Fragmentation of presidential elections and governability crises in Latin America: a curvilinear relationship?

ABSTRACT

Part of the literature views high numbers of presidential candidates as a threat to political stability in presidential democracies. A contradictory model proposes that an overconcentration of the presidential party system is problematic. Both models are hard to reconcile. We approach this puzzle by arguing that the relationship between the level of presidential election fragmentation and governability crises is curvilinear: both very low and very high effective numbers of presidential candidates increase the risk of governability crisis. We test this theoretical claim with ordered logit models drawing on a sample of 108 presidencies in Latin America between 1978 and 2013 and using an ordinal index of the intensity of crisis as the dependent variable. We explore the operation of the theorized causal mechanisms through case studies and argue that they are different at both extremes, high and low levels of fragmentation. Finally, we formulate implications for the design of presidential electoral rules drawing on the debate contrasting runoff and plurality rules.

KEYWORDS:

• governability crises; Latin America
• presidential politics
• electoral competition
• democracy

Acknowledgements

We thank the two anonymous reviewers at Democratization for their valuable comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Karel Kouba http://orcid.org/0000-0002-0974-9144

Tomáš Došek http://orcid.org/0000-0003-4297-8456

Notes

1. Valenzuela, “Latin American Presidencies Interrupted.”

2. Marsteintredet and Berntzen, “Reducing the Perils of Presidentialism”; see also Hochstetler and Samuels, “Crisis and Rapid Reequilibration.”

3. See Pérez-Liñán, “Evaluating Presidential Runoff Elections.”

4. Valenzuela, “Latin American Presidencies Interrupted,” 12.

5. Linz, “The Perils of Presidentialism.”

6. Linz, La quiebra de las democracias; Linz and Valenzuela, The Failure of Presidential Democracy.

7. Mainwaring, “Presidentialism, Multipartism, and Democracy.”

8. Stepan and Skach, “Constitutional Frameworks.”

9. Cheibub, Presidentialism, Parliamentarism, and Democracy; Chasquetti, “Elecciones presidenciales mayoritarias.”

10. Álvarez and Marsteintredet, “Presidential and Democratic Breakdowns.”

11. Mejía Acosta and Polga-Hecimovich, “Parliamentary Solutions.”

12. Buitrago, “Civil Society.”

13. In the post-1988 period, Brazil was considered an exception (together with Chile) and a prime example of a coalitional presidentialim (see note 27). After the impeachment of Fernando Collor de Mello in 1992, Brazil enjoyed considerable presidential stability until the removal of Dilma Rousseff in 2016. In fact, some authors contrasted precisely the Brazilian experience in governability and more stable public policies with the pattern of institutional instability and volatile public policies in Ecuador (see Pereira and Mejía Acosta, “Policymaking in Multiparty”), despite the fact that both countries suffer from a highly fragmented party system (the differences are owed to divergent nature of the governing coalitions). In the end, with the impeachment of Rousseff, Brazil seems to have converged to the Ecuadorian case, as Rousseff was removed due to the absence of her own legislative shield and in the context of economic crisis, corruption scandals and social mobilizations (all commonly mentioned variables triggering governability crisis).

14. Pérez-Liñán, Presidential Impeachment.

15. Chasquetti, “Elecciones presidenciales mayoritarias,” 49.

16. Pérez-Liñán, “Evaluating Presidential Runoff Elections.”

17. Colomer, “Non-Median and Condorcet-Loser Presidents”; Lehoucq, “The Third Wave of Democracy,” 253; McClintock, “Reevaluating Runoffs.”

18. Jones, “Electoral Institutions”; Golder, “Presidential Coattails and Legislative Fragmentation.”

19. McClintock, “Reevaluating Runoffs,” 98–100.

20. George and Bennett, Case Studies.

21. Samuels, “Presidentialized Parties,” 468.

22. Hicken and Stoll, “Presidents and Parties,” 848; Golder, “Presidential Coattails and Legislative Fragmentation.”

23. Golder, “Presidential Coattails and Legislative Fragmentation.”

24. Hicken and Stoll, “Presidents and Parties.”

25. Williams-Wyche, “An Empirical Test.”

26. Chaisty et al., “Rethinking the ‘Presidentialism Debate’” propose a more optimistic view. In their argument, presidents in highly fragmented context can make use of a diverse “set of tools” in order to forge a coalititional dynamics that mitigate the inherent perils of Mainwaring’s “difficult combination” (see Mainwaring, “Presidentialism, Multipartism, and Democracy”) of presidentialism and multipartism. The authors take Brazil as the most illustrative case in the Latin American region. However, even when the president has a considerable toolbox to temper the perils of multipartism (as in the Brazilian case), Dilma Rousseff was finally ousted from the presidency in 2016.

27. Golder, “Presidential Coattails and Legislative Fragmentation,” 47.

28. Mainwaring, Rethinking Party Systems, 338.

29. Corrales, “Latin America’s Neocaudillismo,” 19.

30. Jones, “Electoral Institutions,” 75.

31. Ibid.

32. Shugart and Carey, Presidents and Assemblies, 215.

33. McClintock, “Reevaluating Runoffs,” 101.

34. Ibid.

35. Ibid., 98–100.

36. We do not delve further into these cases as the mechanisms are well-known. See Freidenberg, El sueño frustrado or Mejía Acosta and Polga-Hecimovcich, “Parliamentary Solutions,” for the landmark case of Ecuador, Ollier, “Argentina: Up a Blind Alley Once Again” for Argentina, and Pérez-Liñán, Presidential Impeachment, Linz and Valenzuela, The Failure of Presidential Democracy, Valenzuela, “Latin American Presidencies Interrupted” or Llanos and Marsteintredet, Presidential Breakdowns, for a more comparative view.

37. We thank the anonymous reviewer for suggesting both points.

38. Elgie et al., “Proximity, Candidates, and Presidential Power,” 468.

39. The following analyses of such a relationship were considered: Neto and Cox, “Electoral Institutions”; Mozaffar, Scarritt and Galaich, “Electoral Institutions, Ethnopolitical Cleavages”; Golder, “Presidential Coattails and Legislative Fragmentation”; Golosov and Kalinin, “Presidentialism and Legislative Fragmentation”; Elgie et al., “Proximity, Candidates, and Presidential Power”; Stoll, “Presidential Coattails.”

40. Neto and Cox, “Electoral Institutions”; Jones, “Electoral Laws”; Golder, “Presidential Coattails and Legislative Fragmentation,” 45; Dickson and Scheve, “Social Identity, Electoral Institutions and the Number of Candidates.”

41. Ordeshook and Shvetsova, “Ethnic Heterogeneity”; Clark and Golder, “Rehabilitating Duverger’s Theory”; Neto and Cox, “Electoral Institutions”; Birnir and Van Cott, “Disunity in Diversity.”

42. Golder, “Presidential Coattails and Legislative Fragmentation.”

43. Neto and Cox, “Electoral Institutions”; Mozaffar, Scarritt and Galaich, “Electoral Institutions, Ethnopolitical Cleavages”; Golosov and Kalinin, “Presidentialism and Legislative Fragmentation”; Stoll, “Presidential Coattails.”

44. Hicken and Stoll. “Are all Presidents Created Equal?”

45. Elgie et al., “Proximity, Candidates, and Presidential Power.”

46. Hugo Banzer (Bolivia), Jaime Roldós (Ecuador).

47. Pérez-Liñán, “Evaluating Presidential Runoff Elections.”

48. Ibid., 139.

49. We take this variable and its operationalization from Pérez-Liñán and Polga-Hecimovich, “Explaining Military Coups.”

50. Doyle and Elgie, “Maximizing the Reliability.” The authors construct a new and more reliable measure of presidential powers based on weighting all existing measures in the literature and thus create a measure available for a greater number of countries and longer time period. Here we use the prespow1 scores that are more reliable for the Latin American cases.

51. We thank an anonymous reviewer for highlighting this point.

52. Unless otherwise stated, the values of all institutional variables come from official electoral results available online and collected by the authors.

53. We also calculated the effective number of electoral parties and included this measure in separate models replacing the number of legislative parties. The results did not influence the main interpretations of the models. We consider legislative party competition as a more adequate measure for theoretical reasons, so we opted for its inclusion into the final models. All the data come either from the Indicadores section of the Observatory of Latin American Political Parties (Observatorio de Partidos Políticos de América Latina, OPAL) at http://americo.usal.es/oir/opal/indicadores.htm or from our calculations based on official data available from the electoral institutions of each country.

54. Álvarez and Marsteintredet, “Presidential and Democratic Breakdowns”; Basabe-Serrano and Polga-Hecimovich, “Desempeño económico.”

55. Hochstetler, “Rethinking Presidentialism.”

56. We use the data from Pérez-Liñán and Polga-Hecimovich, “Explaining Military Coups.” The authors also discuss more in depth the limitations of this way to measure social protests.

57. Martínez, “Presidential Survival in South America.”

58. Tomz et al., “CLARIFY: Software for Interpreting.”

59. George and Bennett, Case Studies.

60. This idea applies to different explanations at the two extremes of the fragmentation spectrum as well as to diverse pathways to the same result at one of the spectrum ends.

61. The destitution of Lugo in 2012 in Paraguay exemplifies more neatly the second path and the previous governability crisis in this country illustrates the third path. Indeed, Paraguay has an important history of these crises and interrupted presidencies in the last two decades. See Nolte, “Paraguay: The President” for the Paraguayan case specifically, or Marsteintredet et al., “Paraguay and the Politics” and Valenzuela, “Latin American Presidencies Interrupted” for a comparative perspective.

62. See e.g. Pachano, “Ecuador: El nuevo sistema” on the Ecuadorian case.

63. Mainwaring et al., The Crisis of Democratic Representation.

64. Coppedge, Strong Parties.

65. McCoy, “Chavez and the End.”

66. Wills-Otero, Latin American Traditional Parties, 88–92.

67. McCoy, “Chavez and the End,” 67.

68. Corrales and Penfold, “Venezuela: Crowding Out.”

69. McClintock, “Reevaluating Runoffs.”

70. Ibid., 106.

71. Abente-Brun, “‘People Power’”; Valenzuela, “Latin American Presidencies Interrupted.”

72. Otero Felipe, “El sistema de partidos”; Rodríguez, “Honduras: Towards a Reconfiguration.”

73. Taylor-Robinson, “La política hondureña”; Llanos and Marsteintredet, “Ruptura y continuidad.”

74. Llanos and Marsteintredet, “Ruptura y continuidad.”

75. Cunha Filho et al., “A Right-to-Left Policy Switch?”

76. Llanos and Marsteintredet, “Ruptura y continuidad.” The re-election was prohibited under any circumstances in Honduras and formed part of the constitution that did not allow for reform.

77. Cunha Filho et al., “A Right-to-Left Policy Switch?”

78. Kim and Bahry, “Interrupted Presidencies”; Kim, “Impeachment and Presidential Politics.”

79. Concurrent or temporally close legislative elections and re-election rules allowing an incumbent president to run reduce the number of presidential candidates. See: Jones, “Electoral Laws”; Jones, “Electoral Institutions.”

80. Linz, “Presidential or Parliamentary Democracy,” 22; Valenzuela, “América Latina,” 20, Shugart and Carey, Presidents and Assemblies, 210.

81. Mainwaring and Shugart, “Conclusion”; Shugart and Carey, Presidents and Assemblies; Jones, “Electoral Laws”; Carey, “Presidentialism and Representative Institutions,” 13; Payne et al., La política importa.

82. Williams-Wyche, “An Empirical Test,” 173.

83. Peru is often cited as an example of such effects. The runoff contributed there to the inability of the opposition to unite against Alberto Fujimori in 2000 because it gave diverse opposition parties the hope to place second in the first round and reverse the result in the second. The Peruvian 1990 election, which saw Fujimori elected after outcome inversion between the first and second rounds, is often cited as an example of the runoff facilitating the election of political outsiders responsible for future governability crises. See Levitsky and Cameron, “Democracy Without Parties?”

84. Colomer, “Non-Median and Condorcet-Loser Presidents.”

85. Martínez, La elección presidencial; Crespo “Ballotage”; McClintock, “Reevaluating Runoffs.”

86. McClintock, “Reevaluating Runoffs,” 104–6.

Additional information

Funding

Tomáš Došek is grateful to CONICYT for funding his doctoral studies (CONICYT-PCHA/Doctorado Nacional/2016-21160096) and also acknowledges the support of the Millennium Institute for Foundational Research on Data.

Notes on contributors

Karel Kouba

Notes on contributors

Karel Kouba is an assistant professor at the Department of Political Science, Philosophical Faculty, University of Hradec Králové, Czech Republic. He specializes in voting behaviour and electoral institutions in Latin American and post-communist countries.

Tomáš Došek

Tomáš Došek is a Ph.D. candidate in Political Science at the Instituto de Ciencia Política, Pontificia Universidad Católica de Chile, Chile. His work focuses on political parties, electoral reforms and subnational politics. He is co-editor of Women, Politics, and Democracy in Latin America (Palgrave Macmillan, 2017) and has published peer-reviewed articles in Taiwan Journal of Democracy, Política y gobierno, Dados – Revista de Ciências Sociais, and others.