Accountable for what? Regime types, performance, and the fate of outgoing dictators, 1946–2004
2013; Taylor & Francis; Volume: 20; Issue: 1 Linguagem: Inglês
10.1080/13510347.2013.738866
ISSN1743-890X
Autores Tópico(s)Electoral Systems and Political Participation
ResumoAbstract For some political leaders losing power might entail further punishment than just being replaced. During the period 1946–2004, 47% of dictators have been jailed, killed, or had to go into exile as a consequence of losing power. The possibility that such punishment occurs can shape and condition present-time decisions rulers make. This article investigates the factors which may affect the fate of outgoing dictators. I find that regime types determine, to a high extent, the fate of rulers as regimes shape internal dynamics and incentives and decisively affect the way rulers are replaced. Due to the lack of regularized succession procedures and to the concentration of power and patronage-based loyalty, personalist leaders are more likely to be killed and to go into exile upon their exit than other autocrats. Military and personalist leaders are also more likely to be imprisoned than monarchs or rulers of single-party regimes. If replaced by democracies, rulers of party and military regimes have a much higher likelihood of not being punished than personalists. It is also found that performance matters for explaining what happens to rulers after the loss of power. Economic crises and defeats increase the likelihood that an authoritarian ruler is imprisoned. Keywords: dictatorspunishmentpost-exit fateexileregime types Notes See Zeliger, “ICC Indictment of Qaddafi.” Dix, “The Breakdown of Authoritarian Regimes,” 567–568. O'Donnell and Schmitter, Transitions from Authoritarian Rule, 29. Przeworski, Democracy and the Market. Huntington, The Third Wave, 231. Escribà-Folch, “Maten al León”; Debs, “Living by the Sword.” Sutter, “Settling Old Scores.” Geddes, “Democratization after Twenty Years.” Huntington, The Third Wave, 120. Debs, “Living by the Sword.” Wright and Escribà-Folch, “Authoritarian Institutions.” Ibid. Nalepa, Skeletons in the Closet. See Snyder and Vinjamuri, “Trials and Errors.” Escribà-Folch and Wright, “Human Rights Prosecutions.” Goldsmith, “International Criminal Court”; Nalepa and Powell, “To Stay or To Go.” Kim and Sikkink, “Explaining the Deterrence Effect.” Escribà-Folch, “Potential Punishment under Dictatorship.” Bueno de Mesquita and Siverson, “Survival of Political Leaders”; Chiozza and Goemans, “Peace through Insecurity”; Chiozza and Goemans, “Tenure of Leaders”; Weeks, “Accountable Autocrats?”; Weeks, “Rulers, Risk, and Restraint.” Downs and Rocke, “Gambling for Resurrection.” Chiozza and Goemans, “Peace through Insecurity.” Goemans, “Fighting for Survival.” Debs and Goemans, “Regime Type,” 430. Peceny, Beer, and Sanchez-Terry, “Dictatorial Peace?”; Reiter and Stam, “Identifying the Culprit”; Peceny and Beer, “Peaceful Parties.” Weeks, “Rulers, Risk, and Restraint.” Debs and Goemans, “Regime Type,” 434. Geddes, “Democratization after Twenty Years”; Ezrow and Frantz, Dictators and Dictatorships. Goemans, ‘Which Way Out?’. The number of those removed by foreign forces or who left office due to natural death or ill health was included in the denominator but it is not shown in the bar graph. Therefore, the percentages in the figure do not add up to 100%. Regular exits are those which occur “according to the prevailing rules, provisions, conventions, and norms of the country,” while “a loss of office is considered irregular when the leader was removed in contravention of explicit rules and established conventions. Most irregular removals from office are the result of the threat or use of force as exemplified in coups, (popular) revolts, and assassinations […] and occur at the hands of domestic opponents”; Goemans, Gleditsch, and Chiozza, “Introducing Archigos,” 273. Geddes, “Democratization after Twenty Years.” Besley and Kudamatsu, “Making Autocracy Work.” Quinlivan, “Coup-Proofing.” Iqbal and Zorn, “Sic Semper Tyrannis?.” Goodwin, No Other Way Out. Bratton and Van de Walle, “Political Transitions in Africa,” 460. According to Svolik and Akcinaroglu's data, 13.5% of personalist rulers were ousted by civil wars, while no monarchy and only about 3% of military and party-regime rulers experienced that end. As for revolts, more than 12% of personal rulers are ousted by them, but only 7.7% of monarchs, 4.9% of party-regime leaders, and 9.5% of military rulers. Svolik and Akcinaroglu, “Government Change.” See also Gurses and Mason, “Weak States.” During the popular uprising that started in 2011, Yemen's President Saleh was severely injured after an attack with an explosive projectile on his palace, which took place on 3 June. He left the country for medical treatment and there were doubts about whether he was ever going to return. He eventually did so in September. Iqbal and Zorn, “Sic Semper Tyrannis?.” Ezrow and Frantz, Dictators and Dictatorships, 90. Almost 33% of monarchs manage to retain power until they die of natural causes or are unable to rule due to ill health. Tullock, Autocracy. Cox, “Authoritarian Elections”; Gandhi and Lust-Okar, “Elections under Authoritarianism.” Magaloni, “Credible Power-Sharing.” Brownlee, Age of Democratization; Gandhi, Political Institutions under Dictatorship; Smith, “Life of the Party.” Magaloni, Voting for Autocracy; Greene, Why Dominant Parties Lose. Wright and Escribà-Folch, “Authoritarian Institutions.” Ibid. Geddes, “Democratization after Twenty Years.” Dix, “The Breakdown of Authoritarian Regimes.” Sutter, “Settling Old Scores.” Nobles, “Human Rights Violations.” Debs, “Living by the Sword.” In fact, most military rulers, 67.6%, are succeeded by another dictator. Dix, “The Breakdown of Authoritarian Regimes.” Ezrow and Frantz, Dictators and Dictatorships, 85–86. Debs, “Living by the Sword.” Norpoth “Guns and Butter,” 949; cited in Bueno the Mesquita and Siverson, “Survival of Political Leaders.” See Acemoglu and Robinson, “Democratic Transitions” for a theoretical explanation. For empirical evidence, see Gasiorowski, “Political Regime Change”; Geddes, “Authoritarian Breakdown”; Haggard and Kaufman, “Democratic Transitions”; Wright, “Exit During Crisis.” Haggard and Kaufman, “Democratic Transitions,” 267. Bueno de Mesquita and Smith, “Leader Survival, Revolutions.” Londregan and Poole, “Poverty, the Coup Trap”; Galetovic and Sanhueza, “Citizens, Autocrats, and Plotters.” Finer, The Man on Horseback; Nordlinger, Soldiers in Politics. Goemans, ‘Which Way Out?’. Bueno de Mesquita and Siverson, “Survival of Political Leaders.” Chiozza and Goemans, “Tenure of Leaders”; Debs and Goemans, “Regime Type.” Goemans, “Which Way Out?.” Weeks, “Accountable Autocrats?.” Chiozza and Goemans, “Tenure of Leaders.” Ironically, Galtieri had been cleared of the charges of human rights violations in a previous trial in 1984. Goemans, Gleditsch, and Chiozza, “Introducing Archigos”; Goemans, “Which Way Out?.” Geddes, Wright, and Frantz, “Authoritarian Regimes.” See also Geddes, “Democratization after Twenty Years.” Both variables are taken from Maddison, “World Population.” Compiled from Goemans, “Which Way Out?.” The data on interstate conflict are compiled from PRIO's (Peace Research Institute Oslo) Armed Conflict Dataset. See Gleditsch et al., “Armed Conflict 1946–2001.” I further distinguished between French and British former colonies, but this separation did not yield any significant results either. I also coded two additional variables: the total number of international alliances a country has in a given year, and the total number of countries with which a country has alliances in a given year (so each alliance is weighted by the number of countries it includes). However, neither of these variables is significant and their explanatory power is small when region fixed effects are included. A formal alliance is defined as “a written agreement that identifies at least the members and the obligations of each alliance member,” and “the alliance commitment must be a defense pact, neutrality or non-aggression pact, or an entente.” Gibler and Sarkees, “Measuring Alliances,” 212. The data on intrastate wars are compiled from PRIO's Armed Conflict Dataset. See Gleditsch et al., “Armed Conflict 1946–2001” for a description. Data are from Goemans, see http://www.rochester.edu/college/faculty/hgoemans/data.htm Gandhi and Lust-Okar, “Elections under Authoritarianism”; Cox, “Authoritarian Elections.” When only pure types are considered, hybrid regimes with personalistic traits are not included in the military and party-based categories. This increases the differences between pure personalists and party-regimes. Another noticeable difference between both models, besides the size of the coefficients, is that in model 2 monarchs are significantly more likely to exile than party-regime leaders. Once the full set of controls is included (model not reported but available from the author), the dummy for monarchies ceases to be significant. 71.7% of personalist rulers, 87.7% of party-regime leaders, and 67.6% of military rulers are. Alternatively, models 1, 2, and 4 in Table 1 include those replaced by other dictators as well as those followed by a democratic regime. Monarchs are excluded, as only one was replaced by a democracy in our sample: King Birendra of Nepal in 1991. To check this I have also re-run the basic multinomial logit model, splitting the sample and thus running two separate regressions: one for those leaders replaced by dictators and a second one including only those succeeded by democratic regimes. These results are available from the author. Conversely, the percentage of those imprisoned decreases from 24% to 11.1%. This is consistent with Debs' findings. Debs, “Living by the Sword.” Due to the use of lags and some other data availability limitations, the period covered in this last model is 1948–2003. Again, the other variables are held constant at their mean values. In a robustness check I allowed the defeat variable to decrease over time, as defeats may lose importance as time passes. The results, however, were very similar, that is, defeats significantly increase the probability that an outgoing ruler is imprisoned. The impact in all models is driven by defeats in wars. Some region dummies are significant: rulers in South America have been significantly more likely to go into exile, while leaders in Africa and the Middle East have had a higher probability of being killed. “Libyan Convoys in Niger, May Be Gaddafi deal.” Reuters, September 6, 2011. Available at http://www.reuters.com/article/2011/09/06/us-libya-idUSTRE7810I820110906 Including a variable indicating whether in a given year a parliamentary election was held yields extremely similar results.
Referência(s)