Between Presidential Power and Legislative Veto: The Impact of Polity and Politics on Economic Reforms in Brazil

Contents

Abbreviations

1. Confronting the Challenges of the “Lost Decade” - Reform Dynamics in Brazil bet- ween Heterodoxy and Orthodoxy
1.1 Introduction - Structure of the Dissertation
1.2 The Relevance of the Brazilian Economic Reforms as Empirical Foundation

2. Crafting Market Reforms with Multiple Veto-Players - The Theoretical Conditions of Reform Capability
2.1 The Rational-Choice Perspective
2.1.1 General Approaches
2.1.2 Stages of Reform
2.1.3 Conclusion
2.2 Elaborating a Basic Institutional Perspective with Rational-Choice Elements
2.2.1 Separation of Powers: Presidentialism versus Parlamentarism
2.2.2 The Separation of Purpose: Challenging Formal Institutions
2.2.3 Decisiveness versus Resoluteness - Rapid versus Sustainable Policy-making
2.2.4 Public- versus Privat-Regardedness
2.2.5 Endogenous Institutions and the Coalitional Nature of Policy-making
a. From Political Spot-Markets to Long-term-contracts
b. The Notion of the Pivotal Player
c. De-Constitutionalization and Re-Ronstitutionalization: Gaining and Forgoing Flexibility?
2.2.6 Preliminary Conclusion
2.2.7 Theses and Conditions

3. Endogeneous Institutions and the Political System: Increasing Decisiveness and Reform Capability
3.1 Blurring Policy and Polity - The Brazilian Constitution
3.1.1 Fatally Flawed or Historical Second-best Solution - the Creation of the Constitution
3.1.2 The Potential Impact of the Constitution on Policy-making and Reforming
3.2 Between Gridlock and Electoral Success - Brazilian Politics after the Enactment of the 1988s Constitution
3.2.1 Indecisiveness and Fragmentation as the Incumbent's Burden
a. The Legacy of Decentralization
b. Deficiencies of the Electoral System - Weak Parties and Strong Locals
c. Over-Representation in Congress
d. Coattail Effects and Electoral Interdependencies
e. Decentralization and Bicameral Incongruencies
f. Conclusion:
3.2.2 Elements of Decisiveness and Coherence
a. The Constitutional Powers of the Executive
b. The Negotiating Powers of the Executive - Assembling Majorities and Pork Barrel Trade
The Relations between Agenda-setter and Veto-player in the Brazilian Congress
The Role of Individual Budget Amendments
The Pension Reform as Evidence of the Possibilities and Limits of Patronage
c. The Missing Powers of the Legislative - Stability and Reliance by Submission and Executive Dominance
The College of Leaders
Empirical Evidence
3.3 The Growing Leeway of the Executive in the 1990s: Decisiveness and Indecisiveness in
a Dynamic Perspective
3.3.1 Executive Disenchantment: Limiting the Executive’s Powers and Negotiating Abilities
3.3.2 The Empowering of the Executive: Recentralization and the Revamping of the Inter-State Relations
a. Interjurisdictional Rivalries and Governmental Unity
3.4 Increasing Decisiveness and the Tightening of the Federal Budget by Reconstitutionalization
3.4.1 The Constitutional Status Quo of the Fiscal Relations
a. Stressing the Federal Budget I: Financial Transfers and Revenue-Sharing
b. Stressing the Federal Budget II: The Hardwiring of Social Expenditures
c. Stressing the Federal Budget III: the Borrowing Autonomy of the States
d. Parametric Conditions - The “Fiscal Straightjacket”
e. Preliminary Conclusion: The Earmarking of Revenues and Budget Rigidities
3.4.2 The Art of Recentralization - De-linking and the Social Emergency Fund
a. The Increase of Social Contributions
b. The Social Emergency Fund - Paving the Way for Fiscal Recentralization
The Significance of the FSE
c. The Law of Fiscal Responsibility - Glorious Peak of Re-centralization?
Main Contents and the Impact of the Law
The FSE and the LRF in Comparison - the Stages of Evolution
Conclusion

4. Gridlock, Delay and Success of Particular Reform Projects
4.1 Success of Privatization - Few Veto-Actors, Strong Consensus and Re- constitutionalization
4.1.1 The Privatization of CVRD - Benefitting from Previous Re-Constitutionalization
4.1.2 The Privatization of the Telecommunications Sector
4.2 Delay of Social Security Reform - Numerous Veto-Players, Weak Consensus and Constitutional Status-Quo-Orientation
4.2.1 The Phases of Social Security Reform
4.2.2 The Dynamics of Social Security Reform Negotiation: Adverse Conditions and Strategic Mistakes
4.3 Failure of Tax Reform - Few, yet Strong Veto-Players and Missing Political Will
4.3.1 The Course of Tax Reform
4.3.2 The Dynamics of Tax Reform: Lack of Compensation and Missing Political Will
4.4 Conclusion

5. Summary and Central Findings

6. Bibliography

Abbreviations

Abbildung in dieser Leseprobe nicht enthalten

Dedication

To write a dissertation - as with many other challenges and endeavours in life - is never only the work of one person. It may certainly be one person who writes the words, phrases and pages down, however always influenced and greatly supported by the thoughts and words of others. As each piece of art or literature, the attentious reader can find numerous imprints of those who contributed to the work by good advice and those who delivered the sources of knowledge the work is based upon.

First of all, I want to thank my supervisor Prof. Dr. Tobias Debiel for his great support and - as immensely rare trait these days - his enormous patience also in periods when I was hardly able to move along with my dissertation. Due to his early interest in my work the whole project was just made possible.

Additionally, Prof. Franz Nuscheler signalled great interest in this dissertation and provided much-needed support in evaluating this dissertation and providing interesting ideas. It was a great pleasure to me that he consented immediately to examine this dissertation which proved to me that the subject of this work is in fact appealing.

Furthermore Prof. Dr. José Maria Ramos deserves my gratitude. In many very specific problems and complexities he stood by my side, answering my questions patiently and also giving me good methodological advice.

I am proud to thank my family, in particular my parents, for their immense support, for always filling me with great optimism and encouraging me to write this paper despite a time-consuming job and many other interests and activities in life. And in several occasions my partner in life had the difficult task to cheer me up in difficult times - an always well-succeeded undertaking despite the immense geographical distance.

Finally, the memory of Prof. Dr. Christian Fenner is strongly present in this work, who inspired me vividly to go forward and who always urged me to pursue a doctoral project. It is a particular want on my behalf to stress that he was in many ways one of the best teachers I ever had. But there was a particular attitude which deserves highest esteem: Prof. Fenner belonged to the few who placed the needs and problems of his students, collegues and assistants above his own ambitions, always admonishing us that teaching is at least as important as research. His demise in January of 2006 was a tragic loss to all of his collegues, students and, in particular, friends.

1. Confronting the Challenges of the “Lost Decade” - Reform Dynamics in Brazil between Heterodoxy and Orthodoxy

1.1 Introduction - Structure of the Dissertation

The pace and scope of the Brazilian economic development in the 1980s and 1990s is intrinsically linked to the wide-ranging discussion of political and economic reform projects. In the wake of the crisis of the import-substitution economic model a wide array of approaches and theories revolved around one basic question: When there occurs a crisis in the economy - whatever definition of crisis may be applied in a particular case - which are the best ways of overcoming anti-reform resistances, regain economic growth and promote sustainable and “future-approved” development? In the most general sense a crisis of the political or economic system can be referred to as two-dimensional: on the one hand it means the shattering and dis-equilibrating of a formerly successful status quo, on the other hand it opens up the necessity to find a new status quo (or status quo post) that can be deemed compatible to the new economic, domestic as well as international, circumstances. In fact, the longer a prosperous status quo lasts, the more difficult a subsequent change will get. Path- dependencies develop, stable institutions arise, group interests and organizations take root and expectations about growth, inflation, external trade and other variables cement into place.

Since 1930 Brazil has been pursuing an amazingly stable and successful course of economic development. Initiated by the Vargas-Administration with its open commitment to import- substitution and promotion of industrialization, the subsequent decades were decades of high growth and rising - albeit unequal - per-capita rents. Between 1950 and 1960 its first peak was reached by Joscelino Kubitschek's visionary proclamation of the “fifty years in five” developmental program, legendarily symbolized by the construction of Brazil's contemporary capital Brasilia and underscored by the insight that development can be possible without tears (desenvolvimento sem lagrimas). The beginnings of the military government (1964) proved a catastrophe for the democratic and pluralistic culture, but only a minor nuisance for the pace of economic development at best. After short years of restrictive economic policies the development accelerated even faster, with the Brazilian economy opening up to foreign capital, the repression of labor demands and the establishment of an authoritarian regulatory framework.

However, several decades of development without tears required its price, in the sense that the tears had not utterly disappeared, but rather been held in a latent stadium: the high dependency on international capital flows and lending together with growing inertial inflation led finally to the economic downturn in the so-called “Lost Decade” which started about 1980 and lasted until later president’s Cardoso successful reform efforts. The term “Lost Decade”, originally emerged in Mexico, was based on the continuous downturn of the majority of economic indicators, such as growing employment, the worsening of the the fiscal budget as well as deterioration of the terms of trade. A particular expression of this overall deterioration was the debt crisis, mainly caused by higher international interest rates subsequent to Ronald Reagan’s reformulation of American monetary policy and put the whole Brazilian economic model in serious danger. Another aspect, deeply connected to the worsening fiscal situation and one whose impact on overall crisis perception in Brazil was hightest, was the annual inflation, starting at 226 % in 1985 and reaching its peak in 1990 with 30370 % (Samuels 2002). Though growth rates could be maintained for the time being (1985: 8,4 %/1986: 7,9 %), the annual inflation intensified and reached hyperinflationary heights which could not be confined until the enactment of the Real Plan in 1994 (Franco 1993). Besides, efforts to improve market and production efficiency have done little to reverse the country’s continuous economic downturn.

After the return to democracy in 1985 the subsequent administrations confronted the crisis of the economic model by applying various heterodox stabilization programs which in a certain way mirrored similar developments in Argentina by that time. Heterodox reform attempts are characterised by policies which aim at freezing prices, controlling the monetary influx into the economy and enforcing a countercyclical industrial policy. It contrasts with orthodox reform attempts that are rather based on reversing the fiscal deficit and restraining credit supply.

The main focus of heterodoxy laid on the assumption that inflation is intrinsically inertial: the problem thus lies not with the federal budget deficits but is rather due to “distributional dynamics” (Franco 1993) caused by a sequence of rising salaries and rising prices. The essential elements of this kind of stabilization comprise income policies that aim at freezing prices as well as monetary reforms and verbal (but frequently not abided) commitments to fiscal restraint. Indeed, all the heterodox programs, partially motivated by political electoral cycles, proved after short periods of relief complete failures and levelled the path to hyper- inflation. One example was the Plano Cruzado¹, an economic emergency program launched by former finance minister Dilson Funaro in 1986, which replaced the former currency, abolished temporarily the general indexation mechanisms and tried to anticipate salary adjustments.

The result was obvious: inflation decreased from 60 % in the first quarter of 1986 to 2,1 % in the third quarter, but regained high levels in the following year up to 44 % in the first quarter and 95 % in the second quarter of 1987. Since heterdox inflation did not tackle the problem of the fiscal operational deficit², it proved a highly inept and inconsistent measure to confine inflation and get rid of the crisis (Rezende 2006).

As many scholars noted, the late 1980s were coined by a political climate in which the political and economic elite, still profiting of the rent-seeking opportunities provided by the economic model of import-substitution, did not really suffer the effects of rising inflation. Though the old developmental coalition, characterised by reliance on import-substition and state interference in the economy, was in decline, a new hegemonic coalition did yet not exist - as exemplified by the 1988 Constitution that cemented Brazil’s transition to pluralistic democracy, yet lacked a stringent and consistent pace for the political and economic life of the country.

One of the main reasons which mitigated the immediate effects of the upcoming crisis was the fact that the nominal deficit was not equal to the core or operational deficit: The high inflation disguised the serious disequilibria existing in the public budget.³ The public receipts were in the comfortable situation of being indexed against inflation, whereas the real expenses could mostly not recur to indexation mechanisms - which caused the so-called Reversed Oliveira- Tanzi-Effect⁴. Thus inflation helped to reduce the real expenses and enabled the build-up of arrears (in relation to salaries and other public expenses): the payouts of salaries were simply delayed until the end of the year and thus diminished significantly in their real value. The original or core deficit must have been very high, and the higher it got, the more elevated inflation had to be in order to mask its detrimental effects.

Thus, by means of the inflation tax, classical instruments of interest accommodation and compensation had not yet ceased to exist, and the frequent recurrence to the inflation tax could easily be maintained. With other words: induced by inflationary dynamics, a deeprooted veto-coalition against anti-inflationary and structural reforms was established and could not be undone until the more successful and sustainable Plano Real.

At some point the hyperinflation and its detrimental effects could not be overcompensated by the gains which accrued to the federal budget and to reform adversaries. The late 1980s were still characterised by the basic notion that - built upon inflation-financing - the social debt should be redeemed, and social policy should be granted high priority in public expenditures. This agenda was to be replaced by a new one focussed on market reforms. Central to this new agenda was the notion of reducing the Custo Brasil⁵ - an imperative for the country’s competitive integration into world markets.

Thus the 1990s should become a period of intense reform efforts that - as an ever-growing legion of literature testifies - can be deemed as ambivalent at best. The Lost Decade was followed by a Difficult Decade (Montero 2003). However, Brazil did not follow the neoliberal blue-print sequence, prescribed allegedly by the Washington-Consensus-Institutions. In general the 1990s contributed to Brazil’s image - from the point of view of international, but also domestic investors - as an underperformer.

The effects and consequences of the Real Plan - or Plano Real - proved crucial for the subsequent reform efforts. Frequently mistaken as a pure orthodox monetary reform, it proved a successful mix of orthodox and heterdox measures which enabled the then minister of finance Cardoso to tackle hyperinflation and with that shielded the nation from the most severe consequence which would have proven fatal for years to come. Advocated by its prime architects - famous economists - like André Lara Resende and Edmar Bacha - the Real Plan when tax receipts are somewhat indexed - as it was the case in Brazil - whereas expenses are not indexed. In consequence the revenues remain constant while the value of expenses decreases with increasing inflation. rooted in the insight that the inflation in Brazil was inertial. The former heterodox stabilization plan had this insight in common, but lacked the commitment to fiscal consolidation and the monetary adjustment technology which made the plan sustainable. The “Unidade Real de Valor” (Real Unit of Value), conceived by former Finance Minister Rubens Ricupero, served as a key step to the implementation of the new currency, the Real. Additionally, by doing away with inflation and thus eliminating the inflation tax, many previously practiced compensation schemes could not be maintained. Fiscal discipline developed into a core norm which touched the federal as well as the state level and put into a fiscal straight-jacket (Samuels 2002).

The monetary stabilization served as prelude that facilitated virtually everything about reforms what ought to follow. The worst excesses of a crisis scenario were overcome, but these short-term gains - as underlined by a bunch of similar reform stories in the 1980s and 1990s - had to be accompanied by more profound, structural and long-lasting reforms which would place Brazil on a path to sustainable development amidst the central pillars of fiscal solidity, economic growth, price-stability and rising per-capita-incomes.

However, it is intriguing to note the very diverse success, as well as the highly different pace and scope of the reforms of the 1990s. The reform experiment in Brazil can be best illustrated by a pattern of muddling towards adjustment, in which conflicting political and economic views over policy making alternatives did not defeat efforts to adjust, but delayed its implementation and limited its success. My chief concern will be to explain why certain reforms were successful, while others got delayed or failed - and this in a context of constant institutional arrangements and even expanding decisionary leeway of the executive which would actually foresee a rather linear development.

Thus a central thesis of the dissertation refers to the continuous consensual enhancement of presidential powers under the Cardoso government and his immediate predecessor Itamar Franco - contrary to what most of the literature states - which was accompanied by various legislative successes. This enhancement of consensual presidential power was best reflected in the development of the horizontal and vertical fiscal relations which - more in terms of procedures than in outcomes - improved significantly. However, despite growing leeway of the executive several reforms could not be enacted, were continuously watered down or had to be postponed to the subsequent Lula government. Whereas the monetary reform as well as privatization proved successful and sustainable, albeit at the expense of economic growth which was highly restrained by a restrictive anti-inflationary and overvaluation policy, other structural reforms like tax or pension reform faced much larger obstacles. With other words: Brazil found itself in a situation characterized by a partial reform equilibrium⁶ (Hellmann 1998) in which crucial reforms were addressed, some finished and the majority halfway stopped.

In the second chapter I shall review existing theoretical approaches that are frequently applied in order to explain variances in reform outcomes. Explanations for the failed or successful adoption of reforms are found in different theoretical schools, including structural, ideational, institutional, rational choice, and cognitive-psychology approaches. In general, however, since most of them deal with the incentive structure of veto-actors, I shall refer to them as adherent to the Rational-Choice School. The major points these theories have in common is the conviction that the enduring puzzle for the politics of neo-liberal reform implementation in Latin America lies in the fact that those who have the most to gain from reform are members of a diffuse, differentiated mass in society that face high costs to collective action and must wait for the benefits of reform to develop over long periods (Corrales 1997/1998). This speaks clearly against the probability of reforming, especially if we base our considerations on a War-of-Attrition model - another Rational-Choice-based approach in Prisoner's Dilemma fashion - which picks up the notion of two competing groups that both suffer the costs of reform delay, albeit do not want to assume the respective costs. Consequently they hope for the other group to give up resistance and thus assume a major part of the costs of reform. If none of the groups concede they enter a war of attrition until one of them surrenders (Alesina/Drazen 1991).

However, collective action problems and resistances may be overcome by the occurrence of severe economic crisis, in which the procrastination costs of reform surpass the total and net transition costs of transformation (Bresser-Pereira/Apud 1997). The proponents of this branch see the occurrence of a crisis in a positive light: in the long run it may lead to faster reforms and thus higher and more sustainable growth. Related to the crisis-argument is the notion that a successful stabilization effort prior to structural reforms creates a so-called “Early Victory- effect”: presidents claim credit for their stabilizing efforts and - as prospect theory states - set the political and economic actors into the “realm of gains” with a high distributional mass. This facilitates subsequent reforms and should it make easier to overcome veto-coalitions.

While the above mentioned Rational-Choice approaches focus on incentives, the macro- institutional setting provides enabling and constraining devices. Several institutional variables of the Brazilian political system possess explanatory power: the fragmented legislature, which indeed in the Brazilian case is not conducive to stable, nationally oriented policy-making; the open-list electoral system, which contributes to cultivating personal votes (Carey/Shugart 1995) and augments the costs for side payments; the particular nature of federal-state relations which are rather characterized by conflict than coordination - in one of the world’s most federalist countries. So, what does this mean concerning policy and reform? Or rather: what do we know about institutions? In fact, the formulation and enactment of policies are highly contingent on political institutions. Alston and Mueller summarize in a convincing way:

“The political institutions determine who are the key players, the payoffs to the players, the forum in which they interact and the frequency of their interaction (Alston et al. 2005b).”

Thus, it is evident that one needs to analyze

1. The key players: who makes policy, and who opposes - in the most general sense it refers to the juxtaposition of executive and legislative powers.

2. Which payoffs can they expect? For example a particular electoral system determines quite significantly how an incumbent or a deputy has to formulate, propose and resist certain policies in order to gain the president’s favour, the electoral success or the discretionary transfers to his constituency.

3. The forum in which they interact and the frequency of interaction - mostly the national parliament, but also the assemblies, the committees and caucuses.

Pure institutionalist approaches, however, tend to overemphasize their role and exclude other intervening variables. The sole focus, for example, on the classical debate between presidential and parliamentary democracy or the recurrence on pure Separation of Powers - concepts does not bring us any further. Macro-institutional settings constitute the frame of reference for policies, but do not determine them. Rather the rules which regulate the relation between executive and legislative are much more apt to explain the different pace and scope of policies in general as well as reforms in particular. An institutional approach with rational- choice elements, giving due attention to the actor’ perspective, is consequently chosen.

For example, in presidential democratic systems it is always argued that the presidents, which by definition have a strong incentive to pursue stable fiscal and monetary policies, act in a quasi-authoritarian manner and impose their agenda onto the legislative. We need to make, however, an analytical distinction between mere constitutional delegation and process-driven delegation, and understand that the frequent contradictions between these concepts are precisely what is most interesting in the contemporary debate over which actor - the president or the legislature - controls the legislative agenda. Institutions and formal as well as informal rules are deeply intertwined. One has to take into account the informal rules of the game, a set of practices and unwritten norms that are commonly known to political actors and set the conditions under which the game is actually played. Building on Rational-Choice theories, this approach recovers the perspective of the politician as an agent capable of crafting optimal game strategies given the formal and informal constraints posed by the rules of the game. In fact, due to a high degree of Separation of Purpose - a concept introduced by Cox/McCubbins (2001) - and varying degrees of decisiveness and resoluteness the construction of successful governing coalitions is vital. It is not the grand institutional design that matters, but the endogenous nature of institutions, beginning from the coalitional dynamics in congress to the more or less institutionalized relations between federal government and the states which decide whether a political system is decisive or indecisive, apt at achieving initiation as well as sustainable implementation of policies.

In terms of material policies, initially indecisive and gridlock-prone political systems need to gain higher decisiveness - in the case of high degrees of (constitutional) policy-hardwiring one of the most important instruments is deconstitutionalization. But, as we shall argue, even crafting optimal game strategies and deconstitutionalization are not always sufficient: sometimes it is necessary to re-constitutionalize in order to lock in certain provisions, even if this means to tie partially its own hands. This is mainly due to the fact that pivotal players can emerge which endanger the coherence of the governing coalition.

These theoretical insights and the focus on endogenous institutions get us closer to resolving two central puzzles of this dissertation:

1. How can the apparent success and the growing leeway the President enjoyed during the 1990s be explained, despite constant (= invariant) macro-institutional settings, an overall missing consensus about the general pace of reforming and apparently contradicting characteristics of the political system? The relation of passed to unpassed laws rose significantly and the impression prevailed, that the President obtained higher degrees of freedom, thus a larger leeway. Here we offer an explanation that focusses on the endogenous nature of institutions.

2. Observing a strengthening of the President’s policy-making capacities contrasts with the view of the observed erratic reform performance. How then can we explain these variances in view of the increasing reform capabilities of the political system?

To the first puzzle we shall turn in the third chapter which aims at examining the reform capability of Brazil. An evaluation of whether elements of decisiveness or resoluteness predominate will lead us to the several steps necessary for successful policy-making. While in a static perspective the apparently contradictory tendencies of the political system (strong executive prerogatives versus numerous legislative veto points) appear to underline the erratic patterns of reforming, a more dynamic view will reveal an increasing reform capability under former president Cardoso.

Each evaluation of any political system has initially to consider the grand constitutional design - and the embodiment of this design is the constitution. A constitution is by definition a meta-institution that defines the rules of the game which can change in certain limits, but keeps its hands out of material policymaking. Whenever this is not the case, i.e. when a constitution is over-determined and “trespasses” into actual policy-making, the latter is partially excluded from the negotiating table - a fact which complicates successful and adaptable policy-making and makes reforms even harder. I shall show that this is exactly the case in Brazil. The Brazilian Constitution served as an instrument which included not only polity, but many policy-related prescriptions which narrowed the subsequent negotiating range. It contributed to the hardwiring of particular interests and codified the overrepresentation of smaller provinces in the national legislature which made the President’s agenda a hostage of local interests. It intensified the existing common-pool-problem and determined the character of the subsequent reforms, namely by creating a “Constitutional Agenda” which had to be transformed gradually into a non-constitutional one (Couto 1998).

Such a status-quo oriented constitutional document naturally complicates subsequent reform. To make up for missing reform capabilities, the only way to pave the way for reform are either equally strong powers for the executive or an overwhelming pro-reform consensus. The provisions of the Constitution, in fact, proved highly contradictory, being mirrored in several scholarly reflections of the political system in the 1990s: Some scholars argue that features of the political system as the open-list electoral system are responsible for gridlock, delay and eliminating of policies from the legislative agenda (Mainwaring/Shugart 1997). Namely Ames (2001), Mainwaring (1995) and Carey/Shugart (1998) stress the highly fragmented legislative, the detrimental effects of the open-list electoral system and the contradicting issues of a divided government - aspects not very amicable to reform. Others, most prominently figuring Limongi and Figueiredo⁷, come to the opposite conclusion: they attest a high decision power of the executive from which infers stable policy coalition and a submissive role of the legislature. On the rational-choice level several recent works (Alston/Mueller 2001) propose a model of legislative bargaining that explores the formal and informal mechanisms and payoffs that presidents used to compensate potential coalition partners and push their agenda for economic reforms. Reciprocal voting coalitions arise which will prove crucial for the enactment of policies.

Getting a dynamic point of view, this contradiction slowly gets resolved. An empowerment of the executive took in fact place - due to strategically prudent acting of the president, mainly reflecting tendencies of re-centralization and inter-jurisdictional rivalries, which provided a greater leeway to the federal government and moved the Brazilian decision-making process on federal level to higher degrees of decisiveness. Two developments reinforced this tendency:

Ad-hoc bargaining and “spot-market-politics” (Rodden/Arretche 2003) provided an initial leeway for the executive, yet could only yield partial reform equilibria and many legislative projects would have had the character of a mere sunset-legislation (Samuels 2003b). It was rather necessary to de-constitutionalize certain issues and by that make open negotiation and legislation possible. De-constitutionalization is per se only possible in a consential environment due to supermajority requirements.

A reverse process of “constitutionalization” benefiting the executive occurred in the 1990s when reforms were to put through efficiently and when overwhelming societal consensus was lacking. That means that after de-constitutionalizing certain passages of the Constitution it was highly recommendable to re-constitutionalize certain policies which served the preferences of the executive and proved mainly disadvantageous for the legislative and the states.

Subsequently in chapter 3 the thesis of the increasing reform capability in a dynamic perspective shall be underlined by the help of empirical examples of the Brazilian reform period. The heretofore elaborated arguments shall be applied to the general relations between executive and legislative in relation to the federal budget.⁸ The financial relations between those two powers were in the beginning characterized by the overwhelming influence of the Constitution. By intelligent bargaining, benefiting from high formal powers and a coherent agenda-setting strategy, the Brazilian executive managed to gain higher leeway during the 1990s and appropriated a larger bunch of resources formerly destined to states and regions. The 1988 Constitution incorporated many provisions highly detrimental to the executive and beneficial to the territorial entities (e.g. the higher share of tax receipts went to the states and municipal regions while the federal government disposed of a minor share). In the 1990s the executive managed to de-constitutionalize several items and thus shaping the financial relations to its advantage. This can be shown by the debt negotiations with the states which changed their fashion significantly. Finally the Law of Fiscal Responsibility (Lei de Responsibilidade Fiscal, LRF) represented the culmination of a re-constitutionalization process in which the higher leeway of the executive found itself “quasi-constitutionally” codified.

While the Judicial Power - The Supreme Court - served as an important brake and counterweight to the ambitions of the President, on the other hand the process of “gaining leeway” and appropriating higher resources was quite stony. It got only possible by intense bargaining and granting several side-payments. The history of the Social Emergency Fund (Fundo Social de Emergência, FSE)⁹ will make this evident. However, the development from the FSE to the LRF is enormously impressing and fortifies our thesis of growing presidential leeway.

In chapter 4 we will finally turn to the question why the reform performance was so different amongst specific reform projects, especially privatization, tax- and pension reform. The principal results of chapter 3 might insinuate that things got pretty easy for the executive in the 1990s. However, despite an overall tendency towards decisiveness and resoluteness in the financial relations, the picture is different in relation to single, distinguished reform projects. While privatization proved a major success story, the pension reform could be pursued in a rather thin-stretched version with many objections from veto-players (partial success). Finally the tax reform was completely abandoned after a time and was just resumed after the takeover of current President Lula da Silva (failure).

Thus, the rather linear development we observed does not provide us with a universal explanatory blueprint. It is erroneous to assume that the development which characterised the fiscal relations “trickled down” to each single reform project. Quite the contrary, we can observe rather different outcomes which prove a high degree of path dependency and dynamically varying support and adversarial coalitions. Each project requires thus a distinct process-tracing (George 1979).

1.2 The Relevance of the Brazilian Economic Reforms as Empirical Foundation

There are two principal reasons why I chose the Brazilian reform experience as principal empirical foundation for my study:

On the one hand the end of import-substituting industrialization initiated a bunch of heterodox and orthodox reforms in all Latin America as well as - on an earlier historical stage - in some Asian countries. Brazil was - as many scholars underline - even in the Latin-American context a relative “late-comer”, especially compared to Chile, but also in relation to Argentina, Peru or Uruguay. Thus certain crisis-conditions worsened and gained a particular momentum, as exemplified by the high levels of inflation. This caused a particular need for reforming and pressed a relatively “new-born” political system - after years of autocracy - to come to terms with a very demanding economic environment. I believe that the Brazilian experience suits very well to demonstrate the intrinsic difficulties, but also chances, which arise in rather fragmented, heterogeneous political systems when being confronted with the need for rapid change. The need for reform stresses political institutions, places them onto the test rig and thus shows certain mechanisms in a more poignant way.

On the methodological side, this study is originally conceived as a structured and focussed comparative case-study. The justification of case studies is evident: case studies can be applied to develop and refine theory. Additionally, there are in fact significant comparative elements included. However, comparison always occurs when various parameters and potential explanatory factors are held constant. Usually this is achieved in cross-national or cross-regional comparisons, in rarer cases in diachronic studies. For answering the first question - How the increased executive leeway can be explained despite apparent contradictions of the Brazilian political system - I chose a diachronic study which traces the process to its final outcome, accounting for the dynamic and time-variant nature of Brazilian politics, based on time-invariant macro-institutional settings.

For answering the second question - explaining the erratic pattern of policy-making and reform in particular fields - a structured and focussed comparison is in order. In a structured, focussed comparison the analysis of each case follows a rather common scheme. The base of such a comparison constitutes general questions that derive from theoretical assumptions (Beckmann 2008: 103). In short, the growing executive leeway should lead to more decisiveness and thus facilitate policy-making and the enactment of reforms.

2. Crafting Market Reforms with Multiple Veto-Players - The Theoretical Conditions of Reform Capability

Crafting Market reforms has in the last two decades undergone a miraculous transformation in perception: While in the late 1980s certain blueprint patterns emerged about how reforms have to be executed, some years later these apparently too simple and under-complex patterns lacked viability (Mukand/Rodrik 2002). Suddenly - especially in the Latin-American context - successful reforming was a rare, case specific occurrence and, of course, political art. Success and setbacks characterized the most salient reform processes, from Argentina to Venezuela, from Peru to Brazil¹⁰. Those who advocate reform are opposed to those who object: about whether reforms have to take place, whether they ought to be gradual or in “bigbang”-fashion and son on.¹¹

By emphasizing explanatory deficits of the most renowned and elaborated approaches to explaining pace and scope of reforms this dissertation hopes to achieve a first scholarly contribution. We can basically distinguish between different single substantial approaches as well as classificational schemes and reviews that attempt to organize the wide array of approaches into some distinguished branches. One of the most renowned classifications is Kurt Weylands classification into economic, political-institutional, rational-choice as well as ideational explanations (Weyland 1998).

- Economic theories refer to structural explanations, based on dependence-theory, that see the scope and pace of reforms mainly constrained by exogenous variables like international constraints, changes in the trade-regime, globalization as well as the 28 2. Crafting Market Reforms with Multiple Veto-Players occurrence of the debt crisis. Structural adjustment and market-oriented reforms are seen as mainly propelled by international organizations (Stallings 1992).
- Political-institutional approaches tend to emphasize the regulatory functions of institutions as well as the enabling and constraining incentives posed by the executive, legislative and judicial branches of governments (see e.g. Haggard/Kaufman 1995; Haggard/Webb 1994; Acemoglu/Robinson 2002; Alesina/Rodrik 1991; Geddes 1992; for an ecompassing overview: Cox/McCubbins 2001),
- Genuine rational-choice and psychological approaches enter the domain of individual cost-benefit-calculations that shape the perception of reforms. Benefits are weighed against costs, and based on its outcome different actors of society assume a pro- or antireform stance (Alesina/Tabellini 2005; Alston/Mueller 2005; Olofsgard 2002; Olson 1965/1982; Souza 1998)
- Psychological theories, including Weylands own prospect theory, consider individuals as risk averse or risk prone actors whose risk propensity varies with the proper situation in which they are in. While, according to Weyland, individuals embrace bold steps in the domain of losses, in the domain of gains people attempt to keep what they have and oppose severe further changes (Weyland 1998).
- Finally, ideational approaches recur to the diffusion of ideas, insights and time-variant convictions - especially after the downfall of communism in 1989 (Kahler 1992).

While these five branches encompass the majority of approaches that are significant for our theoretical fundament, we shall focus in particular on rational-choice and institutional approaches. While the analytical distinction between psychological and rational choice approaches is not always clear cut, the former can be seen as intrinsic part of the latter. Furthermore, in the wide array of accessible literature about the variables influencing reforming processes (see the above mentioned as Alston/Mueller; Alesina et al.; Haggard/Webb/Kaufman; Shugart/Carey) ideational and economic factors are considered as exogenous factors initiating or creating the basic need for reform (perception/reality), yet not attempt to explain the proper dynamics of reform. Admittedly, economic or structural and ideational explanations cannot be underestimated. Especially structural approaches usually represent the root cause, the basic condition why reforms are initiated at all. The structural necessities in the end open up the road for reforming by demonstrating that a political and economic status quo is not sufficient to cope well with the future. Insofar they bear some tautological traits - when there are some structural “hard facts” at odds, or, expressed differently, when there is a crisis, the necessity for reform is evident. The principal objective of the dissertation is not to determine the exact economic causes of reforms, but the way they are pursued and the mechanisms to which they are subdued. Furthermore, in a dissertation which focuses on institutional explanations and their deficiencies we shall treat structural explanations as exogenous variables which influence the pace and scope of reform in a parametric way. Shugart/Haggard (1997) and Haggard/Kaufman (1992b) focus on rational- choice as well as institutional approaches as endogenous variables explaining policy-making and reform. I shall follow this approach here.

The focus of the dissertation is clearly rooted in an actor-centered institutionalist tradition which perceives institutions as constraining and enabling devices that can offer obstacles or catalysts to policy-making and reforming. One perception that sometimes is missing is that they have an exogenous as well as endogenous dimension and are foremost expressions of human social engineering. They are actively suggestible and can be subject to constructive criticism. Other than economic, cultural and societal circumstances institutions can be “engineered” in the medium term - despite ample path dependencies and a natural institutional stickiness. Due to the frequent focus on their exogenous character institutional theory is inherently difficult to explicate, “because it taps taken-for-granted assumptions at the core of social action” (Zucker 2004). Thus by elaborating a study about the Brazilian reforming process I shall attempt to make institutional theory more accessible, especially with regard to the policy-making process in transitional periods. On the other hand I believe that an institutional focus always has to take into consideration the actor-level, i.e. the motivational patterns by which actors show compliance with or resistance to existing institutions. The endogenous dimension of institutionalism is thus deeply rooted in the rational-choice- perspective which - in the reform- and policy debate - focuses mainly on costs and benefits or different societal groupings. However, following the debates of the recent years it is justified to state that pure rational-choice-approaches were somewhat overvalued and treated reform delay on a rather too abstract and unempirical level. I shall thus focus on these pure approaches which explain the propensity for reform on a cost-benefit-scale and identify some of their shortcomings.

Chapter 2 - Crafting Market Reforms with Multiple Veto-Players - has the following principal objectives: On the one hand I shall perform a theoretical review describing the most important theories and approaches explaining reform dynamics. In a second step, it is crucial to show that common convictions have to be scrutinized - relating to rational-choice as well as pure institutionalist approaches. In a third step we formulate some basic insights by focussing on a modified version of an institutionalist approach, stressing furthermore some particular mechanics by which the reform process can be understood better. This shall be achieved by identifying a decision-making arena below the macro-institutional setting, in which coalitions are formed and decisions made in a decisive or resolute fashion. It is not only the formal separation of powers determining certain policies, but also the separation of purpose on an infra-institutional level which influences strongly whether the political system tends to decisiveness or resoluteness. Furthermore institutions are no exogenous constructs - they possess an endogenous character and reflect distributional struggles.

Theoretical Approaches for Explaining Reform in Weyland (1998)

Abbildung in dieser Leseprobe nicht enthalten

Classification and the Synthesis of Rational-Choice and Institutionalist Perspectives applied in the present dissertation (based amongst others on Shugart/Haggard 1997 and Haggard/Kaufman 1992b)

Abbildung in dieser Leseprobe nicht enthalten

2.1 The Rational-Choice Perspective

The basic and overwhelming notion of judging reform effects and causes is the insight that reforms are hardly ever Pareto-optimal.¹² Even in the best possible scenario with a minimum of transaction-costs and high societal consensus it can reach the quality of an overall positive- sum-game, but with some, individual losers which might not be compensated after all. Their pace and scope is determined by its short-term or long-term consequences for the different political and economic actors - and the perception of these consequences shapes incentives and aversions. Thus, from a certain point of view, cause and consequence of reform can be congruent and do not necessarily need to be analytically distinguished.

In the short term perspective theoretical considerations as well as empirical affirmations show that the losses or costs caused by a particular reform are frequently carried by a few members or groups of society (Olson 1968; Rodrik 1994), while the benefits accrue only in a diffuse manner or merely in a long-term perspective. Every system creates vested interests which are rooted in a status-quo situation and which will have to bear direct costs. On the other hand, not only are benefits and costs of reform unequally distributed, but also frequently invisible (Rodrik/Fernandez 1991). The consequence is that those, who oppose reform, gain leeway and become substantial veto-players. Thus reforms that are deemed beneficial can ultimately fail.

Veto-players find themselves confronted by the pro-reform-faction, amongst others the executive, the president or, for example, the finance ministries or central banks. If, in reverse and counterfactually, these costs were negligible and overall benefits highly specified and Pareto-optimal, there would not be any reform obstacles. There is an indirect way to reach Pareto-Optimality: by compensating the losers¹³. This thought can be object to political engineering - which will in fact represent an important ingredient of policy making in Brazil and be referred to in the later chapters.

Rational-Choice Approaches

Abbildung in dieser Leseprobe nicht enthalten

2.1.1 General Approaches

One branch of theoretical explanations thus refers in a rather mathematical sense to above- mentioned costs, which Bresser-Pereira/Abud (1997) refer to as transitional costs of reform. Transition costs (or total transition costs) are the cumulative costs that a country faces from the time economic distortions emerged, causing inflation and eventually an output gap, till the moment economic reforms allow the country to return to stability and growth (ibidem: 906). These costs can be subdivided in adjustment and procrastination costs. Whenever economic adjustment is pursued, particular groups and sections of society have to bear the costs of adjustment. Increased rents, for example, can lead to an increase in rent seeking activity, and a decrease in productive activity, in a common pool framework (Tornell/Lane 1999). Denying access to this common pool is equivalent to a setback of benefits and thus an elevation of costs (Braun 2007). Immediate public expenditure cuts and/or tax increases, and, eventually, temporary recession and unemployment belong to the most severe areas where adjustment takes place and which are felt not only by particular groups, but by society in general.

A rational-choice related concept with a higher degree of formalization refers to the Political Cost and Benefit Ratio (PCBR) elaborated by Dani Rodrik (1991). In most simple terms this relation juxtaposes political costs of reform in the numerator and political benefits, perceived as efficiency gains, in the denominator. It takes into account that each major reform has two impacts: a general, benefit-enhancing impact that accrues to the whole population as well as a genuine redistributive content that is only felt by a part of the population, in different degrees. The higher the redistributive content is, i.e. the higher a reshuffling of entitlements, and the lower the general benefits are, the more difficult a reform will be a priori and the higher the probability of reform reversion a posteriori.

2.1.2 Stages of Reform

Beyond general rational choice approaches there exists a certain array of phase-specific theories that attempt to explain actor behaviour in the different stages of reform. They thus provide a more dynamic view of the reform process which cannot be taken for static.

The “War of Attrition” - approach is one that focusses exclusively on the initiation stage of economic stabilizations and adjustments.

[...]

---

¹ The “Plano Crusado” was modelled after the “Plano Austral”, the first major heterodox inflation-targeting in Argentina, which equally failed and paved the way for the later reform efforts of the Meném - administration. Its major instrument was the freezing of prices and salaries in order to mitigate the inertial dimension of inflation. Due to its populist nature, its inconsistent execution and the distortion of prices it finally failed and resulted in a much higher annual inflation. The subsequent heterodox plans shared mainly this fate and lead in the long term to higher rather than lower inflation.

² The operational deficit can be also referred to as the Public Sector Borrowing Requirement (PSBR) and is defined - in opposite to the Nominal Deficit - as the deficit created by subtracting the expenses from income without considering inflation and money correction effects.

³ About the deficit-disguising effects of inflation, see Roubini/Sachs 1989.

⁴ The Oliveira-Tanzi-Effect occurs when during a period of high inflation the volume of tax collection declines relative to the expenses and thus contributes to a rising budget deficit. A reversed Oliveira-Tanzi-Effect occurs

⁵ “Custo Brasil” is a general term referring to overall structural, economic and bureaucratic difficulties that investors encounter in Brazil, restraining growth perspectives and increasing unemployment, the informal labor sector, capital flight and tax evasion.

⁶ Instead of forming a constituency in support of advancing reforms, short-term winners of reform may assemble to stall the economy in a partial reform equilibrium. One reason can frequently be found in the intention to generate concentrated rents for the short-term winners, while imposing high costs on the rest of society. A more common reason, however, is simply the fear that far-reaching and complete reforms might give a disadvantage to those who suffer relative losses. Promises of compensation might not be credible and later reversibility not be feasible.

⁷ Prominently figuring their first major essay “Constitutional Change, legislative Performance and Institutional Consolidation” (Figueiredo/Limongi 1995) which was followed by various other essays undermining their central argument of a predominance of the executive in the Brazilian policy-making process.

⁸ I chose to focus on the federal budget as well as the general fiscal relations because its state can be seen as a “meta-condition” of successful reforms in other areas. Expressed differently: the fiscal relations mirror the principal distributional issues in a political system. It further reflects more than anything else the dynamic (conflicts/cooperation) that take place between the different levels of government, i.e. between executive and legislative (horizontal) as well as the central government and the single states (vertical). Thus, according to Baumann (2001), high political consensus as well as cooperation between the executive and legislative get best reflected in relatively peaceful fiscal relations and moderate distributional conflicts.

⁹ The Social Emergency Fund - as shown in chapter 4 of the dissertation - was one of the principal instruments of the Brazilian executive to de-link a part these resources destined for sharing agreements between the federal and the state level. It disconnected 20 per cent of the central government revenue from constitutionally mandated spending (see Samuels 2003b).

¹⁰ A short overview about reform dynamics in Argentina and Venezuela in Tommasi/Velasco (1995).

¹¹ For more detailed information about the reform process in general in Latin-America see Singh et al. 2005: Stabilization and Reform in Latin America: A Macroeconomic Perspective on the Experience since the early 1990s, International Monetary Fund. This anthology comprises different aspects of reforming and provides an impressive array of examples from different countries. Also very recommendable is the paper by Karen L. Remmer (The Politics of Neoliberal Economic Reform in South America, in: Studies in Comparative International Development, Vol. 33: 3-29) which provides a rather critical approach. An ample overview of different theoretical approaches for explaining reform provides Kurt Weyland (The Politics of Neoliberal Reforms in Latin American Democracies: Argentina, Brazil, Peru and Venezuela, 1998).

¹² Pareto-optimality is a concept named after the Italian economist Vilfredo Pareto (1848 - 1923). It describes a situation in which not a single individual situation can be improved without worsening another one´s utility.

¹³ The notion of compensating losers of a policy change is better - and analytically more correct - described by the Caldor-Hicks-Criterium.