<%BANNER%>

Reassessing the Impact of Institutions on Economic Reform in Brazil


PAGE 1

REASSESSING THE IMPACT OF INSTITUTIONS ON ECONOMIC REFORM IN BRAZIL By BRANDON KNOX A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2001

PAGE 2

Copyright 2001 by Brandon Knox

PAGE 3

To my parents, who have always shown love and support

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank the Center for Latin American Studies at the University of Florida for the financial assistance provided to me during my time as a graduate student. My experiences at the Center have exceeded all my expectations and I am truly grateful to all the faculty and staff with whom I have had the pleasure to work. The Center awarded me a Title VI fellowship for the study of Portuguese at the Instituto Brasil Estados Unidos in Rio de Janeiro during the summer of 1999, which b egan what I hope will be a life long relation with this wonderful country. During the summer of 2000 the Center for International Business Education and Research (CIBER) provided me with an award to study in the newly formed Business in Brazil program at the Pontifcia Universidade Catlica do Rio de Janeiro. I would also like to thank the members of my thesis committee, David Denslow, Terry McCoy, and Philip Williams. I want to extend a special thanks to Terry McCoy. From the beginning of my inquiries into the MALAS program at UF, he displayed an incredible degree of attention and helpfulness. I had the privilege to work as his graduate research assistant for the duration of my time in the MALAS program, and my work with him has benefited me immensely both academically and professionally. I am extremely grateful to him for his advice and guidance. Finally, none of what I have done would have been possible without the unfailing love and support of my parents, Steve and Rhinda Knox.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ iv ABSTRACT ................................ ................................ ................................ ...................... vii CHAPTERS 1 INTRODUCTION: INSTITU TIONS AND POLITICS ................................ ................... 1 The Brazilian Puzzle ................................ ................................ ................................ ....... 2 Theoretical Framework ................................ ................................ ................................ ... 6 Political Science and Institutionalism ................................ ................................ ......... 8 Institutions and the Study of Latin American Politics ................................ .............. 12 Structure of the Study ................................ ................................ ................................ ... 19 2 THE ROLE OF THE STATE IN THE BRAZILIAN ECONOMY ............................... 21 Early History: Economic Cycles and External Vulnerability ................................ ....... 24 Vargas and the Origins of State Capitalism ................................ ................................ .. 27 The Military Regime: From Boom to Bust ................................ ................................ ... 34 Democratization and Demise of the State Led Model ................................ .................. 39 3 BRAZILIAN POLITICAL INSTITUTIONS ................................ ................................ 43 Political Institutions and Economic Reforms ................................ ................................ 47 The 1988 Constitution and the Reform Process ................................ ........................ 48 Federalism ................................ ................................ ................................ ................. 49 Presidentialism ................................ ................................ ................................ .......... 54 The Party System ................................ ................................ ................................ ...... 57 The Electoral System and Congress ................................ ................................ .......... 60 Conclusion ................................ ................................ ................................ .................... 61 4 ACCOUNTING FOR ECONOMI C REFORMS IN BRAZIL ................................ ....... 64 The Changing Brazilian Political Economy ................................ ................................ .. 65 The Vagaries of Economic Reforms in the New Republic ................................ ........... 66 Cardoso and the Economic Reform Agenda ................................ ................................ 70 The Big Five Constitutional Amendments of 1995 ................................ ...................... 73 Accounting for Economic Reforms in Brazil ................................ ............................... 76 Timing ................................ ................................ ................................ ....................... 77 First vs. Second Generation Ref orms ................................ ................................ ...... 79 The Power of Ideas ................................ ................................ ................................ ... 82

PAGE 6

vi Influence of Political and Economic Variables ................................ ......................... 84 Conclusion ................................ ................................ ................................ .................... 87 5 CONCLUSION ................................ ................................ ................................ ............... 88 Main Findings ................................ ................................ ................................ ............... 88 Suggestions for Future Research ................................ ................................ ................... 92 BIBLIOGRAPHY ................................ ................................ ................................ .............. 95 BIOGRAPHICAL SKETCH ................................ ................................ ........................... 103

PAGE 7

vii Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts REASSESSING THE IMPACT OF INSTITUTIONS ON ECONOMIC REFORM IN BRAZIL By Brandon K nox December 2001 Chairman: Terry L. McCoy Major Department: Latin American Studies During the1980s and early 1990s Brazil, like much of Latin America, passed through a period of economic crisis, as the country struggled to pay its foreign debt and in flation reached four digit levels. While most large countries in Latin American began to enact structural, or neoliberal, reforms in response to this crisis, Brazil was often seen a laggard. Political scientists looked at the countrys political instit utions for reasons for Brazils delay in passing economic reform. This emphasis on political institutions mirrored a general trend in political science, which is often referred to as the new institutionalism. The problematic design of Brazilian politica l institutions, such as federalism, presidentialism, multipartism, the constitution of 1988, the electoral system, and Congress, played a role in Brazils late start with economic reform. However, extensive reforms have been implemented in recent years, m ost notably during the presidency of Fernando Henrique Cardoso. Despite the extensive economic reforms that have been

PAGE 8

viii passed since the mid 1990s, political scientists, especially those from North America, still stress the deleterious nature of Brazils in stitutional design. I argue for a reassessment of the impact of institutions on economic reform in Brazil. Specifically, I look at four factors that help to account for economic reform in Brazil: the timing of reform; the type of reform proposed (first versus second generation reforms); the role of ideas and ideology among the political class; and the role of economic and (non institutional) political factors. Collectively, these factors help to explain how economic reforms have been so extensive in rec ent years, in spite of the problematic design of the Brazilian political system.

PAGE 9

1 CHAPTER 1 INTRODUCTION: INSTITUTIONS AND POLITICS This study will examine the relationship between Brazils political institutions and efforts to implement economic reforms. While there have been attempts to restructure the countrys economy since the advent of th e New Republic in 1985 1 this study will concentrate on developments during the 1990s, principally the presidency of Fernando Henrique Cardoso (1995 present). There have been numerous studies analyzing how certain institutional features of the Brazilian p olitical system make it difficult to implement needed economic (as well as political and social) reforms (Bresser Perreira 1996; Fleischer 1998; Lamounier and Bacha 1994; Mainwaring 1999: chap. 10; Roett 1997; Selcher 1998; Stepan 2000). But this focus on the institutional obstacles to reforms in Brazil generally seems to ignore the extensive economic reforms that have occurred to date. As a result of this perspective, there has been little systematic examination of how such reforms have been possible. Th is study does two things. First, it reviews the current literature on Brazilian politics, most of which uses an institutionalist perspective. This literature is, in general, highly pessimistic about the possibilities for deep reforms in Brazil because of the 1 The New Republic refe rs to the political regime that has been in place in Brazil since the end of military rule. This term is employed almost exclusively within academic circles and is not used frequently by the Brazilian public. There is much disagreement about the extent t o which the New Republic meets the criteria of a consolidated democracy (see Linz and Stepan 1996: chap. 11). Regardless of ones position in this debate, however, all cite 1985 (the year in which the first civilian president assumed office) as the beginn ing of the New Republic.

PAGE 10

obstacles emanating from the political system. Although the first decade of the New Republic was characterized by a litany of unsuccessful stabilization plans and piecemeal economic reforms, the latter half of the 1990s produced extensive changes. M ost analyses see these advances as anomalies, instances in which Cardoso has been able to overcome political institutional constraints to advance painful but necessary changes. Second, this thesis will highlight the shortcomings of the institutionalist cr itique of Brazilian politics. I contend that Brazil is still plagued with numerous problematic institutional arrangements, yet recent developments force us to reassess their effect on economic reform. After reviewing the course of the reform agenda durin g the four presidencies of the New Republic, I will discuss what factors help explain how economic reforms have been implemented successfully in Brazil. Specifically, I analyze four factors: the timing of reform; the type of reform proposed (first versus second generation reforms); the role of ideas and ideology; and political and economic pressures. The Brazilian Puzzle At the beginning of the 1990s analysts considered Brazil a laggard in implementing structural economic reforms, going so far as to bra ndish it the sick man of Latin America. ( Forbes 1993). For most of Latin America the 1980s had been a lost decade during which living standards fell precipitously and countries struggled to meet crushing debt obligations. At that time all countries i n the region, with the sole exception of Colombia, defaulted on their debt and most were forced to seek assistance from the U.S. government or from international financial institutions (IFIs) led by the International Monetary Fund and the World Bank. In o rder to avert economic crises like that which

PAGE 11

3 struck the region during the 1980s, economists, especially those from IFIs and the U.S. government, began to advocate structural, or neoliberal, reforms. According to John Williamson (1990) of the Institute for International Economics, one of the leading think tanks advocating neoliberal reforms, countries in Latin America should pursue nine policies to stabilize and restructure their economies. These measures included: (1) fiscal discipline, (2) changing pr iorities for public spending, (3) tax reform, (4) financial liberalization, (5) competitive exchange rates, (6) trade liberalization, (7) promotion of foreign direct investment, (8) deregulation and (9) privatization of state owned enterprises (SOEs). T he core of the Washington Consensus was the reduction of the role of the state in the economy. During the Great Depression many political leaders in Latin America, most notably in the larger republics such as Brazil, Argentina, and Mexico, began to advo cate an activist role for the state in order to reduce their dependence on export earnings. Although policies varied greatly, the core of this activism lay in the policy of import substitution industrialization (ISI). ISI entailed the state working to fo ster, through a variety of incentives, the development of industries for goods that traditionally had been imported. The creation of SOEs was an integral part of ISI in most countries, as the state came to assume control of the commanding heights of the e conomy. Over time, however, many of these enterprises became highly inefficient and instruments of political patronage. Many economists argued that misguided policies pursued during the era of state activism following the Second World War were in part res ponsible for the travails of the 1980s. While many countries in Latin America began to adopt neoliberal measures

PAGE 12

4 during the late 1980s, Brazil had progressed little towards reforms in these various areas by the early 1990s. When examining the reasons for Brazils delay in implementing reforms, increasing attention was given to political obstacles. First, in 1985 the armed forces, which had been in power since 1964, allowed for the transfer of power to a civilian (though not popularly elected) president, J os Sarney. It was not until the election of Fernando Collor de Mello in 1990 that the chief executive was directly elected, the first direct presidential election in the country since 1960. Due to the transitional and delicate nature of politics at tha t time, some argued it was extremely difficult for a newly installed civilian regime to implement extensive economic reforms, which would likely entail a high degree of austerity for much of the population (Sola 1994). Second, many pointed to institutiona l aspects of the Brazilian political system that hampered the reform process: the new constitution of 1988 and its nationalist economic mandates, the federal system, the weakness of political parties and the party system, and the electoral system, among ot hers (Lamounier and Bacha 1994). Economic reform became all the more pressing as inflation continued to accelerate during the early 1990s. Inflation had surpassed four digit figures by 1989 and worsened during the early 1990s (Rigolon and Giambagi 1999). Furthermore, since the country had the largest economy in Latin America and was a major trading partner with its neighbors, the state of the Brazilian economy affected events beyond its own borders. While President Collor did successfully undertake refor ms in certain areas, notably with the liberalization of trade and the beginning of privatization of SOEs, it was not until Fernando Henrique Cardoso was elected president in 1994 that a coherent economic

PAGE 13

5 reform platform was pursued. During Cardosos tenur e Brazil has deepened its privatization program, record levels of foreign investment have entered the country, and inflation, which had exceeded an annual rate of one thousand percent during the early 1990s, has been under ten percent for several years. I n sum, Cardosos achievements in the economic realm have been extensive. These reforms are all the more noteworthy in light of the political obstacles that the president had to overcome to achieve them. Most analysts agree that Brazils difficulties with passing economic reforms during since the return to democracy in 1985 had their roots in the countrys political institutions. Yet, as Cardosos presidency attests, it has been possible to implement an array of reforms in Brazil in spite of the institutio nal barriers posed by the countrys fractious political system. In 1995 he secured several important amendments to the constitution that ended the states monopoly over certain sectors of the economy (mining, coastal shipping, gas distribution, and, in a different form, oil exploration) and also eliminated the distinction between domestic and foreign capital. This legislation allowed for the subsequent privatization of SOEs in these sectors and helped to attract record levels of foreign investment into Br azil. The constitutional amendments of 1995 mentioned above, also known as the Big Five, will provide a case study for analyzing not how political obstacles confound reforms in Brazil, but rather how these obstacles can and have been overcome. In orde r to address this question, it will be necessary to examine the role of institutions in politics. The following section provides an overview of the new institutionalism literature in political science and how this school of thought pertains to economic reforms in Brazil.

PAGE 14

6 Theoretical Framework Since the 1980s political scientists have paid increasing attention to institutions and the ways they affect politics. According to Bo Rothstein (1996: 135 6), Political scientists ask three different but interr elated questions about political institutions. One is normative: Which institutions are best suited for creating good government and societal relations. The other two are empirical: What explains the enormous variation in institutional arrangements? A nd what difference those differences make for political behavior, political power and the outcome of the political process? In this study I will explain how the design of many Brazilian political institutions has impeded the passage of reform legislation. At the same time, I will also argue that the institutionalist literature does not adequately take into account the extensive reforms that have been implemented to date. Such successful reforms force us to reassess the common notion that Brazils fractio us political system makes it extremely difficult to implement contentious measures. Institutionalism is a common framework with which scholars have analyzed the political economy of economic reforms in Brazil because it is an intermediate level of anal ysis of political phenomena (Just 1999: 17 19). A macro structural framework is inadequate in this instance because of the dependent variable being examined, namely, the implementation of specific economic reforms. A structural analysis would perhaps be useful if one were performing a cross country analysis of economic reforms. Structural factors in particular countries could help to account for why economic reforms began at a certain time. In fact, such a comparison with Brazil and other Latin America countries could help elucidate the larger economic factors at play in this process. Structural analyses still must account for the processes that shape outcomes. In this sense, the study of institutions is apposite, for institutions shape behavior and

PAGE 15

7 s tructure preferences. An institutional analysis also resists the temptation to reduce outcomes to an individual level. Some scholars of Brazilian politics, such as Fleischer (1998), view the outcomes of the political process as contingent on the idiosync rasies of key politicians and factors unique to certain times. While all political outcomes depend to some degree on idiosyncratic factors, in the end such methodological individualism does not yield any larger patterns of behavior. By allowing for an im mediate level of analysis between macro structures and individual particulars, institutionalism is the most common methodology for recent studies of economic reforms. For Douglass North (1990:1), institutions, in their broadest sense, are the rules of th e game in a society or, more formally, the humanly devised constraints that shape human interaction. In consequence, they structure incentives in human exchange, whether political, social, or economic. According to this definition an institution could i nclude cultural customs, codified laws, formal organizations, or any of a plethora of other things. However, if the operational definition of an institution were kept so broad as to signify cultural norms and other informal patterns of behavior, it would be practically impossible to identify which institutions affect certain political outcomes. For our purposes here, then, it is important to specify what constitutes a political institution. According to Margaret Levi (1990: 405), institutions are the f ormal arrangements for aggregating individuals and regulating their behavior through the use of explicit rules and decision processes enforced by an actor or set of actors formally recognized as possessing such power. In this study, institutions will ass ume this narrower definition offered by Levi. Hence, in my discussion of the Brazilian political system I will discuss the constitution of 1988, the federal system, presidentialism, and the party and electoral

PAGE 16

8 systems. These institutions structure politi cal processes and frame politicians behavior, thus helping explain political outcomes. However, it is important to note that institutions alone cannot account for outcomes. In fact, one of the great weaknesses of the literature on Brazilian politics is that it often overemphasizes the explanatory power of institutions. Political Science and Institutionalism Modern political science arose during the late nineteenth and early twentieth centuries. At that time attention was given primarily to the forma l structures of government, such as legislatures, constitutions, and the courts. In many ways, during its early days the discipline was more descriptive than analytic; many studies compared the political and legal systems of different countries. This ol d institutionalism, as it is known, saw political institutions as dependent, rather than independent, variables (Just 1999: 18). Following World War II scholars began to question the utility of the prevailing institutionalist paradigm. Rather than solel y examining formal governmental structures, increasing attention began to be paid to the micro foundations of politics, that is, individual behavior. This focus on political behavior subsequently became known as the behavioral revolution. Behavioralism eschewed the emphasis on the formal institutions and instead sought to understand individuals preferences, which behavioralists believed could be surmised by examining observable behavior. In addition to behavioralism, other schools of thought within political science, notably Marxism and structural functionalism, sought to discredit the old institutionalist approach. These two frameworks of analysis were not mutually exclusive, nor did either in and of itself represent a unified way of analyzing pol itics. Since its beginnings, Marxism has undergone substantial refinement and evolution, yet despite the conceptual

PAGE 17

9 differences that separate the various strands of Marxism they all criticized the traditional focus on institutions as inappropriate. Since the state was the expression of the interests of economic elites, political institutions were simply the means by which these elites (the bourgeoisie) dominated the masses. Institutions did not have any explanatory power of their own, but were epiphenome nal. Structural functionalism, on the other hand, held that the systemic needs of the social system (for political stability or social stratification, for example) tended to produce political institutions that were more or less automatically functional i n relation to those needs (Rothstein 1996: 139). Starting in the 1980s, political scientists began to reassess the import of political institutions. From Latin America to Eastern Europe to Asia, politics across the globe were in transition. Along wit h these rapid developments there also came a growing dissatisfaction with the explanatory power of behavioralism and Marxism and other structural theories. According to Rothstein (1996: 141), this renewed interest in institutions took place across diverse subfields of political science, as well as in other social sciences 2 This renewed focus on political institutions is commonly referred to as the new institutionalism. Unlike the old institutionalism, the new variant holds that institutions themselves have explanatory power. One well known collection of essays urged for bri nging the state back in (Evans et al 1985), urging scholars to move beyond seeing the state as the black box it had become during the height of behavioralism and Marxism. The renewed focus on political institutions has to some degree turned upside down behavioral, group theoretical or Marxist modes of explanation 2 For example, Selznick (1996) and Bardhan (1989) describe how institutionalism influenced organizational analysis and economics, respectively.

PAGE 18

10 stressing social and economic forces as explanations for political processes (Rothstein 1996: 144). Rather than a unified mode of thinking, new institutionalism in fact is a catch all term for a variety of approaches. As Immergut (1998: 5) explains, The new institutionalists do not propose one generally accepted definition of an institution, nor do they share a commo n research program or methodology. In fact, three separate branches of scholarship rational choice, organization theory, and historical institutionalism all lay claim to the label, seemingly without adhering to an overarching theoretical framework. Rati onal choice theories take their theoretical underpinnings from microeconomics. While there are several variants of rational choice theory, most of them share the basic assumption that individuals have fixed preferencesbehave entirely instrumentally so a s to maximize the attainment of these preferences, and do so in a highly strategic manner that presumes extensive calculation (Hall and Taylor 1996: 944 5). According to rational choice theories of politics, actors preferences are shaped by the institut ions within which they interact with others. Preferences are exogenous, shaped primarily by the incentives presented by certain institutional arrangements. Over time the institutions that come to predominate do so because they maximize the utility of the greatest number of actors involved. As discussed in Chapter 3, rational choice is often employed, either explicitly or implicitly, in explaining how political institutions in Brazil provide certain incentives for politicians to behave in ways that hinder the passage of reforms. Rational choice has created enormous controversy within political science because of its extensive use of economic models in the analysis of political behavior, and many justifiable criticisms have been laid against it (Green and Shapiro 1996). One criticism is that rational choice perspectives are static, and hence do not help explain

PAGE 19

11 dynamic situations. What is important to draw from rational choice institutionalism for this study, however, is that it can help to explain polit ical behavior when the rules of the game have been fairly well established. This criterion has been present in Brazil during the period under examination, namely since the promulgation of the 1988 constitution, which established many of the important in stitutions under consideration (Mainwaring 1999). A second institutionalist approach, organization theory, is also referred to as sociological institutionalism. Instead of viewing individuals as calculating utility maximizers, sociological institutionali sm argues that institutions must be examined within their particular cultural context. Institutions are more than means towards maximizing ones utility; they are primarily an expression of cultural and societal norms. Because of this outlook, sociologic al institutionalism seeks explanations for why organizations take on specific sets of institutional forms, procedures or symbols (Hall and Taylor 1996: 947). In contrast to rational choice theories, sociological institutionalism holds that actors prefe rences are not the result of the incentives created by institutions. While this approach is more sociological than political, and is concerned primarily with how institutions form and change over time, it can help explicate important differences in Brazil ian institutions vis vis other countries, thus lending insights into why Brazilian politics is so seemingly fractious. The third and final approach is historical institutionalism. Adherents of this perspective emphasize the role of institutional choic es made early in the development of policy areas, or even of political systems. The argument is that these initial choices (structural as well as normative) will have a pervasive effect on subsequent policy

PAGE 20

12 choices (Peters 1996: 210). Initial institutio nal choices are important, argue historical institutionalists, for even if attempts are later made to alter the institutional structure, it will be difficult if not impossible to change it; this phenomenon is referred to as path dependency Historical ins titutionalism has enormous explanatory power when examining contemporary Brazilian political history, especially in this case study. The transition from military to civilian rule during the 1970s and 1980s profoundly influenced the existing political orde r, from the constitution to the party system to the economic order. As will be explained in the following chapters, many of the legacies of this transition have been deleterious for the Brazilian economy (and society) and have proven highly difficult to c hange. In this study I will not employ one single institutionalist perspective, but instead will implicitly draw upon all three branches to help to explain the ways in which institutions do and do not influence outcomes. Institutions and the Study of L atin American Politics Regional studies, which generally fall under the rubric of comparative politics, often follow the larger trends within political science yet modify them to the particular characteristics of the political systems being examined. The study of politics in Latin America certainly conforms to this pattern. Just as in the United States, early political studies of countries in the region focused principally upon formal government structures (Ames 1999). Over time, however, it became clear that much more than formal institutions were at play in the politics of the region. As a result, less attention was devoted towards institutions, as scholars began to utilize other methods of analysis. While in the United States behavioralism shifted att ention away from institutions starting in the 1950s, in Latin America a growing critical perspective turned attention away from

PAGE 21

13 institutions towards issues such as domestic class struggles and global economic inequalities. The intensification of the Cold War and the Cuban Revolution had a profound impact on the intellectual mood of Latin America. Marxism, which has deep intellectual roots in the region, provided an appealing and effective framework with which to analyze both domestic and international po litics. Variants of Marxism, such as world system and dependency theories, also gained popularity starting in the 1960s. This was a time of great social and political upheaval in Latin America. While leftist parties and revolutionary movements made inro ads during the 1950s and early 1960s, by the early 1970s most countries in the region had succumbed to some form of authoritarian rule. Countries in most of Central America and the Caribbean had a tortuous political history, with authoritarian rule being almost a constant throughout their history. In contrast, the countries of the Southern Cone of South America were traditionally more democratic and economically developed. As Brazil and then Chile, Argentina, and Uruguay all fell to military rule, schola rs began to examine the causes for the breakdown of democracy. Some scholars attempted to explain this wave of authoritarianism by exploring the structural economic relationship between the region and the developed world. To many the events of this peri od served as proof of the effects of capitalist expansion into the developing world. Guillermo ODonnell (1973; 1978) saw a connection between bureaucratic authoritarian regimes in the Southern Cone and these countries level of dependent capitalist d evelopment. ODonnell believed there to be a structural relationship between these countries position in the international economy and their authoritarian political systems; there was a mutual indispensability between the

PAGE 22

14 exigencies of international c apital and the bureaucratic authoritarian state. Peter Evans (1979), in another seminal work of the period, spoke about the alliance of national, state, and multinational capital in Brazil. According to Evans this triple alliance, in which multinationa l and state capital predominated over Brazilian private domestic capital, helped to explain the countrys recent authoritarian political history. Works such as those by ODonnell and Evans, which looked for structural variables in the rise of Latin Ameri can authoritarianism, were more common than micro behavioral analyses. Most political scientists studying Latin American politics saw the state as a dependent variable; Marxists saw it as the reflection of class relations, while dependistas saw it as a re flection of a countrys position in the global economy. Alfred Stepan, in The State and Society: Peru in Comparative Perspective (1978 a ), viewed the Peruvian state as an independent variable, arguing that the state was itself an autonomous actor rather tha n the mere aggregation of the dominant classs interests. This study helped spark a debate about the state and its autonomy vis vis social forces like the military and business. Many scholars criticized structural studies as methodologically weak, fo r they failed to establish empirical links between dependent development (another problematic concept) and political authoritarianism. With the democratization of most of the region during the 1980s, structural analyses came under further attack. Besides the methodological weaknesses inherent in dependency theory and Marxism, these theories were of little value in analyzing the substantive functions of government. In essence, these schools of thought had little to say about institution and institutional design. What could Marxism say about how to organize democratic political parties? An analysis of

PAGE 23

15 the world system could reveal structural economic inequalities, yet it would be unable to elucidate debates over the relative merits of presidentialism and parliamentarism, electoral system design, and other such substantive issues. According to Barry Ames (1999: 221), Institutions are in. Not that Latin Americanists ever ignored them completely, but in the old days of military authoritarianism, power and international influences seemed more important than governmental structures. With the transition to, and in most cases consolidation of, democratic rule, a sizable number of publications focusing on Latin American political institutions appeared during the past decade. Economic stabilization and restructuring was a theme present in much literature, and many scholars, such as those in the edited collection by Haggard and Kaufman (1992), stressed the role of institutions (both domestic and international) in influencing the various paths to stabilization followed by Latin American countries. Yet Latin America was not the only region undergoing the dual processes of economic restructuring and political liberalization. During the 1980s Eastern Europe was al so tackling many of the same issues as Latin America. This led many scholars to undertake comparative studies between the two regions, often focusing on the relationship between political and economic reforms. Przeworski (1991), in one of the most cogent of these comparative studies, uses a game theoretic perspective to explain potential courses for economic and political reforms in Latin America and Eastern Europe, dedicating much attention to institutional factors in these transitions towards market dem ocracies. Another comparative study, edited by Lijphart and Waisman (1996), examined the role of institutional design in new democracies, examining

PAGE 24

16 specifically electoral systems, executive legislative relations, and the type of economic market pursued fo llowing democratization. One of the most widely debated areas of institutional design has centered on presidential systems. Juan Linz (1990) spearheaded an extensive debate on the merits of the presidential system. Linz contends that presidential systems of government are less attractive than parliamentary systems for various reasons. Since a president is elected separate from the legislature, it is common for a chief executive not to command control of the legislative body. When the executive and legis lative bodies are highly polarized, this can lead to a crisis of dual democratic legitimacy since both bodies were democratically elected and therefore can claim a mandate from the public. Furthermore, because presidents are elected for fixed terms, thi s creates rigidity during times of great dissatisfaction with executive leadership. Whereas in parliamentary systems executives can be removed with a simple vote of no confidence, it is generally much harder to remove a sitting president. Linz and Arturo Valenzuela (1994) explored the failures of presidential democracy in Latin America and inspired other studies of presidentialism in Latin America, such as the edited volume by Mainwaring and Shugart (1997). Their survey of presidentialism across the reg ion shows how it has been less problematic in some countries than in others. As will be explicated in greater detail in Chapter 3, presidentialism in Brazil is complicated by the countrys federal system and by the weakness and multiplicity of parties. I n line with this general trend in the study of Latin American politics, there has been abundant research on Brazilian political institutions. Being the largest country in the region in both economic and demographic terms, Brazil always warranted the

PAGE 25

17 atten tion of scholars. The transition towards democracy in the country heightened the concerns about institutional design and how it could possibly affect the future course of democracy in Brazil (Lamounier 1989). The deepening of the countrys economic troub les was another impetus for research (Kaufman 1990). While most major countries in Latin American had successfully stabilized their economies by the beginning of the 1990s, Brazil was still suffering from hyperinflation. The persistent problem led schola rs to focus attention on how institutions complicated the process of economic reform (Sola 1994; Bresser Perreira 1996). How did institutions affect economic stabilization and reform in Brazil? Before answering that question, it will first be necessary t o explain the processes and institutions at work in Brazilian politics. This will require a brief discussion of the transition to democratic rule during the 1980s, which resulted in the presidency of Jos Sarney in 1985 and the creation of a new constitut ion in 1988. 3 This new document both established new institutions and complicated existing institutional arrangements (such as the skewed system of representation in the lower house of Congress), a fact that has confounded recent efforts at economic refor ms. For example, the federal system mandated by the constitution forced the federal government to give a larger share of tax revenues to local levels of government, most notably the states, without transferring a commensurate number of fiscal responsibili ties. As a result, the federal government has been forced to continue to provide the same level of services without the necessary fiscal 3 See the special issue of the Journal of Interamerican Studies and World Affairs Brazil: The Challenge of Constitutional Reform, (Winter 1998, Vol.40, No.4) edited by David Fleischer, for an extensive discussion of the enduring problems created by the 1988 constitution.

PAGE 26

18 resources. These transfers to the states and municpios (which are similar to counties in the United States) have bee n a windfall, helping to grease the local political machines that are the heart of Brazilian politics. Not only is the fiscal situation of the central government damaged, but so too is its very authority vis vis the states and municpios Furthermore, as the world economy has become increasingly globalized, this tense fiscal relationship between Braslia and the states has been a persistent concern of international investors, who fear the consequences of recurring, unchecked fiscal deficits on the part of local governments. Beyond the Brazilian system of federalism, the party and electoral systems have created institutional rules of the game [that] have given politicians incentives to cultivate a personalistic relationship with the electorate, to focu s on local constituencies, and to pay attention to state and local politics (Mainwaring 1999: 10). These incentives weaken the coherence and discipline of parties, making legislative efforts cumbersome and highly costly in terms of time and pork. Chapte r 3 will offer a thorough discussion of the Brazilian political system, problems with the design of key institutions, and their effects on reform efforts. As stated above, in this study I will not rely upon one exclusive method of institutional analysis. Instead, throughout I will draw upon all three approaches. My conclusion is that in Brazil it has been possible, despite political institutional obstacles, to implement significant economic reforms. While many reforms have floundered during the past few years, one must take account of successful reforms. This, I conclude, calls for a reappraisal of institutionalist perspectives of Brazilian politics.

PAGE 27

19 Structure of the Study Following the theoretical framework of analysis laid forth here, Chapter 2 prov ides an overview of the economic development of Brazil and the role of the state in the economy. With its history of Portuguese monarchical control, the colony of Brazil developed a heritage that had a profound impact on economic development. Most notice able were the reliance on a few primary product exports and the consequences of such dependence on growth. One of the poorest countries in Latin America at the beginning of the twentieth century, Brazil experienced some of the highest growth rates in the world from 1900 until the debt crisis of the 1980s (Rigolon and Giambiagi 1999). During this period of rapid development, the state came to play a central role in the economy. One of the clearest manifestations of the states active role in development c ame by means of state owned enterprises, which proliferated following World War II. In short, to understand modern Brazilian economic history, one must understand the vital role the state played in development and how and why that role has changed over ti me. As this is a study of political institutions and their effects on the reform process, it is necessary to examine Brazilian political institutions and processes in some depth. This will be the focus of Chapter 3. Many aspects of the countrys politic al system have proven extremely deleterious to the countrys fiscal situation and continue to hamper efforts at reform of the state. According to one common interpretation, the Brazilian state in the post authoritarian period has been not been a powerful Leviathan, but rather a growing source of patronage, favors, and rents for those with the access to such resources (Roett 1997). The constitution of 1988 codified a federal system that forced Braslia to earmark revenues to states and local governments wi thout a commensurate increase in their responsibilities in areas of public spending. This has entailed a fiscal windfall for

PAGE 28

20 state and local politicians, but has created enormous strain on the fiscal health of the federal government. Beyond the fiscal pr oblem emanating from the political system, there is a myriad of features in the party system, the countrys particular form of presidentialism, and electoral system that create hurdles for reforms. In Chapter 4 I will analyze how, in spite of Brazils prob lematic institutional arrangements, economic reforms have been successfully implemented. I will begin with a discussion of the various reform measures put forward by the four presidents of the New Republic. Particular attention will be given to current P resident Fernando Henrique Cardoso, who has been the most vigorous reformer to date. I will also explore one of the most important sets of measures passed by Cardoso at the beginning of his time in office, known as the Big Five. I will conclude the chapt er by discussing what factors help account for economic reforms in Brazil. Specifically, I look at four factors: the timing of reform; the type of reform proposed (first versus second generation reforms); the role of ideas and ideology; and political and economic pressures. Collectively, these four factors help to explain the success of economic reforms in Brazil. Finally, in the conclusion I reiterate the main findings of this study and offer suggestions for future research.

PAGE 29

21 CHAPTER 2 THE ROLE OF THE STATE IN THE BRAZILIAN ECONOMY This chapter will explore the history of state involvement in the Brazilian economy. Since the discovery of Brazil by the Portuguese in 1500 the state (or the crown) has always played some role in the econom y. Until the early twentieth century that role was largely confined to supporting or promoting in some fashion primary products for sale on the international market. During the early twentieth century Brazil began to industrialize, and following the Seco nd World War the state played a primary role in this process. This chapter is arranged chronologically, tracing the changing relationship between the state and the economy from colonial days until the demise of the state led model in the late twentieth ce ntury. (Table 2.1 provides a list of the key political and economic events in Brazilian history since the early nineteenth century.) An understanding of Brazilian economic history is necessary in order to comprehend the enormous economic changes that hav e occurred during the past decade. It will also help one understand why economic reforms are such a politically charged issue. When examining the economic history of Brazil during the past century, three salient themes emerge. The first is rapid growth. Brazil was one of the fastest growing countries during the twentieth century, with only Japan and Finland growing more rapidly between 1900 and 1970 (Rigolon and Giambiagi 1999: 6). Although the 1990s was a period of stabilization and resumed growth in B razil, overall economic growth

PAGE 30

22 Table 2.1 Key Political and Economic Events in Brazilian History Period Year Colony 1808 Portuguese court moves to Rio de Janeiro, which becomes seat of Empire 1820 Royal court returns to Portu gal Empire 1822 Pedro I declares Brazilian independence 1888 Abolition of slavery First Republic 1889 Overthrow of monarchy, establishment of republic Second Republic 1930 Revolution of the tenentes brings Vargas to power 1937 V argas establishes the Estado Novo 1941 Creation of the Companhia Siderrgica Nacional (Volta Redonda) 1942 Creation of the Companhia Vale do Rio Doce (CVRD) 1945 Vargas, under pressure from armed forces, steps down 1950 Vargas elected p resident 1952 Creation of the Banco Nacional de Desenvolvimento Econmico (BNDE) 1953 Creation of Petrobrs 1954 Vargas commits suicide while still in office 1955 Election of Kubitschek, Plano de Metas 1960 Quadros elected Construction of Brasilia completed 1961 Quadros resigns; Goulart assumes presidency Military Regime 1964 Military coup against Goulart 1968 Institutional Act 5 signals start of repressive period 1974 Start of political liberaliz ation ( distenso ) 1975 Second National Development Plan introduced, which intensifies ISI process 1979 Second oil shock 1982 Direct election of governors New Republic 1985 Military leaves power, Sarney assumes presidency 1988 Ne w constitution adopted 1989 Collor elected in first direct presidential election since 1960 1990 National privatization program initiated by Collor 1992 Collor impeached 1994 Real Plan; Cardoso elected president 1998 Cardoso r e elected 1999 Devaluation of the real

PAGE 31

23 during the past two decades was low by historical standards. Second, with economic growth came a concomitant demographic shift, as Brazil was transformed from a largely rural country into a highly urb anized one during the span of about two generations. 1 A third theme is the importance of the state in the process of economic development. Since colonial days the state was a central actor in economic activity. Its role, however, changed dramatically ov er the course of the twentieth century, notably during the rule of Getlio Vargas (1930 45 and 1950 54), as the state began actively to promote industrialization. The economic role of the state assumed various forms, including that of regulator, financier and producer (Baer et al, 1976). During both democratic and authoritarian periods in Brazil, desenvolvimentismo or developmentalism, was an important ideological force, one which called for the state to champion industrialization and foster the grow th of the economy (Lima 1999). SOEs played a key role in this developmentalist strategy, as the state became involved in sectors such as steel production, oil exploration and refining, mining, and electricity production and distribution. However, due to the energy crises of the 1970s and increased indebtedness during that period, the state found itself increasingly incapable of making the investments necessary to sustain growth. This downturn in the fiscal position of the state, coupled with changes in t he international political economy the rise of neoclassical economics and its embodiment in Reaganism and Thatcherism, the decline in financing from IFIs, the collapse of the Soviet bloc, the consequent discrediting of socialism as a guiding force of devel opment set the stage for a reassessment of the relationship between the state and the economy. This reexamination was particularly

PAGE 32

24 marked in Latin America, where most states had played an active role in development since the Second World War. Although t his neoliberal trend reached Brazil later than most countries in the region, in Brazil its effects have been among the most pronounced. As Luiz Carlos Bresser Perreira (1996) explains, the country was facing a crisis of the state and therefore required an extensive reshaping of both the political and economic spheres of life. The result has been a dramatic restructuring of the economy as well as a drive to reform the political system. Privatization of state owned enterprises embodies both of these phen omena perhaps more so than any other reform. Alluding to Mexico, Judith Teichman believes that privatization programs have been the most important way in which the states role has been reduced and redefined (1995: 3). The same could said of Brazil. Ea rly History: Economic Cycles and External Vulnerability At the beginning of the twentieth century Brazil was one of the poorest countries in Latin America, itself one of the poorest regions in the world. Since its initial colonization by the Portuguese cr own during the sixteenth century, Brazil relied on the export of a few primary products for its economic growth. The colonys original name, Terra da Vera Cruz or Land of the True Cross, reflected the religious sentiments that played a role in the expans ion of the Portuguese empire. However, even more important than religious zeal was the quest for wealth. The Portuguese established a series of feitorias or trading outposts, on coastal areas from Africa to Indonesia. Brazil, in 1 According to Baer (1995: 3), the population living in urban areas was 30 percent in 1940, 56 percent in 1970, and 76 percent in 1991.

PAGE 33

25 contrast to the crowns other possessions, did not possess any known mineral wealth nor did it have a large domestic population with which trade could take place. Because of this lack of natural resources, Brazil initially received scant attention from Lisbon. In fact, Portugu ese control of this vast territory initially was extremely precarious. During most of the sixteenth century the Portuguese intermittently fought with the French for control of key ports in the colony. While the original name of the colony had a religiou s origin, it is telling that the name that has endured to this day came from the colonys first major commodity export, pau brasil or brazilwood. Economics, not religion, would be the driving force in the formation of the colony. Brazilwood became an im portant ingredient in clothes dye at that time and grew in abundance in the lush Atlantic rain forest around Rio de Janeiro. The exploitation of brazilwood contributed to the destruction of the Atlantic rain forest and also marked the first commodity cycl e in Brazilian history (Dean 1995) 2 According to historian E. Bradford Burns, From the beginning, extreme economic fluctuation has been a major characteristic of the Brazilian economy and a powerful influence on the course of Brazilian history (1993: 6 3). This fluctuation was directly related to the commodity export cycles that punctuated the entirety of Brazilian colonial history and lasted until the beginning of industrialization. While the extraction of brazilwood provided the impetus for initial e xploration of the new colony, it was not until the advent of sugar cultivation along the southeastern, and especially the northeastern, coasts that modern day Brazil began to be heavily colonized. The fertile 2 In this sense, one sees early roots to two of Brazils recurring problems: a reliance on a small number of primary product exports and a pattern of massive environmental degradation.

PAGE 34

26 lands of the littoral proved to be ideal for t he cultivation of cane sugar. As demand for the sweetener grew in Europe, the Brazilian sugar sector expanded exponentially. With time competitors emerged to challenge Brazilian predominance of sugar production. As plantations became operational through out the Caribbean, Brazils market share slipped, setting the stage for an economic decline. While its is beyond the scope of this study to trace the vagaries of all of Brazils economic cycles, it is worth noting that what happened with sugar would recur with future commodities gold, cacao, cotton, and especially coffee. The structural position of Brazil in the world economy and the tumult it created was to have a profound effect on developments during the twentieth century. In 1822 Brazil achieved its i ndependence from Portugal. Unlike its Spanish speaking neighbors, Brazil was able to wrest control from its former rulers without bloodshed. Furthermore, independence was achieved without territorial disintegration, a far cry from developments in Spanish America. Another difference separating Brazil from its neighbors was its monarchical form of government, as the country was ruled by the house of Bragana (Pedro I and Pedro II) until the latter part of the nineteenth century. During the imperial era, c offee came to dominate the economy and coffee interests became increasingly influential in the nations politics. Slavery, which had long been an integral facet of the coffee economy, was abolished in 1888, which in part led to the deposal of Pedro II in 1889. Following his deposal, the centralized monarchical system was replaced with a highly decentralized republican system. The advent of the republic also marked an important point in Brazilian economic history, as the country became increasingly integr ated into the world economy through the growth of the coffee trade.

PAGE 35

27 In the economic realm, the republican regime created following the overthrow of the monarchy was most concerned with those areas related to export agriculture, especially coffee. Accord ing to Steven Topik, during the early republican era the state began to play an active role in the economy, namely through its involvement in railroads and the creation of financial institutions. Through [its] haphazard, almost involuntary, procedure, c laims Topik (1979: 326), the First Republics 3 State set the precedents that have greatly affected the nature of current interventions. While the Brazilian state did become more involved in the economy during the end of the nineteenth and the beginning of the twentieth centuries, Topik overstates its role. The First Republic was a classical, liberal regime, in which the state rarely became actively involved in the economy. When it did, it pursued activities, namely those related to public goods, which the market failed to provide adequately. To say that the steps undertaken during the First Republic greatly affected the nature of current interventions is to take the argument too far. It was not until the rule of Getlio Vargas that such a claim coul d be substantiated. Vargas and the Origins of State Capitalism The growth of state involvement in the Brazilian economy became pronounced during the era of Vargas, president from 1930 to 1945 and again during the early 1950s. The government Vargas overth rew was based on an alliance between agrarian interests in So Paulo and Minas Gerais ( caf com leite ) and had intervened frequently to support the price of coffee, Brazils leading export during that period, on the international market. The political sys tem established in 1889 had changed little, even as Brazil itself was 3 The republic established in 1889 is commonly referred to as the Old Republic and First Republic. The Vargas era, which began with the Revolution of 1930, ushered in the so called Second Republic.

PAGE 36

28 changing rapidly. With the support of the military and disgruntled political elites in Rio Grande do Sul Minas Gerais, and the Northeast, Vargas was able to lead a nearly bloodless co up. The time of Vargass rule was one of tremendous political and social change in Brazil. As a result of the Great Depression the country was suffering from a decline in world demand for its primary commodities, most notably coffee. At the same time t hat the economy was adjusting to the new international environment, social forces were changing the nature of Brazilian politics. During the 1920s and 1930s nationalism, especially in the economic sense, was on the rise. So too were labor movements, whic h to varying degrees were based on anarcho sindicalism, Marxist, and fascism. These various social forces helped shape the Vargas regime. On the other hand, Vargas worked to contain these emerging social forces so as to lessen their risk to the political status quo. In 1937 he created the Estado Novo an authoritarian, fascist inspired corporatist state. During the Estado Novo the state began to play a growing role in economic development, as some of the first SOEs were created. Under this regime Econo mic, political, and ideological disorganization gradually placed economic policy and planning in the hands of the government, strengthening its interventionism, without threatening the power of the dominant groups. Thus it presented a conservative pattern of modernization which was to characterize t he countrys future deve lopment. (Becker and Egler 1992: 47) This pattern of conservative modernization lent a unique face to economic development in Brazil, for it preserved traditional power structures at t he same time that the state was spearheading economic development. One of the effects of the Great Depression was that it revealed the precariousness of countries that relied on external markets for their economic wellbeing. This situation

PAGE 37

29 forced leaders across Latin America to reexamine their position in the world economy. Most countries in the region had suffered from a deterioration in their terms of trade with the developed countries of North America and Western Europe, as the prices of Latin American primary products stagnated or declined relative to the prices of manufactured goods from the developed world. Ral Prebisch of Argentina, one of the early leaders of the United Nations Economic Commission for Latin America (CEPAL), was the most prominent economist to investigate and criticize this phenomenon. Assailing the theory of comparative advantage the sacred tenet of classical economics Prebisch posited a theory of unequal exchange. According to Prebischs theory, the developed economies benefi ted from international trade over time, while growth in the countries of the periphery stagnated. Moreover, he claimed that this stagnation was linked to poorer countries structural relationship to the industrial countries (Love 1995). Since all Latin A merican countries were still primarily agrarian at the time of the Depression, Prebisch and his followers believed that it was necessary for the countries of the region to industrialize in order to combat the erosion in their terms of trade and, by extensi on, make themselves less vulnerable to external shocks. In order to achieve this industrialization, Latin American leaders decided to raise protectionist barriers in their economies, thus limiting the level of foreign goods and capital that could enter. C ountries began to produce domestically some goods that they previously had imported from abroad, usually light consumer goods. This policy, which became the dominant development model in Latin America following the Great Depression, was known as import su bstitution industrialization (ISI). While this was not a monolithic policy, there was a great deal of convergence in the policies undertaken

PAGE 38

30 across the larger Latin American republics. The level of state investment of the economy grew rapidly, as develop ment banks appeared in many countries with the goal of providing capital for targeted sectors of the economy. 4 Protective measures, such as tariff and non tariff barriers to trade, were enacted in order to protect nascent industries from foreign competiti on. Another key aspect of ISI was the creation of SOEs. Dominant economic thinking today assumes that the state plays a negative role in the economy. By undertaking relatively inefficient activities, the state is believed to crowd out more efficient private activity. However, during the era of state led growth in Brazil the state was viewed as a positive force in the economy. During the Vargas era, when state economic planning began, it was the state elites who championed a growing role for the stat e in the economy (Skidmore 1967: 46). Rather than crowding out private economic activity, state planners believed that the state played a crowding in role by providing cheap credit (often with negative real interest rates) and cheap inputs. Moreover, in those days the growth objective outweighed efficiency concerns (Baer 1996: 366). Baer (1994: 509) identifies three factors that influenced the growth of SOEs in Latin America: (1) the existence of externalities related toindustrialization objectiv es; (2) externalities related to equity considerations; and (3) nationalism. One of the largest obstacles to the advancement of ISI was the unwillingness of local private capital to participate in key long term investments, such as public utilities and c ommunications infrastructure, at levels the state desired. Investment in these sectors was essential if 4 In Brazil some of the most important development banks included the Banco Nacional de Desenvolvimento Econmico (BNDE), created in 1952, and the Banco do Nordeste, created in 1954.

PAGE 39

31 economies were going to be able to industrialize rapidly. The risk involved in many of these activities forced private companies to provide services to limited, mainly urban, areas, which in turn stunted the development of rural areas. Likewise, these sectors had traditionally been controlled by foreign interests, which helped to fuel populist and nationalist sentiments in favor of state involvement. T o remedy these externalities the state took control of activities such as public utilities, nationalizing most of these from the private sector. 5 Suffering from a lack of private domestic capital, many Latin American governments also increased their invol vement in financing particular sectors of the economy. Having this kind of control allowed governments to take equity considerations into account by devoting resources to areas of the economy which had been largely neglected in the past, be they in depr essed regions (such as the Brazilian Northeast) or in particular industries (such as steel production). While the 1930s were a period of increased government involvement in economic affairs, it was during the 1940s that the first of Brazils major SOEs ap peared. For decades there had been discussion about how best to develop a steel industry in Brazil. With the rise of Vargas and the consequent increased influence of the armed forces within the highest circles of the government, the steel question assume d a new urgency. To the military the development of the steel industry was more than an issue of economic development; it was also a question of national security. In 1941, in exchange for a security agreement from Brazil to aid the U.S. against Nazi Ger many, the U.S. government agreed to provide funding for the creation of the Companhia Siderrgica 5 For example, 64 percent of utilities in Brazil were controlled by the private sector in 1962. That figure dropped to 20 percent in 1977, and by 1982 the s tate controlled all public utilities (Baer 1995: 258).

PAGE 40

32 Nacional (CSN). Located on an old coffee plantation at Volta Redonda in the state of Rio de Janeiro, the CSN was one of the earliest and most powerful symbol s of the states championing role if development (Wirth 1970). The company was also one of the largest SOEs in Brazil for decades to come. Closely linked to the steel industry was the mining sector. In 1942 the Companhia Vale do Rio Doce (CVRD) was cre ated, largely in response to the demand for iron ore for the nascent steel industry. In time it became the countrys largest mineral exporting firm, and has been one of the most powerful of the states public enterprises (Baer 1995: 247). By 1985 it ha d become the seventh largest firm in the country and had acquired or created a number of subsidiaries outside the mining sector, including sectors such as consulting, shipping, pulp and forestry, marketing, and fertilizers (Raw 1987). The CVRD was a stat e conglomerate whose development demonstrated that state firms could be highly profitable and could compete according to market forces despite its public ownership structure. In fact, it became the largest mining firm in all Latin America. (It is ironic that it was during the military regime of 1964 85, which seemed at least superficially committed to the market, that CVRD became increasingly politicized. As a result, by the late 1970s the company had become highly indebted and uncompetitive, like most other state enterprises at that time.) During the 1950s more economic activities came under the purview of the state. In 1952 one of the most important state institutions was created, the Banco Nacional de Desenvolvimento Econmico (BNDE). 6 The BNDE was created to serve as the countrys principal development bank for infrastructure, but over time this task expanded

PAGE 41

33 to include a variety of financing responsibilities. Over time the BNDE became the primary shareholder in many SOEs (and some originally priva te enterprises); in exchange for providing a firm with financing, the development bank was given a percentage of ownership of the firm. If some firms encountered financial difficulties, the bank would provide assistance in exchange for greater control. During the early 1950s there were growing public pressure in favor of creating a state oil company. The motto in favor of the creation of a state oil company O petrleo nosso, or The oil is ours underscored the growing sense of economic nationalism i n Brazil during that period. In 1953 Congress mandated the creation of Petrobrs, which was given monopoly powers in the most aspects of oil exploration and refining. According to the historian John Wirth (1970: 217), the Vargas era [1930 1954] saw a mar ked increase in the authority and capacity of the national government. Brazilian policy makers developed a future orientation and a confidence in national purpose that in the broadest sense can be called developmental nationalism. In many ways the creat ion of Petrobrs marked the height of economic nationalism in Brazil, but also the beginning of a decline in Vargass influence in national political life. Threatened by the military, who were about to depose him, Vargas committed suicide in 1954 while st ill in office. The 1950s were also a time of increased price controls in Brazil, as the government attempted to control inflation. These controls gave private firms little incentive to participate in sectors such as utilities, and as a result the state be came in involved in these sectors by default. Starting at this time the government began to create 6 The BNDE was renamed the BNDES, or Banco de Desenvolvimento Econmico e Social, during the 1980s.

PAGE 42

34 SOEs in utilities and communications. The state electric conglomerate Eletrobrs was established in the early 1960s, as was the state telephone company Emb ratel. During the late 1950s, Brazil underwent a period of rapid growth and industrialization. Under the direction of President Juscelino Kubitschek, the country experienced what Thomas Skidmore (1967) has called years of confidence. Kubitscheks Plano de Metas or Targets Plan, laid out development goals for the country that, the president argued, would spur fifty years of development in five. The new national capital, Braslia, was being constructed and enormous investments were undertaken by the s tate in a variety of industries. It also marked the beginning of the countrys automobile industry, today one of the largest in the world. While Brazil grew rapidly during the Kubitschek era, the eventual result of this prodigious growth was a worsening of the countrys balance of payments and an acceleration of inflation. By the early 1960s, the economy was slowing, inflation was worsening, and society was becoming ever more polarized. The Military Regime: From Boom to Bust In 1964 a military coup ov erthrew the government of President Joo Goulart, whose actions had incensed the military and alarmed the middle class (Stepan 1978b). The new military regime continued the past policies of state led developmentalism, yet under far different political con ditions. In the past there was a pattern of state coordination of private domestic capital, public domestic capital, and foreign capital. This pattern was solidified into a triple alliance under the military regime, which promoted the expansion of mult inational and state capital, often to the detriment of private domestic capitalists (Evans 1979). At the same time the government assisted in

PAGE 43

35 the suppression of wages and the weakening of organized labor. As a result of the militarys deliberate developm ent plans the economy began to grow at impressive rates. This led to the so called Brazilian miracle of the late 1960s and early 1970s. Between 1967 and 1973 the country averaged annual growth rates of 12.9 percent. This growth was driven largely by a promotion of exports, which increased from $1.9 billion in 1968 to $6.2 billion in 1973 (Roett 1992: 164). Furthermore, this period of rapid growth was also marked by an expansion SOEs and overall state involvement in economic activity through a variety of mechanisms: Government expenditures . as a proportion of GDP grew from 17.1 percent in 1947 to 22.5 percent in 1973. State enterprises dominated in steel, mining, and petrochemicals. They controlled over 80 percent of power generating capacity and most public utilities. It has been estimated that in 1974, for the 100 largest firms (in value of assets), 74 percent of the combined assets belonged to state enterprises, whereas for the 5,113 largest firms, 37 percent of assets belonged to state enterprises (Baer 1995: 79 80) The impressive growth of the Brazilian economy during the miracle of the late 1960s and early 1970s, then, owed much to the actions of the state. However, this growth came at the cost increased foreign indebtedness. Rapid industrial ization required large amounts of physical and financial capital from abroad. By 1973 Brazil was suffering from a balance of payments crisis due to the huge inflow of imports, usually capital goods, that sustained industrial growth. Growth continued thro ughout the 1970s, but at slower rates and by means of greater foreign indebtedness, especially through commercial (not public) loans. By the beginning of the 1980s growth rates had dipped below zero, inflation accelerated (due in great degree to increases in energy costs) and Brazil slipped into recession with the rest of the world. As happened throughout the entire developing world, the energy crises of the 1970s put Brazils statist

PAGE 44

36 developmental plans under great stress. During this time the state pur sued a course of heavy industry ISI (such as steel production). Moreover, in response to the countrys dependence on foreign sources of energy, several large scale public utilities projects were initiated, including the construction of the worlds largest hydroelectric dam at Itaip. The result of such initiatives was increased foreign indebtedness. It also contributed to the increasing stagnation of the economy. In great contrast to other bureaucratic authoritarian regimes in South America, the military government of Brazil greatly expanded the reach of the state in the countrys economy. In 1974, 39 percent of net assets and 16 percent of sales in the Brazilian economy belonged to state firms. By 1985, those figures had risen to 48 percent and 26 perc ent, respectively (Baer 1995: 257). The energy crises of the 1970s strained Latin American governments abilities to continue at the commanding heights of their economies. Increased energy costs caused an increase in production costs for SOEs, forcing cen tral governments to borrow yet more money from the international financial community. There was no shortage of money to be sought by the Latin American countries. The influx of funds from OPEC nations into the banks of Europe and North American induced t hese banks aggressively to pursue borrowers. This process of using funds from the oil producing countries in loans to developing countries came to be known as petrodollar recycling. Latin America was the preferred destination for these funds, satisfyin g the needs of cash strapped central governments in the region, as well as those of the international financial community. Furthermore, the perception of riskless soveriegn debt provided a strong incentive to lenders to continue making loans Latin America n governments.

PAGE 45

37 Coinciding with the energy crises of the 1970s was en explosion in both the number of SOEs in their level of foreign indebtedness. During the 1970s alone, 251 SOEs were created in Brazil (Bresser Pereira 1998: 65). Furthermore, the importan c of SOEs in Table 2.2 Number of Federal SOEs in Brazil, 1900 1981 Number of federal SOEs Period by date of creation 1900 1930 25 1930 1939 8 1940 1949 33 1950 1959 55 1960 1969 180 1970 1979 259 Total (as of January 1981) 560 Sourc e: Lamounier and Bacha (1994: 148) the overall economy had grown greatly during the military regime. In 1962 12 of the largest 30 Brazilian firms were SOEs; by 1979, 28 of the largest Brazilian firms were state owned (Bulmer Thomas 1994: 356). Long pro tected from foreign competition, most SOEs had become highly inefficient. Since they were under governmental control, most became instruments of political patronage and corruption. As a result, SOEs accounted for a large portion of government debt during the period. Bela Belassa and Gerardo Bueno write that in Latin America The deficit of state enterprises in the mid 1970s accounted for one quarter of the public sector deficit but rose to about one half in 1980 1982, while the overall public sector defici t doubled from about 4 percent to 8 percent of GNP. Partly because of price controls on basic products, state enterprises thus accounted for three quarters of the deterioration in public sector finances (1986: 139) Thus, there were already signs during the mid 1970s that SOEs were becoming a major drain on state resources in most countries in Latin America. However, it was not until

PAGE 46

38 after the energy crisis of 1979 that the fiscal stance of most states such as Mexico and Brazil became critical, largely b ecause of their overextended and heavily indebted state enterprise sector. The expansion of the state in the economy during the 1970s had serious implications. The Second National Development Plan (1975 1979), under the leadership of President Ernesto G eisel, supported increased investments in large scale projects. Since it was impossible for Brazils exports to finance these projects, the country looked abroad for funding. The Eurodollar market, which was awash in money following the oil crisis of 197 3, provided a steady and reliable source of foreign finance. With the second oil shock of 1979, Brazil found itself increasingly vulnerable as servicing of the foreign debt amounted to 67 percent of export earnings (Roett 1992: 167). By 1980, it was clea r that Brazil was going to face enormous difficulty in servicing its foreign debt in the near future. The situation was exacerbated by international conditions. By the early 1980s, the world economy was suffering its worst downturn since the Great Depres sion. Economic hard times in the developed world led to a decline in demand for exports in developing countries. Complicating matters even further was the monetary policy adopted by the U.S. Federal Reserve at that time, which increased interest rates. The effect of this policy was an increase in the debt service requirements of developing countries. The combination of global recession and increased short term debt burdens led to a financial crisis across Latin America. The critical moment came in 1982 when the Mexican government announced that it was unable to service its debt payments, causing a panic throughout the international financial community. Since Mexico had one of the

PAGE 47

39 largest economies in Latin America, and supposedly one of the healthiest, what happened there was seen as a portent for the rest of the region. Soon after the Mexican default, investors began withdrawing billions of dollars from all countries in Latin America. The combination of economic contraction, capital flight, and result ant decreased living standards turned the 1980s into the lost decade for the region. Brazil was forced to enter into tedious and contentious negotiations with foreign creditors. Resolution of the debt crisis was complicated by political developments wi thin Brazil, as the country was in the process of political opening ( abertura ) at that time. In 1985 the military stepped down from power, leaving the first civilian government in 21 years the formidable task of resolving the countrys pressing economic p roblems. Democratization and Demise of the State Led Model The new president, Jos Sarney, was confronted with economic pressures from various sources. First, Brazil was yet to come to a final agreement with its foreign creditors. Second, the advent of civilian rule brought with it long repressed popular grievances to address social issues, which had been ignored by the military government. While the international financial community was pressing for an orthodox plan of fiscal discipline, and many Brazi lian economists supported such a plan as a means to control inflation, the Sarney government found itself beholden to popular pressures. Sarney introduced the Cruzado Plan in 1986, a heterodox set of measures aimed at controlling inflation without exact ing austerity on the population. Although the plan was initially successful in slowing inflation, after less than a year inflation was on the rise again and Brazil found itself unable to meet its debt obligations. In early 1987 the government declared a moratorium on its foreign debt payments. Economic strategy remained

PAGE 48

40 inconsistent, and the following years witnessed a string of unsuccessful heterodox stabilization plans. Thus Brazil entered the 1990s with a growing sense of crisis. This mood allowed fo r the election of Fernando Collor de Mello in 1990, an anti establishment candidate and political unknown from the northeastern state of Alagoas. As will be discussed in more detail in the following chapters, Collor initiated the first significant set of economic reforms, including the beginning of the national privatization plan and the lowering of restrictions on trade. Yet Collor was impeached after serving less than two years in office, and Brazil drifted further. Collors successor, Itamar Franco, p roved to be a highly ineffective leader. During his tenure inflation reached record levels, exceeding two thousand percent in 1994. It was not until the introduction of the Real Plan in 1994, under the leadership of Finance Minister Fernando Henrique Car doso, that stabilization was finally achieved. Cardosos popularity was so high from his role in implementing the Real Plan that he was able to win a first round victory in the presidential election on 1994. During his tenure as president Brazil has unde rgone its most sweeping economic reforms. Many observers point to Brazils political system as the cause of the countrys poor economic performance since the return to democracy. On closer inspection, one sees that growth remained rather high at the beginning of the civilian regime, yet entered a period of stagnation in 1988 (see figure 2.3). Brazils elaborate system of indexation allowed for growth in the face of inflation, yet confidence was eroded dur ing the late 1980s and early 1990s as inflation surpassed four digit levels. Critics cited numerous institutional arrangements for complicating efforts to implement much needed economic

PAGE 49

41 Figure 2.1 Brazilian GDP Growth rates, 1981 2000 reforms. Instead of promoting competition and modernization, the 198 8 constitution enshrined a number of nationalist economic tenets. Furthermore, it created a problematic system of revenue sharing between the federal government, on one hand, and state and local levels of government on the other. By forcing Braslia to t ransfer resources to lower levels of government without changing the federal governments spending obligations, the constitution created a problematic system which led to a structural fiscal imbalance. The constitution, as well as other features of the Br azilian political system, led scholars to -4.3 0.8 -2.9 5.4 7.9 7.5 3.5 -0.1 3.2 -4.4 1.0 -0.5 4.9 5.9 4.2 2.8 3.7 0.2 0.9 4.0 -6 -4 -2 0 2 4 6 8 10 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Percentage Annual Change

PAGE 50

42 focus on the institutional obstacles to economic reforms. It is to these institutions that I now turn my attention.

PAGE 51

43 CHAPTER 3 BRAZILIAN POLITICAL INSTITUTIONS In Chapter 1 I examined the evolution of the discipline of political science during the past two decades, emphasizing the rise of the new institutionalism. While it is not a monolithic school of thought, the new instituti onalism has as a foundation its emphasis on the role of institutions in influencing political outcomes. Rational choice proponents hold that actors preferences are themselves shaped by the incentives presented by institutions. In contrast, historical in stitutionalists generally examine the ways in which early decisions regarding institutional design have (often unintended) consequences in the future. According to historical institutionalists, once certain institutions come into being they display a sti ckiness which makes institutional change difficult. Douglass North maintains that initial institutional configurations lead to path dependence, which he defines as network externalities that limitedactors choices and prevented them from radically alte ring the institutional framework (1991: 115). Finally, sociological institutionalists contend that persistent institutional patterns underlying the frequently changing formal rules deeply shape individual actors and their interests (Weyland 1996: 200). The previous chapter ended with a brief discussion of the link between Brazils difficulties with stabilization and structural economic reforms during the 1980s and the countrys political system. In the present chapter I will move into an examination o f the main political institutions in contemporary Brazil. Chief among these institutions is the Brazilian variant of presidentialism, the system of federalism, party and electoral

PAGE 52

44 systems, and the national congress. All of these formal political institut ions are closely linked, making it difficult if not impossible to separate their effects on the overall functioning of the political system. For example, the electoral system for the Cmara de Deputados (the lower house of Congress) is based on an open li st proportional representation system. This system, which presents politicians with incentives to run personalistic campaigns against members of the same party and to utilize a party label opportunistically, in turn has had a direct effect on the party sy stem by fostering fragmentation and indiscipline (Ames 1995a). Furthermore, to understand fully these institutions and their origins, one must also examine the constitution of 1988, which established the institutional framework of the New Republic (Souza 1997; Reich 1998). There is much evidence to support the thesis that the institutional structure of the Brazilian political system encumbered the implementation of economic reforms following the advent of the New Republic in 1985. However, I will argue that this focus on Brazils apparently problematic institutional arrangements has caused scholars to pay insufficient attention to the countrys impressive track record with economic reforms. Successes on the reform front are usually examined anecdotally and seen as cases in which executive resolve has been able to overcome the obstacles to reform presented by the political system. What is needed, I will argue in Chapter 4, is a reassessment of Brazilian political institutions and how they influence ref orms. In Chapter 2 I briefly detailed the historical role of the state in the Brazilian economy. Starting in the 1950s the country embarked on a course of state led industrialization, passing from a largely agricultural country to the largest industrial country in the developing world in the span of two generations. According to most

PAGE 53

45 scholars, including Trebat (1983), Shapiro (1994) and Baer (1995), such rapid development would have been impossible without the active intervention of the Brazilian state. State owned enterprises (SOEs) played an indispensable role in this process. Due to this particular relationship between the state and the economy, some scholars described the Brazilian economy as a case of state capitalism, in which market forces pre dominated, but the state played a larger role in capital accumulation and industrialization than in Western Europe and North America (Baer et al 1976; Hagopian 1986). One could utilize any of the three main institutionalist approaches when studying the va garies of recent economic reforms in Brazil. An historical institutionalist approach would highlight the institutional framework created during the course of Brazils development since the beginning of state led industrialization. One could argue that th e Brazilian states developmentalist efforts during the latter half of the twentieth century created an institutional matrix closely linked to the economy, through myriad direct and indirect political and bureaucratic channels. Such channels included fisc al policy, the central bank, federal and state development banks, SOEs, price controls, and corporatist labor relations (Baer, et al 1973). This complex web between the political and economic systems consequently made it very difficult to reform the state following the crisis of the 1980s. According to Frances Hagopian (1986: 2), by concentrating most economic resources within the public sphere, [state capitalism] enhances political clientelism, which privileges political elites who control the distribut ional arm of the State. Hence, the original political economic institutional structure, which provided key politicians with important patronage resources, would display a great degree of rigidity (or stickiness)

PAGE 54

46 in the face of efforts to reform it. Th e entire economic system proved highly resistant to change precisely because it had long been highly politicized. Rational choice institutionalism can also be useful in explaining the difficulties encountered in passing reforms in Brazil. Barry Ames (1 994; 1995a; 1995b), for example, has studied the effects of the Brazilian electoral system on the strategies employed by candidates for the lower house of Congress, the Cmara de Deputados. He concluded that the system, which is based on open list proport ional representation (see more below), gives politicians incentives to concentrate on small, targeted constituencies during their campaigns for elections. Once in office, politicians behave in such a manner as to funnel as much pork as possible to these c onstituencies, which is fundamentally at odds at efforts to address the states fiscal imbalances. In his study of the party system Mainwaring (1999) employs a rational choice methodology. He defends the use of rational choice theory for this case study, saying: Brazilian catch all parties usually behave like self interested actors whose behavior is shaped by formal rules of the political game. Such rules help explain the individualism of politicians, the decentralization of parties, the weakness of nati onal parties, the lack of party discipline, and the limited allegiance of politicians to their parties (1999: 10) In the discussion of political institutions that follows, I will cite literature from all three institutionalist schools of thought rather t han utilizing one sole perspective. After first discussing the general links between political institutions and economic reforms in Brazil, I then analyze particular political institutions: the 1988 constitution, federalism, presidentialism, the party sys tem, the electoral system, and the Congress. In each section I will describe the function of these institutions and then will explain how they hinder the implementation of economic reforms.

PAGE 55

47 Political Institutions and Economic Reforms Brazil possesses one of the most problematic political systems in Latin America, and there is widespread acceptance among scholars that the Brazilian political system contains several institutional features that have impeded the passage of much needed economic reforms. By ma ndating a transfer of revenues to state and local governments without a commensurate decrease in the federal governments obligations, the 1988 constitution lay at the heart of the countrys persistent fiscal problem. Correcting this imbalance, as well as a host of other structural and political reforms, has proven to be difficult. According to Figueiredo and Limongi (1995), politics in post authoritarian Brazil has been marked by the existence of two separate competing agendas. On the one hand, presiden ts have generally been interested in economic reforms, while on the other hand the legislature has been more concerned with social issues. These distinct policy orientations have often pitted the executive against the legislative, making reforms time cons uming. Once implemented, many reforms barely resemble the original legislation submitted and are not effective at attacking the problems they are meant to address. Tim Power (2000) has described politics in the New Republic as a permanent constitution al convention. Since the transition from authoritarianism and the start of the New Republic, the very institutional framework of the political system has been called into question. The formation of the National Constituent Assembly to draft the new cons titution was itself a highly politicized process, and there has been widespread dissatisfaction with the final document from a number of groups since its promulgation in 1988. As a result, Power explains, Brazil witnessed a permanent constitutional conve ntion as legislators and the public mulled over the desirability of tinkering with the constitution. One of the most important events in this process was the plebiscite in 1993

PAGE 56

48 to decide whether to maintain the existing presidential system, or to switch to either a parliamentary or a monarchical system. Despite the discontent with the presidential system and its supposed deleterious effects, which I will outline below, 55 percent of voters opted for the existing system. The 1988 Constitution and the Ref orm Process One of the pressing issues surrounding the transition to democracy during the early 1980s concerned how the post authoritarian political order would be structured, as well as by what means this structure would be established. The legislature elected in 1986 was to serve as a National Constituent Assembly (Assemblia Nacional Constituinte ANC), which convened in February 1987. The Assembly included representatives ( deputados ) elected in 1986 and senators elected in 1982; together in a unicam eral body they served as the ANC. The ANC shunned any major political reforms, instead focusing principally on social and economic issues (Fleischer 1998: 120). In effect, the document produced by the ANC constitutionalized numerous provisions that g enerally are the preserve of normal legislation. The new constitution placed stiff restrictions on foreign investment, sanctioned state monopolies in sensitive sectors of the economy, and granted public sector employees job security and an array of bene fits. Another important provision of the constitution established a new revenue sharing arrangement between the federal government and the states and municipalities. The new arrangement forced the federal government to transfer significant revenues to lo cal governments without a commensurate decrease in the federal governments spending obligations. According to Roett (1997: 25), What it actually did was to open the floodgates for local corruption and mismanagement, as local and state authorities

PAGE 57

49 enjoye d a financial bonanza without any programmatic responsibilities in critical areas of social investment. To understand the uphill battle fought by the Cardoso administration (as well as his predecessors), one must understand the consequences of the 1988 constitution. Because many of the economic changes Cardoso has tried to enact are in sectors in which the state has constitutionally mandated monopoly power, the constitution itself must be revised to reform these areas. In order to amend the constitutio n, both houses of Congress must approve the amendment two times by a majority of sixty percent or greater. In 1995, soon after taking office, Cardoso was able to pass several key constitutional amendments. Known as the Big Five, these amendments ended several state monopolies (mining, coastal shipping, telecommunications, natural gas distribution) and eliminated the distinctions on domestic and foreign capital. This was one of the greatest achievements of the Cardoso presidency, for it cleared the way for the continuation of the stalled national privatization program and helped to attract record levels of foreign investment, both of which were essential in alleviating balance of payments problems and modernizing the economy. The next chapter will exam ine in greater detail how Cardoso was able to secure these amendments, with the hopes of shedding light on how, in spite of political institutional barriers, widespread economic reforms have been possible in Brazil. Federalism Like the United States, Br azil is a federal republic. The federation is composed of twenty seven states and a Federal District, Braslia. Although federalism has a long history in Brazil, the present form of federalism is the result of the 1988 constitution. Due to the new const itution, the executive has lost power to the legislative and judicial

PAGE 58

50 branches, the union to the states and municipalities, and the state to society, private enterprise, and market forces (Selcher 1998: 25). Throughout Brazilian colonial and imperial hi story, there was a balance between an ostensibly powerful central authority and a de facto diffusion of power. In many ways, this decentralization of power assisted in maintaing the countrys territorial unity following independence. Modern federalism i n Brazil has its roots in the early republican era of the late nineteenth century. Following the deposal of the Bragana monarchy in 1888, those actors involved in the republic movement worked to create a federal system that granted individual states a hi gh degree of autonomy vis vis the central government. The constitution of 1891 reflected the interests of the most powerful economic elites, which were also those linked to the anti monarchist movement in 1888. This document granted states extensive po wers. As a result, the most powerful states So Paulo, Minas Gerais, and Rio Grande do Sul were able to limit the power of the central government. In practice they did so by maintaining sizable state militias (often much larger than the federal forces st ationed within these states), taxing exports, and establishing their own international diplomatic missions. 1 One important implication of the early history of modern Brazilian federalism is that it created a contentious and at times contradictory divisio n of powers. As Alfred Stepan (2000) explains, these initial arrangements have displayed a high degree of stickiness, which has made reforming the more problematic aspects of Brazilian federalism very difficult. This is not to say that there has not be en significant change in 1 This was important especially for the state of So Paulo, which aggressivel y promoted immigration during the coffee boom of the late 1800s.

PAGE 59

51 the structure of the Brazilian political system since the advent of the republic. In fact one sees both centralizing and decentralizing trends during the past century. The era of Getlio Vargas (1930 1945) and the military regime (1964 1985) displayed a tendency to centralize power, while periods of democratic rule (1945 1964, 1985 present) have shown a movement towards political decentralization. These shifting trends are the result of a dialectic between two separate forces wit hin Brazilian political culture: one based on local patron clientelism, the other based on the belief in the efficacy of greater federal authority. While many today feel that the devolution of political power away from the center is beneficial for democr acy, Stepan (2000) persuasively argues that in the New Republic decentralized federalism has been injurious to democracy. The Brazilian system of federalism is demos constraining, according to Stepan, for it has distorted to a higher degree than any oth er federal system in the world the basic democratic idea of one person, one vote. The Senate, based on the principal of serving as a voice for less populated states, is by its nature demos constraining. In Brazil, however, the Senate possesses signif icant powers compared to upper chambers in other federations. With three seats granted to each state, the Senate further skews the representation of the national population. Furthermore, the creation of several lightly populated states in the North and C enter West of the country during the past two decades has greatly enhanced the power of conservative interests in the Senate, with congressmen from these areas often opposed to the executives reform measures. The Cmara de Deputados, which in theory is s upposed to approximate more closely the actual population distribution within a federation, greatly over represents rural states and underrepresents more populous ones.

PAGE 60

52 While similar arrangements can be found during the whole of the republican era (Soare s forthcoming: chap 12), the 1988 constitution mandated the most skewed system to date. It set a minimum of eight deputados and a maximum of seventy per state. Figure 3.1 illustrates the degree of disproportionality created as a result of this arrangment in the Cmara de Deputados. By calculating the percentage of seats awarded to separate geographical regions and contrasting it with each regions share of the total national population, Lamounier and Bacha (1994: 175) revealed that in 1991 the Southeast, the most economically advanced region of the country, was under represented by 51 seats (of a total of 503). Stepan (2000: 150) further elaborates this dilemma: The one person, one vote rule would have given the states of Acre, Amap, and Roraima 1 feder al deputy each in 1994 and the large state of So Paulo 114, whereas the Brazilian system actually produced 8 deputies for each of the three sparsely populated states while the citizens of the huge state of So Paulo were represented by only 70. This over representation of rural states has largely benefited conservative political interests, many of which have been consistently hostile to reform efforts during the New Republic. Aside from the impact on the composition of the national congress, Brazils syste m of federalism has had a deep impact on the relationship between the federal government, the states, and the municpios During the transition to democracy state governors played an important role in contesting the powers of the military regime. In 198 2 the military regime, well along in its plan to liberalize political activity, allowed for the direct elections of governors. This in turn granted the governors a great amount of leverage during the transition to democracy. Since the military regime, wh ich centralized

PAGE 61

53 Table 3.1 Disproportionality between Regional Shares of Federal Chamber Seats and Population, 1986 1991 Chamber seats 1986 1991 Number of Number of % of Ideal Region seats % seats % population number a Di fference b North 47 9.65 65 12.9 6.55 33 32 Northeast 149 30.60 151 30.0 28.50 143 8 Southeast 169 34.70 169 33.6 43.63 220 51 South 82 16.84 77 15.3 15.15 76 1 Center West 40 8.21 41 8.2 6.17 31 10 Total 487 100.00 503 100.0 100.00 503 a: Number of deputies each region would have if regional shares were exactly proportional to their respective populations b: Difference between the ideal and the number in fact prevailing during the 1991 1994 legislature Source: Lamounier and Bacha (1994: 175) power within the federal government, had curtailed the traditional powers of the states, the governors (and their state political machines) were eager to enhance their powers following the end of the military regime (Abrcio 1994). Wi th their power bolstered following the elections of 1982, the governors became influential actors during the constitutional conventional of 1987 88. Partly as a result of their influence, the federal system created by the new constitution gave the states greater autonomy vis vis Braslia, notably in fiscal terms. The new constitution forced the federal government to redistribute a greater share of revenues to the states and municpios than had been the case previously. However, a transfer of responsib ilities did not accompany this transfer of resources. The federal government therefore has been forced to provide the same level of services, yet without the necessary resources. This has been a boon for the states and municpios with enhanced tax reven ues from Braslia aiding in local patronage politics. Two analysts of the current system go so far as to say that

PAGE 62

54 municipal governments in Brazilshould be the envy of all governments in the developing (or even developed) world (Bomfim and Shah 1994: 53 6). By most accounts, Brazils particular system of federalism lay at the heart of many of the countrys current problems (Stepan 2000; Mainwaring 1997; Abrcio 1994; Roett 1997). The federal system is the root cause of the countrys chronic fiscal prob lems, and as such it will be impossible to improve the federal governments fiscal position without altering the revenue sharing arrangement with the states and municpios Furthermore, until recently states had great autonomy in contracting their own sov ereign debt. The federal government has gained control of profligate state spending in recent years, but only by bailing out bankrupt state banks at a cost of $15 billions (Samuels 2000: 17). The Fiscal Responsibility law, passed in April 2000, marked a turning point in the fight for fiscal discipline. The law places stiff limits on the level of discretionary spending allowed by states and placed penalties on those states that do not maintain sound fiscal policies. Presidentialism With the wave of democr atization that occurred during the 1970s and 1980s, there was a growth of research examining questions of institutional design in nascent democracies. One of the central arguments in this debate centered on the efficacy of presidential systems of governme nt. Juan Linz (1990; 1994) was at the fore of this debate. He argues that presidentialism contains several inherent flaws that hinder effective governance in nascent democracies. The first problem associated with presidentialism is what Linz calls dual democratic legitimacy. Having come to power through the ballot, both the president and the legislature possess democratic legitimacy. Yet when there is conflict between these two branches of government, does one possess

PAGE 63

55 more legitimacy than the other? In the absence of any logical principle to define who really has democratic legitimacy, it is tempting to use ideological formulations to legitimize the presidential component of the system and delegitimize those opposing him, transforming what is an ins titutional conflict into serious social and political conflicts (Linz 1994: 8). Many times in the past such executive legislative conflicts have precipitated coups dtat in Latin America. A second problematic aspect of presidentialism is its rigidity. Presidents are elected to serve fixed periods of time. While constitutional provisions providing for the removal of incompetent or corrupt presidents usually exist, they are rarely invoked; when they are invoked they often signal political crises and vi rtually paralyze the political system. Some observers justly maintain that democracy ultimately prevailed during the impeachment of President Collor in 1992, yet during the impeachment process legislative activity came to a virtual standstill. Under parl iamentary systems, such crises can be solved within a relatively short time frame and without as much interruption of the legislative process. More importantly, under parliamentarism such crises of government are less likely to lead to the crises of regim e that have plagued Latin America. For example, Mainwaring (1997) believes that the coup of 1964 was caused in large degree by the problematic structure of Brazilian presidentialism. This debate over the merits of presidentialism has been the subject of intense public discussion in Brazil more so than in any other country in Latin America. In 1993 there was a plebiscite in which voters had to vote to maintain the presidential system, or adopt either a parliamentary or monarchical system. While presiden tialism won by a wide majority, the plebiscite vividly demonstrated the dissatisfaction with the countrys

PAGE 64

56 institutional arrangement. This dissatisfaction arose from the protracted stalemate between the executive and legislature, especially concerning eco nomic issues. Brazil possesses one of the most powerful presidencies in the world. The 1988 constitution granted presidents sweeping reactive and proactive powers. Presidents may issue both a complete and partial (line item) vetoes of congressional leg islation. Even more significant than these reactive powers are the proactive powers. Brazilian presidents have the right to initiate legislation in several areas, from the size of the armed forces to areas of public administration and ministries. The pre sident also wields significant powers over the federal budget. Perhaps the most important resource available to presidents by the constitution is the medida provisria or provisional measure. These measures immediately become law and are effective for thirty days. During these thirty days Congress has the power to reject or pass the measure; if no action is taken during that time, the law expires. The medida provisria has become the most important tool available to the executive branch. Although th e constitution states that they are only supposed to be issued in cases of relevance and urgency, in reality presidents have adopted them as a means of controlling the legislative agenda. Between October 5, 1988, when the new constitution went into effec t, and late May 1995, the four presidents is sued 1,004 provisional measures . Of [these], 640 had previously been issued. Congress had approved 274 provisional me asures and had rejected only 18. (Mainwaring 1997: 62 63) Furthermore, the frequency with which presidents have utilized the provisional measures has increased with each successive administration. This concentration of power within the executive branch serves to counteract the factors that work against the executive in Brazil: federalism and multipa rtism. As

PAGE 65

57 discussed above, the Brazilian system of federalism grants significant powers to the states, and the present system of revenue sharing weakens the fiscal position of the central government. The fragmentation and indiscipline of the party system is in some ways a concomitant of federalism, as congressional politicians seek to funnel patronage to their constituencies at home. Sweeping presidential powers may not bode well for the quality of democracy (ODonnell 1994), but they have been an essent ial means by which Brazilian presidents have been able to implement economic reforms in the face of congressional resistance. In fact, both Collors economic program in 1990 and the Real Plan in 1994 were enacted principally through the use of provisional measures. The Party System Political parties are indispensable elements in the functioning of democracy. They both shape the political arena and are shaped by the political system as a whole; they shape the nature of interests in society and respond to demands and interests of societal groups (Mainwaring and Scully 1995b: 459). During the period of military authoritarianism, political parties did not receive much attention from scholars, who often viewed them as peripheral to the political process in m ost Latin American countries. With the demise of authoritarianism, there was a growing awareness of the centrality of parties to politics. While in the past Marxist or dependency analysis discounted their role, it is now clear that, in short, parties mat ter. They are no longer seen simply as the by products of class forces, but are viewed as important variables in their own right. Compared to other countries in Latin America, Brazil is striking for its level of party underdevelopment. According to Mainwa ring and Scully (1995a: 7), the volatility of the Brazilian party system is comparable to that of Peru, Ecuador and Bolivia, all countries of with substantially lower levels of economic development.

PAGE 66

58 Historically Brazil has never possessed highly instituti onalized parties. During the entire Old Republic (1889 1945), there were no parties of national scope. Getlio Vargas, who served as president from 1930 until 1945 (and later from 1950 to 1954) and was instrumental in shaping the modern political system, was hostile towards parties. In fact, during his first presidency he governed without belonging to a party. Before leaving office in 1945 Vargas created two parties, the Partido Trabalhista Brasileiro (PTB) and Partido Social Democrtico (PSD), establish ing a pattern of state manipulation of the party system. Another pattern to emerge during the period between World War II and the coup of 1964 was the use of electoral alliances among parties. Electoral laws encouraged crossparty alliances and the indep endence of candidates from parties, and rewarded small regional parties (Coppedge 1998: 198). National parties worked against the driving forces of Brazilian politics: localism and patron clientelism. Unlike other bureaucratic authoritarian regimes, th e military regime that came to power in 1964 in Brazil took steps to maintain the trappings of democracy. Whereas the congress was closed and parties were forbidden in Chile, Argentina, and Uruguay, in Brazil the regime organized a two party legislative s ystem. ARENA (Aliana Renovadora Nacional) was the pro government party, while the MDB (Movimento Democrtico Brasileiro) served as the officially sanctioned opposition party. However, unlike its South American neighbors, which saw the resurgence of trad itional parties at the end of authoritarian rule, in Brazil none of the parties founded before the coup survived into the post military era. In 1979 the military regime allowed for a liberalization of the laws governing party formation, with the intention of splintering the opposition. As a result, the MDB succumbed to infighting, with most members joining

PAGE 67

59 the PMDB (Partido do Movimento Democrtico Brasileiro). Likewise, ARENA also fractured, with most members joining the PDS (Partido Demcrata Social). After 1979 there was a rapid growth in the number of parties, and was hastened in 1985 by the passage of laws that further facilitated party formation. The result of these permissive laws has been an unabated fragmentation of the party system. For examp le, none of the parties serving in Cardosos coalition existed before 1970. There are other aspects of the Brazilian political system which adversely affect the party system. Brazil has a combination of presidentialism and multipartism, a mix that is e specially unconducive to party building and to democratic stability. Thisis further complicated by the strong federalist bases of Brazilian politics. Power in concentrated in the national government, but political careers are based on the local or state level (Mainwaring 1992 93: 700). Furthermore, the electoral system serves to further weaken party coherence. As Ames (1995a, 1995b) shows, the system of open list proportional representation provides incentives for politicians to focus on local issues and deliver pork to small, targeted constituencies. Such behavior inevitably weakens party formation and discipline. The structure of the Brazilian party system, in turn, has had seriously implications for the implementation of economic reforms. Because of the fragmentation of the party system and lack of discipline among party members, the passage of contentious legislation by the president entails a cumbersome process of negotiation with individual politicians. One of the functions of parties is the ag gregation of interests; these groupings facilitate the negotiation of issues by lowering, in effect, the transaction costs of negotiation. But in Brazil, there are extensive costs for executives who are attempting

PAGE 68

60 reform. Pork and other perquisites must be distributed, thus the fiscal impact of securing support is high (Kaufman Purcell and Roett 1997: 16). Cardoso has had some success in passing elements of his reform agenda, especially at the beginning of his presidency, but he has had to accede to nu merous demands for patronage from congressional supporters. Winning this support has worsened the deficit of the central government, which, as mentioned above, had already deteriorated due to the Real Plan. The net public sector debt doubled under Cardo so, surpassing $300 billion (or about 36 percent of GDP) in early 1998 (Smith and Messari 1998: 9). In recent years government revenues have reached their highest levels in history. However, expenditures are growing even more quickly and these new expen ses, rather than being used in critical areas like education and health care, are devoted to interest payments, social security payments, and salaries and entitlements of public sector employees. The Electoral System and Congress Barry Ames (1995a: 407) ma kes a persuasive argument that at the core of Brazils institutional crisis lies the electoral system. He employs a rational choice argument in his analysis, highlighting the incentives that the electoral system presents to candidates to the national co ngress. The Chamber of Deputies (the lower house of congress) is based on an open list proportional representation PR system. Under this system, each state is a single, at large, multimember district. In a system of closed list PR, voters are presented with a list of candidates from each party. Candidates are listed in the rank by which they will be selected. Hence, if a party lists ten candidates and is awarded three, the first three candidates will be granted seats. Under the Brazilian system of ope n list PR, parties do not determine which candidates will represent them; the voters do. This forces candidates from the same party to run against each other,

PAGE 69

61 serving to further erode party coherence and discipline. Party fragmentation in turn makes it d ifficult for presidents to negotiate legislation. They are forced to dole out patronage such as appointments, cabinets positions, and government funds for pet projects to politicians in order for their support of the presidents agenda, a very costly and time consuming process. Besides the impact of the electoral system, there are other aspects of the national congress that hinder the passage of reforms, which are almost universally proposed by the president. As mentioned above, the federal system establi shes upper and lower limits for the number of deputies from each state, ranging from eight to seventy. This has the effect of bolstering poorer, less populated regions, which are those most dependent on financial support from the federal government. Henc e, more often than not they are the ones most opposed to efforts to correct the governments fiscal imbalances. Furthermore, several new states have been created during the past two decades in the North and Center West, themselves poor and sparsely popul ated areas. Since all states are granted three senate seats, this has made the Senate even more averse to some reforms than the Chamber of Deputies. Most national politicians spend only a few years in Braslia before returning home to run for local offic e. This focus on local politics, rather than national issues, further complicates efforts to pass reforms in areas such as the economy, social security, and public administration. Conclusion The institutional design of Brazils political system has had adverse effects on the passage of reforms. The 1988 constitution created a problematic revenue sharing arrangement between the federal government and lower levels of government. It also

PAGE 70

62 mandated state monopolies in various areas of the economy. Since t he only way to amend the constitution is with two majority votes of sixty percent or greater in both houses, change proved difficult. The constitution also created lax rules for the formation of political parties and an open list PR system that further er odes party discipline. Party underdevelopment has been offset to some degree to the Brazilian variant of presidentialism, which possesses extensive reactive (veto) and proactive (legislative) powers. Still, presidents cannot always govern by decree, and the fragmentation and weakness of parties makes negotiation and coalition building tedious, if not impossible at times. All these factors taken together help to explain why Brazil delayed so long in implementing reforms that would stimulate its stagnant e conomy. Since the start of the New Republic in 1985, Brazil was faced with a deteriorating economic situation. The first two civilian presidents, Sarney and Collor, introduced numerous stabilization plans, all of which failed. Most explanations of these failures pointed to the relationship between the still tentative transition to democracy and Brazils political system, which, most agreed, possessed institutional aspects that encumbered the implementation of much needed economic reforms. It was not unt il the introduction of the Real Plan in 1994, under the leadership of then finance minister Cardoso, that the economy was finally stabilized. After his election as president, Cardoso was able to pass an array of reform legislation, which will be the focus of the next chapter. Still, other facets of his reform agenda including reform of the social security, tax, and judicial systems as well as the public sector in general have fallen short of the presidents expectations. Again, scholars point to institut ional factors as causes for these shortcomings. What these explanations lack, however, is precisely a

PAGE 71

63 framework with which to account for Brazils extensive economic reforms. This chapter has sought to demonstrate the relationship between political insti tutions on the one hand, and the difficulty of implementing reforms on the other. As I will argue in the next chapter, what is needed is a reassessment of the relationship between political institutions and outcomes in the Brazil. There I will what facto rs account for the reforms that have been successfully implemented during recent years, thus calling into question the standard institutionalist critique of Brazilian politics.

PAGE 72

64 CHAPTER 4 ACCOUNTING FOR ECONOMIC REFORMS IN BRAZIL In the previous chapter I analyzed the principal political institutions of Brazil and the ways in which they have impeded efforts to pass economic reforms. As mentioned in that chapter, there is consensus among s cholars that the institutional design of the Brazilian political system has complicated the passage of reform legislation. Many point to this institutional design as the root cause for Brazils late start with structural economic reforms vis vis other L atin American countries. Despite the insights provided by this institutionalist literature, it too often focuses on the failures of reform in Brazil rather than the successes. What is needed, I argue in this chapter, is a reassessment of the impact of po litical institutions on the passage of economic reform legislation in Brazil, for extensive structural reforms have been enacted during the past decade, especially since the start of the Cardoso presidency in 1995. In this chapter I will examine the effor ts made by presidents in the New Republic to pass economic reforms. The administration of President Fernando Henrique Cardoso will receive most scrutiny, for it has been during his time in office that reforms have proceeded furthest. After outlining the vagaries of the reform process since the transition to democracy, I will analyze one set of reforms from early in the Cardoso administration. Know as the Big Five, these reforms were a set of five separate constitutional amendments that ended the states monopolies in several sectors of the economy. One of the amendments also eliminated the legal distinction between foreign and domestic capital. I will examine the Big Five in detail because they were one of the earliest, and

PAGE 73

65 most important, successes of the presidents economic agenda. Finally, I will conclude the chapter with observations about how, in spite of the institutional obstacles discussed in Chapter 3, economic reforms have been possible in Brazil. Too much emphasis has been placed on the obs tructionist nature of the countrys political system, I argue, and scholars must therefore reassess the impact of political institutions on economic reforms in Brazil. The Changing Brazilian Political Economy I contend that significant changes have occu rred in the Brazilian political economy during the past decade, especially during Cardosos tenure. What evidence is there to support this contention? Writing in 1992, Riordan Roett provided the following description of the Brazilian economy: In spite of [President Collors] efforts, Brazil remains one of the worlds most protected economies with an average import tariff of 40 percent (compared with Argentinas 11.5 percent). Business remains under the control of cartels and does not cooperate with the g overnments economic stabilization efforts. Only at the end of the year did the first privatization take place with widespread protest and violence. The public sector deficit is a giant obstacle to efficient policy making (Roett 1992: 1 79) In 1993 an a rticle in Forbes disparaged Brazil as the sick man of Latin America, as inflation reached over 2000 percent and no solution to the problem seemed in sight ( Forbes 1993). Yet during the following year, 1994, inflation, the persistent economic demon of Br azil, was successfully controlled by the Real Plan. The stabilization caused by the Real Plan paved the way for resumed growth and an influx of foreign investment. Just as impressively, the national privatization program has generated more revenue in Bra zil ($102 billion between 1991 and 2001 [BNDES 2001]) than in any other county in the developing world. Foreign trade has been greatly liberalized, with an average import

PAGE 74

66 Table 4.1 Key Economic Events in the New Republic, 1985 2001 President Period Re forms Sarney 1985 1990 Four stabilization plans, all of which failed Privatization on very small scale Collor 1990 1992 Initiated national privatization program Began liberalization of trade Failed stabilization plans Franco 1992 1994 Tepid towards the reforms introduced by Collor Overshadowed by then finance minister Cardoso Introduction of Real Plan Cardoso 1995 present Deepening of privatization program Deepening of trade liberalization End of state monopoly on reins urance (1996) Administrative reform (March 1998) Watered down social security reform (Nov. 1998) Fator Previdencirio (Nov. 1999) Law of Fiscal Responsibility (April 2000) tariff of 14.3 percent in 1999 (IDB 2000: 125). And in sp ite of many setbacks, the fiscal deficit is slowly coming under control. Within a decade, then, the Brazilian economy has undergone extensive restructuring. Yet these changes have not happened by chance. They are the direct result of changes within the political system. And as I shall argue, these changes call into question the prevailing institutionalist critique of Brazilian politics. The Vagaries of Economic Reforms in the New Republic Almeida (1996: 223) cites three factors that account for the del ay in economic reforms in Brazil: the power and profile of interest groups; the decentralization of governmental structures; and changes in the structure and capacity of the federal

PAGE 75

67 government. Because of these factors, economic reforms in Brazil have been guided, in Almeidas words, by pragmatism by necessity. Rather than disparaging the slow pace of change, Almeida posits that Brazil has in fact benefited from it for several reasons. First, the country has avoided large scale de industrialization, as has occurred in some countries that undertook a more orthodox approach to reforms 1 Second, she believes that the slowness of reform bodes well for democracy in Brazil. Rather than having a rubber stamp Congress, the Brazilian legislature has given v oice to many concerns about the consequences of economic reforms. While this may have slowed the pace of reform in Brazil, she contends, such a situation is preferable to cases in which presidents have autocratically implemented draconian stabilization m easures 2 As touched upon in Chapter 2, Brazil underwent the transition from authoritarian to democratic rule at the same time that the economy was suffering from the debt crisis. When the first civilian president, Jos Sarney, took office in 1985, there was still no final agreement with Brazils foreign creditors, nor was there any clear plan on how to tame inflation. Sarneys answer to the countrys economic dilemmas came in February 1986 with the introduction of the Cruzado Plan. The plan introduced a new currency, the cruzado a temporary wage and price freeze, and a temporary freeze to indexation. While the plan initially brought about price stabilization and a boom in consumption, producers were not able to maintain the price freeze for long. Furt hermore, the government was accumulating a growing deficit. The problem for the Sarney government was that it had 1 For a criticism of the quick move towards market reforms in Chile, see Collins and Lear (1994). 2 Chile under Pinochet Peru under Fujimori, and Argentina under Menem are three prominent examples of the autocratic implementation of neoliberal reforms, indicative of what ODonnell (1994) calls delegative democracy.

PAGE 76

68 neither the political will or the political capacity to reduce its budget deficitthe Sarney government could function only by paying out pa tronage resources and by preserving congressional access to those resources (Kingstone 2000: 189). After less than a year, then, it was clear that the Cruzado Plan would not be effective at stabilizing inflation. During 1987 88 Congress met as the Natio nal Constituent Assembly (ANC), which meant that other legislative matters were put on the back burner. The final result of the ANC, the constitution of 1988, was a further setback towards stabilization, for, as mentioned in Chapter 3, it mandated a great er role for the state in the economy (by means of monopolies in several sectors of the economy) and also created a fiscally deleterious revenue sharing arrangement between Braslia and lower levels of government. Sarney introduced three other stabilizatio n plans during his term, all of which failed. At the end of his term public dissatisfaction was extremely high, leading people to look for someone who could improve the deteriorating economy, as well as control what most perceived as a highly corrupt poli tical system. With inflation accelerating and the economy stagnating, the Brazilian public found such a person in Fernando Collor de Mello. It was during the administration of Collor (1990 1992) that Brazil made its first, hesitant steps towards economic reform. Sarney left office in ignominy, having introduced four unsuccessful stabilization plans. It was under these conditions that the dark horse candidate Collor, a former governor of the poor northeastern state of Alagoas, was able to rise to power. Collor won the presidency by emphasizing his commitment to fight corruption and stabilize the economy. He created his own party, the Party of National Reconstruction (PRN), to run for office, eschewing alliances with any established parties. Upon taking office, he had no firm congressional base with which to

PAGE 77

69 work. Furthermore, his harsh rhetoric against the political class did little to enamor him with members of Congress. With an extremely conflictual relationship with Congress, Collor found himself be sieged when allegations about corruption by him and his family and associates surfaced in 1992. Later that year he became the only president in Brazilian history to be impeached. Despite the ignoble circumstances under which Collor left office, during h is brief tenure in office Collor implemented several reforms that laid the groundwork for subsequent reform efforts. His two greatest accomplishments were spearheading the national privatization program in 1990 and opening the Brazilian economy to greater foreign competition. It should be noted that these were areas in which the president could act alone, through his decree power, to effect change; privatization and trade liberalization were two areas where the president has a free hand vis vis Congress Yet Collors habit of riding roughshod over Congress, such as making frequent use of decrees, did little to help him as the corruption allegations began to arise (Weyland 1993). The interregnum of President Itamar Franco, Collors successor, was a time of growing uncertainty about the future of Brazil. Franco was not a supporter of Collors earlier reforms measures, but at the same time Franco did not work actively to reverse these reforms. Francos most important move in the economic realm was appoint ing Senator Fernando Henrique Cardoso as finance minister in early 1993. As will be discussed in greater detail below, Cardosos role in crafting the Real Plan was to have profound political ramifications during the early days of his presidency.

PAGE 78

70 The Real Plan, which Cardoso devised and implemented as finance minister to President Franco, sought to cure the countrys hyperinflation by creating a new currency, the real and linking it to the dollar with a crawling exchange rate (managed float) In order to address successfully the hyperinflation problem, it would be necessary to rein in public expenditures. In this respect privatization was to play an important role. The divestiture of state assets would add additional revenues to the national treasury, th us alleviating the governments deficit. Furthermore, if foreign participation in the program were expanded, foreign investment could help the countrys balance of payments. In order to attract foreign capital, however, the laws governing foreign investm ent had to be changed so as to end discrimination between foreign and domestic capital. Without such changes, it would be difficult to attract the investments needed to modernize the countrys infrastructure, much of which had been under control of the st ate for decades and was in desperate need of improvements. As we shall see shortly, President Cardoso was acutely aware of these facts as he quickly pushed for an extensive array of economic reforms upon taking office. Cardoso and the Economic Reform Age nda Since taking office on 1 January 1995, President Fernando Henrique Cardoso has worked to promote legislation that would liberalize and make more competitive the Brazilian economy; to reform the political system, ranging from the laws governing party fo rmation and representation in Congress, to the system of revenue sharing between the federal government and lower levels of government; and to restructure the countrys bloated social security and other pension programs that are near bankruptcy.

PAGE 79

71 Table 4.2 Monthly Inflation Rates, 1994 January 42.2 February 42.4 March 44.8 April 42.5 May 40.9 June 46.6 July 5.5 August 3.3 September 1.6 12 months to end Sept 3748.1 January December 1993 2708.6 Source: LARR Brazil (1994b: 7) Figure 4.1 details t he dramatic drop in inflation brought about by the Real Plan. When the plan was implemented in June 1994, monthly inflation exceeded 40 percent. By September of that year, it had dropped to less than 2 percent. As the Real Plan was implemented and infla tion slowed dramatically, millions of Brazilians quickly saw a rise in their purchasing power. The stabilization plan had, in short, ended the onerous inflation tax that had haunted the poor for years. The rapid improvement in the populations living s tandards, especially the poor who worked outside of the formal economy and whose wages were not indexed to inflation, gave Cardoso enormous clout as he decided to run for president. He warned the public that his opponent, Luis Incio Lula da Silva of th e Workers Party (PT), would return Brazil to its previous days of statism and fiscal irresponsibility, thus erasing the recent gains made in the fight against inflation. Cardoso won the support not only of the nations business leaders (Kingstone 2000), but also the overwhelming support of the population at large, who had seen the tangible benefits of Cardosos stabilization plan. On 3 October 1994, Cardoso garnered a

PAGE 80

72 first round victory in the presidential election, receiving 53 percent of the vote (LAR R Brazil 1994a). Thus, as he assumed power Cardoso enjoyed immense popular support, and, as the president would soon argue, a mandate to reform the state and its role in the economy. His hope was to translate his popularity from the Real Plan into politi cal capital in his attempt to promote economic reforms early during his administration. Before entering a life of politics, Cardoso established his reputation as one of Brazils foremost social scientists. His writings on dependency theory placed him at t he fore in the debate about capitalist development and its effects on underdeveloped countries. He was exiled along with other leading Brazilian social scientists during the late 1960s, during which time he lived in Santiago, Chile, with other Latin Ameri can intellectual exiles and also spent time at universities in the United States and Europe. Following his return to Brazil, Cardoso launched his political career in the 1970s, running for the Senate in his home state of So Paulo. Over time his political views moderated, moving to the center on economic issues while maintaining a social democratic philosophy on social policy (Goertzel 1999). During the 1980s he was a founding member of the Brazilian Social Democratic Party (PSDB), which also counted amon g its ranks long time Cardoso friends Jos Serra (Cardosos current Health Minister) and Mrio Covas (the late governor of So Paulo). As his stock rose during the early 1990s, Cardoso was to draw upon his extensive academic knowledge and political expe rience, both of which influenced his thinking. He acknowledged that the root of many of Brazils problems lay within the nations political system. As Peter Evans, writing about the necessary conditions in which to undertake economic adjustment, states, Dismantling the state is not the answer. It must be

PAGE 81

73 reconstructed (1992: 141). Evans highlights the following paradox arising from recent attempts at economic reforms in developing countries: the state, which is believed to be the root problem of recen t economic crises, is now trusted with the task of implementing solutions to these crisis. Cardoso has been acutely aware of this paradox (Kingstone 2000: 201). Reform of the Brazilian economy and an improvement in social welfare would necessarily entail a reform, and hopefully strengthening, of the state itself. Reflecting his avowed social democratic philosophy, Cardoso pledged during his 1994 campaign to make agriculture, employment, security, education, and health the main priorities of his administ ration (Kingstone 2000: 197). Upon assuming office, however, the new president dedicated himself to swift economic reforms, arguing that it would be impossible truly to improve social welfare until structural adjustments has been undertaken and the state itself had been reformed. One of the presidents first, and most successful, moves was an attempt to end the states monopolies in several areas of the economy and to end discrimination against foreign capital. It is to this subject that I now turn. The Big Five Constitutional Amendments of 1995 Cardoso assumed the presidency in early 1995 with a resounding mandate. Having authored and implemented the Real Plan the previous year as finance minister under President Franco, Cardoso was able to capitalize upon his success in tackling inflation during his bid for the presidency. Upon taking office, he attempted to make use of his enormous popularity by pushing for a series of economic reforms, later dubbed the Big Five, that would require amending the const itution.

PAGE 82

74 The passage of these amendments was important to Cardosos economic plan for several reasons. First, and most importantly, they would end the states mandated monopoly in key sectors of the economy, therefore opening them to foreign participati on. This step had two purposes, one purely economic, the other economic and psychological. From a purely economic standpoint, these constitutional changes would allow for the infusion of badly needed capital into key sectors of the economy. With the exc eption of the mining sector, these were areas in which the state had performed poorly in recent years and as result had not received adequate investment. Furthermore, the administration and even many within the business community realized that competition was needed in order for Brazil to grow rapidly. Another purpose of these amendments was to send a clear and strong signal to international investors that Brazil was serious about structural reforms. The simultaneous end of the discrimination between fo reign and domestic capital and the opening these sectors to private (foreign and domestic) participation consequently assisted in attracting record levels of foreign investment to Brazil. Despite the new presidents popularity, the passage of these pivotal measures was in no way assured. First, they were likely to draw criticism on nationalist grounds, as these amendments would pave the way for the privatization of SOEs in politically sensitive sectors 3 Second, there were possible political obstacles. A s discussed earlier, amendments to the constitution require two separate majority votes of sixty percent by 3 See the collection edited by Barbosa Lima Sobrinho (19 94) for a scathing criticism of the Brazilian privatization program, manily on nationalist grounds.

PAGE 83

75 Table 4.3 Cardoso's Legislative Coalition, 1994 Party Chamber Senate PMDB 107 22 PFL 89 18 PPB 88 11 PSDB 62 11 PTB 31 5 Coalition Total 377 67 Total Seats in Each House 513 81 Coalition Percent 0.73 0.83 Source: Kingstone (1999: 200) both houses of congress. Hence, tampering with the constitution in Brazil requires a rather broad consensus among legislat ors. The 1988 constitution grants presidents some of the most sweeping decree powers in the world (Power 1998b), the most important of which is the medida provisria (provisional measure). However, the constitution also codified a number of nationalistic economic tenets, such as granting the state monopolies over traditionally sensitive sectors, namely those related to sectors that had once been controlled principally by foreign companies. In order to liberalize these sectors of the economy, therefore, Ca rdoso could not utilize his own presidential legislative powers but would instead have to jockey with Congress. Cardosos electoral alliance included his own Brazilian Social Democratic Party (PSDB), the conservative Liberal Front Party (PFL), and the smal ler Brazilian Workers Party (PTB) and the Liberal Party (PL). Following the election, the president included all these parties except the PL in his coalition. It also expanded to include the catch all centrist Brazilian Democratic Movement Party (PMDB) a nd the Progressive Brazilian Party (PPB). Coalition members held 73 percent of seats in the lower house and 83

PAGE 84

76 percent of seats in the Senate (see Figure 4.2). This was a solid majority, yet the breadth of the coalition meant that certain parties would be less amenable to certain measures. For example, the PFL, controlled by the powerful senator Antnio Carlos Magalhes from the state of Bahia, favored a neoliberal economic agenda but was extremely conservative on social issues. The presidents party, in contrast, was one of the most reserved in its support of economic reforms but favored a progressive social agenda. From the outset Cardoso realized that he would have to create sets of shifting alliances, looking for support on the right for his econo mic proposals and turning to the left for help in passing his social agenda (LARRB 1994b: 7). Despite his avowed credentials as a social democrat, Cardosos first priority upon taking office was economic reform. Towards this end, the Big Five amendments stand out as one of the earliest and boldest of the presidents proposals. Table 4.4 The Big Five Constitutional Amendments of 1995 Amendment 5 Allowed for private participation in natural gas distribution Amendment 6 Eliminated legal distincti on between Brazilian and foreign capital Amendment 7 Opened internal Brazilian shipping to foreign participation Amendment 8 Permitted private participation in telecommunications services Amendment 9 Allowed for private participation in petroluem explor ation Source: BNDES (1996: 37) Accounting for Economic Reforms in Brazil To this point I have attempted to justify my rationale for reassessing the impact of institutions on economic reforms in Brazil. The evidence demonstrates that significant economic reforms have been passed during the past decade, which has helped restore

PAGE 85

77 growth and has increased investor confidence. Now I want to suggest some reasons for how economic reforms have been possible in Brazil. I look specifically at four factors : timing; the type of reform proposed; ideology and ideas; and the influence of political and economic factors. Taken together, I believe, these four variables help to explain the depth of economic reforms in Brazil. Timing When looking at attempts to pas s economic reforms in the New Republic, it is clear that timing plays an important role. More specifically, economic reforms have been easier to pass at the beginning of a presidents time in office. This is not to say that it has been impossible to pas s reforms well into a presidential mandate. However, an analysis of successful reforms shows that they are overwhelmingly stacked towards the beginning of a presidents tenure. In fact, all presidents from Sarney to Cardoso have proposed drastic reforms near the beginning of their terms, though not all have had the same level of success. The timing of reform is linked to Brazils system of presidentialism. The combination of presidentialism, a fragmented multiparty system, undisciplined parties, and ro bust federalism is often difficult[and] makes it diffult for presidents to establish reliable bases of support (Mainwaring 1997: 56). Because of this institutional structure, since the restoration of the democracy presidents have attempted to use their honeymoon period to push for contentious economic reforms. In presidential systems the transfer of power from one executive to another is a momentous occasion. Parliamentary systems generally provide relatively easy mechanisms for the removal of unpopular executives, such as the vote of no confidence. Parliamentarism also tends to be more flexible vis vis presidentialism because under the former executives do not have to serve for a fixed mandate. Popular prime ministers

PAGE 86

78 can call for early elections if it is thought opportune, or the opposition can attempt a vote of no confidence if it senses that the government has lost popular support. Moreover, under parliamentarism prime ministers inherently possess a congressional majority, either with a coalition of parties or by just their own party. In presidential systems, it is common for the executive and legislative branches to be controlled by different parties (Linz 1994). Brazil has a long history of presidentialism, yet its post authoritarian variant has been attenuated by the fragmentation of the party system. The high number of parties, as well as their general lack of discipline, forces Brazilian presidents to rely on often unstable and undependable congressional coalitions to support the president ial legislative agenda (Mainwaring 1997). Even if a presidents coalition possesses a majority (or super majority) in the Congress, there is still fear that coalition members will break ranks and go against the wishes of the president and party leaders. Brazils particular system of federalism further undermines a presidents authority. A further hindrance to presidents is the weakness of the coattails effect, in which presidents are generally able to muster support for their agendas following an elect ion. Samuels (2000) finds that in Brazil governors, rather than presidents, possess powerful coattails. As he explains, In Brazil federalism cuts presidential coattails short or simply rips the presidents coat off his back, cuts it up, and divides it am ong the governors. The root of this problem and the reason why it remains so intractable lies in the impact of the institutions of Brazilian federalism on the party system in limiting the linkage between national executive and legislative elections (Samue ls 2000: 17) Because of the volatile nature of coalition politics and the limited coattails effect, Brazilian presidents must adroitly take advantage of the political capital afforded to them during their honeymoon.

PAGE 87

79 All presidential systems witness the honeymoon effect to some extent, as the new leader is initially given some leeway by the opposition. However, in Brazil this period signifies an important window of opportunity in which it has been possible to muster support for important, yet contentious legislation. This phenomenon has been apparent with every president in Brazil since the end of the military regime, from Sarney to Cardoso. One of Sarneys first moves was the introduction of the Cruzado Plan, which failed to stabilize inflation. On his first day in office, Collor announced the firing of thousands of public sector employees, introduced a draconian stabilization plan (which also failed), and took the first steps towards liberalizing foreign commerce. Finally, Cardoso achieved most suc cess with his reform agenda during his first year in office. In short, all post transition presidents have used their honeymoon as a period to attempt to pass economic reforms, though with varying levels of success. With items still left on Cardosos econ omic and political reform agenda, it is possible that the next president (to be elected in October 2002) will also pass through the same experience. With the passing of time it is possible that this phenomenon will be less powerful than to date. Nonethel ess, recent history has demonstrated the importance of this period for new presidents. First vs. Second Generation Reforms In Brazil it has been easier to pass first generation as opposed to second generation economic reforms. First generation reforms re fer to areas such as stabilization, liberalization, and exchange rate policy, all of which are directly related to macroeconomic activity. These were generally the first measures undertaken in countries during the 1980s in response to the debt crisis (Hag gard and Kaufman 1992). In contrast, second generation reforms aim at changing areas such as the tax system, public

PAGE 88

80 administration, social security and other social programs, and the political system. While these areas do not have an immediate impact on the macroeconomy, collectively they are important because they form the root cause of the states fiscal imbalances. When speaking of reforms, it is easy to compress first and second generation reforms into one monolithic whole. Differentiating between these two categories is necessary, for different kinds of reforms face different kinds of political obstacles. This is because the ultimate success or failure of a particular reform depends much on the relationship between the various branches of governme nt. Some political systems allow presidents personally to implement certain types of legislation, sometimes without the consent of the legislature. This is the case in Brazil, where presidents wield significant legislative powers, such as the medida prov isria Two of the most significant achievements in the economic realm during the past decade the start of the privatization program under Collor and the implementation of the Real Plan by Franco were legislated by means of presidential decree. In other cases, the legislature is given the exclusive prerogative over certain types of laws. In Brazil has been the case when changes were sought in the constitution. Amendments must be approved twice by a three fifths vote of both houses of Congress. While C ongress supported Cardoso with the Big Five, it has been much less receptive to second generation types of reform, such as altering the social security system. The judicial branch can also be a key player by deciding on constitutionally questionable refor ms, by itself ordering certain measures, or by some other related means. In Brazil the judiciary has often been ultimate arbiter of several economic reforms. For example,

PAGE 89

81 the Supreme Court struck down in early 1999 a proposal from the Cardoso government that forced retired public workers to contribute to the social security system. Hence, when speaking of reforms, one must always be cognizant of the specific kind of reform being proposed, as well as the legislative means by which such a reform is to be implemented. Second generation reforms are politically less palatable in Brazil compared to first generation reforms. The latter, unlike those pertaining to social security or the bureaucracy, are subject to fewer veto players, individual or collective actors whose agreementis required for a change of the status quo (Tsebelis 1995: 289). This is partially due to the legislative powers available to the executive, as presidents are able to implement many laws by decree. Even in those areas in which the executive cannot act alone, congressional agreement with economic reform is often more forthcoming than in other areas of reform. For example, during the 1990s the Brazilian Congress generally sided with Presidents Collor and Cardoso in their plans for p rivatization. Since much of the privatization program required congressional approval (and not just presidential decrees), it would have been impossible for the program to have progressed as far as it has: $102 billion in divestitures from 1991 2001 (BNDE S 2001). Such success is directly attributable to the relative absence of what Tsebelis (1995) terms veto players with privatization vis vis broader, second generation reforms. The case of social security reform further illustrates the role of veto p layers. Figueiredo and Limongi (1998) contend that reform of the social security system ( previdncia ) failed not only because of the content of the proposal itself, but also because of the rules governing Congress. Reforming the social security system wo uld have required amending the constitution. As mentioned previously, amendments require

PAGE 90

82 two separate votes of three fifths of both houses of Congress. The reforms proposed by Cardoso would have an immediate impact on a large number of people (pensioners) which dissuaded politicians from voting for most of the presidents proposals. Furthermore, the procedures governing social security, which allowed the opposition to force a vote on piece by piece basis, further hindered the passage of the reform. The case of social security reform is an some ways the opposite of the Big Five, which involved a much smaller number of veto players, namely workers in particular SOEs. The Power of Ideas Changes in politicians opinions about the proper role of the state in the economy are another reason for the success of economic reforms. The spectacular growth of the Brazilian economy and the countrys rapid industrialization following the Second World War were directly attributable to the states active involvement in th e economy (Trebat 1983; Baer 1995). Although a states involvement in the economy inherently involves political considerations, a balance must be established between satisfying political considerations and the need for economic growth. Over time the Braz ilian economy became increasingly politicized, with the result being a worsening of performance. During the heyday of the military regime the number of SOEs expanded dramatically, large scale inefficient investments were undertaken, and by the late 1970s growth stagnated. Following the debt crisis of the 1980s, most Latin American countries implemented structural reforms with the aim of restoring growth. Brazil was the last major country in the region to pursue such reforms. There were several causes f or this late start. A concern for social issues and a desire to address the countrys enormous inequalities marked the transition to democracy and the promulgation of the 1988 constitution. The military regime was viewed as

PAGE 91

83 hostile towards the poor and t he working class. The drop in living standards and the worsening of income distribution during the 1970s and 80s further fueled resentment at past economic policies. Hence, there were pressures to address social ills either brought about or ignored by th e previous, military led government. Yet another reason for the delay with economic reforms, however, came from within the political class. Politicians were long accustomed to using economic tools as patronage. Attempts to liberalize the economy, privat ize SOEs, reform the system of revenue sharing between the federal government and lower levels of government all these signified a reduction in the instruments of patronage available to politicians. Although this was (and still is) the case in countries t he world over, many believed the problem to be intractable in Brazil, since corruption and cronyism grease the wheels of national politics. Despite this pessimistic prognosis, recent research indicates a change in Brazilian politicians attitudes toward th e states role in the economy. Timothy Power (1998a) analyzed several data sets to discover whether there has been an attitudinal shift among Brazilian congressmen (both deputies and senators) in regards to economic issues. Power was able to reach severa l conclusions from his analysis. First, he found that neoliberalism has made inroads among congressional politicians since the late 1980s. Second, politicians from the center left to the far right generally accept neoliberalism, although there are still questions about the depth of politicians commitment to neoliberalism. Third, President Cardoso has achieved an impressive string of legislative victories in his pursuit of neoliberal reform, mainly through skillful coalition management (Power 1998a: 67 ). According to Powers study, party discipline within the Cardoso coalition has been remarkably high, in contrast to what most analysts claim.

PAGE 92

84 Finally, the Left is isolated in its economic views, a fact that may bode poorly for its future political pros pects. What do these findings have to say about the future course of economic reforms in Brazil? Power cautions that his study is preliminary, yet believes that further research will support these initial findings. His belief is that with the exception o f the far Left the political class in Brazil will continue to accept neoliberal economic tenets, at least during the short term. Studies like Powers are important, for they reveal the underlying modes of thinking within the political class. As recent s cholarship has shown, ideas play an important role in politics (Hall 1989). In one case study, Sikkink (1991) describes the way in which the idea of desenvolimentismo or developmentalism, took hold in Brazil during the Kubitschek era, leading to the crea tion of state development oriented institutions. Neoliberalism likewise brings with it repercussions in the political sphere. President Cardoso, while ostensibly a social democrat, has aggressively pursued a neoliberal agenda. He has successfully enacte d most first generation reforms and consequently has focused his attention to reform of the state, which is directly linked to the nagging problem of the fiscal deficit. However, as discussed above such second generation reforms have been more difficult t o implement in Brazil. Influence of Political and Economic Variables One criticism lodged against institutionalism is that it overemphasizes the explanatory power of institutions in political outcomes. This has been the case in much of the scholarship on Brazilian politics in recent years. For example, Brazil is said to possess a system of paralyzed federalism, with a paralyzed competitive arena characterized by intractable tensions between the federal government on the one hand and

PAGE 93

85 the state and muni cpios on the other (Souza 1997: 170). The result of this paralyzed federalism has been a lack of governance, a highly problematic structural fiscal problem, and a lack of authority for Braslia vis vis lower levels of government. While Brazilian pre sidents, especially Cardoso, have attempted to recover the federal governments authority, such efforts have been hampered by weak parties, clientelistic politicians most interested in securing patronage for allies at the local level, and an apathetic poli tical culture that is distrustful of the political system. Considering this gloomy scenario, it should seem a miracle that economic reforms were ever passed in Brazil. Yet they have been passed, and it is doubtful that the country will experience a large scale regression to the statism of the past. These developments clearly call into question the institutionalist critique of Brazilian politics. First, the political landscape is changing, as evinced by politicians shifting attitudes towards neoliberalis m. Hence our past assumptions about the obstacles, i.e. political resistance, to reforms must be revised. Second, despite the influence of institutions, one must still pay heed to the power of political and economic variables. For example, by most acc ounts President Cardoso has been a skillful if cautious negotiator, displaying a high degree of political acumen when pursuing his reform agenda. We may legitimately question how different events would have been in Brazil since 1995 if Cardoso or a simila rly capable politician had not been in the Planalto. Besides political skill, there are a host of other factors at play in the course of the reform process. For example, economic pressures, both domestic and international, have been decisive in influenc ing the passage of economic reforms. According to Haggard and Kaufman (1992: 4), decision makers operate in a matrix of international, social, and

PAGE 94

86 institutional incentives that sets limits on the range of policy alternatives. In the scholarship on econ omic reforms in Brazil, institutional factors have been given almost exclusive attention while others, such as international and social ones, have received inadequate consideration. An important case demonstrating the influence of external economic press ures was the international financial assistance package of late 1998. Totaling $41.5 billion, the package was tied to an agreement between IMF and Brazil. The distribution of these funds was contingent on Brazils reaching specified primary account surpl uses, which many doubted could be accomplished. Yet, to the surprise of many, the country has consistently met its targets with the IMF. Cardoso has vigorously sought measures that would enhance fiscal discipline, both at the federal and local levels. T owards this end he secured passage of the Fiscal Responsibility Law in April 2000, which puts tight limits on discretionary social spending at the federal level and imposes penalties on fiscally profligate states (EIU 2000: 9). International pressures p layed a role in imposing fiscal discipline on Brazil. Following the devaluation of the real in early 1999, analysts were deeply concerned about the future course of the Brazilian economy. It was imperative to assuage the fears of international investors, and towards this end Cardoso was able to meet the fiscal targets agreed upon with the IMF, as well as the Fiscal Responsibility Law even though such measures limited the discretionary spending available as pork to congressional politicians.

PAGE 95

87 Conclusion De spite the bleak assessments offered by most political scientists, Brazil has undergone widespread economic reforms during the past decade. Factors such as the timing of reforms, the kind of reform proposed, changing ideas about the role of the state in th e economy, and (non institutional) political and economic pressures all help to explain how economic reforms have been possible in Brazils inchoate (Mainwaring and Scully 1995a) political system. Rather than criticizing the speed of reform in Brazil, K ingstone (2000) believes that this success is directly attributable to the slow pace of reform. According to him each of the New Republics presidents has been able to pass some small portion of the overall [reform] program. In turn, each successor has b een able to use those reforms as a base on which to build further . which has permitted the continuation of the process and has consequently kept Brazil muddling through gridlock (Kingstone 2000: 186) He recognizes the problems presented by Brazils insti tutional design and is doubtful that Brazil will be an immediate resolution of the economic problems emanating from the political system, such as a persistent fiscal deficit. He adds that long term resolution is likely to come, if at all, only with slow, incremental reforms (Kingstone 2000: 202). Reform has occurred and will continue to occur in Brazil, yet subject to the constraints of the countrys political system. This is not to say that further reform will not be possible. As recent years have de monstrated, reforms in Brazil occur at a slower pace than in other Latin American countries. This alone is not reason for pessimism.

PAGE 96

88 CHAPTER 5 CONCLUSION The purpose of this study has been twofold. First, it has provided an overview of contemporary scholarship on Brazilian politics. Much of this literature draws on institutionalism in its various guises to explain the difficulties encountered in implementing reforms in Brazil. Second, this study has attempted to highlight the gap between the predominant interpretation of Brazils reform efforts what I termed the institutionalist critique of Brazilian politics and the countrys successful recor d with economic reforms, which has seem much progress under Cardosos leadership. Main Findings Specifically, I argued that four factors help account for economic reforms in Brazil. First, the timing of a proposed reform can influence its ultimate succ ess or failure. Brazils combination of presidentialism, federalism, and multipartyism complicates the passage of economic reforms (Mainwaring 1997). Hence, Brazilian presidents at times have made use of their honeymoon period in office in order to pass reform legislation. In Chapter 4 I analyzed in detail the passage of some of the most important reform measures during the New Republic, especially the government of Fernando Henrique Cardoso, showing that most significant ecoomic reforms were proposed at the beginning of a presidential term. Not coincidently, must successful reforms occur during this period.

PAGE 97

89 Second, first generation economic reforms have been easier to pass than second generation reforms. This fact is due partially to the powers bestowe d upon the Brazilian president. The constitution of 1988 provided significant reactive (veto) and proactive (legislative) powers to the executive. Many first generation reforms have been realized by means of presidential decree, thus circumventing the Co ngress. In those areas in which Congressional support has been necessary, one sees that the areas in which significant reform has occurred entails fewer of what Tsebelis (1995) calls veto players. That is, those areas with fewer actors whose interests involved are easier to reform. Thus privatization has been impressive in Brazil, while social security reform, which effects millions of people, has proven much more difficult to change. Third, reforms have been possible thanks in part to changing attitud es among politicians. As discussed in Chapter 2, Brazil has a history of heavy state intervention in the economy. The states role had not only an economic logic, but also played an important political tool by providing the state and key politicians with patronage resources. Following the debt crisis of the 1980s and the hyperinflation of the early 1990s, it proved difficult to alter this traditional relationship between the state and the economy. In the words of historical institutionalism, the Brazili an political system displayed a high degree of stickiness in regards to economic policy. However, as Power (1998) has shown, only the Left still supports the traditional statist economic model, while most others sectors of the political class support, at least in theory, neoliberal ideals. This shift in politicians opinions has played a role in securing economic reforms in Brazil.

PAGE 98

90 Fourth and finally, institutionalist studies have too often emphasized domestic political factors while ignoring or paying in adequate attention to domestic and international economic forces. Brazils vulnerability to international forces has been a key explanatory factor in the passage of several pivotal reforms in recent years, such as the Law of Fiscal Responsibility. There is growing recognition that the countrys persistent fiscal deficit, as well as profligate spending by the states, harms the country by raising interest rates and worrying international investors. Beyond economic forces, I also argued that one must still pay heed to non institutional, political factors such as leadership. For example, one could argue that much of the success with economic reforms since the mid 1990s is attributable to the acumen of President Cardoso. He has been to overcome the problem s emanating from Brazils political institutional design, thus furthering his economic reform agenda. Collectively these factors help explain the passage of extensive economic reforms in Brazil. Although studies of Brazilian politics since the transition to democracy have been fixated on the institutional inadequacies of the New Republic, one can now perceive a shift in scholars attitudes towards the effect of institutions on Brazilian politics in general. In a recent review of the literature on Brazilia n politics and political economy, one scholar noted that works on Brazilian party systems and legislative executive relations continue to be dominated by outdated, impressionistic, and overdrawn arguments and evidence, with the result being the perpetuat ion of pessimism regarding the nature and accomplishments of post transition democracy in Brazil and the countrys efforts to reform its economy (Resende Santos 2001: 209). This shifting attitude is most

PAGE 99

91 apparent among Brazilian political scientists, wi th many North American scholars still subscribing to the institutionalist critique laid forth in Chapter 3. One of the problems with the prevailing institutionalist critique of Brazilian politics may not be so much overemphasizing the power of institutions as it is examining the wrong institutions. With so much attention given to the political institutions examined in the Chapter 3, scholars have often overlooked the impact of other variables. Such a move has already started to occur, again spearheaded by Brazilian scholars. According to Figueiredo and Limongi (2000: 168), the characteristics of the decision making process the legislative powers of the president and the legislative organization may be more appropriate determinants of governability than t he form of government, the characteristics of party system, or the electoral system. They find that since the promulgation of the 1988 constitution, presidents have had a very high rate of success in implementing their agendas, due principally to the le gislative powers of the president and the centralized organization of the legislative work (Figueiredo and Limongi 2000: 151). Their findings counter the conventional description of Brazilian political parties. Riordan Roett, one of the foremost U.S. a nalysts of Brazilian politics, states that Brazilian political parties are best understood as shifting groups of self interested individuals who find it convenient to come together under a bland banner that usually includes the words progressive, social, or democratic (1997: 21). While Roetts criticism still possesses some truth, research like that done by Figueiredo and Limongi demonstrate that common notions of the Brazilian political system are becoming increasingly outdated.

PAGE 100

92 Suggestions for Future Research The conclusions of this study point to topics for future research. Considering the outmoded perspective of much of contemporary scholarship on Brazilian politics and political economy, future studies must move beyond the conventional notion that the Brazilian political system inherently impedes reform. Scholars must take it upon themselves to reassess the impact of institutions on economic policy in general and to look for answers to how reforms are passed. In some cases, studies could benefit from a comparative, cross country perspective. First, future research should explore in greater detail the attitudes of politicians towards the proper role of the state in the economy. Powers (1998) analysis, as he himself states, only offers tentativ e conclusions. While his survey data from members of Congress suggest that politicians have increasingly accepted neoliberal ideals, only time will show the depth of their commitment to less state intervention. With a new administration to be elected in late 2002, it will be interesting to see the level of continuity in economic policy between the present and incoming administrations. The success of the Cardoso government has been due in large part to the maintenance of a broad coalition, which has remai ned united principally through the dispensation of favors and pork to coalition members. Further studies of the party system are needed, as presently there are conflicting accounts of the level of coherence and discipline of Brazilian political parties. On the one hand, scholars such as Mainwaring (1999) stress the low level of institutionalization of parties in Brazil. However, others such as Figueiredo and Limongi (2000) stress that parties have displayed a high degree of discipline since the start o f the New Republic, thus aiding presidents in promoting their agendas. Certainly, the proliferation of parties

PAGE 101

93 has stabilized in recent years, but problems such as party switching by politicians remain. Understanding the party system is essential for th e reasons addressed in Chapter 3. Political parties are an integral part of a democratic system, facilitating the functioning of government by aggregating citizens interests and allowing for bargaining between the executive and the legislative branches. One hopes that Brazilian political parties are in fact becoming more stable and programmatic, yet much research still needs to be done on the topic before we can know with certainty. One of the most important areas for greater study is the role of interna tional economic factors on contemporary economic policy making. Brazils position in the global economy makes it vulnerable to external pressures, as the devaluation of the real in early 1999 and subsequent IMF agreement attest. During the 1980s and earl y 1990s Brazil often eschewed the prescriptions of the Washington Consensus. Now the current government is staffed with several prominent neoliberal adherents. There are even signs that a future PT government would be sensitive to international economic forces ( Jornal do Brasil 2001). As Haggard and Kaufman (1992) state, economic policies are shaped by a mix of domestic, international, and institutional factors. Recent scholarship on Brazil has focused heavily on domestic institutional factors, yet int ernational economic forces seem to have significantly influenced economic policy in Brazil during the past several years. In November 1999 the Fator Previcdencirio was passed, which created a new system for calculating pension benefits for private secto r workers. In April 2000 the government successfully passed the Fiscal Responsibility Law, which placed limits on spending by states and imposes penalties on those that do not maintain fiscal discipline. And while

PAGE 102

94 much work still remains to be done with shoring up the social security system, measures have been enacted that at least temporarily ease pressure on the governments coffers. All these measures have helped to improve the governments accounts. Understanding the relationship between these measu res and the exigencies of international economic and financial forces should be the subject of future enquiry. * As the past century has shown, Brazil is a country of immense potential. Between 1900 and the 1980s, Brazil had one of the highest rates o f growth in the world (Rigolon and Giambiagi 1999: 6). The country was transformed from a poor, agricultural country into the industrial powerhouse of the developing world. During the past two decades, however, economic growth has remained below historic averages. Political scientists saw Brazils political institutions as the cause for the countrys poor reform record. While institutionalism helped explain Brazils late start with reforms, it has been less useful in explaining the reforms that have occ urred since Fernando Henrique Cardoso took office in 1995.

PAGE 103

95 BIBLIOGRAPHY Abrcio, Fernando Luiz. 1994. Os Bares da Federao. Lua Nova 33: 165 83. Almeida, Maria Hermnia Tavares de. 1996. Pragmatismo por Necessidade: Os Rumos da Reforma Econmica no Brasil. Dados 39(2): 213 34. Ames, Barry. 1995a Electoral Strategy under Open List Proportional Representation. American Journal of Political Science 39(2): 406 33. ------. 1995b. Electoral Rules, Constituency Pressures, and Pork Barrel: Bases of Voting in the Brazilian Congress. Journal of Po litics 57(2): 324 43. ------. 1999. Approaches to the Study of Institutions in Latin American Politics. Latin American Research Review 34(1): 221 36. BNDES (Banco Nacional de Desenvolvimento Econmico e Social). 1996. Relatrio de Atividades 1 995, Programa Nacional de Desestatizao ------. 2001. Homepage < http://www.bndes.gov.br/pndnew/already.htm> Accessed 15 May. Baer, Werner. 1994. Privatisation in Latin America. The World Economy 17(4): 509 28. ------. 1995. The Brazilian Economy: Growth and Development 4 th ed. Westport, CT: Praeger. ------. 1996. Changing Paradigms: Changing Interpretations of the Public Sector in Latin Americas Economies. Public Choice 88: 365 79. Baer, Werner, Isaac Kerstenetzky and Anniba l Villela. 1973. The Changing Role of the State in the Brazilian Economy. World Development 1(11): 23 34. Baer, Werner, Richard Newfarmer, and Thomas Trebat. 1976. On State Capitalism in Brazil: Some New Issues and Questions. Institute of Latin American Studies, University of Texas at Austin, Technical Papers Series No. 1 Bardhan, Pranab. 1989. The New Institutional Economics and Development Theory: A Brief Critical Assessment. World Development 17(9): 1389 95.

PAGE 104

96 Becker, Bertha and Claudio Egler. 1992. Brazil: A New Regional Power in the World Economy Cambridge: Cambridge University Press. Belassa, Bela and Gerardo Bueno. 1986. Toward Renewed Economic Growth in Latin America Washington, DC: Institute for International Economics. Bomfim, Antlio and Anwar Shah. 1994. Macroeconomic Management and the Division of Powers in Brazil: Perspectives for the 1990s. World Development 22(4): 535 42. Bresser Perreira, Luiz Carlos. 1996. Economic Crisis and State Reform in Brazil: T oward a New Interpretation of Latin America Boulder: Lynne Rienner. ------. 1998. A Economia Brasileira: Uma Introduo Crtica 3 rd ed. Rio de Janeiro: Editora 34. Bulmer Thomas, Victor. 1994. The Economic History of Latin America since Indep endence New York: Cambridge University Press. Burns, E. Bradford. 1993. A History of Brazil 3 rd ed. New York: Columbia University Press. CEPAL. 2001. Balance Preliminar de las Economas Latinoamericanas Santiago: CEPAL. Collins, Joseph and John Lear. 1994. Chiles Free Market Miracle: A Second Look Oakland: Food First Books. Coppedge, Matthew. 1998. The Evolution of Latin American Party Systems. Pp. 171 206 in Scott Mainwaring and Arturo Valenzeula, ed. Politics, Society, and Demo cracy: Latin America Boulder, CO: Westview Press. Dean, Warren. 1995. With Broadax and Firebrand: The Destruction of the Brazilian Atlantic Forest Berkeley: University of California Press. EIU (Economist Intelligence Unit). 2001. Brazil Country P rofile 2000. Evans, Peter. 1979. Dependent Development: The Alliance of Multinational, State and Local Capital in Brazil Princeton, NJ: Princeton University Press. ------. 1992. The State as Problem and Solution: Predation, Embedded Autonomy, and Structural Change. Pp. 139 81 in Haggard and Kaufman (1992). Evans, Peter, Dietrich Rueschemeyer, and Theda Skocpol, ed. 1985. Bringing the State Back In Cambridge: Cambridge University Press.

PAGE 105

97 Figueiredo, Argelina Cheibub and Fernando Limongi. 199 5. Partidos Polticos na Cmara de Deputados, 1989 1994. Dados 38(3): 497 525. ------. 1998. Reforma da Previdncia e Instituies Polticas. Novos Estudos 51: 63 90. ------. 2000. Presidential Power, Legislative Organization, and Party Be havior in Brazil. Comparative Politics 32(2): 151 70. Fleischer, David. 1998. The Cardoso Governments Reform Agenda: A View from the National Congress, 1995 1998. Journal of Interamerican Studies and World Affairs 40(4): 119 136. Forbes 199 3. The Sick Man of Latin America. September 13. Gamarra, Eduardo. 1994. Market Oriented Reforms and Democratization in Latin America: Challenges of the 1990s. Pp. 1 15 in William Smith, Carlos Acua, and Eduardo Gamarra, ed. Latin American Politic al Economy in the Age of Neoliberal Reform Boulder, CO: Lynne Rienner. Goertzel, Ted. 1999. Fernando Henrique Cardoso: Reinventing Democracy in Brazil Boulder: Lynne Rienner. Goodin, Robert and Hans Dieter Klingemann, ed. 1996. A New Handbook o f Political Science New York: Oxford University Press. Green Donald and Ian Shapiro 1996 Patholog ies of Rational Choice Theory: A Critique of Applications in Political Science New Haven : Yale University Press Haggard, Stephen and Robert Kaufman, ed. 1992. The Politics of Economic Adjustment: International Constraints, Distributive Conflicts, and the State Princeton, NJ: Princeton University Press. Hagopian, Frances. 1986. State Capitalism and Politics in Brazil. Working Paper #63, Kellogg Institute for International Studies. University of Notre Dame. Hall, Peter, ed. 1989. The Political Power of Ideas: Keynesianism across Nations Princet on, NJ: Princeton University Press. Hall, Peter and Rosemary Taylor. 1996. Political Science and the Three New Institutionalisms. Political Studies 44: 936 57. Immergut, Ellen. 1998. The Theoretical Core of the New Institutionalism. Politics and Society 26(1): 5 34. Jornal do Brasil 2001. PT se preocupa em negar calote e quer alongamento. 16 June.

PAGE 106

98 Just, Diane. 1999. The Influence of State Institutions on the Implementation of Neoliberal Economic Reforms: Evidence from the Mexican Case Ph.D. dissertation, University of Florida. Kaufman, Robert. 1990. Stabilization and Adjustment in Argentina, Brazil, and Mexico. Pp. 63 111 in Joan Nelson, ed. Economic Crisis and Policy Choice: The Politics of Adjustment in the Third World Pr inceton, NJ: Princeton University Press. Kaufman Purcell, Susan and Riordan Roett, ed. 1997. Brazil Under Cardoso Boulder: Lynne Rienner. Kingstone, Peter. 1999. Crafting Coalitions for Reform: Business Preferences, Political Institutions, and Ne oliberal Reform in Brazil University Park, PA: Pennsylvania State University Press. ------. 2000. Muddling through Gridlock: Economic Policy Performance, Business Responses, and Democratic Sustainability. Pp. 185 203 in Kingstone and Power (2000) Lamounier, Bolvar. 1989. Brazil: Inequality against Democracy. Pp. 119 69 in Larry Diamond, Juan Linz, and Seymour Lipset, ed. Democracy in Developing Countries: Latin America Boulder: Lynne Rienner. Lamounier, Bolvar and Edmar Lisboa Bacha. 1994. Democracy and Economic Reform in Brazil. Pp. 143 205 in Joan M. Nelson, ed. A Precarious Balance: Democracy and Economic Reforms in Latin America San Francisco: ICS Press. LARRB (Latin American Regional Reports Brazil). 1994a. Clear first r ound victory a double edged sword for Cardoso. 27 October: 1. ------. 1994b. Cardoso will have to make deals. 27 October: 6 7. Levi, Margaret. 1990. A Logic of Institutional Change. Pp. 402 19 in The Limits of Rationality ed. K.S. Cook and Lev i. Chicago: University of Chicago Press. Lijphart, Arend and Carlos Waisman, ed. 1996 Institutional Design in New Democracies: Eastern Europe and Latin America Boulder, CO: Westview Press. Lima, Paulo Tarso Flecha de. 1999. Liberalism versus Nationalism: The Prodevelopment Ideology in Recent Brazilian History. Presidential Studies Quarterly 29(2). Linz, Juan. 1990. The Perils of Presidenti alism. Journal of Democracy 1(1): 51 69.

PAGE 107

99 ------. 1994. Presidential or Parliamentary Democracy: Does it Make a Difference? Pp. 3 87 in Linz and Valenzuela (1994). Linz, Juan and Alfred Stepan. 1996. Problems of Democratic Transition and Consoli dation Baltimore: Johns Hopkins University Press. Linz, Juan and Arturo Valenzuela, ed. 1994. The Failure of Presidential Democracy: The Case of Latin America Baltimore: Johns Hopkins University Press. Love, Joseph. 1995. Ral Prebisch and the Origins of the Doctrine of Unequal Exchange. Pp. 99 120 in James Dietz, ed. Latin Americas Economi c Development: Confronting Crisis Boulder: Lynne Rienner. Mainwaring, Scott. 1992 93. Brazilian Party Underdevelopment in Comparative Perspective. Political Science Quarterly 107(4): 677 707. ------. 1997. Multipartism, Robust Federalism, and Pr esidentialism in Brazil. Pp. 55 109 in Mainwaring and Shugart (1997). ------. 1999. Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil Stanford, CA: Stanford University Press. Mainwaring, Scott. and Timothy Scully. 1995a. Introduction: Party Systems in Latin America. Pp. 1 34 in Mainwaring and Scully, ed. Building Democratic Institutions: Party Systems in Latin America Stanford: Stanford University Press. ------. 1995b. Conclusion: Parties and Democracy in Latin America Different Patterns, Common Challenges. Pp. 459 74 in Mainwaring and Scully, ed. Building Democratic Institutions: Party Systems in Latin America Stanford: Stanford University Press. Mainwaring, Scott and Matthew Shugart, ed. 1997 Presidentialism and Democracy in Latin America New York: Cambridge University Press. North, Douglass. 1990. Institutions, Institutional Change and Economic Development New York: Cambridge University Press. ODonnell, Guillermo. 1973. Moderniza tion and Bureaucratic Authoritarianism Institute of International Studies, University of California, Berkeley. ------. 1978. Reflections on the Patterns of Change in the Bureaucratic Authoritarian State. Latin American Research Review 13(1): 3 3 8. ------. 1994. Delegative Democracy. Journal of Democracy 5(1): 55 69.

PAGE 108

100 Peters, B. Guy. 1996. Political Institutions, Old and New. Pp. 205 20 in Goodin and Klingemann (1996). Power, Timothy. 1998a. Brazilian Politicians and Neoliberalism: M apping Support for the Cardoso Reforms, 1995 1997. Journal of Interamerican Studies and World Affairs 40(4): 51 72. ------. 1998b. The Pen is Mightier than the Congress: Presidential Decree Power in Brazil. Pp. 197 230 in John Carey and Matthew Shugart, ed. Executive Decree Authority Cambridge: Cambridge University Press. ------. 2000. Political Institutions in Democratic Brazil: Politics as a Permanent Constitutional Convention. Pp. 17 35 in Power and Kingstone, 2000. Power, Timothy a nd Peter Kingstone. 2000. Democratic Brazil: Actors, Institutions, and Processes Pittsburgh: University of Pittsburgh Press. Przeworski, Adam. 1991. Democracy and the Market New York: Oxford University Press. Raw, Sylvia. 1987. The Making of a State Owned Conglomerate: A Brazilian Case Study. Working Paper #97, Kellogg Institute, University of Notre Dame. Reich, Gary. 1998. The 1988 Constitution a Decade Later: Ugly Compromises Reconsidered. Journal of Interamerican Studies and World Aff airs 40(4): 5 24. Resende Santos, Joo. 2001. Democracy, Equity and Governance in Brazil. Latin American Research Review 36(1): 207 37. Rigolon, Francisco Jos Zagari and Fabio Giambagi. 1999. The Brazilian Economy: General Overview. Monograph BNDES. Roett, Riordan. 1997. Brazilian Politics at Centurys End. Pp. 19 42 in Kaufman Purcell and Roett (1997). Rothstein, Bo. 1996. Political Institutions: An Overview. Pp. 133 66 in Goodin and Klingemann (1996). Samuels, David. 2000. Concu rrent Elections, Discordant Results: Presidentialism, Federalism, and Governance in Brazil. Comparative Politics 33(1): 1 20. Selcher, Wayne. 1998. The Politics of Decentralized Federalism, National Diversification, and Regionalism in Brazil. Journ al of Interamerican Studies and World Affairs 40(4): 25 50.

PAGE 109

101 Selznick, Philip. 1996. Institutionalism Old and New Administrative Science Quarterly 41: 270 77. Shapiro, Helen. 1994. Engines of Growth: The State and Transnational Auto Compani es in Brazil Cambridge: Cambridge University Press. Sikkink, Kathryn. 1991. Ideas and Institutions: Developmentalism in Brazil and Argentina Ithaca: Cornell University Press. Skidmore, Thomas. 1967. Politics in Brazil, 1930 1964: An Experiment i n Democracy Oxford: Oxford University Press. Smith, William. and Nizar Messari. 1998. Democracy and Reform in Cardosos Brazil: Caught Between Clientelism and Global Markets? North South Agenda Papers, No. 33, North South Center, University of Miami Soares, Glucio Ary Dillon. Forthcoming. A Democracia Interrompida Rio de Janeiro: Editora FGV. Sola, Lourdes. 1994. The State, Structural Reform, and Democratization in Brazil. Pp. 151 181 in William Smith, Carlos Acua, and Eduardo Gamarra, ed. Democracy, Markets, and Structural Reform in Latin America Miami: North South Center Press. Souza, Celine. 1997. Constitutional Engineering in Brazil: The Politics of Federalism and Decentralization New York: St. Martins Press. Stepan, Alf red. 1978a. The State and Society: Peru in Comparative Perspective Princeton: Princeton University Press. ------. 1978b. Political Leadership and Regime Breakdown: Brazil. Pp. 110 37 in Juan Linz and Alfred Stepan, ed. The Breakdown of Democra tic Regimes: Latin America Baltimore: Johns Hopkins University Press. ------. 2000. Brazils Decentralized Federalism: Bringing Government Closer to the Citizens? Daedalus 129(2): 145 69. Teichman, Judith. 1996. Privatization and Political Chan ge in Mexico Pittsburgh: University of Pittsburgh Press. Topik, Steven. 1979. The Evolution of the Economic Role of the Brazilian State, 1889 1930. Journal of Latin American Studies 11: 325 42. Trebat, Thomas. 1983. Brazils State Owned Enterpr ises: A Case Study of the State as Entrepreneur New York: Cambridge University Press.

PAGE 110

102 Tsebelis, George. 1995. Decision Making in Political Systems: Veto Players in Presidentialism, Parliamentarism, Multicameralism and Multipartyism. British Journa l of Political Science 25:289 325. Weyland, Kurt. 1993. The Rise and Fall of President Collor and its Impact on Brazilian Democracy. Journal of Interamerican Studies and World Affairs 35(1): 1 37. ------. 1996. Development, Democracy, and Ineq uity: New Contributions on the Brazilian Political Economy. Latin American Research Review 31(3): 189 202. Williamson, John. 1990. Latin American Adjustment: How Much Has Happened? Washington, D.C.: Institute for International Economics. Wirth, J ohn. 1970. The Politics of Brazilian Development, 1930 1954 Stanford: Stanford University Press.

PAGE 111

103 BIOGRAPHICAL SKETCH Brandon Lee Knox received his Bachelor of Arts in economics from Stetson University in 1999. During his time as an undergraduate he spent a semester at the Universidad de Guanajuato in Mexico, as well as a month in Peru studying the a ctivities of no n governmental organizations. While a t the University of Florida he studied at the Instituto Brasil Estados Unidos and the Pontifcia Universidade Catlica, both in Rio de Janeiro, Brazil. During the fall of 2001, he will participate in a Univer sity of Florida sponsored internship program with Prudential Icatu Gesto de Investimentos Imobilirios, a real estate investments firm in Rio de Janeiro. After receiving his masters degree in December 2001, he hopes to pursue a career in international b usiness.


Permanent Link: http://ufdc.ufl.edu/UFE0000325/00001

Material Information

Title: Reassessing the Impact of Institutions on Economic Reform in Brazil
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000325:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000325/00001

Material Information

Title: Reassessing the Impact of Institutions on Economic Reform in Brazil
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000325:00001


This item has the following downloads:


Full Text











REASSESSING THE IMPACT OF INSTITUTIONS ON
ECONOMIC REFORM IN BRAZIL

















By

BRANDON KNOX


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS

UNIVERSITY OF FLORIDA


2001




























Copyright 2001

by

Brandon Knox



























To my parents, who have always shown love and support















ACKNOWLEDGMENTS

I would like to thank the Center for Latin American Studies at the University of

Florida for the financial assistance provided to me during my time as a graduate student.

My experiences at the Center have exceeded all my expectations and I am truly grateful

to all the faculty and staff with whom I have had the pleasure to work. The Center

awarded me a Title VI fellowship for the study of Portuguese at the Instituto Brasil-

Estados Unidos in Rio de Janeiro during the summer of 1999, which began what I hope

will be a life-long relation with this wonderful country. During the summer of 2000 the

Center for International Business Education and Research (CIBER) provided me with an

award to study in the newly formed Business in Brazil program at the Pontificia

Universidade Cat6lica do Rio de Janeiro.

I would also like to thank the members of my thesis committee, David Denslow,

Terry McCoy, and Philip Williams. I want to extend a special thanks to Terry McCoy.

From the beginning of my inquiries into the MALAS program at UF, he displayed an

incredible degree of attention and helpfulness. I had the privilege to work as his graduate

research assistant for the duration of my time in the MALAS program, and my work with

him has benefited me immensely both academically and professionally. I am extremely

grateful to him for his advice and guidance.

Finally, none of what I have done would have been possible without the unfailing

love and support of my parents, Steve and Rhinda Knox.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

A B ST R A C T ...................................................................................................................... vii

CHAPTERS

1 INTRODUCTION: INSTITUTIONS AND POLITICS ...............................................1
The B razilian Puzzle ........................................................... .... .......... 2
T theoretical F ram ew ork ......................................................... .................................. 6
Political Science and Institutionalism.................. ............................. 8
Institutions and the Study of Latin American Politics.................................... 12
Structure of the Study .... ............................ .... ............................ .............. 19

2 THE ROLE OF THE STATE IN THE BRAZILIAN ECONOMY ..............................21
Early History: Economic Cycles and External Vulnerability .......................... 24
V argas and the O rigins of State Capitalism ............................................ .... .. .............. 27
The Military Regime: From Boom to Bust ................. .............................. 34
Democratization and Demise of the State-Led Model............................................ 39

3 BRAZILIAN POLITICAL INSTITUTIONS................................ ............................43
Political Institutions and Economic Reforms ...................................................... 47
The 1988 Constitution and the Reform Process..................................................... 48
F ed eralism ................................................. ....... ......... ...... 4 9
Presidentialism ................... .. ......... 54
The Party System ......... ...... ... .................... ........ .. .... ...... .. .......... .. 57
The Electoral System and C ongress............................................... ... ................. 60
C o n c lu sio n .............................................................................. 6 1

4 ACCOUNTING FOR ECONOMIC REFORMS IN BRAZIL.................. ............64
The Changing Brazilian Political Economy..................... ......................................... 65
The Vagaries of Economic Reforms in the New Republic................ .................... 66
Cardoso and the Economic Reform Agenda.................................................... 70
The Big Five Constitutional Amendments of 1995 ............................................... 73
Accounting for Economic Reform s in Brazil .............. .............................. ....... ....... 76
Timing ....... ................................................ ............ ........... 77
First- vs. Second-Generation Reforms ........................................................ 79
T he P ow er of Ideas .............................. ................................... ........ ................... 82









Influence of Political and Economic Variables....................................................... 84
C onclu sion .................... ................................ ....... .......... ..... 87

5 CON CLU SION .................. .................. .................. .. ......... ... ............ .. 88
M ain F findings ................ .................................... 88
Suggestions for Future R esearch..................................................................................... 92

BIBLIO GRAPH Y .................. ...................................................... .. ......... 95

BIOGRAPHICAL SKETCH ............................................................. ..................103













































vi















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts

REASSESSING THE IMPACT OF INSTITUTIONS ON
ECONOMIC REFORM IN BRAZIL

By

Brandon Knox

December 2001

Chairman: Terry L. McCoy
Major Department: Latin American Studies

During the1980s and early 1990s Brazil, like much of Latin America, passed

through a period of economic crisis, as the country struggled to pay its foreign debt and

inflation reached four-digit levels. While most large countries in Latin American began

to enact structural, or "neoliberal," reforms in response to this crisis, Brazil was often

seen a laggard. Political scientists looked at the country's political institutions for

reasons for Brazil's delay in passing economic reform. This emphasis on political

institutions mirrored a general trend in political science, which is often referred to as the

"new institutionalism."

The problematic design of Brazilian political institutions, such as federalism,

presidentialism, multipartism, the constitution of 1988, the electoral system, and

Congress, played a role in Brazil's late start with economic reform. However, extensive

reforms have been implemented in recent years, most notably during the presidency of

Fernando Henrique Cardoso. Despite the extensive economic reforms that have been









passed since the mid-1990s, political scientists, especially those from North America, still

stress the deleterious nature of Brazil's institutional design. I argue for a reassessment of

the impact of institutions on economic reform in Brazil. Specifically, I look at four

factors that help to account for economic reform in Brazil: the timing of reform; the type

of reform proposed (first- versus second-generation reforms); the role of ideas and

ideology among the political class; and the role of economic and (non-institutional)

political factors. Collectively, these factors help to explain how economic reforms have

been so extensive in recent years, in spite of the problematic design of the Brazilian

political system.














CHAPTER 1
INTRODUCTION: INSTITUTIONS AND POLITICS

This study will examine the relationship between Brazil's political institutions and

efforts to implement economic reforms. While there have been attempts to restructure

the country's economy since the advent of the New Republic in 1985,1 this study will

concentrate on developments during the 1990s, principally the presidency of Fernando

Henrique Cardoso (1995-present). There have been numerous studies analyzing how

certain institutional features of the Brazilian political system make it difficult to

implement needed economic (as well as political and social) reforms (Bresser Perreira

1996; Fleischer 1998; Lamounier and Bacha 1994; Mainwaring 1999: chap. 10; Roett

1997; Selcher 1998; Stepan 2000). But this focus on the institutional obstacles to reforms

in Brazil generally seems to ignore the extensive economic reforms that have occurred to

date. As a result of this perspective, there has been little systematic examination of how

such reforms have been possible.

This study does two things. First, it reviews the current literature on Brazilian

politics, most of which uses an institutionalist perspective. This literature is, in general,

highly pessimistic about the possibilities for deep reforms in Brazil because of the



1 The "New Republic" refers to the political regime that has been in place in Brazil since
the end of military rule. This term is employed almost exclusively within academic
circles and is not used frequently by the Brazilian public. There is much disagreement
about the extent to which the New Republic meets the criteria of a consolidated
democracy (see Linz and Stepan 1996: chap. 11). Regardless of one's position in this
debate, however, all cite 1985 (the year in which the first civilian president assumed
office) as the beginning of the New Republic.









obstacles emanating from the political system. Although the first decade of the New

Republic was characterized by a litany of unsuccessful stabilization plans and piecemeal

economic reforms, the latter half of the 1990s produced extensive changes. Most

analyses see these advances as anomalies, instances in which Cardoso has been able to

overcome political-institutional constraints to advance painful but necessary changes.

Second, this thesis will highlight the shortcomings of the institutionalist critique

of Brazilian politics. I contend that Brazil is still plagued with numerous problematic

institutional arrangements, yet recent developments force us to reassess their effect on

economic reform. After reviewing the course of the reform agenda during the four

presidencies of the New Republic, I will discuss what factors help explain how economic

reforms have been implemented successfully in Brazil. Specifically, I analyze four

factors: the timing of reform; the type of reform proposed (first- versus second-generation

reforms); the role of ideas and ideology; and political and economic pressures.


The Brazilian Puzzle

At the beginning of the 1990s analysts considered Brazil a laggard in implementing

structural economic reforms, going so far as to brandish it the "sick man of Latin

America." (Forbes 1993). For most of Latin America the 1980s had been a "lost decade"

during which living standards fell precipitously and countries struggled to meet crushing

debt obligations. At that time all countries in the region, with the sole exception of

Colombia, defaulted on their debt and most were forced to seek assistance from the U.S.

government or from international financial institutions (IFIs) led by the International

Monetary Fund and the World Bank. In order to avert economic crises like that which









struck the region during the 1980s, economists, especially those from IFIs and the U.S.

government, began to advocate structural, or "neoliberal," reforms. According to John

Williamson (1990) of the Institute for International Economics, one of the leading think-

tanks advocating neoliberal reforms, countries in Latin America should pursue nine

policies to stabilize and restructure their economies. These measures included: (1) fiscal

discipline, (2) changing priorities for public spending, (3) tax reform, (4) financial

liberalization, (5) competitive exchange rates, (6) trade liberalization, (7) promotion of

foreign direct investment, (8) deregulation and (9) privatization of state-owned

enterprises (SOEs).

The core of the "Washington Consensus" was the reduction of the role of the state

in the economy. During the Great Depression many political leaders in Latin America,

most notably in the larger republics such as Brazil, Argentina, and Mexico, began to

advocate an activist role for the state in order to reduce their dependence on export

earnings. Although policies varied greatly, the core of this activism lay in the policy of

import-substitution industrialization (ISI). ISI entailed the state working to foster,

through a variety of incentives, the development of industries for goods that traditionally

had been imported. The creation of SOEs was an integral part of ISI in most countries, as

the state came to assume control of the commanding heights of the economy. Over time,

however, many of these enterprises became highly inefficient and instruments of political

patronage. Many economists argued that misguided policies pursued during the era of

state activism following the Second World War were in part responsible for the travails

of the 1980s. While many countries in Latin America began to adopt neoliberal measures









during the late 1980s, Brazil had progressed little towards reforms in these various areas

by the early 1990s.

When examining the reasons for Brazil's delay in implementing reforms,

increasing attention was given to political obstacles. First, in 1985 the armed forces,

which had been in power since 1964, allowed for the transfer of power to a civilian

(though not popularly elected) president, Jose Sarney. It was not until the election of

Fernando Collor de Mello in 1990 that the chief executive was directly elected, the first

direct presidential election in the country since 1960. Due to the transitional and delicate

nature of politics at that time, some argued it was extremely difficult for a newly installed

civilian regime to implement extensive economic reforms, which would likely entail a

high degree of austerity for much of the population (Sola 1994). Second, many pointed

to institutional aspects of the Brazilian political system that hampered the reform process:

the new constitution of 1988 and its nationalist economic mandates, the federal system,

the weakness of political parties and the party system, and the electoral system, among

others (Lamounier and Bacha 1994).

Economic reform became all the more pressing as inflation continued to

accelerate during the early 1990s. Inflation had surpassed four-digit figures by 1989 and

worsened during the early 1990s (Rigolon and Giambagi 1999). Furthermore, since the

country had the largest economy in Latin America and was a major trading partner with

its neighbors, the state of the Brazilian economy affected events beyond its own borders.

While President Collor did successfully undertake reforms in certain areas, notably with

the liberalization of trade and the beginning of privatization of SOEs, it was not until

Fernando Henrique Cardoso was elected president in 1994 that a coherent economic









reform platform was pursued. During Cardoso's tenure Brazil has deepened its

privatization program, record levels of foreign investment have entered the country, and

inflation, which had exceeded an annual rate of one thousand percent during the early

1990s, has been under ten percent for several years. In sum, Cardoso's achievements in

the economic realm have been extensive. These reforms are all the more noteworthy in

light of the political obstacles that the president had to overcome to achieve them.

Most analysts agree that Brazil's difficulties with passing economic reforms

during since the return to democracy in 1985 had their roots in the country's political

institutions. Yet, as Cardoso's presidency attests, it has been possible to implement an

array of reforms in Brazil in spite of the institutional barriers posed by the country's

fractious political system. In 1995 he secured several important amendments to the

constitution that ended the state's monopoly over certain sectors of the economy (mining,

coastal shipping, gas distribution, and, in a different form, oil exploration) and also

eliminated the distinction between domestic and foreign capital. This legislation allowed

for the subsequent privatization of SOEs in these sectors and helped to attract record

levels of foreign investment into Brazil.

The constitutional amendments of 1995 mentioned above, also known as the "Big

Five," will provide a case study for analyzing not how political obstacles confound

reforms in Brazil, but rather how these obstacles can and have been overcome. In order

to address this question, it will be necessary to examine the role of institutions in politics.

The following section provides an overview of the "new institutionalism" literature in

political science and how this school of thought pertains to economic reforms in Brazil.









Theoretical Framework

Since the 1980s political scientists have paid increasing attention to institutions

and the ways they affect politics. According to Bo Rothstein (1996: 135-6),

Political scientists ask three different but interrelated questions about political
institutions. One is normative: Which institutions are best suited for creating
"good" government and societal relations. The other two are empirical: What
explains the enormous variation in institutional arrangements? And what
difference those differences make for political behavior, political power and the
outcome of the political process?

In this study I will explain how the design of many Brazilian political institutions has

impeded the passage of reform legislation. At the same time, I will also argue that the

institutionalist literature does not adequately take into account the extensive reforms that

have been implemented to date. Such successful reforms force us to reassess the

common notion that Brazil's fractious political system makes it extremely difficult to

implement contentious measures.

Institutionalism is a common framework with which scholars have analyzed the

political economy of economic reforms in Brazil because it is an "intermediate level" of

analysis of political phenomena (Just 1999: 17-19). A macro-structural framework is

inadequate in this instance because of the dependent variable being examined, namely,

the implementation of specific economic reforms. A structural analysis would perhaps be

useful if one were performing a cross-country analysis of economic reforms. Structural

factors in particular countries could help to account for why economic reforms began at a

certain time. In fact, such a comparison with Brazil and other Latin America countries

could help elucidate the larger economic factors at play in this process.

Structural analyses still must account for the processes that shape outcomes. In

this sense, the study of institutions is apposite, for institutions shape behavior and









structure preferences. An institutional analysis also resists the temptation to reduce

outcomes to an individual level. Some scholars of Brazilian politics, such as Fleischer

(1998), view the outcomes of the political process as contingent on the idiosyncrasies of

key politicians and factors unique to certain times. While all political outcomes depend

to some degree on idiosyncratic factors, in the end such methodological individualism

does not yield any larger patterns of behavior. By allowing for an immediate level of

analysis between macro-structures and individual particulars, institutionalism is the most

common methodology for recent studies of economic reforms.

For Douglass North (1990:1), institutions, in their broadest sense, "are the rules of

the game in a society or, more formally, the humanly devised constraints that shape

human interaction. In consequence, they structure incentives in human exchange,

whether political, social, or economic." According to this definition an institution could

include cultural customs, codified laws, formal organizations, or any of a plethora of

other things. However, if the operational definition of an institution were kept so broad

as to signify cultural norms and other informal patterns of behavior, it would be

practically impossible to identify which institutions affect certain political outcomes. For

our purposes here, then, it is important to specify what constitutes apolitical institution.

According to Margaret Levi (1990: 405), institutions are the "formal arrangements for

aggregating individuals and regulating their behavior through the use of explicit rules and

decision processes enforced by an actor or set of actors formally recognized as possessing

such power." In this study, institutions will assume this narrower definition offered by

Levi. Hence, in my discussion of the Brazilian political system I will discuss the

constitution of 1988, the federal system, presidentialism, and the party and electoral









systems. These institutions structure political processes and frame politicians' behavior,

thus helping explain political outcomes. However, it is important to note that institutions

alone cannot account for outcomes. In fact, one of the great weaknesses of the literature

on Brazilian politics is that it often overemphasizes the explanatory power of institutions.


Political Science and Institutionalism

Modern political science arose during the late nineteenth and early twentieth

centuries. At that time attention was given primarily to the formal structures of

government, such as legislatures, constitutions, and the courts. In many ways, during its

early days the discipline was more descriptive than analytic; many studies compared the

political and legal systems of different countries. This "old" institutionalism, as it is

known, saw political institutions as dependent, rather than independent, variables (Just

1999: 18). Following World War II scholars began to question the utility of the

prevailing institutionalist paradigm. Rather than solely examining formal governmental

structures, increasing attention began to be paid to the micro-foundations of politics, that

is, individual behavior. This focus on political behavior subsequently became known as

the "behavioral revolution." Behavioralism eschewed the emphasis on the formal

institutions and instead sought to understand individuals' preferences, which

behavioralists believed could be surmised by examining observable behavior.

In addition to behavioralism, other schools of thought within political science,

notably Marxism and structural-functionalism, sought to discredit the old institutionalist

approach. These two frameworks of analysis were not mutually exclusive, nor did either

in and of itself represent a unified way of analyzing politics. Since its beginnings,

Marxism has undergone substantial refinement and evolution, yet despite the conceptual









differences that separate the various strands of Marxism they all criticized the traditional

focus on institutions as inappropriate. Since the state was the expression of the interests

of economic elites, political institutions were simply the means by which these elites (the

bourgeoisie) dominated the masses. Institutions did not have any explanatory power of

their own, but were epiphenomenal. Structural-functionalism, on the other hand, held

that "the systemic needs of the social system (for political stability or social stratification,

for example) tended to produce political institutions that were more-or-less automatically

functional in relation to those needs" (Rothstein 1996: 139).

Starting in the 1980s, political scientists began to reassess the import of political

institutions. From Latin America to Eastern Europe to Asia, politics across the globe

were in transition. Along with these rapid developments there also came a growing

dissatisfaction with the explanatory power of behavioralism and Marxism and other

structural theories. According to Rothstein (1996: 141), this renewed interest in

institutions took place across diverse subfields of political science, as well as in other

social sciences.2 This renewed focus on political institutions is commonly referred to as

the "new institutionalism." Unlike the old institutionalism, the new variant holds that

institutions themselves have explanatory power. One well-known collection of essays

urged for "bringing the state back in" (Evans et al. 1985), urging scholars to move

beyond seeing the state as the "black box" it had become during the height of

behavioralism and Marxism. "The renewed focus on political institutions has to some

degree turned upside down behavioral, group-theoretical or Marxist modes of explanation




2 For example, Selznick (1996) and Bardhan (1989) describe how institutionalism
influenced organizational analysis and economics, respectively.









stressing social and economic forces as explanations for political processes" (Rothstein

1996: 144).

Rather than a unified mode of thinking, new institutionalism in fact is a catch-all

term for a variety of approaches. As Immergut (1998: 5) explains,

The new institutionalists do not propose one generally accepted definition of an
institution, nor do they share a common research program or methodology. In
fact, three separate branches of scholarship-rational choice, organization theory,
and historical institutionalism-all lay claim to the label, seemingly without
adhering to an overarching theoretical framework.

Rational choice theories take their theoretical underpinnings from microeconomics.

While there are several variants of rational choice theory, most of them share the basic

assumption that individuals "have fixed preferences.. behave entirely instrumentally so

as to maximize the attainment of these preferences, and do so in a highly strategic manner

that presumes extensive calculation" (Hall and Taylor 1996: 944-5). According to

rational choice theories of politics, actors' preferences are shaped by the institutions

within which they interact with others. Preferences are exogenous, shaped primarily by

the incentives presented by certain institutional arrangements. Over time the institutions

that come to predominate do so because they maximize the utility of the greatest number

of actors involved. As discussed in Chapter 3, rational choice is often employed, either

explicitly or implicitly, in explaining how political institutions in Brazil provide certain

incentives for politicians to behave in ways that hinder the passage of reforms.

Rational choice has created enormous controversy within political science

because of its extensive use of economic models in the analysis of political behavior, and

many justifiable criticisms have been laid against it (Green and Shapiro 1996). One

criticism is that rational choice perspectives are static, and hence do not help explain









dynamic situations. What is important to draw from rational choice institutionalism for

this study, however, is that it can help to explain political behavior when the "rules of the

game" have been fairly well established. This criterion has been present in Brazil during

the period under examination, namely since the promulgation of the 1988 constitution,

which established many of the important institutions under consideration (Mainwaring

1999).

A second institutionalist approach, organization theory, is also referred to as

sociological institutionalism. Instead of viewing individuals as calculating utility

maximizers, sociological institutionalism argues that institutions must be examined

within their particular cultural context. Institutions are more than means towards

maximizing one's utility; they are primarily an expression of cultural and societal norms.

Because of this outlook, sociological institutionalism "seeks explanations for why

organizations take on specific sets of institutional forms, procedures or symbols" (Hall

and Taylor 1996: 947). In contrast to rational choice theories, sociological

institutionalism holds that actors' preferences are not the result of the incentives created

by institutions. While this approach is more sociological than political, and is concerned

primarily with how institutions form and change over time, it can help explicate

important differences in Brazilian institutions vis-a-vis other countries, thus lending

insights into why Brazilian politics is so seemingly fractious.

The third and final approach is historical institutionalism. Adherents of this

perspective "emphasize the role of institutional choices made early in the development of

policy areas, or even of political systems. The argument is that these initial choices

(structural as well as normative) will have a pervasive effect on subsequent policy









choices" (Peters 1996: 210). Initial institutional choices are important, argue historical

institutionalists, for even if attempts are later made to alter the institutional structure, it

will be difficult if not impossible to change it; this phenomenon is referred to as path

dependency. Historical institutionalism has enormous explanatory power when

examining contemporary Brazilian political history, especially in this case study. The

transition from military to civilian rule during the 1970s and 1980s profoundly influenced

the existing political order, from the constitution to the party system to the economic

order. As will be explained in the following chapters, many of the legacies of this

transition have been deleterious for the Brazilian economy (and society) and have proven

highly difficult to change. In this study I will not employ one single institutionalist

perspective, but instead will implicitly draw upon all three branches to help to explain the

ways in which institutions do-and do not-influence outcomes.


Institutions and the Study of Latin American Politics

Regional studies, which generally fall under the rubric of comparative politics,

often follow the larger trends within political science yet modify them to the particular

characteristics of the political systems being examined. The study of politics in Latin

America certainly conforms to this pattern. Just as in the United States, early political

studies of countries in the region focused principally upon formal government structures

(Ames 1999). Over time, however, it became clear that much more than formal

institutions were at play in the politics of the region. As a result, less attention was

devoted towards institutions, as scholars began to utilize other methods of analysis.

While in the United States behavioralism shifted attention away from institutions starting

in the 1950s, in Latin America a growing critical perspective turned attention away from









institutions towards issues such as domestic class struggles and global economic

inequalities.

The intensification of the Cold War and the Cuban Revolution had a profound

impact on the intellectual mood of Latin America. Marxism, which has deep intellectual

roots in the region, provided an appealing and effective framework with which to analyze

both domestic and international politics. Variants of Marxism, such as world-system and

dependency theories, also gained popularity starting in the 1960s. This was a time of

great social and political upheaval in Latin America. While leftist parties and

revolutionary movements made inroads during the 1950s and early 1960s, by the early

1970s most countries in the region had succumbed to some form of authoritarian rule.

Countries in most of Central America and the Caribbean had a tortuous political history,

with authoritarian rule being almost a constant throughout their history. In contrast, the

countries of the Southern Cone of South America were traditionally more democratic and

economically developed. As Brazil and then Chile, Argentina, and Uruguay all fell to

military rule, scholars began to examine the causes for the breakdown of democracy.

Some scholars attempted to explain this wave of authoritarianism by exploring the

structural economic relationship between the region and the developed world. To many

the events of this period served as proof of the effects of capitalist expansion into the

developing world. Guillermo O'Donnell (1973; 1978) saw a connection between

"bureaucratic-authoritarian" regimes in the Southern Cone and these countries' level of

"dependent capitalist" development. O'Donnell believed there to be a structural

relationship between these countries' position in the international economy and their

authoritarian political systems; there was a "mutual indispensability" between the









exigencies of international capital and the bureaucratic-authoritarian state. Peter Evans

(1979), in another seminal work of the period, spoke about the alliance of national, state,

and multinational capital in Brazil. According to Evans this "triple alliance," in which

multinational and state capital predominated over Brazilian private domestic capital,

helped to explain the country's recent authoritarian political history.

Works such as those by O'Donnell and Evans, which looked for structural

variables in the rise of Latin American authoritarianism, were more common than micro-

behavioral analyses. Most political scientists studying Latin American politics saw the

state as a dependent variable; Marxists saw it as the reflection of class relations, while

dependistas saw it as a reflection of a country's position in the global economy. Alfred

Stepan, in The State and Society: Peru in Comparative Perspective (1978a), viewed the

Peruvian state as an independent variable, arguing that the state was itself an autonomous

actor rather than the mere aggregation of the dominant class's interests. This study

helped spark a debate about the state and its autonomy vis-a-vis social forces like the

military and business.

Many scholars criticized structural studies as methodologically weak, for they

failed to establish empirical links between dependent development (another problematic

concept) and political authoritarianism. With the democratization of most of the region

during the 1980s, structural analyses came under further attack. Besides the

methodological weaknesses inherent in dependency theory and Marxism, these theories

were of little value in analyzing the substantive functions of government. In essence,

these schools of thought had little to say about institution and institutional design. What

could Marxism say about how to organize democratic political parties? An analysis of









the world system could reveal structural economic inequalities, yet it would be unable to

elucidate debates over the relative merits of presidentialism and parliamentarism,

electoral system design, and other such substantive issues.

According to Barry Ames (1999: 221), "Institutions are in. Not that Latin

Americanists ever ignored them completely, but in the old days of military

authoritarianism, power and international influences seemed more important than

governmental structures." With the transition to, and in most cases consolidation of,

democratic rule, a sizable number of publications focusing on Latin American political

institutions appeared during the past decade. Economic stabilization and restructuring

was a theme present in much literature, and many scholars, such as those in the edited

collection by Haggard and Kaufman (1992), stressed the role of institutions (both

domestic and international) in influencing the various paths to stabilization followed by

Latin American countries.

Yet Latin America was not the only region undergoing the dual processes of

economic restructuring and political liberalization. During the 1980s Eastern Europe was

also tackling many of the same issues as Latin America. This led many scholars to

undertake comparative studies between the two regions, often focusing on the

relationship between political and economic reforms. Przeworski (1991), in one of the

most cogent of these comparative studies, uses a game theoretic perspective to explain

potential courses for economic and political reforms in Latin America and Eastern

Europe, dedicating much attention to institutional factors in these transitions towards

market democracies. Another comparative study, edited by Lijphart and Waisman

(1996), examined the role of institutional design in new democracies, examining









specifically electoral systems, executive-legislative relations, and the type of economic

market pursued following democratization.

One of the most widely debated areas of institutional design has centered on

presidential systems. Juan Linz (1990) spearheaded an extensive debate on the merits of

the presidential system. Linz contends that presidential systems of government are less

attractive than parliamentary systems for various reasons. Since a president is elected

separate from the legislature, it is common for a chief executive not to command control

of the legislative body. When the executive and legislative bodies are highly polarized,

this can lead to a crisis of "dual democratic legitimacy" since both bodies were

democratically elected and therefore can claim a mandate from the public. Furthermore,

because presidents are elected for fixed terms, this creates rigidity during times of great

dissatisfaction with executive leadership. Whereas in parliamentary systems executives

can be removed with a simple vote of no confidence, it is generally much harder to

remove a sitting president. Linz and Arturo Valenzuela (1994) explored the "failures of

presidential democracy" in Latin America and inspired other studies of presidentialism in

Latin America, such as the edited volume by Mainwaring and Shugart (1997). Their

survey of presidentialism across the region shows how it has been less problematic in

some countries than in others. As will be explicated in greater detail in Chapter 3,

presidentialism in Brazil is complicated by the country's federal system and by the

weakness and multiplicity of parties.

In line with this general trend in the study of Latin American politics, there has

been abundant research on Brazilian political institutions. Being the largest country in

the region in both economic and demographic terms, Brazil always warranted the









attention of scholars. The transition towards democracy in the country heightened the

concerns about institutional design and how it could possibly affect the future course of

democracy in Brazil (Lamounier 1989). The deepening of the country's economic

troubles was another impetus for research (Kaufman 1990). While most major countries

in Latin American had successfully stabilized their economies by the beginning of the

1990s, Brazil was still suffering from hyperinflation. The persistent problem led scholars

to focus attention on how institutions complicated the process of economic reform (Sola

1994; Bresser Perreira 1996).

How did institutions affect economic stabilization and reform in Brazil? Before

answering that question, it will first be necessary to explain the processes and institutions

at work in Brazilian politics. This will require a brief discussion of the transition to

democratic rule during the 1980s, which resulted in the presidency of Jose Sarney in

1985 and the creation of a new constitution in 1988.3 This new document both

established new institutions and complicated existing institutional arrangements (such as

the skewed system of representation in the lower house of Congress), a fact that has

confounded recent efforts at economic reforms. For example, the federal system

mandated by the constitution forced the federal government to give a larger share of tax

revenues to local levels of government, most notably the states, without transferring a

commensurate number of fiscal responsibilities. As a result, the federal government has

been forced to continue to provide the same level of services without the necessary fiscal




3 See the special issue of the Journal of Interamerican Studies and World Affairs, "Brazil:
The Challenge of Constitutional Reform," (Winter 1998, Vol.40, No.4) edited by David
Fleischer, for an extensive discussion of the enduring problems created by the 1988
constitution.









resources. These transfers to the states and municipios (which are similar to counties in

the United States) have been a windfall, helping to grease the local political machines

that are the heart of Brazilian politics. Not only is the fiscal situation of the central

government damaged, but so too is its very authority vis-a-vis the states and municipios.

Furthermore, as the world economy has become increasingly globalized, this tense fiscal

relationship between Brasilia and the states has been a persistent concern of international

investors, who fear the consequences of recurring, unchecked fiscal deficits on the part of

local governments.

Beyond the Brazilian system of federalism, the party and electoral systems have

created "institutional rules of the game [that] have given politicians incentives to cultivate

a personalistic relationship with the electorate, to focus on local constituencies, and to

pay attention to state and local politics" (Mainwaring 1999: 10). These incentives

weaken the coherence and discipline of parties, making legislative efforts cumbersome

and highly costly in terms of time and pork. Chapter 3 will offer a thorough discussion of

the Brazilian political system, problems with the design of key institutions, and their

effects on reform efforts.

As stated above, in this study I will not rely upon one exclusive method of

institutional analysis. Instead, throughout I will draw upon all three approaches. My

conclusion is that in Brazil it has been possible, despite political-institutional obstacles,

to implement significant economic reforms. While many reforms have floundered during

the past few years, one must take account of successful reforms. This, I conclude, calls

for a reappraisal of institutionalist perspectives of Brazilian politics.









Structure of the Study

Following the theoretical framework of analysis laid forth here, Chapter 2

provides an overview of the economic development of Brazil and the role of the state in

the economy. With its history of Portuguese monarchical control, the colony of Brazil

developed a heritage that had a profound impact on economic development. Most

noticeable were the reliance on a few primary product exports and the consequences of

such dependence on growth. One of the poorest countries in Latin America at the

beginning of the twentieth century, Brazil experienced some of the highest growth rates

in the world from 1900 until the debt crisis of the 1980s (Rigolon and Giambiagi 1999).

During this period of rapid development, the state came to play a central role in the

economy. One of the clearest manifestations of the state's active role in development

came by means of state-owned enterprises, which proliferated following World War II.

In short, to understand modern Brazilian economic history, one must understand the vital

role the state played in development-and how and why that role has changed over time.

As this is a study of political institutions and their effects on the reform process, it

is necessary to examine Brazilian political institutions and processes in some depth. This

will be the focus of Chapter 3. Many aspects of the country's political system have

proven extremely deleterious to the country's fiscal situation and continue to hamper

efforts at reform of the state. According to one common interpretation, the Brazilian state

in the post-authoritarian period has been not been a powerful Leviathan, but rather a

growing source of patronage, favors, and rents for those with the access to such resources

(Roett 1997). The constitution of 1988 codified a federal system that forced Brasilia to

earmark revenues to states and local governments without a commensurate increase in

their responsibilities in areas of public spending. This has entailed a fiscal windfall for









state and local politicians, but has created enormous strain on the fiscal health of the

federal government. Beyond the fiscal problem emanating from the political system,

there is a myriad of features in the party system, the country's particular form of

presidentialism, and electoral system that create hurdles for reforms.

In Chapter 4 I will analyze how, in spite of Brazil's problematic institutional

arrangements, economic reforms have been successfully implemented. I will begin with

a discussion of the various reform measures put forward by the four presidents of the

New Republic. Particular attention will be given to current President Fernando Henrique

Cardoso, who has been the most vigorous reformer to date. I will also explore one of the

most important sets of measures passed by Cardoso at the beginning of his time in office,

known as the Big Five. I will conclude the chapter by discussing what factors help

account for economic reforms in Brazil. Specifically, I look at four factors: the timing of

reform; the type of reform proposed (first- versus second-generation reforms); the role of

ideas and ideology; and political and economic pressures. Collectively, these four factors

help to explain the success of economic reforms in Brazil. Finally, in the conclusion I

reiterate the main findings of this study and offer suggestions for future research.














CHAPTER 2
THE ROLE OF THE STATE IN THE BRAZILIAN ECONOMY

This chapter will explore the history of state involvement in the Brazilian

economy. Since the discovery of Brazil by the Portuguese in 1500 the state (or the

crown) has always played some role in the economy. Until the early twentieth century

that role was largely confined to supporting or promoting in some fashion primary

products for sale on the international market. During the early twentieth century Brazil

began to industrialize, and following the Second World War the state played a primary

role in this process. This chapter is arranged chronologically, tracing the changing

relationship between the state and the economy from colonial days until the demise of the

state-led model in the late twentieth century. (Table 2.1 provides a list of the key

political and economic events in Brazilian history since the early nineteenth century.) An

understanding of Brazilian economic history is necessary in order to comprehend the

enormous economic changes that have occurred during the past decade. It will also help

one understand why economic reforms are such a politically charged issue.

When examining the economic history of Brazil during the past century, three

salient themes emerge. The first is rapid growth. Brazil was one of the fastest growing

countries during the twentieth century, with only Japan and Finland growing more

rapidly between 1900 and 1970 (Rigolon and Giambiagi 1999: 6). Although the 1990s

was a period of stabilization and resumed growth in Brazil, overall economic growth















Table 2.1 Key Political and Economic Events in Brazilian History


Period


Year


Colony


Empire

First Republic
Second Republic

















Military Regime







New Republic


1808
1820
1822
1888
1889
1930
1937
1941
1942
1945
1950
1952
1953
1954
1955
1960


1961
1964
1968
1974
1975
1979
1982
1985
1988
1989
1990
1992
1994
1998
1999


Portuguese court moves to Rio de Janeiro, which becomes seat of Empire
Royal court returns to Portugal
Pedro I declares Brazilian independence
Abolition of slavery
Overthrow of monarchy, establishment of republic
Revolution of the tenentes brings Vargas to power
Vargas establishes the Estado Novo
Creation of the Companhia Siderurgica Nacional (Volta Redonda)
Creation of the Companhia Vale do Rio Doce (CVRD)
Vargas, under pressure from armed forces, steps down
Vargas elected president
Creation of the Banco Nacional de Desenvolvimento Econ6mico (BNDE)
Creation of Petrobras
Vargas commits suicide while still in office
Election of Kubitschek, Piano de Metas
Quadros elected
Construction of Brasilia completed
Quadros resigns; Goulart assumes presidency
Military coup against Goulart
Institutional Act 5 signals start of repressive period
Start of political liberalization (distensdo)
Second National Development Plan introduced, which intensifies ISI process
Second oil shock
Direct election of governors
Military leaves power, Sarney assumes presidency
New constitution adopted
Collor elected in first direct presidential election since 1960
National privatization program initiated by Collor
Collor impeached
Real Plan; Cardoso elected president
Cardoso re-elected
Devaluation of the real









during the past two decades was low by historical standards. Second, with economic

growth came a concomitant demographic shift, as Brazil was transformed from a largely

rural country into a highly urbanized one during the span of about two generations. A

third theme is the importance of the state in the process of economic development. Since

colonial days the state was a central actor in economic activity. Its role, however,

changed dramatically over the course of the twentieth century, notably during the rule of

Getulio Vargas (1930-45 and 1950-54), as the state began actively to promote

industrialization. The economic role of the state assumed various forms, including that of

regulator, financier, and producer (Baer et al, 1976).

During both democratic and authoritarian periods in Brazil, desenvolvimentismo,

or "developmentalism," was an important ideological force, one which called for the state

to champion industrialization and foster the growth of the economy (Lima 1999). SOEs

played a key role in this developmentalist strategy, as the state became involved in

sectors such as steel production, oil exploration and refining, mining, and electricity

production and distribution. However, due to the energy crises of the 1970s and

increased indebtedness during that period, the state found itself increasingly incapable of

making the investments necessary to sustain growth. This downturn in the fiscal position

of the state, coupled with changes in the international political economy-the rise of

neoclassical economics and its embodiment in Reaganism and Thatcherism, the decline

in financing from IFIs, the collapse of the Soviet bloc, the consequent discrediting of

socialism as a guiding force of development-set the stage for a reassessment of the

relationship between the state and the economy. This reexamination was particularly









marked in Latin America, where most states had played an active role in development

since the Second World War.

Although this neoliberal trend reached Brazil later than most countries in the

region, in Brazil its effects have been among the most pronounced. As Luiz Carlos

Bresser Perreira (1996) explains, the country was facing a "crisis of the state" and

therefore required an extensive reshaping of both the political and economic spheres of

life. The result has been a dramatic restructuring of the economy as well as a drive to

reform the political system. Privatization of state-owned enterprises embodies both of

these phenomena perhaps more so than any other reform. Alluding to Mexico, Judith

Teichman believes that "privatization programs have been the most important way in

which the state's role has been reduced and redefined" (1995: 3). The same could said of

Brazil.


Early History: Economic Cycles and External Vulnerability

At the beginning of the twentieth century Brazil was one of the poorest countries

in Latin America, itself one of the poorest regions in the world. Since its initial

colonization by the Portuguese crown during the sixteenth century, Brazil relied on the

export of a few primary products for its economic growth. The colony's original name,

Terra da Vera Cruz, or Land of the True Cross, reflected the religious sentiments that

played a role in the expansion of the Portuguese empire. However, even more important

than religious zeal was the quest for wealth. The Portuguese established a series of

feitorias, or trading outposts, on coastal areas from Africa to Indonesia. Brazil, in



1 According to Baer (1995: 3), the population living in urban areas was 30 percent in
1940, 56 percent in 1970, and 76 percent in 1991.









contrast to the crown's other possessions, did not possess any known mineral wealth nor

did it have a large domestic population with which trade could take place. Because of

this lack of natural resources, Brazil initially received scant attention from Lisbon. In

fact, Portuguese control of this vast territory initially was extremely precarious. During

most of the sixteenth century the Portuguese intermittently fought with the French for

control of key ports in the colony.

While the original name of the colony had a religious origin, it is telling that the

name that has endured to this day came from the colony's first major commodity export,

pau brasil, or brazilwood. Economics, not religion, would be the driving force in the

formation of the colony. Brazilwood became an important ingredient in clothes dye at

that time and grew in abundance in the lush Atlantic rain forest around Rio de Janeiro.

The exploitation of brazilwood contributed to the destruction of the Atlantic rain forest

and also marked the first commodity cycle in Brazilian history (Dean 1995).2

According to historian E. Bradford Bums, "From the beginning, extreme

economic fluctuation has been a major characteristic of the Brazilian economy and a

powerful influence on the course of Brazilian history" (1993: 63). This fluctuation was

directly related to the commodity export cycles that punctuated the entirety of Brazilian

colonial history and lasted until the beginning of industrialization. While the extraction

of brazilwood provided the impetus for initial exploration of the new colony, it was not

until the advent of sugar cultivation along the southeastern, and especially the

northeastern, coasts that modern-day Brazil began to be heavily colonized. The fertile



2 In this sense, one sees early roots to two of Brazil's recurring problems: a reliance on a
small number of primary product exports and a pattern of massive environmental
degradation.









lands of the littoral proved to be ideal for the cultivation of cane sugar. As demand for

the sweetener grew in Europe, the Brazilian sugar sector expanded exponentially. With

time competitors emerged to challenge Brazilian predominance of sugar production. As

plantations became operational throughout the Caribbean, Brazil's market share slipped,

setting the stage for an economic decline. While its is beyond the scope of this study to

trace the vagaries of all of Brazil's economic cycles, it is worth noting that what

happened with sugar would recur with future commodities-gold, cacao, cotton, and

especially coffee. The structural position of Brazil in the world economy and the tumult

it created was to have a profound effect on developments during the twentieth century.

In 1822 Brazil achieved its independence from Portugal. Unlike its Spanish-

speaking neighbors, Brazil was able to wrest control from its former rulers without

bloodshed. Furthermore, independence was achieved without territorial disintegration, a

far cry from developments in Spanish America. Another difference separating Brazil

from its neighbors was its monarchical form of government, as the country was ruled by

the house of Braganca (Pedro I and Pedro II) until the latter part of the nineteenth

century. During the imperial era, coffee came to dominate the economy and coffee

interests became increasingly influential in the nation's politics. Slavery, which had long

been an integral facet of the coffee economy, was abolished in 1888, which in part led to

the deposal of Pedro II in 1889. Following his deposal, the centralized monarchical

system was replaced with a highly decentralized republican system. The advent of the

republic also marked an important point in Brazilian economic history, as the country

became increasingly integrated into the world economy through the growth of the coffee

trade.









In the economic realm, the republican regime created following the overthrow of

the monarchy was most concerned with those areas related to export agriculture,

especially coffee. According to Steven Topik, during the early republican era the state

began to play an active role in the economy, namely through its involvement in railroads

and the creation of financial institutions. "Through [its] haphazard, almost involuntary,

procedure," claims Topik (1979: 326), "the First Republic's3 State set the precedents that

have greatly affected the nature of current interventions." While the Brazilian state did

become more involved in the economy during the end of the nineteenth and the beginning

of the twentieth centuries, Topik overstates its role. The First Republic was a classical,

liberal regime, in which the state rarely became actively involved in the economy. When

it did, it pursued activities, namely those related to public goods, which the market failed

to provide adequately. To say that the steps undertaken during the First Republic

"greatly affected the nature of current interventions" is to take the argument too far. It

was not until the rule of Getulio Vargas that such a claim could be substantiated.


Vargas and the Origins of State Capitalism

The growth of state involvement in the Brazilian economy became pronounced

during the era of Vargas, president from 1930 to 1945 and again during the early 1950s.

The government Vargas overthrew was based on an alliance between agrarian interests in

Sdo Paulo and Minas Gerais (cafe cor leite) and had intervened frequently to support the

price of coffee, Brazil's leading export during that period, on the international market.

The political system established in 1889 had changed little, even as Brazil itself was


3 The republic established in 1889 is commonly referred to as the Old Republic and First
Republic. The Vargas era, which began with the "Revolution of 1930," ushered in the
so-called Second Republic.









changing rapidly. With the support of the military and disgruntled political elites in Rio

Grande do Sul Minas Gerais, and the Northeast, Vargas was able to lead a nearly

bloodless coup.

The time of Vargas's rule was one of tremendous political and social change in

Brazil. As a result of the Great Depression the country was suffering from a decline in

world demand for its primary commodities, most notably coffee. At the same time that

the economy was adjusting to the new international environment, social forces were

changing the nature of Brazilian politics. During the 1920s and 1930s nationalism,

especially in the economic sense, was on the rise. So too were labor movements, which

to varying degrees were based on anarcho-sindicalism, Marxist, and fascism. These

various social forces helped shape the Vargas regime. On the other hand, Vargas worked

to contain these emerging social forces so as to lessen their risk to the political status quo.

In 1937 he created the Estado Novo, an authoritarian, fascist-inspired corporatist state.

During the Estado Novo the state began to play a growing role in economic development,

as some of the first SOEs were created. Under this regime

Economic, political, and ideological disorganization gradually placed economic
policy and planning in the hands of the government, strengthening its
interventionism, without threatening the power of the dominant groups. Thus it
presented a conservative pattern of modernization which was to characterize the
country's future development. (Becker and Egler 1992: 47)

This pattern of "conservative modernization" lent a unique face to economic

development in Brazil, for it preserved traditional power structures at the same time that

the state was spearheading economic development.

One of the effects of the Great Depression was that it revealed the precariousness

of countries that relied on external markets for their economic wellbeing. This situation









forced leaders across Latin America to reexamine their position in the world economy.

Most countries in the region had suffered from a deterioration in their terms of trade with

the developed countries of North America and Western Europe, as the prices of Latin

American primary products stagnated or declined relative to the prices of manufactured

goods from the developed world. Raul Prebisch of Argentina, one of the early leaders of

the United Nations Economic Commission for Latin America (CEPAL), was the most

prominent economist to investigate and criticize this phenomenon. Assailing the theory

of comparative advantage-the sacred tenet of classical economics-Prebisch posited a

theory of "unequal exchange." According to Prebisch's theory, the developed economies

benefited from international trade over time, while growth in the countries of the

periphery stagnated. Moreover, he claimed that this stagnation was linked to poorer

countries' structural relationship to the industrial countries (Love 1995). Since all Latin

American countries were still primarily agrarian at the time of the Depression, Prebisch

and his followers believed that it was necessary for the countries of the region to

industrialize in order to combat the erosion in their terms of trade and, by extension,

make themselves less vulnerable to external shocks.

In order to achieve this industrialization, Latin American leaders decided to raise

protectionist barriers in their economies, thus limiting the level of foreign goods and

capital that could enter. Countries began to produce domestically some goods that they

previously had imported from abroad, usually light consumer goods. This policy, which

became the dominant development model in Latin America following the Great

Depression, was known as import-substitution industrialization (ISI). While this was not

a monolithic policy, there was a great deal of convergence in the policies undertaken









across the larger Latin American republics. The level of state investment of the economy

grew rapidly, as development banks appeared in many countries with the goal of

providing capital for targeted sectors of the economy.4 Protective measures, such as tariff

and non-tariff barriers to trade, were enacted in order to protect nascent industries from

foreign competition. Another key aspect of ISI was the creation of SOEs.

Dominant economic thinking today assumes that the state plays a negative role in

the economy. By undertaking relatively inefficient activities, the state is believed to

"crowd out" more efficient private activity. However, during the era of state-led growth

in Brazil the state was viewed as a positive force in the economy. During the Vargas era,

when state economic planning began, it was the state elites who championed a growing

role for the state in the economy (Skidmore 1967: 46). Rather than "crowding out"

private economic activity, state planners believed that the state played a "crowding in"

role by providing cheap credit (often with negative real interest rates) and cheap inputs.

Moreover, "in those days the growth objective outweighed efficiency concerns" (Baer

1996: 366).

Baer (1994: 509) identifies three factors that influenced the growth of SOEs in

Latin America: "(1) the existence of externalities related to... industrialization objectives;

(2) externalities related to equity considerations; and (3) nationalism." One of the largest

obstacles to the advancement of ISI was the unwillingness of local private capital to

participate in key long-term investments, such as public utilities and communications

infrastructure, at levels the state desired. Investment in these sectors was essential if



4 In Brazil some of the most important development banks included the Banco Nacional
de Desenvolvimento Econ6mico (BNDE), created in 1952, and the Banco do Nordeste,
created in 1954.









economies were going to be able to industrialize rapidly. The risk involved in many of

these activities forced private companies to provide services to limited, mainly urban,

areas, which in turn stunted the development of rural areas. Likewise, these sectors had

traditionally been controlled by foreign interests, which helped to fuel populist and

nationalist sentiments in favor of state involvement. To remedy these externalities the

state took control of activities such as public utilities, nationalizing most of these from

the private sector.5 Suffering from a lack of private domestic capital, many Latin

American governments also increased their involvement in financing particular sectors of

the economy. Having this kind of control allowed governments to take "equity

considerations" into account by devoting resources to areas of the economy which had

been largely neglected in the past, be they in depressed regions (such as the Brazilian

Northeast) or in particular industries (such as steel production).

While the 1930s were a period of increased government involvement in economic

affairs, it was during the 1940s that the first of Brazil's major SOEs appeared. For

decades there had been discussion about how best to develop a steel industry in Brazil.

With the rise of Vargas and the consequent increased influence of the armed forces

within the highest circles of the government, the steel question assumed a new urgency.

To the military the development of the steel industry was more than an issue of economic

development; it was also a question of national security. In 1941, in exchange for a

security agreement from Brazil to aid the U.S. against Nazi Germany, the U.S.

government agreed to provide funding for the creation of the Companhia Siderurgica



5 For example, 64 percent of utilities in Brazil were controlled by the private sector in
1962. That figure dropped to 20 percent in 1977, and by 1982 the state controlled all
public utilities (Baer 1995: 258).









Nacional (CSN). Located on an old coffee plantation at Volta Redonda in the state of

Rio de Janeiro, the CSN was one of the earliest and most powerful symbols of the state's

championing role if development (Wirth 1970). The company was also one of the largest

SOEs in Brazil for decades to come.

Closely linked to the steel industry was the mining sector. In 1942 the

Companhia Vale do Rio Doce (CVRD) was created, largely in response to the demand

for iron ore for the nascent steel industry. In time it became the country's "largest

mineral exporting firm," and has been one of the most powerful of the state's public

enterprises (Baer 1995: 247). By 1985 it had become the seventh largest firm in the

country and had acquired or created a number of subsidiaries outside the mining sector,

including sectors such as consulting, shipping, pulp and forestry, marketing, and

fertilizers (Raw 1987). The CVRD was a "state conglomerate" whose development

demonstrated that state firms could be highly profitable and could compete according to

market forces despite its public ownership structure. In fact, it became the largest mining

firm in all Latin America. (It is ironic that it was during the military regime of 1964-85,

which seemed at least superficially committed to the market, that CVRD became

increasingly politicized. As a result, by the late 1970s the company had become highly

indebted and uncompetitive, like most other state enterprises at that time.)

During the 1950s more economic activities came under the purview of the state.

In 1952 one of the most important state institutions was created, the Banco Nacional de

Desenvolvimento Econ6mico (BNDE).6 The BNDE was created to serve as the

country's principal development bank for infrastructure, but over time this task expanded









to include a variety of financing responsibilities. Over time the BNDE became the

primary shareholder in many SOEs (and some originally private enterprises); in exchange

for providing a firm with financing, the development bank was given a percentage of

ownership of the firm. If some firms encountered financial difficulties, the bank would

provide assistance in exchange for greater control.

During the early 1950s there were growing public pressure in favor of creating a

state oil company. The motto in favor of the creation of a state oil company-"O petroleo

e nosso," or "The oil is ours"-underscored the growing sense of economic nationalism

in Brazil during that period. In 1953 Congress mandated the creation of Petrobras, which

was given monopoly powers in the most aspects of oil exploration and refining.

According to the historian John Wirth (1970: 217), "the Vargas era [1930-1954] saw a

marked increase in the authority and capacity of the national government. Brazilian

policy makers developed a future orientation and a confidence in national purpose that in

the broadest sense can be called developmental nationalism." In many ways the creation

of Petrobras marked the height of economic nationalism in Brazil, but also the beginning

of a decline in Vargas's influence in national political life. Threatened by the military,

who were about to depose him, Vargas committed suicide in 1954 while still in office.

The 1950s were also a time of increased price controls in Brazil, as the

government attempted to control inflation. These controls gave private firms little

incentive to participate in sectors such as utilities, and as a result the state became in

involved in these sectors by default. Starting at this time the government began to create




6 The BNDE was renamed the BNDES, or Banco de Desenvolvimento Econ6mico e
Social, during the 1980s.









SOEs in utilities and communications. The state electric conglomerate Eletrobras was

established in the early 1960s, as was the state telephone company Embratel.

During the late 1950s, Brazil underwent a period of rapid growth and

industrialization. Under the direction of President Juscelino Kubitschek, the country

experienced what Thomas Skidmore (1967) has called "years of confidence."

Kubitschek's Plano de Metas, or Targets Plan, laid out development goals for the country

that, the president argued, would spur "fifty years of development in five." The new

national capital, Brasilia, was being constructed and enormous investments were

undertaken by the state in a variety of industries. It also marked the beginning of the

country's automobile industry, today one of the largest in the world. While Brazil grew

rapidly during the Kubitschek era, the eventual result of this prodigious growth was a

worsening of the country's balance of payments and an acceleration of inflation. By the

early 1960s, the economy was slowing, inflation was worsening, and society was

becoming ever more polarized.


The Military Regime: From Boom to Bust

In 1964 a military coup overthrew the government of President Jodo Goulart,

whose actions had incensed the military and alarmed the middle class (Stepan 1978b).

The new military regime continued the past policies of state-led developmentalism, yet

under far different political conditions. In the past there was a pattern of state

coordination of private domestic capital, public domestic capital, and foreign capital.

This pattern was solidified into a "triple alliance" under the military regime, which

promoted the expansion of multinational and state capital, often to the detriment of

private domestic capitalists (Evans 1979). At the same time the government assisted in









the suppression of wages and the weakening of organized labor. As a result of the

military's deliberate development plans the economy began to grow at impressive rates.

This led to the so-called "Brazilian miracle" of the late 1960s and early 1970s. Between

1967 and 1973 the country averaged annual growth rates of 12.9 percent. This growth

was driven largely by a promotion of exports, which increased from $1.9 billion in 1968

to $6.2 billion in 1973 (Roett 1992: 164). Furthermore, this period of rapid growth was

also marked by an expansion SOEs and overall state involvement in economic activity

through a variety of mechanisms:

Government expenditures ... as a proportion of GDP grew from 17.1 percent in
1947 to 22.5 percent in 1973. State enterprises dominated in steel, mining, and
petrochemicals. They controlled over 80 percent of power-generating capacity
and most public utilities. It has been estimated that in 1974, for the 100 largest
firms (in value of assets), 74 percent of the combined assets belonged to state
enterprises, whereas for the 5,113 largest firms, 37 percent of assets belonged to
state enterprises. (Baer 1995: 79-80)

The impressive growth of the Brazilian economy during the "miracle" of the late 1960s

and early 1970s, then, owed much to the actions of the state.

However, this growth came at the cost increased foreign indebtedness. Rapid

industrialization required large amounts of physical and financial capital from abroad.

By 1973 Brazil was suffering from a balance-of-payments crisis due to the huge inflow of

imports, usually capital goods, that sustained industrial growth. Growth continued

throughout the 1970s, but at slower rates and by means of greater foreign indebtedness,

especially through commercial (not public) loans. By the beginning of the 1980s growth

rates had dipped below zero, inflation accelerated (due in great degree to increases in

energy costs) and Brazil slipped into recession with the rest of the world. As happened

throughout the entire developing world, the energy crises of the 1970s put Brazil's statist









developmental plans under great stress. During this time the state pursued a course of

heavy industry ISI (such as steel production). Moreover, in response to the country's

dependence on foreign sources of energy, several large-scale public-utilities projects

were initiated, including the construction of the world's largest hydroelectric dam at

Itaipu. The result of such initiatives was increased foreign indebtedness. It also

contributed to the increasing stagnation of the economy. In great contrast to other

bureaucratic-authoritarian regimes in South America, the military government of Brazil

greatly expanded the reach of the state in the country's economy. In 1974, 39 percent of

net assets and 16 percent of sales in the Brazilian economy belonged to state firms. By

1985, those figures had risen to 48 percent and 26 percent, respectively (Baer 1995: 257).

The energy crises of the 1970s strained Latin American governments' abilities to

continue at the commanding heights of their economies. Increased energy costs caused an

increase in production costs for SOEs, forcing central governments to borrow yet more

money from the international financial community. There was no shortage of money to

be sought by the Latin American countries. The influx of funds from OPEC nations into

the banks of Europe and North American induced these banks aggressively to pursue

borrowers. This process of using funds from the oil-producing countries in loans to

developing countries came to be known as "petrodollar recycling." Latin America was

the preferred destination for these funds, satisfying the needs of cash-strapped central

governments in the region, as well as those of the international financial community.

Furthermore, the perception of riskless soveriegn debt provided a strong incentive to

lenders to continue making loans Latin American governments.









Coinciding with the energy crises of the 1970s was en explosion in both the

number of SOEs in their level of foreign indebtedness. During the 1970s alone, 251 SOEs

were created in Brazil (Bresser Pereira 1998: 65). Furthermore, the importance of SOEs in




Table 2.2 Number of Federal SOEs in Brazil, 1900-1981
Number of federal SOEs
Period by date of creation
1900-1930 25
1930-1939 8
1940-1949 33
1950-1959 55
1960-1969 180
1970-1979 259
Total (as of January 1981) 560
Source: Lamounier and Bacha (1994: 148)


the overall economy had grown greatly during the military regime. In 1962 12 of the

largest 30 Brazilian firms were SOEs; by 1979, 28 of the largest Brazilian firms were

state-owned (Bulmer-Thomas 1994: 356). Long protected from foreign competition,

most SOEs had become highly inefficient. Since they were under governmental control,

most became instruments of political patronage and corruption. As a result, SOEs

accounted for a large portion of government debt during the period. Bela Belassa and

Gerardo Bueno write that in Latin America

The deficit of state enterprises in the mid-1970s accounted for one-quarter of the
public sector deficit but rose to about one-half in 1980-1982, while the overall
public-sector deficit doubled from about 4 percent to 8 percent of GNP. Partly
because of price controls on basic products, state enterprises thus accounted for
three-quarters of the deterioration in public-sector finances. (1986: 139)

Thus, there were already signs during the mid-1970s that SOEs were becoming a major

drain on state resources in most countries in Latin America. However, it was not until









after the energy crisis of 1979 that the fiscal stance of most states such as Mexico and

Brazil became critical, largely because of their overextended and heavily indebted state

enterprise sector.

The expansion of the state in the economy during the 1970s had serious

implications. The Second National Development Plan (1975-1979), under the leadership

of President Ernesto Geisel, supported increased investments in large-scale projects.

Since it was impossible for Brazil's exports to finance these projects, the country looked

abroad for funding. The Eurodollar market, which was awash in money following the oil

crisis of 1973, provided a steady and reliable source of foreign finance. With the second

oil shock of 1979, Brazil found itself increasingly vulnerable as servicing of the foreign

debt amounted to 67 percent of export earnings (Roett 1992: 167). By 1980, it was clear

that Brazil was going to face enormous difficulty in servicing its foreign debt in the near

future. The situation was exacerbated by international conditions. By the early 1980s,

the world economy was suffering its worst downturn since the Great Depression.

Economic hard times in the developed world led to a decline in demand for exports in

developing countries. Complicating matters even further was the monetary policy

adopted by the U.S. Federal Reserve at that time, which increased interest rates. The

effect of this policy was an increase in the debt service requirements of developing

countries.

The combination of global recession and increased short-term debt burdens led to

a financial crisis across Latin America. The critical moment came in 1982 when the

Mexican government announced that it was unable to service its debt payments, causing a

panic throughout the international financial community. Since Mexico had one of the









largest economies in Latin America, and supposedly one of the healthiest, what happened

there was seen as a portent for the rest of the region. Soon after the Mexican default,

investors began withdrawing billions of dollars from all countries in Latin America. The

combination of economic contraction, capital flight, and resultant decreased living

standards turned the 1980s into the "lost decade" for the region. Brazil was forced to

enter into tedious and contentious negotiations with foreign creditors. Resolution of the

debt crisis was complicated by political developments within Brazil, as the country was

in the process of political opening (abertura) at that time. In 1985 the military stepped

down from power, leaving the first civilian government in 21 years the formidable task of

resolving the country's pressing economic problems.


Democratization and Demise of the State-Led Model

The new president, Jose Sarney, was confronted with economic pressures from

various sources. First, Brazil was yet to come to a final agreement with its foreign

creditors. Second, the advent of civilian rule brought with it long-repressed popular

grievances to address social issues, which had been ignored by the military government.

While the international financial community was pressing for an orthodox plan of fiscal

discipline, and many Brazilian economists supported such a plan as a means to control

inflation, the Sarney government found itself beholden to popular pressures. Sarney

introduced the Cruzado Plan in 1986, a heterodoxx" set of measures aimed at controlling

inflation without exacting austerity on the population. Although the plan was initially

successful in slowing inflation, after less than a year inflation was on the rise again and

Brazil found itself unable to meet its debt obligations. In early 1987 the government

declared a moratorium on its foreign debt payments. Economic strategy remained









inconsistent, and the following years witnessed a string of unsuccessful heterodox

stabilization plans.

Thus Brazil entered the 1990s with a growing sense of crisis. This mood allowed

for the election of Fernando Collor de Mello in 1990, an anti-establishment candidate and

political unknown from the northeastern state of Alagoas. As will be discussed in more

detail in the following chapters, Collor initiated the first significant set of economic

reforms, including the beginning of the national privatization plan and the lowering of

restrictions on trade. Yet Collor was impeached after serving less than two years in

office, and Brazil drifted further. Collor's successor, Itamar Franco, proved to be a

highly ineffective leader. During his tenure inflation reached record levels, exceeding

two-thousand percent in 1994. It was not until the introduction of the Real Plan in 1994,

under the leadership of Finance Minister Fernando Henrique Cardoso, that stabilization

was finally achieved. Cardoso's popularity was so high from his role in implementing

the Real Plan that he was able to win a first-round victory in the presidential election on

1994. During his tenure as president Brazil has undergone its most sweeping economic

reforms.

Many observers point to Brazil's political system as the cause of the country's

poor economic performance since the return to democracy. On closer inspection, one

sees that growth remained rather high at the beginning of the civilian regime, yet entered

a period of stagnation in 1988 (see figure 2.3). Brazil's elaborate system of indexation

allowed for growth in the face of inflation, yet confidence was eroded during the late

1980s and early 1990s as inflation surpassed four-digit levels. Critics cited numerous

institutional arrangements for complicating efforts to implement much-needed economic














10






59

44-

a)
$ I II I ,

S2




.-,
C
a)
B 0*
1li :1 1 il : i : 1 :: i~ : i : 1: :-- 1 :1: :i :li~ : 1:1 i :li : : 1 ~~:1 : 11- 1 1: 1 1 i:i : i 1 ~:1 : 1 i:i : i 1 ~:1 : 1 i:i : i: 1 :1 : 1 i :i:i : 1II : 1 :1 :1 _i-i-i





-4
-4 .44

-R


Figure 2.1 Brazilian GDP Growth rates, 1981-2000




reforms. Instead of promoting competition and modernization, the 1988 constitution


enshrined a number of nationalist economic tenets. Furthermore, it created a problematic


system of revenue-sharing between the federal government, on one hand, and state and


local levels of government on the other. By forcing Brasilia to transfer resources to lower


levels of government without changing the federal government's spending obligations,


the constitution created a problematic system which led to a structural fiscal imbalance.


The constitution, as well as other features of the Brazilian political system, led scholars to






42


focus on the institutional obstacles to economic reforms. It is to these institutions that I

now turn my attention.














CHAPTER 3
BRAZILIAN POLITICAL INSTITUTIONS

In Chapter 1 I examined the evolution of the discipline of political science during

the past two decades, emphasizing the rise of the new institutionalism. While it is not a

monolithic school of thought, the new institutionalism has as a foundation its emphasis

on the role of institutions in influencing political outcomes. Rational choice proponents

hold that actors' preferences are themselves shaped by the incentives presented by

institutions. In contrast, historical institutionalists generally examine the ways in which

early decisions regarding institutional design have (often unintended) consequences in the

future. According to historical institutionalists, once certain institutions come into being

they display a "stickiness" which makes institutional change difficult. Douglass North

maintains that initial institutional configurations lead to path dependence, which he

defines as "network externalities that limited...actors' choices and prevented them from

radically altering the institutional framework" (1991: 115). Finally, sociological

institutionalists contend that "persistent institutional patterns underlying the frequently

changing formal rules deeply shape individual actors and their interests" (Weyland 1996:

200).

The previous chapter ended with a brief discussion of the link between Brazil's

difficulties with stabilization and structural economic reforms during the 1980s and the

country's political system. In the present chapter I will move into an examination of the

main political institutions in contemporary Brazil. Chief among these institutions is the

Brazilian variant of presidentialism, the system of federalism, party and electoral









systems, and the national congress. All of these formal political institutions are closely

linked, making it difficult if not impossible to separate their effects on the overall

functioning of the political system. For example, the electoral system for the Camara de

Deputados (the lower house of Congress) is based on an open-list proportional

representation system. This system, which presents politicians with incentives to run

personalistic campaigns against members of the same party and to utilize a party label

opportunistically, in turn has had a direct effect on the party system by fostering

fragmentation and indiscipline (Ames 1995a). Furthermore, to understand fully these

institutions and their origins, one must also examine the constitution of 1988, which

established the institutional framework of the New Republic (Souza 1997; Reich 1998).

There is much evidence to support the thesis that the institutional structure of the

Brazilian political system encumbered the implementation of economic reforms

following the advent of the New Republic in 1985. However, I will argue that this focus

on Brazil's apparently problematic institutional arrangements has caused scholars to pay

insufficient attention to the country's impressive track record with economic reforms.

Successes on the reform front are usually examined anecdotally and seen as cases in

which executive resolve has been able to "overcome" the obstacles to reform presented

by the political system. What is needed, I will argue in Chapter 4, is a reassessment of

Brazilian political institutions and how they influence reforms.

In Chapter 2 I briefly detailed the historical role of the state in the Brazilian

economy. Starting in the 1950s the country embarked on a course of state-led

industrialization, passing from a largely agricultural country to the largest industrial

country in the developing world in the span of two generations. According to most









scholars, including Trebat (1983), Shapiro (1994) and Baer (1995), such rapid

development would have been impossible without the active intervention of the Brazilian

state. State-owned enterprises (SOEs) played an indispensable role in this process. Due

to this particular relationship between the state and the economy, some scholars described

the Brazilian economy as a case of "state capitalism," in which market forces

predominated, but the state played a larger role in capital accumulation and

industrialization than in Western Europe and North America (Baer et al 1976; Hagopian

1986).

One could utilize any of the three main institutionalist approaches when studying

the vagaries of recent economic reforms in Brazil. An historical institutionalist approach

would highlight the institutional framework created during the course of Brazil's

development since the beginning of state-led industrialization. One could argue that the

Brazilian state's developmentalist efforts during the latter half of the twentieth century

created an institutional matrix closely linked to the economy, through myriad direct and

indirect political and bureaucratic channels. Such channels included fiscal policy, the

central bank, federal and state development banks, SOEs, price controls, and corporatist

labor relations (Baer, et al 1973). This complex web between the political and economic

systems consequently made it very difficult to reform the state following the crisis of the

1980s. According to Frances Hagopian (1986: 2), "by concentrating most economic

resources within the public sphere, [state capitalism] enhances political clientelism,

which privileges political elites who control the distributional arm of the State." Hence,

the original political-economic institutional structure, which provided key politicians with

important patronage resources, would display a great degree of rigidity (or "stickiness")









in the face of efforts to reform it. The entire economic system proved highly resistant to

change precisely because it had long been highly politicized.

Rational choice institutionalism can also be useful in explaining the difficulties

encountered in passing reforms in Brazil. Barry Ames (1994; 1995a; 1995b), for

example, has studied the effects of the Brazilian electoral system on the strategies

employed by candidates for the lower house of Congress, the Camara de Deputados. He

concluded that the system, which is based on open-list proportional representation (see

more below), gives politicians incentives to concentrate on small, targeted constituencies

during their campaigns for elections. Once in office, politicians behave in such a manner

as to funnel as much pork as possible to these constituencies, which is fundamentally at

odds at efforts to address the state's fiscal imbalances. In his study of the party system

Mainwaring (1999) employs a rational choice methodology. He defends the use of

rational choice theory for this case study, saying:

Brazilian catch-all parties usually behave like self-interested actors whose
behavior is shaped by formal rules of the political game. Such rules help explain
the individualism of politicians, the decentralization of parties, the weakness of
national parties, the lack of party discipline, and the limited allegiance of
politicians to their parties. (1999: 10)

In the discussion of political institutions that follows, I will cite literature from all three

institutionalist schools of thought rather than utilizing one sole perspective. After first

discussing the general links between political institutions and economic reforms in Brazil,

I then analyze particular political institutions: the 1988 constitution, federalism,

presidentialism, the party system, the electoral system, and the Congress. In each section

I will describe the function of these institutions and then will explain how they hinder the

implementation of economic reforms.









Political Institutions and Economic Reforms

Brazil possesses one of the most problematic political systems in Latin America,

and there is widespread acceptance among scholars that the Brazilian political system

contains several institutional features that have impeded the passage of much-needed

economic reforms. By mandating a transfer of revenues to state and local governments

without a commensurate decrease in the federal government's obligations, the 1988

constitution lay at the heart of the country's persistent fiscal problem. Correcting this

imbalance, as well as a host of other structural and political reforms, has proven to be

difficult. According to Figueiredo and Limongi (1995), politics in post-authoritarian

Brazil has been marked by the existence of two separate competing agendas. On the one

hand, presidents have generally been interested in economic reforms, while on the other

hand the legislature has been more concerned with social issues. These distinct policy

orientations have often pitted the executive against the legislative, making reforms time-

consuming. Once implemented, many reforms barely resemble the original legislation

submitted and are not effective at attacking the problems they are meant to address.

Tim Power (2000) has described politics in the New Republic as a "permanent

constitutional convention." Since the transition from authoritarianism and the start of the

New Republic, the very institutional framework of the political system has been called

into question. The formation of the National Constituent Assembly to draft the new

constitution was itself a highly politicized process, and there has been widespread

dissatisfaction with the final document from a number of groups since its promulgation in

1988. As a result, Power explains, Brazil witnessed a "permanent constitutional

convention" as legislators and the public mulled over the desirability of tinkering with the

constitution. One of the most important events in this process was the plebiscite in 1993









to decide whether to maintain the existing presidential system, or to switch to either a

parliamentary or a monarchical system. Despite the discontent with the presidential

system and its supposed deleterious effects, which I will outline below, 55 percent of

voters opted for the existing system.

The 1988 Constitution and the Reform Process

One of the pressing issues surrounding the transition to democracy during the

early 1980s concerned how the post-authoritarian political order would be structured, as

well as by what means this structure would be established. The legislature elected in

1986 was to serve as a National Constituent Assembly (Assemblkia Nacional

Constituinte, ANC), which convened in February 1987. The Assembly included

representatives (deputados) elected in 1986 and senators elected in 1982; together in a

unicameral body they served as the ANC. "The ANC shunned any major political

reforms," instead focusing principally on social and economic issues (Fleischer 1998:

120). In effect, the document produced by the ANC "constitutionalized" numerous

provisions that generally are the preserve of normal legislation. The new constitution

placed stiff restrictions on foreign investment, sanctioned state monopolies in "sensitive"

sectors of the economy, and granted public sector employees job security and an array of

benefits. Another important provision of the constitution established a new revenue-

sharing arrangement between the federal government and the states and municipalities.

The new arrangement forced the federal government to transfer significant revenues to

local governments without a commensurate decrease in the federal government's

spending obligations. According to Roett (1997: 25), "What it actually did was to open

the floodgates for local corruption and mismanagement, as local and state authorities









enjoyed a financial bonanza without any programmatic responsibilities in critical areas of

social investment."

To understand the uphill battle fought by the Cardoso administration (as well as

his predecessors), one must understand the consequences of the 1988 constitution.

Because many of the economic changes Cardoso has tried to enact are in sectors in which

the state has constitutionally mandated monopoly power, the constitution itself must be

revised to reform these areas. In order to amend the constitution, both houses of

Congress must approve the amendment two times by a majority of sixty percent or

greater. In 1995, soon after taking office, Cardoso was able to pass several key

constitutional amendments. Known as the "Big Five," these amendments ended several

state monopolies (mining, coastal shipping, telecommunications, natural gas distribution)

and eliminated the distinctions on domestic and foreign capital. This was one of the

greatest achievements of the Cardoso presidency, for it cleared the way for the

continuation of the stalled national privatization program and helped to attract record

levels of foreign investment, both of which were essential in alleviating balance-of-

payments problems and modernizing the economy. The next chapter will examine in

greater detail how Cardoso was able to secure these amendments, with the hopes of

shedding light on how, in spite of political-institutional "barriers," widespread economic

reforms have been possible in Brazil.

Federalism

Like the United States, Brazil is a federal republic. The federation is composed of

twenty-seven states and a Federal District, Brasilia. Although federalism has a long

history in Brazil, the present form of federalism is the result of the 1988 constitution.

Due to the new constitution, "the executive has lost power to the legislative and judicial









branches, the union to the states and municipalities, and the state to society, private

enterprise, and market forces" (Selcher 1998: 25). Throughout Brazilian colonial and

imperial history, there was a balance between an ostensibly powerful central authority

and a de facto diffusion of power. In many ways, this decentralization of power assisted

in maintain the country's territorial unity following independence.

Modem federalism in Brazil has its roots in the early republican era of the late

nineteenth century. Following the deposal of the Braganca monarchy in 1888, those

actors involved in the republic movement worked to create a federal system that granted

individual states a high degree of autonomy vis-a-vis the central government. The

constitution of 1891 reflected the interests of the most powerful economic elites, which

were also those linked to the anti-monarchist movement in 1888. This document granted

states extensive powers. As a result, the most powerful states-Sao Paulo, Minas Gerais,

and Rio Grande do Sul-were able to limit the power of the central government. In

practice they did so by maintaining sizable state militias (often much larger than the

federal forces stationed within these states), taxing exports, and establishing their own

international diplomatic missions.1

One important implication of the early history of modem Brazilian federalism is

that it created a contentious and at times contradictory division of powers. As Alfred

Stepan (2000) explains, these initial arrangements have displayed a high degree of

"stickiness," which has made reforming the more problematic aspects of Brazilian

federalism very difficult. This is not to say that there has not been significant change in




1 This was important especially for the state of Sao Paulo, which aggressively promoted
immigration during the coffee boom of the late 1800s.









the structure of the Brazilian political system since the advent of the republic. In fact one

sees both centralizing and decentralizing trends during the past century. The era of

Getulio Vargas (1930-1945) and the military regime (1964-1985) displayed a tendency to

centralize power, while periods of democratic rule (1945-1964, 1985-present) have

shown a movement towards political decentralization. These shifting trends are the result

of a dialectic between two separate forces within Brazilian political culture: one based on

local patron-clientelism, the other based on the belief in the efficacy of greater federal

authority.

While many today feel that the devolution of political power away from the center

is beneficial for democracy, Stepan (2000) persuasively argues that in the New Republic

decentralized federalism has been injurious to democracy. The Brazilian system of

federalism is "demos-constraining," according to Stepan, for it has distorted to a higher

degree than any other federal system in the world the basic democratic idea of "one

person, one vote." The Senate, based on the principal of serving as a voice for less

populated states, is by its nature "demos-constraining." In Brazil, however, the Senate

possesses significant powers compared to upper chambers in other federations. With

three seats granted to each state, the Senate further skews the representation of the

national population. Furthermore, the creation of several lightly populated states in the

North and Center-West of the country during the past two decades has greatly enhanced

the power of conservative interests in the Senate, with congressmen from these areas

often opposed to the executive's reform measures. The Cdmara de Deputados, which in

theory is supposed to approximate more closely the actual population distribution within

a federation, greatly over-represents rural states and underrepresents more populous ones.









While similar arrangements can be found during the whole of the republican era

(Soares forthcoming: chap 12), the 1988 constitution mandated the most skewed system

to date. It set a minimum of eight deputados and a maximum of seventy per state. Figure

3.1 illustrates the degree of disproportionality created as a result of this arrangement in the

Cdmara de Deputados. By calculating the percentage of seats awarded to separate

geographical regions and contrasting it with each region's share of the total national

population, Lamounier and Bacha (1994: 175) revealed that in 1991 the Southeast, the

most economically advanced region of the country, was under-represented by 51 seats (of

a total of 503). Stepan (2000: 150) further elaborates this dilemma:

The one-person, one-vote rule would have given the states of Acre, Amapa, and
Roraima 1 federal deputy each in 1994 and the large state of Sao Paulo 114,
whereas the Brazilian system actually produced 8 deputies for each of the three
sparsely populated states while the citizens of the huge state of Sao Paulo were
represented by only 70.

This overrepresentation of rural states has largely benefited conservative political

interests, many of which have been consistently hostile to reform efforts during the New

Republic.

Aside from the impact on the composition of the national congress, Brazil's

system of federalism has had a deep impact on the relationship between the federal

government, the states, and the municipios. During the transition to democracy state

governors played an important role in contesting the powers of the military regime. In

1982 the military regime, well along in its plan to liberalize political activity, allowed for

the direct elections of governors. This in turn granted the governors a great amount of

leverage during the transition to democracy. Since the military regime, which centralized












Table 3.1 Disproportionality between Regional Shares of Federal Chamber Seats and
Population, 1986-1991
Chamber seats
1986 1991
Number of Number of % of Ideal
Region seats % seats % population number Differenceb
North 47 9.65 65 12.9 6.55 33 32
Northeast 149 30.60 151 30.0 28.50 143 8
Southeast 169 34.70 169 33.6 43.63 220 -51
South 82 16.84 77 15.3 15.15 76 1
Center-West 40 8.21 41 8.2 6.17 31 10
Total 487 100.00 503 100.0 100.00 503
a: Number of deputies each region would have if regional shares were exactly proportional
to their respective populations
b: Difference between the ideal and the number in fact prevailing during the 1991-1994 legislature
Source: Lamounier and Bacha (1994: 175)


power within the federal government, had curtailed the traditional powers of the states,

the governors (and their state political machines) were eager to enhance their powers

following the end of the military regime (Abrucio 1994).

With their power bolstered following the elections of 1982, the governors became

influential actors during the constitutional conventional of 1987-88. Partly as a result of

their influence, the federal system created by the new constitution gave the states greater

autonomy vis-a-vis Brasilia, notably in fiscal terms. The new constitution forced the

federal government to redistribute a greater share of revenues to the states and municipios

than had been the case previously. However, a transfer of responsibilities did not

accompany this transfer of resources. The federal government therefore has been forced

to provide the same level of services, yet without the necessary resources. This has been

a boon for the states and municipios, with enhanced tax revenues from Brasilia aiding in

local patronage politics. Two analysts of the current system go so far as to say that









"municipal governments in Brazil... should be the envy of all governments in the

developing (or even developed) world" (Bomfim and Shah 1994: 536).

By most accounts, Brazil's particular system of federalism lay at the heart of

many of the country's current problems (Stepan 2000; Mainwaring 1997; Abrucio 1994;

Roett 1997). The federal system is the root cause of the country's chronic fiscal

problems, and as such it will be impossible to improve the federal government's fiscal

position without altering the revenue-sharing arrangement with the states and municipios.

Furthermore, until recently states had great autonomy in contracting their own sovereign

debt. The federal government has gained control of profligate state spending in recent

years, but only by bailing out bankrupt state banks at a cost of $15 billions (Samuels

2000: 17). The Fiscal Responsibility law, passed in April 2000, marked a turning point in

the fight for fiscal discipline. The law places stiff limits on the level of discretionary

spending allowed by states and placed penalties on those states that do not maintain

sound fiscal policies.

Presidentialism

With the wave of democratization that occurred during the 1970s and 1980s, there

was a growth of research examining questions of institutional design in nascent

democracies. One of the central arguments in this debate centered on the efficacy of

presidential systems of government. Juan Linz (1990; 1994) was at the fore of this

debate. He argues that presidentialism contains several inherent flaws that hinder

effective governance in nascent democracies. The first problem associated with

presidentialism is what Linz calls "dual democratic legitimacy." Having come to power

through the ballot, both the president and the legislature possess democratic legitimacy.

Yet when there is conflict between these two branches of government, does one possess









more legitimacy than the other? "In the absence of any logical principle to define who

really has democratic legitimacy, it is tempting to use ideological formulations to

legitimize the presidential component of the system and delegitimize those opposing him,

transforming what is an institutional conflict into serious social and political conflicts"

(Linz 1994: 8). Many times in the past such executive-legislative conflicts have

precipitated coups d'etat in Latin America.

A second problematic aspect of presidentialism is its rigidity. Presidents are

elected to serve fixed periods of time. While constitutional provisions providing for the

removal of incompetent or corrupt presidents usually exist, they are rarely invoked; when

they are invoked they often signal political crises and virtually paralyze the political

system. Some observers justly maintain that democracy ultimately prevailed during the

impeachment of President Collor in 1992, yet during the impeachment process legislative

activity came to a virtual standstill. Under parliamentary systems, such crises can be

solved within a relatively short time frame and without as much interruption of the

legislative process. More importantly, under parliamentarism such crises of government

are less likely to lead to the crises of regime that have plagued Latin America. For

example, Mainwaring (1997) believes that the coup of 1964 was caused in large degree

by the problematic structure of Brazilian presidentialism.

This debate over the merits of presidentialism has been the subject of intense

public discussion in Brazil more so than in any other country in Latin America. In 1993

there was a plebiscite in which voters had to vote to maintain the presidential system, or

adopt either a parliamentary or monarchical system. While presidentialism won by a

wide majority, the plebiscite vividly demonstrated the dissatisfaction with the country's









institutional arrangement. This dissatisfaction arose from the protracted stalemate

between the executive and legislature, especially concerning economic issues.

Brazil possesses one of the most powerful presidencies in the world. The 1988

constitution granted presidents sweeping reactive and proactive powers. Presidents may

issue both a complete and partial (line-item) vetoes of congressional legislation. Even

more significant than these reactive powers are the proactive powers. Brazilian

presidents have the right to initiate legislation in several areas, from the size of the armed

forces to areas of public administration and ministries. The president also wields

significant powers over the federal budget. Perhaps the most important resource

available to presidents by the constitution is the medidaprovis6ria, or provisional

measure. These measures immediately become law and are effective for thirty days.

During these thirty days Congress has the power to reject or pass the measure; if no

action is taken during that time, the law expires.

The medidaprovis6ria has become the most important tool available to the

executive branch. Although the constitution states that they are only supposed to be

issued in cases of "relevance and urgency," in reality presidents have adopted them as a

means of controlling the legislative agenda.

Between October 5, 1988, when the new constitution went into effect, and late
May 1995, the four presidents issued 1,004 provisional measures Of [these],
640 had previously been issued. Congress had approved 274 provisional
measures and had rejected only 18. (Mainwaring 1997: 62-63)

Furthermore, the frequency with which presidents have utilized the provisional measures

has increased with each successive administration.

This concentration of power within the executive branch serves to counteract the

factors that work against the executive in Brazil: federalism and multipartism. As









discussed above, the Brazilian system of federalism grants significant powers to the

states, and the present system of revenue sharing weakens the fiscal position of the

central government. The fragmentation and indiscipline of the party system is in some

ways a concomitant of federalism, as congressional politicians seek to funnel patronage

to their constituencies at home. Sweeping presidential powers may not bode well for the

quality of democracy (O'Donnell 1994), but they have been an essential means by which

Brazilian presidents have been able to implement economic reforms in the face of

congressional resistance. In fact, both Collor's economic program in 1990 and the Real

Plan in 1994 were enacted principally through the use of provisional measures.

The Party System

Political parties are indispensable elements in the functioning of democracy.

They "both shape the political arena and are shaped by the political system as a whole;

they shape the nature of interests in society and respond to demands and interests of

societal groups" (Mainwaring and Scully 1995b: 459). During the period of military

authoritarianism, political parties did not receive much attention from scholars, who often

viewed them as peripheral to the political process in most Latin American countries.

With the demise of authoritarianism, there was a growing awareness of the centrality of

parties to politics. While in the past Marxist or dependency analysis discounted their

role, it is now clear that, in short, parties matter. They are no longer seen simply as the

by-products of class forces, but are viewed as important variables in their own right.

Compared to other countries in Latin America, Brazil is striking for its level of party

underdevelopment. According to Mainwaring and Scully (1995a: 7), the volatility of the

Brazilian party system is comparable to that of Peru, Ecuador and Bolivia, all countries of

with substantially lower levels of economic development.









Historically Brazil has never possessed highly institutionalized parties. During

the entire Old Republic (1889-1945), there were no parties of national scope. Getulio

Vargas, who served as president from 1930 until 1945 (and later from 1950 to 1954) and

was instrumental in shaping the modern political system, was hostile towards parties. In

fact, during his first presidency he governed without belonging to a party. Before leaving

office in 1945 Vargas created two parties, the Partido Trabalhista Brasileiro (PTB) and

Partido Social Democratico (PSD), establishing a pattern of state manipulation of the

party system. Another pattern to emerge during the period between World War II and the

coup of 1964 was the use of electoral alliances among parties. Electoral laws

"encouraged crossparty alliances and the independence of candidates from parties, and

rewarded small regional parties" (Coppedge 1998: 198). National parties worked against

the driving forces of Brazilian politics: localism and patron-clientelism.

Unlike other bureaucratic-authoritarian regimes, the military regime that came to

power in 1964 in Brazil took steps to maintain the trappings of democracy. Whereas the

congress was closed and parties were forbidden in Chile, Argentina, and Uruguay, in

Brazil the regime organized a two-party legislative system. ARENA (Alianga

Renovadora Nacional) was the pro-government party, while the MDB (Movimento

Democratico Brasileiro) served as the officially sanctioned opposition party. However,

unlike its South American neighbors, which saw the resurgence of traditional parties at

the end of authoritarian rule, in Brazil none of the parties founded before the coup

survived into the post-military era. In 1979 the military regime allowed for a

liberalization of the laws governing party formation, with the intention of splintering the

opposition. As a result, the MDB succumbed to infighting, with most members joining









the PMDB (Partido do Movimento Democratico Brasileiro). Likewise, ARENA also

fractured, with most members joining the PDS (Partido Dem6crata Social). After 1979

there was a rapid growth in the number of parties, and was hastened in 1985 by the

passage of laws that further facilitated party formation. The result of these permissive

laws has been an unabated fragmentation of the party system. For example, none of the

parties serving in Cardoso's coalition existed before 1970.

There are other aspects of the Brazilian political system which adversely affect

the party system. Brazil has a combination of presidentialism and multipartism, a mix

that "is especially unconducive to party building and to democratic stability. This...is

further complicated by the strong federalist bases of Brazilian politics. Power in

concentrated in the national government, but political careers are based on the local or

state level" (Mainwaring 1992-93: 700). Furthermore, the electoral system serves to

further weaken party coherence. As Ames (1995a, 1995b) shows, the system of open-list

proportional representation provides incentives for politicians to focus on local issues and

deliver pork to small, targeted constituencies. Such behavior inevitably weakens party

formation and discipline.

The structure of the Brazilian party system, in turn, has had seriously implications

for the implementation of economic reforms. Because of the fragmentation of the party

system and lack of discipline among party members, the passage of contentious

legislation by the president entails a cumbersome process of negotiation with individual

politicians. One of the functions of parties is the aggregation of interests; these groupings

facilitate the negotiation of issues by lowering, in effect, the transaction costs of

negotiation. But in Brazil, there are extensive costs for executives who are attempting









reform. Pork and other perquisites must be distributed, thus "the fiscal impact of

securing support is high" (Kaufman Purcell and Roett 1997: 16). Cardoso has had some

success in passing elements of his reform agenda, especially at the beginning of his

presidency, but he has had to accede to numerous demands for patronage from

congressional supporters. Winning this support has worsened the deficit of the central

government, which, as mentioned above, had already deteriorated due to the Real Plan.

"The net public sector debt doubled under Cardoso, surpassing $300 billion (or about 36

percent of GDP) in early 1998" (Smith and Messari 1998: 9). In recent years government

revenues have reached their highest levels in history. However, expenditures are growing

even more quickly and these new expenses, rather than being used in critical areas like

education and health care, are devoted to interest payments, social security payments, and

salaries and entitlements of public-sector employees.

The Electoral System and Congress

Barry Ames (1995a: 407) makes a persuasive argument that "at the core of

Brazil's institutional crisis lies the electoral system." He employs a rational choice

argument in his analysis, highlighting the incentives that the electoral system presents to

candidates to the national congress. The Chamber of Deputies (the lower house of

congress) is based on an open-list proportional representation PR system. Under this

system, each state is a single, at-large, multimember district. In a system of closed-list

PR, voters are presented with a list of candidates from each party. Candidates are listed

in the rank by which they will be selected. Hence, if a party lists ten candidates and is

awarded three, the first three candidates will be granted seats. Under the Brazilian

system of open-list PR, parties do not determine which candidates will represent them;

the voters do. This forces candidates from the same party to run against each other,









serving to further erode party coherence and discipline. Party fragmentation in turn

makes it difficult for presidents to negotiate legislation. They are forced to dole out

patronage such as appointments, cabinets positions, and government funds for pet

projects to politicians in order for their support of the president's agenda, a very costly

and time-consuming process.

Besides the impact of the electoral system, there are other aspects of the national

congress that hinder the passage of reforms, which are almost universally proposed by the

president. As mentioned above, the federal system establishes upper and lower limits for

the number of deputies from each state, ranging from eight to seventy. This has the effect

of bolstering poorer, less-populated regions, which are those most dependent on financial

support from the federal government. Hence, more often than not they are the ones most

opposed to efforts to correct the government's fiscal imbalances. Furthermore, several

new states have been created during the past two decades in the North and Center-West,

themselves poor and sparsely populated areas. Since all states are granted three senate

seats, this has made the Senate even more averse to some reforms than the Chamber of

Deputies. Most national politicians spend only a few years in Brasilia before returning

home to run for local office. This focus on local politics, rather than national issues,

further complicates efforts to pass reforms in areas such as the economy, social security,

and public administration.


Conclusion

The institutional design of Brazil's political system has had adverse effects on the

passage of reforms. The 1988 constitution created a problematic revenue-sharing

arrangement between the federal government and lower levels of government. It also









mandated state monopolies in various areas of the economy. Since the only way to

amend the constitution is with two majority votes of sixty percent or greater in both

houses, change proved difficult. The constitution also created lax rules for the formation

of political parties and an open-list PR system that further erodes party discipline. Party

underdevelopment has been offset to some degree to the Brazilian variant of

presidentialism, which possesses extensive reactive (veto) and proactive (legislative)

powers. Still, presidents cannot always govern by decree, and the fragmentation and

weakness of parties makes negotiation and coalition-building tedious, if not impossible at

times. All these factors taken together help to explain why Brazil delayed so long in

implementing reforms that would stimulate its stagnant economy.

Since the start of the New Republic in 1985, Brazil was faced with a deteriorating

economic situation. The first two civilian presidents, Sarney and Collor, introduced

numerous stabilization plans, all of which failed. Most explanations of these failures

pointed to the relationship between the still-tentative transition to democracy and Brazil's

political system, which, most agreed, possessed institutional aspects that encumbered the

implementation of much-needed economic reforms.

It was not until the introduction of the Real Plan in 1994, under the leadership of

then-finance minister Cardoso, that the economy was finally stabilized. After his election

as president, Cardoso was able to pass an array of reform legislation, which will be the

focus of the next chapter. Still, other facets of his reform agenda-including reform of

the social security, tax, and judicial systems as well as the public sector in general-have

fallen short of the president's expectations. Again, scholars point to institutional factors

as causes for these shortcomings. What these explanations lack, however, is precisely a






63


framework with which to account for Brazil's extensive economic reforms. This chapter

has sought to demonstrate the relationship between political institutions on the one hand,

and the difficulty of implementing reforms on the other. As I will argue in the next

chapter, what is needed is a reassessment of the relationship between political institutions

and outcomes in the Brazil. There I will what factors account for the reforms that have

been successfully implemented during recent years, thus calling into question the

standard institutionalist critique of Brazilian politics.














CHAPTER 4
ACCOUNTING FOR ECONOMIC REFORMS IN BRAZIL

In the previous chapter I analyzed the principal political institutions of Brazil and

the ways in which they have impeded efforts to pass economic reforms. As mentioned in

that chapter, there is consensus among scholars that the institutional design of the

Brazilian political system has complicated the passage of reform legislation. Many point

to this institutional design as the root cause for Brazil's late start with structural economic

reforms vis-a-vis other Latin American countries. Despite the insights provided by this

institutionalist literature, it too often focuses on the failures of reform in Brazil rather

than the successes. What is needed, I argue in this chapter, is a reassessment of the

impact of political institutions on the passage of economic reform legislation in Brazil,

for extensive structural reforms have been enacted during the past decade, especially

since the start of the Cardoso presidency in 1995.

In this chapter I will examine the efforts made by presidents in the New Republic

to pass economic reforms. The administration of President Fernando Henrique Cardoso

will receive most scrutiny, for it has been during his time in office that reforms have

proceeded furthest. After outlining the vagaries of the reform process since the transition

to democracy, I will analyze one set of reforms from early in the Cardoso administration.

Know as the Big Five, these reforms were a set of five separate constitutional

amendments that ended the state's monopolies in several sectors of the economy. One of

the amendments also eliminated the legal distinction between foreign and domestic

capital. I will examine the Big Five in detail because they were one of the earliest, and









most important, successes of the president's economic agenda. Finally, I will conclude

the chapter with observations about how, in spite of the institutional obstacles discussed

in Chapter 3, economic reforms have been possible in Brazil. Too much emphasis has

been placed on the obstructionist nature of the country's political system, I argue, and

scholars must therefore reassess the impact of political institutions on economic reforms

in Brazil.


The Changing Brazilian Political Economy

I contend that significant changes have occurred in the Brazilian political

economy during the past decade, especially during Cardoso's tenure. What evidence is

there to support this contention? Writing in 1992, Riordan Roett provided the following

description of the Brazilian economy:

In spite of [President Collor's] efforts, Brazil remains one of the world's most
protected economies with an average import tariff of 40 percent (compared with
Argentina's 11.5 percent). Business remains under the control of cartels and does
not cooperate with the government's economic stabilization efforts. Only at the
end of the year did the first privatization take place-with widespread protest and
violence. The public sector deficit is a giant obstacle to efficient policy-making.
(Roett 1992: 179)

In 1993 an article in Forbes disparaged Brazil as the "sick man of Latin America," as

inflation reached over 2000 percent and no solution to the problem seemed in sight

(Forbes 1993). Yet during the following year, 1994, inflation, the persistent economic

demon of Brazil, was successfully controlled by the Real Plan. The stabilization caused

by the Real Plan paved the way for resumed growth and an influx of foreign investment.

Just as impressively, the national privatization program has generated more revenue in

Brazil ($102 billion between 1991 and 2001 [BNDES 2001]) than in any other county in

the developing world. Foreign trade has been greatly liberalized, with an average import












Table 4.1 Key Economic Events in the New Republic, 1985-2001
President Period Reforms
Sarney 1985-1990 Four stabilization plans, all of which failed
Privatization on very small scale

Collor 1990-1992 Initiated national privatization program
Began liberalization of trade
Failed stabilization plans

Franco 1992-1994 Tepid towards the reforms introduced by Collor
Overshadowed by then-finance minister Cardoso
Introduction of Real Plan

Cardoso 1995-present Deepening of privatization program
Deepening of trade liberalization
End of state monopoly on reinsurance (1996)
Administrative reform (March 1998)
Watered-down social security reform (Nov. 1998)
Fator Previdencidrio (Nov. 1999)
Law of Fiscal Responsibility (April 2000)



tariff of 14.3 percent in 1999 (IDB 2000: 125). And in spite of many setbacks, the fiscal

deficit is slowly coming under control. Within a decade, then, the Brazilian economy has

undergone extensive restructuring. Yet these changes have not happened by chance.

They are the direct result of changes within the political system. And as I shall argue,

these changes call into question the prevailing institutionalist critique of Brazilian

politics.


The Vagaries of Economic Reforms in the New Republic

Almeida (1996: 223) cites three factors that account for the delay in economic

reforms in Brazil: "the power and profile of interest groups; the decentralization of

governmental structures; and changes in the structure and capacity of the federal









government." Because of these factors, economic reforms in Brazil have been guided, in

Almeida's words, by "pragmatism by necessity." Rather than disparaging the slow pace

of change, Almeida posits that Brazil has in fact benefited from it for several reasons.

First, the country has avoided large-scale de-industrialization, as has occurred in some

countries that undertook a more orthodox approach to reforms.1 Second, she believes

that the slowness of reform bodes well for democracy in Brazil. Rather than having a

rubber-stamp Congress, the Brazilian legislature has given voice to many concerns about

the consequences of economic reforms. While this may have slowed the pace of reform

in Brazil, she contends, such a situation is preferable to cases in which presidents have

autocratically implemented draconian stabilization measures.2

As touched upon in Chapter 2, Brazil underwent the transition from authoritarian

to democratic rule at the same time that the economy was suffering from the debt crisis.

When the first civilian president, Jose Sarney, took office in 1985, there was still no final

agreement with Brazil's foreign creditors, nor was there any clear plan on how to tame

inflation. Sarney's answer to the country's economic dilemmas came in February 1986

with the introduction of the Cruzado Plan. The plan introduced a new currency, the

cruzado, a temporary wage and price freeze, and a temporary freeze to indexation. While

the plan initially brought about price stabilization and a boom in consumption, producers

were not able to maintain the price freeze for long. Furthermore, the government was

accumulating a growing deficit. "The problem for the Sarney government was that it had



1 For a criticism of the quick move towards market reforms in Chile, see Collins and Lear
(1994).
2 Chile under Pinochet, Peru under Fujimori, and Argentina under Menem are three
prominent examples of the autocratic implementation of neoliberal reforms, indicative of
what O'Donnell (1994) calls "delegative democracy."









neither the political will or the political capacity to reduce its budget deficit...the Sarney

government could function only by paying out patronage resources and by preserving

congressional access to those resources" (Kingstone 2000: 189). After less than a year,

then, it was clear that the Cruzado Plan would not be effective at stabilizing inflation.

During 1987-88 Congress met as the National Constituent Assembly (ANC),

which meant that other legislative matters were put on the back burner. The final result

of the ANC, the constitution of 1988, was a further setback towards stabilization, for, as

mentioned in Chapter 3, it mandated a greater role for the state in the economy (by means

of monopolies in several sectors of the economy) and also created a fiscally deleterious

revenue sharing arrangement between Brasilia and lower levels of government. Sarney

introduced three other stabilization plans during his term, all of which failed. At the end

of his term public dissatisfaction was extremely high, leading people to look for someone

who could improve the deteriorating economy, as well as control what most perceived as

a highly corrupt political system. With inflation accelerating and the economy

stagnating, the Brazilian public found such a person in Fernando Collor de Mello.

It was during the administration of Collor (1990-1992) that Brazil made its first,

hesitant steps towards economic reform. Sarney left office in ignominy, having

introduced four unsuccessful stabilization plans. It was under these conditions that the

dark horse candidate Collor, a former governor of the poor northeastern state of Alagoas,

was able to rise to power. Collor won the presidency by emphasizing his commitment to

fight corruption and stabilize the economy. He created his own party, the Party of

National Reconstruction (PRN), to run for office, eschewing alliances with any

established parties. Upon taking office, he had no firm congressional base with which to









work. Furthermore, his harsh rhetoric against the political class did little to enamor him

with members of Congress. With an extremely conflictual relationship with Congress,

Collor found himself besieged when allegations about corruption by him and his family

and associates surfaced in 1992. Later that year he became the only president in

Brazilian history to be impeached.

Despite the ignoble circumstances under which Collor left office, during his brief

tenure in office Collor implemented several reforms that laid the groundwork for

subsequent reform efforts. His two greatest accomplishments were spearheading the

national privatization program in 1990 and opening the Brazilian economy to greater

foreign competition. It should be noted that these were areas in which the president could

act alone, through his decree power, to effect change; privatization and trade

liberalization were two areas where the president has a free hand vis-a-vis Congress. Yet

Collor's habit of riding roughshod over Congress, such as making frequent use of

decrees, did little to help him as the corruption allegations began to arise (Weyland

1993).

The interregnum of President Itamar Franco, Collor's successor, was a time of

growing uncertainty about the future of Brazil. Franco was not a supporter of Collor's

earlier reforms measures, but at the same time Franco did not work actively to reverse

these reforms. Franco's most important move in the economic realm was appointing

Senator Fernando Henrique Cardoso as finance minister in early 1993. As will be

discussed in greater detail below, Cardoso's role in crafting the Real Plan was to have

profound political ramifications during the early days of his presidency.









The Real Plan, which Cardoso devised and implemented as finance minister to

President Franco, sought to cure the country's hyperinflation by creating a new currency,

the real, and linking it to the dollar with a crawling exchange rate (managed float). In

order to address successfully the hyperinflation problem, it would be necessary to rein in

public expenditures. In this respect privatization was to play an important role. The

divestiture of state assets would add additional revenues to the national treasury, thus

alleviating the government's deficit. Furthermore, if foreign participation in the program

were expanded, foreign investment could help the country's balance of payments. In

order to attract foreign capital, however, the laws governing foreign investment had to be

changed so as to end discrimination between foreign and domestic capital. Without such

changes, it would be difficult to attract the investments needed to modernize the

country's infrastructure, much of which had been under control of the state for decades

and was in desperate need of improvements. As we shall see shortly, President Cardoso

was acutely aware of these facts as he quickly pushed for an extensive array of economic

reforms upon taking office.


Cardoso and the Economic Reform Agenda

Since taking office on 1 January 1995, President Fernando Henrique Cardoso has

worked to promote legislation that would liberalize and make more competitive the

Brazilian economy; to reform the political system, ranging from the laws governing party

formation and representation in Congress, to the system of revenue sharing between the

federal government and lower levels of government; and to restructure the country's

bloated social security and other pension programs that are near bankruptcy.












Table 4.2 Monthly Inflation Rates, 1994
January 42.2
February 42.4
March 44.8
April 42.5
May 40.9
June 46.6
July 5.5
August 3.3
September 1.6
12 months to end-Sept 3748.1
January-December 1993 2708.6
Source: LARR-Brazil (1994b: 7)


Figure 4.1 details the dramatic drop in inflation brought about by the Real Plan.

When the plan was implemented in June 1994, monthly inflation exceeded 40 percent.

By September of that year, it had dropped to less than 2 percent. As the Real Plan was

implemented and inflation slowed dramatically, millions of Brazilians quickly saw a rise

in their purchasing power. The stabilization plan had, in short, ended the onerous

"inflation tax" that had haunted the poor for years. The rapid improvement in the

population's living standards, especially the poor who worked outside of the formal

economy and whose wages were not indexed to inflation, gave Cardoso enormous clout

as he decided to run for president. He warned the public that his opponent, Luis Inacio

"Lula" da Silva of the Workers' Party (PT), would return Brazil to its previous days of

statism and fiscal irresponsibility, thus erasing the recent gains made in the fight against

inflation. Cardoso won the support not only of the nation's business leaders (Kingstone

2000), but also the overwhelming support of the population at large, who had seen the

tangible benefits of Cardoso's stabilization plan. On 3 October 1994, Cardoso garnered a









first-round victory in the presidential election, receiving 53 percent of the vote (LARR-

Brazil 1994a). Thus, as he assumed power Cardoso enjoyed immense popular support,

and, as the president would soon argue, a mandate to reform the state and its role in the

economy. His hope was to translate his popularity from the Real Plan into political

capital in his attempt to promote economic reforms early during his administration.

Before entering a life of politics, Cardoso established his reputation as one of

Brazil's foremost social scientists. His writings on dependency theory placed him at the

fore in the debate about capitalist development and its effects on underdeveloped

countries. He was exiled along with other leading Brazilian social scientists during the

late 1960s, during which time he lived in Santiago, Chile, with other Latin American

intellectual exiles and also spent time at universities in the United States and Europe.

Following his return to Brazil, Cardoso launched his political career in the 1970s, running

for the Senate in his home state of Sao Paulo. Over time his political views moderated,

moving to the center on economic issues while maintaining a social democratic

philosophy on social policy (Goertzel 1999). During the 1980s he was a founding

member of the Brazilian Social Democratic Party (PSDB), which also counted among its

ranks long-time Cardoso friends Jose Serra (Cardoso's current Health Minister) and

Mario Covas (the late governor of Sao Paulo).

As his stock rose during the early 1990s, Cardoso was to draw upon his extensive

academic knowledge and political experience, both of which influenced his thinking. He

acknowledged that the root of many of Brazil's problems lay within the nation's political

system. As Peter Evans, writing about the necessary conditions in which to undertake

economic adjustment, states, "Dismantling the state is not the answer. It must be









reconstructed" (1992: 141). Evans highlights the following paradox arising from recent

attempts at economic reforms in developing countries: the state, which is believed to be

the root problem of recent economic crises, is now trusted with the task of implementing

solutions to these crisis. Cardoso has been acutely aware of this paradox (Kingstone

2000: 201). Reform of the Brazilian economy and an improvement in social welfare

would necessarily entail a reform, and hopefully strengthening, of the state itself.

Reflecting his avowed social democratic philosophy, Cardoso pledged during his

1994 campaign to make agriculture, employment, security, education, and health the

main priorities of his administration (Kingstone 2000: 197). Upon assuming office,

however, the new president dedicated himself to swift economic reforms, arguing that it

would be impossible truly to improve social welfare until structural adjustments has been

undertaken and the state itself had been reformed. One of the president's first, and most

successful, moves was an attempt to end the state's monopolies in several areas of the

economy and to end discrimination against foreign capital. It is to this subject that I now

turn.


The Big Five Constitutional Amendments of 1995

Cardoso assumed the presidency in early 1995 with a resounding mandate.

Having authored and implemented the Real Plan the previous year as finance minister

under President Franco, Cardoso was able to capitalize upon his success in tackling

inflation during his bid for the presidency. Upon taking office, he attempted to make use

of his enormous popularity by pushing for a series of economic reforms, later dubbed the

Big Five, that would require amending the constitution.









The passage of these amendments was important to Cardoso's economic plan for

several reasons. First, and most importantly, they would end the state's mandated

monopoly in key sectors of the economy, therefore opening them to foreign participation.

This step had two purposes, one purely economic, the other economic and psychological.

From a purely economic standpoint, these constitutional changes would allow for the

infusion of badly needed capital into key sectors of the economy. With the exception of

the mining sector, these were areas in which the state had performed poorly in recent

years and as result had not received adequate investment. Furthermore, the

administration and even many within the business community realized that competition

was needed in order for Brazil to grow rapidly.

Another purpose of these amendments was to send a clear and strong signal to

international investors that Brazil was serious about structural reforms. The simultaneous

end of the discrimination between foreign and domestic capital and the opening these

sectors to private (foreign and domestic) participation consequently assisted in attracting

record levels of foreign investment to Brazil.

Despite the new president's popularity, the passage of these pivotal measures was

in no way assured. First, they were likely to draw criticism on nationalist grounds, as

these amendments would pave the way for the privatization of SOEs in politically

sensitive sectors.3 Second, there were possible political obstacles. As discussed earlier,

amendments to the constitution require two separate majority votes of sixty percent by






3 See the collection edited by Barbosa Lima Sobrinho (1994) for a scathing criticism of
the Brazilian privatization program, manily on nationalist grounds.












Table 4.3 Cardoso's Legislative Coalition, 1994
Party Chamber Senate
PMDB 107 22
PFL 89 18
PPB 88 11
PSDB 62 11
PTB 31 5
Coalition Total 377 67
Total Seats in Each House 513 81
Coalition Percent 0.73 0.83
Source: Kingstone (1999: 200)


both houses of congress. Hence, tampering with the constitution in Brazil requires a

rather broad consensus among legislators. The 1988 constitution grants presidents some

of the most sweeping decree powers in the world (Power 1998b), the most important of

which is the medidaprovis6ria (provisional measure). However, the constitution also

codified a number of nationalistic economic tenets, such as granting the state monopolies

over traditionally sensitive sectors, namely those related to sectors that had once been

controlled principally by foreign companies. In order to liberalize these sectors of the

economy, therefore, Cardoso could not utilize his own presidential legislative powers but

would instead have to jockey with Congress.

Cardoso's electoral alliance included his own Brazilian Social Democratic Party

(PSDB), the conservative Liberal Front Party (PFL), and the smaller Brazilian Workers

Party (PTB) and the Liberal Party (PL). Following the election, the president included all

these parties except the PL in his coalition. It also expanded to include the catch-all

centrist Brazilian Democratic Movement Party (PMDB) and the Progressive Brazilian

Party (PPB). Coalition members held 73 percent of seats in the lower house and 83









percent of seats in the Senate (see Figure 4.2). This was a solid majority, yet the breadth

of the coalition meant that certain parties would be less amenable to certain measures.

For example, the PFL, controlled by the powerful senator Ant6nio Carlos Magalhdes

from the state of Bahia, favored a neoliberal economic agenda but was extremely

conservative on social issues. The president's party, in contrast, was one of the most

reserved in its support of economic reforms but favored a progressive social agenda.

From the outset Cardoso realized that he would have to create sets of "shifting

alliances," looking for support on the right for his economic proposals and turning to the

left for help in passing his social agenda (LARRB 1994b: 7). Despite his avowed

credentials as a social democrat, Cardoso's first priority upon taking office was economic

reform. Towards this end, the Big Five amendments stand out as one of the earliest and

boldest of the president's proposals.




Table 4.4 The Big Five Constitutional Amendments of 1995

Amendment 5 Allowed for private participation in natural gas distribution
Amendment 6 Eliminated legal distinction between Brazilian and foreign capital
Amendment 7 Opened internal Brazilian shipping to foreign participation
Amendment 8 Permitted private participation in telecommunications services
Amendment 9 Allowed for private participation in petroluem exploration
Source: BNDES (1996: 37)



Accounting for Economic Reforms in Brazil

To this point I have attempted to justify my rationale for reassessing the impact of

institutions on economic reforms in Brazil. The evidence demonstrates that significant

economic reforms have been passed during the past decade, which has helped restore









growth and has increased investor confidence. Now I want to suggest some reasons for

how economic reforms have been possible in Brazil. I look specifically at four factors:

timing; the type of reform proposed; ideology and ideas; and the influence of political

and economic factors. Taken together, I believe, these four variables help to explain the

depth of economic reforms in Brazil.

Timing

When looking at attempts to pass economic reforms in the New Republic, it is

clear that timing plays an important role. More specifically, economic reforms have

been easier to pass at the beginning of a president's time in office. This is not to say that

it has been impossible to pass reforms well into a presidential mandate. However, an

analysis of successful reforms shows that they are overwhelmingly stacked towards the

beginning of a president's tenure. In fact, all presidents from Sarney to Cardoso have

proposed drastic reforms near the beginning of their terms, though not all have had the

same level of success. The timing of reform is linked to Brazil's system of

presidentialism. "The combination of presidentialism, a fragmented multiparty system,

undisciplined parties, and robust federalism is often difficult... [and] makes it diffult for

presidents to establish reliable bases of support" (Mainwaring 1997: 56). Because of this

institutional structure, since the restoration of the democracy presidents have attempted to

use their honeymoon period to push for contentious economic reforms.

In presidential systems the transfer of power from one executive to another is a

momentous occasion. Parliamentary systems generally provide relatively easy

mechanisms for the removal of unpopular executives, such as the vote of no confidence.

Parliamentarism also tends to be more flexible vis-a-vis presidentialism because under

the former executives do not have to serve for a fixed mandate. Popular prime ministers









can call for early elections if it is thought opportune, or the opposition can attempt a vote

of no confidence if it senses that the government has lost popular support. Moreover,

under parliamentarism prime ministers inherently possess a congressional majority, either

with a coalition of parties or by just their own party. In presidential systems, it is

common for the executive and legislative branches to be controlled by different parties

(Linz 1994).

Brazil has a long history of presidentialism, yet its post-authoritarian variant has

been attenuated by the fragmentation of the party system. The high number of parties, as

well as their general lack of discipline, forces Brazilian presidents to rely on often

unstable and undependable congressional coalitions to support the presidential legislative

agenda (Mainwaring 1997). Even if a president's coalition possesses a majority (or

super-majority) in the Congress, there is still fear that coalition members will break ranks

and go against the wishes of the president and party leaders. Brazil's particular system of

federalism further undermines a president's authority. A further hindrance to presidents

is the weakness of the "coattails effect," in which presidents are generally able to muster

support for their agendas following an election. Samuels (2000) finds that in Brazil

governors, rather than presidents, possess powerful coattails. As he explains,

In Brazil federalism cuts presidential coattails short-or simply rips the
president's coat off his back, cuts it up, and divides it among the governors. The
root of this problem-and the reason why it remains so intractable-lies in the
impact of the institutions of Brazilian federalism on the party system in limiting
the linkage between national executive and legislative elections. (Samuels 2000:
17)

Because of the volatile nature of coalition politics and the limited coattails effect,

Brazilian presidents must adroitly take advantage of the political capital afforded to them

during their honeymoon.









All presidential systems witness the honeymoon effect to some extent, as the new

leader is initially given some leeway by the opposition. However, in Brazil this period

signifies an important window of opportunity in which it has been possible to muster

support for important, yet contentious, legislation. This phenomenon has been apparent

with every president in Brazil since the end of the military regime, from Sarney to

Cardoso. One of Sarney's first moves was the introduction of the Cruzado Plan, which

failed to stabilize inflation. On his first day in office, Collor announced the firing of

thousands of public sector employees, introduced a draconian stabilization plan (which

also failed), and took the first steps towards liberalizing foreign commerce. Finally,

Cardoso achieved most success with his reform agenda during his first year in office.

In short, all post-transition presidents have used their honeymoon as a period to

attempt to pass economic reforms, though with varying levels of success. With items still

left on Cardoso's economic and political reform agenda, it is possible that the next

president (to be elected in October 2002) will also pass through the same experience.

With the passing of time it is possible that this phenomenon will be less powerful than to

date. Nonetheless, recent history has demonstrated the importance of this period for new

presidents.

First- vs. Second-Generation Reforms

In Brazil it has been easier to pass first-generation as opposed to second-

generation economic reforms. First-generation reforms refer to areas such as

stabilization, liberalization, and exchange rate policy, all of which are directly related to

macroeconomic activity. These were generally the first measures undertaken in countries

during the 1980s in response to the debt crisis (Haggard and Kaufman 1992). In contrast,

second-generation reforms aim at changing areas such as the tax system, public









administration, social security and other social programs, and the political system. While

these areas do not have an immediate impact on the macroeconomy, collectively they are

important because they form the root cause of the state's fiscal imbalances. When

speaking of reforms, it is easy to compress first- and second-generation reforms into one

monolithic whole.

Differentiating between these two categories is necessary, for different kinds of

reforms face different kinds of political obstacles. This is because the ultimate success or

failure of a particular reform depends much on the relationship between the various

branches of government. Some political systems allow presidents personally to

implement certain types of legislation, sometimes without the consent of the legislature.

This is the case in Brazil, where presidents wield significant legislative powers, such as

the medidaprovis6ria. Two of the most significant achievements in the economic realm

during the past decade-the start of the privatization program under Collor and the

implementation of the Real Plan by Franco-were legislated by means of presidential

decree.

In other cases, the legislature is given the exclusive prerogative over certain types

of laws. In Brazil has been the case when changes were sought in the constitution.

Amendments must be approved twice by a three-fifths vote of both houses of Congress.

While Congress supported Cardoso with the Big Five, it has been much less receptive to

second-generation types of reform, such as altering the social security system. The

judicial branch can also be a key player by deciding on constitutionally questionable

reforms, by itself ordering certain measures, or by some other related means. In Brazil

the judiciary has often been ultimate arbiter of several economic reforms. For example,









the Supreme Court struck down in early 1999 a proposal from the Cardoso government

that forced retired public workers to contribute to the social security system.

Hence, when speaking of reforms, one must always be cognizant of the specific

kind of reform being proposed, as well as the legislative means by which such a reform is

to be implemented. Second-generation reforms are politically less palatable in Brazil

compared to first-generation reforms. The latter, unlike those pertaining to social

security or the bureaucracy, are subject to fewer veto players, "individual or collective

actors whose agreement...is required for a change of the status quo" (Tsebelis 1995:

289). This is partially due to the legislative powers available to the executive, as

presidents are able to implement many laws by decree. Even in those areas in which the

executive cannot act alone, congressional agreement with economic reform is often more

forthcoming than in other areas of reform. For example, during the 1990s the Brazilian

Congress generally sided with Presidents Collor and Cardoso in their plans for

privatization. Since much of the privatization program required congressional approval

(and not just presidential decrees), it would have been impossible for the program to have

progressed as far as it has: $102 billion in divestitures from 1991-2001 (BNDES 2001).

Such success is directly attributable to the relative absence of what Tsebelis (1995) terms

"veto players" with privatization vis-a-vis broader, second-generation reforms.

The case of social security reform further illustrates the role of veto players.

Figueiredo and Limongi (1998) contend that reform of the social security system

(previdencia) failed not only because of the content of the proposal itself, but also

because of the rules governing Congress. Reforming the social security system would

have required amending the constitution. As mentioned previously, amendments require









two separate votes of three-fifths of both houses of Congress. The reforms proposed by

Cardoso would have an immediate impact on a large number of people (pensioners),

which dissuaded politicians from voting for most of the president's proposals.

Furthermore, the procedures governing social security, which allowed the opposition to

force a vote on piece by piece basis, further hindered the passage of the reform. The case

of social security reform is an some ways the opposite of the Big Five, which involved a

much smaller number of veto players, namely workers in particular SOEs.

The Power of Ideas

Changes in politicians' opinions about the proper role of the state in the economy

are another reason for the success of economic reforms. The spectacular growth of the

Brazilian economy and the country's rapid industrialization following the Second World

War were directly attributable to the state's active involvement in the economy (Trebat

1983; Baer 1995). Although a state's involvement in the economy inherently involves

political considerations, a balance must be established between satisfying political

considerations and the need for economic growth. Over time the Brazilian economy

became increasingly politicized, with the result being a worsening of performance.

During the heyday of the military regime the number of SOEs expanded dramatically,

large-scale inefficient investments were undertaken, and by the late 1970s growth

stagnated. Following the debt crisis of the 1980s, most Latin American countries

implemented structural reforms with the aim of restoring growth. Brazil was the last

major country in the region to pursue such reforms.

There were several causes for this late start. A concern for social issues and a

desire to address the country's enormous inequalities marked the transition to democracy

and the promulgation of the 1988 constitution. The military regime was viewed as









hostile towards the poor and the working class. The drop in living standards and the

worsening of income distribution during the 1970s and 80s further fueled resentment at

past economic policies. Hence, there were pressures to address social ills either brought

about or ignored by the previous, military-led government. Yet another reason for the

delay with economic reforms, however, came from within the political class. Politicians

were long accustomed to using economic tools as patronage. Attempts to liberalize the

economy, privatize SOEs, reform the system of revenue sharing between the federal

government and lower levels of government-all these signified a reduction in the

instruments of patronage available to politicians. Although this was (and still is) the case

in countries the world over, many believed the problem to be intractable in Brazil, since

corruption and cronyism grease the wheels of national politics.

Despite this pessimistic prognosis, recent research indicates a change in Brazilian

politicians' attitudes toward the state's role in the economy. Timothy Power (1998a)

analyzed several data sets to discover whether there has been an attitudinal shift among

Brazilian congressmen (both deputies and senators) in regards to economic issues. Power

was able to reach several conclusions from his analysis. First, he found that

neoliberalism has made inroads among congressional politicians since the late 1980s.

Second, politicians from the center-left to the far right generally accept neoliberalism,

although there are still questions about the depth of politicians' commitment to

neoliberalism. Third, "President Cardoso has achieved an impressive string of legislative

victories in his pursuit of neoliberal reform," mainly through skillful coalition

management (Power 1998a: 67). According to Power's study, party discipline within the

Cardoso coalition has been remarkably high, in contrast to what most analysts claim.









Finally, the Left is isolated in its economic views, a fact that may bode poorly for its

future political prospects.

What do these findings have to say about the future course of economic reforms

in Brazil? Power cautions that his study is preliminary, yet believes that further research

will support these initial findings. His belief is that with the exception of the far Left the

political class in Brazil will continue to accept neoliberal economic tenets, at least during

the short term. Studies like Power's are important, for they reveal the underlying modes

of thinking within the political class.

As recent scholarship has shown, ideas play an important role in politics (Hall

1989). In one case study, Sikkink (1991) describes the way in which the idea of

desenvolimentismo, or developmentalism, took hold in Brazil during the Kubitschek era,

leading to the creation of state development-oriented institutions. Neoliberalism likewise

brings with it repercussions in the political sphere. President Cardoso, while ostensibly a

social democrat, has aggressively pursued a neoliberal agenda. He has successfully

enacted most first-generation reforms and consequently has focused his attention to

reform of the state, which is directly linked to the nagging problem of the fiscal deficit.

However, as discussed above such second-generation reforms have been more difficult to

implement in Brazil.

Influence of Political and Economic Variables

One criticism lodged against institutionalism is that it overemphasizes the

explanatory power of institutions in political outcomes. This has been the case in much

of the scholarship on Brazilian politics in recent years. For example, Brazil is said to

possess a system of "paralyzed federalism," with a "paralyzed competitive arena"

characterized by intractable tensions between the federal government on the one hand and









the state and municipios on the other (Souza 1997: 170). The result of this "paralyzed

federalism" has been a lack of governance, a highly problematic structural fiscal problem,

and a lack of authority for Brasilia vis-a-vis lower levels of government. While Brazilian

presidents, especially Cardoso, have attempted to recover the federal government's

authority, such efforts have been hampered by weak parties, clientelistic politicians most

interested in securing patronage for allies at the local level, and an apathetic political

culture that is distrustful of the political system.

Considering this gloomy scenario, it should seem a miracle that economic reforms

were ever passed in Brazil. Yet they have been passed, and it is doubtful that the country

will experience a large-scale regression to the statism of the past. These developments

clearly call into question the institutionalist critique of Brazilian politics. First, the

political landscape is changing, as evinced by politicians' shifting attitudes towards

neoliberalism. Hence our past assumptions about the "obstacles," i.e. political resistance,

to reforms must be revised. Second, despite the influence of institutions, one must still

pay heed to the power of political and economic variables. For example, by most

accounts President Cardoso has been a skillful if cautious negotiator, displaying a high

degree of political acumen when pursuing his reform agenda. We may legitimately

question how different events would have been in Brazil since 1995 if Cardoso or a

similarly capable politician had not been in the Planalto.

Besides political skill, there are a host of other factors at play in the course of the

reform process. For example, economic pressures, both domestic and international, have

been decisive in influencing the passage of economic reforms. According to Haggard

and Kaufman (1992: 4), "decision makers operate in a matrix of international, social, and









institutional incentives that sets limits on the range of policy alternatives." In the

scholarship on economic reforms in Brazil, institutional factors have been given almost

exclusive attention while others, such as international and social ones, have received

inadequate consideration.

An important case demonstrating the influence of external economic pressures

was the international financial assistance package of late 1998. Totaling $41.5 billion,

the package was tied to an agreement between IMF and Brazil. The distribution of these

funds was contingent on Brazil's reaching specified primary account surpluses, which

many doubted could be accomplished. Yet, to the surprise of many, the country has

consistently met its targets with the IMF. Cardoso has vigorously sought measures that

would enhance fiscal discipline, both at the federal and local levels. Towards this end he

secured passage of the Fiscal Responsibility Law in April 2000, which puts tight limits

on discretionary "social" spending at the federal level and imposes penalties on fiscally

profligate states (EIU 2000: 9). International pressures played a role in imposing fiscal

discipline on Brazil. Following the devaluation of the real in early 1999, analysts were

deeply concerned about the future course of the Brazilian economy. It was imperative to

assuage the fears of international investors, and towards this end Cardoso was able to

meet the fiscal targets agreed upon with the IMF, as well as the Fiscal Responsibility Law

even though such measures limited the discretionary spending available as pork to

congressional politicians.









Conclusion

Despite the bleak assessments offered by most political scientists, Brazil has

undergone widespread economic reforms during the past decade. Factors such as the

timing of reforms, the kind of reform proposed, changing ideas about the role of the state

in the economy, and (non-institutional) political and economic pressures all help to

explain how economic reforms have been possible in Brazil's "inchoate" (Mainwaring

and Scully 1995a) political system. Rather than criticizing the speed of reform in Brazil,

Kingstone (2000) believes that this success is directly attributable to the slow pace of

reform. According to him

each of the New Republic's presidents has been able to pass some small portion
of the overall [reform] program. In turn, each successor has been able to use
those reforms as a base on which to build further .. which has permitted the
continuation of the process and has consequently kept Brazil "muddling through
gridlock." (Kingstone 2000: 186)

He recognizes the problems presented by Brazil's institutional design and is doubtful that

Brazil will be an immediate resolution of the economic problems emanating from the

political system, such as a persistent fiscal deficit. He adds that "long-term resolution is

likely to come, if at all, only with slow, incremental reforms" (Kingstone 2000: 202).

Reform has occurred and will continue to occur in Brazil, yet subject to the constraints of

the country's political system. This is not to say that further reform will not be possible.

As recent years have demonstrated, reforms in Brazil occur at a slower pace than in other

Latin American countries. This alone is not reason for pessimism.














CHAPTER 5
CONCLUSION

The purpose of this study has been twofold. First, it has provided an overview of

contemporary scholarship on Brazilian politics. Much of this literature draws on

institutionalism in its various guises to explain the difficulties encountered in

implementing reforms in Brazil. Second, this study has attempted to highlight the gap

between the predominant interpretation of Brazil's reform efforts-what I termed the

institutionalist critique of Brazilian politics-and the country's successful record with

economic reforms, which has seem much progress under Cardoso's leadership.


Main Findings

Specifically, I argued that four factors help account for economic reforms in

Brazil. First, the timing of a proposed reform can influence its ultimate success or

failure. Brazil's combination of presidentialism, federalism, and multipartyism

complicates the passage of economic reforms (Mainwaring 1997). Hence, Brazilian

presidents at times have made use of their honeymoon period in office in order to pass

reform legislation. In Chapter 4 I analyzed in detail the passage of some of the most

important reform measures during the New Republic, especially the government of

Fernando Henrique Cardoso, showing that most significant economic reforms were

proposed at the beginning of a presidential term. Not coincidently, must successful

reforms occur during this period.









Second, first-generation economic reforms have been easier to pass than second-

generation reforms. This fact is due partially to the powers bestowed upon the Brazilian

president. The constitution of 1988 provided significant reactive (veto) and proactive

(legislative) powers to the executive. Many first-generation reforms have been realized

by means of presidential decree, thus circumventing the Congress. In those areas in

which Congressional support has been necessary, one sees that the areas in which

significant reform has occurred entails fewer of what Tsebelis (1995) calls "veto

players." That is, those areas with fewer actors whose interests involved are easier to

reform. Thus privatization has been impressive in Brazil, while social security reform,

which effects millions of people, has proven much more difficult to change.

Third, reforms have been possible thanks in part to changing attitudes among

politicians. As discussed in Chapter 2, Brazil has a history of heavy state intervention in

the economy. The state's role had not only an economic logic, but also played an

important political tool by providing the state and key politicians with patronage

resources. Following the debt crisis of the 1980s and the hyperinflation of the early

1990s, it proved difficult to alter this traditional relationship between the state and the

economy. In the words of historical institutionalism, the Brazilian political system

displayed a high degree of stickiness in regards to economic policy. However, as Power

(1998) has shown, only the Left still supports the traditional statist economic model,

while most others sectors of the political class support, at least in theory, neoliberal

ideals. This shift in politicians' opinions has played a role in securing economic reforms

in Brazil.









Fourth and finally, institutionalist studies have too often emphasized domestic

political factors while ignoring or paying inadequate attention to domestic and

international economic forces. Brazil's vulnerability to international forces has been a

key explanatory factor in the passage of several pivotal reforms in recent years, such as

the Law of Fiscal Responsibility. There is growing recognition that the country's

persistent fiscal deficit, as well as profligate spending by the states, harms the country by

raising interest rates and worrying international investors. Beyond economic forces, I

also argued that one must still pay heed to non-institutional, political factors such as

leadership. For example, one could argue that much of the success with economic

reforms since the mid-1990s is attributable to the acumen of President Cardoso. He has

been to "overcome" the problems emanating from Brazil's political institutional design,

thus furthering his economic reform agenda.

Collectively these factors help explain the passage of extensive economic reforms

in Brazil. Although studies of Brazilian politics since the transition to democracy have

been fixated on the institutional inadequacies of the New Republic, one can now perceive

a shift in scholars' attitudes towards the effect of institutions on Brazilian politics in

general. In a recent review of the literature on Brazilian politics and political economy,

one scholar noted that "works on Brazilian party systems and legislative-executive

relations continue to be dominated by outdated, impressionistic, and overdrawn

arguments and evidence," with the result being the perpetuation of "pessimism regarding

the nature and accomplishments of post-transition democracy in Brazil" and the country's

efforts to reform its economy (Resende-Santos 2001: 209). This shifting attitude is most









apparent among Brazilian political scientists, with many North American scholars still

subscribing to the institutionalist critique laid forth in Chapter 3.

One of the problems with the prevailing institutionalist critique of Brazilian

politics may not be so much overemphasizing the power of institutions as it is examining

the wrong institutions. With so much attention given to the political institutions

examined in the Chapter 3, scholars have often overlooked the impact of other variables.

Such a move has already started to occur, again spearheaded by Brazilian scholars.

According to Figueiredo and Limongi (2000: 168), "the characteristics of the decision-

making process-the legislative powers of the president and the legislative

organization-may be more appropriate determinants of governability than the form of

government, the characteristics of party system, or the electoral system." They find that

since the promulgation of the 1988 constitution, presidents have had a very high rate of

success in implementing their agendas, due principally to "the legislative powers of the

president and the centralized organization of the legislative work" (Figueiredo and

Limongi 2000: 151).

Their findings counter the conventional description of Brazilian political parties.

Riordan Roett, one of the foremost U.S. analysts of Brazilian politics, states that

"Brazilian political parties are best understood as shifting groups of self-interested

individuals who find it convenient to come together under a bland banner that usually

includes the words progressive, social, or democratic" (1997: 21). While Roett's

criticism still possesses some truth, research like that done by Figueiredo and Limongi

demonstrate that common notions of the Brazilian political system are becoming

increasingly outdated.









Suggestions for Future Research

The conclusions of this study point to topics for future research. Considering the

outmoded perspective of much of contemporary scholarship on Brazilian politics and

political economy, future studies must move beyond the conventional notion that the

Brazilian political system inherently impedes reform. Scholars must take it upon

themselves to reassess the impact of institutions on economic policy in general and to

look for answers to how reforms are passed. In some cases, studies could benefit from a

comparative, cross-country perspective.

First, future research should explore in greater detail the attitudes of politicians

towards the proper role of the state in the economy. Power's (1998) analysis, as he

himself states, only offers tentative conclusions. While his survey data from members of

Congress suggest that politicians have increasingly accepted neoliberal ideals, only time

will show the depth of their commitment to less state intervention. With a new

administration to be elected in late 2002, it will be interesting to see the level of

continuity in economic policy between the present and incoming administrations. The

success of the Cardoso government has been due in large part to the maintenance of a

broad coalition, which has remained united principally through the dispensation of favors

and pork to coalition members.

Further studies of the party system are needed, as presently there are conflicting

accounts of the level of coherence and discipline of Brazilian political parties. On the

one hand, scholars such as Mainwaring (1999) stress the low level of institutionalization

of parties in Brazil. However, others such as Figueiredo and Limongi (2000) stress that

parties have displayed a high degree of discipline since the start of the New Republic,

thus aiding president's in promoting their agendas. Certainly, the proliferation of parties