The Legitimacy of Interest Groups
Constituencies, advocacy styles and legitimacy sources
M.L. Kievit (10674977)
Master Thesis Political Theory and Political Behaviour (GSSS, UvA) Final version (resit)
Research Project: Advocacy in EU and World Politics: Who is Represented (and who is not)? Lecturer: Dhr. M.C. Hanegraaff
2
Contents
Introduction 3
Sources of the Legitimacy of Interest groups 6
Research Design 14
Results 21
Conclusion 32
References 35
Appendix A: Coding Manual 38
3
Introduction
Low levels of trust in the current political system, its institutions and its personnel, i.e. politicians, are widespread. This distrust is apparent in many developed countries and at all political levels: Local, national and international (e.g. Gallup, 2015; Campbell, 2009; den Ridder & Dekker, 2015). Different scholars have attempted to provide solutions or at least ways forward, in the sense of improving representation and citizen participation, and rebuilding trust. Some of the proposed ideas are more theoretical, such as Habermas’ deliberative democracy (1994), some are more practical, such as Van Reybrouck’s G1000 citizen meetings (G1000, 2015). A group of actors that could contribute positively to the political process are interest groups. Interest groups are increasingly seen as very important to a well-functioning democratic political process (e.g. Halpin, 2010: 3).
There are many different interest groups active in modern society and they busy themselves with a variety of activities. Per definition, interest groups defend interests and are involved in political advocacy, i.e. interest groups attempt to convince and motivate political actors to support the group’s cause (Beyers et al., 2008: 1106-7). Sometimes, interest groups also provide services for their members or a wider constituency. For example, while trade unions defend and promote the interests of workers, they also provide training, education and advice for their members. By performing all these activities, interest groups fulfil crucial functions in the political process, because they give voice to grievances in society, aggregate concerns, and direct political attention (Halpin, 2010: 3). In a time when party politics are declining (Mair, 2006) and many have lost faith in the current political system, the position of interest groups is deemed increasingly important. Where political parties fail to perform their democratic functions, interest groups can fill part of that vacuum.
In order to be able to perform all these activities, interest groups do not only need resources and opportunities, but also a kind of legitimacy. If the public and government do approve of the activities of an interest group, i.e. these activities are being viewed as legitimate, this can greatly increase the interest group’s capacity to perform the aforementioned functions in the political process. Public support and acknowledgement by the government can both contribute to an interest group’s resources, while legitimacy can also help an interest group to convince the government to support the group’s goals. If a group lacks legitimacy, it will lack
4
support, which often means a lack of resources, and its actions will receive little attention or response.
Interest groups can increase their legitimacy in different ways, which are more extensively described in the next part of this paper. In short, some groups can rely on democratic representation and some cannot. Whether interest groups can employ democratic representation depends on their constituency. An interest group’s constituency exists of those whose interests the group promotes or defends. If the members of the constituency could conceivably (fully and actively) participate in the decision-making process of the group, then the group could potentially utilise democratic representation. However, earlier research has pointed out that interest groups rarely fulfil this potential. If the constituency of an interest group cannot be involved in the decision-making process, then the group will have to look for other ways to attain legitimacy in the eyes of the government and the general public. Not involving the constituency can also be a conscious choice by an interest group. Alternatively, a group might attempt to include constituency members, but simply fail to achieve implementing actual member participation. Interest groups that cannot, choose not to, or fail to employ democratic representation have to use other sources, such as expertise and experience, to attain legitimacy. For example, by showing that they possess relevant expertise and experience, interest groups can increase their legitimacy. However, expertise or experience in a specific area will only increase the interest group’s legitimacy in that specific field of interest group activity. For example, if Greenpeace shows they possess expertise on how to protect whales, they are not automatically also granted legitimacy when talking about household economics. To summarise, interest groups that cannot rely on representation will put more emphasis on expertise and other possible sources of legitimacy. This is what can be expected on basis of the existing theory and research.
The assumption that those interest groups that cannot rely on representation will put more emphasis on expertise and other possible sources of legitimacy has not been extensively tested, however. The aim of this paper is to test whether the strategies of interest groups do indeed follow this logic. Do interest groups that cannot rely on internal democracy indeed rely more on other sources of legitimacy and vice versa? A model was developed on the basis of research by Halpin (2010) and Van Rooy (2004) (see p. 10). The model shows which legitimacy sources interest groups of different categories are expected to use more or use less. The interest groups in the sample were divided into three categories on basis of their constituency: Potentially representative groups (e.g. trade unions), potential solidarity groups (e.g. Greenpeace) and groups representing disenfranchised human constituencies (e.g. Amnesty
5
International). The model is further explained in the subsequent sections of this paper. The research question of this project is: To what extent can the usage of different legitimacy sources by an interest group be predicted by the mobilisation potential of their constituency?
Data from fifty Dutch interest groups has been collected into a dataset. The data exists of website pages and annual reports published by the selected groups. The prominent legitimacy source in each document (web page or annual report chapter) was coded. In this way, insight was created into which legitimacy sources these groups use and in which relative proportions the different sources are used by different kinds of interest groups. This is further explained under the research design.
The paper will start with the theoretical framework explaining the issues with the representation of constituencies by interest groups, and introducing (more extensively) the other possible sources of legitimacy. Next, the research design will be explained, including the models to be tested, hypotheses, the data gathering, data analysis, and the issues that arose while gathering the data. After that, the results will be presented and checked against the hypotheses. The conclusion will summarise the findings and present suggestions for further research.
6
Sources of the Legitimacy of Interest Groups
A definition of legitimacy
Legitimacy is a central concept in this paper, therefore a clear definition is necessary. Unfortunately, legitimacy is a contested concept, i.e. there are multiple definitions and these are sometimes conflicting. There is no overall consensus on what legitimacy entails. In his paper on this particular problem, Stillman offers the following definition: “The compatibility of the results of government output with the value patterns of the relevant systems, i.e. those affected by the results” (1974: 33). As with most definitions of legitimacy, Stillman’s is related to a legitimate government. Nonetheless, with some adjustments, it can be applied to interest groups, especially when they are seen as part of a larger political system, which can be deemed legitimate to a higher or lesser degree. Interest groups are defined as such by three key factors: Organisation, political advocacy, and informality. First, interest group politics concern aggregated individuals’ political behaviour and organised forms of political behaviour. Second, these groups are involved in political advocacy, i.e. they attempt to use political means to reach their goals. Third, interest groups do not aim to have their members elected in parliament or government offices. In this sense they are ‘informal’ (Beyers et al., 2008: 1106-7). However, interest groups can have internal elections. Stilman’s definition can be altered to: The compatibility of (the results of) interest group activities with the expectations of, firstly, the constituency and the group’s members, but also with the expectations of the government and the general public.
Why legitimacy is important for interest groups
Interest groups are often seen as crucial to the political process. They have a function to identify problems within society which require the attention of the government. Interest groups solidify concerns and amplify them, so they receive political attention, and thus focus political attention and action. They can even propose and provide solutions for specific problems (Halpin, 2010: 3). Interest groups can also contribute to creating space for new democratic practices (Kayuni, 2013: 202).
If interest groups are to play such an important role in politics, they need to gain both the attention and the approval of both the government and the general public. Not only for practical reasons, i.e. material support, but one could also argue that interest groups should act in a legitimate way if they are to fulfil such a crucial part in a system which needs to be deemed
7
legitimate by the public in order to function. However, it is clear that not every interest group is regarded equally by the government, nor by the public. For example, the activities of the Ku Klux Klan1 will probably receive a quite different government response than those of a group
defending the interests of Alzheimer patients. This seems obvious, but what causes this difference? Legitimacy is one of the factors which influences how interest groups and their activities are perceived by the public and the authorities.
Interest groups can basically have two kinds of legitimacy: Vis-à-vis their constituency (those whose interests the group promotes or defends), their affiliates (subscribing members) 2 or the broader public; and legitimacy vis-à-vis the government and government agencies. Legitimacy among either a group’s constituency, their members or the broader public means that the group’s activities are seen by one or more of the populations mentioned as legitimate and possibly beneficial, not only to the constituency, but potentially also to the broader political process or to society as a whole. Legitimacy toward government means that the group is regarded by the government and government agencies as a legitimate actor in the political process, and the group consequently gains access to this process. However, the sources of legitimacy investigated in this research could potentially increase both kinds of legitimacy, especially since legitimacy among a broader public increases legitimacy towards government. Not all groups approach the issue of legitimacy in the same way. Groups might not even think in terms of ‘legitimacy’ when devising lobbying tactics and other strategies. In the following paragraphs the different ways in which interest groups can make claims to legitimacy are explained.
Representation and internal democracy
The most obvious way for an interest group to assert legitimacy, is to show that it represents the actual interests of a clearly delineated constituency. This constituency is often simply defined as the affiliates of the group. Groups can show they represent the interests of their affiliates by using internal democracy procedures. Internal democracy procedures exist of opportunities for member participation and mechanisms for linking member input to group policy. Elections for offices can also be held. Multiple authors have noted that relatively little scholarly attention, within the field of interest group research, has been paid to the internal decision-making processes of groups. Our understanding of the internal political machinery of groups remains shallow (Barakso, 2005: 318-19; Bosso, 2003; Kayuni, 2013: 202; Klüver and
1 This might be stretching the definition of ‘interest groups’, but it serves the example here. 2 An interest group’s constituency and the group’s affiliates do not necessarily overlap.
8
Saurugger, 2013: 185; Knoke, 1986: 11). For example, little is known about the practical concerns that can dominate discussions within advocacy organisations (Bosso, 2003: 4). This may be due to the fact that, as Long noted, “people will readily admit that governments are organizations. The converse – that organizations are governments – is equally true, but rarely considered” (ibid., 1962: 110). Without assessing the internal political system of these groups, it is very difficult to determine whether and how well they perform mediating functions between government and citizens (Barakso, 2005: 318-19; Bosso, 2003; Lehman Schlozman and Tierney, 1986: 121; Halpin, 2010: 69). Despite the crucial role of the internal structure of interest groups, “analyses of the relationship between interest groups and civic engagement typically reveal little about the political infrastructure of groups” (Barakso, 2005: 317).
However, the research that has been undertaken has pointed out that the internal structure of interest groups is often not very democratic at all. There is nothing intrinsically democratic about interest groups, and Michels’ ‘iron law of oligarchy’, predicting that, given time, the power within an organisation will come to fall into the hands of a small number of people, seems to have a profound effect on interest groups (Binderkrantz, 2009: 657-59; Halpin, 2010: 24). Although appearing democratic on the outside, many groups are rather oligarchic internally, and only a fraction of the members participates actively. Some groups cannot even be considered democratic in the formal sense, since office holders are not elected or because group policies are not subject to approval by the members (Binderkrantz, 2009: 657-59; Berry, 1977: 187; Halpin, 2010: 18). In a similar vein, Grant concludes that “most pressure groups are not very democratic internally”. For example, executive officers are often elected through non-transparent processes, if they are elected at all. Additionally, the consultation of members may be very limited, and organisations can be very resistant to appeals for more democracy by their members (Grant, 2002: 7). Warleigh (2001) similarly concluded that the internal governance procedures of NGOs are insufficiently democratic. Decisions are often made by key officers, with very little – if any – supporter input (ibid.: 623). According to Grant, all this calls the representative legitimacy of groups into question (Grant, 2002: 7; see also: Lehman Schlozman and Tierney, 1986: 136). If groups are to make a positive contribution to the democratic process, it is argued, they should themselves be democratic (Halpin, 2010: 17).
Democratic representation is not always possible
Not all interest groups can or attempt to be representative in the sense that they directly represent their constituency or members. Many advocacy groups have few members or none at all. A survey of pressure groups in the United States showed that eighty percent of those groups
9
did not have a membership structure (Lowery & Gray, 2000: 8). These groups usually receive their funding from a handful of large donors (Halpin, 2010: 72-73). For these groups, there is no membership whose interests they could represent, though they may have a constituency. Additionally, groups often do not even claim to act as representative of their members’ interests (ibid.: 73). Furthermore, representation can be hard to put into practice. Constituencies can be geographically dispersed, lack resources, be disenfranchised, incapacitated, confined or marginalised otherwise (ibid.: 75-76). Alternatively, interest groups sometimes claim to represent interests of those who have no possibility to speak in their own voice at all. This is the case with groups advocating for non-humans or future generations. Take environmental groups for example. We cannot ask ‘the environment’ what its interests are. In other words, the environment cannot authorise representation (O’Neill, 2001: 494). Hobbes held a similar view, stating that inanimate things cannot grant authority to other beings or institutions, only humans can (1968: 219).
Expected advocacy styles and legitimacy sources
If they do not or cannot meet the democratic demands of the representational frame, then on what other basis can interest groups claim to represent genuine interests and gain access to the political process? As was explained in the introduction, interest groups that cannot rely on internal democracy will make more use of other sources of legitimacy, such as expertise and experience. The expectations are schematised in the following model:
10
The model used in this research is based on the framework proposed by Halpin in which groups appear along a continuum between representation and solidarity (2010: 102). This distinction was suggested earlier by O’Neill (2001: 492). Representational groups represent constituencies which have the possibility to speak with their own voice. Members of these constituencies have the possibility to be included in the decision-making process of the group. For representational groups, affiliates (subscribing members) and constituency (possible beneficiaries of group action) are the same (Halpin, 2010: 100-3). By contrast, solidarity groups represent constituencies that have no possibility at all to speak their own voice, i.e. non-humans and future generations (ibid.: 98-100). These constituencies have no possibility to authorise representation (O’Neill, 2001: 494). Groups can be solidary to these constituencies, but cannot represent them in the traditional sense. Halpin expects that groups positioned towards the solidarity end of the spectrum rely on expertise, solidarity, and empathy to create legitimacy for themselves (Halpin, 2010: 102). However, the latter two terms are hardly explained. Solidarity seems to overlap with the depth of commitment of members (see below). In this research, it will be understood as the commitment of the organisation to the cause. Empathy requires a deeper understanding (possibly through emotional connection) of the interests of the constituency. This understanding might be facilitated by expert knowledge and experience.
Interest groups are expected to adopt either a representational or solidarity advocacy style on the basis of their constituency. Groups that represent a constituency that can be involved in decision-making (e.g. trade unions) are to the right side of the model and thus will use internal democracy and membership arguments more often when compared to potential solidarity groups (e.g. Greenpeace) and groups representing disenfranchised human constituencies (e.g. Amnesty International). Furthermore, potentially representational groups are expected to make more use of internal democracy and membership arguments in comparison to their usage of the other legitimacy sources. Obviously, this argument also works vice versa. Potential solidarity groups are expected to use claims based on solidarity and empathy the most and to make little use of internal democracy and membership arguments. The definitive hypotheses that follow from this model are listed in the at the end of this chapter. Some additional legitimacy sources have been added to Halpin’s model on basis of Van Rooy’s research on international activism (2004). These are explained below.
Alternative sources of legitimacy
If they cannot use internal democracy, because they cannot enfranchise their constituency, interest groups have to look for other sources of legitimacy. First of all, even though they might
11
not be democratically enfranchised in group decision-making, having members alone already grants groups “with a sort of automatic credibility”. The larger the membership of a group the more legitimate it will be in the eyes of policy-makers3 (Van Rooy, 2004: 64). Therefore, even
groups that do not represent the interests of their affiliates might seek to increase their membership, not only for funding, but also to increase their legitimacy. The depth of commitment of members to the cause, i.e. participation in group activity (not necessarily in decision-making), can also lend legitimacy to a group (ibid.: 68-70). Additionally, arguments about the size of the constituency (possible beneficiaries) of the group can also increase legitimacy. General support from the larger public was also included in this category, since the membership argument is based on the assumption that support grants legitimacy.4
Alternatively, Van Rooy suggests that groups can rely on victimhood, expertise, experience and moral authority to increase their legitimacy (ibid.: 77-78; Risse, 2006: 190). Relying on victimhood means that a group represents (or claims to do so) the interests of persons whose lives have been negatively affected by the activities of an outside agency (Van Rooy, 2004: 78).
Expert evidence is assessed on the basis of rarity, validity, comparative credibility, balance, and disciplinary might (ibid.: 81). The higher information scores on these criteria, the more legitimacy it lends to those who have it at their disposal. Rarity concerns whether the information is rare, valuable, comprehensive, specific or exclusive (ibid.: 81-83). However, the information must also be valid, if any legitimacy is to be derived from it. Invalid information, when discovered, can decrease legitimacy (ibid.: 83-85). Disciplinary might concerns the disciplinary credentials of groups and individual activists. For example, the opinion of financial experts on the height of inflation is considered more useful than their view on wild life protection (ibid.: 85-86). Alternatively, groups might simply benefit from a lack of trust in government and business organisations (ibid.: 88). If interest groups are able to weigh a wide range of considerations in their argumentation, this also adds to their legitimacy (ibid.: 90-91).
NGOs can accumulate experiential evidence by working at grassroots level when implementing projects or by collecting witness testimonies. First-hand experience with the context of the interests the group defends lends it more credibility (Van Rooy, 2004: 92-93).
3 Whether this is because politicians are concerned about votes or genuinely care for society’s concerns is
difficult to establish, and irrelevant for this research.
4 Whether the configuration of the legitimacy source categories was adequate is reviewed in the conclusion of
12
NGOs can also rely on their history, i.e. institutional survival, track record and reputation to claim legitimacy (Van Rooy, 2004: 93-94; Hudson, 2002: 414).
Finally, groups can gain legitimacy by claiming to represent the public interest or claiming to defend universal values. When such a standard is adopted into international law, it lends even more legitimacy to the group (Van Rooy, 2004: 94-97). Van Rooy mentions a number of other possible sources of legitimacy, but these are either difficult to effectively employ by groups or are difficult to observe or test for (‘hidden rules’) (2004: 77-79, 98-102, 103-26). Hudson’s research confirms that NGOs rely on their history, moral authority, expertise and experience as well as internal democracy (2002: 414).
Expectations and hypotheses
As explained above, a number of expectations follow from the existing literature about the use of legitimacy sources by different interest groups. These will form the hypotheses to be tested in this research. First of all, the expectation is that potentially representational groups will rely more on internal democracy than on other sources of legitimacy. Thus, for the group of potentially representational groups the proportion of documents pertaining to internal democracy will be larger than the proportion pertaining to other sources of legitimacy (H1a). Second, the expectation is that potentially representational groups will make less use of a legitimacy source the closer it is placed to the solidarity end of the continuum. Thus, potentially representational groups will more often use membership arguments than victimhood, expertise, experience or moral authority, and they will use solidarity and empathy least of all (H1b).
Third, potential solidarity groups are expected to rely mostly on solidarity and empathy (H2a). Fourth, potential solidarity groups are expected to make more use of a legitimacy source the closer it is to the solidarity end of the spectrum. Thus, solidarity groups will more often use victimhood, expertise, experience or moral authority than membership arguments, and they will use internal democracy least of all (H2b).
Fifth, the expectations about which legitimacy sources are used by groups representing disenfranchised human constituencies are the same as for potential solidarity groups, because the representational potential of these groups can only partially be exploited, or sometimes not at all. Thus groups representing disenfranchised human constituencies will more often use solidarity or empathy than victimhood, expertise, experience or moral authority, and they will use victimhood, expertise, experience or moral authority more than membership arguments, and they will use internal democracy least of all (H3).
13
Finally, potentially representational groups are expected to make proportionally more legitimacy claims based on internal democracy and membership than the two other kinds of interest groups (H4a). Potential solidarity groups and those representing a disenfranchised human constituency are expected to make proportionally more statements based on victimhood, expertise, experience, moral authority, solidarity and empathy (H4b). The data is expected to show an association between expected advocacy style and the legitimacy sources used (H4c).
Hypothesis 1a Potentially representational groups: Internal democracy most used legitimacy source
Hypothesis 1b Potentially representational groups: Membership arguments > victimhood, expertise, experience or moral authority > solidarity and empathy
Hypothesis 2a Potential solidarity groups: Solidarity and empathy most used legitimacy sources
Hypothesis 2b Solidarity groups: Victimhood, expertise, experience or moral authority > membership arguments > internal democracy
Hypothesis 3 Groups representing disenfranchised human constituencies: Victimhood, expertise, experience or moral authority > membership arguments > internal democracy
Hypothesis 4a Potentially representational groups: Proportionally more legitimacy claims based on internal democracy and membership than the two other kinds of interest groups
Hypothesis 4b Potential solidarity groups and those representing a disenfranchised human constituency: Proportionally more statements based on victimhood, expertise, experience, moral authority, solidarity and empathy
Hypothesis 4c Association between expected advocacy style and the legitimacy sources used
14
Research Design
Data collection
Thirty-seven interest groups have been selected from those that contributed to the Dutch government report on social organisations from 2014 to which ninety-one organisations in total contributed (Posthumus et al., 2014: 208-10). Those groups that fit the definition of interest groups (Beyers et al, 2008: 1106-7) were selected and they formed the original sample for this research. The condition for being included in the report was that the organisation needed to have at least fifty-thousand members (Posthumus et al., 2014: 68). Since this means that the key social organisations in the Netherlands, i.e. those with assumedly the largest influence and impact, were included, this sample is representative for the interest group population in the Netherlands. The sample was complimented with thirteen environmental groups randomly selected from a list of Dutch environmental groups based on an internet search. The aim of adding these environmental groups was to enlarge the dataset and to add groups to the potential solidarity category, since these were somewhat underrepresented in the original sample. All selected groups have a membership structure, which means they at least possess the possibility for internal democracy and to use membership arguments.
Web pages and annual report chapters were each coded regarding the interest group, their constituency, expected advocacy style and legitimacy sources prominently present in the document. The annual reports are all from 2013 or 2014. Unfortunately, not all groups provided an annual report; of these groups only their relevant web pages have been coded. The annual reports that were available differ greatly. Some consist of a few pages, while others contain more than a hundred. Therefore, the amount of documents per organisation differs greatly.
The constituency of each interest groups was determined on the basis of mission statements of the organisations on their websites. As explained in the theoretical section, the expected advocacy style was dependent on the mobilisation potential of the constituency. Groups whose constituency could be involved in decision-making were coded as potentially representational, groups whose constituencies had no opportunity to speak their own voice whatsoever as potential solidarity groups, and the final category existed of groups representing disenfranchised human constituencies. For some groups, it was not immediately clear who their constituency was. For example, at the beginning of the research it was assumed that the constituency of Milieudefensie was the environment, but this was later changed to the human population as a whole. Other constituencies had more than one defining characteristic; the most
15
prominent one has been selected. For example, the constituency of the Lilianefonds exists of handicapped children. The fact that these children are handicapped is what sets them apart, rather than the fact that they are children, therefore this constituency was categorised under patients and not under children. The constituency of an interest groups leads to expectations about which kind of advocacy it will pursue, either representation, solidarity or representing a disenfranchised human constituency. For example, a group defending the interests of employees is expected to use a representational style, while an environmental group is expected to use solidarity. Some constituencies could be seen as marginalised, e.g. women or the elderly, but in the Netherlands they are certainly not disenfranchised as it is defined in this paper. Therefore, groups defending these constituencies are expected to use a representational style.
For each document - website page or annual report chapter - the most prominent legitimacy source was coded. Each document was entered as a separate item in the dataset. Determining which legitimacy source was prominent was straightforward for most
documents, since only one legitimacy source was mentioned in that document. In other instances, chapter or paragraph titles, or paragraphs introducing the chapter gave a clear indication of the content and which legitimacy source was the most important in a particular document. For documents where there was mention of more than one legitimacy source and the most prominent legitimacy source could not be identified, multiple legitimacy sources were entered into the dataset. This meant that sometimes all legitimacy sources present in a document were coded or only the prominent ones among those.
First the of all, the homepage of each group’s website was analysed. The focus was on the main body of text. The legitimacy source that was most prominent in that text on the homepage was coded. Subsequent website pages providing further information about the group (Over ons, Wie wij zijn, Organisatie), and their mission and method of action (Missie,
Werkwijze) were also coded for each group, if available. For these pages the focus was also on
the main body of text. These pages were selected because they were considered likely to provide legitimacy claims. Coding whole websites (all subpages) was unfeasible timewise and unlikely to provide much information about the use of legitimacy claims, since a lot of the pages purely provide practical information for constituents and members.
Most of the annual reports were downloaded from the group’s websites. This
amounted to thirty-nine of the fifty groups. Where an (recent) annual report was not (readily) available, an email was send to the group requesting a recent annual report. In some instances the group did indeed send back a report (five groups), others had none available (four groups), and a few did not answer (two groups). Each annual report chapter was coded individually,
