Interim Evaluation Of Reconstruction And Return Task Force (RRTF)
14 September 1999
Click here to view this document as a PDF file in A4 format. To download the PDF file to your computer, right-click here, select "Save Link/Target As..." in the contextmenu and choose a destination on your hard disk. If you are having problems opening this PDF file, please click here for help. Please note that the PDF version of this document also contains footnotes that are not used in its web version.
Executive Summary

Assisting displaced Bosnians return to their homes in areas controlled by the military of another ethnic group, so-called minority returns, has proved one of the most difficult challenges of the peace process. In the face of systematic obstructionism by local power brokers, the Office of the High Representative (OHR) and the office of the United Nations High Commissioner for Refugees (UNHCR) formed in January 1997 the Return and Reconstruction Task Force (RRTF) as an inter-agency forum to co-ordinate international efforts to this end. The RRTF brings together all agencies working on return-related issues. Following the adoption of the 1999 RRTF Action Plan, the RRTF is today managed by a secretariat within the OHR with 12 employees, with a field network of a further 25 international staff -- hired for their local expertise -- working from 6 regional and 7 field offices.

Building on the often frustrating experience of the first two years of the peace process, the RRTF developed a methodology to support minority returns which, in the course of this year, following the expansion of the RRTF field network, it has applied consistently throughout the country. This methodology consists of locating "return axes", identifying groups of displaced persons eager to return and working with them; following the flow of returns, responding to unexpected movements of displaced persons when they attempt to return home off their own bat with flexible, financial aid; negotiating "beach-head" returns, brokering deals by which returnees can establish themselves in their former communities; identifying security risks, anticipating problems so as to be able to defuse delicate situations; and information management, monitoring events on the ground closely and sharing knowledge among the international agencies.

According to the UNHCR, only 11,666 and 33,837 managed to return to their homes in 1996 and 1997 respectively, of whom only some 2,200 minorities returned to Republika Srpska. In 1998 some 35,000 minorities returned home, again almost exclusively in the Federation. Thus, by the end of 1998, the total number of persons who had been able to return to their homes in areas not under the control of their ethnic group was estimated to be around 79,500. By 30 June 1999 the total had reached 94,567 persons, giving return figures for the first six months of 1999 of only 15,000 an extremely modest result, given a projected target of more than 120,000 in the 1999 RRTF Action Plan.

A hiatus of two months as a result of the war in Kosovo, when international agencies withdrew from Republika Srpska, and slow donor response to refugee needs, in particular by the European Commission, have held up the process. Nevertheless, observation in the field suggests that the UNHCR estimate of 15,000 for the first six months of the year is low, in part, because it is calculated according to the numbers of returnees registering with local, often hostile authorities.

During 1999, accelerated returns have taken place to numerous areas previously considered inaccessible for minorities, including Prijedor, Kotor Varos, Zvornik, Teslic and Doboj in Republika Srpska, and Stolac, Prozor-Rama, Drvar, Grahovo, Glamoc and West Mostar in Croat-controlled areas. Moreover, if current trends continue, and if donor resources can be found to close the funding gaps identified by the RRTF, it is conceivable that by the end of the year 2000, the RRTF will have produced similar results in the remaining black spots, such as Eastern Republika Srpska and Posavina.

Returning minorities to vacant, destroyed areas is the easier half of the RRTF's task. The more difficult task is reintegrating minorities into town centres. These so-called contested returns involve complex legal and political issues which are not encountered in returns to empty villages. They also require successful implementation of property laws, which were only adopted, following strong pressure by the OHR in 1998, including evictions of members of the ethnic majority. To date, breakthroughs have only taken place in Sarajevo and other Bosniac-controlled areas.

Where the obstacle to the implementation of the property laws is at a high political level, as in Croat- and Serb-controlled areas, there is little that the RRTF can accomplish in the field. Instead, it is up to the OHR to take the initiative. The policy of the Banja Luka government and of HDZ authorities has so far been completely obstructionist, with less than two percent of claims resolved. Otherwise, the RRTF's techniques for supporting implementation of the property laws are still being developed. They should benefit from shifting public attitudes to the property laws. Field staff in both Entities report that displaced persons feel increasingly uncomfortable living in property belonging to others, and would like to regularise their housing status.

Although the overall number of returnees remains modest, their political significance is extremely important, perhaps indicating that local politicians are losing the capacity to resist the demands of the displaced. For the first time, there are signs of a multiplier effect, where successful returns in one area are encouraging spontaneous returns to the surrounding areas. The RRTF should therefore continue its brokering activities, trying to achieve breakthroughs in the remaining strongholds of Eastern Republika Sprska and the Posavina, and consolidating the returns which have already been achieved.

As a result of its hands-on activities and its focus on achieving clearly defined outcomes the RRTF is the most effective international field structure currently operating in Bosnia, both through its experienced staff and its inter-agency co-ordinating mechanisms. It has developed considerable expertise in brokering returns, and a substantial knowledge base on local political and social conditions effecting the returns process. As the level of both reconstruction aid and international peace-keepers declines, this expertise could become even more important as an early-warning mechanism for problems occurring in the return process.


In response to the many obstacles to refugee return encountered during the first year and a half of the Bosnian peace process, an inter-agency body, the Reconstruction and Return Task Force or RRTF, was created to oversee the process. Largely as a result of the work of this body and the efforts of its staff, the numbers of minority returnees, that is members of an ethnic group other than that controlling the territory, as well as the destinations to which they are going have steadily increased in the course of the past two years. While overall figures remain disappointing, 1999 has seen the opening of many hitherto closed areas for minority returnees. This paper seeks to assess the RRTF's performance during 1999. It examines RRTF structures, that is the origins and concepts behind the RRTF; the methodology and delivery mechanisms employed by the RRTF; relations between the RRTF and donors; and the various external constraints on operations. It evaluates the RRTF's 1999 programmes considering returns to vacant areas, so-called phase one returns, as well as returns to contested space. And it concludes with an analysis of the tasks facing the RRTF next year.

This interim assessment forms part of the European Stability Initiative (ESI) Bosnia Project and was prepared at short notice for the meeting of Political Directors of the Steering Board countries on 17 September 1999. A further paper on the results of and options for international policy in Bosnia, are currently being prepared for the Meeting of Foreign Ministers of Steering Board countries in New York on 23 September.


The RRTF was established in January 1997 by the Office of the High Representative (OHR) and the office of the United Nations High Commissioner for Refugees (UNHCR) as a forum for co-ordinating international efforts in support of the return of displaced persons to their homes. It arose from the need to link the return-related field programmes of the various international agencies with the provision of international reconstruction aid, together with the need for more focused political pressure in support of return, and better targeted use of economic conditionality. Two years later, its membership includes all the main international agencies engaged in implementation of Annex 7 of the Dayton Peace Agreement (DPA), working under the leadership of OHR. Through a well-developed field network, it represents the primary mechanism through which the international community is able to promote the return of displaced persons.

At the end of the war, there were more than a million displaced persons in Bosnia and Herzegovina (Bosnia), with further displacement occurring in 1996 following the transfer of territory around Sarajevo, and continuing campaigns of ethnic cleansing in a number of locations, including Croat-controlled West Mostar and Serb-controlled Teslic. Spontaneous returns to vacant housing reduced this figure to some 800,000 by mid-1997, where it has more or less stabilised ever since. The difficulty of accomplishing minority returns in the face of often hostile local authorities has made progress very difficult to achieve. What minority return has been achieved since 1996 -- some 90,000 returnees according to UNHCR estimates by June 1999 -- has been more than balanced by refugees repatriated to Bosnia but unable to return to their homes of origin, and therefore joining the ranks of the internally displaced.

The development of the RRTF methodology was prompted by a number of clearly unsuccessful return strategies in 1996 and 1997. Attempts to negotiate returns through high-level agreements between the leaders of the three nationalist parties, beginning with the agreement between Presidents Franjo Tudjman of Croatia and Alija Izetbegovic of Bosnia at Dayton for UNHCR-led 'pilot'return projects, failed to yield results. The return of 100 Bosniac families to Croat-controlled Stolac was achieved only in 1998, after consuming enormous UNHCR resources. Negotiating regional return plans proved equally difficult, with obstructionist authorities often including elements in the plans which were certain to fail. Second, it became apparent that reconstruction of destroyed housing would not necessarily lead to minority returns. If international agencies selected their reconstruction projects on purely technical criteria, such as degree of damage or condition of local infrastructure, the beneficiaries of the reconstruction programmes might, nonetheless, be prevented from returning by political obstruction, or refuse to accept local living conditions, leaving the reconstructed house empty or partially occupied. This led to widespread wastage of international resources, and in certain parts of the country, has left a stock of unoccupied, reconstructed dwellings. Third, early attempts to co-ordinate the application of economic conditionality to reward municipalities which were open to return, and punish those which were obstructionist, proved difficult to implement. Various schemes, beginning with the UNHCR 'target areas' and culminating in the 'Open Cities'programme of 1997, were largely unsuccessful. Such was the level of obstruction of minority return across the country during that period that it was difficult to distinguish between local regions. These co-ordinated initiatives were not supported by the donors, with the result that there was established a form of 'market-place' in international aid, where local authorities facing the loss of a project could seek to replace the funds from other sources.

These lessons led to several basic elements of the RRTF methodology. First, because obstruction to minority return tends to be concentrated at the local level, breakthroughs must be achieved by political pressure and brokering agreements at the local level, by individuals familiar with the local situation. Second, in order to use international reconstruction assistance effectively, the money must 'follow the flow'of returns, focusing on areas where displaced persons have already demonstrated that they have the will to return. The return process is only feasible where it is a result of genuine commitment by the displaced persons themselves as the primary actors. Thus, the RRTF directs international resources into areas where return is already occurring, to support the process and maximise the number of returns that can be achieved. This led the RRTF to promote two modes of reconstruction assistance. The first is the 'return axis'concept, which is the predictive element in RRTF planning, directed at donors such as the European Commission who work with longer planning cycles. It involves identifying areas where a process of return is pending, based upon the expressed wishes of displaced persons communities. The second mode is flexible funding available for quick disbursement, which is used to support unexpected return movements. Third, the RRTF has introduced more subtlety into the use of conditionality, using local infrastructure projects and the threat of suspension of smaller projects as a negotiating tactic. The effectiveness of this is dependent on the flexibility of various donors and implementers.

RRTF methods also developed as a result of certain key successes in 1997. One example took place in the Croat-controlled municipality of Jajce in Central Bosnia in the summer of 1997, when a few hundred Bosniac returnees were driven out through orchestrated violence. SFOR troops were immediately deployed to provide area security so that the victims of violence could return within days, breaking for the first time the reluctance of the peace-keeping mission to become involved in refugee return, and showing that the international forces could perform effectively when faced with organised violence. Secondly, the OHR played a robust political role in ensuring not only that the violence did not continue, but that municipal leaders throughout Central Bosnia had to indicate areas in which they would ensure security for immediate minority returns. This combination of strong political intervention and area security proved very successful and within weeks the return process to Jajce had resumed and surpassed its mid-1997 levels. There were also important breakthroughs achieved by the Banja Luka-based North-Western RRTF, which led the way on close inter-agency co-ordination in the field, and successfully supported returns to key areas -- the most notable of which was the return of nearly 2,000 Serbs to Croat-controlled Drvar municipality during the winter 1997/98. The Banja Luka RRTF also adopted a clear approach towards obstructionist authorities. While they were informed of planned return initiatives, and every effort was made to include them in the preparations, they were told unmistakably that they did not hold veto power on returns.

These lessons showed that achieving returns was, above all, a political process, rather than technical or humanitarian, in which strong political leadership by the OHR could make a substantial contribution.


By the middle of 1998, the RRTF had turned this experience into a general methodology for supporting minority return. However, it was only in 1999, following the expansion of the RRTF field network, that this methodology began to be applied consistently throughout the country. This methodology includes the following elements:

  • Locating return axes: The RRTF identifies where there is a genuine desire on the part of displaced persons communities to return to their homes of origin. This occurs most commonly where displaced persons from a specific area have remained together during their displacement, and can express the desire to return as a group. Identifying return axes depends on working closely with displaced persons communities.
  • Following the flow of returns: RRTF identifies wherever return movements are already occurring, and advises donors to concentrate resources to support and maximise these return movements. This often involves identifying funding gaps, such as where returnees have repaired their own housing, but require international assistance in repairing infrastructure. The RRTF then draws up 'funding-gap tables'(RRTF Mostar) or 'needs overviews'(RRTF Travnik) which are regularly up-dated.
  • Negotiating 'beach-head'returns: Where return potential has been identified, the RRTF works to begin the process by negotiating agreements with local authorities for the first returns to take place. Once initial returns have been brokered, this may clear the way for subsequent returns with reduced international involvement. Promoting the first returns may involve assisting the displaced persons themselves to organise and appoint representatives, and convening and attending the first meetings with local authorities. Often strong political leadership from the OHR applied over a significant period of time is needed to create the first breakthrough. In order for this form of negotiating to be successful, the RRTF officers require deep understanding of the issues at stake, and must ally themselves closely with the displaced persons themselves. At the same time, they must also try to win the confidence of displaced person in the receiving communities.
  • Identifying security risks: Potential violence against returnees is identified through maintaining contact with returnees, municipal authorities and local police, and wherever possible prevented through timely political pressure, and close co-ordination with IPTF and SFOR. At times, this requires postponing return movements while tensions are calmed, and ensuring that they do not occur in a provocative manner.
  • Information management: All of these activities require substantial information on conditions in the field concerning return movements; distribution of displaced person; damage to housing and infrastructure; local political developments; and past international programmes. The RRTF provides the best forum where the field staff of the various international organisations can share information and thus build institutional memory and expertise.

The RRTF Action Plan for 1999 identifies three main pre-conditions for minority return, to which the efforts of the RRTF are to be directed: space (making housing available for return, through reconstruction of damaged property and implementation of the property laws); sustainability (ensuring that returnees have access to educational facilities and the necessary minimum of income and social services to make family life in the return destination viable); and security (preventing threats and violence from being used to prevent returns). While the 1999 Action plan is strong in its operational plans and the description of the delivery mechanism these require it is not wholly convincing in its policy considerations and proposed remedies. In particular, it does not make clear that minority return to empty, destroyed housing is a very different proposition to minority return to occupied housing. In the former case, the local authorities may be required to do little more than passively accept reconstruction and return. In the latter case, by contrast, the authorities are required to undertake the politically difficult task of evicting the current occupants of claimed property, in favour of members of ethnic minorities. The latter is therefore more akin to rule of law and institution-building programmes. Most of the techniques described above are designed for the former mode of return. Because of its importance to the return process, RRTF has been drawn into the implementation of the property laws. The Action Plan also fails to set out clearly the extent and nature of RRTF involvement in the difficult area of sustainability of returns, with the result that RRTF officers have become involved in complex areas such as reintegrating the education systems and addressing ethnic discrimination in employment. They bring to these areas an essential understanding of local conditions few other international officials have, but since strategy decisions on these issues are made outside the RRTF framework there has been no consistent methodology developed to deal with sustainability issues.


Although the RRTF was initiated as an inter-agency co-ordinating mechanism, it is chaired and directed by Deputy High Representative Andy Bearpark, and has become in practice and theory (first clearly spelled out in the 1999 Action Plan) an instrument of OHR policy. The effectiveness of the RRTF has increased as the High Representative has emerged as the strongest political player among the international community, and most of the agencies working within the RRTF are willing to submit to the OHR's political guidance on most issues. That said, the RRTF would not function on the local level without the expertise of the OSCE and UNHCR field officers, the technical support provided by the International Management Group and the close linkages to both SFOR and IPTF and to NGOs working in the field.

The RRTF is managed by a Secretariat in OHR Sarajevo, with 12 employees, and a field network of a further 25 RRTF international staff working from 6 regional and 7 field offices. Almost all of its staff have had prior experience of fieldwork in Bosnia working for the OSCE, the UNHCR or SFOR and have been hired because of their specific local expertise. This field network remains small and tightly managed, capable of higher levels of consistency and professionalism than any other international field operations in Bosnia.

According to the 1999 Action Plan, the regional offices are the key level for co-ordination of planning and implementation, meeting with member agencies on a monthly basis. The agencies work together on identifying and resolving the practical obstacles to return, identifying for example the dividing line between bureaucratic incompetence and political obstruction, and developing appropriate intervention strategies. Information sharing is a crucial element of the RRTF co-ordination strategy, as differences in approach among the different member organisations often result from discrepancies in information. Local RRTF offices are located according to the key return axes, and have the task of identifying return movements, both actual and pending, and conducting negotiations with local officials and displaced persons leaders.

While field co-ordination mechanisms have developed well in 1999, donor co-ordination has been less successful. The RRTF acts as a information service for donors, alerting them to existing return movements and trying to securing funding for key projects to support those movements. A number of key donors, notably the European Commission, have built the RRTF axes concept into their planning at the beginning of the funding cycle. However in 1999, the European Commission's funding, representing a substantial part of the RRTF's foreseen reconstruction work, is yet to be paid. The extensive delay to planned projects has meant that, although 'beach-head' returns have been achieved in 1999, opportunities have been missed and the numbers of returns have been less that expected.

In addition, with the exception of the Netherlands, donors have generally failed to respond to RRTF proposals for flexible funding sources for quick disbursement. In 1998, the Dutch Government made funds available for quick release through its Embassy, and was able to support effectively minority return breakthroughs, in particular in Central Bosnia, on short notice. The RRTF asked in its Action Plan for 10 million German Marks of quick disbursement funds in 1999, but none has been forthcoming. RRTF officers have been frustrated by their inability to provide already promised reconstruction support as well as aid to spontaneous return movements, thus following up on their success in opening new areas to negotiated return. In light of its original aims, the lack of co-ordinated reconstruction works must be judged an important failure of the RRTF mechanism, although not one which is attributable to the secretariat itself.

a) The 1999 Return Scorecard

Bosnia had a pre-war population of 4.3 million. According to UNHCR, 43 months of war resulted in over 1 million people being displaced within Bosnia, and more than 1.2 million seeking refuge abroad. UNHCR estimated that, at the end of 1997, there were 816,000 Bosnians displaced internally (366,000 in the Federation and 450,000 in Republika Srpska), plus 40,000 Serb refugees from Croatia in Republika Srpska.

In the first two years the international community made little effort to promote minority returns, resulting in only 11,666 returns in 1996 and 33,837 in 1997. In these two years only some 2,200 minorities returned to Republika Srpska. 1998 saw about 35,000 minority returns, again almost exclusively in the Federation. Thus, by the end of 1998, the total number of persons who had been able to return to their homes in areas not under the control of their ethnic group was estimated to be around 79,500.

According to UNHCR figures again, the total number of minority returnees by 30 June 1999 reached 94,567 persons, giving return figures for the first six months of 1999 of only 15,000. That is an extremely modest result, given the projected target of more than 120,000 returnees which the 1999 RRTF Action Plan had foreseen for the whole of 1999. The RRTF's own estimates, on the other hand, suggest that the figure is substantially higher.

There are several possible explanations for this discrepancy. The basis for UNHCR numbers are registrations of returnees with municipal administrations. With the vast majority of minority return movements taking place to vacant, destroyed housing, into areas under the control of often hostile administrations, the number of individuals who register is small. The more contentious the areas targeted by returnees, the less likely that returns will be registered. In addition, there are discrepancies in criteria as to what counts as a return. The RRTF figures, for instance, may include heads of households who have returned to commence reconstruction work, even if their families remain in displacement.

For example, according to Republika Srpska government numbers, the number of returnees to Zvornik stands at 1,000. According to the estimates of the RRTF field officers, by 10 September "more than 2,000 Bosniacs were cleaning destroyed houses and/or staying overnight in a total of 10 local communities and 26 villages in Zvornik municipalities."

While ESI has not undertaken an independent investigation, it does appear from observation in the field that the UNHCR estimate of 15,000 for the first six months of the year is too low. The figure for 1999, especially taking into account the two-month delay caused by the withdrawal of all international organisations from Republika Srpska caused by the Kosovo crisis, are a substantial improvement on 1998, but still -- for the first six months of 1999 -- well short of the predictions of the Action Plan.

b) 1999 - Opening up Closed Areas for Returns

The successes for 1999 can best be understood by analysing the areas to which minorities have begun returning, and those which remain closed to minority returns. During 1999, returns to numerous areas previously considered inaccessible for minorities took place. All over the country areas run by hard-liners opened up. In others, building on previous beach-hear returns, they accelerated. In Republika Srpska the West Bank area in Prijedor, Kotor Varos, the Zvornik area, Teslic and Doboj (including the hills overlooking the town); in Herzegovina Stolac, Prozor-Rama, Drvar, Grahovo, Glamoc, West Mostar and Serb-controlled Eastern Herzegovina; in Una-Sana Canton Bosanski Petrovac; new return movements, including outward movements of Serbs into Bugojno, Travnik and Croat-controlled Herzegovina, accelerated.

In Prijedor, returns of Bosniacs throughout 1998 had focused on the area of Kozarac, with a lot of international pressure required to enable reconstruction to begin. By the summer of 1999, some 90 families had returned to Kozarac. However, by that time the geographical barrier to returns had fallen, and spontaneous minority returns of Bosniacs took place to 11 villages, with estimates of up to 700 people sleeping each night in vacant, damaged houses.

In Kotor Varos, as late as this summer, returns of Bosniacs to many villages had met bitter resistance. Then one of the most difficult villages, Vecici, saw a successful Bosniac return on 4 August 1999. Within a month, about 800 Bosniacs had visited the area, more than had visited the area in all the time since the end of the war. One group of 80 Bosniacs, visiting the village of Vranic in August to spend the week-end cleaning houses, ended up eating lamb with their Serb neighbours.

In Teslic municipality, the first assessment visit of Croat displaced to the area of Komusina passed well in February 1999. By April 1999, a date had been set to start with house-cleaning activities. Once Croats from Zepce started to go to Komusina to do this, Bosniac displaced from the neighbouring village of Kamenica began to follow suit. What makes this remarkable is that as late as 1996 the SDS/SRS leadership there, known as one of the most hard-line in Bosnia, had expelled Bosniacs from their homes, following a horrendous war-time history of atrocities and ethnic cleansing. Indeed, merely to discuss the issue of a visit of 200 Croat pilgrims to the Komusina parish in August 1998 required OHR intervention with the national leadership of the SDS and the Interior Minister in Banja Luka. The prospect of so many Croats visiting Komusina was considered too provocative and dangerous only one year ago.

In some areas, returns, house-cleaning and the beginning of reconstruction activities have started in all potential return villages within the municipality. The final Phase One Return of Bosniacs to Prozor-Rama (settlements Varvara, Ripci and Hamici) took place on 3 August 1999. The final Phase One return to Stolac occurred on 2 July 1999 (settlements Gorica and Kraisjna). In many areas, the 'bridge-head'function of Phase One returns has thus come to an end. This is already the case in the municipalities of Prozor-Rama, in Stolac, in Vitez and indeed in most Bosniac-controlled municipalities in Central Bosnia. In Herzegovina-Neretva Canton and in North-Western Republika Srpska Bosniac returnees had started to settle in almost every area. In the Canton 10 municipalities Drvar, Grahovo and Glamoc Serb returns to vacant housing accelerated further during 1999, with an estimated 2,000 out of a pre-war population of 7,000 Serbs back in Bosansko Grahovo.

However, some areas with significant return potential remain closed. The most significant omissions from this year's map of breakthroughs are Posavina (Brod, Derventa and Modrica), Capljina, and parts of the Drina Valley, from Visegrad and Bratunac to Foca. These areas will continue to require sustained focus to generate similar results. In Eastern Republika Srpska there are already first signs of cautious opening.

None of these returns could have happened without the support and involvement of the international community. And all of them ended up transforming the general atmosphere. In Kotor Varos, the Serb Mayor announced that a special day would need to be organised on 5 September 1999 to welcome the population back. "We want to do everything so that these people do not feel any difficulty due to the fact that they had been away for some years," he announced on local radio B-99. In Prozor-Rama, the further return of Bosniacs to the Croat-controlled villages of Lapsung, Druzinov and Slimac in early July was accompanied by a Franciscan priest (attending at the request of UNHCR and OSCE) preaching reconciliation and the need for Croats to leave homes they occupied illegally. During a meeting on 1 June 1999, the SDS Mayor of Doboj told a Western ambassador that returns to Doboj were going to happen "whether the politicians liked it or not".

The success of Phase One returns even in hard-line areas represents a continuing process. It has led to mounting pressure from other displaced persons to begin Phase Two returns to homes which are presently occupied. As a report from the Mostar Regional RRTF from September 1999 notes: "Bosniac displaced persons returns to homogeneous areas in this region are not exhausted, and more and more returns to mixed areas are planned... The newest planned return movements include people whose houses are occupied and who nevertheless wish to return to collective centres in their pre-war settlements." Returns scheduled for later in September will also include more movements of Bosniacs into areas largely occupied by displaced Croats (for instance in Pocitelj). This highlights the growing pressure on the international community to move from returns to vacant spaces to successful implementation of property laws, beginning with effective double occupancy evictions. It is a pressure based on a growing sense among displaced persons in the regions where Phase One returns have peaked that the promise of Annex 7 might actually be achievable.

These successful returns have also begun to undermine the credibility of the assertions by the dominating nationalist Croat and Serb parties that only Bosniacs are interested in returns. In fact, the return of Serbs from Eastern Herzegovina to the Neretva valley and of Croats from Zepce to Komusina in Teslic, happened against the explicit policy of both the SDS and the HDZ. As late as this spring, HDZ politicians tried to discourage Croats from returning to Teslic instead of waiting for a "revision of the IEBL", intimidating displaced persons with house-to-house visits. It was only when the spontaneous return movement, supported by the Catholic Church and the opposition Nova Hrvatska Inicijativa (NHI), crossed the point of no return, that HDZ president Ante Jelavic hastened to add his blessing and promise a contribution of building material for an initial five houses.

c) Future Directions for Phase One Returns

The completion of the task the RRTF set for itself in facilitating these types of return movements has come within sight. If things continue as they have until now, and if donor resources can be found to close the funding gaps identified by the RRTF, then it is conceivable that by the end of the year 2000, the RRTF objective, set in 1997 for parts of the country and then expanded to all of Bosnia, of returns to all vacant areas can be achieved. In the wake of breakthroughs in 1999 in Prijedor, Doboj, Kotor Varos and other hard-line areas, there is no reason to believe that the RRTF cannot produce the same results in Eastern Republika Srpska, in Foca and Bratunac, or in the Posavina, if the right political and security resources are applied.

Clearly, there is still a long way to go. The RRTF has developed methods for brokering returns to empty, destroyed housing even in areas governed by political forces opposed to the return of other ethnic groups. However, although the original breakthroughs in Drvar and Jajce took place in 1997 and 1998, the RRTF has failed to achieve returns to contested, i.e. occupied housing in the town centres. Moreover, this remains the case today, with returns to Drvar and Jajce continuing into vacant areas, but not into the towns.

This leaves two clearly defined tasks to the RRTF to complete in 2000: to unlock returns to the last closed areas of the country and to maximise the use of international reconstruction funding in areas to which returns have already taken place. With the prospect of reduced funding for reconstruction, the efficient allocation and co-ordination of what remains is even more crucial.

The coming year is also likely see cuts in the number of peace-keepers deployed in Bosnia. The impending reduction of SFOR will have serious implications for the RRTF. In particular, it will increase demand for on the organisation's information-gathering and political-brokering functions, to identify and defuse potential conflicts.

a) Introduction

By contrast to return programmes to empty, destroyed villages, return to occupied housing has been disappointing, and the RRTF is yet to develop an effective method for breaking the deadlock. There have been isolated successes, notably in Sarajevo Canton and other Bosniac-controlled municipalities. In other places, however, particularly in Croat-controlled Herzegovina, Croat enclaves in Central Bosnia, and much of Republika Srpska, the process has not yet begun. The development of local administrative structures for managing housing is an extremely important part of the refugee return process. Progress must therefore be measured in terms of institution building, as well as in numbers of returns.

Return to contested space requires implementation of the property laws in both Entities, which create administrative claims processes for refugees to repossess abandoned houses and apartments. These laws were adopted only under strong OHR pressure after long delays (April 1998 for the Federation; December 1998 for Republika Srpska). Return through this mechanism raises complex legal and political issues which are not encountered in returns to empty villages. It involves the eviction of members of the majority ethnic group in favour of returning minorities, requiring local authorities to disregard ethnic allegiances in the neutral application of the law. It involves addressing multiple occupancy, upgrading of apartments, divided families, and other forms of misallocation of housing, often targeting the politicians who control local administrative structures. It involves identifying alternative accommodation for displaced persons required to leave their temporary housing, but not yet able to return to their homes. It involves restoring legal systems and systems for managing public housing which have suffered both from political manipulation and loss of capacity during the conflict. The mode of return therefore requires different activities from those described in the previous section.

During 1999, the RRTF has experimented with a number of field implementation mechanisms. However, no consistent mode of operation has yet been developed, although it is not too early to begin drawing general conclusions.

b) Developments During 1999

Most of the progress in 1999 has been in legislative and administrative matters. The High Representative has used his powers liberally to accomplish a number of key reforms, including:

  • cancelling the illegal reallocation of apartments since the Dayton Peace Agreement came into force;
  • setting aside court decisions which cancelled occupancy rights to apartments;
  • narrowing the criteria for access to humanitarian accommodation, to minimise abuse; and
  • placing unclaimed apartments under municipal administration for use as temporary accommodation.

In the Federation, the administrative mechanisms and housing management systems established in 1998 have been gradually strengthened, particularly in Bosniac areas. In Republika Srpska, implementation of the property laws began only at the beginning of 1999. The Ministry for Refugees and Displaced Persons has established a network of municipal offices for processing property claims, although these remain severely under-resourced.

The claims process has three stages: the registration of return claims; the production of decisions confirming the right to return; and the reinstatement of the claimant into the property, which requires the eviction of the current occupant, and where appropriate the provision of alternative accommodation. The process is complex and bureaucratic, and easily subject to delays and obstruction by local officials. The following table gives approximate figures on progress in each of the two Entities:




Positive decisions


% of claims resolved






















Republika Srpska

























Total for BiH






Registration of claims is proceeding at a good rate, with the majority of claims already registered in the Federation, and the process well underway in Republika Srpska. Issuing of decisions on the claims, however, is running considerably behind (15 per cent of registered claims). This figure hides substantial regional variations: in HDZ-controlled municipalities, a number of housing authorities have failed to issue a single decision recognising the right to return. Reinstatements are proceeding even more slowly, and forced eviction of current occupants remains rare.

Analysed regionally, the figures send a very clear message. The most successful region is Sarajevo Canton, which dominates figures on decisions and reinstatements. Most of the implementation has occurred in Bosniac-controlled areas, including Una-Sana Canton, Bosniac-controlled parts of Central Bosnia Canton and Zenica-Doboj Canton. In both Croat- and Serb-controlled areas, the process cannot be said to be functioning effectively.

Field studies indicate that the problem is still one of direct political influence on the work of the administrative authorities. In Bosniac-controlled areas, although implementation is happening, it is progressing at a rate conditioned by international pressure and power-struggles between moderates and hard-liners within the SDA-led ruling Coalition. In Croat areas, the HDZ is explicitly preventing administrative authorities from taking any steps towards implementation. In Stolac municipality not a single property decision has been issued. In Capljina, not even the appropriate department has been established. HDZ officials in Mostar are open about their determination not to see any evictions. At the same time, they refuse action even against proven double-occupants, as in the case of Croats whose houses have been reconstructed in other municipalities but who continue to live in Mostar.

In Serb areas, political manipulation takes a different form depending on the region. The municipal structure of the Ministry of Refugees and Displaced Persons has not been given nearly enough resources to implement the property laws, and the government makes no effort to remove obvious obstacles to the process. This enables the government of Milorad Dodik to attribute its poor record on implementation to financial or organisational problems. Indeed, ESI was told by the responsible Minister that Republika Srpska's 1999 budget did not foresee budget items such as postage to mail property decisions to Bosniac claimants, fees for locksmiths to carry out evictions or indeed petrol for municipal commission members investigating the status of property for the Ministry.

The duplicity of the Dodik government, which promised "70,000 minority returns in 1998" when it came to power, becomes apparent in analysing the 'floater issue'. Floaters are members of ethnic minorities who have been forced out of their apartments in Banja Luka, although they have never left Republika Srpska. The handful of reinstatements which have taken place have been the result of an exhausting case by case fight by international organisations, not a result of a due legal and administrative process. The resolution of the floaters issue featured at the Banja Luka Return Conference in April 1998 which affirmed that this was a conditio sine qua non for substantial financial assistance to Banja Luka. Instead, the Republika Srpska National Assembly issued a resolution in 1998, with the votes of Dodik's ruling SLOGA coalition, ordering a halt to the execution of Republika Srpska court orders to evict the illegal occupants. More than 300 cases are still pending, and most of 1999 has seen no progress.

c) The RRTF's Contribution

During 1999, RRTF field officers have become increasingly engaged in promoting implementation of the property laws. Local political pressure and deal-brokering is less successful in this mode of return, particularly in Croat- and Serb-controlled areas where the larger political climate is unfavourable. Where resistance takes the form of bureaucratic obstruction, strong political pressure can successfully halt improper practices, but in the end may merely move the obstruction to another point in the process (in Sarajevo this recently took the subtle form of reducing the numbers of municipal inspectors). As a result, field strategy has shifted towards establishing monitoring mechanisms that can follow the entire process, from claim registration through identification of multiple occupancy and alternative accommodation to the point of evictions.

Because of the difficulty of securing evictions of current occupants who have no other accommodation to go to, RRTF strategy has focused on achieving secondary returns, i.e. returns to properties vacated by other displaced persons due to their return to a reconstructed home. There has also been a special emphasis on eliminating multiple occupancy (where families are in possession of more than one housing unit). The former involves identifying where displaced persons have regained possession of their homes, and ensuring that all members of their family vacate their temporary accommodation, making it available for the return of the original owner. Here the information network of the RRTF, tracking returns and locating absent owners, is critical, often ensuring that reconstruction of a damaged property leads to further consequential returns. These cases are a logical point to begin implementation in reluctant municipalities, as there is no humanitarian or political argument to set against the requirements of the law.

Multiple occupancy is a deeper-rooted problem, which requires strong local political pressure to resolve. While most municipalities are committed publicly to addressing this problem, the responsible administrative bodies often lack the authority to move against powerful local figures. The RRTF has established multiple-occupancy committees in a number of municipalities, including Travnik, Zenica, Tuzla and Bijeljina, which provide a forum where the responsible local officials meet regularly with international field officers, identifying multiple-occupancy cases and co-ordinating action. The multiple occupancy committees provide a combination of sustained international pressure and political cover to help officials carry out their responsibilities. Although progress continues to be slow, the political benefits can be substantial, helping to place the return process in the context of restoration of the rule of law, rather than in its ethnic dimension.

The most advanced field model used by the RRTF is the Sarajevo Housing Committee (SHC) which was a product of the Sarajevo Declaration of February 1998, and has achieved a much higher rate of property law implementation in Sarajevo than elsewhere in the country. The SHC works at the Cantonal level, providing a forum where local officials meet regularly with representatives of the international organisations. The SHC has co-ordinated strong political pressure on the Sarajevo Cantonal Government to ensure that the housing authority is staffed by co-operative officials, and to improve its management systems. It provides a forum where the local authorities responsible for different parts of the process -- claims processing; reconstructing damaged apartments; finding alternative accommodation; policing evictions -- share information with each other and the international community. The idea is to encourage the local authorities to perform their functions as efficiently as possible, while protecting them from the political repercussions of their work. The SHC has achieved an average rate of 30 evictions per week during most of 1999. It has taken care to limit the political repercussions of the evictions by targeting multiple occupants and divided families, and by reassuring the public that displaced persons who are not able to return -- particularly Bosniacs from Eastern Republika Srpska -- will not be evicted unless they are offered alternative accommodation.

d) Future Directions for Returns to Contested Space

Returns to contested space is the most difficult aspect of the return process, and progress is only just beginning to materialise. Although difficult to achieve, it provides the vital institution-building and rule of law element to the return process, which should continue to produce results after international supervision is withdrawn.

Analysis suggests certain priorities for the coming period. First, where the obstacle to the implementation of the property laws is at a high political level, as in Croat- and Serb-controlled areas, there is little that the RRTF can accomplish in the field. Its information networks can simply offer detailed accounts of how political obstruction occurs in Herzegovina, and of the poor design and under-resourcing of administrative processes in Republika Srpska. This information must be used by the OHR to develop a political strategy to clear the deadlocks.

Second, the RRTF's techniques for supporting the implementation of the property laws are still being developed. Unlike return to empty property, which is now in its fourth year, the legal conditions for return to occupied property are a recent development. The key elements of the OHR's property law reform have been in place only a few months. While the RRTF has correctly identified a number of starting points, it is still in the process of experimenting with programmes and field structures. As part of its planning process, the RRTF should look carefully at the experiments undertaken in 1999, with a view to developing models that can be applied consistently in different regions. In particular, it should examine how to export the SHC structure to areas where the political conditions are open to return, to maximise the results. To avoid wasting resources, these efforts should follow a similar philosophy to the RRTF's reconstruction aid strategy, following return movements that have already begun to occur.

Third, the successes which have been achieved in this area are in part due to a change in public perception of the property laws and returns' process. In the past, the arbitrary exercise of public power and widespread discrimination led many individuals to believe that they would never be called to account for taking property that did not belong to them, contributing to the breakdown in the rule of law. Slowly but steadily, that attitude is being eroded and field staff in both Entities are reporting that displaced persons feel increasingly uncomfortable living in property belonging to others, and would like to regularise their housing status. The RRTF should try to maximise this change in public attitudes by shifting the focus of its pressure away from the inter-ethnic element (reversing ethnic cleansing) towards the rule of law element (eliminating misuse of housing). For example, the SHC has so far avoided eviction of Srebrenica displaced persons, but has cracked down on multiple occupancy even among protected groups such as demobilised soldiers. By reassuring vulnerable displaced persons that they will not be left without accommodation, but being strictly impartial in application of the law in other circumstances, the RRTF can help build a consensus among officials and the general public behind restoring the rule of law in property.


Within the scope of this paper, it is not possible to make detailed recommendations as to how the RRTF should set priorities and develop strategies for the coming year. However, ESI can offer the following general conclusions about the successes of the RRTF, and its importance for the coming period.

The RRTF is yet to achieve a dramatic breakthrough in the numbers of minority returns for a number of reasons, including delays in the disbursement of major reconstruction funding. Even allowing for under-estimation of unregistered returns, however, the figures for the first half of 1999 suggest that the goal of 120,000 minority returns for the year will not be reached.

Nonetheless, the RRTF has achieved significant breakthroughs in 1999 in returns to key parts of Republika Srpska and Herzegovina, as well as important progress in Sarajevo. Although the number of returnees to these places remains modest, their political significance is extremely important, perhaps indicating for the first time that local politicians are losing the capacity to resist the demands of the displaced person population. For the first time, there are signs of a multiplier effect, where successful returns in one area are encouraging spontaneous returns to the surrounding area. If the return process is to continue into the future, then these dynamics should be nurtured as far as possible. The RRTF should continue with its brokering activities, trying to achieve breakthroughs in the remaining strongholds of Eastern Republika Sprska and the Posavina, and consolidating the returns which have already been achieved.

Without doubt, the RRTF is the most effective international field structure currently operating in Bosnia, both through its experienced staff and its inter-agency co-ordinating mechanisms. It has developed considerable expertise in brokering returns, and a substantial knowledge base on local political and social conditions effecting the returns process. It offers an important service to donors and reconstruction agencies, enabling them to target their activities efficiently. Moreover, it is worth noting that the RRTF field structure has only become fully effective during the course of 1999, and is still developing. The full benefits of its efforts to date may yet be seen.

The more the flow of international reconstruction aid and the level of international troops is reduced in the future, the more important it will be to have an effective co-ordinating and early-warning mechanism for problems occurring in the return process. To date, donors have failed to make full use of the RRTF structure to maximise the return on their investments in Bosnia. As the scale of the reconstruction effort declines, donors may be more willing to offer the kind of flexible approach to funding which the RRTF has advocated.

Sarajevo, 14 September 1999
Click here to view this document as a PDF file in A4 format. To download the PDF file to your computer, right-click here, select "Save Link/Target As..." in the contextmenu and choose a destination on your hard disk. If you are having problems opening this PDF file, please click here for help. Please note that the PDF version of this document also contains footnotes that are not used in its web version.