All hands on deck: levels of dependence between the EU and other international organizations in peacebuilding

ABSTRACT

The EU seeks extensive partnership with other international organizations when it comes to security challenges. This is puzzling as the EU relies for its resources mostly on its member states. The relations between the EU and other international organizations have thoroughly been studied, yet scholars rarely question the actual rationale for partnership. We start from resource dependency theory which explains that almost all organizations are dependent on the resources of their partners. Yet we extend this theory by distinguishing between macro, meso and micro-level dependencies. To illustrate resource dependencies between the EU and other international organizations, we analyse EU’s peacebuilding policies in Kosovo, Mali and Armenia. By accounting for macro- and micro-level dependencies we provide a more holistic perspective than conventional meso-level explanations. Our contribution is therefore to expand the scope of the resource dependency theory and provide a framework to analyse dependencies between the EU and other international organizations.

KEYWORDS:

• European Union
• international organizations
• partnership
• resource dependence
• peacebuilding

1. Introduction

Partnership with other international organizations (IOs) has long been a key objective of the European Union (EU) in addressing international security challenges. As the EU Global Strategy (European Union 2016, 18) states ‘[t]he EU will be a responsible global stakeholder, but responsibility must be shared and requires investing in our partnerships. Co-responsibility will be our guiding principle in advancing a rules-based global order’. The Joint Communication on the EU’s Comprehensive Approach (High Representative and European Commission 2013, 11) similarly notes ‘the EU needs to engage and work together with other international and regional actors. The role of the EU is linked – to a greater or lesser extent – to the action (or non-action), resources and expertise of others’.

How organizations interact with other organizations has long been a topic of inquiry in sociology. Resource dependency theory, in particular, notes that to fulfil their mandates, almost all organizations are dependent on the material and immaterial resources of their partners (e.g. Levine and White 1961; Benson 1975; Pfeffer and Salancik 1978). While some scholars have used resource dependency theory to explain the relations between IOs (Harsch 2015; Biermann and Harsch 2017), the extent to which the EU seeks partnership with other IOs remains puzzling. After all, the EU and other IOs are normally dependent for their resources on their members rather than on other organizations (Patz and Goetz 2019). In addition, if the EU gets resources from other IOs, member states will lose some of their control over the organization. We are thus left with the question why the EU focuses so much on partnerships when it comes to addressing security challenges.

Surprisingly, this question is not extensively addressed. The preference for partnerships is often assumed. Scholars focus on why the EU and other IOs are (not) successful in coordination (Tardy 2005; Morsut 2009; Dursun-Ozkanca 2010; Rein 2015; Gebhard and Smith 2015). They identify obstacles for interaction (Stewart 2008; Biermann 2008; Smith 2011; Gebhard and Smith 2015) and they discuss whether the mode of interaction can be conceptualised around cooperation, competition and conflict (Duke 2008; Brosig 2011, 2014; Van Willigen and Koops 2015). The main contribution of this article is therefore to reconsider resource dependency perspectives to provide a more nuanced explanation as to the reasons behind the EU’s preference for partnership.

This article argues that although resource dependency theory provides an important starting point, we need a broader conceptual model to better account for the explanatory factors behind the EU’s preference for partnerships. In particular, we argue that we need to distinguish between macro, meso and micro-level dependencies. At the macro-level, the EU operates independently from other IOs to achieve its formal mandate, but requires the success of other IOs to achieve broader objectives. At the meso-level, the EU and other IOs contribute directly to the achievement of each other’s mandates. Resources are exchanged in a regulated, institutionalised, manner according to predefined inter-organizational agreements. At the micro-level, the EU and other IOs contribute on an ad hoc basis allowing each of them to better implement their mandates. This conceptual model provides a more holistic view on dependency.

To illustrate how resource dependence works at these levels, this article illustrates the dependencies between the EU and other IOs in the area of peacebuilding. Peacebuilding refers to the stages in the conflict cycle after violent conflict (war) has ended (Ramsbotham, Woodhouse, and Miall 2011, Figure 5.1). It therefore covers much of what the EU does in the Common Foreign and Security Policy (CSDP), mediation and dialogue activities and also includes project funding, such as the Instrument contributing to Stability and Peace and part of the Instrument for Pre-Accession Assistance.¹ Many of the activities of the EU’s partners, including the United Nations (UN) and Organization for Security and Co-operation in Europe (OSCE), also fall under peacebuilding. While peacebuilding in post-conflict countries is a policy area in which many IOs normally interact (McEvoy 2017), macro, meso and micro-level dependencies are also present in other areas, such as development, humanitarian assistance, environment, human rights and non-proliferation.

This article compares macro, meso and micro-level resource dependencies in the cases of Kosovo, Mali and Armenia. In Kosovo, the EU is the lead international actor in peacebuilding and its commitment is multi-dimensional and long-term. In Mali, the EU is a key supporting actor to the UN. While it has deployed military and civilian operations, its commitment is more narrow than in Kosovo. In Armenia it is a secondary actor. While the EU engages in peacebuiling, it does not have a CSDP presence and Armenia itself is strongly dependent on Russian support. In these cases, the EU correspondingly uses a variety of peacebuilding instruments (CSDP, neighbourhood policy, pre-enlargement policy). Such variation provides us with insights to relevant dependencies on partners. These cases show that while macro-level dependencies can clearly be identified and are often explicitly acknowledged, the EU and other IOs find it particularly difficult to establish institutionalised resource exchange at the meso-level. At the micro-level, the cases reveal a plethora of informal one-off instances in which the EU and other IOs support one another. These case studies are informed by empirical data from 29 interviews² and relevant secondary sources.

This article starts with an overview of resource dependence theory and makes a case why we need to consider different levels of dependencies. It continues by providing empirical examples of such dependencies between the EU and other IOs in peacebuilding at the macro, meso and micro-levels. For each level, the article discusses the situation in Kosovo, Mali and Armenia.

2. Resource dependency between IOs: macro, meso and micro levels

The starting premise of resource dependence theory is that organisations are embedded in their environments and depend on external resources to operate and survive (e.g. Levine and White 1961; Pfeffer and Salancik 1978). Resources are scarce and few organisations produce and control all the resources they need themselves. There are two components to dependence: the essentiality of a resource for organizational operation and the substitutability of the resource with resources from alternative sources (Jacobs 1974, 51). Organizations want to avoid relying on other organizations for their essential non-substitutable resources. The core dilemma is thus that organisations are dependent on each other to operate, but that too much dependence puts their operations at risk.

The extent to which organizations develop their relations is affected by their mutual dependence and power imbalances (Casciaro and Piskorski 2005). For instance, when resources are non-specific and widely available on the market, organizations can engage in ad hoc exchanges. On the other hand, if resources are essential and specific, organizations can even opt for vertical integration (Williamson 1979). The institutionalisation of inter-organisational relations therefore varies and this variation can be usefully captured in four dimensions: the formalisation of the exchange of resources between organizations, the intensity of the exchange of resources, the reciprocity in terms of resources, and the standardization of the exchange of resources (Marrett 1971; Aldrich 1979; see also Dijkstra 2017).

The resource dependence perspective is therefore an established theory that helps explain why organizations interact and how their interactions are structured. This also makes it potentially important for studying relations between the EU and other IOs. Yet despite a wealth of empirical studies on interactions between IOs, resource dependence approaches remain relatively rare in international relations (e.g. Biermann 2008, 2015; Harsch 2015; Biermann and Harsch 2017; Koops 2017). This is puzzling, because if not for resource dependence theory, then what explains why IOs interact?³ In other words, why exactly do we empirically witness continuous attempts by the EU and other IOs to reach synergies and coordinate their activities, such as for example peacebuilding missions in post-conflict countries?

While all organizations try to avoid becoming overly dependent on others, this is particularly the case with IOs. Member states worry about the autonomy of IOs, as it may result in agency loss on the side of the member states or even institutional pathologies (Barnett and Finnemore 2004; Hawkins et al. 2006). They will try to prevent that IOs obtain resources through alternative channels thereby increasing their autonomy in relationship to the membership. Indeed, IOs normally get their resources from their membership with clear strings attached (Patz and Goetz 2019). In other words, IOs seeking outside resources for the fulfilment of their own mandates is generally considered a second-best option. For some immaterial resources, such as the legitimacy to engage in peacebuilding (Biermann 2017), the IOs may have to look elsewhere. But when it comes to material resources, the resource-rich member states of the EU should be able to provide most themselves.

We argue that the focus on observable and formalized exchange of resources between IOs provides a too narrow application of resource dependency theory. In other words, if we are to understand why the EU consistently seeks partnerships, we need instead to amend the conceptual model underpinning the theory. To be sure, Harsch (2015) has distinguished between ‘thin’ and ‘thick’ cooperation, whereas Gebhard and Smith (2015) have usefully pointed at the scope of informal cooperation. We propose, however, an all-encompassing approach. When thinking about resource dependencies between the EU and other IOs, it is useful to consider different levels of interaction: macro, meso and micro-level dependence.

At the macro-level, the EU and other IOs operate independently to achieve their (limited) mandates, but require each other to achieve overall (broader) objectives (see Table 1 for examples). The rationale behind macro-level interaction is that many international problems (e.g. climate change, development or peace and security) are too large for single IOs to address resulting in a need for collective action. Furthermore, different IOs may have different sorts of strengths, expertise and authority (cf. Barnett and Finnemore 2004). The result is a division of labour between IOs, sometimes explicit but regularly implicit. These IOs may not need each other on an everyday basis. After all, in fulfilling their (limited) mandates they can rely on the resources of the membership. Yet in terms of achieving overall objectives (e.g. solving climate change, economic growth or achieving a sustainable peace), resource dependence is unavoidable: the EU cannot succeed in its overall objectives without the parallel effort of other IOs. This creates a relationship of dependence.

Table 1. Examples of resource dependencies in the area of peace and security at the macro, meso and micro level.

Macro	The EU deploys a military stabilization force and the UN does parallel policing within refugee camps.
Meso	The EU makes use of NATO military assets through the Berlin+ agreement.
Micro	The EU provides extrabudgetary project funding for OSCE peacebuilding.

By identifying the macro-level, we significantly increase the scope of resource dependency theory. One can argue indeed that this goes beyond the theory in its classic form. At the same time, a macro-level perspective also speaks to established international relations concepts, such as role theory (Holsti 1970) or even hierarchies in international relations (Lake 2009). Among IOs there is a strong implicit sense about which organization can contribute what to the broader objectives of the international community. The macro-level helps us to explain why there is considerable coordination, even in the absence of formal resource exchange. IOs need to coordinate the broader objectives, position themselves and carve out a role for themselves.

The meso-level focuses instead on institutionalized dependencies between IOs contributing directly to the implementation of the (limited) mandate of other IOs. IOs need one another, because otherwise they cannot even achieve their own mandates provided to them by their own member states. For instance, NATO-led forces in Kosovo stand by when EU police can no longer handle a situation. The level includes many of the examples which we typically find in the literature on inter-organizational relations. This is also the level that resembles traditional resource dependency theory (e.g. Biermann and Harsch 2017; McEvoy 2017). Institutionalized dependence indicates a degree of formalization, standardization and continuity (Abbott et al. 2000; see also Dijkstra 2017).

Finally, we can identify micro-level dependencies. These are ad hoc, one-off, contributions allowing other IOs to better implement their mandates. For instance, one IO can provide airlift to another IO. Because resource dependence inevitably touches upon questions of control, it is not surprising that scholars have found more instances of micro-level than meso-level dependence (e.g. Harsch 2015; Biermann 2015; Gebhard and Smith 2015). Particularly the member states of IOs find it difficult for IOs to sign formal agreements with other IOs, if this negatively affects their control. Organizations tend to value their autonomy, and this is particularly so with IOs which include many different principals. Informal, one-off, support to other IOs can therefore be more appealing than formalized dependence at the meso-level (Dijkstra 2017; Vabulas and Snidal 2013).

The remainder of this article provides empirical examples of these dependencies between the EU and other IOs in peacebuilding. Peacebuilding in post-conflict countries is a policy area in which many IOs normally interact and are dependent on one another. In Kosovo the EU acts as lead international actor in a politically and geo-strategically important region, which includes EU candidate countries. Unsurprisingly, its presence in Kosovo is multi-dimensional and long-term, involving one of the largest EU-led civilian missions to date. In Mali the EU has deployed operations since 2013 and as such it is a key supporting actor alongside the UN in another long-term, strategically important for the EU region. In Armenia the EU is a secondary actor. This case is interesting as it regards a country strongly dependent on Russian support for its political and economic survival, where the EU has less leverage and yet an interest to stay engaged due to the strategic importance of the region, not at least as a result of the ongoing conflict with Azerbaijan. It is also an important case to probe into the EU’s ability to influence the peace process in a country, in which the EU has not deployed a CSDP mission. While the article focuses on these three countries, peacebuilding cannot be seen in isolation of related regional conflicts (respectively with Serbia, in the Sahel and with Azerbaijan). The three cases are unique in that they provide for strong examples of the three different levels of engagement of the EU with other IOs.

3. Macro-level dependencies

At the macro-level, different IOs often provide complementary peacebuilding support to post-conflict countries. This often results in a (informal) division of labour between IOs. These IOs may not need each other on an everyday basis to get their work done, yet in terms of achieving overall objectives, such as a sustainable peace, individual IOs cannot succeed without the effort of other IOs. This section shows that in all three cases – Kosovo, Mali and Armenia – there are identifiable macro-level dependencies. Some are explicitly acknowledged, for instance in UN Security Council resolution 1244 which set roles for the UN and NATO in Kosovo (see McEvoy 2017). Most often, however, macro-level dependencies remain implicit. Nevertheless divisions of labour exist and it is clear that collective effort of IOs is needed to bring about the objectives of the international community.

3.1. Kosovo and its statehood

Kosovo unilaterally declared independence from Serbia in 2008, after having been de facto independent since 1999 (on the role of IOs: Papadimitriou, Petrov, and Greicevci 2007; Papadimitriou and Petrov 2012; Eckhard 2016). Even though statehood remains contested, the implicit objective of the international community is to provide stability in a volatile region. Bringing Kosovo (and Serbia) into the orbit of European integration is an important means, which in turn requires a functioning economy, good neighbourly relations and professional state institutions.

The EU has been the primary international actor since 2008. The EU has, for instance, facilitated the high-level political dialogue between Belgrade and Pristina since 2011. Through the EU Special Representative, it has been the main political contact point for the local authorities. The European Commission has long had a key role in economic development. It has also led on all pre-enlargement measures since the signing of the Stabilisation and Association Agreement in 2015. The EU mission in Kosovo (EULEX) has been monitoring, mentoring and advising the Kosovar rule of law institutions since 2008 (Dijkstra 2011; Papadimitriou and Petrov 2012).

The EU nonetheless remains dependent on other international actors. The United States ultimately guarantees Kosovar independence. NATO’s Kosovo Force (KFOR) has had troops on the ground since 1999. In addition to providing stability, the NATO Advisory Liaison Team, situated in the Ministry for Kosovo Security Forces headquarters, has an advisory role in the Kosovo Security Forces (KSF) capacity building (Ministry for the Kosovo Security Forces 2016).

The UN Mission in Kosovo (UNMIK), set up in 1999, continues to play a role. Even though it was significantly downsized after 2008 (Eckhard and Dijkstra 2017) and now only has several residual functions, UNMIK initially provided the umbrella for other IOs to work alongside each other (McEvoy 2017). This included the United Nations High Commissioner for Refugees (UNHCR), OSCE, and also the EU. Apart from UNMIK, the UN system comprises several agencies, funds, and programmes, notably the UN Development Programme (UNDP), which provides support for democratic governance and peacebuilding.

Despite the fiercely contested statehood, and the fraught relations between IOs (notably the EU and UN), at a macro-level we can therefore identify a division of labour and dependencies. Part of these relationships are formal and defined in UN Security Council resolution 1244 (McEvoy 2017). Another part has developed informally after 2008. While all IOs implement their own mandates, they are ultimately dependent on each other to reach the overall objective of providing stability in a volatile region. While the EU takes the lead in peacebuilding, it is still dependent on a variety of partners to succeed in Kosovo.

3.2. Mali and the Sahel

After a military coup in which the Malian President Touré was deposed in 2012 (on the conflict: Down and Raleigh 2013; Bøås and Torheim 2013), an insurgent alliance of Tuareg nationalists and Islamist groups exploited the disruption of government and seized several provincial capitals. Following the immediate condemnation by the international community, the coup-makers and President Touré agreed to an interim government. The Economic Community of West African States (ECOWAS) then deployed troops in support of Malian territorial integrity (Oluwadare 2014). This led to a strong international involvement with the aims of preserving Malian territorial integrity, disrupting Islamist terrorism and rebuilding a functional state authority.

While ECOWAS, supported by France, was a key actor in the immediate aftermath of the coup, authority was quickly transferred to the UN Multidimensional Integrated Stabilization Mission in Mali (MINUSMA) in April 2013 (UNSC Resolution 2100). MINUSMA acts as an integrated UN presence and provides a coordinating platform for the international community (ibid.). MINUSMA now has a wide mandate ranging from providing security, to reconciliation, and rebuilding of the Malian security sector (UNSC Resolution 2164).

France has been a key actor, carrying out various unilateral military counterterrorism operations (Operation Serval and Operation Barkhane). The EU launched its EU Training Mission (EUTM) Mali in February 2013 and the EUCAP Sahel Mali in March 2014 (European Union 2013, 2014). EUTM provides advice and training to Malian armed forces. It complements MINUSMA, which is not directly involved with capacity building. EUCAP Sahel Mali provides advice and training to the different components of Malian internal security forces. The division of labour is, however, less clear, as the MINUSMA Police Component includes a capacity-building pillar as well. In practical terms, EUCAP has focused mostly on the strategic and national level training, whereas MINUSMA mainly engages in capacity building at the tactical level.

The international community is actively involved in Mali. While the different IOs work on their own mandate, the ultimate exit strategy is for the Malian government and security services to be able to exert their own authority. Even though their interaction is obvious, the dependencies between MINUSMA and the EU remain implicit. While the EU Council Decisions refer to the other IOs and stress the need for coordination, they do not provide a division of labour. This is a clear example of a macro-level dependence which has not been formalized.

3.3. Armenia and the Nagorno-Karabakh conflict

Armenia has been in conflict over Nagorno-Karabakh, which the international community recognises as a region of Azerbaijan, for nearly 30 years and both countries have been in prolonged peace talks in the framework of the OSCE Minsk Group (on the conflict: Melander 2001; De Waal 2010; Cornell 2017). This remains the principal platform for the peace negotiations. While the OSCE is a logical institution, it does not have the weight of larger IOs, such as the EU and UN. The role of the OSCE in monitoring the conflict is limited and not permanent. Six monitors routinely deploy to Nagorno-Karabakh and write up a report (De Waal 2010, 160, 166). Relations between both countries have recently considerably deteriorated as a result of the four-day April 2016 war (International Crisis Group 2017).

The EU’s role in addressing the conflict is limited. The EUSR for the South Caucasus supports the Minsk Group (Paul and Sammut 2016, 3), but has no independent role. Most significantly, the EU pays for the European Partnership for the Peaceful Settlement of the Conflict over Nagorno-Karabakh (EPNK), which focuses on civil society thereby complementing the work of the OSCE. It is also widely recognized that if Armenia and Azerbaijan were to reach an agreement, the implementation would probably require donor money and a peacekeeping force. While there are no explicit plans, it is understood that the EU and UN would step in. This gives credibility to the OSCE process.

In addition to OSCE-led peace negotiations over Nagorno-Karabakh, the EU and other IOs are also active in domestic Armenia security matters. While the conflict with Azerbaijan casts a shadow over many domestic security sector reforms in Armenia, this is essentially a separate matter of analysis. Russia remains the most important international actor in Armenia. Through the Collective Security Treaty Organization and its military base, it is the ultimate guarantor of the Armenian state. With few friends in the neighbourhood, Armenia does not have alternatives. The Russian dependence also makes the relations with the EU complicated, resulting in Armenia walking away from key trade deals.

While the geopolitical situation overshadows domestic security sector reform, it is still worth to point at the bilateral efforts of the various IOs. Separate from the Minsk Group, the OSCE office in Yerevan provides bilateral support with an emphasis on democratisation and politico-military support. OSCE SSR activities tend to be rather limited and focus on workshops, study trips, and most prominently the introduction of community-based policing. The OSCE has also focused on the Armenian police training centre, the police college and the police academy. Much of the EU effort involves funding the Council of Europe through a range of projects related to human rights, rule of law, and democratic governance. This complements the work of the OSCE but remains separate (Council of Europe 2015). The role of the UN family is largely limited to development assistance.

4. Meso-level dependencies

At the meso-level, different IOs institutionalize how they exchange resources and contribute to the implementation of each other’s mandate. While dependencies at the macro-level often force IOs to develop institutionalized mechanisms to coordinate their actions at the meso-level, this section shows that on the whole the degree of formal institutionalization remains limited. The most obvious dependencies occur when IOs do not have a field presence and rely on other IOs for information as well as security and other enablers. This has resulted in some form of institutionalized resource exchange.

4.1. Kosovo and its statehood

UNSCR 1244 formally serves as a basis for the division of labour among IOs in Kosovo. In addition, some memorandums of understanding (MoUs) have been signed, even though they do not appear as a regular tool for arranging inter-organizational relations. The two most important meso-level dependencies in Kosovo have to do with day-to-day security provided by NATO and the information-sharing between IOs both through formal reports and coordination meetings.

One key meso-level dependency is the provision of security by NATO. There is a ‘protocol of understanding’ between the Kosovo Police, the EULEX formed police units, and NATO about how to address crowd control through a three-tiered response. According to this protocol, the Kosovo Police acts as the first responders. If they are not capable of addressing the situation EULEX formed police units step in. NATO’s KFOR gets involved when riots turn into paramilitary violence. While it is a clear example of a macro-level type of dependency, this protocol of understanding has not been officially formalized due to the EU-NATO conundrum (Cyprus-Turkey). In addition to this protocol, KFOR also provides an ‘evacuation plan’ for UNMIK and EULEX employees ‘in case something happens’ (interview #6).

To fulfil their mandates, IOs in Kosovo are also dependent on each other’s information. An obvious example is reporting on the situation in north Kosovo, where the EU’s field presence is weaker. The EU Office therefore makes use of the OSCE and UNMIK presences in the north for monitoring and reporting, in particular on the position of minorities. The information from the OSCE and UNMIK monitoring reports is integrated into the reports that the EU Office sends to Brussels. There is a similar relationship between the OSCE and UN, as the OSCE also contributes to the reports of the UN Secretary-General on UNMIK.

Beyond formal reports, information-sharing takes place during coordination meetings. The best known are UNMIK interagency Monday morning meetings, which originally brought together all the IOs under the UNMIK umbrella. While these formal coordination meetings are the most institutionalized form of coordination found in all three cases studies, they also show the limits of the meso-level institutionalization. From the EU’s side, they are seen as pre-2008 leftovers and the EU is rarely represented at the level of head of mission (interviews #6, #9 and #10). In Kosovo, there are also a plethora of coordination meetings dealing with more practical issues, such as situational relations and staffing (interviews #9 and #10). The EU, EULEX, the International Criminal Investigative Training Assistance Program (ICITAP) and UNMIK have monthly coordination meetings at which they share information on their work with Kosovo Police. In the rule of law area, UNMIK, EU Office, EULEX, OSCE and the United States have their own working group.

4.2. Mali and the Sahel

Following the coup in 2012, a Joint Task Force for Mali was established by the AU, ECOWAS, the EU and the UN. This task force and the subsequent Support and Follow-up Group have tried to coordinate a coherent international initial response to the conflict (UNSC Resolution 2100, 3). The overarching framework for international support became Resolution 2085 and currently Resolution 2164, under which MINUSMA has taken on responsibility for coordinating international support with respect to the stabilization of Mali. This is mirrored in the mandates for EUTM and EUCAP Mali, where both missions are tasked to coordinate and harmonze their actions with MINUSMA and other international actors (European Union 2013, article 7(4); European Union 2014, article 2(2)).

Under this framework, extensive formal coordination mechanisms have been established between MINUSMA, the EU Delegation, the CSDP missions, and other international actors. For instance, the MINUSMA Force Commander and the Commander EUTM meet at bi-weekly intervals for coordination (interviews #12, #14 and #19). At lower levels, an exchange of liaison officers has been established between MINUSMA Military HQ and EUTM (interviews #14 and #19). And, to coordinate intelligence assessments, a direct coordination and exchange of information has been established between EUTM and the MINUSMA All Source Intelligence Fusion Unit (ASIFU), which is tasked with collecting and providing military intelligence products to partners, including embassies. With respect to police, there is a quarterly strategic meeting between EUCAP Head of Mission and MINUSMA Police Commissioner and a monthly technical meeting between EUCAP Head of Operations and MINUSMA Deputy Police Commissioner. In addition to formal and informal meeting structures, liaison officers have been appointed to improve communication between the UN and EU missions (interviews #15, #17 and #19).

There is a critical structural dependence between MINUSMA and EUTM for which both IOs have tried to find an institutionalized solution. Both IOs have an interest in the training of Malian armed forces, by EUTM, to be conducted closer to the actual area of operations (training on the job). Since EUTM does not have a physical presence beyond the capital city, it is reliant on MINUSMA to provide logistical support and local facilities for EUTM to extend its presence into the north-eastern areas of Mali (Gao and Kidal) (interviews #12, #13, #14 and #19). Yet while the EU has long had this ambition, the implementation has suffered delays, primarily due to the difficulty of coordinating the extensive support required from MINUSMA Mission Support Division (interviews #14 and #19). It has not been straightforward to put the necessary formalized exchange of capabilities in place.

Dependency also goes into the other direction. As MINUSMA is conducting missions with Malian forces, the UN has a significant dependency on EU missions in terms of training quality. The EU missions have been responsive to suggested training requirements being put forward by MINUSMA, yet at the same time significant concerns have been voiced towards the lack of on-ground training and follow-up from both CSDP missions. Part of the problem is the inability of the EU to move around the country, but there is also a strong emphasis on the risk averseness by EUTM. This undermines effective inter-organizational relations.

4.3. Armenia and the Nagorno-Karabakh conflict

While interviewees uniformly commented on the good collaboration of the international community in Armenia, there are hardly any formal mechanisms for liaison and there is only a limited interaction between IOs. The division of labour is largely tacit and based on both path dependence (i.e. who came first) and specialized expertise. Perhaps also because the international engagement is relatively small, actors do not get into each other’s way and provide largely a complementary effort. There are, however, few meso-level dependencies.

The most institutionalized forms of dependence are on both extremes of the security-development nexus. It is uncontested that the OSCE Minsk Group leads the peace negotiations over Nagorno-Karabakh. This creates a focal point for other international actors. For instance, any time the OSCE Minsk Group meets, the EUSR flies in to have consultations with all actors in the margins of the meeting. The EUSR supports, in this respect, the work of the OSCE from a diplomatic standpoint. Lending the weight and authority of the EU to the OSCE process – on a continuous basis – provides an important dependency. At the other side of the spectrum, the development agencies have clearly worked out profiles. Local donor coordination follows standard UNDP template, which is present in most development countries across the globe, and also involves the local authorities. There is no formal coordination between the OSCE, EU and Council of Europe in the area of security sector reform.

5. Micro-level dependencies

Finally, we can identify micro-level dependencies. These are ad hoc, one-off, contributions allowing other IOs to better implement their mandates. There are many different ways in which dependencies of IOs occur at the micro-level. Micro-level dependencies also often substitute meso-level interactions, when IOs cannot come to institutionalized agreement. Yet significant challenges remain with micro-level dependencies, particularly the ad hoc nature of engagement.

5.1. Kosovo and its statehood

Project funding is an example of how IOs contribute to the mandates of partners. International actors often co-fund projects or even take over follow-up projects started by other international actors when they run out of money. In Kosovo, the EU often relies on project implementing partners, such as the UN agencies and the Council of Europe. There are no clear guidelines for selecting an implementation partner, but it depends on the field of work and capacities of an international actor. The position of IOs towards the status of Kosovo also plays an important role. For instance, United Nations Office for Project Services (UNOPS) has been selected by the EU to construct integrated border crossings between Serbia and Kosovo (UNOPS n.d.). Not only for its previous experience, but also because a UN agency is more acceptable for Serbia.

Despite formal channels, the bulk of communication and information exchange among international actors in Kosovo runs informally on an ad hoc basis. Crucial for UNMIK’s relation with the EU was the fact that the latter employed many former UNMIK staff. On the other hand, interpersonal tensions are said to have made cooperation between the two missions uneasy at the beginning of the EULEX mandate. Physical proximity is regarded as important for efficient coordination. The size of Pristina is deemed to be a factor facilitating communication: ‘This is a small city. They [internationals] meet in the same restaurants; everyone knows what everyone else is doing’.

On an ad hoc basis, IOs also lend diplomatic support for one another. A specific practice identified could be described as ‘logo switching’. IOs tend to shift visible support for certain programmes, depending on whom they are primarily seeking to address. The UN flag is auspicious for the Serbian side, but on the other hand, UNMIK does not enjoy popularity among central authorities in Pristina. There have been instances of UNMIK and OSCE co-organising an event, when UNMIK would provide more funding, but only the OSCE logo would be visible, in order to attract Kosovar institutions. Similarly, the EU may use its logo to lend support to the other IOs, even if it is not funding their activities. For instance, the EU provided the Media Justice Transparency Initiative of the OSCE with such support, while its contribution was limited to the simple participation of EULEX judges in panel discussions.

5.2. Mali and the Sahel

As in Kosovo, a wide range of ad hoc and informal exchanges take place between staff officers in the different missions. For instance, to avoid a potential inflation of validity assessments on duplicate intelligence reports, military analysts informally coordinate between mission headquarters. There is no formal structure and it is managed on an ad hoc basis. Similarly, interviewees (#15 and #17) indicated that on a by-project and location basis ad hoc coordination is conducted between the UN police officers and EUCAP experts, where close-coordinated cases meetings takes place several times a week.

Important in terms of micro-level capability dependence is the provision of equipment for the Malian armed forces. The EU can provide the best possible training, but it does not have the mandate to provide local forces with equipment. This creates a dependency on partners. The unilateral ad hoc contributions of individual member states and non-member states, such as the United States and Switzerland, are particularly significant. This requires informal outreach on the side of the EU to other partners (interview #19).

5.3. Armenia and the Nagorno-Karabakh conflict

In the case of Armenia, there are also clearly identifiable forms of micro-level dependence. One example is the public support that the EU and the UN provide for the OSCE Minsk Group. This gives the OSCE additional authority as the only feasible forum for negotiations. For instance, High Representative Federica Mogherini attends all OSCE Ministerial Council meetings. Similarly, during the April War, she noted that ‘The European Union fully supports the efforts of the OSCE Minsk Group and the three Co-Chairs’ (Mogherini 2016). Similarly, UN Under-Secretary-General for Political Affairs, Jeffrey Feltman (2016) stated that ‘There is no alternative to a political process as proposed by the Minsk Group Co-Chairs and to restore trust between the sides’. Such public support is significant, also because it shows the two parties that there is no alternative.

In the area of SSR, the EU is providing significant project funding to the Council of Europe in the area of rule of law and justice. The Council of Europe has become the implementing partner of choice for the EU in these areas, with the EU making 90% of the funding available for co-financed projects (Council of Europe and European Union n.d.). While the EU is also the largest funder of OSCE extra-budgetary projects, the EU does not significantly support the OSCE office in Yerevan in terms of SSR. Indeed, this also shows that drawback of temporary ad hoc project support. The OSCE staff, in particular, feel they are underfunded and recall with some nostalgia the times that the EU provided significant funding for the OSCE projects.

6. Conclusion

It is well known that the EU coordinates a lot with other IOs, including on peacebuilding. In the academic literature, much attention is being paid to the reasons why the EU and its partners are sometimes (not) successful in coordination and other forms of interaction. The discussions about the EU and its partners are furthermore often conceptualized around cooperation, competition and conflict.

What much of the literature fails to address are the reasons why the EU consistently interacts with other IOs in peacebuilding. Indeed, resource dependency theory, the most convincing account in organization theory explaining why organizations interact seems difficult to apply to IOs. Whereas this theory stresses the resource dependence of organizations on their environment, IOs tend to rely mostly on their membership. The main contribution of this article is therefore to expand the scope of the resource dependency perspectives and make them applicable to interactions between the EU and other IOs and thus provide a more nuanced explanation as to the reasons behind the EU’s cooperative preference.

This article has argued that one needs to account for macro-level and micro-level dependencies in addition to the more conventional meso-level dependencies. While the EU in Kosovo, for instance, might not be formally dependent on other IOs – apart from NATO’s security provisions and UNMIK’s information about northern Kosovo – for the implementation of its EULEX mission, the EU can only succeed in bringing stability and progress in Kosovo, if the other IOs are also effective in fulfilling their own mandates. Despite the EU’s extensive commitment, the full challenge of Kosovo is too large for the EU to address, which has resulted in a division of labour across IOs.

Similar to such macro-level dependencies, we should neither underestimate the importance of micro-level dependence. While ad hoc, one-off, contributions by partners might not be essential for IOs to fulfil their core mandate, such contributions are often very welcome. It has been long stressed that particularly on the ground, there tends to be a large degree of practicality to interactions between IOs based on informal and personal relations. Even if not institutionalized through formal and permanent agreements, ad hoc contributions are often the bit of oil that keep the machinery running.

While this article has given illustrations through insights from Kosovo, Mali and Armenia, the article also triggers new questions. First, while it is useful to distinguish between these levels, a key question is how the levels relate to each other. UNMIK, as originally established, sought to create a formalized division of labour amongst IOs through its pillar structure. This is where the macro and meso levels intersect. Similarly, there are different degrees of formality that may distinguish between the meso and micro levels. Second, we have not tried to explain variation across cases within the levels. For instance, macro-level interaction is more problematic in Kosovo than Mali. At the same time, the pre-accession process in Kosovo provides a macro-level focal point for the international community. The situation is less clear in Armenia where the EU is a secondary actor. The new framework we have put forward should facilitate further research into the dependencies of IOs across the different levels.

Acknowledgments

The empirical section of this article builds on a previous policy report (Dijkstra et al. 2017). The article was previously presented at 11th Pan-European Conference on International Relations of the European International Studies Association (EISA) in Barcelona, 13–16 September 2017 and the Annual Convention of the International Studies Association (ISA) in San Francisco, 4–7 April 2018.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Research for this article has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no.: 653227. The content reflects only the author’ views, and the European Commission is not responsible for any use that may be made of the information it contains.

Notes

1. The EU itself confusingly uses the term crisis management as an equivalent for the CSDP missions, whereas most CSDP missions are about post-conflict capacity-building and security sector reform.

2. These interviews took place in Kosovo (11), Mali (9) and Armenia (9) from November 2016 until March 2017. Interviewees included senior officials from the EU, UN, OSCE, NATO, European embassies, local government and think tanks. These interviews were semi-structured and the average length was 60 minutes.

3. Constructivist and psychological theories provide alternative accounts, such as the idea of partnership as an end in itself (Biermann and Koops 2017, 12, 18). We are, however, concerned with resource dependency theory.