Electoral rules of the authoritarian game: undemocratic effects of proportional representation in Rwanda

Abstract

Much has been written about the special design of Rwanda's judiciary in order to handle the aftermath of the genocide in 1994. By contrast, other institutional elements of the 2003 constitution have rarely been addressed in research. The second (partial) parliamentary elections in September 2008 revealed some of the implications which the carefully designed electoral system has for Rwanda's political development. As a starting point, the paper emphasises the need to link the debate about institutional design in divided societies with the debate about elections in authoritarian regimes. Under different regime types, “institutional engineers” may pursue different goals. The paper argues that proportional representation (PR) can foster undemocratic outcomes: PR impedes the local accountability of politicians in a political environment in which the government is not held in check by a democratic opposition. Thus, the current PR system facilitates the maintenance of authoritarian power in Rwanda whereas small constituencies would establish closer links between the local populations and their representatives.

Keywords:

• elections
• Rwanda
• proportional representation
• institutional design
• authoritarianism

Acknowledgements

I would like to thank J⊘rgen Elklit, Svend-Erik Skaaning, my colleagues at the GIGA Matthias Basedau, Martin Beck, Gero Erdmann and Sebastian Elischer as well as two anonymous reviewers for their comments on earlier versions of this paper. Much of the data used for this article (local publications, documents, interviews, observations) would not have been available without field research in September 2008, funded by a private research donation to the GIGA.

Notes

1. Lindberg, Democracy and Elections; Mehler, “Es gibt keine verfrühte Demokratie.”

2. Bratton and van de Walle, Democratic Experiments, 272.

3. Grofman and Reynolds, “Electoral Systems and the Art of Constitutional Engineering.”

4. Certainly, an innumerable number of studies have examined institutional settings as independent variables, often with a focus on systems of government and often with democratic breakdown versus consolidation as the dependent variable (presidentialism vs. parliamentarism, inter alia see Linz and Valenzuela, The Failure of Presidential Democracy; Shugart and Carey, Presidents and Assemblies; Lijphart, “Constitutional Design”). This has remained a lively debate, although some have declared massive doubts about the real importance of the effects (e.g. Thibaut, “Regierungssysteme”). However, the meaning of electoral system choices in authoritarian regimes that are not substantially democratising has remained understudied.

5. Karl, “Imposing Consent”; Levitsky and Way, “The Rise of Competitive Authoritarianism”; Schedler, Electoral Authoritarianism.

6. Cf. Brownlee, Authoritarianism; Gandhi and Przeworski, “Authoritarian Institutions.”

7. Stroh, “Crafting Political Institutions.”

8. E.g., Lijphart, Thinking About Democracy; Reynolds, “The Case For Proportionality.”

9. Grofman and Reynolds, “Electoral Systems and the Art of Constitutional Engineering,” 127–8.

10. Santiso and Loada, “Explaining the Unexpected.” Burkina Faso has been labelled “partly free” by Freedom House but shows particular deficits in the electoral regime, as detailed assessment data from Freedom House and the Bertelsmann Transformation Index reveal.

11. Stolz, “Rwanda.” The founding name of the MDR was PARMEHUTU, the Party of the Hutu Emancipation Movement, which clearly underlines its main political goal. This is why the party is also referred to as MDR-PARMEHUTU. For historical details on the ethnicisation of PARMEHUTU see Saur, “La frontière ethnique.”

12. Population estimates of the United Nations show a total population of about 2.7 million in 1960 and 9.2 million in 2005.

13. Horowitz, Ethnic Groups in Conflict.

14. Basedau and Stroh, “Ethnicity and Party Systems”; Erdmann, “Social Cleavages, Ethnicity and Voter Alignment”; Mozaffar, Scarritt, and Galaich, “Electoral Institutions.”

15. Burnet, “Gender Balance”; Rafti, “Perilous Path to Democracy”; Reyntjens, “Post-1994 Politics in Rwanda”; Reyntjens, “Rwanda, Ten Years On”; ICG, Fin de transition.

16. Buckley-Zistel, “Dividing and Uniting.” On the one hand, official documents such as the national identity cards must not indicate ethnic affiliations to one of the three Rwandan groups – Hutu, Tutsi or Twa – as formalised by the colonial powers and continued until the genocide. On the other hand, a recent constitutional amendment defines the crimes of 1994 as “genocide of the Tutsi.” Vandeginste and Huyse, Consociational Democracy for Rwanda?, 107–16, give evidence for the lack of inclusiveness of Rwanda's current system.

17. Niesen, “Demokratieerhalt durch Parteiverbot?”

18. Kiwuwa, “Democratization and Ethnic Politics,” 462.

19. Meierhenrich, “Presidential and Parliamentary Elections in Rwanda, 2003,” 633.

20. The second parliamentary elections took place after the constitutional five-year term ended. The next presidential election is scheduled for 2010, since the head of state is elected for a term of seven years that is renewable once. The current head of state is Paul Kagame, who is, simultaneously, the RPF party president and has been the military commander of the Rwandan Patriotic Army since 1990.

21. Bills concerning some “softer” policies are excluded from the compulsory approval by the senate. In return, the senate has some important exclusive rights such as the approval of high-level government appointments or investigations into the necessity of a party ban.

22. Examples are given by Reyntjens, “Chronique politique,” 3–4. The increasing influence of deputies was also underlined by several of the author's interview partners in Kigali, September 2008, including high-level Western diplomats.

23. All other votes and selection procedures for public office in Rwanda are either not secret, highly indirect, or a privilege of exclusive groups. Local elections are, for instance, made by public queuing behind the “preferred” candidate. As another example, eight senators are appointed by the head of state. Provincial electoral colleges choose 12 senators, a consultative constitutional body called the “Forum of Political Parties” appoints four senators and university faculties send two. The Chamber of Deputies fills a third of its ranks through indirect elections in three parliamentary segments: 24 female deputies receive their office by special provincial bodies, which themselves emerge from the public “elections” at the municipal level. Two representatives of the youth are elected by the National Youth Council. Similarly, the national association of handicapped people is a privileged group which sends one deputy to parliament.

24. See Munyarugerero, Réseaux, pouvoirs, oppositions, 231–62.

25. Bertrand, Rwanda.

26. Ibuka is the largest association of genocide survivors in Rwanda.

27. For the last election see Table 1, for the previous one see Kiwuwa, “Democratization and Ethnic Politics,” 464.

28. RPF, Parti Démocrate Centriste (PDC), Parti Démocrate Idéal (PDI), Parti pour le Progrès et la Concorde (PPC), Parti de la Solidarité et du Progrès (PSP), Parti Socialiste Rwandais (PSR), Union Démocratique du Peuple Rwandais (UDPR); this is, all accredited parties besides PSD and PL.

29. Explicitly admitted by the vice president of the PDI, Mukamah Abbas, in an interview with the author, Kigali, September 16, 2008.

30. Interviews with RPF and alliance party officials, Kigali, September 2008. The author personally observed, in an unsystematic and non-representative manner, during the electoral period that campaign rallies and printed posters exclusively presented the colours and name of the RPF. However, representatives of the alliance partners said they would have regularly participated and spoken at RPF rallies. The EU observation report cannot confirm this (EU-EOM, Final Report, 26).

31. Kagame, “Address by His Excellency Paul Kagame”; Interview with the former head of the constitutional commission, Tito Rutaremara, Kigali, December 8, 2004.

32. Kimonyo, “Analyse comparative”; IRDP, La Démocratie au Rwanda; similarly Vandeginste and Huyse, Consociational Democracy for Rwanda?, 115.

33. RoR, Rwanda: Genocide Ideology and Strategies for its Eradication; cf. also Kimonyo, “Analyse comparative.”

34. Cf., among others, Vandeginste and Huyse, Consociational Democracy for Rwanda?

35. This is, at least, the common assessment of all major democracy indices such as Freedom House, Bertelsmann Transformation Index, Polity IV or Economist Intelligence Unit's Index of Democracy. However, not being that explicit, EU electoral observation reports have also stated that Rwanda missed minimal standards for democratic elections in recent years (EU-EOM, Rapport Final; EU-EOM, Final Report).

36. Lijphart, Patterns of Democracy; Lijphart, Thinking About Democracy.

37. Only one factor is obviously favourable: the absolute number of politically relevant groups in society, which must not be too large. Paradoxically, it was this particular social constellation which was mobilised in order to plan and execute the most terrible events in Rwanda's history.[0] See Appendix 1.

38. See Jeune Afrique, No. 2493/2494, October 19–November 1, 2008, 9.

39. Lemarchand, “Consociationalism and Power Sharing.”

40. Burnet, “Gender Balance.”

41. Cf. Rafti, “The Rawandan Political Opposition in Exile.”

42. Reyntjens, “Rwanda, Ten Years On”; Lemarchand, “Consociationalism and Power Sharing”; Rafti, “Perilous Path to Democracy.”

43. E.g., Lijphart, Thinking About Democracy; Reynolds, “The Case For Proportionality”; Doorenspleet, “Electoral Systems and Good Governance”; Bos and Schmidt, “Politisierte Ethnizität und Verfassungsstaat.” Different positions favouring majoritarian formulas are prominently represented in the debate (Barkan, “Elections in Agrarian Societies”; Reilly, Democracy in Divided Societies; Horowitz, “Constitutional Design”). However, the point here is that the Rwandan government itself proclaims PR as the most suitable pro-democratic institution for its own divided society (Stroh, “Crafting Political Institutions”).

44. Lust-Okar, “Elections under Authoritarianism,” 458.

45. See Uvin, “Difficult Choices.”

46. Nohlen, Wahlrecht und Parteiensystem.

47. Lust-Okar, “Elections under Authoritarianism.”

48. Axtmann, Reform autoritärer Herrschaft in Nordafrika.

49. Fish, “Creative Constitutions.”

50. Bielasiak, “Regime Diversity and Electoral Systems.”

51. Barkan, Densham, and Rushton, “Space Matters,” 926f.

52. The term “general formula” refers to the basic distinction between proportional and majoritarian systems.

53. Hartmann, “Paths of Electoral Reform”; Nohlen, Krennerich, and Thibaut, “Elections and Electoral Systems in Africa.”

54. Erdmann and von Soest, Diktatur in Afrika.

55. The dichotomy of PR versus majoritarian is virtually useless without supplementary information on the district magnitude, that is, the number of seats to be allocated in one constituency. Only a combination of both allows for a helpful placement of electoral systems on a continuum between proportionality and majority building (see Grofman and Reynolds, “Electoral Systems and the Art of Constitutional Engineering” or Nohlen, Wahlrecht und Parteiensystem).

56. Detailed information on the constituency size in Equatorial Guinea was not available. Only one press agency report could be found indicating the number of constituencies (AFP, “Guinée équatoriale”).

57. Cf. Hartmann, “Paths of Electoral Reform”; Basedau, “Zum Zusammenhang von Wahlsystem, Parteiensystem und Demokratiestabilität in Afrika.”

58. Cf. Axtmann, Reform autoritärer Herrschaft in Nordafrika; Stroh, “Crafting Political Institutions.”

59. In autocracies, the official results do not reflect the free and fair will of the population. Thus, it is more interesting to highlight when, where, and how these results are manipulated.

60. Stroh, “Crafting Political Institutions.”

61. As mentioned above, the minor alliance partners of the RPF had virtually no influence on where the RPF executive committee placed their candidates on the common list.

62. Interview with a party official at PSD headquarters, Kigali, September 5, 2008.

63. Umuseso, No. 312, Kigali, September 24–30, 2008, 13.

64. Rwanda sharply criticised the 2003 European Union report which uncovered heavy deficiencies and, therefore, could not declare the last election “free and fair.” The distrustful allegations and public warnings by the NEC, which, inter alia, urged the 2008 mission not to misuse its mandate, apparently shaped the relationship between Rwandan officials and the electoral observers (cf. Ngabonziza, “NEC Warns”). Some inconsistencies in the final report point to political decisions which could have led to the suppression of insights that support Umuseso's allegations (see Reyntjens, “Rwanda: A Fake Report”). Officially, the EU states only that the decision was taken to discount “any discrepancies between the provisional figures, the EU EOM figures and the final official figures” due to the mission's “low statistical sample” (EU-EOM, Final Report, 42). Surprisingly, the EU employs the same sample for all other analyses in the same report without reservation (EU-EOM, particularly 37–9).

65. EU-EOM, Statement of Preliminary Findings. EU-EOM, Final Report.

66. Schedler, “Menu of Manipulation.”

67. Latest data available to the author.

68. EU-EOM, Final Report, 10.

69. Incidentally, even Lijphart has claimed that the constituencies should not be too large (Lijphart, “Constitutional Design”).

70. Nohlen, Wahlrecht und Parteiensystem; Taagepera, “Designing Electoral Rules.”

71. Barkan, “Elections in Agrarian Societies”; Barkan, Densham, and Rushton, “Space Matters”.

72. Due to the lack of detailed election data, an objective recalculation is impossible.

73. Stroh, “The Power of Proximity.”

74. See Bako-Arifari, “Démocratie et logiques du terroir au Bénin”; Beck, Brokering Democracy; Hilgers and Jacob, “Anthropologie de la candidature.”

75. Barkan and Okumu, “‘Semi-Competitive’ Elections.”

76. Reynolds, “The Case For Proportionality”; cf. also Posner, Institutions and Ethnic Politics in Africa, 217–49.

77. The “yellow dog syndrome” describes a situation in which a locally very popular and deeply rooted party can present whomever it wants and the candidate will be elected, even if he is a “yellow dog.” The stronger institutionalisation of dominant parties in Africa supports this caveat (see Basedau and Stroh, “Measuring Party Institutionalization”; Brownlee, Authoritarianism).

78. Bratton and van de Walle, Democratic Experiments.