The Kosovo Refugee Crisis: An Independent Evaluation of UNHCR’s Emergency Preparedness and Response
by Guy S. Goodwin-Gill, Professor of International Refugee Law, University of Oxford, Oxford, UK December 2012

This is a welcome report; it highlights successes, but also failings and weaknesses. It asks whether Kosovo refugees obtained appropriate protection and assistance, and whether UNHCR met its own standards. It looks at five areas in particular, namely context, including background, preparedness and initial responses; management; assistance and coordination; protection; and relations with the military. This short review touches only a few.

Kosovo was not unique, even though no one disputes that the exodus was unusually large and swift – some 500,000 refugees fled within two weeks, rising to a high probably in the region of 850,000. No one disputes, either, that UNHCR was constrained by circumstance. But that aside, all the errors recognised in its own Great Lakes report (EC/47/SC/CRP.11, 2 January 1997) are repeated: structure that did not ensure effective coordination; poor communication; weak information management; lack of sufficient senior protection staff; protection problems inadequately dealt with and not given sufficient priority; and so on.

UNHCR’s weak presence in neighbouring states is noticed early, and its long involvement with IDPs in Kosovo criticised for distracting it from a possible substantial external exodus (95). Its failure, predating the crisis, to develop relationships with Albania and Macedonia is hard to understand. Common sense and local history should have put UNHCR on notice of external movement (81), but the evaluators do not excuse it for doing no better than any one else. Mandated to protect and assist refugees, UNHCR ‘might be expected… [to] be more institutionally inclined… to consider the possibility of a massive refugee outflow’ (92). It did not, and the report finds a ‘near complacent attitude’ on its part.

If lack of information and analysis contributed to this attitude, then UNHCR has only itself to blame. It had long since cut back resources for its Centre for Documentation and Research (CDR), whose rather comprehensive background paper on Kosovo from February 1996 has never been updated.

Against this state of unreadiness, the report finds substantial management weaknesses and a staggering lack of in-country and headquarters support (288). Once again, staff were deployed too late and too slowly, and mid-level and senior management were missing. Initial staff movements, quick enough in parts, were based on the emergency response team roster, but the report highlights why the ‘volunteer’ system did not work thereafter, including managers resisting release of personnel and lack of recognition or credit for those going to the field. Systemwide demoralisation likely also played a part.

On protection, the report finds that UNHCR ‘performed as well as the situation permitted, within the framework of traditional approaches to universal refugee protection’ (438, 504). This is usefully ambiguous, but ought to be read as criticism.

Thus, UNHCR tried to promote the principle of mass admission on the Kosovo-Macedonia border, but apparently paid no heed to the context in which the relevant international legal principles have emerged, or to their application in a dynamic sense. Its performance on ‘first asylum’ strongly implies that staff initially responsible for protection in the region, and possibly also in Headquarters, are unfamiliar with basic principles. If, as the report suggests (482, 511), UNHCR seriously considers that ‘first asylum should be considered as an absolute and unconditional legal obligation consistent with the 1951 Refugee Convention’, then there is something wrong with its perception of international law. UNHCR’s primary responsibility is to provide international protection to refugees and not, as such, ‘to promote the right to seek asylum in a Convention signatory State’ (459). Its first concern, therefore, is not with ‘choice’, but with non-refoulement, which can as easily be offered in a country of ‘first asylum’ as in a neighbouring country willing and able to grant refuge. History is full of reluctant hosts, where basic principles of protection could only be maintained contingently; that is, through the negotiated support of other states (451-6).

The report rightly identifies the ‘Humanitarian Transfer Programme’ (ie, transfer of Kosovar refugees from Macedonia to Albania and Turkey) and the ‘Humanitarian Evacuation Programme’ (ie, transfer of Kosovar refugees from Macedonia to other European countries and beyond the region) as ‘innovations… that generally enhanced protection’ (507). That more use was not made of the former again appears due to lack of a clear position of principle on the part of UNHCR (475, 478), and even to obstruction in the field (478). The most practical approach, consistent with the goal of protection, seems to have come from the Director of the Europe Bureau (478, note 48). 

What can be drawn from the report? First, there is no evidence of a protection strategy, let alone a plan. Indeed UNHCR, as presently constructed, appears still to offer no place in which a system-wide protection strategy can be devised and implemented.

Second, first asylum is a fine ideal but it does not reflect general international law or the requirements of the 1951 Convention/1967 Protocol. The practice of states – the heart of customary international law – cannot be disregarded when devising a protection strategy or plan.

Third, in relation to emergency response, UNHCR is not capable of infinite and immediate expansion which means contraction and standby arrangements with other, particularly non-governmental, actors play a critical role. However, they ought to be complementary to the development and strengthening of a national (ie, refugee receiving country) capacity to deal with refugee emergencies; expatriate solutions may be necessary from time to time or for particular purposes, but rarely offer long term ‘value added’, especially for developing countries or countries in transition.

Fourth, in the matter of early warning and preparedness, whether UNHCR might or might not have predicted the exodus with greater accuracy, it could certainly have made better use of internal and external resources. The report suggests a certain imperviousness to information and analysis – both in protection and operations – which needs urgent correction.

Kosovo may have been special, but it was hardly unique. Overall, UNHCR has not learnt from the mistakes, or the experience of the past, including earlier evaluations. Whether such evaluations actually have any impact in effecting institutional change may be seriously doubted, however, and the impetus for true reform must come now both from within and from without; there may be an opportunity here for the new High Commissioner. 

UNHCR’s growth over the past decade has been accompanied by an apparently fruitless, indeed damaging, search for appropriate structures. Bureaucratisation now prevails over both protection and operations, and the organisation no longer meets its responsibilities to the international community.

The report’s elucidation of errors suggests a serious failure to identify and transmit, both internally and externally, a clear vision of UNHCR’s role, especially in the provision of international protection; it seems to be confused as to its legal standing and as to the international law which it is supposed to apply; it seems unable or unwilling to take note of what is going on in the real world, at ground and field level; it lacks the structural capacity to integrate information and analysis into protection decision-making and policy and programme formulation; and it seems to have put aside those principles which in turn would enable it to lay down clear goals in relation to donors, states, the military, NGOs (whose voice, inexplicably, is difficult to hear at any point in the evaluation), and, above all, refugees.

These weaknesses themselves allow the inference of yet greater problems – namely, no transparency, little if any sense of accountability, and a tendency to submerge all and any criticism of performance in someone else’s absence of political will. Indeed, the report itself disappoints at this juncture; no one is identified save by post or acronym, credit and criticism fall indiscriminately, and one may well emerge with the feeling that, though a lot went wrong, no one ultimately was responsible. If that’s another lesson to be learnt, then perhaps we do need to return to the drawing board.

The report can be found at <www.unhcr.ch/evaluate/kosovo/toc.htm> For more information see ICVA’s newsletter: ‘Talk Back’ vol. 2 (1) ‘UNHCR’s Kosovo Evaluation’ at <www.icva.ch>