The Balkan Refugee Crisis: Regional and long-term perspectives

from International Crisis Group
Published on 02 Jun 1999

Executive Summary
The magnitude and complexity of the unfolding refugee crisis in the Balkans is hard to overstate. One and a half million people have been forced to flee their homes in Kosovo since the start of this year. These latest victims of Balkan conflict join the ranks of a further one and a half million other refugees and internally displaced persons (IDPs) from earlier wars in Croatia and Bosnia.

While many of those expelled from Kosovo are anxious to return home as soon as possible, the obstacles in the way of their return are formidable. Creating the conditions necessary for large scale return1 will take a long time and require enormous resources.

This report argues in favour of providing temporary protection2 for refugees in the region, with the aim of them returning home at the earliest opportunity. Temporary protection is necessary to maintain pressure on Belgrade and demonstrate our commitment to reverse the effects of ethnic cleansing. But this report argues for more realism in relation to the length of time it will take to reverse the present refugee flow. Lessons from Croatia and Bosnia have demonstrated that there is no such thing as fast voluntary return in the wake of war and ethnic cleansing3 . Perhaps induced, but most likely not voluntary. Non-voluntary return of refugees is a very sensitive issue. The international community can only try to circumvent it by striving to put in place the necessary conditions that would make return acceptable to Kosovo refugees. This report discusses these key conditions and calls for the establishment of a comprehensive repatriation plan. Strong regional management structures must be established by the Organisation for Security and Co-operation in Europe (OSCE) and the United Nations High Commissioner for Refugees (UNHCR) in order to develop, co-ordinate and implement the strategy for the return process.

Specifically, the report recommends that the international community focus on the following action points:

  • Maintain and promote the temporary status of refugees;
  • Develop a comprehensive strategy for the return of all refugees and IDPs;
  • Keep the refugees in the region, in so far as possible;
  • Plan for the early return for refugees evacuated to third countries;
  • Prepare for spontaneous return;
  • Plan according to realistic time frame;
  • Keep refugees informed;
  • Give equal attention to short- and long-term needs;
  • Involve the local population in the return process;
  • Develop regional humanitarian solutions and structures;
  • Mobilise up-front funding of return efforts;
  • Include the whole region in economic recovery planning;
  • Keep the roles of humanitarian aid workers and the military separate;
  • Include binding return mechanisms in the future peace agreement; and
  • Synchronise European refugee policy.


In a speech to high-ranking military officers in December 1998, President Tudjman of Croatia declared that "we have resolved the Serb question in Croatia", clearly indicating that the majority of the former Croatian Serb population never will return - in spite of the Dayton and Erdut agreements. Will Yugoslav President Slobodan Milosevic also 'resolve' his Albanian question by letting time work toward his long-term goals?

NATO air strikes against the Federal Republic of Yugoslavia (FRY) were launched to halt Milosevic's brutal ethnic cleansing campaign against the ethnic Albanians in Kosovo. Now that the majority of this population has been forced to flee the province despite NATO's action, the West must help these refugees and displaced persons find a future. Failure to address their plight will have significant political consequences for a fragile region that already harbours more than 1.6 million refugees and displaced persons from earlier conflicts in Croatia and Bosnia. With an additional 850,000 Kosovars in exile4 and more than 500,000 people internally displaced in Kosovo, the total number of refugees and displaced persons from and within the territory of the former Yugoslavia is at present (May 1999) estimated at more than three million people5.

The problem cannot be solved by a piece-meal approach. Representatives of the international community need to adopt a regional and long-term perspective in handling the Kosovo refugee crisis. Decision-makers and the media both tend to focus narrowly on Kosovo and on short-term humanitarian needs. There is one exception, namely the recently launched "Concept paper on a proposed framework for the return of refugees and internally displaced persons to Kosovo"6 , but the paper, drafted by UN agencies, also tends to neglect the regional context and some of the long-term obstacles to return.

This report proposes measures to address the human tragedy, political instability and economic chaos ensuing from the Kosovo refugee crisis and outlines a long-term approach to solving the region's problems. It also discusses the key conditions that must be in place if refugees are to return to Kosovo.

Displacement from and within the region

If one includes the victims of the current Kosovo crisis, the Balkan region now harbours more than three million refugees and internally displaced persons from Bosnia-Herzegovina, Croatia and the former Yugoslavia. What follows is an approximate breakdown of the various groups of these displaced persons in terms of numbers, current status and prospects for return7.

Displaced persons from Bosnia-Herzegovina

836,500 people are still internally displaced within Bosnia-Herzegovina. This group consists of ethnic Croats, Serbs and Muslims. Most of them come from places where they now would belong to an ethnic minority if they went home. Many of them had houses which have either been destroyed or occupied by other people. There are very few return movements at present. These uprooted people constitute a heavy political and social burden on many local communities and on Bosnia-Herzegovina as a whole.

223,000 Bosnian Serbs are still refugees in FRY. Although some of these still live in collective centres, most are lodged in private homes. Their halting return back to Bosnia has gathered pace since the NATO bombing campaign started on 24 March 1999. Among the returnees are young men who are leaving FRY to avoid being drafted to the FRY army. Many of the returnees add to the number of IDPs in Republika Srpska, since most of them originally come from areas where they would now belong to a minority. This kind of disorganised return represents no durable solution. It only adds to the existing tension in Republika Srpska. It is possible that steadily deteriorating conditions in FRY caused by the air strikes will push even greater numbers of people to return to Bosnia.

30,000 Croats, mainly from Bosnia8 , are still registered as refugees in Croatia. In addition, 140,000 Bosnian Croats have received Croat citizenship since the end of the Bosnian war, many of them still live in homes belonging to Croatian Serbs expelled from Croatia in 1995. A highly politicised group, most of its members come from communities that now constitute a local minority in Bosnia, or where minimal reconstruction has taken place. Very few have expressed any interest in returning to Bosnia. Tension around this group will increase sharply if the bombing of FRY leads to an increased return to Croatia of Croatian Serbs.

128,000 people from Bosnia-Herzegovina are still living as refugees in Western Europe. The largest group lives in Germany, but their number has declined since the German government decided to start sending them back to Bosnia. Of the 100,000 sent back so far, many originally came from communities that now constitute a local minority in the area. To a large degree, these returnees have preferred to relocate to other parts of Bosnia-Herzegovina where they are part of the ethnic majority, thereby adding to the number of IDPs. 1,250 refugees from Bosnia-Herzegovina live presently in Macedonia and 3,500 live in Slovenia.

The Dayton Peace Agreement intended to regulate return to Bosnia-Herzegovina through the establishment of key principles for voluntary return and procedures for the restitution of property. But the agreement lacks provisions for implementing these principles and procedures, which is one of its major shortcomings.

Displaced persons from Croatia

Up to 300,000 Croatian Serbs are still refugees in FRY after the exodus caused by Croatian military operations Flash and Storm in 1995. As with the Bosnian Serbs, most of this group lodge with Serb families having abandoned their properties in Croatia. Very few have returned in spite of international efforts to force the Croatian government to adopt non-discriminatory procedures for return, the restitution of property and reconstruction.

Deteriorating living conditions in FRY caused by the NATO bombing campaign might create severe social tension around this group. But if, for this reason, many now choose to return, political tension will certainly increase in Croatia.

62,000 Croats are still internally displaced within Coratia after the Serb offensive of 1992. This is a highly politicised group that takes an aggressive stance toward Croatian Serbs. This group is still in the process of returning to their former homes. Croatian Serbs now occupying these houses are often thrown out without alternative accommodation and forced to live either as internally displaced persons in Croatia or as refugees abroad. 30,000 Croats are still refugees in Bosnia-Herzegovina.

Displaced persons from the Federal Republic of Yugoslavia

Already, before NATO bombing began on 24 March 1999, 100,000 Kosovo Albanian refugees were scattered in 27 European countries as asylum seekers.

Over the past two months, 750,000 Kosovo Albanians have left the province for Albania, Macedonia, Montenegro and Bosnia-Herzegovina. Of these, more than 30,000 have so far been evacuated from Macedonia to other countries. The intention is to evacuate at least 100 000 more9. An estimated 500,000 Kosovo Albanians are displaced within the borders of Kosovo. These enormous numbers constitute more than two-thirds of the whole Albanian population in Kosovo. The ethnic cleansing is presently continuing.

The situation of the 200,000 mainly ethnic Serbs in Kosovo is presently unknown. A key question is whether this population will feel sufficiently safe to stay on after a cease-fire or peace settlement, in the face of a significant return of Kosovar Albanians. If not, we may see a considerable movement of Kosovar Serbs into a Serbia that is already hard pressed to cope with the needs of its existing population in the wake of the bombing and many years of economic sanctions. If the Kosovar Serbs relocate to Serbia, they would be likely to stay for the foreseeable future, adding to social and political instability in FRY.

Temporary Measures and Durable Solutions

The reception and accommodation of Kosovo refugees now underway in Macedonia and Albania is explicitly seen as a temporary relief measure and not as a long-term solution. Other states that have offered to burden-share by taking in Kosovo refugees - through a so-called humanitarian evacuation - also consider this arrangement temporary.

Temporary relief measures aside, there are a limited number of durable solutions potentially available to the region's refugees and displaced persons:

  • return;
  • settlement in countries neighbouring the conflict;
  • resettlement in third countries; or
  • relocation within the country of origin, as practised in Bosnia.

The following section discusses the merits and pitfalls of these different refugee scenarios in the Kosovo context.


Return is always the preferred solution to a refugee crisis, for humanitarian as well as political, legal and financial reasons10. Where ethnic cleansing is the reason for the exodus, return also becomes an imperative for the international community as a way of demonstrating to the perpetrators' the unacceptability of their behaviour. This is reflected in the fact that NATO lists the 'unconditional and safe return of all refugees and displaced persons' as one of five conditions to end the military offensive against FRY. Voluntary return is also prescribed in the UNHCR's mandate, and forms the basis of both the Erdut and Dayton agreements. The UNHCR concept paper for refugee return to Kosovo11 lists several essential protection standards:

  • Refugees and IDPs must be able to make a free, voluntary and well-informed decision regarding repatriation or return;
  • Refugees and IDPs who are not willing or able to return to Kosovo must continue to receive protection and necessary assistance;
  • Returns must take place under conditions of legal safety, physical security, material security and dignity;
  • The basic humanitarian needs of returnees and other affected populations for protection and assistance must be met;
  • Returnees enjoy and may exercise their fundamental human rights without discrimination;
  • New displacement is prevented through protection monitoring, reporting, intervention and related activities;
  • Sustainable reintegration is achieved through the restoration of national protection mechanisms, reconstruction and reconciliation.

Settlement in countries neighbouring the conflict

Settlement in countries neighbouring the conflict will always have delicate political and financial implications. The presence of large numbers of refugees is prone to destabilise the existing political and ethnic balance. Sometimes this can strengthen the ruler's position (Albania), other times it can be perceived as a real threat to the ethnic status quo (Macedonia). But in both cases a mass influx of refugees represents a heavy financial burden on the host country and thus in itself cause tension. With the exception of Croatia's naturalisation of the majority of Bosnian Croats, settlement in neighbouring countries has so far not been seen as a durable solution for refugee problems in the Balkans region.

Resettlement in third countries

In the past, many refugees have been permantently resettled to third countries willing to take refugees, in particular USA, Canada, Australia, New Zealand, the Scandinavian countries. This option is limited in principal to those few cases where refugees need to leave a country or region to obtain international protection. While resettlement solves some problems, it can create others. The option of resettlement can accelerate refugee exodus by providing a gateway to attractive immigration states12.


Where ethnic cleansing was the reason for exodus, refugees may shy away from returning to their country of origin, even if parts of the country of origin represent a safe environment. Though not the preferred option in the Dayton Agreement, the great majority of recent returns to and within Bosnia have occurred through relocation to places where the returnee forms part of the ethnic majority. Among the disadvantages of relocation are that it enforces ethnic divisions and creates problems of overpopulation in the most 'popular' majority areas.

Humanitarian evacuation

Transporting refugees out of their region of origin for protection on a temporary basis is called 'humanitarian evacuation' to make the distinction from more permanent solutions such as resettlement. Humanitarian evacuation and resettlement programs are extremely expensive compared to relief efforts in the vicinity of the crises13 .

Another disadvantage is the psychological effects of these programmes, which select few individuals, often by unclear criteria, and offer them much higher standards of living compared with those left behind. The likelihood that these people will return, having been offered full integration along with other refugees on a permanent status in Western countries, is slight. Governments have widely differing policies in this area.

In the case of Kosovo, past experience suggests that even if refugee status is eventually withdrawn with the aim of returning evacuees to Kosovo, these people will be the last to go home. For various reasons many of them will wish to settle permanently in their countries of temporary refuge, and they will receive support from their new social networks and pressure groups. Any form of mandatory return, even once safe conditions have been created in Kosovo, will be labelled inhumane. The evacuees may also be seen to have received an unfair advantage, which could act as a divisive factor between them and friends and relatives who stayed in the region and had to face the difficulties there.

Evacuees will also be the last group to go home if return is planned in phases as recommended in the UNHCR concept paper. Phased return means that one seeks to repatriate internally displaced persons before those from neighbouring countries, and then, as the last group, the refugees "temporarily" lodged further away. This report argues that the reverse should happen, that is, that the evacuees should be the first ones to be repatriated. The fact that these refugees have had better material conditions in exile should be used as an argument in support of early return, rather than against it.

Temporary protection

In the West, 'asylum' has traditionally been a method to provide refugees with permanent international protection. Protection granted on a temporary basis has so far mainly been practised in parts of the world that have suffered situations of mass flight. Following the Bosnian war, when European countries were faced with the first real mass flight since World War II, the question of a harmonised approach to temporary protection was placed on the agenda of the European Union. To date, EU member states have not reached any agreement on this matter, and indeed have shown great reluctance to make any commitment. As a result, when the Kosovo crisis broke out, EU countries did not know what kind of protection to grant new refugees and how to share the burden with the countries neighbouring the crisis.

'Ad hoc' solutions continue to be the rule. The criteria for temporary protection now being granted to the Kosovars vary from one asylum country to another. This also explains in part the delay in the international community's reaction to the Kosovo refugee crisis. The UNHCR has attempted to issue multilateral guidelines, but these only have the status of recommendations and do not regulate crucial issues such as the timing for the lifting of the temporary refugee status and the question of voluntary or non-voluntary return.

It is also important to keep in mind that many people in Western countries will agree to receive refugees 'temporarily' only so many times, as they come to perceive that the notion of temporary status is just an illusion to permit another burden on domestic budgets. In dealing with one refugee crisis, states should keep in mind the political contingencies of the next one.

Voluntary versus non-voluntary return

Now that the mass exodus of Albanians from Kosovo is a grim reality, the key question is whether these refugees will move back voluntarily when the war is over and the dust has settled. This report not only doubts that any massive repatriation will take place in the short-term (within the next twelve months), but also that wide-scale voluntary repatriation will take place at all.

If this negative scenario is correct, what are the alternatives? Only two: either wait and see, and take the risks associated with an additional 1.5 million people displaced in the Balkans and elsewhere; or create the conditions necessary to justify a massive organised return by introducing elements of inducement. While acknowledging the sensitivity of the issue, this report will argue for the second alternative. If the international community wants Kosovar Albanians to live in Kosovo, and wishes to demonstrate unequivocally that ethnic cleansing is unacceptable, no matter how high a price the perpetrator is willing to pay, then their return must be orchestrated14 .

Several criteria, however, must be fulfilled if induced repatriation - repatriation not initiated by the refugees themselves - is to take place:

  • essential preconditions for return must be in place at the local level (see chapter IV);
  • Standards and procedures must be developed by the UNHCR, which must be clearly mandated to undertake this task;
  • All involved states must follow the same process;
  • Returns must take place multilaterally15 , in conditions of safety and dignity and on the basis of an independent and impartial human rights assessment;
  • Regular asylum procedures must eventually be made available to provide those unwilling to return with the possibility of remaining.

It is necessary to have a discussion on time limits. How long can people who already have suffered be kept in limbo about their future? By not having strong governmental commitment on return, most time limits will be exceeded by the psychological mechanisms involved. The concept of voluntary return often serves as a recipe for inaction by states. However, by accepting the consequences of temporary protection and by introducing a comprehensive plan for the return of all refugees and IDPs, donors will find themselves under greater pressure to come up with the necessary funding for reconstruction and rehabilitation.

A comprehensive repatriation plan which ensures the return of all refugees and IDPs would represent a new paradigm in European refugee policy which could prove helpful in preventing future situations of mass exodus.

Return to Kosovo - preconditions and obstacles

The statement on Kosovo issued by the Heads of States and Governments participating in the NATO summit in Washington (23-24 April 1999) reiterates that "we will fulfil our promise to the Kosovar people that they can return to their homes and live in peace and security", and also pledges "the deployment of an international military force to safeguard the swift return of all refugees and displaced persons..."

The ambitions are clearly spelt out, both in terms of numbers: 'all', and time: 'swift'. NATO may well be able to fulfil these obligations for its own part, but what about the other preconditions that have to be fulfilled in order for people to return? And what about regional implications? Patterns of prolonged displacement elsewhere in the Balkans and the rest of the world have taught some lessons about the standards necessary for people to return to their homes - and return for good.


Refugees who consider going home invariably say that the most important condition for their return is the security situation. In strictly military terms, this may not be seen as the most important problem in Kosovo, as Kosovo will most likely be under some form of international control in the foreseeable future. But in any post-war situation there are many different security threats. Weapons, ammunition and explosives are rife, and a disorganised society creates space for all kinds of criminal acts. Milosevic's army has planted thousands of mines in strategically important areas, and mines have a strong psychological impact alongside the physical. Another security threat is the real or imagined fear of harassment, and the uncertainty of the future relationship between the Kosovar Serbs and the returnees.


Closely linked to the security question is the issue of personal documentation and the documentation of property. As has been the case in the ethnic cleansing campaigns against Serbs in Croatia and minority groups in the different territories of Bosnia-Herzegovina, the Kosovars were deprived of their personal documents before being forced out of the country. This obstacle will take a long time to remedy. To issue new documents is in itself a time-consuming and difficult job, involving the restoration of rights and assets that may be disputed. It will probably be even more time-consuming to establish necessary legal structures to issue documents. Should they be issued by FRY or a new protectorate, on a temporary or permanent basis? To illustrate the difficult nature of such tasks, the restitution of property in Bosnia-Herzegovina as prescribed in the Dayton Agreement has been underway for years, and will have to operate into the next century to process all cases presented to the Bosnia-Herzegovina Commission for Real Property Claims.


According to reports coming out from Kosovo, thousands of homes have been destroyed, looted, burnt or blown up in much the same way as seen in the former Krajina area and Eastern Slavonia in Croatia or in many places in Bosnia-Herzegovina. The outlook for Kosovo may well be similar to that of the other areas in the region: many tens of thousands of homes destroyed in the period 1992-95 in Croatia and Bosnia-Herzegovina are still in ruins - in spite of promises and massive reconstruction efforts by the states themselves and the international community. Reconstruction is unlikely to be any swifter in Kosovo. The same problems, costs donors and budget lines are at issue.


In addition to the destroyed houses, much of Kosovo's infrastructure has also been destroyed in Kosovo, due to fighting, sabotage by Milosevic's army and NATO's bombing. Once again, the experience of neighbouring Bosnia-Herzegovina and Croatia show how long it takes to rebuild infrastructure necessary to normalise daily life. In addition to having to compete for funding with Croatia and Bosnia-Herzegovina, Kosovo will also have to compete for money with the rest of FRY for reconstruction of damage inflicted by NATO bombing.

Income and social security

Even when security, documentation, shelter and infrastructure are in place, people cannot return to a place where there is no likelihood of economic survival. The economic situation in Kosovo has been extremely difficult for many years. What kind of income and living conditions for the thousands of returnees do NATO and the international community foresee when they talk about swift return? A massive investment in jobs in the corporate as well as the public sectors will be necessary.

Prospect of lasting peace and commitments

In addition to the above-mentioned problems, factors other than those based on economic and technical capacity may discourage return. One important element is the prospect of a lasting peace. If that does not prevail, few people will opt for repatriation. Sustained political resolve on the part of donors and the Western Alliance, to enforce any peace agreement and defuse threats to peace will be critical. Promises made on CNN during a military offensive are one matter; a domestic discussion on budget deficits in the parliament one year later is quite another.


There is also the problem of possible internal conflicts among the returning population. Will those who stayed on and fought the war welcome all those who fled? Will we see the necessary political consensus among the different political factions? One factor that will make return a little easier to Kosovo, compared to Croatia and many places in Bosnia-Herzegovina, is that the Kosovar Albanians will in the main return to local communities where they themselves constitute the ethnic majority. The situation will be far more complicated if Milosevic manages to repopulate Kosovo with ethnic Serbs, in the same way that his colleague Tudjman effectively undertook ethnic engineering by letting Bosnian Croats move into empty houses after the expulsion of the majority of the Serb population from Croatia.

Motivation and information

Correct, reliable, concrete and comprehensive information must be made available to refugees, IDPs and all those involved in relief and support efforts. The existence of accessible, independent media can make a decisive difference compared to a situation where IDPs, refugees and returnees are left victims of propaganda, misinformation and prejudice. Constructing an effective media and information capability is however both a politically sensitive and time-consuming task.


For obvious reasons, time is a key factor in any repatriation process. The longer people are away from home, the more they integrate and get accustomed to their new environment. It may seem that this factor is over-emphasised when one looks at repatriation taking place many years after the initial flight, but then it is important to note that repatriation after more than a few years is often induced or seen as without alternative. In some refugee situations, for example in the case of the Palestinians, people wish to keep their refugee status and their intention of going back for generations - for political reasons. This could also become the case for some of the groups of displaced persons from the Balkans.


The further away refugees have been settled from their homeland the less chance remains for voluntary repatriation. Distance creates both physical and psychological barriers to return. In addition, countries that traditionally offer humanitarian evacuation also offer local integration and more material support than the poorer countries neighbouring a conflict. The mechanism of 'phased return'16 emphasises this effect.

For the many thousand Kosovar Albanians who have been and will be evacuated to places far away from the Balkans, talk of voluntary repatriation is likely to be more rhetoric than reality17 . In addition, another possible obstacle to return occurs when neighbouring countries decide to retain refugee populations for political reasons, as is the case for the Bosnian Croats in Croatia.

Overcoming trauma

Ethnic cleansing as it is being conducted in Kosovo is a deeply traumatic experience. For many, this creates psychological wounds that can last a lifetime. Going back to the place where the atrocities occurred is extremely difficult for some people. When refugees say that they never want to go back to their place of origin, the reason is frequently trauma.

Impact on the region

The bleak prospect of a swift return to Kosovo give rise to several questions as to the regional implications of the Kosovo refugee and IDP crisis. Together with the existing refugees and IDPs, the uprooted and unsettled people represent a major source of instability for the security in the region.

Political consequences

Politically, a prolonged refugee situation will almost certainly destabilise Albania, Macedonia and Montenegro18 . In Albania, the influx of refugees has already contributed to increased polarisation of the internal political landscape. In Macedonia, non-Albanians fear negative consequences for the fragile ethnic balance where the Albanian proportion (25%) of the pre-war population has increased to 35% as a result of the arrival of refugees. In Montenegro, with its small population and extremely delicate position in relation to Belgrade, a 15-20% increase in the population might be the one extra burden that causes the downfall of the government, which is presently struggling to keep a separate line from Milosevic.

For Bosnia too, any significant influx of people is likely to contribute to destabilisation. The disorganised return of Bosnian Serbs temporarily exiled in FRY is adding to local problems. Most returnees choose to relocate to already overpopulated places within Republika Srpska, rather than to become minorities in their pre-war homes - if their homes exist at all. The humanitarian crisis in Kosovo may also fuel negative elements, as extremists operate more freely in the media shadow of Kosovo. In addition to the present critical situation in Republika Srpska, some Bosnian Croats are increasingly arguing for a third (Croatian) entity - a policy that would dramatically undermine the Dayton agreement.

On a more positive note, however, the crisis in Kosovo and the rest of FRY may also help to strengthen the integration of civil society in Bosnia-Herzegovina - both in the Federation and in Republika Srpska19 . People are frightened by developments in neighbouring FRY and may choose to actively support more positive alternatives in Bosnia-Herzegovina, and the disintegration of the FRY economy seems to promote economic integration of the two Bosnian entities.

Unlike Bosnia and FRY, Croatia has not received much critical international media coverage in recent years. This is partly due to Croatia having long been in the media shadow of Bosnia and partly also to the fact that President Tudjman's regime is seen as less of a problem than others in the Balkans20. This is despite the fact that 300,000 Croatian citizens are still refugees in FRY, and Serbs still feel forced to leave Croatia. As mentioned above, Croatia also hosts 180,000 Bosnian Croats, most of whom are still in need of durable solutions. The present regime has, in ethnic terms, managed to make Croatia one of the most homogenous states in the Balkans. Any increased refugee influx may give further ammunition to the nationalism that has so far formed the basis for Tudjman's rule .

For FRY, NATO's bombing campaign has created a dramatic situation for many civilians and the large refugee population from Bosnia-Herzegovina and Croatia. Until now, these refugees have been tolerated, however, the increasing scarcity of resources may leade to a change in popular attitudes. In addition, there is a real danger that the existing population of 200,000 Serbs in Kosovo will feel forced to flee that territory, irrespective of the design of the peace arrangements.

Humanitarian consequences

The cost of rebuilding and rehabilitating Kosovo and FRY, in addition to the damages still prevailing in Bosnia-Herzegovina and Croatia, will be many times as high as the cost of the humanitarian short-term efforts. If this increased burden on the traditional donors should lead to a policy of spreading existing funding more thinly over the whole region, the already very slow progress in Croatia and Bosnia will not only come to a standstill, but may lead to new crises. Swift reconstruction and rehabilitation is possible, however, if there is sufficient and sustained political will and commitment on the part of key members of the international community.

No matter when and how a peace settlement in Kosovo comes about, it is necessary to start planning now at the regional level for ways to deal with the massive displacements throughout the region. So far the OSCE, which has a region-wide mandate, has operated country-by-country with very little regional co-ordination. The UNHCR, which has overall responsibility for refugee protection and co-ordination of the humanitarian effort for refugees and IDPs22 , is also weak on regional co-ordination. These two structures, faced with the enormous challenge of coping with more than three million refugees and IDPs, have no time to lose in finding a strong and well co-ordinated mechanism23. A regional structure to deal with displacement is urgently needed, irrespective of the future security and political arrangements for Kosovo.

A structure based on the complementary nature of the OSCE's political mandate and the UNHCR's humanitarian mandate24 has to take the lead in co-ordinating the refugee- and IDP-related activities of all other international, governmental and non-governmental bodies in the region. In setting up such a structure, several elements should be taken into consideration, such as:

  • Loose Permanent Council resolutions will not provide what the OSCE needs. They must be tight and well defined in order to provide the OSCE with real power, rather than close association with whichever embassy appears to have the greatest influence;

  • The UNHCR cannot operate in an environment where there is no clearly articulated supportive and co-operative political agenda;

  • The authorities in FRY can be expected to play the same role as the authorities in Croatia and Bosnia-Herzegovina, where officials pay lip service to international commitments and obligations, but ignore their implementation.


The Kosovo refugee crisis is an integral part of a regional situation

The geographical vicinity, common history, common attempts to cleanse whole territories of one ethnic group and overlapping settlements all reinforce the point that the Kosovo refugee crisis is an integral part of a precarious situation faced by practically all states and territories of the Former Republic of Yugoslavia. Every action by the international community in one part of the region will, directly or indirectly, influence the situation in other parts of the region.

There will be no swift voluntary repatriation

Kosovo may, in a strict military sense, represent safe territory for returning refugees and IDPs soon after a peace settlement. The ethnic composition in places of return may also be favourable, but voluntary mass repatriation to Kosovo is unlikely to happen in the foreseeable future unless significant and determined steps are taken to tackle the obstacles to return head-on. Several alternative strategies have to be developed.

Governments must commit themselves to a comprehensive return plan

To establish durable solutions for the large number of displaced individuals; to limit further negative political consequences for the region; and to clearly demonstrate that no one gets away with ethnic cleansing, a commitment to the safe return of all refugees and IDPs from Kosovo must be made. If the temporary nature of the protection now being provided is to have any meaning beyond rhetoric, a comprehensive return plan needs to be developed and pursued vigorously.


Maintain and promote the temporary status of the refugees

No matter how long it will take to establish conditions for safe return, it is important to maintain the perspective of return for the great majority of Kosovar Albanians as the only acceptable solution to this crisis. This can be done by faithfully supporting a policy of temporary status for the refugees in exile. Tendencies so far, especially in resettlement and evacuation countries, of giving up the insistence on temporary status, and thereby the idea of return, have to be rebuffed. Otherwise Milosevic, as Tudjman and others before him, will manage to "resolve his minority question". It is also important to keep in mind that people in many Western countries will agree to receive refugees 'temporarily' only so many times, as they come to perceive that the notion of temporary status is just an illusion to permit another burden on domestic budgets. In dealing with one refugee crisis, states should always keep in mind the political contingencies of the next one.

Develop a comprehensive plan for the return of all refugees and IDPs

A comprehensive plan that ensures the safe and sustainable return of all refugees and IDPs once the necessary conditions are in place is imperative. Certain degrees of induced repatriation will need to be considered. UNHCR and OSCE have key roles to play in this endeavour. All states involved must synchronise the chosen policy. Altogether this would represent a new paradigm in European refugee policy that could also prove helpful in preventing future attempts at ethnic cleansing.

Keep the refugees in the region

For economic, political and psychological reasons, refugees should be kept in the region by alternative use of the resources which receiving countries are willing to spend domestically. Real burden-sharing mechanisms must be developed among involved states. Humanitarian evacuation should only take place for refugees who fall into one of the following categories:

  1. People in need of international protection outside the region;
  2. Necessary medical evacuation;
  3. Family reunification for children under age of 18 separated from their families.

(These criteria should be subject to a restrictive, common interpretation).

Plan for early return of refugees evacuated to third countries

Contrary to the plans for phased return under the Dayton agreement, many of the refugees from Kosovo who are temporarily settled in third countries outside the region should be the first to return25 . Due to the support they have received, they should be in a better mental and most likely also financial situation to return than many of the refugees in the countries neighbouring Kosovo. This also would free resources from domestic spending in host countries to contribute toward the reconstruction and rehabilitation of Kosovo.

Prepare for spontaneous return

Regardless of the steps the international community takes to orchestrate return of refugees and IDPs will begin to move spontaneously. They will move through and to places where they will need organised commodities like food and shelter, and some times also security protection. It is therefore important for the humanitarian agencies as well as for the peacekeepers to have contingency plans in place for these eventualities immediately after a peace agreement is reached.

Plan according to a realistic to time frame

Unrealistic assumptions regarding the time frame and effort needed to resolve this refugee and IDP crisis will backfire on all participants. Wishful thinking about swift return may lead to frustration once refugees, aid workers, host countries and donor countries realise that this will take more time than envisaged. It is better to prepare people from the outset for tough years ahead. For the donors, it is important to realise that the Kosovars will need humanitarian assistance both in exile and inside Kosovo in the foreseeable future - in addition to massive reconstruction efforts.

Finally, winter programmes for displaced Kosovars and people elsewhere in FRY have to be prepared now, in order to avoid new humanitarian catastrophes only six months down the road26 .

Keep the refugees informed

Propaganda and misinformation must be counterbalanced by offering support to national and local independent media in neighbouring states, FRY and, where possible, inside Kosovo. All displaced persons should be well informed about the prospects, plans, conditions and procedures for return. This is as important for the dispersed evacuated refugee population as for refugees and IDPs in the region.

Give equal attention to short- and long-term needs

Relief efforts have to be based on investment rather than consumption. For the planners and organisers of the humanitarian operation, it is important to start investing in more permanent or semi-permanent settlements in order to be prepared both for a possible prolonged stay in exile. This is of particular importance psychologically for the makeshift refugee communities. It is important to support the social fabric of exiled populations through schools, kindergartens and so on. This will represent a much-needed mental capital when people return to rebuild their war-torn societies. Investment in neighbouring countries will also alleviate the pressure on the local economy and help create goodwill vis-à-vis the refugees.

Involve the local population in the return process

In order to create understanding and ownership, and utilise all existing resources, the refugees and IDPs themselves, their representatives, their organisations, community leaders and local hosts should all be extensively involved in the return process, from the planning through to implementation.

Develop regional humanitarian solutions and structures

The OSCE and the UNHCR need to considerably strengthen their capacity to deal with the regional aspects and consequences of the unsolved crises in the Balkans27 . While most participants seem to agree on the need for regional political solutions and institutions to end the Kosovo crisis, it is important to apply the same logic when it comes to humanitarian efforts. Durable solutions can only be found and reached by political means. But in the meantime, a large part of the uprooted population needs humanitarian aid and support. Investment in this support will, in addition to alleviating the suffering for the individuals, also help avoid a new crisis stemming from despair and frustration among the displaced.

Mobilise up-front funding of return efforts

Humanitarian efforts inside Kosovo are necessary before a sustainable return can take place and these will need considerable up-front funding. For example, some vital commodities will need to be available to avoid new crises, especially in case of a large spontaneous return.

Include the whole region in economic recovery planning

With more than three million people in internal and external displacement, there is a tremendous need for increased economic support throughout the region. Funding to Kosovo and FRY has to come on top of existing programmes. Economic recovery plans following a peace deal for Kosovo must be prepared now and should include the whole region. Region-wide programmes will help prevent new conflicts brewing in low-priority areas. The UNHCR, as the one agency specialised in refugee questions, should provide funding in all parts of the former Yugoslavia in a coherent and transparent manner.

Keep the roles of humanitarian work and the military apart

In situations where military resources have been used as part of the humanitarian operation, it is imperative to clearly separate the roles of civil and military actors in order to stabilise war-torn societies. There are many good reasons for the use of military assets in a crisis where civil structures have too limited a capacity. But the price for this is necessarily an increased politicisation of the situation. The Kosovo crisis is a case in point, where NATO plays a key role in providing relief for the refugees while being party to the armed conflict. A massive military presence will most likely be necessary for the foreseeable future. It is therefore important to identify and support civil society organisations and structures that deal with the humanitarian priorities - in neighbouring countries and inside Kosovo.

Include binding return mechanisms in the peace agreement

In addition to the return mechanisms included in the Dayton Peace Agreement28 , any peace arrangement for Kosovo must include binding provisions, commitments and timetables for the implementation of the return process. Unclear designation of responsibility and division of labour between the international community and national/local authorities must be avoided.

Synchronise European refugee policy

In the short term, there is an urgent need for a synchronised policy with regard to burden-sharing, engagement in the region, humanitarian evacuation and both the aims and means of return. In a longer-term perspective, the momentum created by the Kosovo situation should be used to develop a much-needed pan-European refugee convention. A major reason for the lack of European preparedness in handling the Kosovo refugee crisis - in a timely and co-ordinated manner - is the lack of a harmonised legal basis for refugee protection which goes beyond the traditional instruments. The Refugee Convention of 1951 has already been supplemented in other regions of the world, Europe should be prepared to take the steps necessary to create a coherent and cohesive refugee policy. The working definition of refugee status must be expanded to cover victims of generalised violence and others in need of international protection. Furthermore, the obligation of states to grant protection according to need on a permanent or temporary basis must equally be clearly defined. To this end, the necessary mechanisms for international co-ordination must be established.


  1. 'Return' refers to all return back to place of origin, from exile as well as internal displacement. 'Repatriation' is only used for return from exile.
  2. Refugee status was never intended to be permanent. The 1951 Refugee Convention gives room for granting of international protection on a temporary basis through its 'cessation clause'.
  3. Only 20% of the refugees and IDPs had returned to their homes in Bosnia-Herzogovina 16 months after the signing of the Dayton Peace Agreement. In the same period only 10 % of damaged houses were repaired. ICG report: "Going nowhere fast", 1 May 1997.
  4. Based on UNHCR Kosovo Crisis Updates, 10 May 1999.
  5. See breakdown below.
  6. UNHCR concept paper of 12 April 1999, with inputs from IOM, UNDP, UNFPA, UNHCR, UNOCHA, WHO and WFP.
  7. It is impossible in a brief report to present all nuances; for example where different groups of refugees are being held up and used against the interest of other groups. It is therefore important to underscore already from the outset that all these groups are victims of war.
  8. Some ethnic Croatian refugees also come from FRY.
  9. According to the Humanitarian Evacuation Program.
  10. Voluntary return is the form of return prescribed in the UNHCR's mandate. It is also the basis for both the Erdut and Dayton agreements. However, the UNHCR has often been involved in repatriations where the means used to return large numbers of refugees have contained elements of inducement and force.
  11. "Concept paper on a proposed framework for return of refugees and internally displaced persons to Kosovo", UNHCR, 12 May 1999.
  12. The example most often referred to in this circumstance is the case of many Vietnamese boat people, who left for neighbouring states in hope of resettlement at the end of the 1980s, but who were eventually turned back, as the program was drastically changed in 1989.
  13. Norway is one example: 6,000 people - less than 1% of the whole caseload of Kosovo Albanian refugees - have been moved out of the region, at a cost five times as high as Norway's contribution to the humanitarian programs in the region. On top of this, the price per shelter per family is many times higher in Norway than in Macedonia or Albania.
  14. There is perhaps a parallel with the circumstances of 1991-92, when several hundred thousand Kurds were kept close to the Iraqi borders and sent home when an allegedly safe haven was created in Northern Iraq. The Kurdish population here would probably have been much smaller, and security problems in the region would have been even worse today, if this had not happened.
  15. This is not a question of getting people out of asylum countries; it is a question of making it possible for the great majority to move back to their homes. Germany's ill-prepared unilateral action to send back Bosnian refugees is an example of how things should not be done.
  16. See footnote 13.
  17. In a press briefing 30 April on the reception of 20 000 Kosovar refugees to the US, Assistant Secretary of State Julia Taft said: "When it is safe for them to return, they will be encouraged to do so and assisted in being able to. However, if they chose to stay under our laws - they may stay."
  18. See also ICG report 'War in the Balkans: Consequences of the Kosovo Conflict and Future Options for Kosovo and the region', 19 April 1999.
  19. Interview with the Deputy Head of Mission, Democratisation, OSCE in Bosnia-Herzegovina, Ms Elisabeth K. Rasmusson.
  20. Some key NATO Member States seem to team up with Tudjman in a tactical alliance to avoid too many lines of conflict in the Balkans.
  21. See also: 'Change in the Offing', ICG Report 14 December 1998.
  22. The UNHCR is given an extended mandate to also include IDPs in the Former Yugoslavia.
  23. See inter alia 'Put OSCE in Charge of Balkan Policy' by Ambassador Robert Barry, Wall Street Journal, 3 May 1999.
  24. To provide protection and assistance to refugees and, when given a particular mandate, to IDPs.
  25. Exemptions must be made for those who were evacuated because their continued stay in the region is impossible both from a short- and a long-term perspective, unless conditions preventing their return no longer exist.
  26. "The frightening thing is that unless a solution is found quickly, there will be a major humanitarian disaster in Yugoslavia. Winter will come and, with refineries gone, bridges gone, water gone, you will have a mass exodus, It will be another catastrophe." Secretary General of the UN, Kofi Annan, quoted in the International Herald Tribune, 3 May 1999.
  27. The main principles for regional solutions are already developed in the UNHCR document: 'A regional strategy for sustainable return of those displaced by conflict in the Former Yugoslavia', 9 June 1998.
  28. Annex 7 of the General Framework Agreement for Peace in Bosnia and Herzegovina.
  29. See also Vigdis Vevstad: 'Refugee Protection. A European Challenge,' Oslo 1998.