# On Two Sides of the Smoke Screen: How Activist Organizations and Corporations Use Protests, Campaign Contributions, and Lobbyists to Influence Institutional Change

ISBN: 978-1-78754-350-8, eISBN: 978-1-78754-349-2

ISSN: 0733-558X

Publication date: 6 August 2018

## Abstract

We explore the simultaneous influence of activist organizations and corporations on institutional change. Focusing on protests, campaign contributions, and lobbyists as the strategies used by activist organizations and corporations to influence institutional change, we study the dynamics between movements and counter-movements and their influence on the probability of institutional change. In the context of the US tobacco industry, the results shed light on the effectiveness of these strategies and uncover potential moderators of this relationship. Overall, we demonstrate the simultaneous and asymmetric effects of activist organizations and corporations that use conspicuous and inconspicuous strategies to change institutions.

## Keywords

#### Citation

Aranda, A. and Simons, T. (2018), "On Two Sides of the Smoke Screen: How Activist Organizations and Corporations Use Protests, Campaign Contributions, and Lobbyists to Influence Institutional Change", Briscoe, F., King, B. and Leitzinger, J. (Ed.) Social Movements, Stakeholders and Non-Market Strategy (Research in the Sociology of Organizations, Vol. 56), Emerald Publishing Limited, pp. 261-315. https://doi.org/10.1108/S0733-558X20180000056011

### Publisher

:

Emerald Publishing Limited

## Introduction

Decades after the 1964 US Surgeon General concluded that smoking causes lung cancer and other diseases, and after the accumulation of scientific evidence on the health risks and costs associated with smoking, tobacco remains the leading cause of preventable death in the United States (McKinsey Global Institute (MGI), 2014). Although legislation has been implemented to control the epidemic of tobacco-related diseases, the history of tobacco control has been a continuous struggle between anti-smoking groups focused on the hazards created by smoking on the one hand and tobacco companies focused on the economic benefits associated with tobacco production and consumption on the other. This struggle has been fueled by the apparent success of the anti-smoking groups, whose actions increased the social unacceptability of smoking and fostered the enactment of laws to reduce cigarette consumption by highlighting the harmful effects of smoking. The anti-smoking movement, thus, has threatened tobacco companies and has sparked the emergence of the smokers’ rights movement, a group aimed at resisting the enactment of smoking bans by representing smokers and others whose interests have been threatened by the success of anti-smoking activists (Meyer & Staggenborg, 1996; Zald & McCarthy, 1987). Two sides characterize the struggle between these actors: “the projection of honesty and trustworthiness for the tobacco industry, countered by public health’s master frame of distrust of the industry” (Derry & Waikar, 2008, p. 102). In this contested industry, where anti-smoking groups and tobacco companies interact and compete for support, we seek to understand how different actors use various strategies to shape the enactment of tobacco control regulations.

We build on research in the intersection between social movement theory, institutional theory, and non-market strategy. Traditionally, these streams of research have focused either on the broad impact of social movements on institutions (Amenta & Caren, 2004; Andrews, 2002; King & Pearce, 2010; Olzak & Soule, 2009; Schneiberg & Lounsbury, 2008) or on the influence of social movements on corporations (de Bakker & den Hond, 2008; McDonnell & King, 2013). However, to the best of our knowledge, the simultaneous influence of social movements (hereafter activist organizations) and corporations on institutions remains understudied (Vogus & Davis, 2005). In fact, as Edelman, Leachman, and McAdam (2010) posit, we know relatively little about the interplay among corporations, activist organizations, and institutions, whereas we know more about each of these dyadic relationships. Considering this interplay is of importance as it has been recognized that institutional change is rarely achieved without contests between actors that have divergent goals (O’Mahony & Bechky, 2008) and that are differently affected by such change (Kim, Shin, Oh, & Jeong, 2007). Hence, we examine the actions of activist organizations to promote their interests, mobilize public opinion, and obtain political support, while simultaneously explore the actions of corporations that counter-mobilize to promote their own, and often times contrary, interests (Weber & King, 2013). Specifically, we aim to uncover the role of activist organizations and of corporations that counter-mobilize as producers of institutional change (Schneiberg & Lounsbury, 2008).

In this study, we are interested in contributing to research on how institutional change comes about. To do so, we consider how activist organizations and corporations bring about changes in the regulative pillar of institutions, which is the most common target of their actions (Bosi & Uba, 2009; Weber &King, 2013). That is, we understand institutional change as the enactment of new laws. Therefore, we recognize that activist organizations and corporations contribute to institutional change by promoting laws that further their interests or by obstructing laws that challenge or disregard their interests. Specifically, we argue that contests between activist organizations and corporations result from innate differences in their respective objectives that lead activists to recognize the need to promote laws which embed their interests over the interests promoted by corporations, and vice versa (de Bakker, den Hond, King, & Weber, 2013). For instance, in November of 2002 Delaware became the first state to enact a comprehensive smoking ban (i.e., a ban in restaurants, bars, and workplaces). IMPACT – a social movement of tobacco control and smoking prevention organizations – successfully organized to encourage the passing of the ban by gathering public support with the message: “Second-hand smoke causes diseases and death!” (Boyer & Ratledge, 2009). The tobacco industry fiercely opposed the passage of the ban on the grounds of its possible negative economic consequences, yet the ban was enacted. Thus, we study contests between activist organizations and corporations as central precursors of an industry’s particular regulative structure.

We study the US tobacco industry from 2000 to 2012. Although scarcely studied (Hsu & Grodal, 2015; Simons, Vermeulen, & Knoben, 2016), the tobacco industry provides a suitable empirical setting for this study for two reasons. First, mobilization in the tobacco industry has involved two sides: anti-smoking groups that aim to advance the enactment of tobacco control regulations, on the one hand, and tobacco companies that aim to prevent regulations, on the other hand (Pertschuk, 2001). Moreover, as smoking is a disputed social issue, actors on each side of the smoking debate play a central role in shaping societal perceptions and attitudes toward smoking, and ultimately, in influencing tobacco control regulations (Derthick, 2002). Second, the dynamics occurring in the tobacco industry shed light on the pursuit of institutional change through conspicuous (e.g., protests) versus inconspicuous (e.g., campaign contributions and lobbyists) strategies. Specifically, the tobacco industry is known for spending considerable efforts and for investing substantial resources on campaign contributions and lobbyists to undermine the enactment of tobacco control regulations (CRP, 2013); while, at the same time, both anti-smoking organizations and tobacco companies – through Astroturf groups 1 – have used protests in order to either encourage or promote the enactment of tobacco control regulations, or to oppose or thwart the enactment of such regulations (Derry & Waikar, 2008; Nathanson, 1999). Thus, this industry not only provides an opportunity to examine the effect of simultaneous conspicuous and inconspicuous means to influence institutional change but also represents a particularly relevant empirical setting to study the unexplored dynamics underlying the interaction between movements and counter-movements and their impact on institutional change.

By answering the question of how activist organizations and corporations change institutions our contribution to the literature is threefold. First, we respond to the call for research on identifying the causal influences of activist organizations and corporations on institutions and shed light on the differential effects that these actors have on institutional change (Walker & Rea, 2014). Specifically, studying the interplay between activists that support and corporations that counter-mobilize to oppose institutional change is of interest as the importance of examining the “countervailing effects” of movements and counter-movements has been recognized (Soule & King, 2006), yet understudied (Vogus & Davis, 2005). Second, we respond to the call for research on “how lobbying and business collective actions shapes government” (Walker &Rea, 2014, p. 35), by uncovering the effects of the strategies used by activist organizations and corporations to shape laws. Finally, we provide a longitudinal conceptualization of the process of institutional change at the industry level, which is pertinent because contests between activist organizations and corporations often take place within industries (Hiatt, Sine, & Tolbert, 2009). Overall, this study presents a comparison between movements and counter-movements, and between their different strategies in an effort to contribute to the understanding of the process of institutional change.

## Theory and Hypotheses

Our work differs from prior research by focusing on institutions as targets of the simultaneous actions of both activist organizations and corporations. Previous research has found that activist organizations have been successful in their attempts to change institutions (Bosi & Uba, 2009; Earl, 2004; Giugni, 1998; Guérard, Bode, & Gustafsson, 2013; Hiatt et al., 2009; McCammon, Campbell, Granberg, & Mowery, 2001; McVeigh, Myers, & Sikkink, 2004; Schneiberg & Soule, 2005; Soule & Olzak, 2004) and has made some progress toward understanding the influence of activist organizations on institutions (Amenta, Caren, Chiarello, & Su, 2010). Comparatively, there is much less research on the influence of corporations on changing institutions (Hillman, Keim, & Schuler, 2004), a notable exception being the work of Davis (1991) on the adoption of the poison pill as a takeover defense. Within the organizations and management literature, most research on this topic has taken a non-market strategy perspective (Walker & Rea, 2014). Work originating from this stream of research has discussed how corporations influence institutional change by implementing non-market strategies (de Figueiredo, 2009; Keim & Zeithaml, 1986), has uncovered the factors that predict the use of these strategies (Boies, 1989; Grier, 1994; Meznar & Nigh, 1995; Schuler, Rehbein, & Cramer, 2002), and has identified their antecedents and performance consequences (Lux, Crook, & Woehr, 2010). Although prior studies have contributed to our understanding of how corporations influence institutions (Bonardi, Hillman, & Keim, 2005; Davis & Thompson, 1994; Hillman, 2003), few empirical tests of non-market strategies are available (Soule, 2012).

With respect to research that examines the actions of both activist organizations and corporations, the bulk of research causally linking the strategies used by these actors to institutional change has mainly been from the political science and sociology literatures (Baumgartner & Leech, 2001; Baumgartner, Berry, Hojnacki, Kimball, & Leech, 2009, 2014; Hojnackia, Kimball, Baumgartner, Berry, & Leech, 2012; Hojnackia et al., 2015; Kimball et al., 2012; Leech, Baumgartner, La Pira, & Semanko, 2005). Previous work has examined the different tactics of organized interest groups and social movements (Gilens & Page, 2014), and the outcomes of their efforts (Hojnacki et al., 2012). However, within this line of work, the simultaneous influence of different actors using several strategies to influence institutional change remains to be explored. Therefore, this study follows the suggestion by Hojnacki et al. (2012) and differs from previous work by specifying the direct interconnectedness of actors and strategies, and by studying the strategies of actors to achieve their goals, not in isolation but, with special attention to context.

### Hypotheses

We identify two central actors in the debates over institutional change: activist organizations and corporations. As the struggle “between the public interest and corporate interests is perhaps nowhere better illustrated than in the decades-long struggle between the tobacco industry and advocates for public health” (Pertschuk, 2001), we follow prior research that has recognized two main sides in the tobacco control debate (Derry & Waikar, 2008; Nathanson, 1999). On the one hand, activist organizations represented by “anti-smoking groups” (e.g., coalitions on smoking or health, groups against smoking pollution, tobacco control and prevention organizations) that advocate for restrictions on smoking or that encourage the enactment of tobacco control regulations. On the other hand, corporations that represent the counter-movement and oppose the enactment of tobacco control regulations. Although tobacco companies play a leading role in organizing efforts and funding campaigns to oppose tobacco control regulations (Blanke & da Costa e Silva, 2004), the counter-movement is a broad group that features tobacco companies, smoker’s rights groups, and other associations (e.g., tobacco growers). Hence, we use the label “Big Tobacco” to broadly represent the counter-movement. By identifying the struggle over tobacco control regulations as being between “anti-smoking groups” (i.e., tobacco control organizations or groups that support the enactment of bans) and “Big Tobacco” (i.e., organizations or advocates representing the tobacco industry who fight the enactment of laws), the two sides that comprise of the “tobacco wars” are comprehensively represented (Pertschuk, 2001). This classification is also in line with the characterization of the media, where the struggle has been portrayed as being between anti-smoking groups on one side, and the tobacco industry – which has been commonly labeled “Big Tobacco” – on the other (Menashe & Siegel, 1998). 2 We model the interaction between activist organizations and corporations that counter-mobilize taking into account that these actors use various strategies to change institutions. Scholars have identified the most prominent strategies used by contesting actors to be conspicuous (e.g., protests) and inconspicuous (e.g., campaign contributions and lobbying) strategies (Andrews & Caren, 2010; Soule, 2009; Walker & Rea, 2014).

### Protests

Following prior research, protests are defined as manifestations or demonstrations by which actors voice their claims, convey their aims, and attract support to disrupt or maintain the status quo by promoting or resisting institutional change (Earl, Soule, & McCarthy, 2003; King & Soule, 2007). Research has traditionally considered the regulative pillar of institutions to be the central target of protests (Walker, Martin, & McCarthy, 2008). 3 Specifically, prior work (Luders, 2006; McAdam & Su, 2002; Olzak & Soule, 2009) has found that protests have a role in changing the regulative pillar of institutions by making an issue more salient, by generating public support for the issue, and by putting it on the political agenda (Snow, Della Porta, Klandermans, & McAdam, 2013). Therefore, in this study, we categorized protests as a conspicuous constituency building strategy (Hillman & Hitt, 1999).

In line with the tenants of political mediation theory, the ability of protests by activist organizations to achieve their goals comes not only from their visibility, but also from the public support these organizations receive and are able to mobilize. We posit that by being visible actions, protests by activist organizations have a significant potential to bring about institutional change, understood as changes in the regulative pillar of institutions. Protests are visible in that they allow actors to voice their claims, interests, and goals publicly, since they are not only performed on the streets but also covered by mass media. The visibility of protests means that they appeal simultaneously to policymakers and to the general public (King & Soule, 2007; Olzak & Soule, 2009) and that they are a source of information for both audiences (Ingram, Yue, & Rao, 2010). In the context of this study, for instance, anti-smoking groups have used protests to mobilize and receive support to achieve legislative change by making visible the importance of regulating tobacco use. That is, these groups have used protests as a mechanism to communicate their health message and to disrupt the status quo by altering the perception of smoking. By doing so, anti-smoking groups have been essential in the formation and change of tobacco control regulations by developing, constructing, and reinforcing a set of expectations about smoking. Thus, protests constitute a visible strategy used by activist organizations to achieve legislative change by voicing and articulating societal views about a certain issue (McAdam & Su, 2002) and by enlisting the support of policymakers (Snow et al., 2013). Therefore, we expect as a starting premise that protests are impactful in promoting the activist organizations’ agenda and hypothesize:

H1a. Protests by activist organizations increase the probability of institutional change.

Although prior research has identified that protests by activist organizations foster and inspire corporate counter mobilization (Jasper & Poulsen, 1993), there are very few studies that explicitly consider the effects of the protests by movements and counter-movements (Rao, 2004; Schneiberg, King, & Smith, 2008). Some current work has posited that corporations that are highly scrutinized, and are large and visible, are more likely to counter-mobilize (de Bakker & den Hond, 2008). In line with these efforts, we posit that although both activist organizations and corporations use protests to promote the visibility of their claims and to win supporters who would mobilize for their cause, there are two fundamental differences between the protests of activist organizations and those of corporations. First, whereas activist organizations protest in an attempt to disrupt the status quo, corporations in contested industries protest to maintain the status quo. Therefore, the protests of corporations are more likely to occur when the issue at hand is already on the political agenda but before the law is enacted, mainly because once the law has been passed it is more difficult for corporations to influence policymakers to alter or overturn it (Amenta, Caren, & Olasky, 2005). Second, corporations in contested industries need to devise protests that offset the constant threats to their legitimacy by, for instance, allying with groups whose claims are deemed as deserving of recognition and attention, and that are successful in attracting and mobilizing public support.

In the context of study, protests by anti-smoking groups have encouraged the counter-organization and resistance of Big Tobacco. At the same time, however, protests by anti-smoking groups have put tobacco companies in a vulnerable position as they have pressed the industry to keep a low profile and to minimize its direct involvement in pro-tobacco protests. The ability of anti-smoking groups to invalidate Big Tobacco’s protests by highlighting the dangers associated with smoking has been crucial in forcing Big Tobacco to use Astroturf groups to deliver its messages (Derry & Waikar, 2008; Nathanson, 1999). These Astroturf groups mitigate the inherent lack of consonance between Big Tobacco’s interests and the public interest:

because opposition is most credible if it appears to come from independent sources, tobacco companies prefer to work through surrogates. They do this by mobilizing others whose interests are aligned with their own and by creating or co-opting “front groups.” (Blanke & da Costa e Silva, 2004, p. 135)

For instance, Big Tobacco has used a myriad of concealed counter-mobilization efforts to stop the enactment of laws: from funding think thanks such as the Tobacco Institute to funding advocacy groups like the National Smokers Alliance or engaging in mobilization by, for example, facilitating own employees’ participation in protests against tobacco control regulations or supporting bars that resist smoking bans (Samuels & Glantz, 1991; Sweda &Daynard, 1996). The Astroturf efforts conceal Big Tobacco’s involvement in the movement against smoking bans in an effort to present their claims as appropriate and to increase their pool of supporters (Cardador, 1995; Smith &Malone, 2007).

In view of the above, the occurrence of protests by activist organizations represents a challenge for corporations, as their protests indirectly affect the ability of corporations to gain and mobilize supporters (McDonnell & Werner, 2016). Therefore, we posit that protests by activist organizations compel corporations to counter-mobilize in an effort to maintain the status quo; however, corporations in a contested industry are obliged to do so by mobilizing Astroturf groups on their behalf instead of engaging in it themselves. Nevertheless, protests, which advance the corporations’ positions, are a visible strategy that targets policymakers in an attempt to maintain the status quo through mobilizing constituent support. Thus, analogous to the prediction concerning activist organizations, as a baseline expectation, we posit that:

H1b. Protests by Astroturf groups that represent corporations and aim to maintain the status quo decrease the probability of institutional change.

As mentioned before, protests involve interactions between movements and counter-movements around different claims (Earl, 2004) and constitute a strategy used to change institutions (Snow et al., 2013). Resource mobilization theory suggests that the efficiency of protests as a strategy to change institutions depends on the potential of the movement to gather and mobilize support (Amenta et al., 2005). However, in the case where both the social movement and the counter-movement use protests as a strategy to achieve institutional change, given that resources are limited, we expect the overall efficiency of protests to be diminished as the visibility and impact of the protests by each side lessen. In the context of this study, for example, anti-smoking groups emerged along with the perceived opportunity to enact laws given the assessment of smoking as a hazard to health (Nathanson, 1999). As anti-smoking groups have made their message more visible and have gained public support, they have hindered the ability of Big Tobacco to deter the advancement of laws that threaten the status quo and have obliged Big Tobacco to use Astroturf groups to protest on their behalf. Simultaneously, though, the concealed participation of Big Tobacco in protests by relying on societal members whose claims are more in line with those of the general public (i.e., employees) or who appear to be more legitimate in the eyes of society members (i.e., scientists) has helped Big Tobacco not only in its efforts to maintain the status quo but also in visibly questioning the validity of the claims made by anti-smoking groups. Hence, when both sides of the smoking debate protest, their individual ability to gain public support and to foster further mobilization diminishes due to the reduced visibility of each protest, and to the mixed claims conveyed by the protestors. In other words, when the protests of the social movement make an issue visible, they result in protests on the part of the counter-movement that confront their message, which lowers the effectiveness of the single protest, and vice versa. Thus, we formulate:

H2. The efficiency of protests by the social movement to influence institutional change is negatively moderated by the counter-movement’s protests.

### Campaign Contributions

Campaign contributions are an inconspicuous financial strategy linked to individual interests because through contributions actors articulate their demands in a more concealed or hidden form compared to protests. Campaign contributions have been identified as a financial strategy that targets policymakers through monetary incentives (de Figueiredo & Edwards, 2015). Therefore, we study campaign contributions as a strategy by which agents offer monetary incentives in an attempt to persuade policymakers to propose or (not) support a given policy (Baron, 2006). Research on the use of campaign contributions has looked at how corporations use financial incentives to influence legislators in diverse settings such as trade policy (Schuler, 1996), electric utilities (Bonardi, Holburn, & Vanden Bergh, 2006), and universities (de Figueiredo &Silverman, 2006). Despite its importance, little attention has been paid to contributions by Big Tobacco, whereas Big Tobacco is known to expend substantial sums in offering monetary incentives to influence legislators.

By using campaign contributions, activist organizations and corporations engage directly with policymakers in a clear effort to shape laws that sustain and advance their interests (Keim & Zeithaml, 1986). Campaign contributions by activist organizations are aimed at disrupting the status quo and contribute to the likelihood of institutional change by performing several functions. First, campaign contributions provide activist organizations access to the policymaking process: “at the very least, contributors are more likely to have their phone calls returned” (Hall & Deardorff, 2006, p. 80). Thus, monetary resources are expected to buy attention from legislators, which gives access to the policymaking process. Second, campaign contributions align the incentives of legislators with those of activist organizations. This is because contributions can provide useful support to legislators on matters that they care about, and by making a contribution, activist organizations assert their own interests on the same matter (Hillman & Hitt, 1999). Third, campaign contributions encourage policymakers to prioritize the claims of activist organizations over other antagonistic claims by recruiting the support of the policymaker (Baron, 2006). Thus, campaign contributions of anti-smoking groups have the potential to disrupt the status quo in order to influence the enactment of bans through financial incentives that persuade policymakers to support their claims. In other words, campaign contributions provide monetary incentives to policymakers to adopt the preferences or the claims of the anti-smoking groups. Based on the above, we expect as a starting premise that the more anti-smoking groups use campaign contributions, the more likely it is that legislators are sensitized to their claims and hypothesize:

H3a. Campaign contributions by activist organizations increase the probability of institutional change.

Resource mobilization theory suggests that actors that have more resources at their disposal are those that are better able to achieve their goals and succeed (McCarthy & Zald, 1977). This, combined with the idea that resource imbalances affect an actor’s potential to influence institutional change suggests that during an episode of contention, the actor that controls the most resources is more likely to affect institutional change in its favor (Baumgartner et al., 2009). Thus, even in the face of tireless pressure for change from activist organizations, corporations are able to maintain the status quo by continuously making substantial campaign contributions. The strong interest of corporations in undermining institutional change is primarily financial. In the case of the tobacco industry, the significant resources that Big Tobacco expends in campaign contributions reflect its concern with the negative impact of tobacco control regulations on smoking rates, and thus, with the potential reduction in its profits as a consequence of a decline in the usage of tobacco products (Smith, Savell, & Gilmore, 2013). Hence, Big Tobacco remains able to secure (some of) its influence in the policymaking process by using its knowledge and experience in influencing public policy, and its substantial financial resources (Jacobson, Wasserman, & Anderson, 1997). Therefore, we posit that corporations attempt to maintain the status quo via significant campaign contributions against the enactment of laws that negatively affect their business. Analogous to the prediction concerning activist organizations, as a baseline expectation, we predict:

H3b. Campaign contributions by corporations that aim to maintain the status quo decrease the probability of institutional change.

Traditionally, movements and counter-movements are more likely to contribute to legislators that support their individual interests (Hall & Deardorff, 2006). However, when the campaign contributions made by a movement increase, the counter-movement may also attempt to contribute to legislators who have opposing interests to the movement’s positions. In this case, the ability of the movement to recruit legislators not only requires higher monetary incentives, but may also be at stake, as legislators may simultaneously receive contributions from the counter-movement, which reduces the effectiveness of the specific contribution (Baron, 2006). Moreover, the effectiveness of contributions by corporations is threatened by the fact that policymakers are less willing to be associated with corporations that are targeted by activist organizations (McDonnell & Werner, 2016). Hence, policymakers may avoid contributions from corporations in contested industries in order to minimize the negative repercussions elicited by sustaining the industry’s interests. For instance, given the history of lies and deception of the tobacco industry (Proctor, 2011), Big Tobacco’s contributions may be discredited by some policymakers. Evidence of this is the “Free from Tobacco Money” award, which is given yearly by ASH (Action on Smoking and Health), a nonprofit anti-smoking organization that recognizes politicians who have not accepted Big Tobacco’s contributions (ASH, 2014). Still, Big Tobacco is not fighting a lost battle, as the sizable resources it commands allow it to deter the enactment of laws by means of continuous and generous contributions to its allies. Hence, corporations remain a strong opponent for activist organizations, particularly because of having deep pockets that fund campaign contributions targeted at politicians that support the industry’s efforts and which are aimed at maintaining the status quo. Therefore, we posit that in the case where both the social movement and the counter-movement use campaign contributions as a strategy to achieve institutional change, the efficiency of their individual contributions is diminished. Thus, we formulate:

H4. The efficiency of the campaign contributions by the social movement to influence institutional change is negatively moderated by the counter-movement’s campaign contributions.

### Lobbyists

Lobbying is the activity by which individuals representing an organization (i.e., lobbyists) convey information to policymakers with the objective of influencing institutional change (Baron, 2006). The use of lobbyists, thus, is an important strategy by which activist organizations and corporations attempt to influence institutions (Hillman & Hitt, 1999), and it is an inconspicuous information strategy linked to the pursuit of private goals. That is, lobbyists seek to provide specific information directly to policymakers about the preferences of the group of societal members they represent. Moreover, lobbyists are also involved in providing information to policymakers on the potential impact of proposed regulations to persuade them to support their clients’ claims. Although, lobbyists and campaign contributions are linked (Ansolabehere, Snyder, & Tripathi, 2002), by defining lobbyists independently from campaign contributions, we are incorporating the distinct effects of information and financial strategies on institutional change (Hall & Deardorff, 2006; Hillman & Hitt, 1999). In the case of the tobacco industry, for instance, anti-smoking groups and Big Tobacco tend to use both strategies simultaneously, as campaign contributions grant them access to the policymaking process and employing lobbyists grant them influence over policymakers. Therefore, we analyze these strategies separately, in an effort to contribute to a more nuanced understanding of each strategy.

Previous research has argued that lobbyists draw attention to how the goals of policymakers align with those of activist organizations, as such, hiring lobbyists is an attempt by activist organizations to benefit from the actions of like-minded policymakers that represent their claims (Roscoe & Jenkins, 2005). In this sense, research has shown that lobbyists tend to concentrate their efforts on likeminded policymakers because lobbying is a relational strategy that involves reciprocity and trust (Hall & Deardorff, 2006). In the tobacco industry, anti-smoking groups hire lobbyists to persuade policymakers to enact tobacco control regulations by highlighting the need to protect the public’s health. Thus, we expect anti-smoking groups to be able to persuade policymakers who are concerned with protecting public health of the need to enact tobacco control regulations. In fact, lobbying efforts related to health issues have been reported as a top priority for lobbyists, meaning that they are commonplace in the legislative system (Baumgartner &Leech, 2001; Baumgartner et al., 2009) and are more likely to resonate with policymakers given the positive public health outcomes of tobacco control regulations (Burstein, 2014). Hence, we argue that when the claims of lobbyists representing activist organizations are deemed as appropriate and consequential by policymakers, these lobbyists are more likely to be aided in their efforts by policymakers (Robert Wood Johnson Foundation (RWJF), 2011). Hence, we expect as a starting premise that:

H5a. The number of lobbyists representing activist organizations increases the probability of institutional change.

Prior work has shown that lobbying efforts increase as the government’s activity on a specific issue intensifies (Leech et al., 2005). In this sense, corporations are expected to use as much resources as possible in hiring lobbyists to persuade policymakers to oppose regulations put forth by activist organizations if those regulations would negatively affect their business. Moreover, supported by the monetary resources they command, corporations have been known for dominating the lobbying landscape and for their ability to recruit lobbyists that become effective proponents of their claims (Kimball et al., 2012). This is especially so in the case of the tobacco industry, where Big Tobacco has hired numerous lobbyists to buttress their political allies. Big Tobacco’s lobbyists have transmitted a message focused on the rights and freedom arguments for smoking, and on the allegedly negative economic consequences to tobacco-related businesses of the enactment of tobacco control regulations (ASH, 2014). In short, we posit that lobbyists that represent corporations potentially coax legislators not to enact a certain law by conveying information on the foreseeable negative impact of a law. Therefore, we hypothesize:

H5b. The number of lobbyists representing corporations that aim to maintain the status quo decreases the probability of institutional change.

Conflicting results on the effectiveness of lobbying efforts have been documented by prior research (Baumgartner et al., 2009). Although prior work has posited that on average actors who face no opposition are more likely to achieve their goals by lobbying (Hojnackia et al., 2015), it remains unclear who is more effective in the case when there is a dispute between the interests of corporations and activist organizations, and both rely on lobbyists. While corporations are generally perceived to be relatively more successful in their lobbying efforts, prior scholarly work has shown that that is so when there is no opposition from other actors. However, in the face of opposition, corporations are less likely than activist organizations to achieve their goals by lobbying (Hojnackia et al., 2015). This is because the conflict between activist organizations and corporations draws the public interest and attention, which then constrains the advantage of corporations in the policymaking process by making the issue more salient and visible (Smith, 2000). Thus, the efficiency of using lobbyists is negatively affected for both sides when the lobbying efforts of the social movement are challenged by the lobbying efforts of the counter-movement. Thus, we formulate:

H6. The efficiency of lobbyists representing the social movement to influence institutional change is negatively moderated by the counter-movement’s lobbyists.

As the previous discussion highlights, protests, campaign contributions, and hiring lobbyists are strategies used by activist organizations and corporations that play a critical role in disrupting or maintaining the status quo and in bringing about or repelling institutional change. However, the effectiveness of a strategy can depend on various contextual influences. In the case of study, for example, anti-smoking groups’ success in persuading policymakers to enact tobacco control regulations has been closely tied to their ability to frame their claims in terms of the health costs of smoking. This is because smoking-related health costs weigh heavily on the states’ healthcare systems: “of every 10 spent on healthcare in the U.S., almost 90 cents is due to smoking” (Xu, Bishop, Kennedy, Simpson, & Pechacek, 2016, p. 331). Hence, we posit that two framing strategies explain the ability of activist organizations to persuade policymakers to change the status quo: relevance and credibility. The former refers to the ability of activist organizations to effectively and strategically frame their claims by linking them to the economic benefits for the state. For example, relevance in the tobacco setting is evidenced by the fact that anti-smoking groups have framed their claims in relation to the substantial burden that smoking inflicts on the state’s healthcare institutions, especially on those that are funded by public tax dollars (Xu et al., 2016). Thus, relevance has been a crucial framing strategy for anti-smoking groups to portray their goals as being worthy of recognition and attention since their aim is to stop the costly “tobacco epidemic” (Derry & Waikar, 2008; Nathanson, 1999). Credibility relates to the ability of activist organizations to gain public support from legitimate organizations that attest to the economic benefits of their claims. In the case of study, for instance, credibility is related to the ability of anti-smoking groups to gain public support not only from legitimate health organizations such as the American Lung Association or the American Cancer Society but also from those societal members that are most impacted by and/or that are most aware of the health costs associated with smoking (e.g., medical professionals). These organizations reinforce the economic benefits related to the enactment of tobacco control regulations by emphasizing the “true” costs of smoking as related to lost productivity and direct healthcare expenditures (American Cancer Society (ACS), 2015; Surgeon General (SG), 2014). Hence, we expect that the strategies used by activist organizations to change institutions result in a greater impact when they are capable of steering the attention of policymakers to their claims while undermining the claims made by corporations and that this happens when they frame their claims as having a relevant and credible impact on the economy of the state. Based on this argument, we predict: H7. The different strategies used by activist organizations will further increase the probability of institutional change for states in which the potential economic benefits of disrupting the status quo are higher. As mentioned before, corporate counter mobilization takes place through strategies similar to those utilized by activist organizations, the differences being that corporations in contested industries face threats to their legitimacy, which challenge their ability to maintain the status quo. However, the ability of corporations in contested industries to deter institutional change is facilitated when their interest in maintaining the status quo is aligned with the goals of policymakers and society members (Smith, 2000). That is, we posit that the perception of alignment between the preferences of corporations and those of society members is a necessary condition for the efforts of corporations to be effective in persuading policymakers to maintain the status quo for two reasons. First, policymakers that aim to be reelected rely on voters for their election; thus, they are more inclined to support laws that are supported by societal members and to avoid laws that may result in low public credibility (Keim & Zeithaml, 1986). For example, it has been recognized that disparities between tobacco control regulations across states partly depend on societal attitudes toward smoking, as regulations in tobacco growing states are less strict or widespread than regulations in states where smoking is less socially accepted. Second, policymakers represent their constituents, which means that they are more likely to support policies that are linked to the preferences of these constituents (de Guia et al., 2003). Therefore, a more significant influence of Big Tobacco in the legislative process can be expected in states with higher smoking rates given the resonance of their claims among society members (McVeigh et al., 2004; Pedriana, 2006). In other words, we argue that when the states’ constituency supports the strategies of corporations to maintain the status quo, they favorably dispose policymakers toward the interests of corporations, which poses a significant threat to institutional change (Dhalla & Oliver, 2013; Oliver, 1991). Thus, we predict: H8. The different strategies used by corporations will further reduce the probability of institutional change for states in which constituencies support maintaining the status quo. ## Methods ### Data Sources and Measures We gathered state-level data from several data sources for the period 2000–2012. During this period the number of states with comprehensive smoking bans increased from 0 to 27 (see Fig. 1). This period is especially relevant for our study because it allows us to minimize censoring problems in our models, as the first comprehensive smoking ban was enacted in 2002 and no comprehensive smoking ban has been enacted after 2012. Moreover, although during the period of study the tobacco industry in the United States was operating under the conditions of the 1998 Master Settlement Agreement, the “tobacco wars” in the United States are far from being over. In fact, contestation over the tobacco industry is still present, while some experts maintain that “America is still beholden to Big Tobacco” (Cohen, 2016), others contend that “investors and incorrigible puffers, are its last remaining friends” (Economist, 2014). We use the state level of analysis not only because in the United States the decentralization of politics allows exploiting between state differences but also because the strategies studied are known to be more powerful at the state level (Marquis, Guthrie, & Almandoz, 2012). Moreover, in the United States, state legislatures are important arenas for tobacco control because they have the power to enact laws related to tobacco taxes, youth access to tobacco, and smoking bans, and to pass preemption laws. State legislatures are, however, prohibited from taking legal action on cigarette advertising. ### Fig. 1 State-Level Comprehensive Smoking Bans Enacted by Year. Dependent variable. A comprehensive smoking ban is one that prohibits smoking in all restaurants, bars, and workplaces – with no exceptions – because these are the major venues of exposure to secondhand smoke for nonsmokers (Centers for Disease Control and Prevention (CDC), 2012). Comprehensive smoking bans have been recognized to be the most effective tobacco control regulations, as they make smoking less attractive and less visible (SG, 2014). We cannot observe a states’ probability of enacting a comprehensive smoking ban, instead, we only observe the actual enactment of a ban. Hence, we create a binary variable that takes the value of “1” for a state in a given year if a comprehensive smoking ban is enacted, and “0” otherwise. We collected data on the enactment of comprehensive smoking bans by US states from the State Tobacco Activities Tracking and Evaluation System and from the ANR Tobacco Control Laws Database. Once a state enacts a comprehensive smoking ban it is excluded from the data set. The sample consists of 540 state-year observations. Independent variables. We model the probability of the enactment of a comprehensive smoking ban as a function of time-varying independent and control variables. To test H1a, H1b, and H2, we followed prior research that has commonly used the media as a source of protests event data (Earl, Martin, McCarthy, &Soule, 2004). We conducted a search for newspaper articles about tobacco-related protests using Lexis–Nexis database and assembled the articles in 2014. In order to avoid possible selection bias in the reporting of protests, we did not focus on a single national outlet as has been commonly done, but instead included articles from both national and state sources. The combination of national and state sources has been recognized to be a more credible source of protests data when studying state-level protests (McAdam & Su, 2002). This triangulation of data sources is likely to result in more protests being reported, as protests closer to the media source are more likely to be covered than those further away (Earl et al., 2004; Inclán, 2008; Oliver & Maney, 2000). Moreover, using a combination of sources also attenuates description bias as it provides a more complete depiction of the protests (Earl et al., 2004; Soule, 2013). We operationalized protests as occurring when one of the following events took place: protests, strikes, riots, boycotts, marches, sit-ins, demonstrations, manifestations, attacks, rallies, vigils, occupations, campaigns, petitioning, or activism (Earl et al., 2003; King & Soule, 2007; Walker et al., 2008). An initial search on Lexis-Nexis using a combination of the search terms listed before and Boolean operators resulted in 909 newspaper articles. After similarity analysis to eliminate duplicated articles and after reading all the retrieved articles to exclude those that did not refer to public demonstrations or manifestations, the final sample resulted in 207 newspaper articles. All the articles in the final sample were read to extract the relevant information on each protest (i.e., participants, state, and year) and to validate that only protests that occurred before the enactment of a ban were included in the models. Each protest was coded as either anti-smoking or Big Tobacco based on the actors that initiated and participated in the protest and their claims. An anti-smoking protest includes, for instance, manifestations against smoking or calls to prevent tobacco use (see Fig. 2). A protest by Big Tobacco is, for example, one in which demonstrators urge the defeat of a bill or light up cigarettes to campaign against laws that restrict their freedom of choice (see Fig. 3). We utilize two state-level measures of protests in the analyses, the first measure is the count of anti-smoking protests and the second is the count of Big Tobacco’s protests. 4 To assess inter-coder reliability, a second coder read all the newspaper articles and coded each protest as either anti-smoking or Big Tobacco, the rate of interrater agreement (90.53%) suggests almost perfect agreement (Landis & Koch, 1977). 5 ### Fig. 2 Anti-Smoking Protests by State between 2000 and 2012. In Each, Alaska and Hawaii, There Was Only One Protest by Activist Organizations. ### Fig. 3 Big Tobacco Protests by State between 2000 and 2012. Neither in Alaska nor in Hawaii Were There Protests by Corporations. The Lobbying Disclosure Act of 1995 mandated corporations to disclose their federal lobbying practices. State legislatures followed suit and extensive data on lobbying activities at the state level are available from the National Institute on Money in State Politics (http://www.followthemoney.org/). 6 We collected data on state-level campaign contributions and lobbyists from this institute by retrieving information listed under the tobacco, tobacco companies, and tobacco product sales categories. To test H3a, H3b, and H4, we include a variable that captures campaign contributions for both sides at the state level (see Fig. 4). Since this variable is highly skewed, we used the log-transformed measure in the analyses. To test H5a, H5b, and H6, we include a variable that captures the number of lobbyists for both sides at the state level (see Fig. 5). For both variables, campaign contributions and lobbyists, we classified actions either as anti-smoking or Big Tobacco based on the information on the industry and recipient provided. Specifically, activities by tobacco control groups are classified as anti-smoking, whereas activities by tobacco companies or by organizations involved or individuals employed in the tobacco business were categorized as Big Tobacco. ### Fig. 4 Total Tobacco-Related Campaign Contributions by State between 2000 and 2012. ### Fig. 5 Total Tobacco-Related Lobbyists by State between 2000 and 2012. To test H7, we interact anti-smoking protests, campaign contributions, and lobbyists with health costs. The health costs variable measures the capital expenditures on health expenses as a percentage of the general expenditures of a given state. This variable accounts for the potential economic benefits related to a lower incidence of smoking in a given state since it captures the state’s expenditures on health costs associated with smoking. Data for this variable were gathered from the US Census Bureau. To test H8, we interact Big Tobacco’s protests, campaign contributions, and lobbyists with smoking prevalence. This variable measures the yearly percentage of adults who are smokers in a state and indicates the size of the market for cigarettes. Hence, this variable reflects societal preferences towards smoking as it directly captures the extent to which smoking is (or is not) a widespread habit in a given state. Data on cigarette smoking come from the CDC Behavioral Risk Factor Surveillance System. Control variables. We include one-year lagged control variables in the models to better capture causality. First, we include the log of state tobacco-related tax collections to capture the economic pressures against the enactment of laws derived from cigarette excise tax revenues. These data were obtained from the Tax Burden on Tobacco (Orzechowski & Walter, 2014). Second, in order to control for the fact that the so-called “tobacco states” are less likely to enact smoking bans, we account for tobacco harvested area per state (acreage). Data for the tobacco crops variable were collected from the National Agricultural Statistics Service. Third, we control for the percentage of sales to minors in a given state. Data were obtained from the Annual Synar Report, a report that details trends in sales of tobacco to minors. Fourth, to capture the anticipation that liberal governments are more likely than conservative governments to support the enactment of comprehensive smoking bans, we control for state ideology by using the measure developed by Berry, Ringquist, Fording, and Hanson (1998). Data are available at: https://rcfording.wordpress.com/state-ideology-data/. We also control for the anticipation that democrats are more likely than republicans to support the enactment of comprehensive smoking bans by controlling for unified democratic gubernatorial and legislative control of state government. Data were gathered from the US Census Bureau. Fifth, we include the potential for mobilization represented by a count of the number of tobacco-related social movements in a state for a given year. The existence of social movements has been recognized to be a necessary condition for influence, as the circumstances that spur the formation of social movements and that promote mobilization are also necessary for movements to gain influence. Data for this variable were collected using annual volumes of the Encyclopedia of Associations from 1998 to 2012; we consulted yearly volumes in order to avoid the measure to be shaped by entries and exits (Walker, 2009). Sixth, to capture regional diffusion effects we compute the number of geographically contiguous states that had previously enacted comprehensive smoking bans by including a neighboring effects variable. Moreover, to tap into the states that have already enacted a ban, we include a variable that captures the cumulative number of comprehensive smoking bans previously enacted. Finally, to control for the visibility of protests, we include media coverage, a variable that counts the number of newspapers reporting a given protest. The models also include the health costs and smoking prevalence variables as controls, and a time trend variable to control for the upward sloping path of the dependent variable. ### Model and Analyses We use a probit model to estimate the probability of state i enacting a comprehensive smoking ban at time t, as follows: (1) P(yit=1|xit)=Φ(xitβ) The probability of the enactment of a comprehensive smoking ban is given by Eq. (1), where β is the vector of coefficients to be estimated and xit is the vector of variables discussed in the previous section. As mentioned before, within this set of variables there are potentially endogenous variables. In particular, we identify protests as an endogenous variable given that protesting efforts may be driven by the expectation of whether a ban will be enacted (Oliver & Maney, 2000). Although event history or hazard models have traditionally been used to estimate the likelihood of occurrence of an event, estimating the models without correcting for endogeneity will result in biased coefficients (Alvarez & Glasgow, 1999). 7 To solve for the potential endogenous nature of the protest variable, we use a two-stage residual inclusion (2SRI) estimation method (Hausman, 1978; Terza, Basu, & Rathouz, 2008; Wooldridge, 2015), also known as a control function approach (Heckman & Hotz, 1989; Petrin & Train, 2010). 8 The selection of the 2SRI estimation over traditional event history or hazard models is motivated by the possibility that some omitted variables that determine the probability of enacting a comprehensive smoking ban may be correlated with protests by anti-smoking groups or Big Tobacco. The essential idea of the 2SRI estimation is that although there are some unobserved variables that may be correlated with protests by anti-smoking groups or Big Tobacco (e.g., the expectation of enactment of a ban), we can estimate these unobservables and thereby obtain consistent estimates for the probability of the enactment of a comprehensive smoking ban. To do so, the 2SRI estimation suggests a two-stage procedure: (i) estimate the endogenous variable(s) by regressing it on a set of independent variables that do not relate to the expectation of an enactment to obtain the residuals and (ii) estimate the original regression with the inclusion of the residuals as additional explanatory variable(s) (Wooldridge, 2015). In short, to estimate the probit coefficients using 2SRI we first estimate the endogenous variable(s) (i.e., protests), then obtain the residuals from the first-stage regression, and add these residuals as an additional variable in the second-stage to estimate the probit equation. This two-step procedure allows for the estimation of unbiased coefficients using a nonlinear model (Terza et al., 2008). In the first stage, we model the endogenous variables (i.e., protests by anti-smoking groups – yAS – and Big Tobacco –yBT ), as follows: (2) yAS=τAS+zitπAS+uAS (3) yBT=τBT+zitπBT+vBT where zit is a vector of exogenous variables that have been identified as determinants of public protests by prior research, namely: population, GDP per capita, poverty rate, unemployment rate, size, and number of state legislators (Lehman-Wilzig & Ungar, 1985; Meyer, 2004; Snow, Soule, & Cress, 2005; Soule, Mcadam, Mccarthy, & Su, 1999). State-level data for these variables were gathered from SAGE Stats. By running a regression for each of the two endogenous variables, yAS and yBT , on the aforementioned variables, zit , we explain the exogenous variation in these variables. Next, we can consistently estimate the residuals of Eqs. (2) and (3). In the second stage, we use the 2SRI estimator to correct for endogeneity by including the retained residuals (u^AS and v^BT) as additional variables with corresponding parameters to be estimated in Eq. (1), leading to: (4) P(yit=1|xit,uAS,vBT)=Φ(xitβ+u^ASδ+v^BTθ) The inclusion of the residuals in Eq. (4) accounts for endogeneity, and Eq. (4) results in unbiased and consistent estimates (Terza et al., 2008). We estimate the models using clustered standard errors to allow for correlation across observations within states. ## Results The descriptive statistics and correlations among the variables included in the analysis are shown in Tables 1 and 2. All correlations indicate that there are no multicollinearity problems in the data. As mentioned in the methods section, we use a 2SRI estimation procedure to test our hypotheses: the first-stage estimates protest by anti-smoking groups or Big Tobacco to recover the residuals entering in the second-stage probit model. Table 1 Descriptive Statistics. Obs Mean Std. Dev. Min Max Protest anti-smoking 540 0.119 0.356 0 2 Protests Big Tobacco 540 0.080 0.389 0 6 Contributions anti-smoking 540 1.025 2.810 0 12.892 Contributions Big Tobacco 540 6.122 5.616 0 18.045 Lobbyists anti-smoking 540 1.156 2.357 0 19 Lobbyists Big Tobacco 540 3.270 6.942 0 48 Health costs 489 0.031 0.012 0.012 0.102 Smoking prevalence 489 21.682 3.481 9.3 32.6 Tax collections 489 18.382 2.491 0 21.166 Tobacco crops 489 2.696 4.286 0 12.086 Sales to minors 489 14.924 7.977 1.6 55.8 State ideology 482 46.509 27.525 0 97.5 Democrats 489 0.552 0.498 0 1 Social movements 489 1.851 2.304 0 18 Neighboring effects 489 0.149 0.425 0 3 Cumulative bans 489 7.959 8.763 0 26 Media coverage 489 0.078 0.477 0 6 Population 540 0.593 0.682 0.049 3.804 GDP per capita 540 10.708 0.232 10.277 12.068 Poverty rate 540 0.123 0.033 0.055 0.222 Unemployment rate 540 5.617 2.062 2.3 14.7 Size 540 0.008 0.01 0 0.067 Number of legislators 540 148 63 0 424 Table 2 Correlations. First-Stage Variables 1 2 3 4 5 6 1. Population 1 2. GDP per capita 0.063 1 3. Poverty rate 0.135* −0.258* 1 4. Unemployment 0.237* −0.062 0.413* 1 5. Size 0.125* 0.180* −0.049 0.137* 1 6. Number of legislators 0.138* −0.232* −0.168* −0.036 −0.278* 1 Second-Stage Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1. Protest anti-smoking 1 2. Protests Big Tobacco 0.106* 1 3. Contributions anti-smoking 0.015 −0.023 1 4. Contributions Big Tobacco 0.036 −0.006 0.346* 1 5. Lobbyists anti-smoking −0.011 0.191* 0.136* 0.127* 1 6. Lobbyists Big Tobacco 0.131* 0.076 0.062 0.106* 0.586* 1 7. Health costs 0.026 −0.043 0.049 −0.023 0.075 −0.043 1 8. Smoking prevalence −0.110 0.011 −0.148* −0.025 −0.167* −0.105* −0.064 1 9. Tax collections −0.029 0.099* 0.091* 0.179* 0.183* 0.191* −0.051 −0.269* 1 10. Tobacco crops −0.062 −0.014 0.016 0.080 0.151* 0.088 0.263* 0.087 0.146* 1 11. Sales to minors 0.041 −0.001 −0.102* −0.160* −0.229* −0.186* 0.167* −0.220* 0.071 0.057 1 12. State ideology 0.030 −0.022 0.022 0.035 −0.014 0.006 0.000 0.200* 0.017 −0.087 0.171* 1 13. Democrats 0.018 0.004 0.042 0.005 −0.006 0.045 −0.080 −0.136* −0.074 0.047 −0.299* −0.114* 1 14. Social movements 0.096* 0.210* 0.033 0.025 0.010 0.070 0.050 0.193* 0.151* 0.125* 0.056 0.010 0.162* 1 15. Neighboring effects −0.047 −0.001 0.024 0.010 0.073 0.063 −0.177* 0.067 −0.066 −0.135* 0.039 −0.032 −0.034 −0.082 1 16. Cumulative bans −0.075 −0.047 0.077 0.216* 0.233* 0.136* −0.249* 0.162* 0.006 −0.474* −0.018 −0.010 −0.111* −0.302* 0.222* 1 17. Media coverage 0.019 0.243* 0.048 0.018 0.112* 0.158* −0.058 0.097* −0.027 −0.018 0.025 −0.026 −0.024 0.087 0.044 0.067 1 Note: *Correlations are significant at the 0.05 level. The first-stage estimates of the determinants of anti-smoking and Big Tobacco protests are reported in Table 3. From Table 3 we see that population size has the expected positive effect on protests by both groups (Lehman-Wilzig & Ungar, 1985). We also find that GDP per capita increases protests by anti-smoking groups. This result supports the central tenant of resource mobilization theory stating that resources facilitate protests (Snow et al., 2005), and for the specific case of study, corresponds with evidence on the negative relationship between smoking and income (Humphreys, 2015). Moreover, we find that protests by Big Tobacco decrease with the size of the state and increase with the number of legislators. These findings suggest that the larger the state legislative structure, the greater the political opportunity for counter-mobilization by Big Tobacco (Soule et al., 1999). Lastly, we find no impact of the economic indicators included in the models. From these models, we obtain the residuals to be included in the main probit regression. Table 3. First-Stage Results: Drivers of the Occurrence of Protests. Protests Anti-Smoking Protests Big Tobacco Population 0.076** 0.102** (0.033) (0.040) GDP per capita 0.246*** 0.101 (0.085) (0.076) Poverty rate −0.680 0.223 (0.610) (0.471) Unemployment rate −0.006 −0.001 (0.007) (0.006) Size −2.076 −1.327* (1.385) (0.748) Number of legislators −0.000 0.000* (0.000) (0.000) Constant −2.394** −1.144 (0.937) (0.884) Observations 540 540 Notes: Robust standard errors in parentheses. ***p < 0.01; **p < 0.05, *p < 0.1. Table 4 reports the results of the probit estimates. The results are estimated using a 2SRI estimator where the residuals obtained from the first-stage regressions enter without transformation in the second stage. When included, the residuals capture the endogenous part of the protests by both groups. Models 1–12 present a series of nested models with the correction for endogeneity using the 2SRI estimator. Positive (negative) coefficients mean that the probability of the enactment of a smoking ban increases (decreases) with the focal variable. Marginal effects are evaluated by setting all other variables at their means (Hoetker, 2007). Table 4 Second-Stage Results: Drivers of Enacting a Comprehensive Smoking Ban. Model 1 Controls Model 2 Protests Model 3 Protests Interaction Model 4 Contributions Model 5 Contributions Interaction Model 6 Lobbyists Model 7 Lobbyists Interaction Protest anti-smoking 7.331*** 7.076** (2.688) (2.868) Protests Big Tobacco −9.754*** −10.941*** (2.352) (2.791) Contributions anti-smoking 0.092*** 0.161** (0.033) (0.081) Contributions Big Tobacco −0.040* −0.034 (0.024) (0.025) Lobbyists anti-smoking 0.122*** 0.220*** (0.044) (0.074) Lobbyists Big Tobacco −0.014 0.014 (0.022) (0.018) Interaction term 0.504 −0.007 −0.008*** (1.325) (0.008) (0.003) Health costs 20.233** 21.440* 21.646 −0.013 20.503** 0.009 19.679** (9.096) (12.539) (13.206) (0.070) (8.705) (0.061) (9.199) Smoking prevalence −0.092*** −0.104*** −0.113*** 0.010 −0.097*** −0.066 −0.073** (0.032) (0.040) (0.043) (0.106) (0.033) (0.121) (0.034) Tax collections −0.007 0.301* 0.296 −0.114*** −0.003 −0.142*** −0.098 (0.105) (0.175) (0.181) (0.036) (0.105) (0.044) (0.122) Tobacco crops −0.094** −0.082 −0.083 −0.008 −0.098** −0.005 −0.121*** (0.041) (0.059) (0.064) (0.025) (0.038) (0.025) (0.039) Sales to minors −0.006 −0.012 −0.013 0.011*** −0.003 0.012*** −0.003 (0.026) (0.030) (0.031) (0.004) (0.027) (0.004) (0.026) State ideology 0.011** 0.014** 0.016** 0.250 0.012*** 0.214 0.012*** (0.004) (0.006) (0.006) (0.232) (0.004) (0.226) (0.004) Democrats 0.276 0.287 0.406 21.531** 0.228 17.511** 0.155 (0.243) (0.260) (0.263) (8.635) (0.242) (8.776) (0.224) Social movements −0.027 −0.034 0.007 0.327* −0.028 0.375* −0.012 (0.076) (0.086) (0.086) (0.196) (0.074) (0.203) (0.069) Neighboring effects 0.277 0.194 0.238 −0.072 0.259 −0.089 0.300 (0.190) (0.206) (0.214) (0.053) (0.188) (0.060) (0.204) Cumulative bans −0.086 −0.058 −0.064 0.376*** −0.069 0.391*** −0.045 (0.057) (0.063) (0.067) (0.137) (0.055) (0.129) (0.058) Media coverage 0.389** 0.331** 0.400** 0.376*** 0.383*** 0.391*** 0.376** (0.157) (0.151) (0.160) (0.137) (0.142) (0.129) (0.151) Constant −2.408 −8.658*** −8.674*** −4.564** −2.260 −3.497* −0.946 (1.985) (2.989) (3.184) (1.887) (1.825) (2.111) (1.990) Time trend Included Included Included Included Included Included Included Residuals Included Included Observations 482 482 482 482 482 482 482 Model 8 AS Health Costs Model 9 BT Smoking Prevalence Model 10 Conspicuous AS Inconspicuous BT Model 11 Conspicuous BT Inconspicuous AS Model 12 Full Model Protest anti-smoking 11.137*** 7.232** 2.262 7.558*** (3.351) (3.012) (2.382) (2.877) Protests Big Tobacco −9.884*** −10.607*** −6.445*** −10.243*** (3.479) (3.307) (2.410) (2.913) Contributions anti-smoking 0.196** 0.121*** 0.081*** 0.108*** (0.087) (0.037) (0.031) (0.036) Contributions Big Tobacco −0.043 −0.047 −0.016 −0.044 (0.030) (0.031) (0.021) (0.029) Lobbyists anti-smoking −0.243 0.122** 0.089** 0.107** (0.251) (0.047) (0.043) (0.046) Lobbyists Big Tobacco −0.002 0.016 0.001 0.001 (0.021) (0.016) (0.015) (0.018) Moderation protest −145.667*** 0.012 (55.612) (0.067) Moderation lobbying −2.850 0.023*** (2.313) (0.007) Moderation lobbyists 10.247 0.002 (6.366) (0.008) Health costs 14.096 18.714 21.693** 13.140 17.282 (14.827) (12.101) (10.313) (8.461) (11.675) Smoking prevalence −0.148*** −0.119*** −0.078** −0.126*** −0.104*** (0.046) (0.042) (0.037) (0.036) (0.040) Tax collections 0.066 0.136 −0.031 0.126 0.198 (0.206) (0.191) (0.133) (0.190) (0.192) Tobacco crops −0.170*** −0.149*** −0.101** −0.094** −0.122** (0.049) (0.057) (0.046) (0.039) (0.048) Sales to minors −0.004 0.001 −0.006 −0.004 −0.005 (0.033) (0.031) (0.024) (0.033) (0.031) State ideology 0.025*** 0.023*** 0.011** 0.016*** 0.020*** (0.005) (0.005) (0.005) (0.004) (0.005) Democrats 0.270 0.232 0.330 0.035 0.110 (0.243) (0.243) (0.237) (0.206) (0.219) Social movements −0.043 0.021 −0.066 0.053 −0.023 (0.107) (0.082) (0.083) (0.073) (0.086) Neighboring effects 0.260 0.300 0.252 0.286 0.238 (0.242) (0.234) (0.191) (0.196) (0.221) Cumulative bans 0.005 −0.002 −0.081 −0.031 −0.001 (0.064) (0.068) (0.058) (0.058) (0.061) Media coverage 0.468*** 0.433*** 0.397** 0.307** 0.355** (0.127) (0.144) (0.156) (0.139) (0.142) Constant −3.327 −5.527* −2.503 −3.755 −6.420** (3.518) (3.264) (2.083) (3.022) (3.245) Time trend Included Included Included Included Included Residuals Included Included Included Included Included Observations 482 482 482 482 482 Notes: Clustered robust standard errors in parentheses. ***p < 0.01; **p < 0.05; *p < 0.1. Model 1 is the baseline model as it only includes the control variables. Most of the results for the control variables are consistent across models. We find that the probability of enacting a comprehensive smoking ban significantly increases with the health costs associated with smoking. According to the US Surgeon General, smoking costs the US more than300 billion a year, including $170 billion in direct healthcare expenditures and$156 billion in lost productivity (SG, 2014). Thus, the healthcare costs associated with smoking have an essential role in guiding governmental efforts to enact smoking bans. Moreover, the results suggest that the probability of enacting a comprehensive smoking ban decreases with smoking prevalence. This is because cigarette usage remains socially accepted in states with high smoking rates, and thus, the public tends to oppose the enactment of bans; whereas in states with low smoking rates, population attitudes toward cigarettes are more negative and the public tends to favor the enactment of bans (RWJF, 2011). We also find that the probability of the enactment of a comprehensive smoking ban decreases with the production of tobacco, given the economic dependence of the focal state’s economy on this crop. Furthermore, the results show that the probability of the enactment of a comprehensive smoking ban increases as the government’s ideology moves from zero (the most conservative value) to one (the most liberal value). This finding is in line with the assumption that for conservative governments the ultimate responsibility for health lies with the individual, whereas liberal governments tend to regulate the behavior of individuals and organizations to protect and promote public health (Cohen et al., 2000). Finally, the results suggest that the scale of protests matters, as protests that are covered by several newspapers are more likely to foment institutional change.

Model 2 tests H1a and H1b. This model shows the effect of protests by activist organizations and corporations on the probability of enactment of a law. We find evidence to support both hypotheses. On the one hand, we find that protests by activist organizations increase the probability of a law being enacted; specifically, one anti-smoking protest increases the probability of enactment of a comprehensive smoking ban by 11.2%. As the claims by anti-smoking groups become more visible and salient, their claims are deemed as less challenging or contentious and they grow and gain supporters to change the status quo. On the other hand, we find that the probability of a law being enacted decreases with protests by corporations; that is, one protest by Big Tobacco decreases the probability of enactment of a comprehensive smoking ban by 15%. This result reveals that the organization of a counter-movement is a powerful deterrent for the enactment of a law. As most of the protests by corporations occur when the regulative discussions are taking place, this result shows that when there is a real threat of change, reactive protests by corporations have the power to maintain the status quo. The results presented in Model 3 do not support H2, as the interaction coefficient for protests is not statistically significant. However, further support for H1a and 1b is provided, as the main effects of the protests variables remain significant.

Model 4 introduces the campaign contributions by activist organizations and corporations to test their effect on the probability of a law being enacted. The results provide support for H3a and H3b. They indicate that a 1% increase in campaign contributions by anti-smoking groups corresponds to an increase in the probability of enacting a ban by 0.3%, while a 1% increase in campaign contributions by Big Tobacco decreases the probability of enactment by 0.2%. These results suggest that anti-smoking groups are more efficient in their campaign contributions despite Big Tobacco’s significant financial resources. The test for H4, presented in Model 5, shows that the interaction for the campaign contributions variables is not significant. However, although in this model the main effects of campaign contributions remain positive and significant for activist organizations, the marginal effect of campaign contributions by corporations disappears; which suggest that when both actors attempt to influence the enactment of regulations by means of campaign contributions, anti-smoking groups are more effective than Big Tobacco given their access to complementary external resources.

Model 6 introduces the number of lobbyists representing each side of the debate to tests H5a and H5b. The results only provide support for H5a and suggest that for each extra lobbyist that represents activist organizations, the likelihood of enactment of a comprehensive smoking ban increases by 0.4%. These results indicate that even though corporations have been widely known for their extensive lobbying efforts, lobbyists that represent activist organizations are more successful in increasing the likelihood of enactment of a comprehensive smoking ban because their position is compatible with the government’s mandate to protect public health. Model 7 presents the results of the interaction test for H6. The interaction coefficient is negative and significant, which suggests that the positive effect of anti-smoking lobbyists is reduced by 0.03% for each extra lobbyist working for Big Tobacco.

Models 8 and 9 present the effects of the moderation analyses testing the effects of contextual variables. For ease of comparison, we have included all control variables, main effects, and interactions. Model 8 includes the interactions of anti-smoking protests, campaign contributions, and lobbyists with the health costs variable. Model 9 includes the interactions of Big Tobacco’s protests, campaign contributions, and lobbyists with smoking prevalence. Hoetker (2007) suggests that the interpretation of interaction terms in nonlinear models cannot be evaluated by simply looking at the coefficients, as their magnitude, sign, and significance can change across observations. Instead, given that the interaction effects depend on all the covariates in the model, the best practice is to provide a graphical interpretation (Hoetker, 2007). Therefore, Figs. 6 and 8 provide a graphical presentation of the moderation analyses, and Figs. 7 and 9 provide histograms that show the distribution of the moderator variables.

Fig. 6 illustrates the moderation effects of health costs for anti-smoking protests, campaign contributions, and lobbyists. The first plot shows that the positive moderation effect of protests by anti-smoking groups on the probability of enactment of a comprehensive smoking ban is present in states where the healthcare costs of smoking are less than 5.5% of the state’s general expenditures. Similarly, the second plot shows that health costs positively moderate the relationship between anti-smoking campaign contributions and the probability of enactment of a comprehensive smoking ban when health costs are between 1% and 4.5% of the state’s general expenditures. In contrast, the third plot shows that health costs do not moderate the effect of lobbyists on the enactment of a law. Overall, Fig. 6 provides support for the positive moderation effect of health costs stated in H7 for the case of anti-smoking protests and campaign contributions. Given that in most states the health costs associated with smoking represent more than 1% and less than 6% of the state’s general expenditures (see Fig. 7), the results suggest that the healthcare costs associated with smoking weigh heavily on the ability of anti-smoking groups to disrupt the status quo.

### Fig. 6

Average Marginal Effects of Anti-Smoking Strategies with 95% CIs (Health Costs).

### Fig. 7

Health Costs Histogram.

Fig. 8 illustrates the moderation effects of smoking prevalence for Big Tobacco’s protests, campaign contributions, and lobbyists. The first plot shows that smoking prevalence moderates the relationship between Big Tobacco’s protests and the probability of enactment of a comprehensive smoking ban when the state’s smoking rate is between 1% and 25%. Similarly, the second plot illustrates that smoking prevalence moderates the effect of Big Tobacco’s campaign contributions on the probability of enactment of a comprehensive smoking ban for states where the smoking rate is lower than 25%. In contrast, the third plot shows that smoking prevalence does not moderate the effect of lobbyists on the enactment of a law. Given that most states have smoking rates lower than 25% (see Fig. 9), the results suggest that societal smoking habits are important determinants of Big Tobacco’s ability to maintain the status quo. Overall, these results provide support for H8, which suggests that Big Tobacco’s strategies are effective in deterring legislative change in states where constituencies support smoking.

### Fig. 8

Average Marginal Effects of Big Tobacco Strategies with 95% CIs (Smoking Rate).

### Fig. 9

Smoking Prevalence Histogram.

Models 10 and 11 further test the simultaneous effects of protests, campaign contributions, and lobbyists by activist organizations and corporations. Model 10 presents the case in which each actor uses the strategy that is expected or associated with its traditional tactics to change institutions; namely, it presents the results when activist organizations use protests, and corporations use campaign contributions and lobbyists. Surprisingly, the results of this model are not significant. Therefore, the results of this model suggest that for actors aiming at either disrupting or maintaining the status quo, employing their traditional strategies does not prove to be effective in fomenting institutional change. The results of Model 11, which presents the results when activist organizations use campaign contributions and lobbyists, and corporations use protests, are statistically significant. Specifically, the results indicate that campaign contributions and lobbyists are effective strategies for activist organizations to encourage legislative change, and that protests are an effective strategy for corporations to discourage the enactment of laws. This unexpected finding suggests that actors are effective in fomenting institutional change when they adopt the strategy that has been traditionally used by their counterpart. Taken together, the results of models 10 and 11 suggest not only that it is important for actors that aim to change institutions to be able to use a wide range of strategies (including those in which their counterpart traditionally has had an advantage), but specifically, that if corporations are to maintain the status quo, constituency building strategies are more effective in deterring institutional change than financial or information strategies. These results are in line with Smith (2000), who argues that policymakers are more responsive to the preferences of society members when corporations show unity of claims or seem to have the same objective.

Model 12 presents the results of the full model. The results remain remarkably similar when compared to those of the nested models, although campaign contributions by Big Tobacco are no longer significant. Overall, the results provide substantial evidence that activist organizations and corporations use different strategies to change institutions. Specifically, they show that Big Tobacco’s power to inconspicuously thwart the enactment of comprehensive smoking bans has decreased along with the changing societal perception of tobacco products, which has led to Big Tobacco’s dependence on the use of conspicuous means of influence. In contrast, anti-smoking groups have gained visibility and public support, which has strengthened both their conspicuous and inconspicuous impact on the legislative process. In other words, as knowledge about the health risks of smoking spreads, and as anti-smoking groups have gained ground to advance tobacco control regulations, Big Tobacco’s influence on the legislative process has waned. Overall, the results illustrate how corporations counter-mobilize by imitating and adopting the strategies of activist organizations that are successful in changing institutions.

### Robustness Checks

In order to test for sensitivity of our results due to model choice, we performed several tests which are reported in Table 5. First, we re-estimated Eq. (4) including an interaction term between the residuals of Eqs. (2) and (3) (u^AS * v^BT), as reported in Model 1, Table 5. This interaction term allows for a joint effect of the residuals, which takes into account that either they may influence each other or be driven by a common unobserved latent factor that may also influence the probability of the enactment (Wooldridge, 2015). Analogous to Petrin and Train (2010), we interpret the sign and significance of residuals. The residuals are significant and with the expected signs across models. In particular, the positive coefficient of the residual for protests by corporations (β = 12.123, p-value: 0.000) suggests that if corporations’ protests increase, it is more likely to expect the enactment of a law. In contrast, the residual for activist organizations is negative (β = –7.927; p-value: 0.007), which suggests that these groups would protest less when there is a higher likelihood of enactment. We observe that our findings are robust to the inclusion of the interaction term between the residuals. 9

Table 5

Robustness Checks.

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Protest anti-smoking 6.970** 6.714** 8.626*** 7.232** 8.592*** 11.911**
(2.942) (2.832) (3.012) (2.826) (6.004)
Protests Big Tobacco −11.272*** −9.952*** (2.479) −10.607*** −11.151*** −18.380***
(3.159) (2.829) (3.307) (3.511) (5.481)
Contributions anti-smoking 0.114*** 0.099*** −0.045* 0.121*** 0.111*** 0.193**
(0.036) (0.036) (0.037) (0.039) (0.077)
Contributions Big Tobacco −0.048 −0.040 (0.026) −0.047 −0.062** −0.087
(0.030) (0.029) (0.031) (0.029) (0.060)
Lobbyists anti-smoking 0.123*** 0.103** −0.019 0.122** 0.136*** 0.164
(0.046) (0.043) (0.047) (0.043) (0.123)
Lobbyists Big Tobacco 0.018 0.001 (0.023) 0.016 0.004 0.005
(0.016) (0.017) (0.016) (0.020) (0.043)
Moderation protest −0.000 −0.001
(0.000) (0.032)
Moderation lobbying −0.000 0.001
(0.000) (0.001)
Moderation lobbyists 0.000 0.003**
(0.000) (0.001)
Health costs 18.582 16.016 21.255* 14.096 17.703 27.077
(12.183) (11.899) (12.538) (14.827) (12.401) (24.896)
Smoking prevalence −0.118*** −0.110*** −0.095** −0.148*** 0.017 −0.221**
(0.042) (0.041) (0.040) (0.046) (0.010) (0.099)
Tax collections 0.148 0.235 0.277** 0.066 0.151 0.432
(0.191) (0.189) (0.132) (0.206) (0.182) (0.447)
Tobacco crops −0.144*** −0.117** −0.094 −0.170*** −0.173*** −0.217*
(0.054) (0.047) (0.058) (0.049) (0.053) (0.121)
Sales to minors −0.001 −0.002 −0.008 −0.004 −0.015 −0.028
(0.031) (0.032) (0.028) (0.033) (0.025) (0.084)
State ideology 0.023*** 0.022*** 0.017*** 0.025*** 0.024*** 0.037***
(0.005) (0.006) (0.006) (0.005) (0.006) (0.011)
Democrats 0.236 0.088 0.350 0.270 0.288 0.127
(0.240) (0.223) (0.276) (0.243) (0.256) (0.511)
Social movements 0.009 −0.022 −0.062 −0.043 0.038 −0.024
(0.093) (0.086) (0.081) (0.107) (0.083) (0.229)
Neighboring effects 0.301 0.225 0.193 0.260 0.272 0.481
(0.234) (0.212) (0.213) (0.242) (0.249) (0.452)
Cumulative bans 0.001 −0.014 −0.049 0.005 0.034 −0.025
(0.067) (0.057) (0.067) (0.064) (0.065) (0.141)
Media coverage 0.431*** 0.330** 0.374*** 0.468*** 0.539*** 0.664**
(0.143) (0.134) (0.136) (0.127) (0.138) (0.271)
Constant −5.586* −7.054** −8.552*** −5.527* −8.193*** −12.020
(3.277) (3.147) (2.999) (3.264) (3.116) (7.418)
Time trend and residuals Included Included Included Included Included Included
Observations 482 482 482 482 482 482

Notes: Clustered robust standard errors in parentheses.

***p < 0.01; **p < 0.05; *p < 0.1.

Second, we checked whether the results would still hold with a different specification of the ideology variable. For this purpose, we used Berry, Fording, Ringquist, Hanson, and Klarner’s (2010) alternative nominate measure of government ideology which the authors argue is an improved measure of state ideology. The results reported in Model 2, Table 5, are similar to those reported in Model 12, Table 4. This implies that our analysis is not sensitive to changes in the measurement of ideology, which rules out the option that the specification of this variable influences the results.

Third, in an effort to further assess the interaction between activist organizations and corporations and their simultaneous actions to change institutions, we ran the full model including the deltas of protests (i.e., anti-smoking protests–Big Tobacco protests), campaign contributions (i.e., Big Tobacco contributions–anti-smoking contributions), and lobbyists (i.e., Big Tobacco lobbyists–anti-smoking lobbyists) instead of the level. The results are reported in Model 3, Table 5. The results for protests show that an increase in the gap of protest in favor of anti-smoking groups raises the probability of enactment of a comprehensive smoking ban by 10.4%, whereas an increase in the gap of campaign contributions in favor of Big Tobacco decreases the probability of enactment of a comprehensive smoking ban by 0.1%. A change in the gap of lobbyists has no effect. The results provide further support for the interaction between strategies and actors in the case of protests and campaign contributions. 10

Fourth we test our moderation analyses with a different specification of the moderator variables. Model 4, Table 5, presents the results of an alternative moderator variable for H7. Instead of using health costs to measure the potential economic benefits of a smoking ban, we use Medicaid expenses. Data on total Medicaid spending were retrieved from the Henry J. Kaiser family foundation. Although Medicaid as a program mainly targets low-income individuals, this measure directly relates to the state’s expenses on smoking-related diseases because of the negative correlation between smoking and income (Goszkowski, 2008). The results of the moderation analysis suggest that Medicaid expenses positively moderate both, anti-smoking protests and campaign expenditures, in states where the expenses are lower than 12.000 million per year (in our data, only California, Florida, and Texas have expenses higher than this figure). Model 5, Table 5, presents the results of an alternative moderator variable for H8. Instead of using smoking prevalence to capture tobacco’s constituency influence, we use the yearly change of the Gross State Product (GSP) from tobacco-related businesses. Data on the GSP by industry were gathered from the US Bureau of Economic Analysis. The results of the moderation analysis suggest that the percentage GSP change negatively moderates the impact of both, anti-smoking protests and campaign expenditures, on the probability of enactment of a comprehensive smoking ban; such that the efficiency of Big Tobacco’s strategies increases as the industry’s economic impact on a given state increases. Consequently, these results strengthen our confidence in the robustness of the chosen measures for the moderator variables.

Finally, we estimate our full model using a logit specification (Petrin & Train, 2010). Conventional wisdom suggests that probit and logit models give essentially similar results. The results obtained are presented in Model 6, Table 5. The analysis yielded substantively similar results; all the coefficients have the same sign and significance levels, though some of the magnitudes slightly change because of expected scale differences. Thus, the results indicate that our model choice to analyze the enactment of a comprehensive smoking ban does not provide substantially different outcomes in comparison to other candidate models.

## Discussion

Our study disentangles the impact of activist organizations and corporations as participants and actors of institutional change. It shows that both activist organizations and corporations shape the enactment of laws and uncovers the strategies used by these movements and counter-movements to influence institutional change. The results show that activist organizations and corporations shape institutions through conspicuous (e.g., protests) and inconspicuous (e.g., campaign contributions and lobbyists) strategies. Specifically, the results suggest that whereas both actors use conspicuous and inconspicuous strategies to influence institutional change, activist organizations are successful in their use of both strategies to disrupt the status quo. In contrast, given the threats to their legitimacy, corporations in contested industries are mainly successful in maintaining the status quo when using conspicuous strategies. Therefore, the study shows that actors use multiple strategies to change institutions, and as such, the influence of actors cannot be captured through examining a single strategy. Furthermore, the results provide evidence that although constituency building has been considered the main strategy by which activist organizations aim to change institutions, and financial and/or information strategies have traditionally been linked with the attempts of corporations to influence institutions, in the face of changing societal expectations, the proven (in)effectiveness of these respective “traditional” strategies for each actor does not lend support to this belief anymore.

In the case of protests, the results highlight how in order to protest successfully, the ability to mobilize and to receive public support is crucial for activist organizations and corporations to successfully protest. Specifically, the results show that anti-smoking groups have been successful in disrupting the status quo and have effectively persuaded policymakers to enact tobacco control regulations. Simultaneously, Big Tobacco’s reactive protests by Astroturf groups that mobilize on its behalf have been effective in thwarting the enactment of laws. However, given the threats to Big Tobacco’s legitimacy, Big Tobacco is forced to conceal its involvement in protests that confront the anti-smoking message. Therefore, the capacity of Big Tobacco’s protests to maintain the status quo is significantly dependent on the acquisition of allies that legitimize their claims and on gaining public support. Hence, the results shed light on the effectiveness of constituency building strategies to change institutions, especially for counter-movements that face a successful movement. In other words, the results provide preliminary evidence for a lifecycle of protests, suggesting that for nascent movements protests are a successful strategy to disrupt the status quo, and that once the movement’s claims gain visibility and public support, the efficiency of protests by the counter-movement to maintain the status quo increases.

With respect to campaign contributions, the results show that controlling resources is not a sufficient condition for actors to be able to influence institutions, and provide evidence that financial strategies are not as effective when the actor that controls the monetary resources faces threats to its legitimacy. Specifically, the results show that when actors aim to advance a claim that goes against prevailing societal expectations, money is not enough to influence policymakers. Even though Big Tobacco commands more financial resources than anti-smoking groups, the latter are rooted in cultural and social structures that support their goals in the legislative process. Thus, we find limited support for the efficacy of Big Tobacco’s campaign contributions and suggest that as a social movement gains prominence because of the alignment between its goals and the goals of policymakers, the relative power of the counter-movement is reduced, and thus, while it is still more powerful in terms of financial resources, its influence through campaign contributions diminishes.

Concerning lobbying, the results suggest that the influence of lobbyists on institutional change is dependent on having interests that are aligned with those of policymakers, as is the case for activist organizations. Put differently, the results show that in the face of legitimacy threats, lobbyists that represent corporations in a contested industry are less influential than those that represent activist organizations whose claims are deemed as socially acceptable and aligned with the interests of policymakers. In the tobacco industry, anti-smoking groups have been successful in hiring lobbyists that frame their claims as being compatible with the preferences of societal members, and thus, their claims resonate with policymakers. In contrast, for corporations, the ability to align their claims with the priorities of policymakers is paramount, but Big Tobacco has been unable to do so given the increasing negative societal perception of the industry. These results are in line with work that has found little impact of corporate lobbying efforts on institutional change (Baumgartner et al., 2014).

A preliminary insight into the adjusted efficiency of protests, campaign contributions, and lobbyists by the social movement conditional upon the level of each of these tactics by the counter-movement yields no apparent conclusion. This, though, does not exclude more complex and intricate forms of dependency of the efficiency of the use of each strategy by the social movement based on the use of the same strategy by the counter-movement. In other words, the preliminary results for the adjusted efficiency of each strategy do not exclude that the efforts of activist organizations are contingent upon the efforts of corporations and vice versa. Nevertheless, as the moderation analyses suggest, societal preferences are important determinants of corporations’ ability to maintain the status quo, while the ability of activist organizations to disrupt the status quo is deeply linked to the use of framing strategies that highlight the relevance and credibility of their claims by linking them to economic benefits for the state. That is, given that often times the interests of corporations are opposed to those of activist organizations, their parallel actions result in a contested field were activist organizations aim to challenge the status quo, while corporations seek to preserve the status quo. In this fragmented system, policymakers tend to be more responsive to the preferences of activist organizations while the preferences of corporations are less likely to prevail (Smith, 2000).

Together, the results raise awareness of the interconnectedness of actors and strategies engaged in institutional change. That is, our results illustrate that the actions of corporations aiming to change institutions take place within industries alongside the efforts of activist organizations, and that the characteristics of the environment in which these actors operate shape their potential to change institutions. Moreover, our results contribute to the literature by illustrating that actors use conspicuous strategies like constituency building (e.g., protests), and inconspicuous strategies such as financial (e.g., campaign contributions) and information (e.g., lobbyists) propagation to influence institutional change in their favor (Hillman & Hitt, 1999). Interestingly, our results show that money and lobbyists are not as effective in bringing about or deterring institutional change as constituency building is, which leads us to believe that the ability of actors to influence institutional change in their favor comes more so from being able to attract and mobilize public support, and less so from having deep pockets or from controlling an army of lobbyists.

### Limitations and Future Research

The dependent variable is the enactment of a comprehensive smoking ban; thus, the models explain the final outcome of the legislative process. However, future research could divide the legislative process into different phases (i.e., agenda setting, passage, enactment, and possible repeal) in order to shed light on how the identified actors and strategies affect each phase separately and possibly differentially (Soule & King, 2006). Furthermore, our data collection efforts end in 2012; although no comprehensive smoking ban has been passed since then, it could be interesting for future research to study the states in which protests, campaign contributions, and/or lobbyists have not resulted in the enactment of comprehensive smoking bans.

Our level of analysis is the state. Although the state is a relevant level of analysis for the reasons mentioned above, future research could disaggregate the influence of each actor and strategy at a more local level, since in states where there is no preemption there may be a difference between city, county, and state laws (Vasi, Walker, Johnson, & Tan, 2015). Using the city or the county as the level of analysis may also allow researchers to uncover the local conditions or configurations that alter the rate at which states pass anti-smoking laws. Moreover, the change of level of analysis could also shed light on the differential effectiveness of various strategies in the enactment of local bans and possibly enhance our understanding of the micro processes of institutional change.

As mentioned before, in our analysis we defined two main sides (i.e. anti-smoking groups and Big Tobacco), each of which is comprised of various independent actors. Although we identify substantial different stances on tobacco control regulations between these groups, future research may disentangle variations within these groups to identify how differences in legitimacy or power, or the use of intra-coalition strategies, lead to different outcomes, as some actors within each group may be more efficient than others in bringing about institutional change.

There has been little research connecting institutional change to the simultaneous actions of activist organizations and corporations. This is partly because institutional change may result from a multitude of causes. Although we control for several factors that may have an effect on institutional change, we may have not fully considered all possible influences on the probability of enacting laws, as actors can certainly use other parallel strategies in order to change institutions, such as CSR and litigation. These strategies also target institutions but we do not include them in our study for two reasons. First, in the tobacco industry CSR contributions can also be understood as tax deductible expenses which result in disclosure issues because corporations are not obliged to report charitable contributions (Tesler & Malone, 2008). Thus, CSR contributions are potentially problematic in the context of study since Big Tobacco is known for using creative ways to disguise their efforts and to exert influence out of the public sight. Second, as Polletta (2000) posits, it is hard to uncover the role of litigation in relation to the efforts of activist organizations and corporations because litigation is primarily used after the enactment of laws has taken place. Nonetheless, future efforts that uncover the influence of actors in advancing or hindering institutional change by relational strategies that directly engage policymakers to discuss and craft legislation or to influence their voting, along with the differential impact of these and other non-market strategies such as CSR or litigation on institutional change, would be beneficial.

Finally, despite the fact that the tobacco industry displays a number of unique features that make it different from other noncontested industries, its structure is relatively similar to that of other contested industries that exhibit increased concentration, face extensive regulations, and exert political influence through extensive lobbying efforts mainly by the largest actors in the industry. Thus, our results are mainly generalizable to other contested industries such as weapons and arms, alcohol, or gambling. The results could also be generalizable to contentious issues where there are two sides: on one side, a rich corporation aiming to preserve the status quo or to manipulate institutions in its favor, and on the other side, society members aiming to disrupt the status quo and are urging for institutional change. Furthermore, the results are generalizable to situations where regulators have the capacity to legislate and face opposite demands; on the one hand, a corporate actor that may be harmed by a change in the status quo, and on the other hand, a social movement interested in bringing about institutional change. The results have limited generalizability, though, to issues where society members fight against the government, since in that case there is no clear counter-movement led or funded by corporations. Moreover, the results have limited generalizability to industries that do not operate across borders, as in the case of study, the tobacco industry was allegedly able to alleviate some of the national pressures by partly shifting its efforts abroad. Overall, although our study of the tobacco industry may be somewhat of a “polar type” when compared to other industries, the mechanisms that we uncovered have implications for scholars whose interests center on contestation, social movements, and non-market strategies. In addition, though we believe that the tobacco industry is particularly suited for exploring the strategies used by different actors to influence institutions, given the particular features of this industry, the generalizability of our findings to other noncontested industries needs to be established by future research. Lastly, besides the issue of generalizability, we believe that there is value in studying the tobacco industry in its own right given its considerable size, global presence and reach, as well as the health and other consequences of its products. That has been hardly done in management scholarship.

## Notes

1

Astroturf, contrary to grassroots organizing, is the practice by which organizations secretly support or create a counter-movement to articulate their interests (Davis, Morrill, Rao, & Soule, 2008).

2

Although it should be noted that the group collations described above occur in an ad hoc basis, this grouping structure recognizes that the individual efforts of the actors on each side of the tobacco control debate focus on advancing a common cause, on defending a common interest, and on defeating a common opponent. That is, the actors on each side share the same claim and show high consensus because the issues that they represent are beyond the interests of a single actor (e.g., the right to breath clean air vs. smoking as a right), which facilitates the pursuit of a shared goal. Moreover, as actors come together around the issues they aim to advance, they display unity by taking identical positions on the smoking debate, and their unified interests are stronger than the influence of a single actor (Baumgartner et al., 2009).

3

An exception is the work of Pescosolido, Grauerholz, and Milkie (1997), who study the normative pillar of institutions as target of protests by examining how the representation of African Americans in children’s books changes when racial images are challenged by, among others, protesting.

4

It is not possible to know the size of the protests. About 30% of the coded articles provide an indication of the number of participants, and less than 20% provide an exact or estimated number of participants.

5

As a rule-of-thumb, values greater than 75% represent excellent agreement beyond chance (Fleiss, Levin, & Cho-Paik, 2003). Disagreements occurred in the coding of 23 articles in which either protests from both sides or splits in support were reported. These coding discrepancies were resolved by discussion.

6

Astroturf efforts – “in which the firm secretly subsidizes a group with similar views to lobby” (Lyon & Maxwell, 2004) – are not included as the disclosure of these efforts is not mandated.

7

Handling endogeneity in time hazard models is an issue under development. In particular, the suitability of the 2SRI approach for duration/survival models remains an open question. The results from the estimation of the Extended Cox model with correction for endogeneity are available from the authors. However, these results should be interpreted with caution.

8

The 2SRI approach to control for the endogenous nature of explanatory variables has recently gained popularity, as more traditional approaches (i.e., instrumental variables) are limited in their applicability to nonlinear models. We refer the interested reader to Wooldridge (2015) for an excellent overview.

9

We note that Eqs. (2) and (3) do not allow for any cross-over effects, but given that we use the same set of regressors, the estimation of Zellner’s seemingly unrelated regression or of a bivariate probit model would not yield different coefficients. Thus, including the correlation in the first stage of the model is not relevant.

10

Alternatively, we tested the restriction of equality for the coefficients of these variables (Model 10). The results suggest that the coefficients are not equal (Protests: Prob > χ 2 = 0.000. Contributions: Prob > χ2 = 0.008. Lobbyists: Prob > χ2 = 0.060).

## References

Alvarez, & Glasgow, 1999Alvarez, R. M. & Glasgow, G. (1999). Two-stage estimation of nonrecursive choice models. Political Analysis, 8(2), 147165.

Amenta, & Caren, 2004Amenta, E. & Caren, N. (2004). The legislative, organizational, and beneficiary consequences of state-oriented challengers. In D. A. Snow, S. A. Soule & H. Kriesi (Eds.), The Blackwell companion to social movements (pp. 461488). Malden, MA: Wiley.

Amenta, Caren, Chiarello, & Su, 2010Amenta, E., Caren, N., Chiarello, E. & Su, Y. (2010). The political consequences of social movements. Annual Review of Sociology, 36(1), 287307.

Amenta, Caren, & Olasky, 2005Amenta, E., Caren, N., & Olasky, S. J. (2005). Age for leisure? Political mediation and the impact of the pension movement on U.S. old-age policy. American Sociological Review, 70 (3), 516538.

American Cancer Society (ACS), 2015American Cancer Society (ACS). (2015). The true costs of smoking. Retrieved from https://www.cancer.org/research/infographics-gallery/tobacco-related-healthcare-costs.html

Andrews, 2002Andrews, K. (2002). Movement-countermovement dynamics and the emergence of new institutions: The case of “White Flight” schools in Mississippi. Social Forces, 80(3), 911936.

Andrews, & Caren, 2010Andrews, K. T. & Caren, N. (2010). Making the news: Movement organizations, media attention, and the public agenda. American Sociological Review, 75(6), 841866.

Ansolabehere, Snyder, & Tripathi, 2002Ansolabehere, S., Snyder, J. M. & Tripathi, M. (2002). Are PAC contributions and lobbying linked? New evidence from the 1995 Lobby Disclosure Act. Business and Politics Journal, 4, 131155.

de Bakker, & den Hond, 2008de Bakker, F. G. A. & den Hond, F. (2008). Introducing the politics of stakeholder influence: A review essay. Business and Society, 47(1), 820.

de Bakker, den Hond, King, & Weber, 2013de Bakker, F. G. A., den Hond, F., King, B. & Weber, K. (2013). Social movements, civil society and corporations: Taking stock and looking ahead. Organization Studies, 34(5–6), 573593.

Baron, 2006Baron, D. P. (2006). Competitive lobbying and supermajorities in a majority-rule institution. Scandinavian Journal of Economics, 108(4), 607642.

Baumgartner, Berry, Hojnacki, Kimball, & Leech, 2009Baumgartner, F. R., Berry, J. M., Hojnacki, M., Kimball, D. C. & Leech, B. L. (2009). Lobbying and policy change: Who wins, who loses, and why. Chicago, IL: University of Chicago Press.

Baumgartner, Berry, Hojnacki, Kimball, & Leech, 2014Baumgartner, F. R., Berry, J. M., Hojnacki, M., Kimball, D. C. & Leech, B. L. (2014). Money, priorities, and stalemate: How lobbying affects public policy. Election Law Journal: Rules, Politics, and Policy, 13(1), 194209.

Baumgartner, & Leech, 2001Baumgartner, F. R. & Leech, B. L. (2001). Interest niches and policy bandwagons: Patterns of interest group involvement in national politics. Journal of Politics, 63(4), 11911213.

Berry, Fording, Ringquist, Hanson, & Klarner, 2010Berry, W. D., Fording, R. C., Ringquist, E. J., Hanson, R. L. & Klarner, C. E. (2010). Measuring citizen and government ideology in the U.S. states: A re-appraisal. State Politics and Policy Quarterly, 10(2), 117135.

Berry, Ringquist, Fording, & Hanson, 1998Berry, W. D., Ringquist, E. J., Fording, R. C. & Hanson, R. L. (1998). Measuring citizen and government ideology in the American states, 1960–93. American Journal of Political Science, 42, (1), 327348.

Blanke, & da Costa e Silva, 2004Blanke, D. D. & da Costa e Silva, V. (2004). Tobacco control legislation: An introductory guide (p. 261). Geneva: World Health Organization.

Boies, 1989Boies, J. (1989). Money, business, and the state: Material interests, Fortune 500 corporations, and the size of PACs. American Sociological Review, 54(5), 821833.

Bonardi, Hillman, & Keim, 2005Bonardi, J., Hillman, A. & Keim, G. (2005). The attractiveness of political markets: Implications for firm strategy. Academy of Management Review, 30(2), 397413.

Bonardi, Holburn, & Vanden Bergh, 2006Bonardi, J., Holburn, G. L. F. & Vanden Bergh, R. (2006). Nonmarket performance: Evidence from US electric utilities. Academy of Management Journal, 49(6), 12091228.

Bosi, & Uba, 2009Bosi, L. & Uba, K. (2009). Introduction: The outcomes of social movements. Mobilization, 14(4), 409415.

Boyer, & Ratledge, 2009Boyer, W. W. & Ratledge, E. C. (2009). Delaware politics and government. Lincoln, NE: University of Nebraska.

Burstein, 2014Burstein, P. (2014). American public opinion, advocacy, and policy in congress: What the public wants and what it gets. Cambridge: Cambridge University Press.

Cardador, 1995Cardador, M. (1995). Tobacco industry smokers’ rights publications: A content analysis, American Journal of Public Health, 85(9), 12121217.

Centers for Disease Control and Prevention (CDC), 2012Centers for Disease Control and Prevention (CDC). (2012). Comprehensive smoke-free laws. Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/mm6145a3.htm

Cohen, 2016Cohen, J. (2016). Let smokers see the warning they need. The New York Times, June 3. Retrieved from https://www.nytimes.com/2016/06/03/opinion/let-smokers-see-the-warning-they-need.html

Cohen, Milio, Rozier, Ferrence, Ashley, & Goldstein, 2000Cohen, J. E., Milio, N., Rozier, R. G., Ferrence, R., Ashley, M. J. & Goldstein, A. O. (2000). Political ideology and tobacco control. Tobacco Control, 9(3), 263267.

CRP, 2013CRP, (2013). OpenSecrets, The Center for Responsive Politics. Available at: http://www.opensecrets.org/

Davis, 1991Davis, G. (1991). Agents without principles? The spread of the poison pill through the incorporate network. Administrative Science Quarterly, 36(4), 583613.

Davis, Morrill, Rao, & Soule, 2008Davis, G., Morrill, C., Rao, H. & Soule, S. (2008). Introduction: Social movements in organizations and markets. Administrative Science Quarterly, 53(3), 389394.

Davis, & Thompson, 1994Davis, G. & Thompson, T. A. (1994). A social movement perspective on corporate control. Administrative Science Quarterly, 39(1), 141173.

Derry, & Waikar, 2008Derry, R. & Waikar, S. V. (2008). Frames and filters: Strategic distrust as a legitimation tool in the 50-year battle between public health activists and big tobacco. Business and Society, 47(1), 102139.

Derthick, 2002Derthick, M. (2002). “Up in smoke”: A scathing critique of the role of anti-smoking groups, attorneys general, and private tort lawyers in protecting Big Tobacco. Washington, DC: CQ Press.

Dhalla, & Oliver, 2013Dhalla, R. & Oliver, C. (2013). Industry identity in an oligopolistic market and firms’ responses to institutional pressures. Organization Studies, 34, 132.

Earl, 2004Earl, J. (2004). The cultural consequences of social movements. In D. A. Snow, S. A. Soule & H. Kriesi (Eds.), The Blackwell companion to social movements (pp. 508530). Oxford: Blackwell.

Earl, Martin, McCarthy, & Soule, 2004Earl, J., Martin, A., McCarthy, J. D. & Soule, S. A. (2004). The use of newspaper data in the study of collective action. Annual Review of Sociology, 30(1), 6580.

Earl, Soule, & McCarthy, 2003Earl, J., Soule, S. A. & McCarthy, J. D. (2003). Protest under fire? Explaining the policing of protest. American Sociological Review, 68(4), 581606.

Edelman, Leachman, & McAdam, 2010Edelman, L. B., Leachman, G. & McAdam, D. (2010). On law, organizations, and social movements. Annual Review of Law and Social Science, 6(1), 653685.

de Figueiredo, 2009de Figueiredo, J. (2009). Integrated political strategy. In J. A. Nickerson, B. S. Silverman (Eds.), Economic Institutions of Strategy (Advances in Strategic Management, Volume 26, pp. 459486). Bingley: Emerald Group Publishing Limited.

de Figueiredo, & Silverman, 2006de Figueiredo, J. & Silverman, B. (2006). Academic earmarks and the returns to lobbying. Journal of Law and Economics, 49(2), 597625.

de Figueiredo, & Edwards, 2015de Figueiredo, R. J. P. & Edwards, G. (2015). The market for legislative influence over regulatory policy. In J. M. De Figueiredo, M. Lenox, F. Oberholzer-Gee, R. G. Vanden Bergh (Eds.), Strategy Beyond Markets (Advances in Strategic Management, Volume 34, pp. 193232). Bingley: Emerald Group Publishing Limited.

Fleiss, Levin, & Cho-Paik, 2003Fleiss, J. L., Levin, B. & Cho-Paik, M. (2003). Statistical methods for rates and proportions (3rd ed.). New Jersey, NJ: John Wiley & Sons.

Gilens, & Page, 2014Gilens, M. & Page, B. I. (2014). Testing theories of American politics: Elites, interest groups, and average citizens. Perspectives on Politics, 12(3), 564581.

Giugni, 1998Giugni, M. (1998). Was it worth the effort? The outcomes and consequences of social movements. Annual Review of Sociology, 24, 371393.

Goszkowski, 2008Goszkowski, R. (2008). Among Americans, smoking decreases as income increases. Washington, DC: Gallup.

Grier, 1994Grier, K. (1994). The determinants of industry political activity, 1978–1986. American Political Science Review, 88(4), 911926.

Guérard, Bode, & Gustafsson, 2013Guérard, S., Bode, C. & Gustafsson, R. (2013). Turning point mechanisms in a dualistic process model of institutional emergence: The case of the diesel particulate filter in Germany. Organization Studies, 34(5–6), 781822.

de Guia, Cohen, Ashley, Ferrence, Rehm, Studlar, & Northrup, 2003de Guia, N. A., Cohen, J., Ashley, M., Ferrence, R., Rehm, J., Studlar, D. & Northrup, D. (2003). Dimensions underlying legislator support for tobacco control policies. Tobacco Control, 12(2), pp. 133139.

Hall, & Deardorff, 2006Hall, R. & Deardorff, A. (2006). Lobbying as legislative subsidy. American Political Science Review, 100(1), 6984.

Hausman, 1978Hausman, J. (1978). Specification tests in econometrics. Econometrica, 46(6), 12511271.

Heckman, & Hotz, 1989Heckman, J. & Hotz, V. (1989). Choosing among alternative nonexperimental methods for estimating the impact of social programs: The case of manpower training. Journal of the American Statistical Association, 84(408), 862874.

Hiatt, Sine, & Tolbert, 2009Hiatt, S. R., Sine, W. D. & Tolbert, P. S. (2009). From Pabst to Pepsi: The deinstitutionalization of social practices and the creation of entrepreneurial opportunities. Administrative Science Quarterly, 54(4), 635667.

Hillman, 2003Hillman, A. J. (2003). Determinants of political strategies in U.S. multinationals. Business and Society, 42(4), 455484.

Hillman, & Hitt, 1999Hillman, A. J. & Hitt, M. A. (1999). Corporate political strategy formulation: A model of approach, participation, and strategy decisions. Academy of Management Review, 24(4), 825842.

Hillman, Keim, & Schuler, 2004Hillman, A. J., Keim, G. D. & Schuler, D. (2004). Corporate political activity: A review and research agenda. Journal of Management, 30(6), 837857.

Hoetker, 2007Hoetker, G. (2007). The use of logit and probit models in strategic management research: Critical issues. Strategic Management Journal, 28, 331343.

Hojnacki, Kimball, Baumgartner, Berry, & Leech, 2012Hojnacki, M., Kimball, D. C., Baumgartner, F. R., Berry, J. M. & Leech, B. L. (2012). Studying organizational advocacy and influence: Reexamining interest group research. Annual Review of Political Science. Annual Reviews, 15(1), 379399.

Hojnackia, Marchettib, Baumgartner, Berryd, Kimballe, & Leechf, 2015Hojnackia, M., Marchettib, K. M., Baumgartner, F. R., Berryd, J. M., Kimballe, D. C. & Leechf, B. L. (2015). Assessing business advantage in Washington lobbying. Interest Groups and Advocacy, 4(3), 120.

Hsu, & Grodal, 2015Hsu, G. & Grodal, S. (2015). Category taken-for-grantedness as a strategic opportunity: The case of light cigarettes, 1964 to 1993. American Sociological Review, 80(1), 2862.

Humphreys, 2015Humphreys, K. (2015). Why the wealthy stopped smoking, but the poor didn’t. The Washington Post, January 14. Retrieved from https://www.washingtonpost.com/news/wonk/wp/2015/01/14/why-the-wealthy-stopped-smoking-but-the-poor-didnt/?utm_term=.879cb12ccdf1

de la Inclán, 2008de la Inclán, M. L. (2008). From the ¡Ya Basta! to the caracoles: Zapatista mobilization under transitional conditions. American Journal of Sociology, 113(5), 13161350.

Ingram, Yue, & Rao, 2010Ingram, P., Yue, L. & Rao, H. (2010). Trouble in store: Probes, protests, and store openings by Wal-Mart, 1998–2007. American Journal of Sociology, 116(1), 5392.

Jacobson, Wasserman, & Anderson, 1997Jacobson, P. D., Wasserman, J. & Anderson, J. R. (1997). Historical overview of tobacco legislation and regulation. Journal of Social Issues, 53(1), 7595.

Jasper, & Poulsen, 1993Jasper, J. & Poulsen, J. (1993). Fighting back: Vulnerabilities, blunders, and countermobilization by the targets in three animal rights campaigns. Sociological Forum, 8(4), 639657.

Keim, & Zeithaml, 1986Keim, G. & Zeithaml, C. (1986). Corporate political strategy and legislative decision making. Academy of Management Review, 11(4), 828843.

Kim, Shin, Oh, & Jeong, 2007Kim, T., Shin, D., Oh, H. & Jeong, Y. (2007). Inside the iron cage: Organizational political dynamics and institutional changes in presidential selection systems in Korean universities, 1985–2002. Administrative Science Quarterly, 52(2), 286323.

Kimball, Baumgartner, Berry, Hojnacki, Leech, & Summary, 2012Kimball, D. C., Baumgartner, F. R., Berry, J. M., Hojnacki, M., Leech, B. L. & Summary, B. (2012). Who cares about the lobbying agenda? Interest Groups and Advocacy, 1(1), 525.

King, & Pearce, 2010King, B. & Pearce, N. A. (2010). The contentiousness of markets: Politics, social movements, and institutional change in markets. Annual Review of Sociology, 36(1), 249267.

King, & Soule, 2007King, B. & Soule, S. (2007). Social movements as extra-institutional entrepreneurs: The effect of protests on stock price returns. Administrative Science Quarterly, 52(3), 413442.

Landis, & Koch, 1977Landis, J. & Koch, G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159174.

Leech, Baumgartner, La Pira, & Semanko, 2005Leech, B. L., Baumgartner, F. R., La Pira, T. M. & Semanko, N. A. (2005). Drawing lobbyists to Washington: Government activity and the demand for advocacy. Political Research Quarterly, 58(1), 1930.

Lehman-Wilzig, & Ungar, 1985Lehman-Wilzig, S. & Ungar, M. (1985). The economic and political detemrinants of public protest frequency and magnitude. International Review of Modern Sociology, 15, 6380.

Luders, 2006Luders, J. (2006). The economics of movement success: Business responses to civil rights mobilization. American Journal of Sociology, 111(4), 963998.

Lux, Crook, & Woehr, 2010Lux, S., Crook, T. R. & Woehr, D. J. (2010). Mixing business with politics: A meta-analysis of the antecedents and outcomes of corporate political activity. Journal of Management, 37(1), 223247.

Lyon, & Maxwell, 2004Lyon, T. P. & Maxwell, J. W. (2004). Astroturf: Interest group lobbying and corporate strategy. Journal of Economics and Management Strategy, 13(4), 561597.

Marquis, Guthrie, & Almandoz, 2012Marquis, C., Guthrie, D. & Almandoz, J. (2012). State activism and the hidden incentives behind bank acquisitions. Social Science Research, 41(1), 130145.

McAdam, & Su, 2002McAdam, D. & Su, Y. (2002). The war at home: Antiwar protests and congressional voting, 1965 to 1973. American Sociological Review, 67(5), 696721.

McCammon, Campbell, Granberg, & Mowery, 2001McCammon, H., Campbell, K., Granberg, E. & Mowery, C. (2001). How movements win: Gendered opportunity structures and US women’s suffrage movements, 1866 to 1919. American Sociological Review, 66(1), 4970.

McCarthy, & Zald, 1977McCarthy, J. & Zald, M. (1977). Resource mobilization and social movements: A partial theory. American Journal of Sociology, 82(6), 12121241.

McDonnell, & King, 2013McDonnell, M.-H. & King, B. (2013). Keeping up appearances: Reputational threat and impression management after social movement boycotts. Administrative Science Quarterly, 58(3), 387419.

McDonnell, & Werner, 2016McDonnell, M.-H. & Werner, T. (2016). Blacklisted businesses: Social activists’ challenges and the disruption of corporate political activity. Administrative Science Quarterly, 61(4), 584620.

McKinsey Global Institute (MGI), 2014McKinsey Global Institute (MGI). (2014). How the world could better fight obesity. Retrieved from http://www.mckinsey.com/industries/healthcare-systems-and-services/our-insights/how-the-world-could-better-fight-obesity

McVeigh, Myers, & Sikkink, 2004McVeigh, R., Myers, D. J. & Sikkink, D. (2004). Corn, Klansmen, and Coolidge: Structure and framing in social movements. Social Forces, 83(2), 653690.

Menashe, & Siegel, 1998Menashe, C. L. & Siegel, M. (1998). The power of a frame: An analysis of newspaper coverage of tobacco issues–United States, 1985–1996. Journal of Health Communication, 3(4), 307325.

Meyer, 2004Meyer, D. S. (2004). Protest and political opportunities. Annual Review of Sociology, 30(1), 125145.

Meyer, & Staggenborg, 1996Meyer, D. S. & Staggenborg, S. (1996). Movements, countermovements, and the structure of political opportunity. American Journal of Sociology, 101(6), 16281660.

Meznar, & Nigh, 1995Meznar, M. & Nigh, D. (1995). Buffer or bridge? Environmental and organizational determinants of public affairs activities in American firms. Academy of Management Journal, 38(4), 975996.

Nathanson, 1999Nathanson, C. A. (1999). Social movements as catalysts for policy change: The case of smoking and guns. Journal of Health Politics, Policy and Law, 24(3), 421488.

Oliver, 1991Oliver, C. (1991). Strategic responses to institutional processes. Academy of Management Review, 16(1), 145179.

Oliver, & Maney, 2000Oliver, P. E. & Maney, G. M. (2000). Political processes and local newspaper coverage of protest events: From selection bias to triadic interactions. American Journal of Sociology, 106(2), 463505.

Olzak, & Soule, 2009Olzak, S. & Soule, S. (2009). Cross-cutting influences of environmental protest and legislation. Social Forces, 88(1), 201225.

O’Mahony, & Bechky, 2008O’Mahony, S. & Bechky, B. (2008). Boundary organizations: Enabling collaboration among unexpected allies. Administrative Science Quarterly, 53(3), 422459.

Orzechowski, & Walter, 2014Orzechowski, W. & Walter, R. C. (2014). Tax burden on tobacco. Arlington, VA. Retrieved from https://data.cdc.gov/Policy/The-Tax-Burden-on-Tobacco-Volume-51-1970-2016/7nwe-3aj9.

Pedriana, 2006Pedriana, N. (2006). From protective to equal treatment: Legal framing processes and transformation of the women’s movement in the 1960s. American Journal of Sociology, 111(6), 17181761.

Pertschuk, 2001Pertschuk, M. (2001). Smoke in their eyes: Lessons in movement leadership from the tobacco wars. Nashville, TN: Vanderbilt University Press.

Pescosolido, Grauerholz, & Milkie, 1997Pescosolido, B. A., Grauerholz, E. & Milkie, M. A. (1997). Culture and conflict: The portrayal of blacks in U.S. children’s picture books through the mid-and late-twentieth century. American Sociological Review, 62(3), 443464.

Petrin, & Train, 2010Petrin, A. & Train, K. (2010). A control function approach to endogeneity in consumer choice models. Journal of Marketing Research, 47(1), 313.

Polletta, 2000Polletta, F. (2000). The structural context of novel rights claims: Southern civil rights organizing, 1961–1966. Law and Society Review, 34(2), 367406.

Proctor, 2011Proctor, R. (2011). Golden holocaust. Berkeley, CA: University of California Press.

Rao, 2004Rao, H. (2004). Institutional activism in the early American automobile industry. Journal of Business Venturing, 19(3), 359384.

Roscoe, & Jenkins, 2005Roscoe, D. & Jenkins, S. (2005). “A meta analysis of campaign contributions” impact on roll call voting. Social Science Quarterly, 86(1), 5268.

Samuels, & Glantz, 1991Samuels, B., & Glantz, S. (1991). The politics of tobacco control, JAMA, 266(15), 21102117.

Schneiberg, King, & Smith, 2008Schneiberg, M., King, M. & Smith, T. (2008). Social movements and organizational form: Cooperative alternatives to corporations in the American insurance, dairy, and grain industries. American Sociological Review, 73(4), 635667.

Schneiberg, & Lounsbury, 2008Schneiberg, M. & Lounsbury, M. (2008). Social movements and institutional analysis. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin (Eds.), The handbook of organizational institutionalism (pp. 648670). Thousand Oaks, CA: Sage.

Schneiberg, & Soule, 2005Schneiberg, M. & Soule, S. (2005). Institutionalization as a contested, multilevel process. In G. F. Davis, D. McAdam, W. R. Scott & M. N. Zald (Eds.), Social movements and organization theory (pp. 122160). Cambridge: Cambridge University Press.

Schuler, 1996Schuler, D. (1996). Corporate political strategy and foreign competition: The case of the steel industry. Academy of Management Journal, 39(3), 720737.

Schuler, Rehbein, & Cramer, 2002Schuler, D., Rehbein, K. & Cramer, R. (2002). Pursuing strategic advantage through political means: A multivariate approach. Academy of Management Journal, 45(4), 659672.

Surgeon General (SG), 2014Surgeon General (SG). (2014). The health consequences of smoking: 50 years of progress. Retrieved from https://www.surgeongeneral.gov/library/reports/50-years-of-progress/full-report.pdf

Simons, Vermeulen, & Knoben, 2016Simons, T., Vermeulen, P. & Knoben, J. (2016). There is no beer without a smoke: Community cohesion and neighboring communities’ effects on organizational resistance to anti smoking regulations in the Dutch hospitality industry. Academy of Management Journal, 59(2), 545578.

Smith, & Malone, 2007Smith, E. A., & Malone, R. E. (2007). We will speak as the smoker: The tobacco industry’s smokers’ rights groups, European journal of public health, 17(3), 30613.

Smith, Savell, & Gilmore, 2013Smith, K. E., Savell, E. & Gilmore, A. B. (2013). What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control, 22(2), 144153.

Smith, 2000Smith, M. A. (2000). American business and political power: Public opinion, elections, and democracy. Chicago, IL: University of Chicago Press.

Snow, Della Porta, Klandermans, & McAdam, 2013Snow, D. A., Della Porta, D., Klandermans, B. & McAdam, D. (2013). Encyclopedia of social and political movements. Oxford: Wiley-Blackwell.

Snow, Soule, & Cress, 2005Snow, D. A., Soule, S. A. & Cress, D. M. (2005). Identifying the precipitants of homeless protest across 17 U.S. cities, 1980 to 1990. Social Forces, 83(3), 11831210.

Soule, 2009Soule, S. (2009). Contention and corporate social responsibility. Cambridge: Cambridge University Press.

Soule, 2012Soule, S. (2012). Social movements and markets, industries, and firms. Organization Studies, 33(12), 17151733.

Soule, 2013Soule, S. A. (2013). Protest event research. In D. A. Snow, D. Della Porta, B. Klandermans & D. McAdam (Eds.), Encyclopedia of social and political movements (Volume 3, pp. 10191022). London: Wiley-Blackwell.

Soule, & King, 2006Soule, S. A. & King, B. G. (2006). The stages of the policy process and the Equal Rights Amendment, 1972–1982. American Journal of Sociology, 111(6), 18711909.

Soule, Mcadam, Mccarthy, & Su, 1999Soule, S. A., Mcadam, D., Mccarthy, J. & Su, Y. (1999). Protest events: Cause or consequence of state action? The U.S. women’s movement and federal congressional activities, 1956–1979. Mobilization: An International Journal, 4(2), 239256.

Soule, & Olzak, 2004Soule, S. A. & Olzak, S. (2004). When do movements matter? The politics of contingency and the Equal Rights Amendment. American Sociological Review, 69(4), 473497.

Sweda, & Daynard, 1996Sweda, E., & Daynard, R. (1996.) Tobacco industry tactics, British Medical Bulletin, 32(1), 183192.

Terza, Basu, & Rathouz, 2008Terza, J. V, Basu, A. & Rathouz, P. J. (2008). Two-stage residual inclusion estimation: Addressing endogeneity in health econometric modeling. Journal of Health Economics, 27(3), 531543.

Tesler, & Malone, 2008Tesler, L. E. & Malone, R. E. (2008). Corporate philanthropy, lobbying, and public health policy. American Journal of Public Health, 98(12), 21232133.

Vasi, Walker, Johnson, & Tan, 2015Vasi, I. B., Walker, E. T., Johnson, J. S. & Tan, H. F. (2015). “No fracking way!” Documentary film, discursive opportunity, and local opposition against hydraulic fracturing in the United States, 2010 to 2013. American Sociological Review, 80(5), 934959.

Vogus, & Davis, 2005Vogus, T. & Davis, G. F. (2005). Elite mobilizations for antitakeover legislation, 1982–1990. In G. F. Davis, D. McAdam, W. R. Scott & M. N. Zald (Eds.), Social movements and organization theory (pp. 96121). Cambridge: Cambridge University Press.

Walker, 2009Walker, E. T. (2009). Privatizing participation: Civic change and the organizational dynamics of grassroots lobbying firms. American Sociological Review, 74(1), 83105.

Walker, Martin, & McCarthy, 2008Walker, E. T., Martin, A. W. & McCarthy, J. D. (2008). Confronting the state, the corporation, and the academy: The influence of institutional targets on social movement repertoires. American Journal of Sociology, 114(1), 3576.

Walker, & Rea, 2014Walker, E. T. & Rea, C. M. (2014). The political mobilization of firms and industries. Annual Review of Sociology, 40, 35.

Weber, & King, 2013Weber, K. & King, B. (2013). Social movement theory and organization studies. In P. Adler, P. du Gay, G. Morgan & Reed, M. (Eds.), Handbook of sociology, social theory and organization studies (pp. 487509). Oxford: Oxford University Press.

Wooldridge, 2015Wooldridge, J. F. (2015). Control function methods in applied econometrics. Journal of Human Resources, 50(2), 420445.

Xu, Bishop, Kennedy, Simpson, & Pechacek, 2016Xu, X., Bishop, E. E., Kennedy, S. M., Simpson, S. A. & Pechacek, T. F. (2016). Annual healthcare spending attributable to cigarette smoking. American Journal of Preventive Medicine, 48(3), 326333.

Zald, & McCarthy, 1987Zald, M. N. & McCarthy, J. D. (1987). Social movements in an organizational society. New Brunswick, NJ: Transaction Books.

# Acknowledgments

We would like to thank the volume editors, Forrest Briscoe, Brayden King, and Jocelyn Leitzinger, as well as an anonymous reviewer for their insightful comments. Comments and suggestions from Rodolphe Durand, Candace Jones, Xavier Martin, Christopher Marquis, Renate Meyer, William Ocasio, and Patrick Vermeulen helped us improve this study. We also thank participants of the 2015 NIT workshop in Vienna, participants of the 2015 EGOS and Academy of Management Conferences, and seminar participants at Cornell University and Tilburg University for their insightful comments.

Prelims
Introduction: Integrating Research Perspectives on Business and Society
Section I Social Movements and Organizational Theory
Chapter 1 Plug Power. Social Movements and Electric Vehicle Charging Stations in California, 1995–2012
Chapter 2 Negotiating Moral Boundaries: Social Movements and the Strategic (Re)Definition of the Medical in Cannabis Markets
Chapter 3 Movement-Led Institutional Change: Uncertainty, Networks, and the Diffusion of Contentious Practices in Organizational Fields
Chapter 4 Social Movement Theory’s Contribution to Understanding Activism Around Corporations and Markets
Section II Social Movements and Stakeholders
Chapter 5 The Negative Relationship Between Event-Specific Corporate Social Responsibility and Shareholder Value
Chapter 6 Bridging Social Movement and Industrial Relations Theory: An Analysis of Worker Organizing Campaigns in the United States and China
Chapter 7 Not a Drop to Drink? Drinking Water Quality, System Ownership, and Stakeholder Attention
Chapter 8 Influence Stakeholders, Influence the World
Section III Social Movements and Non-Market Strategy
Chapter 9 On Two Sides of the Smoke Screen: How Activist Organizations and Corporations use Protests, Campaign Contributions, and Lobbyists to Influence Institutional Change
Chapter 10 Failure or Success? Defensive Strategies and Piecemeal Change Among Racial Inequalities in the Brazilian Banking Sector
Chapter 11 Non-Market Strategy and Social Movements Research: What are the Gains from Trade?
Afterword Broadening Business And Society Research: A Postscript on the Limits of Strategic Action
Index