The regulation of private military and security companies: Analyzing power in multi-stakeholder initiatives

ABSTRACT

This article studies the limitations of multi-stakeholder initiatives (MSIs) relating to Private Military and Security Companies (PMSCs). It draws attention to three distinct ways in which power operates in and around MSIs: rules, structural positions, and discourses. Based on an analysis of two MSIs, it shows that these governance initiatives strengthen the perspectives of stakeholders that consider PMSCs as normal and legitimate security actors. Western governments and like-minded actors have used the Swiss Initiative and the International Code of Conduct for Security Service Providers to bypass the less privatization-friendly process in the United Nations. MSIs equally perform an important legitimizing function for PMSCs through their discourses and practices. Finally, participants of the MSIs have relegated critical voices, weakening their ability to partake in governing the PMSC industry. By studying the limitations of MSIs through a power-analytical lens, this article therefore points at an important but overlooked dimension.

KEYWORDS:

• Multi-stakeholder initiatives
• power
• Private Military and Security Companies
• transnational governance

Since the 1990s, multi-stakeholder initiatives (MSIs) have emerged as an increasingly common mode for regulating the conduct of business in a range of sectors and issue areas (Dingwerth & Pattberg, 2009). Private Military and Security Companies (PMSCs) have not been immune to this trend as evidenced by the launch of the Swiss Initiative in 2005, the International Code of Conduct for Security Service Providers (ICoC) in 2010, and its oversight body, the ICoC Association in 2013, all of which bring together representatives from states, civil society, and PMSCs. For their proponents, these initiatives promise some sort of regulation after years of unregulated growth (Buzatu, 2015). Their multi-stakeholder character is, in this regard, frequently lauded for leading to effective governance of the industry (Avant, 2016). In addition, as a new mode of governance, there is normative appeal both in the deliberative, consensual character that allows for participation by all categories of stakeholders, and the ability to solve problems through concerted action (Schäferhoff et al., 2009; Van Tulder, 2012).

The strong focus on open deliberation, consensus, and problem-solving, however, fails to recognize that MSIs do not only entail relations that promote free and unconstrained action, but also those that approximate states of domination. As Westerwinter (2016) notes, “[n]ew modes of security governance are not necessarily less hierarchical and more inclusive than traditional state-based forms of political regulation” (p. 222), but replete with power. Exercises of power, if not defined in terms of coercion or open conflict, may be less visible in MSIs; but power is present all the same. By studying the prospects and limitations of multi-stakeholder security regulation through a power-analytical lens, this article directs us to an important but overlooked dimension of the MSIs.

Based on an analysis of the Swiss Initiative and the ICoC—the first and most prominent MSIs that set standards for PMSCs and states that use, host or are home to these companies—this article argues that these governance arrangements are replete with power. They are heavily biased towards those actors that accept PMSCs as normal and legitimate security actors; impose constraints on the behavior of their NGO participants; marginalize dissenting voices; and legitimize the status quo of security privatization. It arrives at this conclusion by looking at the power dynamics in and around MSIs from different angles—including institutional, co-constitutive, and productive types of power (Barnett & Duvall, 2005)—and the outcomes they generate. It shows that institutional power manifests itself in the ability of major states and the PMSC industry to bypass the United Nations (UN), which has been more reluctant to legitimize the large scale use of PMSCs, in favor of the Swiss Initiative and the ICoC. Institutional power equally works through the membership rules of the two MSIs, which have a disciplining effect on their NGO participants. The concept of co-constitutive power, in turn, draws our attention to how MSI-critics are placed in a structurally disadvantaged position of “non-experts.” Finally, the Montreux Document and the ICoC are productive in that they allow PMSCs to portray themselves as good, respectable and legitimate actors.

Scholars have long debated the shifting location of power in the field of security from the public/state to the private/market (Abrahamsen & Williams, 2011). This shift has been linked to the productive power of PMSCs, that is, their ability to shape discourses and identities in a way that legitimizes their use (Joachim & Schneiker, 2019; Kruck & Spencer, 2013; Leander, 2005; Prem, 2020). Others have shown that PMSCs derive power from prevailing norms and ideas that favor private, market-based modes of governance (Cutler, 2010; Kruck, 2014; Leander, 2005; Leander & van Munster, 2007). So far, one important source and site of power in private security governance has not been investigated: multi-stakeholder regulation of PMSCs. The article suggests that the oft-cited normalization of PMSCs is not only a result of these companies cultivating positive self-images; it can also be traced to the power dynamics of MSIs. They empower PMSCs in various ways: by diverting attention and energy away from more intrusive kinds of regulation, constituting them as legitimate governors, and muting critics of the industry. These findings provide a different angle from where we can assess the desirability and effectiveness of MSIs and other transnational governance initiatives in the realm of security (and beyond). While scholars are indeed critical of MSIs and their performance (Baumann-Pauly et al., 2016; Mena & Palazzo, 2012), they have neglected a central source of those deficiencies: the play of power.

The article proceeds as follows. The first section introduces Barnett and Duvall’s taxonomy of power and discusses its relevance for the analysis of MSIs. Section two goes on to operationalize institutional, co-constitutive and productive types of power for the subsequent analysis. The third section examines the power dynamics of the Swiss Initiative. It argues that the United States and the United Kingdom have used the Swiss Initiative to sideline the UN Draft Convention process and pursue a privatization-friendly agenda with like-minded states. The Swiss Initiative also performs an important legitimizing function by accepting PMSCs as an unavoidable feature of contemporary warfare. The fourth section shows how dissatisfaction with the process in the UN has driven PMSCs to create an alternative institution, the ICoC. By giving PMSCs an equal standing at the negotiating table, the ICoC process allows the industry to steer the regulatory agenda into a direction that better serves its interests. Moreover, membership status implies a “social upgrading” of the controversial industry. Section five sheds light on the operations of power that cut across the two MSIs: the power relations that exist between members and “outsiders” and the indirect control of NGO participants through the membership rules of MSIs. Section six concludes with a summary of the key findings and discusses their relevance beyond the issue of private security.

Conceptualizing power in MSIs

This article takes Barnett and Duvall’s (2005) typology of power as a starting point for exploring the multiple operations of power in and around MSIs. They define power as “the production, in and through social relations, of effects that shape the capacities of actors to determine their circumstances and fate” (p. 44). MSIs, as will be argued, produce particular kinds of effects by differentially enabling and constraining their participants to partake in the governance of the PMSC industry. Barnett and Duvall (2005, pp. 45–48) divide power into two categories: the specificity of social relations (diffuse/direct) and how power is expressed (interaction/constitution). This renders four types of power, out of which three will be discussed in detail below: institutional, structural (or co-constitutive), and productive.¹

Institutional power

Institutional power is an indirect way of exercising control, namely through the rules and procedures, formal and informal, that guide, steer, and constrain the behavior and conditions of existence of an actor (Barnett & Duvall, 2005, p. 51). Institutional power analyses often focus on the behavioral constraints and governing biases of institutions that favor some groups or actors over others (Barnett & Duvall, 2005, p. 52; Steinberg, 2002, p. 341). The design of institutions and rules is not a contingent choice but reflects the ability of certain actors “to further or preserve their interests and positions of advantage into the future” (Barnett & Duvall, 2005, p. 58; de Souza, 2015; Ikenberry, 2001). “[L]aw,” as Abbott and Snidal (2013, p. 39) put it, “is always to some extent ‘victor’s law,’ reflecting the values and interests of those who were best positioned to affect its development.” Here, power is manifesting itself in the ability of prominent actors to influence rule-setting to further their own strategic and economic interests (Cheyns & Riisgaard, 2014, p. 413).

Yet institutional power is not limited to direct-rule setting power. It also includes an actor’s ability to deny other actors the opportunity to vote for alternative types of regulation that would be detrimental to them (Bachrach & Baratz, 1962; Gruber, 2005). Rather than applying direct pressure, an actor can wield indirect agenda-setting power by preventing certain issues (or regulations) from being taken up by an institution (Gruber, 2005, p. 205). Where does this kind of power derive from? Forum shopping, defined as the “strategic selection and use of policy venues by actors in order to advance their policy goal,” is one way to influence which issues enter into or which decisions are made in an institution (Murphy & Kellow, 2013, p. 139). The proliferation of international institutions creates a powerful incentive for state and non-state actors to forum shop, providing them with an ever-greater number of venues to select from in order to pursue their goals (Raustiala & Victor, 2004).

Private security regulation is a case in point. Absent binding regulation, a multiplicity of soft law initiatives has mushroomed, including MSIs like the Swiss Initiative and the ICoC (DeWinter-Schmitt, 2017; Leander, 2016a). Morse and Keohane (2014) distinguish two strategies: shifting the focus from a status quo institution to another that would serve one’s interests better (forum shifting) or, if no alternatives are available, creating a competing institution (forum creation). Braithwaite and Drahos (2000, p. 564) identify a third strategy which may be undertaken in parallel to the other two options: forum blocking. In this case, an actor or a coalition of actors prevents an institution from becoming a forum for an agenda that threatens its interests. The rationale behind these strategies is that an actor or a group of actors tries to evade unfavorable outcomes or institutional characteristics (e.g., in terms of membership, mandate, decision making rules and procedures) by shifting to a “new game” more in line with their preferences (Morse & Keohane, 2014; Murphy & Kellow, 2013, p. 145).

Co-constitutive power

Co-constitutive power works through direct interactions between/among actors that simultaneously create their identities and social capacities for action. This type of power sheds light on the way in which actors relate to one another, namely through their structural position in a system like, say, the global capitalist system (capital/labor). The key premise is that neither category (of states, etc.) can exist without the other and that the structural position (and identity) of actor A is always defined by the position (and identity) of actor B and vice versa (Barnett & Duvall, 2005, p. 53). The relative position of actors matters because it generates differential privileges and capacities to the occupants of these positions or identities—it generates power. It constrains or increases actors’ options by shaping who they are and what they want. Structural positions therefore have a socializing and self-disciplining effect (Barnett & Duvall, 2005, p. 46). They “teach” actors to play certain roles and behaviors to the point that they eventually come to accept their position in the “order of things” as normal and inevitable (Mattern & Zarakol, 2016, p. 638).

Productive power

In contrast to the co-constitutive variant, productive power is much more pervasive and decentered. It is not limited to hierarchical and binary relations of domination in which it is readily apparent who the “powerful” and “powerless” agents are (Hayward, 2000, p. 36), but it can operate in the absence of actors designated as “powerful” and without an apparent connection between A and B. In other words, productive power entails more generalized and diffuse social processes that produce all social subjects and the space of world politics in which they act (Hayward, 2000, p. 6). Productive power analyses take an interest in how historically and contingently produced discourses and practices shape the subjectivities of all social beings, including their desires and capacities for action (Barnett & Duvall, 2005, p. 56). Discourse is central to the operation of power; it is socially productive for all subjects by shaping how they define and organize both themselves and their social world (Butler, 1999). The performative aspect of language suggests that discourse is neither a neutral practice of stating what is, nor based on some kind of universal truth; rather, it purveys a univocal and contingent representation of reality that may temporarily gain the status or currency of truth (Mattern & Zarakol, 2016, p. 641).

Law is a particularly powerful discourse because of its claim to truth, rationality and objectivity and its ability to depoliticize certain issues (Barnett & Finnemore, 1999, p. 708; Smart, 1990, p. 197). It fixes and freezes certain meanings which themselves become an authoritative point of reference until the law is altered or repealed. This claim to objectivity and rationality merely disguises the extent to which law itself is deeply implicated in making, rendering and knowing the social world in particular way rather than in another. Studying law as a system of meaning that escapes the “provenance of formal lawmaking” has therefore taken a prominent place in constructivist research (Johnson, 1996/1997; Krahmann, 2012; Leander, 2012). From this point of view, legal concepts such refugees, mercenaries or criminals represent the construction or reinforcement of social categories and relationships, including the construction of “others,” the justification of the status quo and the legitimization of certain policies (Van Leeuven & Wodak, 1999).

Operationalizing power in MSIs

Institutional power can be approached from two angles. From the vantage point of “rule takers”—those located at the receiving end of the power relationship—the rules and norms of MSIs may work as constraints on their behavior. Following Joachim and Schneiker (2015) who have argued that MSIs have a disciplining effect on their NGO participants, I will focus on the informal membership rules of MSIs and the way they impact on NGOs’ ability to criticize and contest PMSCs. A first good measure in that regard is to compare NGOs’ positions on the PMSC issue before and after the set-up of the Swiss Initiative and the ICoC process. To this end, I have analyzed the advocacy material of five NGOs that have championed the issue of PMSCs: Amnesty International USA, Human Rights First, the ICRC, Global Policy Forum, and War on Want. This sample covers both organizations that have been members of at least one of the MSIs considered here and those that have tabled alternative regulatory arrangements and who did not partake in these arrangements. This allows for a differential analysis of the way (non-)membership affects NGOs. For our basic assumption about “power as constraints on behavior” to hold, we would expect insiders to town down their message, while outsiders framing is likely to remain unaltered.

From the perspective of power holders, institutional power denotes the ability to block or shift the focus away from an institution that threatens the interests of an actor or a coalition of actors. This article focuses on the regulatory competition unfolding between the UN, which has long been the central venue in which regulation of PMSCs could be discussed, and the Swiss Initiative and the ICoC. Although the two MSIs are by no means the only governance initiatives in the field of private security regulation, they have by far attracted the highest level of recognition and adoption. To make the case for institutional power, the article compares the UN Draft Convention and the ongoing work of the Intergovernmental Open-ended Working Group (OEWG) with the Swiss Initiative and the ICoC, focusing on the following institutional characteristics of each forum: membership, mandate, and decision-making rules.

The central premise is that the two MSIs have become the preferred venues for certain kind of actors to pursue their regulatory agenda. It will then trace the agency of the most-likely candidates—the United States, United Kingdom, and PMSCs—in accomplishing the shift away from the UN.² The United States and United Kingdom can be considered “crucial cases” insofar as any regulation of the PMSC industry will critically depend on their support. They are not only home to some of the largest PMSCs worldwide, but also major consumers of their services. This puts them in a favorable position to resort to forum shopping. My analysis of both variants of institutional power rests on documentary analysis, expert interviews and, where primary data is missing, secondary sources (including accounts of scholars who attended the negotiations leading to the Montreux Document or the ICoC) to provide us with an expert assessment of the extent to which these actors were able to forum shift, block, and create—to exercise institutional power.

Co-constitutive power is the product of actors’ direct relations that shape both their understanding of who they are, what they want and, by extension, their capacities for action. This article puts particular emphasis on the relationship between MSI members and non-members. The assumption is that exclusion, whether voluntary or not, places non-members in a subordinate position, namely that of non-experts. From the potential universe of non-members, I will confine my analysis to actors which are active in the broader field of private security regulation but work outside the confines of MSIs. This includes not only the UN Working Group but also the two NGOs mentioned earlier: War on Want and the Global Policy Forum. While it is difficult to measure the co-constitutive dimension of power directly (i.e., how it shapes the position-contingent identities of non-members and their capacity to partake in the governance of the PMSC industry) a good approximation is to examine self-representations and processes of “othering” (Casier, 2017, p. 107; Joachim & Schneiker, 2019, p. 7). How do non-participants conceive of their role in relation to these multi-sectoral forums? And how do participants construct categories of the other? Based on interviews that I have conducted in 2014 and 2017 for a larger research project on private security regulation with both members of MSI and organizations that have continued to work outside these fora, I will look for patterns that fit into the categories “inside” and “outside.” In a second step, the paper will discuss the (likely) consequences of being grouped into either category in terms of actors’ ability to partake in the governance of the PMSC industry.

Finally, a consideration of productive power in MSIs calls for an analysis of the meanings that inhere in the Montreux Document and the ICoC and that have served to constitute and legitimize (or delegitimize) PMSCs as a social and legal category. Following Wodak’s discourse analytical approach, this involves the identification of three strategies (Van Leeuven & Wodak, 1999): constructive strategies which “build” and “bring into being” particular groups, such as “private military and security companies” or the “socially responsible company”; strategies of perpetuation and justification which seek to preserve and justify the status quo of security contracting and privatization; and strategies of transformation which attempt to redefine well-established understandings. It bears noting, though, that the performative dimension of MSIs has not only a discursive/textual dimension; it also involves certain (technical) practices such as certification, monitoring or performance assessments that produce the very actors they seek to “govern” (Krahmann, 2017). Therefore, the article seeks to reconstruct how these practices produce a particular identity for PMSCs that have underwent certification or monitoring in MSIs. According to a performative definition, certification and monitoring produce already what they seek to achieve: the creation of the “compliant” and “socially responsible” PMSC—irrespective of how these companies perform on the ground.

Forum shifting: The Swiss initiative

Despite mounting allegations of contractor misconduct, regulation of PMSCs has long proved elusive because it was unclear which laws, if any, applied to them and who was legally responsible for their conduct. The Montreux Document is the first international attempt to flesh out the responsibilities of states in an issue area where they have not been considered before: the operation of PMSCs in an armed conflict environment. It is born out of a joint initiative, sponsored by the Swiss government and the ICRC, to reiterate the rules and principles under international humanitarian law and human rights law that apply if PMSCs operate in zones of armed conflict. Although the Swiss Imitative was essentially state driven, it has benefited from the input of a range of other stakeholders, including states, international organizations, industry representatives, civil society groups and academics, who were consulted through four expert meetings between January 2006 and April 2008. The outcome of these efforts, the Montreux Document, was endorsed by 17 states in September 2008.

At roughly the same time, two consecutive UN working groups have explored the possibilities for creating binding regulation of the industry through international conventions. The UN has a long pedigree in dealing with private force, reaching back as far as in the era of decolonization when mercenaries were hired to destabilize newly independent states in Africa. It was largely the experience of these states that fueled anti-mercenary sentiment in the UN and led to the adoption of the UN International Convention against the Recruitment, Use, Financing, and Training of Mercenaries in 1989.³ The appearance of PMSCs on the international stage reinvigorated international regulatory discussions within the UN which continued through the existing office of the Special Rapporteur for mercenaries. In 2005, this one-(wo)man position was replaced by the UN Working Group on the Use of Mercenaries, a collective body comprising five independent experts, each representing five geographical regions. The UN Working Group’s mandate is to study the impact of new manifestations of mercenary or mercenary-related activities (in particular the activities of PMSCs) on the enjoyment of human rights, monitor such activities, and prepare a draft of international basic principles to encourage respect for human rights by those companies. The work of the Working Group culminated in the submission of the UN Draft International Convention on the Regulation, Oversight and Monitoring of Private Military and Security Companies (hereafter: the UN Draft Convention) to the Human Rights Council (HRC) in 2010. On this basis, the HRC established the OEWG tasked with further pursuing the treaty option.

From its inception, the UN process was scorned by industry representatives and prolific users and suppliers of PMSC services, especially the United States and United Kingdom. The Working Group’s continued use of the term “mercenary” was one source of concern. In an open letter submitted to the Working Group, head of the industry association International Stability Operations Association (ISAO), Brooks, recommended removing the term “mercenary” from both its name and mandate. The group’s continued use of the “m-word” was “derogatory” and would stand in the way of collaboration on regulatory issues (Brooks, 2009). Similarly, the UK and US have repeatedly disputed the competence of the Working Group to deal with the issue on the grounds that PMSCs “could not be considered mercenaries” (see statement by Peter Gooderham for the UK Human Rights Council, 2010b).

Another concern was the hard law approach which, until quite recently, has prevailed within the UN as both the aim of the Draft Convention and the mandate OEWG is to create a legally binding instrument. Such an instrument would have placed severe limitations on what functions would be allowed to be contracted out to PMSCs. The Draft Convention adopts a very restrictive understanding of the state monopoly on the legitimate use of force. It is based on the premise that there are inherently governmental functions that should not be delegated or outsourced to PMSCs, including

direct participation in hostilities, waging war and/or combat operations, taking prisoners, law-making, espionage, intelligence, knowledge transfer with military, security and policing application, use of and other activities related to weapons of mass destruction and police powers, especially the powers of arrest or detention including the interrogation of detainees. (Human Rights Council, 2011, p. 5)

If enacted, these limits would have amounted to nothing less than a partial ban on PMSC activities. Moreover, for powerful states such as the United States and United Kingdom, the UN process had the effect of diluting their overall influence due to its consensual, all-inclusive approach and the variety of dissenting voices in the HRC which has the final authority to decide over the recommendations of the Working Group and the OEWG. As Percy (2007) notes, the HRC has traditionally taken a critical view of private force in all its incarnations, resulting from a majority of states that were directly affected by mercenary or PMSC activity. It is therefore no coincidence that the Draft Convention represented the views of states which were more reluctant to legitimize the large-scale use of PMSCs, to include those representing the African Group and the Organization of Islamic Conference, but also Russia and China (Seiberth, 2014, p. 129; White, 2011, p. 150).

The efforts of the Working Group and its successor, the OEWG, were therefore met with resistance from the major home and contracting states of PMSCs. They were anxious to retain their flexibility in using PMSCs. The United States, in particular, was relying on the industry to an extent that going to war without them has become virtually impossible (Commission on Wartime Contracting, 2011, p. 19). Both states favored a soft law approach that would legitimize (the use of) these companies (Foreign and Commonwealth Office, 2009; see statement by Mark Cassayre from the US in the Human Rights Council, 2010b). Thus, under the impression of mounting regulatory activity from the UN, they pursued two complementary strategies: forum blocking and, once the Swiss Initiative took off, forum shifting.

Regarding the former, Western governments have repeatedly obstructed the labor of the Working Group and the OEWG by voting against its establishment, mandate, and recommendations (del Prado, 2010, p. 1). The debates in the HRC following the submission of the Draft Convention are case in point. State delegates from the United States and United Kingdom made clear that they did neither support the Draft Convention proposed by the Working Group nor the call for an open-ended Working Group (see country statements in Human Rights Council, 2010a; Human Rights Council, 2010b). These objections were not only based on the link between PMSCs and mercenaries made in the resolution extending the mandate of the Working Group. They also bore on the binding character of a future UN convention and the competence of the HRC which, in some countries’ views, was not the appropriate forum for addressing the matter of PMSC regulation. In the end, the resolution inaugurating the OEWG was passed without support by Western states, with 32 in favor, 12 against and 3 abstentions (Human Rights Council, 2010a). Instead, these states shifted their focus to the Swiss Initiative (and later the ICoC) which aligned more closely with their interests for a number of reasons.

First, although the Montreux Document contains elements of hard law, it is not a biding international treaty, but a but a piece of soft law. In fact, a number of caveats were introduced throughout the negotiation process to stress its essentially hortatory nature. An earlier draft prepared by the Swiss government and the ICRC that had foreseen a binding framework was abandoned under the pressure from a number of states, including the United States, United Kingdom, and Canada which were unwilling to accept a legally binding treaty (Cockayne, 2008, p. 423). Since the Swiss Initiative has involved only a limited number of mostly like-minded Western states, namely those falling into the category of home and contracting states, the latter could make their voices more easily heard than in the UN process. At the same time, the inclusion of industry representatives and NGOs, particularly of a neutral institution like the ICRC, gave the Swiss initiative an inclusive and legitimate imprint, even if these actors did not enjoy equal decision-making rights. On the contrary, states have used their convening power to exclude non-state actors from the final round of negotiations. This step was taken to bypass oppositional input from human rights groups who urged to retain the obligatory character of the draft and pushed for a stronger human rights component (Cockayne, 2008). The Document differs from the UN draft in other important respects. It is conservative insofar as its primary purpose is to restate existing legal obligations and practices and consider them for a new context (the operation of PMSCs in an armed conflict environment), not to create new ones. This means that Montreux did not impose any new constraints on states beyond the commitments they had already made (Percy, 2012, p. 954). Institutional power, as these examples suggest, derives not only from states’ ability to forum shift, but also from their disproportionate influence in setting the rules of the game.

Finally, the document performed an important legitimizing function by ridding PMSCs of the mercenary image (transformation) and normalizing their use (perpetuation). This brings us to another facet of power, that of productive power. While the Working Group long suggested that PMSCs were simply a new incarnation of mercenaries, Montreux did not make any reference to mercenarism or anti-mercenary law in relation to PMSCs. Instead, the drafter agreed that PMSCs and their employees represented a distinct phenomenon (Tougas, 2009, p. 323). Moreover, the Montreux Document allowed a wider range of activities to be outsourced than the Draft Convention, reifying the state of security privatization in a seemingly unquestioned manner. Despite its declared intent of remaining neutral, the document accepts PMSCs as a “a fact of international life” (Stürchler, 2008, p. 10), as something that cannot be ignored but has to be dealt with in a pragmatic fashion. This statement in itself performs an important legitimating function. It turns PMSCs into an indisputable “given” and places them beyond the possibility of dissent. Such taken-for-grantedness constitutes a particularly effective source of power precisely because it prevents people from imagining alternatives (Lukes, 2005, p. 27).

Since the launch of the Swiss Initiative and the ICoC process, the UN has struggled to remain relevant as a regulatory institution. Once the Montreux Document and the ICoC were completed, the US and UK disputed the need for a new treaty which, in their view, “would divert resources, time and attention from more constructive approaches [the Montreux Document and the ICoC]” (statement by Eileen Chamberlain Donahoe from the US in the Human Rights Council, 2010a). Without such backing from key home and client states, the Draft Convention process is likely to suffer the same fate as the 1989 Mercenary Convention which has become largely irrelevant because it failed to elicit sufficient state support. The current stalemate in the OEWG, which remains hamstrung by two seemingly irreconcilable blocs—one skeptical of PMSCs and the other taking more “pragmatic” positions—does not bode well for the UN process. At the same time, the Working Group has been excluded from discussions about the Swiss Initiative and the ICoC despite its continued interest in these processes (del Prado, 2008, p. 443; Leander, 2016b, p. 225). In light of these developments, the recent departure of the OEWG from one its core aims—the elaboration of a legally binding convention—can be read as a belated recognition that its regulatory agenda has reached an impasse. At the sixth session of the OEWG, state delegates decided “to freeze discussion of whether to recommend the creation of a legally binding instrument” (Human Rights Council, 2017, p. 7).

Forum shifting and forum creation (through the ICoC) have also created additional bargaining leverage vis-à-vis the UN process. Despite their continued skepticism of this forum, leading contracting and home states engaged in the sessions of the OEWG to influence the design of a future regulatory framework. They wanted to make sure that a new draft would include references to the Montreux Document and the ICoC as the only two documents that dealt specifically with PMSCs (Human Rights Council, 2017, p. 6). In this sense, regime shifting may have additional feedback effects in the UN venue by enabling “insiders” to introduce “specifics standards and tools that had not been universally drafted and accepted and were not part of the UN process,” but that “outsiders” are subsequently compelled to observe, as some state representatives have bemoaned (Human Rights Council, 2017, p. 7).

Forum creation: The International Code of Conduct

For the PMSC industry, the question was not whether they should be regulated but how. While industry representatives welcomed the Montreux Document, they articulated a need for industry best practices, particularly in zones of weakened governance where states were unable to uphold their obligations under the document (Buzatu, 2015, p. 27). Growing dissatisfaction with the UN process created additional impetus for pursuing regulation through a new institution. From its beginning, industry representatives conceived of the ICoC an alternative to and “competing with” the ongoing work of the UN which they felt was unduly biased against PMSCs (Interview#4, 2014). Thus, when the Montreux Document was drawing to an end, industry representatives solicited Switzerland to help them develop standards that would be both practical and less biased. Switzerland answered this call by kicking off consultations with the most implicated stakeholders: the industry, civil society/academic, and various governments during the first part of 2009. During these consultations it became clear that a multi-stakeholder approach would lend the initiative further credibility. Thus, even if the idea of an industry-wide code was “essentially industry-driven” (Buzatu, 2015, p. 28), forum creation was not a unilateral move. Instead, PMSCs benefited from the support of a number of governments and NGOs who joined the undertaking for different reasons.

In a series of multi-stakeholder meetings from 2009 to 2010 the final ICoC document was produced and ultimately signed on November 9, 2010. To lend the Code “teeth,” the initiators took additional efforts to develop an oversight and governance body, the ICoC Association, which would take over three core functions: (1) certification of signatory companies’ compliance with the Code and with the standards derived from it, (2) auditing and monitoring of PMSCs’ performance in the field, and (3) a mechanism to address violations of the code. The ICoC Association takes up a multi-stakeholder approach. It has a three-pillared governance structure that allows for equally weighted participation and decision-making from the three stakeholder groups: civil society, PMSC industry, and governments.

As much as the ICoC implied a wholly new level of scrutiny and constraints to which the industry was subjected, it allowed PMSCs to retain the ownership over their fate. Industry figures have repeatedly claimed that “we do not have a problem with regulation, we would just like to be part of it” (Spicer, 1999, p. 170). Contrasting the ICoC initiative with the UN Working Group process, an industry representative resumed that

[t]he good thing about the Code process is that we, industry, have a seat at the table to talk about what makes sense, what doesn’t make sense, what’s doable, what’s not doable. Whereas, when you’re looking at the UN Working Group process, it’s going to be the UN […] and all the folks who are all largely academics or government members. I met with the UN Working Group two or three times and it was never what I would call collaborative discussion. (Interview#4, 2014)

Even though the UN-led process includes consultations with PMSCs, they remain embedded within a fundamentally state-centered framework.⁴ In contrast, the distinct appeal of the ICoC is that it raises PMSCs to equal partners in the regulatory process. This is reflected in the voting structure of the board, the primary decision-making body of the ICoC Association, in which decisions require a majority of two thirds board members, including a minimum of two votes from each pillar. This procedure prevents that the industry pillar (or any other stakeholder group) can be overruled by a voting bloc of the two others. Moreover, participation in the ICoC process has been a priori limited to actors who shared a basic understanding about PMSCs and how they should be regulated: the belief that the industry is here to stay and therefore should be permissively regulated rather than entirely banned (Schneiker & Joachim, 2018). Such exclusivity of membership meant that PMSCs could advance a policy agenda that aligned more closely with their interests than the Draft Convention. As a consequence, the Code reflects the privatization-friendly sentiment of those stakeholders that were instrumental in bringing it about. It does not take issue with PMSCs per se, but tries to turn them into “better companies.”

This directs our attention to another power dimension—that of productive power. Like the Montreux document, the ICoC performs an important legitimating function. In many ways, it is even more outspoken in its judgement on the legitimacy of the industry. The preamble of the Code (art.1) starts out by recognizing that

Private Security Companies and other Private Security Service Providers (collectively “PSCs”) play an important role in protecting state and non-state clients engaged in relief, recovery, and reconstruction efforts, commercial business operations, diplomacy and military activity. In providing these services, the activities of PSCs can have potentially positive and negative consequences for their clients, the local population in the area of operation, the general security environment, the enjoyment of human rights and the rule of law.

In other words, the ICoC reaffirms that PMSCs have a valid role to play in security governance (perpetuation). Unlike the Montreux Document, the ICoC has also dropped any reference to the “military” attribute in describing the industry, which had long dominated the legal debate (transformation). This choice of terminology reflects the industry’s desire to avoid any association with proper military operations, including interrogation and training assignments which would have revived older images of mercenarism (Federal Department of Foreign Affairs, 2010, p. 27; Jerbi, 2013, p. 9, 18). During the drafting process, PMSC representatives thus made plain that they did not consider themselves “military” companies in any way and pressed to erase the label from the original draft (Federal Department of Foreign Affairs, 2010, p. 27). Restricting the Code’s applicatory scope to “security” contractors implies that military companies have actually ceased to exist or are confined to relatively unproblematic niches such as transport, catering, laundry, sanitation or camps, maintaining or operating weapons systems. This assumption is problematic for a number of reasons. One is that support functions such as training, moving military troops and war equipment, maintaining weapons systems and intelligence provision are all “key aspects of waging war” that are no less integral to the use of force than actually pulling the trigger (Hilary, 2007; War on Want & Campaign Against Arms Trade, 2006, p. 8). The other reason is that the Code’s concern with private security contractors obfuscates the new activities that PMSCs now engage in: the provision of military services in the cyberspace and the operation of drones, both of which carry the prospect that PMSC personnel directly participate in hostilities (Liu, 2015; Prem, 2018). In other words, the Code creates the illusion that only security contractors are problematic and require regulation, whereas the military category ceases to be a relevant descriptor for the industry.

MSIs are performative in yet another way. They shift company’s status from “deviant” to “compliant” and allow them to rectify their tarnished image as Harald Koh, then legal adviser to the U.S. Department of State, summarized at the Code’s singing ceremony:

By demonstrating a commitment to high standards and an effective compliance mechanism, signatory companies will be more attractive to all potential clients and address the negative perceptions that have followed this industry. (Koh, 2010)

Signing the Code can be interpreted as a performative act that brings into being a new kind of actor: the socially responsible PMSC (strategies of construction). It is a crucial marker of respectability, professionalism and ethical conduct that helps to distinguish “good” and “bad” companies (Interview#3, 2014). The normative commitment to the ICoC may even compensate for genuine changes in policy or behavior. For example, adherence to the Code has long been merely symbolic because there were no initial barriers for determining eligibility to sign. As a consequence, the initial number of signatory companies was heavily inflated, with 708 PMSCs having signed by February 2013. Now that the status of “signatory company” has been replaced by “transitional membership” in the Association (International Code of Conduct Association, 2017), this number has gone down to 95 PMSCs. Transitional members can maintain membership for two years during which they have to work towards full membership status by obtaining certification by the Association. These examples illustrate the performative dimension of the ICoCA seal which fulfills an important legitimating function as long as the Association’s monitoring, reporting, and grievance arrangements are still underway.

Power’s effect on “insiders” and “outsiders”

The growing pro-privatization consensus has gained currency way beyond MSIs. This has been achieved by marginalizing actors which have taken strong and non-pragmatic positions—to include War on Want, Global Policy Forum, and the UN Working Group. Their critical work and refusal to engage in MSIs were repeatedly dismissed as unrealistic, dogmatic and counterproductive by those participating in MSIs (Interview#10, 2014; Interview#2, 2014; Interview#4, 2014). This dismissal of critique as unreasonable is illustrative of patterns of subordination which work through the mutual constitution of an in-group of MSI supporters—those “groups who understood that the industry was going to exist” and were therefore willing to engage in discussions (Interview#4, 2014)—and an out-group of MSI critics. It showcases the workings of co-constitutive power. The group of MSI critics is not only defined by the way it relates to the ICoC, but also by a specific kind of expertise on the PMSC issue, characterized by a hard-law approach and a refusal to accept the terms of debate set by MSIs—the idea that it is necessary or vital to delegate military and security functions to profit-making entities. War on Want exemplarily represents such a stance. The NGO sees MSIs as a retrograde step to regulate PMSCs and presses for an end of the privatization of security (Interview#6, 2014). The neat dichotomy between outsiders/non-experts and insiders/experts is summarized by an MSI participants as follows:

There are organizations [that] accept that there is a PMSC industry and therefore seek […] to help that industry perform in a proper fashion. And there are some NGOs who would still almost deny the fact that there is an industry and would like to see it all closed down. That sort of position I don’t think is strictly helpful because, to my mind, it’s not going to happen. […] So whether people like it or not, to be frank, […] the industry is there to stay and the important thing is making sure that those companies that deliver services to a high standard are recognized doing so […] (Interview#10, 2014)

As this citation suggests, expert status accrues to actors which take “realistic,” “pragmatic,” and “constructive” positions as opposed to those exhibiting “unrealistic,” “dogmatic,” and “counterproductive” views. Whether an actor is positioned at the pragmatic or radical end of the spectrum is deeply consequential. For one, it defines which actor is authorized to speak or being listened to. At this point, it is also helpful to recall the fate of the UN Working Group which has disqualified itself from taking part in the larger regulatory debate because of its non-pragmatic, abolitionist positions. In a similar fashion, a participant of the Code process has made clear that he considered it “[not] even worth talking to them [War on Want]” (Interview#4, 2014). Thus, “outsider” status weakens the position of actors which are interested in the ongoing regulatory discussion but refuse to accept the terms of the debate set by MSIs. For another, the marginalization of dissenting voices makes it possible to present MSIs and their permissive thrust of regulation as if they were consensual.

As much as NGOs outside MSIs were struggling to prevail against the growing pro-privatization bias, NGOs were facing different, often unanticipated odds within the two MSIs. According to my interview partners, membership in MSIs did require all sides to internalize some unwritten rules of conduct: To engage respectfully with each other, search for compromise, and take pragmatic positions that would not overly alienate the other stakeholders (Interview#1, 2014; Interview#4, 2014; Interview#8, 2014). This expectation was particularly upheld by PMSCs for whom a conditioning factor to engage with NGOs in the first place was a commitment to “talk […] about solving the problems that come up” rather than pursuing radical campaigning strategies (Interview#2, 2014; Interview#4, 2014). This is acknowledged by industry figures such as Doug Brooks who urged that “[i]nstead of obstruction, energies should focus on ensuring the most, ethical, transparent, accountable and effective use of the private security companies in the field” (Brooks & Chorev, 2008, p. 116). This may have curbed NGOs’ willingness to speak up and criticize, let alone to campaign against the industry, as confirmed an NGO representative for whom “the price of being involved in those initiatives is that one is not allowed to speak outside to the public” (as cited in Joachim & Schneiker, 2015, p. 198).

In other words, membership in MSIs acts as an aggregate constraint to forgo radical campaigning against companies (institutional power). This is also apparent from my content analysis of NGO publications which, from 2009/10, have become more moderate over time. To track changes or continuities in the advocacy material of both MSI-participants and non-members, I have looked for recurrent presentations of PMSCs and recorded the overall tone of framing (positive/negative). While the initial publications of Amnesty International and Human Rights First (both participants of MSIs) have formulated sharp criticism bearing on PMSCs’ track record of human rights violations, their impunity, a lack of transparency and democratic control, as well as their threat to the state monopoly over the use of force, the tone of framing has become more appreciative, addressing PMSCs also in a positive way—as being part of the solution. According to Amnesty, the industry “has a contribution to make” in ensuring respect for human rights (Amnesty International, 2009). In a similar vein, Human Rights First has lauded the industry’s desire and commitment to regulation (Human Rights First, 2011; Levine, 2010). Since 2012, both NGOs have ceased active campaigning altogether. This coincides with the development of the ICoC. The ICRC represents a somewhat special case because it has made virtually no use of public naming and shaming of “bad industry players,” but remained measured in its response to the industry. Even before it took the lead in the Swiss Initiative, the organization has reached out to the most important and prominent PMSCs based in the United Kingdom and United States (Voillat, 2012), which has put in a similar position to being in a MSI.

The (mounting) silence of “insiders” can be contrasted with the critical work of groups that have resisted the pull into MSIs, to include NGOs like War on Want and Global Policy Forum. They have kept up both a stable level of output and their call for an end of the privatization of security (Pingeot, 2012, 2014; War on Want, 2013). This lends further force to the assumption that the silencing of critique can be linked to workings of institutional power in MSIs. NGOs advocating outside the structures of the MSIs were “insulated” from the pressures that many of their peers have encountered once they got involved.

Conclusion

This article set out to analyze the power dynamics of multi-stakeholder security governance. Overall, its findings caution against the optimistic undertones that usually surround debates on multi-stakeholderism as the most desirable and effective form of governance. As far as the realm of private security is concerned, one of the most fundamental challenges with the Swiss Initiative and the ICoC is that they privilege and reinforce the views and interests of those actors who share the idea that PMSCs are here to stay and therefore should be permissively regulated rather than entirely banned. Western governments and the PMSC industry have used the Swiss Initiative and the ICoC to bypass and challenge alternative options and avenues for regulating PMSCs—namely the UN Draft Convention process which has traditionally held an uncompromising statist, hard-law only position and which represented the view of states that were more skeptical of PMSCs (institutional power). In contrast, participation in the Swiss Initiative and the ICoC has been a priori limited to groups that held a favorable view regarding PMSCs and MSIs. Their ideas about “good governance” of the PMSC industry have been “baked in” the two MSIs from the beginning. This is not only evident in the rules that are supposed to govern PMSCs but also in the discourses and practices that inhere in the Montreux Document and the ICoC and that have erased earlier, less positively connoted representations of PMSCs (productive power). The marginalization of critical voices as inferior types of expertise has made it possible to present hybrid public-private rules through multi-stakeholder deliberations as if they were the only game in town, even though a number of alternative options and perspectives have been under consideration (co-constitutive power).

If we move beyond this article’s focus on MSIs, there is a sheer variety of codes of conduct, benchmarks and best practices worthy of further exploration. On the productive side of power, this ever-expanding web of voluntary standards is likely to reinforce the empirical trend identified above—the entrenchment of PMSCs in private security governance—by generating new legitimate areas of activity and roles for the industry. From an institutional power perspective, they signify a shift away from state-centered and binding regulation through the UN or other intergovernmental bodies towards decentralized, voluntary, and self-regulatory frameworks that serve a neo-liberal, privatization-friendly agenda of north Western governments and like-minded actors. These power dynamics are by no means unique to the field of private security regulation. Consider the activities of the U.S. government in issue areas such as the governance of the Internet where the United States has consistently opposed centralizing regulation via the UN in favor of a multi-stakeholder approach (Carr, 2015). Finally, the emergence of new modes of governance is likely to further weaken the position of actors who fall outside the supposed consensus in favor of voluntary self-regulation, relegating them to the positions of non-experts (Leander, 2016a). As Barnett and Duvall (2005) put it, acknowledging the different plays of power, whether in multi-stakeholder (security) regulation or elsewhere, “makes it much more difficult to approach global governance purely in terms of cooperation, coordination, consensus, and normative progress” (p. 62). It is also a site where institutional biases, hierarchies, and hegemonic discourses play out.

Acknowledgements

I would like to thank Elke Krahmann, Claire Cutler, the editor, and the two anonymous reviewers for their helpful comments.

Disclosure statement

No potential conflict of interest has been reported by the author.

Additional information

Funding

This research was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Project-ID 5485778 – SFB 597.

Notes on contributors

Berenike Prem

Berenike Prem is a postdoctoral researcher at the University of Bremen, Germany. She holds a Ph.D. from the University of Bremen where she worked in research project on the privatization and internationalization of security. Her research concentrates on non-state actors’ involvement in security governance, with a focus on the activities of Private Military and Security Companies (PMSCs), and emerging military and security technologies. She is the author of Private Military and Security Companies as Legitimate Governors: From Barricades to Boardrooms (Routledge, 2020).

Notes

1 The first type mentioned by Barnett and Duvall, compulsory power, works through coercion or violence in the sense that powerful agents use or threaten to use specific resources in order to alter the free action of the “powerless.” Since this type of power works largely independent of MSIs, it will not be subject to further analysis in this article.

2 That said, a focus too strong on the United States and United Kingdom risks neglecting the agency of non-Western states, particularly those falling into the category of “host” or “territorial” states (van Meegdenburg, 2015). Their role will be reflected upon in discussing the UN process.

3 The Convention entered into force in 2001.

4 This is also true for the Swiss Initiative which can be described as essentially state-driven.