7 - Crisis and Change at IOM

Summary

This chapter investigates how the International Organization for Migration (IOM) dramatically expanded its involvement in humanitarian emergencies over the past three decades. Building on insights from historical institutionalism in international relations, we hypothesize that crises which touch upon matters of migration may constitute opportunities for IOM to expand the range of its activities as contingencies call for flexible responses that the organization is (the only one) apt to deliver. The 1990-91 Gulf War served as a ‘critical juncture’ in this regard, where IOM started to expand more forcefully into the broad realm of humanitarian assistance. It set a precedent that served as best-practice example and led to an ex post formalization of the institutional expansion through corresponding frameworks for action. As we show in case studies of the 2011 Libyan civil war and the 2014-16 Ebola crisis, this pattern holds across a variety of crisis contexts: humanitarian emergencies expose gaps in the governance architecture that IOM is quick to fill, thereby increasing the range of its activities which is later normalized in institutional rules and practice. Today’s vast array of humanitarian and other crisis-related tasks fulfilled by IOM attest to the lasting ‘power of precedent’.

7.1 Introduction¹

The Constitution of the International Organization for Migration (IOM) lists the purposes and functions of the Organization in Article 1. In particular, they encompass the organized transfer of migrants, refugees, and displaced persons in agreement with the states concerned as well as the provision of broader ‘migration services’ ranging from language training to advisory functions. IOM and its predecessor organizations have often deployed these functions in the context of a humanitarian crisis, particularly in post-conflict settings – starting with refugees in post-War Europe. For a long period, however, the organization’s activities were restricted to migration management, that is, the logistical support for migration at the request of member states,² leading to its depiction as nothing but a better travel agency.³ By contrast, IOM’s functions today comprise frontline emergency relief and a staggering variety of humanitarian activities that often only remotely link to migration issues.⁴ In fact, its institutional development in the post-Cold War era seems to be one of the most intriguing features in the history of IOM. It is characterized by rapid organizational growth and task expansion, a shift in the allocation of resources from migration facilitation to the provision of humanitarian aid in emergencies, and an engagement with an ever-wider range of policy fields, now encompassing issues as diverse as climate change and border control.⁵ How can these dramatic developments be explained?

Literature on IOM has identified important facilitating conditions for its task expansion: First, its ‘non-normative mandate’⁶ and functional organization type certainly represent a driver. Unlike many other international organizations (IOs) (incl. esp. UNHCR), IOM is not tasked to oversee and help with the implementation of international legal rules in its field. This makes it more flexible to go for new and rather unrelated tasks.⁷ Second, its projectized funding structure plays a role. Since IOM only has a very small core budget and receives funding almost exclusively for concrete projects, it has a financial incentive to broaden the scope of its activities – and convince member states and other donors of the necessity to operate in new fields.⁸ These important insights notwithstanding, we are still lacking a clear understanding of how IOM took hold in a growing number of areas and what institutional mechanisms underpinned this development.

In this contribution, we develop a historical institutionalist argument that combines the concepts of critical juncture and path dependency with agency-driven accounts of institutional change in IOs.⁹ Historical institutionalism assumes that institutional trajectories are path-dependent, that is, their development is conditioned by original decisions that introduce either self-reinforcing or self-undermining reactive sequences.¹⁰ While a strong focus thus lies on the relative stability of institutions and their gradual change, historical institutionalism also theorizes the original moments that create path dependencies in the first place. These critical junctures are conceived as situations in which the structural constraints on political action are significantly reduced and ‘the range of plausible choices open to powerful political actors expands substantially’.¹¹ Which political actors at the IO level can be expected to benefit from such conditions is subject to theoretical controversy in international relations. On the one hand, as expected by much rationalist theorizing on IOs, the most powerful member states might seize the opportunity to shift the institution in their desired direction.¹² On the other hand, as anticipated particularly by constructivists, it might also be the bureaucratic IO organs that attain institutional change through organizational entrepreneurship.¹³ Drawing on recent accounts of crisis-induced authority expansions by IOs, we assume that both may be possible, but hold that the strongest institutional ruptures can be expected where organizational entrepreneurship is met with tacit or explicit support by the most powerful member state(s).¹⁴

We submit that the metamorphosis of IOM in the past 30 years can be understood as a path-dependent development rooted in a critical juncture at the beginning of the 1990s. At the level of the international system, this period was marked by the end of the Cold War that infused international politics with a large degree of fluidity in general. At a situational level, the 1991 Gulf War represented a contingent window of opportunity for IOM to change its role from post-conflict migration manager to active humanitarian emergency responder. The shift was premised on the coincidence of the organization’s willingness to assume responsibility in this area and the United States’ active enlistment of IOM to fulfil crisis management tasks on the ground. This decision proved momentous as it set the organization on a path that has shaped its development to the present day. Not only did the Gulf intervention leave a lasting imprint on IOM’s institutional structure, but it also provided a blueprint for institutional expansion that would be reactivated time and again over the next decades: Humanitarian crises expose governance gaps that IOM is ready to fill on an ad hoc basis which member states accept ex-post or even invite ex-ante. This repeated match of demand and supply creates social precedents for IOM that widen its practical and operational experience and hence increase the range of tasks that ‘naturally’ fall within its remit over time.

The analytical narrative we provide in this chapter on IOM’s institutional evolution since the entry into force of its Constitution contributes to a better understanding of the organization’s changing character, transitioning from a foremost migration manager to a provider of humanitarian assistance in active crises. By focusing on the institutional mechanisms underlying this process, we shed light on IOM’s internal dynamics that so far have remained ‘almost completely unexamined’.¹⁵ The remainder of this chapter is structured as follows: First, we provide the theoretical background to our argument by providing theoretical building blocks from historical institutionalism and developing expectations about IOM’s institutional development in times of crisis. In the main part of the chapter, we first analyse the critical juncture at which IOM’s institutional path initially deviated (the 1990–1991 Gulf War) and show how it set in motion mechanisms of reproduction which reinforced the expansionary logic of IOM’s crisis interventions. Second, we illustrate how this logic of mandate extension through precedent setting has taken hold in the organization in two important crisis interventions by IOM in the more recent past: the Libyan civil war (2011), and the 2014–2016 Ebola crisis. In the concluding section, we discuss our findings with a view to their implications for the organization’s ethos, obligation and accountability.

7.2 Historical Institutionalism and International Organizations

In this section, we first introduce concepts from historical institutionalism, especially critical junctures and path-dependent processes of self-reinforcement, that provide analytical tools to understand long-term institutional developments. Second, we build on theories of international organizations to derive concrete expectations about the actors and conditions driving change at IOM.

7.2.1 Critical Junctures and Path Dependence

Historical institutionalism is rooted in comparative politics. More recently, its use has been extended to international institutions and IR more generally.¹⁶ The core insight of historical institutionalist thought is that institutional outcomes at a given point in time are regularly not the product of exogenous factors and independent actor choices at that moment, but follow from path-dependent processes of reproduction and change endogenous to the institution itself. While not oblivious to mechanisms of gradual transformation,¹⁷ historical institutionalists usually take a ‘punctuated equilibrium’ view on institutional change. That is, long periods of relative stability are only interrupted by rare moments of contingency in which new institutional paths are chosen. These moments are called critical junctures. Here, actor decisions evoke reactive sequences which set in motion self-reinforcing (or self-undermining) mechanisms of path-dependent institutional development.¹⁸

Historical institutionalist explanations thus gravitate towards the concepts of critical junctures and path dependence. Generically, critical junctures can be defined as ‘relatively short periods of time during which there is a substantially heightened probability that agents’ choices will affect the outcome of interest’.¹⁹ The distinctive feature of such historical junctures in which actor choices matter more than usual ‘is the loosening of the constraints of structure to allow for agency or contingency to shape divergence from the past’.²⁰ Often, critical junctures are equated with crises or turning points. They are not necessarily instantaneous events, but can represent ‘short phases that may actually last for a number of years’.²¹ The main challenge in the analysis of critical junctures is to identify cases in history and to explain why these moments in time are characterized by weaker constraints on agency than others.

According to Soifer (2012), critical juncture accounts need to identify and distinguish permissive and productive conditions. ‘Permissive conditions can be defined as those factors or conditions that change the underlying context to increase the causal power of agency or contingency and thus the prospects for divergence’.²² The focus thus lies on structural shifts, unintended consequences, exogenous shocks, etc., that interrupt the previously established processes of institutional reproduction. Productive conditions, on the other hand, are those factors that – in the possibility space created by the permissive conditions – cause divergent institutional outcomes that then represent the starting point for new institutional equilibria.²³ Often, productive conditions will combine with so-called ‘critical antecedents’, that is, factors preceding the historical juncture that unfold different causal effects under the changed conditions.²⁴ For instance, if an institutional equilibrium is unsettled by permissive conditions, agents that are at the right place at the right time (the productive condition) may effectuate change by redeploying long-established institutional capacities (the critical antecedent) for new purposes.

Once a critical juncture ends, historical institutionalists expect the deviant outcome to trigger mechanisms of reproduction that create new path dependencies. Most often, these are mechanism of institutional self-reinforcement. Here, positive feedback effects change actors’ attitudes in favour of an existing institutional practice. As Rixen and Viola explain:

The process is reinforcing because it is subject to increasing returns, that is, a situation in which the returns to engaging in a certain behavior or from adopting a certain rule increase over time and make the adoption of alternatives less attractive. The process is self-reinforcing, because it is reinforced through variables endogenous to the institution.²⁵

From a utilitarian perspective, institutional reproduction is the result of a cost-benefit imbalance of transformation. Given the investments sunk into setting up the institution as well as the learning and coordination effects produced by the institution once in place, opportunity costs for drastically altering the existing or creating an alternative institution are high and increasing over time.²⁶ Moreover, as highlighted by Zürn with a specific view to IOs, there are also increasing returns through cognitive effects. Both institutional actors and IO members engage in increasingly close interaction, producing convergent understandings (learning) and adaptive expectations, that is their belief in the success of the institution leads to adaptive behaviour which reinforces the institution’s ability to develop in the desired direction.²⁷

7.2.2 Assumptions about International Organizations and IOM

From the perspective of historical institutionalism, then, a long-term institutional development such as IOM’s rapid expansion in the area of humanitarian emergencies is likely to be rooted in a contingent starting point, a critical juncture, that sets in motion a process of institutional reproduction. As a general model of institutional change, however, it naturally lacks action and actor-theoretic specifications that would allow deducing concrete expectations for either the outcome of critical junctures or the drive behind its reproduction.²⁸ In the specific context of IOs, the question is who are the ‘powerful political actors’ for whom the range of available options increases during a critical juncture, who benefits and consequently whose repeated interactions increase the returns of institutional practice over time.

Most theories about IOs differentiate between IOs’ member states on the one hand and IOs’ supranational bodies such as secretariats and judicial entities on the other, and hold specific views on their respective role and influence on the design and direction of IOs. At one end of the spectrum are rational institutionalists who contend that all power lies with member states: IO bureaucracies are conceived as agents fulfilling tasks on behalf of their principals without much independent power of their own.²⁹ While often understood as one collective principal, a distributive variant of the theory highlights that member states differ in their capacity to wield control over policy which is why institutional choices will typically reflect the interests of the most powerful among them.³⁰ Moments of crisis and contingency, then, should represent opportunities for powerful states to steer IOs in their preferred direction.³¹ At the other end of the spectrum are sociological institutionalists who emphasize the ability of IOs to wield independent power: IOs are conceived as partially autonomous bureaucracies influencing member state behaviour through their delegated, moral, and epistemic authority.³² Importantly, this literature argues that ‘IOs tend to define both problems and solutions in ways that favour or even require expanded action for IOs’.³³ Mission creep is a distinct possibility. Seen from this perspective, crises could represent an opportunity for entrepreneurial IO staff to push their organization in an expansionary direction.³⁴

We adopt a middle-ground position between these two poles.³⁵ There is no compelling theoretical reason to treat the influence of powerful states and that of entrepreneurial IO staff as mutually exclusive or either as individually exhaustive in accounting for all patterns of institutional choice and change at IOs. It is much more plausible to entertain the possibility that both play a role to varying degrees depending on empirical conditions. At IOM, these conditions generally seem to favour a strong role for powerful states.³⁶ Compared to the specialized agencies of the UN, for instance, IOM has very small headquarters (both in terms of staff and funding), it lacks an appreciable amount of delegated authority, and its formal role in policy coordination among member states is minuscule. Moreover, states’ power to choose the projects they want to fund puts them in a prime position to control the organization. On the other hand, the general shortage of funds also fosters organizational entrepreneurialism,³⁷ and the lack of clearly mandated tasks opens the way for IOM to venture into various areas.³⁸ Additionally, IOM has a high number of relatively autonomous country offices with skilled and experienced staff whose expertise can be decisive for the decision to launch a new project.³⁹ Hence, even though it ultimately always depends on member state approval, the organization has both motive and opportunity to push its institutional path towards expansion.

In sum, our theoretical conjecture thus holds that IOM’s task expansion will be marked by both push factors on the part of the organization and pull factors on the part of powerful member states. We suppose that a critical juncture proves especially momentous if it creates conditions under which these factors align and show the actors that they may both profit from the expansionary path taken. For the case at hand, we refer in particular to IOM’s ability and eagerness to provide operative crisis management capacities in new areas that are largely ungoverned by any other actor, and powerful member states’ desire to leverage this capacity in situations that they care about. Any such situation, we argue, creates a social precedent through which IOM gains experience, knowledge and reputation as a flexible crisis manager. After the fact, we expect the organization to entrepreneurially foster an institutionalization of the precedent by creating corresponding programs, divisions, or operational frameworks which normalize the new-found tasks. Such formal and informal institutional devices can be used to signal to member states that IOM is ready to take on similar jobs in the future and that a wider than previously considered range of situations falls within its remit. The process is thus foremost a cognitive one by which mutual expectations among state and organizational actors converge and create increasing returns from expansion.

7.3 The Critical Juncture: IOM in the Gulf War

Operations in the context of crises have always been part of IOM’s activities, especially when large numbers of refugees were involved. In fact, the organization portrays its own history as one tracking man-made and natural disasters in which it provided help to migrants.⁴⁰ However, until ‘the late 1980s, IOM’s emergency responses were traditionally focused on movements and medical checks related to the resettlement of refugees and displaced persons. In the 1990s that situation changed’.⁴¹ It expanded the array of its crisis-related activities to encompass an ever-wider range of services such as humanitarian evacuation, camp management, and border control. During the 1990–1991 Gulf War, IOM for the first time adopted the role of first emergency responder evacuating displaced persons in an active crisis context. In this section, building on the concept of critical juncture, we analyse how this decisive precedent came about and what short and long-term institutional effects it produced.

7.3.1 Permissive and Productive Conditions: Understanding IOM’s Gulf War Operations

Arguably, a number of exogenous factors eased the constraints on political agency at IOM in the early 1990s, creating the possibility space for its expansion in the realm of humanitarian emergency assistance. One such permissive condition certainly was the end of the Cold War which created a moment of malleability in international politics more generally. Most importantly for our purposes, the fall of the Soviet Union and the temporary cessation of great power rivalry allowed for a surge in Western-led, liberal forms of institutionalized cooperation around the globe.⁴² New and more capable organizations were created (e.g. WTO, OPCW, etc.), existing ones started tapping the potential of their original mandates (e.g. UN Security Council), or received additional authority.⁴³ Similarly, for IOM the end of the Cold War created a window of opportunity to transition from a Western or US-led service organization to an IO with global ambition. It soon expanded its membership base to the East and it suddenly seemed possible to more fully live up to the global aspiration included in the 1989 Constitution.⁴⁴

The IOM Constitution itself represents an additional factor that opened the range of available options and increased the possibility for agency. It is an important historical coincidence that the amendment to the 1953 ICEM Constitution, debated since 1975 and adopted in 1987, entered into force on 14 November 1989, five days after the fall of the Berlin Wall.⁴⁵ The new Constitution was supposed to reflect a broadened field of activities that the organization had come to occupy and the changed geographical focus since its creation as an ad hoc Committee to deal with post-war refugees in Europe in the 1950s. In combination with the (in-)famous lack of a formal protection mandate given to IOM by its member states, that is, the fact that there is no set of norms and rules that the organization is supposed to observe and help implement,⁴⁶ the result was a constitutional text that merely states very broad objectives for the organization without clearly defining either the scope of these goals or the way that they should be realized. Article 1 says that IOM shall ‘make arrangements’ and ‘concern itself’ with the ‘organized transfer’ of migrants in need of assistance as well as refugees and displaced persons. This can mean virtually anything. While there is no indication that the Constitution drafters intended to carve out space for the organization to expand into new areas, this imprecision and rule ambiguity factually provided IOM with the legal flexibility to engender policy innovations.⁴⁷

Finally, the turn of the decade also saw a steep rise in regional inter- and intra-state armed conflicts that strongly influenced population movements by generating huge numbers of refugees and displaced people. The 1990–1991 Gulf War was the first such conflict. After the Iraqi invasion of Kuwait, the United States launched the first UN-sanctioned military campaign to liberate Kuwait and protect Saudi Arabia. Moreover, amidst the hostilities, Iraqi Kurds attempted a secession from Iraq that was quashed by air and ground attacks of the Iraqi military. At both fronts, thousands of refugees and displaced people were left in dire conditions. Kuwait, in particular, had hosted a large number of migrant workers from South-East Asia that were displaced within Kuwait or fled to neighbouring Saudi Arabia. In a strict legal sense, these were not refugees according to the Geneva Convention that pertains to individuals being forced out of their country of citizenship (Art. 1). The movements of people during and after the Gulf War thus did not fall squarely and exclusively within the mandate of UNHCR⁴⁸ – a condition that opened the door to IOM.

There was nothing necessary about IOM’s subsequent involvement in the humanitarian emergency response, however. For one, UNHCR actually offered its services to the UN Secretary General and thus signalled its readiness to take the lead in the humanitarian emergency response.⁴⁹ That the offer was refused and IOM given the lead⁵⁰ instead is a puzzle to be explored. Moreover, nothing in IOM’s mandate and previous practice would have made it seem necessary or logical for the organization to stage a big emergency relief effort. Similar crises in the previous decades had not triggered that kind of response and the new Constitution did not specifically ask for it either. The question is thus what productive conditions caused IOM’s surprisingly intensive engagement in the context of the Gulf War. After all, IOM became active in the region at an extremely fast pace and immediately started to evacuate displaced persons and stranded migrant workers by air, land, and sea routes. IOM set up offices in Kuwait and Southern Iraq and moved as close to areas where hostilities were ongoing to identify and assist people willing but so far unable to leave the countries. As early as 3 September 1990, a month after hostilities had started, IOM had organized the first ‘humanitarian repatriation flight’⁵¹ and evacuated about 155,000 people by the end of the year. Later, it also cooperated with UN Blue Helmets to facilitate the safe repatriation of more than 600,000 displaced Kurds that were transported in a fleet of locally rented trucks and buses.⁵²

What drove IOM to take on this new role? The official account tells a rather formalistic story of streams of forced migrants causing the affected governments to call on the UN for help which then asked IOM to take the lead in providing transportation and return-related services.⁵³ However, the account given to us in an interview by Bill Hyde,⁵⁴ the head of IOM’s emergency response team in the Gulf War, has a strikingly different tone to it. In his recollection, it was especially the coincidence of IOM’s willingness and ability to act and the double leadership role of the US in the coalition forces as well as in IOM that facilitated its entry to the scene. The US happened to have the authority both within IOM to sanction a certain course of action and on the ground in the conflict region to allow actors of their choice to become active. The particularly dominant position of the US in IOM, which the US valued for its managerial and outcome-oriented style of operation, contrasted with its rather complicated relationship with UNHCR which it deemed too liberal and politically entangled at the time, may explain how IOM got into the central position.⁵⁵ Hence, before any other agency apart from the Red Cross had reached the region, the first IOM team was already on its way. On board the US ambassador’s aeroplane it landed in Kuwait City⁵⁶ and was introduced to the Kuwaiti government to whom the IOM officials explained what they had to offer and were authorized to carry it out.⁵⁷

What made IOM an attractive cooperation partner for all concerned governments and gave it a competitive advantage over other IOs was basically two critical antecedents. On the one hand, IOM was not constrained by a mandate bound to legal definitions of who could be assisted under what conditions. According to Hyde,⁵⁸ ‘IOM has always been doing things on a timely basis for the greater good’ – a notion that was ‘a bit nebulous without being illegal’. In this sense, IOM showed an amount of flexibility much required in the complex Gulf War crisis that was ‘not very much in the DNA of established UN organizations’. On the other hand, IOM possessed the technical expertise needed for the task at hand. While it had never operated under these precise circumstances and had never used its tools for the exact same purposes, it was still very used to organizing the logistics of people's movement. Accordingly, the main operative task in Kuwait and Iraq ‘fit right into our ballpark’.⁵⁹ In the end, IOM was already operating an ad hoc but functional system of emergency evacuations when other actors entered the scene and inter-agency coordination started. Due to the organization’s first-mover advantage, its leadership position in the area of emergency evacuations and the provision of shelter was never questioned. ‘Needs were so immediate that there was never the question if we should have the lead … it was ‘you have a plan, you have the resources, it’s within your broad mandate, you can do this, you can do it now, so please do it’.⁶⁰

7.3.2 The Short- and Long-Term Institutional Consequences of IOM’s Gulf War Operations

Many things had to come together for IOM to adopt this outstanding role on the humanitarian assistance front in the Gulf War. Important permissive conditions such as the end of the Cold War and the outbreak of the Gulf War created a possibility space in which productive conditions such as the US’s dissatisfaction with UNHCR and IOM’s flexible problem-solving approach allowed for an unprecedented institutional outcome. But how did the Gulf War episode affect the organization’s institutional development on the long run? According to Georgi, ‘the First Gulf War in 1990/1991 was the single most important event at that time for IOM’s subsequent expansion’.⁶¹ As we argue, it set in motion a path-dependent process of institutional growth in the area of humanitarian assistance by ex-post formalizing competence in the area and creating organizational capacity which would be redeployed to different contexts, thus facilitating a cognitive normalization over time.

While IOM’s Gulf operations were initially conceived as a unique and one-off engagement, Director-General Purcell recognized the potential for a recurrence of comparable scenarios and tasked the head of IOM’s Gulf operations with the establishment of the Emergency Response Unit (ERU) which became operative in 1992.⁶² This was a completely independent process without member state interference as the ERU at first did not require any new resources. Its working method was to connect, train, and equip standing staff for future emergency interventions by IOM.⁶³ Over the next few years, the Unit developed IOM’s emergency preparedness and put it to tests in a number of refugee- and displacement-generating conflicts such as in Yugoslavia (1992), Rwanda/Zaire (1994), and Chechnya (1994). Building on this increasingly frequent involvement in humanitarian assistance, the IOM Secretariat in 1995 proposed a ‘strategic plan’ supposed to formally include for the first time a task to provide migration assistance to persons affected by emergencies. Reportedly, this step was not unequivocally supported by member states who feared overlaps and duplications with other IOs in this area.⁶⁴ However, while IOM needed to officially recall that it did not view itself primarily as an emergency response organization, none of the member states were seriously opposed to its substantive work in the realm of humanitarian assistance.⁶⁵ Accordingly, IOM continued to step up its crisis response activities. The increasingly extensive involvement of IOM in conflict regions such as East Timor and Kosovo towards the end of the 1990s, for instance, led to an institutional solidification of these efforts in the larger Emergency and Post-Conflict Division in 2000, a precursor of today’s Department of Operations and Emergencies (DOE) that firmly enshrined humanitarian assistance in emergencies in IOM’s institutional structure.⁶⁶

Beyond the immediate impact that the Gulf War intervention had on IOM’s organizational structure, it also influenced the organizational culture and its perception by its environment. As an exemplary precedent, IOM’s Gulf operations changed how IOM’s role was perceived internally and externally. The precedent suggested a pattern that was transferrable: A crisis exposes governance gaps in terms of timing and functions; IOM has some capacity in its portfolio that can be used to fill such gaps; IOM immediately and actively offers and advertises its services to member states who value the organization’s flexibility and low expected normative costs; IOM moves in before anyone else and sets another precedent for a new kind of activity; if carried out effectively, there is recognition at both IOM and its member states that this type of activity may be useful in other contexts, too, which leads to its ex-post institutionalization. At the level of organizational culture, this produced and over time reinforced an ‘esprit de corps’ among IOM’s civil servants that help would be provided wherever help was needed, irrespective of formal responsibilities and conventional views of the boundaries of migration management.⁶⁷ At the level of organizational environment, member states and relevant non-state actors grew increasingly accustomed to IOM’s flexibility and started to use its fungible capacities for crisis-related activities that were ever more remote from the organization’s previous focus on migration management and, sometimes, even from the issue of migration altogether. For instance, the Organization for Security and Cooperation in Europe (OSCE) enlisted IOM to facilitate out-of-country-voting for citizens of Bosnia and Herzegovina in 1996–1999,⁶⁸ and, starting with Mozambique in 1992, several member states made use of IOM’s field presence to assist post-conflict disarmament, demobilization, and reintegration (DDR) campaigns in by now over 120 projects.⁶⁹

With both sides learning how to profit from each other in a growing array of activities and building on a consolidating base of experiences, we may conclude that a mutually reinforcing cognitive process of convergence underlies a mechanism of increasing returns that reproduces the institution’s path towards horizontal task expansion. In the following, we use two important cases in the more recent history of IOM to underscore the claim that this logic of institutional expansion through precedents has taken hold in the organization’s development: the 2011 civil war in Libya and the 2014–2016 Ebola crisis.

7.3.3 Path-Dependent Reproduction of IOM’s Expansionary Logic in Libya and West Africa

Both the civil war and foreign intervention in Libya and the Ebola crisis in West Africa gave rise to further emergency operations by IOM that covered partly new terrain and led to ex post institutional accommodations of its practice. While the Libyan case was marked by the creation of new best practices by IOM as a now focal manager of migration crises, the case of Ebola saw IOM redeploy its emergency toolkit to a new type of crisis context, namely one caused by the spread of a contagious disease.

7.3.3.1 Setting New Best Practices in Libya

In February 2011, civil unrest erupted in Libya in the context of the so-called Arab Spring. The situation quickly escalated into a civil war between the Libyan army of the Gaddafi government and rebels supported by NATO air forces. In terms of the number of people displaced, the civil war caused one of the worst migration crises in the region since the first Gulf War. Before the war, the Libyan economy had heavily relied on migrant workers with foreigners making up about 21–35% of the Libyan population.⁷⁰ When the war broke out, both Libyan citizens and migrant workers tried to escape the violence and flee the country. Many of the foreigners who wanted to leave Libya were (mostly undocumented) manual labourers from sub-Saharan Africa.⁷¹ Soon, a severe governance gap was exposed: While the migrant workers’ countries of origin lacked the capacity to bring home their citizens, UN agencies were, at first, prevented from providing assistance due to the strict security protocols and the escalating violence on the ground.⁷² Moreover, the UN was in a weak position to negotiate access to the country as the Security Council had authorized military action against the Gaddafi regime. The situation called for an actor to coordinate with both, the Libyan government and NATO, which was trusted by the migrant workers’ countries of origin,⁷³ and able to enter the dynamic and dangerous environment in Libya. IOM fulfilled these criteria.

IOM was the first responder on site.⁷⁴ Its field office in Tunisia, which conducted most of the emergency response, consisted of two to three employees on the day the war broke out. Within a week, IOM had deployed about 1000 staff to the Tunisian country mission who were working on the ground at the Libyan border.⁷⁵ Soon after the onset of the crisis, IOM coordinated with UNHCR to set up the ‘Humanitarian Evacuation Cell’ (HEC), a liaison body of the two organizations at headquarters’ level.⁷⁶ It coordinated with the Libyan government to obtain the needed clearances and access to regions affected by ongoing fighting and with NATO to fly out migrants through the no-fly zone.⁷⁷ Moreover, IOM coordinated with humanitarian organizations, set up temporary camps, performed necessary health checks, and transported large numbers of migrants out of Libya.⁷⁸

Most of IOM’s operational activities in the early phase of the 2011 Libyan migration crisis can be considered part of what had become the organization’s core crisis portfolio. At that point, IOM was used to negotiate with warring parties to gain access to conflict zones and its abilities as a facilitator of mass transport were well known. Two aspects of IOM’s crisis response in Libya were unprecedented, however. One was that beyond evacuation, IOM also started building capacities to support migrants once they disembarked their means of transportation outside the conflict zone.⁷⁹ By creating transition camps, integration programs, and community projects, IOM assumed tasks typical for a development agency. The second was IOM’s focal position as a coordination hub between all parties involved. In the past, IOM had operationally assisted UN-coordinated efforts on the ground, especially in tandem with UNCHR.⁸⁰ Over the course of the Libyan crisis, however, it became a key coordinator, eventually co-leading the Refugee and Migrant Platform and preparing a Joint Operational Framework for Humanitarian Response in Libya.⁸¹

IOM’s initial response to the Libyan migration crisis ‘was unanimously welcomed abroad’,⁸² as the migrant workers’ countries of origin praised the organization’s swift action on the ground. Additionally, and in contrast to what had previously been understood as a rather tense relationship,⁸³ UNHCR acknowledged the improved partnership between the two organizations that proceeded to co-publish joint statements at a distinctly accelerated rate.⁸⁴ This positive feedback notwithstanding, a few months after the start of its operations IOM actually encountered an unprecedented funding lag.⁸⁵ At the peak of the crisis, the organization ran out of funds to charter all the planes necessary to transport migrants to diverse locations on different continents.⁸⁶ Even though most of IOM’s typical donor states were willing to finance the efforts, the US preferred to fund efforts in Iraq rather than in Libya,⁸⁷ causing a crunch in the operations. IOM and its member states thus had to learn the hard way that the organization’s short-term project-based budget proved insufficient to fund emergency evacuations of such a scale.⁸⁸

After the acute phase of the 2011 Libyan migration crisis had subsided, IOM translated such lessons into prescriptions for the handling of future crises. First, it used the Libyan example to sell the idea of a new funding mechanism to its member states.⁸⁹ With success: The IOM Council created the ‘Migration Emergency Funding Mechanism’ in December 2011, a permanent fund to finance IOM’s widened set of humanitarian evacuation efforts in future similar situations.⁹⁰ Second, the complex and multi-layered crisis in Libya arguably served as an eye-opener demonstrating the need for a structured and concerted approach to the governance of migration crises.⁹¹ The Libyan experience thus provided the spark for the development of the ‘Migration Crisis Operational Framework’ (MCOF), which was approved by the IOM Council in 2012.⁹² MCOF has since become a centrepiece of IOM’s emergency responses. The document describes a variety of activities to be undertaken in crisis situations by IOM staff on the ground. While the document rationalizes a task expansion beyond mere migration matters, it was justified as enabling the organization to even better respond to such situations in the future.⁹³ Third, IOM’s improved relationship with UNHCR and generally the functioning inter-agency coordination during the Libya crisis also spurred lasting institutional change. In particular, the HEC, which was originally ‘thought to be time-limited’ became a permanent mechanism ensuring sustained cooperation with UNHCR.⁹⁴ Moreover, in 2016, IOM became a related organization to the UN, formalizing the ever-closer embeddedness of IOM within the UN framework⁹⁵ and thus allowing IOM to assume the role of the central coordinator in future crises.

7.3.3.2 IOM’s Venture into Global Health Crisis Management: The 2014–2016 Ebola Outbreak

Another illustrative example of IOM’s expansion into a new area that few would have associated with the portfolio of the organization is its involvement in the 2014–2016 Ebola epidemic in West Africa. In March 2014, an outbreak of the Ebola Virus Disease (EVD) was detected in Guinea. At that time, the virus had already spread to neighbouring Liberia, Sierra Leone, and Mali.⁹⁶ While certainly propelled by the fact that various cultural communities in the region span borders, the resulting health crisis had little to do with migration.

Similar to the Gulf War and the civil war in Libya, the situation in West Africa was perceived as very dangerous and unclear. Given the magnitude of the problem and the limited governance capacities of the states involved, some of IOM’s most influential donor states, the United Kingdom, France, and especially the United States, asked the organization for assistance.⁹⁷ At first, IOM hesitated to get involved due to safety concerns for its staff⁹⁸ in light of what was perceived as a ‘completely new threat’.⁹⁹ However, the US government under the Obama administration insisted,¹⁰⁰ referring to IOM’s proven ability to move into extremely difficult situations with speed and quick adaptability.¹⁰¹ Since the organization lacked a health-related framework for operation at the time, IOM started projects under its recently established ‘Humanitarian Border Management’ (HBM) framework, which intends to prepare governments and border authorities for crisis-induced mass movements and displacement.¹⁰² IOM joined a cluster of IOs responding to the EVD crisis, which included the World Health Organization (WHO) and was coordinated at headquarters’ level by the UN Mission for Ebola Emergency Response (UNMEER).¹⁰³

Contrary to IOM’s previous crisis responses, it is worth noting that the organization did not jump on the opportunity to enter uncharted territory in the case of the Ebola epidemic. Of course, the organization’s hesitation was not based on concerns about potential mandate violations or organizational over-stretch, but about its staff security. The fact that IOM’s most influential donor states still enlisted the organization for sake of its flexible crisis management capabilities illustrates the advancement of the cognitive mechanism of self-reinforcement. After repeated demonstrations of its usefulness for varied crisis governance tasks, IOM’s member states seem to have internalized an impression of the organization as a quasi-universal tool deployable in any kind of crisis context. IOM does not necessarily have to push for its involvement anymore – it is being pulled in.

Once the decision was taken, the organization repeated the same pattern in had established in previous crises. It quickly deployed its staff and started out with its core activities – the documentation and transportation of people – in order to ensure that virus testing results got to the correct individuals and that patients could reach health facilities.¹⁰⁴ Soon, IOM identified governance gaps on the ground which were not filled comprehensively by any other responder and could be addressed by IOM. Thus, IOM started to conduct health screenings,¹⁰⁵ sanitized migrants,¹⁰⁶ and provided psychosocial counselling sessions at schools in areas affected by EVD.¹⁰⁷ It set up clinics and emergency treatment centres based on a US model¹⁰⁸ and managed its own Ebola treatment units (ETU).¹⁰⁹ IOM was the first to conduct trainings of border officials¹¹⁰ on health screening¹¹¹ and later formally took over the full management of the Ebola training academies from the United Kingdom’s Ministry of Defense in Liberia¹¹² and Sierra Leone.¹¹³ Moreover, IOM built structural improvements to border checkpoints, creating so-called flow monitoring points (FMPs)¹¹⁴ to collect data on people’s movements based on the HBM framework.¹¹⁵ These border surveillance measures led to the creation of an unprecedented collection of data mapping population flows in the region.¹¹⁶ Finally, IOM carried out a comprehensive public health information campaign, including radio spots,¹¹⁷ town hall meetings, billboards, posters, and comics,¹¹⁸ to inform the public on matters like EVD prevention measures, immunization campaigns, the ETUs, and the fight against the stigmatization of EVD survivors.¹¹⁹ The information campaign also involved consultations with local authorities and community leaders¹²⁰ who were trained on community preparedness for EVD.¹²¹

While the documentation, logistics, and transportation parts of the operation were IOM’s core business, many of IOM activities during the Ebola crisis appear to be new endeavours for the organization, at least at such a scale and with such intent.¹²² It was the first time that IOM engaged in border surveillance with FMPs according to the HBM framework and it had neither conducted a major public health campaign nor taken over the management of entire emergency treatment centres before.¹²³ To be sure, IOM was also used to health-related activities inasmuch as ground staff often performed routine health checks for migrants before boarding transportation and the organization had been involved in the cholera outbreak in Haiti a few years prior. In the Ebola crisis context, however, IOM conducted health checks and treatments at a new scale.¹²⁴

While the World Health Organization (WHO) was criticized for its (mis)management of the Ebola epidemic,¹²⁵ IOM received mainly positive reactions for its involvement in West Africa, even though it was not expressly mandated to respond to health emergencies. For example, UNMEER repeatedly expressed its appreciation of IOM’s activities¹²⁶ and the WHO Director-General praised the Ebola-related cooperation with IOM in a speech at the IOM Council.¹²⁷ Moreover, at the IOM Council, an African Union spokesperson thanked IOM for the swift response to the crisis.¹²⁸ However, some member states, especially the Netherlands, also voiced concerns about the apparent mandate violations in the Ebola crisis.¹²⁹ IOM’s leadership retorted that ‘migration is a cross-cutting issue’ and that it was able to ‘tie all its activities to migration’.¹³⁰ In the debates, it received backing by the US as one of IOM’s most influential member states and major donors. That the Americans praised the organization’s operation in the context of the Ebola crisis successfully muted concerns regarding legal issues and mission.¹³¹

Again, IOM followed the pattern it had established with its involvement in the Gulf War and used the positive feedback from its major donor state to create a new framework called ‘Health, Border and Mobility Management’ (HBMM)¹³² in order to formalize its expanded portfolio.¹³³ Based on the understanding that diseases do not stop at borders,¹³⁴ HBMM was considered a reiteration of the HBM framework.¹³⁵ It includes a diverse set of tasks ranging from ‘operational research, evidence, data gathering and sharing’, which normalizes the surveillance aspects of the Ebola response and underlines IOM’s continued effort to expose governance gaps during crisis, to ‘enhanced capacity of health systems and border management services’.¹³⁶ Sources inside IOM maintain that the main lesson learned from the Ebola response was the realization of a ‘continued need for capacities’ to respond to health crises¹³⁷ and that the HBMM framework was a direct result of IOM’s Ebola crisis response.¹³⁸ Since then, IOM has continued to perform health-related activities in crisis response operations around the world based on the HBMM framework.¹³⁹ In the context of the contemporary COVID-19 pandemic, IOM is promoting its ‘Global Strategic Preparedness and Response Plan’, which is anchored in the HBMM framework.¹⁴⁰ Starting with the Ebola crisis response, IOM has thus successfully established itself as a player in yet another policy field not originally covered by its mandate.

7.4 Conclusion

Today’s institutional design and policy of IOM is heavily influenced by the historical legacies of its earlier crisis interventions. As we argued in this chapter, IOM’s contingent emergency response in the Gulf War marked a critical turning point in the organization’s evolution. Ever since, it has embarked on an institutional trajectory branching out further and further into the realm of humanitarian assistance in an ever-wider range of crisis contexts and in an ever more central role. The case study attests to the power of precedents in IOM’s development. As predicted by historical institutionalism, once institutional choices provide increasing returns over time, they are not easily undone. In the remainder of this conclusion, we shall reflect on the implications of our findings for IOM’s ethos, obligations, and accountability.

It seems most relevant for IOM’s task expansion in the field of humanitarian emergency assistance that the organization is underpinned by an ‘esprit de corps’ in its staff that seems to prioritize hands-on assistance to people in need over broader normative or legal concerns. While arguably part of the organization’s DNA from the beginning,¹⁴¹ this practical helper ethos not only facilitates flexible crisis interventions in uncertain circumstances, but it is itself also reinforced in tune with the number of social precedents set by IOM in this area of activity. As suggested by the accounts of our interview partners, every new crisis intervention following the pattern provides arguments to rationalize (any other case of) humanitarian emergency assistance in terms of the organizational ethos: ‘because this is what we do’.

The small regard for mandate violations or legal ramifications at IOM hints at a conflict that its ethos may create with obligations and accountability. What our account of IOM’s near-exponential growth in the area of humanitarian emergency assistance has revealed in this regard, is that its expansions generally predate the adoption of any clear policies to reflect the pertinent normative principles. The Gulf intervention predated the adoption of any humanitarian policy principles, the Libya intervention predated the formalization of MCOF and MICIC¹⁴² policies, and the Ebola response predated the adoption of the HBMM framework. Indeed, on the long run, these steps lead to a normative regulation of IOM activities. In the moment of expansion, however, IOM acts in a normative void ruled by facticity only. In this void, it is hard to discern forms of accountability that go beyond answering to donor states. In fact, as our model suggested, IOM often works at the behest of particularly powerful donor states on the territory of weaker states, without any clear foundation in a multilaterally endorsed set of principles. While the organization is thus strongly accountable to a few states, the countries and societies most affected by IOM’s interventions lack the means to hold the organization to account. From a constitutionalist perspective, this is hard to reconcile with normative legitimacy requirements. However, legitimacy assessments need to consider both the input and output dimensions. To what extent IOM’s achievements in living up to its ethos balances these normative problems, is a question we can only allude to here.