Bunce, Valerie and Sharon Wolchik. "Defeating Dictators: Electoral Change and Stability in Competitive Authoritarian Regimes". World Politics, Vol.62, No.1 (2010): 43-86.

Bunce, Valerie and Sharon Wolchik. "Defeating Dictators: Electoral Change and Stability in Competitive Authoritarian Regimes". World Politics, Vol.62, No.1 (2010): 43-86.

• Competitive authoritarian regimes allow nominal competition for political office, but create an uneven playing field so that the regime -- and often the autocrat that regime is built around -- continues to remain in power (43). 
• Leaders in authoritarian regimes are often nominal bound to certain laws and legal limitation on their power, but these can be subverted. The restriction on term limits is often undermined in one of three ways: amending the constitution to allow additional terms, handpicking a trusted replacement, or allowing a replacement but placing severe constitutional limits that the replacement's power (44).
• However, the continuation of the ruling regime is not always the result of elections in a competitive authoritarian regime. This article examines the factors contributing to electoral outcomes in competitive authoritarian systems. The authors examine eleven elections held in the former Warsaw Pact and analyze the factors associated with electoral continuity and electoral change in each case (44-45).
• These cases can be collectively analyzed because they share many features in common, with all measured elections being in competitive authoritarian regimes in former communist countries with high levels of corruption (46).
• The authors have generated two hypotheses, the first being that the strength of the regime will directly determine its performance in elections. Regimes are weaker if they are facing internal struggles or poor economic conditions, and if more areas of the state and civil society are democratically contested (48-49). 
• Since domestic regime security is not the only relevant factor, the authors also propose a second hypothesis that looks at the centrality of elections and analyses factors for electoral weakness in regimes, namely transitions from one ruler to his successor, and unity from the opposition. The authors also hypothesize that a history of fraudulent elections and strong international support for democracy both contribute to electoral change (50).
• A description of three traits of developing democracy, along with a sizable source mine on the topic: "We build on three arguments in the literature that link elections to democratic development: (1) that elections have often served as key contributors to democratic progress, (2) that elections, rather than, say, civil liberties, are understood by citizens in countries outside the West as the defining feature of democracy, and (3) that elections feature conditions conducive to political protests" (49).
• Initial results would not indicate a direct correlation between the authoritarianism of the electoral system and electoral change, as, although all of the regimes withe electoral continuity are squarely authoritarian, so are many of the regimes which experienced electoral change without a clear relationship between the two (51-52).
• Additionally, although all systems became more democratic following electoral change, all countries experienced increased authoritarianism prior to the elections of interest. A relative increase or decrease of authoritarianism was not a factor in the continuity or following elections (52-53).
• There is also significant overlap between the groups which experienced electoral continuity and those experiencing electoral change in terms of mass incidents and protests, and history of successful electoral transitions. Both groups had countries with many and few mass incidents, and countries with and without histories of successful transition. This means that these are likely not as large factors as traditionally expected (54).
• A precedent for previously fraudulent elections also had mixed results on the chance for electoral change. In some cases in acted as a successful nexus for mobilization to demand electoral change, whereas in some cases the previous history had disillusioned the population and reduced the chance for electoral change (54).
• While all of the regimes which experienced electoral continuity shared a number of traits, the sample of states which experienced electoral change is extremely varied. Whereas the states with electoral continuity where generally authoritarian with strong economic records, states with both strong and weak economies have experienced electoral change (56, 72).
• Former communist states in particular face difficulties in establishing clear democratic institutions, due to a legacy of strong state control over civil society. Since most regimes inherited their political and societal infrastructure from the Communist era, these competitive authoritarian regimes continue to dominate courts, media, and civil society. From this same legacy, opposition parties are also weak, underdeveloped, and centered around personality; this leaves these organizations particularly ill equipped to challenge the state (60).
• The nature of competitive authoritarian regimes, with their inconsistent mix of democratic and non-democratic systems of control and governance leaves opposition parties with difficult choices in how to react to the regime. Accordingly, participation and non-participation, or cooperation, are undertaken by different elements of the opposition each electoral cycle (60-61).
• The division of opposition movements does not seem to be a structural factor in the elections observed. In all observed elections, the opposition managed to form a united bloc against the regime, including those were the regime maintained electoral continuity (63). Rather than the belief that a united opposition signals strength, these results would suggest that weak opposition groups are forced to band together to pool limited resources, just resulting in further difficulties should they become a governing coalition (74).
• It is important that electoral change is not equivalent to democratic change and that most autocratic regimes are followed by other autocratic regimes, a transition most likely to occur in regions with weak opposition movements like the former Soviet Union (62).
• The effect of term limitations and a perceive inability to continue to maintain existing systems of social control and patronage have mixed effects on outcomes of electoral change. Some long-ruling leaders lost elections, and in some countries the new elements of the regime manage to maintain electoral continuity (63).
• Support by Western countries, focused only on measuring American aid in this study, also appears to have had a mixed result on electoral continuity in competitive authoritarian regimes. In all categories of support, American efforts have resulted in electoral change and electoral continuity (64-66).
• The methods with which the opposition challenged the regime was extremely important in determining the likelihood of electoral change. All of the successful instances of electoral change mobilized large segments of the population through widespread political activities beyond simple electoral disputation -- these included protests, rallies, active campaigning, and the mobilization of youth movements (67-70).
• The increased mobilization of popular support by opposition groups is linked to an increased role played by civil society during the electoral process. This was driven by both domestic actions cooperation between civil society and opposition groups, and directly linked to American support for the development of civil society (71).
• "One factor provided a clear line of demarcation between our two sets of elections, that is, whether the opposition in collaboration with civil society groups and regional and Western-based democracy activists used an ensemble of sophisticated, intricately planned, and historically unprecedented electoral strategies to maximize their votes and, if necessary, to support public protests demanding a change in leadership, or whether they worked largely alone and settled for a more passive approach to contesting the election. [...] The key was not simply unifying the opposition but also running ambitious political campaigns, orchestrating elaborate voter registration and voter turnout drives, and putting in place electoral monitoring procedures that in combination made oppositions more effective and more politically attractive to voters, created a widespread sense that victory was possible, and made it much harder for the regime [...] to win the election and remain in office after it had lost. The successful defeat of dictators [...] rested on the deployment of a set of innovative, well-planned, detailed, and sometimes dangerous strategies for winning political power" (73).
• Competitive authoritarian regimes are, by their nature, both extremely vulnerable to electoral change, and well insulated against it. The functioning of these regimes divides opposition groups and civil society on issues of cooperating with the semi-legitimate regime institutions, undermines cooperation between these sectors of society, and depoliticizes potential opposition voters by demonstrating the inadequacy of the opposition (74).
• This paper provides key evidence in a central argument within political science on the role of structural factors versus agency in political change. The lack of clear structural indicators for electoral change, compared to the strong correlation between forms of electoral agency and electoral outcomes, would indicate that the agency of opposition actors plays a very large role in regime change (76).