Review of the CIREFCA Process

EVAL/CIREF/14

SCOPE OF THE REVIEW

(1) This review examines the role of the CIREFCA process in providing durable solutions to the plight of refugees and other displaced people in Central America. There were three objectives to the review: to analyze CIREFCA as a regional model for solving the problems of uprootedness; to explore the institutional relationship between UNHCR and UNDP in the framework established for the CIREFCA process; and to assess the impact of the process at the political and project levels.

(2) The CIREFCA process was evaluated on the basis of five criteria, representing the various facets of the process:

• as a model for regional solutions for the uprooted;

• as an example of inter-agency collaboration within the United Nations system;

• as an interlocutor between the capabilities and needs of the uprooted and the resources and solidarity of the national and international community;

• as a vehicle to promote protection of the uprooted; and,

• as a facilitator of solutions for the uprooted.

(3) The evaluation was funded by UNHCR and the European Community, and undertaken by the Refugee Policy Group. UNDP officially indicated its partnership in the evaluation once it was underway. The evaluation was led by Mary Am Larkin, in collaboration with Marie-Chantal Barre and Elizabeth Hayek-Weinmann. The evaluation team was assisted by Kate Lawler.

(4) The time period of the evaluation begins with the 1989 conference and ends in June 1993. Fieldwork for the review was undertaken in each of the seven CIREFCA countries between March and June 1993. Discussions were also held in Geneva, New York, Brussels, Stockholm and Washington DC. Over 200 interviews were held with representatives of UNHCR, UNDP, host governments, donor states and NGOs, as well as scholars and beneficiaries The evaluation team also conducted a detailed examination of CIREFCA conference documents and other relevant literature. Some updated information received prior to finalizing the report is footnoted.

(5) It is impossible in a review of this type to do justice to the full complexity of the CIREFCA process, the differences among country programmes and conditions, and the changes in actions and perceptions of participants over time. The report does, however, reflect the overall strengths and weaknesses of the process, indicates lessons learned, and includes recommendations for the final year of the process in Central America and for replication elsewhere.

SUMMARY

FINDING DURABLE SOLUTIONS FOR UPROOTED POPULATIONS

(6) The International Conference on Central American Refugees, known by its Spanish acronym CIREFCA, was held in May 1989, culminating attempts over the previous decade to resolve the regional crisis of refugees, displaced persons and returnees. Approximately two minion people were uprooted by the civil conflicts which raged over that time and affected in one way or another each country in the region. CIREFCA's purpose was to seek a durable solution for the problems of refugees, returnees and displaced persons within the framework of social and economic development in the region.

(7) With the coordination of UNHCR and UNDP, and the resources of the international community, the governments of the region committed themselves to the goal of linking emergency assistance to on-going development plans. There was an understanding in CIREFCA that truly durable solutions to the crisis of displacement were a key to the restoration of lasting peace. In this way, CIREFCA went from a conference to a process, inextricably linked to the ongoing peace process in the region. CIREFCA, intended to be a three-year endeavour, was soon extended by two years owing to the complexities of programme development and implementation.

(8) The CIREFCA process reflected both the political win of the Central American governments to find durable solutions to uprooted populations, and the policy commitment of UNHCR and UNDP to work collaboratively toward that same goal. CIREFCA became exceedingly complex, with differences in almost every aspect of the process among the seven participating nations. The process also has been in constant evolution as governments changed, conflicts subsided, and refugees repatriated.

ESTABLISHING A COMPREHENSIVE APPROACH

(9) While there were many obstacles to translating this conceptual framework into reality, CIREFCA represented an important step forward in finding a comprehensive way to address the inter-related issues of population displacement, reconciliation and development. If CIREFCA had not brought the problem of displacement to the forefront of the peace initiative in central America, it is unlikely that uprooted populations in the region would have received the degree of protection, attention and funding that was directed towards them.

(10) This summary outlines the overall findings of an evaluation of the CIREFCA process conducted by the Refugee Policy Group for the Central Evaluation Section of UNHCR between March and June 1993. The evaluation focused on five aspects of the CIREFCA process: CIREFCA as a regional model for durable solutions to uprootedness; as an example of inter-agency collaboration within the U.N. system; as an interlocutor between the capabilities and needs of the uprooted and the resources and solidarity of the national and international community; as a vehicle to promote protection; and as a facilitator of durable solutions for the uprooted. The discussion in this summary is admittedly brief and is intended to outline only the most basic issues addressed later in the report.

ACHIEVEMENTS AND SHORTCOMINGS OF THE PROCESS

(11) The evaluation highlights the achievements and shortcomings of CIREFCA, as well areas that were a combination of the two. CIREFCA's successes can be summarized as follows:

• CIREFCA reinforced the regional peace process and, in turn, was strengthened by that process;

• CIREFCA achieved the commitment of governments in the Concerted Plan of Action (CPA) to respect human rights and to attend to the needs of the uprooted;

• CIREFCA raised consciousness about the responsibilities of states toward uprooted populations and reinforced their legal protection;

• CIREFCA attracted additional resources to the region and directed them towards uprooted populations.

(12) In other respects, CIREFCA produced a mixture of successes and shortcomings:

• CIREFCA as a forum for dialogue among various sectors in society, contributed to reconciliation in war-torn nations and to increased understanding in countries of asylum. NGOs in particular found unprecedented and gradually increasing access to official channels through CIREFCA. Nevertheless, the differences in approach between governments and NGOs, and the donors' willingness to fund NGO projects outside official channels, led to a two track process: projects were presented to donors in official meetings by governments having achieved varying degrees of consensus with NGOs and other projects were presented by NGOs to donors outside official fore.

• CIREFCA also provided NGOs with increased international funding and a framework to see their work within the broader context of peace and development in the region. However, the increased funding was in many cases not accompanied by a proportionate amount of training and institution building. CIREFCA reflected a recognition by both UNHCR and UNDP that their missions could complement one another in addressing the inter-related issues of displacement, peace and development. Differences in the two-organizations' structure and style, however, often hindered the implementation of this policy commitment especially at the beginning. In some countries, the relations have been cordial and collaborative throughout CIREFCA, while in others, initial problems have improved over time.

• CIREFCA further advanced the concept that emergency relief and development should be seen as a single continuum rather than a two-step process. The CIREFCA process provided an important practical lesson in the obstacles that must be overcome for this theory to be fully operationalized.

(13) Finally, them were certain shortcomings to the process:

• CIREFCA did not establish appropriate mechanisms at the start to track funding and monitor projects, which complicated coordination and evaluation of its impact;

• CIREFCA was also largely unsuccessful in adequately or systematically addressing the protection and assistance needs of internally and externally displaced persons, despite the contributions of UNDP's region-wide programme, PRODERE, and other governmental and private relief efforts.

BRINGING OPPOSING PARTIES TOGETHER

(14) CIREFCA excelled as a forum for dialogue in a war-torn region, and will be remembered most for its political contribution in bringing opposing parties together. By furthering peace and reconciliation, CIREFCA contributed to the immediate protection of and long-term solutions for refugees, displaced and returnees.

(15) On the programme side, the accomplishments of CIREFCA were less clear for several reasons. Proposals for funding were prepared and presented by governments at international meetings of the Follow-up Committee, after some level of consultation with UN agencies and NGOs. However, many other proposals were funded by donors that did not undergo consultation, including many NGO projects, development projects and food aid, which were termed projects "in the framework of CIREFCA."

(16) There was little coordination of activities among all the national and international actors working in protection and assistance activities, with both positive and negative consequences. This lack of coordination was positive in that it allowed substantial autonomy to the governments and donors, which otherwise might not have been able to commit themselves to CIREFCA. However, the flexibility did not permit monitoring and evaluation or centralization of data on funding levels or programmes. Thus, it is not possible to measure the impact of CIREFCA's investment of resources and personnel.

(17) UNHCR's Quick Impact Projects (QIPs) in Nicaragua have been replicated in other countries in central America and both QIPs and UNDP's PRODERE have been replicated elsewhere in the world. Both are grassroots approaches to reintegration which emphasize beneficiary participation and the involvement of local government.

STRAINED RELATIONS GRADUALLY OVERCOME

(18) While CIREFCA highlighted many of the differences between UNHCR and UNDP and often led to strained relations between the two agencies, the quality of inter-agency cooperation and coordination has improved with time in most countries. In particular, the two agencies worked together well in the Joint Support Unit (JSU) This and other examples of collaboration between the two organizations indicate that joint efforts are possible, particularly when they are working toward a shared goal.

(19) UNDP, which took over the lead role in CIREFCA in June 1993, shifted its focus to the sustainability of peace and reconciliation by broadening consultations in its development work. The Sustainable Human Development approach that is emerging as UNDP strategy, emphasizing grassroots participation and human rights, should not be seen as unrelated to the organization's experience with PRODERE CIREFCA and the overall Special Program for Economic Cooperation in Central America (PEC)

(20) UNHCR's initiative with the QIPs illustrates the agency's increased appreciation for the linkages between today's immediate relief work and tomorrow's reconstruction and development. The lessons CIREFCA has provided on the potential for cooperation and the enhanced understanding that each organization has gained of the other's work should be seen as a foundation upon which to build and improve future collaboration.

RECOMMENDATIONS OF THE REVIEW

(21) The evaluation presents a wide range of recommendations, the most important of which are listed below:

• First, UNHCR and UNDP should intensify already collaborative efforts in Guatemala, paving the way for future repatriation and reconciliation when conditions permit. In this way, UNHCR and UNDP will make concrete strides in effecting the relief to development continuum under CIREFCA.

• Second, priority focus should be placed on internally displaced persons throughout the region, who have received relatively less attention in the process.

• Third, UNHCR and UNDP should establish an inter-agency team at headquarters level to ensure communication and coherence on the numerous collaborative efforts now underway and proposed in Central America and elsewhere.

• Fourth, future attempts to replicate aspects of CIREFCA should consider, as appropriate, the key elements of the process: secure political commitment of the parties involved; establish follow-up mechanisms to encourage compliance, including flexible systems for tracking and evaluating projects to ensure needs are covered and resources are used effectively and efficiently; establish a neutral coordinating mechanism such as a UN joint support unit; and ensure access to adequate international funding.

• Fifth, both UNHCR and UNDP should engage in an internal campaign to assure understanding and agreement on institutional priorities in any new CIREFCA-like endeavours. This is particularly important for UNDP whose structure permits Resident Representatives substantial autonomy from headquarters in their country programming.

• Sixth, UNHCR and UNDP staff should always be open, as they were in Central America, to the possibilities for creating the political space for protecting and assisting uprooted peoples, even before peace accords are signed.

BACKGROUND TO THE CIREFCA PROCESS

FINDING SOLUTIONS TO A REGIONAL CRISIS

(22) The International Conference on Central American Refugees (CIREFCA) was convened in Guatemala in May 1989 by the governments of Central America and Mexico in collaboration with UNHCR and UNDP.1

(23) The purpose of the meeting was to initiate a process to find lasting solutions to the regional crisis of refugees, displaced persons and returnees within the context of national development. The conference was actually just one step in a larger process which began before and continued after the conference. Intended to be a three-year programme, CIREFCA was extended two more years until 1994 to achieve its goals.

(24) The CIREFCA initiative was a response to ever more protracted conflicts which raged in El Salvador, Guatemala and Nicaragua throughout the 1980s. The effects of clashes between the military and opposition forces went far beyond clashes between armed elements. Tens of thousands of civilians were killed and approximately two million people were forced out of their homes by the violence and its aftermath.

(25) Assisted refugees were only a small part of the problem. In 1989, some 150,000 Central American refugees were assisted by UNHCR, according to conference documents. They resided primarily in camps in Mexico (43,000), Honduras (37,000), and Costa Rica (40,000) with "smaller but nevertheless significant numbers in Belize, El Salvador, Guatemala 2and Nicaragua." Approximately 61,500 persons had repatriated. The remainder of the uprooted were internally displaced persons (approximately 900,000)3 and externally displaced persons (900,000).4

(26) At the time that CIREFCA was convened, there was general agreement that these were vulnerable populations. UNHCR has a mandate to assist refugees who had crossed international borders and are so recognized; it was operating in all seven CIREFCA countries in 1989. No international agency had or has a specific mandate to assist or protect internally displaced persons, although some assistance was offered by governments and NGOs, often in an ad hoc fashion. Similarly, there is no international agency mandate referring to externally displaced persons. UNDP, in its work with governments to address poverty and of populations at the time, was beginning to reflect on how it could address the needs of those uprooted by conflict generally. Still, there was an imbalance in the assistance afforded to these vulnerable groups.

(27) Though the conflicts were internal, the pattern of population displacement they caused linked the countries to one another, as sending or receiving countries. For this reason, there was a recognition in CIREFCA that the crisis of displacement required a regional solution.

CONVERGING EFFORTS TO BRING ABOUT PEACE

(28) The conference reflected the convergence of political, economic and humanitarian efforts to bring about peace in Central America. in the political sphere, most importantly, Central American heads of state had been meeting annually to forge agreements. The 1987 Esquipulas II agreement made an unprecedented official call for, among other issues, attention to the needs of refugees and displaced persons in the region. The existence of this regional, political process was perhaps the single most important factor behind CIREFCA's accomplishments.

(29) The Central American crisis played out in an international arena of East-West tensions. CIREFCA reflected these divisions despite its stated intention to be "strictly an and non-political." Individual countries often took sides by intervening politically and economically in the region.

(30) The largest CIREFCA donors were European governments. The European Community in the San Jose process annually pledged economic support to the war-torn national economies in a strategy designed to seek measures to promote democracy and human rights. The Nordic countries, not members of the Community, also contributed heavily in Central America and responded enthusiastically through CIREFCA. In certain ways, some donors used CIREFCA to counterbalance US policy in the region. The United States directed substantial sums bilaterally, but was a relatively small contributor through the CIREFCA process; the United States did have a significant role, nonetheless, as the major funder of the Quick Impact Projects (QIPs) in Nicaragua. Its early coolness toward the CIREFCA process has lessened over time as politics have changed in the region and as QIPs were considered a success.

(31) In the humanitarian sphere, CIREFCA represented the culmination of a series of earlier efforts to extend protection to refugees, starting with the 1984 Cartagena Declaration. Based on the Organization of African Unity's definition, this declaration expanded upon the refugee definition contained in the 1951 UN Convention Relating to the Status of Refugees to include those fleeing generalized conflict.5 In this way the non-binding Cartagena Declaration became an important complement to the 1951 Convention and 1967 Protocol by responding to broad protection needs in the particular context of Central America.

(32) UNHCR staff and donor representatives, in addition to other organizations, took advantage of a particular historical moment within these political spheres to increase protection for the uprooted. One of a number of advisory groups that was called together by UNHCR to study the issue included influential government refugee officials serving in their personal capacity.

(33) In mid-1987, this advisory group asked the UN High Commissioner for Refugees to "convene a regional conference on Central American refugees in 1988, in order to examine needs and specific proposals aimed at achieving practical solutions to refugee problems."6 The meeting was eventually held in 1989, but with the important sponsorship of the Central American and Mexican governments themselves. UNHCR was the principal promoter and facilitator.

UN ESTABLISHED A SPECIAL PLAN FOR CENTRAL AMERICA

(34) During the period leading up to the convening of the 1989 CIREFCA meeting in Guatemala there were two lines of action to attend to the humanitarian issues of war and peace. One was led by the UNHCR and donor country representatives, described above. The other involved the United Nations', specifically UNDP's efforts to support economically the emerging regional peace process. As a result of UN General Assembly action in 1987 to support the Central American peace process and to respond to a call for resources also made in Esquipulas, UNDP was ordered to coordinate Special Plan of Economic Cooperation for Central America (PEC).7 Among the PEC programmes was assistance to uprooted populations as the first priority.

(35) Prior to CIREFCA, UNDP worked essentially with governments on development planning. There had been little overlap between its work and that of UNHCR. However, there was some collaboration among top staff in the development of the PEC, particularly with respect to projects for refugees. As UNDP strove through PEC to address the development needs of displaced populations and to support the peace process, its work in Central America came closer to coinciding with the priorities of UNHCR in the region. Individuals in UNDP and UNHCR were close to the economic and/or political consultations and therefore were part of the larger efforts to bring about peace. Thus, it was the social and economic reality of the Central American region that pushed the two institutions toward common goals and collaborative efforts, emphasizing the connections between population displacement and development.

(36) It was not until 1989, just months before the opening of CIREFCA, that the UN General Assembly officially called for coordination between CIREFCA and PEC and for the participation of UNDP in CIREFCA.8 The UN Secretariat continued to follow CIREFCA through periodic, joint meetings with UNHCR and UNDP CIREFCA staff

(37) The two courses of action led by UNHCR in CIREFCA and UNDP in PEC were sufficiently separate in practice, if not on paper, that there was not a great deal of collaboration on the activities of each institution in the early years. Numerous UNHCR staff participated in the development and founding mission of UNDP's new programme: the US$115 million, Italian government-funded, Programme for Displaced Persons, Refugees and Returnees in Central America (PRODERE). Notwithstanding this collaboration, by the time the 1989 CIREFCA conference convened, and for some time thereafter, UNDP, particularly its Office of Special Projects (OPS), was focused on the start-up of PRODERE and less on other aspects of CIREFCA. The process of merging the work of the two institutions was a challenging one and was only beginning to evolve at the time of the evaluation in 1993.

RESOLVING THE DILEMMA OF DISPLACEMENT

(38) In conclusion, CIREFCA had its origins within an on-going political process in the region, in which the Central American presidents increasingly recognized that resolution of the dilemma of displacement was integral to the peace process and that solutions for uprooted populations must be consistent with national development plans.

(39) For the resources to bring this about, governments in the region called for a commitment by the community. The international community responded, but effectively used its resources as leverage to ensure that governments committed themselves to refugee and human rights issues. On the humanitarian side, CIREFCA firmed up earlier protection commitments made informally by the Cartagena and Esquipulas II documents.

(40) When CIREFCA opened in May 1989, there were seven regional governments participating - Belize, Costa Rica, El Salvador, Guatemala, Honduras, Mexico and Nicaragua. The conference attracted the attention of the highest political levels in Central America, international donor countries, the related offices of the UN system, non-governmental organizations - both and local - academics and the refugees themselves.

(41) Each participant carried its own image of what CIREFCA was about and what might be expected from the process. For governments in the region, CIREFCA was expected to bring resources to address the crisis of the uprooted. UNHCR and UNDP expected a resolve by governments of the region to provide greater protection and a commitment by donors to fund the projects and solutions to be proposed. As a whole, both donors and NGOs expected to influence a more humane turn in government policy toward the uprooted, and to grant wider latitude for NGOs working with these populations.

CIREFCA AS A MODEL AND PROCESS

THE PROCESS HAD MANY DIMENSIONS

(42) The CIREFCA conference secured the commitments necessary for an ongoing, multidimensional process. The process operated at national and international levels and among numerous actors. The complexity of CIREFCA defies simple statements about how it worked and its impact.9 It had political, economic, humanitarian, and institutional dimensions in each country and in the regional as a whole. CIREFCA took on and evolved in somewhat different ways in each of the seven member countries. Thus, it is difficult to refer to CIREFCA as a "model," but rather, it has been a "process" that can provide inspiration for addressing problems of the uprooted in other regions. The country differences and changes over time contributed to the different paths taken and the individualized nature of each country's participation in CIREFCA. Comparisons are complicated by the fact that there are few parallels among asylum countries or among countries of origin in CIREFCA. This has made the task of evaluation more difficult. There are, nonetheless, key elements and reasons for CIREFCA's successes and shortcomings. These elements will be highlighted in this report. In analyzing how the process worked, illustrations will be drawn from the various country experiences.

(43) Over the years, CIREFCA became many things to many people. At times, it was overvalued by some headquarters staff who saw it as a Panacea for resolving protracted refugee crises. It was not fully appreciated by some governments which resented the fact that they did not receive greater resources as a result of it. On the whole, donors were pleased that CIREFCA represented a forum for discussions that contributed to peace efforts and gave them the confidence to invest additional sums in the region. Some NGOs complained of being excluded, but in the end they received increased funding through the process. Although a handful of uprooted individuals attended the meetings, most beneficiaries interviewed were unaware that CIREFCA was anything more than an international meeting in the capital or that it was linked to assistance they received. Thus, they had no expectations to be met or broken.

KEY ELEMENTS CHARACTERIZE THE PROCESS

(44) In order to comment on the lessons of the CIREFCA process and how they might be useful elsewhere, it is first necessary to understand its mechanisms and how they worked in Central America. From the outset, CIREFCA planners believed there were several key elements in assuring protection and assistance for uprooted populations. These key elements included:

• securing the political commitment of the governments of the region to address the situation of uprooted populations

• establishing follow-up mechanisms to monitor compliance;

• including the United Nations as a respected neutral participant; and,

• assuring access to adequate international funding.

(45) Early UNHCR-led planning for CIREFCA focused on the protection and political sides of the refugee crisis. It was not until later that the funding side of CIREFCA was deemed necessary to complement PRODERE in the PEC process and to maintain government involvement over time.

(46) As a word of caution for those seeking to learn the lessons of CIREFCA, it should be remembered that CIREFCA's strengths and weaknesses must be understood in the context of the peace process led by Central American heads of state and in the context of the last years of the Cold War when considerable international financial and political support was showered on the relatively small, geographically well circumscribed region.

CIREFCA WAS AN EXPRESSION OF POLITICAL WILL

(47) The Concerted Plan of Action (CPA) is the centerpiece of CIREFCA. At the inauguration of CIREFCA in 1989, the seven member governments approved the plan which provided the basis for commitment to finding lasting solutions to the problems of four categories of uprooted persons: refugees, repatriates, internally and externally displaced persons. It was the first time in an international effort of this kind that internally displaced persons were fully included. This was the legacy of the Esquipulas, H Agreement.

(48) The CPA is an important expression of the political will of the governments to resolving the crisis of displacement and of their understanding that this was a necessary step toward attaining and maintaining peace in the region. The CPA also made explicit the need to find a regional solution due to the region-wide nature of the displacement winch affected both the countries in conflict that produced displaced persons and the neighbouring countries that received them.

(49) The funding levels awarded by some donors were ultimately affected by the degree to which they considered that governments were respecting their commitments in the CPA.

(50) The states pledged their "commitment to the fundamental principles of human rights and protection of refugees." In addition, the CPA recommended a legal framework for the protection and assistance of refugees. Included in this framework was the 1951 Convention and the 1967 Protocol on refugees and the non-binding, 1984 Cartagena Declaration. The intention was to increase awareness of and compliance with the international norms and to discourage violations.

(51) It is this political commitment that sets CIREFCA apart from other regional efforts to address refugee crises10 One must also remember that the displacement in Central America was dwarfed by any comparison with other refugee crises in Africa or Asia, for example, which may have made the prospects for solutions to the Problem of uprootedness in Central America seem more manageable. In addition, the relatively small geographic area and similarities in language and. culture also contributed to the Perception that the Central American crises could be positively affected with sufficient regional and international support.

CIREFCA STRATEGY LINKED RELIEF TO DEVELOPMENT

(52) CIREFCA, then, was a package: it implied a series of humanitarian commitments in the CPA and commitment to a particular strategy of assistance to bring about lasting solutions. This strategy was described as "link[ing] solution programmes for refugees, returnees, and displaced persons with economic and social development in the region".11 It was apparent by 1989 that the ad hoc, short-term strategies and even self sufficiency programmes promoted until then did not adequately respond to the refugee problem, given the protracted nature of the conflicts and displacement in the region. Today, this strategy is commonly referred to as the "relief to development continuum." Crucial to the operationalization of this strategy was to link the emergency network represented by UNHCR to the development network represented by UNDP.

(53) In practical terms, this commitment required governments to consider assistance to the uprooted as part of. their national development planning. Specific country plans, based on "diagnostic" assessments, were brought to the 1989 meeting. Eventually, these plans led to concrete funding proposals presented in later years by governments. Government approaches to this process varied. In one example, Nicaragua presented its comprehensive plans for post-war reconstruction, initially totalling in the hundreds of millions of dollars, all under the rubric of finding a Lasting solution. Government representatives consistently approached CIREFCA as a potential funding channel, hoping that the more projects they presented, the more funding they could capture from donors. Other governments presented more modest plans which related more directly to uprooted populations. In Costa Rica or Honduras, these plans were seen as additional to their absolute needs and were generally only carried out when external funding was made available. Thus, the proposals appear not to have been a "matter of priority" in development plans, as indicated in the CPA.

(54) Finally, the CPA specifically encouraged NGO assistance efforts within the framework of government plans, and recognized UN and other initiatives in support of peace in Central America.

(55) In evaluating CIREFCA, it was important to examine how far the process advanced in its strategy of linking emergency to development efforts, of encouraging NGO participation, and of linking UNHCR and UNDP efforts in the overall coordination of the process. The national and international follow-up committees established in the CPA were the means by which the strategy was implemented.

FOLLOW-UP MECHANISMS WERE ESTABLISHED

(56) CIREFCA went from a conference to a process by building follow-up mechanisms into the CPA. These follow-up mechanisms facilitated dialogue on solutions at international, regional, and national levels. Ultimately, CIREFCA's most important achievements fall on the political side by virtue of these encounters. The CPA set out the commitment to consensus-building or "concertación." CIREFCA is most highly regarded for its achievements in this area. Here, governments agreed to consult with other national and international agencies, and for the first time with non-governmental organizations (NGOs), on planning and implementing projects. This measure was made possible by donor insistence that government funding would hinge on NGO participation in the negotiated process.

(57) By setting up a regional structure with international participants, CIREFCA had the effect of de-polarizing issues which were too volatile to discuss at the national level. Refugees and other uprooted persons were considered subversive by many national authorities, as were agencies, individuals, or programmes that assisted them.

(58) The forum for dialogue created under CIREFCA played a mutually reinforcing role in the peace process. El Salvador is the case par excellence of the successful interrelationship between CIREFCA consultation forums and the peace process. As the CIREFCA process played itself out in El Salvador, CIREFCA brought opposing parties together to discuss concrete refugee and humanitarian action. These first meetings that included the government and opposition forces helped pave the way to their own peace negotiations.

(59) The International Follow-up Committee, composed of representatives of member governments, UNHCR and UNDP, presided over regional meetings held several times throughout the year, as well as occasional international meetings. The day-to-day organizational and operational tasks were assumed by the UNHCR/UNDP Joint Support Unit, which served as a neutral Secretariat to the Committee.

(60) As proposed by the CPA, national committees were formed that were composed of the appropriate governmental authorities who would work on plans and implement projects for the uprooted. Periodic meetings were held both in the region and at the international level to report on each country's progress and to seek international funding.

(61) In the CIREFCA strategy, the national committees were an important guarantee of state involvement and "ownership" of the process. Indeed, each country held national committee meetings, but they were often a mere formality. For the national committees to have worked as planned, one would have expected active participation in these meetings by government relief, development and planning personnel; exchange of information on plans and implementation of their respective programmes; and on-going communication among themselves and with UN personnel. The National Committee can be said to have operated as planned in Honduras and by 1991-92 in E Salvador and to a degree in Nicaragua.

(62) More often than not, however, a few key technicians gathered to prepare for regional meetings and international funding conferences, with the assistance of CIREFCA consultants and UN staff. There was little on-going exchange of information to allow plans to be developed in complementary fashion. And, except for the countries and time periods noted, development and planning personnel were scarce in this national level work. Therefore, it is not clear, in many cases, how the projects proposed as a result of the national level process could have been consistent with or integral to national development planning, as proposed in the CPA.

(63) In addition to the national committees, the CPA suggested that each country form a CIREFCA support group for consultation in the planning and implementation of projects. The support groups added donor representatives, other international agencies, and NGOs to the circle. All the member countries organized such a group, but the purpose and frequency of meetings varied. In Costa Rica, Honduras, and Nicaragua, for example, the groups did not appear to exert great influence in the country presentations and were said only to have operated with the active intervention of the JSU.

(64) In two other countries, donor representatives. brought their influence to bear: in Mexico, one support group, organized by UNHCR, travelled seven times to Chiapas and was credited with helping to resolve intractable problems in a project there. In Guatemala, the International Group of Consultation and Support for the Return (GRICAR) played a key role in negotiating repatriation.

(65) UNHCR and, in some countries, UNDP officials played an important role in encouraging the governments to form and consult with these groups. However, in most cases, the meetings were held to inform the other CIREFCA players of decisions taken, rather than to solicit their opinions. While governments were willing to comply with CPA requirements to a certain extent in the hope of getting funds, they were reluctant to lose control over programmes for sensitive populations.

(66) One of the outstanding examples of collaboration occurred in El Salvador as the final peace accords were being negotiated there. In true concerted fashion, the government worked with NGOs representing the opposition to present a joint proposal at the 1992 international meeting of CIREFCA. By 1992, Honduras, Nicaragua and Belize were also able to present a joint package for funding.

(67) Even in Costa Rica, a refugee receiving country in which public debate was not a particular problem, CIREFCA facilitated greater coordination among government agencies with direct CIREFCA responsibilities. The Inter-Agency Commission was created including UNHCR, the official refugee agency DIGEPARE, the Ministry of Education and the local affiliate of the International Rescue Committee. Prior to CIREFCA, DIGEPARE did not attempt to coordinate or communicate with other involved ministries.

(68) The role of these consultations was far more important in the larger political sense - and in keeping the plight of uprooted peoples before the policy community - than for the specific programmes planned or funded out of the neutral process. To support the process, the CPA promised "continuous" accompaniment by UNHCR and UNDP. The role played by the United Nations was particularly important as it gave confidence to donors, NGOs and governments alike and lent credibility to the process. Nevertheless, the CPA may have too simplistically expected the process to be carried out and the relief and development arms to join just by ordering UNHCR and UNDP to coordinate it jointly. It gave little guidance as to how the two would share the responsibility.

THE JSU PROVIDED INSPIRATION AND COORDINATION

(69) In order to facilitate the events and coordinate the many regional and international activities, UNHCR and UNDP designated personnel to staff a Joint Support Unit. Based in Costa Rica, the JSU worked directly on CIREFCA project planning, promotion, presentation and reporting, in collaboration with the governments in the region. JSU staff played critical roles from the beginning in setting strategy and maintaining the momentum in CIREFCA. It played an active, not passive, role as trouble-shooter, smoothing relations among CIREFCA actors in the region, serving as a liaison to funders, and providing technical support as necessary. While the JSU had no direct fund-raising role, which is left to individual governments or agencies, it did have considerable influence by its reporting and liaison role to suggest priority areas and projects.

(70) For most of the last few years, the JSU consisted of a four-person staff in Costa Rica and consultants in most CIREFCA countries. It facilitated the consultation process at the national, regional and international levels and responded to problems as they arose. It received the important support and trust of both UNHCR and UNDP.

(71) Hazy lines of authority and communication had both good and bad consequences for the JSU. For example, the JSU operated somewhat ;independently of both UNHCR and UNDP headquarters. This independence gave it great flexibility to respond quickly to problems by providing technical assistance or resolving conflicts. The office of the UN Secretary-General, which played an active role in Central American peace negotiations, was informed of CIREFCA progress in periodic task force meetings including UNHCR, UNDP, United Nations and JSU representatives.

(72) The JSU was relatively successful in its mandate, despite occasional government complaints that it acted without authority. The JSU worked with both agencies and governments to achieve regional coherence and harmonization among national plans and programmes. It displayed the neutrality of the United Nations in CIREFCA. The JSU quickly developed into a united and highly dedicated tem.

(73) UNDP and UNHCR provided support for the JSU. UNDP and later UNHCR also housed CIREFCA consultants for the JSU in most of the member countries. These consultants worked with governments to prepare and present projects for funding, but once projects were funded, consultants had little or nothing to do with the projects. This was the case, despite their terms of reference, which spelled out the follow-up to be conducted. Consultants indicated that no funds were made available from the country office to conduct follow-up or site visits. It is unclear why the JSU did not provide funds for this purpose.

(74) A problem arose with the consultants which was due in part to both the independent nature of the UNDP Resident Representative position and the ambiguous authority of the JSU. JSU consultants were frequently called to work on non-CIREFCA programme activities. The fact that these consultants often had a long-time association as' a consultant or employee of either UNDP or UNHCR facilitated their being assigned other tasks. The JSU often worried whether it would receive consultant reports or project proposals on time for CIREFCA meetings. In Nicaragua, for example, CIREFCA apparently held a low priority on the UNDP workload until recently. Even in those countries in which the respective agency heads were fully supportive of CIREFCA and took collaboration seriously, CIREFCA work often took a back seat to other urgent deadlines. Though the consultants were under contract to the JSU, in effect, they reported to the country Representative or Chief of Mission. The JSU took measures to resolve this problem in 1993 by hiring other short-term consultants as necessary.

RESOURCE MONITORING MECHANISMS WERE WEAK

(75) The final key element of the CIREFCA process was resource mobilization. Through the meetings, CIREFCA provided an opportunity for donors and those seeking funding to meet. The process was intended to yield proposals that were the result of wide consultation among the official and private sectors.

(76) In reality, while CIREFCA attracted new resources to the region, little coordination took place around the acquisition and deployment of these resources. There was no specific discussion much less agreement in the CPA or CIREFCA generally about how resources would be mobilized. CIREFCA became essentially a dual process. On the project side, governments worked with the UNDP and UNHCR to present proposals. The amount of NGO involvement in this process varied, but generally increased over time; by 1991-92 key NGOs were participating in the process in El Salvador, Honduras and Belize. Specifically, UNHCR and donors were responsible for persuading governments to accept NGOs into the consensus-building process. Nevertheless, this same group of NGOs, and others that attended regional and international meetings of CIREFCA, took advantage of traditional donor ties, and the new ties created in CIREFCA, to present their own action plans outside formal channels, separate from government. Many of these NGO proposals were funded by donors from their CIREFCA pledges. The definition of "CIREFCA projects" is thus subject to interpretation. The lack of clear, consistent funding and project data resulting from this voluntary exchange of information with the JSU made evaluation of this aspect of CIREFCA very tentative, as will be addressed further in the chapter on Progress Towards Solutions.

(77) The first international CIREFCA conference in 1989 was a high level, official meeting. In many Central American countries, the refugee assistance agencies reported to the Office of the President or Vice-President, because of the security threat often believed to be posed by refugee presence in country.

(78) Over time, participation in CIREFCA reflected the interest of governments in the process. The high level government participation at the beginning reflected the refugee-security crisis and expectations of major funding. Later, as the issues in CIREFCA were related more to integration, repatriation, and reintegration, ministerial and technical personnel participated more in the national level forums. This was the case in Mexico, Nicaragua, Costa Rica, and Guatemala. With the exception of Honduras, most development or planning ministries played a minor role in CIREFCA in the early years. This may reflect the emergency orientation in the early years, represented today only in Guatemala and possibly Belize. Changes in government in Nicaragua, the peace accords in El Salvador, and repatriation to Guatemala have ushered in renewed interest in rehabilitation and (re)integration that now attract development experts.

(79) The CIREFCA process has been in constant evolution. The CPA provided a framework that was flexible enough to allow for substantial national variation. At first this was seen negatively, as internal UNHCR reports suggested prior to the 1989 meeting12.

PARTICIPANTS HAD DIVERSE OBJECTIVES

(80) It was apparent from early visits to member states preparing for the conference that each held a different idea of the purpose or objective of CIREFCA. Combining these interpretations with the varied national contexts, the flexibility of the process, and changes over time, it becomes easier to understand how the seven member countries arrived at such different CIREFCA programmes.

(81) This diversity was reflected throughout CIREFCA, from the initial country needs assessments to the projects ultimately funded (directly and indirectly) by CIREFCA donors. In a positive sense, CIREFCA allowed each country programme to be formulated more on the basis of real need and local circumstances than totally on stated donor interests.

(82) CIREFCA's key contribution was to provide a forum for dialogue in a war-torn region. While the subject was population displacement, CIREFCA often gave opposing parties their first experience in dealing directly with each other that served them in later discussions on peacemaking and programme development. This was made possible by the political commitments of governments to find viable solutions for uprooted populations. Political leaders found in CIREFCA a way to fulfil this commitment of Esquipulas, which essentially recognized that these solutions were necessary for maintaining peace.

(83) It should be recognized that the success of the CIREFCA process very much depended on the peace process. The follow-up mechanisms, or numerous national and regional level consultations, provided the meeting space for a wider sector of society and kept the uprooted on the national agenda.

(84) CIREFCA represented a unique opportunity to outline strategies in the relief to development continuum. It was in this regard that many differences arose between UNHCR and UNDP that provoked internal debate about how to respond in such humanitarian crises. Nevertheless, occasional conflictive relations often do lead to changes and resolutions. At the time of the evaluation, however, the bigger problem in merging relief and development work came from governments which were generally not disposed to use their own funds to implement programmes for uprooted.

CONCLUSIONS AND RECOMMENDATIONS ON THE CIREFCA PROCESS

(85) CIREFCA provides inspiration for addressing the problems of the uprooted in other protracted regional conflicts. one must be keenly aware of the importance of the on-going peace process in CIREFCA's successes, however. Elements of the process can and should be replicated as they are found appropriate to another context. What made CIREFCA successful was the interplay of its key elements within a flexible process that allowed for substantial variation among countries and participants.

(86) The following recommendations are based on the essential elements of the process in Central America that may be applicable to UNHCR/UNDP work in other regions of the world. In Central America these elements formed a package which made a profound contribution. The context of a particular region or country should be considered in any attempts to replicate the elements of CIREFCA. In doing so, UNHCR and UNDP should:

• seek political commitment to resolving the refugee crisis within the framework of peace; one should not wait for political changes to begin to be aware of opportunities to further peace through humanitarian assistance;

• organize regional, multi-country or sectoral responses which will tend to downplay normally polarized national debates;

• propose commitment to a framework of protection and respect for human rights that will condition financial support;

• secure the assurance of sufficient international financial support to address the magnitude of the problem and to entice government participation; one should promote multi-donor participation in project funding to avoid the precariousness of dependency on a single donor or over-identification with a particular project and to promote de-politicization;

• establish follow-up mechanisms of consultation and monitoring to be called by member governments, but with participation by the UN, donors, and NGOs;

• create a joint technical support and coordination unit, like the JSU, that can facilitate the process, representing neutral bodies, such as the UN;

• encourage linkage between government counterpart agencies and ongoing planning and development institutions of the country; and,

• promote full participation in discussion by all official and private sectors in refugee and development work.

THE INSTITUTIONAL FRAMWORK: NEW APPROACHES WERE NEEDED

(87) Throughout the 1980s, a recognition was building in the international system that organizational relationships were not adequately responding to the complex emergencies that resulted in the large-scale displacement of populations.

(88) By the end of the decade, it was becoming clear to relief organizations, such as UNHCR, that new approaches had to be found to link solutions for uprooted populations to the broader framework of reconciliation and development. Progressive elements within development agencies, such as UNHCR, became aware that the promotion of development similarly depended upon the establishment of peace and the return and (re) integration of uprooted populations.

(89) In Central America, the peace process provided the political climate in which relief and development agencies could begin to build bridges between their efforts. During the period preceding the 1989 CIREFCA meeting, UNDP and UNHCR were following parallel courses of action with regard to displaced populations in Central America: UNDP through PEC and PRODERE; and UNHCR through its central role in promoting the CIREFCA conference.

(90) In 1989, when the CIREFCA conference launched the formal process, the two agencies faced the challenge of translating their shared mission into relationships and mechanisms that worked in practice. From the beginning these progressive forces struggled to overcome the inertia of large, long-standing bureaucracies. Often it was not a matter of wanting to work toward a goal or greater collaboration, as much as the challenge of how to effect it. In each country the relationship between UNHCR and UNDP evolved somewhat differently for reasons related as much to country conditions as to personalities.

(91) UNHCR held the lead role from the start in 1989, expecting to turn over that responsibility to UNDP at a later point. In mid-1993, UNDP assumed that lead role. As noted earlier, the responsibilities involved in each role were never clearly assigned in the CPA nor by subsequent agreement. The timing of the hand-over was not defined at first, but was expected to roughly correspond to the nature of the crises-focused early on refugees, repatriates and emergency assistance to them, followed by reconstruction and development assistance. This turnover, or emphasis on development, was not early enough to help with repatriation and reconstruction in 1991-92 in Nicaragua, where UNHCR and UNDP had a particularly difficult time communicating with each other. However, the lessons of cooperation have been learned in the Salvadoran case and are evident in the Guatemalan repatriation.

(92) The collaboration of UNHCR and UNDP in CIREFCA has underscored the differences in structure and style between the two organizations over the years and has often has led to strained relations between them. Given the complexity of the issues upon which their collaboration has been based, this was not surprising. CIREFCA has provided an important regional context for the two organizations to shape a mutual working relationship. Particularly since UNDP began preparing in 1992 to take the reigns, the level of interaction and cooperation between UNHCR and UNDP has increased both at the field level and at headquarters.

ORGANIZATIONAL DIFFERENCES LED TO TENSIONS

(93) The pairing of UNHCR and UNDP in CIREFCA was at once deliberate and circumstantial. With CIREFCA's development focus, it would have seemed logical for the UN's development agency to be involved. The two agencies began to collaborate on plans for refugee assistance under PEC. The individuals first involved in forming CIREFCA saw the potential for joining forces. Translating these personal efforts into institutional collaboration in each country and headquarters has taken time, but continues to be a headquarters priority.

(94) The two agencies differ significantly in their operational structures and institutional cultures, and they clashed frequently in CIREFCA as a result. UNDP is a planning and technical assistance agency of the UN working with member governments on the full range of national development issues; it is not generally considered a development implementing agency, although its Office of Special Projects does implement major programmes including PRODERE, PRORAAS in Nicaragua, among others. UNHCR is an implementing agency which often implements its programmes through non-governmental organizations or government entities. It is mandated to provide refugee protection and assistance around the world. Unlike UNDP, it often finds itself in an antagonistic relationship with respect to governments as it tries to defend refugee rights.

(95) These differences in the nature of the organizations mean UNDP's outlook and strategies are more influenced than UNHCR's by the priorities and policies of governments (often in the countries of origin) and by development theory and practice than by the situation on the ground or needs of any particular population. UNHCR, on the other hand, traditionally relates to governmental refugee offices as well as non-governmental organizations in countries of asylum. Furthermore, only recently has UNHCR had greater contact with governments of origin because of the increased opportunities for refugee repatriation. Therefore, UNHCR and UNDP came to CIREFCA with different perspectives; it would only be expected that they would need time to forge common approaches for their joint work.

(96) On another level, UNDP's focus is geographic while UNHCR primarily focuses on population groups, specifically refugees and repatriates. These dichotomies became less clear over time in some countries. UNDP is essentially a planning agency. The flexibility of its resource availability is limited by the small number of mechanisms through which it can make funds available for development. As decisions regarding the use of funds through its principal funding mechanism, the Indicative Planning Figure (IPF), are made in conjunction with governments, the use of these funds for initiatives that are not necessarily supported by the governments is made more difficult. Its funding is allocated within the United Nations system and, while it has administered large voluntary contributions for major initiatives, it has had little need to develop an ongoing capacity for special fund-raising. UNHCR sponsors and implements projects with its programme support, but regularly raises additional special funding to carry out its mission.

(97) These differences in focus are brought into relief when one considers how the two organizations are structured. UNHCR field staff take their lead from headquarters, but the structure is flexible enough to adapt to country differences and to rapidly changing conditions. UNHCR's flexibility of response is enhanced by its on-going fund-raising capacity. UNDP, rather, gives tremendous autonomy to its Resident Representatives, who, in coordination with the governments of the countries where they reside, may develop quite diverse programmes. Its relatively lengthy (five year) planning and funding cycles, do not lend themselves to the changing conditions of war and, later, reconciliation. Therefore, the two organizations did encounter some friction when trying to agree on priorities.

(98) In those countries in which the Resident Representative chose priorities compatible with CIREFCA and UNDP's institutional commitment to it, relations between UNHCR and UNDP were productive. This was the case in El Salvador and Guatemala during most of CIREFCA; it was only recently the case in Nicaragua.

(99) In addition to these structural differences, the staff of each agency often manifested a competitive attitude toward the other that impeded effective coordination. UNHCR staff were criticized as "do-gooders," i.e. well-meaning, but unenlightened; UNDP technicians were criticized for their lack of field knowledge and their over-reliance on technical studies. UNHCR action in dealing with emergencies was seen as too quick and without thought; UNDP was seen as too slow and bureaucratic to be helpful in emergencies. Indeed, especially at the outset, there were differences in style and operation that coloured each agency's perception of the other's capacity. Some UNHCR staff saw themselves in the vanguard and were impatient to get along with the process, lest they lose the possibility to link refugees to the emerging peace process. While certain regional staff of UNDP were equally aware of the possibilities of a brokered peace, the bureaucracy of the institution seemed only prepared for the longer-term view. UNDP/OPS worked to eliminate some of these same problems in PRODERE.

PERCEPTIONS OF CIREFCA WERE INCONSISTENT

(100) Another reason for the sometimes strained relationship between UNHCR and UNDP related to misunderstandings between the two over the nature of their respective commitments to the CIREFCA process. The Concerted Plan of Action did not define roles or objectives for either agency. Their commitments differed, as shown below.

(101) UNHCR was the lead agency involved in the genesis of CIREFCA. There was great overlap in the objectives of UNHCR and CIREFCA. CIREFCA therefore became an agency priority, involving personnel at all levels. UNHCR still has personnel in the JSU, in-country, and at headquarters whose job is CIREFCA. UNHCR became the lead agency in CIREFCA at the start. As UNDP normally holds the UN flag in international operations, this somewhat unusual move may explain some of the distance from CIREFCA that UNDP has shown at the country level. Nevertheless, UNHCR expected to "pass" the lead agency role to UNDP for the development phase of CIREFCA.

(102) The line between what is CIREFCA and what is UNHCR's own work became less clear at the field level. Over time and as staff changed, CIREFCA became "another project" to report on and "another meeting" to prepare for. Complaints have been voiced regarding the lack of resources in the programme for this follow-up. Indeed, the only CIREFCA projects tracked by UNHCR are those funded directly through it; projects funded through governments or NGOs do not receive UNHCR (or UNDP) follow-up. Protecting and assisting refugees and repatriates, and at times, displaced persons - the objectives of CIREFCA - are carried out by UNHCR as its own work for which it does not necessarily see the need to consult with UNDP.

(103) High level UNDP and UNHCR officials collaborated in the early strategizing for PEC and CIREFCA. In the move of these officials from the field to headquarters, they brought the commitment to CIREFCA with them, resulting in both policy and financial resources flowing in its favour. For UNDP, however, CIREFCA was viewed as but one humanitarian programme in its wide portfolio (including PEC) of development plans in the region and treated it accordingly. Addressing the problems of the uprooted was considered just one aspect of reconstruction and reconciliation efforts.

(104) As noted earlier, UNDP came formally into CIREFCA relatively late and was at the time focused on PEC and PRODERE. The PEC programme was valued at US$2.9 billion, of which approximately US$500 million were estimated to have been mobilized by the end of 1991.13 Of that amount, CIREFCA projects amounted to only US$65 million (although they had increased by the time of evaluation to approximately US$250 million).14 The funds mobilized for CIREFCA were dwarfed by comparison to the full PEC load for UNDP.

(105) Under these conditions, it was unrealistic to expect UNDP to give CIREFCA the same relative level of attention as UNHCR gave it. Despite UNDP's of large funded programmes, it did not have an ongoing fund-raising capacity to dedicate to CIREFCA.

A CLEAR DIVISION OF RESPONSIBILITY WAS LACKING

(106) There was no clear division of responsibility in CIREFCA. Both UNHCR and UNDP participated in CIREFCA meetings at all levels, but the JSU work on the day-to-day coordination made up for the lack of clear institutional responsibilities. The JSU requested back-up as needed from the respective headquarters and country offices.

(107) The country-level officials of both institutions played a significant behind-the-scenes role in advising governments on their proposals and strategy in CIREFCA. Together, they were instrumental in pressuring unwilling governments to consult with NGOs, both as a minimum requirement of donors for funding and as an official commitment made under the CPA. In some countries, there was minimal involvement of UNDP Resident Representatives. In Mexico, UNDP priority areas did not overlap with CIREFCA. in Belize, UNDP had stationed no international representative. Also, there was no UNDP CIREFCA consultant in these countries.

(108) Each institution held distinct visions of how to integrate relief into national development plans. UNHCR's vision was that of a two-step process involving relief then development which was reflected in the CPA, written essentially by UNHCR staff in the planning for the conference. Over time, documents written by UNDP and the JSU described the strategy in different terms. They described the strategy as one continuum from initial emergency aid through to development assistance. In theory, decisions at each stage were to be made with the goal of long-term development always in mind.

(109) While UNHCR staff appeared to embrace the continuum concept, no one was clear as to how it should be carried out. This difference became important in how the two agencies approached CIREFCA and each other. As lead agency in CIREFCA between 1989 and 93, UNHCR approached the implementation of the strategy as more of a two-step process than as a continuum: UNHCR took responsibility for the first step of relief and expected to "turn over" its beneficiary population to UNDP for the next step of inclusion in development programmes.

(110) In Belize, Costa Rica, Honduras Mexico and Nicaragua, for example, UNDP never really became involved in the planning or implementation of relief with UNHCR, despite UNDP's expressed understanding of the single continuum theory. Thus, they were not able to reap the benefits of the continuum strategy. In El Salvador, it was PRODERE that worked with UNHCR in some of the emergency aspects of repatriation from Honduras. It is important to the fulfilment of the strategy that this field capacity of UNDP be maintained. Today, Guatemala represents the possibility for applying the continuum strategy. Relations between UNDP and UNHCR have been cordial and cooperative there throughout CIREFCA, and the recent repatriation of refugees has provided an area for even more intense collaboration as to where and how to best assist repatriates in a way that will be helpful for reconciliation in the short-medium term and development in the long term.

(111) Despite memoranda of understanding between the two institutions giving UNDP responsibility to promote assistance to internally and externally displaced persons and UNHCR prime responsibility for refugees and repatriates under CIREFCA, the distinctions were never clear.15 Expectations on the part of UNHCR that UNDP should initiate more projects to attend to the internally displaced were not fulfilled. Both institutions claimed it was not their role - UNDP, because it could not focus on specific populations groups; UNHCR, because its mandate extended only to refugees.

(112) UNDP appeared to regard PRODERE as a concrete response to its commitment. However, PRODERE was focused on two to three sites per country where a concentration of displaced and other uprooted groups were found; it did not directly address the situation of displaced persons on a national level.

(113) Indeed, the situation of internally displaced persons continues to be an enigma even under CIREFCA. Little is known about their conditions, numbers and needs; the few studies and little information that do exist indicate more attention is required than PRODERE, governments or NGOs have been able to provide.16 The task of assessing and attending to the needs of internally displaced persons was a specific objective in CIREFCA, which was confounded by the difficulties of collecting such information, institutional gaps, and an effective lack of political will.

(114) Further, while it is known that donor countries sponsored projects to attend to displaced, the information available to the evaluation team was not disaggregated to know how much was spent or what the impact was of the funds directed to internally displaced persons, specifically. Early government proposals from Guatemala and El Salvador to address the internally displaced went largely unfunded in CIREFCA.

(115) The issue was not black and white, however. UNHCR argued that its mandate extended only to refugees, but it did take upon itself to address the needs of internally displaced persons alongside assisted repatriates in Nicaragua and El Salvador. In El Salvador and for a time in Belize, there was friendly collaboration between UNHCR and PRODERE in these efforts. UNDP recognized the competence of UNHCR in the protection area and urged UNHCR to share greater responsibility for the protection and documentation. This has provided an area for even more intense collaboration as to where and how to best assist repatriates in a way that will be helpful for reconciliation in the short-medium term and development in the long-term.

(116) With UNDP limiting its responsibility to PRODERE and UNHCR addressing displaced persons in an ad hoc fashion, the needs of displaced persons appeared to have received comparatively far less attention than other groups in CIREFCA. To identify, quantify and address the needs of internally displaced is a complicated matter differing throughout the region, that requires further attention from CIREFCA and beyond.

(117) More than anything else, the result of the lack of full understanding and collaboration between UNHCR and UNDP in CIREFCA has been lost opportunities. For example, opportunities were lost in Nicaragua to consolidate the gains by UNHCR's early intervention, economic and political conditions in Nicaragua notwithstanding. Here and in other countries, the comparative advantage of each institution could have brought improvements in the planning and implementation of projects and better coordination among their respective networks, including government ministries and NGOs. UNDP and UNHCR could have learned from each others' experiences and opened opportunities for collaboration in the new, participative grassroots, intervention strategies of UNHCR (Quick Impact Projects) and UNDP/OPS (PRODERE). As noted earlier, greater attention to internally and externally displaced may have been possible with greater institutional collaboration. Closer working relationships also could have helped bridge the gaps between the plans of governmental development agencies and those of ad hoc refugee agencies, public and private.

(118) UNDP and UNHCR cooperated well in some instances and in several countries. Examples of the on-going cooperation in El Salvador and Guatemala, particularly with regard to repatriation and reintegration, are cited throughout this report. In addition, there were other instances of collaboration well worth mentioning. In Costa Rica, when the defeat of the Sandinista government in elections in Nicaragua posed the possibility of unexpected large-scale repatriation, the two agencies negotiated a temporary halt to PRODERE's programme in Costa Rica; this halt allowed refugees to make a decision on return without being influenced by asylum country conditions.

(119) In Belize, UNDP/PRODERE and UNHCR signed the first cooperative agreement between the two agencies, allowing PRODERE to compensate UNHCR for documentation assistance. This agreement was the precursor of a successful agreement for collaborative work in other countries of Central America.

(120) In Nicaragua in 1993, UNDP agreed to assume oversight of UNHCR's projects, as the latter prepared its phased withdrawal from that country.

JOINT ACTION REQUIRES SHARED COMMITMENT AND PERSONAL CONTACT

(121) One of the brightest spots in the relationship between the two agencies was the Joint Support Unit (JSU). The JSU consisted of an equal number of staff from each agency and has been headed by senior staff of the lead agency. The transfer from UNHCR to UNDP in 1993 involved the same staff shifting roles. It is testimony to the team spirit and commitment to collaboration they have attained that the transfer took place so smoothly.

(122) Ultimately, as the JSU demonstrated, the differences in substance and orientation between UNHCR and UNDP can be overcome when personnel, especially agency heads in a given country, are convinced of the importance of a programme. This involvement can be achieved through Personal, informal contact, but also through clearly stated agency priorities communicated down the line.

(123) One of the impediments to cooperation in Central America was the UNDP structure which gave significant latitude to its Resident Representatives in cooperation with governments to choose the priorities of each country programme. However, cooperation and mutual understanding were more likely when the agency heads in country were committed to the goal and when there was greater familiarity and informal, personal interaction between the staffs, as was the situation with the JSU.

(124) Overall, the experience of working together and the personal contacts made over the years within the context of CIREFCA has affected each agency. Relations have improved since CIREFCA began. Each agency developed an expanded vision of its own work, as well as a greater understanding for the work of the other. UNDP developed and began to implement some aspects of its new Human Development strategy in Central America, with emphasis on human rights and grassroots participation. The recent human rights work of PRODERE reflected this new approach.

(125) UNDP also replicated some of the CIREFCA consultation processes in its national reconstruction planning with governments. For its part, UNHCR staff became more aware of the need to relate emergency relief within the context of development plans and follow-up. This was one of the principal lessons of the Quick impact Projects in Nicaragua and has been evident in UNHCR's plans to replicate lessons of CIREFCA in other countries.

(126) The transfer of the lead agency role from UNHCR to UNDP was controversial, because of the differences in structure, style and commitment. First proposed by UNHCR to take place in mid-1992, it was delayed one year. Concerns about the transfer were raised by governments, NGOs and donors. They worried about losing the protection role of UNHCR and questioned the fund-raising and field capacity of UNDP. For the most part, the concerns were based on a lack of familiarity with UNDP and a failure to realize UNHCR would maintain a presence in the region, albeit with a lower profile.

(127) Once UNDP agreed to the transfer, this concern subsided. UNHCR however continued to press UNDP for a commitment to raise funds for remaining projects. The matter caused some consternation in the transfer of the lead role and was reportedly resolved in late 1993.17

(128) Having taken over the lead role, UNDP is making efforts now to steer CIREFCA and post-CIREFCA work toward the consolidation of peace in the region. A headquarters-led campaign has been evident, in 1993, to enlist governments, donors, NGOs and other agencies in the implementation of a broad-based "human development" effort.

CONCLUSIONS AND RECOMMENDATIONS ON UNHCR/UNDP RELATIONS

(129) To summarize, the relationship between UNHCR and UNDP in CIREFCA, not always easy, has improved with time and is the source of very important and constructive lessons for the future. The differences in organizational structure and culture were difficult to overcome in most countries. UNDP was more focused on the specific implementation of PRODERE than the general coordination of CIREFCA the transfer of leadership was planned. Fulfilment of CIREFCA objectives was agency policy for UNHCR, which had initiated the process and dedicated numerous personnel at various levels in the agency to its implementation. Nevertheless, the joint posting in the JSU is the outstanding example of teamwork, although other examples of collaboration between country staff were based on personal relationships between agency heads and commitment to CIREFCA.

(130) Recommendations for improving the relationship between UNHCR and UNDP follow:

• UNHCR and UNDP must share commitment to the goal in order to collaborate effectively in CIREFCA and similar endeavours. In situations in which either institution has decided upon a regional or approach that may differ from prior commitments or standard practice, the field staff, as well as headquarters, should be informed about and follow such institutional priorities. This is especially important in the UNDP structure which has allowed substantial autonomy in setting country level priorities to its Resident Representatives in the field even when those priorities may not have been in line with regional commitments;

• Where UNHCR and UNDP collaborate w projects in a given country, personnel of the two institutions should share offices as a proven means of facilitating cooperation and understanding between these two very different institutions;

• Both agencies should continue the level of support and autonomy given to the JSU, which has been an important part of its success as a unit in promoting CIREFCA;

• UNHCR and UNDP should work closely at the headquarters and field levels, and in the JSU, to ensure the progress and transition of CIREFCA projects in the next year and post-CIREFCA;

• Relationships and understanding between the two organizations can be improved by joint training and seminars on topics of concern to both. Possible joint seminars on the grassroots strategies of PRODERE and QIPs, involving senior strategists, technical and field personnel in Central America, would be as instructive about methodologies as about their respective institutions.

• Both UNHCR and UNDP should give priority to collaborative efforts in Central America and elsewhere. The roles and responsibilities of each institution should be clearly defined, within the limits of their respective mandate or purpose. Such collaboration will contribute further to the understanding of how to bridge emergency programmes on one side and development programmes on the other.

RESOURCE MOBILIZATION

CIREFCA ATTRACTED ADDITIONAL FUNDS

(131) According to donors, the CIREFCA process attracted funds to Central America, and particularly to uprooted populations. While there is no way to prove that CIREFCA attracted additional funds than would otherwise have been directed to the region, most donors indicated that the regional aspects of the process and the coordination by the United Nations inspired them to invest more in the region.

(132) A comparison of funding levels prior to and during CIREFCA would have helped to show CIREFCA's role in attracting resources to the region. However, it was. impossible to find comparable data among donor sources. To the extent that these data were made available, they either represented overlapping or multi-year figures, or were not disaggregated to show assistance to uprooted populations within overall commitments to the region.

(133) In general, data on resource mobilization were not complete, as of the time of the evaluation, and often not detailed enough to discern types of projects funded, implementing agencies, or categories of uprooted persons as beneficiaries. Thus, the review is not definitive on tins point and lists of projects by country are not included here. Instead, estimates are compiled on the basis of information presented by the JSU through 1992. As appropriate, some updates were added in Notes in the final version of this report.

(134) It can be said that the mobilization of international resources was the glue that held much of CIREFCA together. Without the possibility of funding, CIREFCA may have been only an interesting meeting that highlighted the problem of refugees and displaced persons, but it would not have sustained interest as an ongoing process. The promise of funding kept governments interested, secured legal protection pins as collateral, and promoted dialogue among disparate and opposing sectors of society.

(135) There are at least two interpretations of CIREFCA's scope. In a strict sense, CIREFCA projects were those that underwent consultation in the various national fora of CIREFCA and were then presented for funding at pledging conferences. In general, donors were reluctant to fund directly some Central American governments and preferred to channel resources through UNHCR, UNDP or NGOs. Approximately one-third. of government proposals went unfunded on technical or political grounds.

(136) The way in which money was channelled led the governments of Honduras, El Salvador and Guatemala to charge that they received little or no funding directly as a result of CIREFCA. It was their way of showing ownership of the CIREFCA process by denying that funds not directly received and administered by the government, or at least in consultation with it, could be considered part of the CIREFCA process.

(137) As CIREFCA was a dynamic and multifaceted process, its scope is open to another interpretation. Donors and the JSU, at least, also considered as CIREFCA projects those that were never presented officially by governments, but which were funded from the pledges made by donors in CIREFCA fora.

(138) Many of these projects were presented by NGOs directly to donors without passing through the CIREFCA consultation mechanisms. This second category of projects was termed by the JSU as projects "in the framework of CIREFCA," even though they had not been discussed with governments. Because they were funded from official donor pledges in CIREFCA, they were considered CIREFCA projects by criteria of the donors. Because the CIREFCA process respected each funder's autonomy to commit pledged funds according to their own criteria and mechanisms, there was no single interpretation of what constituted CIREFCA projects.

(139) Thus, the process of categorizing CIREFCA projects was somewhat arbitrary. In addition to these projects, millions of dollars were directed by the United States and others to other projects benefiting uprooted populations in the region which were not considered as CIREFCA by the JSU, member governments or donors.

(140) Ultimately, every participant had different criteria for defining CIREFCA. In this way, the resources mobilized by CIREFCA did not constitute the total resources brought to bear on the problem of displacement. This has direct implications for an evaluation of solutions and makes it nearly impossible to isolate CIREFCA's impact.

DATA AND TRACKING PROBLEMS WERE ENCOUNTERED

(141) Providing a precise tally of the resources brought to the region by CIREFCA is not possible with the information at hand. There was an expectation that CIREFCA would provide a mechanism for defining priorities and coordinating their implementation, ensuring coverage of problems areas and eliminating

(142) However, reliable data on donor pledges and on projects funded was not centralized in CIREFCA. This was no fault of the JSU, but rather, reflected a structural problem related to the lack of effective or compulsory mechanisms on project development, fund-raising and monitoring of projects.

(143) Funds were channelled through each donor's own funding and implementation mechanisms, at times passing through various government offices or NGOs before reaching the beneficiary. In other cases, funds were awarded through UNDP or UNHCR. Funds were not channelled through the JSU or other CIREFCA mechanisms. Thus, the JSU was handicapped in its ability to report on the progress of CIREFCA. Curiously, a number of active donor and regional country representatives interviewed were not aware that CIREFCA itself did not receive and allocate funds.

(144) The relatively loose coordination of project funding and implementation had both positive and negative consequences. The lack of strict reporting requirements and rules allowed both donor governments and member states the freedom to participate in CIREFCA. The autonomy of each was respected. Stricter follow-up mechanisms may have resulted in the break up of CIREFCA.

(145) The evaluation revealed tremendous uncertainty about the total dollar value of donor commitments, despite the efforts of the JSU to collect this information over the years. In this situation, no single actor in CIREFCA has an overall picture of CIREFCA's work or impact.

FUNDING FOCUSED ON REPATRIATION

(146) The following table represents CIREFCA projects approved through 1992, (data received as of July 1993 are as yet incomplete).

Approved CIREFCA Projects (through 1992)

	NUMBER OF PROJECTS	AMOUNT US$ (MILLIONS)
Official	67	95.0
Initiatives in the framework of CIREFCA *	50	154.7
SUB-TOTAL	117	249.7
PRODERE **	9	115.7
TOTAL	126	365.4

* This is a minimum figure, as an undetermined number of individual projects are often grouped together in JSU reports as a single initiative.

** PRODERE is implemented through six country programmes and three regional sub-Programmes. In addition to the US$115 million contributed by the government of Italy, PRODERE received additional funds from UNDP totalling US$770, 000, as of the end of 1992. Not included in these totals are additional contributions of up to US$30 million by the UN Fund for Capital Development, Holland, and the International Fund for Agricultural Development for "similar and complementary 11 projects administered by UNDP's Office of Project Services.

(147) For lack of clear data, the JSU catalogued projects in the framework of CIREFCA as "initiatives. " It is therefore difficult to say exactly how many projects were financed through each initiative. A minimum of 126 projects were funded, however, totalling at least US$365 million. Over and above this figure, unaccounted programme funds and time were spent by UNHCR and UNDP in CIREFCA activities (additional to and not to be confused with project follow-up listed below).

(148) At least 14 governments outside the region, as well as multilateral agencies, contributed to the CIREFCA process. The table below shows the largest donors18.

Major Donors

Donor	US$ millions committed through 1992
Italy (PRODERE)	115
European Community	88
World Food Programme	40
Sweden	54
TOTAL	297

(149) The total pledged by these four donors alone constituted 80 percent of the total commitments to CIREFCA. The other major donors included Australia, Canada, Denmark, Finland, France, Germany, Holland. Japan, Norway, Spain, Switzerland, and the United States.19

(150) Sixty-seven projects were funded at least partially. After undergoing consultation and being officially presented, another 29 Projects remain unfunded totalling, US$54.8 million. An additional US$39.7 million is required to fund completely the approved projects, summing to a total funding requirement of US$95 million.

(151) To the extent that information was available, the table below shows the implementing agencies to which it appears that the contracts were awarded. Project funds were awarded to both UNHCR and UNDP (PRODERE and 12 other national level projects administered by UNDP). Not all governments received funding directly, as shown, but some received funding indirectly. For example, UNHCR received millions each year for the Mexican governmental agency, COMAR, to distribute to Guatemalan refugees. NGOs, as a group, were awarded funding for the most number of projects.

Implementing agencies	Number of projects and initiatives
UNHCR	25
UNHCR/UNDP	4 (follow-up, JSU, etc)
UNDP	21 (9 of PRODERE)
Governments	17 (2 Belize, 3 Guatemala, 1 Honduras, 1 Honduras, 11 Nicaragua)
NGOs	56 (minimum figure)
Other	3

(152) Overall, directed most CIREFCA resources to Nicaragua, El Salvador and Guatemala, in that order. Honduras received the least overall funding from this process, not counting the related funds for repatriation of Nicaraguan Contra forces, under CIAV, the International Commission for Assistance and Verification of the Organization of American States and the U.N. Although it is not possible to say exactly how many projects or resources were dedicated to each population group under CIREFCA because of the current data problems, the ranking of the three (ex-)conflict countries confirms the current focus on repatriation.

(153) Of the funds provided by the top four donors, El Salvador received the largest share and Belize received a far smaller share than Honduras. This highlights the importance of a large and varied donor pool.

(154) Belize, Honduras and Costa Rica, the countries which bore the highest refugee per population load, received the least resources from CIREFCA, according to the JSU data and analysis of the four largest donors. This was especially important Belize with upwards of 20 percent of its population foreign-born.20 Prior to CIREFCA and the repatriations, Costa Rica and Honduras were privileged with funding as countries of first asylum. The focus shifted as refugees returned to their countries of origin.

(155) During the CIREFCA period, some donor countries continued restrictions which allowed humanitarian relief only to refugees in camps. Such action penalized both the refugees and government which had permitted them or other uprooted populations to live outside camps. The practice countered CIREFCA's attempts to find the most inclusive and viable solutions to the problems of the uprooted. It was one of the reasons Belize had such difficulty attracting funding, despite the acute situation caused by the refugee influx. Belize's comparatively high per capita income and early indications by the government that it could handle the influx were also cited as reasons for lower funding to Belize through much of the CIREFCA period.

NGO'S RECEIVED A LARGE PROPORTION OF RESOURCES

(156) The above figures show that most of the resources mobilized by CIREFCA were not directed to the official country presentations. Rather, most resources were provided to projects presented by NGOs "in the framework of CIREFCA. " Even the projects funded through governments, UNHCR and UNDP are implemented by local NGOs (whether they be traditional, local NGOs, ad hoc NGOs created in the political environment, or newly created NGOs for the purposes of a particular programme).

(157) Thus, on all fronts, NGOs appear to have gained the most resources from CIREFCA. An analysis of the funding of CIREFCA's larger donors shows this even more clearly: 80% of Swedish funding went to NGOs, with no funding of governments; 30% of EC funding went to NGOs. Having funded NGO presentations and having left open the possibility of funding governments, the donors effectively pressured governments to widen the civil consultation process to include NGOs.

FUNDING WAS POLITICIZED

(158) Commitments to CIREFCA by European donors often were based on a political commitment to pressure the governments, of the region to continue to dialogue for peace, human rights, democracy, and solutions for uprooted peoples.

(159) In the context of the Cold War and its aftermath, some donors even manifested support for particular warring factions by use of CIREFCA funding. This was particularly clear in some donors' funding of Salvadoran NGOs related to the Farabundo Marti National Liberation Front, as a counterbalance to other funding of the Salvadoran government and military.

(160) At times, donors and NGOs deliberately held back details of funding and projects. Concern over the security of NGO workers and of uprooted beneficiaries in conflict conditions was generally given as the reason for this "non-transparency. " However, the practice has continued even in post-conflict times. Such a policy of unaccounted transfers leaves wide open the opportunities for inefficient use of funds, as well as misuse for personal or political purposes.

(161) The political use of humanitarian funding during the conflicts was widely alleged by both sides in the region. In order to provide humanitarian assistance in conflict zones, where many agencies could not operate, donors worked through those institutions at hand, often those of the opposition, whether they agreed with their politics or not. Under these circumstances, humanitarian assistance by private groups or governments almost inevitably takes on political colouration, as in the case of the collective return of Guatemalan refugees in January 1993 or the collective return of Salvadorans from Colomoncagua, Honduras. A distinction should be drawn between assistance for expressed political purposes and assistance to NGOs whose activities in" have both social and political consequences.

(162) Regardless of their intentions, donors and their local representatives must be aware that even the appearance of abuse can jeopardize programmes and future funding. While it occurs frequently in this and other conflicts, funds for express political purposes should not be confused with humanitarian channels. Greater transparency in funding will minimize the chances that the use of resources will be subject to suspicion in the future.

GOVERNMENT PROPOSALS REMAIN UNFUNDED

(163) There are two issues yet to be addressed with respect to resource mobilization. One refers to governments' intentions in CIREFCA, the other with those of certain donors. While programmes and funding for uprooted peoples were integrated into development plans, this aid was usually viewed as "additional" to their development requirements. The concept of additionality, as at extra support to the region, was so stated in the PEC strategy.21 Unfortunately, such a position obstructs the planning and funding for continuity to individual programmes and the CIREFCA process itself As conditions changed in the region, those programmes for uprooted populations included in national reconstruction and reconciliation plans in El Salvador, and to some degree in Nicaragua and Guatemala were viewed with more importance.

(164) The second issue refers to donors. By continuing to fund NGOs outside the CIREFCA consultation process, one must ask if certain donors misled UNHCR and UNDP on their intentions to fund governments if and when they improved their human rights records and submitted to an open consultation process with NGOs. That many government proposals remain unfunded was also the result of other technical and political factors.

CONCLUSIONS AND RECOMMENDATIONS ON RESOURCE MOBILIZATION

(165) In conclusion, CIREFCA did attract additional resources to benefit refugees and displaced persons in the region. The lack of centralized and rigid reporting requirements proved to be both a virtue and drawback for CIREFCA. On the one hand, it allowed governments in the region and outside donors to participate in the process without undue restrictions on their autonomy. On the other hand, CIREFCA did not coordinate and rationalize the use of resources and failed to assure non-duplication of funds. At the time of this review, CIREFCA has no mechanism to evaluate the impact of the effort at the level of projects and resource investments.

(166) In general, resource mobilization carried political implications. Many European donors supported CIREFCA as a way to pressure Central American governments to take steps toward greater respect for human-rights and democracy. It sometimes appeared that this objective was more important than the effective and efficient implementation of projects.

(167) Following are recommendations for resource mobilization: Donors must recall their commitment in CIREFCA to finding solutions within the context of development. With peace yet so fragile, they should continue to support those countries which emerged from the emergency phase, in their efforts to assist the uprooted through rehabilitation and development.

• Donors and recipients should be more forthcoming with the JSU about the projects they sponsor under CIREFCA in post-conflict situations. Such sharing among official and private agencies working in reconstruction, reintegration and development will only improve efficiency and reduce the odds for political manipulation of funding. Promoting greater transparency and the central monitoring of project and funding data will eliminate costly duplication of efforts and enable the JSU to identify geographic regions, sectors, or population groups which yet require attention.

• Donors should be attentive to the timing of funding. The bureaucracy involved in making payments on approved projects should be appropriate to the size and scope of projects and timed to avoid problems caused by funding delays.

• Donors, governments, and NGOs should not use humanitarian funds for expressly political purposes. Such abuse, or even the appearance of abuse of funding, jeopardizes other legitimate activities and may affect the level and conditions for humanitarian assistance in a given country.

CIREFCA'S PROTECTION ROLE

CIREFCA STRENGTHENED THE LEGAL FRAMEWORK

(168) The CIREFCA Concerted Plan of Action referred to the legal framework for protection of refugees and displaced persons. This framework was based on a body of commitments and legal principles from the 1951 Convention and 1967 Protocol to the 1984 Cartagena Declaration.

(169) The CIREFCA process, in effect, made compliance with these principles a condition for funding. Operating within the CIREFCA framework, UNHCR encouraged countries to sign the Convention and its Protocol and to pass domestic laws protecting refugees. However, it is not what was achieved in the CIREFCA process in a direct sense to protect refugees, repatriates and displaced persons that is most important. CIREFCA gave legitimacy to their claims as civilians caught in conflict, where prior to CIREFCA they were considered by governments as the enemy - guerrillas or their sympathizers.

REFUGEE PROTECTION WAS REINFORCED

(170) Prior to CIREFCA, refugee protection was not as high a priority for receiving governments as it was for UNHCR and non-governmental organizations. Costa Rica, El Salvador, Guatemala and Nicaragua were signatories to the international conventions on refugees. UNHCR was also invited to operate in Belize, Mexico, and Honduras which were not signatories at the time.

(171) Following a decade of legal studies and initiatives promoting rights of refugees, CIREFCA provided the catalyst that sensitized governments to the fact that displacement was a consequence of war that required direct attention. As a result, Honduras signed the Convention and Protocol in 1991 and amended its national legislation accordingly. Although Belize was considering its adherence to the international instruments prior to CIREFCA, they were not acceded to or promulgated into national law until 1991.

(172) In an already ongoing reform of the population law in Mexico, authorities were convinced to amend a category in the law which accorded refugee status for the first time; Mexico also continues its long tradition of diplomatic asylum. Mexican non-governmental organizations that had organized around CIREFCA provided a particularly effective lobby for this change. The wording of the 1990 Mexican law conformed to the refugee definition adopted at Cartagena. In 1992, Costa Rica decreed that refugees could become permanent residents - a move considered by authorities as the completion the refugee assistance cycle: provide asylum, help them toward self-sufficiency, facilitate voluntary repatriation, and then integrate those who decide to remain.

RETURNEES BENEFITED FROM INCREASED DIALOGUE

(173) Voluntary repatriation began prior to CIREFCA, but was not complete. Over time CIREFCA committed the majority of its resources to return and reintegration programmes. Providing protection for individual repatriates continued to be difficult logistically. Nevertheless, UNHCR's response in-country to collective returns became a firm commitment, and less of an ad hoc extension of its mandate.

(174) In addition, CIREFCA fora facilitated an open environment for dialogue and negotiation which brought the actors in the repatriation process together. Examples include the tripartite commissions; the quadripartite meetings (among COMAR, NGOs, UNHCR and refugees) in Mexico regarding the repatriation of Guatemalan refugees, where refugees were included in such talks for the first time; the Guatemalan Mediating Commission; and the Guatemalan GRICAR support group of donor governments and international NGOs. These efforts increased protection for returnees. Guatemala provides the ideal case study for identifying the specific impact of refugees participating in negotiations on their own behalf.

(175) Salvadoran refugees in Colomoncagua, Honduras used the occasion of the 1989 inauguration of CIREFCA to announce their collective repatriation. Their action was at once a political statement and a plea for protection and assistance.

THE INTERNALLY DISPLACED REMAIN LARGELY UNPROTECTED

(176) No international agency has a clear mandate to protect and assist internally displaced persons. CIREFCA, which was charged with protecting and assisting them, has done little to change this. UNHCR claims it has no official mandate to work with them. In a 1992 report of the High Commissioner to the Executive Committee, however, she admits that UNHCR should take responsibility in some cases.22 Under PRODERE, UNDP did much to reach areas of high concentration of displaced persons. It provided early assistance to returning refugees and displaced persons in one region of Nicaragua following the change of government in 1990, for example. PRODERE has also been working in the Ixcan, Ixil, and the outskirts of Guatemala City where internally displaced persons have returned home or relocated. Still, UNDP's style of operation in PRODERE is to focus only on particular regions of a country, not specific population groups. Assistance to other dispersed populations would not be possible under these limitations.

(177) By including internally displaced persons as a category under CIREFCA, following the recommendation of Esquipulas II, however, CIREFCA did further raise awareness of their plight and offered the promise of protection. Prior to CIREFCA, internally displaced persons in Central America were almost always considered combatants by local military and other authorities.

(178) Under PRODERE, there were projects that provided human rights 41 education to internally displaced persons. Other projects addressed their documentation needs, giving them freedom of movement and facilitating their work and voting privileges. As noted, UNHCR assisted in this effort. Still, the coverage was limited to certain geographic regions.

(179) Current data do not permit an assessment of the specific impact on the displaced. In the first place, the data on their numbers, location and condition are generally inconsistent and unreliable. Second, data received from donors generally puts repatriates and internally displaced indiscriminately in one category. Therefore, it has been difficult to ascertain exactly how many projects or how much money has been directed to the internally displaced or its impact. It is generally acknowledged that internally displaced persons still remain a vulnerable category.

(180) At a minimum, the CIREFCA process screened out government attempts to secure outside funding under CIREFCA for pacification programmes in the guise of humanitarian programmes. Such programmes were common in several countries during the 1980s.

(181) The CIREFCA process, not including PRODERE, has made little direct contribution to the protection and assistance of internally displaced persons that was not already underway. Perhaps because of this lacuna, there has been some momentum in Central America toward reaching the displaced. A group of lawyers, scholars, diplomats and international agency personnel who were active in CIREFCA have joined forces with a broader international movement on behalf of internally displaced persons. Individuals involved are studying ways that the mechanisms and institutions of the international system could respond in an ad hoc fashion to the needs of internally displaced persons.23 UNDP and UNHCR should take advantage of the new Opportunities for information exchange and increased dialogue on the needs of the internally displaced that this process has initiated. Providing protection for populations within a sovereign state has been the major challenge in beginning to address this exceedingly complicated issue in Central America and elsewhere.

THE EXTERNALLY DISPLACED HAVE BENEFITED LITTLE FROM CIREFCA

(182) Externally displaced persons received little protection from CIREFCA. Both before and after CIREFCA, they have most often been treated as undocumented or illegal aliens and have been subject to deportation, according to local policy. This was certainly the case among Salvadorans and Guatemalans in Mexico and Honduras. While not necessarily deported in some countries, some displaced groups had no legal rights and many still lacked access to services because of their illegal status; these groups included Nicaraguans in Honduras and Costa Rica and, to some extent, Salvadorans and Guatemalans in Belize.

(183) There were some exceptions to this lack of attention. There were small attempts to include them in integration projects for refugees and nationals in Honduras and Costa Rica, particularly by PRODERE, which was working to resolve their illegal migration status. It is unclear how many others fell outside such protection. In addition, UNHCR has followed the practice for a number of years in Central America of assisting externally displaced persons who approach the institution for the purpose of repatriation.

(184) CIREFCA made a conceptual contribution to the legal area by the identification of "externally displaced persons" as a vulnerable category of persons. While the category was apparently suggested by One government in its attempt to justify a higher funding request, the literature on migrants and refugees in Central America proves that this group is indeed vulnerable. They live and work in precarious conditions fearing detection and deportation.24 Not recognized as refugees, for many reasons, they are not generally eligible for international protection. The new international convention on migrant workers may offer some coverage, however.25

CONCLUSIONS AND RECOMMENDATIONS ON PROTECTION

(185) Summarizing, CIREFCA made important gains for the uprooted by linking, in the CPA, human rights commitments to participation in the process. CIREFCA's achievements in the protection area focus on legal progress for refugees. Repatriates received the benefits of CIREFCA Protection in facilitating broad-based negotiations and assistance for return. Internally and externally displaced persons received only recognition as vulnerable civilian populations and a proportionately small amount of assistance by their inclusion in the CIREFCA process. While legal advances may not translate into immediate protection for all groups, it is a step in the direction of recognizing and respecting their human rights.

(186) Neither internally nor externally displaced persons received many concrete protection benefits from CIREFCA. The process, however, should have been able to do more for internally displaced persons than PRODERE and a few other programmes were able to accomplish. There was wider acknowledgment in the community of the need to respond to the internally displaced and they were specifically mentioned in Esquipulas II to which CIREFCA was a response.

(187) Recommendations on protection follow:

• Experience in Central America has underscored the need for a more effective system for assisting and protecting internally displaced persons. The UN system must continue efforts already underway to establish mechanisms and open up dialogue aimed at resolving the protection and assistance problems that these populations face. In the meantime, UNHCR is and potential drawbacks of such an approach, with regional and international repercussions. encouraged to extend its mandate to reach the internally displaced in Central America, particularly in areas where they are living together with repatriates.

• Mexico is encouraged to become a signatory to the 1951 Convention and 1967 Protocol Relating to the Status of Refugees. All CIREFCA member countries, that have not yet done so, are equally encouraged to bring their national legislation on refugees in line with the Cartagena Declaration and the Convention and Protocol.

• Those countries with a residual population of externally displaced persons are encouraged to legalize their migration status, with consideration given to the ability of those persons to return safely and voluntarily to their own countries, the time spent in country, and the degree to which they have already become incorporated into the economy.

• The direct participation of Guatemalan refugees in negotiations for repatriation presents the opportunity for studying the beneficial elements.

PROGRESS TOWARDS SOLUTIONS

PROJECT REVIEW MUST BE QUALIFIED FOR LACK OF DATA

(188) Our efforts to evaluate the impact of CIREFCA on the reintegration of uprooted persons are constrained by several factors. The analysis herein is based on information collected from March-June 1993, and may not reflect changes since that time.

(189) This review initially was intended to include a technical evaluation of the impact of CIREFCA projects, as explained earlier. However, it was discovered early in the interviews that this would not be possible given the time and resource limits of the evaluation. Data on projects was not centralized and there was no requirement or procedure within the CIREFCA process to do so. Information on funding and projects flows between the parties to the contract, that is, between donor and implementing agency (government, international organization, or NGO). While the JSU was able by the end of 1992 to compile a fairly comprehensive listing of projects funded by country, donor, and amount pledged, the JSU indicated it had no reports on project implementation or progress at the time of this evaluation. The project list is known to be incomplete with respect to certain donors and NGO projects.

(190) The evaluators attempted to identify several projects for in-depth review. However, the wide variation in projects defied categorization in any representative fashion by type of project, size, donor source, country of origin/asylum, implementing agency or beneficiary group. Simply, the available lists of projects were so diverse within each country and in the region as a whole, that any truly representative sampling was impossible.

(191) The financial and programme data on CIREFCA-inspired projects often does not permit an analysis of the attention given a specific beneficiary population.

(192) Earlier attempts to evaluate CIREFCA projects were hindered for some of these same reasons. More importantly, the lack of definition in project and funding mechanisms resulted in delays in the initial presentation and funding of projects. Early projects often needed to be revised before donors would consider them, because of technical problems or government attempts to take economic or political gain from the projects. Both UNDP and UNHCR staff were instrumental in steering governments toward more appropriate and technically sound presentations. The improved quality of the projects presented in the 1990 New York Follow-On Meeting was reflected in increased funding. Many projects with CIREFCA funding did not really get underway until 1990-91. While some projects were implemented earlier, implementation of others was delayed while plans received technical improvements.

(193) Before CIREFCA, attempts were made to better protect and assist refugees and displaced persons throughout the 1980s. Each new initiative built on the lessons of previous efforts. Still, refugees in camps were ever more dependent on outside aid. Voluntary repatriation had begun, but international agencies found it difficult to respond under conflict conditions. The beneficiary population of CIREFCA projects was often composed of up to 50 percent local populations. (Re)integration and development therefore were intertwined with larger issues of marginalization in society and endemic poverty. CIREFCA was not expected to resolve these structural issues.

(194) In some ways, CIREFCA served as a laboratory for institutions to attempt new strategies to reach the needs of uprooted populations. in any discussion of CIREFCA with UNDP, the subject turned to PRODERE, just as with UNHCR, it turned to Quick Impact Projects (QIPs). Both PRODERE and QIPs represented new ways of working for their respective implementing agencies and new ways of working with uprooted persons. These programmes are not the only or most important ones funded by CIREFCA, but they do merit separate discussion here.

PRODERE DEVELOPED A REGIONAL MODEL

(195) The weight of PRODERE's budget, approximately one-third of the estimated CIREFCA total, and its regional character, in six countries, signal it for special mention in the evaluation. However, PRODERE was treated ambiguously within the CIREFCA process. PRODERE both secured its funding and began operating before the 1989 CIREFCA conference; its projects did not undergo the same kind of consultation process of other CIREFCA projects, and it received separate treatment in meetings and in JSU information reports. Regional government and donor representatives interviewed also placed PRODERE in a special category. Thus, while PRODERE was considered. UNDP's major contribution to CIREFCA, it did receive consistently separate treatment. This review does not comprise an in-depth evaluation of PRODERE; nevertheless, the size, duration and importance of PRODERE would indicate that a separate and detailed evaluation should be conducted.

(196) PRODERE was the one region-wide programme utilizing a uniform, grassroots strategy intended to promote socio-economic integration. Despite severe cash flow and political problems during its first years (1989-90), PRODERE began to consolidate its work in 1991. In the early years a substantial portion of funds was spent on and feasibility studies rather than on project implementation. In part this imbalance was due to the early uncertainty, inadequacy, and irregularity of its funding, as indicated in PRODERE's annual progress reports. But it also appeared to reflect an approach consistent with the importance UNDP places on technical support, feasibility studies, and planning documentation before it begins implementation. Tins emphasis appeared to change somewhat over time as cash-flow problems were worked out, making funds available for project implementation; the end of the conflicts in Nicaragua and later El Salvador created conditions in which projects could be implemented; and PRODERE itself took on more field-based character.

(197) PRODERE's technical strategy was to promote community participation to meet the needs at the local and departmental levels. Unfortunately, the early top-down implementation of the strategy often hindered community participation. For example, according to many individuals interviewed, decisions about project sites were made in the capital without local input; local individuals and groups were invited to hear presentations of pre-set plans and feasibility studies; and decisions about project activities were based more often on studies or political interests than on local consultations. With experience, PRODERE learned how to incorporate more effectively local participation in developing plans, thus bringing the implementation of the programme more in line with the model's original intent. The delays in funding further left some communities skeptical of PRODERE, which required their full participation to succeed. These debilitating delays in funding were due to the reliance on the programme's single donor; while large contributions are not to be discouraged, the problem does point out the advantages of diversity in funding sources for projects as large as PRODERE.

(198) PRODERE's early emphasis was framed in a broad, multisectoral development plan, including production, infrastructure, agriculture, health, and education. The programme did support significant infrastructure in 78 municipalities and provided specialized training including human rights, municipal administration, and agriculture. The gains in the productive area were limited, however, as on-going conflict often impeded access and prevented the minimum stability required for operations. PRODERE systematized project data to a much greater extent than other projects reviewed, but despite numerous feasibility studies, it is aware that it lacks baseline data against which to measure its own progress.

(199) Although a relatively small part of its budget, human rights education will be one of its major contributions. It has been highly developed and often conducted in collaboration with UNHCR. In Guatemala, for example, PRODERE trains human rights promoters, distributes literature on international human rights instruments in several languages, and works to strengthen an independent Judiciary. Over time, the approach by PRODERE articulated the need to repair the " social fabric" of the society at the same time as other productive interventions were organized. Their work has helped to further and put into practice the Sustainable Human Development approach pursued world-wide by UNDP.

(200) Basically, PRODERE was a facilitator, rather than an implementor, in the words of the field staff. An important legacy of PRODERE, when it terminates in 1994, will be the creation of local development mechanisms in some countries. Facilitating the organization of local public and private efforts for training, technical assistance and credit, to which representatives of uprooted populations participate, will ensure on-going impact in economically (and politically) marginalized areas after the programme ends.

(201) In implementing this and other PRODERE programmes, UNDP/OPS has worked collaboratively with numerous other local agencies, both official and private, as well as international agencies including UNHCR, UNICEF, WHO/PAHO, WFP and the ILO.

QUICK IMPACT PROJECTS HAVE PROVED EFFECTIVE

(202) The QIPs, developed by UNHCR in Nicaragua, were an attempt to provide a solution for a large number of communities impacted by uprooted populations. The approximately 320 small projects were seen to be low-cost and effective in providing immediate, focused support at a cost of approximately US$9.6 million. Projects included replacing or repairing damaged infrastructure, investment in small livestock or other productive community enterprises. They were intended to serve as a bridge between emergency and development assistance. In particular, QIPs gave returning populations a chance to unite with others in the community, find employment to prevent further migration, and attempt to re-build their lives. However, the lack of follow-up and a larger framework in tins case made it difficult to build on the progress that they achieved.

(203) The speed with which the QIPs were implemented in Nicaragua had positive immediate effects in a country that lacked alternatives, long-term impact may have been lack of planning.26 The country had been racked by war and was only beginning the proms of reconciliation at the time. Still, the QIPs did promote participation, interaction, and reconciliation among repatriates, displaced and others in the community, and did "leave the seeds" for follow-up development when resources are available.27

(204) Both PRODERE and QIPs are being seen as prototypes to promote integration. QIPs are being replicated in El Salvador and Guatemala and used to integrate refugees in Belize. UNDP and UNHCR are collaborating in Cambodia to replicate appropriate aspects of both programmes, as well as CIREFCA mechanisms, and their replication in other areas such as Southern Africa and the Horn of Africa is being assessed.

(205) One of the most attractive features of each of these programmes is the element of beneficiary participation to promote reconciliation. In Belize, the QIPs have been implemented with speed, showing their importance also in countries of asylum. Unfortunately, the element of community participation, particularly by refugees and externally displaced persons, appeared to be downplayed in the rush to reach implementation goals. These problems were being addressed at the time of the evaluation and were being taken into consideration in planning in other countries. Without the participation of the affected populations, the programmes lose sight of their original aims - to promote reconciliation and reintegration.

CIREFCA PROJECTS HAVE HAD POSITIVE CONSEQUENCES

(206) Not all CIREFCA projects were new and not all were successful. The variety of projects in the region and even within each country made it impossible to select "representative" projects in this evaluation. Included in CIREFCA were projects as diverse as settlement projects for refugees and repatriates; integrated rural development projects; health and sanitation projects and infrastructure construction; schools; training; a land survey; a survey of externally displaced persons; housing; production; credit; legal protection and documentation; and micro-enterprise development.

(207) The following analysis highlights some impacts of these and other CIREFCA projects according to criteria which assess how effective solutions supported by CIREFCA have been. Due to a scarcity of data, these assessments are based on field observations, interviews, and available supporting documents.

PHYSICAL REINTEGRATION IS PROMOTED

(208) As noted, CIREFCA concentrated resources on repatriation-related activities in Nicaragua, El Salvador and Guatemala. CIREFCA projects - official, UN and NGO - supported collective returns of Salvadorans and Nicaraguans and the return of individual repatriates. CIREFCA projects through international agencies and NGOs were instrumental in facilitating the first collective return of some 500 Guatemalan refugee families from Mexico in early 1993. Both before and since CIREFCA, UNHCR programme funding contributed significantly to cover the costs of repatriation.

(209) CIREFCA provided important support for the Tripartite, Quadripartite, and other mediating processes which facilitated land settlements for returnees in Guatemala and earlier in El Salvador and Nicaragua.

(210) Secure access to land in each of these countries has been a principal factor in the conflicts themselves and it remains a stumbling block in achieving a lasting peace. Aside from PRODERE's resolution of some land tenure issues, and a land study commissioned in Guatemala by UNDP/PRODERE to identify lands for repatriates, CIREFCA did little to open dialogue on the roots of this structural problem. Though the possibilities for such discussion at the national level were slim in all but El Salvador after the accords, it remains a question whether CIREFCA's regional level discussions might not have de-politicized this debate, as it did other sensitive human rights issues, for example.

LOCAL INFRASTRUCTURE REPAIRED

(211) Local infrastructure that was damaged or destroyed in the conflicts was improved with the assistance of PRODERE in 78 municipalities throughout the region and in each of the regions of Nicaragua to which QIPs were directed.

(212) In refugee host countries, PRODERE also made infrastructure investments in areas where refugees and externally displaced were concentrated. CIREFCA's other major donors, the European Community, in particular, channelled millions into reconstruction of infrastructure. It is likely that through CIREFCA programmes, marginalized areas received far more attention than they would otherwise have, given the overall scarcity of resources in these war-ravaged economies and the numerous competing demands on the state. CIREFCA did not support the rebuilding of all those communities in need, but did give a glimmer of hope to a select number of them.

SOME PROJECTS SUPPORT REINTEGRATION INTO CIVIC LIFE

(213) Identity documents are one of the most important needs of all uprooted populations. The UNHCR has supported Salvadoran government efforts to document repatriates and displaced in former conflict zones; in these areas, PRODERE and UNHCR collaborated on documentation activities there. In Guatemala, PRODERE has given documentation and human rights education a priority in all programmes. Additionally, UNHCR and PRODERE are working together in Guatemala to resolve the complex logistical and human rights situation posed by the first mass repatriation in 1993.

(214) Also, in a project with its share of understandable technical and political difficulties, the civil registries destroyed during the conflict in El Salvador are being replaced under CIREFCA funding, a contribution to the return to normalcy and lasting peace. But, one of the most significant indicators of the positive impact of CIREFCA in leading to solutions can be seen in the Salvadoran repatriates participation in election-year activities, not just by voting, but by presenting their own candidates for local office.

REINTEGRATION RESULTS ARE DIFFICULT TO QUANTIFY

(215) Most of the projects funded under CIREFCA were aimed at socio-economic (re)integration. As mentioned, it is not possible to quantify the impact of these investments. Only general observations may be made about how the funding appears to have been directed. For example, once relocated to their homes, repatriates and internally displaced received significant aid from CIREFCA in Nicaragua and El Salvador. Individual returnees received relatively less attention than did collective returnees in all countries, primarily because of security and logistical difficulties in reaching them. There was no indication, however, that the availability of assistance through CIREFCA served as a primary motive for their actual return.

(216) Financial credit was provided to individuals and communities in all countries . The use of credit programmes under such precarious conditions had varied results. There are many examples of individuals successfully setting up their own small businesses and of communities that used the funds for production or other investment.

(217) Although default rates for negligence were often low, at about 4 percent, revolving funds and other credit schemes became an on-going burden for some. Beneficiaries were frequently poorly prepared to understand and manage their credit, often unable to calculate total indebtedness on multiple loans (housing, production, etc.). These problems were compounded by agricultural conditions in recent years. The training and follow-up required was not always available in the projects visited.

(218) There are some shining examples of successful, multi-sectoral integration programmes for Guatemalan refugees in Campeche and Quintana Roo, Mexico. There, refugees received land from the state and made it produce enough not only to secure their self-sufficiency, but to sell in the local market. The programme has been effective in large part because of the participation of the local and national government, as well as the need for labour in this isolated and desolate region. The multi-annual programme began before CIREFCA, but subsequent contributions were considered under the CIREFCA umbrella. As opportunities for repatriation to Guatemala began to open in 1992-93, many of these refugees were expected to return home.

SUSTAINABILITY WAS NEGLECTED

(219) Frequently, the sustainability of the various projects proposed was not given explicit attention in their planning or implementation. Productive or income generating activities were often taken to be Permanent solutions without consideration for factors winch would condition their future, such as land tenure, access to markets and product demand. This frequent failure to address the numerous structural, environmental and other barriers impeded the progression from relief to (re)integration within a strategy of sustainable development.

(220) A question was raised about the appropriateness of the agrarian focus in reintegration efforts. Such efforts may only vitiate the cycle of poverty and dependency because of limited land availability, scarce resources, and the difficulty farmers have in achieving self-sufficiency. While for political, cultural, and economic reasons attention must include the rural areas where the majority of the uprooted still reside, future development and poverty alleviation call for greater diversification within each repatriation programme.

(221) On the whole, the projects that focus on socio-economic integration of specific population groups, promoted the participation of the beneficiaries in the production of goods and services for their own consumption. However, the population tended to remain dependent on outside inputs and financial assistance to produce them.

PARTICIPATION AND CONSULTATION HAVE BEEN EMPHASIZED

(222) In El Salvador, the involvement of NGOs and beneficiaries in the design and implementation of programmes to aid them reintegrate has been particularly strong. The Salvadoran people deserve the credit for what they have been able to achieve. There was tremendous enthusiasm at all levels for reconciliation and reconstruction. Although certain NGOs comment that consultations at the national level do not continue to be as open and frequent today as at the time of the accords, the levels of dialogue among public and private entities are evident throughout society, even at the community level.

(223) In the case of Guatemalan repatriation, a CIREFCA-inspired innovation is the Quadripartite process, described earlier. Refugees have forcefully and successfully represented themselves in negotiations with the governments. The exceedingly complex mass repatriation of Guatemalans in January 1993 - in which attending to humanitarian and protection concerns were complicated by the politicization of the return itself - almost required the participation of all parties in what were quite difficult negotiations. In traditional repatriation situations, it has been assumed that UNHCR represented the refugees' interests. The changes implied by direct refugee participation in repatriation negotiations is a subject for further discussion and analysis.

(224) In Central America, UNHCR has generally looked to NGOs as the advocates for refugees and other uprooted populations. This tendency was reinforced in CIREFCA as NGOs gained ever greater participation in what was initially perceived as an official process. More recently, particularly in Guatemala, NGOs are passing through a "credibility crisis" with certain repatriate groups that resent and reject the notion that NGOs might presume a role appropriately belonging to the repatriates themselves.

INSTITUTIONAL CAPACITY HAS BEEN STRENGTHENED

(225) When CIREFCA called on governments to identify the needs of uprooted populations and to develop projects to respond to these needs within national development plans, there were varied responses and capabilities. It had never been done in Central America before. With the technical assistance of consultants for UNDP and UNHCR during the first year, the technical and feasibility level of plans presented improved.

(226) At another level, CIREFCA funds strengthened the technical capacity of government counterpart agencies in the planning and implementation of programmes, such as in Belize. UNDP programmes, particularly under PRODERE, and UNHCR's QIPs in Nicaragua supported the development of local level government.

(227) A weakness in this process was UNHCR's almost exclusive relationship with ad hoc, emergency systems or counterparts set up by the governments. To the extent that these counterparts were not integrated or had little interaction with the principal ministries of planning or development, there was a failure to impact permanently on national response capacity. UNHCR might have advised governments on the merits of such linkages; whether governments then followed such advice differed by country.

NGO'S ENJOYED NEW OPPORTUNITIES

(228) Non-governmental organizations working with uprooted populations were considered a nuisance, if not a security threat, by governments throughout the 1980s. CIREFCA opened opportunities for NGOs to participate with governments in the definition of problems and solutions. Having been invited as mere observers to the 1989 CIREFCA conference, their influence grew over time.

(229) Traditional NGOs organized and matured in their professional capacities in that decade with exposure to new projects, funding and contacts.28 Their outlook broadened and both national and regional associations of NGOs working on refugee concerns were created, with CIREFCA fora as their prime outlet. NGOs received greater exposure to international funders from this experience and their funding levels grew from the tens of thousands of dollars to the millions.

(230) The availability of funding for NGOs spurred the creation of new NGOs by governments, UNHCR, and political forces which expected the NGOs to implement projects according to their respective criteria. UNHCR argued that the new NGOs (or local affiliates of NGOs) were necessary to assure apolitical, quality services.

(231) To the extent that ad hoe NGOs were created, rather than boosting the capacities of existing NGOs, international assistance hindered the development of permanent national response capacity. The question is whether it was possible for UNHCR to turn, at least in part, to local NGOs as new and technically capable groups emerged over time. It should be noted that in at least two examples, ASAI and CIRES in El Salvador, these ad hoc agencies became independent and established their own right.

(232) There continues to be a critical need for technical and administrative training in the region, particularly for NGOs. Having had their effective operation during conflict recognized and rewarded in CIREFCA, NGOs are now finding themselves subject to a new set of criteria for peacetime.

(233) The activities of the last few years did not necessarily result in the strengthening of the NGOs institutional capacities. Few donors contribute to institution building, yet some donors are now beginning to impose more rigorous accounting and other restrictions. Many ad hoc NGOs may not survive under these conditions, but other established NGOs are currently receiving millions of dollars without apparent knowledge of standard accounting procedures. The increased funding was not accompanied by proportional support for institution building.

WOMEN AND CHILDREN HAVE RECEIVED SPECIAL ATTENTION

(234) The Concerted Plan of Action called for solutions to consider the particular situation of uprooted women and children. The centerpiece of CIREFCA's work on women was the First Regional Forum on Gender Focus in Working with Refugee, Returnee, and Displaced Women (FOREFEM), held in early 1992.

(235) Controversial in its methodology, which among other things first excluded men from planning meetings, the conference did raise awareness among all CIREFCA actors of the need to address the issue. Both UNHCR and UNDP collaborated in the formation of regional and national study groups. Governments, NGOs, donors and international agencies attended the conference. The conference document was not approved, probably more because of a lack of preparation by conference organizers than because the recommendations were radical. The process was too much, too fast for the governments.

(236) However, FOREFEM had its impact. First, gender became an integral component in the development of the QIPs which were being started in Nicaragua. Numerous QIPFEMs were funded and are now planned for other countries. Second, after inordinate delay, the JSU/UNDP entered into an agreement in 1993 with UNIFEM to provide consultants who have begun to review and proposed strategy for inclusion of the gender focus in United Nations and collaborating government programmes throughout the region.

(237) In other work with women, it is important to highlight CIREFCA's support of Mama Maquin in Mexico. This 8,000 member organization of Guatemalan refugee women is a model of the strength of women's organizing, having achieved a capacity for project generation, implementation and training unequalled among other Guatemalan refugees.

(238) Specifically for children, CIREFCA directed resources in all countries to construct and rebuild schools and to health and nutrition programmes that had child health components. These facilities would not have been possible without the resources from outside, but they do represent now an on-going burden to governments that must staff and provide equipment for them.

CONCLUSIONS AND RECOMMENDATIONS ON SOLUTIONS

(239) By the end of 1992, CIREFCA facilitated no less than 126 projects in seven countries with a total investment of not less than US$365 million.29 There were no representative CIREFCA projects; rather, the process allowed for tremendous national and regional variation. The projects were further differentiated by funding source and levels and by implementing agencies, all with their own methodologies. This is testament to the degree to which CIREFCA projects were driven by beneficiary needs, perhaps, and were not totally donor-driven.

(240) Refugees and repatriates appear to have received most attention. Despite the expectations that projects would benefit internally displaced persons, there were insufficient efforts to do so under CIREFCA. Internally displaced persons were assisted, however, by governments, PRODERE and other UN agencies and NGO projects in localized regions where they mixed with repatriates.

(241) Externally displaced persons continue to be excluded from national and international benefits in most of the region. Nicaraguans in Costa Rica and, to some extent, Honduras, represent the exceptions. It can only be concluded that there really were no intentions to address their needs in a direct way by most CIREFCA actors.

(242) In assessing the durable impact of CIREFCA, one must consider that even with the extensive funding attracted to numerous projects in the region, there was an isolated nature to their implementation. Projects did not build on one another. PRODERE is the only region-wide programme, operating in six countries with a uniform strategy and within a structure of planned, follow-up activities.

(243) Whether CIREFCA projects will result in permanent solutions rests, in part, with the follow-up action that is taken at the national level. The current attitude, still held by some governments, that assistance for uprooted is not an essential item in their national budgets or funding requests, is detrimental to the continuation of current efforts and the gains made under CIREFCA.

(244) Following are recommendations for solutions:

• UNHCR and UNDP should implement strategies to bridge the relief to development divide by requiring the presence of development experts in emergency planning and implementation;

• All Project Planners and implementors, should ensure that projects be presented in a framework of sustainable development, with particular attention to environmental concerns;

• Agencies should collaborate early in the documentation of uprooted persons, bearing in mind that this facilitates their voluntary return and freedom of movement;

• From the QIP example in Nicaragua, UNHCR should learn to involve both new and old residents of a community in reintegration projects to facilitate reconciliation and future project maintenance;

• Repatriation strategies should consider the viability of non-rural and non-agrarian options, as well as rural agrarian approaches to (re)integration;

• UNDP should be working on operational strategies to link its planning and technical assistance at the national level with its own and others' efforts to implement development projects at the local or departmental level; in this regard, an institutional evaluation of PRODERE's impact would prove useful to planning in UNDP and to development intervention generally by governments and NGOs.

• Governments should be encouraged to invest also from their own resources to cover programmes that incorporate uprooted populations into national development plans;

• International agencies and governments should consider a new balance between the need to create ad hoc NGOs to implement projects and the need to build the capacities of local institutions; over time the balance should favour the strengthening of local capacities;

• Attention in funding should be given to the administrative and technical training needs of implementing agencies. Funding levels should be concomitant to the capability of a given agency to administer the funds;

• UNHCR and UNDP are encouraged to build upon their efforts, with the constructive assistance of UNIFEM, to address women's specific needs in their programmes and to promote such attention within their respective institutions and on the part of governments and private agencies with which UNHCR and UNDP collaborate.

• Both UNHCR and UNDP are encouraged to continue their attention to refugee and displaced children and to promote even greater assistance and protection to the needs of unaccompanied minors.

CONCLUSIONS AND RECOMMENDATIONS

REGIONAL ASSISTANCE PROCESS

(245) CIREFCA is not a "model, " but it does provide inspiration for resolving the problems of displacement in other regional conflicts. Elements of the process should be replicated only when adapted to meet specific local conditions. CIREFCA was inextricably linked to the on-going peace process in Central America, which accounts for many of its successes. Nonetheless, elements of CIREFCA, including PRODERE and QIPs and a functioning joint support unit, are being replicated with apparent success.

(246) The international community should be attentive to the possibilities of political commitments to extend protection and support for integration to uprooted populations, even before peace is signed. Also, new attempts to replicate CIREFCA's elements should include flexible tracking mechanisms to best coordinate and evaluate the effort. Governments should be encouraged to link ongoing planning in development ministries with the often ad hoc agencies providing assistance to the uprooted.

INTER-AGENCY COLLABORATION

(247) Despite difficult relations throughout much of CIREFCA, UNHCR and UNDP are able to work well in some circumstances where unity of purpose has been achieved through personal contacts between staff or through commitment to CIREFCA's goals. The JSU represents the most unified and cohesive example of this collaboration.

(248) UNHCR and UNDP should promote strategies of joint posting or other joint exercises to bring staff together and allow each to better understand the mission and operations of the other. Additionally, the establishment of an inter-agency team would facilitate and standardize communication between the two agencies on its joint programmes and promote further collaboration.

UPROOTED PEOPLE AND GOVERNMENTS

(249) CIREFCA attracted resources to the region which went to benefit uprooted populations. A minimum of US$250 million were raised through CIREFCA and related processes, in addition to PRODERE funds, which were secured before CIREFCA with a US$115 million pledged by Italy. A of 126 projects were funded through governments, international agencies and NGOs. The latter received unprecedented recognition in CIREFCA, ultimately most of CIREFCA funds, either directly or indirectly. Donors effectively pressured governments into allowing NGOs into the consultation processes in an effort to promote reconciliation and growth of democratic systems.

(250) In a world of competing demands, new campaigns to work with marginalized and uprooted people need to prioritize and present coordinated efforts. The CIREFCA process required substantial resource commitments to ensure the political commitments and collaboration of all actors. While humanitarian assistance in conflict situations often takes on political overtones, it should not be used as a channel for expressed political activities.

THE PROTECTION OF DISPLACED PEOPLE

(251) CIREFCA raised awareness about the rights and needs of uprooted peoples. Legal advances were made to protect refugees directly; repatriates and internally and externally displaced persons benefited indirectly from their coverage in CIREFCA. CIREFCA encouraged governments to sip the covenants, promoted compliance with international commitments, and inhibited gross violations and manipulations of these populations.

(252) Sustained efforts should be made to secure legal advances, especially with regard to internally displaced persons. Additionally, all CIREFCA actors should promote greater coherence between protection commitments and day-to-day practice.

DURABLE SOLUTIONS

(253) CIREFCA's project side was poorly coordinated. CIREFCA projects did encourage (re)integration of uprooted populations and wider participation by beneficiaries, non-government organizations and governments, but it appears to have had only partial success in promoting local governmental capacities to respond to these emergencies and in promoting projects consistent with sustainable development. It cannot be known how much would have been accomplished anyway, without CIREFCA, and it remains to be explored how much these projects alleviated the critical needs of uprooted populations. Despite policy-level commitment, UNHCR and UNDP were not able in most cases to make operational a strategy that linked relief to development efforts, and therefore, the continuity and sustainability of some efforts were jeopardized. However, the lack of data made definitive statements about CIREFCA's impact impossible.

(254) CIREFCA's efforts were scattered throughout the region, thus, impacts were quite localized. Most projects focused on repatriation to countries of origin, in order to support peace, reconciliation, and reconstruction. Displaced persons were attended only with respect to their location in areas of repatriation. Attention to externally displaced persons was ad hoc, framed only in projects intended to integrated refugees and local populations.

(255) UNHCR and UNDP should develop and implement strategies to operationalize the relief to development continuum. The sustainability of the projects and development proposed will be enhanced through the promotion of participation of the beneficiaries themselves in the design and implementation of solutions. CIREFCA should arrange for a current assessment of the needs and possibilities of assisting internally displaced persons through regional consultation and collaboration.

CIREFCA'S FINAL YEAR

(256) During 1994, CIREFCA should focus on a number of unresolved and unaddressed issues.

(257) Priority should be given to the situation of the uprooted Guatemalans, who face continued uncertainty about their options. The Guatemalan case represents CIREFCA in a microcosm and prospects for improving agency collaboration and coordination are already evident.

(258) A regional conference should be called to address the needs of internally and externally displaced persons. The forum, which would highlight the conditions of this little-studied population group and propose concrete action and follow-up, would begin to rectify the imbalance in attention given these priority groups until now under CIREFCA.

(259) In a forum such as this, with CIREFCA's tendency to de-polarize issues, land questions may also be raised, as they represent one of the keys to solutions for these and other populations in poverty. Participation by the affected populations would be essential for success of such an endeavour. In addition, as a part of the preparations, unfunded or partially funded projects directed at the displaced should be reviewed, revised for presentation, or new projects elaborated for funding.

(260) Both of these priority areas should be pursued in the spirit of the relief to development continuum and involve both UNHCR and UNDP.

(261) UNHCR and UNDP should take immediate steps to establish mechanisms of joint cooperation so that development experts will be present as resource persons in the planning and implementation of emergency aid, in attempts to implement the relief to development continuum.

(262) The Joint Support Unit should continue its valuable work of coordination and reporting on CIREFCA.

(263) In order to finish the remaining year of CIREFCA, UNHCR and UNDP should work closely at the headquarters level, in country, as well as in the JSU to ensure the transition of CIREFCA projects. An inter-agency team responsible for coordinating and communicating agreements might be established. In addition, some documentation of inter-agency discussions and agreements should circulate to key headquarters and country staff, which may inspire cooperation at other levels.

(264) The two agencies should come to an agreement on the priority and feasibility of pending projects within the context of their other regional and international work. Those priority projects should then be presented to funders.

(265) The JSU should step up its attempts to organize information about the funding and projects for uprooted people in the region, as a way of facilitating the most efficient and effective responses. However, it should not attempt another technical evaluation of the impact of CIREFCA at the project level without the appropriate data.

(266) The current lack of information about project funding and the controversy over which projects should be considered CIREFCA are just two of the more difficult obstacles to carrying out such an evaluation. Additionally, projects presented and funded are a reflection of the interests of governments and donors as much as they may be a response to demands or needs in the community.

(267) UNHCR should continue its efforts to improve and share the methodology of the Quick Impact Projects used in Nicaragua. These efforts included the sponsorship of internal and external evaluation, seminars for agency and government officials (including field visits), exchanges of personnel involved in the development of the QIPs to other countries.

(268) These experiences should be made cross-regional, however, by conducting such seminars also in Geneva and by sending the original QIP team members on temporary duty to countries considering applying the methodology. Donor interest in QIPs should extend to this type of training.

(269) UNHCR should consider the expanded use of its dual personnel strategy, that rotates some personnel but allows others to build up time, experience and contacts in a region that displays a protracted refugee crisis. Such a dual strategy allows for cross-regional exchange of ideas which was important in the very genesis of CIREFCA. Moreover, it would also permit others to gain an in-depth knowledge of a country or region to best respond and carry forward processes such as those begun under CIREFCA.

(270) The importance of key UNHCR officials in the start-up of CIREFCA calls into question the wisdom of frequent rotation of personnel. CIREFCA and its results owe their existence to a corps of seasoned professionals with many years experience and even deeper knowledge and understanding of Central America. One contributing factor to their success, but by no means the only one, was that many of these individuals were Latin American or Hispanic, for whom there was no need to translate language or culture.

(271) As UNDP proposes to make Human Development the framework for its next step in CIREFCA and beyond, it must first look inside the institution to build consensus around this new approach to meet modem issues of poverty, exclusion and politics; it must allocate appropriate funding and fund-raising efforts to the drive; and it must make the structural changes necessary to permit application of the framework at the grassroots, as well as national levels.

(272) UNDP's efforts to extend the merits of PRODERE to other regions in innovative cross regional programmes should continue. The effort will also provide insights for overall improvements in the programme.

(273) UNDP must reduce the layers of bureaucratic oversight that may be appropriate in other peacetime development conditions. In particular, field staff must be given greater latitude in decision-making and application of funds to respond effectively to constantly changing situations.

(274) Before the termination of PRODERE in 1995, UNDP should work to integrate the field capacity achieved by that regional team. Ibis may require some restructuring to ensure compatibility between national planning work and field operations.

---

1 For other official evaluations conducted on CIREFCA see, Jorge Santiestevan, CIREFCA y la application de su Plan de Acción, Consultant Report, mimeo, 24 February, 1992; Patricio Auda, Estudio de los impactos principales de los proyectos en ejecución que forman parte del Plan de Acción de CIREFCA, Consultant report to JSU, February 1992; Bryan Deschamp, The Role of Follow-up and Promotion Mechanisms in the CIREFCA Process and of Inter-Organizational Cooperation in Support of that Process, Report, February 1992; CIREFCA/CS/92/10 March 31, 1992. Outside assessment were conducted by Dennis Gallagher and Janelle M. Diller, CIREFCA: At the Crossroads Between Uprooted People and Development in Central America (Washington DC: Commission for the Study of International Migration and Cooperative Economic Development), No. 27 (March 1990); Adolfo Aguilar Zínser, CIREFCA: The Promises and Reality of the International Conference on Central American refugees (Washington, DC: CIPRA/Georgetown University), 1991.

2 CIREFCA/89/14, pp.6-7, items 4-5.

3 CIREFCA/89/14, February 1989, p.7; CIREFCA/89/3, March 1989. P.4; CIREFCA/89/8, February 1989, p. 2.

4 CIREFCA/89/4, p.32; By externally displaced persons, CIREFCA refers to those undocumented persons are outside their countries of origin and for various reasons did not seek or receive refugee status.

5 UNHCR, Cartagena Declaration on Refugees, 22 November 1984; Organization of African Unity, 0AU Convention Governing the Specific Aspects of Refugee Problems in Africa, adopted in Addis Ababa on 10 September 1969 (UNTS/45/14691)

6 UNHCR, Report, Advisory Group on Possible Solutions to Central American Refugee Problems, Geneva 25-27 May 1987, UNHCR/50B/17/87; GE.87-01381/8877E.

7 The developments in the Central American peace process were followed by the General Assembly of the United Nations. In 1987, the General Assembly expressed its support for the peace process, and in particular to the commitment in Esquipulas H to refugees (GA42/110, December 7, 1987) and in 1988, the Special Plan of Economic Cooperation for Central America (PEC) (GA42/23 1, May 12, 1988).

8 GA/43/118, February 17, 1989.

9 CIREFCA, implementation of the Concerted Plan of Action of CIREFCA: From Conflict to Peace and Development, CIREFCA/CS/92/10, March 31,1992.

10 In the case of ICARA I and II, regional efforts to respond to the refugee crises in Africa, the objectives were humanitarian and related to coordination of relief and development. There were no peace processes to influence, nor the national political will to attend to refugees, then considered a mere symptom of larger political conflicts.

11 CIREFCA/89/14, p.9, Item 16.

12 12 UNHCR, International Conference on Central American Refugees, Evaluation Mission, 8-21 December 1988, San Jose, 3 January 1989; and UNHCR/TTS Mission Report 43/88 Geneva.

13 UNDP/PEC, PEC 1989-1991: Achievements and Prospects. NY: UNDP, August 1992, Annex I.

14 UNHCR/UNDP, Joint Support Unit, Estado del financimiento y localizacion de los proyectos CIREFCA y de otras iniciativas en el marco de CIREFCA, San Jose, February 1993

15 Roger Guarda and Philippe Lavanchy, Note for the File: Meeting between UNDP and UNHCR on the CIREFCA Profiles, Geneva, March 17,1989.

16 CIREFCA, From Conflict to Peace and Development, 1992; Asociación para el Avance de las Ciencias Sociales en Guatemala (AVANCSO), ¿Donde esta el futuro? Procesos de reintegración en conumidades de retornados (Guatemala: AVANCSO), 1992; Marvin Ortega, Reintegration of Nicaraguan Refugees and Internally Displaced (Washington DC: CIPRA/Georgetown University), 1991.The Permanent Consultation on Internally Displaced Persons in the Americas, an informal group composed of international agencies (including UNHCR and UNDP), non-governmental organizations, and individual experts, was formed in 1992. The Consultations's objective is to provide concerted action in response to the problems presented by displaced populations in Latin America. The Inter-American Institute for Human Rights serves as the Secretariat to the Consultation. See Inter-American Institute on Human Rights, Memoria, 1993; Roberta Cohen, Human Rights Protection for Internally Displaced Persons. Washington, DC: Refugee Policy Group, 1991.

17 Before the final version of this report was completed, evaluators were informed that a re-priontization and technical review of unfunded projects had been conducted to the agreement of both institutions.

18 The JSU provided updated 1992 data as the evaluators were finishing the final version of this report. Through 1992, actual contributions to CIREFCA by Sweden totalled $56.2 million and by WFP, $33.1 million. Such corrections late in the process highlight the difficulty of getting accurate, specific data in this evaluation. It is also illustrative of the fluid basis upon which the JSU has had to operate throughout CIREFCA.

While these differences do not significantly change the analysis herein, they do slightly affect calculations regarding numbers of projects, implementing agencies and sector. These corrections are not incorporated in the text at this late time, as they would necessitate recalculation of all project lists by implementing agency, country and sector.

19 Before presentation of the final version of this report, evaluators received updated financial information in February and March 1994 from the JSU Data Bank; the additions do not substantively affect the analysis in the report. According to these data, additional funds mobilized in 1993 total a i i of US$45.6 million. Approximately one-half of the funds mobilized in 1993 came from the European Union (formerly European Community) that is, $20 million ECU (U.S.$23.5 million).

Thus, the total mobilized in the framework of CIREFCA through 1993, with available information is amended to US$411 million, including PRODERE.

With these additional pledges, the corresponding changes to the table of top donors through 1993 is as follows:

Major Donors

Donor	Committed US$ (millions) through 1993
Italy European Union Sweden World Food Programme Provisional Total (Minimum)	115 115.5* 60.7** 34.7*** 321.9

Sources

* European Union, Belgium.

** Update by JSU; does not include funds directed to non-government organizations.

*** Update by JSU.

By the time of the next meeting of the International Follow-Up Committee, the European Union will move into the top donor position after consideration of 1994 contributions. Total contributions from Sweden and WFP will also increase in 1994.

Even these latest totals are provisional and apparently incomplete. They are subject to revision as new information is made available to the JSU.

This latest updating of information by the JSU reveals how insecure their reporting base has been throughout the CIREFCA years, for lack of discussion or agreement on procedures at the outset of CIREFCA. Some donors have been forthcoming and regular in their voluntary reporting about their contributions, while others have been less than forthcoming. Oftentimes, data presentations confused "proposed" contributions with "actual" contributions without annotation.

20 UNHCR/Belize, Field Survey of Refugees and Immigrants from Central America in Belize (Belmopan: UNHCR), February 1993. UNHCR/Belize, 1993.

21 UN General Assembly A/42/949, April 26, 1988.

22 UNHCR, Note on International Protection, A/AC.96/799,25, August 1992.

23 See Note 16.

24 Inter-American Institute for Human Rights, Undocumented and Illegal in Central America (Washington, DC: CIPRA/Georgetown University), 1991.

25 The International Convention on the Protection of the Rights of All Migrant Workers and Members of their Families is reproduced in the International Migration Review, Vol. 25, No. 4, Winter 1991.

26 See, Jeff Crisp and Lowell Martin, Quick Impact Projects: A Review of UNHCR's Returnee Reintegration. Programme in Nicaragua, UNHCR/EVAL/QIP12, November 1992

27 Centro de Investigación y Asesoría Socio-Económica, Instituto de Promoción Humana, and Instituto de Investigación Itztani, Evaluación de los proyectos de unpacto rapido (PIR) en Nicaragua, Final Report to UNHCR, Managua, May 1993.

28 See, Aguilar, op. cit.; Sergio Aguayo, From the Shadows to Center Stage: NGOs and Central American Refugee Assistance in the 1980s (Washington DC: CIPRA/Georgetown University), 1991.

29 See Note 19.