Sage Journals: Discover world-class research

Abstract

Whereas protests have been discussed predominantly in terms of collective action issues, achieving coordination does not always guarantee success. Protest groups must also back their demands with sufficient threats. Some assert that threats are enhanced by the mobilization of more resources. Yet this conventional wisdom fails to explain why not all large-scale protests win government concessions or why some protest groups spend resources on their organizational infrastructure even though it will not inflict immediate damage on the government. Formalizing protest in a bargaining model, I show that investing in organizational infrastructure improves the impact of protest groups’ threats by lowering the probability that a counter-protest will offset the impact of the original protest.

Keywords

Counter-protest formal model organizational development protest protest group

Introduction

To be successful, protest groups must back their demands with sufficient threats to sway political leaders. When civic groups seeking political change employ different forms of disruptive interference such as demonstrations and violent campaigns, the threats they pose are claimed to be enhanced when a bargainer can mobilize more resources (e.g. participants, money, facilities, violence), demonstrating its ability to attack a competing bargainer in case negotiations fail (Banks, 1990; Fearon, 1997; Schelling, 1960). Consistent with these insights from bargaining theory, much of the protest literature explicitly argues or implicitly depends on the assumption that protest groups mobilizing more resources have a higher ability to force political leaders to the bargaining table and are more likely to obtain concessions (Bell et al., 2014; DeNardo, 1985; Gamson, 1975; Lipsky, 1968; McCarthy and Zald, 1977; Schumaker, 1975).¹

Although the assumption concerning resources is widely accepted, it is contradicted by two sets of empirical phenomena. The first puzzle is why some large-scale protests fail to win political leaders’ concessions. Consider the cases of antinuclear protests in the 1970s. France and West Germany achieved higher levels of participant mobilization than other antinuclear movements in Europe and the US (Kitschelt, 1986: 71–72),² but made little impact on government policies. Their American counterparts, however, succeeded in forcing the abandonment of several nuclear power plants planned or under construction (Kitschelt, 1986: 71–72; Kolb, 2007), although protests in West Germany and France should have posed greater threats to the governments and therefore won government concessions if the conventional wisdom is true. Further, one group’s excessively large mobilization promotes counter-mobilization by another seeking to maintain the status quo (Tilly and Tarrow, 2015: 38), making bargaining with government less straightforward.

The second puzzle is why some protest groups reduce the amount of resources available for disruptive activity by investing in their organizational development. The antinuclear protesters in the US bargained with the government by spending some of their resources on building their organizational infrastructure rather than only on street demonstrations (Kitschelt, 1986: 68). For unstructured, grassroots protest groups to transform into structured organizations, they must invest in organizational infrastructure. For example, the US recycling movement hired paid professional staff and incorporated key activists during its loose process of structuring into a national recycling association in the 1960s, rather than using volunteers (Lounsbury, 2005: 77–83); successful antinuclear movements made efforts to win legal incorporation and transform from grassroots to regional or national groups founding umbrella and branch offices (Nelkin and Fallows, 1978); and lobbyists are obliged to register and pay the fee in most US states. However, simply hiring full-time staff and acquiring office facilities does not immediately inflict damage upon governments. From a resource-mobilization perspective contending that the more resources are mobilized for disruptive interference in what governments are responsible for, the more powerful a protest becomes, the allocation of resources to organizational development should not immediately improve a group’s ability to influence the policymaking process. Moreover, the formation of organized advocacy groups can be threatening to political leaders in future periods as it may facilitate coordination among protesters with stronger leadership. Such long-term effects are not necessarily promising, however, owing to a potential decrease in public support or issue salience compared with the effect of disruptive action that can punish political leaders. Notwithstanding, US antinuclear protests devoted considerable effort towards organizational development to great success. Moreover, civic advocacy groups investing in both protest activity and organizational development usually have a larger impact on the policymaking process than episodic protest groups with few organizational foundations.

Relatedly, it is also puzzling that despite several benefits of organizational development claimed thus far, not all protest groups mobilize resources for their organizational evolution. If organizational development benefits organized groups, every group should have invested in it.³ Therefore, certain short-term incentives must cause civic advocacy groups to allocate resources to organizational infrastructure.

By formalizing a one-shot bargaining model of protests between a political leader and two sets of audiences to whom the political leader must appeal, I will clarify a mechanism by which those puzzling processes can go hand in hand and show that protest groups do have incentives to distribute some resources to building organizational infrastructure, even in a one-shot game. This article proves that excessively high investment of resources in protest activity perversely undermines a protest group’s ability to influence the government policymaking process because it facilitates a counter-protest that offsets the impact of the original protest. This result challenges the conventional wisdom that the greater the resources mobilized for protests the more successful the protests will be.

The findings demonstrate that protest groups can benefit from distributing resources between both protest activity and organizational development, although the more resources groups use for organizational infrastructure, the fewer costs they can then impose upon government today. Weakening a protest group’s capacity to sway the government in a credible, observable manner (i.e. organizing groups) makes a pivotal contribution to a competing protest group’s willingness to stage a counter-protest and thereby improve its odds of winning concessions. This finding explains why the US antinuclear protests in the 1970s led to substantial government concessions despite their relatively low level of resource mobilization for protest activity whereas those in West Germany did not. This result can also explain why radical civil resistance, such as terrorism, tends to backfire by hardening the population’s stance (Abrahms, 2013; Fortna, 2015; Gould and Klor, 2010). This article presents, to the best of my knowledge, the first attempt to theorize the significance of a strategy to reduce counter-protests.

This article speaks to the literature on civic advocacy groups. Although protest groups and interest groups attempt to influence policymaking processes in essentially the same way (Berry, 1999: 142), these groups have been examined separately: sociologists have predominantly examined protests (e.g. Gamson, 1975) and interest group politics have primarily been studied in economics (e.g. Grossman and Helpman, 1994; Tullock, 1980). With the premise that organized advocacy groups, including interest groups, and episodic protest groups are on a continuum representing the degree of organizational development and by endogenizing civic groups’ decision on their organizational forms, this article bridges several groups of literature that have often failed to talk to each other.

Literature

Protests are predominantly discussed in the collective-action framework (e.g. Gavious and Mizrahi, 1999; Karklins and Petersen, 1993). Although the resolution of coordination problems is essential for protests to emerge, the cases of unfruitful protests demonstrate that coordination does not always guarantee success. To better understand protest consequences, protests also need to be discussed in terms of threats. Protesters have opportunities to punish political leaders ex post if the leaders withhold concessions, for example, by voting against the leaders, cutting off financial support for state agencies and resorting to violence.

Only a few studies formalize the bargaining aspect of protests (Lindvall, 2010; Scartascini and Tommasi, 2012). The inter-state bargaining literature, however, provides insights into the role of resource mobilization as threats (e.g. Fearon, 1997; Schelling, 1960; Slantchev, 2005) to suggest that protesters should impose a sufficient magnitude of threats towards government to make the government prefer the retraction of threats to their fulfillment. A number of studies on civil resistance depend on the assumption that the more resources a group mobilizes for protest activity, the greater the group’s threat becomes (Bell et al., 2014; Gamson, 1975; McCarthy and Zald, 1977; Schumaker, 1975; Tenorio, 2014). However, this conventional wisdom is now under increasing challenge.

First, the amount of resources mobilized for protest activity does not always positively correlate with protest outcomes. Empirical studies have yielded mixed (Colby, 1982; Giugni, 2004) and contradictory (Taft and Ross, 1969) results concerning the efficacy of a high level of mobilization. Cases question the applicability of a costly signalling framework to the context of protests. Some protests using significant resources failed to win political leaders’ concessions and often provoked counter-protests to offset the efforts of the initial protests. For example, rightist protests such as segregationist protests (Andrews, 2002) and antiabortion protests (Meyer and Staggenborg, 1996) often faced counter-protests, and antigovernment protests in countless countries such as Russia, the US and Thailand have also propelled rival movements. Meanwhile, a small number of protesters have made ‘a big splash’ such as those during the New York rent strikes (Nie and Verba, 1975: 24) and piquetero (road blockade) protests in Argentina (Benclowcz and Breña, 2011).⁴

Second, the literature on protests and on bargaining in general tends to overlook the possibility of counter-mobilization when analysing the interaction between protesters and government, relying on the assumption that political leaders have only one set of audiences to appeal to. This assumption may hold in the context of inter-state bargaining in which domestic audiences uniformly do not want their home leader to concede,⁵ but not in the context of domestic bargaining in which public preferences about their leaders’ concessions are diverse. The effects of concessions are essentially twofold (Schelling, 1960: 34). Suppose that an incumbent leader modifies a given status-quo policy that she has implemented. Protesters dissatisfied with the policy may construe her concessions favourably, whereas people who prefer the previous policy may think that the leader has broken her prior commitment and may become suspicious of any new pretence at commitment. This situation indicates that, in domestic bargaining, larger mobilization in protest activity aimed at increasing the probability of executive concessions may incentivise counter-mobilization seeking to offset the impact of the initial mobilization. Although a few exceptions directly deal with counter-protests such as Meyer and Staggenborg (1996) and Zald and Useem (1987), they do not analyse how the emergence of counter-protests constrains political leaders’ decision-making. Little is known about such countervailing strategies that one group employs against another.⁶

Furthermore, a theoretical shortcoming of protest literature has yet to be solved. Most literature on protest takes it for granted that protest groups have already formed, although a number of protests were staged without back-offices or clear leadership. In other words, people should have options regarding the formation of protest organizations, the extent to which they make their organizations institutionalized, and the amount of resources they will utilize for protest activities. One reason for the lack of perspective on endogenous institutionalization is in part attributed to the fragmentation of scholarship. Traditionally, ad hoc protest groups and well-structured civic advocacy groups such as interest groups were studied in isolation, although how those groups attempt to influence policies is the same in that both forms of groups pay costs to pressure governments and to influence their decision-making process (Andrews and Edwards, 2004; Berry, 1999). The major difference is how they distribute resources between protest activity and organizational development. Historically, some interest groups emerged as an institutional outlet for social movements such as those that arose after the 1960s in the US (Berry, 1999: 142).

The second reason for this shortcoming is that the effect of resource mobilization for protests and organizations is not clearly discerned. Protest literature assumes that using resources benefits protest groups no matter the purpose. Yet protests and organizations are different in their observability and uncertainty. Protesting on the street is intended to be seen by the public and government to improve the policy influence and social presence whilst organizing civil groups is less visible, often behind the curtain. Furthermore, groups’ organizational development does not necessarily guarantee their future ability to threaten a government and to survive.⁷ Given governments’ limited ability to address threats, ongoing threats should be immediate threats to government unlike potential threats in future periods. Without dissecting the effect of protest groups’ resource mobilization, we may miscalculate the effect of protests.

Thus, the literature on protest may not have overcome the selection bias inherited in the categorization of civic groups. Selection effects present an inferential problem because citizens should have incentives to form better organized groups such as lobbying groups if they provide better bargaining leverage. From a resource-mobilization perspective, then, no persuasive answer as to how civic groups rationally determine their organizational forms and what endogenous factors produce such different degrees of bargaining leverage between civic advocacy groups of different levels of organizational development has been found. If civic groups that devote resources to organizational development are more likely to succeed than those that do not, it would follow that those protest groups that the protest literature analyses fail to achieve their goals despite mobilizing all resources for protest activity. Placing episodic protest groups and organized civic groups on a continuum representing different degrees of organizational development, then, matters for more accurately assessing the effect of resource mobilization and more coherently understanding ordinary citizens’ political participation.

The model

The model highlights strategic interaction between a political leader and two sets of audiences, which is sketched in Figure 1. Two groups of citizens are potentially in conflict over a given status-quo policy I has implemented (SQ: status-quo policy) represented by $x_{SQ} = 0$ , and a political leader (I, shorthand for incumbent) who is in charge of policy choice and implementation may face protests by either or both. Let the respective groups be called the anti-SQ group (A), which is potentially dissatisfied with the SQ and the pro-SQ group (P) whose ideal policy is identical to the SQ. Each ideal policy is defined as $x_{A} > 0$ and $x_{P} = 0$ , respectively. Protesters can simultaneously set a wide variety of agendas in reality, but I assume that these three players are bargaining over a single issue at one time. Each group is assumed to be a unitary actor, and I refer to A as ‘she’ and P as ‘he.’

Figure 1.

Sequence of play.

The game begins with Nature randomly selecting I’s type and P’s type. I’s type as c is drawn from a uniform distribution on interval $[0, \bar{c}]$ . The parameter c measures the amount of technical costs I must pay, such as the structural costs of taking legislative procedures and coordinating between coalition parties. Neither A nor P knows the true value of c, but they know that it is drawn from the aforementioned uniform distribution.⁸P’s type as $m_{P}$ is drawn from a uniform distribution on interval $m_{P} \in (0, 1]$ . $m_{P}$ denotes P’s maximum capacity to mobilize resources. P and I know the true value of $m_{P}$ , but A does not. This is based on the fact that pro-status-quo-policy groups tend to work with a government to achieve their policy goals while attempting to hide their resource mobilization capacity from their rival groups. Being a challenger to the status quo, A must engage in policy bargaining with these uncertainties.

Second, A decides whether to stage a protest as a threat to I; by staging a protest, A can show that she will punish I if I does not accommodate A’s demand. $m_{A} \in (0, 1]$ represents A’s maximum capacity to mobilize resources. Since a government is usually capable of collecting information about dissident activities for security reasons, the value of $m_{A}$ is assumed to be common knowledge. $m_{A}$ could be a parameter measuring her maximum possible impact on political, economic and social issues: it could represent the number of participants, amount of financial resources or damages from protest activity.⁹ More resources allow A to pose a greater threat and provide an incentive for I to change a policy over which they compete. However, mobilizing also imposes costs directly on A in the form of time, lost resources and possible wages, and the like.

When A decides to stage a protest, A simultaneously makes two other decisions. One is the level of organizational development. A chooses how much resources to distribute to building organizational infrastructure, $α \in [0, m_{A})$ . A spends $α$ out of her resource pool of size $m_{A}$ and spends the rest, $m_{A} - α$ , on pressuring I. Note that, in this one-shot game, resources mobilized for protest activity, $m_{A} - α$ , will be translated into costs inflicted on I but those for organizational development, $α$ , may not. A is not assumed to pay $α$ when she does not engage in protest activity. $α$ is assumed to be common knowledge. No matter how she distributes her resources, resource mobilization always imposes costs. The other decision is on a policy proposal. A proposes a take-it-or-leave-it offer $x \in X = (0, x_{A}]$ .

Third, P decides whether to organize a counter-protest by using his resources $m_{P}$ . P’s counter-protest also serves as a threat to I and shows that P will punish I when I concedes to A. The cost of counter-protests also sinks. Neither A nor P is assumed to organize a protest when they are indifferent. Finally, I either accepts or rejects A’s offer and is assumed to accept when indifferent between accepting it and not.

The payoffs for both groups are determined by policy outcomes and costs of mobilization. Benefits from policy outcomes depend on the distance between the final policy outcome and the ideal point of each group. If the game ends in the status quo, A and P get $- x_{A}$ and 0, respectively. I’s acceptance of the offer x brings $- | x - x_{A} |$ and $- x$ to A and P, respectively. Mobilization costs are added to the benefits above if and only if each chooses to undertake a protest.

Since I is assumed to be an office seeker, I’s payoff consists of political costs inflicted by the protest groups and the technical costs of changing the SQ. When the game ends with no protest, I gets 0 since I neither increases nor decreases public support. When A organizes a protest and I rejects the offer, A’s protest activity inflicts costs upon I as punishment, for the amount of $- (m_{A} - α)$ to handle the destruction that mobilization of size $m_{A} - α$ could cause.¹⁰ This cost could correspond to expenditures on policing, damage to the economy and obstruction to the legislative process. Moreover, A’s investment in organizational infrastructure becomes a threat to I with probability p. That is, I also incurs $- p α$ when rejecting A’s offer, where p means the exogenous probability of A’s organizational survival. Thus, A’s survival may be subject to unexpected shocks but should be more likely as A pays more resources on organizational infrastructure. Meanwhile, accepting A’s offer imposes technical costs of changing the SQ on I, which is given by $c > 0$ , regardless of whether P stages a counter-protest.¹¹ Accepting A’s offer also carries political costs from P as punishment if and only if a counter-protest occurs. The size of costs imposed by P are determined by the distance between P’s ideal policy and a new policy, $- x$ .¹² This cost increases as I makes a greater policy concession because a greater decline in support from P’s members is expected in a next election.

Before turning to formal results, it is important to assess whether the model’s structure and assumptions are sufficiently reasonable. One of the key assumptions is that A and P will use all resources when they decide to stage a protest. A is allowed to allocate her resources to protest activity and organizational development. That is, when A stages a protest, she will either use all her resources only for protest activity or allocate her resources between those two purposes and exploit them all. I make this assumption because although an alternative structure may allow the groups to endogenously choose their size of investment in protest activity, allowing each group to choose the amount of mobilization does not produce any substantively peculiar dynamics. Everything else being equal, protest groups tend to mobilize as much as possible to maximize their ability to control the bargaining.

The second assumption is that A is allowed to organize herself but P is not because it is often the case on the ground that the pro-status-quo-policy groups achieve much better organizational development than anti-status-quo-policy groups and are often backed by state institutions and industries. For example, as I will introduce later, the 1970s pro-nuclear counter-protesters in West Germany were mainly led by scientists and trade unions (Rucht, 1990: 204). Similarly, the US pro-nuclear movements were often led by industry firms (Useem and Zald, 1982). These episodes exemplify that pro-status-quo-policy groups indeed have an advantage over anti-status-quo-policy groups in terms of their organizational development. Given this, if I allowed P to invest in his organization as well as A, P would gain a more advantageous position than actual pro-status-quo-policy groups, which may contradict the empirical fact that pro-status-quo policy groups tend to achieve organizational development before anti-status-quo-policy groups. Hence, my model seeks to describe such a common type of dispute between civil groups, rather than interaction between civil groups competing for a newly emerging issue on which government has no stance. In line with these, my formal model could be interpreted to assume that P has already organized when the game begins.

The third assumption is that $α$ is common knowledge. It comes from the empirical observations that the foundation of interest groups is usually publicly revealed in democracies and that organized advocacy groups disclose information on their membership, offices, asset size and campaigns, and publicize their settlement reports and other information on their websites.

It should be noted that collective action problems are not assumed to occur for either side in this game. A theoretical rationale for this is elite leadership theory (e.g. Calvert, 1992), in which members of each group are assumed to participate in protests if their leaders stand up. Furthermore, formal results of this model could capture elements of collective action problems in interpreting the results, despite not explicitly formalizing.

Finally, I’s payoff structure needs justifying. Admittedly, the assumption that I, as an office-seeker, cares about support from A and P only is too simple, but not too unrealistic. Although A and P, in reality, seldom comprise the whole of the electorate in democratic systems, those not interested in a policy these two groups are bargaining over may not consider I’s reactions to the protests but, instead, emphasize other issues more in deciding whether to reelect I.

Equilibrium

To gain a better understanding of how domestic audiences’ heterogeneous preferences affect equilibrium outcomes, I first solve the baseline model assuming that the audiences are monolithically against the SQ (i.e. no P) and that A maximizes the disruptive influence of her protest activity without investing in organizational infrastructure ( $α = 0$ ). I then show the equilibrium of the entire game with the two protest groups in two steps. The solution concept employed is perfect Bayesian equilibrium. All proofs are in the Online Appendix.

Bargaining before an audience

The equilibrium in which every citizen prefers the executive concession is equivalent to a simple take-it-or-leave-it game between A and I.

Proposition 1

Suppose that P does not exist and $α = 0$ . A undertakes a protest and proposes her ideal policy if $1 \leq x_{A}$ ; otherwise, she does not undertake a protest. I accepts any $x \in (0, x_{A}]$ if $c \leq m_{A}$ and rejects otherwise.

Proposition 1 provides simple comparative statics concerning the effect of resource mobilization. The minimal amount of resources A needs to mobilize to make I accommodate increases as c becomes higher. This shows that investment in protest activity is always rational if the magnitude of A’s policy goal is worth the price of a protest. She does not stage a protest when her policy offer is sufficiently small. Since $m_{A}$ is a sunk cost, it does not affect A’s decision. In contrast, I’s decision depends on $m_{A}$ . As Figure 2 displays, A’s mobilization capacity ensures the equivalent probability of success under limited but fairly reasonable conditions.

Figure 2.

Equilibrium policy outcome in Proposition 1 if $x_{A} \geq 1$ .

Consider a hypothetical situation in which all the populace has an identical preference over a certain policy and no group opposes A’s effort to change the policy, say, an ethnically homogenous society fighting for secession. If a secession movement uses violent tactics (high $m_{A}$ ) to acquire complete independence from the state (large $x_{A}$ ), the government probably considers its movement as rational and sufficiently threatening. Meanwhile, if the movement threatens to resort to violence to acquire a relatively small magnitude of political goals such as government permission for the use of local languages in schools and offices, the government could deem the threat a bluff because the use of violence could not be payable. An anti-status-quo-policy group simply mobilizes the minimum amount of resources needed to make I reluctant to reject and therefore pursue her political goal.

Bargaining before two audiences: the escalation of conflict

I now solve the entire game introduced above, in which two protest groups with different political goals bargain with the political leader. In this subsection, I solve the baseline model assuming $α = 0$ : A expends all her resources on protest activity.

Proposition 2

Suppose that no resources are spent on organizational infrastructure. A never undertakes a protest if

0 \leq m_{A} \leq 4 (1 - \sqrt{m_{P}})

while she stages a protest if

4 (1 - \sqrt{m_{P}}) < m_{A} \leq 1

proposing

x^{*} = \frac{m_{A}}{2 (1 - \sqrt{m_{P}})}

P undertakes a counter-protest if $x^{*} > \sqrt{m_{P}}$ and does not otherwise. In the absence of a counter-protest, I accepts A’s offer if and only if $c \leq m_{A}$ and rejects otherwise. In the presence of a counter-protest, in contrast, I accepts A’s offer if and only if $c \leq m_{A} - x$ and rejects otherwise.

When a potential opposition exists, the equilibrium behaviour of A and I are conditioned by P’s behaviour. Figure 3 displays A’s equilibrium offer when staging a protest, P’s strategy and I’s strategy with sample parameters. The values of $m_{P}$ and c are fixed. The vertical axis means A’s offer (x), and the horizontal axis means A’s resource mobilization capacity ( $m_{A}$ ). The solid line represents A’s equilibrium offer (x*). The dotted horizontal line, meanwhile, represents P’s indifference line between staging and withholding a counter-protest. P stages a counter-protest if A’s offer is above this dotted line while he does not do so when A’s offer is on or below this. The other two dotted lines mean I’s strategy. In the absence of a counter-protest, I accepts x when $m_{A}$ is equal to or exceeds the vertical line while I rejects otherwise. In the presence of a counter-protest, in contrast, I accepts x when x is to the right side of the dotted positive line. The dotted positive line means I’s indifference line between accepting and rejecting in the presence of a counter-protest. Comparing these two lines for I’s strategy, it is obvious that I is less likely to accept A’s offer in the presence of a counter-protest ( $c \leq m_{A} - x$ ) than in its absence ( $c \leq m_{A}$ ). That is, P has the ability to offset the effect of A’s protest by placing pressure on I through counter-protests.

Figure 3.

(i) $m_{P} = 0.4$ (ii) $m_{P} = 0.7$ . A’s equilibrium offer and equilibrium strategies of P and I with sample parameters. $x^{CP}$ and $x^{NP}$ respectively denote the maximum offer that I can accept in the presence and the absence of a counter-protest. $α = 0$ , and $c = 0.2$ .

Figure 3 shows that a nonmonotonic relationship between $m_{A}$ and the probability of $x^{*}$ being accepted can exist. When $m_{A}$ is small, A’s protest does not provoke opposition from P but is not threatening enough for I to concede. Once $m_{A}$ passes the dotted vertical line, A’s protest becomes a sufficient threat to I. Since a counter-protest is still deterred in this range of $m_{A}$ and does not offset the effect of A’s threats, x will be accepted. When $m_{A}$ passes the horizontal line, however, a counter-protest always occurs. Because I finds it costlier to accommodate A’s demand in this circumstance, x will be rejected if $x > \sqrt{m_{P}}$ and if the other parameters take the value given in Figure 3.¹³ That is, I accommodates only the middle range of x, which is bold in Figure 3. Therefore, even though A is resourceful enough to coerce I to compromise in the absence of a counter-protest, she cannot always translate her maximum resource mobilization capacity into an equivalent level of threats to I in the presence of a counter-protest.

Let me consider A’s maximum possible offer that I accepts. By definition, A does not propose an offer that exceeds her ideal policy $x_{A}$ . The magnitude of $x_{A}$ is significant because it determines the amount of resources A should have and whether it facilitates opposition from P. Figure 4 shows how two different magnitudes of $x_{A}$ change bargaining outcomes. Each axis and line represents the same as before. The values of $m_{P}$ and c are fixed. When $x_{A}$ is lower than $\sqrt{m_{P}}$ (i.e. $x_{ALow}$ ), a counter-protest will not occur, and A’s capacity to mobilize resources can be minimal. In contrast, when A’s ideal policy is much further away from the status quo and higher than $\sqrt{m_{P}}$ (i.e. $x_{AHigh}$ ), she needs to mobilize much more resources. If she mobilizes the necessary amount of resources to demand $x_{AHigh}$ , it will provoke a counter-protest. Comparing Figures 2 and 4 shows that an increase in A’s resource mobilization capacity does not always enhance A’s chance of winning I’s concessions. In the one-audience game, A’s offer is always accepted if $c \leq m_{A} < x = x_{A}$ . As long as her policy goal is worth the price of protest, mobilizing more resources is consistently rational. This sharply contrasts with the result of the two-audience game, in which A’s ability to mobilize resources is not necessarily positively correlated with A’s maximum possible offer to be accepted. Even though A’s capacity to mobilize resources is enough to extract concessions in the one-audience situation, the presence of a rival group prevents it from translating its capacity into an equivalent level of bargaining power.

Figure 4.

Equilibrium strategies of P and I and different magnitudes of A’s ideal policy with sample parameters. $α = 0$ , $m_{P} = 0.4$ , and $c = 0$ .

Proposition 3

Suppose $α = 0$ . The probability of a counter-protest:

(i) is increasing in $m_{A}$ ; and

(ii) is decreasing in $m_{P}$ when m_P < ¼ and increasing in $m_{P}$ when m_P > ¼.

Proposition 3 shows comparative statics regarding the probability of a counter-protest. For larger $m_{A}$ , as Figure 3 shows, P is more likely to stage a counter-protest. This is intuitive because if P infers A has a larger amount of resources available for protest activity, P is more likely to fear that I might renege on her previous promise about implementing P’s ideal policy.

The effect of $m_{P}$ on the probability of a counter-protest is more indirect. When $m_{P}$ is sufficiently small, more resourceful P is less likely to stage a counter-protest. When $m_{P}$ passes ${\hat{m}}_{P}$ , meanwhile, more resourceful P is more likely to counter-mobilize, which Figure 3 visualizes. With an increase in $m_{P}$ , the intersection of

x^{*} = \frac{m_{A}}{2 (1 - \sqrt{m_{P}})} and x^{*} = \sqrt{m_{P}}

moves to the left, and the minimum $m_{A}$ which provokes a counter-protest lowers. That is, the stronger the opposition becomes, the greater the risk that an anti-status-quo-policy group will be exposed to counter-protests. As noted, the emergence of counter-protests not only has a negative effect on government willingness to concede to the anti-status-quo-policy group but also makes it more difficult for the anti-status-quo-policy group to achieve its policy aim.

Thus, A with larger $m_{A}$ is more likely to face problems. I tends not to seriously consider a threat sent by more resourceful A because protests by A with larger $m_{A}$ are more likely to fuel counter-protests that offset the impact of A’s threat at any rate. Accordingly, the occurrence of counter-protests has a negative effect on the probability of I’s concessions and the maximum offer that I accepts. This result is inconsistent with the conventional theory that protests mobilizing greater resources are more likely to be rewarded. When both sides of the protest groups have high resource mobilization capacity, resources will be inefficiently consumed.

The initial stage of antinuclear protests in West Germany illustrates the equilibrium when the investment of full resources in protest activity fuels counter-protests and undermines the protest group’s odds of winning government concessions. The antinuclear movement in Germany began in the Wyhl area and fled to grassroots protest groups in many other cities (Rucht, 1990: 204). Many antinuclear movements in the early and mid-1970s employed conventional means of protests. For instance, in the Brokdorf and Grohnde areas, the antinuclear groups mobilized mass demonstrations and even militant cooperation, but they ended in direct and violent battles with police (Karapin, 2007: 115–121). Thus, the antinuclear protest groups gained a good deal of resources (high $m_{A}$ ) and exploited them predominantly for protest activity rather than making their groups more institutionalized ( $α = 0$ ). Prior to the mid-1970s, the antinuclear protest groups in West Germany believed that maximizing the disruptive impact of their protests would lead to government concessions. As Figure 3 implies, in 1975 the antinuclear protests provoked pro-nuclear counter-protests led mainly by scientist and trade unions (Rucht, 1990: 204). This coevolution of both sides of protests was responded to with relatively low procedural impact.¹⁴ The construction of nuclear power plants was, on average, delayed for 6.1 months in 1974 and 13.8 in 1977 whereas in the American cases, which is discussed further below, plant construction faced delays almost three times longer despite the smaller mobilization of participants (Kitschelt, 1986: 80).

Moreover, several other studies lend empirical support to the formal result that counter-protests impede political changes. Andrews’s (1997, 2001) quantitative studies on the Mississippi civil rights movement show that white resistance had a significant negative impact on political change. McCright and Dunlap (2000) examine conservative movements against global warning issues in the US and contend that they successfully halted the government endorsement for the Kyoto Protocol in 1997.

Nonetheless, these equilibrium results call into question the considerable impact that civic advocacy groups with large resource pools can exert on the policymaking process. How can resourceful protest groups overcome the problem that their protests may provoke counter-protests?

Organizational diversification: effective deterrence of counter-protests

Comparing the following equilibrium, in which A allocates resources to organizational development ( $α > 0$ ), with Proposition 2, where A never invests in organizational development ( $α = 0$ ), demonstrates how investment in organizational infrastructure affects the impact of A’s protest on a policy change and the other players’ decisions even in a one-shot game involving a rival group.

Proposition 4

Suppose that $α > 0$ . A never undertakes a protest if $0 \leq m_{A} \leq 2 x^{*} (1 - \sqrt{m_{P}})$ , whereas she undertakes a protest paying

α^{*} = \frac{m_{A} - 2 x^{*} (1 - \sqrt{m_{P}})}{1 - p}

for organizational infrastructure if $2 x^{*} (1 - \sqrt{m_{P}}) < m_{A} \leq 1$ . In staging a protest, A offers

x^{*} = \frac{m_{A} - (1 - p) α}{2 (1 - \sqrt{m_{P}})}

P undertakes a counter-protest if $x^{*} > \sqrt{m_{P}}$ and does not otherwise. In the absence of a counter-protest, I accepts any offer if and only if $c \leq m_{A} - (1 - p) α$ and rejects otherwise. In the presence of a counter-protest, in contrast, I accepts A’s offer if and only if $c \leq m_{A} - (1 - p) α - x$ and rejects otherwise.

Proposition 5

Suppose $α > 0$ . The probability of a counter-protest is decreasing in $α$ .

Proposition 4 demonstrate the players’ best responses when $α > 0$ . Intuitively, A never organizes a protest when she is not resourceful enough ( $0 \leq m_{A} \leq 2 x^{*} (1 - \sqrt{m_{P}})$ ) whilst she stages a protest paying for building organizational infrastructure ( $α^{*} > 0$ ) when she has enough resources ( $2 x^{*} (1 - \sqrt{m_{P}}) < m_{A} \leq 1$ ). Figure 5 illustrates how different levels of $α$ affect players’ equilibrium behaviour. Each axis and line means the same as before. Figure 5 demonstrates that, all else being equal, an increase in $α$ moves all lines to the right. As a result, as the bold line in Figure 5 highlights, more resourceful A who could otherwise provoke counter-protests will be allowed to achieve her policy aim without provoking counter-protests.

Figure 5.

Players’ best responses with sample parameters. $m_{P} = 0.6$ , and $c = 0.2$ .

Consider why more resourceful A who otherwise fails to coerce I into compromise succeeds in doing so by spending $α > 0$ . Figure 6 helps us understand the twofold effect of $α$ . First, larger $α$ makes counter-protests less likely (Proposition 5), which is attributed to the impact of $α$ on I’s type because I’s type is determined by the magnitude of $α^{*}$ and $x^{*}$ relative to c. Larger $α$ makes the two thresholds in the figure move to the left and I more intransigent. Even though neither A nor P knows the exact value of c, they know how an increase in $α$ changes the expected probability that I concedes to A. Thus, A can manipulate P’s expectation about I’s willingness to accept A’s proposal by controlling the amount of $α$ ex ante and thereby control the probability of P staging a counter-protest. Through the change in P’s expectation about I’s behaviour, spending $α$ can benefit A though doing so seems at first glance irrational.¹⁵

Figure 6.

An illustration of I’s equilibrium behaviour.

Yet another effect of $α$ is that larger $α$ weakens A’s power to threaten I. The intersection of $x = \sqrt{m_{P}}$ and $x^{*}$ in Figure 5 moves to the right with an increase in $α$ . As a result of this move, the leftist range, in which I rejects A’s offer, expands. Figure 6 also demonstrates the same trend. That is, greater $α$ makes I more intransigent even though organizational development may become a future threat to I. Without this effect, nonetheless, investment in organizational infrastructure cannot have the first effect of reducing counter-protests. Thus, the effect of investing in organizational infrastructure is indirect in that such investment does not straightforwardly strengthen A’s ability to send threats.

This result suggests that there is a tradeoff between A’s ability to threaten I and A’s ability to deter a counter-protest. The greater resources A pays to establish her organizational infrastructure, the more likely she is to deter P’s interruption. Meanwhile, the larger the resources mobilized are, the lower the levels of her threats to I becomes. Put substantively, A needs to make a tradeoff decision between pursuing ideal policies and bargaining smoothly without interruption from the outside. Instead of simply strengthening the levels of protest, contra the conventional wisdom, anti-status-quo-policy groups need to be aware of this tradeoff.

Proposition 6

The amount of A’s equilibrium investment in organizational infrastructure is

(i) increasing in $m_{A}$ ;

(ii) increasing in $m_{P}$ ; and

(iii) decreasing in $x^{*}$ .

Comparative statics in Proposition 6 reveal the tradeoff that A faces more directly. If A is more resourceful, she needs to invest more in their organizational development (Proposition 6(i)) to deter counter-protests, but she needs to decrease the level of investment in organizational development if her policy goal is larger (Proposition 6(iii)) to keep her protests threatening enough. This result complements the findings in the previous section, namely a better ability to mobilize resources does not simply provide protest groups with greater bargaining power. That the effects of $m_{A}$ and $x_{A}$ on $α^{*}$ are the opposite makes A’s decision-making harder. Furthermore, Proposition 6(ii) states that larger m_P is more likely to facilitate A’s organizational development. Thus, the presence of a potentially stronger pro-status-quo group is more likely to promote the formation of an organized anti-status-quo group and anti-status-quo groups need to sacrifice their ideal policy goals. Thus, managing this tradeoff could be tricky. By deterring a counter-protest from a potentially strong opposition group, A needs to allocate more resources to organizational development but simultaneously runs a larger risk of failure. Yet given the results of the two-audience game without organizational development (Proposition 2), the prevention of potential opposition matters.

The formal results enable us to answer the second question: why does investing in organizational development have a positive effect on their bargaining leverage despite reducing the amount of resources available for protest activity? Proposition 2 demonstrates that making demands through protest activity without building an organizational foundation is not necessarily the best way to negotiate with I, given the diversified political preferences of the public, since doing so can trigger P’s interruption that negates A’s efforts to mobilize resources for protest. Particularly, this tendency is higher when A’s resource mobilization capacity is higher. A’s investment in her organizational development can remedy this problem since A will be able to bargain with I more smoothly without P’s opposition. It is worth noting that A still cannot resolve all the problems that she would face. One drawback with paying for the organizations is that it will restrict A’s ability to threaten I. If A is sufficiently resourceful, she might be able to achieve her policy goal, but the emergence of a counter-protest is inevitable, and the bargaining becomes harder. This tension between policy aim and bargaining smoothness has existed in many cases of domestic resistance.

The organizational evolution of US antinuclear movements exemplifies how investment in organizational infrastructure contributes to the deterrence of counter-protests and the improvement of the organized protest group’s bargaining power. Antinuclear protests prior to 1974 were fragmented, and the groups often conducted sit-ins in specific nuclear facilities and voiced their concerns mainly by participating in public hearings (Nelkin and Fallows, 1978: 279). In the mid-1970s, they changed from grassroots civic groups to regional or national groups (Nelkin and Fallows, 1978: 280–281) and changed how they allocated resources. Instead of allocating resources to protest activity in their initial stage ( $α = 0$ ), they began to distribute resources between protest activity and organizational development ( $α > 0$ ). With this change, they gradually started to bargain more directly about energy policies through means such as using courts to tackle issues of administrative policy and lobbying state and local legislatures. For example, the Western Bloc, an umbrella organization that antinuclear groups in 19 states founded towards the end of 1975 contributed to getting the issue on the ballot for referenda held in eight states. Although all the referenda were defeated, some state legislatures voted for tighter standards to review and license nuclear-power plants (Nelkin and Fallows, 1978: 282–284). As Proposition 3(i) implies, by the 1970s, the widespread organizational development of the antinuclear movement weakened pro-nuclear movements. Pro-nuclear forces started to shrink with the rise of the antinuclear lobbying groups in the mid-1970s. They were gradually decentralized, and no single organization tied all the movements together in their final stage (Useem and Zald, 1982: 153). This widespread organizational development had a substantial impact on the nuclear programmes. In 1974, the US experienced 20-month delays in the construction of nuclear power plants, compared with France (0.7) and West Germany (6.1) (Kitschelt, 1986: 80). Cancellation of new nuclear reactor orders increased after 1974, while the construction of only seven reactors was cancelled in 1972 and no cancellation was decided in the rest of the early 1970s. Despite the rapid increase in the number of orders,¹⁶ more than 10 plans for building reactors were stopped in the late 1970s and the early 1980s (Duffy, 1997: 175).

Discussion

Comparing the propositions so far offers new insights regarding the effects of resource mobilization. First, the amount of resources that protest groups possess is not the most important determinant of protest success. As Propositions 2 and 3 show, counter-protests can rarely be deterred when resourceful protest groups invest no resources in organizational infrastructure. What matters more is how protest groups allocate their resources between protest activity and organizational development. As Propositions 4 and 5 reveal, investment in organizational infrastructure helps protest groups deter potential counter-protests and increase the range of $m_{A}$ , which will not facilitate counter-protests. As a result, paying for organizational infrastructure benefits more resourceful protest groups that would otherwise trigger a rival’s interruption.

Second, the allocation of resources towards organizational development signals that the protest group will not undertake too assertive tactics. In equilibrium, A’s investment in organizational development decreases her maximum power to pressure I. When P recognizes that A has reduced the amount of her resources available for protest activity, it makes a counter-protest less likely; otherwise, P might find A’s protest potentially too threatening down the road and attempt to impede a policy change. In this way, the de-radicalization of dissident means can make protest groups more influential. Interestingly, investment in organizational development has an immediate effect on protest groups’ policy influence. Existing studies often highlight the result of organizational infrastructure in the long run, such as stronger leadership, better coordination between members, more stable financial resources from the membership base and tighter informal political ties (Berry, 1999; Boulding, 2010). In contrast, my model proves that the process of organizational development in and of itself plays a role.

Specifically, what makes the process of organizational development inherently special? How is investing in organizational development different from simply burning resources? First, the strategy of burning resources cannot apply to every type of resource. Destroying banknotes and facilities is possible but not destroying human resources. The option to consume resources for organizational capacity is thus plausible and consistent with norms. Second, simply burning resources would not have a lasting effect on protest groups’ ability to pose threats. As previously noted, organizing protest groups can benefit them on a long-term basis despite weakening their ability to threaten governments, and governments’ anxiety about protest groups’ future capability will improve their bargaining leverage today.

Third, and most importantly, consuming resources through organizational development communicates the reduction of resources to other actors in more credible and observable ways than other means. Democracies emphasize information disclosure and ensure formal procedures. Once a protest group has obtained corporate status, they must publicly disclose information about membership and organizational size and issue settlement reports. High observability backed by institutional setup enables protest groups to credibly show their commitment to the de-radicalization of their activity.

This implication provides a logical prediction that if the investment is made behind closed doors, it might not have such effects on potential opposition. Suppose that a protest group pays to make people stay home rather than joining counter-protests. It may reduce the number of potential protesters who will participate in counter-protests, but such a payment would not alleviate a competing group’s fear concerning the original protest group’s potential ability to force policy changes. The original protest group may not reveal that it has bought off potential protesters, and even if it does so, it would be difficult for a competing protest group to check all the individual payments made and accurately assess the remaining amount of resources the original protest group possesses. Another example is terrorist organizations. Lake (2002) and Abrahms (2008) indicate that terrorist groups often prioritize improving their organizational survival over accomplishing their political goals; however, the organizational development of terrorist groups is unlikely to operate in the same manner since it occurs stealthily. Thus, the formal findings dovetail with empirical observations that more organized protest groups tend to gain more bargaining leverage without resorting to radical means.^17,18

Furthermore, the results above provide additional empirical implications. First, we cannot measure the strength of protest groups merely by observing the size of resources mobilized for protest activity or by observing the magnitude of policy changes demanded. As the bold line in Figures 3 and 5 shows, protest groups with relatively intermediate resource mobilization capacity successfully moderate a counter-protest and achieve their policy goals whereas those possessing high capacity cannot necessarily do so because of a high probability of fuelling counter-protests. Rather, the amount of resources spent on founding non-transitory protest groups would be a better proxy to infer protest groups’ potential policy influence.

Second, protest tactics that include direct costs on political leaders, such as demonstrations and violent campaigns, give protesters only limited policy influence. Instead, protest groups’ ability to influence policies can be enhanced by using resources in such a way that it does not directly attack political leaders. This implication comports with the empirical findings by scholars like those of Stephan and Chenoweth (2008), Franklin (2009) and Chenoweth and Stephan (2011), namely that non-violent resistance is more effective than violent resistance. The mechanism by which radical tactics tend to backfire by hardening the stance of the populace is also consistent with recent studies on terrorism that call into question the conventional wisdom that violence allows civil resistance to achieve its policy aims (Abrahms, 2006, 2013; Fortna, 2015; Gould and Klor, 2010).

Although I highlight the importance of resource allocation to organizational development, my results could potentially be due to other causal explanations, like principal–agent issues. Protest groups that pay less in organizational infrastructure may be more likely to suffer from supervision of their participants, which may lead to violent clashes and lower success rate. Admittedly, the strength of leadership is sometimes correlated with the levels of organizational development. However, several cases show otherwise. For example, the People’s Alliance for Democracy (PAD), an unorganized antigovernment protest group in Thailand,¹⁹ started to protest peacefully in 2006, changing to more radical tactics in 2008 such as seizing an international airport and government buildings. After successfully overthrowing Prime Minister Thaksin, the PAD leaders had been training security guards who could serve as shock troops when they chose to take the offensive (Ockey, 2009: 322). Given that the number of PAD demonstrators had decreased dramatically in the two years (Ockey, 2009: 322–332), it should have been easier for its leaders to facilitate coordination among the members. This episode demonstrates that the leaders did consider using violence for their political goal, rather than proving that such radical means of protesting were attributed to the failure of the principal control. Hence, violent protests and the lower success rate of such protests cannot simply be attributed to principal–agent problems.

Despite not explicitly formalizing, this model could capture elements of collective action problems. The lower threshold of $m_{A}$ for A to have for sending sufficient threats can be understood as the minimum amount of resources that A must have to mobilize enough participants. Considering that providing positive selective incentives is asserted to be an effective way to achieve coordination (Olson, 1965), this interpretation is plausible. If sufficient resources should reduce the possibility of coordination failure, the equilibria discussed so far would not change critically.

Conclusions and extensions

This article suggests a framework to scrutinize the effects of resource mobilization on protest consequences. It does so by formalizing a bargaining game between a political leader and two competing protest groups and by endogenizing a protest group’s choice about its organizational form. The model proves that the greater mobilization of resources for protest activity does not consistently improve a protest group’s ability to achieve policy changes because it raises the probability of a counter-protest offsetting the impact of original protest. This result suggests that protest groups’ neglecting to consider the heterogenous preferences of domestic audiences can lead to inefficient mobilization of resources.

The model sheds light on the double-edged effects of investing resources in organizational infrastructure. On the one hand, investment in organizational infrastructure reduces the amount of resources available for pressuring activities and hence decreases protesters’ ability to carry costs on a political leader. On the other, investment in organizations decreases the probability of counter-protests by affecting a competing group’s expectation that the political leader is less likely to accept a policy change. By credibly showing that the amount of resources available for protest activity has been reduced, resourceful protest groups are more likely to translate much of their resource mobilization capacity into equivalent levels of threats towards the political leader. Thus, protest groups would face a tradeoff between the pursuit of ideal policies and smooth bargaining without rival groups’ intervention. If protest groups seek extensive policy changes, they need to mobilize a larger amount of their resources for protest activity to pose greater threats to political leaders, but it occasions rival groups’ interventions. Some protest groups might be better off by limiting the levels of protest activity and offering a compromise policy deal to government. A precise evaluation of the impact of organizing protest groups on political leaders’ decision-making would require further research, but differentiating the process from the product of organizational development conceptually and studying organizational evolution as commitment devices to choose de-radicalized dissent tactics could advance our understanding of the organizational diversification of civic advocacy groups.

Taken together, the findings challenge and are consistent with conventional wisdom: the more resources are mobilized, the more powerful the protest is. Better access to resources allows protest groups to afford to develop in an organizational sense and to commit to the de-radicalization of their activities, not because a larger resource pool allows protest groups to impose heavier costs on political leaders. Maintaining the central idea of resource mobilization theory, this article provides a new way of interpreting the formation of interest groups and bridges a gap between the literature on protest groups and on interest groups. While it is difficult to measure resources of organized groups and hence organizational development in any meaningful and comparable way (Berry, 1999: 147), future research should overcome this difficulty to pursue systematic, empirical evidence for the formal findings.

A way to produce more insightful models would be to highlight structural effects on the levels of transparency in the process of organizational development. I assume that the amount of resources paid for organizational infrastructure is known to every player, but communicating resource consumption in credible ways may not be easy for some types of groups or in some types of regimes. Incorporating characteristics of dissident groups and regimes might further clarify the complicated diversification of civic advocacy groups and the effects of political rules on the consequences of protests.

Supplemental Material

190506_online_appendix – Supplemental material for Protest, counter-protest and organizational diversification of protest groups

Supplemental material, 190506_online_appendix for Protest, counter-protest and organizational diversification of protest groups by Kana Inata in Conflict Management and Peace Science

Footnotes

Acknowledgements

I would like to thank Dominik Duell, Gina Yannintell Reinhardt, Jonathan Slapin, Hugh Ward, Daniel Berger, Masaru Kohno, Shuhei Kurizaki and participants at seminars at University of Essex, Waseda University and the 2017 EPSA Annual Convention for invaluable comments and discussion.

Funding

The author disclosed receipt of the following financial support for the research, authorship and/or publication of this article: I gratefully acknowledge the financial support of the Grant-in-Aid for Young Scientists (no. JP19K13605) from Japan Society for Promotion of Science.

ORCID iD

Kana Inata

Supplemental material

All data, replication materials, and instructions regarding analytical materials upon which published claims rely are available online through the SAGE CMPS website:

Notes

References

Abrahms

(2006) Why terrorism does not work. International Security 31(2): 42–78.

Abrahms

(2008) What terrorists really want: Terrorist motives and counterterrorism strategy. International Security 32(4): 78–105.

Abrahms

(2013) The credibility paradox: Violence as a double-edged sword in international politics. International Studies Quarterly 57(4): 660–671.

Andrews

(1997) The impacts of social movements on the political process: The civil rights movement and black electoral politics in Mississippi. American Sociological Review 62(5): 800–819.

Andrews

(2001) Social movements and policy implementation: The Mississippi civil rights movement and the war on poverty, 1965 to 1971. American Sociological Review 66(1): 71–95.

Andrews

(2002) Movement–countermovement dynamics and the emergence of new institutions: The case of ‘white flight’ schools in Mississippi. Social Force 80(3): 911–936.

Andrews

Edwards

(2004) Advocacy organizations in the U.S. political process. Annual Review of Sociology 30: 479–506.

Banks

(1990) Equilibrium behavior in crisis bargaining games. American Journal of Political Science 34(3): 599–614.

Bell

Bhasin

Clay

Murdie

(2014) Taking the fight to them: Neighborhood human rights organizations and domestic protest. British Journal of Political Science 44(4): 853–875.

10.

Benclowcz

Breña

(2011) Continuities, scope, and limitations of the Argentine Piquetero movement: The cases of Tartagal and Mosconi. Latin American Perspectives 38(1): 74–87.

11.

Berry

(1999) The New Liberalism: The Rising Power of Citizen Groups. Washington, DC: Brookings Institution Press.

12.

Boulding

(2010) NGOs and political participation in weak democracies: Subnational evidence on protest and voter turnout from Bolivia. Journal of Politics 72(2): 456–468.

13.

Calvert

(1992) Leadership and its basis in problems of social coordination. International Political Science Review 13(1): 7–24.

14.

Chenoweth

Stephan

(2011) Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict. New York: Columbia University Press.

15.

Colby

(1982) A test of the relative efficacy of political tactics. American Journal of Political Science 26(4): 741–753.

16.

DeNardo

(1985) Power in Numbers: The Political Strategy of Protest and Rebellion. Princeton, NJ: Princeton University Press.

17.

Duffy

(1997) Nuclear Politics in America: A History and Theory of Government Regulation. Kansas: University Press of Kansas.

18.

Fearon

(1997) Signaling foreign policy interests tying hands versus sinking costs. Journal of Conflict Resolution 41(1): 68–90.

19.

Fortna

(2015) Do terrorists win? Rebels’ use of terrorism and civil war outcomes. International Organization 69(3): 519–556.

20.

Franklin

(2009) Contentious challenges and government responses in Latin America. Political Research Quarterly 62(4): 700–714.

21.

Gamson

(1975) The Strategy of Protest. Homewood: The Dorsey Press.

22.

Gavious

Mizrahi

(1999) Two-level collective action and group identity. Journal of Theoretical Politics 11(4): 497–517.

23.

Giugni

(2004) Protest and Policy Change: Ecology, Antinuclear, and Peace Movements in Comparative Perspective. Lanham, MD: Rowman & Littlefield.

24.

Gould

Klor

(2010) Does terrorism work? Quarterly Journal of Economics 125(4): 1459–1510.

25.

Grossman

Helpman

(1994) Protection for sale. American Economic Review 84(4): 833–850.

26.

Karapin

(2007) Protest Politics in Germany: Movements on the Left and Right since the 1960s. University Park, PA: Pennsylvania State University Press.

27.

Karklins

Petersen

(1993) Decision calculus of protesters and regimes: Eastern Europe 1989. Journal of Politics 55(3): 588–614.

28.

Kitschelt

(1986) Political opportunity structures and political protest: Anti-nuclear movements in four democracies. British Journal of Political Science 16(1): 57–85.

29.

Kolb

(2007) Protest and Opportunities: The Political Outcomes of Social Movements. Frankfurt: Campus.

30.

Lake

(2002) Rational extremism: Understanding terrorism in the twenty-first century. Dialogue IO 1(1): 15–28.

31.

Langguth

(1984) The Green Factor in German Politics: From Protest Movement to Political Party. Boulder, CO: Westview Press.

32.

Lindvall

(2010) Pressure group strength and the politics of protests. Working paper.

33.

Lipsky

(1968) Protest as a political resource. American Political Science Review 62(4): 1144–1158.

34.

Lounsbury

(2005) Institutional variation in the evolution of social movements: Competing logics and the spread of recycling advocacy groups. In: Davis

McAdam

Scott

Zald

(eds) Social Movements and Organization Theory. New York: Cambridge University Press, pp. 73–95.

35.

McCarthy

Zald

(1977) Resource mobilization and social movements: A partial theory. American Journal of Sociology 82(6): 1212–1241.

36.

McCright

Dunlap

(2000) Challenging global warming as a social problem: An analysis of the conservative movement’s counter-claims. Social Problems 47(4): 499–522.

37.

Meyer

Staggenborg

(1996) Movements, countermovements, and the structure of political opportunity. American Journal of Sociology 101(6): 1628–1660.

38.

Nelkin

Fallows

(1978) The evolution of the nuclear debate: The role of public participation. Annual Review of Energy 3(1): 275–312.

39.

Nie

Verba

(1975) Political participation. In: Greenstein

Polsby

(eds) Handbook of Political Science Vol. 4: Nongovernmental Politics. Reading, MA: Addison-Wesley, pp. 1–74.

40.

Ockey

(2009) Thailand in 2008: Democracy and street politics. Southeast Asian Affairs315–333.

41.

Olson

(1965) The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press.

42.

Rousselin

(2014) Modern communication technologies and the extension of the territory of struggle: Conceptualising Tunisia’s Jasmine Revolution. New Media and Society 8(7): 1201–1218.

43.

Rucht

(1990) Campaigns, skirmishes and battles: Anti-nuclear movements in the USA, France and West Germany. Industrial Crisis Quarterly 4(3): 193–222.

44.

Scartascini

Tommasi

(2012) The making of policy: Institutionalized or not? American Journal of Political Science 56(4): 787–801.

45.

Schelling

(1960) The Strategy of Conflict. Cambridge, MA: Harvard University Press.

46.

Schumaker

(1975) Policy responsiveness to protest-group demands. Journal of Politics 37(2): 488–521.

47.

Scruggs

(1999) Institutions and environmental performance in seventeen Western democracies. British Journal of Political Science 29(1): 1–31.

48.

Slantchev

(2005) Military coercion in interstate crises. American Political Science Review 99(4): 533–547.

49.

Snyder

Borghard

(2011) The cost of empty threats: A penny, not a pound. American Political Science Review 105(3): 437–456.

50.

Stephan

Chenoweth

(2008) Why civil resistance works: The strategic logic of nonviolent conflict. International Security 33(1): 7–44.

51.

Taft

Ross

(1969) American labor violence: Its causes, character, and outcomes. In: Graham

Gurr

(eds) Violence in America: Historical and Comparative Perspectives. New York: Bantam, pp. 221–301.

52.

Tenorio

(2014) Social spending responses to organized labor and mass protests in Latin America, 1970–2007. Comparative Political Studies 47(4): 1945–1972.

53.

Tilly

Tarrow

(2015) Contentious Politics. New York: Oxford University Press.

54.

Tullock

(1980) Efficient rent-seeking. In: Buchanan

Tollison

Tullock

(eds) Toward a Theory of the Rent-seeking Society. College Station, TX: Texas A&M University Press, pp. 97–112.

55.

Useem

Zald

(1982) From pressure group to social movement: Organizational dilemmas of the effort to promote nuclear power. Social Problems 30(2): 144–156.

56.

Ward

(2004) Pressure politics: A game-theoretical investigation of lobbying and the measurement of power. Journal of Theoretical Politics 16(1): 31–52.

57.

Zald

Useem

(1987) Movement and countermovement interaction: Mobilization, tactics, and state involvement. In: Zald

McCarthy

(eds) Social Movements in an Organizational Society. New Brunswick, NJ: Transaction, pp. 245–272.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

Protest,counter-protest and organizational diversification of protest groups

Abstract

Keywords

Introduction

Literature

The model

Equilibrium

Bargaining before an audience

Proposition 1

Bargaining before two audiences: the escalation of conflict

Proposition 2

Proposition 3

Organizational diversification: effective deterrence of counter-protests

Proposition 4

Proposition 5

Proposition 6

Discussion

Conclusions and extensions

Supplemental Material

190506_online_appendix – Supplemental material for Protest, counter-protest and organizational diversification of protest groups

Footnotes

Acknowledgements

Funding

ORCID iD

Supplemental material

Notes

References

Supplementary Material