Why do some authoritarian states have competitive elections? This study shows that whenever there is a balance of power between candidates, competitiveness will ensue. Electoral fraud is often widespread in autocratic states, but if no single candidate or party is in a position to monopolize electoral support the result will be competitive. The contribution here is to analyze the relative strength of all the actors involved in a parliamentary election and to show that electoral returns reflect the district level balance of power, even in autocracies. Three main sources of candidate-level electoral power are identified: state, market, and society. State affiliated candidates in authoritarian states perform well due to favorable treatment by state institutions. Market actors perform well due to financial resources. These actors arise when market reforms create a class of entrepreneurs that defend their interests by running for public office, often challenging state sanctioned candidates. The strength of candidates using social cleavages, here mainly ‘clan’ and ethnic, is found to be exaggerated in the literature. The study also confirms that competitiveness did not result from an active civil society. Competitive  elections matter because they can severely destabilize the regime, as was the case in Kyrgyzstan in 2005. However, electoral competitiveness that is the result of an intra-elite balance of power should not be confused with democracy. This form of self-interested competitiveness where clientelism is pervasive and accountability mechanisms are weak is an affront to the democratic ideal. For those of us who advocate democracy and genuine political participation competitive authoritarian regimes can be used as an informative cautionary tale. Power matters, and especially so in authoritarian states. Understanding the logic behind competitive authoritarianism helps us revise strategies for lasting democratic reforms.

Ce mémoire a pour but d’étudier si le fait d’avoir des élections compétitives au Brésil a mené à des politiques de redistribution de revenus plus efficaces. Le chapitre 1 présente les critères retenus pour mesurer la compétitivité électorale au poste de Gouverneur. À partir des 27 États du pays, on distingue un groupe de 5 États où la compétition électorale est forte et un autre groupe de 5 États pour lequel elle est faible. Pour chacun de ces groupes, un État a été retenu. Le chapitre 2 détaille les facteurs qui expliquent les inégalités au Brésil. La suite du mémoire se consacre à l’analyse des politiques favorisant une meilleure redistribution de revenus pour l’État retenu de chaque groupe (chapitre 3) et pour l’ensemble des deux groupes (chapitre 4). L’analyse révèle qu’un plus important niveau de compétition électorale ne mène pas nécessairement à de meilleures politiques de redistribution de revenus. / This study aims to look at whether electoral competitiveness in Brazil led to more efficient income redistribution policies. Chapter 1 takes a look at which criteria were used to measure electoral competitiveness for Governor. From the country’s 27 States, we have identified a group of 5 States considered to have a high level of electoral competitiveness and another group of 5 States with low levels of competitiveness. We have then outlined a single State for each of these two groups. Chapter 2 details the factors explaining income inequality in Brazil. The remainder of the study focuses on the analysis of policies favouring better income redistribution for the single State selected for each group (chapter 3) and for both groups as a whole (chapter 4). The analysis reveals that a higher level of electoral competitiveness does not necessarily lead to better income redistribution policies.

Usually, the explanations about the little electoral competitiveness of peruvian political parties have emphasized the unlikelihood that this situation could be reversed. However, when each case is analyzed carefully, we can see that groups like APRA and the Fujimorista Movement have developed some features beyond their leaders which, in certain situations, have provide them a little more electoral competitiveness than others parties, such as the Popular Christian Party (PPC). From this point, the following article argues that this difference in the competitiveness of the peruvian parties is basically for the type of strategies that they have been implemented. While some of them have seen in the permanent positioning a way to be more competitive in the medium term, others have believed that the competitiveness is achieved more quickly when political parties reduce their intervention in political debates. In that sense, this article tested an actor-center explanation to understand the little competitiveness of peruvian parties and also will risk a less pessimistic view of their future. / Las explicaciones sobre la poca competitividad electoral de los partidos políticos peruanos han enfatizando de manera predominante la improbabilidad de que ello se revierta. Sin embargo, cuando cada partido es visto con detalle se puede notar que algunos como el APRA y el Movimiento Fujimorista han desarrollado características más allá de sus líderes, los cuales en determinadas coyunturas, les han brindado una mayor competitividad electoral frente a los demás, como el Partido Popular Cristiano (PPC), por ejemplo. A partir de esta observación, el siguiente artículo argumentará que esta ligera –aunque no menos importante– diferencia en las capacidades competitivas se debe principalmente a las estrategias que han aplicado. Mientras unos han visto en el posicionamiento claro y permanente una forma de ganar competitividad en el mediano plazo, otros han creído que ello reside en la poca presencia en el debate político. Es así que, junto a una explicación voluntarista para entender la poca competitividad electoral de los partidos peruanos, se arriesgará una mirada menos pesimista sobre su futuro.

Submitted by Oliveira Santos Dilzaná (dilznana@yahoo.com.br) on 2016-04-11T14:14:48Z No. of bitstreams: 1 Dissertação de Vladimir Meira Nunes.pdf: 2467427 bytes, checksum: f43b436137fb8e26bfd9f1e82c1d40e7 (MD5) / Approved for entry into archive by Ana Portela (anapoli@ufba.br) on 2016-04-28T18:16:51Z (GMT) No. of bitstreams: 1 Dissertação de Vladimir Meira Nunes.pdf: 2467427 bytes, checksum: f43b436137fb8e26bfd9f1e82c1d40e7 (MD5) / Made available in DSpace on 2016-04-28T18:16:51Z (GMT). No. of bitstreams: 1 Dissertação de Vladimir Meira Nunes.pdf: 2467427 bytes, checksum: f43b436137fb8e26bfd9f1e82c1d40e7 (MD5) / FAPESB / O estudo analisa a Geografia do Voto dos Deputados Estaduais e Partidos na Bahia, nas eleições de 2002, 2006 e 2010, buscando captar e avaliar supostos impactos produzidos sobre a localização das suas bases eleitorais, por força de seus reposicionamentos face às variáveis condições de ser governo, ou oposição, nos planos nacional e estadual. Além disso, procurou correlacionar as votações agregadas dos principais partidos no Estado com indicadores sociais, como IDH e População Rural dos municípios, visando mapear os perfis de suas bases, no que diz respeito às características urbanas e sociais e saber se eles sofreram modificações no período, também tendo em vista os referidos reposicionamentos. Buscou, ao mesmo tempo, problematizar essas questões a partir de referenciais da Teoria Democrática, assim como na literatura institucional acerca dos Sistemas Partidário e Eleitoral. Foram identificadas mudanças significativas na geografia das bases eleitorais dos principais partidos a partir dos reposicionamentos, mesmo com a manutenção dos respectivos padrões de votação antes predominantes. Nesse sentido, a análise das mudanças na geografia eleitoral dos deputados estaduais e partidos baianos ao longo das três eleições estudadas, apontou que o padrão de votação horizontalmente “disperso” é mais vinculado a deputados e partidos de governo, enquanto o padrão vertical “dominante” não é necessariamente governista, sendo afetado, também, pela trajetória anterior de cada partido. Os resultados sugerem também que o Legislativo baiano não possui um padrão oligárquico de competição, já que 52% dos seus parlamentares possuem votação Não-Dominante, ou seja, compartilham seus redutos eleitorais. Apesar disso, observou-se que o reposicionamento estadual provocado pelos resultados das eleições de 2006 atenuou, sem reverter, a tendência à ampliação da competitividade eleitoral nos municípios baianos, provocada pelo realinhamento nacional ocorrido em 2002.The study analyzes the geography of the Vote of Parties and State Representatives in Bahia in the 2002, 2006 and 2010 elections, seeking to capture and evaluate alleged impacts produced on the location of their constituencies , by virtue of its repositioning in the face of changing conditions to be government or opposition, in the national and state plans. In addition, sought to correlate the votes aggregates of the major parties in the state with social indicators such as HDI and Rural Population of the municipalities, aimed at mapping the profiles of their bases, with regard to urban and social characteristics and whether they have been changed in the period also with a view to repositionings. Search at the same time, discuss these issues from references of Democratic Theory, as well as in institutionalist literature on the Partisan and Electoral systems. Significant changes were identified in the geography of constituencies of the major parties from the repositioning, even with the maintenance of their voting patterns before prevailing. In this sense, the analysis of changes in electoral geography of state representatives and Bahian parties over the three studied elections, pointed out that the voting pattern horizontally "dispersed" is more linked to MPs and parties of government, while the vertical standard "dominant" is not necessarily ruling, being affected also from the previous path of each parties. The results also suggest that the Bahian legislative does not have an oligarchic pattern of competition, since 52% of its parliamentarians have vote Non- Dominant, ie shared their bailiwicks. Nevertheless, it was observed that the in state repositioning, caused by the results of the 2006 elections, were attenuated, without reversing the trend to expand the electoral competitiveness in Bahia municipalities, caused by national readjustment in 2002.

Partido Popular Cristiano (PPC) es un partido peruano de derecha fundado en 1966. Sin embargo, nunca ha ganado ninguno de los cargos más importantes del país pese a que, espe- cialmente en la última década, se lo propuso con gran determinación. La literatura sobre la poca competitividad electoral de los partidos políticos peruanos ha señalado que ello se debería a factores históricos, estructurales, institucionales u organizacionales, difícilmente superables. En esa línea, los pocos autores que han arriesgado hipótesis sobre el PPC han resaltado una supuesta ideologización y poca apertura a sectores populares como los factores que lo han hecho poco competitivo. En cambio, en este trabajo se argumenta que en la última década la poca competitividad del PPC ha sido profundizada por la adopción por parte de sus élites partidarias de una estrategia atrapa-todo, la cual implicó que no se posicionara en el debate político programático existente en el Perú de forma clara y permanente y, por tanto, perdiera identidad política, crucial en un contexto con altos niveles de volatilidad, dificultades para la agregación de intereses y en el que no siempre se necesitan de grandes mayorías para ganar elecciones. La falta de una identidad política enraizada en una parte de la ciudadanía le impidió conservar un voto duro y una logística organizacional, factores que en el Perú brindan una mayor capacidad competitiva.