The Review of International Organizations

, Volume 10, Issue 4, pp 465–488 | Cite as

Does stakeholder involvement foster democratic legitimacy in international organizations? An empirical assessment of a normative theory



The involvement of non-state organizations in global governance is widely seen as an important step toward global democracy. Proponents of “stakeholder democracy” argue that stakeholder organizations, such as civil society groups and other non-state actors, may represent people significantly affected by global decisions better than elected governments. In this article we identify a particularly promising sociological variant of this argument, test it against new evidence from a large-scale survey among stakeholder organizations with varying levels of involvement in international organizations (IOs), and find that the suggested stakeholder mechanism for producing democratic legitimacy in global governance does not work. Stakeholder involvement is unproductive for democratic legitimacy in IOs as perceived by stakeholders themselves. We suggest alternative explanations of this finding and argue that empirical analysis is useful for adjudicating normative arguments on the viability of stakeholder democracy in global governance.


Democracy Democratic theory Normative theory Civil society Stakeholder democracy Global democracy Legitimacy Global governance International organizations Non-state actors UN 

A growing proportion of all political decisions are taken through international procedures.1 More than ever, such decisions impact on people and politics within states. This development over recent decades raises new questions about democracy and legitimacy. In so far as democracy is valued among people in domestic politics, it ought to play an important role for the legitimacy of global governance as well (Held 1995). Yet, once the problem of how democracy ought to be reflected in international institutions is confronted, whatever consensus may exist on the general importance of democracy breaks down into a number of alternative conceptions and proposals.2

The debate over democratically legitimate institutions beyond the state has conventionally been summed up as suggesting partly competing “models” of global democracy (Archibugi et al. 2012a; McGrew 2002), that is, simplified packages of ontological assumptions, positive expectations, and normative principles that underpin alternative prescriptions of what would make world politics democratically legitimate. A first wave of debate centred attention on cosmopolitan and federal models of democracy.3 On these views, international politics can be democratized by constructing global political institutions similar to those found within liberal democratic states, for example a parliament, a constitution, and a citizenship for humanity as a whole. Rather soon, however, critics began to object to the lack of realism in these suggestions, and developed alternative models for thinking about global democratization, such as polycentric, pluralist, postmodern, transnational, deliberative, and stakeholder models of democracy.4

The purpose of this article is to examine a key empirical assumption that is common to several of the models invoked as alternatives to those of cosmopolitan and federal democracy. The assumption is that strengthened opportunities for involvement of self-organizing non-state actors, such as civil-society groups, advocacy organizations, and self-appointed representatives, in political decision-making enhance the democratic legitimacy of these decision-making procedures. While shared by several traditions for thinking about global democracy, this assumption is particularly central to the idea of stakeholder democracy in global governance (e.g., Scholte 2004; Bäckstrand 2006; Macdonald and Macdonald 2006; Dingwerth 2007; Steffek et al. 2008; Macdonald 2008; Tallberg and Uhlin 2012).

The stakeholder model of democracy views international politics as the regulation of global affairs by state and non-state actors, including governments, international organizations (IOs), non-governmental organizations (NGOs), companies, and expert networks. These actors meet and decide global issues within a continually evolving set of institutions, partnerships, forums, and formal or informal processes of varying composition and policy scope. The power they exercise can be democratized, according to the stakeholder model, by allowing significantly affected people to become directly or indirectly involved in the exercise of this power. More generally, stakeholder democracy represents a particular attempt at realizing the all-affected principle in normative theory, namely, that everyone affected by a political decision should be able to participate in making that decision.5 In the context of regional and global IOs, where the number of significantly affected people exceeds the number of possible participants, stakeholder involvement is necessarily indirect or representative (e.g., Macdonald 2008: 145). Significantly affected people may be represented or served in politics beyond the state by NGOs (Bäckstrand 2006; Macdonald 2008; Scholte 2004; Sechooler 2009; Steffek et al. 2008), but also by actors such as corporations (Bexell and Mörth 2010; cf. Fuchs et al. 2010), social movements (O’Brien et al. 2000; Smith 2008), and self-appointed spokespersons (Montanaro 2012; Saward 2010, 2011). We use the term ‘stakeholder organization’ to refer to self-organizing non-state actors, including actors of all aforementioned kinds.

By pointing to elements of democratization in stakeholder dialogues, participatory arrangements, and accountability experiments in global governance today, stakeholder theorists have fought off staple criticisms directed against the cosmopolitan model of democracy, namely that democracy beyond the state is a utopian project and that it misses the importance of existing national communities (e.g., Bäckstrand 2006; Dingwerth 2007; Macdonald 2008). The sensitivity in the stakeholder model towards empirical realities is reflected also in its preferred conception of legitimacy. Rather than assessing the legitimacy of global governance institutions through comparisons with moral ideals of justice and democracy in the abstract, some leading proponents of the stakeholder model have recently made explicit an assumption that democratic legitimacy should be evaluated in light of sociological features of stakeholder involvement. Such sociological matters include democratic credentials in political institutions: as perceived by ordinary people (Saward 2010); as corresponding to norms held by real stakeholders (Macdonald 2012); and as reflected in conceptions of democracy practiced by global activists (Scholte 2013, 2014). From the perspective of this sociological variant of stakeholder theory, inclusion or representation in politics of significantly affected people is not an end in itself (or not only an end in itself). Instead, it is a strategy to create democratic legitimacy in the sense of experiences and perceptions among stakeholders that politics is adequately democratic, for instance, with regard to democratic credentials such as representation, accountability, and deliberation in global governance.6

In this article, we address the sociological variant of stakeholder democracy and examine the central assumption needed for that theory to make sense, namely, that greater opportunity for involvement of self-organized stakeholders in global policy-making contributes to experiences and perceptions among these actors that global policy-making is democratic. In contrast to previous assessments of the stakeholder model, which have examined its conceptual, descriptive, or normative propositions,7 we analyse this causal assumption empirically. To this end, we surveyed almost 300 randomly selected stakeholder organizations with varying levels of involvement in one global and three regional IOs, namely, the United Nations (UN), the African Union (AU), the Organization of American States (OAS), and the Council of Europe (COE). These stakeholder organizations, which represent various stakeholder groups, were asked to evaluate their opportunities to participate as well as to express their perceptions of the democratic credentials of these IOs in terms of deliberation, accountability, and representation.

We find that stakeholder organizations recognize high levels of democracy in these IOs. However, in contrast to the assumption underlying sociological stakeholder theory, our results reveal that beliefs in the democratic legitimacy of IOs among stakeholder organizations are not strengthened by opportunities for participation in these IOs. Instead, democratic legitimacy in these IOs is shaped by the level of democracy in the home countries of stakeholder organizations and the self-perceived influence of these organizations over policy outcomes. Hence, the stakeholder strategy of creating democratic legitimacy in global governance by strengthening opportunities for involvement of stakeholder organizations appears ineffective for creating experiences of democracy in IOs.

What do these results imply for the viability of sociological variants of stakeholder democracy in global governance (e.g., Macdonald 2012; Saward 2010; cf. Scholte 2013)? We suggest two alternative interpretations. If the reason why stakeholder involvement does not produce democratic legitimacy is that stakeholder groups presently are a weak force in global governance, the principles of the stakeholder model need not be revised, since the normative ideal can be protected by promoting the model more forcefully in political practice. (The advantage of the stakeholder model to construe democracy as a realistic ideal in today’s global governance disappears on this interpretation of the results, however.) On the other hand, if the reason why stakeholder involvement is unproductive for democratic legitimacy is that stakeholder organizations are unrepresentative of relevant global constituencies, then sociological stakeholder democrats need to reconsider their core theoretical idea of creating democratic legitimacy through direct involvement of self-organizing agents.

This critique of the sociological variant of stakeholder theory has implications for stakeholder democracy more generally (e.g., Steffek et al. 2008; Bäckstrand 2006; Macdonald 2008; Tallberg and Uhlin 2012; Dingwerth 2007). Sociological stakeholder democracy is not just any variant of the stakeholder model, but constructed to neutralize objections that risk undermining non-sociological variants of the model as well, notably, that stakeholder democracy does not sufficiently protect the value of political equality (cf. Macdonald 2012) or distinguish democratic from non-democratic political representation (cf. Saward 2010).8 Our results suggest that the strategy of turning to sociological variants of stakeholder theory offers limited or no sustainable solution to those problems in non-sociological variants of the model.

The argument follows in three parts. First, we elaborate on the definition and causal assumptions of the stakeholder model in light of its merits in normative theory. Second, we explain the design of the empirical study and report the results of our statistical analysis. Third, we suggest different explanations of our results and discuss their normative implications.

1 The theory of stakeholder democracy

This section introduces the theory of stakeholder democracy in four steps. We define the stakeholder strategy of democratization, present its merits in normative theory, recognize a sociological variant of stakeholder theory, and present our approach for assessing this version of the stakeholder theory empirically.

1.1 The stakeholder strategy of democratization

Different variants of stakeholder theory are unified by the assumption that strengthened opportunities for involvement of self-organized stakeholders in political procedures hold significant promise for making those procedures more democratic. This assumption in stakeholder theory also defines what we refer to as the stakeholder strategy for global democratization, namely to strengthen opportunities for stakeholders to engage directly or indirectly in the political procedures of global governance. Stakeholders are persons or groups with significantly affected interests, who may be directly included in political procedures or indirectly represented by NGOs, philanthropic foundations, business associations, labor unions, and even private companies. Because of the close affinities between stakeholder and civil society models of global democracy (e.g., Scholte 2004; Bäckstrand 2006; Macdonald 2008), the concept of stakeholder representatives is commonly understood as excluding state actors, such as governments and intergovernmental IOs. The stakeholder model of global democracy offers an alternative to, not a justification for, purely inter-governmental procedures (e.g., Macdonald 2008: 141).

What we designate as the stakeholder strategy of global democratization is a general and widespread idea. Scholars suggest that “the political activities of non-state actors [in global governance] need not threaten democracy, but rather have the potential to satisfy certain rigorous democratic standards” (Macdonald 2008: 6),9 and that “civil society associations do indeed offer significant possibilities to increase democratic accountability in global regulatory arrangements” (Scholte 2004: 213). More specifically for IOs, it is allegedly “essential for a democratic procedure” that civil society organizations have “institutionalized access” because this is “the only way to secure that stakeholders’ arguments can be voiced” (Steffek et al. 2008:10). The same general idea receives support among policy-makers as well. For instance, former UN Secretary-General Boutros Boutros-Ghali characterized NGOs as “a basic form of popular representation in the present-day world,” arguing that “their participation in international organizations is, in a way, a guarantee of the political legitimacy of those international organizations” (quoted in Götz 2008: 244).

Beyond involvement of self-organizing non-state actors in global governance, the literature on stakeholder democracy features a range of context-specific democratization strategies, including economic support for stakeholder representatives, legal protection of citizen rights, corporate codes of conduct, and information-sharing systems (Bäckstrand 2006; Macdonald and Macdonald 2006; Scholte 2011). For purposes of theory development, however, privileging opportunities for stakeholder involvement in policy-making is motivated for two reasons: this strategy is central in political and academic debates on stakeholder democracy (as seen above) and it speaks directly to the unique merits of stakeholder democracy in normative theory (highlighted in the next section).

1.2 The normative merits of the stakeholder strategy

The normative principle underpinning the stakeholder strategy of democratization is that people significantly or potentially affected by decisions should be included or represented in the making of those decisions (e.g., Bäckstrand 2006: 474; Macdonald 2012: 47; Montanaro 2012: 1094). The aim of the formulation is to protect the value of autonomy or self-determination also in contexts where traditional principles of democracy, like political equality and control of the agenda by the people, may lack straightforward application or imply undesirable actions.

To begin with, in the present global governance system, it may seem unfeasible to copy paste institutions from national democracies into global politics. To approximate the normative ideal of stakeholder democracy by means of electoral institutions (as suggested in cosmopolitan and federal models, e.g., Held 1995; Tännsjö 2008; Marchetti 2012), the electorate would ideally have to change depending on the issue to be decided in order to reflect the significantly affected. Making such changes every time political issues with new patterns of decision-makers and decision-takers appear in global governance seems unrealistic. In comparison, the stakeholder strategy of strengthening opportunities for involvement is a more feasible way to attain the stakeholder normative ideal of including or representing all significantly affected.

Opportunities for stakeholder involvement may also seem preferable to establishing a world government by general elections for reasons in ideal normative theory. All people in global politics do not affect each other in significant ways, and many people do not want to be affected by others more than they already are. If we impose the same electoral rights and legal duties on everyone, or on some predefined group of people, under such conditions, many people will be included in global political organizations from which they prefer to be excluded. The stakeholder strategy may then seem normatively more attractive in comparison with general elections and universal suffrage even as an ideal, because it allows people who perceive themselves to be significantly affected by global decisions (that is, not all people) to self-organize and involve in the making of those decisions (that is, not in all global decisions).

For reasons of both feasibility and normative justifiability in ideal theory, therefore, opportunities for stakeholder organizations to become involved in global governance may be a preferable alternative to electoral representation. Moreover, from a liberal perspective on stakeholder democracy (e.g., Macdonald 2008), all people – rich and poor, women and men – are the best judges of their interests and will organize politically if and to the extent that those interests are significantly affected. Hence, self-organized participation yields fair representation. From a more radical perspective (e.g., Hardt and Negri 2012), all political power corrupts and creates a social class of its own, and non-elites must continually self-organize and re-conquer political institutions in order to constitute democracy again and again. On different theoretical accounts, therefore, involvement of self-organizing social actors seems useful for creating democratically legitimate global governance.

By recognizing what institutions seem unrealistic to establish, while at the same time yielding recommendations for greater democracy supported in normative theory, the stakeholder model appears sensitive to real-world dynamics in global governance. Because of this strength, the model is typically presented as a promising candidate for the organization of democracy beyond the state. However, its ultimate success remains to be proven in practice (see, in particular, Bäckstrand 2006; Scholte 2004, 2011).

1.3 A sociological variant of stakeholder theory

In recent years, prominent theorists have developed the idea of stakeholder democracy by recognizing its empirically contingent sociological components and, more specifically, the normative value of democracy as experienced by political actors themselves. This introduction of sociological arguments has occurred in response to theoretical problems rarely addressed in earlier accounts of stakeholder democracy.

Familiar objections against stakeholder democracy in normative theory include the limited capacity of the model to promote political equality, to distinguish democratic and non-democratic representatives, and to formulate a desirable ideal in the first place (e.g., Agné 2006; Christiano 2012; Gould 2004; Marchetti 2012; Näsström 2011). More specifically, it has been questioned: whether political involvement of stakeholders can occur without transposing economic inequalities in civil society into political inequalities (Bäckstrand 2006; Scholte 2014); whether it is legitimate that, for example, Bono plays a role in global governance simply because he claims to “represent a lot of people [in Africa]” (cf. Dryzek and Niemeyer 2008: 481; Montanaro 2012 and Saward 2010 use the same example, emph. added); and whether it is really undemocratic or intrinsically bad to be affected by decisions that you have not participated in making (Agné 2006; Näsström 2011).

It is in response to these and similar critiques against the model that some leading stakeholder theorists recently have preferred a sociological understanding of democratic legitimacy. Democratic legitimacy is then viewed as an experience or a perception of democracy among real political actors, while imposing few or no a priori conceptual limitations on their judgments about what democracy may and may not entail (e.g., Macdonald 2012; Saward 2010, 2011). Democratic legitimacy in this sociological sense is not necessarily vulnerable to the problems of stakeholder democracy identified in ideal normative theory, since the normative value of stakeholder democracy first and foremost is determined by real political experiences. For example, if the principle of political equality is violated in practice, it does not affect legitimacy as understood in sociological stakeholder theory as long as the violations do not engage real people. And if in fact those violations of political equality mobilize real people, the stakeholder model has the conceptual resources needed to suggest how global governance institutions should be designed to channel those mobilized interests, that is, by strengthening opportunities for political involvement of self-organized actors. Turning towards a sociological understanding of democratic legitimacy thereby helps stakeholder theorists to defend the model against criticism in ideal normative theory.10

Some of the foremost contributions to the field well illustrate the adoption of this sociological variant of stakeholder theory.11 Saward (2010: 145), for instance, recognizes the complexity of representation in stakeholder democracy: “Political representation can occur in a number of different and unpredictable places and spaces, involve a wide range of unelected as well as elected actors, and in principle can work through different sorts of institutional structures (or none).” Observing this complexity leads him to notice what has led others to question the theory itself, namely, that the unpredictability of representation seems to “undermine efforts to clarify how democratically legitimate representation might occur” (ibid.: 145). Saward himself, however, makes a partially sociological proposition to clarify the normative issue in question: “provisionally acceptable claims to democratic legitimacy across society are those for which there is evidence of sufficient acceptance of claims by appropriate constituencies under reasonable conditions of judgments” (ibid.: 146, italics in original). What matters for democratic legitimacy, in this view, is not fulfilment of a specific criterion, but acceptance as democratic by particular groups of people on the basis of criteria decided in large parts by the people themselves.

Macdonald (2012) takes a similar stance when first admitting that there is “some real bite” in the critique that the stakeholder model is “poorly equipped to ensure equal input into political decision-making in reality for all individuals” (Macdonald 2012: 54–55, italics in original), and by then invoking as a solution a definition of legitimacy in terms of real actor perspectives: “legitimacy is concerned not with the abstract moral justifiability of institutions as formulated in philosophical terms, but rather with their actual political acceptability, given some background set of sociological facts about the identities that political actors possess through their shared participation in these institutions” (ibid., 57, italics in original). She makes the same point in more concrete terms when arguing that the “the stakeholder strategy of subjecting existing political agencies (states, international organizations, corporations, NGOs and so on) to direct legitimacy through democratic control is compatible with legitimacy being a normative quality concerned with political justifications offered among existing political agencies, rather than with political implementation of more deeply transformative moral projects” (ibid., 62). It is clear that these arguments do not distinguish between normative and sociological conceptions of legitimacy (pace, e.g., Buchanan and Keohane 2006). On these accounts, sociological and normative legitimacy imply each other (cf. footnote 6 above).

1.4 Assessing the sociological legitimacy of stakeholder democracy

Because the sociological variant of stakeholder democracy avoids problems detected in the normative theory of non-sociological variants, it is particularly interesting to put the sociological variant of the model to an empirical test. To do so, we compare global governance institutions with varying opportunities for stakeholder involvement and the democratic credentials of those institutions as perceived by stakeholder organizations. Perceptions of democratic credentials among these organizations are a first and critical step in a chain of legitimation that ends with perceptions among stakeholders themselves. If involvement of stakeholder organizations does not create legitimacy for global governance even among those organizations, it will have little chance to make a difference for legitimacy in the eyes of the actual stakeholders further down the road. A link between stakeholder involvement and legitimacy in this sense is a necessary but insufficient condition for the stakeholder argument to hold.12 Empirical support for that link would not ultimately confirm the theory, while lack of empirical support would disconfirm it. For instance, the validity of our test does not depend on whether stakeholder organizations are accurate and disinterested transmitters of information between IOs and wider publics (cf. Steffek et al. 2008), as our test addresses an earlier step in the causal chain that sociological stakeholder theory assumes.13

In this light, we distinguish between two elements in the stakeholder model. First, there is an institutional element, namely, that stakeholder representatives are allowed to be involved in policy-making. This is what constitutes the stakeholder strategy of democratization. Second, there is an element of perception, namely that the institution is experienced as democratic by stakeholder organizations. This is the sociological concept of democratic legitimacy in the stakeholder model. The key question, then, is whether the stakeholder strategy of democratization fosters democratic legitimacy for an institution among stakeholder representatives. Hence, we are interested in effects of a feature of political institutions – their openness to involvement by stakeholder organizations – on the perceived democracy of those institutions among stakeholder organizations.

This methodology ought to be considered fair, or even generous, for testing sociological variants of the stakeholder model. Directly asking stakeholders about their political perceptions is motivated either because their experiences of democracy are what counts (Saward 2010; cf. Scholte 2013, 2014) or as a means to factor their roles and identities into the analysis of what democracy demands in the global context (Macdonald 2012; cf. Scholte 2013). If stakeholder involvement does not at all foster democratic credentials as perceived among stakeholder representatives, it follows either that stakeholder involvement is a defective institutional strategy for real legitimacy or that real actors have identities that do not support legitimation through involvement of stakeholder representatives – two alternatives that should raise questions about the usefulness of the stakeholder strategy. In these arguments, therefore, normative commitment to the stakeholder strategy of democratization comes with an implication for positive theory that stakeholder involvement (whether direct or representative) causes democratic legitimacy.

We will address democratic legitimacy based on three prominent values in theoretical and practitioner discussions of stakeholder democracy, namely, accountability, representation, and deliberation (operationalized in the next section). Hardly any discussion of stakeholder democracy ignores all these values, and most discussions address two or all three of them (e.g., Bäckstrand 2006; Dingwerth 2007; Steffek et al. 2008; MacDonald 2008; Uhlin 2010). There are appealing reasons in normative theory for this to be so. Accountability, in the sense of political scrutiny and possibility to sanction power-wielders, contributes to limited government and division of power. Representativeness, in the sense of similarity along politically relevant dimensions among decision-makers and ordinary people, integrates rulers and ruled into a common system of governance. Deliberation, in the sense of communicative rationality fair to all parties in face of a common problem, facilitates political decisions that everyone can accept and are willing to comply with. The three values all belong to the input or throughput dimensions of legitimacy, not to the output dimension (Dingwerth 2007; Uhlin 2010). We thereby focus on perceptions of procedural features inherent in democracy. Output legitimacy, constituted by the quality of public policies, is more commonly treated as external to the concept of democracy (but normatively important for other reasons) (e.g., Dingwerth, 2007: 15; Fuchs et al. 2010: 47).

The popularity in stakeholder theory of these democratic values also proceeds from the commonly held expectation that their manifestations in practice depend on the involvement of stakeholder organizations in political procedures. Stakeholder involvement is expected to raise the accountability of global governance institutions by shortening the link between policy-makers and significantly affected citizens (cf. Dellmuth and Tallberg 2014), facilitating scrutiny through new information on the actions of policy-makers, and enabling stakeholder organizations to participate in the potential sanctioning of policy-makers (Scholte 2004, 2011; Grant and Keohane 2005; Wenar 2006). Likewise, involvement of stakeholder organizations is expected to raise representativeness by including positions and interests otherwise underrepresented in global governance institutions, for example, transnational interests, national minorities or weak groups, and people opposing present territorial or constitutional state structures (Macdonald 2008; Scholte 2014). Finally, involvement of stakeholder representatives is expected to raise deliberative qualities in global governance institutions by: allowing new expertise and perspectives into political discussions; facilitating recognition of similar interests in different states, thereby undermining fixed government interests and predisposing governments towards genuine learning; and teaching state representatives how political problems can be solved through exchange of ideas (Bäckstrand 2006; Dryzek and Niemeyer 2008; Steffek et al. 2008).

2 Empirical analysis

In this section we present our research design, the measurement of our theoretical concepts, and the empirical results.

2.1 Research design

To examine whether stakeholder involvement increases the perceived democratic legitimacy of global governance, we use new data from a survey among stakeholder organizations with an expressed interest in four different IOs. The survey was conducted between December 2011 and July 2012. The survey approach allows us to collect data about the varying levels of involvement of stakeholder organizations in IOs and their evaluations of the democratic credentials of these organizations, as well as relevant background information. We focused on stakeholder organizations’ involvement in IOs, rather than other fora in global governance, since IOs are important sites of international power, and since this choice allows us to control for potentially confounding explanatory factors pertaining to institutional context.

To begin with, we conducted a telephone survey among stakeholder organizations with varying levels of involvement in one global organization, the UN. The UN is a crucial case for examining the link between stakeholder organizations’ opportunities for involvement in an IO and their evaluations of the democratic credentials of that organization. The UN has a long history of interaction with stakeholder organizations (Weiss and Gordenker 1996; Willetts 1997). The possibilities for NGOs to become involved in the work of the UN through consultative status with the Economic and Social Council (ECOSOC) was one of the earliest examples of stakeholder access to IOs, and the UN remains among the most accessible organizations (Tallberg et al. 2013). Hence, the UN offers an organizational context with unusual room and a long-term strategy for stakeholder input (Lenz et al. forthcoming, Patterns of international organization: Task specific vs. general purpose, unpublished). In addition, it is the global IO for which both policy-makers and academics have most often highlighted stakeholder involvement as a strategy of democratization (e.g., Willetts 1997; UN 2004; Götz 2008). Specifically, we asked 303 stakeholder organizations to participate in our UN survey and arrived at 224 completed questionnaires, which amounts to a completion rate of 74 %.

In addition, we conducted a survey among stakeholder organizations with an expressed interest in three other IOs: the AU, the COE, and the OAS. The inclusion of these IOs allows us to assess whether the results from the UN hold for multi-issue IOs in different world regions, with variation in levels of domestic democracy and economic development. Due to feasibility constraints, this survey was conducted online. We asked 221 stakeholder organizations interested in the policy-making of these three IOs to participate in the survey, and received 67 completed questionnaires, corresponding to a completion rate of 30 %.14

With regard to the phone survey, we contacted stakeholder organizations interested in UN policy-making in order to identify the person that is most competent to answer questions about the organization’s involvement in a particular IO. As a result, most respondents have a long-standing experience in working in international affairs, with more than half of the respondents having worked in their organization for more than a decade, increasing our confidence in the validity of the data.15 For the web survey, we identified the liaison officer or head of stakeholder organizations by looking at the organizations’ websites, where possible. Otherwise we contacted stakeholder organization secretariats via email and provided a description of the study and a link to the online questionnaire. Each respondent was asked 21 closed- and open-ended questions pertaining to the body it was most involved in within an IO. In this way we capture the respondents’ opinions on processes they have experiences of, rather than general impressions from the IO as a whole (see Online Appendix C for a list of these bodies).

We selected respondents for the survey by randomly sampling from two specific lists of stakeholder organizations. The first list includes all stakeholder organizations accredited to the AU, the COE, the OAS, and the UN in 2011.16 Once accredited, these actors enjoy varying levels of formal and informal involvement in the IO, depending on the body they are most active in. The second list contains stakeholder organizations that indicated an interest in one of the four IOs in the Yearbook of International Organizations Online in 2011 (Union of International Associations 2011), but are not accredited to the organization. By combining random samples from these two lists, we arrived at a total sample of stakeholder organizations with varying levels of involvement in these four IOs, from formally accredited actors with extensive involvement to non-accredited actors with no involvement.17

2.2 Measurement of the democratic legitimacy of IOs

To capture democratic legitimacy in a sociological sense, we code three different variables based on respondents’ agreement with specific statements about the democratic character of IO bodies. First, we code a variable representation, based on the statement that “most people significantly affected by decisions taken by a body are represented on that body.” Second, we create a variable deliberation, using the responses to the statement that “actors in the body usually listen to the views and arguments of others, even if they do not agree.” Third, accountability is a variable based on the statement that “decision-makers in the body usually are held accountable for their decisions.” All three variables take on four values, ranging from 0 (strongly disagree) to 3 (strongly agree).18 These variables are only weakly positively correlated, indicating that they capture empirically distinct aspects of democratic legitimacy (see Table B2 in Online Appendix B). Moreover, each statement reflects a weak interpretation of representation, deliberation, and accountability in order to make it easy for respondents to evaluate democratic qualities of IOs positively and, in this respect, to capture even minor effects of stakeholder involvement.

2.3 Measurement of IOs’ openness to stakeholder involvement

Having established measures of democratic credentials of IOs, we turn to the measurement of potential explanatory factors. To measure IOs’ openness to stakeholder involvement, we asked respondents to evaluate their organizations’ opportunities to be involved in different IO bodies on a scale from 0 (no opportunities) to 3 (many opportunities).19 We asked respondents about the opportunities for involvement of their own organizations and not about involvement opportunities in general in order to have well-informed and valid answers. At the same time, this information is indicative of opportunities for stakeholder involvement in particular IO bodies more generally, as the access arrangements that shape opportunities for involvement normally apply to all stakeholder organizations in a particular IO body (Tallberg et al. 2013). This is an advantage in the context of our study since we seek to capture – in line with sociological stakeholder democratic theory – the opportunities for involvement of stakeholder organizations in general, not the opportunities for involvement of particular stakeholder organizations. In addition, the survey approach (as opposed to coding of formal access rules in IO bodies) has the advantage of allowing us to capture opportunities for involvement as shaped by both formal rules and informal practices, both of which are relevant for stakeholder democracy. In sum, the variable stakeholder organization involvement intends to capture the general opportunities for stakeholder organizations to be involved in particular IO bodies. We generate this information through a survey question posed to stakeholder organizations about their perceived level of involvement in the IO- bodies in which they are most involved.

2.4 Measurement of control variables

In addition to the variables of main theoretical interest mentioned above, we include a range of control variables. In doing so, we test in the first place whether there is an effect of opportunities for stakeholder involvement under some conditions only. It would set unreasonably high standards for stakeholder theory, as we see it, to expect that stakeholder involvement fosters democratic legitimacy under all possible conditions. More realistic and fruitful would be to assess whether such effects arise depending on contextual factors specific to stakeholder organizations, the home countries of these organizations, and the policy context in which they are active. Consequently, we analyze interaction effects between stakeholder involvement and four actor characteristics, broadly conceived: whether the stakeholder organization is an NGO or not (stakeholder organization type); how big it is in terms of staff (stakeholder organization size); how much policy influence the stakeholder organization thinks it has (stakeholder organization influence); and how important the political issues under debate are to state actors within the body (issue salience). Moreover, we analyse interaction effects between stakeholder involvement and a set of institutional features, namely the level of democracy in the home country of the stakeholder organization (domestic democracy), and whether stakeholder involvement takes place in the phases of policy formulation, decision-making, implementation or enforcement.

While the main reason for including these factors is to examine their conditioning effects – that is, whether involvement of stakeholder organizations fosters legitimacy under some but not all conditions – there are theoretical reasons to expect that these factors have independent effects as well. Hence, we include all these variables as controls in the regression models. NGOs, in contrast to business associations or banks, may pursue normative democracy-enhancing agendas (O’Brien et al. 2000), which is likely to make them relatively more sensitive to both democratic credentials and deficits in IOs. Moreover, large and resource-rich stakeholder organizations may be less attentive to democratic qualities of IOs than smaller actors whose influence depends on the extent to which the IO is democratic. Stakeholder organizations that perceive their own influence over IO polices to be relatively high may rate the democratic legitimacy in these organizations higher. Policies that reflect one’s preferences are a source of legitimacy in itself (Hurd 2007). Furthermore, when a topic is politically salient to states, national governments are more likely to behave and organize on the basis of self-interest and relative power than on the basis of democratic principles, which may reduce stakeholder organizations’ perceptions of democratic legitimacy.

With respect to institutional characteristics, democracy in the home country of stakeholder organizations may influence their perceptions of democratic credentials in IOs. A highly (poorly) democratic country of origin may raise (lower) the democratic standards against which a stakeholder organization evaluates an IO. Finally, stakeholder organizations may have different perceptions of democracy in IOs depending on the policy phase in which they are involved. For example, deliberation may be perceived as a particularly relevant democratic value in agenda-setting, representation in decision-making, and accountability in decision-making and enforcement.

The controls are measured as follows. The stakeholder organization type is measured through a dichotomous variable that equals one if a stakeholder organization is an NGO and 0 if otherwise. Stakeholder organization size is measured through its number of full-time employees. Stakeholder organization influence is measured through a survey question that asks respondents to rate the overall impact of their organization on policy-making on a scale from 1 (no impact at all) to 10 (extremely high impact) in the IO body they are most involved in. Issue salience is a categorical variable that captures the extent to which respondents agree with the statement that most issues debated are extremely important to many states in the body in question. Responses range from 0 (strongly disagree) to 3 (strongly agree). The four different policy phases – policy formulation, decision-making, implementation and enforcement – are measured through a survey question on whether a stakeholder organization is active or not active in each policy phase. Domestic democracy in the home country of the stakeholder organization is measured through levels of democracy in the countries of stakeholder organisations’ headquarters, using an additive eleven-point index derived from the Polity IV dataset. This index is based on measures of the competitiveness of political participation, the openness and competitiveness of executive recruitment, and constraints on the chief executive (see Teorell et al. 2013).

2.5 Results

Table 1 reveals that respondents, irrespectively of the IO they evaluate, perceive the deliberative qualities of IO bodies to be the most developed, while accountability comes in second place and representation in third. A vast majority of respondents (50–100 %) somewhat or strongly agree that decision-makers listen to the views of others even if they disagree. Furthermore, a clear majority of respondents across all IOs (63–73 %) somewhat or strongly agree that decision-makers are held accountable. Finally, less than half of the respondents across all IOs (33–45 %) somewhat or strongly agree that the body they are most involved in performs well in terms of representing those significantly affected by its decisions. These differences among aspects of legitimacy may reflect that accountability and especially representation require demanding institutional reforms while deliberation occurs also in non-institutionalised practices. Furthermore, stakeholder organizations in the AU tend to be the most positive, and stakeholder organizations in the OAS the least positive. Conformant with the multivariate results reported below, this might be because stakeholder organizations in the AU are more satisfied with the democracy they get through the IO, as they have less democracy at the domestic level than stakeholder organizations in the other IOs.
Table 1

Evaluations of representation, deliberation, and accountability in four IOs

Items measuring legitimacy

Share of respondents that somewhat or strongly agree with one of the three statements, respectively (Number of observations)





Representation. “The significantly affected are represented.”

45 % (5 out of 11)

35 % (7 out of 20)

33 % (2 out of 6)

43 % (53 out of 124)

Deliberation. “Decision-makers listen to other views even if they don’t agree.”

100 % (11 out of 11)

100 % (20 out of 20)

50 % (3 out of 6)

89 % (110 out of 124)

Accountability. “Decision-makers are held accountable.”

73 % (8 out of 11)

65 % (13 out of 20)

67 % (4 out of 6)

63 % (78 out of 124)

N = 161. The other two answer categories that respondents could choose were somewhat disagree or strongly disagree

We now examine whether the democratic legitimacy of IO bodies depends on the opportunities for involvement that they offer stakeholder organizations. Do stakeholder organizations with no or few opportunities for involvement perceive IO bodies to be less democratically legitimate than stakeholder organizations with some or many opportunities for involvement?

Table 2 reveals that there are no systematic differences in the assessments of stakeholder organizations with limited or extensive involvement For example, about 44 % of respondents with no or few opportunities for involvement agree that the body they are most involved in performs well in terms of representation while about 41 % of those with some or many opportunities for involvement think positively of the body’s performance in terms of representation. The same holds for the democratic criteria of deliberation and accountability, and it is confirmed when calculating correlations. Stakeholder involvement is neither correlated with representation (r = 0.015) nor with accountability (r = –0.005), and only very weakly correlated with deliberation (r = 0.203).
Table 2

Relationship between opportunities for involvement and democratic legitimacy

Items measuring legitimacy

Share of respondents that somewhat or strongly agree with one of the three statements, respectively (Number of obs.)

Percentage difference


Opportunities for involvement

No or few

Some or many

Representation. “The significantly affected are represented.”

44 % (17 out of 39)

41 % (50 out of 122)


Deliberation. “Decision-makers listen to other views even if they don’t agree.”

85 % (33 out of 39)

91 % (111 out of 122)


Accountability. “Decision-makers are held accountable.”

62 % (24 out of 39)

65 % (79 out of 122)


N = 161. The other two answer categories that respondents could choose were somewhat disagree or strongly disagree

Furthermore, to examine whether opportunities for stakeholder organizations to be involved in IO bodies have an independent effect on the different measures of democratic legitimacy, we estimate a series of ordinary least squares (OLS) regression models that allow us to hold other variables constant. Table 3 shows the regression results.20 We estimate two models for each indicator of democratic legitimacy, one with and one without the controls for the institutional context. The most important result is the absence in all models of a systematic relationship between stakeholder organization involvement and the measures of democratic legitimacy. Furthermore, we do not find evidence that the absent effect of stakeholder involvement on democratic legitimacy is conditional upon specific characteristics of the stakeholder organizations. More specifically, we estimated each model in Table 3 by including interaction terms between stakeholder organization involvement and each of the independent variables separately. None of these interaction terms were statistically significant, indicating that an effect of stakeholder involvement is not contingent on: different types, sizes and influence levels of stakeholder organizations; the level of democracy in the home country of the stakeholder organization; or the phases of the policy cycle in which a stakeholder organization is active. In sum, our results suggest that stakeholder involvement does not improve the perceived democratic legitimacy of the four IOs.
Table 3

Regression analysis of representation, deliberation and accountability











Stakeholder organization involvement

 Stakeholder organization involvement













Stakeholder organization characteristics

 Stakeholder organization type













 Stakeholder organization size













 Stakeholder organization influence













 Issue salience













Institutional characteristics

 Policy formulation




















































 Domestic democracy


























 Explained variation in the dependent variable in % (adj. R2)







N = 161. Figures are unstandardized coefficients, with standard errors in parentheses. Significance levels: * p < .10, ** p < .05, *** p < .01

What accounts for this negative finding? One possible explanation would be that stakeholder organizations, as they become more involved, gain a better picture of the actual practices of IO policy-making, and then become disappointed. Global governance offers many examples of NGOs disillusioned by “participatory governance” in IOs (O’Brien et al. 2000; Steffek et al. 2008). On this interpretation, strengthening opportunities for involvement of stakeholder organizations does not reduce a democratic deficit that precedes their involvement in IOs. An alternative explanation would be that those stakeholders who mobilize to become involved with IOs have particularly high demands on democracy, compared to non-mobilized stakeholders, and therefore are particularly tough on IOs in their assessments. However, this suspicion is contradicted by the very positive overall evaluations of deliberation, accountability, and representation reported earlier (see Table 1). Stakeholder organizations – whether little or much involved – are not disappointed with democracy in the first place. On these grounds, the simplest interpretation of the result seems to be the most accurate: the level of stakeholder involvement does not affect the democratic legitimacy of the IOs under study.

If stakeholder involvement does not matter for the democratic legitimacy of these IOs, then what does? Among the alternative explanations, the coefficients for stakeholder organization size are negatively significant at the 5 % level in models 1, 2, 5 and 6, indicating that larger stakeholder organizations think more negatively of the democratic legitimacy of IO bodies in terms of representation and accountability. To illustrate, a stakeholder organization with more than 201 permanent staff members scores about 0.6 points lower on the representation and accountability scales, respectively, than an organization lacking or having only one or two permanent staff members.21 This supports the expectation that small and less resourceful actors are more appreciative of democratic features that enhance equality and prevent abuse of power. Furthermore, the coefficient for stakeholder organization influence is positively significant at the 5 % level in models 3 and 4. An NGO that thinks it has extremely high impact scores about 0.6 points higher on the deliberation scale than an NGO perceiving to have no impact at all, suggesting that the more influence actors think they have in an IO body, the more positively they evaluate the deliberative qualities of that body. Moreover, we find that issue salience has a positive and significant effect on all measures of democratic legitimacy, contrary to the expectation. Moving issue salience from its minimum to its maximum raises NGOs’ score on the representation scale by about 0.6, on the deliberation scale by 0.4, and on the accountability scale by 0.8. In other words, in the perceptions of stakeholder organizations, IO bodies are more democratic in terms of deliberation, representation, and accountability, the more states care about the policy issues under consideration. This finding contradicts the widespread suspicion that democracy is something that states “can afford” when real interests are not at stake (while the finding supports arguments that states are essential for democratic legitimacy at international levels, e.g., Brown 2011; cf. Mitzen 2005). Finally, we find that the variable domestic democracy is negatively significant at the 5 % level in model 2, indicating that stakeholder organizations lacking experiences with democratic institutions in their home country think more positively of the representative qualities of IOs. Respondents from undemocratic countries (scoring 0 on the democracy variable) evaluate representation with roughly 0.7 points lower than respondents from democratic countries (scoring 10 on the democracy scale). Likewise, domestic democracy and evaluations of accountability are negatively related at the 10 % level in model 6, with NGOs from undemocratic countries scoring about 0.5 points lower on the accountability scale than NGOs from democratic countries. This suggests that previous experiences with low levels of democracy in domestic politics induce stakeholder organizations to evaluate IOs’ accountability more positively. For these stakeholders, democracy in global governance, even when weak, may present a valued rescue from political repression at the national level.

We do not find any empirical support for an effect of stakeholder organization type on perceived democratic legitimacy, or that the policy phase in which stakeholder organizations are involved would shape their assessments.

2.6 Robustness of the results

Several robustness checks were conducted to examine whether the regression results hold. First, all relevant variables have been entered separately in the regression equation, although only some of the independent variables are weakly correlated and none of them are moderately or highly correlated (see Table B2 in Online Appendix B). The results of these analyses are similar to the ones reported in Table 3 and do not change our interpretation.

As a second robustness check, we examined the effect of stakeholder organization involvement on democratic legitimacy within the specific IOs by controlling out the variance in the data that is due to differences between IOs. In other words, we tested whether unobservable institutional factors in the specific IOs may affect stakeholder organizations’ perceptions of the democratic credentials of IOs. We do so by replicating all models with fixed effect dummies for IOs that account for all the IO-specific variance. The results remain robust and suggest that the data can be pooled into a single population, as we did for the purpose of the regression analyses reported in Table 3.

Third, to address the same concern, we re-ran all models with robust standard errors clustered at the level of IOs. This robustness check adjusts the covariance matrices for potential within-IO correlations. The results indicate that potential unobserved dependencies between organizations involved in the same IO do not bias the results reported in Table 3.

Fourth, we adjusted the data for unequal selection probabilities by applying base weights, since stakeholder organizations were selected from two sampling frames.22 All models discussed in this section were conducted with and without base weights. Weighting the data did not affect the results.

Finally, we replicated all models using ordered logit regression analysis given that the dependent variables are categorical, and similar results were found.

3 Conclusions: Scope conditions and normative implications

Recent decades have witnessed an intense debate on whether and how global governance may be democratic. The first wave of this debate was dominated by cosmopolitan and federalist suggestions that democratic institutions, such as a parliament, constitution and citizenry, should be created at the global level. More recently, many scholars have emphasized the democratic merits of ongoing involvement of stakeholder organizations in global governance. In this article, we have evaluated a sociological variant of this stakeholder model by investigating the consequences of stakeholder involvement for perceptions of democracy in the UN, the AU, the COE, and the OAS. Our empirical analysis demonstrates that stakeholder involvement in these IOs has not strengthened their democratic legitimacy as perceived by stakeholder organizations themselves.

How far can we expect these results to generalize? Based on the selection of IOs, we have reason to believe that our findings may say something general about the relationship between stakeholder involvement and democratic legitimacy in global governance. As one of the earliest IOs to open up to stakeholder organizations, and still one of the most generous in terms of opportunities for involvement (Tallberg et al. 2013), the UN is a crucial case among IOs. If opportunities for stakeholder involvement have not improved the democratic legitimacy of this IO, then it is unlikely to generate such effects in other IOs as well. The inclusion of the three regional IOs in the analysis lends additional support to this expectation. Not only do the AU, the COE, and the OAS differ from the UN in their level of governance and histories of stakeholder involvement, but also from each other in terms of institutional design and member state characteristics. While we cannot prove that our results travel beyond these four organizations, the selection of IOs and the robustness of the results suggest that these findings should be taken seriously in debates on democracy in global and regional governance.

For both sociological and non-sociological variants of stakeholder democratic theory, these findings come at a vulnerable moment in time. The concepts of autonomy and participation in non-sociological stakeholder theory have been widely criticised for being vague, unhelpful, and sometimes normatively misleading (see section on theory above). It was in attempts to resolve these problems that some leading authors on stakeholder democracy recently turned to a sociological conception of legitimacy – and this variant of the theory has now failed its first empirical test.

While it is tempting to conclude against this background that the stakeholder model as such is ready for abandonment, we believe this would be premature. Stakeholder theory is a multifaceted construct of which not all variants have been assessed in this article. The potential for democratization through transnational discourses (Dryzek and Niemeyer 2008), social movements (O’Brien et al. 2000; Hardt and Negri 2012), and corporate social responsibility (Bexell and Mörth 2010) has not been analysed here. Global democratization has been treated as a problem of creating legitimate institutions (e.g., Steffek et al. 2008; Macdonald 2008), not as a problem of structural power (e.g., Hardt and Negri 2012) or insufficient normative theory (e.g., Erman and Näsström 2013). We have examined the democratizing effect of involving stakeholder representatives in policy-making (Macdonald 2012), not the responsiveness of policy-makers to stakeholder demands (cf. Steffek et al. 2008: 10-12).

In this light, our aim is limited to stimulating theoretical discussion about the viability of both sociological and non-sociological conceptions of stakeholder democracy. To do so, we end by suggesting two potential reasons why opportunities for stakeholder involvement in IO bodies did not foster democratic legitimacy in our study, and identify their different implications for the viability of sociological stakeholder theory. One should notice, however, that these implications have indirect relevance also for non-sociological variants of stakeholder theory. In case the sociological variant of the model fails, it can no longer be invoked as a solution for the normative problems identified in non-sociological variants of the theory.

The absent effect of opportunities for stakeholder involvement on democratic legitimacy may be explained either by the limited power of stakeholder organizations in relation to states, or by their limited representativeness vis-à-vis stakeholder communities. The stakeholder model of democracy finds itself in more or less trouble depending on which of these explanations we judge to be most accurate.

In the case of the first explanation, the absent effect on democratic legitimacy depends on the inability of stakeholder groups to change the course of action decided by states. According to this explanation, today’s stakeholder organizations cannot engage globally with any realistic prospects of changing policies, at least not within IOs. The destiny of stakeholder organizations in global governance is to become exploited by states to produce information and deliver services that states demand.

This explanation of our empirical result does not contradict the normative logic of the stakeholder model in its sociological variant. What would be needed for stakeholder involvement to foster democratic legitimacy are politically more powerful stakeholders and stakeholder representatives. In other words, the stakeholder model does not work yet because there has been too little stakeholder democracy in global governance so far, and this should not count against the normative attractiveness of the model itself (at least not in discussions of ideal normative theory, where lack of realism in suggested reforms does not pose an insurmountable problem to the validity of arguments). It is possible that stakeholder democracy works better where stakeholder organizations decide on policy among themselves, as in the regulation of internet names and numbers (Chango 2011), or in contexts where structural economic inequalities are less accentuated than in today’s IOs (cf. Scholte 2013). If so, the theoretical challenge for stakeholder democrats is a relatively minor one, namely, to specify the empirical conditions under which stakeholder involvement fosters democratic legitimacy.

In the case of the second explanation, the absent effect is explained by poor representative qualities among the stakeholder organizations. This explanation suggests that stakeholder organizations themselves recognize imbalances in their representation of stakeholders, perhaps due to resource and mobilization restrictions in many countries (cf. Bäckstrand 2006; Uhre 2013; Persson 2007). Stakeholder organizations involved in IOs may therefore be pessimistic in their assessments of the democratic qualities of exactly those IOs in which they are involved. This explanation may be reflected in the relatively low levels of perceived representativeness, in comparisons with the higher levels of perceived accountability and deliberation in IOs.

If this explanation is correct, the normative case for the sociological variant of stakeholder democracy is weakened or perhaps lost, depending on whether the explanation can be successfully integrated with stakeholder theory. On this interpretation, the model must designate institutional means to safeguard against imbalances in representation. However, this step is potentially a slippery slope towards models of global democracy that rely more exclusively on formal institutions, for example cosmopolitan and federal democracy. If the absent effect of opportunities for stakeholder involvement on democratic legitimacy depends primarily on the non-representativeness of stakeholder organizations, the stakeholder model must reconcile its notion of democratic control through self-organizing actors with the use of top-down instruments for guaranteeing the representativeness of stakeholder organizations – and the possibility to do so remains to be proven in theory.

In sum, different explanations of our main empirical finding yield different implications for the viability of the sociological variant of the stakeholder model of global democracy. Whether stakeholder democracy, as a consequence of our results, should be reconceptualised, abandoned, or promoted further in practice, ultimately depends on future empirical and theoretical analyses. Research on stakeholder democracy in global governance therefore needs to broaden its agenda. Unless a theoretically revised version of the sociological variant of the model can be supported empirically in some international political context, it will deserve a less prominent role in future studies of legitimacy beyond the state than it has today. And if so, non-sociological variants of the stakeholder model will face limited opportunities to escape from the troubles that they already confront.


  1. 1.

    Data and supplemental information necessary to reproduce the numerical results are available at the website of this journal and Lisa Maria Dellmuth at Stockholm University ( Earlier versions of the article benefited from reactions to presentations at ECPR (European Consortium for Political Research) in Bordeaux in September 2013, EISA (European International Studies Association) in Warsaw in September 2013, SWEPSA (Swedish Political Science Association) in Stockholm in October 2013, and the Transdemos Concluding Conference in Lund in June 2014. We want to thank, in particular, Thomas Gehring, Lisbeth Hooghe, Mathias Koenig-Archibugi, Jonathan Kuyper, Sofia Näsström, Thomas Risse, Jan Aart Scholte, Jens Steffek, the editor, and three anonymous reviewers of this journal for important comments without suggesting that they agree with our argument. This research was financially supported by Riksbankens Jubileumsfond and the European Research Council (200971DII).

  2. 2.

    E.g., Archibugi et al. (2012b); Bohman (2007); Brown (2011); Bäckstrand (2006); Dingwerth (2007); Dryzek and Niemeyer (2008); Frey and Stutzer (2006); Goodhart (2011); Gould (2004); Hardt and Negri (2005); Held (1995); Keohane et al. (2009); Macdonald (2008); Marchetti (2008); Miller (2010); Scholte (2014); Smith (2008); Scholte (2013); Steffek et al. (2008); Tännsjö (2008); Erman (2013).

  3. 3.

    E.g., Held (1995); Archibugi and Held (1995); Marchetti (2008); Tännsjö (2008).

  4. 4.

    E.g., Bäckstrand (2006); Frey and Stutzer (2006); Gould (2004); Hardt and Negri (2005); Dingwerth (2007); Dryzek (2006); Macdonald (2008); Macdonald and Macdonald (2006); Montanaro (2012); Saward (2010, 2011); Scholte (2013); Sechooler (2009); Smith (2008); Steffek et al. (2008); van Rooy (2004); Kuyper (2014).

  5. 5.

    Previous formulations of this principle vary slightly; see Bäckstrand (2006: 474); Goodin (2007: 51); Macdonald (2012: 47); Montanaro (2012: 1094).

  6. 6.

    It should be clear that both sociological and non-sociological variants of stakeholder democracy are normative theories, in the sense that they claim to specify how political institutions should be constructed and evaluated. Only the sociological variant of the theory, however, makes the causal assumption that the normative legitimacy of global institutions depends on their ability to foster experiences and perceptions of democracy through involvement of stakeholders or their organizations. Whether democratic legitimacy as understood in the sociological variant of the theory is an end in itself, or valuable in light of its consequences or necessity for still other matters, for example stability or effectiveness of global institutions, is a question not addressed in this article. The relevance of our research does not depend on this question since both sociological and non-sociological variants of stakeholder theory assume that democratic legitimacy is valuable (either intrinsically or extrinsically).

  7. 7.

    E.g., Archibugi et al. (2012a, b); Bäckstrand (2006); Dingwerth (2007); Macdonald (2008); Steffek et al. (2008).

  8. 8.

    See theoretical section below for elaboration on both points.

  9. 9.

    While Macdonald writes primarily about democratization of NGOs, she explicitly extends her argument to IOs: “[I]t is not difficult to imagine how the kinds of institutions I have discussed in this book could be employed to democratize their [IOs’] power” (Macdonald (2008: 224).

  10. 10.

    Stakeholder democracy may in theory be defended by arguing that political equality is normatively less important than political influence in proportion to varying stakes (Brighouse and Fleurbaey 2010). However, abandoning political equality as a fundamental principle of democracy is a theoretical weakness in itself (cf. Erman and Näsström 2013).

  11. 11.

    The sociological conception of stakeholder democracy is worth empirical examination because of the questions it addresses and the theoretical problems it may solve, not because of the number of researchers who have explicitly commit themselves to it so far. Nonetheless, sociological observations are commonplace in normative research on stakeholder democracy. Scholte (2013b) suggests that research on global democracy should as a first step inquire into what ordinary people mean by and desire in terms of democracy, and he also emphasises that all people should experience political “circumstances as being democratic in their own terms (Scholte 2013b: 15). Bäckstrand (2006) as well as Tallberg and Uhlin (2012), together with many others, motivate their interest in stakeholder democracy by pointing to the support for this ideal among real actors – as would be normatively irrelevant unless democratic legitimacy is defined partly in sociological terms. More generally, the recent interest in sociology in normative stakeholder theory parallels a shift towards non-ideal theory in debates on global justice (e.g., Sangiovanni 2008).

  12. 12.

    We thank an anonymous reviewer for suggesting this point.

  13. 13.

    The ability of stakeholder organizations to represent stakeholder communities democratically may be questioned (e.g., Pallas 2013). To be of interest in practice and normative theory, however, the stakeholder model of global democracy must assume that on balance, or in important cases, stakeholder organizations have this ability. For-profit actors are no exception. Firms and business associations may represent owners, consumers and experts in different sectors of the economy.

  14. 14.

    A completion rate of 30 % is similar to response rates of previous web surveys conducted among interest groups in the EU (see, e.g., Dür and Mateo 2013); Klüver 2013). In the absence of a 100 % response rate, the question arises whether non-respondents differ systematically from respondents, since this may give rise to a non-response error. Ideally, we would want to compare the characteristics of survey respondents with the characteristics of organizations in the total population that is surveyed (Rogelberg et al. 2003). Since we lack information about the total population of organizations involved and their characteristics, however, we cannot make such a comparison. Instead, we inquired for the reason of non-response when conducting the telephone survey. About half of the organizations that did not respond to the telephone survey indicated a lack of resources as a reason for not participating in the survey, which leads us to raise the cautionary note that organizations with relatively few resources could be underrepresented. Yet, although resourceful organizations may be overrepresented compared to the universe, the dataset involves organizations of all sizes, with a relatively similar distribution that is skewed towards smaller organizations across all IOs.

  15. 15.

    We trained four interviewers to conduct the interviews on the phone.

  16. 16.

    See Council of Europe (2011); Organization of American States (2011); UN DESA (2011). Data for non-state actors accredited to the AU was compiled using documents obtained from the AU Secretariat.

  17. 17.

    Before drawing the random samples, the sampling frames were checked for non-state actors appearing in both frames, but no such problem was detected.

  18. 18.

    The exact question wording of this and other variables in our analysis is summarized in Online Appendix A. See Tables B1 and B2 in Online Appendix B for descriptive statistics and correlations between all independent variables.

  19. 19.

    The exact question wording is as follows: Altogether, how would you rank the opportunities for your organisation to be involved in the following [IO] bodies? 1 “no opportunities,” 2 “few opportunities,” 3 “some opportunities,” 4 “many opportunities” (see Online Appendix C for an exhaustive list of the included IO bodies).

  20. 20.

    We tested whether several assumptions underlying OLS regression hold. Specifically, we examined the distribution of the residuals, which appear to be homoscedastically distributed. Furthermore, we tested for non-normal distribution of the residuals, but no such problem was detected. Last, a test of how much multi-collinearity may cause harm to the precision of the estimates reveals a Variance Inflation Factor (VIF) of less than 2 for all variables, indicating that multi-collinearity should not inflate the coefficient estimates (cf. Fox and Monette 1992).

  21. 21.

    To demonstrate how the predicted evaluation of representation, deliberation, and accountability changes when altering the explanatory variables from their minimum to their maximum value, while holding other variables at their means, we simulated first differences for models 2, 4, and 6. For each stakeholder organizations, we repeat the expected value algorithm M = 1000 times to approximate a 95 percent confidence interval around the expected value of influence, using the software package CLARIFY (King et al. 2000).

  22. 22.

    The base weight was calculated as the reciprocal of the probability of selection, BW=N/n, and reflect a non-state actors’ probability of being selected into the sample.

Supplementary material (62 kb)
ESM 1(ZIP 61.5 kb)


  1. Agné, H. (2006). A dogma of democratic theory and globalization: why politics need not include everyone it affects. European Journal of International Relations, 12, 433–458.CrossRefGoogle Scholar
  2. Archibugi, D., & Held, D. (1995). Cosmopolitan Democracy: An Agenda for a New World Order. Cambrdige: Polity Press.Google Scholar
  3. Archibugi, D., Koenig-Archibugi, M., & Marchetti, R. (2012a). Global democracy: Normative and empirical perspectives. Cambridge: Cambridge University Press.Google Scholar
  4. Archibugi, D., Koenig-Archibugi, M., & Marchetti, R. (2012b). Introduction: Mapping global democracy. In D. Archibugi, M. Koenig-Archibugi, & R. Marchetti (Eds.), Global democracy. Normative and empirical perspectives. Cambridge: Cambridge University Press.Google Scholar
  5. Bäckstrand, K. (2006). Democratizing global environmental governance? Stakeholder democracy after the world summit on sustainable development. European Journal of International Relations, 12, 467–498.CrossRefGoogle Scholar
  6. Bexell, M., & Mörth, U. (Eds.). (2010). Democracy and public-private partnerships in global governance. Basingstoke: Palgrave Macmillan.Google Scholar
  7. Bohman, J. (2007). Democracy across borders. From dêmos to dêmoi. Cambridge, MA: MIT.Google Scholar
  8. Brighouse, H., & Fleurbaey, M. (2010). Democracy and proportionality. Journal of Political Philosophy, 18, 137–155.CrossRefGoogle Scholar
  9. Brown, G. W. (2011). Bringing the state back into cosmopolitanism: the idea of responsible cosmopolitan states. Political Studies Review, 9, 53–66.CrossRefGoogle Scholar
  10. Buchanan, A., & Keohane, R. (2006). The legitimacy of global governance institutions. Ethics and International Affairs, 20, 405–437.CrossRefGoogle Scholar
  11. Chango, M. (2011). Accountability in private global governance: ICANN and civil society. In J. A. Scholte (Ed.), Building global democracy? Civil society and accountable global governance. Cambridge: Cambridge University Press.Google Scholar
  12. Christiano, T. (2012). Is democratic legitimacy possible for international institutions? In D. Archibugi, M. Koenig-Archibugi, & R. Marchetti (Eds.), Global democracy. Normative and empirical perspectives. Cambridge: Cambridge University Press.Google Scholar
  13. Council of Europe (2011). NGO database. Available at:
  14. Dellmuth, L. M., & Tallberg, J. (2014). The social legitimacy of international organisations: Interest representation, institutional performance, and confidence extrapolation in the United Nations. Review of International Studies. Available at: CJO2014. doi:10.1017/S0260210514000230.
  15. Dingwerth, K. (2007). The new transnationalism: Transnational governance and democratic legitimacy. Basingstoke: Palgrave Macmillan.CrossRefGoogle Scholar
  16. Dryzek, J. S. (2006). Deliberative global politics. Cambridge: Polity.Google Scholar
  17. Dryzek, J. S., & Niemeyer, S. (2008). Discursive representation. American Political Science Review, 102, 481–493.CrossRefGoogle Scholar
  18. Dür, A., & Mateo, G. (2013). Gaining access or going public? Interest group strategies in five European countries. European Journal of Political Research, 52(5), 660–686.CrossRefGoogle Scholar
  19. Erman, E. (2013). In search of democratic agency in deliberative governance. European Journal of International Relations, 19, 847–86.CrossRefGoogle Scholar
  20. Erman, E., & Näsström, S. (Eds.). (2013). Political equality in transnational democracy. New York: Palgrave Macmillan.Google Scholar
  21. Fox, J., & Monette, G. (1992). Generalized collinearity diagnostics. Journal of the American Statistical Association, 87, 178–183.CrossRefGoogle Scholar
  22. Frey, B. S., & Stutzer, A. (2006). Strengthening the citizens’ role in international organizations. Review of International Organizations, 1, 27–43.CrossRefGoogle Scholar
  23. Fuchs, D., Kalfagianni, A., & Sattelberger, J. (2010). Democratic legitimacy of transnational corporations.In E. Erman & A.Uhlin (Eds.). Legitimacy beyond the state? Re-examining the democratic credentials of transnational actors, Hampshire and New York: Macmillan, pp. 41–63.Google Scholar
  24. Goodhart, M. (2011). Democratic accountability in global politics: norms, not agents. The Journal of Politics, 73, 45–60.CrossRefGoogle Scholar
  25. Goodin, R. E. (2007). Enfranchising all affected interests, and its alternatives. Philosophy & Public Affairs, 35(1), 40–68.CrossRefGoogle Scholar
  26. Götz, N. (2008). Reframing NGOs: the identity of an international relations non-starter. European Journal of International Relations, 14, 625–48.CrossRefGoogle Scholar
  27. Gould, C. C. (2004). Globalizing democracy and human rights. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
  28. Grant, R. W., & Keohane, R. (2005). Accountability and abuses of power in world politics. American Political Science Review, 99, 29–43.CrossRefGoogle Scholar
  29. Hardt, M., & Negri, A. (2005). Multitude: War and democracy in the age of Empire. New York: Penguin.Google Scholar
  30. Hardt, M., & Negri, A. (2012). Declaration. New York: Argo-Navis Author Services.Google Scholar
  31. Held, D. (1995). Democracy and the global order: From the modern state to cosmopolitan governance. Stanford: Stanford University Press.Google Scholar
  32. Hurd, I. (2007). After anarchy: Legitimacy and power in the United Nations Security Council. Princeton: Princeton University Press.Google Scholar
  33. Keohane, R., Macedo, S., & Moravcsik, A. (2009). Democracy-enhancing multilateralism. International Organization, 63, 1–31.CrossRefGoogle Scholar
  34. King, G., Tomz, M., & Wittenberg, J. (2000). Making the most out of statistical analyses: improving interpretation and presentation. American Journal of Political Science, 44(2), 341–355.CrossRefGoogle Scholar
  35. Klüver, H. (2013). Lobbying in the European Union: Interest groups, lobbying coalitions and policy change. Oxford: Oxford University Press.CrossRefGoogle Scholar
  36. Kuyper, J. W. (2014). Global democratization and international regime complexity. European Journal of International Relations, 20, 620–646.CrossRefGoogle Scholar
  37. Macdonald, T. (2008). Global stakeholder democracy: Power and representation beyond liberal states. Oxford: Oxford University Press.CrossRefGoogle Scholar
  38. Macdonald, T. (2012). Citizens or stakeholders? Exclusion, equality and legitimacy in global stakeholder democracy. In D. Archibugi, M. Koenig-Archibugi, & R. Marchetti (Eds.), Global democracy. Normative and empirical perspectives. Cambridge: Cambridge University Press.Google Scholar
  39. Macdonald, T., & Macdonald, K. (2006). Non-electoral accountability in global politics: strengthening democratic control within the global garment industry. European Journal of International Law, 17, 89–119.CrossRefGoogle Scholar
  40. Marchetti, R. (2008). Global democracy: For and against. London: Routledge.Google Scholar
  41. Marchetti, R. (2012). Models of global democracy. In Defence of cosmo-federalism, D. Archibugi, M. Koenig-Archibugi, & R. Marchetti (Eds.), Global democracy. Normative and empirical perspectives. Cambridge: Cambridge University Press.Google Scholar
  42. McGrew, A. (2002). Transnational democracy: Theories and prospects. In A. Carter & G. Stokes (Eds.), Democratic theory today: Challenges for the 21st century. Cambridge: Polity.Google Scholar
  43. Miller, D. (2010). Against global democracy. In K. O’Neill & S. Breen (Eds.), After the Nation? Critical reflections on nationalism and postnationalism. Basingstoke: Palgrave Macmillan.Google Scholar
  44. Mitzen, J. (2005). Reading Habermas in anarchy: multilateral diplomacy and global public spheres. American Political Science Review, 99, 401–417.CrossRefGoogle Scholar
  45. Montanaro, L. (2012). The democratic legitimacy of self-appointed representatives. Journal of Politics, 74, 1094–1107.CrossRefGoogle Scholar
  46. Näsström, S. (2011). The challenge of the all-affected principle. Political Studies, 59, 116–134.CrossRefGoogle Scholar
  47. O’Brien, R., Goetz, A. M., Scholte, J. A., & Williams, M. (2000). Contesting global governance: Multilateral economic institutions and global social movements. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
  48. Organization of American States (2011) Civil society registry. Available at:
  49. Pallas, C. L. (2013). Transnational civil society and the World Bank: Investigating civil society’s potential to democratize global governance. Hampshire: Palgrave Macmillan.CrossRefGoogle Scholar
  50. Persson, T. (2007). Democratizing European chemicals policy: do online consultations favour civil society participation? Journal of Civil Society, 3, 223–238.CrossRefGoogle Scholar
  51. Rogelberg, S. G., Conway, J. M., Sederburg, M. E., Spitzmüller, C., Aziz, S., & Knight, W. E. (2003). Profiling active and passive nonrespondents to an organizational survey. Journal of Applied Psychology, 88, 1104–1114.Google Scholar
  52. Sangiovanni, A. (2008). Justice and the priority of politics to morality. Journal of Political Philosophy, 16, 137–164.CrossRefGoogle Scholar
  53. Saward, M. (2010). The representative claim. Oxford: Oxford University Press.CrossRefGoogle Scholar
  54. Saward, M. (2011). Slow theory: taking time over transnational democratic representation. Ethics & Global Politics, 4, 1–18.CrossRefGoogle Scholar
  55. Scholte, J. A. (2004). Civil society and democratically accountable global governance. Government and Opposition, 39, 211–233.CrossRefGoogle Scholar
  56. Scholte, J. A. (Ed.). (2011). Building global democracy? Civil society and accountable global governance. Cambridge: Cambridge University Press.Google Scholar
  57. Scholte, J. A. (2013). Democratizing global democracy research, paper for the international political theory section. Stockholm: Swedish Political Science Association Conference.Google Scholar
  58. Scholte, J. A. (2014). Reinventing global democracy. European Journal of International Relations, 20, 3–28.CrossRefGoogle Scholar
  59. Sechooler, A. (2009). Democratizing global governance? Non-state participation in the World Bank Inspection Panel and NAFTA, New Global Studies, 3 Google Scholar
  60. Smith, J. (2008). Social movements for global democracy. Baltimore: Johns Hopkins University Press.Google Scholar
  61. Steffek, J., Kissling, C., & Nanz, P. (Eds.). (2008). Civil society participation in European and global governance. A cure for the democratic deficit? London: Macmillan.Google Scholar
  62. Tallberg, J., & Uhlin, A. (2012). Civil society and global democracy: An assessment. In D. Archibugi, M. Koenig‐Archibugi, & R. Marchetti (Eds.), Global democracy: Normative and empirical perspectives. Cambridge: Cambridge University Press.Google Scholar
  63. Tallberg, J., Sommerer, T., Squatrito, T. & Jönsson, C. (2013). The opening up of international organizations: Transnational access in global governance. Cambridge: Cambridge University Press.Google Scholar
  64. Tännsjö, T. (2008). Global democracy: The case for a world government. Edinburgh: Edinburgh University Press.Google Scholar
  65. Teorell, J., Charron, N., Dahlberg, S., Holmberg, S., Rothstein, B., Sundin, P., & Svensson, R. (2013). The Quality of Government Dataset, version 20 Dec 13. University of Gothenburg: The Quality of Government Institute. Accessed 12 Dec 2014.
  66. Uhlin, A. (2010). Democratic legitimacy of transnational actors: Mapping out the conceptual terrain. In E. Erman & A. Uhlin (Eds.), Legitimacy beyond the state? Re-examining the democratic credentials of transnational actors (pp. 16–37). Hampshire: Macmillan.Google Scholar
  67. Uhre, A. N. (2013). On transnational actor participation in global environmental governance. Diss. Stockholm Stockholm University, Dept. of Pol. Sci., Stockholm University.Google Scholar
  68. Union of International Associations (2011). Yearbook of international organizations online. Available at:
  69. United Nations (2004) ‘We the peoples’: civil society, the United Nations and Global Governance: Report of the Panel of Eminent Persons on United Nations-Civil Society Relations. UN Document A/58.817.Google Scholar
  70. United Nations Department of Economic and Social Affairs (UN DESA). (2011). Integrated civil society organizations system. Available at:
  71. Van Rooy, A. (2004). The global legitimacy game: Civil society, globalization, and protest. Hampshire: Palgrave Macmillan.CrossRefGoogle Scholar
  72. Weiss, T. G., & Gordenker, L. (Eds.). (1996). NGOs, the UN, and global governance. Boulder: Lynne Rienner.Google Scholar
  73. Wenar, L. (2006). Accountability in international development aid. Ethics & International Affairs, 20, 1–23.CrossRefGoogle Scholar
  74. Willetts, P. (Ed.). (1997). The conscience of the world. The influence of non-governmental organizations in the UN system. Washington D.C: Brookings.Google Scholar

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  • Hans Agné
    • 1
  • Lisa Maria Dellmuth
    • 2
  • Jonas Tallberg
    • 1
  1. 1.Departement of Political ScienceStockholm UniversityStockholmSweden
  2. 2.Stockholm Resilience CentreStockholm UniversityStockholmSweden

Personalised recommendations