The Flight of the Beetle:
Party Politics and the Decision-Making Process in the Cardoso Government

Eduardo Graeff
Sociologist, Master in Political Science from the University of São Paulo
Special Advisor to the Cabinet of the President of the Republic

The points of view expressed in this text are the exclusive responsibility of the author, not reflecting, directly
or indirectly, positions of the Brazilian government.

Paper prepared for the
V Congress of the Brazilian Studies Association
Recife, Brazil
June 2000

Translated by Ted Goertzel from the Portuguese original.

The beetle is an affront to the principles of engineering.  With its heavy carapace and peculiar wings, it should never been able to fly, but it does.  Awkwardly, but it flies.

Brazil in the 1990s is a case of the flight of the beetle.  With all its economic and social problems and the proverbial imperfection of its political institutions, it would appear condemned to flail its wings helplessly until it collapsed in chronic ungovernability.    Nevertheless, beginning in 1994, it has managed to pull itself together politically, control in inflation, and carry out a reform agenda which compares favorably with those of other "emerging" countries.

Both Brazilian political scientists and foreign Brazilianists have described the poor functioning of Brazil's political institutions and its governability problems.1 What follows, in general lines, is their diagnosis, with a few touches of personal observation.

The political conditions which made the success of the Real Plan and the advances of the Cardoso government's reform agenda have not been so fully explored, I suppose because of the lack of adequate time for academic analysis.  Here I offer a contribution towards this analysis, more from the perspective of a participant observer than that of a rusty political scientist.


The Brazilian party system is very fragmented.   More than a dozen parties have significant political and electoral weight.  The three largest - PFL, PMDB and PSDB - each has no more than one-fifth of the Chamber of Deputies (see the Table).  In addition to being fragmented, the system is characterized by a low level of institutionalization:  the parties generally have little influence over the behavior of either their supporters or their elected representatives, especially on the national level.

Composition of the Chamber of Deputies in May, 2000
Government: 389 76%
PFL 105 21%
PSDB 103 20%
PMDB 97 19%
PPB 46 10%
PTB 24 5%
Others 14 3%
Opposition: 124 24%
PT 61 12%
PDT 21 4%
Others 42 8%
TOTAL 513 100%

The Brazilian party system is weak, first of all, because it is still developing:  it has not had time to root itself in society.  Brazil urbanized and industrialized very rapidly in the last century.  The formation of modern parties, identified with the urban electorate, was repeatedly cut off by authoritarian interventions.  The Vargas dictatorship kept Congress closed from 1937 to 1946.  The military government, from 1964 to 1985, did not close Congress but they controlled it by canceling the mandates of representatives, opening and closing parties, reshuffling party names, and twisting electoral rules.

Other factors coincided to retard the consolidation of parties and limit their institutional weight.  Brazil is a federation by law and in fact, significantly decentralized and marked by sharp regional disparities.  Local interests often carry more weight in policy than do the kinds of ideological and class divisions that articulate the great European political parties.  An individualist and patrimonialist tradition views the electoral mandate more as a personal sinecure rather than as a public function.  Electoral legislation and party rules privilege individualism, denying the parties effective means to control the action of their elected representatives.

The system of proportional representation for the Chamber of Deputies has had an especially disruptive effect on party life.   In the system of open lists adopted in Brazil, the number of votes obtained by individual candidates determines their rank in the final list of candidates elected by the party.  This makes the competition between candidates of the same party fiercer than competition between parties.  One must be affiliated with a party to contest an election - the law does not permit independent candidates - but the parties generally have little to offer to their candidates, other than the opportunity to register their candidacy and a few minutes of free time on radio and television.  To mount a competitive campaign, each candidate depends on his personal network of supporters, including, typically, political bosses on the state level, mayors, businessmen, media representatives, churches, labor unions, important professional groups such as physicians, lawyers, professors, police and public servants.

It would be difficult under any hypothesis to impose strict party discipline on a deputy elected under these conditions.  To whom will he respond if the party line conflicts with the interests of some important element in his personal network of supporters?  For this reason, as if this were not enough, the norms and customs which regulate the internal functioning of the parties and the Legislature do not facilitate the task of the party leadership.

By a legal disposition which is being questioned in court, but which continues to be followed in practice, every deputy is a "born candidate" for reelection - he has a guaranteed place on his party's list, even if he is in open conflict with the majority of the party.  On the other hand, it is easy for those who are dissatisfied with their party to change parties on the even of the election.  Those who for some reason (usually related to regional disputes) cannot find a place in any of the big parties have the option of the so-called "rental lists," micro-parties that serve as legal umbrellas for these sharp-shooters.

The internal rules of the Congress are exceptionally generous toward the small parties in the allocation of administrative facilities and procedural prerogatives.  An absurd example:  even someone who is the only representative of a party in the Chamber has the right to speak as leader in order to "orientate his delegation" in its voting.  A more serious example:  a procedure called "a motion to separate the vote" allows the leader of any party to require the body to vote on a project piece by piece, multiplying the number of votes necessary for final approval.  This is a powerful instrument for obstruction that can drastically limit the majority's control of the legislative agenda.  Why does the majority not eliminate this internal rule?  Because the majority - any majority - is a mosaic of parties, regional factions and sectoral interests.  In this mosaic, each part sees itself as a minority, for which any veto power may be a significant trump card.

From the point of view of its political parties and its Congress, Brazil is an extreme case of "consensual democracy," in which power is little centralized, many forces have veto power, and relevant decisions require the agreement of almost everyone.

It happens that the President of the Republic - an only he, among all the elected officials - is elected by the majority of the Nation.  The Constitution of 1988 reinforced this principle, requiring a second round of elections between the two leading candidates, unless one of them obtained an absolute majority of the votes in the first electoral round.  For a typical voter, the Presidential election has a strong plebiscitary meaning.  The personalist tradition in Brazilian political culture is foreign to the division of powers:  it finds it a strange idea that there may be other legitimate parties with which the President must negotiate in order to do that which he was elected to do.

The inconsistency between this view of the Presidential mandate as the expression of a majoritarian plebiscite and the consensual logic of the Legislative function is accentuated by two characteristics of the electoral system:  the disproportionality in the representation of the states, and the pulverization of the vote in metropolitan areas.  The Constitution states that no state will have fewer than eight nor more than seventy representatives in the Chamber.  This favors the smaller states in detriment to the others, and leaves the most populous state, São Paulo, underrepresented.  The number of inhabitants per deputy in São Paulo is seventeen times higher than in the smallest state, Roraima.  The open list system, for its part, makes the state capitals and neighboring districts a sort of free hunting ground for all the potential candidates in the state.  In the interior, it is more common to encounter electoral districts that are relatively closed, with one or two dominant candidates.  The pulverized vote in the capitals ends up electing proportionately fewer representatives than the concentrated vote in the interior.  Ultimately, the most populous states and the large urban concentrations, which are decisive in the election of a President, carry much less weight in the composition of Congress.

The President may be tempted to invoke the plebiscitary expression of the polls and the pressure of the streets against the eventual obstacles to his policies in Congress.  This was the pattern of Executive-Legislative conflict that characterized the short trajectory of Brazil's first experience with mass democracy, from 1946 to 1964.

Or the President can negotiate with Congress, which has been the case since the restoration of democracy in 1985.  In principle, he should not have insuperable difficulties in obtaining the support of one or two of the large parties, in addition to his own, in exchange for participation in the cabinet and compromises with regard to government policies.  In the current party configuration, this type of coalition would give the government a majority in both the House and the Senate - but a majority only in name.  Due to the weak control of the parties over their own delegations, in the hour of voting, especially on controversial measures, one can count on a significant number of defections motivated, most of the time, by dissatisfactions that have nothing to do with the matter under discussion.  Attracting more parties to the government's coalition is one way of giving the government the breathing room necessary to compensate for these defections, but at the cost of aggravating another difficulty:  that of harmonizing in Brasilia the factions that are adversaries on the state level.  There is always the option of guaranteeing the vote of congressmen by giving concessions to the individual demands, which usually involve appointments to positions of confidence and application of federal resources in their districts.  The risk is that this kind of bargaining can become the focus of Executive-Legislative relationships, undermining both the weak control of the parties over their representatives in Congress and the fiscal, administrative and political-moral consistency of the government's action.

Social bases disrupted by urbanization, a weak party system, a Congress deeply segmented by regional and sectoral interests, government struggling with unstable party coalitions, the risk of paralysis in decisionmaking:  this was the diagnosis of Brazilian presidentialism at the beginning of the decade of the 1990s.

Not only the political scientists, but the majority of the observers and participants, began to doubt, at that time, whether the patient's evolvement would conciliate democracy and governability.  Brazil resented the failure of successive inflation control plans over a five-year period, leaving a trail of economic stagnation, administrative disorganization and exasperation in society.  Uncontrolled public spending and the bogging down of structural reforms undermined, in the ultimate analysis, the legitimacy of both powers, Executive and Legislative.  Democracy was not the cause of the economic problems of the country.  Fiscal imbalance, internal and external debt, inflation reignited by the indexing of prices and salaries, all this was an inheritance from the military governments.  But the civil governments also revealed themselves incapable of producing solutions for these problems, or even to find ways to keep them from getting worse.


The curve of inflation is the strongest evidence of the turn-around experienced by Brazil in the Fernando Henrique Cardoso administration.  Economic growth was far from spectacular, but it was positive.2

The structural and institutional changes behind the brute economic facts are impressive.  In fewer than ten years, Brazil transformed itself from a closed economy, almost autarchic, to an open economy.  The account of foreign trade doubled, from US$50 billions at the end of the 1980s to more than US$ 100 billion in 1998.  The internal market also expanded with the increase of consumption by salaried workers, especially the poorest ones, whose income was strongly "taxed" by inflation.  Local industry made up a good deal of its backwardness in comparison to world standards, under pressure from imports, stimulated by the increase in consumption, and pushed along by the strong entrance of foreign capital and technology.  Between 1991 and 1997, the total productivity of all the factors in the Brazilian economy grew on average 3.3% per year, well above the level of the OECD countries.

The state as entrepreneur, created by Vargas and expanded by the military governments, finished leaving the metallurgical and petrochemical sectors, and reduced drastically its direct presence in the infrastructure sectors, which were opened to private national and foreign capital.  There was a profound restructuring of the financial sector, as much public as private.  The state banks (belonging to the state governments), which were an uncontrollable source of expansion of public debt, were almost all privatized or closed.  Social security reform, still incomplete, decelerated expansion of the deficit of the National Social Security Institute - the public federal system which provides for the retirement of private sector employees.  The perspective now is for the deficit to be stabilized within fifteen years.  State regulatory agencies were created in the privatized sectors, such as electrical energy and telecommunications, as well as in sectors that were opened to competition, such as petroleum.  The flexibilization of the rules of employment for public servants made it possible to control costs and to modernize the personnel and management policies in the three levels of government - federal, state and municipal.  A new model of the state emerged, focusing on regulation of markets, strategic coordination and the development and universalization of social rights.

All these changes required the intense support of Congress.  All of the implied legislative changes.  Many of them required constitutional amendments.   The Brazilian Constitution covers matters that in other countries are usually regulated by sub constitutional legislation.  Since it was promulgated, in 1988, it has already suffered 32 amendments, of which 28 were passed since March 1994, when a transitory constitutional measure was introduced to give fiscal support to what became the Real Plan.

The Cardoso administration has exercised an unequivocal, but no absolute, leadership over the legislative agenda.  In five years, it suffered a few defeats in Congress, in votes that involve reduction in social security benefits and advantages given to public employees.  But it managed to achieve approval for its constitutional amendments by the necessary 3/5 votes in both the House and the Senate in dozens of "separate votes."  And it counted with a strong majority in almost all the votes on legislation that did not require constitutional amendment.

There seems to be something outrageous in the distinct and long lasting leadership of president Cardoso in the
congressional arena, as contrasted with the difficulties of past administrations. Cardoso's critics complain about a "steam
roller" when the government's majority mobilizes for important votings. They question the legitimacy of the methods
allegedly used for such mobilization: an invisible machinery of distribution of positions and funds would secure the
government's controle over the majority in Congress.

This vision makes an absurdity of the paradigm of consensual democracy.  It is as if there were something inherently censurable about the taking of decision by a majority, and by extension, about all distinctions between majority and minority with respect to the exercise of power.

The idea of the "steam roller" propelled positions and funds is, in any case, very weak as an explanation of political process.  Previous administrations were more likely to see their leadership deteriorate than consolidate, allowing individual bargaining to become the dominant tone of their relationship with Congress.  By contrast, the reforms conducted by the Cardoso government tightly limited and continue to limit the margin for this kind of bargaining.  The privatization of the state enterprises took out of the hands of government hundreds of positions that could have been, as they eventually were in the past, filled by political patronage.  The reform of federal financial institutions imposed strict professionalization in their management.  The management of governmental programs in areas such as education, health, welfare and agrarian reform follow the same road.  The program of fiscal stability, which is managing to cut the growth of public debt in relationship to GDP, leaves only a residual space for "parochial amendments" - the transfer of federal resources for small works in districts indicated by Congressmen.  To be this successful in the same terrain where the previous administrations became bogged down, would require really incredible gifts of prestidigitation.

What other factors, then, can explain the capacity of the Cardoso government to obtain Congressional support for its policies despite the institutional framework that created so many barriers to governability.

A simple explanation, but not a negligible one, emphasizes the personal role of the President in the political process.  In this view, Cardoso was successful, first, because he undertook the reform agenda with more conviction and conducted it with more ability than his predecessors.  His style of exercise of power combined the right doses of strategic obstinacy - going to the point of, in pursuit of fundamental objectives, running risks that other politicians would have a avoided - and tactical flexibility - and with apparently infinite patience in his negotiations with Congress.3

Another explanation places emphasis on historical circumstances.  The gravity of the economic and political crisis broke the resistance of the politicians to structural reform.  At the same time, a consensus in favor of the reforms was created in other segments of the Brazilian elites, including those in the communications media, experts within the government, and business leaders.  The success of the Real Plan anointed Fernando Henrique Cardoso as the political expression of this consensus and guaranteed him a decisive victory in the first round of the presidential election.  At this time, taken together, the consensus of the elites and strong a popular mandate, taken together, gave the president an unprecedented legitimacy to obtain reforms from the Congress.

The two lines of explanation are pertinent, but they can lead to a somewhat hasty conclusion:  that the success of the Cardoso government would be a fortunate case of the right personality at the right place at the right time.  If nothing changes in the political institutions, the difficulties of marriage between democracy and governability would be fated to reappear in the future, under less exceptional circumstances.

The currency devaluation of January of 1999 appeared to foretell this scenario.  The popularity of the President fell, hit by the feeling of broken confidence - he had just been reelected to defend the real - and by the losses of income resulting from the devaluation.  The turbulence of the financial markets gave evidence of the risks of the opening of the economy, and revealed fissures in the consensus of the elites in favor of structural reforms.  In Congress, the opposition, which always dismissed the macroeconomic stabilization policies as an imposition of the "Washington Consensus," was reinforced - more in speeches than in votes, it is true - by segments of the government coalition disquieted by the low popularity of the President and with cuts in spending.  In this scenario, many analysts asked, not whether, but when the President would become a "lame duck," putting in risk the continuity of the reform process.

But this scenario did not prove true.  Although under fire, the government's majority in Congress did not disperse nor was it paralyzed during the economic crisis.  More powerful was the fear of being held responsible by public opinion for aggravating the crisis, and the calculation that the aggravation of the crisis would only benefit the opposition.  Both the opposition and government's reticent allies lacked an alternative economic strategy, other than the reiteration of criticism that had been made before the devaluation:  specifically, that Brazil had paid a very high and unnecessary price to maintain the value of the real against the dollar.  The President, precisely at the moment of greatest lack of confidence in his authority, ran all the risks and showed more obstinacy than flexibility.  Both the President and the country had little margin for negotiation about the additional fiscal effort required by the crisis.  There was, therefore, determination by the Executive branch and support by Congress for carrying out the necessary fiscal adjustment, turning around the negative economic expectations of the beginning of 1999, and keeping up the processing of the items of structural reform that were still on the Congressional agenda.  In May of 2000, after a long and exhausting public debate that exposed differences within the government coalition, the readjustment of the minimum wage proposed by the Executive was passed by 305 votes in the House - three fewer than the quorum needed for constitutional amendments.  The winning argument was that a major readjustment, although desirable in every other respect, might jeopardize the goals of fiscal adjustment, reduction of interest rates, and recuperation of economic activity.

Pressure of circumstances once again?  The idea that the political institutions only react to crises, but are unable to anticipate them, has resonance in the stereotype of Brazil as a country of improvisation, always "on the brink of the abyss."

A sharper analysis of the aftermath of the exchange devaluation could lead, nevertheless, to another line of explanation:  although the rules of the political party game have not changed (except for the possibility of reelection of the President, governors and mayors), there are signs of deeper changes in the functioning of these institutions.

The government coalition kept united by the fear of aggravating the crisis, it is true.  But it did it in circumstances that were not present in 1994.  First, the coalition's parties and factions had basically the same perception of the risks and options of the crisis and placed their bets together, not on a new economic plan, but on the continuity of the strategy under way, at a moment when it faced sharp difficulties.  The consensus around this strategy, although under fire, not only maintained itself but separated itself from reliance of the short-term popularity of the President and the Real Plan.  The reactions of the communications media, the business opinion leaders, the government technobureaucracy and the political leaders themselves indicate - it is my hypothesis - that this consensus came to be a more permanent and structural expression of the ideological center of gravity of the national elites.

Secondly, the government forces kept a minimum of  capacity to carry out decisions in Congress, even in face of the most critical dissatisfaction with the presidential leadership.  They were led to this by the lack of alternatives to the government's economic strategy.  But they also accomplished this for another reason:  because fifteen years of democratic leadership had distilled civic leadership that, for better or worse, had learned to deal together with both the benefits and responsibilities of power.  This signifies - in my second hypothesis - the consolidation of a political center with the vocation of carrying out the objectives of the ideological center of gravity staked out by the elites.


What future tendencies can we project, from this analysis, for the Brazilian party system and Executive-Legislative relationships?  Running the risk of being roundly refuted by the forthcoming curve of history, I am going to sketch a scenario.

Except in the eventuality of a catastrophic development in the world economy (such as a "crash landing" in the United States and the consequent shock waves), the basic consensus around the structural changes initiated by the Cardoso government should be maintained.  Today the debate in the government camp concerns possible variants of changes:  privatization by selling a controlling block or by selling distributed shared?  But the general direction of the changes is not consistently questioned, even by the opposition.  On this particular issue, the PT (Workers' Party) has not been able to state clearly whether or not it would reverse the privatizations already carried out.

Given this, the tendency for the consolidation of a new political center should also be maintained.  If it will continue to be expressed by the current coalition, if in the future it may lead to mergers of parties, if it will contest the next Presidential election with one or various candidates, if President Cardoso will recover sufficient popularity to be able to conduct the succession process, none of these things can be foreseen at this time.  I would bet, under whichever hypothesis, that it is within the field of this center that the options for power will be defined in 2002.

Another foreseeable implication of this scenario:  the discussion of political reform will continue to be on the agenda, but as the perception of risks of ungovernability are attenuated, it will tend to revolve around incremental improvements in the present system.  There will be little room for more ambitions reform proposals, such as parliamentarism and mixed district voting.

The PSDB [Cardoso's Brazilian Social Democratic Party], since its founding in 1988, was the party which most strongly fought for parliamentarism and which has defended electoral reform.  An irony of history:  as the party of President Cardoso, and the core of the governing coalition, the PSDB may be helping to prove that such an awkward political system was, in the last analysis, capable of carrying Brazil somewhere.


1 Scott Mainwaring, Rethinking Party Systems in the Third Wave of Democratization (Stanford University Press, 1999) is, probably, the academic work most up to date on this theme.  It has original research, intelligent analysis and a good balance of bibliography.

2 From June of 1980 until the launching of the Real Plan in June of 1993, Brazil lived with an average inflation of 1,110% a year.  From July of 1994 until the present, the average inflation has been 14.6% annually.  This rate includes the inflationary effect of the exchange devaluation in January of 1999.  The accumulated inflation in 2000, from January to June, was 1.5%.  The rate of growth of the Gross Domestic Product went from an average of 1.4% a year from 1981 to 1992, to 3.5% a year from 1993 to 1998.  It was close to 1% in 1999, under the impact of the world financial crisis and the devaluation of the real.  It should return to a rate of 3.4% to 4% in 2000.

3 See in this respect the biography of Fernando Henrique Cardoso by Ted Goertzel, Fernando Henrique Cardoso Reinventing Democracy in Brazil(Lynne Rienner Publishers, 1999).