UNHCR's Policy and Practice Regarding Urban Refugees, A Discussion Paper


This Discussion Paper is based upon extensive interviews at UNHCR Headquarters, a review of project documents, and an examination of other relevant background literature. It attempts to identify, on a global basis, the main issues of current concern to UNHCR with regard to its work with urban refugees. It assesses the appropriateness of current policy and practice and examines some of the principal problems experienced by field offices around the world. In conclusion, it presents a number of recommendations for the establishment of a more rational and humane approach to this small but complex area of UNHCR's work. The paper has been prepared by Christine Mougne, Senior Officer, Inspection and Evaluation Service.


(1) Unequal development and a growing North-South divide have contributed to a substantial growth in economically driven migrations in the 1990s. Increasingly, such movements are from rural to urban areas. Urban refugees and asylum seekers tend to be influenced by some of the same push and pull factors, which can sometimes make it difficult to identify them from amongst the massive populations of illegal migrants now found in Eastern Europe, the CIS, Southern Africa and elsewhere.

(2) Meanwhile, UNHCR is seeking increasingly to identify ways of responding to ever-growing refugee problems in a broader context, by addressing the needs of entire caseloads rather than dealing with individuals or small groups. Financially, such an approach is becoming inevitable since UNHCR is no longer in a position to continue to support individual refugees as has been evidenced by the difficulties encountered in recent years in seeking funding for the Education Account. While donors are willing to support activities leading to genuine integration or for voluntary repatriation, they may no longer be prepared to support long-term care and maintenance of urban cases. This places UNHCR in a dilemma. Whilst there is no mandate to assist, the question arises as to how broadly the organization should interpret its mandate to protect. Is there a moral obligation, within the framework of providing protection, to keep refugees from falling into destitution? This dilemma lies behind much of the discussion which follows.

(3) UNHCR has no consistently applied definition of an urban refugee. Definitions used vary, not only from region to region, but even between staff members who tend to apply subjective rationale to their personal definitions. Perhaps the most commonly articulated definition of an urban refugee is that of an individual of urban origin, usually a student, former politician, a professional, a trader or a skilled, non-agricultural labourer. In other words, anyone who is not a farmer or a peasant. To this definition is linked the commonly held belief that such individuals would find it difficult to survive in a rural camp or settlement and therefore require individualised assistance in an urban setting.

(4) An important distinction should be made between refugees and asylum seekers from neighbouring countries or from countries within the same region or sub-region, and the refugee groups with a tendency to travel beyond their region of origin. The latter include, notably, Somalis, Zairians, Ethiopians, Eritreans, Afghans, Tamils, Iraqis and Iranians, most of whom have a long history of migration related to trade, and/or a nomadic tradition. Such groups consequently tend to have well-established links world-wide, a factor which inevitably facilitates their intra-regional movement.

(5) An examination of some of these refugee groups reveals a number of interesting patterns which indicate that far from being random and spontaneous, such movements are often organised. For example, some of the movement (such as the Tamils and Iranians) may be seen as being politically manipulated, with small groups spreading out to new asylum countries to establish a fund-raising base in support of the resistance. Another important aspect of the extra-regional movement of refugees and asylum seekers is the existence of a network. The arrival of extra-regional asylum seekers in Phnom Penh during the past year, is a clear example of this international "trade in refugees'.

(6) Urban refugees, while constituting less than 2% of UNHCR's refugee caseload (and less than 1% of the total caseload of concern to the High Commissioner), demand a disproportionate amount (estimated at between 10-15%) of the organization's human and financial resources. Working with individuals is inevitably much more time-consuming than with large groups. However, a number of factors conspire to further unbalance this equation. Urban refugees and asylum seekers tend to include a wide variety of people, some, but by no means all, of whom have genuine refugee claims. They include opportunistic and dynamic individuals as well as those who have failed to survive as part of the normal migration (or refugee) flow - the maladjusted, the social outcasts, etc. - a factor which can make status determination particularly difficult. Furthermore, since such movement is often stimulated, at least partially, by a desire to improve their economic potential, urban refugees and asylum seekers tend to share a culture of expectation, which, if not satisfied, often leads to frustration and violence.

(7) Despite the qualitative importance of urban refugees world-wide, UNHCR has a tendency to neglect such caseloads particularly in situations where there are large rural caseloads within the same asylum country. Organizational policy regarding urban refugees is particularly weak and unclear1, and practice, in terms of both protection and assistance, tends to vary substantially between and within regions and sub-regions. A common factor contributing to this situation is that wherever major refugee populations are present, small urban caseloads tend to be given low priority. In some situations, such as in parts of South-east Asia and southern Africa, there is a perception among some UNHCR staff that urban caseloads may have been used, until recently, as a bargaining point with host governments, whereby special concessions were made in an effort to reach agreement on arrangements for the larger influx. As a consequence, compromises may have been made with regard to UNHCR's approach to the urban caseload, and "bad habits" set in.

(8) At a later stage, when the major refugee populations leave (as recently, for example, the Vietnamese in South East Asia and the Mozambicans in southern Africa), UNHCR's focus tends to turn to the urban caseloads, rather like rocks appearing as the tide falls. The need to redress the damage done by inappropriate compromises reached in previous years is then apparent2. However, since a durable solution has been found for the major refugee population, programming capacity in branch offices tends to shrink. Consequently, UNHCR is often left without the experienced staff required to respond appropriately to the complex needs of a small, neglected, urban caseload.

(9) As noted, this study has identified little, if any, documentation on organizational policy towards urban refugees. A search of IOM/FOMs and ExCom Conclusions revealed nothing on the subject other than the 1979 and 1989 Conclusions on Irregular Movers. The former Central Evaluation Section and the Programme and Technical Support Service have produced an impressive number of reports on urban refugees over the last decade (see Annex 2). However, since neither of these sections has a mandate for policy development, it is perhaps not surprising that an extensive review of these internal reports indicated that their recommendations dealt exclusively with the specific objectives of the programmes under examination, and made no reference to broader policy issues.

(10) Guidelines for practice with regard to urban refugees currently exist only in the social services realm, namely the "UNHCR Handbook for Social Services" (published in 1984) and the more recent "Community Services for Urban Refugees" (1994). While a brief mention is made to "Individual/family support" in the UNHCR Manual (Chapter IV, Sector H, 96), including standards for calculating assistance rates, and time limits for such assistance, the 1995 project descriptions for urban refugee assistance programmes in different regions indicate that such standards are rarely applied. Neither the 1982 Handbook for Emergencies nor the 1991 Social Services in Refugee Emergencies includes any specific reference to urban refugees, although the latter includes some useful general guidelines on working with groups requiring special attention, focusing on working towards self-reliance and self-help from the beginning of an emergency.

(11) In the absence of a comprehensive policy on urban refugees, and given the marginal importance ascribed to urban assistance programmes at times of major refugee influxes, it is not surprising that practice varies enormously, not only between regions, but also within regions and sub-regions. Significant variations in the way in which urban refugees are defined, protected, assisted and offered durable solutions, inevitably create pull factors, with access to better or longer material assistance, particularly tertiary education, and resettlement, being among the principal factors involved.

(12) Variations in standards of assistance and protection between and within regions are a major factor contributing to the important phenomenon of irregular movement3. While irregular movement has been receiving particular attention in recent years, it is far from a recent phenomenon; it was mentioned with concern during discussions on international protection at the thirty-fifth session of the Executive Committee in 1984 and was the subject of a major study conducted the following year. ("Study of Irregular Movements of Asylum Seekers and Refugees", prepared by Gilbert Jaeger, August 19854). Ten years later, the problem of irregular movement continues unabated, but few, if any, of Jaeger's recommendations have been implemented while his study has largely passed into oblivion.

(13) The only other source of practical guidance in dealing with the problem of irregular movement which has been identified by the present study, is in Chapter 5 of the 1984 Social Services Handbook. Section 5.3.4 addresses the issue of "Handling Cases of Constantly Migrating Refugees". After identifying some of the principal causes of such "orbit" cases, and emphasising the need for regional coordination as a preventive measure, the following guideline is given:

"Unless there are strong reasons to the contrary and agreement between field offices concerned, no financial assistance would be provided to a refugee who has left the country of asylum on his/her own initiative. The only exceptions would be to aid return to the country of asylum or voluntary repatriation." (emphasis added)

Again, this guideline has been largely overlooked, and UNHCR staff still debate the appropriateness of such an approach.

(14) A question raised by a number of UNHCR officers is whether or not refugees have the right to expect assistance in urban areas? This relates back to the point raised earlier on how broadly should UNHCR's protection mandate be interpreted (para.2). There is a general reluctance to take a position on this particular matter, perhaps because it is seen as possibly impinging on the principle of freedom of movement. Interestingly, the organization is quite clear with regard to rights of location in the context of voluntary repatriation, but not in the case of refugees and asylum seekers in a country of asylum. In fact, the focus for UNHCR should be placed not so much on where an individual refugee or asylum seeker chooses to reside, but where the host government is prepared to provide assistance and/or where it chooses to establish refugee status determination procedures.

(15) A number of experienced protection officers point to the fact that residence in an urban area is liable to create more protection problems (and work for UNHCR) than it resolves: status determination, detention, papers, freedom of movement issues, etc. It also becomes an economic problem as refugees have to compete for employment. As a logical conclusion, many would argue that assistance should only be provided outside urban areas. Others in the organization, however, maintain that individuals with an urban background cannot be expected to survive in rural settlements and should therefore be provided with whatever assistance is needed to permit them to make a living in town. It is this approach which has recently led to UNHCR resisting host governments' efforts to relocate refugees away from urban areas in a number of southern African countries.5

(16) At this stage the question becomes one of equity. For example, a Somali refugee who manages to leave Dadaab for Nairobi, or continues from Nairobi to Lusaka, New Delhi or Kuala Lumpur, can expect a very much more comfortable life style, thanks to UNHCR assistance, than his compatriots who have remained in northern Kenya. This discrepancy is not simply a matter of differences in the cost of living which require different assistance levels, but is more a question of the scope of assistance provided. Should UNHCR "reward" those who have the financial and personal means to move on from a rural settlement to the city, or from one country of asylum to another, in order to seek better conditions and prospects? Again, this is not a question about freedom of movement, but one of equity. If conditions for Somalis in Kenya are so poor as to justify such irregular movement, should UNHCR not be working to improve conditions for all camp residents rather than providing individual assistance to the privileged few who manage to reach further shores?

(17) As noted elsewhere, it is often those individuals who succeed in moving from one country to another in the hope of gaining a better standard of assistance or access to resettlement, who become aggressive and violent if their expectations are not met. This phenomenon has become a feature of urban refugee programmes in all regions of the world, with hunger strikes, demonstrations, physical assault of UNHCR and operational partner staff, damage to and/or occupation of office premises, and suicide threats now commonplace. While irregular movers are often among the most vehement of protestors, others might include rejected cases, the un-assisted, or the psychologically disturbed. It should be remembered that occasionally violent behaviour can reveal a history of torture - but UNHCR rarely commands the necessary professional skills to make such a determination. Regardless of the cause of the violence, there is no clear policy on how to respond, although individual violence is mentioned briefly in both the Emergency and Social Services Handbooks.

(18) Views differ greatly on this issue. While some believe that UNHCR has the obligation to do everything possible to resolve such incidents to the benefit of the refugees, others have learned from experience that giving in to violent forms of protest does not pay. The events in Islamabad early in 1995 are a case in point. Successive hunger strikes by a group of urban Iraqi refugees, each involving greater demands than the last, were all eventually conceded to by the government and UNHCR, giving the impression that this would always be the case. On the most recent occasion, however, when the demands made were beyond both the government and UNHCR's capacity to respond, the hunger strike continued until one of the protesters died. These events had repercussions among Iraqi caseloads elsewhere, including Moscow, where a carbon copy hunger strike took place with resettlement as the principal demand, and in Beijing, where a group of Iraqis demonstrated in front of the UN compound, occupied UNHCR premises, and made death threats against staff, in solidarity with their compatriots in Islamabad and demanding their own resettlement.6

(19) UNHCR's failure to address appropriately violent protest among urban caseloads can jeopardise the organisation's ability to protect the refugees concerned. In a number of countries in southern Africa, the violence has produced a backlash from both the government and the general public. The result has been a reaction against the presence of refugees in the urban areas, and a push to move them out into rural centres.

(20) In the view of some experienced protection officers and other long-serving colleagues, in situations where there are no asylum problems and where the needy are receiving assistance, UNHCR should not show weakness and give in to such demands, but should simply close down the office and refuse categorically to cooperate in any way with those making threats of, or committing acts of violence. On the other hand, where refugees and asylum seekers protest in the face of legitimate protection concerns, it is, of course, incumbent on UNHCR to 'intervene'.

(21) Finding a solution to the growing violence amongst urban caseloads is not the focus of the current review. It is, however, an important and dramatic symptom of the broader problem under discussion, that of inconsistent practice, unclear policy, and an ad hoc personalised approach to dealing with individual cases. There is ample evidence of the damaging effects of care and maintenance programmes for urban refugees; how open they are to abuse, how they create dependency and how they lead to inequalities between refugees and host populations. Inconsistency of practice has only served to compound the problem. Given the considerable financial and human resources currently invested in urban programmes it is essential that steps be taken to rationalise the approach without delay.

(22) Not surprisingly, concern has been growing during the past few years over the dearth of policy guidelines regarding urban refugees. UNHCR offices in southern Africa and in South and South East Asia have been facing particular difficulties recently with growing numbers and with mounting violence among urban caseloads. Two regional workshops have been held to address some of these concerns: one in Harare in May 1994 and the other in Kuala Lumpur in May 1995. Both workshops produced detailed reports and extensive recommendations, some of which are reflected and expanded upon in the following paragraphs.


(a) In order to ensure a more consistent and equitable approach to the protection and assistance of urban refugees and asylum seekers, UNHCR should seek to establish a clear, rational, and comprehensive policy, based on principles which are applicable and acceptable on a global basis. Implementation of the policy will necessarily involve adaptation of certain elements to regional or sub-regional conditions, according to variations in socio-economic, political and security factors as well as conditions of asylum.

(b) A working group should be established at Headquarters before the end of 1995 to develop policy guidelines on urban refugee protection and assistance, including guidelines for their implementation, for field testing, within six months.

(c) Once established, the details of UNHCR's policy on urban refugees and asylum seekers must be systematically communicated to all players, including refugees and asylum seekers, governments, NGOs and UNHCR staff at Headquarters and in the field. The method of communication should be adapted according to the country, region and target group to ensure that all concerned are fully aware of its contents and implications (see recommendation (h) below).

(d) The UNHCR policy on urban refugees and asylum seekers should be developed on the basis of the principal that assistance should never be provided as a substitute for protection; strenuous and continuing interventions should be made with host governments (particularly, but not exclusively, those which have signed the Convention and the Protocol) to encourage them:

  • (d.1) to accord refugees access to the labour market
  • (d.2) to accord refugees access to national hospitals, schools and other social services (at rates equivalent to those paid by nationals)
  • (d.3) to accord refugees married to nationals access to the naturalisation process


(e) Redirecting long-established practice cannot be achieved overnight. A number of stages will therefore be required, on the one hand to implement the new policy proactively, and on the other, to address some of the problems created in the past. As a first step, where new refugee situations arise involving movement into urban areas, UNHCR should carefully examine all the implications before considering any intervention in the urban context. Appropriate policy guidance and training to field staff should be provided from Headquarters where necessary in order to avoid inadvertent and undesirable precedent-setting.

(f) In order to promote solutions for urban refugees in existing UNHCR programmes, and to break the vicious circle of assistance and dependence, current caseloads should first be analysed to distinguish the different categories involved - individuals who form part of a prima facie caseload in the same asylum country, irregular movers from other asylum countries, high profile political or security cases, etc. This will facilitate the identification of those individuals with a genuine need for protection and assistance in an urban area of the country concerned, and allow for appropriate action to be taken towards solutions (see below). Once implemented, this approach should result in significant reductions in the urban caseload in a number of countries and regions.

(g) For refugees who are part of a prima facie caseload, it is recommended that, as a general principal, no assistance should be provided in urban areas in the country of first asylum7 (or, in accordance with (f), in any other country of asylum to which they might subsequently travel). This approach has already been adopted by a number of field offices over the years, with positive results. Individual refugees wishing, nevertheless, to reside in an urban area (provided this is in accordance with host government policy on the matter) should be clearly informed that no assistance will be available (with the possible exception of life-saving medical care) other than for refugees residing in a camp or rural settlement in which there is an active assistance programme.

(h) The problem of irregular movers will require both preventive measures to discourage future movement as well as curative measures to deal with individuals who have already moved irregularly. With regard to preventive measures, on the basis of a positive experience with mass information campaigns to discourage irregular movement from Viet Nam, Albania, Russia and elsewhere, it is proposed that an information campaign be launched both in countries of first asylum from which irregular movers tend to come and in those countries which tend to attract irregular movement. The information campaign should advise that as of a certain cut-off date, no further material assistance (including education assistance) will be provided to those moving irregularly beyond their first country of asylum8, nor will they be granted access to resettlement processing. Information on conditions in target countries of asylum should also be included.

(i) In order to further discourage future irregular movement, the following steps should be considered as a means to address some of the factors identified as encouraging such movement:

  • (i.1) no further registration of irregular movers in second or subsequent asylum countries
  • (i.2) no further issuance of "attestations" or "to whom it may concern" documents to individuals not fulfilling the criteria for refugee status in the asylum country
  • (i.3) further elaboration of a system to monitor the refugee status determination process in field offices to ensure regional or sub-regional consistency, as well as enhanced dissemination of country of origin information and guidelines on specific caseloads;
  • (i.4) consistent application of UNHCR's guidelines on resettlement
  • (i.5) exclusion of irregular movers from resettlement processing; this will require comprehensive briefing on the issue to resettlement governments in order to elicit their cooperation9
  • (i.6) establishment of clear guidelines on the provision of assistance, specifically excluding irregular movers from assistance (with the possible exception of life-saving medical care);
  • (i.7) establishment of consistent regional (or sub-regional) policy with regard to the provision of secondary or tertiary education grants and the specific exclusion of irregular movers from such assistance.
  • (i.8) further development of mutually-compatible regional or sub-regional IC databases for urban cases (eg. CEERICS) with the facility to cross reference between regional databases for those groups commonly travelling beyond their region of origin (eg. Somalis, Iranians);

(j) For existing irregular movers registered with UNHCR, steps should be taken to promote their return to the country in which they had found protection or, where appropriate, to the country of origin. Pending return, no further assistance should be provided (with the possible exception of life-saving medical care) other than to facilitate travel to the agreed destination. The support of governments in both first and second countries of asylum (and/or the country of origin) is clearly a prerequisite, and note should be taken of the recent successful cooperation between the governments of China and Pakistan over the return of a group of Iraqi refugees who had moved irregularly from Karachi to Beijing in search of resettlement opportunities.

(k) The issue of the return of irregular movers to their country of first asylum (or country where basic protection was available) will require further discussion in-house to determine the appropriate extent of UNHCR involvement in promoting or assisting such returns, including the financial aspects.

(l) Once existing urban caseloads have been thoroughly reviewed, it will be possible to identify those individual refugees and asylum seekers who are not irregular movers or part of a prima facie caseload, and who have valid reasons for remaining in an urban area in the country of asylum concerned. The following recommendations for the provision of assistance for such individuals would also be applicable to new-arrivals who are not irregular movers or part of a prima facie caseload, and who have valid reasons for seeking UNHCR protection in the urban area concerned:

  • (l.1) as a first step, every effort should be made to encourage host government support to refugees; it should be recalled that UNHCR's mandate is to protect, not to assist refugees; rather than automatically providing assistance, UNHCR's focus should be on institution and capacity building in a country of asylum, to strengthen the national social services in order to facilitate the integration of refugees; however, where host government assistance is not forthcoming, or is inadequate, the following steps should be considered;
  • (l.2) in accordance with the 1984 Handbook for Social Services, emergency assistance to new arrivals pending status determination should be provided only in exceptional circumstances, if the protection officer considers that the individual has a meritorious claim and if s/he is found to be in need if assistance; such assistance should be minimal and time-limited;
  • (l.3) assistance should normally only be considered following a positive status determination which has been endorsed by UNHCR (ie. if a field office does not support a positive government determination, there is no obligation to assist);
  • (l.4) individual assistance should be provided only following a thorough needs assessment including home visits which indicate that an IC not only requires temporary assistance but that s/he can/will use the assistance towards finding a solution;
  • (l.5) assistance following positive status determination should be directly linked to solutions and efforts made from the outset to identify potential for economic insertion; assistance to urban refugees should never be embarked upon from the perspective of automatic "care and maintenance"; the solutions-based approach might include job placement schemes, apprenticeships, micro-projects, etc., the nature and scope of which would be determined on the basis of an economic analysis of the location concerned;
  • (l.6) provision of assistance should take into account support available from within the refugee community including family members as well as from the host government and charitable institutions in the host country and should, where feasible, be channelled through the refugee community or an appropriate national body rather than given to individuals;
  • (l.7) assistance to ICs should be reviewed on a regular basis (monthly or quarterly) to determine what efforts and progress are being made by them towards finding a solution, when assistance might be reduced or cut, and/or when other measures might be required;
  • (l.8) duration of assistance can be flexible, but must be limited (eg. maximum six months) to avoid dependency syndrome;
  • (l.9) an individual refugee who is unable to provide for him/herself because of physical or other incapacity and lacks family support, may require exceptional intervention; ideally family reunion would provide the best solution, but where this is not feasible, the host government should be requested to identify a solution, preferably outside of the urban setting;
  • (l.10) a rational basis for determining temporary assistance levels within each region or sub-region should be identified, such as a compilation of basic costs (rental, food, transport, etc.) which would be reviewed (regionally or sub-regionally) at least annually to keep pace with inflation or to respond to significant currency changes affecting costs;
  • (l.11) criteria should be developed for successful urban programmes so that individual programmes can be evaluated rationally.


(23) As noted earlier, UNHCR has no consistent definition of an urban refugee. While in many parts of the world, being of urban origin is a key criterion, in some regions, such as northern Africa and central Asia, any refugee living in an urban area is considered as an urban refugee, regardless of his or her origins. These discrepancies inevitably have an impact on how we respond to the refugees concerned and therefore deserve careful review. The manual "Community Services for Urban Refugees" (op.cit. 1994, p.5) is one of the few documents available which attempts a classification of some of the "main groups represented in urban refugee populations", as follows:

(a) people with an urban background in the country of origin

(b) people who have been politically active (politicians, members of government, students)

(c) professionals and people with higher education

(d) people with rural backgrounds seeking work or education (including those from camps if conditions there do no allow them to meet their basic needs)

(e) one-parent (female) families

(f) sick and disabled people who have been referred from camps and rural settlements for treatment, rehabilitation

(g) refugees who have left the country of first asylum

(24) While further categories might be added to this list, it warrants comment since those included reflect some of the prevailing biases and confusion which tend to shape UNHCR's response to urban caseloads in many parts of the world. Categories a - c include those, as noted earlier, who are traditionally viewed as "urban cases" and who may arguably have special needs (particularly category b), over and above those of a prima facie rural caseload.

(25) Categories e and f on the other hand, are generally considered as so-called "vulnerable" cases, for whom life in urban areas may well be significantly more difficult than in a rural settlement. In the case of medical referrals, these should normally be temporary, lasting only for the duration of treatment. Categories d and g, meanwhile, may be classified as self-selecting irregular movers who already had access to protection and assistance elsewhere, but have chosen to move on in search of something "better".

(26) As noted in the Overview, urban refugees known to UNHCR generally live alongside a much larger population of illegal migrants found in major urban areas around the world. Those individuals who choose to register with UNHCR and/or with host governments invariably have a specific reason for doing so. While in some cases this is certainly the need for protection, it is also, quite commonly, a desire to obtain assistance or access to resettlement in a third country.

(27) Urban caseloads tend to comprise predominantly young, single (or separated) males. Inadequate or incomplete registration of urban caseloads, however, makes it difficult to ascertain accurate statistics. Furthermore, poor registration systems in field offices often make it difficult to distinguish between active and non-active or closed cases, assisted and non-assisted, making comparative data very unreliable.

(28) An effort has been made in the context of this study to ascertain the total number of urban refugees registered with UNHCR world-wide. This has proved to be an almost impossible exercise. The 1994 Statistical Overview of Populations of Concern to UNHCR does not make a distinction between urban and non-urban caseloads. While raw data submitted by field offices to the Food and Statistical Unit each year do differentiate between urban and non-urban refugees, these figures are not always complete, and tend to be mutually incompatible.

(29) Confusion generally arises between four types of figure:

(a) the total urban caseload (which tends to be a very loose estimate)

(b) the total registered caseload (which may be a cumulative figure, not distinguishing between active cases and those who have left the programme over time)

(c) the total assisted caseload (which is sometimes inflated by double counting of individuals who receive more than one type of assistance), and

(d) the total caseload (rural and urban), registered or not

(30) This confusion also gives rise to a fundamental question: what is the appropriate criterion by which to count urban refugees? More often than not assistance is used as the basis, despite the fact that this might exclude substantial numbers seeking (or receiving) protection but to whom assistance is not provided. Thus, if the basis for counting was protection, instead of assistance, the numbers might be substantially higher. Nevertheless, in order to provide an idea of the scale of the urban refugee "problem" within the total population of refugees and persons of concern worldwide, the figures given below, which were provided for the purposes of this review, may be considered. While incomplete, the figures nevertheless represent two regions in which urban caseloads are currently of major concern, plus six of UNHCR's largest urban programmes in other parts of the world.

Number of Beneficiaries (Individuals)

Southern Africa (Botswana, Lesotho, Malawi, 14,215
Mozambique, Namibia, South Africa,
Swaziland, Tanzania, Zambia and Zimbabwe)

East and Southeast Asia (China, Hong Kong, 3,467
Indonesia, Malaysia, Philippines, Singapore,
Thailand and Cambodia)

New Delhi, India 11,295

Cairo, Egypt 8,000

Athens, Greece 7,326

Moscow, Russia 6,000

Rio, Brazil, 3,148

Islamabad, Pakistan 2,600

(31) These various caseloads total some 56,000 individuals. On this basis one can estimate that urban refugees world-wide might number in the region of 200,000 or approximately 1.4% of 14.4 million refugees worldwide or 0.7% of 27.4 million people of concern to the High Commissioner (UNHCR: A Statistical Overview, 1994).


(32) A major problem identified by this review is a serious lack of consistency in the application of convention and mandate status from one field office to another. UNHCR accepts responsibility for individual refugee status determination for a variety of reasons, either because a host country has not yet signed the Convention, because there are no national procedures, or because the procedures are new and not yet fully established. While there are ongoing efforts to address these shortcomings in many asylum countries around the world, UNHCR nevertheless continues to take primary responsibility for refugee status determination in many of the countries in which the organization has a presence. As a consequence, UNHCR is all too often considered, by refugees and host governments alike, as having exclusive responsibility for the protection of and assistance to refugees. The issuing of "attestations" and "affidavits" in particular, tends to raise expectations among refugees and asylum seekers, whilst providing host governments with a pretext to distance themselves from the problem.

(33) Within a UNHCR branch office, meanwhile, responsibility for individual status determination is commonly delegated to junior professional staff with little experience in refugee work. Decisions are often made quickly and under pressure, with little supervision or monitoring by the Division of International Protection. In spite of the guidance provided by the UNHCR Handbook, there is necessarily a personal element in each determination. Different schools of thought exist within the organization regarding the standard of proof required in status determination, so that discrepancies inevitably arise not only within offices, but between offices dealing with asylum seekers from the same country of origin. Protection officers consequently face considerable difficulty in managing their urban caseloads, particularly in ensuring fairness.

(34) In the case of refugees and asylum seekers who present themselves at a UNHCR office far from their country of first asylum, eligibility officers are often particularly handicapped, lacking detailed information on the country of origin. While occasional guidelines on specific caseloads prepared by the Division of International Protection and field offices in countries of origin, and the CDR country of origin database, can go some way to bridging this gap, there remains a problem of objectively applying available information to an individual case outside of the context of a broader outflow. In the view of some senior protection officers, UNHCR does not yet have all the tools needed for accurate individual status determination. As a result, there is a tendency to be over-generous, thereby jeopardising the fate of genuine refugees. The availability of more complete information on the country or origin will benefit refugees and staff alike, and will have a positive impact on access to asylum and to durable solutions.

(35) A serious lack of consistency is reported by protection officers in the application of criteria to refugees and asylum seekers from a particular country of origin who seek asylum in different countries and regions. The case of Afghans is most frequently cited in this regard. Discussion of the broader Afghan caseload also raises the issue of why individuals, who, in their country of first asylum are given prima facie refugee status, require individual status determination elsewhere? While the prevalence of manifestly unfounded and abusive cases among Afghans in the CIS countries, for example, justifies a consideration on the basis of individual merits, this is by no means always the case. The paradox is most patently evident when individuals from a major influx, enjoying prima facie status in their country of first asylum, are subjected to individual status determination when presenting themselves to UNHCR in the capital city of the same country (eg.Zaire, and until recently Rwanda and Burundi).

(36) An attempt to clarify why protection officers should seek to determine the individual status of refugees already enjoying prima facie status in a country of first asylum, reveals a disturbing link with assistance. Indeed, many protection officers, when asked why individual status determination is done in such cases, indicated that it was in order to identify those refugees who might qualify for assistance! Less surprising perhaps, is the response of a social services officer, when asked on what basis urban refugees are considered for assistance: that assistance depends on the conclusion of the protection officer. The lack of communication between protection and social services staff on individual cases continues to be a problem in many field offices.

(37) There are, however, other problematic links between protection and assistance. There is a rather obvious link whereby refugees or asylum seekers ask for protection as a means of gaining access to assistance, or to move on somewhere else, either through resettlement or by obtaining a travel document. A more fundamental problem identified by a number of colleagues is that UNHCR sometimes falls into a trap of providing assistance year after year in order to buy, and retain asylum. Particular instances cited by concerned colleagues include New Delhi, athens, Bangkok, Manila and others, where fundamental protection problems have remained unresolved year after year, and where UNHCR assistance has served to maintain urban refugees, recognised under the mandate, in an unsatisfactory and artificial state of prolonged temporary asylum.

(38) While the provision of assistance undoubtedly helps host governments to tolerate urban cases which they might not otherwise welcome, it does not encourage them to address their international obligations either by moving towards a signing of the Convention, or for those who have already signed, its implementation. Nor does it help the refugees themselves to fend for themselves, but instead puts them on the slippery slope towards dependency. Efforts to resettle such cases provides no solution to the problem either, since on the one hand resettlement allows the host government to absolve itself of any responsibility towards the refugees, and on the other hand it creates a pull factor.

(39) In an attempt to avoid these problems, UNHCR in the Russian Federation has taken the difficult decision not to undertake eligibility determination of individual urban cases, in order to maintain pressure on the Russian authorities to make an effort towards implementation of the Convention, and to avoid the risk of being left with no choice but to seek resettlement for all cases recognised under the mandate. However, in doing so, the office found itself irrevocably drawn into a mushrooming assistance programme - for asylum seekers - in order to avert widespread destitution. In less than three years, this project has grown into the most expensive urban assistance operation in the world, and stringent efforts are now being made to gradually phase it down. Again, this brings the argument back to the fundamental question raised at the beginning of this report: does UNHCR's mandate to protect refugees necessarily imply an obligation to assist?

(40) In the view of some long-serving colleagues, it is essential to reassert protection as the starting point for our approach to urban refugees. The essential issues thus become not whether to assist or not, but whether or not the refugees are recognised by the government, whether or not they have a right of abode, and whether or not they have the right to work. If the answer is no to any of these questions we should seek to put pressure on the government to provide protection within the terms of the Convention rather than rush in with assistance.


(41) The lack of consistency identified in terms of UNHCR's protection of urban refugees is matched only by that of its assistance to the same caseloads. The inconsistency in provision of assistance is multi-faceted. Whilst taking into account significant differences in standards of living between countries and regions, as well as variations in refugees' access to other forms or assistance and/or means of becoming self-sufficient, there remains little, if any, logic for the erratic levels, types or duration of assistance provided to urban refugees around the world.

(42) Such differences relate not only to the level of basic subsistence allowances provided vis-à-vis the local cost of living, but also to the range of additional types of assistance offered. In some countries urban refugees receive a comprehensive range of benefits in addition to basic subsistence: for food, lodging, health care, education, transportation, and even annual clothing allowances, while in others only a bare minimum is provided. Indeed, there is no established basis for determining assistance levels, such as the national minimum wage, nor any criteria for establishing what elements it should cover.

(43) The time frame for providing subsistence allowances is also highly variable. The six-month limit indicated in the Social Services Handbook and the UNHCR Manual is ignored in most countries, but quite scrupulously applied in others10. Criteria used from one office to another to determine eligibility for assistance are particularly inconsistent. While some branch offices (or operational partners) undertake systematic needs assessments in order to determine those individuals who require assistance of one form or another, in some cases reviewing them on a regular basis, others offices assist indiscriminately. The latter approach can be particularly damaging insofar as it reinforces the common perception among refugees and asylum seekers that assistance is a "right", and that it is provided automatically to anyone who gains refugee status. Referring to assistance as a "salary" is not uncommon amongst urban refugees.

(44) Provision of assistance to asylum seekers pending determination of refugee status is another highly contentious issue. Again, much greater efforts need to be made to shift the focus of responsibility to the host government. Available guidelines do not help much here as they are, once again, internally inconsistent. While the 1984 Social Services Handbook stipulates that assistance may provided only in cases of exceptional need and if there is a reasonable assumption that an individual will be screened in, the recent Community Services for Refugees manual recommends provision of assistance to urban asylum seekers on the basis of vulnerability. In branch offices with a large backlog of cases awaiting status determination, the approach taken on this issue can have significant impact on the requirements for assistance projects. The recent experience of Branch Office Moscow provides a salutary lesson on how rapidly assistance to asylum seekers can escalate.

(45) The problem may be particularly acute where provision of urban refugee assistance is delegated to an operational partner, with minimal supervision by UNHCR, or, in countries where UNHCR does not have a presence, is left in the hands of UNDP. The ever-growing incidence of violence among urban caseloads has tended to reinforce UNHCR's desire to physically separate itself from the direct provision of assistance, thereby reducing further its capacity to monitor effectively. It is essential that this tendency be corrected so that monitoring responsibilities are not further undermined.

(46) Paradoxically, many long-serving UNHCR staff believe that the organization has become increasingly paternalistic over the years, and that "flexible" attitudes towards the provision assistance to urban refugees can only serve to promote dependency. What is in no doubt, is that the many inconsistencies identified have a direct impact on promoting irregular movement between asylum countries not only within, but between, regions and sub-regions.

(47) Access to educational assistance is an important pull factor leading to irregular (as well as regular) movement in a number of regions and sub-regions. Since UNHCR has no consistently applied global policy on educational provision, there are, once again, significant variations from one country to another. This is particularly the case with tertiary education for which funds are provided to an urban refugee programme not on the basis of a systematic needs assessment, for example, of all first asylum countries in a given region, but rather in response to requests from individual branch offices. The result is a scattering of programmes providing tertiary education grants which often prove to be a magnet to the many young, single, male urban refugees. The link between irregular movement of refugees and the availability of scholarships is particularly striking in West Africa.

(48) While in principle UNHCR adheres to the educational clause of the Convention on the Rights of the Child with regard to right of access to primary education, the reality is very different, particularly for children within large refugee influxes. For urban refugee children, there is generally some educational provision although this can vary considerably, from covering the costs of local primary schools for just one or at the most, two children per assisted family, to providing private education for all children in assisted families.

(49) Variations in individual assistance to urban refugees undoubtedly present a significant pull factor, drawing not only for refugees from rural settlements into the cities, but also provoking movement between countries, and even between continents. Although such differences are sometimes related to opportunities for self-sufficiency in the countries of asylum concerned (and, of course, to the standard of living of the national population), this is not always the case. Where such differences occur within a region or sub-region (and even between regions), the inevitable result is "assistance shopping" which is clearly of benefit to no-one in terms of finding a durable solution.


(50) An examination of UNHCR's work with urban refugees points to a major weakness in the organization's approach (or, more often, lack of approach) to durable solutions. Perhaps the fact that urban caseloads are not always the major focus of attention for branch offices, explains why little thought is given to their long-term prospects. The project title of many long-standing urban refugee projects, "care and maintenance", is a clear reminder of the static way in which such programmes tend to be viewed. While much is said and written about the dangers of the "dependency syndrome", many urban assistance programmes have been developed in a way that almost guarantees dependency. It may be necessary to revise, or even dispense with the care and maintenance concept in the context of urban refugee assistance, and place the focus instead on solutions.

(51) A significant gap exists, however, between the awarding of refugee status and the identification of solutions. Recognition of a refugee in itself guarantees little more than protection against refoulement. Some would question whether recognising a refugee in a country where there are no guarantees for the future, is really in that individual's best interest. In reality there might be no alternative, if neither voluntary repatriation or resettlement are viable options. Nevertheless, if the link between asylum and solutions is lost, a refugee is condemned to a lifetime of marginalisation and dependancy. Before attempting to develop a coherent approach to identifying solutions for urban refugees, it is therefore essential to first resolve the problem of irregular movers. Without this, it will be impossible to deal with regular urban cases in a satisfactory manner.

(52) The problem is particularly acute for those who choose to move on beyond the first asylum country where they had enjoyed prima facie recognition as part of a larger refugee caseload. Such individuals, who may not qualify for refugee status under the 1951 Convention may nevertheless be recognised under the UNHCR mandate or as "persons of concern" in a second country of asylum. They may then find themselves excluded from resettlement on the basis of a weak individual claim, and consequently become dependent indefinitely on assistance in the absence of any prospects for local integration.


(53) While resettlement in a third country provides a durable solution for a very small minority of refugees world-wide, it is possibly the solution most sought after by urban refugees. Again, inconsistencies in access to resettlement opportunities within and between sub-regions and regions are a major pull factor leading to irregular movement. Such inconsistencies stem from a variety of factors, some of which are internal to UNHCR, including differential implementation of the resettlement guidelines, changing policy towards a particular refugee group, and the presence or absence of staff qualified to identify those in need of resettlement. External factors also play a major part in generating inconsistencies, particularly the location of resettlement country processing centres, availability of quotas, and the attitudes of consular staff.

(54) Resettlement in a third country is viewed by many urban refugees as almost inherent in their recognition, a misconception which both UNHCR and resettlement country governments have played a role in perpetuating by the inconsistent implementation of policy. Individuals and families make major investments towards achieving this goal by travelling long distances across and between continents to reach a place where they perceive that the prospects of resettlement are good. The frustration and disappointment that results when such expectations prove unfounded are a major factor contributing to mounting levels of violence among urban caseloads, particularly among those who are irregular movers. A vicious circle arises when UNHCR offices seek to contain security threats by cutting down on referrals for resettlement. Rather than addressing the root of the problem, this approach is more likely to result in the non-submission of individuals in genuine need of resettlement.

Voluntary Repatriation

(55) Many of today's urban refugees come from chronically unstable countries to which the prospects for return remain dim, such as Somalia, Iraq, Afghanistan, Liberia, Burundi, Rwanda and Zaire. Furthermore, while in certain circumstances UNHCR might facilitate the voluntary repatriation of refugees wishing to return overland from an asylum country adjacent to one or other of these countries of origin, it can be much more complicated to negotiate the formal return of an individual or family travelling by air from another continent outside of the context of an organised repatriation operation. In fact, for many urban refugees return to the country of origin is unlikely to be the preferred solution, either because of serious security concerns, and/or because the economic aspects of a decision to leave would not be satisfied by return.

Local Integration

(56) Before the end of the cold war, when refugee flows tended to be of longer duration than is currently the case, there was a much greater focus on local integration as a durable solution. Furthermore, governments were more willing in the past to permit small, and even large groups of refugees to remain indefinitely on their soil. Today, more and more urban refugees find themselves in countries where they are not permitted by law to find work in urban areas or cannot remain there unless they have work11. In Zambia, for example, urban refugees were transferred to rural settlement, although many subsequently returned illegally or moved on to another country.

(57) Where local integration has been permitted by the host government, UNHCR's record has been quite positive when it comes to large rural caseloads, but for urban caseloads the experience has generally proved disappointing. Enormous amounts of time, funds and energy have been spent over the years trying to find the right formula for successful vocational training and job placement, or small-scale income-generating projects, to assist individual refugees who have been unable to find work in the formal sector.

(58) In general, such efforts have failed, with rare successes occurring in cases where a beneficiary has the right combination of skills and initiative to have found a solution without UNHCR's help. Efforts to encourage self-sufficiency among the so-called "vulnerable" groups, such as the isolated elderly, the handicapped or women heads of family, have rarely born fruit for very long. It is beyond the scope of this study to analyse the reasons for such failures, but they relate generally to extremely poor economic conditions including high levels of unemployment, in the country of asylum. Furthermore, where limited successes have been found, they tend to be in the context of largely rural populations, or very small urban caseloads, where it has been possible to offer a generous combination of loans, scholarships, vocational training and income-generating projects which would be impossible with a large urban caseload.

(59) Apart from the generally unfavourable economic conditions within which UNHCR has made efforts to facilitate the local integration of urban refugees, it should be noted that the organization's approach to implementation has also been deficient. To an extent, this weakness is linked to the lack of a consistent policy on urban refugees, as a result of which interventions have typically been ad hoc. Furthermore, there is a widespread lack of technical expertise in the field of socio-economic planning, and while such expertise exists with PTSS, it is simply not feasible for one or two Headquarters-based experts to provide the comprehensive input and follow-up required by urban programmes worldwide. Consequently, with very few exceptions, the approach to a new urban programme has been to launch into a traditional care and maintenance programme, sometimes introducing "solutions" at a later stage, almost as an afterthought. In the absence of the required technical expertise, assistance programmes for urban refugees have rarely, if ever, been subject to systematic review, so that despite a multitude of initiatives in many parts of the world, few lessons have been learned, and guidelines to improve practice remain to be developed.

(60) This lack of organizational capacity to offer realistic local solutions to refugees must be acknowledged and addressed without delay. When UNHCR fails on the one hand to persuade a host government to remove restrictions on the local integration of refugees and on the other does not provide a viable framework for attaining self-sufficiency, the possibilities for individual "success" are extremely limited. In the circumstances, we should not be surprised when refugees protest if assistance is cut.

(61) In recent years UNHCR has adopted a more structured approach with regard to interventions designed to encourage non-agricultural self-sufficiency within the context of large rural settlement programmes. Particularly encouraging examples include the introduction of micro-projects for Liberian refugees in rural settlements in Guinea-Conakry and the credit scheme programme based on a revolving fund for Indochinese refugees in China. The latter project, set up in early 1993 with the support of PTSS, has subsequently been subject to annual technical review which has permitted appropriate modifications to be made as the situation evolved.

(62) An attempt to introduce a more systematic approach to dealing with an urban caseload was made in the context of a new urban refugee programme developed in Rio de Janeiro in 1993 in response to a mounting influx of Angolan "air people". The programme received substantial initial technical input from PTSS, and a commitment was made to emphasising a solutions-oriented approach from the outset (or "case management by objectives"), including job placement, apprenticeships and micro-projects. Unfortunately, a follow-up mission by PTSS in 1994 found that "the objectives established in 1993 for local integration have not been reached due to a variety of factors including staffing, management and narrow interpretation of the labour insertion strategy both by UNHCR and its implementing partner." Furthermore, "... relations between refugees and the implementing partners have dangerously deteriorated ... ". A further follow-up mission by PTSS, scheduled for 1995, was subsequently cancelled because of financial constraints and other priorities within the Section. Meanwhile, security problems involving the refugees have increased, with demonstrations, occupation of office premises and threats against staff, in protest against UNHCR's efforts to phase out its temporary assistance. The challenge remains for UNHCR to find the right approach to successful case management by objectives in its urban refugee programmes.


(63) An attempt was made, in the context of the current review, to estimate global costs of protecting and assisting urban refugees. This attempt proved even more difficult than anticipated, for a number of reasons. Firstly, since the demise of the urban-specific "SA", "MA" and "CL" projects with the introduction of FMIS, in countries where there are rural as well as urban caseloads, costs are combined within one single project and cannot be desegregated at Headquarters level. Secondly, even in countries where there is no rural caseload, the budgets of urban projects do not reflect all relevant costs.

(64) It is widely acknowledged that wherever urban caseloads exist, a substantial amount of the working time of both local and international UNHCR staff, from the Representative down, is taken up in dealing with such individual cases. This is particularly true of protection and community services officers, who sometimes work exclusively with urban refugees. In some cases, an urban caseload is the entire raison d'être for a Branch Office and so the bulk of the running costs for UNHCR in that country could be ascribed to urban refugees. Desegregating such staff costs is an exercise beyond the scope of this review. Furthermore, costs of associated projects benefiting the urban caseloads, notably education projects and those leading to one of the three durable solutions, have not been included.

(65) In an effort to provide an indicative baseline figure, without involving countless desk officers at Headquarters and programme officers in branch offices around the world in a complicated financial exercise, contact was made with PCBOs who were asked to provide an estimated total figure for urban projects in their respective regions of responsibility. The following figures, related to just two regions where urban refugees are currently of particular concern, give an indication of the scale of the matter:


Southern Africa: 4,127,680
Botswana, Lesotho, Malawi,Mozambique,
Namibia, South Africa, Swaziland,
Tanzania, Zambia and Zimbabwe

East and South-East Asia: 4,777,537
China, Hong Kong, Indonesia, Malaysia,
Philippines, Singapore, Thailand and

(66) If individual project costs of six of UNHCR's largest urban programmes in other regions are added to these two regional figures, the total figures mount rapidly to form a significant proportion of the organization's total annual budget:


Moscow, Russia 4,800,000
New Delhi, India 4,588,853
Islamabad, Pakistan 2,142,040
Cairo, Egypt 2,024,900
Rio, Brazil 1,466,500
Athens, Greece 1,276,900

(67) As already noted, these figures, together amounting to some 25 million USD, are (a) significantly incomplete in terms of countries covered13, (b) do not reflect direct UNHCR staff costs, and (c) do not include costs of associated projects such as education, local settlement, resettlement or voluntary repatriation. It is clear, therefore, that the total costs of current activities in favour of urban caseloads world-wide, as yet to be accurately calculated, many represent as much as 10-15% of UNHCR's overall annual budget which is currently in the region of 1.3 billion USD.

(68) As such, on a purely financial basis, the organization's approach to dealing with urban cases warrants careful examination to identify areas amenable to rationalization. When one further considers that, as noted above, at a very conservative estimate urban refugees are unlikely to exceed 1.7% of the world's refugees or 0.9% of all those of concern to the High Commissioner, such an examination becomes imperative.

Annex 1



Project title: Care and maintenance of refugees, and assistance to forced migrants in the Russian Federation (95/AP/RUS/CM/200)

Project established: 1992
Total Budget: US$ 4,800,000.
No. of beneficiaries: 6,000
Origin of refugees: Mostly Afghanistan, Somalia and Iraq. Also Ethiopia, Angola, Zaire and 25 other countries.
Assistance provided includes: Basic needs of food, shelter, health, skills training, education, legal protection and income -generation
Implementation: BO, Federal Migration Service, NGOs: EquiLibre, LORIEN (Medical Care for Refugees),
Coordination Council for Aid to Forced Migrants and Refugees, Compatriot Fund

2. INDIA, New Delhi

Project title: Assistance for Care and Maintenance of Refugees (95/AP/IND/CM/200)

Project established: 1981
Total Budget: US$ 4,588,853.00
No. of beneficiaries: 11,295
Origin of refugees: Mostly Afghanistan; also Iran, Somalia and others.
Assistance provided includes: Domestic needs, health care, community services, counselling, education and training
Implementation: OCM, UNDP, NGOs: Service Civil International, Discipleship Centre
Recent developments: OCM has started efforts to reduce assistance and encourage self-sufficiency for long-stayers.
OCM is trying to identify new NGOs.

3. PAKISTAN, Islamabad/Rawalpindi

Project title: Assistance to non-Afghan refugees in Pakistan (95/AP/PAK/CM/200)

Project established: 1981
Total Budget: US$ 2,142,040.00
No. of beneficiaries: 2,600
Origin of refugees: Mostly Iraq and Iran. Also Somalia and Bosnia.
Assistance provided includes: Supplementary assistance toward living requirements and assistance towards self-reliance.
Implementation: BO, SO Peshawar, SO Quetta, NGO: SAVERA/Islamabad.
Recent developments: In 1994, Iraqi Kurds demonstrated and held hunger strikes leading to the death of one IC early in 1995.

A plan to promote skills training and to support small business instead of providing financial assistance was resisted by refugees during 1994. The plan will be promoted again in 1995.

4. EGYPT, Cairo

Project title: Care and Maintenance for Refugees in Egypt (95/AP/ARE/CM/201)

Project established: 1990
Total Budget: US$ 2,024,900.00
No. of beneficiaries: 8,000
Origin of refugees: Mostly Somalia and Sudan. Some from Ethiopia, Eritrea, etc.
Assistance provided includes: Basic subsistence (5,800), medical assistance (2,400), medical treatment (300),
primary/secondary education (1,700), skills training (175).
Implementation: RO, UNDP, NGOs: CARITAS, Family Planning Association, St. Andrew's, Cairo.
Recent Developments: In 1994, RO Cairo began interviewing Sudanese asylum- seekers to assess the validity of claims
and has since been overwhelmed with requests for assistance.

Although volrep to Northern Somalia resumed at beginning 1995, the Somali caseload in Egypt originated in southern Somalia and Mogadishu, and could therefore not return.

5. BRAZIL, Rio de Janeiro, Sao Paolo

Project title: Care and Maintenance for Refugees in Brazil (95/AP/BRA/CM/200)

Project established: 1994
Total Budget: US$ 1,466,500.00
No. of beneficiaries: 3,148
Origin of refugees: Africa, Latin America, Asia and Europe.
Assistance provided includes: Care and maintenance (expected to be 2,305), legal assistance (753)
Implementation: OCM, NGOs: Tolstoy Foundation, CARITAS.

6. GREECE, Athens

Project title: Care and maintenance in Greece (95/AP/GRE/CM/200)

Project established: 1984
Total Budget: US$ 1,276,900.00
No. of beneficiaries: 7,326
Origin of refugees: Mostly Iraq and Iran and some from Africa and Asia
Assistance provided includes: Subsistence allowances and emergency assistance
Implementation: BO, NGOs: Social Work Foundation, Greek Council for Refugees, International Social Service

Annex 2


Côte d'Ivoire et Guinée - Autosuffisance des réfugiés Libériens et Sierra Léonais et Assistance Alimentaire (10-24 May 1995) PTSS Mission Report 95/19/N

China - Reorientation of the Assistance Programme towards self-sustainability, (1-18 June 1994) PTSS Mission Report 94/46N

Brazil - Regional Training Workshop on Local Integration. Brief Review of the Refugee Assistance Programme, (15-26 May 1994) PTSS Mission Report 94/43N

Guinée - Mission technique pour la revue des activités d'auto-suffisance pour les réfugiés Libériens, (22 novembre - 5 décembre 1993) PTSS Mission Report 93/60Nx

The Russian Federation Community Services Mission. PTSS, November 1993

Brazil - Review of the Assistance Programme for Urban Refugees, (20 September to 5 October 1993) PTSS Mission Report 93/42

China - Review of UNHCR-Assisted Programme, (6-28 March 1993) PTSS Mission Report 93/22

Kenya, Social Services Mission. PTSS, January/February 1992.

Turkey: Social Services Mission on Assistance to Urban Refugees. A community-based Approach. PTSS, May/June 1992

South Africa Social Services Mission on Repatriation, Community-based Counselling and Training of Counsellors. PTSS, December 1991

Review of UNHCR Assistance in Guinea and Sierra Leone. PTSS and Central Evaluation Unit, May 1991

Review of UNHCR Operations in India. Central Evaluation Unit, October 1990

Review of UNHCR Operations In Nepal. Central Evaluation Unit, October 1990

Review of UNHCR Operations in Kenya. Central Evaluation Unit, July 1990

Review of UNHCR Activities in Botswana. RBA and Evaluation Unit, May 1989

Review of UNHCR Assistance to Refugees in Zimbabwe. Evaluation Unit and RBA, September 1988

Survey of UNHCR Assistance to Urban Refugees in Guatemala. Evaluation Unit, August 1988

Survey of UNHCR Assistance to Urban Refugees in Mexico. Evaluation Unit, August 1988

Argentina (and Uruguay): Redirection of Local Settlement Programmes. PTSS, July 1988

Review of UNHCR Assistance to Refugees in Kenya. RBA, Social Services and Evaluation Unit, July 1987

Review of UNHCR Assistance to Refugees in India. Evaluation Unit and RBAO, March 1987

Review of UNHCR Assistance to Refugees in Italy. Evaluation Unit and RBENA, January 1987

Model for an Integrated Urban Refugee Programme. PPRU, Social Services and Evaluation Unit, September 1985

Review of Assistance to Urban Refugees in Costa Rica and the Dominican Republic. Evaluation Unit, March 1985

Assistance to Urban Refugees in Africa: A review of projects in four countries. Evaluation Unit, December 1984

1 As a result of the lack of a coherent policy on urban refugees, there is an absence of generally accepted objectives for activities intended to protect and assist these caseloads. Without this basic parameter against which to measure achievements, this review should not be considered as an evaluation, but rather as a discussion paper reflecting the views and concerns of a wide range of experienced UNHCR staff.

2 The recent focus of interest on urban refugees in branch offices in southern Africa and in south east Asia is indicative of this problem.

3 A Regional Workshop on Irregular Movers, held in Kuala Lumpur in May 1995, proposed the following definition: an irregular mover is a refugee or asylum seeker who leaves a country where basic protection was available, for reasons other than: (1) family reunion with immediate family members who are not themselves irregular movers in the current country or (2) a threat to his/her physical security.

4 My thanks to Henry Domzalski for bringing Jaeger's report to my attention.

5 Furthermore, it should be noted that experience in southern Africa (such as in Kenya and Zambia), has shown that efforts to move refugees from urban to rural areas can provoke further irregular movement, as individuals leave the country of asylum in search of another, where they can remain in an urban setting.

6 At the height of the hunger strike in Pakistan a decision was made by UNHCR, on security, as well as, humanitarian grounds, to reverse its hitherto firm position on non-submission of the Iraqi cases for resettlement. The non-advisability of such a policy reversal under pressure has since been acknowledged in a document prepared for major resettlement governments ("Assessment of Global Resettlement Needs for Refugees in 1996" paras 116-118).

7 There will, inevitably, be occasional exceptions to this general rule: for example individuals from rural settlements who are sent temporarily to an urban centre for purposes of education or medical care, or permanently, for family reunion. In developing guidelines for implementation, it is important to ensure that such exceptions do not undermine the general rule.

8 As noted earlier (footnote 3), the definition of irregular movers adopted by the Regional Meeting in Kuala Lumpur, refers to leaving a country where basic protection was available. The definition of basic protection, and the extent to which this concept should go beyond protection against refoulement, to embrace quality of life aspects, is subject to on-going debate within UNHCR.

9 The recently issued "Assessment of Global Resettlement Needs for Refugees in 1996" makes a general point regarding the link between resettlement and irregular movement, in the context of a discussion on Iraqi refugees (paras 114-115). The point needs to be made more forcibly and broadened to include groups other than the Iraqis.

10 While there is a general assumption that cutting short subsistency allowances is essential to avoid dependency, no systematic study has been made to examine the consequences of this approach, not to compare it with urban programmes where long-term assistance is provided.

11 As a result of the democratization process, governments in Southern Africa are now accountable to a public opinion increasingly hostile to refugees, who are not differentiated from the mass of illegal immigrants and perceived as unfairly competing for scarce resources.

12 While assistance costs inevitably differ substantially from one country or region to another, a comparison of project costs with the number of beneficiaries (individuals) noted earlier, results in an interesting range of "per capita" costs (in USD): East and South East Asia - 1378, Islamabad - 823, Moscow - 800, Rio - 465, New Delhi - 406, Southern Africa - 290, Cairo - 253 and Athens - 174.

13 In calculating the estimated total number of urban refugees world-wide (see para.31) an assumption has been made that the countries covered here represent approximately 25% of all urban programmes. In addition, the programmes referred to here are larger programmes where staff costs are proportionally low. Most urban caseloads are assisted in small programmes where staff and other costs are proportionally higher.