http://orcid.org/0000-0001-5329-7424
![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
Drawing on two complementary mechanisms, this article explores the question of whether electoral institutions and conditions of electoral competition create incentives to promote electoral misconduct in young or developing democracies. The first mechanism explains how majoritarian institutions like disproportional electoral systems are more likely to trigger electoral fraud than consensus electoral institutions like proportional representation. However, for this mechanism to be activated, the incumbent must feel effectively threatened by the opposition. To better understand the way this mechanism works, the electoral history of the country also needs to be taken into consideration. Democracies which have a historical record of running clean elections are less likely to experience fraud than countries with a history of electoral misconduct. I test these theoretical claims using a dataset that contains relevant information for 323 parliamentary elections in 59 new or developing democracies in the period between 1960 and 2006. The empirical analysis shows a strong and robust empirical support for the two mechanisms.
Acknowledgements
The author would like to especially express his gratitude to Sarah Birch for her constant support and encouragement with this project. Earlier drafts of this article also received valuable feedback from Jennifer Gandhi, Ferran Martinez i Coma, Adam Przeworski, José María Maravall, Adrian Blau, David Skarbek and two anonymous reviewers.
Disclosure statement
No potential conflict of interest was reported by the author.
ORCID
Rubén Ruiz-Rufino http://orcid.org/0000-0001-5329-7424
Notes
1 Flores and Nooruddin, Elections in Hard Times.
2 Norris, Why Electoral Integrity Matters; Birch, Electoral Malpractice.
3 Donno and Roussias, “Does Cheating Pay?”
4 On this, see http://www.eueom.eu/how-do-we-support-elections-/key-publications and http://www.osce.org/odihr/elections/68439
5 Kelley, Monitoring Democracy.
6 Donno and Roussias, “Does Cheating Pay?”; Simpser, “Does Electoral Manipulation Discourage Voter Turnout?”; Birch, “Perceptions of Electoral Fairness”; Kelley, Monitoring Democracy.
7 Lehoucq and Molina, Stuffing the Ballot Box.
8 Varela-Ortega, El Poder de la Influencia.
9 Mares, From Open Secrets to Secret Voting.
10 Ziblatt, “Shaping Democratic Practice and the Causes of Electoral Fraud.”
11 Fortin-Rittberger, “The Role of Infrastructural and Coercive State Capacity in Explaining Different Types of Electoral Fraud.”
12 Birch, “Electoral Systems and Electoral Misconduct.”
13 Kolev, “The Contingent Effect of Institutions.”
14 Magaloni, “The Game of Electoral Fraud and the Ousting of Authoritarian Rule.”
15 Lehoucq and Molina, Stuffing the Ballot Box.
16 Simpser, Why Governments and Parties Manipulate Elections.
17 Anderson and Guillory, “Political Institutions and Satisfaction with Democracy.”
18 Rae, The Political Consequences of Electoral Laws.
19 Anderson et al., Losers’ Consent.
20 Mozaffar and Schedler, “The Comparative Study of Electoral Governance”; Birch, Electoral Malpractice.
21 Svolik, “Equilibrium Analysis of Political Institutions.”
22 Lijphart, Democracy in Plural Societies.
23 Ibid., 168.
24 Kunicova and Rose-Ackerman, “Electoral Rules and Constitutional Structures as Constraints on Corruption.”
25 Persson, Tabellini, and Trebbi, “Electoral Rules and Corruption.”
26 Chang and Golden, “Electoral Systems, District Magnitude and Corruption.”
27 Lehoucq and Molina, Stuffing the Ballot Box.
28 Birch, “Electoral Systems and Electoral Misconduct.”
29 Carey and Shugart, “Incentives to Cultivate a Personal Vote.”
30 On this see, http://www.ipu.org/parline-e/reports/arc/COSTA_RICA_1970_E.PDF
31 Rokkan, Citizens, Elections, Parties; Boix, “Setting the Rules of the Game”; Przeworski, “Conquered or Granted?”
32 Lijphart, Patterns of Democracy.
33 This may differ if elections occur in autocracies. As Simpser, Why Governments and Parties Manipulate Elections, has shown, even when incumbents enjoy great electoral support, they sponsor electoral fraud to send the opposition a signal about their strength.
34 Taagepera and Shugart, Seats and Votes.
35 Lehoucq and Kolev, “Varying the Un-Variable.”
36 North, “Institutions and Credible Commitment.”
37 Magaloni, “Credible Power-Sharing and the Longevity of Authoritarian Rule.”
38 North and Weingast, “Constitutions and Commitment.”
39 Elklit and Reynolds, “The Impact of Election Administration on the Legitimacy of Emerging Democracies.”
40 Hyde, The Pseudo-Democrat’s Dilemma.
41 The evidence testing this hypothesis, however, is mixed and it remains unclear what the real impact, if any, electoral management bodies really have on the quality of elections. See Hyde and Pallister, “Election Administration”; Hartlyn, McCoy, and Mustillo, “Electoral Governance Matters Explaining the Quality of Elections in Latin America.”
42 The variable analysed is v2elembaut_ord.
43 Pemstein et al., “The V-Dem Measurement Model.”
44 The opposition party NPP also won the parliamentary elections by an even smaller margin but obtained fewer seats than the rival NDP. See Pryce and Oidtmann, “The 2012 General Elections in Ghana.”
45 See http://aceproject.org/regions-en/countries-and-territories/KE/reports/Kenya%20-%20EU%20rep02.pdf
46 The difference between the two main parties in the parliament was, however, 10%. See Cheeseman, “The Kenyan Elections of 2007.”
48 Lehoucq, “Electoral Fraud,” 235.
49 Birch, “Electoral Systems and Electoral Misconduct,” 1535.
50 Electoral misconduct actions limiting political competition target both candidates and voters from opposition parties. The 1970 election in Costa Rica is an example of the latter while the violent clashes among voters during the pre-electoral period in the 1989 elections in India is an example of the former. However, actions like clientelism are not considered in this article. Clientelism follows a distinct pattern linked to distributive politics that does not necessarily respond to the same type of actions described here. See Stokes et al. Brokers, Voters and Clientelism.
51 The notes provided by the NELDA dataset show that in the time period covered in this article all fraudulent actions took place during the pre-electoral period.
52 Using the notes provided by NELDA, two examples illustrate this idea of electoral misconduct limiting political competition. During the 1983 elections in Turkey, the military banned some opposition candidates from running in the elections. In Bangladesh, the government arrested and harassed members of opposition parties that were protesting against the 1988 general elections.
53 Hyde and Marinov, “Which Elections Can Be Lost?”
54 The NELDA dataset contains a variable that accounts for electoral fraud. NELDA11 is a subjective indicator that shows whether there was “domestic or international concerns” on the quality of the election. Hafner-Burton, Hyde, and Jablonski, “When Do Governments Resort to Election Violence?,” for example, use this variable as an indicator of fraud. In this article, however, this variable is not used in an attempt to capture objective facts rather than subjective perceptions which can accommodate some level of bias. See Kelley, “Election Observers and Their Biases.”
55 More concretely, I have defined electoral misconduct using two questions in the NELDA dataset. Variable NELDA13 asks: “Were opposition leaders prevented from running” and variable NELDA15 asks: “Is there evidence that the government harassed the opposition?” If a “yes” was coded in either of these questions, then electoral misconduct existed, if a “no” was coded, then I considered that such fraud did not exist. Those “N/A” or “unclear” cases are treated as missing values. Some of the cases identified by NELDA as having this type of misconduct were, however, recoded. This is particularly the case of the elections in Albania (2001) and Latvia (2006). In Albania some form of fraud occurred but only in the fifth round of the legislative elections. In Latvia, the OSCE reported some minor forms of electoral misconduct but political competition was not compromised. The version of the dataset that I have used for this article is NELDA v3 (http://hyde.research.yale.edu/nelda/). NELDA v4 refers mainly to developed democracies which are not covered in this article. Also the relevant variables used from NELDA v3 were cross-checked with NELDA v4 to maintain internal consistency.
56 This selection of countries is similar to the samples used in other comparative studies on electoral fraud. See Donno and Roussias, “Does Cheating Pay?”; Birch, “Electoral Systems and Electoral Misconduct.” Also, this particular sample concentrates most on the variation of the dependent variable as most of the existing reports on electoral observation missions come from this group of countries. In any case, the selection of countries and years used in this article shows large regional and temporal variation.
57 Ruiz-Rufino, “Characterizing Electoral Systems.”
58 ATF data can be accessed at https://dataverse.harvard.edu/dataset.xhtml?persistentId=hdl:1902.1/15528. For elections that happen after 2000, ATF was calculated as suggested in Ruiz-Rufino, “Characterizing Electoral Systems.”
59 Bormann and Golder, “Democratic Electoral Systems Around the World, 1946–2011.”
60 Nohlen, Elections in the Americas; Nohlen et al., Elections in Asia and the Pacific; Perez-Liñán and Wills-Otero, “La Evolución de los Regímenes Políticos y el Sistema Electoral en Latinoamérica, 1903–2004.”
61 Lindberg, Democratization by Elections.
62 Cheibub, Gandhi, and Vreeland, “Democracy and Dictatorship Revisited.”
63 Ibid.
64 Brancati, Peace by Design.
65 Alesina et al., “Fractionalization.”
66 Since the unit of analysis is election year in country c, the vector of years fixed effect refers to decades rather than year to maximize the number of observations used in the analysis. This type of fixed effects constitutes a suitable estimation method when the structure of the cross-section time-series data is unbalanced like this one. The data also contain repeated time values within the panel given that some countries like Thailand held two elections in the same year (1992). See Wooldridge, Econometric Analysis of Cross Section and Panel Data.
67 Birch, “Electoral Systems and Electoral Misconduct.”
68 The sample of countries is classified into six regions: Africa, America, Asia, Europe, Middle East and Post-Soviet.
69 The number of observations reduces given the use of lag variables and lack of information of Agriculture. When Agriculture is excluded, models 4 and 5 remain, however, the same.
70 The models are also tested for serial autocorrelation. The correlation between the residuals and the lag residuals of the model is -0.05.
71 Brambor, Clark, and Golder, “Understanding Interaction Models.”
72 Birch, Electoral Malpractice; Schaffer, The Hidden Costs of Clean Election Reforms.
73 Lijphart, Electoral Systems and Party Systems.
74 Calvo, “The Competitive Road to Proportional Representation.”
75 Marshall and Jaggers, “Polity IV Project.”
76 Fortin-Rittberger, “The Role of Infrastructural and Coercive State Capacity in Explaining Different Types of Electoral Fraud.”
77 Przeworski, “Acquiring the Habit of Changing Governments through Elections.”
Additional information
Notes on contributors
Rubén Ruiz-Rufino
Rubén Ruiz-Rufino is a Lecturer in Comparative Politics at King’s College London – Department of Political Economy. His areas of research interest include the impact of electoral rules on the quality of elections, the relationship between political representation and political stability, the origin of electoral institutions and how globalisation affects political representation in industrialised democracies. His work has appeared in the European Journal of Political Research, West European Politics, Electoral Studies and Political Studies.