Dilemmas and trilemmas
What are the implications of the institutional, dialectical and communicative conception of legitimacy and accountability outlined above for understanding how regulators will respond to multiple and often conflicting accountability and legitimacy claims? As noted above, regulators can face multiple legitimacy claims from within and outside the regulatory regime, some of which may conflict. They can also be proactive in managing their own legitimacy. So how can and will they respond to multiple legitimacy claims? They may be able to meet all, or at least some of them, simultaneously – to please some of the people some of the time. It is not necessary for legitimacy claims to be interlinked or mutually supporting for this to be possible. They simply have to be compatible. The development of management models such as balanced scorecards (Kaplan & Norton 1992), used by a number of government bodies, is just such an attempt to structure, or at least provide a reporting framework for, the organization’s responsiveness to different aspects of its environment.11
But, as suggested above, the demands of legitimacy communities may well be directly opposed – to satisfy one will necessarily lead to dissatisfaction of the other. The incompatability of democratically rooted claims relating to representation and membership with functionally rooted legitimacy claims relating to efficiency or expertise provides a good example. Frequently, in order to satisfy the legitimacy claims of those they are seeking to regulate, regulators’ main decision-making bodies need to be comprised solely or mainly of representatives of those regulatees or those with considerable technical expertise (or both). In contrast, to be legitimate to a wider section of civil society, and indeed to be legitimate to other actors in the regulatory regime that the standard setting organization may be relying on, such as pressure groups, NGOs, or national governments, those decision-making bodies need to be solely or mainly composed of a wider range of representatives. Regulators can attempt to incorporate potentially competing demands within their organizational structures, for example the FSC attempts to balance the voting power (and structure the potential conflict) between different members through a tripartite structure of membership chambers: social, environmental, and economic (Meidinger 1997). But frequently membership remains contentious. Debates on the IASC’s membership have continued since its inception, for example, and despite recent changes to its constitution to broaden membership and enhance consultation (IASC Foundation Constitution Committee 2004), as noted above the pressures on the IASC Foundation to change its structures remain, as they do with respect to the BCBS. The membership of ICANN has also been a constant battleground since its inception.
Conflicts between legitimacy claims based on functionality and those based on democracy are common, but there can be other conflicts between legitimacy claims: such as between those demanding procedural justice (constitutionality) and those demanding maximum speed and efficiency in decision-making (functionality). Even within groups of claims there are conflicts, such as conflicts between constitutional claims rooted in different models of administrative justice, as Mashaw’s familiar typology of bureaucratic, moralistic, and professional models of administrative justice demonstrates (Mashaw 1983). Moreover, once gained, maintaining legitimacy may be difficult, not least because legitimacy communities can change; new actors with different legitimacy claims may become relevant to the organization or legitimacy claims of existing actors in the regulators’ environment may mutate (see e.g. Dent 1991; Ogden 1995; Conrad 2005), as the example of the IASC illustrates.
Faced with incompatible legitimacy claims, organizations face a legitimacy dilemma; what they need to do to be accepted by one part of their environment, within and outside the regulatory regime, is contrary to how they need to respond to another part. Forming one set of accountability relationships can preclude forming others; it simply is not possible for organizations to have complete legitimacy from all aspects of the environment, including all other organizations in the regulatory regime. Even if the conflict between legitimacy communities does not lead to a dilemma, it can have a deleterious effect on the organization as it seeks to respond to the multiple legitimacy and accountability demands being made on it; in Koppell’s evocative phrase, the organization may suffer “multiple accountability disorder” (Koppell 2005). In other words, its attempts to respond to the multiple demands may diminish its chances of survival (Power 1995).
Koppell argues that ICANN has suffered in this respect. Its turbulent history is marked by significant shifts in membership, structures, and procedures as it attempted, in part, to respond to the different legitimacy claims made on it and attempts to forge different types of accountability relationships, with the result that it has ended up with little legitimacy from anyone (Koppell 2005). Other organizations have had similar experiences. In their attempts to meet the legitimacy claims of one community, they lose it from another. Edwards, for example, has noted that the increasing demands on NGOs to develop the accountability trappings of financial audit, transparency, and so on, can result in NGOs becoming more accepted by state or international actors, but also more bureaucratized and increasingly distant from the communities they seek to engage with and represent (Edwards 1999; Slim 2002; Goodin 2005). Research into the effects of the introduction of new public management tools of accountability and evaluation in healthcare and education in a number of European countries is replete with complaints from those within those sectors that the demands of audit, performance targets, and other accountability relationships are distorting organizational priorities away from what others (doctors, teachers) think should be the central role for the organization (e.g. Laughlin et al. 1992; Lindkvist 1996).
In contrast, even though it is faced with multiple and perhaps incompatible legitimacy claims, the organization may not perceive there to be a dilemma at all. Instead, it simply does not respond to a particular claim. In other words the organization does not (perceive a) need to meet the legitimacy claims of a particular legitimacy community in order to pursue its goals or to survive. It perceives that it can ignore the claims of consumers, less developed countries, human rights organizations and so on, because it has greater (perceived) need for recognition by other legitimacy communities. The Church of England has proved almost impervious to attempts to impose hierarchical systems of financial and managerial control, for example. Although their adoption would enhance its legitimacy for some, this has not yet of itself been a sufficient impetus to promote change and the perception within the Church thus far has been attempts to build these accountability relationships can be ignored (Berry 2005).
But regulatory, and indeed other, organizations cannot ignore all legitimacy claims and survive, even if they can ignore some – or perceive that they can. As emphasized above, regulatory organizations have a particular need for legitimacy – it is not enough that they are “generally accepted,” they need to be actively supported. This is true for all regulators, but as noted it is particularly true for non-state regulators who are trying to promote behavioral changes in others who may be under no legal obligation to take any notice of them at all, and moreover who often compete with other regulators to have their norms accepted (Bernstein & Cashore 2007; Meidinger 2007).
This suggests that there can be significant implications for an organization, therefore, in acquiescing in certain legitimacy claims and developing certain accountability relationships rather than others. The question arises as to the claims that they are likely to respond to and those from which they will refrain, and moreover the extent to which their scope for strategic action is bounded by their institutional environment. There has been very little sustained research on how non-governmental regulators respond to competing accountability and legitimacy claims, and even less on how these responses may be affected by their relationship with other regulators in polycentric regulatory regimes.12 There has been research on how other organizations have responded to attempts to render them more accountable through public management techniques on which we can draw, however, as well as the more general literature on compliance. Research on how organizations respond to the accountability and performance mechanisms introduced in the “new public management” show that conflicting pressures on the organization lead to internal conflicts between different parts of the organization and to the adoption of a range of responses, from transformation and acquiescence to defiance and manipulation (Brignall & Modell 2000; Bevan & Hood 2006; Hood 2006). The variety of organizational responses to regulatory requirements has also been noted in the compliance literature (Braithwaite et al. 1994; Braithwaite & Reinhart 2007).
In an attempt to synthesize the strategic and more bounded models of response, and assuming a structurated model of behavior, Oliver identifies five types of response by organizations to institutional processes: acquiescence, compromise, manipulation, avoidance, and defiance (Oliver 1991). Oliver suggests that organizational responses to institutional pressures to conform will depend on five sets of categories, each with two dimensions: cause, constituents, content, control, and context. First, the response will vary with the cause; in other words, the nature of the pressures exerted and what the organization gains from acquiescence, for example social fitness or economic gain. Second, it will vary with the organization’s constituents, or who is exerting the pressures – notably how multiple they are, and the degree to which the organization is dependent on them. Third, the response will vary with the content of the claim or demand; with how consistent it is with the organization’s own goals, and the extent to which compliance with them will constrain the organization’s discretion. Fourth, the response will vary with the means by which the claims are imposed (through coercive means or not) and with the extent to which they are diffused throughout the organization’s environment. Finally, the response will vary with the context of the organization, in particular the degree of interconnectedness of inter-organizational relations within the field, and the degree of uncertainty that exists. Oliver then develops hypotheses on how organizations will respond based on variation in the 10 dimensions of these five categories.
Of most relevance here, she predicts that organizations will always acquiesce when legitimacy gains are high (Oliver 1991). However, as we have seen above, organizations may not always acquiesce in every legitimacy claim, even where that would result in more legitimacy from that particular legitimacy community. Oliver’s analysis therefore provides a useful starting point, but needs considerable refinement (see also Brignall & Modell 2000). Further, Oliver does not distinguish between types of legitimacy claims in predicting organizational response. In contrast, others have suggested that the degree to which organizations respond strategically to a legitimacy claim depends on whether that claim is pragmatic, moral or cognitive, with strategic responses being lowest with respect to cognitive claims (Suchman 1995; Cashore 2002; Bernstein & Cashore 2007). Further, Oliver is not particularly concerned with analysing the organizational field to any great depth. The key point with respect to polycentric regulatory regimes, however, is the composition of that field. For, as noted above, those within the regime potentially have to respond to legitimacy claims made not just by the actors whose behavior they are attempting to regulate (e.g. firms, governments), and wider civil societies, but by other actors within the regulatory regime with whom they are interrelated and/or on whom they may be partly dependent. Finally, more recognition has to be given to the role of discourse in constructing and expressing legitimacy claims and accountability relationships, and its role in shaping organizational responses. There is a strategic dimension to the role of discourse in this dynamic; empirical research suggests, for example, that in order for a regulator to build an accountability relationship with those who lack coercive or other means to pressure the regulator to acquiesce, they have to translate their demands into a discourse that the regulator already recognizes (Hajer 1995; Morgan 2003). This is a strategy which can enhance the congruence of their claims with the regulator’s own cognitive or normative framework.
Nevertheless, we can draw on Oliver’s work to hypothesize that regulators are less likely to acquiesce to legitimacy claims, even if the legitimacy gains for that regulator within that particular legitimacy community may be high, where all or some of six conditions pertain: (i) the regulator perceives itself to have a low dependency or interconnectedness on that legitimacy community for the performance of its regulatory function; (ii) where the content of the claims, and the discourse in which they are articulated, is not congruent or consistent with the regulator’s, or is only so to a moderate degree; (iii) where acquiescence with the claims would pose moderate to severe constraints on the ability of the regulator to determine its own structures, goals and activities; (iv) where the legitimacy community has little means, directly or indirectly, to coerce the regulator to respond; (v) where the claims being advanced are not widely diffused among others within or outside the regulatory regime; and (vi) the consequences of acquiescing are uncertain. Thus, for example, the hypothesis would be that the Basle Committee on Banking Supervision (BCBS) refuses to expand its membership, or grant observer status, to, countries such as India and China because its dependence on them is low; the demands of those countries are not particularly in line with the aims or views of the Committee; the constraints on what it is the BCBS wants to do would be high; they have little means of imposing their demands; their views or demands are not widely shared by communities the BCBS recognizes as relevant; uncertainty as to whether acquiescence would bring any overall legitimacy or other gains for the BCBS is high; and interconnectedness is low. Conversely, a regulator or organization is more likely to acquiesce to legitimacy claims where it perceives itself to have a relatively high dependence on or interconnectedness with the legitimacy claimant; where the content of the claim and the discourse in which it is expressed is congruent with the organization; where acquiescence would not pose severe constraints on its discretion; where the claimant can directly or indirectly coerce the regulator; where the claims being advanced are widely diffused; and the consequences of acquiescing are relatively certain.
Not all of these conditions are likely to be equally relevant at all times. Moroever, it still has to be recognized that for the regulator, meeting different legitimacy claims often involves engaging in different accountability relationships, relationships which can have transformative effects. Engaging in one set of accountability relationships may compromise the regulator’s ability to engage in another set and so to meet the claims of a different legitimacy community. For there are only so many ways an organization can be transformed at once. Something has to give. Multiple legitimacy claims and engaging in multiple accountability relationships can mean that attempts to make an organization accountable end in the accountability equivalent of the “regulatory trilemma” (Teubner 1986): they are ignored, co-opted, or destroy that which it is they seek to make accountable.