Can you revise & add more flesh (especially #4 & 7) to my assignment based on the attached reading? I will tell you the details of the assignment after I hire you.
Can you revise & add more flesh to my article template based on the readings?
Lee 2 Chaerin Lee (student ID#: 1120 19 0006) Introductory Seminar E-1 ILA Spring 2022 Article Analysis Template 2 Bibliographic citation (use APA style): Stephan, A. (2000). Brazil’s Decentralized Federalism: Bringing Government Closer to the Citizens?. Daedalus, 129(2), pp. 145-169. Article’s subject—what is the question, puzzle, or problem examined?: How does decentralization affect democracy, specifically Brazil’s decentralization movement? 3. Main Thesis Thesis statement (quote article): “Are there, for example, some types of decentralization, such as found embedded in Brazil’s 1988 federal Constitution, that worsen the quality of democratic politics rather than improve it? This is the question I explore in this essay” (Stephan 2000). Where thesis first appears (page and location): p. 145, the third paragraph of page 145 or the (almost) beginning of the introduction 4. Variables – (separate author/s thesis from alternative hypotheses)– Independent variable(s) (how are the concepts measured): Dependent variable (how are the concepts measured): 5. Method used (research design — see Lim) + method how they conducted the research) The Most Similar Systems (MSS) when comparing the U.S. and Brazil’s federalism in the section “The Control of Constitutionality and the Demos.” Stephan first acknowledges that the U.S. and Brazil politically hold many similarities; however, throughout the section, he demonstrates the differences in the details of their political system(e.g. Brazil does not implement the U.S.’s four principles of judicial review, such as “no writ of certiorari”), which confirms that Brazil’s government is more ‘demos-constraining’ than the U.S. one. 6. Findings – In 1988, Brazil’s new constitution declared the promotion of decentralization, which later backfired on the Cardoso regime. However, Cardoso’s expansion of primary education as a part of the decentralization movement was quite successful. – As mentioned above, the United States and Brazil politically share many similarities since they both employ federalism, which is originally more ‘demos-constraining’ than other democratic systems. Compared to the U.S. federalism, which is deemed as the “standard model of democratic federalism,” Brazil’s federalism is more ‘demos-constraining.’ This has caused some issues in Brazilian society. For example, since there is “no writ of certiorari (the right not to review a case)”, the Brazilian Supreme Court has a heavy workload although their decisions have ripple effects: “The Brazilian Supreme Court had to decide on 37,555 cases in 1997, and almost none of their decisions were generically binding” (pp. 152-153). Therefore, the Supreme Court holds less power to sanction the law. – Stephan (2000) acknowledges two possibilities of how decentralization would affect a country: (1) More citizens would be benefited from the government’s system and (2) the distribution of power would be unbalanced since decentralization could provide more power to local elites. (5. Quality of Article – if you can – don’t spend too much time here) A strength: A weakness:
Can you revise & add more flesh to my article template based on the readings?
Brazil’s Decentralized Federalism: Bringing Government Closer to the Citizens? Author(s): Alfred Stepan Source: Daedalus , Spring, 2000 , Vol. 129, No. 2, Brazil: The Burden of the past; The Promise of the Future (Spring, 2000), pp. 145-169 Published by: The MIT Press on behalf of American Academy of Arts & Sciences Stable URL: https://www.jstor.org/stable/20027632 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected] Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms American Academy of Arts & Sciences and The MIT Press are collaborating with JSTOR to digitize, preserve and extend access to Daedalus This content downloaded from 18.104.22.168 on Wed, 15 f on Thu, 01 Jan 1976 12:34:56 UTC All use subject to https://about.jstor.org/terms Alfred Step an Brazil’s Decentralized Federalism: Bringing Government Closer to the Citizens? INTRODUCTION Anew “Washington consensus” has been emerging, which says that increased decentralization is not only good for the economy but good for the politics of democratizing countries also. Why? Because it brings government closer to the people. Brazil’s 1988 Constitution mandated significantly expanded decentralization. In January of 1995, Fernando Henrique Cardoso, a world-famous social scientist and architect of the Real Plan that would reduce inflation from four digits to one digit in two years, was inaugurated as president after a sweep ing first-round victory. In January of 1999, Cardoso, again elected in the first round, began his second term. However, the extreme difficulties that the Cardoso administration has en countered in carrying out his reform agenda raises some serious questions about the wisdom of the Washington Consensus. Are there, for example, some types of decentralization, such as are found embedded in Brazil’s 1988 federal Constitution, that worsen the quality of democratic politics rather than im prove it? This is the question I explore in this essay. But first, let us go back to the origins of democracy. How do contempo rary federal systems relate, or not relate, to these origins? One of the earliest ideas of a democracy was that all the citizens of the polis, collectively deliberating together at one forum, were Alfred Stepan is Wallace Sayre Professor of Government at Columbia University. 145 This content downloaded from 22.214.171.124 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 146 Alfred Stepan the demos. As the number of citizens grew, this direct partici pation of the demos in law-making had to give way to the election of representatives. For the analytic purposes of the argument to be explored in this essay, I will assume that the more a single deliberative body is responsible for law-making in the polis, and the more that body approximates the principle of one-citizen one-vote, the less the demos is constrained. Of course, all modern democracies, correctly, are to some extent “demos-constraining.” In order to prevent the people or demos from passing majoritarian laws that could violate the rights of minorities, for example, almost all modern democra cies have a constitution that protects the fundamental rights of individual citizens and a supreme court with the constitutional prerogative to declare laws passed by the legislature null and void if, in that court’s judgment, the laws violate the constitu tion.1 There are further constitutional constraints on the citizens that modern democracies have found acceptable. Societies that are multinational in population, and very large (in territory and/or populationyf K D Y H Q R U P D O O R S W H G I R U D I H G H U D O V V W H P . Unitary systems have “open agendas”?that is, there is no area where the highest legislative body cannot make laws. Federal systems, by definition, must to some extent have “closed agen das”?that is, there are some areas where the central federal legislature cannot make laws because these law-making areas fall within the constitutionally guaranteed and exclusive legis lative competence of the subunits that make up the federation.2 By universal modern practice, as well, all modern federal de mocracies have an Upper Chamber to represent the people of the territorial subunits, as in the U.S. Senate. To the extent that the Upper Chamber participates to any significant extent in law-making, this has the potential to further constrain the demos. All democratic federal systems are thus more demos constraining than democratic unitary systems. This is widely understood. What is less widely understood is the immense variation that exists within democratic federal systems. One can in fact con struct a continuum that runs from the least to the most demos This content downloaded from 126.96.36.199 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 147 constraining federal systems. When we construct such a con tinuum, two things become clear. First, U.S. federalism, which is often depicted historically, politically, and analytically as the standard model of democratic federalism, is, empirically, an outlier, situated at the demos-constraining end of the con tinuum.3 Second, Brazilian federalism, while often seen as being patterned on U.S. federalism, is actually vastly more demos constraining than even U.S. federalism. Many of the difficulties faced by the reform government of President Fernando Henrique Cardoso (1995- yf P L J K W E H E H W W H r understood if we could get greater intellectual purchase on Brazil’s unique demos-constraining federalism. This essay will therefore address three broad aspects of it. First, I provide the evidence showing that Brazil has by far the most demos-con straining democratic federal system in the world. Second, I discuss briefly some of the historical factors involved in the emergence of such a constraining system of federation. And third, I explore how the Cardoso administration has attempted to make the political system less demos-constraining, less cen ter-disabling, and more demos-enabling. MODERN DEMOCRATIC FEDERATIONS: HOW DEMOS-CONSTRAINING ARE THEY? The evidence upon which I base my assertion that Brazil has the world’s most demos-constraining democratic federal system involves seven key factors. The first two factors involve the Upper Chamber, which I will occasionally call the “Territorial Chamber” because it represents the principle of territorial rep resentation. I will at times refer to the Lower Chamber as the “Demos Chamber,” because it represents in principle the one citizen one-vote ideal.4 The Overrepresentation of States in the Upper Chamber My attention to this factor may strike some readers as strange, because they may assume that the U.S. model, in which every subunit in a federation, however large or small, gets the same number of seats in the federal Upper Chamber, is the norm and This content downloaded from 188.8.131.52 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 148 Alfred Stepan therefore easily justified. Actually, the equal representation of the different states in the federation produces a system of massive over- and underrepresentation. California’s popula tion, for example, is sixty-six times greater than that of Wyo ming, yet both states get two U.S. senators, making one vote in Wyoming worth the same in producing a U.S. senator as sixty six votes in California. Brazil’s system of representation to the Upper or Territorial House is even more extreme. There, one vote in the sparsely populated state of Roraima has 144 times the weight in producing a Brazilian senator as does a vote cast in the densely populated state of S?o Paulo. There is nothing intrinsic to democratic federal systems that leads to such malapportionment; Austria, Belgium, and India, for example, all use formulas for the representation to the Upper Chamber that approximate the principle of proportional representation. If we use the Gini index of inequality to mea sure the principle of representation to the Upper or Territorial Chamber, based on a scale of 0 (perfect equality of representa tionyf W R D E V R O X W H L Q H T X D O L W ” R Q H V W D W H K D V D O O W K H V H D W V f, we find that Austria has a score of 0.05, Germany 0.32, and Brazil 0.52, making Brazil among the most unequal of all the federal systems.5 The Policy Scope of the Upper or Territorial Chamber The inequality built into representation in Brazil’s federal sys tem of government also involves the policy scope of the Terri torial Chamber. On grounds of democratic theory and practice, is there any prima facie reason to think that the frequently malapportioned Territorial Chamber should be given the same policy scope as the more proportional Demos Chamber? Or should its power be limited to those issues that relate directly to the relations between the center and the territorial subunits or states? Democratic federal systems vary greatly as to how they answer these questions, but, in general, Territorial Chambers have less legislative power than the Demos Chamber; this is the case in Germany, for example, while in Austria, Spain, Belgium, and India the Territorial Chambers have even less power. Brazil in theory follows the U.S. constitutional formula of relative power symmetry between the two chambers, but in fact This content downloaded from 184.108.40.206 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 149 Brazil’s Upper or Territorial Chamber, equivalent to the U.S. Senate (and also called the Senateyf K D V P R U H X Q L O D W H U D O S R Z H r to kill a bill passed by the Demos Chamber than in any other federal democracy.6 Though senators seldom find they have to exercise this prerogative (because their power is anticipatedyf , the Senate has many other exclusive competencies that are routinely exercised. Senators directly appoint two-thirds of the judges that review federal expenditures. They have the right to deny or confirm the other third. The Senate has exclusive competence to authorize international borrowing by the states. This means that even if the Central Bank says it is strongly opposed to a state contracting an international loan, the Senate can overrule the Central Bank and authorize the loan. The Senate has exclusive competence to approve the central administration’s foreign borrowing levels. In fact, in Brazil, there is no policy area that is beyond the policy-making compe tence of the Senate. Indeed, there are twelve policy areas that are the exclusive law-making prerogative of the Senate. In comparative terms, then, the Brazilian Senate is one of the most malapportioned territorial chambers in the democratic world. It also has the most disproportionate power vis-?-vis the chamber of the demos of any Upper Chamber in the democratic world. The consequences of this are, as we shall see, that a small group of senators have the power to put into the Brazilian Constitution many things that specially favor their states and to block any efforts at constitutional reform that might challenge or change their constitutional prerogatives. Deviation from Proportionality in the Lower Chamber In keeping with the demos principle of one-person one-vote in the Lower Chamber of federal systems, in the United States there is a national census every ten years in order to adjust the number of seats per state and redraw voting districts to reflect changes in population. In Brazil, however, the demos principle of one-person one-vote in the Lower Chamber is heavily diluted by the territorial principle. A unique “floor and ceiling” deci sion rule determines that no state, no matter how small in terms of population, can have fewer than eight representatives in the Lower Chamber, and that no state, no matter how large the This content downloaded from 220.127.116.11 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 150 Alfred Stepan population, can have more than seventy. This decision rule introduces yet another constraint on democracy, this time in the Lower Chamber. For example, the one-person one-vote rule would have given the states of Acre, Amap?, 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 70.7 The Amount of Powers Constitutionally Devolved to Federal Subunits The U.S. Constitution and federal laws are explicit on the right of the Center to retain important powers that relate to the single market of the entire country. For example, the U.S. Constitution forbids any state to impose duties or taxes on imports or exports without congressional approval and pre vents the states from entering into any treaty or alliance with other countries or extending their own bills of credit. Here, once again, Brazil’s federal constitution operates in a different direction, since it gives the individual states enormous scope to influence macroeconomic policies, thereby reducing the power of the democratically elected federal government to manage the economy for all the citizens.8 Virtually every state in Brazil by the early 1990s had, for example, at least one substantially autonomous state bank. Neither the Lower Cham ber nor the Central Bank had authoritative control over the subunits’ international borrowing activities. In effect, the states de facto had the right to print their own money (by the mecha nism of issuing huge bond offeringsyf $ Q H [ W U D R U G L Q D U Q X P E H r of spending rules, involving such things as the exact details of special tax schemes for regional (state-levelyf G H Y H O R S P H Q W S U R M H F W V , or regulations requiring fixed percentages of federal taxes to be handed over to the states and municipalities, or the amount of federal money that must be spent on irrigation in the North and Northeast, are furthermore specified in the 1988 Constitution as such, making them policy areas that lie beyond the scope of ordinary majority legislation. They are constitutionally embed ded and thus require exceptional majorities to change. This content downloaded from 18.104.22.168 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 151 Party-Systems as Potential Demos-Enabling Forces As I have said, all democratic federal systems are inherently more demos-constraining than democratic unitary systems. However, programmatic and disciplined political parties can play an important role in representing and empowering the citizens of the demos. The United States has a relatively loose two-party system in which the parties, by European standards, are neither very programmatic nor very disciplined.9 The Bra zilian party system is much looser in terms of polity-wide cohesion than that in the United States and therefore is not in itself a compensating factor that can help overcome the demos constraining factors I have mentioned so far, especially the power of groups of individual senators to block reforms. The party system existing in Brazil when Fernando Henrique Cardoso became president did not produce predictable, disciplined, and therefore relatively inexpensive (in terms of side-paymentsyf majorities at the Center.10 Enduring programmatic party coalitions are made even more difficult due to the constant movement from party to party of many legislators. In the federal Lower Chamber during the 1991-1994 session, for example, 51.7 percent of the total depu ties changed parties at least once, 18.1 percent to programmati cally noncontiguous parties.11 On the Pedersen Index of elec toral volatility, which measures the net change in the vote or the seat shares of all parties from one election to another, Brazil’s score from 1982-1994 was 33.0 percent. The United States from 1944-1994 was 4.0 percent. In this period, the only federal system in the world with a higher volatility than Brazil was semidemocratic Russia, with a volatility score of 54 per cent for 1993-1995.12 The Control of Constitutionality and the Demos Brazil has a unique combination of legal procedures concerning constitutional review that makes it easier for judges, at any level, to challenge the constitutionality of laws passed by the legislature than in any other federal democracy in the world. For example, in Switzerland there is no judicial review. In Germany, Spain, Belgium, and Austria there is a centralized This content downloaded from 22.214.171.124 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 152 Alfred Stepan system of constitutional review in which only a single body, the constitutional court, can declare a law passed by the federal legislature unconstitutional. The United States and Brazil have defused systems of consti tutional review; that is, judges, at all major levels of the judicial system, can challenge the constitutionality of laws passed by the federal legislature. However, the United States has four principles of judicial review that tend to limit the defused system’s challenge to the right of the federal legislature to pass laws that are considered binding. First, the Supreme Court in the United States has the right to control its own docket. It has the writ of certiori (the right not to review a caseyf 6 H F R Q G L I W K H 6 X S U H P e Court decides to review the constitutionality of a law and declares that the law is constitutional, the law is erga omnes (against everyoneyf 7 K L U G W K H 6 X S U H P H & R X U W V G H F L V L R Q L V V W D U e decisis (a binding precedentyf L Q D O O D Q D O R J R X V F D V H V I R U D O l judges. Fourth, there is a norm, the so-called political question doctrine, that means that inherently political issues, which could in theory be constitutionally settled by carefully crafted laws, should if at all possible be left to legislative and not judicial determination. None of these four principles is in effect in Brazil. The cumu lative result is extremely demos-constraining. Even if the legis lature passes a law, and the Supreme Court reviews a challenge and issues a judgment that the law is constitutional, the absence of the erga omnes and stare decicis principles means that the decision is only binding in the specific case under litigation. The next day, hundreds of plaintiffs, in virtually identical circum stances, can start constitutional challenges to prevent the same taxation law or the same agrarian reform expropriation law that was just declared constitutional from affecting them. Of ten, within a week, scores of lower court judges have issued injunctions preventing the law from going into effect against the plaintiffs in their court as long as litigation is in process.13 With no writ of certiori, most of the cases go all the way back to the Supreme Court. With no “political doctrine” norm, nor mal politics becomes “justicialized.” In the judgment of one of Brazil’s closest observers of this process, “the judicial system, This content downloaded from 126.96.36.199 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 153 especially the judiciary, has become a powerful resource for vetoing majority decisions made in the political sphere.”14 The combined result is a Brazilian Supreme Court that is overwhelmed and whose decisions have little weight. While the U.S. Supreme Court decided on only 87 cases in 1996-1997, the Brazilian Supreme Court had to decide on 37,555 cases in 1997, and almost none of their decisions were generically binding. Thus, even when the legislature passes a law, Brazil’s lower courts can continually create obstacles to the law’s implemen tation, and the Supreme Court can almost never sanction a law as the law of the land.15 Monopoly of the Legitimate Use of Force by the Government Most political scientists accept some version of Max Weber’s definition of an independent state?namely, that the state is the entity that is able to effectively make a claim to the monopoly over the legitimate use of force in the country. Historically, however, a great deal of this control over force has always been devolved in Brazil to the individual states. Though the Brazilian Constitution of 1891 was relatively silent about state militias and who controlled them, the exporting states, espe cially Sao Paulo, Minas Gerais, and Rio Grande do Sul, ben efited from their right to tax exports to finance virtual state armies. Sao Paulo, for example, tripled its revenues in the first year of the federation and quadrupled its military expenditures in three years. In the 1920s, Sao Paulo flaunted a military police force that normally outnumbered the federal army garrisoned in the state by a factor of ten, an air force that outnumbered that of the entire federal air force, and a French military mission.16 The new Brazilian Constitution of 1988, in Article 125, gives states the right to their own military police forces. Crimes committed by the force fall within the jurisdiction of that state’s military court system. Thus, when Fernando Henrique Cardoso became president of Brazil, crimes against human rights, even murders, committed by state military police forces could only be investigated by military police prosecutors and tried in mili tary courts from the same state where the alleged crime oc curred. Many egregious violations of human rights, including This content downloaded from 188.8.131.52 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 154 Alfred Stepan virtual massacres, were committed with de facto impunity be cause federal law enforcement officers or judges were power less to intervene. THE ORIGINS OF BRAZIL’S DEMOS-CONSTRAINING FEDERAL SYSTEM The dominant theoretical model for explaining the origins of federations is William H. Riker’s.17 Riker believed that all enduring federations come into being via the same path, when a variety of political communities, each with a great deal of sovereignty, believe that they could increase their security, and possibly the size and power of their markets, if they entered into a voluntary bargain in which they pooled as much of their sovereignty, but no more than is necessary, to achieve these goals. We might call this a “coming together” federal system. Riker added a second argument to his first: namely, that the interests of the individuals (he often calls them “tastes”yf L Q W K e federation are more important than federal institutions (which he calls “congealed tastes”yf L Q G H W H U P L Q L Q J S R O L F L H V $ F F R U G L Q g to Riker, if the majority of individuals in a polity want to change their institutions, then these institutions will, in not too much time, become uncongealed, because they will reflect the new tastes of the majority of individuals. Both of these argu ments are misleading and make it particularly difficult to un derstand Brazil. The path to federalism in the United States in 1787 approxi mated the Rikerian path, and it is the example he had most in mind when he proposed his theory of federalism. With some differences, Switzerland in 1848 and Australia in 1901 also approximated this path. But there are in fact many more paths to federalism than the bargaining path Riker described, paths that can generate dif ferent outcomes in federal institutions.18 Brazil’s path to feder alism is normally seen as following the U.S. path; however, there are some important differences we should note. First, there was no “coming together” of previously sovereign politi cal communities. The previous political community was a uni tary state, the Brazilian empire under an emperor. Second, This content downloaded from 184.108.40.206 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 155 there was no “bargain” between the future members to create the federation. The military, encouraged by some economic and political leaders mainly from Sao Paulo, executed a coup, set up a military-led provisional government, and declared Brazil a federal republic. Third, it probably strains language to say the members joined the federation “voluntarily.” The Proclamation of the Republic explicitly stated that one of the extraordinary powers of the provisional government was to “defend the integ rity of the country.” Fourth, unlike the U.S. experience, one powerful state and its ally was able to gain great control over the constitution-making and state-building processes. Let us explore this fourth point. During the constitution-making period in 1890-1891, the military began to divide somewhat, and one economically pow erful and politically cohesive state, Sao Paulo, together with its powerful ally, the state of Minas Gerais, was able to control much of the constitutional agenda. As Brazil’s leading export state and a state that wanted to subsidize immigration, Sao Paulo wanted, and got, a constitution that allowed states to tax exports and to negotiate, with Senate approval, international agreements.19 It also wanted, and got, a constitution that did not say too much about the supremacy of federal law over state law and was not too explicit about the control of state militias. The bargain that the state of Sao Paulo, the emerging hege monic power, together with its allies, offered the other subunits was that all members of the federation would receive the same extreme rights of self-management (but without the financial and military resources to protect these rightsyf , Q D G G L W L R Q H D F h subunit would get the same amount of seats as the hegemonic subunit in the territorial chamber of the weak center. The outcome was a federal system with very strong prerogatives for the member states.20 In the beginning, it was a system dominated by a few powerful states. The president of the first constituent assembly of the federal republic in 1891 was from Sao Paulo. The first three popularly elected presidents of the new federal republic were from Sao Paulo. The ministers of finance and justice who put in place a reinforcing network of structures were from S?o Paulo. Significantly, the first minister of justice in the new federation, the Paulista Campo Salles, This content downloaded from 220.127.116.11 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 156 Alfred Stepan “dissolved the commission appointed to draw up a civil code because he considered this a function of the state rather than the federal government.”21 Ironically, this federal constitutional legacy of strong states’ rights, together with authoritarian in terludes, eventually was not only demos-constraining, but S?o Paulo-constraining. This brings us to the question of the unusu ally strong “stickiness” of federal institutions. Let us examine Riker’s second major argument, that institu tions are only the congealed “tastes” of individuals and thus ultimately melt before the new demands of the majority. The entire point of this essay, of course, is that the key political feature of federations is that they can create decision rules whereby minorities can block majorities. From a theoretical perspective, the hardest rules to change are those that give some specific benefits to a minority whose vote is required to change the rules. Changing the rules of a demos-constraining federation is one example. The situation in the Brazilian federation has been further complicated by two additional factors. First, each of the three major Constitutions of 1891, 1946, and 1988 was put together in a democratizing atmosphere following a long period of cen tralized and nondemocratic rule; the spirit of the times in all three cases led to reinforcing states’ rights as a democratic demand. Second, participation in the constituent assemblies of 1946 and 1988 was largely based on seat-allocation rules crafted by the immediately preceding authoritarian regime for its own purposes. These two apparently contradictory factors fused to make Brazil’s Constitutions of 1946 and (especiallyyf H x tremely demos-constraining.22 The Brazilian military regime that came to power in 1964 did not allow a direct election for president until 1989. However, unlike the military regimes of the period in Chile, Argentina, and Uruguay, the Brazilian military regime allowed the Con gress to stay open except for brief periods. To maintain control of the political system, via an elected Congress, the military regime intensified the demos-constraining features of the fed eration. In 1978 the military regime created the state of Mato Grosso do Sul, and, in 1982, another, Rond?nia, out of rela This content downloaded from 18.104.22.168 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 157 tively underpopulated and underdeveloped areas in the North and Center-West. Federal subsidies and a strong military pres ence made these states much more predictably pro-government than those in the more developed South, where civil society organizations, trade unions, and opposition parties were be coming increasingly strong. In the South, the military also managed to eliminate an opposition state by fusing the states of Guanabara and Rio de Janeiro into a single state called Rio de Janeiro. As a result of the creation of small states in the North, twelve new federal deputy and six new federal senator posts were created. Owing to the fusion by the military of two states in the South, the developed states eventually lost three senators. The military also altered the formula for new federal representation to the lower and more democratic chamber, causing a further underrepresentation of the population from the larger states.23 Representation at the constituent assembly of 1986-1988 to a great extent followed these new military-crafted faits accom plis. Thus, the states of the North, Northeast, and Center-West, which together only represented about 40 percent of the popu lation?the demos, if you will?had 52 percent of the votes at the constituent assembly.24 In this context, there was absolutely no effort in the constitu ent assembly to alter the decision rule that every state, no matter how small or large, would receive three senators. Like wise, no prerogatives were taken away from the Territorial Chamber, and some were added. There was also little discus sion by members of the Lower Chamber (the possible one citizen, one-vote chamberyf D E R X W W K H H O L P L Q D W L R Q R U D W O H D V t diminution, of the decision rule giving a minimum of eight deputies to each state. With their 52 percent of seats in the constituent assembly, the states from the North, Northeast, and Center-West voted, al most as a block, to admit three new thinly populated states from their region: Tocantins, Roraima, and Amap?. In the federal elections of 1990 this block of states, with 43 percent of the population of Brazil, controlled 74 percent of the seats in the federal Senate.25 This content downloaded from 22.214.171.124 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 158 Alfred Stepan The other major effect of Brazil’s constitutions being drafted after long periods of authoritarian governments is that the constitution makers, in their desire not to have another non democratic authoritarian government, engaged in a political process that unwittingly intensified the demos-constraining fea tures of the constitution. In the euphoric democratization pe riod of 1987-1988, all groups of the new civil society (as well as many of the old interest groupsyf Z H Q W L Q W R J U H D W F R Q V W L W u tional detail about the rights and entitlements of citizens and groups. The basic human rights provisions of the 1988 Consti tution were considered by many citizens, correctly, to be the culminating achievement of the democratization movement. However, many aspects of public life that are not basic human rights but politically captured entitlements, or entitle ments granted by authoritarian regimes for their own purposes, such as decisions about the exact percentage of public money to be allocated to the states and municipalities, the very generous pension rules for senior government officials, the specific re quirements for state ownership, and the detailed articles about where the central government should create irrigation projects, were also constitutionally embedded. From the viewpoint of democratic theory and practice, everything that is constitution ally embedded is removed from normal democratic majority voting procedures. Exceptional majorities are required to change these procedures. This means that minorities that want to main tain the status quo are given enhanced leverage. The blocking power of constitutionally entrenched minorities has been one of the elements that has most impeded the reform agenda of President Fernando Henrique Cardoso. If we stay with our concept of the demos, in a presidential system with free and universal suffrage a key representative of the demos could be the directly elected president. In Brazil during the military regime, direct elections for the president that were scheduled to be held in 1965, 1970, 1975, 1980, and 1985 were canceled. However, direct elections for governors (last held in 1965yf Z H U H K H O G D J D L Q L Q 7 K H R S S R V L W L R n captured governorships in many key states such as S?o Paulo, Minas Gerais, and Rio de Janeiro in that year. These directly This content downloaded from 126.96.36.199 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 159 elected governors provided the platforms (and the police pro tectionyf I R U W K H P D M R U R S S R V L W L R Q U D O O L H V R I F L Y L O V R F L H W , Q 4 they spearheaded the massive and popular Diretas J? demon strations in support of the direct elections amendment. Gover nors of states thus had great moral and political weight in the constituent assembly. In contrast, once again, the interests of the Center were not strongly represented in the constituent assembly, mainly be cause there had been no directly elected president since 1960. Making matters worse, Jos? Sarney, a traditional politician from the northeastern state of Maranh?o, became the “acci dental president” upon the sudden death of the much more popular?and historically legitimate?figure of Tancredo Neves, from Minas Gerais. In this context, both new (and oldyf F L Y L O D Q G S R O L W L F D O V R F L H t ies, for their own reasons, effectively championed the idea that the more power was devolved to the states and municipalities, the more democratic Brazil would be. In reality, of course, the decentralized constitution, which transferred a significant amount of Brazil’s total federal tax revenues from the Center to the states and municipalities, served many of the governor’s politi cal, financial, and tax interests extremely well. MAKING BRAZIL’S FEDERALISM LESS DEMOS-CONSTRAINING Here I am concerned with what the democratically elected president in Brazil’s federal system can do to garner more power at the Center and make it less constraining on the democratic needs of its citizens. There is a school of thought inside and outside of Brazil that argues that, no matter what the rights of the states or the Senate, ultimately the president can effect the policies he wants. This school of thought is based on the fact that President Fernando Henrique Cardoso is seldom blocked from implementing new measures, because 1yf K H X V H s presidential decree powers (the medida provisoria, or provi sional measuresyf D Q G f most of the non-decree legislation he formally proposed to Congress actually passed. I do not chal lenge any of the data in the meticulous and valuable study by This content downloaded from 188.8.131.52 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 160 Alfred Stepan Angelina Figueiredo and Fernando Limongi that pioneered this thesis.26 However, three observations are in order. First, decree powers do not cover constitutional amendments. Some of the most critical measures that involved constitutional reform, such as social security reforms and administrative re forms, were passed, but in “watered down” forms, and were still being “revised” or subjected to hundreds of constitutional challenges in late 1999. Second, the fact that President Cardoso, who would like to consolidate the institutions of democracy in Brazil, has had to pass so much legislation that was widely supported in the country and was crucial for the efficacy of government via decrees has been an unfortunate way to advance the demo cratic values of liberty, equality, and efficacy. Under a less able and democratically committed president, the demos-constrain ing element of Brazilian federalism might contribute to what Guillermo O’Donnell calls “delegative democracy” of the sort found in Fujimori’s Peru.27 My third point revolves around a question: how does one measure things that do not happen because they have no chance of passing? Political leaders only have so much political capital and resources; they also know how to count. If a powerful minority win-set opposes many of their preferred policy-pro posals, they will be parsimonious in the measures they will attempt to get by this formidable blocking win-set. From this perspective, which is more important: the fact that most of the measures that the president has proposed to the Congress actu ally passed, or the fact that many of the measures that the president would like to have passed in the first five years of his presidency were never put forward at all, because he believed that they would never pass? Based on my study of the policy proposals written before the Cardoso administration assumed office, on interviews with various cabinet ministers at the be ginning of the Cardoso administration about their plans, and on the fact that many of these plans never came to Congress, I believe that the second interpretation is the more politically significant. I have argued, contra Riker, that federal institutions do not simply represent the congealed tastes of the majority. If this This content downloaded from 184.108.40.206 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 161 were so, a president like Fernando Henrique Cardoso, who has been elected twice and who has in the past had approval ratings of over 60 percent, would not have been so repeatedly con strained by win-sets of minorities. However, it may be that the high point of Brazil’s demos constraining federalism has passed, as a consequence of actions the Cardoso administration has been able to take. Here I assess the changes in relation to each of the seven demos-constraining factors with which I began this essay. Powers Constitutionally Devolved to States The most significant changes in these areas have been in the de facto, more than the de jure, prerogatives of the states to make policies, especially the capacities of the states to carry out policies that impede the federal government’s ability to imple ment a reasonable macroeconomic policy. Cardoso became president largely because of his success in reducing Brazil’s annualized inflation rate of over 2000 percent. The Real Plan began officially on July 1, 1994, and by mid-1996 the inflation rate was down to less than 10 percent a year. One of the consequences of this dramatic reduction in the inflation rate was that state banks could not pass off their huge debts on unsound loans via inflationary measures. Virtually all the state banks were facing bankruptcy by 1995; indeed, two of the largest and most abused state banks, S?o Paulo’s Banespa and Rio de Janeiro’s Banerj, saw intervention by the Central Bank on December 30, 1994, two days before Fernando Henrique Cardoso assumed office. One of the main reasons for the exist ence of many unsound loans in the state banks was that the most politically powerful regulator and the most politically powerful borrower had normally been the same person?the state governor.28 In the midst of this impending bankruptcy crisis of the state banks, the federal government became the liquidator of last resort. The state banks have now virtually lost their capacity to issue bonds as de facto money. But the cost of achieving control over Brazil’s money supply has been great for the Center. As of the end of 1999, the Cardoso government had spent at least 50 billion dollars to prepare state banks for privatization or liqui This content downloaded from 220.127.116.11 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 162 Alfred Stepan dation.29 In an effort to make the states’ debt crises more manageable, the federal minister of finance and the Central Bank helped restructure state debts (many to private banksyf L n such a way as to stretch out debt repayments over thirty years, at an interest rate of 6 percent, with a repayment schedule to the federal government. However, opposition governors, when they took office in January of 1999, complained that they could not meet even these generous terms. Indeed, the declaration by the governor of Minas Gerais, Itamar Franco, of a moratorium on his state’s payments to the federal government greatly con tributed to the financial crisis of January of 1999. The Center has thus acquired important, but continually contested, new leverage vis-?-vis the states through its role as the stabilizer of state accounts. Nevertheless, the Senate will remain a major potential veto player in areas concerning state finances because of continuing Senate prerogatives in this area and the politically close connection between senators and gov ernors.30 Monopoly of Legitimate Use of Force Many reformers inside and outside the Cardoso administration would like to curtail the relative autonomy of the militarized state police forces that often commit human rights abuses and kill citizens with impunity. For example, during a peaceful march in Para toward the state capital in April of 1996, nine teen members of the Movement of Landless Rural Workers (MSTyf Z H U H V O D X J K W H U H G D W ( O G R U D G R G R V & D U D M ” V E W K H J X Q I L U e of the military state police force of Para. Four years later, no one has been convicted. Why? Partly because during the mili tary regime each state’s military police forces were put under the legal authority of state military courts. Owing to reforms championed by the Cardoso administration, the trial eventually went to a civilian jury, but most of the evidence collected (and not collectedyf Z D V X Q G H U W K H L Q L W L D O D X W K R U L W R I W K H 3 D U D P L O i tary state police. The judge, in a controversial 1999 decision in the state capital of Para, acquitted the three top officials in volved in the massacre. Many reformers now want to fuse the civil police and the military police and in essence put them under civilian control This content downloaded from 18.104.22.168 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 163 and civilian courts. There is also some discussion of giving some of their functions to an expanded federal police force, along French lines. But all these reforms would require constitutional changes, because the prerogatives of the state military police forces are constitutionally embedded. In 1996, after the Para massacre, President Cardoso proposed strong legislation that would have begun serious reform of the state military police forces. The Chamber of Deputies passed the president’s propos als. However, the Senate, under great pressure from governors, who in turn were under great pressure from their state military police force lobbies, gutted the bill. Two reforms survived the Senate veto. First, particularly egregious crimes involving loss of life can now be judged in civilian courts. Second, massive violations against human rights became a federal crime that can be tried in federal courts.31 The former minister of justice, Nelson Jobim, whose bill was de feated in the Senate, believes that, given Senate resistance to major reforms of the state military police, “we do not have the political conditions to take them out of the hands of the states.”32 The Court System and the Constitutionality of Laws Five years into the Cardoso administration, Brazil continues with its mixed system of constitutional review, which has strong demos-constraining effects. However, there is growing pres sure to introduce the principle of “binding precedent.” Malapportionment and Broad Policy Scope in the Federal Territorial Chamber One of the key initiatives of the Cardoso government in his second administration was to have been political reform. Sena tor Machado, with encouragement from the Cardoso adminis tration, prepared a major report about proposed political re forms. It is significant that in the entire report there was not one word about decreasing the malapportionment of the Senate or reducing its prerogatives.33 Unfortunately, given the theoretical and political arguments advanced previously in this essay, such political untouchability is precisely what one should have ex pected. This content downloaded from 22.214.171.124 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 164 Alfred Stepan Decreasing Malapportionment in the Demos Chamber There is a proposal being discussed to decrease malapportionment somewhat in the Lower House by lowering the “floor” from eight to four and raising the “ceiling” from seventy to ninety. This would still give Brazil a substantially more malapportioned Demos Chamber than Austria, the United States, West Ger many, or India, but would, if passed, be an improvement. How ever, in August of 1999, the College of Party Leaders of the Lower Chamber (most of whom would lose some seats if such a bill were passedyf Y R W H G W R S L J H R Q K R O H W K H S U R S R V D O X Q W L l further notice. Incentives for Fewer and More Disciplined Parties Despite Brazil’s incentives for party proliferation and loose party discipline, the Brazilian federal legislature of 1994-1998 had somewhat fewer parties than in 1990-1994. Also, party switching by federal deputies declined somewhat and party discipline increased somewhat in this period. However, the original Machado Commission proposals to introduce mea sures to constrain party proliferation have also, in essence, been shelved until at least 2003. CONCLUSION: MAKING DECENTRALIZATION WORK This brief assessment of Brazil should be read as a cautionary note about decentralizing reforms in new democracies. Before rushing to espouse decentralization and federalism for their own sakes, we should be aware that decentralization raises not one but two possibilities. The first possibility is that decentrali zation will indeed bring government closer to the people who are the users of the government’s services. The second possibil ity is that decentralization will devolve even more power and resources to local elites. If poverty alleviation is the central objective in the country, does it make more sense to devolve federal money to poor states permanently, or for the Demos Chamber to design temporary programs that directly help the life chances of poor people? This distinction becomes all the more salient if there is evidence that the poorer states have This content downloaded from 126.96.36.199 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 165 particularly unequal income distribution and/or particularly oligarchic local elites. In this regard, Cardoso’s experiments in crafting new forms of political participation are particularly interesting, especially in the area of primary education, considered by many social commentators to be the most important key to improving citi zens’ income and quality of life. The Cardoso government’s experiments in primary education are probably the administration’s greatest success in decen tralization. A number of practices are at work in new programs of primary education that bear special analysis. First, there is a conscious effort on the part of the central government to be partners, with some oversight role, in the decentralization pro cess. The central government becomes a partner by giving extra federal budget allocations to the experiment. However, federal money is only released after new participatory struc tures actually involving the users (in this case parents of chil dren in the local schoolyf K D Y H E H H Q V H W X S 7 K H H [ W U D I H G H U D l money goes directly to the account of the committee that has been elected by the users. The decision on how best to spend this extra money is made at a public meeting, often involving teachers, local officials, older students, and parents. At another public meeting sometime later, the committee explains, and often shows, how the agreed-upon expenditures were made. This set of new structures and practices involved in the decen tralization initiative entails three fundamental aspects of de mocracy. The election of the users committee involves partici pation. Decision-making at public meetings helps ensure that these commitments have transparency. This means that at the follow-up meetings the elected committee must have account ability. Much of the overrepresentation of the states of the North east, North, and Center-West in the Brazilian federal legisla ture has traditionally been justified on the grounds that they are poorer than the underrepresented states of the South and South east. But there is evidence that many of the states that are overrepresented in the federal legislature are precisely those states with particularly unequal income distribution and strong traditions of local oligarchic control. The Cardoso government’s This content downloaded from 188.8.131.52 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 166 Alfred Stepan primary school reform is a nice response to this seemingly intractable problem. The basic premise of the reform is that the federal government will allocate resources to ensure that any child who attends a primary school in Brazil, regardless of whether the child lives in a rich or poor state or comes from a rich or poor family, will go to a school that spends a fixed minimum sum, per child enrolled, on teachers. Most of the school districts in the Northeast are below this national stan dard. Virtually all the school districts in the South and South east are above this standard. Thus, the policy designed to improve poor citizens’ life chances does just that. But the policy also has the indirect effect of transferring resources to poorer regions without that money being spent by local elites in ways that would concentrate, rather than deconcentrate, power, re sources, and income. Decentralized federal systems in democracies are more “con gealed,” and further away from the control of the majority of the citizens, than most proponents of the new Washington Consensus imagine or would like. Precisely because of this, both Brazil’s unexpectedly intransigent, constitutionally em bedded obstacles and the country’s under-observed successes in decentralized primary education deserve increasing attention. ACKNOWLEDGMENTS I would like to thank the Ford Foundation for their grant support, Professor David Fleischer for many suggestions, and Arthur Costa, a Ph.D. candidate at the University of Brasilia, for his research assistance. ENDNOTES Tn some countries the court with these prerogatives is called the Constitutional Court. 2See Robert A. Dahl, “Federalism and the Democratic Process,” in Dahl, Democ racy, Identity and Equality (Oslo: Norwegian University Press, 1986yf 4 126. Citation from ibid., 114. 3For evidence of this assertion, see Alfred Stepan, “Federalism and Democracy: Beyond the U.S. Model,” Journal of Democracy 10 (October 1999yf . This content downloaded from 184.108.40.206 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 167 4In fact, virtually no Lower House is perfectly proportioned. 5I present comparative data on this and other indicators for all the democratic federal systems in the world in Alfred Stepan, “Toward a New Comparative Politics of Federalism, (Multiyf 1 D W L R Q D O L V P D Q G ‘ H P R F U D F % H R Q G 5 L N H U L D n Federalism,” in Stepan, Arguing Comparative Politics (Oxford and New York: Oxford University Press, forthcoming, Summer 2000yf 7 K L V L V S D U W R I a longer book-length work in progress with Juan J. Linz, Federalism, Democ racy and Nation. 6If the Brazilian Senate does not amend a bill but turns it down entirely, the bill is dead for that legislative session. 7For an excellent analysis of the historical evolution of malapportionment in Brazil’s lower federal house, see Jairo Marconi Nicolau, “As Distorc?es na Representa?ao dos Estados na C?mara dos Deputados Brasileira,” Dados 40 (3yf f: 441-464. 8Article 9 in the 1891 Constitution of Brazil gave the states the exclusive right to tax exports. Articles 48 and 65 gave the states the right to negotiate interna tional treaties as long as the federal Territorial Chamber approved. The more authoritarian constitutions created in Vargas’s Estado Novo (1937-1945yf , and during military rule (1964-1985yf J D Y H W K H V W D W H V O H V V S R Z H U L Q P D F U R H F o nomic areas. 9Nelson Polsby calls the U.S. party system 100 state parties flying two national flags. 10On the Laakso/Taagepera Index measuring party fragmentation, or more ex actly the weighted number of parties in the legislature, Brazil in 1992 had 8.5 parties in the Lower Chamber, higher than any democratic federal system in the world. The mid-1980s scores on the same index for the other important democratic federal systems in the world were 1.9 for the United States, 2.0 for Canada, 2.4 for Austria, 2.5 for Australia, 3.2 for West Germany, 3.7 for Spain, and 5.4 for Switzerland. India was 2.1 in the early 1980s, but is much higher now. 11 See Table 5.3 in Scott P. Mainwaring, Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil (Stanford: Stanford University Press, 1999yf . 12Table 2.1, ibid., 29. 13In Brazil this is called the industria de liminares (the industry of injunctionsyf . 14Rog?rio Bastos Arantes, “The Judiciary, Democracy and Economic Policy in Brazil,” paper prepared for the World Congress of the International Political Science Association, 17-21 August 1997, Seoul, Korea. 15The above discussion is partly based on my interviews with two Brazilian Su preme Court Justices, Minister Moreira Alves and Minister Nelson Jobim, in their chambers in Brasilia in December of 1998. Rog?rio Bastos Arantes puts the Brazilian treatment of the judicial control of the constitutionality of laws in comparative perspective in his Judiciario Pol?tica no Brasil (S?o Paulo: Editora Sumar?, 1997yf . This content downloaded from 220.127.116.11 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms 168 Alfred Stepan 16See Joseph Love, S?o Paulo in the Brazilian Federation: 1889-1937 (Stanford: Stanford University Press, 1980yf $ S S H Q G L [ & D Q G $ O I U H G 6 W H S D Q 7 K H 0 L O i tary in Politics: Changing Patterns in Brazil (Princeton: Princeton University Press, 1971yf . 17His two most general and cited works on federalism are William H. Riker, Fed eralism: Origin, Operation, Significance (Boston: Little, Brown, 1964yf D Q d Riker, “Federalism,” in Fred I. Greenstein and Nelson W. Polsby, eds., Hand hook of Political Science, vol. 5 (Reading, Mass.: Addison-Wesley, 1975yf 3 172. 18One such path is what we might call a “holding together” federalism. A previ ously unified state may decide that the only way to hold its multicultural, multilingual, and perhaps multinational population together democratically is to constitutionally devolve power to a federation while retaining significant power at the Center. Federal constitution-making in India in 1949, Spain in 1978, and Belgium in 1993 approximated this path. The United Kingdom is a possible candidate for such a path to federalism in the future. 19S?o Paulo did not necessarily consider secession and independence indispens able. It expected to be able to control the weak federal Center and did not in any case want to upset the federal military needlessly. 20As early as 1870 the motto of some Republicans was “Centralization= Dismemberment, Federalism=Unity.” While S?o Paulo’s revenues tripled in the first two years of the federation, the Center’s revenues decreased by 30 percent. 21Emilia Viotti da Costa, The Brazilian Empire: Myths and Histories (Chicago: The University of Chicago Press, 1985yf . 22Get?lio Vargas ruled without elections in his Estado Novo of 1937-1945, but after Brazil’s participation against Germany in World War II, he prepared for upcoming elections by raising the “floor” to seven in small states because his political agents, who were forming the new PSD party, controlled most of these states. For a painstaking analysis of the manipulation of over- and underrepresentation by Vargas and his allies, see Maria do Carmo Campello de Souza, Estado e Partidos Pol?ticos no Brasil (1930 a 1964yf 6 ” R 3 D X O R : Editora Alpha-Omega, 1976yf . 23For the military regime’s manipulation of the electoral system (manipulation that often backfiredyf V H H ‘ D Y L G ) O H L V F K H U 0 D Q L S X O D W H V F D V X ” V W L F D V G o sistema eleitoral durante o per?odo militar, ou como usualmente o feiti?o se voltava contra o feiticeiro,” in Gla?cio Soares and Maria Celina D’Ar?jo, eds., 21 A?os de Regime Militar: Balan?os e Perspectivas (Rio de Janeiro: Funda??o Get?lio Vargas, 1994yf . 24See the well-documented and richly detailed study by Ana Luiza Backes, Democracia e Sohre-Representa?ao de Regioes: O Papel do Senado, disserta??o de Mestrado en Ciencia Pol?tica, Universidade de Brasilia, 1998, 8. 25Ibid., 128. 26For an impressive argumentation and documentation of this position, see Angelina Cheibub Figueiredo and Fernando Limongi, “Medidas Provisorias: Abdica??o ou Delega?ao?” Novos Estudos 47 (1997yf . This content downloaded from 18.104.22.168 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms Brazil’s Decentralized Federalism 169 27See Guillermo O’Donnell, “Delegative Democracy,” Journal of Democracy 5 (1yf f: 55-69. 28For the extraordinary financial and political prerogatives of governors in this period, see the pioneering book by Fernando Luiz Abr?cio, Os Bar?es da Federa??o: Os Governadores e a Redemocratiza??o Brasile?a (S?o Paulo: Editora Hucitec, Departamento de Ciencia Pol?tica, USP, 1998yf . 29For an extremely detailed analysis and documentation of the costs for the Cen ter of “bailing in” the state banks, see Harry M. Makler, “Bank Transforma tion and Privatization in Brazil: Financial Federalism and Some Lessons about Bank Privatization,” Quarterly Review of Economics and Finance 40 (1yf (April 2000yf . 30In a private conversation, David Fleischer (professor of political science at the University of Brasiliayf H V W L P D W H G W K D W U R X J K O S H U F H Q W R I % U D ] L O L D Q I H G H U D l senators have been governors (many more than onceyf D Q G W K D W P D Q R I W K e senators aspire to be governors. Almost no U.S. senators were once governors and aspire to return as governors. 31Interviews with Jos? Alfonso da Silva, director of public security, state of S?o Paulo, and Nelson Jobim, former minister of justice, 11 December 1998, Brasilia. 32Nelson Jobim, ibid. 33See Senador Sergio Machado, Relat?rio Final da Comissao Temporaria Interna Encanegada de Estudar a Reforma Pol?tico Partidaria (Brasilia: Fed eral Senate, 11 November 1998yf . This content downloaded from 22.214.171.124 on Wed, 15 Jun 2022 02:08:26 UTC All use subject to https://about.jstor.org/terms