![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
We develop a unified theory of presidential instability to explain why presidents are removed from office through military coups or through legal procedures. While some causal factors motivate opponents to overthrow the president irrespective of the specific mechanism employed, other factors expand the relative capabilities of groups inclined to pursue military or civilian action. The first group of variables, including economic recession, protests, and radicalization, explains why presidents fall. The second set of variables, including regional diffusion, partisan support for the executive, and normative support for democracy, explains how they are ousted. We test this theory using discrete-time event history models with sample selection on a novel database for 19 Latin American countries between 1945 and 2010.
Acknowledgements
We are indebted to Gustavo Emmerich, Guillermo Mira, Fernando Pedrosa, Laura Tedesco, Kathy Hochstetler, and two anonymous reviewers for their valuable comments. Earlier versions of this article were presented at the International Political Science Association (IPSA) Conference in Madrid in 2012, at the International Congress of Americanists (ICA) in Vienna in 2012, at the University of Salamanca in May 2013, and at the American Political Science Association (APSA) in Washington, DC, in 2014.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes on contributors
Aníbal Pérez-Liñán is Professor of Political Science and member of the Center for Latin American Studies at the University of Pittsburgh. He is also Distinguished Research Affiliate (2015–2017) at the Kellogg Institute for International Studies, University of Notre Dame. His work focuses on democratization, political institutions, and regime stability. He is author of Presidential Impeachment and the New Political Instability in Latin America (Cambridge, 2007).
John Polga-Hecimovich is an Assistant Professor of Political Science at the US Naval Academy. His research interests include comparative institutions of Latin America, especially the executive and the bureaucracy, as well as presidential instability. He has published peer-reviewed articles in The Journal of Politics, Political Research Quarterly, Electoral Studies, Party Politics, Latin American Politics and Society, and others.
Notes
1. Marsteintredet, “Explaining Variation”; Pérez-Liñán, “A Two-Level Theory.”
2. Belkin and Schofer, “Toward a Structural Understanding”; Collier and Hoeffler, Grand Extortion; Fitch, “Post-Transition Coups”; Nordlinger, Soldiers in Politics; O’Kane, “A Probabilistic Approach.”
3. Baumgartner and Kada, Checking Executive Power; Llanos and Marsteintredet, Presidential Breakdowns; Pérez-Liñán, Crisis Without Breakdown.
4. Álvarez and Marsteintredet, “Presidential and Democratic Breakdowns”; Kim and Bahry, “Interrupted Presidencies.”
5. Needler, “Political Development and Military Intervention,” 616.
6. Fossum, “Factors Influencing the Occurrence.”
7. Lowenthal, “Review,” 107.
8. Dix, “Military Coups and Military Rule”; Powell and Thyne, “Global Instances of Coups.”
9. Croissant, “Coups and Post-Coup Politics,” 267.
10. Kada, “Impeachment as a Punishment for Corruption?”; Pérez-Liñán, “Pugna de Poderes y Crisis de Gobernabilidad.”
11. Llanos and Marsteintredet, Presidential Breakdowns; Valenzuela, “Latin American Presidencies Interrupted.”
12. Fernando Collor de Mello (Brazil, 1992), Carlos Andrés Pérez (Venezuela, 1993), Abdalá Bucaram (Ecuador, 1997), Raúl Cubas Grau (Paraguay, 1999), Lucio Gutiérrez (Ecuador, 2005), Fernando Lugo (Paraguay, 2012), Otto Pérez Molina (Guatemala, 2015), and Dilma Rousseff (Brazil, 2016).
13. Hernán Siles Zuazo (Bolivia, 1985), Raúl Alfonsín (Argentina, 1989), Alberto Fujimori (Peru, 2000), Fernando de la Rúa (Argentina, 2001), and Gonzalo Sánchez de Lozada (Bolivia, 2003). Joaquín Balaguer (Dominican Republic, 1996) resigned after two years of his (seventh) term in office, following an agreement to overcome an electoral dispute.
14. Rosalía Arteaga (Ecuador, 1997), Alberto Rodríguez Saá (Argentina, 2002), Eduardo Duhalde (Argentina, 2003), Carlos Mesa (Bolivia, 2005), and Eduardo Rodríguez Veltzé (Bolivia, 2006).
15. Jorge Serrano (Guatemala, 1993), Jamil Mahuad (Ecuador, 2000), and Manuel Zelaya (Honduras, 2009).
16. Llanos and Marsteintredet, Presidential Breakdowns.
17. Valenzuela, “Latin American Presidencies Interrupted.”
18. Kim and Bahry, “Interrupted Presidencies.”
19. Pérez-Liñán, Crisis Without Breakdown.
20. Rittinger and Cleary, “Confronting Coup Risk”; Taylor-Robinson and Ura, “Public Opinion and Conflict.”
21. For a careful review of theories emphasizing motivations and capabilities, see Zimmermann, Political Violence; Brooker, Non-Democratic Regimes; and Belkin and Schofer, “Toward a Structural Understanding.”
22. This approach does not deny that particular groups will have distinct motivations to overthrow the president. For example, military officers may pursue a coup to protect distinctive institutional interests (see Needler, “Military Motivations”). But general motivations, such as economic recession or mass protests, facilitate the formation of broad social coalitions against the president.
23. See, for instance, Belkin and Schofer, “Toward a Structural Understanding”; Luttwak, Coup d’Etat, 28–56; Finer, The Man on Horseback.
24. For a rare empirical study that compares the causes of both outcomes, see Álvarez and Marsteintredet, “Presidential and Democratic Breakdowns.” For a study that treats those outcomes jointly but without distinction, see Kim and Bahry, “Interrupted Presidencies.”
25. Needler, “Political Development and Military Intervention”; Finer, The Man on Horseback; Needler, Latin American Politics; Luttwak, Coup d’Etat.
26. Nordlinger, Soldiers in Politics; O’Kane, “A Probabilistic Approach.”
27. Londregan and Poole, “Poverty, the Coup Trap, and the Seizure.”
28. Przeworski et al., Democracy and Development.
29. Hiroi and Omori, “Causes and Triggers of Coups d’Etat”; Singh, Seizing Power.
30. Merkx, “Recessions and Rebellions”; Kim, “Revisiting Economic Shocks and Coups.”
31. Álvarez and Marsteintredet, “Presidential and Democratic Breakdowns.”
32. Hochstetler and Edwards, “Failed Presidencies.” But see, Kim, “Impeachment and Presidential Politics.”
33. Helmke, “The Origins of Institutional Crisis.”
34. Huntington, Political Order; Álvarez and Marsteintredet, “Presidential and Democratic Breakdowns”; Casper and Tyson, “Popular Protest and Elite Coordination.”
35. Singh, Seizing Power.
36. Hochstetler, “Rethinking Presidentialism”; Kim, “Impeachment and Presidential Politics.”
37. Pérez-Liñán, “A Two-Level Theory.”
38. Protests may also reflect the political ability of social movement leaders. But a focus on leaders’ capabilities does not yield any general prediction on whether groups supporting a coup or a legal removal would gain an advantage. Elites may use protests as an excuse to demand the legal resignation of the president or to overthrow the government by force.
39. Finer, The Man on Horseback; Putnam, “Toward Explaining Military Intervention.”
40. O’Kane, “A Probabilistic Approach”; Londregan and Poole, “Poverty, the Coup Trap, and the Seizure”; Lehoucq and Pérez-Liñán, “Breaking Out of the Coup Trap”; Singh, Seizing Power.
41. Helmke, “The Origins of Institutional Crisis,” 742.
42. Hochstetler and Samuels, “Crisis and Rapid Reequilibration.”
43. Marsteintredet, “Explaining Variation.”
44. Fossum, “Factors Influencing the Occurrence.”
45. Pitcher, Hamblin, and Miller, “The Diffusion of Collective Violence”; Brinks and Coppedge, “Diffusion is no Illusion.”
46. Gleditsch, All International Politics is Local; Weyland, Making Waves.
47. Mainwaring and Pérez-Liñán, Democracies and Dictatorships.
48. Miller, Joseph, and Ohl, “Are Coups Really Contagious?”
49. Legler and Tieku “What Difference”; “McCoy “International Response”; Pérez-Liñán, Crisis Without Breakdown; Wobig, “Defending Democracy.”
50. Bruneau and Trinkunas, “Democratization as a Global Phenomenon.”
51. Linz, “The Perils of Presidentialism”; Linz and Valenzuela, The Failure of Presidential Democracy; Valenzuela, “Latin American Presidencies Interrupted.” Huntington (Political Order) argued that weak political institutions increase the probability of a coup.
52. Mainwaring, “Presidentialism, Multipartism, and Democracy”; Mainwaring and Shugart, “Conclusion.”
53. Cheibub, “Minority Governments, Deadlock Situations.”
54. Helmke, “The Origins of Institutional Crisis.”
55. Svolik, “Which Democracies Will Last?”
56. Hinojosa and Pérez-Liñán, “Presidential Impeachment and the Politics of Survival”; Hochstetler, “Rethinking Presidentialism”; Kim, “Impeachment and Presidential Politics”; Negretto, “Minority Presidents”; Pérez-Liñán, “A Two-Level Theory”; Pérez-Liñán, Crisis Without Breakdown.
57. Negretto, “Minority Presidents”; Kim, “Impeachment and Presidential Politics.”
58. For example, Linz, Crisis, Breakdown, and Reequilibration. But see also Croissant, “Coups and Post-Coup Politics”; Hochstetler, “Rethinking Presidentialism.”
59. Elite preferences are potentially endogenous to regime instability. Lack of space prevents us from addressing this econometric issue in the article, but to dispel concerns see Mainwaring and Pérez-Liñán, Democracies and Dictatorships, Chapter 4.
60. García Holgado, “Vencedores y vencidos.”
61. Linz, Crisis, Breakdown, and Reequilibration.
62. Mainwaring and Pérez-Liñán, Democracies and Dictatorships; Benavente Urbina, “Populismo Radical.”
63. Presidents were observed at 1 January of each year, and selected only if the political regime was coded by Mainwaring, Brinks, and Pérez-Liñán (Classifying Political Regimes) as a democracy or semi-democracy. The countries covered by the study are Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, the Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay, and Venezuela.
64. Existing datasets, such as Archigos, did not provide sufficient information to test our hypotheses. Exits coded as “regular” by Archigos describe presidents who completed their terms as well as legal removals, including impeachments and forced resignations. See Goemans, Gleditsch, and Chiozza, “Introducing Archigos.”
65. We do not pass judgment on whether the procedure was properly implemented from a legal standpoint. See Marsteintredet, “Explaining Variation of Executive Instability,” on this issue.
66. Venezuela 2002 was coded as a coup event because another administration took office, even though President Chávez returned to power within two days.
67. This indicator is not free from pitfalls. Brockett documents undercounting of political protest and violence in databases drawing events from The New York Times and Keesing’s Record of World Events, while Herkenrath and Knoll find large differences between national and international coverage of protests. Unfortunately, no alternative sources cover such a long span (1945–2010) for all Latin America. Yet, the large mass protests that attract international attention also signal the presence of broad anti-government coalitions. Brockett, “Measuring Politcal Violence”; Herkenrath and Knoll, “Protest Events in International Press Coverage.”
68. Mainwaring and Pérez-Liñán, Democracies and Dictatorships. The authors relied on historical sources to identify 1460 political actors under 290 administrations in 20 countries between 1944 and 2010. Political actors are individuals (the president, prominent leaders) and organizations (military factions, parties, social movements, trade unions) that played an important role in the competition for power.
69. A historical legacy in the form of a coup trap is conceptually distinct from the more direct “regime origin legitimacy effect,” the idea that leaders entering power through extra-constitutional means (for example, a coup) have a higher probability of exiting office the same way. We do not consider this possibility since we exclude authoritarian governments from our empirical database.
70. Box-Steffensmeier and Jones, Event History Modelling, 70.
71. Carter and Signorino, “Back to the Future.” See also Beck, Katz, and Tucker, “Taking Time Seriously,” and Box-Steffensmeier and Jones, Event History Modelling, 75.
72. Moreover, unobserved variables may simultaneously affect the decision to overthrow the government and the modality employed, potentially biasing multinomial estimates because the three choices are not mutually independent.
73. Van de Ven and Van Praag, “The Demand for Deductibles in Private Health Insurance.”
74. Radicalism does not separate coups from impeachments. In a model including radicalism in the outcome equation for Model 3.2, the second-stage estimate for this variable is insignificant. More information is available in the replication file for this article.
75. Hanmer and Kalkan, “Behind the Curve.”
76. Brinks and Coppedge, “Diffusion is no Illusion”; Fossum, “Factors Influencing the Occurrence”; Mainwaring and Pérez-Liñán, Democracies and Dictatorships; Pitcher, Hamblin, and Miller, “The Diffusion of Collective Violence.”
77. Pérez-Liñán, “A Two-Level Theory.”
78. Mainwaring, “Presidentialism, Multipartism, and Democracy.”
79. Llanos and Marsteintredet, “Ruptura y continuidad,” 180.
80. Taylor-Robinson and Ura, “Public Opinion and Conflict.”
81. Pesquisa CNI-IBOPE, Avaliação do Governo, Resultados Gerais. Março 2016.
82. More intriguing was the impeachment of Paraguayan President Fernando Lugo in June 2012, because preconditions for instability, with the possible exception of radicalism, appeared to be absent. See Cerna Villagra and Solís Delgadillo, “La crisis institucional paraguaya”; Pérez-Liñán, “A Two-Level Theory”; Marsteintredet, Llanos, and Nolte, “Paraguay and the Politics of Impeachment.”
83. Linz, Crisis, Breakdown, and Reequilibration.
84. Bermeo, Ordinary People; Mainwaring and Pérez-Liñán, Democracies and Dictatorships.