The Surprising success of
Multiparty Presidentialism
Carlos Pereira and Marcus André Melo
Carlos Pereira is professor of political economy and government at
the School of Administration (EBAPE) at the Getúlio Vargas Foundation
(http://ebape.fgv.br/en) in Rio de Janeiro. He was recently a
visiting fellow at the Brookings Institution. He publishes widely on
comparative government and Latin American politics. Marcus André
Melo, professor at the Federal University of Pernambuco, is currently
a fellow of the John Simon Guggenheim Foundation. He has been a
Coca-Cola Visiting Professor of Political Science at Yale University
and a Fulbright Fellow at MIT.
Maintaining a sustainable democracy in a multiparty presidential regime—that
is, a stable democratic presidential regime in a context where
partisan fragmentation in the legislature abounds—was once thought to
be nearly impossible. Until recently, pundits and scholars seemed as
one in declaring that multiparty presidential democracies suffer from
survival prospects that are poor at best.
In the early 1990s, the critique of presidentialism advanced by Juan J.
Linz and others exerted a broad influence, and scholars viewed the coexistence
of presidentialism and multipartism as “the difficult combination.”
Experts expected multipartism to sharpen the “perils of presidentialism”
by increasing the probability of deadlock in executive-legislative relations,
by promoting ideological polarization, and by making multiparty
coalitions hard to achieve.1
Presidential democracies’ best chance for survival,
these experts argued, lay in the adoption of a U.S.-style two-party
format.
Yet multiparty presidentialism seems to be here to stay. Argelina
Figueiredo and her coauthors have found that during the more than a
quarter-century from 1979 through 2006, Latin America was host to
only two countries (Costa Rica and Mexico) that had single-party governments
the whole time. Six countries (Bolivia, Brazil, Chile, Colombia,
Ecuador, and Panama) had coalition governments throughout that
Journal of Democracy Volume 23, Number 3 July 2012
© 2012 National Endowment for Democracy and The Johns Hopkins University Press
157Carlos Pereira and Marcus André Melo
same period, while four others (Argentina, Paraguay, Venezuela, and
Uruguay) had coalition governments at least some of the time.2
Nor is
the institutional combination of presidentialism and multipartism something
unique to Latin America. Indonesia has been stable as a multiparty
presidential regime since the fall of Suharto’s authoritarian New Order
and the transition to democracy in 1999. Since 2004, the governing coalition
there has consisted of five parties with seats in congress plus
eighteen supporting but unseated parties.3
The unexpected stability of multiparty presidential regimes demands
further research. How have presidents been solving—or at least managing—the
“difficult combination” with multipartism? We still know little
about the tools and institutional conditions that presidents employ in this
process. We have an observed fact, but no theory to explain it. The ability
of multiparty presidentialism to subsist with sustainable democracy
is beyond dispute, yet we lack a comprehensive account of why and how
this institutional arrangement has proven durable.
Multiparty presidential regimes must be seen as special cases that
function neither like parliamentary political systems nor like two-party
presidential models. In other words, the expert literature does not fully
grasp the operational nuances of coalition-based presidential regimes.
Scholars have been misgauging the outcomes of these regimes because
they have been using theoretical and analytical tools designed to analyze
either European multiparty parliamentary regimes or the classic twoparty
presidential system of the United States.
Until the 1980s, experts believed that “the best of all worlds” was
two-party parliamentarism. Neither multiparty parliamentarism nor any
form of presidentialism, they claimed, could equal the combination of
parliamentary government and two parties. The prevailing sentiment
against coalition government, particularly in the Anglo-Saxon world,
was encapsulated in Benjamin Disraeli’s dictum that “England doesn’t
love coalitions.”4
More than half a century’s worth of comparativists, it
seemed, did not love them either.
As well as stability, two-party parliamentary systems supposedly
offered an array of desirable qualities associated with the responsibleparty
model, including clear lines of responsibility, decisiveness, and
responsiveness to citizen demands.5
By contrast, comparativists and
empirically oriented democratic theorists viewed multiparty governments
as likely to lack stability and transparency, and prone to games
of blame-shifting among coalition partners. They also argued that multipartism
led to inefficient governments, paralysis, and an inability to
reach effective, authoritative decisions.
According to this argument, strong presidents should only make
things worse. Will not powerful executives tend unilaterally to impose
their preferences on the legislature, triggering institutional crises along
the way? And will this tendency not become all the more worrisome
158 Journal of Democracy
when such executives enjoy only weak support from parties and face
fragmented legislatures? Thus constitutional-design experts viewed
strong presidencies as terrible things to introduce into multiparty sys-
tems.6
In the 1980s, comparativists went part of the way toward correcting
this bias. Experts learned to appreciate some aspects of multiparty
systems, including the imperatives for compromise, tolerance, and inclusiveness
that they often include. Research on presidentialism followed
the lead of comparative politics, shifting from open criticism of
multipartism to a more moderate view of what Linz called the “perils
of presidentialism.” Most scholars viewed presidentialism as likely to
prove unstable where party systems were fragmented and executives
enjoyed extensive constitutional powers. While this more nuanced view
of presidentialism appeared in the seminal work of Matthew Shugart
and John M. Carey, recent scholarship has argued that the process of
government formation under both presidentialism and parliamentarism
shares the same underlying institutional logic.7
Recent developments in Latin America and Asia have lent support not
only to the general argument that coalition government is both viable and
functional, but also to the more controversial claim that multiparty systems
may require strong presidents in order to generate effective governance.
Brazil and Chile have emerged as the successful models for governance
in Latin America. Each is a multiparty presidential democracy
whose chief executive numbers among the region’s strongest. Bolivia,
Mexico, and Venezuela, by contrast, where presidents enjoy scantier formal
powers, have not performed so well. What explains this variation?
In Bolivia and Venezuela, presidents have tried to concentrate power
and have created instability in the process. In Mexico, successive
governments have suffered from deadlock as a result of divided government.
Ironically, the pundits and academics of the 1980s and 1990s
predicted that today’s top performers would be the least likely to establish
good governance. By contrast, countries that used to have what
was once considered an ideal combination of weak presidentialism and
strong partisan support have not performed well.
Interestingly, countries extolled for their social, political, and economic
outcomes also proved to be pioneers in the introduction of proportional
representation (PR) in the region: Costa Rica (1913), Uruguay
(1918), Chile (1925), and Brazil (1933). These countries also boast the
longest experience with multipartism in Latin America. The region’s
later adopters of PR (Argentina and Mexico, both 1963) have seen the
emergence of hegemonic parties and zero-sum politics, pathologies associated
with sharply divided polities.
Three things explain multiparty presidentialism’s unexpected success:
the powers of presidents; the availability of trade-worthy coalition
goods (cabinet posts, “pork,” patronage, and the like) that executives
159Carlos Pereira and Marcus André Melo
can use to gain legislators’ support; and the strength of institutional
checks on the executive’s discretion. Good governance hinges on the
delegation of extensive powers to both presidents and autonomous institutions
that can and will check the executive branch. Where presidents
concentrate powers without congress’s consent, what is happening is
not delegation but usurpation. Remembering that one is not the other
can help us to avoid overstretching the concept of “strong presidents.”
This expression is too often used to cover both cases when legislatures
democratically delegate constitutional powers to the executive and cases
when powers are unilaterally abused. Some observers focus so intensely
on how power helps presidents to overcome bargaining problems that
they risk overlooking presidential abuses. The concentration of power
may emerge as a result of presidents’ influence over constitutional decisions
in key constitutional moments or “paraconstitutional initiatives.”8
The Multiparty Game Under Presidentialism
Only recently have scholars begun to examine the relevance of coalitions
in multiparty presidential regimes. This newer research has
addressed topics such as coalition voting discipline, the relationship
between coalition type and legislative success, the flexibility that presidents
derive from their powers of appointment,9
the impact of cabinet
formation on presidential survival in times of crisis, and the relationship
between presidential policy-making strategies and cabinet formation.10
Yet none of this research suggests how or why coalition-based multiparty
presidential regimes may lead to sustainable democracy, much
less which political and institutional conditions are necessary to achieve
this outcome.
Coalition-based multiparty presidential regimes succeed if the president
1) is constitutionally strong, 2) has “goods” to trade in order to
attract and keep coalition partners, and 3) faces institutionalized and
effective checks on presidential actions. Some multiparty presidencies
have avoided serious policy-making stalemates by developing complex
and flexible systems of exchange among the various branches of government.
The goods exchanged in return for legislative cooperation might
include cabinet posts, pork projects, and policy concessions. These
goods do not figure independently of one another in presidential strategies,
nor do they produce independent effects on legislative support.11
Thus, to generate sustainable legislative support and stable democracy
in a fragmented environment, multiple goods are needed.
Although the tools used to evaluate two-party presidentialism are
unsuited to taking the measure of multiparty presidentialism’s peculiarities,
so too are the tools used to analyze parliamentary regimes.
Parliamentary systems typically feature certain bargaining threats and
deadlock-resolvers, including the formal ongoing need for majority con-
160 Journal of Democracy
fidence from the legislature, the prospect of reforming government or
calling early elections, and greater clarity of electoral accountability.
Thus no one should assume that multiparty presidentialism works like
parliamentarism, for this is not the case.
In multiparty presidential systems, even constitutionally strong executives
may operate as perpetual formateurs, cobbling together a distinct
voting coalition to support each important initiative. In such institutional
environments, executives may assemble heterogeneous governing
majorities using ideologically diverse political parties. Executives
also must often use particularistic benefits (such as pork-barrel projects)
along with political transfers (cabinet posts, other presidential appoint-
ments,12
and policy concessions) in complex efforts aimed at garnering
needed votes in congress.
In the vast majority of multiparty presidential regimes, it is the provision
of such coalition goods—and not considerations of ideology, partisan
loyalty, or power to set the agenda—that does the most to sway
congress. In this environment, the electoral rules usually give legislators
strong incentives to respond to particularistic monetary and political
transfers. Therefore, the degree of political support in multiparty
presidential regimes and, in turn, the stability of democracy often hinge
on the extent to which a president can give legislators the goods they
want on an ongoing basis.
For example, the strategic allocation of coalition goods helped Brazil’s
then-President Luiz Inácio “Lula” da Silva of the Workers’ Party
(PT) to pass pension reform within a year of first taking office on 1 January
2003. This reform represented an extreme departure from his previous
rhetoric because it adversely affected the interests of public-sector
workers by putting a ceiling on their benefits. Therefore it created serious
fissures within both the PT and Lula’s larger governing coalition. As
Eric D. Raile and his coauthors have shown, Lula gathered a supermajority
of the lower house behind his favored reform mainly by rewarding
his own party with cabinet jobs while using pork to induce ideologically
distant coalition partners and even outsiders to join.13
Even strong partisan
opponents such as the centrist Brazilian Social Democratic Party
(PSDB) and the conservative Liberal Front Party (PFL) largely went
along with the initiative. Helping Lula even more than the ideological
acceptability of the legislation was the steering of nearly 41 percent of
all 2003 pork spending to individuals associated with those two parties.
Viewing things at the state level, we see that almost nine-tenths of the
pork went to states governed by parties outside Lula’s coalition (about a
third of the total went to states run by the PSDB or the PFL).
The contrasting example of Ecuador shows how important the absence
of coalition incentives can be. In the years leading up to and
including the abortive presidency of Abadalá Bucaram, who was impeached
and dismissed by Congress for “mental incapacity” in Febru-
161Carlos Pereira and Marcus André Melo
ary 1997 after just six months in office, the Ecuadorean presidency
lost so much control over such incentives that presidents found it
nearly impossible to form coalitions. In Brazil, the trend ran the other
way as presidents and party leaders gradually improved their ability to
offer—and honor—incentives that served to cement durable governing
coalitions.14
The Value of Checks on Executives
The most common type of presidential regime in Latin America is
one in which the president lacks the support of a single majority party in
congress. The fragmented party systems typical in Latin America make
exchange mechanisms crucial. If democracy is to work—indeed, if it is
to last—the government in general and the president in particular must
build and keep winning coalitions. In order to do so, the president needs
to be strong and attractive to different coalition partners. By “strong,”
we mean equipped with extensive constitutional powers that allow the
president to change the status quo while at the same time avoiding unwelcome
initiatives from the opposition.
Executive strength in this sense does not mean that congress issues
the president blank checks or that oversight institutions are helpless to
constrain executive-branch actions. On the contrary, in order for a multiparty
regime with a strong presidency to sustain robust democracy,
there must be an extensive system of checks and balances.15
It is political
competition and the fragmentation of power—both characteristics
of multipartism—that will give rise to effective mechanisms for holding
the executive accountable. In other words, party fragmentation itself
works to constrain a powerful executive, functioning as a kind of parliamentary
oversight.
Although presidential abuses of power and conflicts among governmental
branches in Bolivia, Ecuador, and Venezuela have drawn much
recent attention,16
this should not obscure the progress that Brazil and
Chile have made toward good governance. In the early 1990s, scholars
believed these two countries were doomed to failure because of alleged
constitutional-design flaws, including “exaggerated presidentialism,”
large effective numbers of parties and a corresponding need for multiparty
coalition governments, and open-list PR, among other factors.
Paradoxically, presidents in these countries have not only ensured stability,
but have also gained constitutional power through incremental
reform. Contrast with this the Latin American countries (Bolivia, Peru,
Venezuela) that featured the weakest executives during the 1980s: They
have had a much harder time achieving stability and governability. In
these countries, presidents with few de jure powers have nonetheless
imposed their preferences de facto, often by resort to what Carey and
Shugart call “paraconstitutional initiatives.”
162 Journal of Democracy
This comparison suggests that the way to promote effective governance
and democratic stability is to combine a strong presidency with a
robust set of checks and balances. We differ from the expert literature’s
dominant strand in refusing to restrict our discussion of presidential
strength to the question of whether presidents can pass their agendas.
The key to promoting sustainable democracy in countries such as Brazil,
Chile, and Uruguay has been the successful establishment of sturdy
checks and balances. Lively and diverse media are needed, as are legislative
oversight, an independent judiciary, and “horizontal-accountability”
bodies. These can include the contralorias (comptroller-generals’
offices) that exist in a number of Latin American countries, as well as
the Ministério Público (public prosecutor’s office) and tribunais de contas
(audit courts) found in Brazil. In other words, governability requires
that each of the three branches of government be strong. By focusing
exclusively on dealings between presidents and congresses, the extant
literature fails to account for the importance of the accountability institutions,
both in themselves and as factors in shaping executive-legislative
relations.
When it comes to accountability, multipartism can cut two ways. It
may sap the desire of parties in the governing coalition to check the
president, yet it also tends to work against the accumulation of presidential
power by forcing chief executives to negotiate extensively with
party leaders. The upshot can be a system in which clientelism is pervasive
but greases rather than gums up the wheels of governance. Robust
political contestation—often carried on by a viable opposition party or
coalition such as one sees in Brazil and Chile—can keep such a system
from degenerating into dysfunctional clientelism and corruption.
It is no accident that the multiparty presidential democracies which are
best endowed with pluralistic media outlets and independent oversight
institutions are also those which regularly achieve the most impressive
development outcomes.
Although Latin America’s legislatures are generally weak by world
standards, there are great variations in strength among them. Going by
Sebastián Saiegh’s regional index of legislative strength, the congresses
of Brazil, Chile, and Uruguay rank near the top.17
All else being equal, a
more capable congress will not only handle basic lawmaking functions
more smoothly, but will be better able to oversee the executive branch
as well.
As for judicial independence, existing measures suggest that here,
too, Brazil and Chile are among the top performers. According to one
study, Chile earns the top score on five out of seven existing indicators,
whereas Brazil ranks first or second on six of them.18
In the widely cited
index of judicial independence devised by Lars P. Feld and Stefan Voigt
for the Inter-American Development Bank, Uruguay, Chile, Brazil, and
Costa Rica are the top performers.19
At the other end of the spectrum,
163Carlos Pereira and Marcus André Melo
Venezuela, Bolivia, Argentina, Honduras, and Ecuador have the leastindependent
courts.
In the long-term investigation of judicial independence in Latin
America conducted by Anibal Pérez-Li~nán and Andrea Castagnola, Brazil,
Uruguay, and Chile also come out on top as regards the key proxy
measure of judicial job stability.20
In measures of turnover on the highest
national court (the assumption being that the phenomenon of top
jurists frequently leaving the bench suggests a likely judicial-independence
deficit), Brazil does best: The highest Brazilian court evinced a
level of stability (that is, a fixity of judicial tenure) only slightly below
that achieved by the United States over the entire century (1904–2006)
that the study covers. Conversely, El Salvador, Argentina, and Colombia
had the least-stable high courts, particularly during the years from
1945 through 1977.
Among national auditing institutions, the contralorias of Chile, Colombia,
and Costa Rica, along with Brazil’s federal accountability office
(known as the TCU), earn their respective countries high regional marks
when it comes to tracking how public money is spent.21
In the Andean
subregion, the Colombian comptroller-general’s office has become a
model, which makes sense given Colombia’s generally strong recent
performance in the area of democratic governance. Indeed, despite the
instability caused by a guerrilla conflict, the country has experienced
none of the problems of neighboring Bolivia, Ecuador, Peru, or Venezuela.
By contrast, Argentina’s Auditoria General and Peru’s Contraloria
General are among Latin America’s weakest auditing agencies, whether
one examines their autonomy, credibility, timeliness, or enforcement
effectiveness.
When it comes to media outlets, Brazil, Chile, and Uruguay boast the
region’s most diverse and feistily independent. Chile and Uruguay are
the only countries that Freedom House credits with allowing full press
freedom for the entire period from 2002 through 2009. Likewise, Argentina,
Brazil, Chile, and Uruguay have consistently numbered among
the top four Latin American countries in the Reporters Without Borders
(RSF) ranking. Uruguay and Chile were classified as having the highest
press-freedom score, respectively ranking 52nd
and 67th
among the
world’s roughly 190 countries, according to the Freedom House report
covering the year 2009.
Brazil as Paradigm
In order to add depth and dimension to our discussion of how a strong
yet controlled presidency and a multiplicity of parties can bolster democracy,
it may be helpful to turn to the case we know best: Brazil.
Latin America’s largest country, one may safely say, now stands as the
paradigmatic case of a successful multiparty presidential democracy.
164 Journal of Democracy
Brazilians have seen almost two decades of stable governments and have
watched their country pass the test of peacefully passing the presidency
from one party to another after a free and fair election. Brazil possesses
a complex mix of electoral institutions that allow for the representation
of diverse interests within its multiparty system, even if that system has
been blamed for breeding enough fragmentation and decentralization to
make the policy-making process cumbersome.22
There is truth to that critique, yet it is not the whole story. Brazil
also possesses some “majoritarian” institutions that centralize agendasetting
powers and encourage national-level governability. These two
sets of institutions—those that spread power and those that draw it in—
stand as poles of a continuum. Brazil is not a pure specimen of either
extreme, but rather is a hybrid that mixes what Arend Lijphart would
call consensual and majoritarian elements in its constitutional scheme.23
Since the promulgation of the new constitution in 1988, Brazilian
democracy has maintained several consensual features, including
separation of powers under presidentialism, open-list PR voting, a
fragmented party system, federalism, and an independent judiciary.
Yet at the same time, Congress has delegated to the executive massive
authorities including the power to rule by decree, the right to enact
“provisional measures” that have the immediate force of law until
Congress votes on them within sixty days,24
both the full and the lineitem
veto, exclusive rights to legislate on budgetary and administrative
issues, and the power to unilaterally appropriate the budget (the budget
law is not mandatory but merely authorizes spending which the president
can refuse to carry out). This paradoxical combination of consensualism
and centralization has rendered democracy self-enforcing and
consolidated. The combination, in other words, has created circumstances
under which all relevant political forces have found it best to
keep submitting their interests and values to the uncertain interplay of
democratic institutions.
The Congress elected in 1986, acting as a constituent assembly,
wrote and ratified the 1988 Constitution. It gives the executive branch
vast powers in order to forestall a return to the chronic institutional
instability and presidential-congressional deadlock that had led to the
1964 military coup and the more than two decades of dictatorship that
followed it. An institutionally weak president, the framers feared, could
not survive without a capacity to govern, to push through and enforce
an agenda. Legislators decided to retain open-list PR (which encourages
multipartism) because shifting away from it would have bred too much
uncertainty about sitting legislators’ electoral prospects.
In taking stock of the 1988 basic law, it is also important to note—as
many observers fail to, unfortunately—that the document also hands
extensive authority to the judicial branch as well as autonomous institutions
like the Ministério Público and the TCU. Indeed, the autonomy
165Carlos Pereira and Marcus André Melo
and powers that these institutions enjoy are unprecedented. Viewing the
framers’ handiwork as a whole, then, we see a powerful presidency but
also a potent web of watchdogs standing on guard to prevent wrongdoing.
The competitive nature of the
constitution-making process had
much to do with the establishment
of these strong autonomous institu-
tions.
The most important consequence
of the new institutional design has
been the removal of a lurking threat
to Brazilian democracy. Except for
the first directly elected president,
Fernando Collor de Mello, who resigned
in late 1992 just ahead of impeachment
and after less than three
years in office, all presidents since
the end of military rule have been
able to build reasonably stable majority coalitions within Congress,
and have been able to maintain fairly strong party discipline within the
presidential governing coalition. This has resulted in effective levels of
governability. Although no elected president has come from a party that
held a congressional majority at the time of that president’s election,
each president has nonetheless been able to secure congressional support
by wielding the presidency’s extensive legislative and nonlegislative
powers. In fact, under its current institutions, Brazil has yet to face a
true divided-government situation. This is not coincidental. Rather, the
institutional powers and resources that the executive holds and selectively
distributes serve to prevent divided government.
The 1988 Constitution is the bedrock of all this. It defines the political
institutions in Brazil and the powers of the political actors there. It
provides the promising institutional ground upon which political cooperation
can take place. And let it be remembered that the institutional
groundwork which the 1988 Constitution lays down is solely the work
of a democratically elected lawmaking body. In Brazil, there has been
nothing like the illegitimate usurpation of powers that has become sadly
common in some Latin American countries, including Argentina, Bolivia,
Ecuador, and Venezuela.
The 1988 Constitution may be understood in part as the fruit of lessons
learned from previous unsettled institutional experiences. These
include the hyperactively representative and excessively fragmented
democratic period of 1946 to 1964 as well as the centralized military
regime that stretched from 1964 to 1985. The 1988 Constitution preserves
multipartism (which, ironically, the military rulers had originally
introduced as a ploy to split the opposition) yet at the same time opts for
Brazil’s 1988 Constitution
preserves multipartism yet
at the same time opts for
a potent executive with
the tools needed to make
parties gel into coalitions,
while independent oversight
institutions keep an eye on
the whole scene and apply
checks where needed.
166 Journal of Democracy
a potent executive with the tools needed to make parties gel into coalitions,
while independent oversight institutions keep an eye on the whole
scene and apply checks where needed.
The web of accountability institutions clearly plays a role in monitoring
and constraining Brazil’s powerful executive branch.25
It is not
uncommon to see the Supreme Court ruling against presidential preferences.
Not only Lula, but also his predecessor Fernando Henrique
Cardoso suffered important setbacks in court when jurists ruled certain
reform initiatives unconstitutional.26
A very good example of such independence took place during Lula’s
first term at the height of the 2004–2005 scandal over the mensal~ao (a
monthly bribe paid to members of Congress in return for their support of
the administration). Constraining the popular Lula and his ten-party coalition
was a far from easy task. The federal prosecutor, a Lula appointee,
launched an independent investigation and in March 2006 asked the
Supreme Court to open criminal proceedings against forty people linked
to the mensal~ao affair, including the presidential chief of staff and other
powerful figures. In August 2007, the Court unanimously approved all
forty indictments—based on a report by a judicial official who was also
a Lula appointee—and each of the accused is due to go on trial in 2012.27
By saying that Brazil’s accountability institutions are independent of
the executive, we do not mean that they are immune from political influence
or that they are all equally efficient. For instance, Brazil’s court
system is so slow—especially when it comes to addressing political corruption—that
a sense of impunity and loss of confidence in accountability
institutions generally has come to pervade society at large. Contrary
to the most pessimistic analyses, however, there is evidence that checks
on corruption can be effective where there is political competition and
where officials insulated from political influence run accountability in-
stitutions.28
Another key independent player in the web of accountability institutions
is the media. Brazil has a complex, vibrant, and highly competitive
media landscape. It includes a diversified newspaper market with
at least four nationally distributed papers, three weekly newsmagazines
that have played a decisive role in uncovering political scandals, and
several radio networks that have strong journalism departments. The
single most important source of coverage and information is the competitive
television business. Media organizations possess significant autonomy
from the state, and they play active roles in investigative journalism,
denouncing corruption, and publicizing political wrongdoings.
In June 2011, just six months into the term of Lula’s favored successor
President Dilma Rousseff, her chief of staff (influential former
finance minister Antonio Palocci) was forced to resign under fire. His
undoing flowed from charges of illicit self-enrichment that appeared in
Brazil’s premier newspaper, Folha de S~ao Paulo. Leading newsmaga-
167Carlos Pereira and Marcus André Melo
zine Revista Veja recently highlighted a corruption case involving the
payment of bribes (in effect, they were payoffs to one of the presidential-coalition
parties) in return for Transport Ministry contracts. Since
Veja’s revelations of audit-court investigations into the case, President
Rousseff has fired seventeen senior officials, including Transport Minister
Alfredo Nascimento. Episodes such as this bespeak the existence of
an accountability network strong enough to bring down even powerful
executive-branch officials implicated in misdealings.
What Makes the Difference?
What are some testable hypotheses about the determinants of good
governance and democratic stability that we can put forth based on the
analysis so far? The normal idiosyncrasies of various countries aside, it
appears that certain institutional conditions do favor governability and
public integrity and might lead to good governance across multiparty
presidential regimes.
First, an institutionally strong executive mitigates the destabilizing
effects of party fragmentation that can result from minority government
and, at the same time, lends itself to the formation of majority coalition
governments after elections. For various historical reasons, Latin America
seems to have learned that a constitutionally weak president cannot
govern in a context of partisan fragmentation. The institutional choice
of legislators has been to delegate numerous powers—including control
over discretionary coalition goods—to the executive. These powers allow
presidents to frame and enact policy thanks to stable supporting
coalitions in congress. Where the constitutional delegation of extensive
powers to presidents takes place in a competitive environment, the net
governance gains can be positive if this delegation is accompanied by
the strengthening of checks on the executive, as the cases of Brazil,
Chile, and Uruguay have shown.
Second, multiparty democracies can promote good governance if
presidents have “currency” that they can trade in order to secure legislators’
cooperation. The coalition goods that form this currency work as a
mechanism of selective incentives, as a means of rewarding or punishing
legislators contingent upon their support or defection. These gains
from trade tend to be highly effective because they benefit both sides:
The presidential agenda gets enacted, while legislators get pork that they
can bring home to their districts as an aid to reelection.
Third, evidence suggests that strong legislatures act as stabilizers in
the policy-making process. Legislatures capable of mounting strong oversight
(to say nothing of opposition) can render policy enactment costly for
executives, thereby bolstering the stability of the status quo. There is risk
in that, of course, but there is also risk if a legislature fails to serve as a
policy-change gatekeeper and simply lets the executive act without check.
168 Journal of Democracy
In addition to a robust and professional legislature, a presidential system
also needs other independent check-and-balance institutions, including
courts, public prosecutors, accountability agencies, and vigorous media.
By reconciling presidentialism and multipartism and illustrating
how this institutional combination is capable of delivering good governance
and sustainable democracy, our approach has set the stage for
a more integrated and positive view of executive-legislative relations
under multipartism. Because it fosters political inclusion, multipartism
is normatively attractive for a region marked by extreme inequality and
social heterogeneity. We do not claim that multiparty presidentialism
with strong governments is an ideal institutional model. At best, it can
be considered as a suboptimal arrangement that is functional. It may
be plagued by numerous problems including clientelism, lack of transparency,
and corruption (consider those scandals in Brazil again). The
extensive use of pork and other tradeable political currencies for political
bargaining may undermine legitimacy and could point to a serious
downside of this governing method.29
Across much of Latin America,
however, multiparty presidentialism has boosted political stability, and
has not degenerated into systemic corruption as long as robust political
competition and a set of strong autonomous institutions exist alongside
it to keep its potential excesses within bounds.
NOTES
The authors wish to thank Greg Michener, Bernardo Mueller, Lee Alston, Octavio Amorim
Neto, and Maurício Cárdenas for valuable comments and suggestions regarding previous
versions of this essay.
1. Scott Mainwaring, “Presidentialism, Multipartism, and Democracy: The Difficult
Combination,” Comparative Political Studies 26 (July 1993): 198–228; see also Alfred
Stepan and Cindy Skach, “Constitutional Frameworks and Democratic Consolidation:
Parliamentarianism versus Presidentialism,” World Politics 46 (October 1993): 1–22.
2. Argelina C. Figueiredo, Deise L. Salles and Marcelo M. Vieira, “Political and Institutional
Determinants of Executive’s Success in Latin America,” Brazilian Political
Science Review 3, no. 2 (2009): 155–71, www.bpsr.org.br/english/arquivos/BPSR_v3_n2_
ago2010_06.pdf.
3. Larry Diamond, “Indonesia’s Place in Global Democracy,” in Edward Aspinall and
Marcus Mietzner, eds., Problems of Democratisation in Indonesia: Elections, Institutions
and Society (Singapore: ISEAS, 2010), 51–121.
4. Cited in Vernon Bogdanor, The Coalition and the Constitution (Oxford: Hart Publishing,
2011), 61.
5. See G. Bingham Powell, Jr., Elections as Instruments of Democracy: Majoritarian
and Proportional Visions (New Haven: Yale University Press, 2000); G. Bingham Powell,
“The Chain of Responsiveness,” Journal of Democracy 15 (April 2004): 91–105.
6. Juan J. Linz, “The Perils of Presidentialism,” Journal of Democracy 1 (Winter
1990): 51–69; José Antonio Cheibub, Presidentialism, Parliamentarism and Democracy
(New York: Cambridge University Press, 2006).
169Carlos Pereira and Marcus André Melo
7. Matthew Soberg Shugart and John M. Carey, Presidents and Assemblies: Constitutional
Design and Electoral Dynamics (New York: Cambridge University Press, 1992).
8. See the chapter titled “Calling Out the Tanks or Filling Out the Forms,” in John
M. Carey and Matthew S. Shugart, Executive Decree Authority (Cambridge: Cambridge
University Press, 1998).
9. Cecilia Martínez-Gallardo, “Presidents, Posts, and Policy: Ministerial Appointments
and Political Strategy in Presidential Regimes” (PhD diss., Columbia University, 2005).
10. Octavio Amorim Neto, “The Presidential Calculus: Executive Policy Making and
Cabinet Formation in the Americas,” Comparative Political Studies 39 (May 2006): 415–
40.
11. Eric D. Raile, Carlos Pereira, and Timothy J. Power, “The Executive Toolbox:
Building Legislative Support in Multiparty Presidential Regimes,” Political Research
Quarterly 64 (June: 2011): 323–34. On Brazil’s multiparty presidentialism, see Fabiano
Santos, O Poder legislativo no presidencialismo de coaliz~ao (Belo Horizonte: Editora
UFMG, 2003); and Argelina C. Figueiredo and Fernando Limongi, Executivo e Legislativo
na nova ordem constitucional (Rio de Janeiro: Editora FGV, 1999).
12. Sérgio Praça, Andréa Freitas, and Bruno Hoepers, “Political Appointments and Coalition
Management in Brazil, 2007–2010,” Journal of Politics in Latin America 3 (2011):
141–72.
13. Raile et al., “The Executive Toolbox.”
14. Carlos Pereira and André Mejía-Acosta, “Policymaking in Multiparty Systems:
A Comparison between Brazil and Ecuador,” Governance: An International Journal of
Policy, Administration, and Institutions 23 (October 2010): 641–66.
15. Susan Rose-Ackerman, Diane Desierto, and Natalia Volosin, “Leveraging Presidential
Power: Separation of Powers Without Checks and Balances in Argentina and the
Philippines,” Yale Law School, Faculty Scholarship Series, 1 March 2010.
16. Numerous episodes illustrate these developments. In 2005, Ecuador’s President
Lucio Gutiérrez unconstitutionally fired members of the Supreme Court. In Hugo
Chávez’s Venezuela, the government disclosed the identities of voters in a plebiscite.
In Bolivia, President Evo Morales has encouraged mobs to block access to legislative
buildings during key sessions. For more, see Scott Mainwaring, Ana Maria Bejarano, and
Eduardo Pizarro Leongómez, eds., The Crisis of Democratic Representation in the Andes
(Stanford: Stanford University Press, 2007); Fabrice Lehoucq, “Bolivia’s Constitutional
Breakdown,” Journal of Democracy 19 (October 2008): 110–24; and Javier Corrales, “For
Chavez Still More Discontent,” Current History, February 2009.
17. Sebastián Saiegh, “Active Players or Rubber Stamps? An Evaluation of the Policymaking
Role of Latin American Legislatures,” in Carlos Scartascini, Ernesto Stein, and
Mariano Tommasi, eds., How Democracy Works: Political Institutions, Actors, and Arenas
in Latin American Policymaking (Washington, D.C.: Inter-American Development
Bank, 2010).
18. Julio Ríos-Figueroa and Jeffrey Staton, “Unpacking the Rule of Law: A Review of
Judicial Independence Measures,” Committee on Concepts and Methods Working Papers
21 (2008), IPSA.
19. Lars P. Feld and Stefan Voigt, “Economic Growth and Judicial Independence:
Cross Country Evidence Using a New Set of Indicators,” European Journal of Political
Economy 19 (September 2003): 497–527.
20. Anibal Pérez-Li~nán and Andrea Castagnola, “Presidential Control of High Courts
170 Journal of Democracy
in Latin America: A Long-Term View (1904–2006),” Journal of Politics in Latin America
1, no. 2 (2009): 87–114.
21. See the index built by Carlos Santiso, The Political Economy of Government Auditing:
Financial Governance and the Rule of Law in Latin America and Beyond (London:
Routledge, 2009).
22. Barry Ames, The Deadlock of Democracy in Brazil (Ann Arbor: University of
Michigan Press, 2002).
23. Octavio Amorim Neto, “O Brasil, Lijphart e o modelo consensual de democracia,”
in Magna Inácio and Lucio Rennó, eds., Legislativo brasileiro em perspectiva comparada
(Belo Horizonte: Editora UFMG, 2009), 105–31.
24. If Congress fails to vote on the provisional decree within the first forty-five days,
then it automatically goes to the top of the congressional agenda and overrides any other
item of business. By forcing legislators to take a position regarding new provisional decrees
within that period, the president has had incentive to issue a greater volume of (new)
decrees. See Carlos Pereira, Timothy Power, and Lucio Renno, “Agenda Power, Executive
Decree Authority, and the Mixed Results of the Reform in the Brazilian Congress,”
Legislative Studies Quarterly 33 (February 2008): 5–33.
25. Marcus André Melo,“Strong Presidents, Robust Democracies: Separation of Powers
and Rule of Law in Latin America,” Brazilian Political Science Review 3, no. 2 (2009):
30–59. Timothy Power and Matthew Taylor, Corruption and Democracy in Brazil: The
Struggle for Accountability (Notre Dame: Notre Dame University Press, 2011).
26. Lee Alston et al., “On The Road to Good Governance: Recovering from Economic
and Political Shocks in Brazil,” in Ernesto Stein and Mariano Tommassi, eds., Policymaking
in Latin America: How Politics Shapes Policies (Cambridge: Harvard University
Press, 2008), 111–53.
27. Carlos Pereira, Timothy Power, and Eric D. Raile, “Presidentialism, Coalitions,
and Accountability,” in Timothy Power and Matthew Taylor, eds., Corruption and Democracy
in Brazil: The Struggle for Accountability (Notre Dame: Notre Dame University
Press, 2011), 31–55.
28. Marcus Melo, Carlos Pereira, and Carlos Figueiredo, “Political and Institutional
Checks on Corruption: Explaining the Performance of Brazilian Audit Institutions,” Comparative
Political Studies 42 (March 2009): 1217–44. For Daron Acemoglu and his coauthors,
checks and balances are endogenous, and voters in weakly institutionalized environments
may prefer an unchecked president who redistributes over a constrained president
susceptible to elite influence. But political competition is neglected in this framework.
See Daron Acemoglu, James Robinson, and Ragnar Torvik, “Why Do Voters Dismantle
Checks and Balances?” NBER Working Paper # 17293, 2011.
29. This contrasts with the weakness of checks on presidents and of the rule of law
generally in such multiparty presidential regimes as Indonesia’s. See Dan Slater, “Indonesia’s
Accountability Trap: Party Cartels and Presidential Power After Democratic Transition,”
Indonesia 78 (October 2004): 61–92. The fate of Indonesia’s KPK (Commission
to Eradicate Corruption) provides an instructive contrast to Brazil’s. Following a series
of investigations of high corruption, two of the KPK’s deputy chairmen were arrested on
dubious charges of extortion in a maneuver to undermine it. See “The Gecko Bites Back,”
Economist, 5 November 2009.
View publication statsView publication stats