Introduction
Public opinion research is one of the key areas of interest for political scientists. This is not surprising because we evaluate political systems, among other things, on how responsive they are to citizen preferences. Despite its importance, interest group scholars have only recently started to systematically incorporate public opinion in their research agendas (Rasmussen et al., Reference Rasmussen, Carroll and Lowery2014, Reference Rasmussen, Mäder and Reher2018; Giger and Klüver, Reference Giger and Klüver2016; De Bruycker, Reference De Bruycker2017). This is an important development. While organized interests play a vital role in the policy process, acting as transmission belts of societal demands and providing vital resources that policymakers need, these policymakers’ chances of staying in office ultimately depend on their support among voters (Denzau and Munger, Reference Denzau and Munger1986; Giger and Klüver, Reference Giger and Klüver2016). Because policymakers simultaneously have to pay attention to the general public and to the many organized groups that advocate their policy preferences, understanding the multiple linkages that connect advocacy groups and public opinion, as well as how these interactions affect policymaking, is crucial for shedding some systematic light on the politics of representative democracy (Rasmussen et al., Reference Rasmussen, Carroll and Lowery2014).
Existing works have so far largely analyzed the two-way channels connecting interest groups and public opinion by focusing on the important issues of interest groups’ mobilization, strategies, and influence. For instance, one group of studies looks at the causal channels that go from public opinion to interest groups, showing that public salience significantly affects patterns of interest group mobilization (Rasmussen et al., Reference Rasmussen, Carroll and Lowery2014), lobbying strategies (Kollman, Reference Kollman1998; Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016), and chances of advocacy success (Smith, Reference Smith2000; De Bruycker, Reference De Bruycker2017; Rasmussen et al., Reference Rasmussen, Mäder and Reher2018). A second group of work looks into the opposite causal relationship, highlighting the various means and strategies through which advocacy groups can affect public opinion and the public salience of policy issues (Andsager, Reference Andsager2000; Dür and Mateo, Reference Dür and Mateo2014, Reference Dür and Mateo2016; Dür, Reference Dür2019).
This growing body of work, however, has overlooked the questions of whether and how public opinion affects advocacy groups’ concerns about their own survival as organizations and what implications these concerns might have for these groups’ strategic choices (but see Dür and Mateo, Reference Dür and Mateo2016; Flöthe and Rasmussen, Reference Flöthe and Rasmussen2019 for an exception). While the interest group literature has long assumed that advocacy groups are primarily motivated by a desire to influence public policy (Hansen et al., Reference Hansen, Mitchell and Drope2005), a growing number of works stress the importance of reformulating this research agenda by assuming that advocacy groups are fundamentally survival-maximizing organizations (Gray and Lowery, Reference Gray and Lowery1997; Lowery, Reference Lowery2007, Reference Lowery2013; Halpin and Jordan, Reference Halpin and Jordan2009; Halpin and Thomas, Reference Halpin and Thomas2012; Dür and Mateo, Reference Dür and Mateo2016). As Lowery (Reference Lowery2007: 46–47) puts it, ‘the most fundamental goal of organizations must be to survive as organizations. All other goals organizations might have are necessarily secondary considerations since failure to survive will preclude them’.
This call for more attention to advocacy groups’ fundamental concern for organizational survival has important implications for the analysis of the public opinion-advocacy groups nexus. For one, it suggests that public opinion could be considered a potentially critical factor influencing these organizations’ survival chances, not only their decision to mobilize or their chances of success. In addition, it offers an opportunity to theoretically and empirically build on the literature, suggesting that interest groups’ strategies should be understood in light of an organization’s influencing-seeking as well as its survival prospects (Lowery, Reference Lowery2007; Dür and Mateo, Reference Dür and Mateo2016). While former studies mostly use the survival logic as a theoretical construct to explain strategic decisions of interest groups, we include a direct measure of an organization’s fear for survival and analyze how this affects interest group decisions.
In this paper, we address these important issues by investigating whether, and under what conditions, public opinion has a bearing on advocacy groups’ assessment of their own survival prospects and then by assessing the extent to which these concerns affect advocacy groups’ lobbying strategies. Does public opinion affect advocacy groups’ own assessment of their survival chances? And in what ways do these concerns affect their lobbying strategies? We address these questions by relying on data from the Comparative Interest Group Survey (CIGS), which includes over 2500 interest group respondents across five countries in the European union (EU), as well as groups working at the EU level. In particular, we focus on advocacy groups’ mortality anxiety, that is, an assessment of the likelihood that an organization will face a crisis threatening its existence (Gray and Lowery, Reference Gray and Lowery1997; Halpin and Thomas, Reference Halpin and Thomas2012). So rather than directly assessing whether public opinion affects the actual survival chances of these organizations, that is, advocacy groups’ actual birth and death rates, we tap into this question more indirectly by assessing whether advocacy groups consider public opinion an important factor influencing their survival prospects. This choice not only allows us to overcome the inherent difficulties of conducting empirical research on organizations that no longer exist, but also, and perhaps more importantly, it enables us to assess whether mortality anxiety has systematic effects on advocacy groups’ lobbying strategies.
More precisely, we analyze the link between public opinion, concerns about organizational survival, and interest group behaviour in two stages. In the first stage, we assess to what extent (different types of) advocacy groups deem public opinion an important factor in their assessment of how likely they are to survive, relative to other potentially critical factors identified in the literature, such as the level of competition with other organizations, changes in media technology, and reductions in public funding. We first provide illustrative evidence showing that public opinion can be plausibly considered an important factor influencing how advocacy groups estimate their chances of survival as organizations. This is an important finding that underscores the potential of broadening the theoretical and empirical scope of existing studies on the sources of interest groups’ mortality anxiety to include public opinion (Gray and Lowery, Reference Gray and Lowery1997; Halpin and Thomas, Reference Halpin and Thomas2012).
We then move to systematically assessing whether public opinion-related mortality anxiety varies across group types. Our analysis suggests that the relationship between public opinion and mortality anxiety varies systematically across types of substantive interests, more precisely between business organizations and citizen groups. This is also an important contribution to the literature on interest groups’ mortality anxiety because it suggests a plausible explanation, that is, to say the varying impact of public opinion on different groups’ mortality anxiety, for the otherwise unexpected observation that citizen groups display higher levels of mortality anxiety than business groups (Halpin and Thomas, Reference Halpin and Thomas2012).
In the second stage, we investigate whether the fact that some advocacy groups consider public opinion a critical element shaping their own assessment of their survival prospects has systematic effects on their relative use of inside and outside lobbying strategies. We expected that citizen groups concerned about the effects of changes in public opinion on their survival prospects would increase their relative use of outside lobbying more than business groups with similar concerns. Our results confirm this expectation, contributing to expanding the literature on the motives driving different lobbying strategies.
The paper proceeds as follows. In the first section, we briefly review the literature on the relationship between public opinion and advocacy groups and then make a case for the importance of focusing on the links connecting public opinion and mortality anxiety. In the second section, we identify a number of potential factors that, along with public opinion, could plausibly bear on advocacy groups’ mortality anxiety and then derive a number of hypotheses that could account for systematic variation in advocacy groups’ concerns about public opinion’s influence on their chances of survival. The third section discusses data collection and then presents the results of our analyses. The conclusion discusses the main findings of the paper and suggests avenues for further research on this important topic.
Public opinion, advocacy groups, and organizational survival concerns
Interest group research has recently turned its attention to the role of public opinion, seeking to uncover the multiple links that connect public opinion to interest groups’ mobilization, strategies, and influence.
For instance, Rasmussen et al. (Reference Rasmussen, Carroll and Lowery2014) rely on an original data set of contributions to European Commission online consultations and show that interest groups participate more actively on issues in policy areas that are regarded as salient by the general public. Kollman (Reference Kollman1998) and Hanegraaff et al. (Reference Hanegraaff, Beyers and De Bruycker2016) focus on interest group strategies and provide systematic evidence that public opinion stimulates outside lobbying, that is, strategies that seek to raise the awareness of broader audiences by communicating political messages through various sorts of public media. Others have analyzed the links that connect public opinion to advocacy groups’ success. For instance, Smith (Reference Smith2000) provides strong evidence that business success in lobbying is constrained by public opinion, even when businesses act in a unified manner. More recently, Rasmussen et al. (Reference Rasmussen, Mäder and Reher2018), relying on a new data set containing information on public opinion, advocacy positions, and policy outcomes, have shown that public support for advocates’ policy preferences crucially affects their chances of success, particularly for interest groups defending diffuse interests. Similarly, De Bruycker (Reference De Bruycker2017) draws on a large-scale content analysis of statements made by European political elites in different media outlets to show that elites tend to address public interests when policy processes are salient to European citizens.
Moreover, a growing number of works show that the causal channels connecting public opinion and advocacy groups’ activity also work in the opposite direction: advocacy groups not only respond to but also sway public opinion (Dür, Reference Dür2019). For instance, Andsager (Reference Andsager2000) shows that interest groups can influence how journalists report the news and then contribute to create a favourable public opinion on issues they care about. Using original data from a survey experiment on the Transatlantic Trade and Investment Partnership, Dür (Reference Dür2019) reaches similar conclusions, showing that interest groups have an impact on public opinion, mainly via the frames they manage to convey. Interestingly, in a study on the campaign against the Anti-Counterfeiting Trade Agreement carried out by citizen groups in the EU, Dür and Mateo (Reference Dür and Mateo2014) connect these two strands of research by showing both how interest groups managed to raise the public salience of the issue and how, in turn, this increased public salience motivated a growing number of interest groups to mobilize. Despite the fact that these works use slightly different conceptual categories (e.g. public salience, issue salience), they all suggest that public opinion and interest groups are linked in important ways in a two-way relationship.
This important body of scholarly work, however, has so far overlooked the perhaps more essential question of how advocacy groups’ fundamental concern for their own survival as organizations might affect these causal channels. A number of important works have drawn attention to the importance of constructing theories of lobbying centred around the assumption that interest groups are organizations whose overriding priority is to survive in a world of scarce resources and, therefore, whose actions are primarily aimed at securing the resources they need to survive (Gray and Lowery, Reference Gray and Lowery1997; Lowery, Reference Lowery2007, Reference Lowery2013; Halpin and Jordan, Reference Halpin and Jordan2009). While it is certainly true that in many instances interest groups act with a view to maximizing their chances of influencing the content of policies, these studies underscore the importance of focusing on how interest organizations must balance achieving their policy goals with critical needs such as organizational survival. Acknowledging that lobbying is often about something other than influence has proved helpful in making sense of otherwise surprising observations. For instance, one can explain why organizations in many instances engage in lobbying despite the fact that they have little or non-existent chances of influencing policies by considering how lobbying activities have a bearing on these organizations’ survival chances (Lowery, Reference Lowery2007, Reference Lowery2013; Rasmussen et al., Reference Rasmussen, Carroll and Lowery2014). Similarly, the choice of different lobbying tactics, strategies, and agendas can in many instances also be better explained by referring to these organizations’ long-term maintenance needs, rather than solely to their desire to affect policy content (Imig, Reference Imig1992; Dür and Mateo, Reference Dür and Mateo2016; Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016). More generally, a theoretical and empirical focus on interest groups’ struggle for survival is important because the volatility of group populations is a continual finding among more recent studies mapping the activity of organized interests (Halpin and Thomas, Reference Halpin and Thomas2012: 216).
These contributions suggest the importance of focusing on interest groups’ fundamental concerns for survival when investigating their links with public opinion. Two issues stand out as particularly important in this context. First, it is important to assess whether, and under what conditions, public opinion is deemed as a fundamental resource that enables lobbying organizations to survive. Second, it is crucial to understand public opinion’s importance as a fundamental resource that bears on these organizations’ assessment of their survival prospects, thereby determining their selection of lobbying strategies. In this paper, we tackle these questions through a two-way approach (see Figure 1). We first explore whether groups care about changes in public opinion for the assessment of their long-term survival prospects. Is public opinion a key factor influencing interest groups’ assessments of their survival prospects, or are there other contextual factors that are perceived to be more important for the maintenance needs of these organizations? In more detail, we also consider whether public opinion-related concerns for organizational survival vary across group types. For instance, are citizen groups more afraid of changes in public opinion than business organizations, given their stronger ties to societal concerns?
Figure 1. Does fear for changes in public opinion affect strategic action?
Yet, we also want to understand whether this public opinion-driven mortality anxiety is important for interest groups’ actions. In other words, we want to know not only whether advocacy groups care about public opinion when it comes to assessing their survival prospects but also whether these concerns affect the political choices they make. In particular, in this paper we concentrate on the choice between inside and outside lobbying. We explore whether the choice to lobby outside is substantially driven by fear of changes in public opinion.Footnote 1 This would mean that outside lobbying should be understood not only as a means to maximize the political influence of organizations but also as a marketing tool to survive as an organization.
Does public opinion affect mortality anxiety?
Given a focus on organizational survival, a number of contextual forces can be thought to directly bear on interest groups’ ability to secure the environmental resources that are essential for their survival (Lowery, Reference Lowery2007). For instance, population ecology studies of interest communities suggest that the level of competition within the interest community in which a lobbying organization is embedded should critically affect its survival prospects (Gray and Lowery, Reference Gray and Lowery1995, Reference Gray and Lowery1997). This is so because density-driven competition within an interest community triggers higher levels of resource partitioning that may lead individual organizations to acquire realized niche resources that fall below the threshold required for survival (Gray et al., Reference Gray, Lowery, Fellowes and McAtee2004). Another important element relates to the degree of access to policymaking these groups enjoy. For instance, secure access to the policy process contributes to organization identity, which in turn enhances an organization’s prospects of survival (Gray and Lowery, Reference Gray and Lowery1997: 28; Halpin and Thomas, Reference Halpin and Thomas2012). Access can be largely thought of as a function of the legal context, that is, the number of legislators and degree of clarity of the jurisdictional boundaries, within which the policy process unfolds. Also, interest groups’ survival chances may be affected by changes in media technology, which may make older organizations less capable of adapting to a new communication environment and hence less likely to survive in a competitive environment that increasingly requires organizations to make extensive use of new media. Moreover, interest groups’ survival prospects are likely to be crucially affected by the availability of funding, which is a function of both members’ contributions and external forms of subsidization – governmental funding or private patronage. Quite intuitively, the more economic resources these groups dispose of, the less we can expect them to be worried about their survival prospects.
In addition to these important factors, can public opinion also be conceived as an essential resource that bears on interest groups’ survival prospects? The existing literature suggests this might indeed be the case. For instance, Lowery (Reference Lowery2007) includes public opinion among the many contextual forces that have the potential to lengthen or shorten the resource arrays from which organized interests construct viable resource niches. On the one hand, a high degree of congruence between an interest group’s agenda and the issues that are perceived as salient by the general public can stimulate greater support from relevant constituencies and increase the likelihood that these organizations can attract external sources of financing (Rasmussen et al., Reference Rasmussen, Carroll and Lowery2014; Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016). On the other hand, there is evidence that the congruence between the position of an advocate and the proportion of the public that has the same position increases the chances of lobbying success (Rasmussen et al., Reference Rasmussen, Mäder and Reher2018). While lobbying success is an important objective per se for these organizations, it can be viewed instrumentally as a means to increase their survival prospects (Lowery, Reference Lowery2007). So because all these factors – constituency support, financing, and influence – more or less directly impinge on these organizations’ survival prospects, it is plausible to expect them to consider public opinion as an additional important factor affecting their maintenance needs.
In addition to the general observation that interest groups should consider public opinion an important factor influencing their maintenance needs, is there any reason to believe that this relationship may vary depending on the organizational characteristics of these interest groups? In the most comprehensive empirical study on the sources of interest groups’ mortality anxiety, Halpin and Thomas (Reference Halpin and Thomas2012) find that citizen groups experience relatively high levels of mortality anxiety. Building on this finding, and drawing on the existing literature, we contend that it seems plausible that citizen groups should be more worried about the potential effects of public opinion on their survival prospects than organizations representing business interests. The logic underlying our argument here relates to a number of factors. One important element concerns the different dynamics underlying these two sets of organizations’ membership support and the resulting differences in the exchange relationship they establish with policymakers. Citizen groups typically rely on a broad-based membership. As a result, as Flöthe and Rasmussen (Reference Flöthe and Rasmussen2019: 6) argue, ‘failure to adapt their views to a shift in the public mood can potentially be costly, as members can withdraw their membership, possibly selecting another organization that better represents their interests, and potential new members may be disincentivized from joining’. Moreover, citizen groups tend to get access to policymaking because of their strong roots in civil society and their ability to transmit information about public preferences that are valuable to secure public support. This means that for citizen groups public opinion is critical to trigger constituency support and secure access to policymakers, both of which are vital resources for an organization’s survival prospects.
Differently, organizations representing special interests are less likely to act as representatives of the public as a whole. Indeed, for these organizations, ensuring organizational maintenance is frequently a question of delivering certain services to the more specific and narrow economic interests they represent (Klüver, Reference Klüver2011). As a result, exchanges between them and politicians are more likely to be dominated by a logic in which expertise and monetary contributions are more important (Rasmussen et al., Reference Rasmussen, Mäder and Reher2018).
Moreover, citizen groups are also more dependent on public funding than business organizations (Dür and Mateo, Reference Dür and Mateo2016; Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016). This is relevant because public donations are much more closely linked to variation in public opinion than the support business associations get from their members (i.e. firms). Hence, it seems eminently plausible that citizen groups should be more concerned than business groups about changes in public opinion because they fear it might affect their sources of income, especially in the long run. To illustrate, if climate change becomes a less important issue in public opinion or more people start to resist stringent environmental legislation (perhaps because the costs become too high), this would ultimately lead to fewer donations for environmental causes, harming environmental citizen groups’ financial positions. For business associations, the link between public opinion and firm membership is far less direct. For instance, if public opinion on environmental issues changes, car manufacturers, chemical companies, or the apparel industry will not cease being members of relevant business associations; and even if they do, the attrition rate would be much less than would be the case for citizens’ environmental groups. In short, there are a number of good reasons to suspect that citizen groups care more than business organizations about how changes in public opinion might affect their maintenance prospects. This leads us to formulate the following hypothesis:
HYPOTHESIS 1: Citizen groups are more worried than businesses about public opinion’s effects on their survival prospects.
Does public opinion-driven mortality anxiety affect lobbying strategies?
Just knowing whether public opinion is an important indicator for advocacy groups’ assessment of their survival prospects is only one part of the story. Perhaps more important is the question of whether public opinion-driven mortality anxiety matters for their behaviour. In other words, we want to look at not only whether groups care about public opinion but also whether they act according to this fear. This is an important question because it helps shed light on whether, in addition to influence-maximization goals, organizational maintenance needs can affect lobbying strategies in systematic ways. As Lowery (Reference Lowery2007: 83) puts it, ‘the determination of goals that come to dominate the selection of lobbying targets and lobbying tactics depends critically on which of the several resources arrays upon which the organization depends faces the most severe challenge’. Recent systematic work on opinion representation shows that citizen groups’ concerns for their own survival might at least in part explain why these groups’ preferences tend to be more closely aligned to public opinion than those of business groups (Flöthe and Rasmussen, Reference Flöthe and Rasmussen2019), which suggests that public opinion-related survival concerns might indeed crucially affect what interest groups want and how they behave to achieve those goals.
In particular, we focus on how public opinion-driven mortality anxiety affects the choice between inside and outside lobbying. This question has attracted significant attention and it lies at the heart of the interest representation debate because it directly touches upon the logics that drive lobbying activities (Holyoke, Reference Holyoke2003; Beyers, Reference Beyers2004; Dür and Mateo, Reference Dür and Mateo2013; Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016). The use of inside lobbying tends to privatize social conflict by relying on direct contacts with policymakers and making these political activities invisible to broader audiences. Outside lobbying, on the contrary, addresses policymakers indirectly and seeks to raise public awareness on particular issues. Both influence-seeking and organizational maintenance perspectives point to the distinction between diffuse and business interests when accounting for this choice, suggesting that diffuse interests have a greater propensity to rely on outside lobbying than the latter set of groups (Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016). From the influence-seeking perspective this is so because they have more political than technical information that they can exchange with policymakers (Della Porta and Diani, Reference Della Porta and Diani1999; Dür and Mateo, Reference Dür and Mateo2013). From an organizational maintenance perspective, this may be due to the fact that outside lobbying more effectively serves the survival-related purposes of increasing constituency support and attracting funding (Binderkrantz, Reference Binderkrantz2008).
The question we address here is whether an organization that is concerned about the possibility that public opinion represents a threat to its survival chances will systematically prefer inside or outside lobbying. In order to address this question, we start by clarifying how public opinion might affect the choice of lobbying strategies and then derive some expectations about how, more specifically, public opinion-related mortality anxiety influences this choice. Broadly speaking, public opinion generates incentives for organizations to go public to make sure they are not perceived as inactive on salient issues. As Hanegraaff et al. (Reference Hanegraaff, Beyers and De Bruycker2016: 574) put it, ‘the higher the public salience, the more an organizational constituency expects the lobbyists to play a role in the public debate. Remaining inactive on salient issues means that supporters may start asking questions on the passive attitude of their representatives’. While this is plausibly true for all kinds of organizations, however, we expect to observe systematic differences across types of groups in the extent to which public opinion-related mortality anxiety affects groups’ relative use of inside or outside lobbying strategies. This is the case because business associations have a much more clearly defined constituency than citizen groups. The well-delineated membership corpus of business interests enables them to accurately target and market their services among their potential supporters whenever contextual forces (e.g. public opinion) are perceived to negatively affect their survival chances. In addition to increasing public exposure through outside lobbying strategies, business groups can thus also equally rely on inside lobbying to cope with public opinion-related mortality anxiety. Moreover, going public can be problematic for these groups because, as Flöthe and Rasmussen (Reference Flöthe and Rasmussen2019: 6) note, being responsive to the concerns of the public could sometimes even be suicidal for them if it entails the risk of alienating their members and supporters. So, we do not expect significant changes in the relative use of these lobbying strategies for business groups that are worried about public opinion’s effects on their survival chances.
Citizen groups, on the other hand, cannot rely on such a stable and defined constituency and therefore need to address a broader audience in their hunt for membership and supporter contributions. Because satisfying both existing and potential members in the general public is more likely to affect their survival (Flöthe and Rasmussen, Reference Flöthe and Rasmussen2019), increasing the relative use of outside lobbying is likely to be considered a suitable approach for citizen groups seeking to cope with fear of survival. As Dür and Mateo (Reference Dür and Mateo2016: 6) argue, ‘citizen groups invest a larger share of their material resources in activities aimed at ensuring survival, that is, retaining existing or recruiting new members and supporters’. Public visibility increases awareness about an organization’s views, which contributes to a positive image among prospective donors, and also decreases the chances that existing members withdraw their membership. So especially when citizen groups face considerable fears of changes in public opinion, it is plausible to expect them to increase their relative use of outside strategies, while business groups, which can also target their constituency more directly, are less likely to increase the relative use of such outside lobbying strategies. Hence, we expect that
HYPOTHESIS 2: Citizen groups put more emphasis on outside lobbying when they are concerned that a change in public opinion will affect their survival chances.
Research design
For this paper, we make use of data gathered in the CIGS (https://www.cigsurvey.eu/). This survey addresses the organizational characteristics and political activities and strategies of interest groups. At the moment, the survey has been sent out and completed in Belgium, Italy, Lithuania, the Netherlands, Slovenia, and Sweden. While it will be conducted in the UK, Germany, Spain, the Czech Republic, and Poland in the upcoming years. Therefore, the countries we take into account in this paper – together with the EU – represent all the polities for which data are currently available. The country coverage of the survey allows us to significantly expand the empirical scope of existing studies focusing on interest groups’ mortality anxiety, which have so far focused only on the US and Scotland (Gray and Lowery, Reference Gray and Lowery1997; Halpin and Thomas, Reference Halpin and Thomas2012). So, we do not only contribute to this literature by considering an important additional variable omitted by earlier works on interest groups’ mortality anxiety, that is, public opinion, but also test our arguments in the context of a broader set of countries.
As for our unit of analysis – individual interest groups – the literature usually distinguishes between a ‘behavioural definition’ (Baumgartner et al., Reference Baumgartner, Berry, Hojnacki, Kimball and Leech2009) of an interest group and an ‘organizational definition’ (Jordan and Greenan, Reference Jordan, Greenan, Jordan and Halpin2012). In the first case, groups are defined based on their observable, policy-related activities and could include firms; in the second case, the ‘interest group’ term is reserved only for non-profit organizations. As for the CIGS, the latter definition is used, and only non-profit organizations have been included in the survey. Overall, the response rate of the survey was 38%, which is relatively high compared to other online surveys (Marchetti, Reference Marchetti2015). Moreover, the response rate is quite evenly distributed across countries. More precisely, response rates were as follows: from lowest to highest: Italy (32%), Slovenia (36%), the Netherlands (38%), Lithuania (40%), and Belgium (41%).
In our analyses we rely on several questions from the survey, which allow us to tap empirically into the questions we are interested in, namely, the relationship between public opinion and advocacy groups’ assessment of their survival chances. More specifically, we carry out two analyses that allow us to address the two main questions in this paper. In the first part of the analysis, we seek to explain whether changes in public opinion are considered relevant for the maintenance of (different types of) organizations. For this, we first rely on a specific question to highlight which factors groups fear the most for their long-term survival. More specifically, the following question was asked: ‘How important are the following challenges for the maintenance of your organization?’ The respondents could indicate a list of 11 items, ranging from ‘not important’ to ‘very important’: legal uncertainties within your areas of interests, changes in media technology, decreased subsidies and public funding, change of public opinion, ageing of constituency, individualization, other ways of spending free time, competition from other organizations, Europeanization/globalization, dealing with cultural diversity, and other forms of political participation. We therefore first explore what factors are mentioned as the most important for the maintenance of the organization. With this exercise, we do not aim to systematically test a hypothesis in the conventional sense, that is, the existence of a relationship between public opinion changes and mortality anxiety. More humbly, we seek to get a sense of whether, in addition to the contextual factors usually considered relevant in the literature, public opinion can also plausibly be considered an important factor influencing how interest groups perceive their chances of survival. This is in itself an important exercise because existing studies on the determinants of interest groups’ mortality anxiety have so far overlooked the role of public opinion (Gray and Lowery, Reference Gray and Lowery1997; Halpin and Thomas, Reference Halpin and Thomas2012), and also because it helps us make a plausible case for looking more systematically into how group-level variables affect the relationship between public opinion and mortality anxiety.
So, in addition, we carry out a systematic empirical test of Hypothesis 1 by assessing whether organizations fear changes in public opinion more than others. For this, we rely on the fourth item of the list, that is, changes in public opinion, to highlight which organizations are most in fear of changes in public opinion. In the explanatory part, we focus on our key explanatory variable, namely, group type. The group-type variable was coded on the basis of the websites of each organization by the researchers of the respective national teams. The coding scheme includes eight categories stemming from the INTERARENA coding scheme: business groups, professional associations, labour unions, identity groups, public interest groups, leisure groups, associations of institutions and public authorities (labelled institutional associations), and a ‘rest’ category. For our purposes, we are interested in the difference between business groups and citizen groups, so we recoded these organizations into three categories (see Beyers and Hanegraaff, Reference Hanegraaff, Poletti and Beyers2017; Berkhout et al., Reference Berkhout, Hanegraaff and Braun2017): business groups (first category), citizen groups (containing identity and public interest groups), and a rest category (all other categories not included in the business or citizen group categories).
In the second part of the analysis, we systematically subject Hypothesis 2 to empirical scrutiny by assessing whether, among the groups that fear changes in public opinion the most, citizen groups are more inclined than business groups to increase their relative use of outside lobbying strategies. In this analysis, we thus use the variable ‘fear of changes in public opinion’ as an independent variable to explain variation in the use of outside strategies. We rely on a question in the survey asking groups about their relative use of outside vs. inside strategies. More precisely, ‘In influencing policy, how do you divide your time between inside strategies, such as contacts with politicians and civil servants and outside strategies, such as media coordinating, letter writing, e-mailing or social media campaigns, rallies and demonstrations?’ The total adds up to 100%.
In both analyses, we control for various factors. First, we control for resources, where respondents could indicate on a seven-point scale the amount of resources the organization disposes of. Second, we control for the level of professionalization of an organization by considering the way it makes decisions, the criteria it uses when hiring staff, the training of staff, and whether employees are inclined to pursue a career within the organization. These are based on the following questions: (1) ‘Organizations like yours can make decisions in different ways, such as consensus among individual members or board members or by voting procedures. Can you please indicate below how your organization primarily makes decisions in the following areas?’ (responses were recoded into decisions made by members, board members, or senior staff); (2) ‘In general, when organizations like yours recruit new staff, they often seek to ensure that the following criteria are met: (a) “an understanding of and a commitment to organizational objectives” and (b) “professional qualification and expertise”. If a candidate does not meet these criteria equally well, which of them do you prioritize?’ (3) ‘Does your organization engage in any of the following staff development activities?’ (listing items answered with no or yes); and (4) ‘In general, do staff tend to advance their careers through your organization or leave it for positions in other organizations?’ (recoded 0 for leaving for other organizations and 1 for advancing a career in the organization). All measures are standardized and then combined. We also control for the subsidies groups receive. For the subsidies variable, we rely on a question where respondents indicated whether they relied on government funding. It could be that groups adjust their strategies to maximize funding opportunities.
Finally, we control for the possibility that interest groups across different countries rely on a different mix of inside and outside strategies (see Weiler and Brändli, Reference Weiler and Brändli2015; Beyers and Hanegraaff, Reference Beyers and Hanegraaff2017). This is important because interest groups based in countries where resources (e.g. potential membership donations and government subsidies) are scarcer, such as Lithuania and Slovenia, may be characterized by a higher overall fear for their survival. For this reason, we control for the country in which interest groups are based (Italy, the Netherlands, Belgium, Lithuania, and Slovenia).
Results
The results consist of two parts, each of which highlights a different perspective on the relationship between public opinion and organizational survival. The first part focuses on the importance of public opinion for interest groups’ assessment of their survival chances vis-à-vis other potential threats interest groups identify. In the second part, we explore the effects of fear of changes in public opinion in terms of the outside strategies interest groups employ.
Does public opinion affect mortality anxiety?
We start with an overview of the many reasons that respondents indicate fearing with regard to the maintenance of their organizations (see Table 1 for summary statistics). As discussed above, a focus on organizational survival suggests that a number of contextual forces can be thought to directly bear on interest groups’ abilities to secure the environmental resources that are essential for their survival (Lowery, Reference Lowery2007). Several of the factors are directly derived from the population ecology studies of interest communities, which suggests that the level of competition within the interest community in which a lobbying organization is embedded, as well as funding opportunities or other forms of organizing activity, should critically affect the survival prospects of advocacy groups (Gray and Lowery, Reference Gray and Lowery1995, Reference Gray and Lowery1997). Others are added based on earlier conversations with interest group leaders, such as dealing with legal uncertainties or changes in media technology. Unfortunately, given the nature of our data, we are unable to consider other potentially important group-type factors bearing on these groups’ mortality anxiety, that is, organizational policy profile, organizational identity, and organizational change (Halpin and Thomas, Reference Halpin and Thomas2012).Footnote 2
Table 1. Summary statistics for analyses in paper
What do respondents indicate that they fear the most, according to the survey results? The list of issues can be seen in Table 2, ranging from the factors mentioned as either ‘important’ or ‘very important’ the most often. The results are based on an extensive set of respondents – 2564 representatives of interest groups – across five countries: Belgium, Italy, Lithuania, the Netherlands, and Slovenia. Interestingly, the issues we expected to be higher on the list on the basis of population ecology theories, that is, competition with other groups and the introduction of other forms or political participation, are rather low. Competition ranks eighth on the list, and other forms of political participation are listed last (11th). Funding opportunities are listed somewhat higher, at number four. Almost half of the organizations see decreasing funding opportunities as a high or very high risk for their survival. Other notable factors are changes in legal uncertainties (second) and changes in media technologies (third). Yet, a change in public opinion is considered, among all factors, as the most important factor that potentially affects the future survival of these organizations. No less than 54% indicate this as an important or very important threat to the maintenance of their organizations.Footnote 3
Table 2. What factors do organizations fear the most for the maintenance of the organization?
Moreover, as indicated by Table 3, this is also the main concern across countries considered, and especially in the Eastern European/Baltic countries in our sample. In Slovenia and Lithuania, no less than two thirds of the organizations consider changes in public opinion as a serious threat to their existence. In Italy, Belgium, and the Netherlands, this is around half. The differences are likely explained by the fact that groups active in Slovenia and Lithuania have a higher overall fear for their survival than in Italy, the Netherlands, and Belgium. In the latter countries, membership and government subsidies should be higher, given their higher levels of domestic wealth. Importantly, however, despite these cross-country differences, changes in public opinion were considered by all these organizations as the number one potential threat to their survival. This clearly indicates the importance of public opinion for interest groups across each of the countries. Of course, our data do not allow the causal mechanism that leads these groups to fear changes in public opinion so much for their survival prospects to be pinpointed exactly. That is, whether a high degree of congruence between an interest group’s agenda and the issues that are perceived as salient by the general public can stimulate greater support by relevant constituencies and increase the likelihood that these organizations can attract external sources of financing (Rasmussen et al., Reference Rasmussen, Carroll and Lowery2014; Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016). Neither does our data allow us to grasp whether the congruence between the position of an advocate and the proportion of the public that has the same position increases the chances of lobbying success, hence securing their long-term survival chances (Rasmussen et al., Reference Rasmussen, Mäder and Reher2018). Despite these limitations, overall these results lend support to Lowery’s (Reference Lowery2007) intuition that, in addition to the population and group-level variables considered so far in the literature, public opinion might be a key factor influencing these organizations’ assessment of their long-term survival chances, thus suggesting it would be worth looking into this question in a more systematic way (see also Dür and Mateo, Reference Dür and Mateo2016).
Table 3. Fear for changes in public opinion for the maintenance of organizations across countries
We do so by turning to discuss whether the relationship between public opinion and mortality anxiety varies across group types. We anticipated observing variation between business and citizen groups in terms of how much they fear public opinion for their survival prospects. For this reason, we use the variable ‘fear for changes in public opinion’ as the dependent variable and include various variables as independent variables. Given that our dependent variable is an ordinal score, we rely on an ordered logistic regression analysis.Footnote 4 Furthermore, as explained in the research design, we control for the resources of an organization, the level of professionalization, whether groups specifically indicate representing members, whether they get government subsidies, and which country they stem from. The results are portrayed in Table 4.
Table 4. Predicting fear of changes in public opinion for the maintenance of organization (ordered logistic regression)
The model is an ordered logit regression with robust standard errors. The dependent variable is the fear respondents indicate to have for changes in public opinion for the survival of the organization. Significance:*P < 0.1, **P < 0.05, ***P < 0.01.
What are the findings? First, it is clear that not all organizations fear changes in public opinion equally. Some stand out. To get a better grasp on these results, we plotted the predicted values of group type (see Figure 2, based on Table 4). The results show that citizen groups fear changes in public opinion the most and significantly more than business groups. Overall, these results are in line with our expectation that citizen groups, which have strong roots in civil society and a greater need to transmit information about public preferences to gain access to policymaking, should have a greater propensity to rely on lobbying strategies that allow them to trigger constituency and public support.
Figure 2. Predicted level of fear for changes in public opinion, by group type.
Again, future research should analyze more precisely what are the exact causal mechanisms at play here. On the one hand, it could be that citizen groups anticipate that changes in public opinion would limit their chances of getting access to the policymaking process, given their claims to represent the majority of public opinion (Rasmussen et al., Reference Rasmussen, Mäder and Reher2018). On the other hand, it could be that citizen groups have a greater fear for their survival through changes in public opinion because this could threaten the financial support citizens provide to them. Either way, the results clearly show that citizen groups fear changes in public opinion more for their survival prospects than business organizations, thereby confirming Hypothesis 1.
Does public opinion-driven mortality anxiety affect lobbying strategies?
In the final section, we explore the other side of the coin, namely, whether public opinion-driven mortality anxiety affects the strategic choices of organizations. More specifically, we subject the proposition that public opinion-related mortality anxiety is more likely to lead to an increase in the relative use of outside lobbying for citizen groups than for business groups to systematic empirical scrutiny. To answer this question, we focus on the relative emphasis with which groups rely on outside lobbying strategies compared to inside lobbying strategies. Because this variable is a proportion, we rely on fractional logit regression. Our independent variable is the fear of changes in public opinion as well as other factors often mentioned as influencing the relative use of inside vs. outside lobbying, including resources, professionalization, whether groups indicate that they represent members, government subsidies, and the country of origin. Table 5 summarizes the results. We present two models: first, the entire model without the interaction between group type and survival prospects (Model 1) and then a model that includes this interaction variable to test Hypothesis 2 (Model 2).
Table 5. Predicting the relative use of outside strategies compared to inside strategies
The model is a fractional logit regression with robust standard errors; the dependent variable is the relative use of inside vs. outside lobbying; significance: *P < 0.1, **P < 0.05, ***P < 0.01.
The table displays a number of interesting results. First, as indicated by Model 1, fear of changes in public opinion would seem to matter for all groups’ relative use of outside lobbying. To visualize this effect, we plotted the predicted values (see Figure 3). Here one can see that there is a significant increase in the relative use of outside strategies when groups fear that changes in public opinion could bear on their existential prospects. More precisely, groups that have a limited fear of public opinion’s effects on their survival prospects display a rather even predicted use of outside and inside lobbying – employing slightly more inside lobbying than outside lobbying. Yet, groups that fear public opinion very much have a predicted use of outside lobbying of 55%. This is a substantial difference, namely, a predicted 10%, in the use of outside lobbying strategies between groups that fear changes in public opinion to a limited extent and groups that fear it very much. This would seem to suggest that variation in fear of public opinion for the interest groups’ perceptions of their survival chances affects the relative use of outside lobbying substantially across all group types.
Figure 3. Predicted use of outside strategies by varying levels of fear for changes in public opinion.
However, the picture changes when we try to disentangle the relative contribution of each group type to this observed overall effect. To do this, we plotted an interaction between group type and public opinion-related mortality anxiety. The results are portrayed in Table 5, Model 2, and in Figure 4 below. Interestingly, and in line with Hypothesis 2, only citizen groups fearing that changes in public opinion might affect their survival chances increase their relative use of outside lobbying. Public opinion-related mortality anxiety does not lead to an increase in the relative share of outside lobbying for business groups. More precisely, at low levels of fear for changes in public opinion, we see no difference in the use of outside and inside strategies between business groups and citizen groups, that is, predicted scores of around 50% inside and 50% outside strategies employed. Yet, at very high levels of fear for public opinion, citizen groups are predicted to rely over 60% on outside lobbying (compared to inside lobbying), whereas for business groups the level remains equal. That is, we see that at low levels of fear for changes in public opinion, there is no difference in the predicted relative use of outside and inside strategies between citizen groups and business organizations. The difference only emerges once groups become more afraid of their survival chances. So the overall effect of public opinion-related mortality anxiety on the interest groups’ relative use of outside lobbying suggested by Model 2 and Figure 3 is actually completely driven by citizen groups. This finding is in line with Hypothesis 2 and has significant consequences for the debate on the determinants of interest groups’ lobbying strategies. For instance, these results suggest that when we want to explain why citizen groups rely relatively more on outside lobbying than business organizations, we should look at not only the influence strategies these groups employ, the type of information they possess, or the type of members they have but also at the broader resource dependencies they face. More specifically, we should consider that what citizen groups fear for their survival prospects, in casu public opinion, could also critically determine their lobbying strategies. Without this knowledge, we could end up with invalid results.
Figure 4. Predicted use of outside strategies by citizen groups and business organizations at varying levels of fear for changes in public opinion.
Robustness checks
To make sure the results are robust, we also checked whether the increase of outside lobbying by interest groups might be associated with a broader interpretation of mortality anxiety, and not just public opinion. Moreover, we wanted to see whether this has a different effect on business groups and citizen groups. To this end, we ran separate analyses for each of the mentioned fears groups indicated to have and analyzed whether they might lead to an increase of outside lobbying efforts for business and citizen groups. This would be a strong indication that it is not necessarily fear of public opinion that triggers outside lobbying but the broader, yet latent, notion of mortality anxiety. The results can be seen in Appendix 1 (only predicted probability plots presented). The results are very interesting. First, it seems that which type of fear groups have matters a lot. In some cases, it leads citizen groups to put more emphasis on outside lobbying, yet in other cases it does not. The same goes for business groups. In some instances we see no effect on their strategic choices, while in other cases it forces them to put more emphasis on inside lobbying. Overall, a clear trend is visible, namely, that citizen groups respond to threats by either doing nothing or focusing more on outside lobbying. Business groups respond to fears by either doing nothing in terms of strategic variation or focusing more on inside lobbying. It seems that fear, if imminent, pushes interest groups towards their more natural habitat. For citizen groups this is the public sphere, while for business groups this is the political sphere. Overall though, these results confirm our general trend that citizen groups, which have more fear for their own survival, are more likely to put more emphasis on outside lobbying. Yet, at the same time, it also highlights that the type of fear is crucial for understanding whether citizen groups will indeed do this or not.
Conclusion
We started this paper by noting that public opinion significantly affects how interest groups behave. This observation is supported by several recent studies showing both how public opinion affects interest groups’ mobilization, strategies, and influence and how interest groups themselves seek to affect public opinion to sway policymakers to pursue their preferred policies. We sought to contribute to this important strand of research by investigating how interest groups’ fundamental concerns for organizational survival may affect the links that connect public opinion to interest groups. Our analysis adds to this debate by showing that (1) public opinion is considered as a crucial contextual factor that bears on these organizations’ survival prospects, particularly citizen groups and (2) public opinion-driven mortality anxiety incentivizes outside lobbying among citizen groups.
Our findings have important implications and suggest avenues for further research. For one, our analysis highlights the importance of shedding light on how public opinion directly bears on interest groups’ assessment of their own existential prospects, not only on their patterns of mobilization, strategies, and influence. So starting from the assumption that interest groups are fundamentally survival-maximizing actors, we show that public opinion represents a key resource and/or contextual factor shaping these organizations’ assessment of how to deal with the ever-present struggle for survival in a world of scarce resources. Conversely, these findings also contribute to expanding the theoretical scope of existing studies focusing on the sources of interest groups’ mortality anxiety, which have so far neglected the role public opinion plays. Future research might want to further investigate to what extent different contextual factors – competition with other organizations, legal context, legislative activity etc. – interact and affect each other in shaping how interest groups strike a balance between their survival and policy-influence concerns.
In addition, we contribute to broadening our understanding of the multiple goals, in addition to hunger for influence, that might be driving the selection of lobbying strategies and tactics (see also Dür and Mateo, Reference Dür and Mateo2016, who come to similar conclusions). Further research could explore three additional important aspects. First, more work is needed to further disentangle the different causal mechanisms connecting public opinion, mortality anxiety, and lobbying strategies. While we have shown here that public opinion-related mortality anxiety incentivizes outside lobbying in the case of citizen groups, the opposite relationship also deserves attention. Perhaps engaging in outside lobbying – which requires adapting to a very competitive new media technology environment – is a reason why citizen groups tend to be more worried about the effects of public opinion changes on their survival chances. The established evidence that citizen groups have a greater propensity to rely on outside lobbying, read in light of this article’s findings, suggests that indeed the causal mechanisms linking public opinion, mortality anxiety, and outside lobbying may go in different directions.
Second, this article does not touch upon the question of whether there is congruence between the positions of groups and public opinion and how this may affect groups’ lobbying strategies. It is prima facie plausible that it should matter whether there is much support for a group’s position in society or whether, on the contrary, the majority of the public is against a particular group’s position. For instance, in cases in which public opinion is substantially different from a citizen group position, one might even expect that these groups will have an interest in avoiding increasing media exposure. Further research could add this important dimension to the analysis of the links that exist between public opinion, mortality anxiety, and interest groups’ strategies.
Third, one could think of additional factors that might fundamentally affect the resource arrays upon which these organizations depend and that might, as public opinion does in the case of citizen groups, come to dominate the selection of lobbying targets. Overall, there is now an increasing literature that has shown the importance of survival imperatives to explain why interest groups do what they do, when they do it, and how they do it (Gray and Lowery, Reference Gray and Lowery1997, Reference Gray and Lowery1998; Nownes, Reference Nownes2004; Lowery, Reference Lowery2007; Halpin and Jordan, Reference Halpin and Jordan2009; Halpin and Thomas, Reference Halpin and Thomas2012, Hanegraaff et al., Reference Hanegraaff, Beyers and De Bruycker2016; Beyers and De Bruycker, Reference Beyers and De Bruycker2018, see overviews in Halpin and Jordan, 2011 and Halpin et al., Reference Halpin, Lowery and Gray2015; Dür and Mateo, Reference Dür and Mateo2016). We think such explanatory factors should be explored more systematically in future research, either directly as independent variables or at least as important control variables when analyzing interest groups’ mobilization, strategies, and influence patterns. Otherwise, there are important reasons to believe that these models could remain underspecified.
Finally, our results have certain normative implications. Perhaps the most striking of these is that interest groups’ public personas might not always be in line with their actual concerns. Other research has already shown that interest groups adjust issue selection to increase funding opportunities (see Bob, Reference Bob2005) or to increasing institutional access opportunities. Our results highlight that interest groups do not always seek the media out of genuine concern for an issue but also to survive as organizations. In other words, they often follow the money, rather than the cause. While this is certainly understandable – without survival, activism or lobbying becomes impossible – it does warrant our attention to critically review what interest groups say in the media, and why they say it.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1755773919000031
