Protest, Collective Action, and Regime Change Mauricio Rivera Celestino Department of Government University of Essex marive@essex.ac.uk Kristian Skrede Gleditsch Department of Government University of Essex & Centre for the Study of Civil War, PRIO ksg@essex.ac.uk Abstract 1
Introduction In this paper, *** Protest and Regime Change Empirical research on democratization has offered a variety of explanations for why political regimes move from authoritarianism to democracy and why some autocracies are more stable than authors. Macro-focused explanations tend to emphasize the role of structural social and economic factors as the underlying causes of transitions to democracy. By contrast, microfocused explanations tend to see the decisions of individual actors and their strategic interactions as the key determinants of regime transitions, and sometimes stress their relative autonomy from social and economic structural conditions. There are two central and distinct research traditions within micro-focused approach on democratization. Research on the transitions to democracy during the so-called third wave in Latin America and southern Europe has emphasized the role of elite divisions as the main triggers of regime failure. The transition from above approach posits splits within authoritarian elites provide windows of opportunity for more moderate actors within and outside the elite, thereby facilitating agreements on transitions to democracy (e.g., O Donnell and Schmitter 1986; Przeworski 1991). However, subsequent transitions in African, Asian, and post-communist countries have inspired a new wave of research, emphasizing the role of popular collective action in undermining autocratic regimes. This transition from below approach argues that organized movements and popular mobilization strongly influence authoritarian stability, and posits that collective action causes democratization and that protest is not merely a symptom of weakening autocracies (Collier 1999). Most of the empirical research that has tried to examine the relationship between collective action and regime transitions has relied on small-n comparative analyses (e.g., Bratton and van de Walle 1992, 1997; Collier and Mahoney 1997; Bermeo 1997; Wood 2001; Tucker 2007; Slater 2009). The existing research has presented case studies that are helpful to disentangle how contentious collective action may destabilize dictators and fuels 2
democratization. Two more recent large N studies support the notion that protest and collective action may be important precursors to democratic transition. Ulfelder (2005) argues that autocratic regimes are structured in very different ways, and that the effect of protest and collective action protest on their stability will vary considerably. He tests his general argument by analyzing how violent and non-violent protest influences the risk of regime breakdown in personalist, single-party, and military autocratic regimes, using data on riots, strikes, and antigovernment demonstrations from Banks Cross-National Time- Series Data archive. His results indicate that while economic growth increases the stability of all autocracies, general strikes and antigovernment demonstrations increase the likelihood of regime breakdown in single-party autocracies. Likewise, strikes increase the risk of failure for military regimes, but antigovernment demonstrations do not have any effect on the likelihood of autocracies failing. Interestingly, protest does not appear to influence the stability of personalist regimes. Teorell (2010) examines the effect of different forms of popular mobilization, using the same data from Banks, on democratization, defined as the movement toward greater democracy. 1 His empirical results suggest that antigovernment demonstrations promote democratization, but neither riots nor strikes affect the likelihood of democratization. 2 Interestingly, if we compare the estimated effects of antigovernment demonstrations to the estimate effect of riots, it appears that non-violent protest is more effective in promoting democratization than violent protest. Teorell (2010: 116) concludes that there is a strong force from below, when peaceful. However, the alleged non-effect of strikes on democratization is not explained, and clearly not consistent with the previous quantitative and case studies research that point to labor movements as a key influence on regime change during the third wave of democratization. Both small-n and large-n studies demonstrate that contentious collective action can destabilize authoritarian regimes, even using the very crude annual event counts in the Banks data. However, much remains to be understood about why this would be the case. Existing research lacks specificity both on: (a) the actual impact of protest on transitions to democracy 1 Teorell combines and transforms the Freedom House and Polity indexes in a ten-point scale. 2 The collective action variables are lagged one year. 3
and transitions from one autocracy to another and, (b) the relationship between protest and repression and how this influences the prospects for regime change. Research so far has not identified when protest may lead to democracy and when the fall of an autocracy simply leads to another autocratic regime. For example, Ulfelder s (2005) finds that general strikes and antigovernment demonstrations increase the risk of regime breakdown in single-party regimes and that strikes promote the collapse of military autocracies, but do not speak to whether antigovernment demonstrations and strikes conduce to another type of autocracy or to a new democracy. Likewise, Teorell s finding that antigovernment demonstrations have a positive effect on democratization involves three very different outcomes. Antigovernment demonstrations may lead an authoritarian regime to liberalize. Alternatively, protest may spur the fall of one autocracy, which could lead to another autocracy or a new democratic government, possibly through a transition government. We still lack an understanding of under what conditions collective action protest increases the likelihood of democracy and under what conditions it leads to the emergence of new autocratic regimes. 3 Przeworski and Limongi (1997) famously argued that transitions to democracy were inherently random, and we could only consider the longevity of autocracies. However, protest may also entail organization that strengthens certain political factions and hence increase the likelihood of democracy. Second, research on protest and democratization has not examined government repression of popular mobilization and how this influences the prospects for regime. The literature on human rights offers seemingly contradictory findings. On the one hand, there is some evidence that mobilization and collective action increases repression in authoritarian regimes (e.g., Davenport 1995; Carey 2006). On the other hand, empirical research also shows that democratization reduces repression (Davenport 1999; Zanger 2000). Hence, the net effect of protest depends on whether we see repression, and whether repression suffices to increases the prospects for regime survival. Explaining Protest Outcomes: Regime Change and Repression 3 This may implies to reformulate hypothesis 2. 4
Although some research suggests that protest can promote transitions to democracy, the existing literature is rarely explicit on the causal mechanisms. By protest, we here mean popular dissent, directed against a government or its policies. It is common to distinguish between routine political behavior, i.e., actions sanction by the government such as voting, and non-routing direct action, where individuals either refuse to do something they are not sanctioned to do or refuse to do something that they are expected to do. In autocracies, the options for dissent tend to be extremely limited, so almost all political opposition is likely to involve direct action. We also distinguish non-violent protest from violent conflict or terrorism, where actors use violent methods to impose costs on the government or try to unseat the government. Note that we do not consider protest violent if followed by violent repression by the regime, unless the protesters also turn to use violence. Theories of non-violent action suggest a number of ways by which protest can help make democratization more likely. Sharp (1973) proposed a theory based on withdrawal of dissent, where all governments ultimately are sensitive to widespread public nonparticipation and refusals to obey their orders. As such, protest can at least in principle unseat a government through massive participation and making a country ungovernable. Moreover, initial protest can serve to highlight the extent of dissatisfaction with a regime, and encourage greater participation. However, collective action and mobilization tends to be difficult, and few protests ever manage to generate particularly large numbers. Moreover, initial protest is often met by government repression, which may discourage participation and make further protest less likely, at least at a large scale. Still, the effects of repression on subsequent protest are ambiguous, as government repression in many cases turn out to lead to an escalation in protest, especially if repression is seen as excessive and alienates people from the government. Protest can also spur the fall of autocracies through its effect on elites and leaders control of the coercive apparatus. The emergence of protest may serve as a signal to members of the elite that a government may be vulnerable to challenges, and which can encourage some factions to defect. Finally, although governments often wish to repress, they rely on individuals in their coercive apparatus to actually carry out these orders. Many regimes have 5
fallen when the military and the police have refused to carry out government orders, for example Iran in 1979 and in the German Democratic Republic in 1979. In order to explain the effects of protest on regime change and repression we examine how the structure of authoritarian regimes influences the prospects of dictators after the breakdown of autocracy. Whereas a great deal of research emphasizes simply that all leaders always want to stay in office (e.g., Bueno de Mesquita et al 2003), Geddes (2003: 49) emphasize how the costs of losing power differs for leaders as well as how the officers, parties and cliques supporting authoritarian leaders may not always wish a regime to remain in power (see also Goemans et al. 2009). We start by examining how different types of authoritarian regimes affect the costs leaders face from losing power and how this affects their sensitivity to protest (Table 1 offers some descriptive numbers on protest among autocracies). In general, we expect that lower costs from loss of power will decrease the risk of repression against dissidents and a greater likelihood that protest will promote transitions to democracy. Drawing on existing research on political regimes and democratization, we offer a stylized account about the effects of protest collective action on regime change and repression across party, military, and personalist autocratic regimes. 4 Single-party autocracies Scholarship on political regimes emphasizes that single-party autocracies often hold multiparty elections. Although there are important differences between single-party regimes, 5 electoral autocrats typically allow opponents to form independent political parties and contest elections. Another remarkable feature of authoritarian elites in dominant party systems is that they tend to remain in the political arena, even after transitions to democracy. Put differently, once electoral autocrats have been removed from office, they can often still compete for power in a new democracy (Geddes 2003:50). Indeed, if other parties are weakly organized, post dominant parties may have a considerable advantage over competitors after democratic 4 This argument resembles Ulfelder explanation. Nonetheless, we derive more concrete hypothesis with regards the effect of protest on both transition to democracy and transition to another form of democracy. Also we provide expectation on the influence of protest on repression. 5 For instance, see Diamond (2002), Schedler (2002), and Levitsky and Way (2002). 6
transitions. This in turn should make them less likely to resort to extreme repression when faced with challenges. The Mexican case illustrates well how dominant party elites can survive regime breakdown and continue competing for power in post-authoritarian politics. The Mexican revolution unseating Porfirio Díaz (1876-1911) was followed by a period of prolonged instability, with several failed coups, and culminating in the assassination of President Obregón (1928). In 1929, General Calles founded the National Revolutionary Party (PNR), intended to stabilize political competition through institutionalizing power. Paradoxically, a party structure that emerged as a provisional solution to political instability turned into the most long-lived party autocracy in the 20 th century. Indeed, the PNR and its successor Institutional Revolutionary Party (PRI) ruled for six decades, until the government accepted electoral defeat to opposition candidate Vicente Fox in the 2000 elections. 6 Not only did the PRI survive transitions to democracy, but the party has continued to play a prominent role afterwards, controlling controlling 47% and 45% of the seats in the chamber of deputies and the senate in 2003. Moreover, by 2010, the PRI controlled 16 out of 31 federal states. 7 Wright and Escribà-Folch (forthcoming) provide further comparative evidence of how old elites in single-party regimes often emerge as serious competitors for office after democratization. More specifically, of the 19 cases they study, former dominant parties won at least the second largest share of seats in at least one lower house legislative election after the transition to democracy, and 12 out of 19 won at least once. Thus, as the costs of exiting power are relatively low for dominant party elites, we expect that mass popular mobilization is less likely to be repressed and that transitions to democracy should be less likely in the aftermath of protest: 6 Prior elections tended to be non-competitive, with the exception of the 1988 presidential election, when it is widely believed that massive fraud was perpetrated to prevent the opposition candidate Cuauhtémoc Cárdenas (Magaloni 2010:757). However, Cárdenas himself came from the ranks of PRI, and was the son of a former president. 7 Electoral data from CIDAC. 7
H1 Nonviolent protest in dominant party regimes increases the likelihood of transition to democratic rule. H2 Repression will decrease the effect of nonviolent protest in dominant party regimes but not fully undermine the increased likelihood of transition to democratic rule. Military autocracies Military regimes tend to have more state repression and human rights violations (Davenport 2007), often attributed to their desire to remain in power. Other scholars have pointed to how military elites and professional soldiers place a higher value on the survival and efficacy of the military itself (Geddes 2003:54). This suggests that military elites, when faced with mass mobilization, may be inclined to step down from power and negotiate their status in a subsequent democracy. In exchange for democratization, military elites may seek concessions on three major issues: to avoid prosecution for human rights crimes, preserve independence from civilian power, and ensure a satisfactory budget for the armed forces. From this perspective, ex-ante transitions accords between the military and the opposition significantly reduce the costs of leaving power (cf. the Brazilian transition to democracy). Pacted transitions may allows the military elite to settle with opponents and potentially avoid mass mobilization and public demonstrations. Since military regimes usually emphasize public order, they are otherwise likely to respond to protest with repression and halt democratization. One might argue that protest thus would be unlikely to affect the stability of military governments. However, this both overstates the power of the military and understates the potential influence of social actors. Indeed, research suggests that popular mobilization in general and labor movements in particular may foster democratization process through two channels (Collier and Mahoney 1997). First, protest can weaken the legitimacy of incumbents and provoke instability by generating divisions amongst the elite. In Peru, for example, after a general strike in 1977, members of the traditional political parties and economic elite argued that a return to democracy was necessary to restore political order and economic growth as a result of the strike Morales Bermúdez was obligated to announce a timetable for the return to civilian rule (Collier and Mahoney 1997: 288). 8
Second, in countries where military elites started to liberalize the political arena, popular mobilization contributed to keep democratic transitions moving forward. In words of Collier and Mahoney (1997:296), the Brazilian labor movement turned the more limited process of liberalization into one of democratization, in which the party and electoral systems were opened. This process ultimately culminated in the (indirect) election of a civilian president in 1985. Though the authors acknowledge that it is complicated to assign a specific weight to different forces, they argue that labor mobilization fundamentally shaped the transition process. In addition, in spite of the finding that democratization tends to decrease repression (Davenport 1999; Carey 2000), case study research shows that popular mobilization is often followed by repression over the course of liberalization and democratization processes, where long dynamics of protest, repression, and accommodation precede and shape the final outcome (see Bratton and van de Walle 1997; Collier and Mahoney 1997). 8 For instance, altough Peru under General Velasco initially did not repress the labor movement, the subsequent government of General Morales Bermúdez changed to more repressive strategies in the mid-seventies once the workers movement acquired support from broader social segments and protest became more widespread. The opposition and the labor movement responded to repression with more confrontational strategies until the military regime finally returned to the barracks after the elections in May 1980 (see Collier and Mahoney 1997: 288-9; Bermeo 1997: 312-3). Based on the above we propose the following two hypotheses: H3 Nonviolent protest in military regimes increases the likelihood of transition to democracy. H4 Repression will decrease the effect of nonviolent protest in military regimes but not fully undermine the increased likelihood of transition to democratic rule. 8 As pioneering works neglected the importance of social forces in explaining regime change, subsequent research mainly focused on East Europe tended to underrate long-term transition process. Moreover, given data quality and availability, it has been difficult to model such processes statistically. 9
Personalist autocracies Personalist regimes differ from dominant party and military autocracies as leaders maintain power through a broad network of individual patronage, primarily in their closed networks, and through repression (Geddes 2003; Snyder 1999). A single actor monopolizes state power, including its repressive apparatuses. Because of their structure, personalist regimes are often believed to be likely to fall when leaders die and seen as more resilient that military regimes, but less stable than dominant party autocracies (Geddes 2003, Chapter 2). Based on previous research, we are skeptical of the prospects for mobilization and protest alone to be able to overthrow personalist regimes and boost democratization. Recent events in Libya suggest a role for mass mobilization under personalist rule, but we are more inclined to consider this case as an outlier, and that regime change would be unlikely in the absence of elite defections. Personalist leaders are unlikely to be able to compete for office after regime change, and unlike military leaders they can rarely pact their exit from power. As Geddes (2003: 53) points out, after having clawed his way to the top in intense and often deadly struggles among regime insiders, it seems reasonable to assume that personalist dictators have a strong and abiding determination to remain in office. Overall, the costs associated of leaving office for personalist dictators and their patronage networks are comparatively higher. 9 Not surprisingly, Ulfelder has found that protest has no effect on the likelihood of personalist regimes falling. We believe that mass mobilization in these regimes are likely to lead to conflict trajectories where opposition movements become more likely to employ violent strategies against the incumbent regime and lead to the emergence of an internal armed conflict. Recent events in the Arab world provide a good example illustrating how protest collective action provokes government repression, and then how opposition backlash increases government opposition tensions which end in civil war. Relative to other regimes, the fall of Muammar Gaddafi regime in Libya exemplifies conflict trajectories in personalist 9 The more general statement is Gaddafi s son: "Muammar Gaddafi is not Zine al-abidine or Mubarak. [He is not] a traditional... leader. The story is extremely dangerous and here I would like to draw to your attention [the fact] that Libya is not Tunisia or Egypt ( ) Libya, unlike Tunisia and Egypt, is about tribes, clans and alliances. Libya does not have a civil society or political parties (The Guardian 2011). 10
regimes after mass mobilization. Indeed, once opposition under personalist rule overcomes the barriers associated to collective action, it employs both peaceful and violent strategies in order to get concessions from the regime. As the Libyan case illustrates, governments in personalist regimes avoid making political concessions to satisfy liberalization demands from opposition. Instead, personalist governments repress opposition These expectations can be summarized as follow: H.5 Protest in personalist regimes increase the risk of repression, and is more likely to lead to internal armed conflict. 11
Table 1: Peaceful and Violent Collective Action among Autocracies, 1946-2003. Political regimes Peaceful and Violent Collective Action Type % of all autocracies Strikes Demonstrations Riots Guerrilla warfare Civil war Single-Party 33.50% 93 678 605 202 36 Military 8.03% 73 276 225 153 36 Personalist 24.66% 75 344 310 274 74 Monarchic 11.39% 18 95 117 75 5 Single-Party-Military 4.31% 8 39 30 17 6 Single-Party-Military-Personalist 4.65% 8 97 86 49 22 Single-Party-Personalist 8.67% 7 51 42 59 9 Military-Personalist 4.80% 39 139 119 52 21 Listed in number of events (strikes, demonstrations, riots, and guerrilla warfare) and country-year units of observation (civil war) General Strikes: Any strike of 1,000 or more industrial or service workers that involves more than one employer and that is aimed at national government policies or authority (from Banks). Antigovernment demonstrations: Any peaceful public gathering of at least 100 people for the primary purpose of displaying or voicing their opposition to government policies or authority, excluding demonstrations of a distinctly anti-foreign nature (from Banks). Riots: Any violent demonstration or clash of more than 100 citizens involving the use of physical force (from Banks). Guerrilla Warfare: Any armed activity, sabotage, or bombings carried on by independent bands of citizens or irregular forces and aimed at the overthrow of the present regime (from Banks). Civil war: Armed conflict between the government of a state and internal opposition groups without intervention from other states (from PRIO). 12
References Bermeo, Nancy (1997). Myths of Moderation: Confrontation and Conflict during Democratic Transition. Comparative Politics 29(3): 305 22. Bratton, Michael and Van de Walle, Nicholas (1992). Popular Protest and Political Reform in Africa. Comparative Politics 24(4): 419-442 Bratton, Michael and Van de Walle, Nicholas (1997). Democratic Experiments in Africa: Regime Transitions in Comparative Perspective. New York: Cambridge University Press. Carey, Sabiney (2010). The use of Repression as a Domestic Dissent. Political Studies 58: 167-87. CIDAC. Base de Datos Electoral CIDAC. http://www.cidac.org/esp/datos_electorales.php Collier, Ruth Berins (1999). Paths Toward Democracy. Cambridge: Cambridge University Press. Collier, Ruth Berins and James Mahoney (1997). Adding Collective Action to Collective Outcomes: Labor and Recent Democratization in South America and Southern Europe. Comparative Politics 29(3): 285 303. Davenport, Christian (1995). Multi-Dimensional Threat Perception and State Repression: An Inquiry into Why States Apply Negative Sanctions. American Journal of Political Science 39(3): 683-713 Davenport, Christian (1999). Human Rights and the Democratic Pacification. Journal of Conflict Resolution 43(1): 92-116. Davenport, Christian (2007). State Repression and the Tyrannical Peace. Journal of Peace Research 44(4): 485 504. Diamond, Larry (2002). Elections without Democracy: Thinking about Hybrid Regimes. Journal of Democracy 13(2): 21 35. Francisco, Ronald (2005). The Dictator s Dilemma. In Christian Davenport, Hank Johnston, and Carol Muller (eds.). Repression and Mobilization. Minneapolis: University of Minnesota Press. Geddes, Barbara (2003). Paradigms and Sand Castles. Theory Building and Research Design in Comparative Politics. Michigan: Michigan University Press. Levitsky, Steven, and Lucan Way (2002). The Rise of Competitive Authoritarianism. Journal of Democracy 13(2): 51 65. 13
Magaloni, Beatriz (2010). The Game of Electoral Fraud and the Ousting of Authoritarian Rule. American Journal of Political Science 54(3): 751 765 O Donnell, Guillermo and Philippe Schmitter (1986). Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies. Baltimore, MD: Johns Hopkins University Press. Przeworski, Adam. (1991). Democracy and the Market. Political and Economic Reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press. Schedler, Andreas (2002). The Menu of Manipulation. Journal of Democracy 13(2): 36 50. Slater, Dan. 2009. Revolutions, Crackdowns, and Quiescence: Communal Elites and Democratic Mobilization in Southeast Asia. American Journal of Sociology 115 (1): 203-54. Snyder, Richard (1999). Explaining Transitions from Neopatrimonial Dictatorships. Comparative Politics 24(4): 379-99. The Guardian (2011). http://www.guardian.co.uk/world/2011/feb/21/gaddafi-son-muammarlibya, February 22 2011. Teorell, Jan (2010). Determinants of Democratization. Explaining Regime Change in the World, 1972-2006. Cambridge: Cambridge University Press. Tucker, Joshua (2007). Enough! Electoral Fraud, Collective Action Problems, and Post- Communist Colored Revolutions. Perspectives on Politics 5(3): 535-51. Ulfelder, Jay (2005). Contentious Collective Action and the Breakdown of Authoritarian Regimes. International Political Science Review 26 (3): 311 334. Wood, Elizabeth (2001). An Insurgent Path to Democracy: Popular Mobilization, Economic Interests, and Regime Transition in South Africa and El Salvador. Comparative Political Studies 34(8): 862-88. Wright, Joseph and Abel Escribà-Folch (Forthcoming). Authoritarian Institutions and Regime Survival: Transitions to Democracy and Subsequent Autocracy. British Journal of Political Science. Zanger, Sabine (2000). A Global Analysis of the Effect of Political Regime Changes on Life Integrity Violations, 1977-93. Journal of Peace Research 37 (2): 213-33. 14