Democratization

ISSN: 1351-0347 (Print) 1743-890X (Online) Journal homepage: https://www.tandfonline.com/loi/fdem20

Presidential hegemony and democratic

backsliding in Latin America, 1925–2016

Aníbal Pérez-Liñán, Nicolás Schmidt & Daniela Vairo

To cite this article: Aníbal Pérez-Liñán, Nicolás Schmidt & Daniela Vairo (2019) Presidential
hegemony and democratic backsliding in Latin America, 1925–2016, Democratization, 26:4,
606-625, DOI: 10.1080/13510347.2019.1566321

To link to this article: https://doi.org/10.1080/13510347.2019.1566321

Published online: 15 Jan 2019.

Submit your article to this journal

Article views: 131

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=fdem20
DEMOCRATIZATION
2019, VOL. 26, NO. 4, 606–625
https://doi.org/10.1080/13510347.2019.1566321

Presidential hegemony and democratic backsliding in

Latin America, 1925–2016
Aníbal Pérez-Liñána, Nicolás Schmidtb and Daniela Vairob
a
Department of Political Science and Keough School of Global Affairs, University of Notre Dame, Notre
Dame, IN, USA; bInstituto de Ciencia Política, University of the Republic, Montevideo, Uruguay

ABSTRACT
Does the executive’s institutional hegemony represent a risk to the survival of
democracy? By hegemony, we refer to the president’s ability to control other
institutions, particularly the legislature and judiciary. To answer this question, we
develop two indices of presidential hegemony and analyze the duration of
democratic regimes in 18 Latin American countries between 1925 and 2016. The
results show that executive hegemony is a major driver of democratic instability.
This finding is robust to non-linear effects and to potential endogeneity in the
relationship between presidential power and democratic backsliding. Our findings
challenge traditional concerns about executive-legislative deadlock, and have
significant implications for the nascent literature on democratic backsliding, which
highlights executive aggrandizement as a risk factor.

ARTICLE HISTORY Received 8 May 2018; Accepted 31 December 2018

KEYWORDS Democratic backsliding; presidentialism; executive-legislative relations; executive-judicial relations;

Latin America

Introduction
Does the executive’s institutional hegemony represent a threat to the survival of democ-
racy? This question has gained new currency as recent studies on democratic backslid-
ing emphasize “incumbent takeovers” as well as “executive aggrandizement”.1 By
executive hegemony we refer, not to the formal constitutional capacity of the executive
branch to initiate or veto legislation, but to its capacity to exercise political control over
other institutions, particularly the legislature and the judiciary.
In the first section of this article, we show that research on political institutions has
advanced two answers to this question. For the classical literature on executive-legisla-
tive relations, coordination among powers is essential to stabilize democracy. For the
literature on executive-judicial relations, in contrast, the independence of judges is
key for protecting individual rights. Integrating insights from both literatures, we
argue that concentration of power in the executive branch creates a double risk for
democracy: the president can violate the rights of the opposition, or the opposition –
anticipating this result – can ally with the military to overthrow the government.

CONTACT Aníbal Pérez-Liñán [email protected]

© 2019 Informa UK Limited, trading as Taylor & Francis Group
 DEMOCRATIZATION 607

To verify this hypothesis, the second section presents four indicators of presidential
control over the legislature and the judiciary, and introduces two aggregate indices of
presidential hegemony. Our sample covers 18 Latin American countries between
1925 and 2016. This empirical focus has two advantages. First, presidential systems
unify the figures of head of state and head of government, facilitating our identification
of the centre of executive power. Second, the study of a long historical period allows us
to document multiple episodes of democratic breakdown. A greater number of break-
downs not only improves the statistical power of our analysis, it also demonstrates that
executive hegemony already constituted a threat to first-wave democracies. Executive
aggrandizement is not a recent phenomenon.
The statistical analysis of the relationship between presidential hegemony and demo-
cratic instability presents a challenge to causal inference because concentration of power
in the executive branch can be the cause, as well as the consequence, of democratic
backsliding. Hegemonic presidents use their control of other branches to weaken oppo-
sition parties, and the weakness of the opposition allows them, in turn, to consolidate
executive dominance over other institutions. This bidirectional relationship creates an
endogeneity problem, which we address in the third section through a survival model
with instrumental variables.
Our analysis shows that presidential hegemony has been one of the main factors of
destabilization for Latin American democracy. In contrast, institutionally weak presidents
facilitate democratic survival because they lack the capacity – and hence the incentives –
to attempt an incumbent takeover. This result challenges an extensive literature that con-
siders executive weakness a problem for governability and a source of democratic instabil-
ity, and thus contributes to the emerging debates on democratic backsliding.

Executive control and democratic stability

The recent literature on democratic backsliding shows a growing concern with execu-
tive hegemony. Svolik distinguishes authoritarian reversals traditionally produced by
coups d’etat, whereby security forces depose the elected government, from episodes
in which the incumbent manipulates or cancels elections.2 His study of 145 countries
indicates that, between 1789 and 2008, 61% of authoritarian reversals occurred
because of a coup and another 30% responded to a seizure of power by the incumbent
(the remaining cases resulted from civil wars or popular revolts). His analysis shows
that the end of the Cold War reduced the risk of coups but not the risk of incumbent
takeovers. It also shows that democratic consolidation is limited: the risk of a coup
declines over time, but “the accumulation of too much power in the hands of an incum-
bent seems to be a persistent threat to democratic stability”.3
According to Bermeo, executive aggrandizement is characteristic of democratic back-
sliding in the contemporary period.4 Traditional military coups, open electoral fraud, and
executive coups have declined in recent decades. However, electoral manipulation and the
aggrandizement of executive power are growing threats. The latter form of authoritarian
regression occurs “without executive replacement and at a slower pace (…) when elected
executives weaken checks on executive power one by one, undertaking a series of insti-
tutional changes that hamper the power of opposition forces to challenge executive pre-
ferences”.5 Levitsky and Ziblatt also notice that “would-be authoritarians” undermine
democracy by controlling the judiciary and law enforcement agencies, undermining
opposition leaders and free media, and altering the constitution and the electoral system.6
608 A. PÉREZ-LIÑÁN ET AL.

Similarly, Corrales argues that power grabs represent the initial step in processes of
democratic backsliding driven by polarization. “A power grab consists of an expansion
of control over crucial political institutions at the expense of the opposition (…)
through formal, informal, and in some cases unconstitutional mechanisms”.7 Power
grabs undermine horizontal accountability and induce moderates to leave the govern-
ment. They also increase the insecurity of the opposition, and promote the formation of
a radicalized opposition bloc against the executive.
This emerging literature suggests that presidential hegemony poses a risk to democ-
racy. In this article, we define presidential hegemony as the capacity of the executive
branch to control the decisions of the legislature and the judiciary. Elected presidents
establish different degrees of control over other institutions through partisan allies,
nominations, and the use of patronage. The literature on executive-legislative relations,
for example, refers to the president’s “partisan powers” to describe the executive’s
influence over congress.8 We refer instead to levels of presidential hegemony, to indi-
cate that the president’s influence is variable, that it is exercised through partisan mech-
anisms or personal loyalties, and that it extends to institutions beyond the legislature.
Two issues complicate the analysis of this topic. First, executive hegemony is a
matter of degree. At what point does presidential control of other institutions constitute
a risk to democracy? The answer to this question is straightforward for extreme cases: in
a presidential system, absolute control of other institutions by the executive implies the
end of separation of powers and thus the elimination of democracy. However, some
degree of coordination among the three branches is necessary to avoid institutional
paralysis. To address this challenge, we present a continuous measure of presidential
hegemony based on multiple indicators. We also analyze the possibility of non-linear
effects of presidential hegemony on democratic stability.
Second, presidential hegemony can be a cause of democratic backsliding as much as
its consequence. Concentration of power by the executive may unleash a dynamic of
conflict with the opposition that – as Corrales suggests – eventually causes polarization
and the downfall of democracy. But it is also possible that other factors (e.g. unscrupu-
lous presidents or weak institutions) cause democratic erosion, with presidential hege-
mony being just the visible result of this process. Both possibilities may in fact coexist.
Our analysis addresses this problem by offering an econometric treatment of
endogeneity.
In the literature that investigates executive control over congress and the judiciary,
there are two conflicting answers to our research question. The first line of research
indicates that executive influence on the legislature prevents deadlock and reduces
the risk of democratic breakdown. Meanwhile, recent investigations into executive-judi-
cial relations affirm that democracy is more stable when the government does not
control judicial decisions. We explore these arguments and integrate the two perspec-
tives to elaborate a hypothesis about the consequences of the presidential hegemony.

Executive control of the legislature

The first answer originates in the classic debate about the perils of presidentialism.9 This
literature argued that democracy is threatened by institutional deadlock when the
executive is unable to coordinate policies with the legislature. In a presidential
regime without a vote of no confidence, inter-branch coordination requires presidential
influence on congress, either through the ruling party or through the formation of costly
 DEMOCRATIZATION 609

coalitions. The absence of presidential majorities is a potential source of gridlock

between two popularly elected branches that have fixed terms in office. Given the
lack of constitutional mechanisms to resolve these conflicts, polarization between gov-
ernment (executive) and opposition (congress) can lead to the use of extra-consti-
tutional means, eventually inviting military intervention and the fall of the
presidential regime.
Linz’s argument inspired a vast literature which alleges that certain institutional
combinations – more than presidentialism itself – generate democratic instability. In
an early work, Mainwaring asserts that the institutional combination of presidentialism
with multipartyism is dangerous for democracy.10 The greater the fragmentation of the
party system, the greater the probability that the president’s legislative block will be a
minority, leading to inter-branch conflict. Two-party systems would be more prone
to preserve a stable democracy because the president has legislative support and
because the system tends to be less polarized, given that electoral competition is centri-
petal and there are limits to the entry of extremist parties.
The early literature recognized that coalitions offer a possible solution to coordi-
nation problems among elected branches, but it assumed that presidential systems do
not generate incentives to form them.11 However, extensive research subsequently
documented the formation of coalitions in presidential systems.12
At the same time, several investigations highlight the importance of the president’s
constitutional powers for the stability of the regime.13 Shugart and Carey argue that
constitutions that grant strong powers to the president tend to produce instability,
since constitutionally powerful presidents have fewer incentives to seek agreements
with congress.14 In a classic contribution, the authors highlighted the importance of leg-
islative (veto, decree, emergency powers and exclusive initiative) as well as non-legisla-
tive powers (formation and dissolution of the cabinet, the dissolution of parliament)
conferred on the executive.
Later empirical studies challenged the connection between presidential powers and
democratic instability in Latin America, given that new constitutions augmented presi-
dential powers while democracy expanded with the third wave of democratization.15
However, Helmke proves that the absence of executive control over the legislature, com-
bined with strong presidential powers, triggers institutional conflict in the form of
impeachments, self-coups, and purges of the judiciary.16 Her study suggests that
formal powers matter when the president lacks control over congress.
This institutional literature posing an association between institutional deadlock and
democratic breakdown generally ignores the final link in the explanatory chain, which
allegedly connects executive-legislative conflict with military intervention. In response
to Linz’s criticism of presidential systems, Cheibub argues that the instability of presi-
dential democracies does not result from inherent characteristics of this form of govern-
ment.17 Democratic instability results from the fact that presidential systems historically
emerged in countries with an activist military. His study shows a link between a history
of military dictatorships and democratic fragility, and emphasizes the impact of author-
itarian legacies beyond constitutional forms of government.

Executive control of the judiciary

The second answer originates in the literature on judicial independence.18 This litera-
ture maintains that effective separation of powers is necessary for the rule of law and
610 A. PÉREZ-LIÑÁN ET AL.

horizontal accountability, fundamental factors for democratic consolidation. Concen-

tration of power in the executive branch creates the conditions for government viola-
tions of individual rights, including property rights. An independent judiciary
protects civil rights, guarantees political and economic agreements among elites, and
stabilizes democracy.19 These works, however, suggest that the stabilizing effects of judi-
cial independence emerge only in favourable institutional or economic contexts.
For Reenok, Staton and Radean, judicial independence promotes democratic stab-
ility in favourable economic contexts, but not when the level of economic development
is low. The authors find that judicial institutions favour the survival of the regime when
macroeconomic conditions favour inter-class commitments. Institutional systems for
the protection of rights are most effective where redistributive commitments are cred-
ible.20 Therefore, independent courts facilitate the maintenance of the regime in devel-
oped democracies and have an almost null effect in democracies with low levels of
economic development.
In turn, Gibler and Randazzo ask if an independent judiciary prevents democratic
backsliding. They argue that concentration of power by the executive undermines
democratic principles, which must be protected by an independent judiciary upholding
individual rights.21 This study finds that an independent judiciary only offers a safe-
guard against democratic backsliding if it is well established (for more than three
years), and concludes that the power of the courts to protect democracy increases
with time. An independent but poorly institutionalized judiciary can be, in fact, a
source of instability. Elites with little democratic experience may feel threatened by
independent but non-institutionalized courts; in unstable contexts, this threat compro-
mises the survival of the regime.

The perils of executive hegemony

The emerging literature on the judiciary and the classical literature on executive-legis-
lative relations offer two conflicting interpretations of the consequences of presidential
power for the survival of democracy. As we show in later sections, this contrast does not
merely reflect that presidential influence has different consequences for different insti-
tutions. The contrast reflects different theoretical assumptions about the causes of
democratic instability, and about the role of the executive in this process.
The literature on executive-legislative relations typically assumes that the danger to
democracy originates in the military, an actor external to elective institutions who can
act in agreement with civilian leadership. The conflict between elected branches gener-
ates paralysis, triggers political polarization, and thus opens the way to military inter-
vention. The result is an abrupt regime breakdown. The literature on executive-
judicial relations, on the other hand, assumes that the danger to democracy originates
within elective institutions. The judiciary’s counter-majoritarian logic restricts unilat-
eral executive action and the tyranny of legislative majorities. Conflict between
elected branches and the judiciary generates horizontal accountability and protects
the constitutional framework. In the absence of this balance, the result is democratic
backsliding caused by government institutions.
Combining these two perspectives, our theory emphasizes the role of the executive as
the main source of democratic instability. The executive is the most powerful actor in all
modern democracies, but constitutional design and the institutional legacies of the
nineteenth century reinforce its relative advantage in Latin America. Wars of
 DEMOCRATIZATION 611

independence and border-defining conflicts fostered actors’ preferences for concen-

trated power, and the diffusion of presidentialism involved the creation of strong execu-
tives. Latin American presidents acquired broad legislative powers (emergency, decree,
and legislative initiative) that distinguished them from their American counterparts.22
The literature on executive-legislative relations never fully clarified the causal mech-
anism through which institutional paralysis translates into military intervention. In
principle, executive-legislative deadlock simply produces stability in public policies,
indicating that contending actors prefer the status quo to any proposed alternative.23
Since deadlock reflects an institutional equilibrium, civilian actors would have
reasons to mobilize the military if they fear that opponents will unilaterally impose
an alternative with irreversible costs. In modern democracies, the main actor with
capacity to implement policies unilaterally is the executive.
The executive’s capacity to impose irreversible costs to the opposition expands when
the president exercises direct control over other branches. The absence of checks and
balances allows the president to adopt policies that undermine the rights of the opposi-
tion without facing institutional resistance. This suggests that institutional deadlock is
not a source of instability, but rather a potential source of protection for parties that do
not control the government. When, in contrast, the executive finds no resistance to
impose costs on its adversaries, two outcomes are possible: If the president is successful,
democratic backsliding undermines the opposition and reinforces presidential control
over government institutions. If the opposition appeals to the military as a preemptive
move, the result is democratic breakdown by a civilian–military coalition.
Therefore, our main hypothesis holds that concentration of institutional power by the
executive leads to democratic instability. Two alternative causal mechanisms produce
this result: an erosion of the rights of the opposition, led by the executive, or regime
breakdown, led by the opposition in alliance with the military.

Measuring the institutional hegemony of the executive

To test this hypothesis, our analysis covers all democratic presidential systems in 18
Latin American countries between 1925 and 2016 (Argentina, Bolivia, Brazil, Chile,
Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Hon-
duras, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay and Venezuela). The
dataset starts in 1925 because comparative data on legislatures are not available for
earlier years. Our sample includes all cases coded as democracies or semi-democracies
by Mainwaring and Pérez-Liñán.24 The units of analysis are country-years (N = 927)
and our dichotomous dependent variable captures the breakdown of democracy into
an authoritarian regime.25 To perform robustness tests, in later sections we distinguish
between coups (episodes of breakdown that overthrow the president) and incumbent
takeovers (episodes of breakdown in which the president remains in power).
Our main explanatory variable, executive hegemony over other institutions, is
measured through four indicators. Two indicators measure the president’s influence
on congress and the remaining two, its influence on the judiciary. In this section, we
describe the general characteristics of the indicators and their aggregation into compo-
site indices.
The first indicator captures the percentage of seats controlled by the president’s party
(or the party’s faction aligned with the president, in cases of factional parties) in the
lower (or only) house of congress. The second indicator reflects the percentage of
612 A. PÉREZ-LIÑÁN ET AL.

seats controlled by the president’s coalition, an indirect form of control that requires the
collaboration of other allies. When a single party supports the president in congress,
both indicators take on the same value. The information on the composition of congress
was obtained from an original database compiled for this project.26
The third indicator reflects the percentage of justices that joined the supreme court
or constitutional court during the president’s term of office. For example, if six magis-
trates out of ten entered the court under the incumbent president (during the current
term or a previous term, if the president has been reelected), the indicator reflects that
the president and his or her allies in congress had the opportunity to influence the
nomination of 60% of the court. If there is a constitutional tribunal separate from
the supreme court, the indicator captures the average value for both courts. The
fourth indicator, similar to the previous one, reflects the percentage of high magistrates
incorporated to the supreme court or constitutional tribunal during any government led
by the president’s party. This variable captures an indirect form of influence through
other party leaders, e.g. past presidents who nominated the justices. Information on jus-
tices was coded from secondary sources.27
Our coding of indicators follows simple rules to minimize endogeneity concerns.
When we observe parliamentary elections in a year marked by a coup, our indicators
of legislative support reflect the new composition of congress only if the election’s
date is prior to the date of the coup. In this case, we assume that the results of the elec-
tion propelled the action of conspirators. If, by contrast, the election took place under
the new regime, the legislative data for this year correspond to the previous election.
When there is a democratic breakdown without a change of government – in the
case of an incumbent takeover – data on the composition of congress always reflect
the situation on January 1. Similarly, we assess the composition of supreme courts as
of January 1. In this way, judges appointed in the aftermath of a coup are not included
in the hegemony index for the outgoing president.
The indicators related to the judiciary measure alignment with the president (or the
ruling party) based on the timing of justices’ nominations, but these indicators do not
reflect the nomination procedure. The effects of formal nomination procedures – e.g. if
the executive needs a majority in the senate to approve the nomination of judges, or if
congress nominates without the president’s formal intervention – depend largely on
executive control over congress, reflected by the first two indicators. Moreover, Latin
American presidents often bypass formal rules to force the resignation of opposition
judges.28
Each of the four indicators measures executive hegemony imperfectly. For example,
the degree of party control over congress varies with the party system, since more frag-
mented systems produce narrower majorities for the executive. Similarly, the possibility
of nominating judges varies with the justices’ terms in office: constitutional judges with
short terms are renewed more frequently than judges with life terms. Given the presence
of measurement error in all items, it is desirable to aggregate the four indicators into an
index.
To minimize validity problems related to the individual indicators, we combine the
four items into two alternative measures of presidential hegemony:

(1) The first index – easier to interpret – is an arithmetic mean of the four indicators.
Given that the four items have equal weight in the construction of this measure, we
refer to this as the unweighted index of presidential hegemony. The average of the
 DEMOCRATIZATION 613

Table 1. Components of the index of executive hegemony.

Type of Mean
Institution control Item (%) Load Uniqueness
Legislature Direct Percentage of seats controlled by the president’s 53 0.767 0.412
party (faction).
Legislature Indirect Percentage of seats controlled by the president’s 64 0.704 0.503
coalition.
Judiciary Direct Percentage of justices nominated by the president. 46 0.537 0.711
Judiciary Indirect Percentage of justices nominated by ruling party 63 0.616 0.620
administrations.

index for cases of democracy and semi-democracy is 47.5%, with a standard devi-
ation of 17.9.
(2) To accommodate the fact that the mean and variance of the four items are not
the same, the second index reflects the result of a factor analysis (non-rotated
components). Since each item has a different load on the extracted factor, we
refer to this measure as the weighted index of presidential hegemony. The
information on the load of each indicator and its unique variance is reported
in Table 1.

Figure 1. Executive hegemony and democratic breakdown, 1925–2016.

614 A. PÉREZ-LIÑÁN ET AL.

The estimation included all available cases (democratic and authoritarian, N = 1459) to
maximize the variance of the components. In this complete sample, the weighted index
has an average of 0 and a standard deviation of 0.89. In the sub-sample of democracies
and semi-democracies used in our empirical analysis, the index has a mean of −0.37
and a standard deviation of 0.68. In order to establish the robustness of our findings,
the empirical analysis presented below compares the results using both indices, and
also analyzes the effect of the four indicators separately. The results are highly consistent
regardless of the aggregation procedure. As we document below, the indicators that
capture the president’s direct control over other branches show more consistent effects
on democratic instability than the indicators that reflect indirect control.
Figure 1 shows the annual values of the non-weighted index (a value of 100 indicates
executive hegemony) for each country, during periods of democracy (and semi-democ-
racy) between 1925 and 2016. Vertical lines mark years of democratic breakdown. The
figure suggests that longer democratic sequences (for example, Costa Rica since 1948,
Chile 1932–1973) tend to occur in systems with moderate levels of presidential
control. Democracy can break down at low levels of presidential hegemony (e.g.
Uruguay in 1973, Peru in 1992), but systems with high degrees of executive hegemony
(e.g. Argentina and Bolivia in the 1950s) have shorter lifespans. Moreover, breakdowns
often take place when presidential hegemony is on the rise.

Alternative explanations
Our explanatory models incorporate additional variables to control for alternative
explanations of democratic instability. To assess Shugart and Carey’s argument, the
models include a measure of the president’s constitutional powers over legislation, ela-
borated by Gabriel Negretto.29 To reflect Mainwaring’s argument, they include the
effective number of parties in the lower house (negatively correlated with the hegemony
index, r = −0.43).30 The models also include the percentage of democracies in the
region, to address democratic diffusion,31 the number of anti-government demon-
strations,32 per capita GDP (natural log), and the rate of economic growth.33 To
account for long-term historical legacies, we incorporate the average Polity score for
the century preceding our sample, 1825–1924.34

Results
Given the structure of the data, we estimate a discrete-time duration model, based on a
probit estimator with random frailties by country. Table 2 compares the results of the
analysis using the weighted and non-weighted indices. Models 2.1 and 2.2 incorporate
only the index measuring executive hegemony. Models 2.3 and 2.4 incorporate the
remaining control variables.
The results are highly consistent in all models and under all specifications: the con-
centration of institutional power in the executive emerges as one of the main factors
linked to democratic breakdown. Against Shugart and Carey, presidents with stronger
constitutional powers have experienced a lower risk of breakdown.35 Constitutions
adopted in the late twentieth century, a period of greater democratic stability, often
granted presidents more influence over the legislative process. The only additional vari-
able contributing to democratic survival is the percentage of democracies in the region.
 DEMOCRATIZATION 615

Table 2. Models of democratic breakdown, 1925–2016.

(2.1) (2.2) (2.3) (2.4)
No controls, No controls, Controls, Controls,
unweighted weighted unweighted weighted
Executive hegemony 0.01** 0.29** 0.01** 0.29**
(0.00) (0.11) (0.01) (0.14)
Constitutional −0.01** −0.01**
powers (0.01) (0.01)
ENP (House) 0.10 0.11
(0.06) (0.07)
Region, t−1 −0.02** −0.02**
(0.00) (0.00)
Demonstrations 0.01 0.01
(0.05) (0.05)
Per capita GDP, ln −0.15 −0.14
(0.12) (0.12)
Growth, t−1 0.01 0.01
(0.02) (0.02)
Polity, 1825–1924 −0.05 −0.05
(0.03) (0.03)
Constant −2.29** −1.69** −0.39 0.15
(0.24) (0.09) (0.94) (0.87)
Countries 18 18 18
18
N 927 927 927 927
Frailty variance 0.074 0.107 0.001 0.002
Log-likelihood −152.2 −151.9 −133.7 −134.1
Entries are probit coefficients (standard errors). *p < 0.1, **p < 0.05.

Coefficients for other predictors – party fragmentation, mass protests, economic devel-
opment, growth, and democratic legacies – are statistically insignificant.
To evaluate the substantive effect of presidential hegemony, Figure 2 simulates the
risk of democratic breakdown at different levels of the index, based on the parameters
estimated in the four models of Table 2. The simulations keep control variables at their
observed values and alter the measures of hegemony to generate a prediction. The four
panels reflect a consistent pattern: the probability of breakdown increases with the
power of the executive, ranging between approximately 2% and 8% per year, or an
expected regime lifetime of roughly half a century when the president is weak versus
just 13 years when the president is dominant.
The results presented in Table 2 invite multiple interpretations. In rest of this
section, we present robustness tests to rule out four alternative readings: (1) that the
effect of presidential hegemony is non-linear; (2) that presidential hegemony is
endogenous to democratic backsliding; (3) that the main causal mechanism is not
incumbent takeovers; and (4) that presidential control over different institutions has
different consequences for democracy.

Non-linear effects
The classical argument about the perils of presidentialism, presented in the first section,
suggests that an excessively weak executive can also be a source of democratic instabil-
ity.36 Thus, the relationship between executive hegemony and the risk of breakdown
could be non-linear, with greater risk for the regime when the executive is very
strong and when it is very weak.
616 A. PÉREZ-LIÑÁN ET AL.

Figure 2. Risk of democratic breakdown at different levels of executive hegemony.

To explore this possibility, we re-estimate the models including a quadratic trans-

formation of the indices of presidential hegemony. If the thesis on the perils of pre-
sidentialism is correct, we should observe a reduction in the risk for democracy when
the president controls moderate majorities in the legislative and judicial branches.
Table 3 presents the results of this analysis. The results do not support this interpret-
ation, since the coefficient for the first-degree term is always positive.
Although the coefficients for executive hegemony are not significant in models 3.1 and
3.3, it is not possible to evaluate their statistical significance directly because, under the
non-linear specification, the primary coefficient captures the marginal effect of presiden-
tial control when the level of hegemony is zero. In the case of the non-weighted index
(models 3.1 and 3.3), a value of 0 means that the president has no allies in congress or
the judiciary, an unusual situation. In the case of the weighted index (models 3.2 and
3.4), a value of 0 indicates that the president approaches the average value of the
sample. This difference explains the apparent discrepancies in statistical significance,
given that the marginal effect changes nonlinearly with the independent variable.
To assess the substantive effect of presidential hegemony under this non-linear spe-
cification, Figure 3 presents the expected risk of breakdown based on the four models.
The substantive predictions of the non-linear specification are very similar to those pre-
sented in Figure 2, confirming that the risk to democracy grows monotonically with the
president’s control of other institutions. Weak presidents do not produce particularly
unstable democracies, while overly strong presidents seem to destabilize the regime.

Endogeneity and reverse causality

A second interpretation of the results in Table 2 suggests that the association between
presidential hegemony and democratic instability is the result of reverse causality. To
 DEMOCRATIZATION 617

Table 3. Models with quadratic effects.

(3.1) (3.2) (3.3) (3.4)
No controls, No controls, Controls, Controls,
unweighted weighted unweighted weighted
Executive hegemony 0.00 0.31** 0.01 0.30**
(0.02) (0.11) (0.02) (0.14)
Executive 0.00 0.11 0.00 0.06
hegemony2 (0.00) (0.12) (0.00) (0.12)
Constitutional −0.01** −0.01**
powers (0.01) (0.01)
ENP (House) 0.10 0.11
(0.06) (0.07)
Region, t−1 −0.02** −0.02**
(0.00) (0.00)
Demonstrations 0.01 0.01
(0.05) (0.05)
Per capita GDP, ln −0.15 −0.13
(0.12) (0.12)
Growth, t−1 0.01 0.01
(0.02) (0.02)
Polity, 1825–1924 −0.05 −0.05
(0.03) (0.03)
Constant −2.14** −1.74** −0.30 0.09
(0.55) (0.10) (1.05) (0.88)
Countries 18 18 18 18
N 927 927 927 927
Frailty variance 0.069 0.098 0.001 0.002
Log-likelihood −152.2 −151.4 −133.6 −134.0
Note: Entries are probit coefficients (standard errors). *p < 0.1, **p < 0.05.
The overall marginal effect of executive hegemony is significant (p < .05) in 3.1 and 3.3.

Figure 3. Risk of breakdown with non-linear effects.

618 A. PÉREZ-LIÑÁN ET AL.

understand this problem, consider a president who manipulates the electoral process,
displaces the opposition in congress, and purges the Supreme Court. This “takeover”
simultaneously produces the consolidation of the president’s hegemony and the
demise of democracy.37 It is clear that the president’s actions undermine the democratic
regime, but it is not clear whether the president’s hegemony is a “cause” of breakdown
or just the endpoint of democratic erosion. Given that the process of backsliding is pro-
gressive, growing levels of executive hegemony can result from a sequence by which the
ruling party dismantles the regime. If so, rising presidential control over other insti-
tutions would be the product, not the cause, of democratic backsliding.
This interpretation suggests a potential endogeneity problem in Table 2. Before
addressing this issue from an econometric perspective, it is nevertheless necessary to
clarify the nature and the scope of the problem. The problem described in the pre-
vious paragraph potentially affects executive power grabs, especially when backslid-
ing occurs progressively. There is no similar concern for coups, given that
opposition forces overthrow the regime to oust the elected president. Executive con-
centration of power may activate a preemptive reaction of the opposition, but the
institutional power of the ousted president does not increase with the end of
democracy.
Most of the events in our sample (26 of 38 breakdowns) represent coups that rule out
the possibility of reverse causality. The remaining 12 cases are incumbent takeovers in
which the president remains in office after the end of democracy. Of these, six cases cor-
respond to self-coups in which the end of the regime occurred in a visible and abrupt
manner (Colombia 1949, Ecuador 1946 and 1970, Peru 1992, Uruguay 1933 and 1973).
Only six cases correspond to instances of progressive backsliding, in which the exact
moment that marks the end of democracy is more difficult to pinpoint (the Dominican
Republic in 1928, Honduras approximately in 1935, Uruguay in 1942, Argentina circa
1951, Venezuela circa 2009, and Nicaragua circa 2011). Therefore, potential endogene-
ity affects only a small number of events in our sample.
To address this possibility, we estimate a survival model with instrumental variables.
Our instrument captures the country’s experience with presidential hegemony before
1925, regardless of the conditions leading to democratic breakdown later in the twen-
tieth century. We used the Banks database to extract a historical measure of checks and
balances: the number of years between 1825 and 1924 that the country had a legislative
body with some degree of effective power.38 Given that this historical variable has fixed
values for each country, its behaviour is independent of the conditions for democratic
breakdown after 1925. However, because institutional legacies are enduring, this vari-
able – like the average Polity score for 1825–1924, already included in the model –
allows us to predict the average level of presidential hegemony for each country in
the twentieth century. For greater efficiency in the estimation, we also include an inter-
action between the two legacy variables, following Lewbel.39
The top panel of Table 4 shows the second-stage models predicting democratic
breakdown, while the bottom panel shows the first-stage models that predict the
degree of executive hegemony (unweighted index) after 1924. The results in the
bottom panel indicate that countries with an effective congress in the nineteenth
century also generated executives with less institutional power in the twentieth
century. They also show that executives with greater formal powers tend to achieve
greater control over other institutions, and that presidents are weaker in systems
with a greater number of parties, greater economic development, and a longer
 DEMOCRATIZATION 619

Table 4. Models with endogenous hegemony (unweighted).

(4.1) (4.2) (4.3)
Breakdowns Coups Takeovers
Second stage
Executive hegemony 0.06** 0.05 0.06**
(0.01) (0.03) (0.01)
Constitutional powers −0.01** −0.02* −0.00
(0.00) (0.01) (0.00)
ENP (House) 0.33** 0.32** 0.30**
(0.05) (0.13) (0.07)
Region, t−1 −0.01* −0.01 −0.01
(0.01) (0.01) (0.01)
Demonstrations −0.02 0.01 −0.07
(0.03) (0.05) (0.08)
Per capita GDP, ln 0.06 0.03 0.05
(0.11) (0.20) (0.14)
Growth, t−1 0.00 0.02 −0.02
(0.01) (0.03) (0.02)
Polity, 1825–1924 0.02 0.02 0.01
(0.03) (0.06) (0.05)
Constant −4.48** −3.78 −5.20**
(1.10) (2.86) (0.92)
First stage
Constitutional powers 0.10** 0.10** 0.11**
(0.03) (0.03) (0.03)
ENP (House) −5.15** −5.14** −5.10**
(0.35) (0.36) (0.35)
Region, t−1 −0.01 −0.02 −0.02
(0.03) (0.03) (0.03)
Demonstrations 0.41 0.41 0.41
(0.28) (0.28) (0.28)
Per capita GDP, ln −1.86** −1.85** −1.84**
(0.82) (0.82) (0.82)
Growth, t−1 0.12 0.12 0.13
(0.14) (0.14) (0.14)
Polity, 1825–1924 −0.71** −0.70** −0.69**
(0.20) (0.20) (0.20)
Effective, 1825–1924 −0.05* −0.06* −0.06**
(0.03) (0.03) (0.03)
Lewbel instrument −0.02** −0.02** −0.02**
(0.01) (0.01) (0.01)
Constant 77.10** 77.10** 77.20**
(5.63) (5.63) (5.63)
N 927 927 927
Log-likelihood −3998.116 −3960.726 −3921.282
Note: Entries for are maximum likelihood coefficients for probit (second stage) and linear (first stage) simultaneous
estimator. *p < 0.1, **p < 0.05.

democratic history. We include the prediction of first-stage models as “proxy” for the
degree of executive hegemony (purged of potential endogeneity) in the second stage
of the estimation.
Model 4.1 in the upper panel of Table 4 shows the results of the analysis with instru-
mental variables, using the instrumented proxy for the non-weighted index of hege-
mony. These results confirm the findings presented in Tables 2 and 3: presidential
hegemony significantly increases the risk of democratic instability, even when we
rule out possible effects of reverse causality. It is important to note that the coefficient
is substantially greater in model 4.1 than in model 2.3, suggesting that the causal effect is
underestimated in Table 2.40
620 A. PÉREZ-LIÑÁN ET AL.

The thesis presented in this article argues that executive hegemony produces demo-
cratic instability because presidents who control other institutions have greater capacity
to neutralize the opposition and articulate a power grab. In some cases, this risk pro-
duces a strategic reaction from the opposition, which conspires preemptively to over-
throw the president. If this interpretation is correct, presidential hegemony should
increase the risk of breakdown directly, augmenting the risk of an incumbent takeover,
and indirectly – and less consistently – augmenting the risk of a coup.
To explore the validity of our interpretation, models 4.2 and 4.3 alter the coding of the
dependent variable to capture cases of coups (N = 26) and takeovers (N = 12). The other epi-
sodes of breakdown (coups or power grabs, respectively) are treated as censored. The size of
the coefficient for the endogenous variable is similar in both models. It fails to achieve con-
ventional levels of significance (p = .108) in the model for coups but it is strongly significant
in the model for takeovers (p < .001). This pattern suggests that executive hegemony consist-
ently facilitates takeovers and it may prompt some coups – for example, in Venezuela in
2002 and Turkey in 2016 – even though not every coup follows this logic.

Different institutions and the validity of the composite index

The analysis presented in this article employs our aggregate measures of presidential
hegemony. However, theories discussed in the first section anticipate contradictory
effects when the president exercises influence on congress and on the judiciary. Part
of literature argues that “coordination” with congress facilitates democratic governance,
while another segment argues that a reduction of judicial independence undermines the
rule of law. To verify if our composite indices hide contradictory effects for congress and
the judiciary, Table 5 replicates Model 2.1 (from Table 2), substituting the composite
index for its four individual components.
Results in Table 5 suggest two important conclusions. First, presidential hegemony
over congress – and not only over the judiciary – has the potential to create instability.
The coefficients in models 5.1 and 5.2, capturing executive influence over congress, are
positive. Their levels of statistical significance are stronger than those for estimates in
models for the courts. Moderately-sized legislative majorities may prevent executive-
legislative deadlock and promote governability, but very large legislative asymmetries
allow presidents to alter constitutional rules and to purge independent courts, com-
pounding the threat of instability.41
Second, Table 5 suggests that a composite index offers a more efficient measure of
presidential hegemony than every individual measure in isolation, because all indicators

Table 5. Replication of Model 2.1 with index components.

(2.1) (5.1) (5.2) (5.3) (5.4)
Unweighted Congress: Congress: Court: Court:
index Party Coalition President Party
Hegemony 0.01** 0.01** 0.01* 0.00* 0.00
measure [0.01] [0.03] [0.05] [0.08] [0.13]
Constant −2.29** −2.13** −2.23** −1.93** −1.97**
[0.00] [0.00] [0.00] [0.00] [0.00]
Countries 18 18 18 18 18
N 927 927 927 927 927
Frailty variance 0.074 0.117 0.161 0.002 0.097
Log-likelihood −152.2 −153.0 −153.4 −153.9 −154.2
Note: Entries are probit coefficients [p-values]. *p < 0.1, **p < 0.05.
 DEMOCRATIZATION 621

contain a certain amount of noise. The combination of the four items allows us to
reduce error in the assessment of the latent variable, the president’s institutional hege-
mony. Thus, the stronger p-value for the composite index, reported in the first column,
reflects a more consistent effect on democratic instability than the effect of any single
indicator.

Conclusions
The president’s hegemony over other branches of government – but not the president’s
formal constitutional powers – represents a major threat to democratic stability.
Although the traditional literature on the “perils” of presidentialism anticipated that
executive influence over the legislature would reduce the risk of institutional deadlock
and democratic instability, the most recent literature on executive-judicial relations
suggests that executive control over the judiciary represents a source of democratic
backsliding. Our empirical analysis shows that greater direct control of the executive
over other branches expands the risk of democratic breakdown. This result holds for
a considerable number of cases (18 countries) over an extended period (1925–2016).
The finding is robust when we consider possible non-linear effects of the independent
variable, and when we discount potential endogeneity in the relation between presiden-
tial hegemony and democratic backsliding.
The literature on executive-legislative relations often ignored that deadlock rep-
resents an institutional equilibrium, and therefore protects actors from an undesirable
change of the status quo. The veto exercised by constitutional courts represents a similar
outcome. Concentration of power in the executive branch eliminates those constraints
and facilitates unilateral implementation of controversial policies. It also allows the gov-
ernment to encroach on the rights of the opposition.
A growing debate on democratic backsliding highlights the role of “executive aggrand-
izement” and the seizure of power by incumbents.42 Our study contributes to this debate
by showing that unified control over the three branches of government effectively enables
rulers to undermine the democratic process. Power asymmetries encourage incumbents
to redefine the boundaries of the constitution.43 When the executive prevails, there is
democratic backsliding, in line with the concerns expressed by the novel literature on
the rule of law. When, on the contrary, the opposition joins the military to launch a pre-
emptive coup, the result is breakdown, in line with the concerns of the classical literature.
These results call for a reevaluation of institutional environments marked by divided
government or recalcitrant courts, which are often labeled “dysfunctional” because of
their inability to produce policy change. Our findings suggest instead that voters should
strive to balance the distribution of institutional powers as a way to protect democracy.
But the results also point to the limitations of this study and open important research ques-
tions for the future. First, “dysfunctional” institutions represent a safeguard when presi-
dents host hegemonic ambitions, but they also impose policy deadlock for governments
that do not purse hegemony. Further research should explore the conditions under
which the risk of executive takeover justifies the costs of policy paralysis. Second, our
study has focused on presidential regimes, where executive hegemony is easy to identify.
However, recent experiences suggest that parliamentary (e.g. Turkey 2003–14) and semi-
presidential (e.g. Hungary since 2010) regimes confront similar risks. Further research
must determine whether our concern with presidential hegemony should extend into a
broader concern with executive hegemony under any type of constitution.
622 A. PÉREZ-LIÑÁN ET AL.

Notes
1. Svolik, “Which Democracies Will Last?”; Bermeo, “On Democratic Backsliding.”
2. Svolik, “Which Democracies Will Last?”
3. Ibid., 735.
4. Bermeo, “On Democratic Backsliding.”
5. Ibid., 10.
6. Levitsky y Ziblatt, How Democracies Die. See also Corrales, Fixing Democracy.
7. Corrales, “Why Polarize?”, 70.
8. Mainwaring and Shugart, Presidentialism and Democracy.
9. Linz, “The Perils of Presidentialism”; Linz, “Presidential or Parliamentary Democracy.”
10. Mainwaring, “Presidentialism, Multipartism, and Democracy.”
11. Valenzuela, “Latin America: Presidentialism in Crisis.”
12. Alston et al., Brazil in Transition; Altman, “The Politics of Coalition Formation”; Amorim Neto,
“Presidential Cabinets, Electoral Cycles”; Carey, “Parties, Coalitions, and the Chilean Congress”;
Chasquetti, Democracia, presidencialismo y partidos; Deheza, Gobiernos de coalición; Lanzaro,
“Tipos de presidencialismo”; Martínez-Gallardo, “Out of the Cabinet”; Mejía Acosta, Informal
Coalitions and Policymaking; Ollier, Las Coaliciones Políticas en Argentina.
13. Helmke, Institutions on the Edge; Mainwaring and Shugart, Presidentialism and Democracy;
Negretto, Making Constitutions; Shugart and Carey, Presidents and Assemblies.
14. Shugart and Carey, Presidents and Assemblies.
15. Mainwaring and Pérez-Liñán, Democracies and Dictatorships.
16. Helmke, Institutions on the Edge.
17. Cheibub, Presidentialism, Parliamentarism, and Democracy.
18. Helmke, Courts under Constraints, Kapiszewski and Taylor, “Doing Courts Justice?”, Pérez-
Liñán and Castagnola, “Presidential Control of High Courts.”
19. Gibler and Randazzo, “Testing the Effects of Independent Judiciaries”; Reenock, Staton, and
Radean, “Legal Institutions and Democratic Survival.”
20. Reenok, Staton and Radean, “Legal Institutions and Democratic Survival,” 491.
21. Gibler and Randazzo, “Testing the Effects of Independent Judiciaries,” 699.
22. Toro-Maureira and Arellano-González, “The Architecture of Governments”; Garcé, “Hacia una
teoría ideacional”; Cheibub et al., “Latin American Presidentialism.”
23. Tsebelis, Veto Players.
24. Mainwaring and Pérez-Liñán, Democracies and Dictatorship; Mainwaring and Pérez-Liñán,
“Cross Currents in Latin America.”
25. Mainwaring and Pérez-Liñán classify as democracies regimes that meet four characteristics as of
December 31 of each year: (1) president and congress are elected in free and fair elections, (2)
there is a “universal” franchise among the adult population, (3) civil liberties are protected, and
(4) the elected government is not controlled by the military or external agents. A minor violation
of these principles leads to the codification of the case as semi-democracy. A significant violation
of any of these principles transforms the regime into authoritarian, and marks the event of inter-
est in our analysis (breakdown). Cases leave the sample once they become authoritarian. See
Mainwaring and Pérez-Liñán, Democracies and Dictatorship.
26. Available at https://doi.org/10.7910/DVN/VSGRJH.
27. Pérez-Liñán and Castagnola, “Judicial Instability and Endogenous”; Pérez-Liñán and Castag-
nola, “Replication Data for: Judicial Instability.”
28. Castagnola, “I Want It All”; Castagnola, Manipulating Courts in New Democracies; Helmke,
Institutions on the Edge.
29. Shugart and Carey, Presidents and Assemblies; Negretto, Making Constitutions.
30. Mainwaring, “Presidentialism, Multipartism, and Democracy.”
31. Weyland, Making Waves: Democratic Contention.
32. Banks and Wilson, Cross-National Time-Series.
33. Przeworski et al., Democracy and Development.
34. Gerring et al., “Democracy and Economic Growth”; Mainwaring and Pérez-Liñán, Democracies
and Dictatorship; Martínez, “Presidential Survival in South America.”
35. Shugart and Carey, Presidents and Assemblies.
36. See also Helmke, Institutions on the Edge.
 DEMOCRATIZATION 623

37. Svolik, “Which Democracies Will Last?”

38. The original variable in the Banks database is polit13. We consider the legislature ineffective
when polit13 < 2 (no legislature, rubber stamp, turmoil, no meetings) and effective when
polit13 > 1 (executive does not dominate legislature, legislature has substantial authority).
Our variable counts the number of years that each country fulfills this condition between
1825 and 1924. Wilson, Cross-National Time-Series Data Archive.
39. Following Lewbel the second instrument is (polity – (m)polity) * (effective – (m)effective) where
(m) indicates the average of each variable for the sample. See Lewbel, “Constructing Instruments
for Regressions.”
40. Model 4.1 also hints at the ambiguous effect of the president’s constitutional powers. The
second-stage equation suggests that stronger constitutional powers may reduce the risk of break-
down by promoting governability. The first-stage equation, however, suggests that strong con-
stitutional powers promote breakdown indirectly, by fostering presidential hegemony.
41. Chavez, The Rule of Law; Castagnola, Manipulating Courts in New Democracies; Corrales, Fixing
Democracy.
42. Bermeo, “On Democratic Backsliding.” Corrales, “Why Polarize?”; Levitsky and Ziblatt, How
Democracies Die; Svolik, “Which Democracies Will Last?”
43. Corrales, Fixing Democracy.

Acknowledgments
This work was supported by Uruguay’s National Research and Innovation Agency (ANII) under Grant
FCE_1_2014_1_103565. Previous versions of this article were presented at the 9th Latin American
Congress of Political Science (Montevideo, July 2017) and the 2nd International Congress of Political
Science (Popayán, Colombia, April 2018). We are grateful to Julián Caicedo, Adolfo Garcé, Gabriel
Negretto, John Polga-Hecimovich, and two Democratization reviewers for their comments.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by Uruguay’s National Research and Innovation Agency (ANII) [Agencia
Nacional de Investigación e Innovación] under grant FCE_1_2014_1_103565.

Bibliography
Alston, Lee J., Marcus André Melo, Bernardo Mueller, and Carlos Pereira. Brazil in Transition: Beliefs,
Leadership, and Institutional Change. Princeton, NJ: Princeton University Press, 2016.
Altman, David. “The Politics of Coalition Formation and Survival in Multi-Party Presidential
Democracies - The Case of Uruguay, 1989-1999.” Party Politics 6, no. 3 (2000): 259–283.
Amorim Neto, Octavio. “Presidential Cabinets, Electoral Cycles, and Coalition Discipline in Brazil.” In
Legislative Politics in Latin America, edited by Scott Morgenstern, and Benito Nacif, 48–78.
Cambridge: Cambridge University Press, 2002.
Banks, Arthur, and Kenneth Wilson. Cross-National Time-Series Data Archive 2016 Edition
CNTSDATA. Accessed July 5, 2018. https://www.cntsdata.com/.
Bermeo, Nancy. “On Democratic Backsliding.” Journal of Democracy 27, no. 1 (2016): 5–19.
Bill Chavez, Rebecca. The Rule of Law in Nascent Democracies. Judicial Politics in Argentina. Stanford:
Stanford University Press, 2004.
Carey, John M. “Parties, Coalitions, and the Chilean Congress in the 1990s.” In Legislative Politics in
Latin America, edited by Scott Morgenstern, and Benito Nacif, 222–253. Cambridge: Cambridge
University Press, 2002.
Castagnola, Andrea. “I Want It All, and I Want It Now: The Political Manipulation of Argentina’s
Provincial High Courts.” Journal of Politics in Latin America 4, no. 2 (2012): 39–62.
624 A. PÉREZ-LIÑÁN ET AL.

Castagnola, Andrea. Manipulating Courts in New Democracies: Forcing Judges off the Bench in
Argentina. New York, NY: Routledge, 2018.
Chasquetti, Daniel. Democracia, presidencialismo y partidos políticos en América Latina: Evaluando la
“difícil combinación”. Montevideo: Cauce, 2008.
Cheibub, José Antonio. Presidentialism, Parliamentarism, and Democracy. New York: Cambridge
University Press, 2007.
Cheibub, José Antonio, Zachary Elkins, and Tom Ginsburg. “Latin American Presidentialism in
Comparative and Historical Perspective.” Texas Law Review 89, no. 7 (2010): 1707–1740.
Corrales, Javier. Fixing Democracy. New York: Oxford University Press, 2018.
Corrales, Javier. “Why Polarize? Advantages and Disadvantages of a Rational-Choice Analysis of
Government-Opposition Relations under Hugo Chávez.” In The Revolution in Venezuela: Social
and Political Change under Chávez, edited by Thomas Ponniah, and Jonathan Eastwood, 67–97.
Cambridge, MA: Harvard University Press, 2011.
Deheza, Grace Ivana. Gobiernos de coalición en el sistema presidencial: América del Sur. Florencia:
Departamento de Ciencias Políticas y Sociales, Instituto Universitario Europeo, 1997.
Garcé, Adolfo. “Hacia una teoría ideacional de la difusión institucional. La adopción y adaptación del
presidencialismo en América Latina durante el siglo XIX.” Revista Española de Ciencia Política 44
(2017): 13–41.
Gerring, John, Phillip J. Bond, William T. Barndt, and Carola Moreno. “Democracy and Economic
Growth: A Historical Perspective.” World Politics 57, no. 3 (2005): 323–364.
Gibler, Douglas M., and Kirk A. Randazzo. “Testing the Effects of Independent Judiciaries on the
Likelihood of Democratic Backsliding.” American Journal of Political Science 55, no. 3 (2011):
696–709.
Helmke, Gretchen. Courts under Constraints: Judges, Generals, and Presidents in Argentina. Cambridge:
Cambridge University Press, 2005.
Helmke, Gretchen. Institutions on the Edge: The Origins and Consequences of Inter-Branch Crises in
Latin America. New York: Cambridge University Press, 2017.
Kapiszewski, Diana, and Matthew M. Taylor. “Doing Courts Justice? Studying Judicial Politics in Latin
America.” Perspectives on Politics 6, no. 4 (2008): 741–767.
Lanzaro, Jorge. “Tipos de presidencialismo y modos de gobierno en América Latina.” In Tipos de
Presidencialismo y Coaliciones Políticas en America Latina, edited by Jorge Lanzaro, 15–49.
Buenos Aires: CLACSO, 2001.
Levitsky, Steven, and Daniel Ziblatt. How Democracies Die. New York: Crown, 2018.
Lewbel, Arthur. “Constructing Instruments for Regressions with Measurement Error When no
Additional Data are Available, with An Application to Patents and R&D.” Econometrica 65, no. 5
(1997): 1201–1213.
Linz, Juan J. “Presidential or Parliamentary Democracy: Does It Make a Difference?” In The Failure of
Presidential Democracy (Vol. 1), edited by Juan José Linz, and Arturo Valenzuela, 3–87. Baltimore:
JHU Press, 1994.
Linz, Juan J. “The Perils of Presidentialism.” Journal of Democracy 1, no. 1 (1990): 51–69.
Mainwaring, Scott. “Presidentialism, Multipartism, and Democracy: The Difficult Combination.”
Comparative Political Studies 26, no. 2 (1993): 198–228.
Mainwaring, Scott, and Aníbal Pérez-Liñán. “Cross-Currents in Latin America.” Journal of Democracy
26, no. 1 (2015): 114–127.
Mainwaring, Scott, and Aníbal Pérez-Liñán. Democracies and Dictatorships in Latin America:
Emergence, Survival, and Fall. Cambridge: Cambridge University Press, 2013.
Mainwaring, Scott, and Matthew Soberg Shugart, eds. Presidentialism and Democracy in Latin America.
New York: Cambridge University Press, 1997.
Martínez, Christopher A. “Presidential Survival in South America: Rethinking the Role of Democracy.”
International Political Science Review 38, no. 1 (2017): 40–55.
Martínez-Gallardo, Cecilia. “Out of the Cabinet.” Comparative Political Studies 45, no. 1 (2012): 62–90.
Mejía Acosta, Andrés. Informal Coalitions and Policymaking in Latin America: Ecuador in Comparative
Perspective. New York: Routledge, 2009.
Negretto, Gabriel L. Making Constitutions. Presidents, Parties, and Institutional Choice in Latin
America. Cambridge: Cambridge University Press, 2013.
Ollier, María Matilde. Las Coaliciones Políticas en Argentina - El Caso de la Alianza. Buenos Aires:
Fondo de Cultura Económica, 2001.
 DEMOCRATIZATION 625

Pérez-Liñán, Aníbal, and Andrea Castagnola. “Judicial Instability and Endogenous Constitutional
Change: Lessons from Latin America.” British Journal of Political Science 46, no. 2 (2016): 395–416.
Pérez-Liñán, Aníbal, and Andrea Castagnola. “Presidential Control of High Courts in Latin America: A
Long-Term View (1904-2006).” Journal of Politics in Latin America 1, no. 2 (2009): 87–114.
Perez-Liñán, Aníbal, and Andrea Castagnola. “Replication Data for: Judicial Instability and
Endogenous Constitutional Change: Lessons from Latin America.” (Harvard Dataverse; accessed
July 5, 2018) V1, 2016. doi:10.7910/DVN/POUBC2.
Przeworski, Adam, Michael E. Alvarez, José Antonio Cheibub, and Fernando Limongi. Democracy and
Development. Political Institutions and Well-Being in the World, 1950-1990. Cambridge: Cambridge
University Press, 2000.
Reenock, Christopher, Jeffrey Staton, and Marius Radean. “Legal Institutions and Democratic Survival.”
The Journal of Politics 75, no. 2 (2013): 491–505.
Shugart, Matthew Soberg, and John M. Carey. Presidents and Assemblies: Constitutional Design and
Electoral Dynamics. Cambridge: Cambridge University Press, 1992.
Svolik, Milan W. “Which Democracies Will Last? Coups, Incumbent Takeovers, and the Dynamic of
Democratic Consolidation.” British Journal of Political Science 45, no. 4 (2015): 715–738.
Toro-Maureira, Sergio, and Juan Carlos Arellano-González. “The Architecture of Governments in
Conflict Environments: The Origin and Crystallisation of Presidentialism in Latin America.” In
Institutional Innovation and de Steering of Conflicts in Latin America, 33–56. Colchester: ECPR
Press, 2017.
Tsebelis, George. Veto Players: How Political Institutions Work. Princeton, NJ: Princeton University
Press, 2002.
Valenzuela, Arturo. “Latin America: Presidentialism in Crisis.” Journal of Democracy 4, no. 4 (1993): 3–16.
Wilson, Kenneth A. Cross-National Time-Series Data Archive. Databanks International. Jerusalem,
Israel, 2013. www.databanksinternational.com.
Weyland, Kurt. Making Waves: Democratic Contention in Europe and Latin America since the
Revolutions of 1848. Cambridge: Cambridge University Press, 2014.