time. The real question is not about democracy, then, but establishing a democratic consensus for a sensible set of policies that will lead to growth and equity; the patience to wait for their positive results; and the virtuous cycle of political support and continued equitable growth that will ensue. Thirty years ago no one expected Chileans would reach a consensus in support of neoliberal policies introduced by a military regime, especially after the democratic transition (Hira 1 998, and Hira and Sanghera 2004). In the same way, my faith is in the power of ideas to move democracies in LA towards a similar consensus towards an activist IP. Historical Differences

Whereas in EA, some argue, the US overlooked industrial policy in the hopes of containing communism and creating stable states that could pay for military defense, in LA, the US was more concerned about expropriation of its direct investments and access to m ineral resources. Part of these sets of stories is that LA is cursed by dysfunctional legacies from the Spaniards, whose colonialism was much worse than thatofJapan or China. While it is true that Japan did set up some colonial infrastructure in Taiwan and South Korea geared towards secondary industrial production, anyone remotely familiar with either Chinese and/or Japanese colonialism would chafe at the idea that such historical experience was benefcent. The Japanese colonial system was set up explicitly to favor the home colony, not to spur indigenous development, entrepreneurship, or participation in decision-making.s Chang and Grabel point out that Korea '8 literacy rate at the end of Japanese colonialism in 1 945 "was only 22 percent . . . not much better than that of many African countries when they emerged from colonialism. By contrast, Argentina's literacy rate was over 90 percent."(Chang and Grabel 2004, 42) Moreover, one can hardly see any sign of nascent national industrialization in Taiwan or S. Korea in the aftermath of World War II, let alone in 5 Thomas B. Gold, State and Society in the Taiwan Miracle, Annonk: ME Sharpe, 1 986, and Meredith Woo-Cummings, "The Political Economy of Growth in East Asia: A Perspective on the State, Market, and Ideology," pp. 323-341 in Masahiko Aoki, Hyung-Ki Kim, and Masahiro Okuno-Fujiwara, eds., The Role a/ Government in East Asian Economic Development: Comparative Institutional AnalYSis, Oxford: Clarendon Press, 1 997, pp. 32930.

Industrial Policy Lessons/or Latin A mericafrom East Asia


Southeast Asia, which was brutalized by Japanese occupation. Their growth began much later. B ruce Cumings, the foremost English-language historian of Northeast Asia, points out that Japan's attraction to Korea was based upon mineral deposits and rich land that could be used to develop soybeans (Cumings 1 990, 1 4 1 -56). He states it best: The Japanese left Korea with many of the trappings of the modem economy when they withdrew in 1 945 . . . a relatively advanced transportation network, with links to the world market system . . . (a) population with modem skills . . . Yet the fundamental fact to remember i s that Japanese-fostered changes . . . did not run their course. Peasants were not torn from the land, destroyed as a class, and integrated into industry; instead they were introduced to industry and then spewed back upon Korean vill ages. Landlords did not tum into commercial and industrial entrepreneurs . . . Korean bureaucrats were not integrated into the career patterns of modern, or rational-legal bureaucracy so much as used as ethnic interlopers in a foreign administration, setting them against their Japanese superiors and their own Korean brethren. Even in the Korean resistance, both nationalist and communist, fissures developed, setting one faction against another with effects that long outlasted the colony itself Finally, the abrupt denouement of the war shattered . . . the incipient transnational unit carved out of northeast Asia. The end of the war divided the Korean nation itself . . . In the process, railroad, road, and sea networks were severed, trailing off into wilderness or emptied of purpose. The Japanese defeat thus destroyed an integral and delicate structure that had linked the metropole to the periphery. The southern half of Korea was left with a disabled economy and chaotic politics that could only fester . . . (Cumings 1 98 1 , 66-67)

One would also have to wonder, in turn, why other Japanese colonies, such as the Philippines and BurmalMyanmar (as well as China) did not reap similar benefts. One would also then have to explain why the Spanish economy has succeeded despite the Iberian heritage, and why there are individual cases of success in LA, such as Embraer and Codelco, where an active industrial pol icy creates their success. C learly, the choice of strategy, as demonstrated by Southeast Asian and now Chinese growth, taking place decades after colonization, matters much more than any questionable benefts of having an Asian colonial master. On the fatalistic side, the argument is often made that EA was "favored" by US foreign policies for development. This argument usually goes something like this: the US wanted EA to industrialize, so that EA, particularly South Korea and Taiwan, could pay for their own defense efforts against communism amidst the Korean and then the Vietnamese conflicts. This argument has more substantive validity, as key to its plausibility is the shift towards export orientation in East Asia in the 1 960s, while Latin America remained stuck on import substituting industrialization ( lSI). The move in East Asia to primary Export-Oriented Industrialization (Eo!, which was the promotion of the production of consumer non-durable items for export) in the early 1 960s, in contrast, was a sea-change in economic orientation. A lthough both Korea and Taiwan recognized the diminishing gains from primary lSI, unlike Latin America, they moved to exporting rather than a higher stage of l S I . This change was associated with the security environment of the 1 950s and 1 9608, when first the Korean and then the Vietnam Wars raged on. Such a scenario necessitated a strongly centralized mil itary command and perhaps a stronger sense of unity among

Industrial Policy Lessonslor Latin Americafrom East Asia

An East Asian Modellor Latin American Success


East Asian elites. It also meant, more fundamentally, that, lacking ready natural resources in Korea or Taiwan, exporting was necessary in order to fund defense efforts. Moreover, Korea, Taiwan, and Japan benefited from the supplies to the defense efforts of the US (particularly during the Korean and Vietnamese wars) and US military aid.6 As can be seen in the following table, all of the East Asian and Southeast Asian countries received more aid than LA counterparts, and received proportionately more military than economic aid. Table


US economic and military assistance and defense expenditures abroad,

1 950-73

U S Economic Aid


US Defense

Argent. Brazil Mexico Indones. Japan Korea Phillip. Taiwan Thailand Turkey Latin Am E. Asia l isted

237.55 25 19.4 285. 1 5 1 758.7 27 12.7 5655.3 1 62 1 .5 2362.4 7 15 . 1 281 9.8 9972.5 14825.7

1 77.4 485 . 1 14.4 1 77.3 1 239.7 621 4.4 741 .3 404 1 . 1 1 3045 42725 1 662.9 1 3 7 1 8.3

nla nla nJa nla 741 0 1 727 1 152 439 956 680 1010 1 1684

Total 4 14.95 3,004.50 299.55 1,936.00 1 1 ,362.40 1 3,596.70 3,514.80 6,842.50 2,975.60 7,772.30 1 2,645.40 40,228.00

In US$ millions current Sources: US Agency for International Development, US Department of Commerce Figures


net and subcontracting of defense procurement is not available, which would significantly raise

amounts received by East Asian countries during the Korean and Vietnam Wars.

There is no evidence that the U S promoted EOI with the idea that Korea and Taiwan would become industrial powerhouses one day; such a contention is nonsensical given the conditions both countries faced in the immediate aftermath of World War II. As Cumings states in regard to Korea, Ameriean planners (in 1 948) did not worry about deelining industries. Their vision was the Smithean, internationalist creed of a world of interdependent free trade, encouraging the export of American plants and technology . . . The core thinking for the whole of the new policy first appeared in an August 3 1 NSC 48 draft . . . (which stated) 'General industrialization in individual countries could be aehieved only at a high cost as a result of sacrifcing production in f elds of comparative advantage' . . . certain parts of the world are


'natural sources o f supply o f strategic commodities and other basic materials,' giving the United States 'a special opportunity for ' (Cumings 1 990, 1 72-3)

Taiwan and South Korea's tum to EOl was inspired not by prescient economic thinking but rather by the United States' Agency for International Development and by the US-dominated International Monetary Fund's push for self-reliance in terms of f nancing the war effort. The change in policy was also lubricated by these organizations' promises of aid to accompany the reforms. In fact, in both countries, economic planners were anything but optimistic about the success of the change in strategy, and were only proven wrong by the ensuing economic growth (Cheng 1 990, 1 44). The security threat therefore helped to solidify these nations ' economic consensus towards export promotion. By contrast, commodity exports, such as wheat and coffee, enabled Latin America to continue funding lSI (Ranis 1 990, 229). The commodity boom which spared Latin America from the economic growth crisis in East Asia in the early 1 9608 lasted until the late 1 970s. During the 1 970s, Latin America supplanted its foreign exchange earnings with borrowing petrodol lars in the private market, which assured reasonable growth rates and politically-adequate social spending. In the early 1 980s, not only did commodity prices crash, but interest rates were cranked up as the United States severely tightened its monetary policy. The loss of foreign exchange earnings led to a period of stagnant or declining growth rates in Latin America. This external shock has led to the most recent shift in economic policy towards including primary and secondary EOl (the latter being the promotion of consumer durables, and industrial and capital goods production for export). However, LA's version of EOr is missing other key components of the EA strategy as outlined below. As such, while in many cases basic consumer product exports were initiated much earlier than 1 980 in Latin America, these exports occurred on top of or along with primarily lSI policies at the time (Villarreal 1 990, 3 1 0). In EA, meanwhile, Korea and Taiwan followed Japan's early footsteps towards secondary EOI. Both Korea and Taiwan faced heightened restrictions on their textile exports to the United States. Both signed an agreement "voluntarily" decreasing exports in 1 97 1 with the United States, as well as the Multi-Fiber Agreements in 1 973 (Cheng 1 990, 1 62). The move also m ade political sense: the export sector continued to be favored. While the success of EOI in East Asia did bring about new oppositional coalitions, such as the neglected agricultural sectors in both countries, neither regime faced a strong enough challenge, partly thanks to land reform under US occupation, to consider modification of economic policies (Cheng 1 990, 1 65-7 1 ). East Asia's economic success of the 1 960s reinforced the EOI policies in place and helped them to spread elsewhere in the region (Cheng 1 990, 1 56). Southeast Asia7, like Korea, Taiwan, and Japan, benefited from supplying US defense efforts in the Vietnamese War and from US military aid. Most of the countries in this region gained their first policy independence due to the breakup of the colonial system in the late 1 950s and early 1 960s. Their initial economic policies stressed commodity exports, such as petroleum ( Indonesia and Malaysia) and agricultural production (e.g. rice in Thailand). They soon moved to a period of easy import substitution by

I am currently working on an extended version of this analysis for a future article on

the differences in aid between the two regions.



Note: Refers to Thailand. Indonesia. Malaysia and Philippines.


An East Asian Modelfor Latin American Success

the mid- to late 1 960s and early 1 9708 (Lee and Naya 1 988). The flare-up of the Vietnam War can be considered akin to the external catalyst that the Korean War was for EastAsia. United States military aid increased massively to Southeast Asia, as did purchases of local material for the war efforts (See Table 3 . 1 above). Beginning in the late 1 970s, many ofthe Southeast Asian countries undertook the policy change to EOr. This timing was also related to the external shock ofthe rise in oil prices. While Indonesia and Malaysia were major oil exporters, in the 1 970s, their oil resources began to decline. The reaction in each case was a tum towards labor-intensive exports in order to make up for the foreign exchange earnings. These countries adopted primary EOl and foreign investment policies which have encouraged both US and Japanese investors to utilize them in global production networks (Rosario 1 99 1 , 57). Unlike East Asia, they have not benefited from extensive land reform, yet, their record on growth and equity is superior in general to that of LA. In sum, the choice ofEor policies is also undoubtedly a result ofthe learning curve by which Southeast Asian countries have attempted to imitate the successes of East Asia. In the next chapter, we demonstrate that, with distinct historical circumstances, China has also been able to adapt the blueprint for its own development, with tremendous results. The only appropriate conclusion, then, is that EA policies, including Japan, were initially the result of an alignment between domestic forces and external pressures, and subsequently the result of conscious choices. South Korea changed its policies only after Park took over in the early 1 960s. Taiwan's miracle stumbled along initially before gaining momentum in the late 1 960s. In the same way, the Southeast Asian countries transformed their commodity-based economies into export-industrializing ones through choice, not US fat. The relations between Malaysia and the US, for example, were consistently strained under Mahathir. When LA turned to lSI, the reasons included internal political pressures and ideology, as well as the United States' role. LA militaries and the prevailing ideas at the time, supported by the World Bank, Marshall Plan, and the lessons oflndia and the Soviet Union's Five Year Plans, were to focus on rapid internally-based industrialization (Rira 1 998). One of the foremost differences between the US roles in East Asia and Latin America is the greater US ownership of Latin American industries. Another difference is Latin America's much greater reliance on foreign, and particularly US, borrowing, in general, for development than East Asia (Coatsworth 1 993, 1 04). Ironically enough, l S I in Latin America, instead of creating greater independence from the First World, increased Latin America's dependence, in different ways. Now, instead of relying on primary commodity exports, Latin America relied on imports of capital goods, raw materials, intermediate inputs, and foreign technology (Jenkins 1 977, 3-8). Not surprisingly, it is this same greater degree of external dependence which opened Latin America up to external pressures to move to the neoliberal-inspired Export­ Led Industrialization in the 1 9805, a policy by then recommended by the United States in order to ensure orderly debt repayments (Coatsworth 1 993, 1 68-9). These point to internal LA weaknesses as well as the lack of consensus, as we discuss through much of the rest of the book. It is important, moreover, to extinguish the odd but pervasive Leftist argument that the US allowed some development for its own selfish purposes or that in many other ways, the EA miracle was based on some form of good fortune in terms of US policy

Industrial Policy Lessons for Latin America from East Asia


alignment. This, of course, relies on the erroneous assumption of US omnipotence (the ability to allow), which has been proven wrong by history from Vietnam to Iraq now. Other facts reveal the ridiculously exaggerated nature ofthis conspiratorial argument. First of all, South Korea was devastated by the Korean War. It was, in the words of contemporary assessments at the end of the civil war, "a basket case," a country with few natural resources whose families had been forcefully divided up. Taiwan was not much better with the fleeing Guomindang (KMT) party from Mainland China serving as an occupation force upon an unwelcoming native population, with the main intent being the retaking of China some day. Therefore, it seems the US is being given credit by the Left for being able to create development where it does not otherwise exist. Given the US's track record in developing its allies where its own strategic interests dictated so, including South Vietnam, the Philippines, Cuba before the Revolution, N icaragua under Somoza, Iran, and present-day Afghanistan and Iraq, such an assumption is widely inaccurate and indeed puts these critics in the same fog of miscalculation as some of the overly exuberant analysts of US foreign policy on the right. Second, first the Korean War then the Vietnam War had profound effects not only on the home countries but also upon the long-tenn stability of the entire region. The various refugee crises, followed by genocidal policies in Cambodia, and the regional arms race are hardly the ideal context for economic growth. On the contrary, one would have put the chances for democratic and equitable growth emerging from such a scenario as highly improbable at best Third, LA did not somehow receive different treatment in the sense of US foreign pol icy goals. In the context of the Cold War, American troops were active in the Korean peninsula as well as Vietnam, in the Dominican Republic as well as Honduras. In this sense, the priorities for the US in LA were not any different than anywhere else to fight the Cold War by any means possible, making alliances with anyone on the same side. The support of various authoritarian figures in EA, including M arcos in the Philippines, parallels the support of similar fgures in LA, including Somoza in Nicaragua. Though it is important to note the proximity of aid from the Soviet Union and China into Southeast Asia, external factors were not the deciding factors in the development traj ectories of the two regions. The same pressures from the Soviet Union and China did not l ead to development in Myanamar, Cambodia, or anywhere in Central Asia. In short, without exonerating the US for its obvious and predictable self-interested motivations, with the usual short-sighted implementations, the facts demonstrate the false fatalism of those who exaggerate the role of external forces generally, and the US in particular, in development choices. This leads us to the next red herring, which is the argument that ENs miracle is due to a homogeneous and highly consensual population. Differences in Handling Internal Dissension, Including Ethnic Strife

Obviously, given the diversity of populations in China, Indonesia, Taiwan, and Malaysia, it makes no sense to sustain a causal argument for culture or other homogeneity, as may be the case for Korea and Japan, as a key source of growth for East Asia. To a greater extent than LA, EA has been highly contested and bitterly conflicted along internal l ines. Whi le not diminishing the internal strife of LA during


An East Asian Model/or Latin American Success

its history, the conficts did not reach anywhere the same levels as those in South Korea, Malaysia, Singapore, Vietnam, or continuing in the current day in China, Indonesia, and Cambodia. In contrast, in the ongoing conflicts between indigenous groups and fairer elites of LA, there was far less parity between the two sides, as well as a spectrum of identities that reduce the dialectical self-identification of each side. What is also different in the Cold War conflicts was the way that the internal dynamics in the different societies played themselves out. Whereas the South Korean, Taiwanese, and Malaysian elites overcame civil war, internal dissension, and fragmentation to develop a sense of national unity through shared economic growth, LA societies divided even further along class lines, with internal groups taking up different lines along the resources provided by Cold War allies. The Latin American Left miscalculated in thinking another Cuban Revolution was both easy and imminent, including Che's blunder in Bolivia, and the total failure of armed struggle, with the late and redacted exception of success in Nicaragua in 1 979.8 The response to these challenges in LA by the elite victors has hardly been beneficent, rather a military repression has been followed by labor repression and other ways of reducing social safety nets. By contrast, the governments of Singapore and Malaysia, in response to civil war, and the KMT in Taiwan set up affirmative action type programs favoring their native populations. The case of Malaysia is a good example of the differences in state approaches between the two regions. Malaysia consists ethnically of a Chinese minority, comprising 34 percent of the population in 1 980, which has been economically dominant in Malaysia, while the Malays, constituting 55 percent of the population, largely rural smallholders, have been the poorest group. East Indians make up only 1 1 percent of the population, and so have not been as significant in the Malaysian political economy (Nash 1 988, 6). The main "axis" of conf ict between ethnic groups is refected in striking differences between Chinese and Malay assets, income, and education levels. The British created an ethnically divided Malaysia with a corresponding "racial division of labor," by importing Chinese and Indian workers and restricting Malays (Schlossstein 1 99 1 , 224) from the lucrative rubber trade, but setting up a Malay administrative elite (Stenson 1 980, 4-7). With Japanese occupation of Malaysia between 1 942-5, open ethnic conf ict between the Chinese and Malaysians began. The Chinese, who were loyal to their native country were executed by the Japanese and attacked by their Malay lackeys, working ostensibly as policemen. The Malays, in tum, benefited from Japanese rule through exclusive access to administrative jobs in the government. In 1 946, a national alliance, with British encouragement, was formed among the United Malays National Organization (UMNO), the Malaysian Chinese Association (MCA), and the Malaysian Indian Congress (MIC) , which became known as "the Alliance" (Schlossstein 1 99 1 , 225-6). The 1 969 parliamentary 8 The Left within LA seems perfectly able to acknowledge these mistakes. Politicians from Lula to Lagos to Kirchner all represent the consensual rejection of armed struggle for socialism. See Jorge Castaneda, Utopia Unarmed: the Latin American Left after the Cold War, New York: Knopf. 1 993.

Industrial Policy Lessons/or Latin America/rom East Asia


election was a major setback for the Alliance, which lost seats to the Chinese­ dominated DAP and to the Islamic party, the Parti Islam Si-Malaysia (PAS). As the Malaysian Chinese Association withdrew its support from the Alliance, presumably in tacit support ofthe victory of the also Chinese DAP, ethnic riots between Chinese and Malays broke out. The Malay elite, which still controlled the government and the armed forces, responded by imposing martial law, in effect taking over the government for themselves. The 1 969 riots' most important catalyzing effect, however, was spurring the Malay-dominated government to politically capitalize on Malay defensiveness by embarking on a new program designed to equalize Malay and Chinese economic statuses. Malay status was to be enhanced through the New Economic Policy (NEP) (Schlossstein 1 99 1 , 227-8). The roots of the NEP may be seen in the nature of Malaysia's economic growth. While overall economic growth may have allowed the maintenance of some degree of harmony among the ethnic groups, since the Chinese business class and foreign capital have been the largest beneficiaries of the economic growth so far, Malays as a group have felt increasingly threatened by, and jealous of, the Chinese success (Jesudason 1 989, 47). Sch lossstein describes NEP as an affirmative action program for Malays. NEP sought to increase Malay ownership in the commercial and industrial sectors to 30 percent by 1 990, with non-Malays being allocated 40 percent and foreign control being reduced to just 30 percent. Government positions, placement and scholarships in education, and other advantages were reserved for Malays. Furthermore, a national authority was set up to fund Malay stock purchases. Malay acquiescence to UMNO rule was gained, in short, not through income or wealth redistribution, but with economic growth and government spending. When the 1 982 world recession reduced Malaysia's foreign exchange earnings through drops in commodity prices, the government began spending programs through borrowing to maintain economic growth. S ince most of the poor in Malaysia are rural Malays, these programs and the continued emphasis that Malaysia has paid to small holder agriculture, in contrast to LA, have paid off in terms of increasing economic development and reducing poverty. The fact that the rural beneficiaries were Malays provides the political reasons for the propitious attention to agriculture (Demery and Demery 1 992, 37-9). While the NEP has been wound down in recent years, the improved participation of Malays in previously Chinese-dominated sectors of the economy as a result of the program has been indisputable (Jomo and Gomez 1 997, 342-372). This success led to a conversation about reducing racial preferences (Rahman Embong 200 1 , 59-85). While the Malaysian experience is distinct in its details and challenges from other EA cases, on the one hand the homogenizing strategy of Singapore and on the other, the negative outcomes and continuing strife in Indonesia (which grudgingly granted autonomy to East Timor) including challenges from newly mobilized Islamic groups, and the relatively homogeneous populations in some states such as Japan and South Korea, it does illustrate some important differences in elite policy choices between the two regions. Where the EA elites generally see shared growth with increasing employment possibilities as a way to win over hostile populations, the mil itary in LA became a club for internal repression, rather than external protection. While the intellectually-based opposition of the Left in LA argued for redistribution


An East Asian Modelfor Latin A merican Success

and revolutionary overturning of elite assets, the ethnically-based opposition in EA has instead pushed for strong national guidance of development, and redistribution in government positions and spending, particularly investments in human capital (education and health), and attention to agriculture where the poorest reside.9 Nonetheless, it is important to stress that high levels of education alone do not lead to improved growth or equity (India, Argentina, Russia, and Egypt being historic counter-examples), and in EA and SEA the emphasis has been on poverty reduction, rather than demanding immediate changes in relative inequality. 10 The EA blueprint as a package including direct efforts at poverty reduction in EA has been more effective and less threatening to ruling business and political elites than those in LA, as they see the long-term benef ts in the improvement of local workforces. By contrast, indigenous groups in LA, the majority in some countries, have created autonomous movements, using external promotion to push their demands upon unresponsive institutions. These facts of comparison deserve deeper political analysis, which we shall get to in Chapter 5, but for now, let us simply acknowledge that certain choices were made in LA by Latin Americans, choices that had devastating consequences upon the region's population in contrast to but can be changed now. Diferences in the Role of Labor Unions and The Natural Resource Curse

In the late 1 990s, when conducting research for another book, 1 had an interesting interview with Chilean sociologist Manuel Antonio Garreton, internationally­ renowned for his work on Latin American democratization and economic development. During our interview about Chilean neoliberalism, particularly in its contrast to the years of struggle by the Chilean Left for equality under socialism (in which he had participated), Dr. Garreton, echoing many other conversations I have had with intellectual leaders in the region, simply said "There is no other model (than neoliberalism)." (my translation). When asked about why LA has not looked at the more equitable growth models of EA, Dr. Garreton replied, "What and build our societies upon extreme labor exploitation?" This short conversation with one of LA's leading intellectuals was one of the seeds for this book. It shows how one negative aspect of the EA development experience has been used to avoid looking at it holistically as a possible alternative. True to its history, LA tends to see ideas about progress only from Northern intellectuals. It is clear that many elites in Northern academies as well as the developing world f nd it hard to cross the Rubicon of looking to countries in the more comparable South for answers, rather than in the

9 In fact, improvements in relative inequality have a mixed record in Southeast Asia, though poverty has decreased significantly. See M.G. Quibria, "Growth and Poverty: Lessons from the East Asian Miracle Revisited," ADB Institute Research Paper no. 33 (Feb.), Tokyo: ADB Institute, 2002. 10 Anne Booth points to important contrasts in the educational equity of Korea and Taiwan vs. Southeast Asian states. See "Education and Economic Development in Southeast Asia," 1 73-95 in Jomo K.S., ed., Southeast Asian Paper Tigers? From Miracle to Debacle and Beyond. New York: RoutledgeCurzon. 2003.

Industrial Policy Lessons for Latin Americafrom East Asia


North, which we have been programmed to think of since colonialism as inherently (and acceptably) superior. Certainly, it is true that labor repression has been an important feature of East Asian industrialization, however, that can not take away from the very real gains in ternlS of standards of living and employment prospects of EA populations. EA analysts have often pointed to the low degree of labor strife in the region, related to low levels of labor legislation and protection and a more paternalistic management culture resulting in weak unions (Johnson 1 987, 1 49). Federic C . Deyo relates the differences between the relative strength of unions in LA and weakness in EA to the necessities of the EOl strategy. In the case of EA, wage rates were a powerful source of competitiveness and so had to be held down, whereas l S I wages could be passed on to domestic consumers (Deyo 1 987b). Certainly, the contrast is important, as labor unions play a significantly more important role in some countries of LA, particularly those with a large mining base, such as Venezuela, Bolivia, and Chile, though i t has diminished over time with globalization. Moreover, under lSI, such unions gained strength, as in the case of Argentina, on the basis of forming a limited pool of highly skilled workers in the privileged enclave of high-level industrial activity. By contrast, EA countries have sought to co-opt unionization through improving worker conditions, not j ust in terms of increasing national access to health care and education, but also in improving compensation and working conditions (Deyo 1 997, 2 1 6). The most prominent case of labor unrest, South Korea, bears out this analysis. According to analysts of the topic, Korean development from the 1 960s led to accelerating decreases in unemployment and increases in per capita GDP and real wages. In line with our observations above about general inequality, however, wage distributions have remained highly differentiated between blue - and white-collar workers. Moreover, South Korean workers continue to work more hours than the workers with whom they compare themselves in the First World. These demands dovetailed with the linking of students and trade unionists focused on pushing for democratization into serious actions in the 1 980s and 1 9908 . 1 1 In sum, it is not j ust labor repression but a desire to gain more of the rewards from economic success that has led to greater demands for participation and compensation in South Korea. While we do not wish to dismiss the serious problems of labor repression in EA, its overall importance should not be exaggerated either. There is a strong spirit of cooperation a belief among workers and management alike in collective sacrifice for in the benefit of the nation and future generations. Moreover, the attention paid to acquiring practical skills and continual upgrading not only benefted the overall social mobility of the whole population in terms of employment and rising living

1 1 David L. Lindauer, Jong Gie Kin, Joung-Woo Lee, Hy-Sop Lim, Jae-Young Son, Ezra Vogel, The Strains of Economic Growth: Labor Unrest and Social Dissatisfaction in Korea, Cambridge: Harvard I nstitute for International Development and Korea Development Institute, 1 997, see esp. 37, 45, 48, 56, 8 1 , 89, 99, 1 1 1-12. See also Jomo K.S., "Growth with equity in East Asia?", 1 96-2 1 6 in Jomo K.S., ed., Southeast Asian Paper Tigers ? From Miracle to Debacle and Beyond, New York: RoutiedgeCurzon, 2003, and Seung Ho Kwon and Michael

O'Donnell. The Chaebol and Labour in Korea: The Development of Management Strategy in Hyundai." New York: Routlcdgc. 200 I . esp 33 .



Industrial Policy Lessonsfor Latin America from East Asia

An East Asian Modelfor Latin American Success

standards, but it also allowed EA states to maintain eost competitiveness as they moved up the value-added curve. In sum, this culture of education, commonality, hard work, and investment for the future are the keys to ENs success in creating a world class labor force, one that has the skills to produce the most sophisticated and demanding products in the world, and one that can now begin to adjust its level of sacrifice downwards in the next generation. Another argument that must be discussed is the notion that LA is "cursed" not only by the proximity of the US, but also by its abundant natural resources. This counter­ intuitive argument suggests that colonial domination and later imperialism was much more oppressive in LA than in EA because of the stronger presence of commercial interests, based on natural resources. While economic theory would suggest that presence of such resources is a boon, allowing for a much easier development, on the basis of transferring earnings in these sectors towards diversif cation, the fatalists would have us believe otherwise. Formerly natural-resource based countries including Malaysia and Indonesia, and agriculturall y-based countries Thailand and Vietnam, have been able to diversifY into industrial and high-tech production. More obviously, the presence of natural resources did not impede the historical development of countries such as the US, Australia, and Canada, or more recently oil-rich Norway. Stil l, the natural resource argument does seem to resonate with the intense production concentration and corruption of some LA countries, such as Venezuela, which recently overcame and broke the back of its petroleum company unions under a nominally socialist President. Moreover, it is important to note, as we explain below, that U S foreign direct investment (FDI) as well as security concerns have and continue to play a dominant role in LA's history. Peter Evans points to this factor, as well as the important fact that while there was US-occupation enforced land reform in EA, there is a continuation of family and land-owning based agricultural elite from past to present times in LA.12 However, this continuation is a result of choices made, not some abstract determination of history. Deeper Roots ofDiference: The Latin American Reality13

We readily acknowledge that LA can not just wholesale adopt the features of the EA modeL The LA state, has, through great pain and sacrifice, re-entered a robust period of democratic governance. This governance is not at all incompatible with an active industrial policy. On the contrary, we have seen that in the rejections by Brazil and others of the Free Trade Area of the Americas without major modifications, and in the re-national izations of petroleum and other sectors by Venezuela, Ecuador,

1 2 Peter Evans, "Class, State, and Dependence in East Asia: Lessons for Latin Americanists," pp. 203-226 in Frederic C. Deyo, ed., The Political Economy ofthe New Asian Industrialism, Ithaca: Comell University Press, 1 987. See also Meredith Woo-Cummings, "The Political Economy of Growth in East Asia: A Perspective on the State, Market, and Ideology," pp. 323-34 1 in Masahiko Aoki, Hyung-Ki Kim, and Masahiro Okuno-Fujiwara, ed s . The Ro'e (�fGovernment in East Asian Economic Development: Comparative Institutional Analysis, Oxford: Clarendon Press, 1 997. 1 3 I thank anonymous reviewer for suggesting the addition of this important subsection. ,


and Bolivia, that democratic governments can be quite decisive without engaging in inflationary populism or short-term and ultimately destructive policies. Indeed, the Brazilian, Argentine, and Venemelan economies have managed to maintain macroeconomic stability, 14 continue economic growth, and carry out redistributive programs without inflationary policies. I would argue that such strength suggests that the potential for a more active industrial policy is much greater than critics think. However, there are i mportant differences in the cultural character of the elites of both regions, as we have discussed. East Asian states have been created in a context of recent wars that have forged strong national collective identities, across the region. Centeno points out that LA states have never really developed either a strong sense of nationhood or of control over the means of coercion or territory. 15 LA as a region has had relatively few interstate wars. Rather, conflict tends to be internal­ based on divisions between ethnicity and class, rather than nationalism. This reflects the the vast geographical distances and many obstacles to interaction and control; the conquest of large indigenous populations; and the importation of large numbers of slaves in some parts of the region, which reflect a society built upon historical divisions, rather than notions of egalitarian ism. We have to remember that LA gained its independence through European wars (Napoleon's invasion ofiberia), rather than through a struggle for independence against colonialism as was the case in East Asia and the US. The fracturing of colonial power led, outside of Brazil, to a conflict between regional caudillos rather than a continuation of centralized power with the ability to extract internal resources. Many ofthe armies at the time of independence were foreign-trained and often officers were imported from abroad. States lacked the administrative capacity or resources to mount strong militaries, and alliances between the military, landowners, regional caudillos, and, later, with industrialists have always been weak and vacil lating. There is little effort at the symbolic and educational forging of national identities in the region. Other than soccer, few things bind together e lites with their populations in the region; the former are more apt to identifY with European trends, own property in Miami, and send chil dren abroad to study. Perhaps because the elite has been able to live off the extraction of natural resource wealth with little violent external threat, there has been no need to establish a strong extractive state. Yet, this does not prevent them from making different choices in the future. Conclusion

We have seen that EA has hardly been immune from political and historical miseries, indeed in some ways it is even more cursed than LA, if we want to think in those terms. Chinese and Japanese occupations and continuing imperialism; colonialism by less than benign powers such as the Dutch and then the French; development 1 4 Anil Hira andAdam J. Morden, "Venezuela's Chavez: the Emperor has no Distribution," Centre for Global Political Economy (SFU) Working Paper, May 2004. 1 5 Miguel Angel Centeno, Blood and Debt: War and Nation State in Latin America, University Park: The Pennsylvania State University Press, 2002, esp. 1 29, 1 4 1 , 1 50, 1 76, 1 89, 239 and 256. ,


industrial Policy Lessons for Latin A merica/rom East Asia

An East Asian ModelforLatin American Success

centered around a handful of commodity products; ethnic and now religiously-based strife and terrorism; and the devastation of civil wars and ongoing tensions spilling across borders in Korea, Vietnam, Cambodia, and Indonesia, are but a handful of the incredibly difficult aspects of the history of the region. Nor has EA escaped corruption. Despite massive corruption, for example, Indonesia was able to re­ invest its petroleum revenues into industrialization. With a heavy concentration in rubber production at the time of its recent independence, Malaysia has been able to diversifY its economy, even entering into high technology industries. In sum, he curse of natural resources turns out to be j ust as exaggerated as the curse of external imperialism- it is an excuse for LA elites to not take responsibility for their own decisions on what to do with those resources. LA elites choose not to invest them in their own popUlations and the diversif cation of their economies. The point of this book and the reason for hope is that there is a strong element of policy choice involved that works in an iterative fashion with elite and national identity. Policy choices affect outcomes and can help to reverse fortunes. Recent elections in LA show that different choices, rather than historical determinants, are establishing the future trajectory of the region. What is needed is to guide those choices towards a long-term strategy of development and thereby a more productive direction for identity development. . Getting to Brass Tacks: Policy Differences

As Alice Amsden points out, East Asian states invested heavily in both egalitarian and directed human capital, and particularly into "mid-technology" industries, such as semi-conductor chips. Their emphasis on shaving off the costs of production, through the training of "shop-floor" engineers, and on creating national leaders who could compete on the world market, contrasts sharply with Latin America on several key fronts. First, Latin America never really focused on developing a national technological capability. Instead l S I was premised on the importation of foreign capital and expertise through foreign MNCs. Second, Latin America never made the same level of investment in human capital. Amsden points out that the level of technical education, particularly applied engineering, lagged severely in Latin America throughout the l S I period. Research and development expenditures have also severely lagged, and much of the research in the region was more ofan academic than applied nature (Amsden 200 1 , 1 4, 58�64, 278�9; Hira forthcoming b). Third, Latin American state-owned enterprises thrived upon the artificial prices of domestic protected markets, never really acquiring the discipline or goal of competing for export. Thus, Latin American national champions provided decreasing or relatively stagnant service, rather than using its protection as a source of investment for improvement. Fourth, Latin American managers tend to be more aloof from shop­ foor production processes, and wider management-labor antagonisms refect social c lass divisions, state-level fragmentation, and cultural predispositions against applied work. How is it, given the widespread corruption in East Asia, particularly in China and Southeast Asia, that growth still achieved such high levels'? Alice Amsden states "Latin America's protracted stagnation probably owed more to the developmental


state's failure to create a new 'leading sector ' than to its corrupt practices." (Amsden 200 1 , 1 1 ) However, it is not enough to simply espouse the obvious superiority of the EA economic policy framework. Let us now take a step back and conduct a more comprehensive comparison of the different economic policy choices made by EA and LA, and their consequences. What empirically explains the differences between East Asian industrialization and Latin American industrialization? We attempt to tackle this question by examining the differences in policies and outcomes, using them to begin f lling in the political economy relationships. This will allow us to build a holistic picture not j ust of the useful elements of EA political economy for LA, but how they work together. I n sum, we frst have to show that EA did outperform LA before continuing to discuss why. The contention here is that the combination of the descriptive quantitative analysis here and our qualitative analysis through the book provides the best approach possible to tackling the issues. The developmentalist model from EastAsia discusses the need for macroeconomic balance, while favoring certain sectors on the microeconomic level. Government budgets should be balanced, tax receipts should be reasonable to cover expenditures, exchange rates should be stable, i f subtly devalued vis a vis export market currencies, and inf ation should be kept under control. A number ofquantitative studies including the East Asian Miracle have been performed at the macro-level. However, these have severe limits (Chang 2003a). For example, Cuyers and Dumont state at the end of their regression analysis (Cuyers and Dumont 2005, 1 3 7): I n spite of some signifcant links that w e fnd, all variables reflecting intemational trade and intraregional interdcpendcnce combined explain less than one-half of variance in economic growth. Possible domestic variables like investment in human capital and cducation were beyond the scope of this study . . . these were probably highly important to economic growth, as they determine the ability to absorb, diffuse and use the knowledge and technology that is transferred or that spills over from the most developed countries.

It may very well be that institutional factors and policy instruments that focused on the capacity to absorb technology, whichever channel it was transferred through, made international trade an intraregional growth dynamics contribute to economic growth . . .

The reason behind these limitations is plain to anyone who has worked with international data related to developing countries; there are surprisingly very limited data from consistent sources, limiting the possibilities of doing a long-term cross-sectional study. With so few countries in the sample, and major changes in the structure of the international economy, there simply are not enough data points to warrant a full-blown panel (cross-country, cross-time series) analysis, so we shall rely upon descriptive statistics, in keeping with the expository, wide-ranging nature of our analysis on a wide variety of variables related to general development strategy. 1 6 16

Such analyses already exist, with ambiguous policy lessons. F o r example, one policy

analysis comparing the two regions cites no effects of initial income or external shocks to

but important differences due to low investment, high fertility rates, low quality of human resources, high government consumption, weak rule of l aw, quality of education, and income distribution among other variahles. See Jose de Gregroio and Jong-

terms of trade,

Industrial Policy Lessonsfor Latin Americafrom East A sia


We also think that examining amounts standardized as ratios, (values divided by gdp (gross domestic product)) and as percentages is a better way to standardize different starting points and to control for national differences, such as inflation rate and size of economy, thus creating a fairer measure of comparative performance, rather than absolute levels. We acknowledge that there are different starting points and different national economic profiles, and therefore, we think it is necessary to present the data on a country-by-country level. Moreover, it does not make sense to include either China or all of the Southeast Asian countries, since their adoption of the East Asian model is relatively recent and was beset by the war in Vietnam for a good part of the period. Still, we include Malaysia, one of the outstanding performers in the sub-region, in some of the tables as a point of interest. Let us reiterate that we do not seek to isolate one particular variable's weight, but rather to elucidate common patterns of macro and micro variables that cumulatively reveal a regional pattern. I f a regional pattern o f relationships can be discerned, this will lend strong weight that regional economic policies (a blueprint) are leading to different outcomes. In order to accomplish this comparison, we must use long-run historical data, which are very limited in scope and availability. We also acknowledge that any conclusions must be limited, as are the data. For example, data on Taiwan are simply not available in many places for political reasons, including World Bank statistics. We start out with a more inclusive regional comparison on the macroeconomic level and then conduct a microeconomic comparison of Argentina, Brazil, Chile, and Mexico, with South Korea, Taiwan, and Malaysia. Macroeconomic Policy

Table 3.2 contains a wealth of information of the most common measure of national wealth, namely Gross Domestic Product per capita (GDP/capita). We can see from the 1 998 fgure that Japan, South Korea, and Taiwan all have significantly higher GOP/capita than Latin American countries at present. We can also see that, in general, there are significantly higher growth rates in East Asia vs. Latin America, from the I 950s onwards. While there are some spurts of growth in Latin American countries in some parts of this period, such as Chile recently, it is also important to note that growth seems more stable in East Asia than in Latin America. Perhaps the most reliable measure of the quality of life is employment statistics. We also note, going baek to our argument at the end of Chapter 1 , that the growth rates under I S I l and I S I2 are generally better than under neoliberalism, a fact that will b e paralleled in much of this exposition.

Wha Lee, "Growth and Adjustment in East Asia and Latin America," ADB Institute Research Paper no. 54, Manila: Asian Development Bank, Feb. 2004, pp. 48-9. The whole enterprise of growth accounting is fraught with hidden assumptions and unwarranted across-the-board comparisons mixing Northern with developing economies. See Ben Fine, "New Growth Theory," pp. 201 - 1 8 in Chang, 2003.


Industrial Policy Lessonsfor Latin Americafrom East Asia

An East Asian Modelfor Latin American Success



Argentina Brazil Chile Mexico China Indonesia Japan Malaysia South Korea Taiwan

E mployment in LA and Asia, % of population,

1 950 39.8 33 37 30.8 33.8 39 42.7 n/a 30.6 36.4

1 973 37.3 32. 1 29.2 26.3 4 1 .1 37.5 48.4 n/a 32.7 34.5

1 990 36.6 37. 1 33.7 29.4 50 42.3 50.6 38.2 42.2 40.9

1 950-1998

1 998 36 37.7 37.5 32 50.4 42.9 5 1 .5 40.9 42.9 42.6

producers to meet international standards of quality and efficiency. Unlike external debt, an export-based engine of economic growth unlike borrowing does not require later repayment, with the ability to pay subject to changing international commodity prices, changing interest rates (set by Washington), and requiring interest premiums. Exports have been the reliable source ofnew investment capital for EA. Mainstream economists argue strongly that East Asia's superior performance has been based on openness to trade, which is def ned as (exports + imports/GDP per capita). Table 3.4 supports this argument, though in practice, with growing intra-firm and intra-industry trade, tracking such results is increasingly complicated.


Maddison, p. 356

Table 3.3 shows that while Latin American and East Asian countries started on a rough par in terms of overall employment, East Asia has consistently developed more employment while Latin American countries have had small or no increases, or actual decreases. Given the need for developing countries to export goods in order to obtain investment capital, this seems like a logical next step in our analysis. This reinforces our conclusion from the frst two tables that EA has achieved both superior growth and equity to LA.



Argentina Brazil C h ile Mexico China SK Malaysia Taiwan

Openness, % GDP per capita,

Ave 1 950�60 8.4 1 2.3 22.7 26.9 8. 1 5.2 8 1 .9 1 5.3

1 96 1 -70 8.9 8.1 26.8 22.3 8.0 8.9 76.8 26.0

1 950-2000 1 97 1-80 8.9 1 0.5 30.6 22.5 9.0 26.7 75.3 59.0

1 98 1 -90 1 1 .6 9.8 39.1 27.7 2 1 .7 39.3 97.8 74.7

1 99 1�2000 20.8 1 5 .4 55.5 60.3 3 9.6 6 1 .7 1 65.9 9 1.0

Sourcc; Auth cales Penn World Tables 6 . 1



Percent growth in value of merchandise exports at constant prices,


Argentina Brazil Chile Mexico China Indonesia Japan South Korea Taiwan

1 870- 1 9 1 3 784% 121% 323% 876% 200% 475% 3202% n/a n/a

1 9 1 3-29 58% 3 7% 93% 57% 49% 1 64% 1 58% 656% 273%

1 929-50 -33% 35% - 1 4% -46% 1% - 14% - 1 9% -91 % -3 1 %

1 950-73 101% 1 87% 74% 1 62% 84% 326% 2588% 6948% 3 10 1 %

1 973-98 46 1 % 399% 798% 1 24 1 % 1 528% 485% 264% 249 1 % 1 647%

1 950-98 1 027% 1 329% 1 463% 34 1 5% 2900% 2395% 9680% 1 82527% 558 1 1%

Source: Author calculations from Maddison

Table 3.4 above shows again that East Asia's performance in exporting has far exceeded that of Latin America from 1 950- 1 998. Of the Latin American countries, only Mexico, probably benefiting from the 1 994 NAFTA agreement, seems to register comparable export growth rates. Obviously, developing economies need some source of revenue to invest in new capacity, both to supply and to grow domestic demand. Ex.ports bring in hard and scarce foreign currency, force domestic

Table 3.5 demonstrates that East Asia has, on the whole, relied more on exports and imports than Latin America from the 1 950s-2000. Obviously Communist China's opening has been relatively recent, however the growth of exports and imports for that country is staggering. We see again here the differentiation of Mexico and Chile from the rest of the region, reflecting the NAFTA effect and Chile's success in increasing its exports. However, it is important to point out that Asians have higher savings rates and lower inflation rates as well, as shown in Table 3.6 below.



Argentina Brazil Chile Mexico S. Korea Malaysia

Gross n ational savings (% of GNI) inf ation

200 1 1 2.8 17 20.3 1 8. 1 28.8 39

Argentina Brazil Chile Mexico Korea Taiwan

ave, 70-75 62.4 2 1 .3 207.2 I Ll 1 5.7 1 1 .7 6.5

ave, 76-80 2 1 1 .2 62.2 82.4 2 1 .4 1 7.4 8.8 4.5

ave, 8 1 -85 3 82.0 151.1 2 1 .5 62.4 7.3 3 .4 5.1

ave, ave, ave, 86-90 9 1-95 96-0 1 1 1 9 1 .9 43.0 0. 1 7.2 1 076.6 1 1 1 3 .8 1 9.4 1 3.9 4.8 75.8 1 8.0 1 7.5 5.4 6.2 3.8 3.8 2.2 1 .6 4.0 1 .4 3.5

Souree: World Bank. World Developmenl lndicators (WDI) 2003 " I M F. World Economic Outlook 2002

Industrial Policy Lessonsfor Latin Americafrom East Asia

An East Asian Modelfor Latin American Success


Though the savings data for Taiwan are not available, we see once again a pattern of regional differences in comparing savings and inflation rates. Obviously, in a capital-short situation, domestic savings is a key to investment for the future. A healthy fnancial climate for local public and private borrowing, as well as confdence to place savings in national, as opposed to foreign banks, can only take place when infation is under control. The mere volatility of infation can lead to inflationary expectations, so that most LA nations have yet to provide a sustained track record in infationary management to the point where the expectations spiral (spending due to a lack of confidence in price levels, exchange rates, and financial institutions) has been broken. We also see again Chile's superior performance in macroeconomic management to LA, though still not on a par with East Asian counterparts.17 The recovery of EA from the 1 999 financial crisis seems to be much more healthy than the more recent crisis in Argentina as a result. The final factor that we must look at is external debt, often cited as the key deterrent to growth in developing countries.

Table 3 . 7 1 8

Argentina Brazil Chile Mexico S . Korea Malaysia


1990-2001 Total External Debt service % ofXs of G&S 48.6 28.6 1 1 .3 1 8. 1 1 0.4 14. 1 8.1 7.1 6.2 3.6 7.8

% of GNI 9.3

Public debt % of central govt rev 32.5 3.9 25.6 1 9.5 1 0.5 3 1 .4

Source: World Bank W D I 2003 Notes: Xs

exports; GNI = gross national income;

govt government; rev = revenues; G& S = goods and services

Table 3.7 shows that while total debt service is higher in LA, the key difference is that East Asian countries have a greater ability to handle their debt because of their superior performance in exporting, and in generating higher domestic savings. In a nutshell, this means EA can invest more for future growth, leading to a virtuous cycle, as opposed to the short-term vicious cycle of consumption and debt repayments in which LA finds itself.

1 7 See Hira and Sanghera for further analysis of Chile's macroeconomic performance. 1 8 I agree with a reviewer that there is probably something wrong with the figure of 3.9 listed for Brazil's public debt - it seems far too low. However, this is the figure reported in WDI 2003, see p. 250. Strangely, this statistic is not repeated in either W D I 2004 or 2005. According to W DI 2000, p. 252. the fgure for Brazil is 1 5.3 for 1 980.


Foreign Direct Investment (FDI)

We should also discuss foreign direct investment (FDI), which naturally links macroeconomic to microeconomic variables. FDI has been a highly controversial topic for decades. Questions surround whether FDI is benef cial and under what circumstances host governments can maximize the benefts ofFD I. A strong part ofthe thought process in LA surrounding imperialism sees foreign investors as inherently exploitative, especially in regard to LA's natural resources. The infamous stories of United Fruit Company's labor conditions in Central America and US involvement in Cuban sugar, among many other examples, give solid reasons for this suspicion. This led to a number of nationalizations throughout LA history, particularly oil and mineral industries. In the 1 960s, the peak of the dependency thinking in the region went hand in hand not only with nationalizations but the setting up of state­ owned enterprises under lSI. Ironically, privatizations in the last two decades have completely reversed this trend in some LA countries, reviving the skepticism of foreign investors' motives and questions of exploitation. It is interesting to note the parallel history of external fnance in the region, with the IMF replacing the hated British gunboats of yesteryear. Obviously, FDI does not necessarily follow the degree of liberalization. 19 Relatively small but liberalized IMF star economies such as El Salvador and Ghana have a hard time attracting foreign investment of any kind, while large economies, such as China, receive the lion's share of all developing country investment despite high restrictions. Investment decision-making is not, as the Left claims, based solely on proftability and/or cheap labor exploitation. If marginal returns on investment and lowest cost labor were the sole criteria, then Bangladesh and the other (capital) poorest countries would receive a much larger proportion. Obviously, risk, knowledge, established supply chains, and other aspects all affect investment decisions. Besides risk factors, investment decision-making may, in some situations, be based as much on the size of the host market as on the potential for production cost savings - hence the superior bargaining power of China with foreign investors. Furthermore, most economists would admit that the effects of foreign direct and portfolio investment are situation-specif c. Moran states: "Turning again to market structure in the developing-country economies that received incoming FDI, there is again pervasive evidence between market imperfections and FDI." He goes on to state that numerous empirical studies have found a high correlation between industry concentration in host markets and levels ofFDI (Moran 2002, 23). Another recent study by economists for a business journal of FDI in LA points to the high level of (profit) repatriation of MNCs as an ongoing concern that host governments are not reaping the benefits of production and sales within their own economies (Baer and M iles 200 I ). The case of Mexico in NAFTA is especially important, as the idea was to integrate into M NC production chains, leading to a natural spillover effect of capital. managerial, marketing. and technology

1 9 See the discussion by Sherif H . Seid. Glohal Regulat/on (�t'F(I�/lln IJirect Investment. Burl ington. VT: Ashgate. 2002. p. 30.

Industrial Policy Lessonsfor Latin Americafrom East Asia

An East Asian Model for Latin American Success


transfer to local firms.2o Yet, most studies indicate that there is little evidence of technology transfer to local f rms, either in the case of N AFTA or in other cases (Patel and Pavitt, CEPALb, Roberts and Tybout, 26). Indeed, it is self-evident that the continuing dominance of MNCs underscores the inability of local entrepreneurs to compete on any of these key axes. Moreover, in regard to portfolio capital, the evidence indicates that Chile and Malaysia's capital controls helped to buffer them from successive regional f nancial crises. Therefore, our conclusion from this admittedly quick review of the facts is that FDI is subject in part to the conditions that the host government sets up for it. This is backed up by the EA literature (Wade 1 990). Our next table confirms that FDI, trumpeted by mainstream economists and institutions as the most important source of growth for LA, may be over-rated.



FDI in LA vs. EA countries

Argentina Brazil Chile Mexico S. Korea Malaysia Taiwan Year Units

2799 1 42530 1 5547 4 1 1 30 999 1 28732 1 5736

Inward %GDP 9.2 2 1 3 .6 1 2.7 0.8 1 9.4 2.5

Outward %GDP 1 .7 0.7 3.8 0.4 1 .4

1 995 $millions

Ave 90�95 %

Ave 90�95 %



Inward Stock

3458 2000 1 499 8080 978

741 676 438 288 1 842 1 050 29 1 7

4655 1 222

Ave, 1 990�95 Ave, 1 990�95 $millions $millions

3.4 6.2

Source: Auth cales World Investment Report 2002, UNCTAD, Geneva, 2002

As with other statistics, the data on FDI for comparative cross-regional purposes are highly limited. However, what we can discern from the Table 3.8 is that EA, with the exception of Malaysia, received far less FDI than LA. Moreover, as numerous regional specialists have pointed out, EA's direct investment comes from Japan and the US, whereas through most of LA (the Southern Cone enjoying some from Europe in recent years), the sole source is the US. Regional specialists claim that this allows EA countries to negotiate better terms on FDl. While being able to play off investors is likely an important bargaining tool, what we see from the above table is that both Korea and Taiwan highly restrict the overall amount of FDI, regardless of the source.


In fact, the literature on FDI in EA points out, in contrast to the haphazard policies of LA, that the EA state has quite consciously guided FDI towards certain targeted sectors; placed heavy and time-dependent restrictions on how it can be used; and generally guided it towards export-oriented sectors. In other words, FDI is seen by EA states as a necessary evil, but one which must be harnessed through tough state­ MNC bargaining with the domestic private sector's collaboration and interests close at hand. For the most part, there is no evidence of a coherent, long-term strategy of FDI in LA by contrast, nor is there a clear sense of using FDI to consciously transfer technology and capital in order to help domestic businesses enter into tough sectors.21 Mainstream economists consistently warn of the dangers of distortions which can lead to inefficiencies, special interests, and corruption. For example, Moran points out that domestic content requirements could shackle a domestic industry from becoming competitive on world markets by raising f nal prices (Moran 2002, 43). This simply underscores our point that FDI must be part of a conscious strategy to move up the value-added ladder in global production networks. Moreover, unlike the lSI plans of LA, in which parts and capital were imported for domestic sale, leading to severe macroeconomic problems (Hira 1 998), a conscious effort must be made to create competitiveness throughout the production and value-added process on the one hand, and the discipline of having to sell on world markets on the other. The examples of South Korea and Taiwan suggest two strategies that could be applied to LA. South Korea's chaebol [large vertically-integrated conglomerate] system follows the lead of Japan's keiretsu [large vertically-integrated conglomerate, often based on layers of closely-tied subcontractors], in creating national champions who have the economies of scale in f nance, output, diversif cation, and management and marketing knowledge to compete with global companies independently. This seems to be the natural model for heavy and chemical industries, and, as we show in the next chapter, is followed in this fashion by China. On the other side is the development in Taiwan of smaller-scale manufacturers, also nationally-owned, but focused more on niche markets or goods such as electronics that require less heavy concentration of investment or vertical integration. The Taiwan model would be more appropriate for LA countries such as Chile that may not be able to achieve global dominance in labor­ and capital-intensive manufacturing, as has also been argued by Katzenstein in the case of Western Europe. Unfortunately, as Amsden points out, LA countries never really had a clear and stable long-term strategy for developing globally competitive companies. The region's governments placed so many restrictions and conf icting orders upon both their national companies and foreign investors that producers ended up getting crowded out by core multinationals. She also makes the crucial point that, contrary to neoliberal assumptions and Mexico's NAFTA strategy, MNCs usually do not invest in local capabilities for research and development. These problems were and are compounded by the lack of any clear strategy or government support for

20 See M. Cimoli, ed., Developing Innovation Systems: Mexico in a Global Context, New York: Continuum, 2000, esp. p. 280 on the lack of MNC technology transfer generally and in Mexico particularly, and Hubert Schmitz and Jose Cassiolato, eds., Hi-tech for Industrial

2 1 Robert Wade, Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization, Princeton: Princeton University Press, 1 990, Dirk Welllem

Development: Lessons from the Brazilian Experience in Electronics and Automobiles, New

te Velde, "Government policies

York: Routledge,

1 992.

towards FDt," pp. 1 66� 1 97 in Ganeshan Wignaraja, ed.,

Competitiveness Strategy in Developing Countries, New York: Routledge, 2003.


An East Asian Modelfor Latin A merican Success

research and development in local product innovation, as we discuss below (Amsden 200 1 , 205-8, 243). Summing Up the lvfacroeconomic Comparison and Important Caveats

We have demonstrated here that EA has consistently out-performed LA on a country­ by-country comparison in terms of macroeconomic development. LA sufTers from tremendous volatility in all of its macroeconomic accounts. It i s important to note that there are some indicators that are less clear, which we have not had time to discuss here. For example, comparing investmentlGDP seems to indicate that East Asian countries are higher, but Mexico is close behind. In sum, this macroeconomic review could be used to support either the developmentalist state model or the World Bank et aL 's argument that free markets and exporting lead to economic development. This is where a comparison of the microeconomic climates can be very instructive and follows Chapter 2 which showed the states always intervenc on the microeconomic level. The question is to differentiate the ways EA did so vs. LA. Fiscal Policy

To start a comparison of microeconomic climate, we would like to prove the capacity and quality of EA governments is higher than those in LA, as this is the basic starting point of the EA governing the market l iterature. It is interesting to note, furthermore, that there is no clear difference in my own comparison of government spending/GOP or total consumption/GOP. In the grossest measure, then, the relative importance of public sector spending is the same in both regions. Better results in terms of social welfare and human capital therefore suggest that EA states are much more efficient in their spending patterns. In general, the charge, which seems j ustified, has been made that much of the spending by LA states, particularly during the l S I period, went and goes to political patronage, and results in "white elephants," by contrast with EA, where the results are more effective (for example, see Teranishi 1 997,304 and Adams 1 99 1 ). Stil l, there is no clear index or way to begin to measure the relative quality of government. There could be a wide variety of reasons for Latin American macroeconomic volatility, ranging from greater dependence upon price­ volatile natural resources to lack of effective Central Bank independence. We do not have the space here to explore all of the possibilities, but one in particular is worth highlighting. The Latin American state's dependence on external debt goes hand-in­ hand with a very low capacity to tax. The f rst and most obvious measure of f scal capacity i s overall government deficit or surplus, with data given in Table 3.9. Table 3.9 shows that South Korea's and Malaysia's performance (the latter from the mid- 1 980s) in terms of fiscal discipline has been better in terms of lower levels and less volatility than most countries in Latin America, with the exception of Chile. Fiscal policy, by all accounts, is at the heart of the weakness of LA economies, with every country in the region struggling with tax reform (Hira and Dean 2004). We now tum to the data we have available on sources of federal tax revenues. Tabl e 3 . 1 0 reinforces the differences in the fiscal picture between the two regions. Direct taxes are a key indicator of state capacity, in the sense that they

Industrial Policy Lessonsfor Latin Americafrom East Asia


refect the ability of the government to collect income, personal, and property taxes, as opposed to relying on customs duties. The Tabl e demonstrates that LA's ability to tax is much l ower, with the exception of Mexico (though some skepticism should be placed towards Mexico's data). And, with the exception of Chile, LA has much heavier social security burdens, refecting a state much more suscepti ble to patron­ client networks and political pressure, a state geared towards middle and upper class expenditures rather than socially-benef cial l ong-term investments. Large persistent government def cits could also speak to an inability to control cronyism, money­ losing state-owned enterprises and other runaway sources of spending as well as an inability to adequately tax. So, i s there a difference in what East Asian vs. LA countries have been spending tax monies on? The best data we can get are again general categories from the IMF.



Overall central government deficit/surplus

Argentina Brazil


- 1 7.21




3.83 ( 1 970)


- 1 7.0

- \ 5.59

-23. 0 1


- 1 2. 7 1 ( 1 974)

1 8.84



1 3. 1 4


-27.76 ( 1 973)

- 1 6.52


- 1 4.06

-3.3 1

S . Korea

-.39 ( J 975)

- 1 1 .05



1 .46


20.9 ( 1 974)

- 1 8.48


-7. 1 9

1 0. 1 1

year in ( ) Source: Auth cales fro I M F Govt. Finance Statistics Note: (% of Total Exp & Lending


3. 1 0



Tax revenue source a s

% of total revenue





1 7.02



1 998

1 9.44




1 8.35



1 999


1 0.88


26.4 1


1 997


S . Korea

Source: Author cales

[\1f' Govt. Finance Statistics 200 I

Note: taxes on income, profits, and capital gains

Table 3 . 1 1 begins to get us even closer to differences in the quality of governance between the two regions. What we see is that while defense expenditures are higher in East Asia, so are education expenditures, with the exception of Mexico.22 There seems to be no clear pattern on health, but another clear difference is the overwhelming burden placed on f scal expenditures by social security payments in Latin America. Forthcoming analysis that I am in the midst of perfonning also shows that more expenditures go towards elitist higher education in LA with a relative neglect of primary education, in contrast to EA. Thus, as we discuss in the next section, the ultimate root of the problem and heart of the difference between the two regions seems to be the much weaker strength, capability, and autonomy of the Latin American state. Is this difference matched by ditferences in private sector perfonnance? We tum to financial policy to begin to answer this question, and return to complete it in our subsequent chapter on sectoral policy differences.

Table 3.11

Korean state has strict and unequivocal control over "every aspect" of the banking industry. Nembhard describes in detail how the Korean financial system was set up to support the chaebols, or national champions. F irst, the banking system was regulated to ensure low interest rates to such enterprises. Second, the government used 'policy loans' directed toward particular industries. Third, the government kept a tight control over foreign financial flows, including aid, in order to channel them in the right directions (Nembhard 1 996, 92-99, 1 9 1 -95, This led, as is well known, to an external dependence on external borrowing responsible for the financial crash ofthe 1 990s, and the more recent deregulation offinance. We note that South Korea's ability to handle its debt crisis has far exceeded its counterparts in Latin America.



Year 2000 1 998 2000 1 999 1 997

Defense 3 .95 3 .49 8.01 3.27 1 6.66

Education 6.28 6.1 4 1 7. 8 1 25.54 20. 5 1

Health 1 .88 6.21 1 2.4 1 4.22 0.78

1 997

1 1 .14



Interest rates a n d business environment

Interest Rates

Expenditures b y fu nction as of total expenditures

Argentina Brazil Chile Mexico S. Korea Malaysia


Industrial Policy Lessons for Latin A mericafrom East Asia

A n East Asian Modelfor Latin American Success


Soc. 48.46 47.26 36.42 21.17 1 0.79 7.2

Argentina Brazil Chile S. Korea Malaysia Mexico Year

Time to start Time to enforce a Time to resolve contract (days) insolvency (days) up a bus (days)

29. 1 46.7 1 0.2 6.3 9.5 7.9 2001

63 86 34 36 56 41 Jan 02

300 1 80 200 75 270 210 Jan 02

730 3650 1 454 535 545 Jan 02

Source: Author cales IMF Govt. Finance Statistics 2001

Source: Author cales fro World Bank WDI 2003

Financial Policy

On the surface, several Latin American countries such as Brazil and Mexico seem to have followed a similarly developmentalist path in tenns of strong control and directing offinance. However, it is important to note several key differences, beyond the macroeconomic and exchange rate differences noted previously. The most important ofthese is that in the Latin American cases, "the triple alliance," meant that the key partnerships for industrialization included foreign multinational companies (MNCs). Secondly, Latin American finance has obviously relied very heavily on external and often short-term sources, as states struggle with both taxation and fiscal responsibility and with low domestic savings rates, as demonstrated in the tables above. These weaknesses are the primary sources of struggles with inflation and massive capital flight over the last four decades. Thirdly, there never was any l ong­ tenn national plan for financing industrialization, for pol itical reasons reviewed in Chapter 5 . As a result, there was no ability or attempt to create clear i ndustry targets or goalposts for reaching world competitiveness. Moreover, Latin American private enterprises have had chaotic and strenuous access to fnance. in sharp contrast to their competitors. A UN report notes that, in t 995. interest rate spreads were "at l east twice as high in developing countries and economies of transition than in the industrialized world." Moreover, this problem is exacerbated by the tendency of

As pointed out in the previous chapter, the control and direction of finance has been a key component of the East Asian miracle. While we have mentioned in Tables 3.6-3.9 that East Asia has lower debt payments and is more capable of handling the external debt burden through exports and taxes, it bears reinforcing the differences in the financial systems of the two regions. The domestic financial situation is vital in making credit available for businesses to grow, and for consumers to purchase products. It is worth examining the levels of national interest rates across the regions. Table 3 . 1 2 demonstrates that businesses operate in far different environments in EA and LA, backing up the findings of international competitiveness surveys. Interest rates, outside ofMexico, are far lower in EA. In tenns ofbusiness environment, South Korea far outpaces all other countries. These observations are supported by analysts of the financial systems in the two regions. For example, in a detailed comparison of South Korea and Brazil, Jessica Nembhard points out that Brazil exhibited much less capability of controlling its national financial system. She notes that the South 22 Though I am taking these f gures at face value, as one must, l do have strong suspicions about the accuracy of statistics from Mexico.

Industrial Policy Lessonsfor Latin Americafrom East Asia

An East Asian Modelfor Latin American Success


banks in developing countries to: a) remain with long-standing and large clients; b) focus on short-term lending and avoid long-term borrowing; c) charge much higher lending rate differentials from both the world and the domestic prime rate, than their counterparts in the North; and d) have considerably less capacity to enable private bond issues, thus effectively closing off the world private capital market for many companies.23 Fourthly, the Latin American penchant to create state-owned enterprises, including political reasons, rather than a partnership between the state and the national private sector, meant that finance could not be used as leveragc to push forward improvements. Nembhard summarizes the problem for Brazil nicely in a way that describes the Latin American case generally: The lack of control over significant institutions in the public f nancial sector and lack of consistent and coordinated fiscal and monetary policy has continued to plague Brazil through the 1 980s. In addition to the weaknesses in execution of monetary policy, the government's f nancial policies in general have never been well unified and its financial command never very effective in part because its scope has been limited and because of the overlapping responsibilities and conf icting functions and objectives of its three major fnancial institutions, the independent policy-making responsibilities of the various public agencies, and the persistence of infation. The government has not successfully operated Brazil's financial system to use monetary and fiscal instruments deliberately and strategically to both reward and punish economic behavior (to the same extent as South Korea . . . ). Its financial institutions compete with the private financial sector more than they guide and regulate private activity. Over the years, it has partly compensated for long-term lacks in private sector financing, but has not compensated for structural weaknesses and general inadequacies and underdevelopments in the private financial system. In fact, it has mostly attempted to augment and empower the private financial sector as a viable separate entity competing with the public sector. Concern has often been focused on providing adequate or equal sources of credit to the private sector, to lessen public sector domination. What has resulted is a large and eomplex system dominated by the federal executive government, although not well controlled by it, and not well unified. The government, therefore, has not always been able to put the credit system effectively in the service of its development plans nor to adequately coordinate its credit policies, capital controls, and development strategies. (Nembhard 1 996, 1 4 1 ) Social Outcomes

Critics of the approach recommended here will point to our emphasis on purely economic measurements of success. While the focus of this study is on economic comparisons, we can also demonstrate that the EastAsian states' approach has resulted in better standards of living as well, as demonstrated in the following table. Table 3 . 1 3 shows that absolute poverty and relative inequality as measured by the Gini coefficient (a measure of the steepness of a curve between the lowest and highest income deciles) are considerably lower in South Korea than in the other �

23 Francisco Sercovich, et aI., Competition and the World Economy: Comparing Industrial Development Policies in the Developing and Transition Economies N orthampton MA: Edward Elgar and United Nations Industrial Development Organization. 1 999. pp. 1 3 1 3 . See also their discussion of Korean industrial policy. pp. 1 94-205. ,


countries. Malaysia's performance, with the exception of the efficiency of education expenditures is also superior. The rest of the table backs up our previous observations that though social expenditures are often lower in EA they bring better results, and that the major differences are LA dedication to social security.



Comparing standards of living and inequality in EA and LA

Popn below $2/day


Child Education Ave H ealth Mortality Expenditures years of Expenditures schooling

Maternal Mortality ratio

National Argentina Brazil


59. 1












1 60 55


1 997-2000

1 990-98








5 1 .9




5 .4


S . Korea


exc elect



Sectoral and Political Foundations for a New Industrial Policy Pathfor LA


that Prebisch pointed out was a major weakness, back in 1 948. Undoubtedly, the value added and export revenues of food products, as well as the median salaries and wages and the number of employees are lower than more sophisticated large scale industries like machinery, autos, and electronics. When we look at a sectoral breakdown of exports and imports, we f nd a reinforcement of these conclusions. Table 5.5 confrms what we have discussed throughout the book: despite the ability to export by some exceptional companies, LA continues to follow the dependency-type pattern of relying heavily upon exports of agricultural goods and minerals and importing f nished goods. It is also interesting to point out that in both LA and EA, inter-industry trade seems to be key. In other words, both regions import heavy amounts of capital goods. The difference is that EA has managed to upgrade its capabilities so that it can participate in the global production of higher value­ added and higher-technology goods which correspond to a higher standard of living and higher wage employment for its workers. The differences between a migrant farm worker or poultry processing worker and a highly skilled welder/craftsman or computer programmer are the differences between the haves and have-nots in the world, namely characteristics of the sectors in which one is employed. Innovation Policy

Ultimately, the advantage EA has in entering new sectors rests in the microeconomic level upon its technological capability based on a conscious and coherent innovation policy, lacking in LA. In fact, Lall makes the point that the upgrading oftechnological capacity went hand-in-hand with the successful shift to EOI (Export-Oriented Industrialization) in EA (Lall 1 996, 1 2- 1 5). I nnovation policy in EA has included both government subsidization and support of technological upgrading, including, for example, the subsidization and development of research and technology parks and related efforts (for example, S ingapore's new biotech effort); the stimulation of foreign investment in high tech sectors (Malaysia's supercomputing IT corridor); reverse engineering and other forms of iearning (South Korea's mastery of the steel industry); and skills upgrading (La1l 200 1 ). There is no question that EastAsians have made a conscious effort to bridge the technological gap, which is one of the keys to developing manufacturing competitiveness and national champions, as discussed in the literature. We now review evidence on innovation in the two regions. Table 5.6 shows that South Korea is far ahead of the other countries in the sample in regard to innovation policies and outcomes. It is important to notice that the percentage of high technology exports from Malaysia is quite high, which indicates that it is targeting those sectors in terms ofprocessing, if not innovation. As Amsden points out in regard to India and Argentina, it is not enough to simply have a literate and well-educated population. Education must be based on actual factory foor experience and otherwise applied research and development activity geared towards production (Amsden 200 1 , 1 57). Though we do not have comparable data on Taiwan, it is clear that the development of the information technology industry o

.� �

An East Asian iVlodelfor Latin American Success


Sectoral and Political Foundationsfor a New Industrial Policy Path for LA

1 13

there i s l inked to direct government action in selecting, subsidizing, and financing the industry before privatizing it to national businesses.3 8 o N

Is East Asia Really More Competitive?

So far, we have shown the macroeconomic and national accounts superiority of East Asia i s matchcd by an intriguingly d ifferent production and sectoral profile on the microeconomic level. But does the fact that EA produces manufactured goods mean that it is more competitive internationally? To answer this question, we examined a number of common scales of competitiveness. These indices tend to rely upon surveys of business executives to report on conditions such as infrastructure, and economic, government, and business efficiency for particular countries. By every survey, EA countries rank far above all LA counterparts. By some measures, Chile rises higher than its counterparts in the rest of LA, though still lagging behind EA.4 How Latin American Can Begin Targeting Sectors

j �

« � .S = Q

... .. ('I

S = =

Though anathema to neoliberal economists, we have discussed the fact that all states have targeted and supported key industries in Chapter 2. Targeting is also a fundamental characteristic of the EA model, as we have pointed out throughout the book. It's no mistake or market outcome that Korean electronics are competing alongside Japanese ones at a lower cost/quality. C learly, sectoral targeting sometimes fails, as with the case of the aircraft industry in Japan and petrochemicals in Korea. Yet, we have shown overall that EA's strategy, even with the expected and inevitable occasional failures, has reaped rich dividends in terms of moving towards new and more highly value-added goods over time, in turn allowing for more resources for investment and higher standards of l iving. The literature on sectoral targeting, particularly in regard to Latin America, is virtually non-existent. M oreover, there is almost no l iterature on the types of government-business-Iabor relations that deals with specific sectors, with a few notable exceptions. The main l iterature tends to talk about general characteristics of sectors, such as "asset-specific" or effects of capital mobility, rather than examining 3 As I point out in my chapter on developing countries' policies to attract IT business, the same types of aggressive government policies have been followed by Ireland, S ingapore, India, and China with success. See Anil Hira, "How Developing Countries Attract Outsourcing Jobs," in Hira and Ron Hira, Outsourcing A merica, New York: Amacom, 2005, and, on Taiwan, Jeff Saperstein and Dr. Daniel Rouach, e ds . , Creating Regional Wealth in the Innovation Economy, Toronto: Financial Times and Prentice Hall, 2002, chapters 1 3- 1 4, pp. 2 1 9-32. 4 Ganeshan Wignaraj a and Ashley Taylor construet a Manufactured Exports Competitiveness Index (MECI) and a separate index of technology-intensive exports for the

developing world, which clearly show East Asian countries at the top. See Wignaraja and Taylor. "Benchmarking competitiveness: a first look at the M Eel," pp. 6 1 -92 in Wignaraja, 00. . Competltivenes.¥ Strategy In Developing CountrJe.f, New York: Routledge. 2003.

1 14

An East Asian Modelfor Latin American Success

and comparing real sectors. However, these studies look at the politics behind the (often disappointing) outcomes, usually from the point of view of businesses lobbying, rather than proposing a pro-active and prospective state-led strategy that could be fol lowed.s There are also a few authors who have done studies of specifc case studies of sectors, including CEPAL (UN Economic Commission for LA) on occasion. The case studies are highly important for bringing to l ight the complexity of state-sector and intersectoral bargaining and coalition-building but they are universally post hoc and occasionaL While the work of these authors is quite important in terms of political economy theorizing, especially in regard to relative strengths of states and private sectors in an international context, and generally placing emphasis on the neglected area of sectors in LA, there is very little in the way of actual and prospective policy guidance that comes from these analyses. Though there is also a growing interest and literature in sectoral competitiveness in an international sense, including the seminal work of Michael Porter, and the ongoing work of organizations such as UNIDO, this literature does not consider the sectors in a LA political context. So, we will provide some guidelines below on how to begin the idea of sectoral targeting for LA, in the hopes of spurring new research agendas in this direction. What Sectors Should Latin A merican Nations Target?

We could start to answer the question in a tautological fashion, by stating that the most attractive industries are those that serve a nation best. We could inventory a particular LA country's strengths, weaknesses, possibilities, and potential competition in terms of assets, overall international position, stage of development, and the like. We could then suggest a number of criteria for decision-making, such as employment creation, revenues creation, amount and depth of linkages, size of current and potential markets; and size and degree of barriers to entry. Ideally, we would be able to create an international profle of all industries and then rank them according to the proportional weights of our decided-upon criteria. Unfortunately, there are no data which looks at industries on a worldwide scale in this fashion ! As a starting point, 1 have been able to accumulate some evidence that helps us to create a global prof le ranking industries according to some of these criteria, recognizing that this exercise would have to be completed by a national­ level political and economic analysis, and that the relative place and definition of different sectors in the world economy is a moving target. Nevertheless, this exposition may provide some suggestive guidelines to help analysts to begin such a national inventory and planning process. We must reinforce important points of caution and guidelines before examining this data. First, obviously, sectoral targeting must be done on a national 5

The case study literature on



sectors and industrial policy in general i s quite

R. Kingstone, Crqfting Coalitions for Reform: Business Preferences, Political Institutions. and Neoliberal Reform in Brazil, University Park: Pennsylvania State University Press, 1 999, and Roy C. Nelson. Industrialization and Political qfnity: lndu.ytrial Policy In Brazil. Now York: Thunderbird/Routledge. 1 995. limited. Two important works


Sectoral and Political Foundations for a New Industrial Policy Path for LA


and regional level, according t o an estimation o f the potential for sustaining a particular industry over the long-run. Such an estimation must include the potential linkages with existing commodity, industrial, or service production and factors of production. Second, the costs of entry and of maintaining market position must be considered, in terms of the opportunity costs for other industries. Third, the levels of complexity of institutional management in both the public and private sectors should be considered. Fourth, some industries, such as labor intensive electroni c components in Taiwan' s early history, or semiconductors for Malaysia now, may be viewed as important stepping stones for gaining entry into a larger industry. Fifth, though we can indicate below the most promising sectors in terms of overall size, all of the above factors as well as the structure of the market itself must be considered. M arket competition, levels of support and rapidity of innovation required, degrees of hidden and overt market protection, access to export markets, and other aspects of the product market itself must be researched and considered quite carefully. S ixth, an emerging market is likely to be easier to enter than a well­ established market. Thus, though we rely upon overall size (accel eration of market size or emerging products would be much more difficult to size up in this general level of survey), potential growth of market and vulnerability of market leaders are equally viable factors to consider. Seventh, as we have claimed throughout this book, the value-added, employment, and learning aspects all have to be considered in terms of what sectors to target. Finally, as we discuss in detail later, the country must arrive at some consensus for an overall developmentalist strategy. There must be a self-study by government, business, labor, and the general population must be persuaded on this long road. Thus, sectoral choices have to be backed by political consensus, as a nation is pooling its investment towards a certain goal, requiring concentration of resources at first for long-term sharing of the gains. We can begin to explore which sectors to target by looking at the relative importance of different sectors in the US economy. While far from ideal, in the sense that the US is the most advanced economy in the world, it also remains the most i mportant destination market in the world. As we have discussed, self­ sufficiency simply does not make sense for Latin Americans i fthey wish to achieve a standard of living comparable to those in the North. The US is also the market with the best economic statistics by sector. I have prepared some rankings of the top sectors in the US according to some of the basic criteria pointed out in the industrial policy l iterature.

An East Asian Afodelfor Latin American Success


Table 5.7

Rank order of leading US manufacturing sectors, 1997


Electronic and other electric equipment

2 Industrial machinery and equipment

Chemicals and allied products


Chemicals and allied products and equipment

5 Electronic and other electric equipment

Industrial machinery and equipment Food and kindred products Fabricated metal products

Accruals ($millions)

and P-T, WOOs)

I Industrial machinery and equipment

Industrial machinery and equipment

2 Electronic and other electric equipment

Electronic and other electric equipment


Chemicals and allied products

Food and kindred products

4 Food and kindred products

Fabricated metal products

5 Fabricated metal products

Chemicals and allied

Source: Author cales BEA statistics :.Iote: Did not include services: notable: govt, real estate, health, construction, in that order

Table 5 . 7 shows that our efforts comparing LA with EA production profi les can now be brought to fruition . We should note f rst of all that we have focused our rankings on goods-producing industries, and not included services. As is well known, services are a fast growing part of mature economies. However, the nature of services and the data on them are still not well-def ned. Moreover, many services that loom large in the statistics, such as government, real estate, health, and construction, are not likely to be sectors that will provide maj or employment outside of the market country, nor are a significant source of export earnings. Our data analysis reveals that while the food products sector appears to be an important one in which LA has an advantage over EA, EA, particularly South Korea and Taiwan, have competitive industries in every other top-ranking sector, particularly electronics. Our chapter on China showed that it is following in its neighbors ' footsteps in terms of targeting similar industries. Sanj aya Lall provides an alternative table ranking the fi fty fastest growing (by 1 996 $ value) manufactures in world trade from 1 980--95 , with the top ten as (Lall 200 1 , 255): 1. 2. 3. . 4. 5. 6. 7. 8.


9. articles of plastic 1 0. furniture, parts thereof

Value Added (% of ODP)

1 Food and kindred products

4 Motor vehicles

Sectoral and Political Foundations/or a New Industrial Policy Path/or LA

transistors, valves, etc automatic data processing equip telecom equip. , parts office, auto data processing machinery electrical machinery medicinal, pharma products switchgear and parts machinery for special industries

So, we see here again from this l ist that the EA's sectoral production profile is much closer to the fastest growing industries in terms of trade than is LA. This would indicate to us that the electronics and machinery industries would be logical sectors to target. We can also conduct a firm-level analysis on the l imited international data available, from Forbes. Table 5 . 8 reveals that the motor vehicles, financial services, and petrochemicals industries are among the most important sectors in the world. Other sectors i nclude information technology (IT), telecommunications, pharmaceuticals, and retail services. Thus, it is no surprise that Japan has some of the top companies in the former two, and that South Korea and China have very deliberately targeted motor vehicles, pharmaceuticals, telecommunications, and petrochemicals as key sectors in which to compete internationally. This points to the importance of inter-industry linkages and hourly wages, which are not covered by our statistics. It would be good to know which industries provide employment to upstream and downstream industries. In this sense, we can obviously differentiate the mass production of food products from automobiles or metals, as the latter are parts of intricate production chains. Unfortunately, it would be impossible to calculate what the relative wage of certain industries is over others, as conditions from country to country, not to mention time to time and company to company can vary greatly. However, undoubtedly automotive workers are paid far superior wages to those working in food production. C learly, the industry wage should be related to the overall price, profitability, and value-added of a product, as well as the level of skill and other factors that constrict labor supply. In this sense, we can easily conclude that developing countries should target sectors that provide massive employment in the short-run, and wage increases in the long­ run. Thus, the EA literature discusses the "industrial ladder" of beginning with textiles, moving to electronics, and then to heavy and chemical industries, before finally tackling services and high technology industries. On our regional level of analysis, we can state that the important point here is not to espouse certain industries for LA, but to suggest that the sectors targeted should all be part of a progression towards higher-value-added, employment, and learning production. Again, this logic will need to be adapted according to both a world industry and market analysis and an intimate knowledge of a particular country's current and potential comparative advantage. It would also require a serious overhaul of national f nancial systems. However, we can provide a few illustrative examples of how one could begin the thought process here. For example, Argentina could begin by recognizing that it is at a distinct disadvantage vis a vis Brazil in labor costs. However, Argentina has a very literate population and the potential therefore for a highly developed services industry, for the region. Moreover, Argentina clearly can rely upon its traditional comparative advantage in cattle, wheat, and its emerging advantages in natural gas. Thus, Argentina could also attempt to increase the value

An East Asian Modelfor Latin A merican Success


The most important firms in the world

Table 5.S

Forbes Ranking of Largest Companies in the World Assets Profit

Sales I


2 ExxonMobil



Mizuho Financial

General Electric

Fannie Mae



General Motors


Royal Dutch/Shell Group BP Allianz Worldwide


Ford Motor


Altria Group


Daimler Chrysler

Sumitomo Mitsui Financial



Toyota Motor

Deutsche Bank Group

Bank of America


General Electric

Mitsubishi Tokyo Financial

Royal Dutch/Shell Group

HSBC Group

SBC Communications

JP Morgan Chasc & Co

Wal-Mart Stores

9 Mitsubishi 1 0 M it sui

& Co.


Note: Mitsui pdces personal and HH pdcts; Altria pdces tobacco Source: Forbes On-line, accessed Jul 04

added that it captures in food products. Instead of simply exporting the raw materials, Argentina could promote the organic nature of its meat products as superior to those in the US and Europe. To date, Argentina has yet to reach out to potential customers to differentiate the superiority of its exports, as it has at last begun to do with wine (lacking the state support afforded to Chile). As Marx pointed out, it is only in the differentiation of commodities that value-added can be created. Thus, Argentina could look into acquiring companies and retail outlets in the US and Europe in food products and expanding the distributional channels for its differentiated products. Mexico is a highly interesting case in that it is tightly linked to US industries through NAFTA. As we illustrated in the last chapter, this has allowed Mexico to distinguish itself from the region both in terms of employment growth and investment, as well as becoming part of key industries, including motor vehicles. Yet, while there are some exceptions, such Vitro in glass and CEMEX in cement, Mexican companies have not yet been able to take advantage of the NAFTA to produce nationally-based competitors. This is where Wade's analysis of Taiwan's policies towards FDI is particularly instructive. Rather than simply opening doors to US investors, the Mexican government could provide assistance and training to local entrepreneurs to begin to develop their own companies. Undoubtedly, they would have advantages in terms of knowledge of the local labor, supply, and consumption markets. Perhaps initially these companies would look to become competitive sub­ contractors or f nd neglected market niches both at home and abroad, but the goal a la EA must be to develop internationally-capable competitors. Moreover, with the multinationals close by, the Mexican government could also make it attractive for the management of U S companies to take the managerial and technical lessons with

Sectoral and Political Foundationsfor a New Industrial Policy Path for LA


them to start their own companies. Considering the huge level of remittances f owing back to Mexico from ex-patriates, there should be available capital and interest in such ventures. The problem to date has been the almost complete lack of pro-active IP by the Mexican government to establish such linkages. Political Differences of Industrial Policy in East Asia and Latin America

We can now summarize our proposal for the harmonious mutual relationships needed for successful economic development in the following diagram: F igure 5 . 1 shows that a successful IP must have state leadership at its core, but that the state must also work with allies in the private sector, in academics, and in the f nancial sector to make the system work. We also noted the importance of a "pilot" agency within the state to act as a long-term planner and coordinator for IP. So why is there so much more political consensus around IP in East Asia as opposed to Latin America? We can begin answering this question by starting to address the easy arguments of both the Left and the Right in dismissing these East Asian relationships as politically impossible for LA. The arguments we often hear from the Left are that EA depends exclusively on labor repression and authoritarian governments or is a unique historical accident/conspiracy (which we addressed in Chapter 3); that LA governments are run by venal elites and patronage-filled bureaucracies; and that the US and the all-powerful multinational corporations would not permit the development of strong, independent Latin American states. 1 . National Innovation System

Heavy Industries (Employment)

Targeted Sectors (Future Growth) (temp., to stri sta

Universities Think tanks Private Research Foundations, etc



technolog ,





Domestic and Foreign Investors, including MNCs

(macroeco stability, risk reduction)

National Financial System

Agricultural, Mining, and Other Commodity Exports + Newly Successful Export Industries


(growing list over time)

income, personal

Banking System



3. Social Welfare and Human Capltal lnvestmenta Education, Health Care, Retraining and other Insulation from changes In domestic and Intematlonal markets

Figure 5.1

2. National System for Global Competition

Key relationships for a true LA I P

domestic public and private savings

] 20

An East Asian Modelfor Latin American Success

has The Right, which has interestin g parallels with the Left in a variety of areas, m mainstrea So, corrupt. as sees also it equal skepticism for the LA state, which s institution of refonn on focused are USAID institutions such as the World Bank and difference key the course, Of law. contract and and regulation, such as the judiciary is that the Left and Right propose diametric ally opposite solutions . The Left wants ts, a purer, revolutionary state or a blossomin g of autonomous grass roots movemen a where ent environm an of creation the rather and while the Right wants less state The . unleashed be can forces creative and thousand LA entrepreneurs can bloom, latter is typifed by the faddish book of Hemando de So1o, The Other Path, which posits that the l ack of enforceable property rights is really the sole crux ofthe problem s (de Soto 1 989). Another argument worth re-mentio ning is that cultural difference dreaded the to contrast in eurship entrepren and , explain Asian savings, education fatal legacy of the Spanish and the imperialis m of the US (Harrison 1 992). We have already dispensed with many ofthese arguments in the previous chapters. The easiest one is the cultural argument . Given the backward nature of the Chinese economy until the last decade, as well as the continuin g backward state of a number of other Asian economies, such as Myanamar, and the stratified timing of EA and SEA growth in tune with policy changes, this is obviously a ridiculou s claim. The new growth spurt of China and Vietnam after adopting the practices of their peers shows that it is the EA model, and not the culture or particularities of the place, that works. We could add to this the fact that LA entrepreneurs have thrived in both the US and Europe, as well as under some circumstances within the region itself, such as Chile. In terms of labor repressio n and authoritarianism, there is no doubt that this has been a feature ofEA developm ent. We have also shown that Latin American countries with strong institutions, particularly Chile and Costa Rica, have developed faster, yet still lack sustainable means of providing equitable and high employment­ providing growth . There is also little evidence for the feasibility of a ful l-scale top , down or bottom up revolution in LA, and even where revolutions have succeeded are countries nary revolutio the that evidence no is there namely Cuba and N icaragua, able to create sustainable growth. However, both the Left and the Right's focus on the state and the elite classes points to some key differences between Latin American and East Asian elites. East Asian elites seem to be more nationalis tic and long-tenn in their views. Though they are also surrounded by corruption scandals, particular ly in Southeast Asia, it does not seem to be as destructive to the national interest as in LA. Though this author has neither the expertise nor the space to ful ly explain A EA culture, we can take a closer look at some of the more troubling aspects of L it. behind politics the and making economic decisionExplanations Why Has LA Industrial Policy Been Dysfu nctional? Political

Existing Explanations

There have been four dominant approaches to a political understanding of Latin American industrial policy. The first set of literature looks at the weakness of the Latin American bourgeoisie as a partial explanation for the classes' inability to

Sectoral and Political Foundations for a New Industrial Policy Path for LA


transfonn their countries, including agrarian reform (as discussed b y Hidrobo, 1 992). The second set of l iterature looks at class analysis and bourgeois-state relations (for example, Conaghan, 1 988). This often includes foreign actors such as multinational actors, as in the case of "the triple alliance" of Evans' famous study of Brazil (Evans 1 979). Some authors, following the lead of Cardoso and Faletto, use a historical­ structural approach to emphasize the interplay between long-run structures, including evolution in changes in the mode of production and class relations to explain industrialization (Cardoso and Falletto 1 979). This literature is i m portant in recognizing the ongoing external dependency and the long-term inequalities and divisions within Latin A merican society, which explain some of the inconsistencies and incompleteness in Latin America's industrialization. However, it works on such a general level that it is difficult to test out, since the history of Latin American industrialization is relatively short in tenns of structural change. Moreover, it misses important differentiation among different Latin American cases, time periods, and within economic classes. A third set of literature (Hidrobo 1 992) follows the classic M ilner, Gourevitch, and Rogowski school of domestic political economy in emphasizing the role and lobbying ofinterest groups for industrial policy. This school therefore sees more fissures within the industrial class than a class-based analysis. I t would perhaps more convincingly explain the change from l S I t o neoliberalism as the result of i ncreases in the strength and greater opportunities for international ly as opposed to domestica l ly-oriented producers. The strength of this literature is also in pointing out that industrial owners and the traditional agrarian elite may have some overlap in their membership, thus explaining why industrialization in Latin America could proceed without support for agrarian reform. However, it would be important to recognize the role of more pure ly pol itical motivations, and the machinations of the anned forces in industrial ization. Furthennore, as I pointed out ( Hira 1 998) earlier, the interest group model has a difficult time explaining change absent a major crisis, thus underestimating the agency aspects of policy decision-making, and therefore the range of possible outcomes, and, consequently, the possibil ities for change. Finally, there is a wide and growing body of literature that can be considered part of the policy dialogue. This school of thought spans paradigms, from structuralism to neoliberalism, in focusing on empirical analysis to evaluate and prescribe industrial policies for particular situations. Most international organizations work in this avowedly non-political vein. Within the actors themselves, there is a great deal ofnegotiation. The state includes the executive and legislative branch, each with pecuniary interests. The trade ministry, the f nance/treasury, the defense, the investment bank, the industry, agricultural, and education ministries will also have different interests. The public sector on a national level will be challenged by counterparts on regional and local levels. There will be different companies, some large, some small, some internationalIy- and some domestically-oriented. There will be differing degrees of ties to foreign companies and to government contracts and assistance. Labor is also divided by region, skill set, and particular union. Academic institutions, especially l arge universities conducting research and development (with ties to private industry), will have a particularly heavy stake in I P.

1 22

An East Asian Modelfor Latin A merican Success

Sectoral and Political Foundationsfor a New Industrial Policy Pathfor LA

1 23

What is the best way to explain these political relationships? The obvious answer is that it depends on the particular issue within IP and which actors are involved. What we can do here is sketch out some of the key areas of concern which seem to explain the problematique [complex of problems and issues] of LA IP, and why the lessons of EA are so resisted.

development of clear and consensually-based national strategies for development, including planning for regional sectoral clusters and providing the long-term stability and support to ensure success. However, the problems for establishing a healthy LA IP go well beyond these technical aspects, as daunting as they are. In the fnal sections, we review some of these deeper problems of LA society.

The NeedJor National Innovation and Technology Systems in Latin America

Lack ofLeadership, Solidarity

We have seen that the key to economic growth in the modem world economy is innovation ofnew manufacturing products. As we pointed out in Chapter 2, analysts recognize that innovation can occur in process as well as product. Late entry into industry involves heavy and increasing costs over time as learning and fnancial accumulation within nations, industries and frms creates a cumulative advantage. Therefore emerging sectors of technology, such as biotechnology, seem to be attractive areas for a developing country to target. As is the case currently with China's strategy, Korea and Taiwan worked their way up the technology ladder, beginning with easier components and relying upon cheaper labor costs before tackling higher technology products and eventually innovating their own products . The logic behind a cumulative strategy seems sound in the sense that developing the institutional and human capacity for innovation in one area based more on comparative advantage (cheap labor) will create know-how for more complex production lines later. In fact, perhaps the hardest knowledge to master is to develop retail networks so that one's products can compete on US and European shelves. In the case of Korea and Taiwan, electronics and electrical machinery provided the entree into IT. Most writing about LA economies is centered squarely on macroeconomic problems, we have pointed out that the microeconomic perspective has been neglected.6 As we pointed out in Chapter 2, what is really needed is a national innovation system in each Latin American country. It is the development of the university­ industry-finance-government linkages that lead to the development of internationally competitive national champions in key sectors. What is sorely needed is a series of careful sectoral studies for each country, that consider the institutional bases and possibilities for entree from the perspective of each nation.7 These could lead to the

In this section we can begin to tackle the question of why LA leadership seems to be dysfunctional in its political decision-making. What we can do here is suggest some patterns in the nature of the political relationships, since each situation will be idiosyncratic. However, it is clear that LA elites do not seem to have the same f delity to long-term national interests as the Koreans or Taiwanese do. The $billions in capital flight from LA alone would be more than enough to spark an investment revolution in the region, creating jobs throughout the local economies. As we saw in Chapter 3, the basic unwillingness of Latin Americans to pay taxes or to reform their middle-upper class government spending systems is symptomatic ofa wider inability to work towards long-term solutions to nagging problems such as land reform. We also saw in Chapter 3 that there is a major difference in the levels of investment in human capital in the region, reflecting a much more profound differentiation between the