Tag Archives: Latin America

Adrián Albala – How Bicameralism patterns the formation and dissolution of coalitions in presidential regimes

This is a guest post by Adrián Albala, University of Brasilia, Brazil. Contact: adrian.albala@gmail.com. It is based on a paper published in the British Journal of Politics and International Relations

In recent years, there have been huge advances in coalition theories, particularly concerning coalitions in parliamentary systems. However, much of the literature dealing with coalitions under presidential regimes has tended limit itself to reproducing and importing models from the work on parliamentary regimes, without considering the particularities of presidentialism.

Three features of presidentialism may interfere with the coalition process. First, the main particularity of presidentialism is the “winner-takes-all” principle. This states that the election determines a clear winner.

This feature makes it almost impossible for there to be any surprise in identifying the president-elect[1]. This is quite different from most parliamentary regimes in which the identification of a clear winner may be more difficult, and the subsequent composition of the government is, often, very hard to predict. Indeed, under parliamentary regimes elections consist of a “photograph” that depicts the strength of each party or coalition in parliament in order to determine which will have a majority for forming a government. Figure 1 sets out this difference between the two processes.

Figure 1: Election processes in parliamentary regimes/ presidential regimes

The example of Belgium in 2011-2012, where negotiations lasted almost a year and a half, during which the country had no formal government, constitutes a paradigmatic example of this feature. However, this is a recurrent feature of parliamentary politics and has happened = again in Belgium (2015), but also in the UK (2010), Ireland (2016), Spain (2016-17), Germany (2017-18), Italy (2013) and Greece (2014-2015).

The second particularity of presidentialism is the principle of the presidential mandate. This implies that both the inauguration and the conclusion of the presidential mandate are settled by the constitution. This supposes that on the day of the inauguration of the presidency the president has to have his government formed. Moreover, the termination of the mandate should not, theoretically, be dependent on a majority (re)alignment in the congress. Recent events in Brazil have shown that this principle is not deterministic: a president can be impeached for political and opportunistic motivations.

Finally, the third particularity of presidential regimes is symmetrical bicameralism. As a matter of fact, bicameral congresses under presidentialism used to be symmetric, i.e. both chambers used to have the same powers and attributes. This is quite different from parliamentary regimes, where most of the upper chambers – except in Italy – have a mere consultative role.

This is of particular importance as bicameralism supposes a two-round procedure in the policymaking process for the president, thus increasing what Lupia and Strøm (2008) call “the shadow of the unexpected”. For this reason, it is reasonable to state that for those polities with a bicameral congress, holding a bicameral majority is a relevant condition for both the policymaking process and coalition governance (Hiroi and Rennó 2014). By the same token, controlling only one of the two chambers by the president might not be sufficient to ensure that policies get approval, or even to guarantee the survival of the coalition.

When considering coalition cabinet formation and dissolution, I argue that this third particularism is strongly linked to the first two. Indeed, for too long, scholars have studied coalitions under presidential regimes as they did under parliamentarism: assuming that the executive needed to look for allies in only one chamber. However, symmetric bicameralism makes such an assumption untenable. In fact, bicameralism, particularly symmetrical bicameralism in presidential regimes, may contain significant constraints for policymaking and coalition duration. Indeed, controlling one of the two chambers may not be sufficient for the president to ensure policy approval.

This misconsideration is particularly true when reviewing the literature about coalition cabinets under presidential regimes. More particularly, an important number of works have modeled the president’s ability to govern based on his/her legislative strength, or on the distribution of portfolios following Gamson’s law (i.e based on the proportionality principal of the strength of each party in the legislative branch), also known as “coalition congruence”. However almost every study dealing with this issue has measured the legislative strength or coalition congruence, based only on their observation of the lower chamber. In other words, almost no study has ever considered the upper chamber (i. e the Senate) as a relevant actor in the coalition process. We need to consider both bicameralism and bicameral majorities as relevant variables for the understanding of coalition cabinets under presidential regimes. The only work on this topic in parliamentary regimes reaches contradictory conclusions (Eppner and Ganghof 2016; Druckman et al. 2005; Diermeier et al. 2007; Druckman and Thies 2002)

This is what this paper tries to explore, focusing on Latin American presidential regimes that have experienced coalition cabinets.

First of all, half of Latin American countries (9/18) have a bicameral legislature. Bicameralism is not a trivial issue. This number is even higher if we compute every government since the third wave of democratization in the region, which began in 1979. Indeed, I have considered 134 governments in the region, i.e., cabinets following electoral processes. Of these, 54.47% (74) were formed in bicameral polities.

Moreover, when focusing on the occurrence of coalition cabinets, the relevance of bicameralism becomes even more central. Indeed, based on a strict but common definition of coalition cabinets (see Albala 2016), I have computed 31 newly formed coalition cabinets since 1980. That is 31 cabinets that were coalitions on the day of the president assumed office (See Table 1).

Of those 31 coalitions, 29 (93,5%) were formed in bicameral polities. Only Ecuadorian coalitions were formed in polities with a unicameral congress. In other words, for every ten coalitions to be formed in Latin American presidential regimes, more than nine occurred in polities with a bicameral congress. Why has no-one ever considered this feature?

The bicameral condition

I stated above that bicameral congresses under presidential regimes used to be symmetric, that is to say that the two houses (House of Deputies and the Senate) used to share similar attributes and powers. Thus, to ensure governability, a president-elect prefers to enjoy a bicameral majority rather than a partial majority (only one chamber) or no majority.

I have stated also the principle of a fixed presidential mandate. The president’s mandate not only concerns the end of the administration but also the beginning. Thus, the process of cabinet formation under presidentialism is limited in time, running from the proclamation of the result of the election to the inauguration day, generally fixed by the constitution. This feature supposes that the president will have formed his/her cabinet by inauguration.

Then, in order to determine how the combination of those two features (bicameralism + fixed mandate) may affect the coalition formation process, I compared the parliamentary strength of the coalitions after the election day of the president, with their strength at inauguration day.

We may, thus, theoretically, expect that presidents-elect who could not get a bicameral majority on the day of their election would seek to enlarge their parliamentary support including newcomers to their electoral alliance.

Results and findings

In Table 2, I identify 29 presidents-elect, comparing their legislative strength at election day with their strength at inauguration day. I simplified the operationalization of the legislative strength into three categories: i) no presidential majority at all ; ii) a legislative majority in one House; and, iii) a bicameral majority.

The data clearly confirms the hypothesis. Indeed, only 37.93% (11/29) of the presidents-elect had a bicameral majority (2) at election day. Nevertheless, the rate raises at 65.6% (19/29) at inauguration day, indicating that 8 presidents-elect proceeded to open negotiations with other parties to form a coalition or enlarge their electoral coalition. In other words, 8 presidents-elect who could not obtain a bicameral majority via the election, decided to include new members before their inauguration in order to get a bicameral majority. Conversely, the rate of minority coalitions (full or partial) fell from 62,06% ( 0= 31,03% + 1= 31.03%) to 34,4% (10,3% +24,13%).

Additionally, among the presidents who failed to obtain bicameral majorities, the first three Chilean presidents since the return of democracy (Aylwin, Frei and Lagos) had to deal with a particular constitutional feature inherited from the Pinochet rule: the existence of 9 designated senators, mostly from the military forces, who prevented the government from reaching a majority in the Senate.

By contrast, the data shows that bicameralism has been a central feature for presidents-elect who were not able to reach a bicameral majority while running alone. Indeed, among the three cases that ran alone on election day, all of them negotiated with new partners and achieved a bicameral majority.

In the cases where the length of time between election day and and inauguration day is longer – Uruguay and Brazil – there was a high degree of coalition enlargement and only one president-elect failed to achieve a bicameral majority (Lula I). However, among the polities with the shortest of time between election day and and inauguration day we can distinguish between Bolivian presidents-elect who have always managed to obtain a bicameral majority and Argentinian presidents-elect, who led coalition cabinets, but who have never enjoyed majorities in the two houses. Therefore, the timing condition deserves further attention.

We also found that in the cases where president-elect won a majority in only one chamber, the chamber in which the president-elect was unable to reach a majority was systematically the Senate. Hence, the upper chamber seems to be harder to conquer for presidents. This finding should also open a new line of investigation.

Finally, we also found that a bicameral majority makes it easier for the coalition governance generally and, thus, constitutes a sufficient, but not necessary, condition for enduring coalition agreements.

In this work, I have highlighted the need to adapt the study of coalition cabinets under presidential regimes to the particularities of such regimes. Indeed, I have shown out that presidentialism has a critical impact on the timing of the coalition formation process. Moreover, bicameralism is a central feature for the presidents-elect. These elements, in turn, open up new fields of study. For instance, no study has ever considered the role of the vice-president. However, vice-presidents can play a key role, especially when when (e.g. Argentina, Uruguay, and Venezuela) they are also the chair of the upper house.

Notes

[1] There are, however, some exceptions. For example, Bolivia used to have a system that could lead to “surprises”. Indeed until 2008, when no candidate reached the absolute majority, the run off used to take place in parliament leading to parliamentary bargains. Sometimes, the president-elect was not the one who won the plurality at the popular election.

References

Albala, A. (2016) Coalitions Gouvernementales et Régime Présidentiel. Sarbruken: Editions Universitaires Europ­éennes.

Diermeier D, Eraslan H, and Merlo A (2007) Bicameralism and Government Formation. Quarterly Journal of Political Science, 2(3): 227–252.

Druckman J N, Martin L, and Thies M (2005) Influence without Confidence: Upper Chambers and Government Formation. Legislative Studies Quarterly, 30(4): 529- 548.

Druckman JN., and Thies M (2002) The Importance of Concurrence: The Impact of Bicameralism on Government Formation and Duration. American Journal of Political Science, 46(4): 760-771.

Eppner S, and Ganghof S (2016) Institutional veto players and cabinet formation: The veto control hypothesis reconsidered. European Journal of Political Research. DOI: 10.1111/1475-6765.12172.

Eppner S, and Ganghof S 2015. Do (weak) upper houses matter for cabinet formation? A replication and correction. Research and Politics. 2(1): 1–5.

Hiroi T and Rennó L (2014) Dimensions of Legislative Conflict: Coalitions, Obstructionism, and Lawmaking in Multiparty Presidential Regimes. Legislative Studies Quarterly, 39(3): 357-386.

Lupia A and Strøm K (2008) Bargaining, Transaction Costs and Coalition Governance. In Strøm K, Müller W and Bergman T (eds) Cabinet and coalition bargaining. Oxford: Oxford University Press, pp. 51-84.

Sarah Shair-Rosenfield and Alissandra T. Stoyan – Gendered Opportunities and Constraints: How Executive Sex and Approval Influence Executive Decree Issuance

This is a guest post by Sarah Shair-Rosenfield and Alissandra T. Stoyan. It is based on their paper in Political Research Quarterly.

Over the last two decades democracies worldwide have elected record-setting numbers of women presidents – in Argentina, Brazil, Chile, Costa Rica, Liberia, Philippines, South Korean, and Taiwan to name just a few. One of the most frequently touted benefits of electing women to any office is the expectation that they tend to rely on or prefer a model of leadership based on negotiation and consensus-building. Indeed, that very quality is often highlighted by journalism about women’s political successes or sometimes promoted by women themselves.

Portrayals like this are typically built on the actions and behaviors of women legislators, or the behavior of legislatures with substantial proportions of female members. Legislatures may lend themselves to studies of gender and leadership styles or preferences because there are relatively more women legislators to evaluate. Legislatures also vary in the size of their female contingents, so it is possible to compare outcomes across different levels of female representation. Perhaps most importantly, it is also easier to understand why negotiation and consensus might be useful for governance: legislatures are themselves collective bodies that must form at least a majority to accomplish most tasks.

Conversely, it has been difficult for political scientists to study how leadership styles might translate to governance strategies of presidents. Although women presidents are more common today, they are still relatively rare. Furthermore, presidents may need to work with legislative counterparts to affect the policy agenda, but they also often have a range of unilateral powers at their disposal. This may reduce their reliance on or preference for negotiation and consensus. How might we expect the assumptions about women’s leadership styles to shape women’s use of their unilateral presidential powers, such as the ability to issue executive decrees?

In our new work, we use a paired-comparative approach to evaluate rates of executive decree issuance in Argentina, Brazil, Chile, and Costa Rica between 2000 and 2014. In each case, a woman president succeeded a man from the same political party. The advantage of this research design is that each pair of presidents faced the same institutional constraints, the same or highly similar partisan opponents, and the same or similar own-party policy preferences. This means we can eliminate a host of alternative factors that might explain variation in decree issuance. Instead, we are able to narrow our focus to the effect of gender on a president’s tendency to make use of her or his unilateral decree power.

We find that gender by itself matters somewhat to rates of decree issuance; women do appear less likely to rule by decree overall. Cristina Fernández de Kirchner (Argentina) and Laura Chinchilla (Costa Rica) are much less likely to use such power compared to their predecessors, while Michelle Bachelet (Chile) is slightly less likely to do so and Dilma Rousseff (Brazil) issues decrees at higher rates than her predecessor. Collectively this provides some evidence that there is a gender-based difference in the use of this type of presidential authority.

However, a more nuanced look at when and why presidents wield such power reveals additional information about the gender-based difference. Presidents are presumed to have the option of “going public” in order to influence the policy agenda. For example, a president may consider that high public approval ratings indicate a public mandate or support for action. Rather than trying to bargain or work with congress to pass legislation, a popular president may feel confident in issuing more decrees to accomplish her or his policy goals. A president motivated to work collaboratively or build consensus should be less interested in this “go public” option, and should rely on it less frequently.

When we account for a president’s approval rating, we see very different trends emerge in the decree issuance of women and men presidents. This figure shows that the (relatively low) rate at which women issue decrees is largely unaffected by how popular they are with the public. In contrast, men become much more likely to issue decrees as they get more popular. The gap in decree issuance by women and men is widest and most consistent with high levels of approval, but this gap narrows as presidents face declining approval that prevents them from being able to assert their will.

Scholars have often assumed that Latin American presidents are prone to abusing their unilateral authority, especially when they are or become more popular. At higher levels of popularity, presidents might be emboldened to “go public” with their policy preferences, rather than wasting their time and resources negotiating with the legislature. What we find suggests that this assumption may be true for Latin America’s presidentes in general, but that its presidentas tend to be less abusive of their authority even when they are popular enough to potentially do so.

As more women run for high office around the world, it seems important to consider this evidence of gendered differences in leadership that point to a new model of presidential self-restraint. Further analysis could illuminate distinctions in women’s motivations for governing as they do, in terms of both their strategic motivations and also the substance of the policies they may pursue.

Manuel Alcantara, Jean Blondel and Jean-Louis Thiébault – The influence of the presidential system on the character of Latin American democracy

This is a guest post by Manuel Alcantara, Jean Blondel and Jean-Louis Thiébault. It is based on their recent book, Presidents and Democracy in Latin America, London and New York: Routledge, 2017.

The aim of this book is to study the effect of the presidential form of government on democracy in Latin America. The adoption of the presidential system, specifically the personality type of those who have occupied the presidential office, the leadership style of those presidents, and the type of government they have led, helps to explain the consolidation of democracy there.

In this study, six countries (Argentina, Brazil, Chile, Colombia, Mexico and Peru) were chosen. They were chosen because they have successfully completed the process of democratic consolidation. Within each of the six countries, two presidents were chosen, reflecting broad trends in the political and electoral life of these countries. The goal was to select presidents belonging to one of the key political ‘families’ of the country, grouped under the banner of a political party, or who were representative of two particular approaches to the same problem in the same political family. These presidents were in office in the 1990s or the first decade of the 21st century. Some were liberal or conservative, left-wing or right-wing populists, socialists or social democrats, leaders of a political party or ‘outsiders,’ members of parliament or technocrats. They are:

  • Carlos Menem (July 1989-December 1999) and Nestor Kirchner (May 2003-December 2007) for Argentina.
  • Fernando Henrique Cardoso (January 1995-December 2002) and Luiz Inacio Lula da Silva (January 2003-December 2010) for Brazil.
  • Patricio Aylwin (March 1990-March 1994) and Ricardo Lagos (March 2000-March 2006) for Chile.
  • Cesar Gaviria (September 1994-September 2004) and Alvaro Uribe (August 2002-August 2010) for Colombia.
  • Ernesto Zedillo (December 1994-November 2000) and Felipe Calderon (December 2006-December 2012) for Mexico.
  • Alan García (July 1985-July 1990 and July 2006 to July 2011) and Alberto Fujimori (July 1990-November 2000) for Peru.

In practice, these presidents were all center-right or center-left leaders. They were not members of the military, dictators, or revolutionaries. Right-wing populist presidents (Menem, Fujimori and Uribe) were chosen based on the idea that populists can be either on the right or on the left. However, García and Kirchner were chosen as moderate populists, claiming to be an Aprist and a Peronist respectively; both represent different periods. These choices make it possible to analyze processes in a consolidated democracy, but not in military regimes or in dominant party systems.

From the 1930s to the 1980s, these Latin American countries had a long period of instability. Argentina wavered between three types of political regimes: military dictatorship, a populist-corporatist regime, and restrictive democracy. From 1930 to the reestablishment of liberal democracy in 1983, there were six major military coups (1930, 1943, 1955, 1962, 1966 and 1976). There were eighteen presidents, and all those elected were overthrown except one, Peron, who died less than a year after his election. Governments in Peru have been more unstable than in any other South American country. Between 1945 and 1992, Peru’s government was civilian and constitutional almost 60 percent of the time, and a military regime 40 percent of the time. There were nearly two decades of military rule in Brazil and Chile. A military coup overthrew President Goulart in 1964 and began the longest period of authoritarian rule in Brazil’s history. With the collapse of democracy in September 1973, Chile was abruptly transformed from an open and participatory political system into a repressive and authoritarian one. General Pinochet was selected as the junta’s president by virtue of his position as leader of the oldest military branch. Unlike many of its continental neighbors, Colombia has avoided military rule, but there was intense violence between members of the two major parties in the late 1940s and 1950s, known as « la violenca » (the violence). A coalition government resulted from party negotiations. From 1958 to 1974, all governments in Colombia consisted of a bipartisan coalition. The main factors commonly associated with good prospects for democracy have long persisted in Mexico without producing full democracy.

Presidential regimes in Latin America are now a success, despite the pessimistic comments directed at this form of government. There are indeed manifest reasons why the Latin American presidential government should be considered a success. Latin American countries have overcome the fundamental dangers to which they were exposed. Although difficulties continue in a number of countries, presidential government in the region is no longer interrupted as it so frequently was in the past. Democratic development also mean that the number of countries regularly holding free and fair elections has increased. Executive governments are often elected by voters mobilized by clientelistic ties or by a candidate’s personality, rather than programmatic, appeal, all in the context of weak parties that are, moreover, rejected by citizens. The presidential elections of Zedillo in 1994 and Calderon in 2006 were intricate and controversial. Both involved critical moments of acute social tension and political instability that produced distinctive results.

Latin American governments have been influenced by the adoption of the presidential system. They set up institutions drawn largely from the US constitutional model. But Latin American presidents represent another type of executive. In the United States, there is a president, but there is no government. Latin America has a large number of presidential regimes characterised by a high degree of consistency and similarity. They constitute a type of intermediary regime, comprising many elements of presidential regimes, but with some of the features of parliamentary systems with coalition governments so as to ensure a majority in congress. For almost twenty years, Brazil has been considered an extraordinary case of « coalition presidentialism ». This explains why the president’s leadership is important and has an impact on the nature of government. The Brazilian party system is highly fragmented. Dealing with loosely disciplined parties is thus a major problem for presidents because it makes the formation of stable congressional majorities much harder to achieve due to the excessive number of party factions. But there were also the broad multi-party coalition governments seen in Chile. Presidents of these countries have demonstrated leadership skills, arising from a good political performance and cohesive majority coalitions that support them: Aylwin and Lagos in Chile, Cardoso in Brazil. It is impossible to explain the stability of these coalitions without referring to the various mechanisms of coalition management and to presidential leadership. Most importantly, these three presidents facilitated the transition to democracy following the failure of authoritarian regimes in Chile and Brazil. They did not have the same authority as Lula, but they showed great skills of conciliation and moderation during the difficult transition period, namely the restoration and the consolidation of the democratic regime in Chile and Brazil.

This explains why the presidential leadership is important and has an impact on the nature of government. The key feature of the popular election of the president has been the inherent tendency of Latin American countries to emphasize the role of personalities in political life. Latin American political regimes have been markedly affected by patronage and clientelism; with the extension of the right to vote, elections were deeply influenced by these practices. The impact of personalities on the political life of Latin American countries has continued to this day, but it is less substantial. There is a decline in the extent to which Latin American presidentialism is personalized compared to the extent it had been previously. In the past personalization undoubtedly rendered presidential rule more chaotic and less rule-based. The fact that, on the whole, presidencies have tended to follow previously adopted rules during the last decades of the twentieth century and the first decades of the twenty-first century has surely resulted in the personalization of presidents being been less marked than in the past. Whereas presidents often enjoy high levels of popularity, these levels vary from president to president as well as over time in the case of each president. One president exhibited exceptional leadership boosted by his personal dominance: Lula. His performance was strengthened by the fact that he had an interesting experience as founder and president of the Workers’ Party. He is often regarded as one of the most popular politicians in the history of Brazil, boasting approval ratings over 80 percent and, at the time of his mandate, one of the most popular in the world.

A new type of personalised populism emerged with the appearance of formulas promoting demobilization and anti-political behavior. Fujimori in Peru, Menem in Argentina, and Uribe in Colombia. These three presidents have adopted a more or less authoritarian manner, being hostile to or even repressing the opposition. They used exceptional means, such as a state of emergency or government by decree, to implement their economic and social policies, as well as the fight against armed rebellions and drug trafficking. However, these exceptional means did not enable them to achieve the expected results. Their presidency was characterized by an authoritarianism and corruption. The populism of Carlos S. Menem in Argentina was strengthened by the political machine of the historic Justicialita Party. Carlos Menem governed within the framework of « peronism » and enjoyed remarkable popular support. Menem’s economic policy involved profound structural reforms, including the privatisation of public enterprises, economic deregulation and the opening up of the economy to foreign trade and investment. This policy created the conditions for monetary stability and remained in force after Menem left office in 1999 and until the crisis of December 2001. However, the policies of the Menem era led to a deepening of social inequality and a rise in unemployment. However he was considered a true peronist. He was the main player in the political regime, with a negative view of parliament and the judiciary. Menem’s leadership has been labelled neopopulist and delegative due to the continuous use of unilateral measures and emergency legislation. It was of a different nature to the populism of Fujimori in Peru. Fujimori sought to distance his government from politics, disdaining the social and/or political mobilization that could have been mounted through some movement or party. Fujimori outlined a strategy in which criticism of the traditional parties was a part of his discourse. He decided to confront the political class instead of building bridges with it. Instead, he expressly renounced such mobilizations, and depoliticized all the other political bodies. Uribe presented himself as the saviour of a Colombia that seemed to be on the brink of destruction. He portrayed himself as a messiah who would redeem Colombia of all its evils and built a strategy around certain core components. He adopted a radical discourse against armed groups and proposed resolving the internal conflict through war and the subjugation of guerrillas. He withdrew from the Liberal Party, to which he had belonged throughout his political life. He spoke out against the parties and the political class despite having belonged to both and adhering to their norms and rules throughout his political career.

Some presidents demonstrated weaker leadership skills (Kirchner in Argentine, Gaviria in Colombia, Zedillo and Calderon in Mexico). They came to power without holding important positions in the governments headed by their predecessors. They have become second-rate candidates, indirectly because of events that have upset or disrupted the appointment of the first office holder. They have never been able to exercise strong authority, muddling through in the face of significant obstacles and divisions.

The fact that the presidential system had become ‘established’ in Latin America by the second half of the 20th century does not mean that these countries have not suffered serious problems. In the 1990s, democracy spread across the region, even if Colombia, Brazil and Mexico experienced marked political violence, the state being unable to maintain order and public security. What is clear is that, in the context of Colombia, Mexico and Brazil in particular, one very serious problem was identified: violence, and this problem affected the regular development of the presidential system in these three countries. The amount of violence that has affected Colombia has been huge, to the extent that it is surprising that the regular conduct of the electoral process has not been prevented from taking place. The policies of the two Colombian presidents, Gaviria and Uribe, were fundamentally different, the first having pursued the ‘war effort’ against the rebels, whereas the second attempted (unsuccessfully) to find a peaceful solution: his successor was able to make substantial progress in that direction, however. In Colombia it has thus been possible, rather surprisingly, to maintain the main electoral rules of the liberal democratic process, although, at least in a substantial part of the country, confrontation has taken in effect the form of a civil war.

In recent decades, presidential elections have taken place regularly in Latin America. Certainly, some presidents have been more popular than others. Some have been unable to conclude their terms. Others have gradually learned to adjust to the particularities of the institutional system. Overall, though, the presidential form of government has gradually begun to function smoothly. The fact that presidents have tended to follow democratic rules has resulted from the presence of patterns of parliamentary presidentialism. In spite of serious problems (political violence, corruption), the emergence of these tools (coalitional presidentialism, the (de-) institutionalization of party systems, the internal organization of the executive branch) must be seen as having constituted the key institutional development of democracy in Latin America.

Miguel Carreras – Presidential Institutions and Electoral Participation in Concurrent Elections in Latin America

This is a guest post by Miguel Carreras, Assistant Professor in the Department of Political Science at the University of California, Riverside (www.miguelcarreras.com). It is based on his recent article in Political Studies.

What is the impact of political institutions on voter turnout in Latin America? Previous studies (Fornos, Power, and Garand 2004, Kostadinova and Power 2007) have addressed this question by replicating a “classic” model (Jackman 1987). This mainstream model evaluates the impact of a series of legislative institutions—district magnitude, the number of parties in the legislature, and unicameralism—on electoral participation. These factors are found to be poor predictors of electoral participation in Latin America. One of these earlier studies concludes that “the classic model provides a weak explanation for turnout in the region” (Pérez-Liñán 2001: 286).

While all these studies have contributed important insights to the literature on electoral participation in Latin America, their assessments of the effect of institutions on turnout have overlooked the fact that Latin American countries have presidential systems of government. In presidential systems, the presidency is the dominant branch of government. Therefore, presidential elections can be described as first-order elections and legislative elections as second-order elections (Reif and Schmitt 1980). The key argument I make in a forthcoming article in Political Studies is that in concurrent elections in Latin America (i.e. when presidential and legislative elections are held on the same day)[1], first-order factors and first-order (i.e. presidential) institutions should have a stronger impact on electoral participation than second-order (i.e. legislative) institutions.

Presidential Institutions and Turnout

The first important factor to consider is the electoral system. Electoral systems regulating the election of the president must determine a threshold of legitimacy considered sufficient for the chief executive to form an authoritative government. Turnout may increase under majority-runoff systems for two reasons. First, voters who support minor or mid-sized parties and realize that their vote will be “lost” may prefer to abstain in plurality systems. Second, under majority-runoff systems, minor parties have more incentives to activate their bases so as to obtain a large share of votes that could be used as an exchange value in the second round (Shugart and Carey 1992).

Hypothesis 1: Turnout is likely to be higher in majority-runoff systems than in plurality systems.

A second institutional characteristic that may be related to electoral participation is term length. All other things being equal, I expect turnout to be higher in countries where the presidential term length is longer for three main reasons. First, the relative costs of voting decrease as the time between elections increases. Second, since dissatisfaction with the political and economic performance of the incumbent government drives electoral participation in developing countries (Aguilar and Pacek 2000), a longer tenure may lead to higher levels of electoral participation by disenchanted citizens who want to punish the president in power. Third, longer presidential terms increase the clarity of responsibility. As a result, it is easier for voters to determine whom to punish or reward for the country’s performance.

Hypothesis 2: Turnout is likely to increase as the presidential term length increases.

The prerogatives vested on the president may also be related to turnout in the region. In fact, concurrent elections in Latin America become more salient when the powers of the president increase. When presidents are more powerful, they are more likely than their weak counterparts in other countries to influence the direction of policymaking and avoid an executive–legislative gridlock. Moreover, when the institution of the presidency carries more powers and prerogatives, presidential elections are more salient to political elites, who are likely to focus efforts on voter mobilization.

Hypothesis 3: Turnout is likely to increase as the legislative powers of the presidents increase.

Political Context and Turnout in Latin American Elections

Previous research has shown that two variables related to the political context in which elections take place have an impact on electoral participation: electoral competition and the number of competing parties (Blais 2006). Surprisingly, previous studies of turnout in Latin America (Fornos, Power, and Garand 2004, Kostadinova and Power 2007) find that competitiveness and the number of parties are unrelated to voter turnout in the region. The Latin American exceptionalism may result from the fact that previous studies have analyzed electoral competition and the number of parties in second-order (i.e. legislative) elections. This study reevaluates the null findings of the literature, applying these two well-known hypotheses of the electoral behavior literature to the first-order rather than to the second-order institution.

Hypothesis 4: Turnout is likely to be higher when the presidential election is close.

Hypothesis 5: Turnout is likely to be lower when the effective number of candidates increases.

Research Design and Results

To test the five hypotheses, this study uses a new cross-national, pooled time series dataset of electoral participation in 102 concurrent elections in 17 Latin American countries between 1980 and 2016. The dependent variable in all of the models presented in this article is turnout as a percentage of voting age population. The data structure is multilevel because there are several observations per country. In other words, election years are clustered within countries. I therefore specify a multilevel model with random intercept coefficients to take into account the hierarchical nature of the data (level 1: country, level 2: election year). The results of the main empirical model in the paper are presented below.

The results provide strong support for my theoretical expectations. In particular, presidential institutions are good predictors of electoral participation in concurrent elections in Latin America. Other things being equal, electoral participation is almost nine percentage points higher in concurrent elections in which there is a majority-runoff system in place for the election of presidents. Term length is positively associated with electoral participation, and the coefficient is statistically significant. An additional year of presidential tenure is likely to increase electoral participation by 4.2 percentage points. In the same vein, the results demonstrate that turnout increases when the legislative powers of the president increase. A 1-point increase in the 10-point presidential power score created by Doyle and Elgie (2016) leads to an increase in electoral participation by 3.2 percentage points. Finally, the effective number of candidates is negatively associated with electoral participation. An increase in one viable candidate in the presidential elections leads to a decrease in turnout in concurrent elections by three percentage points. As expected, in a fully specified institutional model, legislative institutions have a weaker effect on citizens’ decision to turn out on Election Day.

In sum, my findings challenge the conventional wisdom regarding the impact of institutional factors on electoral participation in Latin America. Previous studies of turnout in Latin American elections replicated an institutional model (the “Jackman model”) that is better suited to explain electoral participation in parliamentary systems. By estimating a fully specified model of turnout in concurrent elections in Latin America which includes both first-order (presidential) and second-order (legislative) institutions, I provide the strongest and clearest evidence, to date, of the impact of presidential institutions and the context of presidential elections on turnout in concurrent elections in the region. My empirical results also demonstrate that legislative institutions have minimal effects on voter turnout in concurrent elections in Latin America.

References

Aguilar, Edwin E., and Alexander C. Pacek. 2000. “Macroeconomic Conditions, Voter Turnout, and the Working-Class/Economically Disadvantaged Party Vote in Developing Countries.”  Comparative Political Studies 33 (8):995-1017.

Blais, André. 2006. “What Affects Voter Turnout?”  Annual Review of Political Science 9:111-125.

Doyle, David, and Robert Elgie. 2016. “Maximizing the Reliability of Cross-National Measures of Presidential Power.”  British Journal of Political Science 46 (4):731-741.

Fornos, Carolina A., Timothy J. Power, and James C. Garand. 2004. “Explaining Voter Turnout in Latin America, 1980 to 2000.”  Comparative Political Studies 37 (8):909-940.

Jackman, Robert W. 1987. “Political Institutions and Voter Turnout in the Industrial Democracies.”  American Political Science Review 81 (2):405-423.

Kostadinova, Tatiana, and Timothy J. Power. 2007. “Does Democratization Depress Participation? Voter Turnout in the Latin American and Eastern European Transitional Democracies.”  Political Research Quarterly 60 (3):363.

Pérez-Liñán, Aníbal. 2001. “Neoinstitutional Accounts of Voter Turnout: Moving Beyond Industrial Democracies.”  Electoral Studies 20 (2):281-297.

Reif, Karlheinz, and Hermann Schmitt. 1980. “Nine second-order national elections – a conceptual framework for the analysis of European election results.”  European Journal of Political Research 8 (1):3-44.

Shugart, Matthew S., and John M. Carey. 1992. Presidents and Assemblies: Constitutional Design and Electoral Dynamics. Cambridge: Cambridge University Press.

Notes

[1] The majority of elections in Latin America are concurrent—60% of national elections in the region between 1980 and 2016 were concurrent.

Aníbal Pérez-Liñán and John Polga-Hecimovich – Getting Rid of the President

This is a guest post by Aníbal Pérez-Liñán of the Department of Political Science at the University of Pittsburgh and John Polga-Hecimovich of the Political Science Department at the US Naval Academy at Annapolis. It is based on their paper in Democratization.

Are presidential impeachments modern functional equivalents of old-fashioned military coups? The impeachment of Brazilian President Dilma Rousseff in August 2016 led to an acrimonious debate on whether her removal from office constituted a “soft coup” against an elected leader. Similar concerns were voiced after the impeachment of Paraguayan President Fernando Lugo in 2012. As calls to impeach President Donald Trump multiply, this question appears to gain increasing relevance for US politics as well.For students of presidentialism, the idea of “functional equivalence” between military coups and legal ousters (impeachments, legislative declarations of presidential incapacity, or anticipated resignations of the executive) translates into very specific questions: Are there any historical factors able to explain military coups as well as impeachments? If so, why are some presidents removed following legal procedures while others are removed by force?

In a forthcoming paper in Democratization we develop a unified theory of presidential instability to explain why presidents are removed from office through military coups or through legal procedures.

We identify two sets of historical causes. First, some factors create conditions for presidential instability, irrespective of the mode of premature exit from office. Because they motivate a political opposition to conspire against the government, those factors explain why presidents are likely to fail, but not how they fail. Second, an alternative set of causes accounts for the specific institutional manifestations of presidential instability. Those factors map onto the relative capabilities of groups inclined to pursue a military coup or the legal removal of the president.

The distinction between general motivations to remove the president and the capabilities of specific opposition groups helps us identify the role of different causal explanations in the literature.

Among the common causes of legal removals and coups, we find:

  • Poor economic conditions. Recessions undermine the president and facilitate conspiracies. Studies on military coups argue that negative economic shocks increase the risk of military rebellions, while the literature on impeachments shows that weak economies undermined Latin American presidents in the 1990s.
  • Popular protests. Mass mobilization against the government signals that the president is weak and destabilizes any elected administration. Students of military intervention find that mass protests help elites coordinate in a coup. Students of impeachment emphasize that protests encourage reluctant legislators to act against the president.
  • Radicalization. Radical actors have intense and extreme preferences; they are reluctant to bargain and remain intransigent in defense of their policy goals. Radicalism is therefore a potential cause of military coups, but also an explanation for the role of social movements forcing the resignation of presidents in places like Bolivia and Ecuador.

Given the prior conditions for instability, several factors separate legal removals from coups:

  • The regional context. A long line of research has invoked international diffusion as an explanation for democratic instability – though not necessarily government instability. The regional context may strengthen the position of coup perpetrators or otherwise direct elites towards legal strategies against the president.
  • Legislative support for the president. Two causal mechanisms are discussed in the literature: Linz’s argument that presidentialism itself is a source of instability and the argument that a legislative majority “shields” the executive against impeachment.
  • Elite support for democracy. A strong normative preference for democracy among elites forecloses the possibility of a military coup and leaves legal removal as the only acceptable strategy for the opposition. The government’s normative preferences also matter: a president dismissive of democratic rules may be unwilling to recognize the legitimacy of an impeachment procedure, driving opponents to consider the option of a coup.

To test those expectations, we use discrete-time event history models with selection.  Our sample covers all democratic regimes in nineteen Latin American countries between 1945 and 2010 (N = 729). The dependent variable measures yearly outcomes for each president:  survival, exit via military coup, or exit via legal removal. Our sample includes 21 coups and 15 legal removals. The selection model estimates the risk of president being removed from office (in any way) in the selection stage, and the risk of being removed via coup (as opposed to a legal procedure) in the outcome stage.

The statistical models allow us to estimate the risk of coups and impeachments, plotted in Figures 1 and 2.

Figure 1 underscores the role of common motivations behind coups (in the bottom row) and impeachments (in the top row), as economic recession, demonstrations, and radicalization consistently expand the risk of both outcomes.

Figure 1: Common Causes of Legal Removals and Coups (Predicted Risk)

Figure 2, on the other hand, illustrates the differential impact of variables. The first column illustrates how a large number of coups in neighboring countries expands the risk of military intervention but reduces the probability of legal removal in the observed country.  The second column shows that the risk of military overthrow remains independent from the composition of congress, but impeachment is less likely when the executive controls the legislature.  The third column shows that a military coup is unlikely when political actors are more committed to democracy. By contrast, the risk of legal removal expands as groups operating within the constitution become empowered by the opposition’s reluctance to engage in military conspiracies.

Figure 2: Causes Separating Legal Removals and Coups (Predicted Risk)

Our findings underscore that common causes of presidential instability are not necessarily causes of democratic breakdown, yet crises of government may easily escalate into crises of the democratic regime when legal venues for the removal of the president are blocked.

These findings are increasingly relevant today.

In a global context in which presidents and their adversaries – in Latin America, Eastern Europe, Turkey, and even the U.S. – have displayed growing levels of radicalism, our findings raise concerns. Radical leaders engender polarization, encouraging their opponents to overthrow the government by any means possible. Combined with economic stagnation or social protest, radicalization is likely to trigger presidential instability.

Yet other factors ultimately tip a crisis towards a non-democratic resolution. A regional environment hostile to democracy and a lack of democratic commitment from domestic elites decrease the probability of a legal impeachment and increase the likelihood of a coup.

International policymakers would be wise to consider these findings: long-term efforts to build regional organizations that discourage military intervention and steady support for democratic leaders will prevent future presidential crises from escalating into full crises of democracy.

André Borges and Mathieu Turgeon – Presidential coattails in coalitional presidentialism

This is a guest post by André Borges and Mathieu Turgeon, both of whom are assistant professors of political science at the University of Brasília. It is based on a recent article in Party Politics.

Research on coalitional presidentialism has focused mostly on post-electoral coalition formation, neglecting the  pre-electoral origins of cabinets  in many – if not most – presidential countries with multiparty systems (Albala 2014; Chasquetti 2008; Freudenreich 2016). Kellam (2015) analyzed pre-electoral coalition formation in presidential elections in eleven Latin American countries from the 1980s to the late 2000s, and found that 35% of all presidential candidates that obtained at least 10% of the national vote formed a coalition with one or more parties. Although pre-electoral coalitions in presidential elections are a rather frequent phenomenon, there is a paucity of research on the causes and consequences of these pre-electoral alliances. In particular, the literature on presidential coattails has failed to consider the potential impacts of multiparty alliances on party system formation, assuming that parties entering the presidential race as members of an alliance do not obtain electoral gains (Mainwaring and Shugart 1997; Shugart and Carey 1992; West and Spoon 2015). That is, the coattail effect benefits only parties that enter the race with a candidate of their own, as voters rely on the party of their preferred presidential candidate as an information shortcut to help them decide how to vote in legislative election (Golder 2006). But, if allied parties do not benefit from presidential coattails and they actually risk losing credibility and weakening their party base if the coalition is not perceived as adequate , why would they support a presidential candidate from another party in the first place? Even if parties believe that entering a pre-electoral coalition will increase their chances of entering the presidential cabinet, they cannot be sure of the supported candidate’s victory in the presidential contest (Freudenreich 2016).

In a recent article (Borges and Turgeon 2017), we challenge the conventional wisdom on presidential coattails and pre-electoral coalitions.  By focusing on coattails from the president-elect party—the coalition formateur—we argue that presidential coattails in coalitional presidentialism benefit not only the party of the president-elect but also the coalition party members, which has  important implications for coalition formation in presidential systems. This is what we label a diffused coattail effect.

In multiparty presidential systems, parties that are viable contenders in the presidential election are likely to “presidentialize”, shifting resources away from their legislative campaigns and focusing on the presidential race (Samuels 2002). To secure the necessary votes to win the presidency, large parties form electoral coalitions with smaller parties and adopt broad campaign strategies. Specifically, they avoid pure partisan campaign strategies and campaign, instead, on behalf of the coalition to mobilize as many voters as possible for the presidential election.

Coalition fomateurs understand that there are costs for parties to join their coalition and are disposed to make important concessions to convince them to join forces. These concessions include, in part, supporting coalition party members in simultaneous, lower-level elections and by making sure that candidates from the coalition formateur party do not “invade” the electoral strongholds of the other coalition party members. Moreover, presidential candidates campaign on behalf of the whole coalition and not only for their own party, especially in other simultaneous, lower-level electoral contests like legislative elections. In exchange, coalition party members aggregate valuable organizational and financial resources to help the formateur party reach segments of the electorate otherwise less accessible but necessary to win the presidential election.

We believe coalition party members benefit from presidential coattails because the parties involved in the coalition work together to coordinate their campaign strategies at all levels (presidential, gubernatorial, senatorial and lower chamber races). But coalitions are not all created equal and the effects they carry over election results depend, in part, on the ability of coalition party members to coordinate effectively with the formateur party. Specifically, we believe that coalition party members that coordinate more effectively with the formateur party should benefit more from presidential coattails than those who don’t. We classify coalition party members into core and peripheral coalition party members. Core coalition party members are defined as those that are close ideologically to the formateur party and that have adopted consistent strategies in the governing and electoral arenas in the past.

Coalition party members that have participated in the past governing coalition can benefit from the president’s popularity during the election by claiming credit for key government programs, tying their fortunes with that of the incumbent president. Moreover, coalition party members that have participated in previous electoral coalitions with the same formateur party should be associated more strongly to the said coalition by voters than those coalition party members that have not. Finally, we believe that coalition party members will coordinate more forcefully the closer they are ideologically to the coalition formateur because, in that scenario, both can tailor campaign messages courting ideologically similar voters.

We test two hypotheses. First, we argue that presidential coattails are diffused, benefiting the president’s party but also her coalition party members. Second, we claim that The diffused coattails effect in coalitional presidentialism should benefit more strongly core coalition party members, as compared to peripheral coalition party members.

To evaluate the two hypotheses we analyze data from Brazil and Chile. These two countries are widely studied cases of coalitional presidentialism where multiparty coalitions play a fundamental role in the governing and electoral arenas. Overall, Chile represents a most-likely case for diffused presidential coattails because its governing and electoral coalitions are stable and ideologically coherent. Brazil, on the other hand, represents a least-likely case for diffused presidential coattails because it shows much less congruence between its governing and electoral coalitions and its electoral coalitions are unstable and generally not ideologically coherent. We believe that such design allows for robust testing of our hypotheses of presidential coattails in coalitional presidentialism. Finding only weakly supportive evidence (or no evidence at all) of diffused coattails in Chile would seriously undermine or lead to outright rejection of our theoretical claims.  On the other hand, if we succeeded in finding evidence of diffused coattails in Brazil, this should strongly support the view that presidential coattails exhibit dynamics of their own in coalitional presidentialism.

Our statistical analysis of coattail effects using data on district-level electoral returns in Brazil and Chile shows that presidential coattails in coalitional presidentialism are diffused, benefiting the president’s party and her core coalition party members. Presidential coattails, however, do not affect coalition party members equally. Core coalition party members, that is, those that are more strongly associated with the coalition formateur, are the sole beneficiaries of presidential coattails. No presidential coattail effect is discernible for peripheral coalition parties.

Admittedly, we cannot make claims about the presence or not of similar diffused presidential coattails in other cases of coalitional presidentialism. We have very good reasons to believe, however, that this phenomenon extends beyond the Chilean and Brazilian cases. In particular, both Chile and Brazil are open-list PR systems. In closed-list PR systems, which are most commonly found in other cases of coalitional presidentialism, intra-coalition coordination is profoundly facilitated. Under such electoral rules, parties can more easily divide the expected seats among coalition partners by ordering the candidates’ names on party lists in each district in a way that benefits more fairly coalition party members (Cruz 2010; Leiras 2007).

Future research should explore further the broader implications of the diffused coattail effect for coalitional presidential systems and party systems, more generally. One such possibility deals with the relationship between electoral and governing coalitions. Our results, for example, suggest that the electoral success of peripheral coalition party members is not tied to that of the coalition formateur party. Consequently, their behavior within the governing coalition could be distinct than that of core coalition party members and could potentially affect the stability of governing coalitions. Thus we may ask: are peripheral coalition party members less loyal and possibly more demanding than core coalition party members? Similarly, are threats to leave the governing coalition more credible than those made by core coalition party members? These are other interesting questions to be explored.

Finally, diffused presidential coattails may also contribute to maintain or even increase party fragmentation in the lower chamber. That is, different from traditional arguments on presidential coattails and party systems, the theoretical argument and empirical evidence presented in this paper indicate that presidential coattails, when diffused, foster instead the survival and growth of small parties. Contrary to West and Spoon’s (2015) findings about electoral coalitions, it is not clear whether this will always and necessarily lead to lower fragmentation in legislative elections. These questions should be of great interest to comparativists given the spread of coalitional presidentialism in Latin America, Africa and the former Soviet Union.

Bibliography:

Albala, Adrian. 2014. “The Timing Effect of Presidentialism on Coalition Governments: evidence from Latin America.” In 23rd IPSA World Congress, Montreal, CA.

Borges, André, and Mathieu Turgeon. 2017. Presidential coattails in coalitional presidentialism. Party Politics: 1-11.

Chasquetti, Daniel. 2008. Democracia, presidencialismo y partidos políticos en América Latina: evaluando la” difícil combinación”. Ediciones Cauce-CSIC.

Cruz, Facundo. 2010. Relaciones e interacciones partidarias en coaliciones de gobierno. Los casos de la Alianza, la Concertación y el Frente Amplio. Revista Debates Latinoamericanos 8: 15.

Freudenreich, Johannes. 2016. The Formation of Cabinet Coalitions in Presidential Systems. Latin American Politics and Society 58 (4): 80-102.

Golder, Matt. 2006. Presidential Coattails and Legislative Fragmentation. American Journal of Political Science 50 (1): 34-48.

Kellam, Marisa. 2015. Why Pre-Electoral Coalitions in Presidential Systems? British Journal of Political Science 47: 391-411.

Leiras, Marcelo. 2007. Todos los caballos del rey: la integración de los partidos políticos y el gobierno democrático de la Argentina, 1995-2003. Prometeo libros.

Mainwaring, Scott, and Matthew Soberg Shugart. 1997. Presidentialism and democracy in Latin America. . Cambridge University Press.

Samuels, David. 2002. Presidentialized Parties: The separation of powers and party organization and behavior. Comparative Political Studies 35 (4): 461-83.

Shugart, Matthew, and John M. Carey. 1992. Presidents and Assemblies: Constitutional design and electoral dynamics. Cambridge, UK: Cambridge University Press.

West, Karleen Jones, and Jae-Jae Spoon. 2015. Coordination and presidential coattails Do parties’ presidential entry strategies affect legislative vote share? Party Politics: 1-11.

Christopher A. Martínez – Why political institutions matter for presidential survival

This is a guest post by Christopher A. Martínez, Assistant Professor in the Department of Sociology and Political Science, Temuco Catholic University, Chile

There is no crisis here, nor problems” Fernando de la Rúa (resigned, December 2001)
I’ll continue to govern” Lucio Gutiérrez (dismissed by Congress, April 2005 )

Since 1979, thirteen South American chief executives have been unable to complete their constitutional terms. These failed presidencies occur when a popularly elected president is forced to leave office early, though the ouster is not followed by a democratic breakdown. Some presidents have been impeached (e.g., Collor and Rousseff in Brazil, Pérez in Venezuela, Cubas and Lugo in Paraguay); others could not withstand massive and widespread street protests (e.g., Alfonsín and De la Rúa in Argentina, Siles Zuazo and Sánchez in Bolivia, and Fujimori in Peru); while other leaders were unseated via unorthodox mechanisms (e.g., Bucaram, Mahuad, and Gutiérrez in Ecuador). Being forced to leave office early represents a dramatic deviation from a central goal of all political leaders, which is to maintain power. Thus, failing to fulfil a presidential term should be an exceptional political event in a presidential democracy.

I used survival analysis to quantitatively study 65 South American presidencies between 1979 and 2012. My results show that the most important forces driving presidential survival are institutional ones: legislative support for the president, and a country’s democratic tradition. Interestingly, inflation, economic recessions, and scandals have no significant impact on presidential survival, whereas violent social mobilisations exhibit a rather weak effect.

Some presidents are “safer” than others: Why a country’s democratic tradition matters

Previous studies have not established whether democracy had any impact, be it positive or negative, on the occurrence of presidential failures. Rather than focusing on current levels of democracy, in my research I used a new measure of democracy which represents a country’s past records with democratic and authoritarian politics: democratic tradition. Figure 1 illustrates how different a country’s current level of democracy (Polity2) and democratic tradition truly are. For instance, Argentina, Chile, Paraguay and Uruguay all had low levels of democracy in 1980. Nevertheless, only Chile and Uruguay stand out for their accumulated experience with democracy.

When considering democracy’s effects over extended periods of time, my findings show that the larger the democratic tradition of a country, the more likely the president will “survive.” That is to say, countries with a well-established democratic track provide a “safer,” less threatening environment for presidential survival. Unlike countries with poor democratic experiences, in these type of polities, political actors—presidents, legislators, parties, among others—are discouraged from pursuing questionable courses of action and are more likely to settle political disagreements through negotiation and accommodation, all of which reduces the risk of early government termination.

Figure 1: Democratic Stock and Polity2 Score of Democracy, 1900-2012

Legislative support is central to “survive” in office

In theory, chief executives in presidential systems do not require a legislative majority to stay in office; however, presidents need legislators’ support more than they may think. Passing relevant legislation is a central task for most executives, but hanging on to power is unquestionably a far more important goal for any president. If presidents are to complete their terms in office, they must ensure the backing of a disciplined contingent of members of congress. This “legislative shield” (Pérez-Liñán 2007) would especially come in handy during dire economic circumstances and intense social mobilisations, as loyal legislators may undermine the opposition’s attacks and criticism against the executive. As in previous studies, my research maintains legislative support for the president as the most consistent and strongest predictor of presidential survival in South America.

What do political scandals do?

Results have been mixed about the relationship between political scandals and failed presidencies. Unlike Hochstetler (2006) and Pérez-Liñán (2007), my findings show that corruption scandals do not reduce presidential survival in South America. Presidents’ involvement in scandals may be frowned upon and weaken their approval ratings, yet they do not directly or necessarily force them to step down. True, some presidents have been deposed because of corruption accusations (for example, Collor in Brazil and Pérez in Venezuela). Nevertheless, political scandals are not exceptional in the region, and many of them have not triggered presidential interruptions (e.g., Menem in Argentina, Samper in Colombia, among others). What scandals can do, especially in cases of fragile president-party relations, is to undermine the ruling coalition and/or reduce the president’s chances to form a new one. Such an instance is what I argue occurred with Fernando De la Rúa’s bribery scandal in Argentina (Martínez 2017) and Lucio Gutiérrez’s alleged links with a drug trafficker in Ecuador (Martínez forthcoming).

Social mobilisations

Though it may come as a surprise, my results show that street protests have only a weak—if any—effect on presidential survival. This is true of both general strikes and social mobilisations aimed at the executive. On the other hand, violent demonstrations such as riots do increase the risk of early presidential removals; nonetheless, their impact is significantly weakened when one analyses a president’s legislative support. That is to say, when it comes to “surviving” in office, the role of congress outweighs any type of social mobilisation, even the bloody ones. An alternative explanation for the weaker-than-expected effects of public demonstrations is that it is their intensity, rather than their simple occurrence, that matters.

Final remarks

Even though a president’s popularity may be negatively affected by economic recessions, street protests, and political scandals, their “survival” in office largely hinges upon legislative support and democratic tradition. The role of congress is likened to the proverbial two-edged sword: it may either shield the president or turn against him/her. Presidents, thus, ought to cultivate smooth relations with their ruling partners should they indeed want to hold onto office. Moreover, chief executives ruling over countries with a weak democratic tradition may have fewer chances to “survive” to begin with, as most political actors in those countries may be more accustomed to bend the rules of the game, which would heighten the risk of presidential failures.

Christopher A. Martínez holds a PhD in Political Science from Loyola University Chicago. He is Assistant Professor in the Department of Sociology and Political Science, Temuco Catholic University, Chile. His current research interests include the executive branch, government survival, institutional performance and democratic consolidation in Latin America. He can be reached at christopher.martinez@fulbrightmail.org and @martineznourdin.

Marisa Kellam – Why Pre-Electoral Coalitions in Presidential Systems?

This is a guest post by Marisa Kellam, Associate Professor, Waseda University. It is based on her recent article in the British Journal of Political Science.

Presidential politics goes hand in hand with coalitional politics in Latin America, especially in South America. As recently reported in this blog, presidents in the region often depend on the support of other parties to win election and to govern.

In this post, I will focus on pre-electoral coalitions. [1] To give some recent examples: President Bachelet in Chile, President Santos in Colombia, and former Brazilian President Dilma Rousseff all ran for re-election with a multiparty electoral alliance.  Multiple parties also supported Argentine President Macri’s candidacy. In recent elections in other countries, the incumbent party candidate defeated an opposition pre-electoral coalition, such as in Ecuador’s recent election, Bolivia’s 2014 presidential contest, or the 2013 presidential election in Venezuela.

In fact, pre-electoral coalitions in presidential elections have been a feature of Latin American democracy since the third-wave, and even before. Yet the conventional wisdom has been that these coalitions were “not binding past election day.” [2] However, increasing attention to post-electoral coalition formation in comparative presidentialism research has led to new findings that winning pre-electoral coalitions usually go on to form post-electoral governing coalitions. [3] Does the strong empirical correspondence between pre-electoral coalitions and post-electoral governments call for a revision of the conventional wisdom? My recently published article in the British Journal of Political Science speaks to this puzzle.

Why do parties form pre-electoral coalitions in presidential systems? From the perspective of a presidential candidate, it would seem to be an easy answer—the more in my camp, the merrier—that is, unless she must give something in return. When considering potential partner parties, we might assume that the presidential candidate offers them government positions—just as presidents offer coalition partners in government negotiations—except that pre-electoral agreements involve only promises not actual offers.

Although I set out to overturn the conventional wisdom on pre-electoral coalitions, I found no convincing argument to support a contrary claim that presidential candidates’ promises to distribute government positions and resources to other parties are credible commitments in presidential systems. Presidents alone control cabinet appointments—even their own parties cannot hold presidents immediately accountable. Moreover, presidents’ partners will not necessarily punish them for breaking their pre-electoral commitments. A party that wants access to resources under the president’s control is unlikely to make a loud complaint, much less to pull out of the government completely.  And if parties do not reveal the extent to which presidents fail to honor agreements to share spoils, then neither presidents nor their parties will pay a reputational cost. This isn’t to say that presidents will break their promises; it is only to make the point that candidates’ pre-electoral promises to share spoils are “cheap talk” and party leaders know this.

I find it useful to contrast these behind the door negotiations with presidential candidates’ public campaigns.  A presidential candidate and her political party pays an immediate reputational cost if she publically campaigns on a policy compromise made with another party in order to gain its support. True, a president is not bound to her campaign platforms. But even so, the pre-electoral policy agreement reveals information about her policy positions.  A pre-electoral policy agreement also gives the president and her partners a shared mandate, or common purpose, after the election. And if the president ends up reneging on that policy agreement later, the coalition partner would likely refuse to go along, consistent with their own electoral incentives and policy motivations.

These differences between patronage promises and campaign platforms provide some insight as to why parties join pre-electoral coalitions to support other parties’ candidates in presidential systems. Political parties join pre-electoral coalitions in pursuit of policy goals, but not as part of an office-seeking strategy.  To provide empirical evidence to support this argument, I compare characteristics of the parties that joined pre-electoral coalitions with those that did not.

More specifically, I compare the probability of participation in pre-electoral coalitions of programmatic parties to that of particularistic parties.  Particularistic parties are those that experts classify as having no discernible policy position on the standard, left-right macroeconomic dimension of politics; instead, particularistic parties focus on the distribution of “pork” and patronage or serve single-interests.  According to my reasoning, if these parties do not have policy goals then they should be less likely to join pre-electoral coalitions (unless one of their own members is on the president-VP ticket). Programmatic parties, in contrast, may use pre-electoral coalitions to identify and help elect presidential candidates who are closest to them in terms of policy.

I analyzed coalitions formed, and not formed, in 77 elections held in 11 Latin American countries.  I found that programmatic political parties (i.e. policy-seeking parties) were more likely than particularistic political parties (i.e. office-seeking parties) to join pre-electoral coalitions in support of another party’s presidential candidate.  As expected, I also found that the greater the ideological distance between a programmatic party and the party of a presidential candidate, the less likely they are to join that candidate’s coalition.

While the once conventional thinking that presidents have little incentive to form governing coalitions has been overturned, this does not imply that the conventional wisdom regarding electoral coalitions should also be cast aside. As I have discussed, pre-electoral coalition bargaining differs from post-electoral government negotiations, with important implications for presidential politics in multiparty systems.

In conducting this research, I realized that the reason why parties join pre-electoral coalitions in presidential systems is less obvious than it appeared at first glance. Even if pre-electoral coalitions are not binding commitments to govern together after the election, the coalition formation process itself informs political parties about their respective policy positions and creates a shared mandate.

 

Notes

[[1]] On a side note, the prefix “pre” seems unnecessary to me, but I use it nonetheless because the term pre-electoral coalitions is widely used in the literature.

[2] Mainwaring, Scott, and Mathew Shugart. 1997. Presidentialism and Democracy in Latin America. Cambridge UP, p. 397.

[3] As discussed previously in this blog, see Freudenreich, Johannes. 2016. “The Formation of Cabinet Coalitions in Presidential Systems.” Latin American Politics and Society 58(4): 80-102.  In my own work-in-progress with Cecilia Martinez-Gallardo, we also find a strong empirical relationship between pre- and post-electoral coalitions in Latin America.

Fernando Meireles – Latin American presidents and their oversized government coalitions

This is a guest post by Fernando Meireles, Ph.D candidate in Political Science at Federal University of Minas Gerais (Brazil). E-mail: fmeireles@ufmg.br

In many countries, presidents have a difficult time governing because their parties lack a legislative majority. In fact, because of the combination of separate elections for executive and legislative branches with multiparty systems, this situation is far from uncommon: during the last two decades in all 18 Latin American countries with presidential systems, only 26% of the time has the president’s party had a majority in the lower house. Due to this constraint, as a vast amount of research now highlights, minority presidents usually form multiparty government coalitions by including other parties in their cabinets. Again, only four Latin American presidential countries in the last twenty years were not governed by a multiparty coalition at some point since the 1980s.

However, the need to craft a legislative majority alone does not explain why presidents frequently include more parties in their governments than necessary to obtain a minimum winning coalition – forming what I call an oversized government coalition. The distribution of this type of coalition in Latin America is shown in the graph below. As can be seen, it is not a rare phenomenon.

If government coalitions are costly to maintain, as presidents have to keep tabs on their coalition partners to ensure they are not exploiting their portfolios to their own advantage – not to mention the fact that by splitting spoils and resources between coalition partners, the president’s own party is worse off – then why are these oversized coalitions prevalent in some Latin American countries?

In a recent article in Brazilian Political Science Review, I tackled this puzzle by analyzing the emergence of oversized government coalitions in all 18 presidential countries in Latin America[1], followed by a case study focusing on Brazil, spanning from 1979 to 2012. To this end, I gathered data on cabinet composition[2] from several sources to calculate the size of each government coalition in the sample: if a coalition had at least one party that could be removed without hampering the majority status of the government in the lower house in a given year, I classified it as an oversized coalition.

Specifically, I examined three main factors that, according to previous research, should incentivize presidents to include more parties in their coalitions than necessary to ensure majority support: 1) the motivation party leaders have to maximize votes, which would make joining the government attractive to opposition parties (vote-seeking); 2) the motivation presidents have to avoid coalition defections to implement their policy agendas (policy-seeking); and 3) the institutional context, considering the effects of bicameralism, qualified majority rules, and party system format on government coalition size.

The results support some of the hypothesis suggested by the literature. First, presidents are more prone to form oversized coalitions at the beginning of their terms, which shows that the proximity to the election affects Latin American presidents’ decision to form, and opposition parties to accept being part of, large coalitions – as others studies argue, this is mainly due to parties defecting from a coalition to present themselves as opposition when elections are approaching. Second, party fragmentation also has a positive effect on the emergence of oversized coalitions, consistent with the hypothesis that presidents might include additional parties in their coalitions anticipating legislative defections. Yet on the other hand, presidential approval, party discipline, and ideological polarization do not have the same positive effects on the probability of an oversized coalition being formed.

The factor that has the most impact on the occurrence of oversized coalitions, however, is the legislative powers of the president. As the literature points out, legislative decrees and urgency bills could be used by skilled presidents to coordinate their coalitions, facilitating horizontal bargaining between coalition partners. The comparative results show that this is the case in Latin America: the difference in the predicted probability of a president with maximum legislative powers in the sample forming an oversized coalition and another with minimum powers is about 32 percent points.

By exploring the Brazilian case in more depth, I also found that bicameralism dynamics and qualified majority rules impact the emergence of oversized coalitions. With two chambers elected through different electoral rules, parties in Brazil are often unable to secure the same seat share in both houses; to make things worse for presidents, party switching is still widespread in the country. In this context, as my results uncovered, differences in the number of seats controlled by the government in the Chamber of Deputies and the Senate positively affect the emergence of oversized coalitions. Finally, as some bills require supermajorities to be approved, such as constitutional amendments, reformist presidents also tend to form and maintain larger coalitions: the maximum value in this variable predicts increases by up to 10 percentage points on the probability of an oversized coalition being formed.

Taken together, these results show a more nuanced picture of why and how presidents form multiparty government coalitions in Latin America: often, obtaining a legislative majority is not enough to implement their legislative agendas, and so they might resort to a complementary strategy: to form larger coalitions. And presidents with greater legislative power, at the beginning of their terms or facing fragmented party systems, are in the best position to pursue such a strategy. In this way, both electoral and programmatic factors, as well as the institutional context, become key to understand variations in the size and the composition of government coalitions in presidential countries.

Notes

[1] These countries are Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, El Salvador, Ecuador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Dominican Republic, Uruguay, and Venezuela.

[2] The criteria employed to identify a government coalition is the party affiliation of the ministers of the principal ministerial portfolios in each country – taking into account that ministers are not always recruited due to their connections or their congressional influence, and that in some cases they are not recognized by their parties as legitimate representatives of the same.

Johannes Freudenreich – The Formation of Cabinet Coalitions in Presidential Systems

This is a guest post by Johannes Freudenreich, Postdoctoral research fellow at the Geschwister-Scholl-Institut für Politikwissenschaft at the University of Munich. It is based on an recent article in Latin American Politics and Society

In the beginning of the 21st century, prospects of Latin American presidential democracies were good. The dictatorships of the 1970s and 1980s had vanished, economies were constantly growing, and comprehensive social welfare programs were implemented. Many political scientists link these successes to the ability of Latin American presidents to form, maintain and manage cabinet coalitions (Cheibub 2007). The differences between presidential and parliamentary systems of government seemed to have become rather marginal. Both presidents and prime ministers achieved legislative majorities by forming broad cabinet coalitions and critics of the presidential form of democracy, such as Juan Linz (1994), seemed to be proven wrong. However, soon presidential impeachments became the new pattern of political instability in the region (Pérez Liñan 2007). Cabinet reshuffling remains constantly high and broad corruption schemes, directly linked to coalition politics, have been disclosed, such as the Mensalão Scandal in Brazil, where the ruling party of President Lula da Silva used illegal side payments to secure the legislative support of members of the ruling coalition.

My recent article in Latin American Politics and Society takes a systematic look at the formation of cabinet coalitions in presidential systems over the past 25 years. It analyzes the extent to which presidents in 13 Latin American countries have formed coalitions that increase their law-making capabilities, and whether presidents form coalitions tailored to find majorities in Congress especially when presidents have low independent influence over policy based on their institutional law-making powers.

The study complements the perspective that cabinet coalitions are largely an instrument for finding legislative majorities with the idea that presidents use cabinet posts to honor pre-electoral support. The reason is the following: presidential elections provide strong incentives for electoral coordination because they tend to favor two-candidate competition. In a multi-party setting, this means that parties have incentives to form pre-electoral coalitions to present joint presidential candidates. When negotiating pre-electoral pacts, parties are likely to agree on how to share the benefits of winning including cabinet posts. After the election, presidents find it difficult to abandon these agreements as they need the trust and support of other parties within and outside of their coalition during their presidential term. Thus, it is expected that cabinet coalitions are likely to be based on the electoral team of presidents and that other legislative parties are invited to join the cabinet only additionally to parties of the existing pre-electoral coalition.

The study further argues that parties attractive as pre-electoral coalition partners are not necessarily the ones that would achieve cabinet participation if the negotiations of cabinet posts were an unconstrained post-electoral process. For example, in a one-dimensional policy space, extreme parties, parties more extreme than the president to the median legislator, are relatively unimportant for legislative decisions and thus unlikely to be included in the cabinet for legislative reasons. In a presidential race, however, extreme parties can provide valuable votes and campaign resources and therefore have far stronger blackmailing power. Furthermore, presidential contests produce a strong antagonism between the president and the parties of the president’s electoral rivals. Since the president’s survival in office is not contingent on the support of other parties in parliament, parties that present a strong presidential candidate are likely to be excluded from the cabinet, even if their inclusion is rational from a lawmaking perspective. It is therefore expected that the party of the runner-up is generally excluded from the presidential cabinet and that the overall explanatory power of variables of legislative bargaining increases once one controls for the effects of pre-electoral coalition formation and competition.

The study empirically evaluates this argumentation on the basis of so-called conditional logit models, presenting a new empirical strategy to analyze cabinet formation under this type of regime. The tests are conducted on a new dataset of 107 democratic cabinets in Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, the Dominican Republic, El Salvador, Honduras, Nicaragua, Panama, Uruguay, and Venezuela. Based on the new method and data, this study presents the most comprehensive test yet of the determinants of the partisan composition of presidential cabinets.

The most note-worthy empirical results are:

First, presidents try to form majority coalitions, but it is the upper house majority not the lower house majority which makes cabinet coalitions significantly likely to from. One potential explanation for this phenomenon is that there are generally fewer parties in the upper than in the lower chamber, due to the disproportionality of electoral systems used to elect upper chambers in Latin America. Thus, the president’s party is often overrepresented in the upper house, which makes it easier for presidents to find majorities. Furthermore, upper chambers are generally strong in Latin America (Nolte and Llanos 2004), and controlling an upper chamber is often sufficient for the president to prevent a veto override.

Second, contrary to expectations in the literature, extensive presidential decree powers decrease the probability of the occurrence of cabinets which control only a minority of seats in the lower house of congress. A potential explanation for this phenomenon is similar to the argument developed by Strøm (1990) for minority governments in parliamentary systems. Parties prefer to stay in opposition when the government has a weak independent influence on policy. The other explanation is that pre-electoral coalition formation is more prevalent when presidents’ institutional authority is high, as political actors make a relatively simple calculation about the benefits and the costs of coordination in presidential elections. The more powerful the president, the higher the incentives for pre-electoral coalition formation (Hicken and Stoll 2008; Freudenreich 2013). And if the a coalition is in power anyway, it is easier to extend this coalition to secure a majority in the lower house of congress.

Third, considerations of governability and pre-electoral bargaining describe two distinct yet compatible sets of factors that influence cabinet formation in presidential systems. Many cabinet coalitions in Latin America are congruent or extended versions of the pre-electoral coalition of the president and parties of the main presidential competitor are generally excluded from the cabinet, but these factors are distinct to the incentives of legislative bargaining. The explanatory power of variables associated with governability increases once variables of pre-electoral bargaining are included in the statistical model. For example, cabinet coalitions are more likely to form when they include the median party in the lower chamber of congress, but this effect is only statistically significant when one controls for the effects of pre-electoral bargaining.

Overall, the paper tries to show that an inclusive approach is necessary to study coalition dynamics in presidential systems. Pre-electoral commitments strongly affect cabinet formation and thereby also confound the relationship between cabinet formation, legislative bargaining and governability.

Literature

Cheibub, José A. 2007. Presidentialism, Parliamentarism, and Democracy. New York: Cambridge University Press.

Freudenreich, Johannes. 2013. Coalition Formation in Presidential Systems. Ph.D. diss., University of Potsdam.

Hicken, Allen, and Heather Stoll. 2008. Electoral Rules and the Size of the Prize: How Political Institutions Shape Presidential Party Systems. Journal of Politics 70, 4: 1109–27.

Linz, Juan J. 1994. Presidential or Parliamentary Democracy: Does it Make a Difference? In The Failure of Presidential Democracy: The Case of Latin America, ed. Linz and Arturo Valenzuela. Baltimore: Johns Hopkins University Press. 3–89.

Nolte, Detlef/Mariana Llanos. 2004. “Starker Bikameralismus? Zur Verfassungslage lateinamerikanischer Zweikammersysteme.” Zeitschrift für Parlamentsfragen 35: 113-131.

Pérez-Liñán, Aníbal 2007. Presidential Impeachment and the New Political Instability in Latin America. Cambridge University Press: New York.

Strøm, Kaare. 1990. Minority Government and Majority Rule. Cambridge: Cambridge University Press.