In militaries that are more professionalized at the time of the seizure of power, the likelihood of a “rogue” coup succeeding is low and the cost of a failed attempt very high because careers are predictably ended by serious breeches of discipline. As a result, the high command can credibly promise not to oust the dictator as long as he sticks to prior agreements about sharing, rotation, and policy. If the military can make credible promises, the dictator has more reasons to work out an acceptable sharing arrangement rather than creating a third actor, the party, with which spoils have to be shared as well.
Consequently, on average, more professionalized militaries have been better able to enforce agreements about consultation and executive rotation, and they have been more successful at preventing the personalization of rule – when necessary, by ousting junta leaders who seem to be trying to concentrate power in their own hands. Recently created officer corps, however -- whether created by indigenization at the time of independence or by U.S. forces trying to build “neutral” National Guards during occupations of Central American and Caribbean nations -- have shown less ability to check the ambitions of fellow officers who manage to attain the presidency.
In summary, where the dictator must bargain with a unified military that could oust him with near certainty, he agrees to a more equal distribution of resources and power, consistent with Rubinstein’s (1982) famous result for two-person divide-the-dollar games. Where, however, he can bargain separately with different factions based on ethnicity or personal loyalty within the military, he could exclude some factions altogether, provide minimal resources to the other members of his “minimum-winning” coalition, and keep the rest if he did not fear coups (Geddes 2004).11 To make this a safe strategy, the dictator needs to deter coups, which – as noted above -- can be undertaken by any dissatisfied faction. Party creation, personal control of the security apparatus, and the creation of paramilitary forces are the main deterrence strategies available.
The argument that other officers in post-seizure coalitions resist party creation because it is part of a leader’s strategy for enhancing personal discretion and power implies that other officers would prefer a shorter term for the regime to the personalization of rule, since party creation increases regime durability. Two observations make this idea plausible. First, observers have long noted that most officers prefer returning to the barracks to remaining in power while the military becomes factionalized and politicized.12 Second, when a leader who came to power via coup successfully concentrates power and discretion in his own hands, he marginalizes most of the military. Although casual observers may still label the regime military, those officers excluded may have as little influence as they would have had in a civilian government, and they are worse off professionally because future promotions will depend on the leader’s favor. They thus have little reason to care about the survival of the regime if they have been marginalized.
The argument above implies that a number of relationships should be observable in the real world.
If authoritarian party creation is a strategic choice by incumbent rulers to protect their rule from possible military rivals, we should find that most party creations are initiated by the incumbent. We cannot observe that directly because we cannot tell from public announcements who sought to create this new institution. We should, however, observe the following:
*Most newly created authoritarian support parties would be led either by the incumbent dictator himself or one of his close relatives or allies.
*Authoritarian support parties would often be created as part of a process of marginalization of the military. A concrete indicator of military marginalization is dissolution of military consultative bodies.
*Parties would be created in the run-up to elections that confirm the incumbent as national executive. Such elections create an appearance of popular legitimacy aimed at undermining the feeling among officers that what the military gives, it has the right to take away.
If more professionalized militaries have greater ability to resist the personalization of rule, and hence party creation, by one of their number, then we might expect to see more parties created in situations in which professional norms had not had time to become fully internalized. We cannot measure the internalization of professional norms in a large number of militaries, but it seems reasonable to assume that they take time to develop. Therefore, it also seems reasonable to assume that professionalization is weaker within 20 years of: independence in countries where officer corps were indigenized shortly before independence; or the creation by foreign occupiers of a National Guard or other defense force to replace pre-existing military forces. Seizures of power by junior or non-commissioned officers could also be used as a proxy for low professionalization. Such seizures indicate that the conventional military emphases on hierarchy and discipline – hallmarks of a professionalized military -- have either not yet become internalized by many officers or are breaking down. Consequently, we might also expect party creation to occur more frequently after such seizures of power.
Data to test these arguments will come from expanding the collection of information I have used to classify authoritarian regimes types.13 The original dataset has been updated, expanded to include monarchies, regimes that began after 1989, and regimes of less than three years duration. A number of new variables, including dates of party creation and details about the seizure-group, have been added to the original information used for coding regime types. The addition of a number of the new variables needed to assess the implications noted above is not complete, so all empirical claims made here are preliminary, based on partial coding, and I report only very simple findings that I think are unlikely to disappear as more data accumulate.
Is Party Creation Usually Initiated by the Dictator?
The identity of the first leader of parties created after seizures of power is missing for quite a few cases, but for those available, the regime leader or a close relative of his led the newly created party in about two-thirds of cases. In some of the others, either the regime leader purged the first party leader after a year or two of organizational effort, or the party leader eventually ousted the regime leader. The organization of a new party was always delegated to a close relative or ally if the leader did not keep the post for himself, but even so, delegation seems to have been dangerous. Where the regime leader delegates control over the new party to one of his allies, the ally can become a powerful rival. As Stalin and Nasser showed, the loyal lieutenant who controls the party can build a well-organized network of activists who depend on him for their future jobs and promotions, and who will help to put and keep him in power.
The circumstances under which decisions to create parties are made also suggest that they form part of the consolidation strategy for specific leaders. In non-monarchic regimes, I have found no descriptions of parties created for the purpose of organizing or stabilizing majorities in authoritarian legislatures. Instead, parties are often created as part of the strategy to insure that an authoritarian supermajority is elected in the first place.14 The most common situation in which parties are created is during the run-up to elections to either confirm the incumbent regime leader as president or elect a legislature that will then confirm him. When legislatures have the task of anointing the president, the selection of legislative candidates achieves high importance, even if elections are uncompetitive. Party ruling committees are given the task of choosing candidates, usually in consultation with the regime leader himself. Parties are also expected to turn out the vote for the new leader in presidential elections. Party activists, who are often public employees, are charged with spreading the regime’s messages, distributing benefits, holding campaign events, and making sure that citizens turn out to vote and vote for the right candidate (if they have a choice).
Can More Professionalized Militaries Deter Party Creation?
For this test, countries with recently indigenized or occupier-created officer corps are compared with countries in which the officer corps had had more than 20 years of autonomous development. The assumption underlying this test is that in recently created officer corps, professional norms have not had time to become fully internalized. As Table 6 shows, among the cases in which coding has been completed, parties were created after 58 percent of illegal seizures of power carried out in countries with recently indigenized or occupier-created military forces. Twenty-one percent of such regimes allied with a preexisting party, and an equal number eschewed the use of a support party. After seizures of power carried out in countries with officer corps that had been in place more than 20 years, and can therefore be assumed to be more professionalized, all else equal, only 33 percent of leaders created new parties. Nine percent allied with preexisting parties, and 58 percent made no use of support parties. These findings are consistent with the argument that more professionalized militaries can more successfully resist the personalization of rule after seizures of power.
They are also consistent with several of the results of the analysis of party creation, as shown in Table 4. As noted above, younger dictators are more likely to create parties, and dictators tend to be younger in countries with less professionalized militaries because more coups are led by junior officers in these countries. The passage of time has reduced the incidence of post-seizure party creation, as would be expected if militaries are becoming more professionalized over time. The analysis also shows that dictators in poorer countries are more likely to create support parties. On average, countries with more professionalized militaries are richer and more industrialized than countries with recently indigenized or occupier-created militaries.
We generally expect richer countries to experience fewer coups and have stabler governments, whether authoritarian or democratic (Londregan and Poole 1990). Among authoritarian regimes initiated by seizures of power not led by parties, however, this relationship is reversed. The reversal occurs because professionalized militaries can prevent the creation of authoritarian support parties that would make governments more stable while reducing military influence on them. It may seem surprising that coups are more common when the military is more professionalized, but the difference is quite large in these data. Coups in professionalized military regimes are analogous to votes of no-confidence in parliamentary democracies. They are simply the way the top officers in the military replace executives with whom they are dissatisfied; they are common and usually bloodless. Because it is relatively easy for a unified military to replace incumbents, incumbents who wish to remain in office in such regimes must be responsive to the demands of their military colleagues. Incumbents committed to changes in the status quo or who face serious economic problems, however, may be unable to respond to military demands.
This paper has addressed two questions. Through what means do newly created parties prolong the life of authoritarian regimes? Second, since party creation prolongs regime life, why do nearly 50 percent of dictators who achieve power without being part of a previously organized party not create one? To answer the first question, I have argued that the creation of a party to support a particular leader creates vested interests in his survival among party leaders and activists. These party leaders and activists in turn have the resources to mobilize mass demonstrations that deter coups. The party can thus serve as a counterbalance to other intra-regime factions, especially erstwhile allies in the military. Official party names, platforms, and ideologies claim a broader and more impersonal purpose than support for particular leaders, and parties may in some circumstances come to serve some of these broader purposes, but during the first years after a seizure of power, they are usually controlled by one person and the close allies to whom he delegates day-to-day supervision. Support parties can end up prolonging the life of authoritarian regimes, not just particular leaders, but their creation and early tasks, reorganization, and purging are determined by particular leaders engaged in competition with rivals for power within the regime elite.
Authoritarian parties may also serve other purposes. Regime leaders often assign to them the tasks of building popular support for the government by distributing benefits to citizens in villages and neighborhoods and spreading the regime’s ideology or policy views. We should not assume that the existence of a regime support party implies greater popular support, however, because parties’ actual ability to accomplish these tasks varies greatly. In some dictatorships, parties barely reach beyond the capital city. In others, they penetrate to the smallest village and play an active role in the economic life of ordinary people.
Regardless of whether parties improve the quality of life for ordinary citizens or persuade them to agree with regime views, however, they provide their officials and activists with benefits that give them a stake in the regime. Party workers often draw salaries. They have preferential access to jobs in the state bureaucracy and schooling for their children. They have insider opportunities to form businesses subsidized by the government and to manage or even take ownership of expropriated businesses or land. Their connections help them get lucrative government contracts and to profit from restrictions on trade. They have the possibility of rising in the party to achieve the political power and, usually, wealth associated with high office. Even where party activists enjoy no current benefits, their connections open up future possibilities for rewards and upward mobility. For all these reasons, party militants have a very strong interest in the survival of the party and the party leader to whom they owe their current good fortune. These militants thus have an interest in organizing mass support for their leader in any conflict with other regime insiders. They also control the means to influence ordinary citizens’ behavior since they control the distribution of resources and can report “disloyalty” to security personnel. So they can turn out the masses for either demonstrations or elections. These demonstrations and elections deter rivals. The most dangerous rivals to leaders of regimes that achieved power without party support usually come from within the military.
The evidence shown above shows that creating a party after seizure does deter coups. After controlling for other factors known or believed to affect the incidence of coups, party creation reduces the likelihood that a coup will occur. These results lend some support to the argument that, whatever else parties may be intended to accomplish, they do on average deter coups.
The answer I propose to the second question is that party creation is a strategy used by military officers who have achieved the position of head of state when the factionalization or lack of professionalization of the officer corps that temporarily supports them prevents it from making credible promises either to oust the dictator if he fails to share with them or to support him if he does share. If the dictator cannot secure his position by sharing, then he is better off keeping a larger share of the spoils and other benefits of office so that he can invest in creating a party to support his rule and also in coercive forces under his personal control. Both parties and paramilitary forces, I argue, are created to deter coups. The game in Figure 3 shows the logic underlying this argument.
Strong evidence to support this part of the argument awaits the completion of data collection, but preliminary evidence is consistent with the argument that factionalization and lack of professionalization within the military increase the likelihood of party creation. Parties are more often created by younger dictators and by dictators who rose to power in recently indigenized officer corps or armed forces recently created by foreign occupiers. Some of the world’s most famous dictators, including Mobutu, Trujillo, and Eyadema, arose in these circumstances.
Aldrich, John. 1995. Why Parties? The Origin and Transformation of Political Parties in America. Chicago: University of Chicago Press.
Arriagada, Genaro. 1986. “The Legal and Institutional Framework of the Armed Forces in Chile” in Military Rule in Chile: Dictatorship and Oppositions, ed., Arturo Valenzuela and Samuel Valenzuela
Banks, A. S. 2001. Cross-national time series data archive (data set). Binghamton, NY: Computer Systems Unlimited
Baron, David and John Ferejohn. 1989. “Bargaining in Legislatures,” American Political Science Review 83: 1181-1206.
Belkin, Aaron and Evan Schofer. 2003. “Toward a Structural Understanding of Coup Risk,” Journal of Conflict Resolution 47:5, 594-620.
Boix, Carles and Milan Svolik. 2007. “Non-Tyrannical Autocracies,” presented Comparative Politics Workshop, UCLA.
Box-Steffensmeier, Janet, Suzanna De Boef, and Kyle Joyce. 2007. “Event Dependence and Heterogeneity in Duration Models: The Conditional Frailty Model.” Political Analysis
Bueno de Mesquita, Bruce and Alastair Smith. 2007. “Political Survival and Endogenous Institutional Change,” unpublished manuscript, NYU.
Bueno de Mesquita, Bruce, Alastair Smith, Randolph Siverson, and James Morrow. 2003. The Logic of Political Survival (Cambridge: MIT Press).
Chehabi, H. E. and Juan Linz, eds. 1998. Sultanistic Regimes (Baltimore: The Johns Hopkins University Press).
Cox, Gary and Mathew McCubbins. 1993. Legislative Leviathan: Party Government in the House. Berkeley: University of California Press.
DeNardo, James. 1985. Power in Numbers. Princeton: Princeton University Press.
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper.
Escriba Folch, Abel. 2003. “Legislatures in Authoritarian Regimes,” Working Paper 196, Instituto Juan March de Estudios e Investigaciones, Madrid.
Fontana, Andrés. 1987. “Political Decision-Making by a Military Corporation: Argentina, 1976-83,” Ph.D. dissertation, University of Texas.
Gandhi, Jennifer. 2003. “Dictatorial Institutions and Their Impact on Economic Performance,” presented APSA 2003.
Jennifer Gandhi and Adam Przeworski. 2005. “Cooperation, Cooptation, and Rebellion under Dictatorship,” Economics and Politics
Geddes, Barbara. 2003. Paradigms and Sand Castles: Theory Building and Research Design in Comparative Politics. Ann Arbor: University of Michigan Press.
Geddes, Barbara. 2004. “ Minimum-Winning Coalitions and the Personalization in Authoritarian Regimes,” presented APSA.
Haddad, George. 1973. Revolutions and Military Rule in the Middle East: The Arab States, Pt. II: Egypt, The Sudan, Yemen, and Libya. NY: Robert Speller & Sons.
Hartlyn, Jonathan. 1998. “The Trujillo Regime in the Dominican Republic,” in H. E. Chehabi and Juan Linz, eds., Sultanistic Regimes (Baltimore: The Johns Hopkins University Press).
Janowitz, Morris. 1960. The Professional Soldier: A Social and Political Portrait (Glencoe, IL: Free Press)
Jowitt, Kenneth. 1975. “Inclusion and Mobilization in European Leninist Regimes,” World Politics XXVIII: 69-96.
Londregan, John and Keith Poole. 1990. “Poverty, the Coup Trap, and the Seizure of Executive Power.” World Politics 42: 151-83.
Lust-Okar. 2005. “Elections under Authoritarianism: Preliminary Evidence from Jordon,” presented APSA.
Magaloni, Beatriz. 2006. Voting for Autocracy. NY: Cambridge University Press.
Magaloni, Beatriz. Forthcoming. “Credible Power-Sharing and the Longevity of Authoritarian Rule.” Comparative Political Studies.
Marshall, M. G. and K. Jaggers. 2000. Polity IV project: Political regime characteristics and transitions, 1800-1999 (data set). College Park, MD: Center for International Development and Conflict Management.
Nordlinger, Eric. 1977. Soldiers in Politics: Military Coups and Governments. Englewood Cliffs, NJ; Prentice-Hall.
O’Donnell, Guillermo and Philippe Schmitter. 1986. Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies. Baltimore: The Johns Hopkins University Press.
Rubinstein, Ariel. 1982. “Perfect Equilibrium in a Bargaining Model,” Econometrica 50: 97-110.
Schattschneider, E. E. 1942. Party Government. New York: Farrar and Rinehart.
Schwartz, Thomas. 1989. “Why Parties?” Unpublished manuscript, UCLA.
Wright, Joseph. 2007. “Political Regimes and Foreign Aid: How Aid Affects Growth and Democratization.” Ph.D. dissertation, UCLA.
Table 1 Effect of Party Creation on Regime and Dictator Survival
Average Regime Duration
Average Dictator Survival*
Creates a Party
Does Not Create a Party
Does Not Create (excludes “moderating military interventions)