Brazil: The Costs of Multiparty Presidentialism
Eduardo Mello, Matias Spektor
Journal of Democracy, Volume 29, Number 2, April 2018, pp. 113-127 (Article)
Published by Johns Hopkins University Press
DOI: https://doi.org/10.1353/jod.2018.0031
For additional information about this article
https://muse.jhu.edu/article/690080
Access provided by FGV-Fundacao Getulio Vargas (17 Apr 2018 19:10 GMT)
Brazil: The CosTs of
MulTiparTy presidenTialisM
Eduardo Mello and Matias Spektor
Eduardo Mello is assistant professor of international relations at
Fundaç~a o Getulio Vargas in S~ a o Paulo. Matias Spektor is associate
professor of international relations at Fundaç~a o Getulio Vargas in
S~ao Paulo.
In 2014, a criminal investigation started by a judge in the city of Curitiba touched off a remarkable string of revelations that have tarnished
almost all members of Brazil’s political elite. The investigation, later
named Operation Car Wash (Operaç~ao Lava Jato), remains ongoing,
but the resulting imbroglio is already considered “the largest corruption scandal ever to beset a democratic nation.”1 It revolves around contractors bribing public officials in sums adding up to billions of U.S.
dollars in order to secure construction and service contracts in the oil,
nuclear, and public-infrastructure sectors—contracts that also became
a device for siphoning money from state-run institutions into private
pockets through overcharging. Many of these dealings involved Brazil’s national oil titan Petrobras. The resulting stream of lucre flowed
through the heart of Brazilian politics: Participants in the scheme diverted money to political parties at the federal and state levels, which
then used it both for their members’ personal enrichment and to finance
political campaigns.
The vast trove of evidence accumulated as part of Operation Car
Wash has stunned Brazilians and the rest of the world. Tapped telephone conversations, secretly recorded meetings, and hundreds of hours
of filmed plea-bargain statements involving top-ranking politicians and
leading businessmen all speak to the pervasive role of money—and particularly of illicit funds—in Brazil’s political life.
The result has been an unprecedented challenge for the political class,
compounded by a severe economic slump that began in 2013. In the period from April 2014 to December 2015 alone, 61 people were convicted
of crimes in connection with Operation Car Wash.2 Among those servJournal of Democracy Volume 29, Number 2 April 2018
© 2018 National Endowment for Democracy and Johns Hopkins University Press
Journal of Democracy
114
ing prison time are several business tycoons, members of the National
Congress—including Eduardo Cunha, onetime speaker of the Chamber
of Deputies (the lower house)—and other politicians. While a number
of political parties were implicated, the scandal dealt a major blow to
the leftist Workers’ Party (PT), which held both the presidency and the
largest seat share in the Chamber at its outset. Amid the turmoil, Congress impeached President Dilma Rousseff in August 2016, though on
unrelated charges. The influential PT leader Luiz Inácio Lula da Silva,
Rousseff’s predecessor as president (2003–11), was sentenced in July
2017 to almost a decade in jail (currently pending appeal) for receiving
bribes; he also stands accused of additional Car Wash–linked offenses.
The clouds kicked up by Operation Car Wash still hang over Rousseff’s successor Michel Temer and his ideologically malleable Party of
the Brazilian Democratic Movement (PMDB), the PT’s former coalition
partner. In May 2017, after incriminating recordings suggested that Temer had participated in arranging payoffs to impede the investigation,
violent clashes erupted between the authorities and demonstrators seeking his removal. Temer has been formally charged with offenses that
include bribe-taking and obstruction of justice, although the Chamber of
Deputies, which has the final say on initiating court proceedings during
the president’s term, has twice voted against bringing the case to trial.
Operation Car Wash is the largest corruption scandal to have rocked
Brazil, but it is not the first. Ten years earlier, the public learned that
under a scheme that became known as mensal~ao (big monthly payment),
the executive branch doled out millions of dollars in side payments to buy
votes for the president’s agenda items from coalition members in the legislature. On top of these “allowances,” illegal campaign-finance activities
and the placement of cronies in state-owned companies in return for kickbacks further greased the wheels of Brazilian politics. Brazil’s successive
scandals, and the sheer scope of the Car Wash disclosures, have torpedoed
public trust in the country’s politicians. They have also revealed the centrality of illicit activity to the Brazilian president’s overcoming of a key
institutional challenge: forming and sustaining a coalition.
Questioning the Consensus
The Brazilian executive’s dependence on graft and patronage to govern calls into question the scholarly consensus in favor of the view that
multiparty presidential systems such as Brazil’s are strong and stable.
This body of analysis, which began two decades ago with landmark
studies on Brazil’s democracy, now includes research on cases of coalitional presidentialism around the globe.3 Exponents of this view have
argued that presidents elected in majority contests can coexist with fragmented legislatures without necessarily bumping up against deadlock or
political instability. In this, these exponents argue against the expecta-
Eduardo Mello and Matias Spektor
115
tions of Juan Linz and others who wrote after the global “third wave” of
democratization affected Latin America during the 1980s.4
Multiparty presidentialism, these scholars posited, can provide stable
governments as long as presidents control the legislative agenda and are
able to deploy resources (in the form of “pork,” cabinet positions, patronage, and other coalition-worthy goods) to secure the loyalty of individual
legislators. Researchers found that presidents in multiparty systems from
Argentina to Indonesia have been able to cobble together working majorities to pass significant reforms, all while keeping the military in the
barracks. Indeed, since the 1990s only one large multiparty presidential
system—that of Venezuela—has given way to authoritarian rule.
Yet it is now clear that corruption, rent-seeking, and fundamentally
flawed ways of conducting political business are not exceptional, but
rather integral to managing Brazil’s brand of multiparty presidentialism.5 The chaos roiling Brazil is not the product of individual malfeasance or an innate culture of corruption, but rather of flawed institutional engineering. At its heart lie the rules that govern the relationship
between the executive and legislative branches, which encourage exactly the kind of graft that the Car Wash scandal has revealed.
This insight seriously undercuts the optimistic view of multiparty
presidentialism, with implications well beyond the Brazilian case. Fragmented presidential systems elsewhere in Latin America (in Argentina,
Bolivia, Colombia, Ecuador, Panama, and Uruguay) show signs of the
same core features that mark Brazil’s politics, including its distinctive
brands of rent-seeking and corruption. So, too, do countries with multiparty presidential systems in sub-Saharan Africa and the former Soviet
Union.6 A new research agenda must take into account mounting evidence
that the dynamics of multiparty presidentialism foster, and indeed depend
upon, a political arena rife with rent-seeking and corrupt behavior.
In order to govern, chief executives in multiparty presidential systems have to reconcile two competing goals. On the one hand, they seek
to provide public goods for the majority of voters; on the other, they
must lock in the support of the parties that make up the governing coalition. This key feature of multiparty presidential systems leads to three
interrelated outcomes: a power imbalance between the executive and the
legislature, interest-group dominance, and bad governance.
First, multiparty presidentialism skews the balance of power between
the executive and legislative branches to a degree that corrodes the
checks and balances integral to the proper functioning of a presidential
democracy. With a fractured legislature facing a president who wields
expansive powers, legislators do not have the capacity to deliver on
meaningful programmatic commitments to their constituents. Instead,
they specialize in providing particularistic goods. To obtain these goods,
they come to depend on handouts from the executive and on resources
from private interest groups; these dependencies compete with and un-
Journal of Democracy
116
dermine their accountability to voters. Second, these dynamics allow
interest groups to take on an outsized role in shaping policy decisions at
the expense of the majority of voters. Corruption-fueled relations with
each other and with outside interests give the executive and legislative
branches a shared incentive to limit the reach of the judiciary and other
watchdog institutions, further damaging accountability. Finally, the erosion of checks and balances and of accountability in turn breeds bad
governance, defined as a situation in which rent-seeking and outright
corruption are the rule rather than the exception.
Coalitional presidential systems generate these outcomes by reinforcing deeply entrenched social institutions typical of nondemocracies, and
common in many developing countries. Multiparty presidentialism is
not the original cause of bad governance, nor of widespread recourse to
personalistic forms of redistribution such as patronage (the conditional
exchange of public-sector jobs for political support) and clientelism (the
conditional exchange of government benefits for political support). It
does, however, produce a model of executive-legislative relations that—
by encouraging patronage-based coalitions, limiting the options available to legislators, and presenting openings for interest groups—hinders
any shift from clientelistic to programmatic approaches.7
All these developments are evident in the case of Brazil. In March
1985, Brazil underwent a transition to civilian rule after twenty-one
years of military dictatorship. The new regime introduced free, competitive, universal-suffrage elections and established new institutions that
ushered in a period of stability. The military have remained in the barracks ever since, and when crises did erupt they were resolved through
constitutional means (as with the presidential impeachments of 1992 and
2016). Successive administrations have tamed inflation, raised taxes,
laid the foundations of a minimalist welfare state, opened sectors of the
economy up to international trade and financial competition, privatized
state companies, and established an incipient regional-security community that has begun to dislodge old-time rivalries in South America.
Today, democracy is the only game in town. For over a decade now,
Brazil has consistently received scores of 2 (the second highest) on Freedom House’s annual indices of political rights and civil liberties. Yet the
mensal~ao and Car Wash disclosures suggest that this stability has come at
the price of effective checks and balances, accountability, and good governance—with significant implications for public-service provision, the
national economy, and, not least, trust in the institutions of government.8
Unbalanced Power
If multiparty presidential systems are to be stable, the president needs
to be able to organize a governing coalition in the fragmented legislative branch. To do so, the executive must have at its disposal extensive
Eduardo Mello and Matias Spektor
117
powers, often including control of the legislative agenda; exclusive authority over key issue areas; extensive decree powers; and the capacity
to distribute “pork,” cabinet positions, and a range of public-sector jobs
as patronage. But the executive’s use of these prerogatives to form and
manage coalitions also skews the balance of power in its favor. The
legislature becomes a market in which parties and legislators constantly
bargain to procure the resources on offer from the president, who thereby secures support for his or her agenda.
As a result, the process of building effective coalitions in fragmented
legislatures can run counter to the logic of checks and balances. The
traditional view of checks and balances, elaborated long ago by James
Madison, holds that the need for presidents (who are focused on the
large national issues of the day) and legislators (whose concerns are
primarily local) to collaborate in order to pass laws also allows the two
branches to check one another’s ambitions. The difficult compromises
that follow make it harder for these two centers of power to easily collude to the detriment of citizens.9
In multiparty presidential systems, however, the relationship between
the executive and legislature centers on the core bargain of particularistic goods for political support. Party leaders rarely base their decisions
to back or oppose the president’s agenda on programmatic commitments. Instead, lacking any other means of delivering on their promises
to voters, they lend their support to the executive in exchange for pork,
resources for patronage, and clientelistic benefits. Parties then distribute
these resources to the legislators in the governing coalition, who pass
them on to constituents and supporting interest groups. A division of
labor emerges wherein legislators concentrate on distributing particularistic goods, while presidents focus on programmatic politics and the
quality of governance nationwide. Although the executive takes on this
role to some extent in all separation-of-powers systems,10 the division
under multiparty presidentialism is much more drastic: The president
effectively becomes the only official accountable to the electorate for
the provision of public goods.
This means that for legislators, in stark contrast to Madison’s formulation, the very process of securing the resources needed to win support in
their districts entails ceding control of the broader agenda to the president. A vivid illustration of this dynamic can be seen in Argentina today:
The Peronist Partido Justicialista nominally opposes center-right president
Maurício Macri, but its sizeable legislative delegation votes in favor of
Macri’s reform agenda for fear of losing access to the resources he controls—which it needs in order to maintain its own clientelistic ties to voters. Such a model of executive-legislative relations inevitably weakens the
barriers that prevent the two branches from colluding against the citizenry.
These dynamics are present in acute form in Brazil’s coalitional presidential system. The seats in Brazil’s National Congress (which consists
118
Journal of Democracy
of the 81-member Federal Senate and the 513-member Chamber of Deputies) are split among numerous parties, with 26 represented as of this
writing. Presidents must cope with the fact that their own parties will
lack a majority. As the incentives of multiparty presidentialism demand,
the constitution grants the executive extensive powers that tilt the balance in its favor. The president can issue provisional legislation by decree
(although all laws must eventually be approved by Congress); dislodge
pending legislation from congressional committees; force Congress to
vote on urgent measures; veto bills in part or in whole; and nominate allies to tens of thousands of jobs in the powerful public bureaucracy and in
over a hundred state-owned companies. To top it off, the president alone
can initiate any legislation pertaining to taxation and budgetary matters.
But Brazilian presidents still need Congress to approve bills, and legislators make their support for the president’s agenda conditional on
securing resources controlled by the executive. Legislators’ political
futures hinge on these resources not only because congresspeople use
pork and patronage to fulfill their commitments to constituents, but also
because they deploy these goods to sustain relationships with interest
groups. In exchange for political influence, these groups fund legislators’ campaigns, help to maintain political machines, organize key voters, and, last but not least, cooperate in siphoning off additional wealth
from the state. Whether legislators use the resources that they control
to satisfy constituents or to win the favor of interest groups, bad governance and corruption are often the result.
Whoever sits at the presidential palace will seek both to please the public and to attract coalition partners—the former with public goods such as
low inflation, income redistribution, state welfare programs, schooling,
sanitation, and public-health services, and the latter with particularistic
resources such as preferential access to public services through patronage
networks or the distribution of goods to their local clients. Thirty years
of practice have proven this system to be stable. To push through their
legislative agenda, Brazilian presidents work through party leaderships
and with the speakers of the two houses of Congress. Statistical analyses
have shown that legislators are disciplined in following the instructions of
their parties’ leaderships and that the majority of successful bills originate
at the presidential palace.11
To make it happen, Brazilian presidents must pump vast resources into
the system as pork and patronage. The mensal~ao scandal and Operation
Car Wash have shown that even this may be insufficient to guarantee legislative support. Presidents have therefore found it useful to sweeten the
pot by allowing legislators to appoint their allies to the many plum jobs
at the executive’s disposal in Brazil’s powerful state-owned companies
and regulatory agencies. For legislators, this is more than just a means of
securing comfortable positions for their associates: Crucially, the government jobs at stake confer upon their occupants responsibility for arranging
Eduardo Mello and Matias Spektor
119
state contracts with private companies and the power to provide favors to
key interest groups. This means that the holders of these positions enjoy
manifold opportunities for extracting from the private sector additional
campaign donations or outright bribes, which the officials then share with
their political patrons in the legislature. This is done with the executive’s
participation and consent. Corruption flowing from government contracts
with the private sector become a powerful glue to hold together the unbalanced executive-legislative compact.
While Brazil’s multiparty presidential system benefits legislators when
they are on the prowl for patronage, it effectively takes away any meaningful tools that they could use to establish programmatic links with their
voters. Since presidents control the legislative agenda and the bulk of
state resources without much input from Congress, legislators on their
own cannot credibly promise to pass bills or shift spending. Moreover, the
system has produced many parties that are strong in the legislative arena
but weak in the electoral arena. President Temer’s PMDB, for instance,
is currently the largest force in Congress, yet there is little evidence of
a unifying ideology driving its legislators’ behavior. During campaigns,
party leaders offer candidates access to resources they obtain from the
president and to key local networks and alliances. But the programmatic
dimension of parties matters less: After securing the resources they need
from Brasilia, legislators tend to campaign in isolation, relying on parties’
manifestos seldom if at all.
Interest-Group Dominance
Executives in multiparty presidential systems follow a two-track strategy in order to satisfy resource-hungry legislators without disrupting their
own ability to provide at least some public goods for the electorate at
large. First, the executive dedicates much of its energy to working with
party leaders to distribute resources among clients and supporters. Second, it sequesters and protects key areas of policy from the process of
bargaining with members of Congress whose chief concern is pleasing
their clients. In the Brazilian case the chief operator in charge of distributing benefits and perks is usually the president’s chief of staff (MinistroChefe da Casa Civil). In other words, the same person responsible for
pushing the president’s agenda and establishing policy coherence across
ministerial portfolios is also in charge of making sure that ministries meet
the particularistic demands of coalition members in the legislature. On
the other hand, to insulate critical policy areas, certain programs and departments—the Ministry of Finance, the Ministry of Foreign Affairs, the
Central Bank, and a host of redistributive initiatives aimed at the poorer
sections of the electorate—are normally run by loyal personal appointees
of the president who do not have to submit to the otherwise ubiquitous
logic of horse-trading between the executive and legislature.
120
Journal of Democracy
Starting in the 1990s, a few years after the beginnings of Brazil’s current democratic system, President Fernando Henrique Cardoso (1995–
2003) realized that guaranteeing redistributive policies to poorer voters
would require cutting legislators out of the picture. He proceeded to centralize major social initiatives in his own hands and to bypass the formal
structures of the ministries of education and health, both major magnets
for congressional requests for pork, appointments, illicit side payments,
and other concessions. The takeoff of flagship antipoverty programs such
as Bolsa Escola (latter renamed Bolsa Família) and Saúde da Família—
which reward families who keep their children in school and provide preventive healthcare to poorer neighborhoods—gave Cardoso a boost in his
1998 reelection campaign. Cardoso’s decision to manage these two initiatives beyond the reach of Congress was a rational one if we accept that the
logic of multiparty presidentialism leads to bad policy. While he had little
choice but to resign himself to poor governance in most areas, he could
cherry-pick key fields in which the provision of good governance would
translate into handsome electoral rewards for himself.
Outside such select areas, executive-legislative collusion has left policy making in Brazil subject to strong influence from organized interest
groups. For example, successive presidents appointed political cronies
to key directorships in the oil giant Petrobras so that they could award
contracts to well-connected companies in exchange for bribes. There
was also corruption in the Navy’s submarine program: Politicians used
national-security laws to bypass normal bidding procedures and favor
companies with strong ties to the administration, which then used the
funds they received to help finance political campaigns. Numerous other
examples exist, in sectors as diverse as education, healthcare provision,
transportation, and even the 2016 Olympics in Rio de Janeiro. In each
and every case, presidents and party leaders have worked hand-in-hand
to reward well-connected groups with benefits, as well as sinecures that
have in turn been used to support members of the governing coalition.
Interest groups in Brazil emerge from both the private and the public
sectors. What they have in common is that they see, in the cracks of multiparty presidentialism, a space to extract exclusive benefits. Powerful
Brazilian interest groups such as the Odebrecht construction company or
the JBS meatpacking conglomerate found in the mid-2000s an environment conducive to the purchase of legislation through campaign donations, side payments, and bribes. The items on the political menu ranged
from cheaper government credits and tax breaks to exceptional treatment
in fields as varied as environmental licensing, “national-content” laws,
and beneficial import-tax regimes.12 In exchange for side payments, for
instance, Odebrecht was awarded generous contracts to build infrastructure. Its political connections with the ruling Workers’ Party were so
formidable that the company’s lawyers helped to write emergency presidential decrees that gave the company tax breaks estimated to be worth
Eduardo Mello and Matias Spektor
121
slightly under US$50 billion.13 JBS also used political connections to
extract rents from the Brazilian taxpayer, paying hefty bribes to politicians in exchange for lucrative subsidized loans from the state-owned
Brazilian National Development Bank. These are just two examples of
companies that have been very successful at gaming the system; other
players have been doing the same for decades. Commodity-export associations secured access to cheaper credit at times of harvest, while
industrial-sector clubs, such as the powerful S~ao Paulo Association of
Industries, have obtained the passage of protectionist legislation. Large,
private telecommunication companies, such as the national conglomerate Oi, have won lucrative tax breaks and other financial concessions.
Public-sector unions in such key sectors as education and health achieved
beneficial conditions for retirement, the armed forces have been spared
from the effects of pension reforms, and judicial-branch employees have
won exemption from a cap in salaries that applies to the other branches
of government.
Interest-group dominance reinforces the disconnect between voters
and legislators, who become subject to the dictates of their financial
backers. Dozens of plea-bargain testimonies have provided evidence
of private conglomerates systematically supplying personal rewards to
legislators in return for their support of favorable laws and regulations.
Also in plea-bargain statements, Odebrecht executives have provided
evidence that they asked President Lula, prior to official international
trips, to put in a word on the company’s behalf with his foreign counterparts whenever pending payments or contracts were at stake. Loans
from the Brazilian National Development Bank were provided to foreign governments with the understanding that these governments would
then use these funds to purchase infrastructure work from Odebrecht;
the Bank was also offering credits to Odebrecht itself to help finance the
company’s projects abroad. In sum, interest-group dominance is prevalent not only at home, but also in Brazilian foreign relations.
The Death of Accountability
Watchdog institutions that promote transparency and accountability—including the Office of the Attorney General (Ministério Público
Federal), the Office of the Comptroller-General, and the judicial system—pose perhaps the greatest threat to the dominance interest groups
enjoy in Brazil. It is therefore no wonder that the president and the legislative branch, linked with these groups in a political ecosystem based
on the circulation of side payments, collude to tame and weaken control
institutions. Shared incentives induce them to work together to maintain protection mechanisms for themselves and for the interest groups
on which they depend, in the process curbing investigative powers and
transparency in governmental affairs.
122
Journal of Democracy
In this light, it seems likely that the chief factor behind President
Rousseff’s impeachment by Congress in 2016 was her inability to protect the political class from the progress of Operation Car Wash, which
had by then sent a host of powerful politicians and businessmen to jail.
In one of the secret recordings that has come to light, Senator Romero Jucá, right-hand man to then–Vice-President Temer, confided that
Rousseff had to be removed from office because “we have to fix this
[expletive]. We need to change the government to stop the bleeding . .
. [Temer in office will] build a national pact, with [the support of] the
Supreme Court. With everything we’ve got.”14 Congress’s refusal to allow the prosecution of Temer, despite audio recordings of the president
colluding with the CEO of JBS to obstruct the Car Wash probe, is perhaps the best example of the deleterious effects of the patterns that shape
executive-legislative relations in Brazil.
Operation Car Wash has highlighted the degree to which the “web”
of accountability is both weak and uneven. Contrary to the claims of
earlier works, Brazil’s multiparty presidentialism has not given rise to
powerful watchdogs able to keep presidents and their legislative coalitions in check.15 For all the powers now in the hands of Brazilian control
institutions, they have demonstrated little effectiveness at checking corruption and have at times themselves succumbed to the broader culture
of �pay to play.�
Under the omnipresent influence of coalition politics, Brazil’s judicial institutions—their considerable formal autonomy notwithstanding—have also been prey to politicization and corruption. Never in the
history of the Republic of Brazil, for instance, has a presidential nominee for the Supreme Court been turned down by the Senate. In the course
of Operation Car Wash, a string of declassified recordings of private
conversations involving top-ranking politicians and businessmen have
shown the degree to which the judicial branch bends to political winds.16
Although it has at key moments exercised its constitutional powers to
counter the executive and the legislative, the Brazilian judiciary is not
an effective bulwark against corruption in the political class.
Consider Brazil’s Court of Accounts (TCU). This is an institution
empowered to run audits of all government accounts. Yet politicization at the top-leadership level is rife. In 2017, the president of the
court and two more of its nine judges—all chosen by either the Congress or the executive—were accused by the Federal Police of taking
bribes from a private company that sought judicial approval of contracts for work on the Angra 3 nuclear reactor. 17 The nomination of
judges with strong political connections diminishes the court’s ability
to monitor and investigate corruption allegations. Account tribunals
at federal and state levels have also been found to be rife with corrupt
activity, and regular audits often fail to uncover corruption and inefficiencies or to lead to serious investigations. By the same token, the
Eduardo Mello and Matias Spektor
123
Electoral Court (TSE) has spent billions on improving its system of
campaign-expense accounting, but investigations suggest that illicit
fundraising remains pervasive.
The sense of impunity is widespread. On top of politicization, certain
features of Brazil’s judicial system, which confers special prerogatives
on public officials, have further thwarted anticorruption drives. With a
few notable exceptions, politicians charged in scandals have largely remained out of prison. A complex system of appeals has for decades protected these officeholders from facing jail time, and it is plausible that
many of the politicians convicted in the Car Wash scandal will never
spend a day behind bars. Congressional censure or removal from office of corrupt politicians is rare, although it is becoming less so as the
current scandal unfolds. Even when parties to corruption face judicial
punishment, this seldom removes wrongdoers from the political game:
There is evidence of politicians continuing their corrupt activities and
even running their party machines from prison. Since auditing bodies
and police investigators expect that their targets will be able to stay in
office and might seek revenge, those charged with combating corruption
are often reluctant to do their job.
Given these conditions, how do watchdog institutions ever succeed?
Why has the political class not stopped Operation Car Wash? Recent experience in Latin America from Argentina to Brazil to Guatemala to Peru
shows that networks of bureaucratic entrepreneurs can sometimes overcome the power of the executive and the legislative branches and succeed
in launching major investigations that lead to the prosecution and eventual
imprisonment of corrupt politicians, together with their public- and private-sector purchasers. These entrepreneurs are normally young, educated
abroad, and tied to a transnational network that supplies knowledge and
support. The crackdown on graft during the last few years in Latin America has been the exception rather than the norm, however. Much of it was
made possible by unprecedented popular anger and the worst economic
crisis since 1929. It remains to be seen how successful the push against
corruption will be in the long run. Evidence is mounting that the political
class is ready to fight back insofar as it can do so without alienating public
opinion, which is largely supportive of anticorruption probes.
Bad Governance
While the problems of rent-seeking and endemic corruption do not
exist in Brazil because of multiparty presidentialism, this particular
form of coalitional politics works to perpetuate the prevalence of patronage and clientelistic arrangements: When checks and balances are
weak and accountability is uneven, the relationship between politicians
and voters becomes strained and ever more dependent on nonprogrammatic connections.
124
Journal of Democracy
Multiparty presidentialism’s addiction to corrupt and rent-seeking
practices has important consequences for the regulatory environment
and the overall relationship between companies and the state. As firms
devote more of their energies and capital to securing advantageous deals
through government connections, more of their revenue will come from
rent-seeking and less from improvements in productivity. Regulatory
agencies overseeing key sectors of the economy will be a tempting target for groups seeking to extract easy profits. All of this is likely to corrode economic efficiency in the long term. It will also eat away at equality and fairness in society: Well-organized groups will create economic
opportunities to benefit themselves, while the disorganized poorer majority of voters shoulders the burden.18
Again, Brazil is the paradigmatic case. We now know that executives
at the state oil giant Petrobras were political appointees who saw as
their main job collecting illegal fees from private-sector contractors who
sought to do business with the state—and then channelling those fees to
their backers in government. The contractors in question included many
of Brazil’s mightiest corporations, among them Odebrecht and the multinational conglomerate Andrade Gutierrez. Estimates suggest that since
1997 the companies involved in the graft secured billions of dollars in
government-subsidized credit through Brazil’s National Development
Bank. To ensure continued access to this gold mine, these firms lavished
gifts and other favors on cooperative politicians and contributed large
sums, both on and off the books, to their reelection campaigns.19
In their quest to build up a governing coalition, Brazilian presidents
also used their discretionary powers to meddle in state-owned companies. Operation Car Wash helped to reveal two scandals in which the
executive branch set policies for state-owned banks in ways that favored the interest groups backing coalition legislators. Similarly, the
public-sector pension funds that own large shares in Brazilian private
conglomerates did not base their investment decisions on the interests
of the shareholders; instead, they made them with an eye to bolstering illicit relationships between private firms and public officials. Multiparty
presidentialism fortified the crony capitalism that has persisted in Brazil
since its authoritarian days.
The result for the majority of Brazilian voters is bad governance.
When officials focus their resources and attention on sustaining clientelistic networks, public-service provision and popular well-being suffer. One of the world’s wealthiest countries in terms of GDP, Brazil falls
short at meeting its population’s basic needs. Half of Brazil’s 206 million people lack access to basic sanitation, and 35 million lack access to
clean water. Public spending on education is high, amounting to 16 percent of government revenue, but student performance on international
tests remains disturbingly weak even in comparison to countries that are
far poorer, such as Albania or Jordan.
Eduardo Mello and Matias Spektor
125
Who profits? The main beneficiaries are the organized interest groups
that gain control over the presidency and the legislature through campaign
donations, and use this control to exercise inordinate influence over the judiciary. Both the illegal and the legal mechanisms of interest-group dominance weaken checks and balances and erode accountability. As a result,
Brazil’s powerful executive can secure stable government, but only at an
enormous cost. As every holder of the office from José Sarney (1985–89)
to incumbent Michel Temer has learned, the interest groups upon which a
president depends for keeping a governing coalition together are also the
biggest obstacle to passing any legislation for reform.
Reconsidering Multiparty Presidentialism
The debate about multiparty presidentialism has mostly focused on
the issue of stability. Can presidents form working coalitions in fragmented legislatures, or does the combination of proportional representation and presidentialism eventually lead to deadlock and regime
collapse? An impressive body of literature has shown that powerful
presidents can indeed work with fragmented legislatures to form coalitions, ensure stability, and pass significant legislation.
Yet the case of Brazil shows that a different issue—quality of governance—raises equally pressing concerns. Existing studies have ignored
the degree to which the process of coalition-building under multiparty
presidentialism undermines the checks and balances that are essential
to presidential democracies. Further studies will be needed to examine how these dynamics play out in other cases—for instance, in countries that combine fragmented legislatures with electoral rules that empower parties, such as Argentina with its closed-list electoral system.
Future research should also consider how each of the particular tools
that presidents use to govern in multiparty systems impacts the quality of governance. Are there ways to design institutions that will allow
independently elected presidents to form majorities without undermining legislators’ ability to make programmatic commitments to voters,
and the accountability of legislators to the citizenry? Finally, we recognize that our theory needs to be systematically tested against alternative
explanations. It is conceivable, for instance, that reforms altering the
balance of power between legislature and executive could improve the
quality of governance.
Nonetheless, Brazil’s scandal-ridden politics gives serious grounds
for pause when it comes to the existing consensus about multiparty presidential systems. In Brasilia, successive occupants of the presidential palace and the chambers of Congress have avoided deadlock only by joining
forces in an unsavory compact with corrupt public- and private-sector
interest groups. The purchase of political stability with ill-gotten gains
has come at a cost for Brazil: eroding voters’ control over the politicians
Journal of Democracy
126
they elect, creating economic distortions, and sucking state resources and
official attention away from the provision of much-needed public services. Now that Operation Car Wash has dragged the backroom dealings of
Brazil’s business and political elites into the light of day, it appears that
many Brazilian voters have had enough. But it remains to be seen whether the current anticorruption push can overcome the mutual-protection
pact among these elites and usher in a new approach to politics. If it does
not, the usual cycle of patronage-based alliances, policies-for-purchase,
and eroding accountability will begin anew.
NOTES
1. Ray Fisman and Miriam A. Golden, Corruption: What Everyone Needs to Know
(New York: Oxford University Press, 2017), 13.
2. Marcus André Melo, “Latin America’s New Turbulence: Crisis and Integrity in
Brazil,” Journal of Democracy 27 (April 2016): 50–65.
3. Argelina Cheibub Figueiredo and Fernando Limongi, Executivo e Legislativo
na nova ordem constitucional (Rio de Janeiro: FGV, 1999); Daniel Chasquetti, “Democracia, multipartidismo y coaliciones en América Latina: evaluando la difícil combinación,” in Jorge Lanzaro, comp., Tipos de presidencialismo y coaliciones políticas
en América Latina (Buenos Aires: Clacso, 2001), 319–59; José Antonio Cheibub, Adam
Przeworski, and Sebastian M. Saiegh, “Government Coalitions and Legislative Success
Under Presidentialism and Parliamentarism,” British Journal of Political Science 34 (October 2004): 565–87; Octavio Amorim Neto, “The Presidential Calculus: Executive Policy
Making and Cabinet Formation in the Americas,” Comparative Political Studies 39 (May
2006): 415–40; Argelina Cheibub Figueiredo, Júlio Canello, and Marcelo Vieira, “Governos Minoritários no Presidencialismo Latino-Americano: Determinantes Institucionais e
Políticos,” Dados 55 (October–December 2012): 839–75; Carlos Pereira and Marcus André Melo, “The Surprising Success of Multiparty Presidentialism,” Journal of Democracy
23 (July 2012): 156–70; Paul Chaisty, Nic Cheeseman, and Timothy Power, “Rethinking
the ‘Presidentialism Debate’: Conceptualizing Coalitional Politics in Cross-Regional Perspective,” Democratization 21, no.1 (2014): 72–94.
4. Juan J. Linz, “Presidential or Parliamentary Democracy: Does It Make a Difference?” in Juan J. Linz and Arturo Valenzuela, eds., The Failure of Presidential Democracy: Comparative Perspectives, vol. 1 (Baltimore: Johns Hopkins University Press,
1994). See also Matthew Soberg Shugart and John M. Carey, Presidents and Assemblies:
Constitutional Design and Electoral Dynamics (Cambridge: Cambridge University Press,
1992); Scott Mainwaring, “Presidentialism, Multipartism, and Democracy: The Difficult
Combination,” Comparative Political Studies 26 (July 1993): 198–228; Juan J. Linz and
Alfred Stepan, Problems of Democratic Transition and Consolidation: Southern Europe,
South America, and Post-Communist Europe (Baltimore: Johns Hopkins University Press,
1996); Scott Mainwaring and Matthew Soberg Shugart, eds., Presidentialism and Democracy in Latin America (Cambridge: Cambridge University Press, 1997).
5. A searchable database of declassified evidence collected as part of Operation Car
Wash can be found at: https://jota.info/lavajota.
6. On multiparty presidentialism in Africa and the former Soviet Union, see Chaisty,
Cheeseman, and Power, “Rethinking the ‘Presidentialism Debate’”; Paul Chaisty, Nic
Cheeseman, and Timothy J. Power, Coalitional Presidentialism in Comparative Perspective: Minority Presidents in Multiparty Systems (Oxford University Press, forthcoming).
Eduardo Mello and Matias Spektor
127
7. On the broader debate on why politicians relinquish patronage and clientelism, see,
inter alia, Seymour Martin Lipset, “Some Social Requisites of Democracy: Economic
Development and Political Legitimacy,” American Political Science Review 53 (March
1959): 69–105; Barrington Moore, Jr., Social Origins of Dictatorship and Democracy:
Lord and Peasant in the Making of the Modern World (Boston: Beacon Press, 1966);
James C. Scott, “Corruption, Machine Politics, and Political Change,” American Political Science Review 63 (December 1969): 1142–58; María Candelaria Garay, Including
Outsiders: Social Policy Expansion in Latin America (PhD diss., University of California,
Berkeley, 2010); and Evelyne Huber and John D. Stephens, Democracy and the Left: Social Policy and Inequality in Latin America (Chicago: University of Chicago Press, 2012).
8. The 2017 Latinobarómetro survey showed a staggeringly low 8, 11, and 7 percent
confidence in the government, Congress, and political parties respectively. “Informe Latinobarómetro 2017,” www.latinobarometro.org/latNews.jsp.
9. James Madison, “No. 51,” in George W. Carey and James McClellan, eds., The
Federalist (Indianapolis: Liberty Fund, 2001).
10. On separation-of-power systems more generally, see for example Terry M. Moe,
“Political Institutions: The Neglected Side of the Story,” Journal of Law, Economics, and
Organization 6 (Special Issue, 1990): 213–53; and Terry M. Moe and Michael Caldwell,
“The Institutional Foundations of Democratic Government: A Comparison of Presidential and Parliamentary Systems,” Journal of Institutional and Theoretical Economics 150
(March 1994): 171–95. See also Douglas L. Kriner and Andrew Reeves, The Particularistic President: Executive Branch Politics and Political Inequality (New York: Cambridge
University Press, 2015).
11. Figueiredo and Limongi, Executivo e Legislativo na nova ordem constitucional.
12. See Jota, “Delaç~ao Odebrecht,” www.jota.info/lavajota/videos.html; “Compra de
MPs continuou mesmo após início da Lava Jato, diz Odebrecht,” Folha de S. Paulo, 20
April 2017; “Joesley comprou uma lei por 20 milh~oes de reais,” Veja, 19 May 2017.
13. Flávio Costa and Vinicius Konchinski, “Delaç~ao aponta que Odebrecht agiu por
MPs que deram R$ 140 bi em benefícios a empresas,” UOL, 18 December 2016, https://
noticias.uol.com.br/politica/ultimas-noticias/2016/12/18/delacao-aponta-que-odebrechtagiu-por-mps-que-deram-r-140-bi-em-beneficios-a-empresas.htm.
14. Rubens Valente, “Em diálogos gravados, Jucá fala em pacto para deter avanço da
Lava Jato,” Folha de S. Paulo, 23 May 2016.
15. Pereira and Melo, “Surprising Success”; Melo, “Crisis and Integrity.”
16. “Leia transcriç~ao dos principais trechos da gravaç~ao dos delatores da JBS,” Folha
de S. Paulo, 6 September 2017.
17. Fábio Fabrini, “PF aponta indícios de corrupç~ao e favorecimento de empreiteira no
TCU,” Estad~ao, 19 July 2017.
18. Joseph E. Stiglitz, The Price of Inequality: How Today's Divided Society Endangers Our Future (New York: W.W. Norton, 2013).
19. See Hamer Arteaga and Elizabeth Salazar, “Los millonarios préstamos del BNDES
al club ‘Lava jato,’” OjoPúblico, 30 October 2016, https://ojo-publico.com/318/el-clublavajato-y-los-millonarios-prestamos-del-bndes.