The Regulation of International NGOS: Assessing the Effectiveness of the INGO Accountability Charter

The INGO Accountability Charter is the only global, cross-sectoral regulatory initiative for international NGOs. This is the ﬁrst independent study of perceptions of its effectiveness, based upon 26 in-depth semi-structured interviews with key individuals from 11 leading international NGOs. Firstly, it analyzes interviewees’ beliefs about the motivations of NGOs in joining the Charter. The ﬁndings contribute to the scholarly debate about the key drivers for voluntary regulation between ‘club theorists’ and ‘constructivists’ by demonstrating that NGO behavior in this regard is both self-interested and norm-guided. Secondly, it investigates the extent to which the interviewees believe that the Charter has been effective in enhancing the accountability of its members. Their responses further underline the applicability of club theory and constructivist explanations of NGO behavior, and lead to several policy recommendations about the future direction of Charter.

responsibilities regarding the Charter within their organizations. The latter group of participants was drawn from eleven leading NGOs from the humanitarian, development and advocacy sectors.
Accountability for NGOs has been variously defined by scholars, NGOs and standard-setting bodies over the years. Debates revolve around to whom and for what NGOs should be held accountable (for an overview see Crack 2013). The Charter defines accountability as follows: • Being transparent on what the organization is, what it commits to doing and progress achieved; • Engaging key stakeholders in meaningful dialogue to enable continuous improvement for those we serve; • Using power responsibly and enabling stakeholders to hold us to account effectively (INGO Accountability Charter 2015a).
The Charter stipulates a number of principles, guidelines and policies that member organizations should observe in order to be deemed 'accountable,' and members have to report annually against these commitments and publish the results online. Many peer regulation initiatives are simply codes of conduct that require little from their members other than a self-proclaimed commitment to the standards. However, those with 'global membership are less likely to have formal complaints mechanisms and to punish rule violators than their regional and single-country counterparts' (Tremblay-Boire et al. 2016: 713) The Charter is notable for having a complaints mechanism, an independent vetting procedure, and a sanctions clause that enables it to expel members that are non-compliant. It is therefore of interest not only because of its cross-sectoral positioning and global membership, but also because of its complaints and enforcement procedures.
The first question that this study addresses is: What do the interviewees believe motivated NGOs to join the Charter? The key drivers behind peer regulation have repeatedly been the subject of academic debate. The interviewees' responses provide the opportunity of exploring a puzzle in the literature: Why do NGOs participate in regulatory initiatives? There are two main explanations: club theory (which, put simply, argues that they do so for self-interested reasons, primarily to send a reputational signal to stakeholders) and constructivist 3 theory (which, put equally simply, argues that NGOs are strongly influenced by shared beliefs about accountability norms). The interview data suggest that both theoretical explanations have some traction: Organizations maintain membership of the Charter to satisfy a mixture of 'selfinterested' and 'norm-guided' motivations.
The second question that this study addresses is: To what extent do the interviewees believe that the Charter has been effective in enhancing the accountability of its member organizations? The interviewees were asked to evaluate the effectiveness 4 of the Charter, based upon their experiences of engaging with the reporting process and their understandings about changes in behavior and performance of member organizations. The study finds that the interviewees believe that the Charter provides NGOs with a defense against criticisms of poor accountability from hostile parties, and also helps members to improve performance through feedback and peer learning. However, they felt that the effectiveness of the Charter was limited due to several factors, including poor awareness of the Charter among key stakeholders, variable levels of engagement inside the member organizations, and misaligned understandings of accountability between advocacy/campaigning NGOs and humanitarian/development NGOs. I argue that their critical appraisals of the Charter attest to the influence of both self-interested and normbased considerations.
The discussion proceeds as follows. The first section provides an overview of the rise of NGO peer regulation and outlines the main contentions of club theory and constructivist theory. The second section provides background information on the Charter, and the third section explains the methodology of the study. The fourth section explores the interviewees' opinions on the motivations for NGOs joining the Charter. The fifth and sixth sections turn to consider the perceived efficacy of the Charter, which is discussed in terms of the benefits and challenges of Charter membership. The concluding section offers some policy recommendations and suggestions for future research.

Club Theory, Constructivism and Measures of Efficacy
It has become evident that there is a pressing need for NGOs to raise their standards of accountability and to address perceptions that they are unaccountable (Schmitz et al. 2012;Thrandardottir 2015). Accountability issues are now high on the NGO policy agenda, particularly given that major donors attach more importance than ever before to transparency and evidence of 'value for money.' It is against this background that NGOs have cooperated to establish several peer regulation initiatives in recent decades.
The topic of NGO regulation has attracted some interest from scholars, both within a domestic (Bies 2010;Bloodgood et al. 2014;Gugerty 2008) and international context (Brown 2008). The literature on the efficacy of peer regulation is relatively scant, not least because of the challenges of finding a common measure of effectiveness (Crack 2016;Featherstone 2013). The scholarship therefore focuses on the factors underpinning the emergence and design of regulation mechanisms. There are two main explanatory approaches in this regard: club theory and constructivist theory. 5 Club theory builds upon principal-agent theory, a political economy approach to understand the problems that are posed when a principal contracts an agent to carry out certain tasks in conditions where both parties may have competing interests and asymmetrical information. The principal will have less information than the agent and so will be uncertain about whether the agent is serving the principal's best interests, particularly when it is difficult for a principal to monitor the agent's actions and/or an agent will find it profitable to exploit the principal. There are all manners of ways in which principals and agents might try to ameliorate these problems-regulatory 'clubs' are but one. Clubs serve to provide a reputational signal to principals. Agents may join regulatory clubs to improve their performance. A strong signal is likely to be sent to principals if standards are stringent and compliance is monitored. Membership can signify high levels of accountability and performance if the club is widely regarded as credible. The voluminous literature on clubs mainly focuses on the private sector (Cornes and Sandler 1996;Sandler and Tschirhart 1997), but the perspective has also been used by nonprofit scholars (Gugerty and Prakash 2010;Potoski and Prakash 2009;Prakash and Gugerty 2010). NGOs have accountability relationships with multiple principals (donors, intended beneficiaries, supporters, etc.), and clubs offer the potential to help NGOs to build trust with these different stakeholders. Effective clubs prompt changes in NGO behavior and open possibilities of receiving certain 'rewards.' NGOs may hope that clubs could encourage donors to increase their funding. '[P]roactive voluntary regulation might dampen the demand for new laws that restrict their activities in even less desirable ways' (Gugerty and Prakash 2010: 11). According to this perspective, effective regulatory initiatives are ones that (a) have high levels of compliance; (b) send a credible and widely recognized signal to principals in order to build trust; (c) could lead to increased funding; (d) could help to preempt the threat of government interference and/or regulation (see Table 1). Distinct from this is the constructivist approach, which considers the influence of shared ideas and values as key to understanding what shapes forms of regulation. They acknowledge that self-interest may partly account for a NGO's decision to join a regulatory mechanism, but are also interested in how NGOs are incentivized by a concern for shared norms, a desire to engage in social learning and to share best practice. Deloffre, for example, accounts for the design of the regulatory initiatives that were established after the Rwandan genocide as being shaped by debates among NGOs and key stakeholders 'that created a feeling of mutual engagement and commitment to defining collective accountability practice ' (2016: 22). According to this constructivist perspective, an effective regulatory initiative would be one that shapes understandings about 'rightful conduct' for responsible NGOs among practitioners and stakeholders. Such understandings may correlate with donor expectations, but are not necessarily determined by the preferences of donors (Pallas et al. 2014). For constructivists, 'effective' standards would help to produce an institutional environment that promotes social learning and norm-compliant behavior, by encouraging individuals to internalize and uphold the norms (see Table 1).
Although club theory and constructivism have a different focus, they are not mutually exclusive. The interpretations that each approach generates can be compatible, and provide a nuanced and multidimensional account of actor behavior.
As will be seen, the interviewees' opinions about the effectiveness of the Charter revealed evidence of both selfinterested and norm-guided behavior among member organizations. The next section provides some background on the Charter, before the discussion proceeds to data analysis.

INGO Accountability Charter: Structure and Objectives
The INGO Accountability Charter was established by a consortium of leading NGOs and launched in 2008. It is funded by annual membership fees from NGOs. Since 2010, it has been based upon the Global Reporting Initiative (GRI), which is the world's largest sustainability reporting framework. The GRI is used by corporations and other organizations on a voluntary basis to report on their performance. The Charter commissioned the GRI to produce a 'NGO Sector Supplement,' a modified version of the guidelines designed to 'enable NGOs to demonstrably meet the same standards of transparency…that are demanded by other sectors' (Global Reporting Initiative 2011: 6).
The Charter consists of ten commitments that are intended to promote the goals of 'greater transparency, accountability and effectiveness' (INGO Accountability Charter 2015b). The commitments are summarized thematically in the Charter as follows: respect for human rights; independence; transparency; good governance; responsible advocacy; participation; diversity/inclusion; environmental responsibility; ethical fundraising; and professional management (INGO Accountability Charter 2014a: 2). The Charter Text goes on delineate each theme in terms of specific undertakings. For example, the commitment to 'good governance' requires NGOs to ensure, among other things, 'publication of a clearly defined and transparent mission, governance structure and decisionmaking process at the governance level' (ibid: 6).
Member organizations must produce an annual report to demonstrate that policies and procedures are in place to promote adherence to the Charter. The report framework consists of 36 'profile disclosures' about the organization, and 20 'performance indicators' about program effectiveness, ethical fundraising and communication, and management of issues concerning finance, the environment, human resources and impact on wider society (INGO Accountability Charter 2014b). Members have to account for any failure to report against all the criteria.
The reports are submitted to an Independent Review Panel (IRP), which is composed of 'respected accountability experts' (ibid: preamble). The IRP assesses the strength of the evidence presented and indications of institutional commitment to accountability in the reporting exercise. They provide targeted feedback, advising on how the member's reporting and/or performance should be improved. Organizations are also encouraged to complete a 'gap analysis' exercise to identify areas in need of improvement and to set self-imposed targets for change.
The Panel scrutinizes progress against these targets in forthcoming annual appraisals. The documentation is made available on the Charter Web site.
Member organizations could be expelled if they are found to be in contravention of the Charter commitments or if they fail to submit reports without sufficient explanation. There are nineteen full members at the time of writing: ActionAid; Amnesty International; Article 19; BRAC; Care; CBM; Civicus; Educo; European Environmental Bureau; Greenpeace; Islamic Relief; Oxfam; Plan; Sightsavers; SOS Children's Villages International; Terre des Hommes; Transparency International; World Vision and World YWCA.

Methodology
The findings are based upon semi-structured interviews with 26 participants, during August-November 2014. Sixteen of these were participants from member NGOs who were centrally involved in the decision to join the Charter and/or were closely involved with producing reports for the Charter. They were speaking in a personal capacity rather than on behalf of their organization. All of the relevant NGOs were contacted with requests for interviews, and participants from 11 of the 18 6 'full member' organizations responded. Ten participants were involved with the administration of the Charter, including five current/former Board members, four current/former members of the Independent Review Panel (IRP) and a representative from the Charter Secretariat. Most respondents spoke on condition of anonymity. The data should be Key actors internalize the norms An institutional environment is created that promotes social learning and norm-compliant behavior treated with a degree of caution, since the views of the participants may reflect their interest in appearing to uphold high standards of transparency and accountability. Nonetheless, the participants did express significant reservations about the efficacy of the Charter, as shall be seen. Data were manually sorted into a list of preset codes, derived from keywords used in club theory (e.g., 'reputation,' 'brand,' 'trust') and constructivism (e.g., 'norms,' 'learning,' 'sharing'). Emergent codes were identified when analyzing the data that enabled the capture of recurrent ideas and meanings. Two validation strategies were adopted to improve the rigor of the study (Creswell 2008). Firstly, a preliminary report of the prevailing themes was circulated to participants for feedback. The quotations used in this article have all been approved by the participants concerned. Secondly, claims made by the interviewees were corroborated with document analysis. This included reports from the member organizations along with feedback from the IRP and responses to feedback from the NGO where provided, the minutes from Charter AGMs 2011-2015; the Charter Annual Report 2011-2014, as well as sundry materials relating to the membership criteria and reporting requirements.
The following questions were posed to interviewees: (a) What motivates NGOs to join the Charter? (b) What are the perceived benefits of being a member of the Charter? (c) What are the perceived disadvantages of membership? (d) Bearing in mind the benefits and challenges of membership that you have just described, to what extent do you feel that the Charter is effective in enhancing the accountability of member organizations? The responses to these questions are detailed below.

Motivations for Joining
All the participants agreed that the key incentive for joining the Charter is the legitimacy it promises to bestow upon member organizations, given its self-proclaimed status as the 'only global, cross-sectoral accountability framework for NGOs' (INGO Accountability Charter 2015a). The high profile of the largest member NGOs was acknowledged as a key factor underpinning the credibility of the Charter and the attractiveness of membership to smaller NGOs. In the words of one interviewee: 'I think it helps your organization to build its brand, its reputation, its acceptance by the public and by other constituencies including donors' (Int.13). It was seen as an additional advantage by some that the Charter is an initiative driven by NGOs, rather than by donors, thus enabling NGOs to shape the 'accountability agenda' in a way that reflects common values and priorities across the sectors. The interviewees were not specific about how the agenda might differ if driven by donors.
Thirteen out of sixteen NGO participants admitted that joining the Charter was partly a defensive move on behalf of their organizations to ward off actual and anticipated criticisms of poor accountability from donors, the media and political opponents. Joining the Charter was a way for NGOs to seize the initiative, because it was feared that attacks on their integrity could gain traction if there was not a concrete way to demonstrate their commitment to standards of excellence. To quote an interviewee from Amnesty: 'When we are questioned by government, for instance, with questions about legitimacy and our accountability-particularly if we're pushing for greater accountability by government-there have been times when we've been able to use our membership of the Accountability Charter to strengthen our position and show how we are accountable' (Int.1).
Therefore, a large part of the Charter's appeal to member NGOs is the 'insurance' it provides against possible accusations of poor accountability. This is despite the fact that, even by the Charter's own admission, it has a low profile among those parties that have an interest in holding organizations accountable (INGO Accountability Charter 2014c: 9). For example, no major donors stipulate Charter membership as a precondition of funding. The interviewees generally acknowledged that the documentation on the Charter web site is rarely accessed by external stakeholders. A participant involved in the administration of the Charter argued that this did not detract from the value of the reporting exercise, because the requirements of membership compel organizations to engage with accountability issues and thus raise standards of performance: The general public, quite frankly, are never going to sit and read those reports… I would hope that the civil society department in DFID who are actually giving out the massive amounts of money and so on would actually look at them, but I've no idea whether they do or not. But I think the fact is that they're there, and that's what's important. And also the process that the NGOs have to go through in order to put them there and to get that information is important, because that in itself drives greater accountability and transparency (Int.5).
None of the interviewees suggested that donors have a meaningful appreciation of the Charter; even the Charter's web site will only go so far to claim that it 'has a good chance of reaching donor recognition due to its unique positioning' (INGO Accountability Charter 2015a). All of the interviewees involved in the administration of the Charter spoke of the importance of increasing donor awareness of the initiative in order to maintain its relevance to existing members and to enhance the attractiveness of membership to other organizations. These were Voluntas (2018) 29:419-429 423 sentiments that were echoed by six participants from humanitarian/development NGOs and five participants from advocacy/campaigning NGOs. Club theorists contend that NGOs join regulatory initiatives to send a reputational signal to principals (Connelly et al. 2011;Gugerty 2009;Prakash and Potoski 2006). The Charter's emphasis on reporting and compliance suggests that signaling is important, especially given that the more rigorous GRI framework was incorporated into the standards four years after the Charter's launch. However, by this yardstick, the Charter seemingly has little efficacy if crucial stakeholders such as donors have poor awareness of its existence, which raises the question of why organizations have continued to pay their membership dues for years. Some of the responses above indicate that club theorists are correct to identify signaling as a key incentive, particularly to evade unwelcome government interference, but the data also suggest the presence of drivers other than self-interest. The following two sections turn to consider the interviewees' perceptions of the Charter's efficacy after joining, beginning with their assessments of the benefits of Charter membership.

Perceived Benefits of Membership and Reflections on the Charter's Efficacy
Thirteen out of the sixteen NGO participants identified peer learning opportunities as one of the most valuable aspects of Charter membership. The Charter provides formal occasions for knowledge exchange; for example, they run Webinars and Peer Advice Groups on numerous accountability-related topics. Peer learning also happens informally, such as networking outside of meetings. Indeed, two interviewees stated that they found that formal and informal peer learning that occurs at these events as more useful in developing their thinking about accountability and performance than the actual exercise of compiling the Charter report.
Six NGO respondents claimed that the high-quality feedback from the IRP was one of the most significant benefits of Charter membership, and asserted that it has led to substantive improvements in practice. It was possible to identify several concrete examples of the influence of the Charter on policies of member organizations. Perhaps the most significant is the introduction of a Complaints Handling Mechanism, which has recently been made a prerequisite of Charter membership (INGO Accountability Charter 2015c). For example, one interviewee explained that an anonymous Web-based whistle-blower system had been implemented within two years following feedback from the IRP: 'We would have probably gone into the direction of reviewing our anti-corruption policies at certain stage, no question about it, but to actually boost and to really make that an urgent matter-that is thanks to the expert panel' (Int.9).
Several non-environmental NGOs have also taken measures to reduce their carbon footprint in order to comply with the performance indicators on environmental responsibility. For example, Oxfam International was commended by the IRP for reducing its greenhouse gas emissions by 8.5% from 2010 to 2013 (INGO Accountability Charter 2015f: 96). Here, the value of the peer learning opportunities that the Charter provides was in evidence, since some respondents particularly singled out Greenpeace for praise in assisting other members to take issues of environmental impact seriously. An interviewee from a humanitarian/development NGO acknowledged that working with Greenpeace helped the organization develop climate-sensitive policies, and claimed that they may not have embraced environmental reporting without the impetus of Charter membership: Basically nobody disagrees with this being the right thing, but in terms of priority there are always so many things to be done… if you're working as a NGO with donations and you always have to, you know, justify 36 other top priorities-it really helps if you also get this kind of external push to say, ok, compared with best practice this is where you are behind (Int.9).
To summarize, participants felt that the channels for feedback helped to promote internal learning, and Charter membership helps to maintain focus on obligations to improve aspects of performance that might otherwise be side-tracked. It was said that the impending deadline of the report helped to increase the urgency for changes in practice.
These reflections suggest that a NGO's decision to join a regulatory mechanism may be partly motivated by selfinterest (as club theory would predict) since there was near consensus that NGOs were spurred to join the Charter to defend their operational freedom. Club theorists could also argue that peer learning opportunities are in the interest of NGOs, if it enables them to adopt better practices, increase compliance and send a stronger reputational signal. However, there was general agreement that the Charter has a low profile among key stakeholders such as donors. None of the interviewees suggested that membership had helped them to retain/increase their levels of funding. The interview data therefore present club theorists with a problem: Why should NGOs participate in the Charter if, according to club theory, its efficacy is limited?
The constructivist approach offers an alternative way to interpret the data (see Table 1). There are indications of norm-guided behavior from the policy-making level to the level of the individual staff member. There is evidence that membership does provoke progressive reforms in policy and practice. It opens channels for more informal and participatory forms of learning about best practice, and dialogue between counterparts in different sectors that may not otherwise exist. The value that the participants claim to attach to these interactions suggests that they have internalized accountability norms. It presents a picture that accords with constructivist predictions: NGOs join regulatory mechanisms because it is widely understood to be an inherently 'right' thing to do. Efficacy is partly measured by the extent to which the Charter has shaped the understandings of practitioners about 'rightful conduct,' and helped to foster an institutional environment that promotes accountability norms. This could partly explain why NGOs abide by the Charter, even though membership does not help them to transmit a widely recognized reputational signal.
The participants discussed the disadvantages of Charter membership, and the responses further underlined the applicability of both theories to understanding key drivers behind peer regulation.

Perceived Challenges of Membership and Reflections on the Charter's Efficacy
Club theorists would expect the participants to frame their criticisms of Charter in terms of poor signaling to principals and membership costs. These themes were indeed evident in the data. There was consensus among all participants that the effectiveness of the Charter is impaired by its low profile among donors and within the NGO community. Also, seven NGO participants complained about allocating resources to meet the commitments of Charter membership (including six from humanitarian/development organizations). The membership fee ranges from €1000 for NGOs with an annual income of less than €1 million to €25,000 for organizations with an income of more than €1 billion (INGO Accountability Charter 2015c). The financial commitment extends to staff time devoted to compiling the Charter reports. Participants expressed weariness with bureaucracy and concerns about the time spent on potentially duplicating information for different internal and external reporting frameworks. This was noted by one participant as particularly problematic for humanitarian/development organizations that also seek to comply with other regulatory initiatives. There are far more of these in the humanitarian/development sector than the advocacy sector. Humanitarian/development organizations also have to contend with stringent reporting requirements from donors. Monitoring and evaluating impact, articulating theories of change and completing log-frames have long been a core activity of their work.
The perceptions of NGO participants about the onerous nature of the reporting requirements should be weighed against efforts by the Charter to reduce the workload entailed by membership by streamlining the reporting process. The recently revised Reporting Requirements state that reports should be a maximum of 40 pages long, and that the relevant information can be embedded in the organization's annual report (INGO Accountability Charter 2015d: 5). Further, once an organization has achieved 'a sufficient level of accountability, it only has to submit full reports every two years' and submit a 4-6-page report in the interim (ibid: 2). The recent simplification of reporting requirements was universally welcomed in the interviews. Nevertheless, for club theorists, the grumbles about the resource-intensive nature of the reporting process would be expected since agents have an interest in minimizing the 'costs' of regulation (or at least to the extent that it does not compromise the credibility of the signal sent to principals).
Another prominent theme in the data was frustration with the low profile of the Charter. NGO participants did not just complain that donors were hardly aware of the Charter, but also that NGO staff were similarly under-informed. This is particularly the case for NGOs with a large 'family' structure with many national entities. The problem is exacerbated by high levels of staff turnover, which is commonplace for NGOs and results in persistent problems with knowledge management. Respondents observed that it can be a challenging task to coordinate data collection for the Charter from country offices and even more so given such poor levels of awareness about the purpose of the exercise. Some expressed feelings of disenchantment because so much time was invested in producing the reports, and yet readership is very low, even within their own organizations. For one interviewee, the low rate of access seriously compromised the value of the reporting process: You know, if only four people have read this, does this even remotely mean accountability? Because there's a presumption that when you've written it, people are actually going to read it and take note of it. You know, asking the questions might influence the way we do things internally, but you want people externally to be reading and asking the questions, otherwise you think, well, is this just a scheme for full employment? Are we all just writing reports that no one else reads? (Int.7) This desire to have an internal/external audience reveals that the participant measures the efficacy of the Charter in terms of how well it performs a 'signaling' function, in line with the predictions of club theory. Furthermore, the Charter is currently working on a Global Standard to 'generate public trust and recognition,' which indicates a common desire to signal even though these ambitions have not yet been realized (INGO Accountability Charter 2015e).
Constructivist themes were also evident in the data. Although constructivists do not deny the presence of selfinterest, they focus attention on how actors evaluate efficacy in terms of the extent to which regulatory mechanisms foster an institutional environment that promotes normcompliant behavior. Several interviewees seemed to employ constructivist measures of efficacy by expressing cynicism about the potential of the Charter to produce positive outcomes. Three participants from humanitarian/ development NGOs voiced skepticism over whether any meaningful changes were implemented in their organization as a result of feedback, and suggested that the report could be regarded as a bureaucratic exercise rather than a real driver of change. Meaningful change, it was suggested, can only occur when commitment to accountability is 'embedded in the DNA of the organization somewhere'; the Charter cannot deliver such a shift because it is 'only a reporting tool… and that's all this I think is ever going to be' (Int.7).
Other participants agreed about the importance of encouraging strong engagement with the accountability agenda across the organization, and here the commitment of senior leadership was seen as key. It was observed that the Charter cannot hope to have more than limited effectiveness if accountability is not a strategic priority. Jeremy Hobbs, former Chair of the Charter Board, confirmed that the Board was aware of this problem: 'So very often the CEO intellectually gets it, but is not committed emotionally if you like. Or they are committed, but the next layer of staff are not.' The problem of uneven levels of commitment also happens in reverse. The potential of the Charter to promote change could be neutralized if it is seen as a 'pet project' of the CEO and little valued by staff at lower levels of the organization, as exemplified by the following extract from an interview with a participant from a humanitarian/development NGO: 'At the moment the CEO says we do it, so we do it. But the trouble with that approach is you don't get a very consistent buy-in across the organization' (Int.7). These candid remarks about varying levels of reveal that actors will evaluate the efficacy of the Charter in terms of the extent to which it fosters an environment that promotes norm internalization and norm compliance, as constructivists would predict.
Constructivists would also expect that key actors would evaluate the Charter in terms of how well the standards shape expectations of 'rightful conduct.' Normative measures of efficacy were evident when interviewees complained of disconnect between NGOs from different sectors regarding conceptions of accountability. Participants from the advocacy/campaigning sector felt that conversations tend to revolve around service delivery, to which their organizations cannot always meaningfully contribute or learn from. In the words of Clare Doube, a member of the Charter's Board of Directors and the Director of Strategy and Evaluation at Amnesty International: 'Therefore, in terms of the experience sharing, peer learning aspects, I feel we sometimes don't gain as much as some of the conversations aren't really relevant for us.' Such participants also felt that the discourse about accountability that takes place under the aegis of the Charter is primarily framed around the working model of humanitarian/development organizations. The complex interplay between selfinterested and normative concerns is illustrated by the following quote from Janet Dalziell, the Director of Global Development at Greenpeace International, and a member of the Charter's Board of Directors: 'I really struggle with it because these concerns are so driven by the model that relies on government funding-or other very large donors-and we at Greenpeace don't have any of that, and so it's just irrelevant for us. It drives the conversation into a very Northern-focused set of obsessions and worries and discussion that I find don't actually…help us….It has all the potential to really distract us from some more overarching considerations about what accountability is and should be.' The quote reveals unease about the tension between sending signals to key stakeholders and what the participant regards as 'appropriate' accountability practice. Moreover, it illustrates how constructivism can supplement club theory by providing additional dimensions to interpretations of actor behavior.
Some participants also observed that it is relatively easier for humanitarian/development organizations to identify stakeholders than it is for advocacy organizations. 'Stakeholders,' for humanitarian and development organizations, tend to constitute a more sharply defined group of people-the users of a newly constructed well, for example, or the borrowers in a micro-finance initiative. Advocacy/campaigning organizations have a more difficult time in identifying and justifying their key constituencies and evaluating the impact of their activities on the lives of the people that they claim to represent. This gives rise to recurrent debates about what it means to 'do good,' which are particularly tricky when NGOs claim to work on behalf of constituencies who are 'voiceless' (e.g., animals, 'future generations').
Tensions arising from competing notions of accountability are to be expected in some degree in an initiative that attempts to articulate common standards across different sectors, which is after all the unique selling point of the Charter. An interviewee from a campaigning NGO reflected upon the problems involved in establishing a set of cross-sectoral standards that are suited for a wide diversity of organizations and suggested that it could impact upon Charter recruitment: I understand the need for standardization…but I would like to have seen probably a little more openness to flexibility rather than what could be interpreted as judgments based on a framework which works for probably development but not necessarily for all organizations. And I think that probably could be the reason why some organizations may not want to join, because fear of being judged because they don't fit into the reporting requirements-but that doesn't mean to say they're less connected to accountability than anybody else (Int.22).
In sum, the interviewees evaluated the challenges of membership using rationalist and normative measures of efficacy-thus underlining the applicability and complementarity of both club theory and constructivism in understanding the drivers behind peer regulation. Participants offered reflections on a range of diverse topics that included the integrity of the memberships' involvement with the Charter. Advocacy/campaigning organizations expressed theoretical and practical concerns about the compatibility of certain accountability standards to their work. These extracts revealed disquiet about the potential of signaling to sidetrack organizations from engaging in normative debates about accountability. However, evidence of self-interested behavior can be found in the complaints about poor signaling, the cost of membership and the 'burdensome' requirements of reporting. Participants measured the Charter's efficacy both in terms of the extent to which it enables organizations to 'do accountability well,' and to the extent that it serves their interest in portraying members as credible and trustworthy.

Conclusion
The findings of this study are salient for academics and practitioners. For the former, the interview data cast light on the club theory-constructivist debate about the key drivers behind NGO peer regulation. For the latter, the participant's views on the efficacy of the Charter suggest several policy recommendations.
The first question that this study sought to address was: What do the interviewees believe motivated NGOs to join the Charter? The interviewees' interpretations of what constituted 'effectiveness' were informed by their understanding of the reasons why organizations submit to peer regulation.
The literature offers rival explanations for the drivers behind NGO behavior, which are linked to distinct measures of efficacy. Club theory posits that members join a regulatory mechanism to acquire an exclusive benefit: a signal of 'virtue' that is communicated to important stakeholders. It predicts that informants would regard a regulatory initiative as 'effective' if (a) NGOs are compliant; (b) it is widely recognized as a signal of credibility and helps to build trust with principals; (c) it could boost funding; (d) it could discourage governments from encroaching upon NGOs' operational freedom. The interview data suggests that club theory has some purchase, since there was evidence of self-interested behavior. There was general agreement that organizations joined the Charter to demonstrate that they were being proactive in improving their accountability, and to send a reputational signal to donors. Moreover, the fact that the standards have been progressively strengthened lends credibility to the interpretation of the Charter as a club.
Club theorists argue that agents can gain from positive 'network effects' from club membership, resulting in enhanced standing with their principals (Prakash and Potoski 2006: 33). It could be argued that if member NGOs gain from a generalized perception that they are credible organizations, it may not matter if donors are unaware of the specifics of the Charter. Voluntary regulatory activities may have an indirect influence on principals, perhaps leading to, for example, increased funding. Future research could test such a hypothesis by interviewing donors to establish how funding decisions are made.
In contrast to club theorists, constructivists consider the influence of shared ideas, norms and values as key to understanding what shapes forms of peer regulation. They would predict that informants would regard a regulatory initiative as 'effective' if (a) it shaped shared expectations about 'rightful conduct,' (b) the norms were internalized by key actors and (c) it helped to foster an institutional environment that supported norm-compliant behavior. The interview data contained themes that revolved around the integrity of the memberships' involvement with the Charter. There was little indication that the participants' opinions about the efficacy of the Charter were shaped by the presence or absence of financial 'rewards,' which were not mentioned at all. Interviewees expressed their disappointment with the poor recognition of the Charter inside and outside the organization and cautioned that a reporting procedure could not deliver meaningful change alone. They stressed the importance of organizational culture and of individual engagement with accountability norms. They valued opportunities to learn from their peers. The findings suggest that organizations participate in the Charter to satisfy a mixture of 'self-interested' and 'norm-guided' motivations.
The second question that this study sought to address has direct policy significance: To what extent do the interviewees believe that the Charter has been effective in enhancing the accountability of its member organizations? The interviewees discussed three main benefits of Charter membership: Firstly, it provides NGOs with a defense against actual or anticipated criticisms of poor accountability from the media and political opponents. Secondly, membership provides peer learning opportunities. Thirdly, the IRP provides high-quality feedback that can be a useful impetus to boost standards of performance. The interviewees also listed a series of challenges associated with Charter membership. Concerns were raised that the low readership of the reports makes it problematic to maintain 'buy-in' at all levels of the organization. Respondents from advocacy/campaigning NGOs felt that Charter membership is more relevant to the working model and concerns of humanitarian/development NGOs. Lastly, some participants also perceived the reporting process as resource intensive, and several cautioned of the danger that reporting becomes a bureaucratic exercise rather than a real driver of change.
A number of policy recommendations arise from these findings. The Charter company should invest further efforts into raising the profile of the initiative among stakeholders. It should also explore ways in which it can work with members to raise awareness of the initiative among NGO staff. It should work closely with advocacy/campaigning organizations to identify ways to enhance the relevance and value of Charter membership to their work, and expand opportunities for members to engage in peer learning. The Charter company has recently attempted to simplify the reporting process by setting clear maximum limits on the amount of information required-future research into this area might investigate whether these new guidelines have helped to address perceptions that the process is overly bureaucratic.
This study also has implications for peer regulation initiatives more generally. Firstly, NGOs should consider more extensive consultation with their principals about what constitutes an effective signal. It was notable that interviewees cited the ability to shape the accountability agenda, free of donor influence, as a benefit of Charter membership. However, they were also concerned that the Charter sent weak signals because of its low profile. There seems to be some tension between their desire for autonomy and their ambitions for greater recognition. If member organizations want to signal that they are more credible than non-members, principals should ideally not only know about the club, but also have faith in it. That may be achieved by inviting principals to contribute to how verification and certification mechanisms are designed. Creating a stronger signal serves the self-interest of member organizations, and so the initiative will be more likely to be perceived as effective in club theory terms. However, regulatory mechanisms should also promote social learning if they are to be perceived as effective by actors who are motivated by norm-guided, as well as selfinterested, considerations. Organizational learning is best achieved in a forum where actors can admit to failures without fear of punishment (Crack 2013). This is difficult to achieve within a regulatory initiative, as actors may be disinclined to speak with candor if this will undermine the reputational signal sent to principals. NGOs should explain to their stakeholders that owning up to failure can actually improve accountability, as long as lessons are learned and shared with peers. The willingness to disclose evidence of under-performance should be considered as a sign of credibility as long as the club facilitates dialogue about the best practice. In this way, the measures of efficacy employed by club theory and constructivists can be better aligned.
This article is a starting point in addressing the knowledge gap about the effectiveness of the Charter. It is a timely juncture for further research to be conducted into the Charter as they inaugurate a 'Global Standard for CSO Accountability' with other global networks and embrace an ambitious new strategy to expand its membership (INGO Accountability Charter 2015e). Accountability is a centrally important value for progressive NGOs, so it is in the interests of practitioners and stakeholders to ensure that policy is designed in accordance with a robust evidence base.