Chapter 1 Excerpt for Why Presidents Fail
ONE
Why Parties Matter for Presidential Survival
In 2019, Chile’s president Sebastián Piñera weathered the most violent social outburst since democracy was restored in 1990, while in 2021 President Iván Duque struggled to hold on to power beleaguered by massive street demonstrations in Colombia. Piñera and Duque have fared far better than other Latin American presidents. Only since 2018, Peru has witnessed the removal and resignation of three chief executives (Pedro Pablo Kuczynski, Martín Vizcarra, and Pedro Castillo). Bolivian president Evo Morales—once seen as an all-powerful leader—was forced to leave office and flee the country amid electoral fraud allegations and social turmoil in November 2019, whereas Brazil’s president Dilma Rousseff was ousted in 2016 in one of the most polarizing impeachment proceedings that Latin America has witnessed in the last decades. At the time of writing, a total of twenty-four Latin American chief executives had been forced out of office, whose successors had always been civilians, and without a democratic breakdown. This phenomenon is known by the literature on government instability as “presidential failure.”1
Interestingly, countries with relatively strong parties, such as Chile, Mexico, and Uruguay, have not experienced cases of presidential failure, whereas countries well known for their weak parties, such as Ecuador, Guatemala, and Peru, have seen several of their presidents forced out prematurely. In this book, I show that parties play an important part when it comes to presidential survival. For instance, it does not seem likely that ruling parties would try to unseat the chief executive in Chile, a country with (still) relatively institutionalized parties. I ask Senator Jaime Quintana (PPD), former president of the Chilean Senate, about what might happen if a ruling party turns its back on the president. His answer illustrates the reasoning about immediate rewards versus forthcoming costs in a country where parties are thought to continue to exist in the medium or long term:
I’ve never seen something like that. But even if a party faction might want to break from the government, ultimately, the party board would end up aligning itself with the president to whom they had promised their loyalty. . . . I think the price for breaking from a government to which [the party] swore to support is too high. . . . The price to be paid is to be regarded as an unreliable partner.2
By contrast, Peru is a place where presidents must be particularly wary of their own parties and ruling partners because of the short-term horizons and low levels of party institutionalization (PI) that characterize the country’s politics. For example, Mercedes Aráoz—former Peruanos por el Kambio (PPK, Peruvians for Change) legislator (2016–2019), minister (2006–2010), and vice-president (2016–2020)—says that, now, politicians from the ruling party in Peru do not face important political costs for abandoning the president.3 In fact, distancing from a “beaten” chief executive might serve them well if they want to win elections. Congresswoman Indira Huilca (2016–2020) from the progressive Peruvian party Frente Amplio (FA, Broad Front) states that, in general, parties and their legislative blocs are rather weak and are not necessarily connected with each other, which hinders their role of defending the government.4
Hence, in some countries, presidents are at still risk of failure when they are backed by an important share of members of congress if such a legislative majority is underpinned by rather weak political parties. In other countries, even opposition parties may provide an institutional exit that does not involve an early removal of a minority president in distress. An example of the former occurred in 2015 in Guatemala, where all legislators of the ruling Partido Patriota (PP, Patriotic Party) voted to strip President Otto Pérez Molina (2012–2015) of his immunity from prosecution, even though he was one of the PP’s founders. With a landslide vote (132–0–26) against him, Pérez Molina lost his immunity and resigned days later. Similarly, in Argentina in 2001, President Fernando de la Rúa (1999–2001), who by that time had been left to his own devices by the ruling parties, was forced to step down during the most serious politico-economic crisis in decades. On the other hand, in 2019, Chile’s president Sebastián Piñera (2018–2022) was besieged by massive street protests, riots, and calls for his resignation. Even though Piñera was a minority president who faced the most severe wave of antigovernment demonstrations since 1990, most left-of-center opposition parties were reluctant (and even objected) to unseat him. In these three countries with different levels of PI, presidents found themselves beset by intense social mobilizations, yet political parties behaved in dramatically dissimilar ways, significantly affecting the crises’ outcomes. These three cases illustrate the main point of this book: parties do affect the chances of a president to complete a term or to be ousted from office.
Even though parties undergird democratic politics in most advanced democracies, thus far, not a single comparative study has thoroughly analyzed the part played by parties in presidential survival and failures in Latin America. The book seeks to answer two research questions overlooked by the extant literature: how a country’s level of PI affects the occurrence of presidential failures, and whether comparatively strong ruling parties have any effect on presidents’ chances of fulfilling their terms in office.
The study of presidential survival and failure in Latin America has made significant progress in the last two decades. Among its major findings are the importance of the president’s legislative majority to help her or him stay in office; and how antigovernment demonstrations, economic recessions, and presidential scandals may dramatically increase the risk of early executive termination. The existing scholarship, however, has paid little attention to the part played by political parties. Apart from some cross-national works analyzing the effects of party system fragmentation (Álvarez and Marsteintredet 2010; Kim and Bahry 2008; Martínez 2017), and few country studies (Malamud 2015; Schamis 2002), the comparative examination on whether and how parties and PI may affect presidential survival remains mostly underexplored.
It is worth mentioning that Gretchen Helmke’s (2017) book on interbranch crises in Latin America is perhaps the only recent study that indirectly addresses the link between parties and institutional instability. Helmke demonstrates that presidents belonging to immature parties tend to launch attacks on the legislature and the courts. Additionally, Helmke’s work suggests that institutionalized parties are more successful in screening and isolating politicians with extremist and demagogic tendencies. Helmke’s argument about young parties is tied to institutionalization. Since institutionalization is a process that takes time to develop, young parties are more likely to be weakly institutionalized. However, there are some differences between Helmke’s book and this one. First, the events she focuses on (“attacks” against other branches), while significant from a democratic perspective, do not have the same domestic (or even international) repercussions as a presidential failure. Second, Helmke’s analysis of interbranch attacks overlooks the threats and challenges to presidential survival that originate outside of the congress. Her study does not examine why and how nongovernmental actors, such as social movements, street protestors, the Catholic Church, the Organization of American States (OAS), and business associations, may play a central role in government stability, depending on the level of PI in different countries.
Furthermore, my book’s unique contribution to the extant literature is to offer a theoretical and empirical approach to understanding how parties affect a president’s ability to stay in office. On the one hand, I propose and test an innovative hypothesis that suggests a nonlinear relationship between PI and presidential survival: presidents who rule countries with moderately institutionalized parties are at a greater risk of early removal than presidents governing countries with weak and, especially, strong parties. Complementing this approach, the book also focuses on the differences in strength between government and opposition parties. By combining these two party-driven explanations, the book provides a broader and more comprehensive assessment of how parties affect a president’s odds of staying in office.
Parties and Presidential (In)stability: A Theoretical Approach
Democracy and political parties go hand in hand, or as Schattschneider (2017: xxvii) famously put it, “modern democracy is unthinkable save in terms of parties.” Parties are fundamental for advancing and stabilizing democracy (Linz and Stepan 1996) and improving governance (Jones 2011; Levitsky 2003); are an essential part of government coalition (Abranches 1988; Chasquetti 2008; Cheibub, Przeworski, and Saiegh 2004); and are crucial for filtering out and isolating extremist and demagogic candidates who may be dangerous for democracy (Levitsky and Ziblatt 2018) and taming powerful presidents who might otherwise not resist the temptation to steamroll political adversaries or break democratic rules (Mainwaring and Scully 1995). Note that these tasks are more likely to be accomplished by strong parties than by weak ones. The book’s main argument is rooted in the importance of parties and their levels of institutionalization as unexplored and central factors behind presidential (in)stability.
In general, institutionalization broadly refers to the process through which an organization solidifies and attains stability over time (Huntington 1968; Panebianco 1988). A highly institutionalized organization is characterized by its adaptability, having agreed-upon mechanisms for resolving disputes, autonomy from wealth and violence, as well as subunits that are interdependent, and functionally and hierarchically differentiated (Huntington 1968; Panebianco 1988). Specifically, my definition of PI centers on three key aspects: organizational features, time horizons, and linkage with civil society. First, PI involves the ability of these parties to maintain organizational stability and adaptability over time, along with the presence of robust and well-developed bureaucratic structures throughout the country. Second, PI also refers to the degree to which a country’s major political parties can project a large “shadow of the future” (Axelrod 1984). That is, PI promotes cooperative and mutually beneficial behavior because parties anticipate repeated interactions and expect to face the consequences of their actions in the long term. Finally, PI also revolves around the establishment and maintenance of enduring linkages between political parties and various segments of civil society, such as student organizations and labor unions. In addition, the presence of committed party activists and supporters plays a crucial role. These factors collectively contribute to the major parties of a country becoming deeply entrenched in the minds of the electorate, recognized as indispensable components of the established political system, and fortified by a resilient societal foundation.
In the following paragraphs, I briefly review the literature on parties, upon which I then build the theoretical underpinnings to explain how parties and PI may affect chief executives’ survival. For simplicity, I divide the discussion into three dimensions: organizational capabilities, time horizons, and links with society.
Parties’ Organizational Capabilities
Parties’ organizational strength focuses on whether parties, in general, are able to prevent incompetent or radical politicians from occupying high office, limit potential power struggles, solve conflicts, reconcile divergent interests institutionally, and counter the president’s power. Helmke (2017) argues that well-established parties effectively filter out and isolate radical and demagogic candidates who may harbor antidemocratic tendencies. Moreover, political neophytes and outsiders have fewer chances to win the presidency when parties are strong and well institutionalized (Corrales 2009; Flores-Macías 2010; Samuels and Shugart 2010). By contrast, less-institutionalized parties are associated with the rise of amateur politicians who tend to “lack skills in bargaining and negotiation, coalition building, dealing with opponents, handling media, and other areas that are critical to governing and legislating,” who are more likely to “respond poorly to crises” (Levitsky 2018, 352) and battle established parties in exchange for electoral and political gains (Carreras 2013).
As a space for discussion, institutionalized parties are well equipped to reconcile different and often competing interests (Aldrich 1995; Brownlee 2007; Cox 1997), and “to preserve and advance the interests of politicians and their supporters” (Mozaffar and Scarritt 2005, 408), all of which discourage power struggles and foster stability. As such, parties contribute to holding the ruling elite together by encouraging accommodation among them, thus helping rulers to “survive” in power longer (Brownlee 2007).
Parties’ organizational features also influence their ability to curb (even challenge) presidents’ power. As institutionalized parties are more capable to monitor rulers, parties may act preemptively and address potential conflicts with the president even before they happen. All this disincentivizes confrontation between the president and parties, even his/her own. Moreover, a balance of power between the president and parties promotes cooperation and stability, whereas imbalances of power favoring the chief executive reduce incentives for presidents to compromise and bargain, thus heightening political tensions that could threaten presidential survival (Cox and Morgenstern 2002; Kim 2014; Shugart and Carey 1992; Valenzuela 2004). Institutionalized parties are more able to keep the president accountable and block his/her attempts to overstep his/her boundaries.
Time Horizons: The Shadow of the Future
Time horizons refers to the extent to which interactions among political players are driven by long-term views. Institutionalization is associated with prolonged periods of time during which an indefinite number of interactions take place (Mainwaring, Bizzarro, and Petrova 2018, 22–23). Institutionalized parties cast what Axelrod (1984) called a large “shadow of the future,” which incentivizes cooperation in forthcoming iterations. Vergara (2018, 70) states that “consolidated, stable political parties are institutions that lengthen the time spans in which politicians think. These longer time horizons, in turn, reduce irrationality and uncertainty within the political system.”
Under these settings, it is expected that even rival players come to realize that cooperation, rather than conflict, is more desirable and mutually beneficial. In this scenario, political players are more likely to weigh the potential negative long-term consequences of a decision against its short-term gains. It is also expected that players would work to prevent crises from happening or, at least, try to institutionally funnel them. By contrast, in countries with less-institutionalized parties, politics is usually characterized by short-term objectives with little regard for their future implications. In these countries, political crises may be seen as opportunities to weaken political adversaries and to reap any political benefit without further consideration of potentially damaging consequences that may arise in the future.
Additionally, when it comes to time horizons, we may think of two types of institutional equilibria: one in which politics is dominated by long-term views and another in which short-termism prevails. Institutional equilibria are characterized by the relatively high cost of unilaterally changing one’s behavior when other actors stick to the equilibrium strategy (Calvert 1995; Greif and Kingston 2011). In countries with high levels of PI, the equilibrium strategy is built around long-term horizons. That is, even if actors want to seek immediate gains at the expense of the rest, they are very likely to be punished (e.g., being excluded from future alliances or being the subject of retaliation in forthcoming moves), which would encourage them to adopt a more long-term-oriented behavior. By contrast, when PI is not high, it is more likely that the equilibrium strategy would hinge upon short-term horizons. This means that even if one political player wanted to play long-term politics, it would not pay off to do so alone since the rest would be securing the rewards of pursuing short-term objectives.
Presidential survival should be higher when politics is deeply influenced by long-term horizons. When political actors are to play along with (or against) each other indefinitely, cooperation is easier to achieve. This cooperation may involve refraining from removing a standing president, even if there are evident and immediate political gains from doing so. The opposition may realize that when their time to lead the government comes, it might be subject to similar actions. Political players, thus, may consider that it is in their best interest to sacrifice immediate individual political rewards in exchange for a more mutually beneficial outcome in the future. Hence, this dynamic would foster presidential survival. In countries where long-term horizons do not prevail, political actors are more prone to succumb to their short-term desires. Removing a president, even if not warranted, may result in a high political payoff in the present for those pursuing it, regardless of any future retaliation. Under this scenario, presidential survival is more fragile, as even partners of the ruling coalition may well withdraw their support from a beset president and join the opposition in their attempt to unseat him/her.
Links with Society
Furthermore, parties develop different types of linkage with civil society, some of which are of great importance for government stability. For instance, ruling parties’ linkage with student organizations (Disi Pavlic 2018, 451) and labor unions (Corrales 2002, 34) has been found to reduce the likelihood of street demonstrations against the government, whereas institutionalized parties are found to “diminish ‘bottom-up’ challenges that could come from defecting voters” (Greene and Sánchez-Talanquer 2018, 202). Moreover, after long periods of time, parties tend to also be “reified” in the electorate’s mind—insofar as people come to regard them as an integral part of the established political system (Janda 1980; Randall and Svåsand 2002) as well as having developed “stable roots in society” (Casal Bértoa 2017). This dimension emphasizes how parties may help to reduce the risk of social uprisings and bottom-up crises, which makes presidents’ forced removals less likely.
The Argument
Drawing on the theoretical underpinnings reviewed in the previous section that inspired this work, here I offer specific explanations on how parties may affect presidential survival and failures. I begin by discussing the nonlinear relationship between PI and presidential survival, and how parties and presidents are expected to behave across different levels of PI. I then focus on the ruling parties’ comparative strength versus the opposition.
PI’s Nonlinear Effects on Presidential Survival
One of the book’s main hypotheses suggests that there is a nonlinear relationship between PI and presidential survival. That is, at either end, PI is associated with a lower risk of presidential failure than when PI is moderate. These outcomes are the result of the combination of organizational features, time horizons, and linkage with civil society. In the following paragraphs, I explain how these features affect presidential survival and crises.
As discussed in the previous section, I argue that PI may affect the likelihood of a presidential crisis, as well as its development once it has unraveled. PI changes slowly over time, and not all presidents fail in countries with low or moderate levels of PI. That is why I argue that PI is not a direct cause (trigger) of presidential crises and failures, but rather a structural, contextual force that affects the outcomes more distantly. Triggers for crises may include economic downturns, presidential scandals, massive protests, recalcitrant congressional opposition, and other events. Below, I hypothesize how presidents, parties, and other actors may behave across different levels of PI.
LOW LEVELS OF PI
In countries with low PI, a stable presidency might be the result of the president’s strength vis-à-vis other players. That is, the president may remain in office because parties are too weak to challenge or even constrain him/her. In this scenario, the “shadow of the future” is rather limited; hence, players are very likely to seek short-term goals. One such goal may be unseating a president on very feeble grounds. However, weak parties are probably unable to effectively organize and coordinate the president’s removal.
When PI is low, it is easier for outsiders and political neophytes, who lack negotiation skills and experience, to rise. Also, unconstrained presidents are likely to engage in unilaterally risky behavior, such as “constitutional hardball” (Tushnet 2004).5 Both outsiders and aggrandized presidents are associated with conflict-prone scenarios. Yet, under these settings, the major threat to presidential survival may come from the occurrence of crisis-triggering events (e.g., corruption scandals, economic recessions, and antigovernment demonstrations, among others), which might seriously destabilize the president. Because parties either barely exist or are too weak, they would have limited (or no) impact throughout the crisis. This means that the president is virtually on his/her own once the crisis has unfolded. In this case, it is not unlikely that political actors other than parties (e.g., Catholic Church, international organizations, courts, business associations, etc.) may take on a more active role during presidential crises.
MODERATE LEVELS OF PI
When PI is moderate, presidents can survive in office by establishing ad hoc, yet undependable, alliances with key political players, be they individual politicians or parties. Because under this scenario parties are somewhat institutionalized, they may try to constrain or undermine the president but do not always succeed, which might create severe political friction between them. Because party-based mechanisms—inter-and intraparty—to reconcile competing interests are not developed well enough, these power struggles may eventually escalate into destabilizing conflicts.
Since politics is not characterized by long-term horizons, interactions are more prone to conflict, whereas cooperation is unlikely to emerge when a presidential crisis strikes. In fact, parties might be tempted to seek short-term gains from a crisis at the expense of the president, which might fuel the crisis even further. Even ruling parties may want to withdraw their support from a beleaguered president, thus weakening his/her chances of holding on to power. Yet, last-minute agreements, even if based on self-serving and questionable grounds, could be brokered during a government crisis to enable the president to remain in office. However, by and large, because long-term horizons and cooperation are not well rooted and politics is not dominated by experienced politicians, these alliances are probably unreliable and short lived, especially if a sociopolitical conflict escalates.
HIGH LEVELS OF PI
Finally, in countries with high levels of PI, government stability is the result of well-established long-term informal and formal interactions built upon a sufficiently large “shadow of the future,” which facilitate bargaining and cooperation. Moreover, strong parties are able to reconcile divergent interests, provide politicians with political experience and negotiation skills, curb the president’s authority, and force him/her to negotiate, all of which reduce the likelihood of a crisis.
Yet, a presidential crisis may still occur, sparked by factors other than parties and president-party relations, such as antigovernment mobilizations and economic downturns. When a presidential crisis is in place, it is expected that major political players would attempt to contain it or stop it from escalating. Parties would seek out an institutional exit from the crisis, one that might not involve the president stepping down. Doing so is, clearly, a costly move for parties, especially those in the opposition, and when the president in question is unpopular. One of the reasons for such behavior is that stable parties recognize the potential long-term benefits of preserving stability and avoiding chaotic transitions of power, which can lead to prolonged political and economic instability and damage their standing with the electorate. Perhaps more egoistically, established parties also expect to continue playing the political game against the same competitors indefinitely. This expectation implies that the opposition, even if it had the opportunity, would be reluctant to remove a rival president. They understand that in the future, the political dynamics might shift, and the ousted president could potentially become a member of their own party, creating a precarious precedent that might come back to haunt them. Here, the dominant institutional equilibrium is built upon long-term horizons. In summary, parties are organizationally capable of unseating the president, yet their long-term views deter them from pursuing such courses of action, which fosters presidential survival.
Ruling Parties’ Comparative Strength
It is a no-brainer that in fútbol, basketball, and most team sports, the number of players that each team has is not enough to determine which side will win. Player-coach-management relations, facilities for training, preparation, fan base, economic resources, shared sense of purpose, and skills, among others, also matter for winning. What this example suggests for analyzing the role of parties in government stability is that it is not enough or accurate to consider presidents’ partisan support mostly in terms of the share of the legislative seat that his/her party or coalition nominally holds. It is necessary to take into consideration other party-related attributes.
For example, determining whether legislators of nominal government parties will come to the rescue of a beset president is also influenced by the parties’ characteristics. Institutionalized parties are more able to enforce discipline and infuse their members with a sense of the long term (Alesina and Spear 1988; Levitsky 2018; Stokes 2001; Mainwaring 2018), which incentivizes politicians to stand by their president and increases the costs of (even temporarily) siding with the opposition. Furthermore, presidents would be more reassured if ruling parties were able to solve conflicts, reward disciplined members, punish bad behavior, channel and articulate members’ demands, keep vibrant activism, and develop links with the society, all key features associated with institutionalized parties (Rosenblatt 2018; Mainwaring, Bizzarro, and Petrova 2018; Huntington 1968; Panebianco 1988; Selznick 1992; Casal Bértoa 2017; Janda 1980; Randall and Svåsand 2002). All in all, parties that can muster these attributes, all things being equal, have better chances to defeat those who do not.
Therefore, when it comes to government stability, the success of opposition parties to oust a chief executive depends on the balance of power between them and ruling forces. The Federalist, no. 65 (Hamilton, Madison, and Jay 2009), already addressed this idea when referring to the logic behind impeachments by stating that “there will always be the greatest danger that the decision will be regulated more by the comparative strength of parties [my emphasis], than by the real demonstrations of innocence or guilt.” This conception involves a more comprehensive understanding of parties’ political muscle, one that includes aspects besides the mere share of legislative seats held by the opposition and ruling parties.
Furthermore, Pérez-Liñán (2014, 38) states that “loyal congress members will openly reject calls for the resignation of the president” and that “loyal politicians with a territorial base” may marshal popular support for a besieged president to counter the impact of antigovernment demonstrations. These two statements are also associated with individual parties’ levels of institutionalization, specifically the depth and scope of their organizational capacity (e.g., internal unity, stable bureaucratic structures, and mechanisms to address and solve inner conflicts) and their links with civil society. In fact, to different degrees and among several dimensions, there are parties famous for their strength. This is the case of peronismo and its well-known links with unions and civil society organizations in Argentina, the Partido dos Trabalhadores (PT, Workers’ Party) and its high level of institutionalization in Brazil, or Movimiento al Socialismo (MAS, Movement toward Socialism) in Bolivia and its party structure made up of hundreds of smaller social organizations.
The share of seats in congress held by ruling parties plays a pivotal role in presidential survival, as extensively documented in existing literature. Nevertheless, the strength and reliability of this “legislative shield” can vary based on the institutionalization level of the legislative parties within the ruling coalition. This is why I argue that, drawing from Aníbal Pérez-Liñán’s work (2007), ruling parties can function as a “partisan shield” for presidents.
Strength is not an absolute state that parties can achieve, as it is inherently linked to other actors’ might. Hence, the effectiveness of the “partisan shield” naturally depends on the balance of power between the governing and opposition parties. Simply put, I hypothesize that when ruling parties with legislative presence are more institutionalized than the legislative opposition parties, it would be more difficult to unseat an incumbent president. This is what I label ruling parties’ comparative strength.
Scenarios for Presidential Survival Depending on PI and Ruling Parties’ Comparative Strength
On one hand, PI is conceptualized as a structural force, functioning akin to a baseline risk that presidents face since their inauguration day. PI impacts contextual dynamics among political actors, thus indirectly influencing government stability. On the other hand, the ruling parties’ comparative strength represents the more immediate effects of the tug-of-war between the president’s party (or coalition) and the opposition.
Table 1.1 provides a summary of various theoretical scenarios depicting presidential survival, ranging from adverse and challenging to viable, favorable, and advantageous. To clarify the impact of both party-based variables on government stability, let me focus on their most positive values. This occurs when PI is high or when ruling parties are stronger than the partisan opposition with legislative presence. When any of these conditions is present, presidential survival significantly improves. However, between these two variables, I posit that PI exerts a more substantial influence on presidential survival. In this regard, the most stable scenarios for chief executives arise when PI is high, irrespective of the balance of power between ruling and opposition parties, which is why they vary from favorable to advantageous. The rationale behind this lies in the fact that under high PI levels, a significant “shadow of the future” fosters cooperation even among rival parties. This cooperation serves to prevent destabilizing conflicts or, in case they do occur, channel them through institutional means, thereby reducing the risk of early government termination. As a matter of fact, since 1979 not a single presidential failure has taken place in countries with high PI despite the occurrence of crisis-triggering events such as massive and intense street demonstrations or presidential scandals.
In comparison, possessing a relatively robust “partisan shield,” characterized by ruling parties having greater strength than the congressional opposition parties, provides presidents with a significant advantage in terms of survival. Nonetheless, the risk of being compelled to step down is not negligible. Despite the ruling parties’ superiority over the opposition, there have been instances of three presidential failures under this scenario between 1979 and 2020.6 For example, Evo Morales (Bolivia) in 2019 was forced to leave office prematurely, although his ruling party was stronger than the opposition parties (figure 2.2). This is why I consider that, for example, the existence of strong ruling parties makes it hypothetically viable—at best—for presidents to achieve political survival when PI is moderate, rather than finding themselves in a more promising scenario.
Now, let’s turn our attention to the least optimistic settings. As table 1.1 illustrates, when ruling parties are weaker than the opposition, heads of governments are confronted with adverse or challenging political scenarios when PI is moderate and low, respectively. As previously mentioned, presidents may encounter more significant obstacles for government stability when PI is moderate vis-à-vis when PI is low and, especially, high. Similarly, I expect that presidents would face slightly more difficulties under moderate levels of PI (challenging scenario) compared to under low levels of PI (viable scenario) when ruling parties and opposition parties have similar strength. However, while presidential survival is viable when both governing and opposition parties possess equivalent strength under overall low PI, it is important to note that in such circumstances, both sides might be equally powerless. This equilibrium, while likely to prevent the opposition from overthrowing the president, is fragile and can be easily disrupted by a presidential scandal, antigovernment protests, or economic crises.
TABLE 1.1. Hypothesized scenarios for presidential survival
Source: Elaborated by the author. Finally, this theoretical framework is built on the notion that government survival depends on both attributes of actors and the political system as well as events that are difficult to predict (Warwick 1994).7 Hence, in this approach, PI and ruling parties’ comparative strength are considered as attributes of the political system and actors, respectively, while other known predictors of government survival, such as presidential scandals, antigovernment demonstrations, hyperinflation, and economic recessions—which are put to the test in chapter 2 and examined in chapters 3–9—are understood as events.
Notes
1. Throughout the book, I use the terms “presidential failure,” “failed president/presidency,” “presidential fall,” or “interrupted presidency” interchangeably.
2. Personal interview.
3. Personal interview.
4. Personal interview.
5. Similarly, Pérez-Liñán et al. (2019) show that there is an inherent degree of government instability linked to presidential aggrandizement.
6. In contrast, 50 percent (11 of 22) of presidential failures during this period have taken place when ruling parties are weaker than the opposition. Whereas only eight cases, accounting for 36 percent of all interrupted presidencies, occurred when ruling parties and the opposition were relatively evenly matched in strength.