The perils of personalism

ABSTRACT

What impact does personalism, or presidential dominance of a weakly organized ruling party, have on the level of democracy? We argue that presidents who dominate their own weakly organized parties are more likely to seek to concentrate power, undermine horizontal accountability, and trample the rule of law than presidents who preside over parties that have an independent leadership and an institutionalized bureaucracy. Independent party leaders, we suggest, will often try to curb the excesses of the president in order to protect their own political prospects. We explore these hypotheses through a quantitative analysis of the determinants of the level of democracy in 18 Latin American countries from 1980–2015. We find that personalism has a consistently negative impact on the level of democracy in Latin America.

KEYWORDS:

• Personalism
• presidentialism
• democratic erosion
• Latin America
• political parties

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Matthew Rhodes-Purdy http://orcid.org/0000-0002-5791-3332

Notes

1 Gunther and Diamond, “Species of Political Parties,” 187.

2 Kostadinova and Levitt, “Toward a Theory of Personalist Parties,” 491.

3 Weyland, “Clarifying a Contested Concept.”

4 Hawkins, Venezuela’s Chavismo in Comparative Perspective; Mudde and Rovira-Kaltwasser, “Populism and (Liberal) Democracy.”

5 Barro, “Determinants of Democracy”; Boix, Democracy and Redistribution; Burkhart and Lewis-Beck, “Comparative Democracy.”

6 Diamond and Morlino, “ The Quality of Democracy.”

7 O’Donnell, “Delegative Democracy.”

8 Locke, “Two Treatise on Government”; Madison, The Federalist Papers; Montesquieu, The Spirit of the Laws.

9 Latin American presidents have modestly greater formal powers than the U.S. president (see Ginsberg, Cheibub, and Elkins, “Latin American Presidentialism”); but what truly distinguishes them are their informal powers.

10 O’Donnell, “Delegative Democracy”; O’Donnell, “Horizontal Accountability in New Democracies.”

11 Linz, “The Perils of Presidentialism”; Stepan and Skach, “Constitutional Frameworks and Democratic Consolidation?”

12 Shugart and Mainwaring, “Presidentialism and Democracy in Latin America,” 37–40; Shugart and Carey, Presidents and Assemblies, Ch. 3.

13 Mainwaring, “Presidentialism, Multipartyism, and Democracy”; c.f. Cheibub, Presidentialism, Parliamentarism, and Democracy.

14 Levitt, Power in the Balance, 37.

15 Samuels, “Presidentialized Parties,” 480.

16 Levitsky, Transforming Labor-Based Parties, 35.

17 Navia, “Partidos politicos como antídoto contra el populismo.”

18 Corrales, “Presidents, Ruling Parties, and Elites,” 135; Naim, Paper Tigers and Minotaurs, 52–4.

19 Pérez-Liñán, Presidential Impeachment, 156-158; Corrales, “Presidents, Ruling Parties, and Elites,” 135–6.

20 López Maya, “Venezuela,” 217.

21 Chávez dissolved the MVR in 2007, choosing to replace it with another personalist party, the Partido Socialista Unido de Venezuela.

22 Martinez, “El Papel Del Movimiento Quinta República,” 24–5.

23 Corrales, “The Repeating Revolution,” 30–3.

24 Corrales and Penefold, Dragon in the Tropics, 41–6.

25 Rhodes-Purdy, Regime Support Beyond the Balance Sheet, Ch. 5; Rhodes-Purdy, “Participatory Populism”; Hawkins, Venezuela’s Chavismo and Populism.

26 Corrales, Presidents Without Parties; Madrid, “Labouring against Neoliberalism”; Levitsky, Transforming Labor-Based Parties.

27 Ibid., 221.

28 Carey, “The Reelection Debate,” 128; Weyland, The Politics of Market Reform, 194.

29 Ibid., 197–8.

30 Corrales, Fixing Democracy, 213–4.

31 These data are presented merely to demonstrate the face validity of our theory. Below, we seek to establish causal inference with more sophisticated methods.

32 Mainwaring, Brinks, and Pérez-Liñán, “Classifying Political Regimes”; We opt to use the MBP dataset because it is categorical by design, while Polity IV is not. We could, of course, have used the zero cutoff to dichotomize the Polity data; doing so would only have excluded four observations: Nicaragua from 1985 to 1989, which MPP codes as semi-democratic, but which Polity scores as −1.

33 Munck and Verkuilen, “Conceptualizing and Measuring Democracy”.

34 Kostadinova and Levitt, “Toward a Theory of Personalist Parties”.

35 A table of these dyads with the number of codings is in the Appendix. Full data is publicly available (Rhodes-Purdy and Madrid, Presidential Personalism).

36 Coppedge, “Party Systems, Governability, and the Quality of Democracy”.

37 Madrid, Latin American Ethnicity Database.

38 These techniques are sometimes called Latent Growth Models. See Bollen and Curan, Latent Curve Models, for a detailed explication of these techniques.

39 As a robustness check, we also ran the analyses using the value of these variables in the first year of each term. This did not affect the results.

40 See the Appendix for the equations that specify these models.

41 The time points as labelled on the graph are zero and 1.609, due to the logarithmic scale. These graphs do not reflect the logarithmic shape of the curve, only the total magnitude of change from the first to last time period.

42 The zone of significance begins at about 1.3 standard deviations below the mean of party organizational weakness.

43 As a robustness check, we also analysed fixed-effects models based on the same data, with lagged personalism variables. We also conducted LCA on VDEM’s polyarchy variable, and conducted an analysis of the Polity data using a single objective measure of personalism (whether or not the president founded his or her party). See the Appendix for details.

44 Coppedge et al., “Varieties of Democracy”.

45 There are risks of bias associated with this approach, but we have good reason to believe that such bias is limited here; the reasons are discussed in the Appendix. Furthermore, we are more concerned here with elucidating the specific elements of democratic quality that are negatively associated with personalism, rather than causal inference, which we sought to establish via analyses of the Polity IV composite score. See the Appendix for more detailed discussion and multiple robustness checks, all of which support our approach.

46 Kostadinova and Levitt, “Toward a Theory of Populist Parties”.

47 Levitt, Power in the Balance.

48 Linz, “The Perils of Presidentialism”; Stepan and Skach, “Constitutional Frameworks”; Mainwaring, “Presidentialism, Multipartism, and Democracy”.

Additional information

Notes on contributors

Matthew Rhodes-Purdy

Matthew Rhodes-Purdy is an Assistant Professor at Clemson University. He received his PhD in government from the University of Texas at Austin. His research interests include legitimacy and regime support, populism, and participatory governance. He is the author of Regime Support Beyond the Balance Sheet (Cambridge, 2017), and his articles have appeared in Comparative Politics, Political Research Quarterly, Latin American Research Review, and Political Studies. His current research investigates how economics and culture have interacted to create the recent surge in global populism.

Raúl L. Madrid

Raúl L. Madrid is the Harold C. and Alice T. Nowlin Regents Professor in the Dept. of Government at the University of Texas at Austin. He is the author of The Rise of Ethnic Politics in Latin America (Cambridge, 2012) and Retiring the State: The Politics of Pension Privatization in Latin America and Beyond (Stanford, 2003) and is a co-editor of Leftist Governments in Latin America: Successes and Shortcomings (Cambridge, 2010) and When Democracy Trumps Populism: European and Latin American Lessons for the United States (Cambridge 2019). He is currently working on a book on the origins of democracy in Latin America.