Conflict-Induced LIVING Displacement in Europe and Central Asia IN LIMBO Steven B. Holtzman Taies Nezam Living in Limbo Living in Limbo conflict-induced displacement in europe and central asia Steven B. Holtzman Taies Nezam THE WORLD BANK Washington, D.C. © 2004 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved. 1 2 3 4 05 04 03 The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978- 750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ISBN 0-8213-5850-2 eISBN 0-8213-5851-0 Cover photo: Refugees entering Armenia in July 1992 Photo credit: Hakob Berbervan Library of Congress cataloging-in-publication data has been applied for. Contents Acknowledgments ix Executive Summary xi 1. Introduction 1 Objectives and Scope 4 Discussing Numbers of DPs 5 From Humanitarianism to Poverty Reduction 9 Definitions of Terminology 11 Organization of the Study 16 2. Material Well-Being 21 Consumption 23 Transfers 27 Unconventional Public and Private Transfers 32 Conclusions 34 3. Employment 39 Formal Employment 40 Agriculture (Land and Livestock) 43 Unemployment 47 Obstacles to Employment 49 Conclusions 57 4. Settlement Patterns and Shelter Situations 61 Initial Shelter Situations 62 Settlement Patterns 63 v vi contents Implications of Settlement Patterns 66 Shelter Situations 67 Conclusions 77 5. Human Capital 81 Demographics 81 Education 83 Health 87 Psychosocial Issues 88 Differential Impacts on Gender and Age Groups 95 Conclusions 99 6. Social Capital 103 Social Cohesion within DP Communities 105 Links to Host Societies 112 Choices in Social Networks 115 Conclusions 116 7. States and Displaced Populations 119 Factors Influencing State Response 121 Public Institutions 124 Conclusions 131 8. Conclusions 135 Specific Key Findings 137 The Political Will to Depoliticize Displacement 144 Annex 1: Development Programming and Conflict-Induced Displacement 149 Poverty Diagnostics and Tools 149 Integration of DPs into Social Safety Nets 154 Issues in Social Service Delivery in Displacement Situations in the Europe and Central Asia Region 157 Displaced Populations and CDD (Community-Based Programming, Municipal Development, Area-Based Development) 159 Housing and Shelter 162 Annex 2: Methodology 165 Bibliography 169 Index 183 contents vii boxes 2.1 Kosovar Refugees and Remittances 32 3.1 Continuity in Public Employment during Displacement 42 3.2 Access to Assets in Areas of Origin 44 3.3 Moveable Assets 45 3.4 Slowly Moving toward Self-Reliance in Azerbaijan 52 4.1 Differing Shelter Situations within One Country 70 figures 1.1 Total Displacement in Europe and Central Asia, 1991­2002 2 1.2 Duration of Displacement, Major Groups and Host Countries, December 2002 3 2.1 Poverty in Selected Countries 23 3.1 Unemployment in Selected Countries 49 4.1 Accommodations in Selected Countries 68 tables 1.1 Major Displacements in Europe and Central Asia, 1990­2002 6 1.2 Displacement in Europe and Central Asia, Major Groups, December 2002 14 4.1 Home Ownership in Selected Countries 71 Acknowledgments THIS STUDY HAS BEEN PREPARED BY Steven Holtzman and Taies Nezam of the Social Development Team, ECSSD, with the support of Nils Junge. The study was conducted under the guidance of Alexandre Marc. Alexandre's commitment to pushing the envelope and willing- ness to discuss and provoke has been an integral and essential part of the process. We have also benefited greatly from the comments and re- view of an internal advisory committee of Daniela Gressani, Michael Mills, Ruslan Yemtsov, Markus Kostner, and Elena Correa. The study has been financed through the ECA Vice President's regional studies fund, through additional support of HDNSP, and indirectly through ongoing diagnostic and operational work within the region, particu- larly the poverty assessment work task managed by Ruslan Yemtsov (Bosnia and Herzegovina, Serbia and Montenegro) and Michael Mills (Azerbaijan). We would like to express specific gratitude to Nils Junge for his painstaking collection of reports, surveys, and other data and his time- less efforts in preparing statistics and charts to help reflect a reality that, as we have all learned, is resistant to accurate quantitative analysis, constantly changing, and subject to interpretation. Our gratitude also to Larry Dershem, Nana Gibradze, and Igor Barsegian, who all pro- vided country-specific reports that helped in the preparation of the study. Within the World Bank, Margaret Grosh, Nancy Cooke, Neil Simon Gray, Mary Judd, Judy Deane, Judy O'Connor, Farid Mamedov, Tevfik Yaprak, Philip O'Keefe, Ian Bannon, Michael Cernea, and many others, through discussions and feedback on this and related initiatives, gave us significant food for thought as we prepared various drafts. Because this study draws upon so much of the work prepared by other agencies, it is difficult to name everyone. We owe a great debt to ix x acknowledgments the VAM unit of WFP, to UNHCR, ICRC, IFRC, the Danish Refugee Council, and the many NGOs and international agencies whose analytical work on conflict-displaced populations has been utilized in the preparation of this study. A special thanks is due to the staff of the Norwegian Refugee Council's Global IDP project and the U.S. Committee for Refugees. We have drawn extensively upon the work developed over so many years by these two agencies. Finally, our deep- est gratitude goes to Francis Deng and Roberta Cohen, whose written works and oral views in multiple discussions have stimulated our thoughts and helped us reflect on the subject matter of this study. We thank Beaulah Noble for her good-natured willingness to meet short deadlines and help in the editing and formatting of this text. We also thank Vigen Sargsyan who was kind enough to find the photo- graph that appears on the book's cover. Finally, we thank Patricia and Leo who involuntarily lived with the book for two years and kept us going. Most of all we thank Lula Muska who patiently sat by her mother as she reviewed the final drafts. Executive Summary THE OBJECTIVE OF THE PRESENT STUDY is to analyze conflict-induced dis- placement from the point of view of vulnerability, using a multifaceted definition of vulnerability. As many as 10 million people have been displaced by war in the Europe and Central Asia region since 1990. While many people have been able to return home, approximately half remain displaced, with no available avenues for sustainable reintegra- tion. Currently, in five countries of the region (Armenia, Azerbaijan, Bosnia and Herzegovina, Georgia, and Serbia and Montenegro) dis- placed persons (DPs) represent more than 5 percent of the total popu- lation. In two other countries (Russia and Turkey), they represent high proportions of the population in specific regions (Ingushetia, southeast Turkey). The majority of these populations have been displaced for up to a decade or longer. With the notable exception of the conflict in Chechnya in Russia, all other major conflicts have either resulted in peace accords or represent what have sometimes been termed "frozen conflicts." In the latter conflicts, no final resolution has emerged, but fighting has been dormant for many years. Populations displaced by these conflicts remain in "limbo," with no clear direction for the future. The general conclusion of the study is that most of these populations remain significantly vulnerable and their patterns of vul- nerability represent a specific category with its own characteristics. The levels of external humanitarian assistance targeted to displaced populations have dropped significantly since the late 1990s. This real- ity has increased the pressure for host governments and development agencies to better understand the levels and nature of vulnerability of these groups in comparison with the already appreciable levels of vul- nerability in other groups in affected societies. In some countries, host xi xii executive summary country assistance to DPs represents a significant, and potentially growing, pressure on public expenditure. Perhaps as significant have been the financial and other contributions made to the survival of DPs by the societies of host countries. Populations displaced by conflict are formally characterized as refugees if they have crossed an international border; those who re- main displaced within the territory of their home country are inter- nally displaced persons (IDPs). In this study we use the term displaced person (DP) to refer to both categories, except when the designation refugee or IDP is needed to differentiate multiple groups of displaced in one country or there are specialized sources of vulnerability that emerge specifically from this legal status. The blurring of such status is particularly characteristic of displacement in Europe and Central Asia due to the dissolution of larger states (the Soviet Union and Yugoslavia) in the early 1990s. Because we look at displacement as a phenomenon related to potential vulnerability rather than as a form of legal status, we include in our analysis displaced populations who have been offered citizenship in host countries and those no longer tar- geted by humanitarian assistance. Typically, the terms refugee and IDP have been tacitly accepted as a kind of proxy category for vulnerability. Assistance by host govern- ments and international agencies, where available, has largely been provided on the basis of identity rather than of individually deter- mined need. While there have been significant attempts in some coun- tries to review specific issues contributing to IDP and refugee vulnera- bility, there has been little concerted analysis of the foundations and patterns of DP vulnerability or cross-country comparison in this region. As the years of displacement have worn on (and available ex- ternal funding for displacement programs has decreased), observers have expressed various opinions on the levels of continuing vulnera- bility of DPs in the region. Some view the displaced as a privileged class or at least assume that after such long periods they have more or less adapted. Others view them as unconditionally vulnerable, even in the context of the significant political and economic shocks that have affected local populations in general in the region. For the most part, this study does not develop new sources of data to assess the situation of the region's DPs. Rather it reviews a wide range of existing surveys, assessments, and field studies conducted by UN agencies, NGOs (nongovernmental organizations), host govern- ments, and the World Bank to try to get a sense of the current situation in specific countries as well as broader patterns of vulnerability in the region. The study is not designed as a series of country studies. Instead it uses examples from specific countries as illustrative of general executive summary xiii patterns across the region as well as of certain heterogeneous aspects of the vulnerability of displacement. Given the reliance on secondary and tertiary sources, we cannot come to specific operational conclu- sions on individual situations, although we hope the discussion herein will help inform ongoing operational discourse in some countries. The reliance on existing resources also creates other limits on the study. Very little reliable information exists on the situation of Kurdish IDPs in Turkey or on Chechens who remain displaced inside of Chechnya in the Russian Federation. In addition, we find little analysis of the situation of DPs who returned home several years ago after periods of displacement, notably in Tajikistan and Moldova. It is important to understand whether returning home actually represents an endpoint in displacement-induced vulnerability or whether long periods of dis- placement create lingering patterns of differential vulnerability. Unfor- tunately, this issue is beyond the scope of the current work. The study concludes that the region's displaced remain vulnerable in several different dimensions. While their human capital indicators in terms of education and health appear fairly comparable to those of local populations and their demography in terms of gender, age, and household size is also similar, their situations differ for the worse in many other aspects. The displaced are generally poorer in terms of measures of consumption. They have limited material assets and little, or at best unstable, access to land for cultivation. Their unemployment rates are significantly higher than those of local populations in all countries for which data were available, and their patterns of employ- ment tend to be heavily reliant on ad hoc and fragile areas, such as seasonal employment and day labor. In several countries, the displaced are far more reliant than local populations on public assistance, whether from international agencies, host government public assistance programs, or both. For the most part, these assistance programs target DPs as a group rather than through any individual determination of vulnerability. In some cases, though, various forms of needs-based targeting have been incorpo- rated into international assistance strategies. The area of housing is one in which DPs are particularly disadvan- taged, in terms of both security of tenure and quality of housing. The majority of the population in the region's affected countries live in self- owned accommodations. While most DPs formally own houses or have social rights to housing in their home area (which is largely inaccessible), very few own housing in the areas where they are now settled. Perhaps half of the displaced in the region live in one of three forms of unconventional housing situations: collective centers, shared accommodations with host families, or illegally occupied private or xiv executive summary public housing. Collective centers, a significant part of the landscape in the South Caucasus, form a catchall category of a variety of structures. These include abandoned factories, schools, clinics, spas, dormitories, and hotels, structures that were either never intended for human habi- tation or only seasonally occupied in the past. The deterioration of these facilities due to lack of maintenance and overcrowding repre- sents a factor in DP living situations that increases in severity over time. Up to a third of the displaced in Georgia, Serbia, and Turkey remain living in sometimes difficult circumstances with host families. Even those DPs who have found private rental accommodations frequently are no better off. Evidence from Serbia suggests that the presence of DPs in private rental accommodations is often due, not to self-reliance, but to a lack of alternative sources of shelter. Although the impoverished situation of the region's DPs is to a sig- nificant degree a factor of the difficult economic environment of host countries, their overall pattern of vulnerability is different from that of local populations in both intensity and nature. This is not to say that all the displaced are especially vulnerable or that on the whole they are the only vulnerable group in affected societies. But the majority are vulnerable, and their numbers are high enough to make them a vul- nerable group of significant concern in affected societies. The study further finds that the nature of DP vulnerability emerges from a specific set of factors. These include the levels of different types of capital of populations prior to their displacement and, as im- portant, the resilience of that capital to the impacts of displacement; the specific impacts of displacement as a phenomenon; the nature of the socioeconomic environment in which DPs are settled, including the impact of specific policies and actions of host governments and international assistance agencies; and the impact of duration of dis- placement (time). Time in this context refers not only to chronologi- cal time but also to the number of generations of DPs. The specific impacts of displacement itself referred to above we distill to four broad categories: destruction of assets, denial of access to assets, dis- location from normal socioeconomic environment, and the psycho- logical and material impacts of living in limbo. Displacement has created general patterns of vulnerability common to all displaced pop- ulations in the region, but individual outcomes in each of these sets of factors also create situation specific dynamics. These dynamics result in differing patterns of vulnerability in specific subgroups among DP populations cross-nationally as well as within the same situation. Extant surveys and other assessments are largely too general to bring out these different "subsituations" with clarity. However, sufficient anecdotal illustrations exist to highlight the fact that the process of executive summary xv displacement is not a "one size fits all" category, but a set of socio- economic processes interacting to different degrees and measures in different situations of displacement. The study tentatively introduces a conceptual framework we hope can provide some utility in shaping diagnostic exercises in specific displacement situations. While the study does not incorporate a detailed analysis of existing programs targeted to DPs, it does use the existing information and analysis to offer some preliminary operational recommendations. These relate to enhancement of diagnostic exercises for poverty moni- toring of the displaced, incorporation of DPs into social safety nets, social service delivery, and community-based planning and shelter programs. Observers sometimes explicitly or implicitly assume that the uncer- tain status of DPs--staying where they are or eventually returning home--means that dedicated focus on their situation, beyond immedi- ate humanitarian aid, can wait. The reality is that these populations have lived in this limbo status for long periods of time, and while conflicts can remain frozen, people cannot. Populations affected by displacement have specific patterns of vulnerability, and this limbo sta- tus, present to varying degrees in different situations, is one element of these patterns. In several instances, the de facto or de jure policies of host governments may serve to exacerbate this status, contributing to, or at least not maximizing the potential to mitigate, vulnerability even when DPs are internally displaced citizens of the country in which they are settled. There is a lingering assumption that investments in sus- tainable solutions for DP self-reliance somehow undermine national objectives of facilitating an eventual return home. Yet, it can be equally argued that the long periods of vulnerability of DPs also represent a factor that mitigates against sustainable self-reliance upon return home should an opportunity to do so arise. There is little real evidence that enhancement of self-reliance or even integration in situ stops displaced populations from ever going home. What is certain, however, is that displacements of a decade or longer represent more than simply a transient reality for those affected. chapter 1 Introduction THIS STUDY FOCUSES ON THE SITUATION of populations displaced by con- flict in countries within the Europe and Central Asia region, notably the Balkans, the former Soviet Union, and Turkey. The displacement we examine resulted largely from the many conflicts that occurred during the dissolution of the former Yugoslavia and U.S.S.R (the major exception being the large Kurdish displacement in Turkey). Since 1990, thirteen countries in the region have been affected by conflict-induced displacement through the generation of refugees or IDPs (internally displaced persons) from their own conflicts or the hosting of significant refugee populations from a neighboring country.1 Although the numbers are not always reliable, some 10 mil- lion people may have been displaced in the region during this period. As of this writing, perhaps 5 million remain displaced. That figure is high in both absolute and relative terms. In seven countries, the dis- placed currently either represent 5 percent or more of the population or constitute a group of at least 200,000. The regionwide numbers of displaced, while declining somewhat over the past few years, seem to be stabilizing at a relatively high level (see figure 1.1 below). The complexity of understanding the degree to which DP condi- tions result from displacement or from other factors that affect host communities as well is exacerbated by the difficult economic and po- litical situations in the host countries. Those countries include many of the region's poorest. Of the 10 states engaged in the Poverty Reduction Strategy Paper (PRSP) process in the region, 9 have been affected by conflict-induced displacement during the past decade. (Relatively better-off nations, such as Russia and Turkey, also host major dis- placed populations, generated by internal conflicts.) In addition, as recent political events and transitions in Azerbaijan, Georgia, and Serbia demonstrate, a number of host countries face continuing prob- lems of political instability and governance, challenges that limit the scope for administrative and political action. Typically, when people speak of conflict-displaced populations, they think of a temporary phenomenon occurring during an active 1 2 living in limbo Figure 1.1 Total Displacement in Europe and Central Asia, 1991­2002 Millions 7 6 5 4 3 2 1 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Source: U.S. Committee for Refugees, World Refugee Surveys (Washington, D.C., 1991­2003). conflict and envisage an eventual concerted effort to facilitate a more sustainable solution, through a return home or assisted resettlement. This implicitly linear view of war-to-peace transitions does not corre- spond with the reality on the ground in Europe and Central Asia. With the exception of those in Chechnya, all major populations currently displaced were generated by situations in which most or all active fighting stopped many years ago, either through a peace accord or via a ceasefire. Some conflicts that ended with a peace accord have offered only a partial resolution to the displacement problem; other conflicts, frozen in place since the declaration of a ceasefire, have offered no res- olution at all. It is tempting for policymakers to think a major re- sponse to displacement, beyond humanitarian assistance, can wait until opportunities arise for a return home supported by a major donor financial appeal and a reconstruction program.2 However, while conflicts or displaced situations may remain "frozen," it does not necessarily follow that displaced populations can similarly be kept in cold storage and thawed out when an opportunity to return home presents itself. As figure 1.2 illustrates, most of the DPs in the region have already been displaced for several years, some for a decade or longer. Within a number of host countries in the past several years, there appears to have developed an almost permanent class of introduction 3 Figure 1.2 Duration of Displacement, Major Groups and Host Countries, December 2002 Kurdish IDPs (Turkey) Ethnic Azeri refugees from Armenia and Meskhetian Turk refugees (Azerbaijan) Ethnic Armenian refugees from Azerbaijan (Armenia) Croat IDPs from Vukovar and Krajina (Croatia) Serb refugees from Bosnia and Croatia (Serbia and Montenegro) Croat refugees from Bosnia and Serbia and Montenegro (Croatia) Serb, Croat, Bosniac IDPs (Bosnia and Herzegovina) IDPs from Nagorno-Karabakh (Azerbaijan) Georgian IDPs from Abkhazia and S. Ossetia (Georgia) Albanian IDPs (Kosovo) Serb and Roma IDPs from Kosovo (Serbia and Montenegro) Tajik IDPs (Tajikistan) Chechen IDPs (Ingushetia, Russia) Slav and Albanian IDPs (Macedonia, FYR) Kosovar Albanian refugees (Macedonia, FYR) Majority still displaced No longer displaced Kosovar Albanian refugees (Albania) 0 2 4 6 8 10 12 14 16 18 20 Approximate length of displacement (years) Source: U.S. Committee for Refugees, World Refugee Surveys (Washington, D.C., 1991­2003). population identified as refugees or IDPs who remain in limbo, with no clear direction for their futures. Consequently, such displaced pop- ulations have become part of the long-term landscape of affected soci- eties. Entire generations have grown up and matured under these uncertain conditions. In the past few years, increasing recognition in some countries that leaving large groups in limbo cannot continue has led to a range of 4 living in limbo ideas and programs. We are concerned, however, that with reductions in humanitarian assistance budgets across the region and only limited attention from development agencies, such initiatives may not stand the test of time. The displaced may be ever more likely to fall between the cracks. Our concern stems primarily from a feeling that these groups need and deserve attention because of their vulnerability and the lost human potential they represent. We are conscious as well of the festering wound that populations caught in limbo, with few op- portunities for sustainable self-reliance and little clear hope of a better future, may represent. Unless the situation is addressed, such a wound could help destabilize affected societies, stimulate ethnic and commu- nal violence, and even catalyze a return to violent conflict. The growing need in the region to look at new approaches to find- ing more sustainable outcomes for DPs creates a particular sense of timeliness for this study. A deeper understanding of the situation of displaced populations has relevance for the design of measures to sup- port reintegration of DPs, if the opportunity to return home arises (as for some in the Balkans), as well as measures to aid permanent settle- ment in the current place of asylum or elsewhere, if made possible. Unfortunately, perhaps, the most immanent challenge in the majority of cases is forming strategies when neither return nor permanent set- tlement appears on the horizon, and DPs remain in the varying states of limbo that have characterized their lives for so long. Objectives and Scope The overall objective in this study is to develop a better understanding of the magnitude of displaced groups' impoverishment and vulnerabil- ity as well as of the special nature of their situation. Our intention in preparing this work has been to contribute to ongoing and future discussions among governments, development and humanitarian agencies, and other concerned actors on evolving responses that will mitigate the impacts of the region's long-term displacement. In a larger sense, by reviewing a range of displaced populations within a single region, we hope to provide a resource for use within global strategic discourse on conflict-induced displacement. We also hope to highlight the importance of specificity in the analysis of displacement in partic- ular historical and socioeconomic environments. The groups on which we focus in this study include refugees and IDPs, although at present the majority of the displaced within Europe and Central Asia are IDPs. Due to some ambiguities that creep in, related to the particular historical background of the region's state introduction 5 succession, we will use the terms conflict-displaced population or dis- placed persons (DPs) to refer to both groups, except when it proves necessary to distinguish between IDPs and refugees. Such distinctions are important, for instance, when discussing Serbia and Montenegro3 or Azerbaijan, where both IDPs and refugees are present, or when identifying specific elements of vulnerability that apply more to one category than the other. It should be noted that the subject of this study is displaced populations who remain displaced. Our concern is tied to what happens to DPs and how we understand their situation while they are displaced. This is not a study of refugee returns, neither regarding why displacement persists nor what happens to people when an opportunity to return home or settle permanently elsewhere emerges. All are important topics that have been studied by others and always need further attention. Here we focus explicitly on the situa- tion of those who remain displaced because in many of the countries in question displacement is becoming not so much a transitory stop- ping point as a way of life for generations of affected groups caught between two worlds. The reasons for the inability of displaced populations to go back to their homes vary. They include ongoing political or military conflict, security concerns, destruction of property, occupation of property by others, and in some host countries, an official (or de facto) policy dis- couraging return. In many cases, even when conditions for a return home appear favorable, lingering perceptions of ethnic tension, often buttressed by tangible experience, present major obstacles to such a move. This region, after all, brought the world the term ethnic cleansing. In the South Caucasus countries, territory remains in the hand of secessionist groups. In Bosnia and Herzegovina, although conditions for minority return have improved, issues such as the dou- ble occupation of property and continuing ethnic tension still preclude significant returns. It is sufficient for our purposes to simply note the cumulative result across the region, that millions of people have been displaced since 1990 and approximately half remain displaced. Table 1.1 shows the major populations displaced by conflict within the region during this period. Discussing Numbers of DPs A discussion of the total numbers of DPs in Europe and Central Asia, or in a given situation, is important in highlighting the magni- tude of the displacement problem. However, the numbers of people originally displaced or remaining displaced are subject to dispute in 6 Table 1.1 Major Displacements in Europe and Central Asia, 1990­2002 Total country population Estimated during year(s) Displaced as a largest Year(s) of of maximum percentage of Host country number maximum displacement the total or territory Displaced group displaced displacement (millions) population Albania Kosovar Albanian 465,000 1999 3.4 12% refugees Armenia Ethnic Armenian refugees from 304,000 1995 3.8 7% Azerbaijan Azerbaijan IDPs from Nagorno-Karabakh 670,000 1995 7.7 9% Ethnic Azeri refugees from Armenia 279,000 1994 7.5 4% and Meskhetian Turk refugees Bosnia and Serb, Croat, Bosniac IDPs 1,300,000 1993 3.4a 38% Herzegovina Croatia Croat refugees from Bosnia and 420,000 1992 4.8 8% Serbia and Montenegro Croat IDPs from Vukovar and Krajina 350,000 1993 4.8 7% Georgia Georgian IDPs from Abkhazia and 285,000 1996 5.4 5% S. Ossetia Macedonia Kosovar Albanian refugees 360,000 1999 2.0 15% Slav and Albanian IDPsb 175,000 2001 2.0 9% Ingushetia, Russia Chechen IDPs 170,000 2000 0.3 35% Tajikistan Tajik IDPs 520,000 1993 5.6 9% Turkey Kurdish IDPs 1,000,000 1994­97 65.3b 1% Serbia and Serb refugees from Bosnia and Croatia 621,000 1993 10.5 6% Montenegro Serb and Roma IDPs from Kosovo 277,000 1999 10.6 3% Kosovo, Serbia Albanian IDPs 600,000 1999 2.0 30% and Montenegro Sources: For total populations (except Ingushetia and Kosovo): World Bank, World Development Indicators, 1990­2000. For the total and IDP populations of Ingushetia: Norwegian Refugee Council, IDP Project, Global Database, 2003; http://www.db.idpproject.org. For Kosovo population: Statistical Office of Kosovo, Kosovo and Its Population: A Brief Description (2003). For refugee and IDP populations (except Ingushetia): U.S. Committee for Refugees, World Refugee Surveys (Washington, D.C., 1991­2003). Note: Determining percentage displaced: For IDPs, IDPs/total population; for refugees, refugees/(total population refugees). For Ingushetia, IDPs/(total population IDPs) because only the territory's population is used. a. 1995 population estimate. b. 1994 population estimate. 7 8 living in limbo many countries because of varied definitions of what constitutes dis- placement or difficulty, particularly in internal displacement situa- tions, in tracking the numbers. The situation in Turkey offers an important example of this difficulty. After a visit to Turkey in late 2002, Dr. Francis M. Deng, the representative of the UN secretary- general on Internally Displaced Persons, reported the following to the United Nations Economic and Social Council: According to government figures, by the end of 1999, 378,000 persons had been "evacuated" by the security forces from 3,165 rural settle- ments in the southeast. However, it has been observed that this figure does not include persons who were not evacuated but who left their homes as a result of the general insecurity, or because of conflict with PKK or the "village guards," a state-sponsored civil defense force com- prised of local Kurds formed to defend their villages against attacks by the PKK and to deny PKK logistical support from the villages in the area. While reports from Turkish NGOs claim that between 2 and 4.5 million Kurds have been displaced, outside observers contend that a "credible estimate" of the number of persons who remain displaced in 2001 was around 1 million.4 The difficulty in tracking numbers of displaced in the region emerges from several factors, some related to the dynamics of individual situa- tions and some stemming from practical issues related to the method- ology of how DPs are counted globally. Statistics relating to DPs are often inherently political, with a range of incentives on the part of various actors to inflate or deflate numbers. In addition, it is very likely that by the time this study is published, some numbers cited will be out of date, particularly in Bosnia, where the displacement situation has elements of fluidity. However, since the citing of DP population statistics is intended only as an illustration of magnitude, such changes are not likely to have much impact on this study's overall scope and conclusions. Absolute numbers of DPs, in any case, form only part of the story. The presence of DPs in such large numbers implies that displacement's impacts are related not only to the direct vulnerability of these groups but also to the effects their presence have on already fragile socioeco- nomic situations. Such impacts represent a kind of multiplier effect of displacement. Government safety net expenditure on DPs is one aspect of the impacts that displacement has on affected societies. Another has been the significant amount of assistance in cash and kind provided by the region's local populations to DPs. Such assistance, although diffi- cult to quantify, clearly represents a significant sacrifice in countries such as Serbia and Turkey, particularly during the early years of dis- placement. In addition, the displaced, particularly in the Caucasus, introduction 9 have occupied a wide range of social infrastructure and have put sig- nificant pressure on social services and the labor market in areas with high DP concentrations. From Humanitarianism to Poverty Reduction As international humanitarian assistance to DPs declines and the years of displacement wear on, questions increasingly arise as to whether displaced populations continue to deserve special attention, particularly in societies suffering through the multifaceted effects of economic and political transitions. In the first years of displacement, international humanitarian assistance was a significant factor in the survival strategies of many displaced groups. While international assistance has not diminished for Chechen IDPs, elsewhere in the re- gion the attention of the world has moved on. Humanitarian agen- cies, facing reductions of funds from donors and never having intended to act as long-term safety nets in such situations, are reduc- ing their focus on displacement in the region. The gap they have left in DP aid is being filled, in a number of nations, by an increase in host government assistance and initiatives. In some countries, DP pro- grams represent a major drain on domestic public expenditure with consequent opportunity costs for other social expenditures. The gov- ernment of Georgia is allocating nearly $30 million to support IDPs. This represents a social protection program second only to pensions.5 In Azerbaijan, government expenditures on IDPs have reached over $150 million per annum, 3 percent of gross domestic product, and exceeded the cost of all other individual social protection programs, including pensions.6 Another impetus for developing a clearer understanding of the comparative vulnerability of displaced populations emerges from a gradual shift in the institutional locus of responsibility to line min- istries of host governments and to development-oriented agencies. In the early years of displacement, responsibility for DPs lay within spe- cialized ministries of host governments and international humanitar- ian agencies, and such discussions often took place somewhat in isolation from broader discussions of national priorities in poverty reduction (and the use of international development assistance). As a result, the targeting of such groups for special attention related, on the one hand, to situation-specific political concerns and, on the other, to a rights-based approach embedded within international humanitarian law, more specifically, to humanitarian agency mandates. As the years of displacement have stretched on, the venue for actions concerning 10 living in limbo displaced groups has changed to the sphere of national development planners and international development agencies, including the World Bank and the United Nations Development Programme (UNDP). The core strategic assessments of issues concerning displacement no longer build toward United Nations Consolidated Appeals (CAPs) and dedi- cated humanitarian ministries but rather feed into larger debates about resource distributions and national strategies for poverty reduc- tion (PRSPs). In this new sphere, the identification of a group as refugees or IDPs does not, in and of itself, denote a justifiable focus of attention. Within the overall strategies of host governments and international development agencies such a focus must be based instead upon two demonstrated conditions. First, the group must have significant im- poverishment as compared with that of other groups in a given society. Second, there must be a need for specific attention to address factors excluding the group from channels of assistance and empowerment available to others or a need for dedicated programs to target the group's special needs. It is not sufficient simply to assert the rights of displaced populations, particularly in situations in which other groups within a society are perceived as similarly impoverished and, as is typically the case, public resources are limited domestically and inter- nationally. Nor is a demonstration of a displaced population's com- parative impoverishment alone enough. The policy implications of such a conclusion do not require any special attention to the displaced in terms of programming; they simply contribute to the larger need for poverty reduction strategies in general. In other words, it is necessary to demonstrate that displaced populations, when compared with other groups in an affected society, have a specific pattern of vulnerability. This study speaks to many audiences in both the humanitarian and development communities. With this in mind, we wish to stress that a focus on vulnerability does not preclude or challenge one that is rights- based. We see the former as a complementary, strategic direction that helps operationalize a rights-based approach in specific situations. In the 1970s and 1980s, humanitarian agencies gradually recognized that a mandate for refugee "protection" required a significant shift toward organizing the provision of basic needs such as food and shel- ter in addition to--not the exclusion of--legal protection. Similarly, a focus on vulnerability as a strategic underpinning for developing mitigation strategies contributes to the practical implementation of international humanitarian law. Nor is a focus on vulnerability a new approach. International assistance strategies have for many years, in certain circumstances, tried to target the most vulnerable among the displaced. Agencies have even broadened their assistance strategies introduction 11 beyond the displaced (as in the war years in Bosnia) when a society- wide vulnerability has been recognized. (It is conceivable that there are cases in the region and elsewhere in which a vulnerability lens would not result in prioritization of DPs within a poverty mitigation strategy. In the case of the analysis within this region, the study does, in fact, find ample justification for contin- ued attention to DPs even in comparison with other sources of vulner- ability within affected societies.) Definitions of Terminology The last section of this chapter will discuss the organization of the study and its strategy for reviewing vulnerability issues among DPs. However, before this, it is necessary to define some of the terms that will be used since there are variable definitions and perceptions of dif- ferent terminology both in regard to DPs themselves and the meanings and use of the terms poverty and vulnerability. Refugee and IDP The United Nations High Commissioner for Refugees (UNHCR) and other humanitarian agencies, drawing on the 1951 Convention on Refugees, defines a refugee as someone who has fled his home owing to well-founded fear of being persecuted for reasons of race, reli- gion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that coun- try; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or, owing to such fear, is unwilling to return to it.7 Based on this definition, the United Nations' Guiding Principles on Internal Displacement defines IDPs as persons or groups of persons who have been forced or obliged to flee or to leave their homes or places of habitual residence, in particular as a result of or in order to avoid the effects of armed conflict, situations of generalized violence, violations of human rights or natural or human- made disasters, and who have not crossed an internationally recognized State border.8 Although such definitions help explain how displacement is initi- ated, they do not facilitate our understanding of when displacement ends, which has increasingly become a topic of concern globally. Nor 12 living in limbo do these definitions formally allow for consideration of the possibility that populations can be more or less displaced. In Europe and Central Asia and elsewhere in the world, there are several examples of DP groups who are not entirely displaced. These groups have the capabil- ity, albeit restrictedly, to move back and forth from their homes to their places of asylum for purposes of farming, trade, or active partic- ipation in conflict. Other DPs are completely blocked from their home area for periods of years. If we view displacement as an accumulation of impacts rather than merely the fact of physical displacement from one's home area, there are also varying degrees of displacement in ref- erence to social alienation and other factors. While, from an interna- tional humanitarian legal perspective, it is fully understandable and justified to fix populations into categories as "displaced" or "not dis- placed," from a standpoint of vulnerability, there are nuances and de- grees of displacement. In a sense, no matter how awkward the phrase, the reality is better captured by a spectrum of "displacedness," a process that differs from situation to situation in degrees, rather than a catchall category of displacement. For the purposes of completeness, this study has reviewed information on populations identified as refugees or IDPs by either governments or international agencies and those whose identification by others or themselves is likely to have im- plications for specialized vulnerability. The key elements of the definition of DP, for the purpose of this study, are dislocation from home and the inability to return. Whether a population consists of refugees or IDPs is only meaningful when this status either creates specific obstacles or opportunities with respect to self-reliance or, strategically, creates legal and political tools within either international humanitarian law or domestic legislation to facili- tate strategies for reduction of vulnerability. In the European context, the differences between refugees and IDPs are sometimes blurred. As noted previously, with the exception of the Kurdish population dis- placed in Turkey, the vast majority of the region's displacement re- sulted from the conflicts that preceded and accompanied the breakup of two multiethnic states, the Soviet Union and the Socialist Federal Republic of Yugoslavia. Thus, for example, ethnic Armenians who left Azerbaijan in 1988 and were viewed as IDPs because they remained within the Soviet Union became refugees in 1991 when Azerbaijan and Armenia became independent states. Serbs from Croatia and Bosnia and Herzegovina settled in Belgrade, formally refugees, are living in the capital city of what used to be their country. In fact, the govern- ment recognizes their plight as its responsibility in a way few host gov- ernments of refugees do worldwide. Some refugee groups, such as the same Serbs from Croatia and Bosnia and Herzegovina in Serbia and introduction 13 ethnic Armenians from Azerbaijan in Armenia, have, to varying ex- tents, taken citizenship in their countries of asylum. According to in- ternational law, this situation suggests that they are no longer formally entitled to refugee status and its protection. However, if lingering impacts of displacement remain relevant to an understanding of the vulnerability of these groups, for the purposes of poverty reduction strategies, it is still useful to differentiate them from local populations. For the purposes of this differentiation, in this study we will refer to such groups as refugees or IDPs regardless of whether their status has legally altered. The United States Committee for Refugees (USCR), in its collection of data on displaced populations globally, has recognized a category labeled "refugee-like" situations. In this category, the committee places groups such as refugees in Azerbaijan and Armenia as well as several groups of displaced populations in Russia. The USCR notes: Many peoples live in situations similar to those of refugees, but do not meet the narrow refugee definition. Some are regarded by host govern- ments simply as illegal aliens; others are tolerated or ignored. In many such cases, and often in the absence of credible refugee determination procedures, it is difficult to determine who among them might be refugees. Other refugee-like people are stateless, denied the protection afforded by citizenship.9 Table 1.2, drawn from USCR data, utilizes this terminology, but for the remainder of this study we will refer to all these groups as "refugees" or "IDPs" regardless of whether their status has legally altered. Poverty and Vulnerability Poverty, according to the World Development Report 2000/2001,10 can be summarized in one word: deprivation. More precisely, poverty implies being hungry, lacking shelter and clothing, being sick and not cared for, and being illiterate and not schooled. It also implies being vulnerable to adverse events and being mistreated, excluded, or both by state and societal institutions. Poverty is a multidimensional phe- nomenon that includes not only material deprivation (as measured by income and consumption) but also low achievements in education and health, vulnerability and exposure to risk, and powerlessness (as mea- sured by civil and political liberties). Vulnerability is the risk that a household will experience an episode of poverty as measured by income and consumption or some other di- mension of well-being. Assessing a household's vulnerability requires information on the household itself (physical assets, human capital, 14 living in limbo Table 1.2 Displacement in Europe and Central Asia, Major Groups, December 2002 Country Displaced Armenia refugees 11,000 refugee-like 255,000 IDPs 50,000 Total 316,000 Azerbaijan refugees 11,000 refugee-like 270,000 IDPs 576,000 Total 957,000 Bosnia and Herzegovina refugees 34,000 IDPs 368,000 Total 402,000 Croatia refugees 8,000 IDPs 17,000 Total 25,000 Georgia refugees 4,000 IDPs 262,000 Total 266,000 Macedonia, FYR refugees 3,000 IDPs 9,000 Total 12,000 Russia refugees 17,000 refugee-like 492,000 IDPs 474,000a Total 983,000 Turkey refugees 10,000 IDPs 400,000­ 1,000,000a Total 410,000­ 1,010,000a Serbia and Montenegro refugees 353,000 IDPs 262,000a Total 615,000 Source: U.S. Committee for Refugees, World Refugee Survey 2003 (Washington, D.C., 2003). a. Reliable estimates unavailable. introduction 15 income diversification) and its links with formal (social assistance, pensions, unemployment benefits) and informal (sources of transfers in cash or kind) safety nets. The World Food Programme (WFP) de- fines this more simply: "vulnerability exposure to risk inability to cope."11 Jorgensen and Holzmann extend this definition by emphasiz- ing that vulnerability is not only the risk of falling further into poverty but also the obstacles to bettering a situation.12 Literature on involun- tary resettlement (development-induced displacement), notably within the work of Michael Cernea, speaks in a similar light of a "risk of im- poverishment."13 In the case of DPs, as with many vulnerable groups, vulnerability is not only about the tangible impediments to bettering their conditions but also about the psychological perceptions of the future, the sense of self-empowerment. Within the concept of vulnerability, it has been recognized that sub- groups in a society may be subject to a set of common circumstances and factors that amount to a specialized pattern of vulnerability, with its own characteristics and impacts. In such cases, specific barriers block an entire group from enhancing their resilience to poverty in ways that do not apply to other segments of the population. Group vulnerability then is recognition that not only are members of a group more vulnerable but also their vulnerability is somehow different. Such special characteristics may result from exclusion. Examples include exclusion based on gender, age, or identity (such as ethnicity, religion, or language) and may be the result of formal public policy, in- formal social prejudice, or even a group's internal dynamics. Exclusion can affect social relations, economic relations, or even legal rights. The interaction between these arenas of relationships indicates that exclu- sion may have multifaceted direct and indirect effects in all these areas. Vulnerability may also be increased by dissolution of internal group cohesion. The World Bank recognizes, for example, that the levels of vulnerability among indigenous peoples are closely tied not only to household survival strategies but also to group strategies and cultural mores beyond the importance of social cohesion and collective action recognized among other populations. This study situates itself within the context of these various specialized treatments of vulnerability. It explores the impacts of conflict-induced displacement as a specific type of vulnerability with its own characteristics, assessing the impacts of exclusion of DPs from local social networks in their areas of settle- ment as well as the levels of social cohesion within DP communities. One peculiar aspect of the vulnerability of displaced populations threading itself through the discussions in all the chapters is what might be termed the Janus-faced dilemma of DPs, which stems from their limbo status. They are often caught in a situation that forces them to look in two directions at once. Vulnerability is, by definition, 16 living in limbo related to how groups or households cope with shocks to their sur- vival strategies.14 For DPs, the future not only is uncertain but also incorporates two very different scenarios: return home or integration in situ (or in a third place). Actions that may eventually mitigate vulnerability upon a return home--such as retention of strong social cohesion in DP communities--could undermine efforts to strengthen bonds with local networks in the area of settlement, deterring or slow- ing down integration locally. Similarly, efforts among DPs (particu- larly younger generations) to acculturate themselves and adapt to a new environment (for example, an urban settlement) may decrease vulnerability for the present but create tensions and difficulties should a family reintegration to a much different environment (such as a rural one) eventually occur. Organization of the Study A core task of this study is to review the available information and try to come to some conclusions about the extent of continued impover- ishment and vulnerability of these groups. If a focus on displaced pop- ulations is warranted in the context of overall priorities in addressing issues of poverty and vulnerability in specific countries, a better under- standing of the patterns and characteristics of vulnerability caused by conflict-induced displacement is key in designing individual situational responses. We attempt to clarify the specific nature of displacement- induced vulnerability in this context and the way in which it differs from the condition of host populations. In this regard, we have made an effort, wherever possible, to draw on data and other information that allows for a direct comparison of host populations and displaced populations. Key Questions Using the multidimensional definition of poverty and vulnerability discussed in the previous section, the study draws upon a range of existing surveys and studies on various country situations to attempt to answer four key questions: · Are DPs poorer, more vulnerable, or both than the local population? · Are the consequences of this poverty and vulnerability different? · Are the causes of this poverty and vulnerability different? · What factors enhance or mitigate this poverty and vulnerability? introduction 17 A dual focus on poverty and vulnerability is important in this regard because a focus solely on a measurement and description of poverty of DPs would insufficiently describe their situations. It is not that the displaced are not poor. In fact, the results of the study show that in most situations, even after long periods of displacement, DPs in the region remain poorer, on average, than surrounding popula- tions. However, even when they are not poorer relative to a society's other potentially vulnerable groups, their survival strategies tend to exhibit particular elements of fragility due to material and psycholog- ical factors stemming from their displacement. We note, for example, that DP coping strategies are highly sensitive to sudden shifts in host government and donor policies due to the displaced's significant de- pendence on public transfers and, perhaps in a larger sense, the politi- cal factors at the root of their continued displacement. Methodology The following indicators have been utilized within the study to repre- sent each dimension of poverty: consumption and income, sources of income, housing, education, health, agriculture, employment, social capital, and psychosocial conditions. Our objective has been to find indicators that are fairly constant across many country surveys to an- alyze broader patterns. This being said, all data are not available for all countries. However, we have tried to report data that are available for at least several displacement situations. Where potentially signifi- cant information exists on only one or two countries, we have utilized it but with appropriate caveats on implications for other situations. In addition to a range of interviews and fieldwork, the study draws upon more than 90 major surveys and poverty assessments and 300 field re- ports on displaced populations in 12 countries in Europe and Central Asia. An annex to the study provides a more detailed description of the methodology utilized. Outline of the Study Analysis has been divided into categories corresponding to different dimensions of poverty and vulnerability and the special circumstances of DPs. Each chapter will cite the available data and review the signif- icance of the data and related information for assessing the vulnerabil- ity of the displaced. Chapter 2 will review information on the material well-being of DPs, including overall poverty indicators and the com- position of household incomes. Chapter 3 will explore the employ- ment of DPs, including agricultural employment and access to land, as well as data on rates of unemployment and some of the possible 18 living in limbo factors that increase unemployment among these groups. Chapter 4 will address the issue of shelter and housing and settlement patterns. Chapter 5 will focus on human capital issues, including demographics, education, health, and psychosocial factors. Chapter 6 deals with issues relating to the social capital of DPs, such as social cohesion among DPs and their social inclusion and exclusion in relation to host populations. Chapter 7 will review the relationship between displaced populations and public authorities. Many impacts of displacement are cross-cutting, relating to more than one category. This in itself repre- sents an important conclusion. It also means that some facets of DP issues, such as settlement patterns, that have implications for various dimensions of vulnerability will be discussed in several places. Chap- ter 8 will conclude by summarizing the major findings and discussing some of the major implications. Following the concluding chapter, we have provided a number of sector and issue-specific recommendations. Notes 1. Within this study, we will only elaborate on the details and chronology of a conflict when necessary to explain the background of specific dynamics. 2. In any case, personal preferences among DPs, the high cost of reconstruc- tion, and the challenge of economic revival of home communities suggest that even if all political problems were to be solved in a given situation, the vulnerability of DPs even then might not be adequately addressed. 3. As this study was being finalized, the Federal Republic of Yugoslavia changed its name to Serbia and Montenegro. Throughout this study, we have used Serbia and Montenegro to refer to what was the Federal Repub- lic of Yugoslavia. We have also used "the former Yugoslavia" to refer to what was the Socialist Federal Republic of Yugoslavia. 4. ECOSOC (UN Economic and Social Council), Commission on Human Rights, Specific Groups and Individuals: Mass Exoduses and Displaced Persons. Addendum: Profiles in Displacement: Turkey, report by Francis M. Deng, 59th sess., November 2002 (E/CN.4/2003/86/Add.2). 5. Larry Dershem, Nana Gurgenidze, and Steven Holtzman, Poverty and Vulnerability among Internally Displaced Persons in Georgia. World Bank internal report (Washington, D.C., 2002). 6. 2002 World Bank data. 7. UNHCR (United Nations High Commissioner for Refugees), Convention and Protocol Relating to the Status of Refugees. UNHCR (Geneva, 1996). 8. UNOCHA (UN Office for the Coordination of Humanitarian Affairs), Guiding Principles on Internal Displacement (Geneva: UNOCHA, 1997). introduction 19 9. See http://www.refugees.org/world/statistics/wrs99_table9.htm. 10. World Bank, World Development Report 2000/2001: Attacking Poverty (Washington, D.C.: World Bank, 2001). 11. WFP (World Food Programme), An Overview of Vulnerability Analysis and Mapping (VAM). WFP (Rome, 2002). 12. Robert Holzmann and Steen Lau Jorgensen, Social Risk Management: A New Conceptual Framework for Social Protection and Beyond. Social Protection Discussion Paper 0006, Human Development Network, World Bank (Washington, D.C., 2000). 13. Michael Cernea, Risks, Safeguards, and Reconstruction, in Risks and Reconstruction: Experiences of Resettlers and Refugees, ed. Michael Cernea and Christopher McDowell (Washington, D.C.: World Bank, 2000). Cernea deconstructs this term into the following major compo- nents: "landlessness, joblessness, homelessness, marginalization, food in- security, increased morbidity, loss of access to common property assets, and community disarticulation." 14. This statement also implies that a concern for vulnerability is applicable not only to those who are currently poor but also to those who may become poor. They have a higher risk of a deteriorating position due to specific factors related to their identity, the elements of their survival strat- egy, or other aspects of their household income or assets that may be par- ticularly susceptible to general or specific varieties of external shocks. chapter 2 Material Well-Being Summary: This chapter focuses on the material well-being of DPs. It begins by comparing the poverty of DPs and that of local populations. The chapter then discusses the contribution of public and international (humanitarian assistance) transfers to DP household incomes. It con- cludes by highlighting the role of private, or interhousehold, transfers, such as remittances, and other less conventional contributions to DP livelihood strategies. THIS CHAPTER REVIEWS DATA ON THE material well-being of displaced populations in the Europe and Central Asia region, examining, for different reasons, measures of both consumption and income. Con- sumption data are reviewed as an indicator of poverty. Income data are reviewed as a means of better understanding the elements of DP household budgets and survival strategies. Though not uniform, the available data do seem to indicate that most displaced populations in the region are, on average, worse off in terms of consumption than the host populations surrounding them. This does not mean that all DPs are poor or that they are necessarily the worst-off group in a given society, only that, when compared with national populations of host countries, the displaced appear, on aver- age, to be worse off. The relative contribution of various sources to overall DP household incomes seems to vary significantly, in many cases, from that of local populations. DP household incomes show more reliance on public transfers and humanitarian aid and less on wage labor. DP employment patterns tend to be concentrated in gray economy activities, such as small-scale trading, day labor, and sea- sonal labor. Unemployment rates of the displaced are higher, some- times twice or even three times those of local populations. The higher level of poverty among DPs is especially of concern in light of the following: · This is true even after factoring humanitarian assistance and public transfers into DP consumption data. 21 22 living in limbo · DP households, unlike the households of many other vulnerable groups (pensioners, female-headed households, populations with disabilities), are demographically quite similar to host populations in terms of size and number of able-bodied adult members.1 · In many cases, DPs have already been displaced for periods of a decade or more. Countries playing host to large displaced populations tend to have problems of poverty and vulnerability that extend far beyond the spe- cific situations of DPs living within their borders. When the conflict is taking place on a country's own soil, destruction and deprivation may be affecting the whole population. When the conflict is in a neighbor- ing country (or neighboring region of a country), the spillover effect of instability and disruption caused by the conflict are often significant factors. In the past decade, most of the region's countries hosting refugees or IDPs have undergone what might be termed a "triple transition." Except for Turkey and Albania,2 all the countries cited in this study, as noted previously, are successor states of the former Soviet Union and Yugoslavia. Their economies and populations have suffered from not only the impacts of conflict (their own and those of their neighbors) but also the disruptions accompanying the birth pangs of transitions from large, multiethnic states to the formation and consolidation of independent states, as well as eco- nomic transitions from different varieties of state socialism to greater reliance on a free-market economy.3 It has been estimated that Georgia's national output dropped 80 percent during its first decade of independence.4 These intertwined legacies have led to transitional poverty as well as widespread feelings of loss and nostalgia for the guaranteed employment and social benefits of previous systems. Therefore, any conclusion that DPs suffer from a specialized or more acute vulnerability must take into account the diminished circumstances of the surrounding host populations. Many surveys of the displaced do not contain comparative infor- mation on host populations. Thus, while useful in identifying DPs' absolute levels of poverty, they do not help us understand, in a given situation, whether displacement is the core reason for this poverty or the poverty is widespread, affecting all groups. It proved possible, however, to find comparative data representing most sig- nificant displacement situations. material well-being 23 Consumption In reviewing the range of surveys and studies available, we found data comparing the material well-being of the displaced with that of the local population for seven out of the region's thirteen major displacement- affected countries or territories. Although they used different ap- proaches to poverty measurement, the data generally bore out the hypothesis that the displaced, when compared with local populations, are poorer (see figure 2.1 below). In six of these seven countries or territories (Armenia, Bosnia and Herzegovina, Croatia, Georgia, Figure 2.1 Poverty in Selected Countries Armenia Refugees (private accommodation) Refugees (collective accommodation) Azerbaijan Bosnia Combined IDPs and refugees Kosovo Local Refugees Serbia IDPs 0 5 10 15 20 25 30 35 Poor or extremely poor (percent) Sources: Armenia: UN Office in Armenia, Poverty of Vulnerable Groups in Armenia: Comparative Analysis of Refugees and Local Populations. UN Dept. of Public Information (Yerevan, 1999). Azerbaijan: World Bank data (2002). Bosnia and Herzegovina: World Bank, Bosnia and Herzegovina: Poverty Assessment (Washington, D.C.: World Bank, 2003). Kosovo: World Bank, Kosovo Poverty Assessment (Washington, D.C.: World Bank, 2001), vol. 2. Serbia: World Bank data (2003). Note: Graph refers to "extreme poverty" in Armenia, Azerbaijan, and Kosovo and "poverty" in Bosnia and Herzegovina and Serbia. 24 living in limbo Kosovo, and Serbia and Montenegro), DPs are, on average, clearly worse off. Again, given the shocks that have affected the local popula- tions of host countries and the considerable duration of displacement in these cases, this finding is quite significant.5 Only in Azerbaijan do IDPs, on average, appear to be equal to or slightly better off than locals, at least in terms of household expenditure figures. As will be noted in a later section, this difference in the Azerbaijan situation rests on the availability of relatively high government safety net payments to this group rather than on any greater capability toward self-reliance or in- trinsic characteristic of the group. Below, we review the evidence on a country-by-country basis. As mentioned above, while numerous surveys that address either poverty or displacement issues, or both, exist for each country, very few of them provide data allowing direct comparison between the conditions of DPs and those of non-DPs. For the purposes of the study, we exam- ine those studies that seem the most robust in terms of methodology and sampling. As will become clear from the variety of poverty analy- sis frameworks, even among these studies, cross-country comparisons must be treated with caution. As we have already observed, however, these measurement inconsistencies, while precluding streamlined data output, do not subvert the argument. Our primary objective is not to compare degrees of poverty between DPs in different countries, but to show that the economic and social conditions of DPs are subpar in nearly every country. Serbia and Montenegro (excluding the province of Kosovo) hosts two basic groupings of DPs: refugees that arrived from Bosnia and Herzegovina and Croatia during the early 1990s wars, and ethnic Serb and Roma Kosovars that were displaced during the 1999 Kosovo war. Both refugees and IDPs are poorer than the local population. The recent World Bank poverty assessment for the country found 17 per- cent of IDPs and 21 percent of refugees in Serbia to be poor compared with 10 percent of the local population.6 A less recent International Committee of the Red Cross (ICRC) survey of Serbia and Montenegro found that despite substantial humanitarian assistance, 47 percent of IDP households fell below the poverty line as opposed to only 14 per- cent of local households.7 Moreover, monthly household expenditures for IDPs were about two times less than those for locals. Displacement in other situations in the Balkans (Bosnia and Herzegovina, Croatia, and Kosovo)8 represents, in a sense, residual displacement. In each of these cases, a significant majority of those once displaced have been able to return home. While the numbers are relatively small inside Croatia and Kosovo, in the case of Bosnia and Herzegovina, the internally displaced continue to represent material well-being 25 approximately 10 percent of the country's entire population (al- though this is only a small fraction of the estimated 60 percent of Bosnians who experienced displacement during the war and immedi- ate post­Dayton Peace Accords years).9 A postconflict assessment of Bosnia and Herzegovina found the poverty risk of the displaced-- using income data to measure welfare--to be 37 percent (31 percent for the federation and 56 percent for the Republic of Srpska) as op- posed to only 24 percent for the nondisplaced.10 Similarly, the first World Bank poverty assessment for Bosnia and Herzegovina found 32 percent of the displaced to be poor compared with 19 percent of the general population. Even after controlling for other factors asso- ciated with poverty in Bosnia and Herzegovina (such as education, number of children), the displaced were 8 percent more likely to be poor.11 The recent World Bank poverty assessment for Kosovo found that those who remain internally displaced there were twice as likely as the general population to be extremely poor.12 The World Bank poverty assessment for Croatia estimates the poverty rate of IDPs in collective centers to be 50 percent, as opposed to just 8 percent for the general population.13 In Georgia, while there are three sets of DPs from different con- flicts, over 95 percent of IDPs in the country are displaced because of the still unresolved 1992­93 internal conflict in Abkhazia. The rest of the displaced include a small number of IDPs from South Ossetia and some 7,000 refugees from neighboring Chechnya. Little reliable infor- mation regarding refugees from Chechnya exists. In regard to the much larger population of IDPs, available data show they, too, are poorer than the local population. However, there is some separation in the data depending upon mode of settlement. According to a 2000 International Federation of Red Cross and Red Crescent Societies (IFRC) study,14 IDPs in collective centers, about half the IDP popula- tion, report 20 percent lower monthly expenditures (on a per adult equivalency basis) than the local population, while IDPs settled in pri- vate accommodations report slightly higher (6 percent) expenditures than the local population. The same study, when using food expendi- ture as a proxy for poverty, found both IDP groups to be worse off than the local population. The IFRC study also notes that "from the information available, it is likely that without government assistance, incomes of the local population exceed those of IDPs in private accommodation by about a third and those of IDPs in collective cen- ters by about one-half." Similarly, a more focused recent study of the Zugdidi region in Georgia found monthly expenditures to be lower for IDPs.15 The 1999 World Bank poverty assessment for Georgia found that 4 percent of IDPs were poor, as opposed to 10 percent of the 26 living in limbo general population.16 These findings are, however, more or less consis- tent with other surveys since the poverty assessment only included IDP households in private accommodations and not collective centers. Azerbaijan hosts two DP groups: about 600,000 Azeri IDPs from Nagorno-Karabakh and surrounding districts, displaced in 1992­93, and about 200,000 ethnic Azeri refugees who left Armenia during 1988­90. Observers generally agree that because of the longer displacement and the availability in certain areas of housing left by fleeing ethnic Armenian populations of Baku and Sumgait, the vast majority of these DPs have been integrated into Azeri society. However, since there has been no assessment of the situation of this population, no definitive statements can be made. In regard to the larger population of IDPs, the World Bank has conducted two poverty assessments, one in 1996­97 and the other in 2002, both incorporating special studies of IDP vulnerability. The assessment work completed in 1997, a few years after the initial displacement, concluded that Azeri IDPs were more likely than the general population to be poor (75 percent versus 62 percent) or very poor (38 percent versus 20 percent).17 Similarly, a European Commis- sion Humanitarian Office (ECHO) assessment undertaken during roughly the same period found the monthly expenditures of IDP households to be less than those of the local population.18 The most recent World Bank poverty assessment, conducted in summer 2002,19 however, indicates that IDPs, on average, are now slightly less likely to be poor than the local population.20 As in Georgia, there was some dif- ferentiation by settlement pattern, in this case between IDPs in the capital city and other groups. While IDPs in Baku had the same rates of poverty and extreme poverty as the city's other residents (55 percent and 20 percent, respectively), those in other parts of the country, where about two-thirds of IDPs live, were less likely than their neigh- bors to be poor (35 percent versus 64 percent) and extremely poor (14 percent versus 34 percent). It should be noted, however, that, even in this most recent study, when targeted humanitarian assistance pro- vided by the government and the international community was not considered, IDP household expenditures came out consistently lower than those of other groups in the society. Armenia, similarly, has both IDPs who fled from their homes near the Azerbaijan border in the early 1990s during the Nagorno- Karabakh hostilities and refugees who came from Azerbaijan earlier in the late 1980s. Estimates of IDPs range up to 80,000, and while many observers believe they have generally been reintegrated, we know of no detailed vulnerability assessment. Estimates of the number of refugees from Azerbaijan range up to 200,000. While significant material well-being 27 numbers of refugees remain in collective centers, a large percentage of the total refugee population, in principle, has been more or less inte- grated through the provision of land and housing left by ethnic Azeri refugees. However, we know of only one study comparing refugees with the local population, and it reports that this refugee population is poorer than the local population. Examining all vulnerable groups in Armenia, including refugees, the study found that poverty, as mea- sured by the Human Development Index--which includes unemploy- ment, education, health, and drinking water--was 1.7 times more likely among refugees than locals.21 In addition, using bread con- sumption as a poverty indicator,22 the study discovered the incidence of extreme poverty to be 22 percent among refugees in private accom- modations and 26 percent among refugees in collective centers, as opposed to 13­14 percent for the local population.23 These conclusions regarding the poverty of DP households as per- tains to consumption clearly indicate one of the key elements making the displaced vulnerable. However, even when the incomes of DP households equal those of local households, the displaced's level of vulnerability to external shocks such as natural disasters and medical emergencies (as well as family events such as weddings and funerals) probably is still higher. This is because DPs are likely to have far fewer resources and assets with which to cushion such expenditures.24 Thus far, the surveys in the various countries have told us only that when consumption data are reviewed, DPs appear, on the whole, to be poorer than the local populations. To many observers, this conclusion will not seem surprising. However, given the wide range of crises that have struck the local populations concurrently and the long periods of time since the original displacement of most of the DP groups, it is sig- nificant that the displaced have not been able to make up ground. This is despite the supplementary income provided to DPs in many countries by both governments and international humanitarian assistance agen- cies. The final sections in this chapter will review available information regarding the composition of household incomes of DPs and locals to attempt to clarify how income brackets differ and to see which factors seem most significant in regard to the greater poverty of the displaced. Transfers This section reviews available information on the importance of public and private transfers in DP household incomes in comparison with those in local populations. Specifically, we look at the relative impor- tance of host governments' public transfers, international humanitarian 28 living in limbo assistance, and private transfers. Generally speaking, we find that even in situations such as those in Azerbaijan, in which DPs and local popu- lations have similar levels of expenditure, the composition of the two groups' household incomes differs significantly. As already noted, the displaced receive a considerable amount of in- come from humanitarian aid and government assistance. The limited data available suggest, however, that these sources, while quite sig- nificant, do not compose the majority of the DPs' incomes (as some observers assume). The figures cited below come from a variety of reports, which used different methodologies and definitions for assess- ing the dependence of DP households on public assistance. These are only cited to emphasize the high degree of overall dependence that DPs have on public assistance. Public Transfers Public transfers to the displaced encompass assistance meant specifi- cally for them as well as benefits and entitlements from state sources received for reasons other than DP status, such as pensions, widow's payments, and disabilities, all of which are available to local popula- tions as well. All DPs are, in principle, entitled to the same benefits as other citizens of their home countries. For IDPs, this is a straightfor- ward issue; for refugees, it obviously can present problems since they left their countries for reasons of politics and conflict. It may not be safe or practical to apply for these benefits. For example, many Chechen IDPs in Ingushetia, while entitled to various benefits as Russian citizens, have reported that the expense and security consider- ations in returning to Chechnya to apply for government benefits out- weigh their value.25 However, appreciable numbers of Serb refugees from Croatia, though either unable or unwilling to travel to Croatia during the years of conflict and for a few years afterward, now access entitlements such as state pensions. Given that Croatian pensions are double the value of Serbian pensions, this has invariably represented a significant contribution to survival strategies of refugee households with pensioners. Though accurate data are not available, the UNHCR estimated in 2002 that as many as 40,000 Serb refugees may be enti- tled to Croatian pensions, either from the government or from private companies.26 Hence the importance of such benefits for refugees must be weighed against the transaction costs. Of the region's major host countries, only two governments, Azerbaijan and Georgia, allocate specific benefits for displaced popu- lations.27 In Georgia, IDPs receive monthly cash payments from the national government. In Azerbaijan, in addition to monthly cash material well-being 29 payments, all IDPs are entitled to exemptions from utility payments (notably electricity) and income tax. Currently, those entitlements are available to all IDPs in the two countries, regardless of income or level of poverty. Data regarding the contribution of public transfers to overall DP household incomes could be found only for the three countries of the South Caucasus, but for the reasons stated above, it is likely that the importance of such transfers is higher for DPs there. In Georgia, investigators found government assistance to be of far greater impor- tance as a source of income for the IDP population than for the gen- eral population (50 percent of IDP income as opposed to 13 percent of the general population's income).28 Another statistic supports the distinction, indicating that virtually all IDPs receive some sort of government assistance while 58 percent of the local population do.29 Furthermore, IDPs received larger amounts than locals (GEL 38­40 per month as opposed to GEL 12 per month).30 The Zugdidi region study had similar findings.31 The 1997 poverty assessment for Azerbaijan concluded that government transfers were a more impor- tant source of income for IDPs than for the general population, with 23 percent of IDP incomes and 10 percent of general population incomes coming from government assistance programs.32 The recent poverty assessment in 2002 found that the gap had increased, with 40­47 percent of IDP incomes from government transfers and exemp- tions as opposed to only 9­15 percent of local incomes.33 The study of vulnerable groups in Armenia discovered more refugee households than local ones receiving benefits from the government (43­53 percent versus 31­36 percent). In addition, refugees received larger amounts.34 Another form of government assistance to IDPs in Azerbaijan, Georgia, and Serbia (and to a lesser extent in Bosnia and Herzegovina) bears mentioning, although categorizing it purely as assistance from the concerned governments to the displaced is difficult. This has been the policy of maintaining at least de jure employment of the public sector cadre from the areas of displacement. (See box 3.1 in next chap- ter.) In Georgia, all administrative cadres from Abkhazia retain their formal jobs in displacement and receive a salary. A range of Serb IDPs previously in government employment in Kosovo also receive a percentage (60­80 percent) of their salaries from the government of Serbia, even during displacement. In Azerbaijan, apparently all public employees, including teachers and health workers, are retained on the public payroll. Whether this can be viewed as social safety net assis- tance depends on whether such employees are still working at a level providing public service, a point that, given the limited available information, is difficult for us to judge.35 30 living in limbo International Humanitarian Aid While international humanitarian aid (mostly food) played a signifi- cant role in the survival strategies of many DP households in the first years of displacement, determining how important such aid is to them now and, even more so, how important it will be in the next few years is difficult. Several of the surveys drawn upon for this study were conducted between 1999 and 2002. In the interim, humanitarian assistance targeted to displaced households in a number of countries has declined. For example, in 1997, a Norwegian Refugee Council survey reported that over 70 percent of the IDPs in Georgia said they received humanitarian assistance.36 Three years later, an IFRC survey in the country found just 23­29 percent of IDPs receiving such assis- tance (as opposed to 3 percent of locals).37 A 2002 study of Georgia's Zugdidi region found 43 percent of IDPs receiving humanitarian as- sistance there (as opposed to 28 percent of the local population).38 Even recognizing the possibility of distortions caused by various man- ners of asking different permutations of such a question, the implied reduced aid contributions to DPs over a five-year period do seem to correspond to the overall tracking of major donor contributions for Georgia's IDPs. The WFP, ECHO, and the U.S. Office of Foreign Dis- aster Assistance (USOFDA) have gradually closed down their IDP food assistance programs. The 1997 poverty assessment for Azerbaijan found international humanitarian assistance to be a more significant source of income for IDPs than for the local population.39 Similarly, the ECHO assessment reported that 49 percent of IDPs received humanitarian food assis- tance, as opposed to 10 percent of the general population.40 However, in the 2002 poverty assessment for Azerbaijan, IDP respondents esti- mated that less than 5 percent of their overall incomes came from such assistance. As in Georgia, ECHO had closed its food assistance pro- gram in the intervening period, and most of the WFP food program shifted to broader vulnerability criteria, reducing the focus on IDPs.41 In the Balkans, humanitarian assistance still remains important for a significant portion of refugee and IDP households. In Serbia, the WFP estimated, in 2001, that its food assistance represented 15 percent of expenditures for 200,000 refugees (out of 500,000 refugees in the country) receiving aid from the WFP.42 The limited number of refugees receiving food was due to a hybrid approach to distribution integrating vulnerability criteria into an overall identity- focused refugee support program. The ICRC, which represents the major international assistance program targeting IDPs, reported in the same year that its assistance represented 25 percent of consumption of material well-being 31 IDP households.43 However, in 2003, the committee announced that both programs would end by summer 2004.44 In neighboring Bosnia and Herzegovina, a 1999 social assessment found that more of the dis- placed than the general population received humanitarian assistance (28 percent versus 17 percent).45 Yet between 1999 and 2001, two of the major donors for food assistance in Bosnia and Herzegovina, the ECHO and the WFP, lowered their country budgets, from $20 million to $5 million and from $45.5 million to $700,000, respectively.46 Even taking into account the return home of large numbers of IDPs during this period, the drop is precipitous. Another trend in humanitarian assistance in the past several years, particularly in the Caucasus, further complicates assessment of such aid's contribution to DP survival strategies: a shift toward investment in programs designed to enhance self-reliance. These include business training and microfinance programs for DPs. While the value of such programs lies in supporting eventual increases in DP employment, they do not, per se, represent direct subsidies of living expenses. Thus while representing international investments in DP situations, the programs cannot replace the humanitarian assistance previously provided. The only situation in which humanitarian aid is not declining is that in Chechnya. In 2001, major donors contributed nearly $90 mil- lion. (Again, however, it is difficult to assess the exact value of this assistance in terms of direct contributions to DP household budgets.) No significant international humanitarian assistance and no govern- ment-initiated public transfer scheme aimed at DPs are available to the displaced population in Turkey nor are there reliable data on con- sumption expenditures for this population. Private Transfers There are very few focused data available regarding DPs' private transfers. What data exist do not show significantly different patterns between DPs and other populations. A Save the Children survey in Georgia found that assistance from relatives and friends was a more important source of income for the general population than for the IDP population, comprising 10 percent of the former's income as op- posed to 6 percent of the latter's.47 Similarly, the study of the Zugdidi region reported a greater proportion of locals (36 percent) than IDPs (27 percent) receiving assistance from relatives and friends.48 Like- wise, the IFRC survey not only found that more locals received assis- tance from relatives and friends within the country but also found that 32 living in limbo Box 2.1 Kosovar Refugees and Remittances The impact of displacement on livelihoods assumes that predisplace- ment livelihoods were anchored in the geographic space of the home area. When household income is derived from assets outside the conflict area, such income is likely to be available even during displacement. In Kosovo, a World Bank poverty assessment in 2001 reported that, on average, 40 percent of household incomes came from remittances. Extrapolating from this survey, it is reasonable to conclude that while the Kosovar Albanian displacement was short, if it had endured, it is likely that, even considering potential transaction costs, such income would have been available to DPs, reducing, compared with other displaced populations in other countries, their levels of poverty and vulnerability. they received more remittances from abroad.49 By contrast, an ECHO assessment in Azerbaijan50 reported that the general population and IDPs received similar amounts of food assistance from relatives and friends; the World Bank's 2002 poverty assessment found locals and IDPs to be receiving similar amounts of assistance from relatives and friends.51 Remittances and support from the diaspora have represented im- portant aspects of coping strategies in many conflict situations. It is commonly assumed, for example, that without such aid, many house- holds in Bosnia and Herzegovina, local as well as displaced, would not have survived the war. (See also box 2.1.) A similar level of depen- dence on remittances among displaced and nondisplaced populations underscores the fact that income is not always completely cut off during displacement. Not all assets and income flows are anchored to a location, and displaced persons fortunate enough to have family abroad may continue to receive badly needed funds via wire transfers. Consequently, not all households will be affected to the same degree by the act of displacement. Unconventional Public and Private Transfers The problems in accurately quantifying private transfers from remit- tances, even for local populations, are well known. Measurement of international humanitarian aid has proved complicated for different reasons. Putting a value, as noted above, on income generation, material well-being 33 microfinance programs, or other programs implemented by non- governmental organizations (NGOs), which include various overheads and other expenditures, is the subject of a study in its own right. Even direct assistance in kind, such as food distribution, does not always prove easy to assess because of the multiplicity of donors and varieties of in-kind assistance provided ad hoc or on a regular basis. In addition, typically no reliable centralized database within most host govern- ments exists, and there are problems with verifying that in-kind resources actually reached the intended beneficiaries. The centrality in this process of a few agencies, most notably the WFP, facilitates such measurement to some degree. There are also two other significant forms of assistance to DPs that need to be mentioned but are more difficult still to quantify or measure in any meaningful way. Private Transfers from Host Households There is no reliable way to measure the support that has been provided by private households in several host countries to subsidize the living costs of DP households. Any discussion of DP survival strategies, how- ever, would be incomplete without reference to this phenomenon, which is mentioned in several other contexts in the following chapters. The hosting of DPs in local homes in Georgia, Serbia, Turkey, and elsewhere, if monetized, would clearly represent a major share of DP household expenditures. That being said, in individual cases, such hospitality may have from the beginning, or after a period of time, involved some transactions and contributions to local households in exchange for this hospitality. In addition, it is likely that in many situ- ations local families provided, and may still provide, provisions in both cash and kind to some DP households with family or clan links. Particularly in the first years of displacement of refugees in Serbia, the value of such assistance could conceivably have exceeded public ex- penditure and humanitarian assistance combined and undoubtedly represented a major sacrifice for many local households. Provision of Shelter by Host Governments The provision of shelter to DPs by some governments, in the form of collective centers such as public buildings, abandoned factories, schools, and clinics, is in itself a form of public transfer to DPs. (See chapter 4 for a separate discussion on shelters.) At a minimum, this assistance--most important in the Caucasus but present in the Balkans as well--represents an indirect subsidy to DP households, allowing them to avoid pressure on household incomes through rent payments. However, most types of collective centers are overcrowded 34 living in limbo and dilapidated, and so assessing a value comparable to rental costs of private housing would not lead to a fair estimate. This highlights the need to analyze vulnerability through reference to both material and nonmaterial factors. Although a minority of DPs in collective centers are certainly housed in adequate facilities, this is not the norm. What benefit the displaced receive in terms of avoiding the substitu- tion cost of private rental housing must be weighed against the psychological, social, and health costs of living under the difficult conditions prevailing in many collective centers. In the case of some private facilities, it may be possible to measure the loss of value to pri- vate individuals of DP occupation of their structures. But it is difficult to put an opportunity cost to the state or to local communities of DP occupation of such facilities given the oversupply of public infrastruc- ture, the failure of state enterprises, and other factors that accompa- nied the economic transitions of the 1990s in places such as Georgia and Azerbaijan. Conclusions From the above discussion, we can conclude the following: · DPs seem to be worse off materially than local populations, even after factoring in significant government and humanitarian assis- tance and many years of displacement. · DPs tend to rely more on government and humanitarian assis- tance than do local populations. However, contrary to popular belief, this assistance does not meet all their basic needs and must be supplemented by income from employment. · Humanitarian assistance is on the decline in Europe and Central Asia as displacement has continued for more than 10 years in many places. In a number of countries, this has left a significant gap in DP household incomes. · While there is little information on private transfers, it is clear that remittances played a key role in sustaining DPs in some sit- uations, especially in regard to Bosnians and Kosovar Albanians. · Although little quantitative information on the material contri- bution of households that have hosted DPs exists, this obviously has played a major role in sustaining DPs, especially in the early years of their displacement. In summary, the dependence of DPs on public assistance, whether from host governments, international humanitarian assistance, or both, is an important aspect of household incomes in most situations material well-being 35 of displacement in the region. The major exception appears to be Turkey, where no public assistance has been aimed at IDPs and inter- national humanitarian agencies have not been extensively engaged in assistance. Nonetheless, as has been noted, it is important to recog- nize that while such forms of assistance, where they exist, represent significant components of DP household incomes, nowhere does such assistance represent sufficient aid to fully subsidize even the most basic household survival strategy. The mythology that views dis- placed populations as hapless victims living off public assistance does not accurately reflect the reality of how the region's DPs cope. In places such assistance is available, were it to stop, DPs, on average, clearly would be far worse off than the data currently show. How- ever, reliance on such assistance as a sole source of income would lead to starvation. Despite the obstacles in their way, most DP house- holds are, in one form or another, generating income through employment, even though this employment is frequently in marginal or ad hoc jobs. The next chapter will discuss the displaced's employ- ment situation and review some of the common obstacles limiting employment. Notes 1. As will be noted in chapter 5, most populations of DPs in the region also appear to have similar education and health indicators. 2. Although not one of the successor states, postcommunist Albania is also in a political and economic transition and has experienced major instabil- ity. Multiethnic Turkey, too, does not belong to the category of transition states, but strong similarities--politico-economic reorientation toward the West and territorial dismemberment--may be discerned, dating back to its birth under Ataturk as a modern state. 3. As one observer, commenting on stereotypes of Serbian IDPs from Kosovo, noted: "The problem of unemployment is less about the work ethics of the Kosovo Serbs than it is about the collapse of the Serbian economy." Vladimir Illic, Yugoslavia: Displacement from Kosovo, from Patronage to Self-Help, in Caught between Borders: Response Strategies of the Internally Displaced, ed. Marc Vincent and Birgitte Refslund Sorenson (London: Pluto Press, 2001), 265. 4. Quoted in IFRC (International Federation of Red Cross and Red Crescent Societies), Internally Displaced Persons: A Socio-Economic Survey, Georgia. IFRC (Tbilisi, 2000). Based on World Bank, Transition, the First Ten Years: Analysis and Lessons for Eastern Europe and the Former Soviet Union (Washington, D.C.: World Bank, 2002). 36 living in limbo 5. Available data do not allow for the comparison of the vulnerability of subgroups of the DP population with the exception of housing type. There do seem to be significant differences between IDPs in private accommodations, who appear better off, and those in collective accom- modations in Georgia. However, even in that country, when adding the two groups together, IDPs are worse off than the local population. Issues regarding differences by shelter type and the question of whether shelter type represents a proxy for vulnerability among DPs will be discussed in chapter 4. 6. World Bank data (2003). 7. ICRC (International Committee of the Red Cross), Beneficiary Monitor- ing: A Sample Survey of IDP Households in FRY. ICRC (Geneva, 2001). 8. No reliable comparative data are available for the recent internal dis- placement within Macedonia. 9. Perhaps 600,000, or 15 percent of the country's population, remain abroad within the region and in western Europe, the United States, and elsewhere as refugees; it is difficult to estimate how many of these will eventually return home. 10. World Bank data (1999). Displaced refers to refugees, IDPs, and returnees. 11. World Bank data (2003). Displaced refers to IDPs and refugees. 12. World Bank, Kosovo Poverty Assessment (Washington, D.C.: World Bank, 2001), vol. 2. 13. World Bank, Croatia: Economic Vulnerability and Welfare Study (Washington, D.C.: World Bank, 2001). 14. IFRC, Internally Displaced Persons. 15. George Shakarishvili, Evaluation of Socio-Behavioral Patterns of Local and Internally Displaced Populations in Zugdidi Region of Georgia and Its Applicability for Local Policy Developments. Research Paper, Open Society Institute (Tbilisi, 2002). 16. World Bank, Georgia Poverty and Income Distribution (Washington, D.C.: World Bank, 1999), vol. 1. 17. World Bank, Azerbaijan Poverty Assessment (Washington, D.C.: World Bank, 1997), vol. 1. 18. A. C. de Roos and D. Venekamp, ECHO Food and Medical Assessment in Azerbaijan, including Nakhichevan. ECHO (European Community Humanitarian Office) (Baku, 1996). The assessment covers three regions, including Baku. IDPs living in camps, roughly 10 percent, were not included in the assessment. 19. World Bank data (2002). 20. This may be because of the large increase in government assistance to IDPs since the 1997 poverty assessment. The assessment also notes the possibility that IDP annual incomes may have been overestimated due to material well-being 37 the fact that the survey was conducted during the summer months when incomes for IDPs from seasonal day labor in agriculture and construction work are at their highest. 21. UN Office in Armenia, Poverty of Vulnerable Groups in Armenia: Com- parative Analysis of Refugees and Local Population. United Nations Department of Public Information (Yerevan, 1999). 22. In Armenia, while the share of expenditures spent on food decreases as one moves from the poorest to the richest households, absolute bread consumption increases consistently and smoothly. 23. This raises an interesting question as to why in the exchange of populations between Armenia and Azerbaijan, refugees in Armenia seem to remain poorer while refugees in Azerbaijan are assumed to be well-integrated. This may be due to settlement of refugees in Azerbaijan in urban areas and in Armenia in rural areas (see chapter 4). As already noted, it could also suggest the need to conduct a review of the condition of refugees in Azerbaijan. 24. The 2003 poverty assessment in Bosnia and Herzegovina reported that 70 percent of IDP households had assets of KM 1,000 or less while local households in the country had assets of KM 40,000, 40 times higher. The value of owner-occupied housing for local households represented the majority of this difference. World Bank, Bosnia and Herzegovina: Poverty Assessment (Washington, D.C.: World Bank, 2003). 25. ICRC, Internally Displaced Persons in Ingushetia: Economic Security Survey. ICRC (Geneva, 2002). 26. Personal communication with Fabrizio Hochschild-Drummond, deputy head of office, UNHCR, December 2002, Belgrade. 27. In principle, the federal government of Serbia and Montenegro also pro- vides subsidies to support refugee participation in public health insurance plans. But as of this writing, the payments were several years in arrears. 28. Dershem, Gurgenidze, and Holtzman, Poverty and Vulnerability. 29. IFRC, Internally Displaced Persons. 30. IDPs also received a significant amount of assistance from the Abkhaz government in exile. 31. Shakarishvili, Evaluation of Socio-Behavioral Patterns. 32. World Bank, Azerbaijan Poverty Assessment, vol. 1. 33. This assessment estimates the cash value of electricity exemptions as part of income. In addition, as previously noted, government monthly per capita cash assistance to IDPs in Azerbaijan was increased from 5,000 manat to 25,000 manat in the interval between the two surveys, an increase of 500 percent. 34. UN Office in Armenia, Poverty of Vulnerable Groups in Armenia. 35. The implications for DP relations with the state, particularly in regard to local governments in exile, will be treated separately in chapter 7. 38 living in limbo 36. NRC (Norwegian Refugee Council), Survey of IDPs in Georgia. NRC (Tbilisi, 1997). 37. IFRC, Internally Displaced Persons. 38. Shakarishvili, Evaluation of Socio-Behavioral Patterns. 39. World Bank, Azerbaijan Poverty Assessment, vol. 1. 40. de Roos and Venekamp, Food and Medical Assessment in Azerbaijan. The assessment covers three regions, including Baku. 41. ECHO has drastically reduced its targeting of IDPs for food assistance in the Caucasus, going from an annual budget in 1995 of over $14 million per country in Azerbaijan and Georgia to about $1 million in each coun- try in 2001. 42. WFP, The Survey on Poverty and Living Conditions of the Refugee Beneficiaries of WFP Aid in Serbia. WFP (Belgrade, 2001). 43. ICRC, Sample Survey of IDP Households in FRY. 44. UNHCR-WFP, Joint Assessment Mission Report. UNHCR-WFP (Belgrade, 2003). 45. World Bank, A Social Assessment of Bosnia-Herzegovina (Washington, D.C.: World Bank, 1999). Displaced includes refugees, IDPs, and returnees. 46. Figures drawn from data tables collected from WFP and ECHO for this study. 47. Dershem, Gurgenidze, and Holtzman, Poverty and Vulnerability. 48. Shakarishvili, Evaluation of Socio-Behavioral Patterns. 49. IFRC, Internally Displaced Persons. 50. de Roos and Venekamp, Food and Medical Assessment in Azerbaijan. The assessment covers three regions, including Baku. 51. World Bank data (2002). chapter 3 Employment Summary: This chapter reviews the employment situation of the dis- placed. It begins by comparing the participation of DPs in formal sector employment with that of the local populations. The chapter then discusses access to land and livestock as major determinants of agricul- tural employment. Finally, the unemployment rates of the displaced and locals are compared, and some possible reasons for high DP unemploy- ment offered. IN THE PREVIOUS CHAPTER, WE EMPHASIZED the high level of dependence on public transfers characteristic of DP household incomes in most sit- uations in the region. This dependence has many potential implications for vulnerability. Whether from host governments or international humanitarian assistance, public transfers can provide a short-term stable subsidy to household incomes because they are predictable in regard to amount and, in principle, issued regularly. However, each type of assistance has its drawbacks. Dependence on government transfers means, in effect, dependence on government efficiency. In countries where these payments are important for DP coping strate- gies, late or delayed payments due to shortfalls in government budgets or complications of bureaucracy can seriously affect household con- sumption.1 International humanitarian assistance was mostly stable for many years, but as noted previously, it has been declining in im- portance as donors withdraw support for chronically displaced popu- lations. At any rate, such assistance almost always has been delivered in foodstuffs, which, though essential, cannot be monetized without significant transaction costs and so does not resolve issues regarding other household expenditures. Perhaps the most significant weakness in dependence on these income sources is that they tend to be fixed payments, which may be reduced but have little or no potential for growth.2 Vulnerability, as noted in chapter 1, includes not only susceptibility to shocks, which reduce income and assets, but also obstacles to increasing household 39 40 living in limbo income and assets, which could stabilize coping strategies and bring a household out of poverty. Employment offers DPs' households the only means for increasing their overall incomes to achieve such a step. Unfortunately, the record on DP employment throughout the region is not good. This chapter will look at employment of the displaced, reviewing available data on wage employment and informal economy employ- ment as contributions to household income. It also will examine the availability of agricultural land and overall rates of unemployment for DPs as compared with those for local populations. Finally, we will discuss some of the possible reasons for high DP unemployment. Formal Employment When DPs are employed, it is often in informal sector professions.3 However, a minority of them have been able to access or retain formal wage employment during displacement. Wage employment data were found for IDPs in four affected countries: Azerbaijan, Bosnia and Herzegovina, Georgia, and Serbia and Montenegro. In all four cases, DPs received a much smaller portion of their overall income from wage employment than did local populations; in many cases, it was reported that DPs received lower wages. In the Balkans, data for Serbia and Montenegro and Bosnia and Herzegovina indicate lower reliance on wage income among DPs. The ICRC survey of Serbia and Montenegro found that wages were a more important source of income for residents than for IDPs.4 In addition, the survey reported that, on average, the latter received lower salaries from such employment than did local populations.5 In Bosnia and Herzegovina, a 1997 vulnerability survey found that when DPs and residents were employed in the same professions, the former received lower wages in every category except army and police.6 The social capital study for Bosnia and Herzegovina reports that IDPs there were more likely to have "precarious" jobs, with only 19 percent of them having formal jobs as opposed to 30 percent of the local population.7 In Georgia, two surveys demonstrate similar trends. One conducted by Save the Children found wages to be more than twice as important as a source of income for the general population (39 percent) as for the IDP population (16 percent).8 Similarly, the IFRC survey previously cited found IDPs to have less wage employment (15­20 percent versus 27 percent for locals) and to receive smaller wages (10­18 lari per month versus 64 lari for locals).9 Wage employment for both groups employment 41 mostly meant government jobs in the education, health, and other social services sectors. Among the self-employed, locals were mainly active in the agricultural sector, while a significant number of IDPs were also engaged in various trading activities. Similarly, the study of the Zugdidi region of Georgia10 reported that locals have more access to salaries (74 percent versus 57 percent of the IDPs). The 1997 poverty assessment for Azerbaijan, too, found wages to be a more important source of income for the general population than for the IDP population, composing 38 percent of the former's income and 31 percent of the latter's.11 IDPs were only half as likely to have wage employment, with 36 percent of working age IDP men receiving salaries. Another finding, comparing incomes in rural areas, showed average per capita income among IDPs at US$13.4 per month, a little over half of the US$23.8 average income of local populations.12 The recent poverty assessment found that 76 percent of local incomes, as opposed to 39­48 percent of IDP incomes, are from employment- related activities.13 Box 3.1 discusses the continued employment by several govern- ments of IDPs who had been in government service prior to displace- ment. Two important points on this issue should be noted. First, such employment illustrates the fact that while displacement commonly wipes out the affected populations' previous employment, exceptions do exist. Employment anchored in the necessity of being physically present in the territory of origin (agriculture is the classic example) is completely blocked by displacement. However, some types of employ- ment prove more resilient to displacement, either because the key fac- tors in certain jobs are mobile (such as the skills of a doctor or assets such as a truck or a taxi) or because the locus of the employment always has lain outside the conflict zone. For government employees, even though the locus of employment was, in fact, in the conflict zone (as in Abkhazia, Kosovo, and Nagorno-Karabakh), the employer (the national government) lies outside the zone and thus can choose to re-create the employment or compensate former employees who lost their jobs during the conflict. This leads to another conclusion, raised in the box: this kind of employment--maintenance of jobs by national governments--does not constitute new employment and so does not indicate that affected DPs have been able to adapt to their new envi- ronment and find jobs. For example, in the case of Azerbaijan (the only country for which we have fairly good data on the entire range of DP wage employment) subtracting such employment from overall percentages of DP households with wage employment begins to demonstrate that the displaced have acquired very few new jobs since displacement. 42 living in limbo Box 3.1 Continuity in Public Employment during Displacement As indicated in the previous chapter, host government assistance to displaced populations can take forms beyond humanitarian assistance. Significant numbers of IDPs employed by state structures in Azerbaijan, Georgia, and, to a lesser extent, Serbia retain their salaries and benefits even during displacement. In Azerbaijan, for example, a third of IDP households report having income from government employment.14 In Georgia, about a quarter of the households report such income.15 Each of the three governments has retained on its payroll varying percentages of public employees from original areas of displacement. Such employ- ment does not represent new employment since displacement and further reduces conclusions regarding the proportion of DPs who have successfully found employment after being displaced. It does highlight the fact that displacement does not always disrupt income streams to the same degree in each population. In Georgia, by some accounts, at least 5,000 IDPs remain employed in the Abkhaz government in exile, serving as teachers, health workers, and administrative officials of the government, which receives alloca- tions directly from the Georgian state budget. Universities and public enterprises previously based in Abkhazia are still functioning. In Azerbaijan, over 60 percent of IDPs report being taught in special IDP classes, which create employment for displaced teachers and school administrative staff in a parallel education system.16 In addition, the government has maintained the formal employment of all local-level (raion) administrative officials and employees from the areas under occupation. In Serbia, as indicated earlier, a range of public officials from Kosovo retain 60 to 80 percent of their salaries during displacement. While such populations are in the minority among IDPs in all three countries, their situations offer a greater insight into the varied menu and degrees of vulnerability of DP survival strategies. It is difficult to answer the situational question of why DPs receive lower wages. Is this due to their identity as displaced persons? Or could it be because the desperation of displacement, combined with implicit subsidies to household incomes through humanitarian assistance, al- lows them to accept lower wages, thus undercutting the market? Both reasons probably hold true. In some cases, the wage differential has led to friction between displaced and nondisplaced communities. Anecdo- tal evidence from Croatia suggests that residents were unhappy with the Bosnian Croat (refugee) construction workers, who were "seen as employment 43 responsible for driving down construction wages."17 Similar tensions have been noted in Ingushetia between Chechen IDPs and residents.18 Agriculture (Land and Livestock) Europe and Central Asia's high percentage of DPs with a rural back- ground makes lack of access to land a serious impediment to income stability for many of them. Given the difficult economic transitions in most of the region's countries and the relatively low wages provided within the formal sector, access to land often represents an important supplement to other forms of income in providing food for household consumption. As with residential property, local populations in the Balkans and the Caucasus have benefited from privatization of land in a way the displaced could not. The lack of a mature real estate market, particu- larly in the Caucasus, and the fact that most DPs have minimal savings or other liquid capital also contribute to the displaced's having less stable access to land than do local populations. Even when DPs have sufficient capital to purchase land, if they already own property else- where, they may choose not to buy, viewing such an investment as a long-term prospect in a temporary situation because they plan to return home. The purchase of land, as well as houses, by the displaced is probably an indication that displaced populations, or groups among them, are giving up the possibility of returning home and choosing instead, despite the barriers placed in their way, to attempt to integrate locally. Laws explicitly prohibiting DPs from purchasing land rarely exist; in most countries, however, only citizens can own land, thus blocking refugees. Even with IDPs, who do have the right to purchase land in their own countries, there are de facto constraints (beyond the appreciable question of having sufficient capital). In Georgia and Azerbaijan, for example, although IDPs can purchase land locally, many of them believe that doing so means giving up their IDP status and relinquishing any government benefits meant for their group. In addition, a common perception among IDPs is that open purchase of property may nullify their rights to reclaim property in their home areas. Although it is unlikely that there exists any basis in law for this perception, anecdotal reports suggest that such a view may lead to reluctance to purchase land or secret purchase through proxies (rela- tives, friends). Hence, although we have information from various sur- veys regarding DP land ownership, that information may understate the true degree of land ownership among the displaced. 44 living in limbo Box 3.2 Access to Assets in Areas of Origin Not all displaced populations are completely cut off from their home areas during the period of displacement. Chechen IDPs in Ingushetia and some Georgian IDPs from the Gali District in Abkhazia retain sea- sonal access to home areas and practice cultivation and animal hus- bandry. To varying degrees, such access helps support their survival strategies. At least a third of Gali IDPs have gone home at least once since their displacement.19 There are no reliable data available indicat- ing how important this access is in supplementing household incomes or consumption. Such access often has high transaction costs and may be unpredictable due to security risks. DPs are not landless, per se. Large percentages of currently dis- placed populations do own or have rights to productive agricultural land. However, this land lies in their areas of origin, and they are, with a few exceptions (see box 3.2), denied access to it during displacement. Some countries have, to varying degrees, attempted to temporarily al- locate agricultural land for displaced populations, and DPs themselves have purportedly made efforts to acquire usufruct rights over various sorts of agricultural land. Available data do not suggest that these efforts have been highly successful. Studies comparing access to agri- cultural land among displaced and local populations were found for only three countries: Armenia, Azerbaijan, and Georgia. Some indirect evidence was found for Bosnia and Herzegovina. Given the logic of displacement in denying access to fixed assets and the lack of extensive programs for land distribution in the situations reviewed here, these findings appear likely to apply in other situations as well. In all three Caucasus countries, the displaced, on average, had far less access to land and livestock than local populations. Obstacles to using land can be both legal and practical. In Georgia, IDPs did not have a right to land use from 1993 to 1996; the law then changed to allow IDPs limited and temporary use of land for cultiva- tion.20 Despite the change, an IFRC survey in Georgia reported in 2000 that only 17 percent of IDPs in collective centers and 34 percent in private accommodations had access to land.21 By contrast, about 62 percent of the local population had access. Moreover, almost all locals (99 percent) owned their land partially or entirely, as opposed to only 15­49 percent of the IDPs. Locals also had access to larger plots of land than did IDPs (2,500 versus 400­1,200 square meters). In Azerbaijan, a 1996 ECHO assessment reported that 9­17 per- cent of IDPs owned land, while 42 percent of the general population employment 45 did.22 Data from the World Bank's 2002 poverty assessment show that, outside Baku, only 22 percent of IDP households had access to land as opposed to 49 percent of local households.23 Moreover, when IDPs owned land, they owned smaller plots, with poorer soil, and they cultivated a smaller share of these plots. Consequently, IDPs con- sumed less home-produced foods than the locals. In addition, IDPs tended to have less secure tenure. The study of vulnerable groups in Armenia24 found that 46 percent of locals owned land, while only 33 percent of the refugees in private accommodations did.25 A recent review of microfinance in Bosnia and Herzegovina that had clients among both resident and displaced populations reported less than 10 percent of loans to the displaced going for agricultural purposes. Nearly half of loans to residents did.26 One study in Bosnia and Herzegovina on preconflict and postconflict employment reported that although the percentage of local households deriving income from agriculture stayed the same (18 percent), the percentage of displaced households fell by more than half, from 22 percent to 10 percent, after the war.27 The unavailability of machinery and animals further impedes the displaced's agricultural production. Usually, displacement occurs quickly, giving DPs little opportunity to take their household goods, equipment, tools, and animals. People often flee with only the prover- bial "shirt on their backs." (See box 3.3 for an exception.) In the first months and years of displacement, what assets have been brought along often have to be sold, frequently at deflated prices, to support Box 3.3 Moveable Assets Most DPs tend to lose their productive assets through the act of dis- placement. The extent of this loss depends upon both the speed and violence of displacement and the portability of assets. Most of the IDPs in Azerbaijan had predisplacement employment linked to land or indus- trial facilities that now lie beyond their reach. A minority of the IDPs (approximately 10,000 to 20,000 Azeri and Kurdish herders) tradition- ally relied on animal herding as a key element of their livelihood. They were able to bring substantial numbers of their herds into displacement and settled on winter pasturelands in areas outside the conflict zone. These were lands over which they had traditional usufruct rights prior to the conflict. In this case, the impact of displacement on access to eco- nomic assets was mitigated by the nature of key assets: herds of animals that could move with them and preexisting usufruct rights on pasture land outside the conflict zone. 46 living in limbo the household. While cannibalization of assets through the sale of household goods also proved a necessity during the same period for many local households, because of economic transition, displaced households tended to start with less. Thus in Azerbaijan, the most recent poverty assessment found 67 percent of residents owning cows; the figure for IDPs was less than half that (27 percent). In Georgia, res- idents were found to be three times as likely as IDPs to own cows. Overall, the IFRC study of IDPs in Georgia reported that 52 percent of the local population had livestock as opposed to 13­29 percent of IDPs. Locals also sold or exchanged produce from their land and ani- mals more often than IDPs (28 percent versus 15­22). Likewise, a study of the Zugdidi region reported that the local population had two times the amount of livestock as the displaced population.28 In Armenia, similarly, 21 percent of locals owned livestock as opposed to only 11 percent of refugees in private accommodations.29 Finally, it should be noted that even when DPs can find employment in agriculture working for others, the value of their labor depends not only on the area's available labor market but also on the level of their skills. While agricultural skills appear to be mobile and transferable to other situations, sometimes the differences in agricultural production in terms of crops and necessary skills in areas of origin differ signifi- cantly from conditions in areas of asylum. The displaced frequently find themselves in different ecosystems than those from which they originate. To take one example, the majority of Azeri IDPs come from a mountainous zone in, and surrounding, Nagorno-Karabakh. In this environment, raising crops such as fruit, particularly grapes, and tobacco, as well as animal husbandry, were important occupations as was employment in related processing industries, such as wine pro- duction and hide tanning. These professions required specific, detailed skills for which a premium was paid. While some IDPs were fortunate enough to settle in immediately neighboring areas of Azerbaijan, with similar ecosystems and employment opportunities, many others set- tled in cotton and wheat production areas. Although employment, to a great extent seasonal and ad hoc, is available in these areas, the eco- nomic environment there does not place a high value on the types of skills developed in home areas and thus reduces the type of wages DPs can command. The question of the importance of access to agricultural land for displaced populations is a complicated one. It is clear from the evidence that while the majority of the region's DPs originated in rural environ- ments, most do not currently have access to land either to cultivate as a source of cash income or to supplement their survival strategies through household consumption. Where such access is available, it employment 47 tends to be unstable due to limited tenure. Because of quality issues of available land and limitations on key inputs, the land probably does not maximize potential benefits either. However, it is also true that the majority of DPs in many situations, regardless of their original habitat, are settled in urban environments, in large towns and cities. To some extent the gravitation of many displaced groups to urban environments may be the direct result of the lack of availability of productive land and the consequent absence of employment opportunities. Even should land be available, to take advantage of any such opportunities DPs might be required to shift from their current locations. Another factor needs to be considered. While the majority of the displaced originated in rural areas, the economic development strate- gies of those areas under the former Yugoslavia and Soviet Union gov- ernments dictated a wide range of industrial enterprises and other nonagricultural activities. When the previous occupations of DPs are reviewed, only a minority, roughly 20 percent regionally, specify them as agricultural. (The Kurdish IDPs in Turkey represent an exception to this statistic: over 80 percent reported farming as a primary occupa- tion prior to displacement.)30 This does not mean such populations could not participate in agriculture nor does it mean they did not have some of the requisite skills. Such skills could have come from their ancestral rural background as well as the fact that even those with nonagricultural occupations probably conducted some level of culti- vation to supplement household consumption. However, it does offer a caveat on overly simplistic assumptions regarding the facility with which such populations could adapt to agricultural production with- out some degree of skill training and transaction costs. Additionally, the longer displacement lasts, the more de-skilling takes place (not only in agriculture) among DPs. In Serbia, the average age of refugees who reported being farmers prior to displacement was over 60.31 The high percentage of DPs currently in urban environments likely will accelerate the process of de-skilling, with implications for agricultur- ally oriented self-reliance approaches now as well as for potential bases of economic production in the eventuality of returns home.32 Unemployment Even when the overall rates of unemployment among local populations in affected countries are high, existing data show still higher rates among DPs, sometimes double those of local populations. Consequently, as dis- cussed in the previous chapter, the displaced depend heavily upon public transfers and humanitarian assistance. 48 living in limbo Displacement physically removes populations from their places of employment, whether these be farms, factories, or shops, and creates, almost by definition, automatic unemployment for tens of thousands or even hundreds of thousands. Thus unemployment in the early days of displacement should not come as a great shock. Similarly, in other displacement situations--for example, when the government sequesters refugees in large camps--unemployment again is an understandable consequence. However, as noted, the majority of the region's DPs have been displaced for long periods of time, living in close proximity to the resident populations, not segregated. The extent to which DPs can find new employment in displacement forms a core issue in assessing their vulnerability. This section compares the situation of DP unemployment rates with those of local populations. As noted earlier, the study found fairly con- sistent data across all displacement situations indicating much higher rates of unemployment among the displaced. This section further explores some of the possible reasons for higher DP unemployment. Data comparing the employment situations of displaced and local populations were found for five countries (Armenia, Azerbaijan, Bosnia and Herzegovina, Georgia, and Serbia and Montenegro). In all of them, the displaced proved far more likely to be unemployed and less likely to have stable employment. There is also some indication that the displaced tend to be more involved in trading and less in agri- culture (see figure 3.1). In the Caucasus, surveys of IDPs in Georgia and Azerbaijan, as well as of refugees in Armenia, report significantly higher unemployment among the displaced than among the local population. In Georgia, an IFRC survey found unemployment among IDPs to be 31­45 percent as opposed to 15 percent for locals.33 The most recent poverty assess- ment for Azerbaijan reported that IDPs were twice as likely to be un- employed as locals, with a 47 percent unemployment rate.34 The study of vulnerable groups in Armenia found unemployment among refugees to be 42 percent as opposed to 28 percent among locals.35 The displaced in the Balkans fare no better. Both IDPs and refugees in Serbia have higher rates of unemployment than the local popula- tions. The recent poverty assessment found unemployment among IDPs and refugees to be 11 percent and 19 percent, respectively; among the local population it was only 8 percent.36 The lower rate of unemployment among IDPs, compared with the rate for refugees, probably stems from the fact, as discussed earlier, that the government of Serbia has maintained some civil servants from Kosovo. A 1999 World Bank assessment of Bosnia and Herzegovina reported unemployment among the displaced to be 66 percent, as opposed to employment 49 Figure 3.1 Unemployment in Selected Countries Armenia Azerbaijan Bosnia Georgia IDPs (collective) IDPs (private) Local Refugees Serbia IDPs 0 10 20 30 40 50 Unemployed (percent) Sources: Armenia: United Nations Office in Armenia, Poverty of Vulnerable Groups in Armenia: Comparative Analysis of Refugees and Local Population. United Nations Department of Public Information (Yerevan, 1999). Azerbaijan: World Bank data (2002). Bosnia: World Bank, Bosnia and Herzegovina: Local Level Institutions and Social Capital Study (Washington, D.C.: World Bank, 2002). Georgia: IFRC, Internally Displaced Persons: A Socio-Economic Survey, Georgia. IFRC (Tbilisi, 2000). Serbia: UNHCR and the Commissioner for Refugees of the Republic of Serbia, Registration of Internally Displaced Persons from Kosovo and Metohija. Government of Serbia (Belgrade, 2001). UNHCR and the Commissioner for Refugees of the Republic of Serbia, Refugee Registration in Serbia. Government of Serbia (Belgrade, 2001). 45 percent for the general population.37 The situation there has im- proved, as a 2002 study of local institutions and social capital shows, but the difference persists. This study still found unemployment in Bosnia and Herzegovina to be higher among DPs than among locals (38 percent versus 25 percent).38 Obstacles to Employment Among the many studies and surveys, one consistent finding across all country situations is the high rate of unemployment of displaced populations. Given that the majority of DPs have been receiving 50 living in limbo various forms of public assistance that are, in most cases, declining, in- elastic, or both, an observer can reasonably conclude that employment is a pivotal element for DP material well-being. In other displacement situations around the world, the reasons for unemployment among DPs often are clearer cut. Frequently, a government cordons refugees off in tented camps and formally blocks them from employment or intention- ally settles them in places distant from major population centers. IDPs in many African countries and in Afghanistan are caught in the middle of war zones. Certainly, a key factor in chronic unemployment of the displaced is the weak economic environment in areas of asylum. But why is unemployment among the displaced in the Balkans and Caucasus so much higher than that of the host populations? Even when displaced populations have the requisite skills to com- pete in employment markets in places of asylum, a number of factors combine to limit their access to jobs. There has been no focused study that would allow a detailed understanding of the weight of these fac- tors, although most of them have come up in qualitative research and in interviews with field staff of various organizations as well as DPs themselves. Some of the factors discussed below are environmental or related to intransigent political factors and may not be subject to op- erational approaches. Others may be more amenable to intervention. As noted at the outset, high unemployment and economic collapse in affected countries create fewer opportunities for employment for all populations, and this undoubtedly represents one of the major imped- iments to DP employment in the region. Yet this does not fully explain the magnitude of differences in unemployment among DPs and resi- dent populations.39 The purpose of this section is to identify and examine some of the various factors contributing to DP unemploy- ment that might be addressed through dedicated action. Each of the following nine factors represents, to a certain degree, a potential obstacle to employment for the displaced: · Subnational settlement environments of DP concentrations · Frequent legal inability to own land · The impact of socioeconomic dislocation · Loss or lack of acceptance of legal documents and professional credentials · Linguistic and cultural differences · Limited ability to access informal patron-client networks or other backdoor channels to employment · Diminished access to formal credit · Impacts of "living in limbo" · Dependency syndrome and moral hazard from government assistance employment 51 Subnational Settlement Environments While the economic situation at the national level of host countries is probably the predominant influence on unemployment, DPs tend to be concentrated in what amount to a country's economically weakest areas. There job opportunities are, not surprisingly, less likely. Deter- mining why this is so is difficult, but the pattern appears frequently enough to bear notice. In the case of refugees, it is possible to suggest that border areas (the areas refugees cross) are historically among the poorest, but this is just a supposition, without any real evidence to sup- port it. However, even among IDPs, the pattern holds. Serbs from Kosovo have settled in the most depressed areas of central and south- ern Serbia. Azeri IDPs outside Baku have settled in raions that rank among the highest in poverty in the country. Similar factors hold for the areas of western Georgia, where IDPs from Abkhazia have concen- trated. The Serb refugees from Croatia and Bosnia and Herzegovina are an exception, having settled almost exclusively in the richest province in Serbia (Vojvodina) and in the areas around Belgrade. The subnational concentration of the displaced has another impact. The inflow of DPs into an area increases the population, sometimes by 30 to 50 percent, putting pressure on the employment market there. Such an increase can be beyond the carrying capacity of a specific region in terms of jobs. At the level of individual settlements, a major aspect of DP unem- ployment is very likely to be the economic environment of the imme- diate area. While, as noted in a previous section, many DPs are integrated or situated close to resident populations, others are settled in areas distant from population centers. This is partly because such areas are more likely to have land available for settlements and partly, in many countries, because of a conscious government policy to keep DPs at arm's length from other segments of the population. At the same level, the relative size of settlements can make a differ- ence in employment opportunities. This factor can be seen in clearest contrast in other areas of the world where refugee camps sometimes comprise tens or even hundreds of thousands of people. The advan- tage of larger DP settlements is that public assistance can be delivered in a more efficient manner through economies of scale. To a small de- gree, larger settlements also may offer some degree of employment within the DP population, generated by the need to establish services such as health, education, and camp administration. This sometimes comes close to replicating the institutional structures of a small town or city administration.40 Yet logically the larger the settlement created, the less likely it will be that any appreciable percentage of the dis- placed population can be readily absorbed into the surrounding area's 52 living in limbo workforce. It is also less likely that agricultural land can be made available in sufficient quantities to provide employment, cash, and produce for consumption to the displaced population. By contrast, smaller concentrations of DPs, while less accessible to any group- targeted assistance programs, represent a lesser burden on surround- ing populations and may have a greater chance of absorption into a local workforce.41 Legal Inability to Own Land As noted above, in some affected countries, there exist various oppor- tunities for DPs to rent land or temporarily use it for production of agricultural commodities. However, the inability of most DPs to legally own land undermines their ability to maximize land-based pro- duction. The majority of governments do not allow foreigners, which includes refugees, to own land. The situation with IDPs is more complex. (See, for example, box 3.4.) While in most cases, they can purchase land, to do so would, de jure or de facto, require them to relinquish their status as displaced persons, which might hamper any eventual claim to disputed land or other assets. Azerbaijan and Georgia offer some temporary land usage. Yet frequently such land is less productive public land left over from the privatization process. Box 3.4 Slowly Moving toward Self-Reliance in Azerbaijan In 2001­02, the government of Azerbaijan announced a program to allocate agricultural land with title to IDP families settled in some of the worst conditions. The program also included the provision of inputs for community infrastructure and the construction of houses. The program has to date provided land and housing to more than 20,000 IDPs, with plans for approximately another 20,000 individuals in 2004. The num- bers represent less than 10 percent of total IDPs still displaced, but it is a beginning and represents an innovative strategy. While the program does not preclude a return home for IDPs, it recognizes that 10 years of displacement have expanded the IDP population and the number of nuclear families. Plots were allocated to extended families under the rationale that, in the event of possible return home, sections of families will remain in situ on the allocated land and other sections will return to the original property left behind in 1992­93. employment 53 In the recent Azerbaijan poverty assessment, very few IDPs reported significant income from agriculture. Reports indicate that in Georgia and in Azerbaijan, land often lies too far from IDP settlements for exploitation to be maximized, and there is little access to irrigation or other inputs. It is impossible through available sources to accurately identify how many DPs actually own land through proxies or have access to agri- cultural land but decline to report it in surveys. This is not only due to typical reluctance of households to report income. As noted, DPs sometimes express fear that land ownership (whether agricultural or residential) would signify integration and would result in a loss of DP benefits or even the right to reclamation of property in the area of origin. This fear may not be valid. However, the risk that DPs perceive they incur is as significant as the psychological tension of making such a decision. Socioeconomic Dislocation The settlement of DPs in socioeconomic environments different from their areas of origin is another factor influencing the possibility of em- ployment. Most of the displaced have rural origins, but this does not necessarily imply an agricultural background. In some predisplace- ment economies, particularly in the former Soviet Union, factories and other industrial facilities were frequently situated in rural areas. How- ever, in many countries, DPs have settled in urban environments, due partly to original settlement patterns and partly to the attraction of urban services and perceived potential for employment. Several non- income-based factors, such as the potential for anonymity promised by an urban environment, also play a role. In Azerbaijan, unemploy- ment has been exacerbated by the fact that many IDPs found them- selves in cities where rural skills such as farming and carpet weaving proved of little value.42 Similar patterns can be seen in other situations of displacement, including the rural Kurds' settlement in Turkey's urban centers. Even when DPs of rural origin settle in rural areas, the predominant modes of economic production may not be the same as those that prevailed in their areas of origin. Legal Documents and Professional Credentials In countries outside Europe, where large percentages of the population have relatively low levels of education and professional training, the question of credentials is not likely to represent a significant impedi- ment to employment for the majority of DPs. Given the structure of the economy in this region's displacement-affected countries, however, 54 living in limbo recognition of professional and educational credentials does become an issue for certain segments of the DP population. Even simple docu- mentation such as birth records and secondary school certificates can complicate employment and educational advancement. Quite often these types of documents are lost or destroyed in the conflict or during the displacement itself. Original copies of such certifications may be destroyed as well since during a conflict official archives and records frequently become targets or are collateral damage. Documents may become inaccessible due to the battle lines drawn during the conflict. Sometimes, particularly for refugees, professional standards may be different; specific documents or degrees, even if proof of them exists, may not be accepted in the place of asylum. This may prove particu- larly true if academic instruction in the area of origin was in a different language or emerged from a different professional and administrative tradition. A related difficulty for DPs emerges from the policy of some states, particularly successor states to the Soviet Union, of requiring registra- tion of individuals residing in each subnational administrative unit (for example, district, municipality, city). The propiska system (a sys- tem of Soviet residency permits for specific localities) officially has been abolished throughout the former republics of the Soviet Union. However, in practice, it still carries weight in many instances. DPs who are not formally residents of a particular community and thus do not have such documentation may find this a complication when applying for public employment or acquiring approvals and licenses to engage in specific types of private business. The lack of a residence permit may also complicate DP attempts to move from one place to another in search of work. Linguistic and Cultural Differences While the majority of the region's displaced have been settled in areas where the dominant culture and language are similar to their own, this is not always the case. The impact of linguistic and cultural differences between DPs and host populations is a multifaceted issue in employ- ment. Some impacts may be direct, such as those that occur when DPs are not literate or able to communicate adequately in the area's pre- dominant language. Some Roma DPs from Kosovo, for example, speak Albanian but not Serbian.43 The effects can also be indirect, as when differences set the displaced population apart, making individu- als easily identifiable and thus potentially susceptible to informal or formal patterns of discrimination. The older generation of ethnic Armenian refugees from Azerbaijan often does not speak Armenian. employment 55 Although they speak Russian and thus can communicate with a popu- lation that speaks Russian as a second language, this fact sets them apart and can result in difficulties in the job market. While Georgian IDPs from Abkhazia speak Georgian (with the exception of those from the Abkhaz capital, Sukhumi), they speak the Mingrelian or Svenati dialects, making them readily identifiable in Tbilisi. Kurdish IDPs in Turkish cities, too, can be readily identified; according to recent survey data, significant numbers of women as well as older generations are not fully proficient in Turkish. While in many of these cases (Georgia, Turkey) previous waves of rural immigrants had moved to urban cen- ters and other areas of the affected countries for economic reasons, the massive flow of DPs often qualitatively altered the situation to the extent that even these economic migrants are assumed, a priori, to be DPs in their interaction with host populations. Such differences lead to many displaced populations being set apart from the host population. In extreme cases, when DPs are not fluent in the working language of the host area, this represents a high barrier to local employment. In other cases, it creates a potential for prejudicial treatment of DPs. Patron-Client Networks and Informal Channels The multifaceted implications of this factor will be discussed in greater detail in later chapters, which deal with social capital and relations between the displaced and the state. Here we will simply note that employment in the region's societies, even in public service occupations, often is understood to be related as much to access to informal networks as to formal job procedures. The dislocation of DPs from social networks, including links to local government administration, can act as a factor undermining the ability to be employed. The laws that frequently prohibit refugees and many IDPs from voting in local government elections in areas of asylum distance them further from municipal and other local level governments, which do not recognize them as constituents.44 Diminished Access to Formal Credit One aspect of DP employment for which little detailed knowledge can be found is self-employment through small business generation. Displaced populations have less access to informal forms of credit due to their dislocation from social networks. In all likelihood the members of their social networks are either dispersed widely or, where concentrated, face similar deficit sit- uations, without excess capital to spare. The existence of DP-focused microfinance programs in several countries highlights the fact that entrepreneurial energy among DPs represents a potentially important resource. However, such programs touch relatively small numbers of the displaced. One constraint on DP business investments is the lack of 56 living in limbo access to formal, broader-based sources of capital. This results from prior cannibalization of household assets and lack of access to fixed assets, such as land and structures during displacement, that could have been used for collateral. However, even for settled populations, the availability of formal credit throughout the postsocialist states is not extensive. In some cases, such as Azerbaijan and Georgia, human- itarian agencies have been able to provide microfinance aimed at IDPs that is not available to members of the settled population--but again with a relatively limited pool of beneficiaries. Impacts of "Living in Limbo" The uncertainty of the future, com- plicated by the ups and downs of peace accords (and possibilities of a return home) may affect perceived discount rates for long-term invest- ments by IDPs. They may also affect employment choices. The dis- placed must constantly evaluate whether investments with long-term gestation periods make sense. Sometimes "long term" signifies periods of as little as a year. Even when DPs have access to capital, they must constantly evaluate and reevaluate whether investments in a small business or the purchase of agricultural land will turn a profit before their circumstances change. This can also play a role in longer term human capital investments, such as education and selection of a tech- nical skill, or even in terms of teaching children one language versus another. All of this requires some vision of a future job market, which could be in the area of asylum or could be back in the area of origin. When the economic environments of such areas differ greatly (or require, for example, different languages) this creates a tension in such investment decisions that is a unique characteristic of displacement. Dependency Syndrome and Moral Hazard In closing this chapter, one other element of displacement situations that may have a major effect on finding employment bears mentioning. When tensions arise with host populations, DPs often asserted, in survey responses, that this factor was a cause of unemployment. This might be described as a kind of psychological malaise, the so-called dependency syndrome, as- sociated with dependence upon government handouts. Combined with the effect of public transfers on household economies, it may create disincentives (that is, moral hazard) for IDPs in finding employment. In postsocialist societies, particularly successor states to the Soviet Union, the expectation of government support frequently has been integrated into the nostalgic mindset of even members of the resident population. When the impact of several years of relief assistance is factored into the situation of displaced populations, it is not surprising that this mindset may affect individual decisions to seek employment. However, before emphasizing the impact of such dependency syndrome too far, it is employment 57 important to recall the statistics cited earlier on degrees of public assis- tance to DP families across the region's affected countries. When DP- targeted assistance alone is considered, it almost never represents, on average, a majority of household expenditure nor an amount that by itself could guarantee survival, even at the most minimal living stan- dard. The case could be made, as noted earlier, that assistance acts as a subsidy enabling the displaced to seek employment and accept lower wages, making them more competitive in some segments of local em- ployment markets. There is anecdotal evidence for such an observation but little in the way of reliable information. Conclusions From the foregoing discussion, several conclusions can be made. First, DPs seem to have less access to formal sector employment than local populations. Even when they can secure such employment, they tend to receive lower wages. Hence, it appears necessary for them to rely on informal activities, such as petty trading and day labor. Second, in a few countries (Azerbaijan, Georgia, and Serbia), some DPs can keep their public sector employment during their displace- ment as part of governments or administrations in-exile. Third, the displaced tend to have less access to agricultural land and livestock than locals, in great part because most DPs were not included in the various privatization processes that have occurred across the region. Hence, they appear to cultivate less and have less produce for their own consumption and sale. Fourth, after more than a decade of displacement, DPs seem more likely to be unemployed than the local population, even in a region where most of them are relatively educated and living in close prox- imity to the resident population (that is, not secluded in camps). Finally, some possible reasons for high unemployment among DPs include: · Settlement in less economically vibrant areas · Limited access to land · Rural dwellers settled in urban areas · Inappropriate or missing educational or employment documen- tation · Linguistic or cultural differences · Limited access to social networks · Limited access to credit · The impacts of "living in limbo," that is, inability to commit to a new life or return to a previous one 58 living in limbo · Dependency syndrome, meaning the psychological malaise and inertia associated with receiving government and humanitarian assistance Notes 1. Public transfers provided on a per person basis may also, under some cir- cumstances, decrease in overall relative value with time as households increase in size through births. In some countries, new registration of DPs can engender transaction costs or otherwise represent a complicated bureaucratic process that results in delays or inability to register new members. By the same token, some DP households may benefit by declin- ing to register deaths of household members. 2. Azerbaijan represents a unique case in this regard. While it is a poor coun- try, the beginnings of an oil economy have provided limited but liquid resources to the state. This, combined with the high political priority placed by the government on support of IDPs from Nagorno-Karabakh, has ensured that as humanitarian aid has declined, government expendi- tures have concomitantly increased to fill the gap. Most other countries in the region do not have the available resources to play a similar role or do not have a similar degree of political will to support the DPs for whom they play host. 3. For example, a 2000 study conducted in Turkey by the Beyoglu Centre for Youth and Children in 23 districts of Istanbul found that "99 percent of the children selling handkerchiefs and chewing gum or polishing shoes in the streets are from displaced families." 4. ICRC, Survey of IDP Households in FRY. 5. As indicated in the last chapter, a large number of IDPs who used to work for the Serbian government in Kosovo continue to receive 60­80 percent of their salaries. 6. CIET International, Unemployment, Employment, and Wages in Bosnia- Herzegovina: 1997 Vulnerability Survey. CIET (New York, 1997). 7. World Bank, Bosnia and Herzegovina: Local Level Institutions and Social Capital Study (Washington, D.C.: World Bank, 2002), vol. 1. 8. Quoted in Dershem, Gurgenidze, and Holtzman, Poverty and Vulnerability. 9. IFRC, Internally Displaced Persons. 10. Shakarishvili, Evaluation of Socio-Behavioral Patterns. 11. World Bank, Azerbaijan Poverty Assessment, vol. 1. 12. ARRA (Agency for Rehabilitation and Reconstruction of Areas), Feasibil- ity Study for the Resettlement and Reconstruction Program of the Liber- ated Areas. Unpublished report, ARRA (Baku, 1998). 13. World Bank data (2002). employment 59 14. Ibid. 15. IFRC, Internally Displaced Persons. 16. The implications of such parallel education systems for human capital and social capital will be taken up, respectively, in chapters 5 and 6. 17. Elizabeth Gomart, Social Assessment of Poverty in Croatia. Poverty Reduction and Economic Management in ECA Region Background Paper, World Bank (Washington, D.C., 2000). 18. World Bank data (2002). 19. IFRC, Internally Displaced Persons. 20. However, they were not permitted to build structures on the land or make other permanent improvements. 21. IFRC, Internally Displaced Persons. 22. de Roos and Venekamp, Food and Medical Assessment in Azerbaijan. 23. World Bank data (2002). 24. UN Office in Armenia, Poverty of Vulnerable Groups in Armenia. 25. A certain number of refugees in Armenia were given access to land vacated by ethnic Azeri refugees. However, the majority of Armenian refugees are of urban origin, with little background in agricultural employment. 26. Personal communication with World Bank task manager. 27. World Bank, Social Assessment of Bosnia-Herzegovina. Displaced in- cludes refugees, IDPs, and returnees. 28. Shakarishvili, Evaluation of Socio-Behavioral Patterns. 29. UN Office in Armenia, Poverty of Vulnerable Groups in Armenia. In Armenia, refugees were included when land and livestock were privatized. 30. Göc-Der (Immigrants' Association for Social Cooperation and Culture), The Research and Solution Report on the Socio-Economic and Socio- Cultural Conditions of the Kurdish Citizens Living in the Turkish Republic Who Are Forcibly Displaced Due to Armed-Conflict and Tension Politics. Göc-Der (Istanbul, 2002). 31. UNHCR and the Commissioner for Refugees of the Republic of Serbia, Refugee Registration in Serbia. Government of Serbia (Belgrade, 2001). 32. In Azerbaijan, nine years after displacement, 41 percent of the IDPs in Baku listed agricultural professions as their previous occupation. The true numbers are probably much higher as people in other professions were rurally oriented and 26 percent reported no profession. A large number of these may be rural workers or matured children of agricultural families. While "agricultural occupation," as noted previously, did not necessarily mean direct cultivation, it does suggest a skill mix and cultural and social mores better adapted for a rural setting than for a large city such as Baku. SIGMA, Azerbaijan: Study of IDPs and Refugees. SIGMA (Baku, 2002). 33. IFRC, Internally Displaced Persons. 34. World Bank data (2002). 35. UN Office in Armenia, Poverty of Vulnerable Groups in Armenia. 60 living in limbo 36. World Bank data (2003). 37. World Bank, Social Assessment of Bosnia-Herzegovina. Displaced includes refugees, IDPs, and returnees. 38. World Bank, Bosnia and Herzegovina: Local Level Institutions, vol. 1. 39. It is also possible that the heavy reliance of DPs on gray market occupa- tions is not being reported in many surveys as "employment." Or DPs themselves, regardless of the phrasing of survey questions, may have chosen not to report employment of any kind at a higher rate than resi- dent populations did. 40. It is interesting to note that the types of skills that represent credentials for employment in such situations often were of lesser value in predisplace- ment home areas. The most notable, when significant international humanitarian assistance is present in the form of NGOs, UN agencies, and other Western agencies, is knowledge of western European languages (or sometimes Arabic). This is due to the premium that such agencies tend to put on hiring individuals who can communicate with their foreign staffs. This represents a relatively small, but not negligible, amount of jobs for educated classes of DPs. 41. However, note that one of the few identifiable subgroup differentiations in poverty among IDPs in the recent Azerbaijan poverty assessment was that those IDPs living in individual clusters, that is, in apartments and pri- vate houses, were poorer than other IDPs. 42. World Bank, Azerbaijan Poverty Assessment, vol. 2. 43. UNOCHA, Serbia Situation Report (April 2002). 44. Simon Bagshaw, Internally Displaced Populations and Political Participa- tion: The OSCE Region. Brookings Institution Occasional Paper, Brookings Institution (Washington, D.C., 2000). chapter 4 Settlement Patterns and Shelter Situations Summary: This chapter focuses on the settlement patterns and shelter situations of displaced populations. It begins by describing the predom- inant DP settlement patterns, why they develop, and what their implica- tions are for the vulnerability of DPs. Then it describes the various types of DP shelters, collective and private. The chapter concludes with a review of the quality and sustainability of each type of accommodation. A REVIEW OF THE SHELTER SITUATION of displaced populations repre- sents a significant part of the assessment of their vulnerability, due to both its direct impact on their quality of life and its indirect impact on other aspects of vulnerability. Those aspects include opportunities for income and the environment for the maintenance and formation of social capital. At the beginning of this study, we noted the importance of recog- nizing the multifaceted quality of poverty and vulnerability. In chap- ter 2, we reviewed available evidence regarding the material well-being of DPs across the region. The conclusions of that chapter were that, on average, DPs emerge as poorer than local populations in terms of con- sumption measurements. The poverty of the displaced and, even more important, their vulnerability extend beyond income. Prior to dis- placement, most of the displaced owned their homes or had long-term rights to their housing. While some DPs have found stable, adequate solutions to their shelter problems, the majority remain, after long periods of displacement, caught in a range of housing situations that are unstable, substandard, or both. This chapter deals with two related issues: the spatial patterns of DP settlement, and the variety and quality of DP shelter situations. We deal with these issues together because the type of shelter the displaced end up in is very much tied to the dynamics of spatial settlement patterns and how these occur in the process of displacement. This 61 62 living in limbo chapter will first examine the spatial patterns of settlement, the rea- sons why certain patterns develop, and their implications for DP vul- nerability.1 Then we will look at the types of DP shelters, specifically collective centers and forms of private accommodations. The review of shelter situations examines related issues such as access to water, san- itation, and heating as quality of life issues and as one dimension of poverty. Initial Shelter Situations In most of the situations reviewed in this region, the original displace- ment occurred with little warning, and shelter situations were devel- oped on an emergency basis. Anticipating the onset of conflict, some individual households in the region did arrange for shelter elsewhere in advance--notably a number of Kosovar Serbs and Azeri and Armenian refugees. They proved to be the exceptions. Most popula- tions were displaced rapidly, and their arrival in settlement places was accompanied by the urgent need for a rapid solution to their shelter problem, which was envisaged, in the early days of displacement, to be only temporary, until they could go home again. The solutions typi- cally found, either in collective centers or in the homes of private host families, were usually meant to be short term, perhaps lasting a few months. Instead they have become the shelter situation of DPs for up to 10 years or more. Within this initial emergency context, the settlement of DPs followed fairly distinct patterns, the result of the conjunction of four factors: · Location of available shelter · Demands among DPs, that is, group and household decisions on where to live · DPs' low income and lack of appreciable capital to pay for housing · Host government (and donor) policies toward the displaced The accommodations of DPs differ significantly from those of local populations in that, as noted above, the displaced often live in shared accommodations with local hosts or in collective centers. They also pay rent much more frequently. While rarely owning homes in their current places of occupation, most DPs do own residential property and structures in their places of origin but do not have access to it. Where the displaced live and how they live are important for a discus- sion of poverty and vulnerability for four key reasons. settlement patterns and shelter situations 63 First, the type of shelter situations in which DPs must live has direct implications for their overall quality of life. Shelter represents an im- portant aspect of understanding the full scope of displacement- induced vulnerability in both physical and psychological terms. For the most part, DPs live in lower quality housing than the surrounding populations and have less space. Second, in many cases, DP settlement patterns have frozen households in locations distant from employment opportunities. Or the locations have put DPs in socioeconomic envi- ronments that differ from their areas of origin, thus weakening their ability to achieve self-reliance from employment. Third, DP shelter sit- uations are potentially fragile and can deteriorate without much warn- ing. Up to a third or more of DPs still live as "guests" in households of family members or friends who took them in for a "temporary" period--which has dragged on for years. In some countries, half or more of the DP population has settled en masse in various collective centers, large buildings and other structures drafted into service for human habitation. Furthermore, due to lack of maintenance, the structures are gradually deteriorating. Fourth, subnational concentra- tions of DPs exacerbate the impact of displacement on host popula- tions. The DP shelter situation often contributes to social exclusion and sometimes active tension with the host population. Yet at the same time, the shelter situation frequently weakens the opportunities for enhancing social cohesion within DP groups themselves. Settlement Patterns Almost always, the displaced are unevenly distributed across the national territory of a host society. In most displacement situations in this region, and in most similar situations in other regions, DPs tend to be clustered in two types of locations: an area adjacent to the conflict zone (their home area), and in and around major urban centers.2 This pattern characterizes virtually all displacement situations globally. In Europe and Central Asia, due to the nature of housing in collective centers and in individual accommodations, most settlement situations share another characteristic. Although DPs may be clustered in indi- vidual areas within a country, within the area itself they tend to be deconcentrated, that is, not living side by side with other members of their home communities. This is a result of the supply-driven nature of shelter. 64 living in limbo Clustering Near Conflict Zones and Urban Centers Clustering near a conflict zone results from several factors, including (a) the initial perception that displacement will be temporary and the consequent desire of DPs to be as close as possible to home, (b) the re- ality that this is the first geographic area DPs arrive in, and (c) a policy decision by the host government to keep DPs there either to quaran- tine them from the rest of the host population or to facilitate rapid return home in case of a cessation of the conflict. It is also often the case that the area immediately adjoining the conflict zone is, for his- torical reasons, settled by members of an ethnic or linguistic group similar to the DPs. There may be family connections or, at a minimum, a sense of comfort and a facility of living because the DPs and the locals share a common language and culture and often a sense of empathy. Thus there is hospitality from the resident population. This was the case, for example, with the Kosovar Albanian refugees who were settled in areas of northern Albania and Macedonia, where they shared ethnicity and language with the local populations. As previ- ously indicated, the vast majority of DPs in the region have settled in proximity to clusters of people sharing with them a common language, ethnicity, or both. The pattern of clustering near conflict zones as well as in urban centers is representative of situations in Azerbaijan. Over 80 percent of IDPs are settled in 15 districts in the country's western and central regions, near Nagorno-Karabakh, and in the cities of Baku, Sumgait, and Mingechivir. In Georgia, similarly, the majority of IDPs can be found in Kutaisi and Samegrelo in the west, near Abkhazia, and in the capital city of Tbilisi. The majority of Serb refugees from Croatia and Bosnia and Herzegovina cluster in northwestern Serbia, in Belgrade and Vojvodina, areas geographically close to their areas of origin. The majority of Serb IDPs from Kosovo have settled in central and south- ern Serbian municipalities closer to Kosovo, although, reportedly, as time passes more and more of these IDPs gravitate toward Belgrade in search of jobs. Kurdish IDPs in Turkey, too, can be found in the country's southeastern region and in and around major urban centers such as Istanbul and Izmir. In the latter, they typically live in slums and peri-urban settlements inhabited by Kurds who previously had migrated to these areas for economic reasons. The majority of Chechen IDPs who have left Chechnya are settled in adjoining areas of Ingushetia and, to a lesser extent, in Daghestan and Georgia. In Daghestan and Georgia, Chechens can be found in areas populated by Chechen minorities in those societies. While Chechen settlement in Ingushetia breaks the pattern of settlement settlement patterns and shelter situations 65 among coethnics, there were no other options available to them. In any case, the pattern is not completely broken as close historical and linguistic ties exist between the Ingush and the Chechens. Until 1990, Ingushetia and Chechnya formed a joint republic within the Soviet Union. The settlement patterns of Chechen IDPs also illustrate the in- fluence of host government policy on settlement patterns. The Russian government has tried to keep Chechens in Ingushetia and discouraged their movement to Moscow and other areas of the Russian Federation, at least partially to facilitate a rapid return home. The Georgian gov- ernment has not only made efforts to limit the numbers of Chechens crossing the border but also tried to ensure that they remain in the area of the Pankisi Gorge near the Georgian-Chechen border. Settled in Towns and Cities Clustering around urban centers is usually a result of the same factor that draws rural inhabitants to urban centers around the world: the perception of greater availability of jobs and social services. With the exception of a few situations (Serb IDPs in Bosnia and Herzegovina, Armenian refugees from Azerbaijan in Armenia), the majority of the displaced in every major situation in the region were, before displace- ment, predominantly rural, coming from smaller towns and villages. A high percentage of DPs are currently displaced in urban areas, envi- ronments that ill suit their skill bases and social backgrounds. While the movement of DPs to towns and cities mirrors the path of many other rural migrants in affected countries, the numbers involved and the rapidity of the movement often create a phenomenon of qual- itatively different proportions that, unless addressed, has long-term impacts for urban planning, employment markets, and social service delivery. The city of Dyarbakir, Turkey, for example, reportedly dou- bled in size in the mid-1990s, largely, although not exclusively, due to displacement from the Kurdish conflict. In the Caucasus, over 10 per- cent of the populations of Tbilisi and Baku are DPs. In Bosnia and Herzegovina, it was estimated in 1999 that in the federation, 25 per- cent of Sarajevo and 49 percent of Gorazde were IDPs. In Republic of Srpska, similarly, IDPs represented 20 percent of Banja Luca's popula- tion as well as 45 percent of Pale's.3 Observers often believe the true numbers of DPs settled in urban areas to be much higher than official host government figures indi- cate. This is because, in addition to initial clusters of DPs settled in such areas at the onset of displacement, DPs over time, either as indi- viduals or in family units, tend to shift to urban centers. Such moves occur when (as is often the case) there are insufficient employment 66 living in limbo opportunities in the original areas of settlement. This does not neces- sarily mean that employment is to be found there, only that the perception of the availability of employment results in populations' shifts. These increases in DP numbers are not always accurately reflected in official figures. Implications of Settlement Patterns The subnational clustering pattern of DP settlements has several implications for vulnerability: · Because such settlements usually are not evenly distributed within a host country, poverty-monitoring surveys without a subnational sampling strategy tend to undercount DPs. These surveys likely will not provide accurate diagnostic information for poverty reduction strategies. · Concentration of DPs, and consequently increased populations, in certain regions puts disproportionately high pressure on local social services and labor markets. Yet clustering of DPs in con- centrated geographic areas also creates potential economies of scale in distribution of humanitarian assistance and in targeting DPs for specialized programs. · When DPs represent a relatively high proportion of the popula- tion of a geographic area--whether city, town, municipality, or district--the problem of displacement cannot be seen as only a problem of a "vulnerable group." It has macroimplications for the area's overall development strategies. · When DPs are settled in areas with socioeconomic environments differing from those of their home areas, their skill bases and social customs put them at a disadvantage in employment. Such differences also contribute to their separation and alienation from the populations. Theoretically, the subregional concentrations of DPs should provide, to a certain degree, opportunities for maintenance of social networks and social cohesion among displaced communities. (See chapter 6 for a more in-depth discussion of this topic.) However, though at the macrolevel DPs tend to be concentrated in specific and identifiable geographic zones, this does not automatically mean that members of individual communities of origin have been able to retain strong links with one another. In many other displacement situations, subregional concentrations of DPs signify large refugee camps, where DPs are settled in clusters of settlement patterns and shelter situations 67 tens of thousands or more.4 Under such circumstances, it is not un- common to find whole villages of origin living side by side in one camp or in a specific camp neighborhood. In Europe and Central Asia, there were, in the early days of displacement, some significant concentra- tions, such as tented camps in Central Azerbaijan and, most notably, clusters of Kosovars in Macedonia. For the most part, however, DPs in this region, while concentrated in particular slums or areas of cities or in specific districts, are spread out geographically within these selected territories. This microlevel dispersal is due in large part to the settle- ment of most DPs in private accommodations and the carrying capac- ity of the facilities and areas that became "collective centers." A few of the collective centers, such as those in the South Caucasus, are quite large, housing several thousand DPs in what had been factory com- plexes or military bases. Collective centers also include school or clinic buildings, which house 10 families or hotels or spas with a few hun- dred. In the latter situation, rarely could whole communities settle to- gether. While they might be geographically in close proximity, within say 50 miles of one another, they are usually not in the same location. This microlevel dispersal of settlements has had implications as well for the degree to which NGOs, government agencies, and other insti- tutional actors working with DPs have been able to develop close con- tacts with DPs and consequently an understanding of the dynamics of displaced households and communities. This dispersal, for example, has often forced agencies into centralized distribution of food assis- tance and other supplies because of the logistics and cost of delivering support to every small school building or factory or every neighbor- hood in which DPs are settled. Shelter Situations As noted, DPs in this region live in two categories of shelter situations: collective centers and private accommodations. Most of them live in the latter, either renting space, living with relatives or friends, or ille- gally occupying the housing of others. However, as figure 4.1 demon- strates, this regional generalization masks important differences between individual country situations. In Serbia and Montenegro, well over 90 percent of both refugee and IDP populations can be found in private accommodations. In Bosnia and Herzegovina, collective cen- ters were more common early in the postwar period; currently only about 7,000 of the 440,000 DPs remain in collective centers.5 Simi- larly, according to available information, virtually all Kurdish DPs in Turkey reside in private accommodations. In Ingushetia, 65 percent of 68 living in limbo Figure 4.1 Accommodations in Selected Countries Private Collective Albania (refugees) Azerbaijan (IDPs) Georgia (IDPs) Ingushetia, Russia (IDPs) Serbia (IDPs) Serbia (refugees) 0 20 40 60 80 100 Private versus collective accommodations (percent) Sources: Albania: Kosta Barjaba and Gloria LaCava, Social Impacts of the Kosovar Refugee Crisis in Albania. ECSSD Background Report, World Bank (Washington, D.C., 2000). Azerbaijan: World Bank data (2002). Georgia: IFRC, Internally Displaced Persons: A Socio-Economic Survey, Georgia. IFRC (Tbilisi, 2000). Ingushetia: ICRC, Internally Displaced Persons in Ingushetia: Economic Security Survey. ICRC (Geneva, 2002). Serbia: UNHCR and the Commissioner for Refugees of the Republic of Serbia, Registration of Internally Displaced Persons from Kosovo and Metohija. Government of Serbia (Belgrade, 2001). UNHCR and the Commissioner for Refugees of the Republic of Serbia, Refugee Registration in Serbia. Government of Serbia (Belgrade, 2001). Chechen IDPs are reportedly living in private accommodations of some sort. Even during the relatively brief, but intensive, Kosovar refugee situation, 70 percent of refugees in Albania found accommo- dation with host families. Most refugees in Armenia also reside in pri- vate accommodations.6 However, approximately half of the 290,000 IDPs in Georgia and three-quarters of the nearly 600,000 IDPs in Azerbaijan live in collective centers. Overall, perhaps 800,000 of those currently displaced in the region stay in some sort of collective center. Most are in the Caucasus. Collective Centers Collective centers include a range of accommodations where (a) mul- tiple DP families live in the same structure or settlement, (b) residence settlement patterns and shelter situations 69 in the settlement has typically not been voluntary but rather assigned by government authorities, donor agencies, or both, (c) shelter is al- most always provided free of charge, and (d) settlements are usually set apart from local populations, creating varying degrees of isolation. Collective centers range in size from a few families to several thousand families. The types of structures and facilities that have been drafted into service as collective centers include the following: · Buildings not originally intended for human habitation, includ- ing abandoned factories, unfinished buildings, military bases, and public buildings such as clinics, schools, and administration buildings · Makeshift accommodation such as railway cars, abandoned transport containers, and other structures never meant to hold people · Buildings and facilities originally intended for seasonal or short- term occupation, such as hotels, dormitories, spas, and summer camps or seasonal herder accommodations such as dugouts in Azerbaijan · Camplike settings, usually on public land established explicitly for DP settlement. These may be initiated as tented camps and eventually consolidated into makeshift housing or may be estab- lished from the beginning as more weatherized housing, such as limestone-brick housing and prefabricated units. There is sometimes a sense that the emptying of collective centers and dispersal of DPs into private housing represents a solution to the shelter problem. In some cases, the displaced have been able to pur- chase housing through receipt of compensation for their property in their area of origin, income from employment, or help from family members. In other cases, governments have distributed vacant housing to DPs. However, these situations represent a minority of cases. Private Accommodations Private accommodations signifies individual family units, typically selected by DPs themselves, where they either live separately or with host families in the same dwelling. In the context of displaced popula- tions, private accommodations cover a wide range of household situa- tions (see, for example, Box 4.1). They include DPs living not only in apartments and houses, but also in half-finished apartments in larger buildings, sheds, or in abandoned dachas (summer houses). A recent survey of IDPs in Turkey reported that 29 percent, when given several options from which to choose (including "house" or "flat") defined 70 living in limbo their shelters as "slum."7 Private accommodations does not necessar- ily mean single-family occupation. Perhaps a third of the displaced in Georgia, Serbia, and Turkey remain, a decade after displacement, liv- ing in joint quarters with relatives or friends among the local popula- tions. Private accommodations may be provided for free or rented by DPs for cash. Home Ownership Even when settled in private accommodations, displaced populations rarely own their homes. Home ownership sym- bolizes stability and represents a potentially valuable economic asset. In rural areas, home ownership also tends to correspond with con- comitant ownership of at least some land, which can be used for the production of food for household consumption; thus lack of home ownership has broader, indirect implications for livelihoods. In con- trast to DPs, most of the local populations in settlement countries do Box 4.1 Differing Shelter Situations within One Country Even within the same country, the private accommodation situations of DPs can differ significantly due to various factors. Serbia and Montenegro is home to a fairly heterogeneous group of displaced peo- ple, and variation in their circumstances does appear in the data. Although the majority of DPs are ethnic Serbs, they also include refugees from Croatia and Bosnia and Herzegovina who have been dis- placed for up to a decade or more and IDPs from Kosovo who arrived just a few years ago. Furthermore, conditions for those settling in Mon- tenegro differ from conditions for those in Serbia. Data from surveys and studies reveal disparities in the groups' circumstances. For example, 15 percent of IDPs live in small spaces of 3­10 square meters, while only 7 percent of refugees do. About 77 percent of IDPs in Serbia versus 67 percent of refugees assessed their living conditions situations as "bad" or "very bad." Conditions appear to be somewhat better in Montenegro, where the corresponding numbers are 72 percent and 52 percent, respectively, for IDPs and refugees. Other differences are apparent. More DPs in Montenegro than in Serbia have their own plumbing (78 percent versus 70 percent). In terms of heating, Montene- grins are only slightly better off, 87 percent as opposed to 85 percent. However, in Serbia, 37 percent of respondents had a telephone as opposed to 19 percent in Montenegro; the figure for nondisplaced was 68 percent. A further breakdown reveals that of displaced groups in Serbia, only 17 percent of IDPs in collective centers, as opposed to 41 percent of refugees, had a telephone.8 settlement patterns and shelter situations 71 own their homes. Although the situation has varied from place to place, after the 1990s breakup of the Soviet Union and the Socialist Federal Republic of Yugoslavia, widespread privatization policies re- sulted in the distribution of titles to private houses and apartments.9 As table 4.1 reveals, Armenia represents the only major exception to this finding. Higher home ownership in that country, as noted else- where in this report, results from the ability of refugees coming from Azerbaijan to take over property left behind by ethnic Azeri refugees, who fled in the other direction. This highlights an important distinc- tion among displacement situations, between what might be called "symmetric displacement"10 and "asymmetric displacement." While the table does not include Azeri refugees, their housing situation is gen- erally assumed to be equal to, or better overall than, that of Armenian refugees. This is because in many cases they, in turn, were able to oc- cupy housing of the ethnic Armenians who left Azerbaijan. In addition to these situations of home ownership, 16 percent of Georgian IDPs11 Table 4.1 Home Ownership in Selected Countries Percent of local Percent of DPs population owning owning their Country their homes homes Armeniaa 82­87 66 (refugees) Azerbaijan 83 15 (IDPs) Bosnia and Herzegovina 82 15 (IDPs) Georgia 97 30 (IDPs) Serbia and Montenegro 90 18 (IDPs); 22 (refugees) Sources: Armenia: UN Office in Armenia, Poverty of Vulnerable Groups in Armenia: Comparative Analysis of Refugees and Local Population. UN Dept. of Public Information (Yerevan, 1999). Azerbaijan: World Bank data (2002). Bosnia and Herzegovina: World Bank, Bosnia and Herzegovina: Local Level Institutions and Social Capital Study (Washington, D.C.: World Bank, 2002), vol. 1. Georgia: IFRC, Internally Displaced Persons: A Socio-Economic Survey, Georgia. IFRC (Tbilisi, 2000). Serbia and Montenegro: UNHCR and the Commissioner for Refugees of the Republic of Serbia, Refugee Registration in Serbia. Government of Serbia (Belgrade, 2001). UNHCR and the Commissioner for Refugees of the Republic of Serbia, Registration of Internally Displaced Persons from Kosovo and Metohija. Government of Serbia (Belgrade, 2001). a. Some refugees who had lost their homes when they fled Azerbaijan have occupied the homes of Azeris who are now refugees in Azerbaijan. Because many of them have not received compensation for their homes in Azerbaijan, the government has provided them legal titles to these homes. In addition, refugees have been included in the privatization process. 72 living in limbo and 18 percent of Azeri IDPs,12 while not holding title to property, live rent free in government-provided private accommodations. Most displacement situations, including those in Azerbaijan (IDPs), Georgia, Serbia and Montenegro, and Turkey, are, one might say, "asymmetric" in that displacement did not involve an exchange of populations between combatants. Since displacement only went one way, no empty housing stock awaited DPs in the settlement country. While symmetric displacement does not necessarily mean an equitable exchange of property, in Azerbaijan and Armenia it has at least pro- vided individual shelter for the displaced.13 However, the situation also has created a supply-side dynamic that has helped dictate loca- tions for displaced population settlement. As noted previously, even though over 90 percent of Armenian refugees were urban and Azeri refugees predominantly rural, the presence of vacated property in cer- tain locations led both governments to settle their DP populations in ill-fitting environments. A similar exchange of populations occurred in the Balkans between rural Bosniacs from what is now the Republic of Srpska and urban Serbs from Sarajevo. However, though some examples of property ex- changes exist, in the majority of cases, these exchanges of populations have not, thus far, led to stable housing solutions for those still dis- placed. According to a recent estimate, almost 80 percent of IDPs in Bosnia and Herzegovina remain illegally squatting in the housing of IDPs of another ethnicity or of some of the estimated 600,000 refugees still living outside the country.14 In Bosnia and Herzegovina, these specific situations aside, the prop- erty situation is far too complex to discuss in detail in the context of this study. While the epicenter of the property problem in the Balkans lies in that country, the problem crosses the borders of Croatia and Serbia and Montenegro. The movements of the region's displaced set off a chain reaction of property disposition issues concerning IDPs within Bosnia and Herzegovina, refugees in Serbia and Montenegro, and IDPs and refugees in Croatia. Living with Relatives and Friends Perhaps the most unique element of shelter solutions for Europe and Central Asia's DPs has been the high number of displaced taken into the homes of the local popula- tion. Shared housing with host families represents one of the most im- portant shelter strategies for DPs in the region. Not commonly seen to the same degree elsewhere in the world, this shelter pattern is related to the fact (as discussed earlier) that most of the region's displacement situations have involved internally displaced populations or refugees fleeing to an area with linguistic, cultural, and, frequently, family settlement patterns and shelter situations 73 links. Thus offers were extended due largely to familial obligations and, in part, to feelings of solidarity for the plight of DPs. Conse- quently, in many situations, a significant number of private house- holds opened their doors to the DPs. A report in Turkey notes: [T]he forced migration connected with the conflict in southeastern Turkey has been chaotic and unorganized. Only the extended kinship relations of Kurds in southeastern Turkey have prevented a larger crisis, allowing the displaced to find a shelter with extended family members. Individuals who [fled]or [were] forced out of their villages in southeast- ern Turkey haphazardly sought refuge in already overburdened provin- cial towns and cities within the region or in Turkey's teeming western urban centres.15 One recent report estimates that even a decade after displacement, perhaps 35 percent of Kurdish IDPs still live in shared housing with relatives and kin. Serbia represents a particularly important example of this phenomenon. By UNHCR estimates, in 1994 some 90 percent of Serb refugees were settled in shared housing with relatives. Even now, about a third of refugees and 40 percent of IDPs in Serbia remain with host families. More than 10 years after initial displacement, about a third of IDPs from Abkhazia in Georgia reside with host fam- ilies. Altogether, more than a half million DPs throughout the region remain as guests in the homes of friends and relatives and have been in such a situation from 5 to 10 years or more. The high proportion of DPs in many countries living in such ad hoc situations is of significant concern for their current and future vulnerability. It is extremely un- likely that many of the host families anticipated that the situations would last as long as they have. DP asylum with host families has typically emerged out of solidar- ity and common feeling. Nonetheless, in many cases DPs have, either from the start or gradually over time, made financial or labor contri- butions to host households. In northern Albania, UNHCR provided cash and in-kind payments to host families to subsidize the burden of Kosovar refugees. In the early days of displacement in Serbia and Georgia, host families undoubtedly welcomed the humanitarian assis- tance provided to DPs. Guests may be of help in other ways. In Georgia, for example, there exists anecdotal evidence that some IDP guests have been useful in rural areas because they have replaced the manpower of emigrants with respect to the cultivation of family land. In urban areas, the value of such additional manpower is relatively less, and the generally smaller size of most housing means greater pressure on living space. A 2000 survey conducted by Save the Children in western Georgia reported that nearly two-thirds of host 74 living in limbo families said that if a similar situation arose, they would not be will- ing to provide space, a reflection of some of the costs, both financial and in terms of quality of life, host families have endured.16 Further- more, the longer such situations last, the more demographic factors create additional pressures for both host and guest families, as births occur and children grow up, producing a need for more space, and marriages take place, leading to a desire for separate housing for the new family. A review of the shifts in housing situations for DPs in Serbia since 1993 illustrates the flux that occurs in them. In the post­World War II period, when Yugoslavia was one country, Serbs from present-day Croatia were transplanted into Vojvodina following Tito's expulsion of Germans there. Croatian refugees who fled to Serbia in the early 1990s were drawn to Vojvodina, partly because of its geographic proximity but largely because of strong family links. As discussed above, families in the area opened their doors to the influx of refugees. According to estimates, in Serbia, in mid-1993, 95 percent of the refugee population resided with friends and relatives. Two-thirds of these host families, when surveyed, noted that they housed refugees because they were closely related. Over a quarter of these host families had previously migrated from Bosnia and Herzegovina, and another 10 percent had migrated from other parts of the former Yugoslavia. Three years later, the 1996 refugee census reported that only 52 per- cent of the refugees were housed with relatives and friends, 23 percent in rented accommodations, and 6 percent in their own accommoda- tions.17 By the 2001 census, the numbers in shared accommodations had decreased further as more DPs shifted out of relatives' homes. About 30 percent were reported to still be staying with relatives and friends; the number of those shifting to rental accommodations rose to 44 percent. Since 1996, the number of refugees in rented apartments has more than doubled, and the number in their own accommodations has quadrupled.18 Rental Accommodations Given the few DPs in the region who own their current residences, the remaining numbers in private accommo- dations either rent shelter or are provided it free by relatives, friends, or, in some cases, government authorities. About 44 percent of Serbian refugees and 41 percent of Serbian IDPs pay rent. By contrast, only 6 percent of the Georgians and a little over 3 percent of Azeris do so. When DPs rent accommodations, observers often assume that this fact represents a proxy for economic stability. However, data from two recent censuses of IDPs and refugees in Serbia, for example, show similar levels of unemployment among DPs in rental and in free settlement patterns and shelter situations 75 housing, suggesting that living in rental housing is not necessarily an indicator of self-reliance. Often it is, instead, the result of a lack of other options. The ICRC, comparing consumption expenditures of IDPs in Serbia in rental accommodations with those of local popula- tions, concluded: A major distinguishing factor between residents and IDPs is the expen- diture for accommodation: on average, residents spend only 6 dm for accommodation. This reflects the fact that residents overwhelmingly own their domicile and that only certain categories of people (students, daily workers) rent. In effect, this means that a proper comparison of expenditures between IDPs and residents should not take into account the amount of the rent. The average "available" expenditures of resi- dent households are thus twice higher than those of IDPs (dm 450 against dm 221).19 The World Food Program, citing 2001 data on its refugee benefi- ciaries in Serbia, reports that 44 percent of refugee households receiv- ing WFP food also pay rent.20 Given that WFP provides assistance to less than half the refugee population based on need, this seems to sug- gest that a significant number of the refugees with lower incomes are forced to pay rent and that shifts from such ad hoc shared accommo- dation arrangements are not necessarily the result of increased self- reliance. More often they seem to emerge from pressures by hosts to move out. In Georgia and in Ingushetia, at different points over the years, there have been frequent incidences of DPs trying to shift from private accommodations to collective centers, typically because they could no longer afford rent or have been asked to leave their shared accommo- dations with relatives.21 Given the issues of quality of life in most col- lective centers, decisions to shift to these accommodations highlight the reality that the displaced in private accommodations are not nec- essarily in sustainable situations. That the presence of DPs living in rental accommodations does not necessarily come from a rise in self- reliance is further illustrated by data from Azerbaijan. There the recent poverty assessment noted that IDPs in private accommodations were in one of the poorest groups in terms of consumption.22 The provision of housing to the region's displaced, through both government-provided space in collective centers and the hospitality of civil society, has represented a significant hidden subsidy to DP sur- vival strategies. Even when DPs have managed, by a combination of employment and government or donor assistance, to reach household incomes similar to those of the local population, as the need to shift to rental accommodations grows, so does pressure on this income. Rent 76 living in limbo is not, for the most part, a major expenditure of local populations (who largely own their homes). Yet continued living in collective cen- ters and with relatives has its own costs in regard to quality of life and social tensions. In any case, as the above examples demonstrate, it is realistic to expect that the longer displacement lasts, the more DPs there will be who need to leave shared accommodations and enter the rental market. As for collective centers, particularly in the South Caucasus, barring a major shift in government policy, DPs will be able to remain where they are.23 However, this eventuality, too, has a price in terms of the crowded conditions and the deteriorating infrastruc- ture as well as continued social exclusion and limited employment opportunities resulting from the isolated locations of some centers. Quality of Housing As indicated in the previous sections, typically DPs live in markedly lower quality housing and often suffer from shelter situations with a degree of instability. In terms of quality of shelter, DPs who live in col- lective centers live in much smaller, more crowded quarters than do local populations. The situation of DPs living in private accommoda- tions varies from situation to situation, but they also tend to have slightly less space than local populations.24 In addition, the displaced have less access to basic amenities, such as cooking facilities, toilets, and hot water; often they are, particularly in collective centers, forced to share these facilities. However, generally speaking, DPs in the re- gion do tend to have similar access to electricity and heating. Access to utilities differs from access to shelter in that utilities are relatively indivisible public goods. These goods can be drawn upon by popula- tions at large, whereas shelter is a private good dependent, on the one hand, upon the largess of domestic authorities (collective centers) and, on the other, upon the displaced's own capital resources (private accommodations). While utilities require monthly payment, the his- tory of free or relatively low-cost utilities in the socialist and early years of postsocialist societies facilitated DP access to such facilities. In Georgia and Azerbaijan, the displaced receive subsidies for utilities from public authorities, and in Georgia, they have benefited during some years from heating subsidies to vulnerable groups supported by the U.S. Agency for International Development (USAID). As privatization of utilities increases tariffs and the efficiency of tariff collection, DPs, particularly in large urban areas, are likely to suffer not so much because of their status as DPs but as a derivative impact of their overall income poverty. The presence of the displaced settlement patterns and shelter situations 77 in collective centers and irregular types of private shelter, such as shacks and unfinished buildings, will probably make some clusters of DPs more dependent upon electricity, as opposed to natural gas, for heat. Thus they may be disproportionately vulnerable to real increases in electricity tariffs. Conclusions From the foregoing discussion, we can reach several conclusions. First, DPs tend not to be evenly distributed within countries. They seem to be concentrated in two types of areas: regions adjacent to the conflict zone and their former homes, and large towns and cities. Within these areas, they tend to be deconcentrated, meaning that while they may live close to members of their families, former neighbors, or other DPs, they are not living next to them. Second, this concentration of DPs can put pressure on local social services and labor markets. Yet it can also facilitate the distribution of humanitarian assistance by creating economies of scale for aid agencies. Third, when DPs represent a significant part of the population of an area (such as a village, town, or district), they should not be consid- ered as just a vulnerable group. They need to be included in area development planning and programming. Fourth, the displaced tend to live in lower quality housing and be less satisfied with their housing than the populations surrounding them. Finally, DP housing seems to be less stable than resident housing because DPs rarely own their accommodations. Unlike the local population, they tend to live in col- lective centers or with friends and family or rent their accommoda- tions. Many collective centers were not built for long-term human habitation and are crowded as well as deteriorating from lack of maintenance. Notes 1. In addition to the importance of residential property as a form of asset, the overall issue of shelter represents an especially important aspect of DP vulnerability in Europe and Central Asia as compared with displacement situations in Africa and elsewhere in Asia. While there have been some tented camps and other forms of makeshift housing for DPs in this region, for the most part the colder climate limits the shelter options that are even minimally viable in the medium to long term. 78 living in limbo 2. This does not mean that all DPs fit this pattern but that it tends to be the predominant pattern. There are various reasons why individual house- holds or individuals end up all over a country's territory. Usually, it is because they have found a way to move closer to relatives or friends in the host society. 3. World Bank data (1999). Given the increase in the pace of returns in the past three years, it is possible that, in the interim period, these percentages have been reduced. 4. Examples include the Afghans in northern Pakistan and Iran, Cambodians along the Thai border, and Rwandans in Kivu Province (then known as Zaire). 5. NRC, Internally Displaced People: A Global Survey (Oslo: NRC, 2002). 6. The shelter situations of refugees in Armenia and Azerbaijan are relatively unique in that the simultaneous refugee flows in each country resulted, in many cases, in de facto and, in some cases, de jure exchange of properties between households in the two groups. Thus these two groups alone among DPs have high percentages of shelter ownership in their places of settlement. Estimates are that no more than 25,000 refugees in Armenia live in collective centers, a situation similar to that of DPs in Serbia. 7. Göc-Der, Socio-Economic and Socio-Cultural Conditions of the Kurdish Citizens Living in the Turkish Republic. 8. IFRC/ICRC/YRC (Yugoslav Red Cross), FRY: IDPs and Refugees Living Conditions. ICRC (Belgrade, 2000). 9. The current situation of DPs in regard to housing is perhaps also an early indicator of the growing housing problem countries in the region will face in the coming decade. While the numbers of homeowners of older gener- ations in the former Soviet Union and Yugoslavia have dramatically in- creased due to privatization, a similar windfall is not likely to occur for their children, who are already beginning to face issues of affordability of housing. 10. The most noted historical example of symmetric displacement was the exchange of populations between India and Pakistan in 1947, at the partition of colonial India. 11. IFRC, Internally Displaced Persons. 12. World Bank data (2002). 13. It is not necessarily true that all, or even the most desirable, property of departing DPs is acquired by, or assigned to, arriving DPs. The environ- ment of conflict often breeds opportunism, and it is frequently reported that in such situations the local populations also occupy valuable pieces of property of departing DPs. 14. World Bank, Bosnia and Herzegovina: Poverty Assessment. 15. Human Rights Watch, Summary (June 1996). 16. Dershem, Gurgenidze, and Holtzman, Poverty and Vulnerability. settlement patterns and shelter situations 79 17. UNHCR and the Commissioner for Refugees of the Republic of Serbia, Refugee Registration in Serbia. Government of Serbia (Belgrade, 1996). 18. UNHCR and the Commissioner for Refugees of the Republic of Serbia, Refugee Registration in Serbia, Government of Serbia (Belgrade, 2001). 19. ICRC, Sample Survey of IDP Households in FRY. 20. WFP, Refugee Beneficiaries of WFP Aid in Serbia. 21. An ICRC survey, for example, found that some Chechens have moved into collective centers from private accommodations because they were unable to continue paying rent. ICRC, Internally Displaced Persons in Ingushetia. 22. World Bank data (2002). 23. The same cannot be said of all situations. In Ingushetia, in late 2002, Russian authorities began to shut down utilities and other services to Chechen refugee camps, sometimes even cutting ropes on tented housing, in an effort to close camps and persuade Chechens to return home. 24. An assessment in FRY found that 80 percent of locals have more than 15 square meters per person as opposed to 3­24 percent of the IDPs and 8­31 percent of the refugees. In Georgia, the IFRC reports that locals, on average, have 18 square meters of space versus 13 for IDPs in private ac- commodations and only 7 for IDPs in collective centers. The most recent poverty assessment for Azerbaijan found locals to have about 12 square meters of living space per person, and IDPs only 6. A study of vulnerable groups in Armenia found that locals have 14­17 square meters of living space per person versus 10 for refugees in collective centers and 16 for refugees in private accommodations. chapter 5 Human Capital Summary: This chapter focuses on the impact of displacement on the human capital and, to a lesser extent, on the demographics of DPs. It begins by comparing the education and health indicators of the dis- placed with those of the nondisplaced. The psychosocial effects of dis- placement are then examined. The chapter concludes with a discussion of the differential impacts of displacement on gender and age groups. AS THE PREVIOUS CHAPTERS HAVE DISCUSSED, displacement creates pro- found limitations on the availability of economic resources of DPs. Many of their assets have been destroyed. They are denied access to what, in most cases, are their most important forms of fixed assets, such as land and residential property, because they lie within the area of ori- gin. Under normal circumstances, indicators of human capital repre- sent significant aspects of poverty and vulnerability. Levels of education and health are important in their own right and as indicators of mater- ial well-being. In this context, the most important resources DPs retain possession of are tied up in their human capital, which, by its nature, is mobile or at least appears so. This section discusses the major impacts of displacement on human capital, reviewing available indicators on the condition of the displaced in regard to education and health as well as demographics, which do not seem, in general, to differ a great deal, in this region, from those of local populations. There are some differ- ential effects on gender and age groups; these will also be discussed. Finally, the chapter will consider psychosocial issues that affect DPs. Demographics Conflict-related casualties among civilians can have significant im- pacts on the human and social capital of displaced populations, even after long periods of displacement. Direct effects on human capital in- clude the death or disability of household members. These can reduce 81 82 living in limbo the number of adult wage earners and result in large numbers of female-headed or single-parent households, with consequent impacts on earning capacity, gender roles, and household dynamics. Indirect effects include conflict-related trauma, which can occur even in situa- tions with low numbers of casualties. In terms of demography, displaced populations are often assumed to differ from others due to the impacts of conflict. There have been long-held public perceptions that DPs have higher dependency ratios in terms of numbers of young and old and have a gender imbalance, with more females and more female-headed households, than do local populations. Ongoing conflicts frequently result in patterns of dis- placement in which men are absent either because they are active com- batants or because they have fled in other directions to avoid being conscripted or arrested. Conflicts typically cause higher casualties among males than among females. Thus if casualties are high enough, they create a gender imbalance in the population that may lead to either smaller households, because adult males are missing, or larger households, because extended families group together for security. However, whatever their validity in other situations, in this region most of these assumptions appear not to hold true. The impacts of conflict on the human capital and demographics of displaced populations relate to the degree of violence experienced. That degree varies significantly from conflict to conflict, depending on the duration and nature. In many embattled countries, these impacts have been severe. The loss of 10­20 percent of the population of Rwanda or, in this region, of approximately 250,000 people in Bosnia and Herzegovina will have lasting effects on the societies of those countries. However, there is no a priori evidence that populations who become displaced by conflicts suffer from any greater mortality rates during it than other populations it affects. With the exceptions of Bosnia and Herzegovina and Chechnya, conflicts in this region have, on average, resulted in lower levels of casualties among civilians. In the Nagorno-Karabakh conflict, estimates put the military and civilian deaths on the Azeri side at approximately 30,000. Estimates of deaths in Georgia and in the Kurdish areas of Turkey suggest comparable levels of casualties in the fighting there. In fact, when groups are dis- placed early in a conflict that continues for a longer period of time, it may even be true that displaced populations suffer less from related casualties and trauma than groups who remain within the war zone.1 The mortality rate early in displacement offers another potential factor in creating different demographic patterns of displaced house- holds. For example, in 1994 perhaps as many as 30,000 Rwandan refugees in Kivu Province in Zaire died of cholera and dysentery in the human capital 83 first few months of displacement. In Europe and Central Asia, how- ever, the higher levels of capacity in receiving societies, good transport networks, and the immediate response of international assistance agencies in the first months of displacement (partly stimulated by the geopolitical realities of displacement in areas in close cultural and physical proximity to western Europe) prevented such extreme crises from emerging. In some circumstances, birthrates in refugee camps can be quite high, due to the listless existence of such camps and the need for phys- ical and emotional release. In some cases, a decision by political fac- tions to promote population increases as a way of engendering a new generation of soldiers for the conflict can contribute to birthrates. However, there is little evidence that this region's birthrates among the displaced differ from those of settled populations. The percentage of children in DP households does not appear to be significantly differ- ent. Where higher levels of children in displaced populations exist, such as in the case of IDPs from Kosovo settled in Serbia, this relates more to the demographic patterns in the area of origin than to the im- pact of displacement. If anything, DP households in this region show higher levels of elderly than those in other regions. But this observa- tion is also true of local populations in this region when compared with those elsewhere in the world. Higher numbers of elderly in DP households does not necessarily indicate higher levels of vulnerability, at least in the short to medium term. Given the low contribution of employment to DP incomes and the correspondingly high dependency level of DP household on public transfers, humanitarian assistance, or both, the elderly contribute to household incomes as individuals with entitlements and, in some cases, with pension income. No significant common patterns were found in the demographics of the region's displaced households in relation to household size or gen- der or age distribution, either in relation to cross-national compar- isons of DPs or in individual country comparisons of displaced and local populations. Education Educational outcomes, as measured by levels of educational attain- ment and percentages of children currently enrolled in school, seem very similar for local populations and DPs. When compared with the local population, adult DPs in Armenia, Azerbaijan, Georgia, and Serbia have similar levels of educational attainment. For example, an IFRC study of Georgia found 96 percent of locals and 96­98 percent 84 living in limbo of IDPs (25 years and older) had finished secondary school. Nor were school enrollment rates among the two populations very different, with 82 percent of the locals and 85­87 percent of the IDPs enrolled in primary and secondary school.2 Similarly, a study of vulnerable groups in Armenia found similar educational attainment among the local and refugee populations, with 68­71 percent of the former and 68­69 percent of the latter having finished secondary school.3 Finally, the recent poverty assessment in Serbia and Montenegro4 discovered displaced populations in Serbia to be slightly more edu- cated than the local populations.5 More residents had no or incom- plete elementary education (18 percent compared with 10 percent for IDPs and 14 percent for refugees). Percentages of DP and local chil- dren currently enrolled in primary and secondary school seem to be similar. Given the disruptive nature of displacement, the apparent lack of impact on educational outcomes of DPs requires some explanation. Displaced populations in other regions rarely show such high levels of educational attainment and do not always have good access to educa- tional facilities during displacement. Where such access exists, it is usually in the form of specially established camp schools supported by humanitarian agencies, not through access to the educational system of the host country.6 A large part of the answer on educational attainment relates to the comment at the beginning of this section that human capital is, in a sense, mobile. The majority of adults in the region's displaced popula- tions received most or all of their education prior to displacement. This observation may seem obvious, but it highlights the fact that the condition of DPs after displacement is very much affected by their condition prior to displacement. The displaced from the former Soviet Union and Yugoslavia lived in societies with relatively high levels of economic development, expansive public services, and almost univer- sal literacy. Most other displaced populations in the world have come from societies in which human development indicators were relatively low. While the duration of displacement has been long, lasting several years or more in each case, educational attainment data for adults are still largely dominated by those people who received their education prior to displacement. Thus it is a result of the predisplacement condi- tion of human capital rather than later impacts on that capital. In regard to the displaced's current educational enrollment, the main reason for continued educational advancement among this region's DPs relates to the environment in which they have settled.7 Elsewhere in the world, some DPs have been left in situations in which social services, even for local populations, have deteriorated through human capital 85 countrywide fighting. In the 1990s, IDPs in Afghanistan, Angola, and Somalia were left in situations in which the country or portion of country where they had settled collapsed around them, leaving little in the way of government services. In Europe and Central Asia, all major displacements have resulted from either conflicts in neighboring coun- tries or ones that affected only part of the country, leaving social ser- vices in other areas and adjacent nations relatively intact.8 While the political and economic transitions in the successor states to the former Yugoslavia and U.S.S.R have in many cases left social services in a weak and atrophied state, these conditions affect everyone in these societies, not just the displaced. As discussed previously, DPs in the region are, with few exceptions, either internally displaced, and thus still within their own countries, or refugees settled in neighboring countries that up until the early 1990s composed part of their own nations. This means that, in the case of IDPs, they have retained the right to access social services because, even though displaced, they remain citizens of the countries in which they live. Relatedly, the degree of dislocation and alienation resulting from displacement has been tempered by access to familiar educa- tional systems, in which teachers mostly use the mother tongue of DPs or a language with which they have proficiency. This is the case for most of the region's displaced groups. Over 75 percent of IDPs from Kosovo, for instance, are ethnically Serb and therefore have little problem integrating into schools with Serbian language instruction during displacement. However, exceptions do exist. Some Roma IDP children in Serbia and Montenegro speak Albanian (Kosovar Romas who speak Albanian do not usually speak Romani or Serbian) and thus, in addition to other difficulties, face the problem of a new lan- guage in schools.9 Not all DPs around the world have had these advantages. Particu- larly in Africa, refugees have often found themselves in neighboring countries without a common language, a factor complicating access to social services of all types but particularly education. Examples include Portuguese-speaking Angolans and Mozambiquans in Zambia and Malawi and Francophone Rwandans and Burundese in Tanzania and Uganda. Even when refugees speak the local language, other fac- tors can cause problems in schools. For Pushtoon Afghan refugees set- tled in linguistically Pushtoon areas of northern Pakistan, education was complicated by governmental pressure to include Urdu, the Pak- istani national language, in the curricula for refugees. DP families set- tled among local populations with different languages and educational systems face a dilemma. If they choose to promote measures that facilitate their children's adaptation to the country of displacement, 86 living in limbo they may watch their children grow up with a language and education that may not translate directly into opportunities upon a return home. If, however, they choose to invest in separate education taught in their own language, which they hope will be adaptable to the home coun- try's educational system, they may facilitate a return home but weaken opportunities for their children in the country of displacement. Fre- quently, this decision is taken out of their hands by host governments who do not wish DPs to integrate locally. These governments discour- age access to local educational systems by various de facto or de jure measures. This region's displaced have, for the most part, not had these problems in education for the reasons stated above. How- ever, the broader problems are of differences in generational experi- ence among the displaced and of acclimation of younger DPs to new languages and environments are issues of note. (They will be discussed in a section below.) While recognizing the value of enrollment data as a proxy for edu- cation, it is important to add some caveats. The available data give us no detailed information on regular school attendance by DP children. A similar level of enrollment (or even attendance) does not mean that DP children, given the burden of financial and psychological effects of displacement, learn or advance at the same rate as other children. It is also likely that most DP children have faced significant obstacles to continuation of their studies at the onset of displacement. Educational systems run on strict schedules, with academic years beginning and ending on fixed dates. Even if studies can be continued within a few months after displaced populations are settled, timing is linked to key examination periods or phases of a school year; a short gap can lead to loss of a full academic year. Finally, the pattern of subnational concentration and clusters of the displaced in specific localities in host countries is likely to negatively affect the quality of social services in geographic areas with high concentrations of DPs.10 Although some national governments have provided special funding to subsidize local government for this additional pressure on infrastructure and staff, such increases in population in these areas have probably diminished the quality of social services, including education, through factors such as deterioration of physical infrastructure and additional burdens on teachers, health workers, and other such professionals. From two countries, some limited evidence that enrollment in higher education may be lower among DPs than local populations can be found. In Armenia, for example, it was reported that just over 50 percent of DPs older than 15 were enrolled in educational institutions while the corresponding number for the local population was around 60 percent. In Azerbaijan, it was reported that local human capital 87 populations were twice as likely as DPs to be enrolled in university- level education. Although there are not sufficient data to make general statements regarding higher education and DPs, the data from Armenia and Azerbaijan suggest the possibility that the displaced, in general, may face some disadvantages when it comes to accessing higher education. In other regions of the world, the displaced, particularly refugees, are isolated in camps, and education is provided by camp authorities with support from humanitarian assistance. Therefore it is not surprising that the opportunities for higher education are limited due to the higher levels of investment required for creating parallel systems and the initial assumption that displacement is temporary and does not justify the creation of such expansive systems. The issue of language and legal access to higher education also frequently complicates the ability of refugees elsewhere in the world to access higher education. In Europe and Central Asia, these factors are more or less absent. Thus if such a differential exists it may be an indirect effect of DPs' higher levels of poverty, particularly in terms of cash income. Poverty directly affects the ability of families to invest in higher education and, relatedly, puts pressure on young adults to work instead of study. Other factors that may reduce or complicate opportunities for higher education among DPs include the loss of forms of certification such as school records and diplomas or the rejection of credentials attained from another country. Finally, when competition for places in higher educational programs is fierce, local populations, on average, have better access to social networks that may provide avenues of influence on admission decisions. Health The access of displaced populations to health care varies from situa- tion to situation, depending on host government decisions to allow DPs unrestricted access to public health facilities. In this region, the displaced generally seem to have similar access to health facilities as the local population. Available data do not indicate any major differ- ences in the physical health of DPs versus local populations in Armenia, Azerbaijan, Georgia, or Serbia. While we have no compara- tive data for Turkey, in a recent survey there DPs did not report health issues as a major concern. The relatively comparable physical health is primarily the result of the displaced's ability to access the host country's public health facilities, due either to their rights of citizenship or the government's 88 living in limbo acceptance of their right to access. During early periods of displace- ment in Azerbaijan, Georgia, and Serbia, additional health facilities for DPs were operated by humanitarian agencies such as the ICRC and Médecins Sans Frontieres (Doctors Without Borders). Another reason for similar health outcomes is the nature of most of the region's conflicts. As discussed, they did not, overall, result in high numbers of casualties or create huge numbers of conflict-related dis- abilities specific to displaced populations.11 The issue of health care for DPs in the region is generally related not so much to displacement's impacts on access to public services as to its impacts on income. The deterioration of social services across the region has forced private purchase of medicine. As a consequence, local populations have sought private health facilities or quasi-private arrangements with additional informal payments to public health facility employees. It is probably true that, to some degree, DPs are disadvantaged in that they cannot as easily draw upon social networks as an alternate way to facilitate adequate treatment in public facilities. However, given the nature of the existing de facto health systems, the predominant issue is displacement's impact on income. Furthermore, on average, DP families have smaller cash incomes available to pay for "supplementary" medical service. In this, their situation does not differ considerably from that of other poor populations in affected countries. A special issue in DP health relates to those DPs remaining in col- lective centers not originally intended for long-term human habitation, which are deteriorating over time (see chapter 4). Without some ade- quate solution to the shelter problem of these DP populations, various health problems associated with the conditions in collective centers, such as respiratory problems and diseases stemming from poor sanita- tion, are likely to increase with time. Psychosocial Issues Given the traumatic nature of displacement, it would not be appropri- ate to discuss the health situation of DPs without a discussion of the resulting psychosocial impacts. They relate to the trauma and violence of the conflict that stimulated the displacement and to the psychologi- cal effects associated with the reality of living as a displaced person. As with other interlocking issues discussed in this study, psychosocial fac- tors not only are aspects of quality of life and well-being in their own right, but also represent potentially significant factors influencing, for example, the capability to move toward self-reliance. human capital 89 Because of this study's nature and reliance on secondary sources of information, it has proved difficult to quantify the impact of psycho- logical issues in regard to DP vulnerability. Typically, general surveys incorporate broad questions, measuring, for example, peoples' re- sponses regarding their satisfaction with their current circumstances and their level of optimism toward the future. Not surprisingly, DPs tend to answer these questions more pessimistically than do others. In surveys in Armenia, Azerbaijan, and Georgia, the displaced re- sponded more negatively than local populations to these types of questions. A study in Georgia, which utilized a more comprehensive approach to estimating depression, found higher rates of depression among IDPs in collective centers (90 percent) than among the general population (65 percent).12 Similarly, another study in the Zugdidi region of Georgia13 reported higher rates of family violence (physical and verbal) and more pessimism about the future among IDPs. In Serbia, studies also discovered higher levels of pessimism among DPs.14 In addition, there exists a wealth of anecdotal evidence illus- trating the psychological impacts of displacement on affected popula- tions and some illustrative data regarding DP psychological health as well as the prevalence of illnesses typically associated with emotional stress, such as ulcers. Even if more comprehensive information were available, it is equally difficult to draw cause-and-effect relationships between psy- chological factors and material well-being. How important is depres- sion or trauma-related paralysis in the unemployment of DPs? How does the feeling of being in limbo affect investment strategies of DPs deciding whether to sign a two-year lease, to open a small shop, or continue to work as a small-scale itinerant trader with a lower profit margin? We will suggest some linkages here but cannot provide defin- itive answers. Nonetheless, the psychological factor seems such an important element of the displaced's condition that we feel the need to spend some time discussing it. It is perhaps important to emphasize, as we have elsewhere, that many problems faced by DPs relate directly to the poverty and unem- ployment created by displacement. Much of the pessimism and de- moralization among DPs invariably comes from these factors; other segments of the local population who are also poor are similarly af- fected. A dramatic reduction in poverty and increase in employment among DPs might go a long way toward minimizing some of the psy- chological impacts of their situation. However, such a sea change in material circumstances is unlikely to occur in the immediate future, and there are other relevant factors that are specific to DPs. Thus a better understanding of the range of issues contributing to the 90 living in limbo well-being (or lack thereof) of the displaced provides a valuable foun- dation for policymaking. Conflict-Related Trauma In trying to isolate the psychological impacts on displaced popula- tions, researchers need to understand that some psychological issues of DPs relate not so much to the act of displacement as to the trauma associated with the violence of conflict experienced prior to displace- ment. In comparing DPs with local populations, conflict-related trauma can be separated into two categories of situations: those in which the local population have also experienced conflict, and those in which displaced populations alone have experienced conflict (the local population having been isolated from the fighting, which occurred ei- ther in another part of their country or in another country altogether). Though in both situations DPs as a group have been affected, in the first instance there is no specific reason to single them out in terms of the effects of conflict-related violence. In the second, however, DPs suffer the double burden of displacement and being the only group in the society who have experienced conflict. Trauma from conflict and violence is difficult to discuss dispassion- ately or to measure in comparative terms. It comes from the experi- ence of physical acts, such as rape and mutilation. It comes from witnessing the murder of others, sometimes close family members and neighbors. It comes from the threat of violence and the fear of violence, whether acted upon or not. It comes from the wrenching of closely held ideals and faith in a nation, an ideology, or the value of friendship among neighbors. It comes as well from the knowledge that the normal, everyday things that are cherished and taken for granted have dissolved into nothingness. Measuring such trauma or trying to say that one person's is less or more than another's is a dangerous and contentious task. Yet gradations of conflict-related trauma do exist even if we have no benchmarks or scales by which to measure them precisely. In some countries, such as Bosnia and Herzegovina, it is difficult to determine whether conflict-associated trauma is limited to the dis- placed population. Specific clusters of population (for example, the survivors of a massacre such as Srebenica in 1995) are victims of trauma associated with their particular horrific experience. Yet a country's whole population could be viewed as victims of psychologi- cal trauma, with a variety of complicated long-term effects. How, for instance, do we compare the trauma of a Serb IDP family who fled its home in Sarajevo to that of a Bosniac family forced to stay in the city human capital 91 through three years of siege and daily sniper fire from the hills above?15 However, in most displacement situations in the region, DPs have lived a different experience from the local population. Thus they are doubly singled out, as populations who have been displaced and pop- ulations who have suffered through a conflict. Local populations in Baku, Belgrade, Ingushetia, and Istanbul living side by side with the displaced did not experience the same conflict as the DPs. As noted, the violence toward the region's displaced populations immediately prior to their displacement was not, on the whole, comparable to ex- periences of, for example, Afghans bombed and attacked for two to four years before they made their trek over mountains and deserts into Pakistan. Nonetheless, even a few weeks or months of conflict is suffi- cient to create memories and effects that may linger for long periods of time. Dependency Syndrome Another psychological factor frequently assumed to influence dis- placed populations comes from strategies implemented by govern- ments and donors to support them, that is, providing benefits and relief assistance. It is not uncommon to hear local populations claim that DPs are privileged, that they receive benefits not available to oth- ers also in need. According to this view, DPs consequently are lazy and unwilling to work. Available information does show that in all the region's major situations of displacement, except Turkey, DPs receive significant proportions of their household incomes from public assis- tance, international humanitarian agencies, or both. Unemployment tends to be disproportionately high among DPs, in comparison with unemployment among local populations, despite similar household demographics and numbers of able-bodied adults. However, as em- phasized throughout this study, nowhere in the region do humanitar- ian benefits represent the majority of income for DP households. Yet where benefits are targeted to DPs on the basis of identity as displaced populations, rather than on a needs-based approach, the possibility of a feeling of entitlement among DPs is not uncommon. This may lead to a so-called dependency syndrome, an unwillingness to work or to seek work due to long dependence on external assistance.16 This sense of entitlement is further reinforced when host govern- ments view displaced populations as part of a political dialogue and treat assistance, at least partly, as compensation for economic losses suffered through an ongoing conflict. This is the case, for example, with IDPs in Azerbaijan, Georgia, and Serbia. Under such circumstances, 92 living in limbo the displaced are supported in the belief that their sacrifices for the country as a whole mean public authorities have a responsibility to pro- vide livelihood subsidies. Dependency syndrome is a factor that cannot be ignored. However, as discussed earlier, nowhere in the region do long-term displaced populations receive enough assistance to obviate their need for employment. Despite the high rates of DP unemploy- ment, the readily available evidence of DPs working in menial or ad hoc occupations throughout the region suggests that dependency syndrome is not a primary reason for DP poverty. The Impacts of Living in Limbo A recent UNOCHA report in Serbia remarking on declines in psycho- logical conditions among IDPs notes: There are numerous reasons for the deterioration in psychological well- being, from the actual loss of home and previous routine, to poverty, dire living environment and the resentment of vulnerable locals who see the newly arrived as competitors for scarce job opportunities and assis- tance. The realization that return in the near future is unlikely makes them feel "neither here nor there."17 Psychological impacts of displacement--alienation, socioeconomic dislocation, and feelings of loss and hopelessness--perhaps accompa- nied with a nostalgia for a semi-mythical past, are in certain ways sim- ilar to the types of impacts frequently heard across the former Soviet Union, and sometimes in the former Yugoslavia. The combination of economic, social, and political disruptions that have occurred in these societies since the dissolution have in many ways left the local popula- tions, too, dislocated from their past. Despite this interesting similar- ity, there exists one difference between the impact of state dissolution and economic transition on local populations and the impact of dis- placement on DPs. Citizens of the successor republics of the former Soviet Union and Yugoslavia have no likelihood of recapturing the universal social welfare system or full employment nor the sense of identity and purpose some of them associated with the previous polit- ical systems. Even those of the older generations who look fondly on the past know they can never go back. DPs, by contrast, are caught in a strange sense of limbo, not knowing whether they will stay or go. They suffer from a different nostalgia, one that mixes the real issue of whether they can ever go home with a vaguely remembered under- standing of what "home" actually was during a different period of his- tory. While they believe their nostalgia is about a place alone, in most cases it is about a time as well--not only of a house and land in an area human capital 93 they have not seen in many years but also a time when their homeland formed part of the Soviet Union or a federated Yugoslavia. The impacts of this feeling of "living in limbo" are difficult to mea- sure. However, such impacts undoubtedly are significant and increase feelings of tension, demoralization, and lack of empowerment among DPs. These emotions arise not only because the future disposition of their situations is beyond their control but also because it is unpre- dictable. This unpredictability creates dynamics in many facets of DP daily lives and is likely to be severest in situations in which the socio- economic conditions of displacement differ radically from those of the home area. These situations include, for instance, rural populations being settled in urban areas, and DPs being caught in different cultural or linguistic environments. Decisions on how to educate children and what occupations to pursue (when options exist) are some examples of areas in which the uncertainty of the future creates tradeoffs and ten- sions unique to the DP condition. Another area comprises strategic economic decisions, including whether to invest in businesses or other economic activities such as purchasing or improving agricultural land. These have long gestation periods, do not provide immediate returns on capital, and may not be easily liquidated when opportunities to re- turn home present themselves. Even in regard to social relations and social networks, the feeling of limbo creates a difficult environment for DP actions. How much effort should be invested in social networks in the area of settlement, which may have importance for both psycho- logical stability and economic benefits? How much effort should be put into the maintenance of social networks from the home area, which may have less value in displacement but would rise in impor- tance if DPs return to their home communities? When marriage is viewed as an aspect of familial alliances, should families value marriage within the home community or create alliances with local communities? It is not impossible for DP households to balance these various factors, but such efforts have costs in terms of time and money as well as psychologically. This feeling of limbo and the consequences it has for how DPs live day to day and what factors influence momentous decisions in their lives form the essence of displacement. When this feeling of limbo goes away and a family or an individual decides to settle in the present location is the day when a DP becomes a resident. DPs may be wealthy and well-established residents or they may be poor and alienated ones. However, while individuals in such a situation may continue to be part of a vulnerable group with special needs (due, for example, to their lack of land or residence ownership), they are no longer displaced. 94 living in limbo However, several factors complicate any psychological transitions to integration. The first factor is that the continued state of limbo is only partly an- chored in a DP's personal decision to retain or give up hope in a return home. In most cases in this region, governments have created two pol- icy frameworks with opposing objectives. In one, the government denies members of a group the status of displaced persons and insists that they can, and should, return home. Alternatively, and more fre- quently, the government insists that they remain displaced because that suits its broader objectives. In both situations, the government has decided that local integration of DPs is not a desirable outcome. Under such circumstances, if a return home is not possible or if DPs them- selves do not desire it, the state of limbo is maintained, its parameters bolstered by the host government's actions. Some of these actions are tangible, such as limitations on the ability to purchase property or vote. Often maintaining identity as displaced confers on a household real or perceived benefits that may be removed if members take steps to integrate. Other actions are more subtle, such as political state- ments from high government officials speaking of IDPs as heroes to a national cause, accompanied by bright promises of full support when the time comes to return home. (See chapter 7 for a more analytical discussion of relations between host governments and DPs.) The second problem in assessing when such a state of limbo ends concerns time. How much time must pass before people no longer see themselves as displaced? Even when the displaced are not formally given integration rights, human beings are resourceful creatures and people can frequently find ways around both de facto and de jure obstacles, despite significant transaction costs. How long does it take for an individual to abandon hope of a return home? Commonsense seems to indicate that a long period of displacement would erase any feeling of limbo, and DPs would simply "get on with it," making a new life where they are. But such decisions are individual. Even without considering the actions of host governments, some people may give up after six months, and others may never give up. The Rwandan Tutsis in Uganda returned home (through an armed invasion) more than 30 years after they were displaced. Millions of Afghans are now going home after more than 20 years of displacement. Palestinians still view themselves as refugees more than 50 years after they left their homes. Related to this problem is the fact that even assessing the length of time of displacement as a factor in peoples' actions is problematic. As of this writing, Azeri IDPs have, in principle, been displaced for nearly 10 years. However, on at least three occasions during that time, nego- tiations between Azerbaijan and Armenia seemed close to a resolution human capital 95 of the conflict and, consequently, a potential return home for IDPs. Each time, to varying degrees, IDPs were forced to review their situa- tion, shift mental gears toward preparation for a new reality, and then gear themselves back down to remaining displaced. Relatedly, ethni- cally Armenian refugees from Azerbaijan, despite 12 years of displace- ment, continue to decline offers of citizenship by the Armenian government. Bosnian refugees and IDPs were given a sense of hope in 1995 with the signing of the Dayton Peace Accords. Yet due to inherent contra- dictions in the accords, notably that while guaranteeing the right of return of ethnic minorities, they also legitimized ethnic majority lead- ership patterns in the federation and the Republic of Srpska, it was not until 1999 that any significant returns were possible. Even now, Bosnian IDPs continue to exist in such a state of limbo. By contrast, the results of a recent census of refugees in Serbia indicate that refugees in that country have largely given up on going home.18 Differential Impacts on Gender and Age Groups There are few quantitative data available that demonstrate the differ- ing impacts of displacement on household members. However, a vari- ety of qualitative reports do discuss differential impacts of displace- ment on gender and age groups. There is also something in the logic of the dynamics of displacement that creates differential impacts on spe- cific groups. Impacts on Gender Groups As noted in the beginning of this chapter, observers often assume the displaced are more vulnerable because they have more female-headed households or a higher proportion of females. The vulnerability of female-headed households is assumed to emerge from (a) the tradi- tional role of males as wage earners and the gap that their absence creates in household income, (b) social pressures and prejudice in em- ployment of females in traditionally male-dominated societies, (c) and pressures on single-parent households. Whether female-headed house- holds are consistently more vulnerable than male-headed households is a question worthy of review.19 In any case, when compared with local populations, displaced populations in the region did not show significant variance in the number of female-headed households.20 In fact, based on incomplete data, in some European countries, such as Bosnia and Herzegovina, Moldova, and Serbia and Montenegro, 96 living in limbo women composed less than 50 percent of the displaced population. This is, however, based on smallish samples taken mainly from collec- tive centers and camps, not from urban areas, where most DPs tend to be settled. Were the data more representative, the picture of a slight predominance of females would probably hold true.21 The one demographic area in which some wider patterns of differ- ence appear to arise between local populations and the displaced is in the gender ratios among the elderly. In this age group, in three coun- tries for which comparative data are available, the ratio of females to males among DPs is higher than among the local populations. Similar variations are not present in gender comparisons of younger adults. The differential effects of displacement on gender are dependent to a great degree on gender roles within the family in the predisplacement period. In societies, for example, in which women are typically clois- tered within the family and kin group, the disruption of displacement can result in further efforts by males to impose restrictions on the mo- bility of women. This can be particularly true when rural populations shift to urban areas, large collective centers, or camps, where a wide variety of displaced from different places are settled. Among Kurdish IDPs in Turkey, a specific gender-related problem is associated with language. While males typically could communicate in Turkish prior to displacement, women, more engaged in traditional roles in the home and in agricultural production, typically spoke only Kurdish languages.22 While populations were at home in rural areas of southeast Turkey, this linguistic limitation had implications for women's mobility but not likely serious impacts on daily life in an area where Kurdish speakers were the majority. However, displacement has frequently shifted DPs to urban areas in other parts of Turkey, outside Kurdish settlements; there knowledge of Turkish is essential for access to employment, social services, shopping, and other daily functions. Displacement has also, in certain circumstances, the potential to empower women. For example, though living in collective centers and camplike settings may result in attempts by males to limit the mobility of women, in practice such attempts are difficult to fully impose, given close quarters and the effects of poverty. Such settings may expose rural women to a wider circle of experiences, links with other women, and other circumstances that would not have been possible at home. Poverty may also force women into the employment market. Some- times it may be women more than men who, for varying reasons, can find employment, such as in small-scale trading and sales. Some men find this type of work beneath them. Donor-supported microcredit programs have targeted individual displaced women and women's associations in several countries. In Georgia, groups of women have human capital 97 been able to conduct trade in citrus fruits and other goods between Abkhazia and Georgia in a manner in which men, sometimes per- ceived by Abkhaz authorities as potential guerrillas, cannot.23 In one sense, the burden of displacement falls disproportionately on women in regard to their typical roles as manager of the household. This is particularly the case when DPs are settled in collective centers or other accommodations with shared or minimal facilities for water, san- itation, and cooking. However, in another sense, while such tasks may become more difficult, they are not qualitatively different from predis- placement task environments. In situations of displacement with high numbers of widows and female-headed households, the pressures on women increase. However, as previously noted, there is little evidence in this region's major displacement situations that DPs have signifi- cantly higher numbers of female-headed (or single-parent) households. By contrast, in predisplacement situations in which males were more widely employed outside the household for wages, the disloca- tion and unemployment caused by displacement may represent a radically different environment. In societies in which males are tradi- tionally viewed as the "breadwinner," displacement undermines a key foundation of male identity and perceptions of self-worth. There is also sometimes a sense that males have somehow failed as providers through "losing a war" or otherwise preventing the act of displace- ment from taking place. At the beginning of this section, we noted that there appears to be a gap, at least in the three countries of the South Caucasus, in numbers of DP elderly males. Given that the variance is in the elderly, it seems unlikely that this difference is due to male migration for work. Simi- larly, conflict-related casualties in these three countries did not suffi- ciently account for the gap. One possible conclusion is that male DPs in these countries are dying earlier than males among the local popu- lation. Given that in all three cases displacement has endured for a decade or longer, the cumulative psychological effects of displacement, including those resulting from unemployment and loss of status, could have had a higher impact on the overall health of males than females. While there is no real evidence for such a supposition, it seems one of the only reasonable explanations and, perhaps, bears further research. Impacts on Age Groups In some ways, our first inclination was to raise issues regarding chil- dren or youth in displacement as a separate issue. Displacement does bring up specific issues regarding children and youths, partly related to the poverty of displaced households and consequent implications for 98 living in limbo child labor as well as the multifaceted effects of household instability on children as they mature. However, in thinking further on the long periods of displacement that have characterized most of the region's currently displaced populations, we came to the conclusion that the issue of the impact of displacement on age groups is more complicated. There are, in effect, three groups within displaced populations seg- mented by the age and stage of life when displacement initiated. Some age cadres matured prior to displacement and have a strong memory of the predisplacement situation. They were educated and probably employed under relatively stabler conditions in environments in which, typically, their families had lived for several generations. Other age cadres were adolescents or teenagers when displacement began. Finally, a third age cadre comprises those who were either small chil- dren when displacement occurred or born into displacement. In a sense, the vulnerability of these various age groups is path dependent. If DPs remain in the place of settlement, the youngest among these groups will be the likeliest to adapt easily to the environ- ment there. However, in a future situation in which return home is possible, this group will be the likeliest to be ill prepared, having only vague memories or none at all of the area of origin. The second age group probably faces the most serious adjustment problems, regardless of the future. Those who were adolescents and teenagers during conflict and initial displacement were old enough to begin the process of growing into adults and taking on adult responsi- bilities but too young to fully form themselves. In some situations, they may have been soldiers themselves prior to displacement. In dis- placement, their educational or professional development paths are disrupted, often midstream, and they have few, if any, skills or oppor- tunities for employment. Their elders and role models have diminished relevance for the new environment since they, too, have lost their po- sitions in society and the foundations of much of their authority. While this group begins as adolescents, the problem they face is not only one of adolescence because the longer displacement lasts, the older this group becomes. In early periods of displacement, such an age cadre is a cadre of youths. But after 10 years of displacement, this category includes a cadre of people in their twenties and thirties. This is the group among the displaced truly caught between two worlds. The men among this group are also more likely to get involved in crim- inal behavior similar to that of young men all over the world. Dis- placement may exacerbate this tendency. They also tend to suffer from more discrimination than any other age or gender group. The third group, those who were adults and had families at the onset of displacement, are, in some ways, the ones who have lost the most and retain a knowledge of that loss. human capital 99 Conclusions From the foregoing discussion, several conclusions can be drawn. First, in this region, with the exception of Bosnia, there have been fewer casualties from conflict than in other regions. There have also tended to be fewer casualties during the displacement process. Finally, the birthrates of DPs do not seem to diverge very much from that of the populations surrounding them. The combination of these factors, among others, has led to no major departures in DP demographic profiles, as measured by their household size or their gender and age distributions. Second, educational indicators reviewed for this study seem to indi- cate that DPs and local populations do not differ very much in terms of their educational attainment and enrollment rates. This can be partly explained by two factors: most adult DPs completed their edu- cation prior to their displacement, and DP children have widespread access to public education, often in their own language. However, these indicators do not tell us about the attendance rates of DP chil- dren or the quality of their learning. In addition, there is some evi- dence that at the tertiary level, fewer DPs than locals are enrolled. Third, as with education, the available health indicators seem to show that the health status and access to services of DPs do not dif- fer much from those of the local populations. This can be largely explained by two factors: the relatively few casualties produced by conflicts in the region, and the seemingly widespread access to health services provided by the national governments in the countries of asylum. However, the deterioration and privatization of health care in the region disproportionably affects the poor among both the local and displaced populations. Fourth, DP psychosocial triggers include the violence and trauma brought about by the displacement and the depression and demoral- ization that can come with living as a displaced person. As with other facets of being displaced, they are important in their own right and can contribute to the vulnerability of DPs by, for example, limiting their ability to move toward self-reliance. Fifth, conflict-related trauma comes from, among other factors, (a) directly experiencing rape, mutilation, and physical abuse; (b) hav- ing a loved one experience violence or death; (c) the threat and fear of violence; and (d) the wrenching of ideals and values. Sixth, when benefits are provided to DPs based on their status rather than need, a sense of entitlement can develop, making DPs less willing to become self-reliant, that is, they suffer from a dependency syndrome. However, the displaced are rarely able to meet all their basic 100 living in limbo needs with humanitarian and government assistance. They usually need to supplement this assistance with some form of employment. Finally, as discussed elsewhere in this study, DPs often feel "neither here nor there" or "don't know if they are coming or going." This state of "living in limbo" affects both short- and long-term decisions, such as whether to invest in a business, a relationship, or a language. Should a DP assimilate and make his or her current life more reward- ing or hold on to the traditions of the area of origin to facilitate a pos- sible return? This state is the essence of being a displaced person. Notes 1. They do, however, suffer from the psychological impacts of the process of displacement itself. 2. IFRC, Internally Displaced Persons. 3. UN Office in Armenia, Poverty of Vulnerable Groups in Armenia. 4. World Bank data (2003). 5. Kosovar refugees displaced into northern Albania were also much better educated than the local population. 6. In one major example, Azerbaijan, perhaps 60 percent of the displaced are taught in special classes and segregated from the general population in regard to education. A minority of DP students in Georgia are similarly taught in DP classes. However, most other DPs in the region have been integrated fully into local educational systems. Even in the cases of Azerbaijan and Georgia, DPs are taught with the same curriculum and in the same educational system as local populations. 7. The fact that parents of DP children in the region generally have higher levels of education, and grew up under systems in which education was im- portant and prestigious, may also lead to family ethics valuing education. 8. The major exception would be IDPs within Chechnya. Although there are no reliable data available, assumedly this group is in a less privileged sit- uation due to the impacts of the active conflict. By contrast, Chechen IDPs who settled in neighboring Ingushetia do have regular access to local educational facilities. 9. UNOCHA, Serbia Situation Report (April 2002). 10. Ironically, in some cases, this pressure on infrastructure has been exacer- bated because local school buildings have been taken out of service and utilized as collective centers for DPs. 11. To take the example of landmine-related disabilities: in some countries, such as Afghanistan, Angola, and Cambodia, displaced populations included high numbers of landmine victims. This was a result of the human capital 101 indiscriminate laying of mines and the reality that many DPs had spent long periods of time in the conflict zone prior to displacement. 12. Dershem, Gurgenidze, and Holtzman, Poverty and Vulnerability. 13. Shakarishvili, Evaluation of Socio-Behavioral Patterns. 14. UNOCHA, Serbia Situation Report (various 2001­2002 reports). 15. Even in Bosnia and Herzegovina, the 2003 World Bank poverty assess- ment reported that displaced women were twice as likely to suffer from trauma as local women. The study did not detail whether the cause of the trauma was related to previous violence or to the direct impacts of displacement. 16. One concise definition of "dependency syndrome" is "an attitude and belief that a group cannot solve its own problems without outside help." Phil Bartle, The Dependency Syndrome. Community Management Programme training material, http://www.scn.org/ip/cds/cmp/modules/ pd-dep.htm. 17. UNOCHA, Humanitarian Risk Analysis, no. 18 (April 2002). 18. UNHCR and the Commissioner for Refugees, Refugee Registration in Serbia (2001). 19. In the recent World Bank poverty assessment for Bosnia and Herzegovina (2003) female-headed households, in fact, showed up as relatively less vulnerable than other households. However, the reason was specific to Bosnia and Herzegovina in that such households typically represented war widows who were allocated specific compensation payments by the Bosnian government. 20. UNHCR, Statistics and Registration: A Progress Report. UNHCR (Belgrade, 2000), quoted in USCR (U.S. Committee for Refugees), Refugee Reports, 21(2) (Washington, D.C., 2000). 21. UNHCR, Women, Children, and Older Refugees: The Sex and Age Dis- tribution of Refugee Populations with a Special Emphasis on UNHCR Policy Priorities. UNHCR (Geneva, 2001). 22. In a recent survey of Kurdish IDPs, 77 percent of male adults reported being able to speak Turkish "enough to express themselves in daily life," while only 21 percent of female adults could do so. In addition, 61 per- cent of female adults, but only 6 percent of male adults, reported not knowing any Turkish at all. 23. Julia Kharashvili, Georgia: Coping by Organizing, in Vincent and Sorensen, Caught between Borders. chapter 6 Social Capital Summary: This chapter focuses on the social capital of displaced populations. It is divided into two main sections. The first discusses the factors that influence social cohesion within DP communities, including those that preceded the displacement and the impacts of the displace- ment itself. The second section reviews the set of issues that influence relations between DP and host communities. SOCIAL NETWORKS, WHETHER ASCRIPTIVE or voluntary, are an important factor in understanding the coping strategies of vulnerable popula- tions. Such networks have the potential to provide informal social safety nets, allow for collective action to the benefit of members of the network, and otherwise contribute to the well-being and overall sur- vival strategies of individual households. These networks can also rep- resent important links between public institutions such as local state administration and member households. In the context of populations displaced by conflict, when fixed assets have been destroyed or made unavailable and when income streams are disrupted, an understanding of the potential role of social networks in reinforcing coping strategies seems to have even greater significance. Social capital has been defined as "networks, norms, and values that enable people to act collectively to produce social benefits."1 We use this broad definition of social capital because it captures not only social networks but also their behavioral and attitudinal under- pinnings. Displacement's impact on "norms" and "values" is an important consideration when attempting to understand the degree to which old social networks are maintained and new ones established during displacement. At the same time, we will still refer to "social networks" in relation to the actual tangible networks of relations DPs maintain and create. The displaced face a rather unique environment for the mainte- nance and formation of social capital due to both the effects the process of displacement has on social fabric and the uncertain future 103 104 living in limbo DPs face in terms of a return home. Some of the key considerations include the following: · Special challenges in social inclusion and exclusion. DPs face a dual challenge of maintaining social cohesion within their com- munities in exile as well as building links with local populations. · Predisplacement impacts of conflict on affected communities. DP social networks, even before displacement begins, have typically experienced strain and tension because of emerging or ongoing conflict. · Spatial fragmentation of predisplacement communities. DPs have been settled in groupings that do not necessarily corre- spond to their predisplacement communities. · Impacts of socioeconomic dislocation. Displacement shifts DPs into a new socioeconomic environment in which many pre- vious activities and structures that supported social capital are missing. · Distortion or disenfranchisement of DPs' citizenship rights. Dis- placement delinks DPs from their previous relationships with both local and national state institutions from their home area. The remainder of this chapter is divided into two parts. In the first part, we look at the factors influencing social cohesion within groups and communities of DPs; in the second, we examine the environment for links between DPs and local communities in the host society. In chapter 7, we will discuss a third type of social capital, what some have termed "linking capital." Linking capital comprises patterns of interaction between the displaced and public institutions, including those of host governments as well as international assistance agencies. In some cases, these agencies assume a number of functions typically associated with the governments of states. DPs, like many vulnerable groups, are not completely free actors with respect to the types of social networks they participate in. The en- vironment in which they make such decisions is sometimes highly cir- cumscribed, as when officials cordon off refugees in camp settings and physically or legally block them from interacting closely with local populations. When reviewing the environment for social capital of DPs globally, this important point should be kept in mind because such camplike settings form a significant part of the settlement pat- terns in some other regions. However, as noted earlier, this type of iso- lation does not characterize most displacement situations in Europe and Central Asia. Typically, DPs are physically integrated, to varying degrees, with host communities. Isolation and exclusion form a major part of the environment of DP settlement patterns, but these factors social capital 105 are nuanced and situational rather than the result of complete and absolute exclusion. There is little information available regarding patterns of collective action and social capital among the region's DPs. This is partly be- cause the displaced have not been a major focus of development agen- cies and related researchers who study social capital.2 It is also due to the great number of factors that go into forming the environment for the maintenance and formation of social capital among DPs, even within the same population. Thus there will be a wide range of pat- terns of social interactions and social cohesion within displaced communities and between DPs and local populations. In addition to highlighting the need for further work in this area, the chapter's pur- pose is to provide a series of diagnostic observations about the nature of social capital formation among DPs. These can inform policymak- ers and practitioners concerned with the macroissues affecting DP so- cial networks as well as those involved in planning community-based strategies that target particular DPs or include them among beneficia- ries of larger programs designed for local populations. Social Cohesion within DP Communities An understanding of social capital among DPs requires some knowl- edge of predisplacement patterns of social capital as well as an under- standing of the direct impacts of the process of displacement itself. The sections that follow will discuss both sets of factors.3 Predisplacement Social Capital The character and strength of predisplacement social networks among affected populations influence the patterns of social capital that mani- fest during displacement. Though it may seem obvious, this point harkens back to one of the study's overall objectives: identifying not only the general patterns of DP vulnerability but also the differences among DP populations in different country situations and even within the same displacement. Social networks do not survive displacement without distortions or deterioration. They do, however, represent the starting point for reviewing the social network patterns that develop during displacement and the role social capital plays in mitigating (or exacerbating) DP vulnerability. Five facets of preexisting social capital are especially important: · Levels of cohesion within social networks · Geographic spread of social networks 106 living in limbo · Presence of multiple identity groups · Impacts of conflict on social institutions · Resilience of social capital Levels of Cohesion within Social Networks It is not really possible to evaluate the "strength" or "weakness" of social networks along uni- form scales. Networks develop for specific purposes and in specific times. Some networks represent powerful cohesive institutional pat- terns but only in relation to one specific subject area. However, under- standing the relative cohesiveness of different communities and groups within a society prior to displacement still has some value. The previous chapter noted that the legacy of relatively high levels of economic development and extensive social services in preconflict U.S.S.R and Yugoslavia contributed to high levels of education among the region's DPs. The legacy of socialist states was not limited to ex- pansive social services. In many ways, the nature of state-society rela- tions and public policy in socialist states was antagonistic toward the formation of bonding capital within kinship networks and historical communities. It led to the development of somewhat formalistic struc- tures of bridging institutions (such as professional societies and youth groups). Both the displaced and local populations in the region (with the exception of those in Turkey) inherited this legacy. Thus commu- nity cohesion among this region's DPs was likely to be weaker, partic- ularly in areas formerly part of the Soviet Union, than among some displaced populations coming out of rural areas elsewhere in the world. This being said, many of the zones where conflict and displace- ment have occurred in the former Soviet Union were mountainous areas (Chechnya, Nagorno-Karabakh, Tajikistan), where relatively stronger local traditions and communities existed than in other areas of the U.S.S.R. Geographic Spread of Social Networks The existence of social net- works extending beyond conflict zones and national borders has great significance for the types of social networks that have proved possible during displacement. As noted previously, the presence of family and kin links, as well as broader ethnic, religious, and linguistic ties, in areas outside conflict zones has played an important role in DP sur- vival strategies in most of the region's situations. Such ties have been most significant where these links have been strongest and most recent. For example, social networks of Kurds in southeastern Turkey had expanded, through economic migration, to other parts of Turkey prior to the early 1990s. As IDPs fleeing violence moved to Turkish urban areas, the presence of these earlier, economic migrants clearly had an impact on the ability of Kurdish IDPs to develop survival social capital 107 strategies. Similarly, as discussed elsewhere, strong family ties among Serbs in Croatia with Serb families in northern Serbia greatly facili- tated vulnerability mitigation strategies when these populations fled into Serbia in the mid-1990s. That many of these refugees were dis- placed into the area surrounding the metropole of their once united country (Belgrade) meant that in addition to family, a wide circle of friends, former colleagues, and schoolmates could be drawn upon for support. Though DPs are poor and vulnerable, their situation without these networks, particularly in the cases mentioned above, likely would have been far worse, especially in the first years of displacement.4 Presence of Multiple Identity Groups Generally speaking, when con- flict divides societies along identity lines such as ethnicity and religion, the presence within a single community of groups from both sides seems likely to have detrimental effects on the resilience of such com- munities during displacement. This is because typically the act of dis- placement, with one group being pushed out and another staying or else choosing displacement in another area, physically splits the com- munity. Exceptions do exist, however, with mixed communities able to retain some degree of social cohesion even in the face of macrolevel ethnic conflict. The presence of mixed communities can act in a posi- tive way by keeping members of the community on both sides of the conflict. For example, some Georgian DPs from mixed communities in the Svaneti region in Abkhazia have noted that it was their Abkhaz neighbors who often protected their property and sometimes their per- sonal safety during the initial conflict and continued to maintain their land and houses after the Georgians were displaced.5 Where conflict divides communities along political lines instead of ethnic lines, such balancing acts are perhaps even more common as families strategically (or incidentally) place members of their network on both sides as an insurance to safeguard property and other assets. Impacts of Conflict on Social Institutions "Social capital," a World Bank report notes, "is not just the sum of the institutions which un- derpin a society--it is the glue that holds them together."6 In a period of conflict, this glue has dissolved and the parts it held together have shifted in a wide variety of ways. These days it has become common to speak of a "war economy" or a "conflict economy" to describe the na- ture of transformations and adaptations that occur in economic activ- ity within a conflict zone. There is also a "conflict society," with its own transformations with respect to the nature and variety of social capital. Conflict itself may have emerged from years of socioeconomic tension and dissipation of trust in state and civil institutions. So even 108 living in limbo DPs who left early in a conflict probably still suffered through a volatile period in which various types of social interaction were un- dermined or distorted.7 To the extent that DPs remained within their home area for significant periods of time during the conflict, violence, too, may have had a direct impact on the fabric of social institutions. On a very tangible level, leaders of communities may have been killed, imprisoned, or delegitimized, thus weakening community institutions. Communities, specific age groups, or professionals may have been traumatized and demoralized by violence and atrocities. More gener- ally, during conflict, social institutions at all levels are likely to deteri- orate or transform, and new, potentially competing, forms of social mobilization may emerge, ones linked to political groupings or com- batant organizations.8 Resilience of Social Capital Resilience is a difficult aspect of social networks to measure because it is a notion relative to the nature and magnitude of the shocks placed upon the network. Resilience is rela- tive to the nature of the impacts of conflict and displacement on social capital. It is not possible to completely define or evaluate the resilience of a social network except in the context of specific shocks and events. It is the nexus between the impacts of displacement itself (discussed below) and the predisplacement fabric of social capital of affected populations that defines social capital outcomes in displacement. Postdisplacement Social Cohesion This section reviews the major impacts the process of displacement has on social cohesion within populations who become displaced. These impacts include: · Spatial fragmentation of communities · Undermining the foundations of social cohesion · Dissolution of trust · Changes in leadership patterns · Formation of new social networks · Derivative impacts of poverty on social capital Spatial Fragmentation of Communities Displacement's most tangi- ble impact on social networks in the region is the physical separation of preexisting communities, extended families, and similar networks. When such networks, due to circumstances and degrees of conscious individual effort, can maintain close spatial proximity, this provides some base for perpetuation of the network. However, many factors inhibit these efforts. social capital 109 Global patterns of displacement demonstrate that DPs, like eco- nomic migrants and other groups, try to reform preexisting social net- works and settle together. Even though, as noted, this region's DPs tend to be concentrated in specific subnational areas, within those larger areas the microlevel settlement patterns frequently crosscut existing social networks such as communities, extended families, and identity- based groupings. This type of fragmentation does not always represent the dominant pattern in situations of displacement. Typically, when they settle in camplike environments in other regions, DPs can reform their communities of origin to varying degrees and settle together. Some examples of such spatial concentrations can be found in this region. In Georgia, the majority of DPs from the Galli region of Abkhazia, while not in camps, have concentrated in and around the city of Zugdidi in the west. In Azerbaijan, a survey in 2000 found more than 50 sites where the majority of the population came from one community of origin.9 The same types of examples can be cited in re- gard to other situations. However, generally speaking, the reformation of communities of displaced is much less prevalent in Europe and Cen- tral Asia than in other situations of displacement around the world. This fragmentation occurs for a variety of reasons, many of them discussed in chapter 4. By its nature, the process of displacement usu- ally is disruptive and occurs in a relatively short period of time, split- ting communities and often forcing individual households to fend for themselves. Even when the pace and timing of conflict allow commu- nities to leave in a group, much of the settlement process, particularly in this region, is "supply-driven" rather than "demand-driven." It emerges from the availability of existing infrastructure and land in dif- ferent parts of the host region or country. Where collective centers form an important part of settlement patterns, such as in the South Caucasus, the size of specific facilities is not always large enough to in- corporate an entire community. Thus communities become separated. In the Balkans as well as in Turkey, and to some extent in Georgia, the presence of relatives and kin in the settlement areas draws individual DP families, who often share the former's housing. In such cases, the displaced make seemingly rational choices regarding the availability of assistance, but this also results in their separation from other members of their predisplacement community. Given the high numbers of such shared household situations in these specific countries, this is a major factor in spatial separation of communities and even extended fami- lies. Under certain circumstances, some members of the community re- main in the area of displacement while other individuals or households are able to flee to third countries (such as Australia, Canada, and the 110 living in limbo United States) distant from the conflict. Physical separation does not preclude maintenance of social networks, particularly in the age of the Internet, satellite phone links, and international air travel. Such sepa- ration does, however, raise the complexity and cost of retaining such links. Undermining the Foundations of Social Cohesion Social capital does not exist in a vacuum from a given socioeconomic environment. The socioeconomic dislocation emerging out of displacement separates DPs from their ecosystem, workplaces, cultural and religious centers, and other foundations that anchor their daily lives. It threatens their sense of belonging and cohesion with members of their communities. Even when preexisting social networks can maintain close physical proximity in settlement patterns, many of the underpinnings of social networks, the "rhythms" of everyday predisplacement life, may be weakened or made redundant. For example, social networks in agri- cultural communities are in many ways tied to the mode of agricul- tural production. Forms of collective action and the organization of daily life are tied to the seasons. Social events and festivals also are typically linked to the agricultural cycle. When displacement alienates communities from this environment, the relevance of such foundations erodes in significance. In predisplacement urban environments, too, workplace social networks such as trade unions or informal social groupings of workers disappear with the factories and offices left behind in the area of origin. Their disappearance not only contributes to individual alienation, but also removes from the daily life of DPs the routines and opportunities for social interaction that contributed to the maintenance of networks. Dissolution of Trust In this region, given the wide-ranging upheavals that have affected many countries, it is not easy to discuss the issue of trust as a special factor affecting only the displaced. Anecdotal as well as statistical evidence suggests that trust is a commodity that suffered greatly in the massive disruptions of the transitional years of the 1990s, which occurred among local populations as well. However, even within the overall decline in trust brought about by the conflict in Bosnia, a recent study of social capital found the postwar decline in in- terpersonal trust to be more pronounced among IDPs. IDPs socialized less than the local populations with their relatives (31 percent versus 12 percent), friends (34 percent versus 16 percent), and neighbors of the same ethnic group (48 percent versus 32 percent).10 This pattern proved more pronounced with respect to other ethnic groups: 76 per- cent of IDPs, as opposed to 41 percent of locals, socialize with neigh- bors of a different ethnic group. social capital 111 Changes in Leadership Patterns The physical division of communi- ties discussed above also undermines leadership patterns. However, there are a wide range of other factors imposing themselves on previ- ous links between leaders and social networks. For example, through their inability to cope with the new environment, older leaders may lose respect of community members, particularly when their leader- ship was linked to possession of assets such as land or political power with state authorities in the area of origin.11 New sources of leadership can also arise under the differing condi- tions and stimulus of displacement. In some countries, for instance, "aid entrepreneurs" emerge among displaced populations, parlaying links with donors or host governments into local power bases. Where displacement coincides with active combat, such combat frequently can create new forms of leadership established through warfare or po- litical activity. This is often true for younger men, who may not have had the status or opportunity to rise to positions of authority prior to displacement but who now represent challenge to age-based leader- ship and other traditional norms. Formation of New Social Networks The fragmentation of predis- placement communities often has a parallel impact as it creates new forms of social identity emerging out of the designation "refugee" or "IDP." On a localized level, those displaced from different communi- ties in their places of origin are frequently pushed together into collec- tive centers, camps, or neighborhoods and develop into what might be called a community of interest. On a larger scale, the fact that DPs share common interests and concerns arising from their displacement can sometimes mobilize them around identities not part of their pre- displacement patterns of networks. Thus in Serbia and Montenegro, separate NGOs and various associations have been formed by refugees from Bosnia and Croatia. In Georgia, more than 200 IDP associations have been registered with the government. Many similar associations exist in Azerbaijan, based upon common districts of origin or on pro- fessional or educational criteria. While many groups form in name only, often in response to perceived potential for drawing donor funding, the shared dilemma of displacement can sometimes create a shared identity. Derivative Impacts of Poverty on Social Capital The high levels of poverty among the region's displaced populations also create issues for the maintenance of social networks. Hospitality is an important part of many cultures and serves to reinforce and maintain bonds of social cohesion. Crowded living spaces and low cash incomes make it difficult for DPs to extend hospitality to guests or, perhaps most 112 living in limbo significant, to commemorate key lifecycle events such as weddings and births in the manner previously considered the norm. When marriages require financial commitments such as dowry or bride-price, the poverty stemming from displacement becomes a matter of shame and may complicate family alliances. While poverty has the same impacts on other populations, those impacts are especially severe in situations of displacement for two key reasons. First, as discussed above, all of the other underpinnings of social capital are simultaneously weakened as well. Second, despite the presence of extended family and kin in many of the local populations where the displaced have settled, their primary family networks are usually DPs like themselves. In a normal environment, many poor populations often have broader networks to draw from, some of them well-off and able to volunteer or be pressured into providing financial support for occasions such as weddings and funerals. Displacement, however, has had, to a certain degree, a leveling effect, with some "rich relations" probably having had their incomes decimated at the same time. A related impact of poverty on the social cohesion of the displaced is what might be called the gradual "hollowing out" of DP communi- ties. Given the conclusions of this study regarding the chronic vulner- ability of the region's DPs and the lack of economic opportunity often characteristic of DP settlements, over time, some of the likeliest entre- preneurs and leaders of DP communities are probably gradually drawn away from the community by their own initiatives. This may be through educational or work opportunities in more fertile environ- ments within the host country or through emigration. Links to Host Societies The second area of social capital important to DPs relates to their links with social networks in host societies. As noted in previous chapters, the region's common displacement pattern has led to the settlement of DPs in areas with ethnic, linguistic, and cultural links. This is largely due to the region's predominance of IDPs rather than refugees as well as the particular history of state succession in the territories of the for- mer U.S.S.R and Yugoslavia. Consequently, the displaced have varying degrees of access to both ascriptive and voluntary social networks within host societies. More broadly, the familiarity of DPs with the cultural and linguistic context of their settlement environments has probably mitigated a significant amount of the alienation that would otherwise have been a factor in DP daily lives. social capital 113 The preexisting ascriptive social networks, particularly family links, available in many settlement areas have been a critical element in mit- igating vulnerability, particularly in the early years of displacement. Nowhere has this been more important than in Serbia and Montene- gro, where an IFRC/ICRC/YRC assessment found that 74 percent of IDPs and refugees had relatives in Serbia (16 percent in Montenegro).12 About two-thirds had received help from their relatives, especially with accommodations. About one-third continued to receive some form of assistance from relatives. We have cited this and other similar examples (Georgia and Turkey) of family links elsewhere in this study. While these factors mitigate social exclusion from the host popula- tion, they do not eliminate alienation. Again, as with factors influenc- ing levels of social cohesion within DP communities, a significant amount of heterogeneity invariably exists in the individual situations of DP households in regard to the intensity and quality of links with host populations. Some aspects of these links are homogeneous across a national level, while others are very specific to localized situations. Thus within the same country there may be many different patterns. Four other key factors help determine the nature of relations between local populations and DPs in a given situation.13 It is impor- tant to reiterate that these factors play out at the national level as well as, in different ways, in specific localities of settlement in affected countries: · Predisplacement intergroup dynamics · Size and placement of DP settlements · Socioeconomic environment of settlement area · Duration of displacement Predisplacement Intergroup Dynamics As noted, virtually all displaced populations in the region have settled among populations sharing a common language and culture. Even refugees are settled in countries that only 10 years before formed part of the same country as their own. Nonetheless, beneath these seem- ingly common links can be found specific subnational differences in culture and identity between DPs and host populations that make a difference in how they relate. For example, the majority of IDPs from Abkhazia, while Georgian, come from specific subgroups (Mingrelian and Svan) with readily identifiable dialects and accents. Local popula- tions in other parts of Georgia have stereotypical perceptions of these groups predating displacement. Serbs in Serbia have a predisplacement perception of many Serbs from Kosovo that ties to their view of Albanian Kosovars. Settlement of rural populations in urban areas has 114 living in limbo also represented a challenge to social capital. Perhaps the most egregious example concerns displaced Roma. In the Balkans countries, social exclusion within affected societies relates as much to patterns of Roma exclusion in preconflict societies as it does to their status as dis- placed persons. Size and Placement of DP Settlements The size of DP settlements in relation to that of local populations cre- ates different potential situations for social capital maintenance. While larger settlements, particularly those inhabited by DPs from the same place, can facilitate maintenance of social networks, such larger settle- ments can conversely make it more difficult for the displaced to build relations with the local population. It seems reasonable to assume that smaller clusters of DPs can, other things being equal, achieve closer levels of integration and build social networks with local populations. When DPs are settled in private accommodations or in shared accom- modations with host families, the likelihood of the growth of such links is greater. The placement of DP settlements matters as well. When DPs are separated from the resident population by settlement in an isolated area of a country or in concentrated camps, links with local social net- works may not be an option. As pointed out elsewhere in this study, in many displacement situations around the world, camp settlements are isolated to varying degrees from resident populations due to host gov- ernment policies meant to maintain separation and the logistical issues of finding available land. However, in most cases of European dis- placement, DPs are either intermingled with resident populations or reside in clusters close by. This is particularly true in the Balkans, where, as noted earlier, the majority of IDPs in Bosnia and over 90 per- cent of refugees and IDPs in Serbia are settled in private accommoda- tions. By contrast, in the South Caucasus, the higher number of collective centers also means relatively less opportunity overall for multifaceted interaction between DPs and local populations. The wide range of DP settlements in the region, even within one country, in terms of size and placement, indicates that generalities are unlikely to be valid and patterns of interaction between local popula- tions and DPs will be influenced heavily by local dynamics. Socioeconomic Environment of Settlement Area The socioeconomic environment of settlement areas affects relations in two ways: level of prosperity and stability in area of settlement, and the degree of similarity of the settlement environment to the home environment of DPs. In regard to the first, under conditions of scarcity social capital 115 and unemployment, alienation between local populations and DPs can be exacerbated by competition for limited resources. Such conditions can change over time, and a deterioration in the condition of local populations can, not surprisingly, lead to greater tension with DPs.14 In regard to the degree of similarity, when, as is frequently the case in the region, DPs from rural environments are settled in large urban areas, levels of alienation between the displaced and local populations are similar to the phenomena in which large-scale rural migration to urban areas occurs in other situations. Residents of cities such as Sarajevo and Baku, where rural displaced populations have replaced urban residents of another ethnicity, have in some cases nostalgically wished to have their urban neighbors back despite the history of conflict. Locals frequently mention stories of rural DPs keeping ani- mals in third-floor apartments, squatting on toilets, or similar tales to illustrate the social distance they feel between themselves and more recently arrived DPs. Duration of Displacement The overall perceptions of DPs among local populations vary a great deal from one situation to another and over time. In the beginning of displacement, local populations often viewed the displaced as heroes and martyrs for a national cause. Beyond the links of family and clan, some of the early hospitality extended to DPs related to this factor. However, with time, the perception of the displaced as a privileged class grows. This is a result of the deprivation of the local population and the view that DPs receive special treatment through the distribu- tion of humanitarian assistance from donor agencies or specialized government transfer systems. Factors that often exacerbate these feel- ings include the visible sale of humanitarian goods in open markets, the presence of DP subgroups with significant incomes or conspicuous consumption, and the growing frustration of general unemployment combined with nostalgic visions of past conditions. The presence of DPs in a certain locality can result in lower wages and higher rents. Choices in Social Networks In chapter 5, we discussed the special psychosocial characteristic of dis- placement that forces affected populations into a kind of limbo, in which they do not know whether they will be able to return home or settle permanently in the displacement area. When the impact of this limbo state is combined with the fact that most of the region's DPs are settled in areas where they have a degree of preexisting social links with 116 living in limbo local groups, some of the complexity of the environment for social cap- ital formation of the displaced begins to emerge. Displaced households must face decisions on whether to invest in local social networks pos- sessing a value in terms of employment opportunities, social inclusion, and contacts with state institutions or invest in the social networks of their communities of origin. They deal with this tension in the context of ascriptive bonding networks as well as voluntary bridging networks such as professional associations and business partnerships.15 In addition, the displaced's relations with local and national public institutions are often channeled into exclusive frameworks through specialized institutions, resulting in varying degrees of exclusion from the type of state-society interaction open to local populations. While it is possible to maintain both networks simultaneously, this dynamic creates tangible decisions for DPs. These include decisions such as with which families to make marital and business alliances, whom to invite to weddings, funerals, and other social functions (sim- ple yet financially costly matters), and what educational languages and cultural adaptations should be made. Decisions about the relative value of such social networks also influence residence locations. For example, in some countries, significant numbers of DPs have chosen to move close to friends or family members instead of remaining in geographic areas in which their immediate community members are displaced. A significant part of the parameters of how both DPs and local pop- ulations react to this universe of social relations relates to the actions of public institutions in setting the environment for patterns of social interaction. The dynamics of relations between DPs and the state is the subject of the next chapter. Conclusions Several conclusions can be drawn from the foregoing. First, social networks, whether ascriptive or voluntary, are an important factor in understanding the coping strategies of vulnerable populations. This is because such networks have the potential to provide informal social safety nets, allow for collective action to the benefit of members, and otherwise contribute to the well-being and overall survival strategies of households. These networks can also represent important links between households and public institutions such as local administrations. With respect to the displaced, when fixed assets have been destroyed or made unavailable and income streams are disrupted, an understanding of the potential role of social networks in reinforcing coping strategies seems to have even greater importance. social capital 117 Second, factors influencing social cohesion within displaced com- munities include both predisplacement patterns of social capital and the impact that displacement has on these patterns. With respect to the former, the following are important: · Levels of social cohesion in the predisplacement community · Existence of social links beyond the conflict zone · Presence of multiple identity groups in the predisplacement community · Impacts of conflict on social institutions · Resilience of the social capital to shocks With respect to the latter, significant factors include: · Spatial fragmentation of communities during displacement · Erosion of the foundations of social cohesion · Dissolution of trust · Changes in leadership patterns · Formation of new social networks · Impacts of displacement-induced poverty on social capital Finally, factors influencing links to social networks in the local pop- ulation are largely influenced by the fact that most of the region's DPs have settled in areas where they share ethnic, linguistic, and cultural ties with locals. While this mitigates the alienation of the displaced, es- pecially compared with areas where the local population is ethnically or linguistically diverse, a number of factors are important in deter- mining the nature of the links with locals: · Predisplacement relations between DPs and the local population · Size and location of DP settlements · Socioeconomic environment of the settlement area · Duration of displacement Notes 1. World Bank social capital Web site, http://www.worldbank.org/poverty/ scapital. 2. The 2002 Bosnia and Herzegovina: Local Level Institutions and Social Capital Study, focusing on the situation of IDPs in Bosnia and Herzegovina, represents a key exception to this statement and serves as a valuable resource. 3. Public policy decisions made by host governments and international assistance agencies play a pivotal role in the environment for social capital as well, and this will be addressed in detail in the next chapter. 118 living in limbo 4. Relatedly, one could speculate that a previous history of migration or dis- placement may have subtler positive effects on a group's ability to remain cohesive in a period of displacement. Examples might include Chechens who experienced massive displacement during the Stalinist period or Roma whose current pattern of life and traditions incorporate migration patterns. 5. Personal communication in Tbilisi with IDPs settled in the Immereti re- gion of Georgia in mid-July 2002. 6. World Bank social capital Web site, http://www.worldbank.org/poverty/ scapital. 7. It is perhaps important as well to note that the timing of displacement in the former Soviet Union coincided with a major shift in norms and be- havior in regard to the legitimacy of ascriptive and voluntary social net- works outside state control. Given their concurrent change in status, DPs were less able to adapt to this shift. 8. A breakdown of institutions at the national level could also strengthen social cohesion at the community level through the increased need for autarchy brought on by delinking from external institutions. 9. SIGMA, Social-Economic State of Refugee and Internally Displaced People in Baku. SIGMA (Baku, 2001). 10. World Bank, Bosnia and Herzegovina: Local Level Institutions, vol. 1. 11. This phenomenon may have mixed impacts, including some potential element of empowerment. Displacement, by removing people from old environments, may also remove some of the monopolistic shackles of power that frequently exist in small communities (the capturing of power nexuses by local elites to the detriment of others in the community). 12. IFRC/ICRC/YRC, FRY: IDPs and Refugees Living Conditions. 13. As already noted, an additional factor of major significance relates to how public institutions craft their approach to DPs and how various aspects of public policy set the parameters for interactions between local popula- tions and DPs. Those policy aspects range from the design and placement of settlements to decisions on how to distribute assistance to DPs to po- litical factors relating to the conflict that created displacement. This issue will be a core subject of chapter 7. 14. For example, the large-scale layoffs in state industries in central Serbia, where large clusters of IDPs from Kosovo have settled, are likely to create social tensions previously less evident. 15. Assistance agencies working with the displaced face much the same tension in deciding whether to support the internal cohesion of DP settle- ments or consciously work toward incorporating DPs into community- based programming targeted at local populations. chapter 7 States and Displaced Populations Summary: This chapter focuses on the relationship between states or governments and DPs. It begins with an analysis of the factors, espe- cially political, that influence state policies and responses toward DPs. Then it continues with a discussion of how legal frameworks affect, and how national and local institutions respond to, the displaced. Through- out the chapter, de jure and de facto obstacles to DP self-reliance and integration are identified. THE VULNERABILITY OF DISPLACED populations does not relate only to the loss of material assets or the psychological impacts of conflict and displacement. In many ways, it relates as well to a loss in status and identity as a citizen. Such a loss may appear, at first glance, to only affect refugees, who cross international boundaries and become "stateless." However, a review of the various situations of the region's IDPs and a comparison between the ways in which they now interact with the governments where they reside and the ways they did so pre- displacement reveal extensive patterns of change. IDPs have become, in many ways, a different class of citizens from their nondisplaced compatriots, even though they remain inside the borders of their own home countries. In some aspects, they have achieved benefits from this special status, but they have often, and we would argue predomi- nantly, suffered and been constrained in a wide range of ways. In the beginning of this study, we noted that the category of dis- placed persons included both refugees (those who had crossed an in- ternational border) and IDPs (those who remained displaced within their own countries). We also noted that, particularly given the recent history of the region in regard to state succession, the legal difference between the two was only significant (for the purpose of this study) if this had an impact on DP vulnerability. For most of this study, the difference has not had any uniform pattern of significance. In regard to relations with state authorities, however, logically there should be a difference. While there may be ethical as well as strategic reasons for 119 120 living in limbo host states to care for refugees, state authorities have no responsibility for refugees beyond those imbedded in the 1951 Convention on Refugees (if they are a signatory) and other international commit- ments. IDPs, however, are citizens of the country and, in principle, have the same rights as other citizens and remain a responsibility of the state. This chapter discusses the relationship between displaced popula- tions and public authorities. In doing so, it recognizes that the actions of states are not the only factors affecting the vulnerability of DPs. Many factors, such as the initial displacement (usually) and loss of assets, are beyond the state's control. The strategies DPs pursue to improve their circumstances in their present locations, as well as deci- sions on whether to stay or go home, are determined by individual households and emerge from a host of sector-specific issues discussed in previous chapters. This study would be incomplete, though, with- out some organized treatment of how, in a multitude of ways, state policies and actions (or inactions) affect displaced populations, help- ing to shape the environment in which they make such decisions. The actions and policies of state authorities in displacement situa- tions have three broad categories of impacts on DPs (both refugees and IDPs): · Tangible impacts, both direct and indirect, on settlement pat- terns, freedom of movement, and ability to work as well as on the availability of public resources and access to international assistance · Indirect impacts through the setting of parameters of interaction between DPs and local populations · Impacts on the shaping of the perceptions and self-image of DPs in relation to their displacement and their options for the future How a state responds to the displaced within its borders depends on several factors, including the perception of state leaders of their sense of responsibility to the displaced population, their financial and institutional capacity, and sometimes the accidental nexus between the needs or actions of DPs with the mandates of wider, unrelated state policies. Perhaps the most important determinants of state policy tend to be political in nature, relating to both the nature of the conflict that initiated the displacement and the domestic and international environ- ment in which decisions are taken. These considerations, in turn, dic- tate the ways in which public institutions interact with the displaced on both the national and local levels and shape the laws instituted by national and local governments. Several different patterns of states' policies toward the displaced seem to emerge from a review of the states and displaced populations 121 major displacement situations in the region. What most of these policies have in common is that they are designed to complicate, or inadvertently result in complicating (or both), the possibility of DP self-reliance and integration. Factors Influencing State Response Sense of Responsibility for DPs An important factor in how states respond to the presence of displaced populations within these borders relates to whether state leaders feel a sense of responsibility for the welfare of these populations. It would appear, on the surface, that the major determinant of the nature of this responsibility would be whether DPs are refugees (that is, citizens of another state) or IDPs, who are citizens of their state and for whom they accept, in some sense, a kind of long-term responsibility. How- ever, this distinction between refugee and IDP is somewhat muted in successor states to the former U.S.S.R and Yugoslavia (as discussed in chapter 1). Refugees often were previously citizens of the same states, and thus a greater responsibility for their welfare has typically been ev- idenced by receiving state authorities than one might otherwise expect. A second important factor influencing state response to DP situa- tions concerns the financial and institutional capacity of the state to shoulder the burden of providing for the displaced. In most cases, at least in the early years of displacement, international assistance is available to subsidize or even take over much of this burden.1 Relat- edly, prior experience in the country with conflict-induced displace- ment or large-scale migration also serves to shape state policy in terms of exhaustion of resources,2 institutional capacity, and lessons learned.3 Finally, some aspects of state interaction with the displaced may not emerge as much from the state's capacity or intentions as from specific facets of DP situations that put groups of DPs almost incidentally into the path of larger state policies. Subnational concentration patterns of the displaced also mean that development strategies focusing on spe- cific areas of a country where DPs are concentrated will have dispro- portionately high implications for the displaced as a whole.4 Similarly, when DPs as a group are dependent upon economic activities in specific sectors, state actions can have an unintentional and dispro- portionate effect. One example of this is the move for urban renewal that has caught fire in many of the region's larger cities. The significant presence of DPs in gray market professions, such as unlicensed street 122 living in limbo vending signifies the likelihood that removing vendors from over- crowded streets may frequently have disproportionate effects on DPs. Political Factors While all the above factors influence state policies toward DPs, the dominant factor tends to relate to the political position of the govern- ment in relation to the conflict that created the displacement. Are authorities neutral or partisan?5 If partisan, are they allied or sympa- thetic to the DPs or alternatively aligned or sympathetic with the forces who created the displacement in the first place? Even where states are not openly partisan in regard to conflicts, the actions of pub- lic authorities toward DPs are rarely predicated solely on a humani- tarian basis. While state responses to refugee and IDP populations share some underlying dynamics, there are also some differences that justify treating the two categories separately. Refugees In the long run, few state authorities wish for refugees from another country to remain permanently on their territory. Conse- quently, they do not wish to offer refugees living conditions that compare too favorably with their home country nor open extensive op- portunities to integrate. In the shorter run, at a minimum, states often have legitimate concerns regarding the implications of a spillover of the conflict into their territories, either directly or indirectly. Georgian policies limiting the mobility of Chechen refugees in the Pankisi Gorge is one illustration. As noted in several places in this study, most of the region's cases of cross-border displacement have resulted in refugee flows into an area where the population shares ethnic, linguistic, or re- ligious ties. In many of these situations, the local population represents a minority in the receiving country. The flow of Kosovar Albanians into Macedonia in 1999 offers a good example of how a government's concerns regarding the disposition of refugees can be influenced by a fear that their presence will contribute to domestic instability. What- ever the role of the Kosovar refugees in the conflict that later occurred in Macedonia, the links between conflicts in Kosovo, southern Serbia, and Macedonia during 2000­2002 are difficult to ignore. States also typically have concerns regarding the socioeconomic implications of refugees, including demands for social infrastructure, competition for employment, and increases in crime. This is particu- larly true when refugees are present in relatively large numbers. These factors lead to decisions to limit the freedom of movement of refugees, minimize their capability to interact with the host population, and restrict access to jobs and educational opportunities in the host coun- try. Rarely are such efforts to "insulate" a host society from refugees states and displaced populations 123 completely successful. However, government restrictions have impor- tant implications for DP survival strategies, through such actions as situating refugee settlements far from population centers and limiting eligibility criteria for refugees to work in formal sector occupations or pursue educational opportunities in public institutions. IDPs While state efforts to limit the activities of refugees are, in a cer- tain light, understandable from a national security and self-interest perspective, it is important to understand that state actions toward IDPs are also predicated, to a significant degree, on political motivations. When IDPs are generated from a conflict between government forces and rebels within the same country, and when those rebels share ethnic links with the people displaced, the state may view the IDPs with varying degrees of suspicion. Agents of the state may have even contributed to the factors that led to the displacement in the first place or conversely may minimize the numbers of DPs who are formally reg- istered as displaced or even deny the very existence of conflict-induced displacement in an effort to portray an image of normalcy for the country.6 Where states correctly or incorrectly view DPs as politically linked to anti-state elements, special efforts to facilitate DP self- reliance in their place of settlement may be avoided. In such cases, states may also choose not to provide targeted assistance to IDPs from public resources, regardless of need, and may limit access by interna- tional agencies as well. In some cases, states may adopt measures that pressure IDPs to return home, whether they wish to do so or not. Even when national authorities see IDPs as allies of the state-- describing them publicly as heroes and martyrs for a national cause-- the eventual objective is to create the conditions for them to return home and regain their property and territory. Generally, in pursuit of this objective, states again frequently institute policy frameworks that limit the ability of displaced populations to achieve self-reliance. Self- reliance is often equated with local integration; local integration is perceived as creating conditions under which IDPs will never go home again.7 When IDPs come from a territory that remains in the hands of separatist forces, state authorities often see a widespread decision by IDPs to stay in the settlement area as weakening the nation's ability to gain international support in its effort to reclaim sovereign territory by diplomatic means.8 The reality is that despite the wide range of political motives un- derlying the policies of different states toward the displaced in the end, most motives lead to a common objective: avoiding measures that may encourage integration of DPs in situ. Ironically, the only two major exceptions to this pattern in the region are refugee situations in 124 living in limbo Armenia and Serbia, where DPs at least formally receive full encour- agement to integrate locally, including offers of citizenship.9 Public Institutions We look at three aspects of relations between DPs and public institu- tions: the legal framework for DPs within the host state and how it differs from that applying to local populations; the ways in which gov- ernments organize themselves to deal with DPs at the national level; and relations between DPs and local units of government. In all three areas, there are a range of de jure and de facto patterns of separation and exclusion that invariably create obstacles to DP self-reliance strategies. While such obstacles rarely block DPs completely from specific courses of action in their daily lives, these obstacles do affect transaction costs and thresholds of decision making and contribute to a de facto environment of instability. Legal Frameworks All IDPs in the region are legally recognized as citizens of their home societies. Refugees naturally are not, although, as noted above, in two of the major refugee populations (Serbia, Armenia), state authorities have extended the offer of citizenship to refugees.10 Underneath this umbrella of legal equality, IDPs face many obstacles in accessing the same rights as the local population.11 Documentation The displaced, particularly IDPs, may formally have the same rights as other residents of a country, but conflict and conse- quent displacement may have resulted in the loss or destruction of key documents necessary for accessing such rights, including passports, national identity cards, and educational certificates. In some situa- tions, the public records of these documents also have been destroyed, or left behind in the area of origin and are thus inaccessible.12 If replacement of documentation is possible, doing so still represents an additional hurdle for DPs because of the potential transaction costs involved. Even more challenging for DPs is the practical difficulty of demonstrating to state officials, in the absence of documentation, that details of their predisplacement lives are accurate. The complexity involved in providing documentation is compounded when a state establishes parallel systems of administration for IDPs that differ from the normal structures of government (see below). states and displaced populations 125 Freedom of Movement Because, in some countries, DP settlements are situated in areas distant from population zones or areas with high unemployment, internal migration for work is an essential element of DP family income strategies. However, refugee movements within a host country are limited sometimes by de jure restrictions and fre- quently by the de facto difficulties of traveling without the permission of state authorities. IDPs also face complications in moving from place to place. Historically, in most of the region's countries, even citizens have been required to register with police or other civilian authorities in the locality where they are settled. In the former Soviet Union, this was known as the "propiska" system. Even when such require- ments have been formally done away with, registration practices retain a degree of force. As a report commissioned by the Council of Europe notes about one specific country (but with resonance for many others): Although the Constitution of Azerbaijan has officially abolished the propiska system, a number of laws continue to refer to it so that, in cer- tain regards, the propiska system remains in place. The resulting restric- tions on freedom of movement place particularly undue hardships on the displaced by limiting their ability officially to establish residence in areas, other than those to which they were initially assigned, where they may wish to migrate in search of better economic opportunities. The remnants of the propiska system still evident in Azerbaijan as well as in several other Commonwealth of Independent States (CIS) countries are inconsistent with the right to liberty of movement and freedom to choose one's residence enshrined in article 12 of the International Covenant on Civil and Political Rights . . . Access to social rights and employment is to a large extent preconditioned by registration. More- over, the registration stamp is entered into the identification card which has replaced former passports.13 A recent study on the status of Chechen IDPs in Russia concluded that the realization of the social rights (of IDPs) in Russia to a great extent de- pends upon whether people are registered in places of their habitation (registration) without which it is next to impossible to find employment, enroll in educational establishments, register with an outpatient clinic or get pensions and child allowances.14 While the various surveys reviewed for this study as well as the bur- geoning populations of IDPs in urban centers across the region testify to the fact that such registration requirements do not prevent internal migration, those requirements do create obstacles and transaction costs. They may have practical implications for DPs with respect to 126 living in limbo accessing health and education resources and put the displaced in an ambiguous position in interactions with police and other state offi- cials. International humanitarian assistance is often distributed ac- cording to official registration lists provided by state authorities. When international agencies provide assistance to replace or supple- ment public transfers to DPs, a choice to move from one place to an- other in search of employment may force the displaced to forgo such assistance or return periodically to their original settlement places to receive it. These obstacles to changing residence and the deceptive gymnastics the displaced must consequently practice are among the reasons national statistics show over or under counts of DPs. Right to Own Land The issues surrounding DP rights to purchase land in host countries were discussed in chapters 3 and 4. The issue of land ownership offers a good example of how de jure rights may exist but de facto realities of state policy may force IDPs, and sometimes refugees, into difficult tradeoffs. The national legal frameworks pertaining to IDPs in Azerbaijan, Georgia, and Serbia and Montenegro share certain commonalities that create de facto, if not de jure, restrictions on IDP choices. In all these countries, while IDPs are not formally precluded from such actions as purchasing land or taking other steps to formally register as inhabi- tants of their present localities, to do so seems to require that they give up their IDP status. This in turn removes any entitlement to identity- based assistance from government sources or international agencies.15 While, as noted elsewhere in this study, international humanitarian as- sistance is on the decline, IDPs nonetheless may be influenced by per- ceptions that such aid could once again be available. In Georgia and Azerbaijan, the levels of public assistance, exemptions, and other such benefits for IDPs remain significant elements of household incomes, so giving up IDP status would mean giving up tangible resources.16 As noted in chapter 3, giving up their displaced status is also per- ceived by some IDPs as giving up the right to go home if conditions permit and, just as important, giving up the right to reclaim property. While the national laws of these countries, particularly in the Cau- casian nations, do not bear out this assumption, the perception is not uncommon among IDPs. Given that any return home in post-Soviet societies would also include participation in the privatization and land distribution processes, this view is, perhaps, understandable. In Armenia, for example, the majority of refugees from Azerbaijan have not taken citizenship, despite the availability of this option since 1999. Many believe that such a step would have a detrimental effect on their ability to claim compensation for their property in Azerbaijan. They states and displaced populations 127 also believe that taking citizenship would preclude their attempting to claim political asylum or resettlement in a third country. This is despite the fact that there exists little evidence that other countries would be willing to accept such claims after their long residence as refugees in Armenia and the extension of citizenship rights to them by the government of that country. National Level Institutions At the national level, most host governments in the region tend to ded- icate a specific ministry or department to dealing with IDP and refugee matters. The initial impetus for establishing such an institution often comes from the influx of international humanitarian assistance at the onset of a displacement situation. While the creation of a dedicated core in national institutions has served an important purpose with respect to helping coordinate assistance, pulling together statistical data, and providing a focal point for addressing specific issues, it has also served to compartmentalize DP issues. It can contribute to the displaced's being treated as a vulnerable group outside the debates on national development policy and social safety nets. The incorporation of IDP issues into national development plan- ning is a slow process, one that has been occurring in many countries over the past five years as international assistance for DPs had de- clined. The initiation of the Poverty Reduction Strategy Paper (PRSP) as a central strategy document in the design and implementation of national development planning has represented a major impetus for a shift in the integration of DP issues into national public policy. The five major host countries in the region eligible for International Devel- opment Association (IDA) funding (Armenia, Azerbaijan, Bosnia and Herzegovina, Georgia, and Serbia and Montenegro) have all, to vary- ing degrees, incorporated DP issues into their treatment of vulnerabil- ity in their PRSPs. This shift is important in that it begins to provide a forum for the discussion of DP issues, not in isolation, but in the context of overall issues of vulnerability and development policy in the country as a whole and in a more open debate on prioritization of public expenditure. In addition, the PRSP process helps to encourage transparency in DP policy. Discussing issues of DP vulnerability and self-reliance in such a context, perhaps for the first time in many coun- tries, brings the DP issue into the context of economics and poverty instead of politics and war. Finally, but not insignificantly, the partici- pation of IDPs and refugees in the PRSP process also represents one of the few activities in which such groups are provided with a voice in public affairs. 128 living in limbo While conditions and the strategies being pursued vary from coun- try to country, windows for sustainable strategies for some groups of DPs have begun to emerge.17 Increasingly, as displacements linger on and underlying conflicts remain unresolved, shifts in policy are devel- oping. While this is not true in every case in the region, in many coun- tries governments have begun to explore broader options for enhanc- ing the self-reliance of DPs that would not have been possible a few years ago. The political sensitivity of displacement remains a factor. However, increasingly governments are searching for ideas and fund- ing assistance to support them in this effort. Local Level Institutions Local government represents the nexus of authority, where state insti- tutions link to society on a day-to-day level. In chapter 2, during our discussion of DP employment strategies, we noted that a minority of IDPs in three situations (Azeris, Georgians, Kosovar Serbs) had been able to maintain a percentage, or all, of their salaries from state em- ployment in their predisplacement localities. We did not discuss in that context the larger implications of that observation. Just as states view refugees, regardless of where they live, as citizens of their countries of origin, in several cases in this region, states have created situations in which IDPs remain citizens, in effect, of the localities from which they were displaced. In four countries (Azerbaijan, Bosnia, Georgia, and Serbia and Montenegro), in what is perhaps the clearest example of DPs' being caught between two worlds, local-level administrative institutions from the areas of origin of major IDP groups have continued to play a role in their lives. The administrative cadre of the previous municipal- ities and raions of origin, ranging from the mayors from Kosovo to the heads of the Ex-Coms (executive committees) from each occupied dis- trict or raion in Azerbaijan, have been retained in public service, along with their staffs, and established in an office somewhere under the government's control.18 In Bosnia, "municipalities in exile" were established in places of asylum when minorities fled majority areas during the war. While these were formally abolished in 1997 (two years after the signing of the peace accord), in practice, some aspects of this system still exist. As one recent study concludes: Officially, "municipalities in exile" created during the war have been abolished. But IDPs and minority returnees are still linked to parallel municipal institutions of some kind--departments in the ministries in charge of IDPs, state-subsidized IDPs' associations, or informal net- works of municipal councilors with the same ethnic background.19 states and displaced populations 129 In Georgia, the entire apparatus of a government in exile--including 97 separate public agencies, ministries, police, schools, hospitals, and similar institutions--has been transferred from the territory of Abkhazia along with thousands of employees. Local officials in exile have typically been charged with many of the same responsibilities for their former constituent populations. If the former populations of a locality in the area of origin were all clustered in one area, this arrangement could arguably be viewed as a valuable strategy for retaining measures of predisplacement social capital and preparing the ground, institutionally, for a return. In each of these cases, there are examples of such clusters where the local-authorities- in-exile have established offices close to their constituents or assigned staff members to interact with specific clusters of former residents in different places in the settlement area. However, as we have noted in earlier chapters, most of the region's DPs, while typically concentrated in specific areas of a country, have not necessarily settled in commu- nity clusters that correspond to their home areas. Functions of local government such as issuance of identity docu- ments, birth and death certificates, and licenses, are often the respon- sibility of these parallel administrations, and IDPs cannot have these functions performed by local officials in the areas in which they now live. An IDP in Belgrade, for example, may have to apply to an official based in a southern Serbian town to replace a lost identity document. Yet, at the same time, for other functions of local government such as police and payment of taxes, the IDP would be responsible to the ad- ministration in Belgrade. Similar types of arrangements are present in Azerbaijan and Georgia. In practice, what would under normal condi- tions represent a geographically based administrative relationship be- tween state and society has for IDPs become something of a "virtual" system of local government where an IDP's identity in relation to local authorities is segmented between a range of institutions, some of which are geographically located where the IDP has settled and some of which may be far away. In some countries, this split between place of residence and local state institutions extends to electoral politics as well. Refugees who have not taken citizenship in Serbia and Armenia cannot vote. In Georgia, for the first eight years of displacement, IDPs could not vote in local government elections but have had that right since 2000, when the law changed. In a number of countries, the enfranchisement of the displaced at the national level depends on the structure of the electoral system. When there are mixed proportional representation and geo- graphic systems, IDPs usually can vote only for candidates elected on the national level, not those in single member elections.20 This is the 130 living in limbo case, for example, for Chechen IDPs. Similarly in Georgia, until the law changed in 2000, IDPs could vote for national parliamentary races on a proportional list basis but not for single constituency races in the places they lived. In an arrangement that has not changed since the ini- tial departure from Abkhazia, eight deputies represent the exiled com- munity in the national parliament; these deputies, however, are not subject to election. In Azerbaijan, IDPs continue to vote for parlia- mentary seats based upon their districts of origin rather than their cur- rent places of residence. According to Bosnian electoral laws, a person born in a locality prior to 1991 is entitled to vote there but can also choose to vote in his or her current municipality of residence.21 Bosnians can choose where to vote, their current municipality or their municipality of origin. There are several implications of parallel tracks of administration and the widespread exclusion of IDPs from local government where they now live. First, in practice, such exclusion puts significant logisti- cal constraints on the displaced because, for certain administrative tasks, they must travel to find local government officials. The offices of those officials may be tens or even hundreds of miles from a IDP's res- idence. Second, such exclusion reinforces the sense that the displaced are different from others in the society. Third, and most important, such exclusion delinks IDPs from a range of personal contacts with local appointed and elected officials, with a whole host of implications for their daily survival strategies. This means that such an official does not view IDPs, even when they live in his or her area, as constituents and therefore does not feel accountable to them. Thus when the displaced do not vote in local elections, they have little access to locally elected representatives, who do not need their votes. This situation limits IDP access to the resources of local govern- ment, including patronage, and narrows the displaced's range of op- tions in terms of influencing local government. The relevance of this access is, of course, proportional to the power and resources of local authorities. When combined with IDP exclusion from local informal social networks, these governance dynamics also mean that informal patterns of approach to local authorities (such as police and taxation and licensing authorities) available to many local residents are inac- cessible or difficult for IDPs to tap. Though little detailed information on such dynamics exists, it is not unreasonable to think that such exclusion may affect IDP's abilities to access local jobs, register busi- nesses, or perform similar tasks and may raise transaction costs for a range of daily dealings with local administration. One other element of the relationship between DPs (both IDPs and refugees) and states needs some brief mention. This chapter has for the most part dealt with the impact state policies have on the coping states and displaced populations 131 strategies of DPs. It is difficult to ignore that there is a reverse rela- tionship as well. DPs may have an impact on a state and its con- stituent society. The continued lack of any sustainable resolution of the limbo in which the displaced find themselves can lead to distor- tions and pressures on the governance environment of a society in sev- eral ways. At a minimum, the displacement problem can be used, and has been used, as an excuse by some national governments to divert attention from other difficult domestic political and economic issues. The presence of refugees can also lead to an increase in attention to security or inadvertently draw states into conflicts in neighboring countries. More commonly, the presence of large numbers of IDPs, combined with an unresolved territorial conflict (which describes the situation in several host countries), can represent a complicated ele- ment in terms of national level electoral politics and, in the extreme, state policies toward a return to war. The potential impact of such effects are due not only to the lingering political problems that created the war, but also to the possibility that IDPs will represent a political pressure group that sees the only path to self-reliance as a reversal of fortune in regard to a previous military conflict. Conclusions From the foregoing several conclusions can be drawn. First, the fol- lowing factors seem to influence how states respond to DPs: · Whether the DPs are refugees (citizens of another state) or IDPs (citizens of the state itself) · The institutional and financial capacity of the state to respond · The political position of the government in regard to the conflict Regardless of their political position, most host states tend to avoid measures that facilitate the integration of DPs in the settlement area. Second, DPs face a number of obstacles with respect to exercising civil rights that local populations enjoy. These include the replacement of documentation lost or destroyed during the conflict, freedom of movement within the host country, and legal ownership of land. Third, while the creation of a specific department or ministry to deal with DP issues has proved useful in coordinating international assistance, maintaining statistics, and providing a focal point within the government, this process has tended to marginalize DP issues and exclude the displaced from national development planning debates. The PRSP process, however, is providing some opportunities to rectify this situation. 132 living in limbo Finally, parallel local administrations and the exclusion of DPs from local governmental institutions in their settlement areas can mean · logistical constraints if the local administration from their areas of origin is in another part of the country; · a sense among the displaced and the local population that DPs are "different"; · a sense among the local officials in the place of settlement that because DPs do not vote in local elections, they are not constituents. Notes 1. As was noted in chapter 2, it also must be recognized that much of the assistance to DPs in the region has come from neither state authorities nor international assistance but from civil society in the affected countries. 2. For example, it is often noted in Serbia that IDPs from Kosovo were less able to find places in public buildings used as collective centers because the limited capacity in such buildings was filled by the prior wave of refugees from Bosnia and Herzegovina and Croatia. 3. When states make commitments of public assistance to DPs, the value of such commitments must also be seen over time in the context of transi- tions in the role of the state in economic activity. The decreasing role of the state in providing employment through public enterprises reduces the value of commitments to provide employment for the displaced. Similarly, the increasing de facto privatization of health care makes commitments to DPs of free access to public health facilities less relevant. In addition, the privatization of public utilities limits the state's ability to fulfill commit- ments with respect to subsidized electricity for DPs. Finally, where signif- icant numbers of the displaced live in collective centers, such as in Georgia and Azerbaijan, the process of privatization of state-owned assets may create tensions between the new private owners and DPs and result in pressures for illegal eviction. 4. As already noted, IDPs from Kosovo in central Serbia, for example, find themselves in areas where massive restructuring of state-owned industries means that thousands of layoffs took place in 2003. 5. It seems fair to say, after a review of the region's displacement situations, that while states may (or may not) be neutral when it comes to conflicts in neighboring countries that generate refugees, there are few circumstances in which they are neutral about internal displacement because, in almost all cases, they are somehow a party to the conflict itself. states and displaced populations 133 6. For example, the government may do so by portraying conflict-induced displacement as part of a larger pattern of economic migration. 7. Whether "self-reliance" equates with "integration" and whether integra- tion itself truly means that DPs will never choose to return home if the opportunity arises are questions discussed in the concluding chapter. 8. It is sometimes noted that such objectives also lead to an informal strat- egy to, in effect, keep DPs vulnerable or at least to ensure that a certain degree of public vulnerability among some DPs is maintained in a visible way. The complicated dynamics of donor assistance strategies, including pressure exerted by public opinion in home countries, suggest that such a strategy is not entirely devoid of rationality. 9. In the case of Serbia and Montenegro, this shift in policy only became possible after a complete change in state leadership in 2000. 10. According to studies prepared for the Armenia PRSP, only 20,000 out of a total of more than 200,000 refugees in Armenia had, as of late 2001, availed themselves of the right, for reasons detailed in the previous chap- ter. In addition, the legal framework on citizenship in Montenegro, although it is formally part of the same federal state as Serbia, differs from that in Serbia. It requires 10 years of residency prior to the granting of cit- izenship, a provision that effectively blocks refugees from Bosnia and Herzegovina or Croatia from immediate access to these rights. 11. An additional obstacle to accessing rights may occur when the state is actively engaged in conflict and IDPs are, correctly or incorrectly, per- ceived as sympathetic to opposing forces. In such cases, IDPs, due to a real or perceived fear of interaction with officials of the state, may face the choice of attempting to achieve "invisibility" by not applying to authori- ties for public services or entitlements. They may gravitate toward forms of employment that minimize interaction with state officials and do not require licensing or other documentation. 12. Displacement and its impacts on access to assets and former places of employment may create similar dynamics that negatively affect DPs. For example, the UNOCHA office in Serbia notes that despite high rates of unemployment among IDPs from Kosovo, only 10 percent of IDPs report having successfully registered for unemployment benefits from the gov- ernment of Serbia. Many IDPs from Kosovo cannot apply for unemploy- ment benefits due to their inability to access proof of former employment in Serbia-based enterprises in Kosovo, as required by Serbian law. 13. COE (Council of Europe), Parliamentary Assembly, Committee on Migration, Refugees, and Demography, The Propiska System Applied to Migrants, Asylum Seekers, and Refugees in Council of Europe Member States: Effects and Remedies. Report (October 2001). On the Interna- tional Covenant on Civil Rights, see ECOSOC, Commission on Human Rights, Specific Groups and Individuals: Mass Exoduses and Displaced 134 living in limbo Persons. Addendum: Profiles in Displacement: Azerbaijan, report by Francis Deng, 55th sess., January 1999 (E/CN.4/1999/79/Add.1). 14. S. A. Gannushka, The Internally Displaced Persons from Chechnya in the Russian Federation (Moscow: Memorial Human Rights Center, Migration Rights Network, 2002). http://www.db.idpproject.org/Sites/ IdpProjectDb/idpSurvey.nsf/AllDocWeb/855B64D7DC1970BDC1256C5 B002E7BF5/$file/RF-GannushkinaEngl.pdf. 15. In addition, in Georgia, until 2000 when the law was changed, female local inhabitants who married IDPs legally became IDPs while female IDPs who married local inhabitants lost their IDP status. 16. In some cases, voluntary withdrawal by an individual of IDP status would also imply no longer having access to such resources. In Georgia, for ex- ample, governments have either retained individual IDPs on salary from state institutions from the area of origin or actually re-created these insti- tutions in exile and employed IDPs. 17. National authorities are often careful to distinguish between self-reliance for displaced populations and integration. In practice, the operational difference between the terms is sometimes hard to grasp. 18. In Azerbaijan, national demographic statistics are maintained for occu- pied raions in the same manner, formally at least, as they were prior to the conflict. 19. World Bank, Bosnia and Herzegovina: Local Level Institutions. 20. Bagshaw, Internally Displaced Persons and Political Participation. 21. Ibid. Bagshaw notes, however, one example of a municipality in the Republic of Serbia where Serbian IDPs were informed that to be eligible for humanitarian aid they had to demonstrate intention to vote in their current residences. chapter 8 Conclusions FROM ANALYSIS OF THE EVIDENCE reviewed in this study, it seems clear that, overall, populations remaining displaced throughout Europe and Central Asia do constitute a significant source of vulnerability in affected societies. The numbers of those who fall into this category are high enough to justify a significant concern both in regard to poverty reduction and in terms of the future resilience of affected societies. While available data are not uniformly reliable nor representative for all these societies, the patterns are consistent enough to lead to this conclusion. When multifaceted dimensions of vulnerability are re- viewed in conjunction, it is hard to escape the conclusion that the re- gion's DPs are a group deserving continued significant attention from government and donors. Contributing to the scope of the DP problem is the pressure the displaced put on public expenditure in certain countries, even through the marginal benefits provided by some host governments. The continued dependence of DPs also burdens host societies, both directly through pressure on social infrastructure and indirectly through a range of other impacts. As noted in chapter 7, another reason for working on strategies to address DP situations concerns the potential political volatility and risk of conflict that can be directly or indirectly catalyzed by large groups of the displaced excluded from opportunity and self-reliance. The available data are not sufficiently differentiated to separate out, in any meaningful way, the subgroups of DPs who have fared better or worse. However, there is sufficient anecdotal information to demon- strate that certain categories of the displaced have been able to man- age their situations better than others. We have cited examples of DPs who have continued to receive government salaries and pensions in displacement. We have also discussed those who have retained varying levels of access to resources either because these are mobile (animal herds, remittances, vehicles, professional qualifications) or because the productive assets of the area of origin have not been completely blocked. By dint of entrepreneurial spirit, energy, and talent, some individuals in any social group invariably overcome the most difficult 135 136 living in limbo situations and find a way to move forward. Yet it must be stressed that these subgroups are a minority. Their ability to cope rarely draws them beyond the dubious distinction of relative economic equality with the already difficult situation of local populations. Furthermore, DP coping strategies have a fragility--such as heavy reliance on gov- ernment subsidies, and free housing in rapidly deteriorating makeshift shelters or extended dependence upon shared housing with relatives and friends--that represents the most disturbing and frequently overlooked aspect of DP vulnerability. Given these factors, and the parallel indications that self-reliance through employment and access to land has been slow and uneven, we cannot escape the conclusion that without significant efforts by host governments and donors, DP situations across the region will worsen with time. Extensive patterns of social exclusion from host societies do con- tribute to the vulnerability of the region's displaced. Yet it is also true that some dimensions of relations with host societies--that is, the nature of social capital available to DPs--have nonetheless repre- sented an important qualifier on vulnerability. Many of the displaced have had access to social networks beyond their areas of origin, networks that have provided resources and connections to facilitate better self-reliance. Without the existence of broad family and rela- tional links in the areas of settlement, the condition of DPs in most of the region's countries would be even worse. The number of DPs in Georgia, Serbia, and Turkey who, even a decade after displacement, remain in joint living situations with host families, though a serious concern for stability, illustrates the enormous role these social net- works have played in mitigating some displacement impacts. Subsidies from national and international public resources to DP households, too, have played a critical role in mitigating DP vulnera- bility, forming a part of their survival strategies, particularly in the ear- liest years of displacement. In the context of wider scale deprivation in host societies in the early to mid-1990s, the targeting of the displaced for assistance may have privileged DPs, providing them with subsis- tence resources in short supply even to the general population.1 How- ever, the decline of humanitarian assistance in recent years for most of the region has diminished the importance of this mitigation strategy. And even in the best-case scenario, the presence of such assistance represents only a stopgap; no long-term reduction in vulnerability is achieved. The memory and mythology of these previous benefits, as well as anecdotal reports of some DPs who have achieved a certain self-reliance, have contributed to the perhaps misplaced perception that the situation of the displaced is, on the whole, less severe than that of other groups in the host societies, at least in the context of overall conclusions 137 vulnerability. Relatedly, in a number of countries, the presence of for- mal benefits, such as free access to social services, and other benefits in cash and kind further fuel such perceptions. However, the value of such benefits, coming from domestic public resources, depends greatly on the efficiency and capacity of public administrations to deliver. Though a full stocktaking was beyond this study's capacity, anecdotal evidence suggests that when DPs do receive cash payments, they are frequently delayed or subject to informal transaction costs.2 Benefits such as free health service are of less value in the face of the diminished capacity of public institutions and the de facto privatization of medical services in many countries. Of perhaps equal or greater importance than the general observa- tion of DPs' continued vulnerability is this study's attempts to sketch out the dynamics of the specific origins and manifestations of the vulnerabilities that emerge from long-term displacement. If DPs were identified as a group whose vulnerabilities were largely similar to those of other groups in the society affected by economic transition, special attention or focus would not be justified. The displaced's situation simply would contribute to a larger volume of vulnerability within affected societies. No special, targeted measures would be needed. However, while the displaced do suffer from the same factors as other groups in the region's societies (diminished access to social services, wide-scale unemployment or underemployment), there are specific characteristics of DP vulnerability that differ in quality and intensity. Among those are higher unemployment, loss of assets, substandard and unstable housing, and psychosocial factors. The conjunction of different disruptions that have affected the societies may sometimes obscure these underlying patterns of specific vulnerability within the larger impacts on all members of society. However, careful examina- tion of the DP situation draws out some of the major areas in which the displaced face specialized vulnerability. Specific Key Findings We will not repeat here all of the conclusions that emerge out of the analysis in this study in the context of the different sectoral treatments in the preceding chapters. Instead we will note, in summary, some of the most important findings and areas of specific DP vulnerability. Consumption Expenditures Consumption expenditures of the displaced populations in the region are, on the whole, lower than those of the local populations. Where 138 living in limbo this does not appear to be the case, such as in Azerbaijan, the differ- ence is not because DPs in those situations have higher levels of em- ployment or more private assets than other DPs. Rather, it is because of the decision of the Azerbaijan government to provide comparatively higher levels of public transfers to DPs to subsidize their living costs. Composition of Household Incomes The composition of incomes tends to be different for DP households than for local households. In all major displacement situations in the region, with the exception of the one in Turkey, the displaced are more dependent than local households upon public transfers, inter- national humanitarian assistance, or both. These sources make up a much higher percentage of DP household incomes than they do of local household incomes. For the most part, these sources of assistance are identity-based and do not differentiate between DPs based on com- parative vulnerability. (Nonetheless, the longer displacement lasts, the more likely vulnerability-based targeting is to emerge as a significant or dominant method of targeting assistance.) However, without these transfers, DP households would be much poorer, and many would probably be unable to survive. Even in Azerbaijan, where the dis- placed appear to have levels of consumption expenditures similar to those of local populations, without these special public transfers, DPs would have significantly higher levels of extreme poverty than do the local population. The general pattern of reduction in international hu- manitarian assistance to DPs in the region is thus a cause for concern. Unemployment Unemployment among DPs is higher. Overall, most DPs have not been able to replace the sources of livelihood they lost by being displaced from their home areas. Unemployment rates for the displaced are typically much higher than those for local populations, and DPs have little secure access to arable land and necessary inputs for income or consumption purposes. To some degree, the higher reported levels of unemployment result directly from this lack of secure access to cul- tivable land. However, given the high numbers of displaced in urban areas, this offers only a partial explanation. A small minority of DPs have retained their government employ- ment or at least continue to receive some element of their salaries dur- ing displacement. Generally, however, DPs are caught in lower paying informal economy employment, such as petty trade and ad hoc and sea- sonal employment, which does not maximize the use of their skills and training. Humanitarian assistance may be enough to make a critical conclusions 139 difference in household survival strategies, but in no situation we re- viewed was it of a sufficient magnitude to provide for such strategies without other sources of income. We have highlighted some of the spe- cific factors that appear to contribute to high levels of DP unemploy- ment in chapter 2, but more work is required on a country-by-country basis to understand the range of reasons for this situation. Despite high rates of unemployment, earned income remains an important part of survival strategies, highlighting the fact that DP households typically include able-bodied adults willing and capable of work. Existing Survival Strategies Existing survival strategies are fragile. Simple review of consumption expenditure only tells part of the story of patterns of vulnerability among DPs. Their overall vulnerability is deeper and multifaceted. In many cases, the availability of public transfers, humanitarian assis- tance, or both provides a relatively stable, albeit minimal, base for household incomes in the short run. However, these sources are only sufficient to offer marginal increments in daily survival strategies. The displaced lost most or all of their assets during displacement or imme- diately after through cannibalization of household assets. Conse- quently, there is little in the way of savings or liquid assets to cushion household crises or to invest in productive activities that could improve the situation. DPs in many countries have benefited greatly from support by extended family, clan, and ethnic links. Yet the reality is that in most cases the closest family and kin relations are also poor and displaced. Thus the extended social networks of the displaced have limited capacity to provide safety nets in situations of dire need. While in some countries DPs have limited, specific access to avenues of state employment, to a large extent their access is restricted by a variety of situational factors, such as loss of documentation and inability to access informal networks of patronage. Housing Situation and Vulnerability Perhaps the greatest difference between local populations and DPs lies in the area of housing. In the affected countries, the majority of local households own their housing as a result of the large-scale privatiza- tion of public assets that occurred in the early to mid-1990s. Prior to displacement, DPs, too, typically owned or had social rights to their housing. But this housing, like any agricultural land to which they once had access, is now largely or completely inaccessible because they have been displaced to another area or country. 140 living in limbo While a significant minority of the region's DPs have found rental accommodations, purchased housing, or received secure title to hous- ing, the majority, perhaps 50­60 percent, live in one of three categories of unconventional housing: collective centers, shared accommodations with host families, or illegally occupied housing. Each category, in its own way, contributes to overall DP vulnerability. Collective centers create psychological and social pressures, as well as physical health issues, through overcrowding and the increasing deterioration of phys- ical infrastructure never intended for such intensive human occupa- tion. Shared accommodations bring social tensions and, in many cases, contribute to the vulnerability of host families as well. The reliance of DPs on all these forms of long-term "temporary" shelter, especially shared and illegal accommodations, highlights the volatility of DP situations and consequently the fragility and flux of their survival strategies. The fact that a majority of DPs live in such "free" or low-cost accommodations also highlights a critical factor in accurate diagnos- tics of DP vulnerability and the potential fragility of DP household survival strategies. If DP households had to move out of such accom- modations, they would have to pay rent under market conditions, an expense, in most of the surveys we reviewed, not reflected in current estimates of their poverty. In Serbia, recent analysis shows that DPs in rental accommodations are significantly more vulnerable than those living with host families, likely because of the additional cost of rent.3 This is an important observation because it undermines the wide- spread perception that the presence of large numbers of DPs in private accommodations in some countries represents a proxy indicator of self-reliance. While this is certainly the case for some of the displaced, generalization of this proxy must be treated with caution. Without significant increases in public transfers and humanitarian aid, an unlikely occurrence given current trends in assistance in the region, such rent could only be afforded with a massive expansion in current DP employment. Even if DPs were to meet or exceed levels of employ- ment and income for local populations, they would never be able to purchase housing without some sort of support. Human Capital Indicators It is also important to note that the only key areas of comparison in which the region's DPs appear to be more or less "on par" with local populations is in demographic profiles and in overall levels of health and education. This parity is largely due to three facts. First, conflicts in the region (with a few exceptions) did not produce high levels of conclusions 141 casualties among DPs. High levels would have created households with, for example, more female-headed families. Second, educational outcomes and health partially reflect the high levels of human capital among DPs before displacement. Third, the displaced are settled in countries where social service delivery systems, though deteriorated, still function, and DPs have access to them.4 Since the only available data on education relate to school enrollment rather than attendance or performance, these observations do not necessarily indicate full equality of DPs with local populations. However, to the extent that equal indicators of human capital are the reality, this observation further emphasizes the existence of other factors contributing to vulnerability. Psychosocial Factors The impact of psychosocial factors on the vulnerability of DPs is diffi- cult to assess.5 Furthermore, many of the effects popularly associated with displacement actually stem from conflict and violence rather than the act of displacement. In some cases, populations who remained behind and were not displaced may have experienced more trauma than those displaced early in a conflict. However, with the exception of Bosnia and Herzegovina and Chechnya, the majority of the region's displaced are settled in areas where the local population were not, to significant degrees, affected by the conflict that created the displace- ment. In such cases, the issue of conflict-related trauma, while not a direct result of the displacement itself, remains an issue associated disproportionately with DPs. In addition to conflict-related trauma, all the tensions, hopeless- ness, and feelings of dislocation emerging from the displacement itself, as well as the perpetuation of displacement, are psychological factors that inevitably do have a differential impact on the daily lives of all DPs. The study found limited data in some countries to tentatively support the view that displaced men may have shorter life spans than their local counterparts, even taking into account normal patterns of gender differential in mortality rates. One possible interpretation of this pattern is that for groups with traditional gender roles, displace- ment proves more disruptive to men than women. This is because displacement, while accentuating the difficulties of maintaining a household, completely dissolves the typical workplace environment for men and may represent to some of them a wholesale failure of their abilities to provide security and a minimum living standard for their families. This observation, if true, also emphasizes that gender issues in displacement are not solely, or even predominantly, "women's 142 living in limbo issues." Rather, displacement has gender-specific impacts that affect men and women in different ways. The specific psychosocial impact of displacement on different age cadres of DPs is an area that has deep, but difficult to define, impacts on the present and future situation of displaced populations. Although there is value in targeting analysis at age groups such as children, ado- lescents, youth, and adults, an additional generational factor must be taken into account among DPs. Displacement effectively divides a population into three cadres related to the age of different generations when the initial displacement took place. The first cadre comprises those who matured prior to displacement and became DPs as adults. The second comprises those who were adolescents when displacement began; the third, those who were young children when displaced or born into displacement. In the early years of displacement, these three categories coincide with age categories. As the years of displacement extend, however, the middle category (called a "lost generation" by some observers) is no longer defined by the "youth" category but ages while retaining the baggage of a generation caught between two worlds. Each age cadre has its own specific issues, and it is difficult to make a linear comparison of impacts on each of them. A further factor in DP vulnerability that differentiates the displaced from local populations relates to the psychology and the physical real- ity of displacement, which freezes DPs in a perpetual state of uncer- tainty regarding their futures. The displaced are, as the title of this study says, caught in "limbo" between two worlds. This state, often exacerbated by situation-specific patterns of displacement and reset- tlement, has created a range of material, social, and psychological impacts and patterns of alienation. The fact that DPs cannot predict whether they will remain where they are or return home at any given time has a broad range of effects on their ability to make decisions re- garding household planning. Those decisions may involve education, purchase of shelter or business property, or other investments in pro- ductive activities. This contributes to an overall environment promot- ing depression and undermining feelings of empowerment. Social Capital Following from this limbo state, DPs face an environment with com- plicated effects on their abilities to create and maintain social net- works. These come about partly through displacement's direct impacts and partly through the reality that being in limbo complicates house- hold decisions regarding integration into local social networks versus maintenance of networks from home areas. In effect, the displaced conclusions 143 must deal with dynamics of social inclusion and exclusion in the context of the social networks from their home communities as well as the local communities and social networks in their area of displace- ment. This tension does not exist to the same degree when DPs are concentrated in large camps--where they are de facto often com- pletely isolated from local populations. The majority of this region's DPs are settled in smaller clusters or in private accommodations with little physical distance from local populations. The process of displacement disrupts the underpinnings of social capital through the removal of people from their traditional socio- economic environment, patterns of social organization, and collective action. Thus it erodes the foundations of previous forms of leadership and creates new and competitive ones. Typically, settlement patterns in this region have also physically fragmented communities, spreading community members across broad geographic areas. In addition, there exists some anecdotal evidence that DPs, particularly those clus- tered in camplike settings, may develop a third identity--"IDP" or "Refugee"--and due to commonality of situation and sometimes com- mon interest mobilize to limited localized degrees around this identity. Relations between DPs and social networks of local populations show mixed dynamics. Most of the region's displaced have settled in areas where local populations speak the same language and share cul- tural, and often familial, ties. However, in several cases, most notably in Georgia, Serbia, and Turkey, it is clear that without the support in cash and kind to DPs from civil society in the early days of displace- ment, their situations would be much worse. The existence of ex- tended social networks through family, clan, and ethnic links in areas of settlement has played a major role in the survival strategies of DP households. Yet on a localized level, the displaced are excluded to varying degrees from local social networks, partly through the impact of DP settlement patterns, partly through existing tightly knit local clans, and partly through choice. Response of Host States Governments of countries with large numbers of DPs have rarely viewed displacement as solely an issue of vulnerability. Various factors come into play in government responses, including the institutional and financial capacity of public institutions, experience from previous situations of displacement or mass migration, and the DPs' citizenship status in the host country. However, the dominant factor in how a government responds to DPs appears to be its perspective on the con- flict that created the displacement. 144 living in limbo Most governments of the region's affected countries tend to be sym- pathetic to DPs. Authorities provide aid from public coffers or actively request international assistance, typically because they are either allied to parties to a conflict linked to DPs or else are parties to the conflict themselves. In such situations, governments want the displaced to go home but only under certain conditions. Usually the conditions involve the return to the government of territory currently being occupied by separatist forces. Some governments wish DPs to return home unconditionally as soon as possible and may attempt to encourage this through various actions. This is either because the government is an active combatant itself or, when displacement emerges from a neighboring state, because it fears potential spillover effects of the conflict and the displacement within its own society. In such cases, the state is unlikely to offer ap- preciable material assistance to the displaced and may only begrudg- ingly accept--or actually decline--international assistance for them. What these types of situations have in common is a resistance by governments to local integration of DPs. Often this translates into a series of de facto and de jure impediments to self-reliance and a rein- forcement of the limbo situation of DPs instead of a mitigation of dis- placement's impacts. Such impediments include obstacles to freedom of movement and to purchase of land and disenfranchisement or dis- tortion of voting rights in national and local elections. Such strategies may be understandable, to a degree, in regard to refugees. But in this region, such impediments are often directed as well at IDPs, citizens of the country in which they are settled. In several situations involving territorial disputes and separatist movements, states have supported the reestablishment of various gov- ernment institutions, particularly local units, from the disputed area elsewhere on their territory. The state then requires DPs (typically IDPs) to continue to register with these units for entitlements of bene- fits and applications for documents such as passports, national iden- tity cards, and driving licenses, regardless of where such DPs live within the country. In effect, such a situation creates elements of a state within a state, with parallel channels for a wide variety of administrative functions. The displaced may live in one locality and yet be required to be a member of another "virtual" locality, further reinforcing their sense of living between two worlds. The Political Will to Depoliticize Displacement While the description of the patterns of state response to DP situations detailed above remains generally accurate, the initial impetus for this conclusions 145 study was a recognition that situations are slowly changing in many of the region's countries. Increasingly, as displacements linger on and un- derlying conflicts remain unresolved, government policies are gradu- ally shifting. The shifting is in part due to the significant decline, in most of the region's DP situations, in the availability of international resources to support humanitarian assistance for long-term displaced populations. In many countries, governments have begun to explore broader options for enhancing DP self-reliance, options that would not have been possible a few years ago. The evolution of this environment raises another, larger policy question for donors, and perhaps it is fair to end this discussion with this question. A central premise of many of the region's government policies has been that mainstreaming DP assistance and facilitating self-reliance would ensure that displaced populations never return home, an outcome viewed, for various reasons, as undesirable. Inter- national agencies, for their part, have tacitly accepted this premise or at least facilitated its power through the provision of humanitarian as- sistance. This premise has rarely been questioned. Perhaps it should be. Much of the complexity of the vulnerability of DPs is only resolu- ble through the expenditure of significant financial resources, re- sources not often available from either hard-pressed host governments or the donor community. However, long years of humanitarian assis- tance, though now of a lesser magnitude, have also had a cost. There is an argument that words such as "integration" must be avoided at all costs. Many host governments, even when the displaced are their own citizens, either de jure or de facto avoid promotion of financial, legal, or policy measures that promote integration. The case for this, in terms of national policy, is clearer when it comes to refugees than IDPs. Ironically, the only major efforts at integration in the region have been in refugee situations (Serbia, Armenia). Even if one accepts the premise that should the opportunity emerge, it is essential for DPs to go home, where is the underlying evidence that promotion of self- reliance, to the point of integration, blocks them from doing so? It seems just as plausible that increasing DP income and stability stands a better chance of creating an enabling environment to limit vulnera- bility upon any eventual return home than keeping such populations in an artificial limbo. In many of the region's countries, the impacts on the environment in DP home areas, including the massive destruction, landmine pollu- tion, and hostile social environments engendered by ethnic cleansing, as well as property disputes ensure that some displaced populations will not go home. If they do return, it will not be to sustainable liveli- hoods for, at a minimum, several years, even if conflict is resolved. To put this in some perspective, while there have been impressive shifts 146 living in limbo in the ability of DPs from the wars in Bosnia and Herzegovina and Croatia to return to their homes in the Balkans, these returns only began to become appreciable four years after the Dayton Peace Ac- cords ended the conflict. People who returned were displaced for more years after peace came than they had been during the war itself. Soon the large residual displaced populations in Bosnia and Herzegovina and Serbia and Montenegro will have been displaced twice as long after the war as they were during active hostilities. Kosovo was not the norm. Returns home take time. Even if peace does come, the possibility that all DPs from a certain area, if given a choice, will ever go home is extremely unlikely. This will probably be true for appreciable numbers no matter what steps governments take now. First, the factors mentioned above in many of the areas of origin will limit the carrying capacity of former conflict zones. They may no longer be capable of supporting their original populations. Second, long-term displacements result in natural demo- graphic increases in population, making a full return of population to some zones difficult to support even when previous property and other resources are recovered intact. Third, several years of displacement, particularly in an environment extremely different from that of the home area, may lead to a decline in skills appropriate for the home en- vironment as well as a lack of desire to shift residence once again. Even if DPs are unable to fully integrate or achieve a comfortable living standard in their places of displacement, it is nonetheless the place, if displacement lasts long enough, where children have grown up, married, and borne their own children. Yet there are many people who simply want to go home whatever the cost. The pull of a homeland is difficult to resist, a truth attested to by the Afghans' returning home now after 23 years or the Rwandan Tutsis who came home (with an army) after more than three decades of displacement. Many factors influence a return home, once an opportunity arises. The seeming assumption by governments that creating obstacles to integration of DPs somehow increases their eventual willingness to re- turn home is open to question. Integration, through support of possi- bilities for the displaced to develop skills and acquire savings and other property, could arguably facilitate return by allowing DPs to amass resources that would reduce their vulnerability and provide in- vestment capital for private reconstruction of housing. Such resources could be used to jumpstart economic activities in areas of origin. Aside from human rights considerations, especially in the case of IDPs, the degree to which government policies contribute to the continued DP vulnerability and poverty also may be counterproductive to stated government goals (eventual return home that is sustainable and conclusions 147 cost­effective). Some governments have, after long periods of different policies, recognized this reality. They have taken preliminary steps to- ward removing impediments to DP self-reliance. To do so, they require technical and financial support from donors as well as encouragement to continue further along such paths. Notes 1. It is also probably true that the influx of humanitarian aid provided some marginal benefits to host populations, at least in the short term, through derivative levels of increased short-term economic activity and, in some places (Georgia, northern Albania), direct payments in cash or kind to host families. 2. It is true that delays in payments of benefits and salaries from many of the region's governments affect a wider section of the population (pensioners, government employees) than just DPs. It is also true that DP payments in countries such as Azerbaijan and Georgia are often delivered with, per- haps, more regularity than some public transfers to other groups. How- ever, the purpose of the observation in the text is simply to note that the assumed addition to DP survival strategies of such public transfers is sub- ject to some caveats. 3. It may also be true that, in some cases, living in shared host accommoda- tions may be a proxy for strength of social capital as well. Specifically, those who remain in host accommodations have stronger links to local social networks, allowing them access to such extended hospitality. 4. Two groups for which there are few good data on these observations are DPs within Chechnya (as opposed to those in neighboring Ingushetia) and in Turkey. Factors in both cases may suggest that these conclusions do not hold to the same degree. 5. The text of this study also discusses other psychological impacts, such as trauma from conflict and potential dependency syndrome from reliance on humanitarian assistance. ANNEX 1 Development Programming and Conflict-Induced Displacement THE PARAMETERS OF ACTION REGARDING DP programming are often circumscribed by political dynamics, sustainability, and security of access, as well as by the levels and nature of domestic and external donor resources. Any recommendations must take into account these parameters, which frequently limit what otherwise would appear to be logical courses of action. This study provides recommendations in five areas: · Poverty diagnostics and tools · Inclusion of DPs in social safety nets · Delivery of social services · CDD, municipal, and area-based approaches · Shelter and housing Poverty Diagnostics and Tools The nature of conflict-induced displacement warrants recognition as a specific type of vulnerability with its own characteristics. However, the decision on whether to dedicate specialized attention to DPs in a given country emerges from the juncture of factors discussed throughout this study and summarized in chapter 8. The following recommen- dations are not intended for all countries, or even for all countries with displacement, but rather for situations in which there is a recog- nition, or suspicion, that displacement has resulted in higher levels of vulnerability. Where displacement appears to represent a significant factor in poverty and vulnerability, diagnostic work, including poverty assess- ment and poverty monitoring, needs to incorporate special approaches 149 150 living in limbo to adequately reflect the situation of DPs. Some suggested guidelines for poverty diagnostic tools where displacement is a factor are listed below. Prepare baseline data as quickly as possible after initial displacement: Humanitarian and development agencies, in partnership, should start collecting baseline data on DP conditions early in the displacement. Without such data, it is often difficult to judge changes in these condi- tions, whether they be for the better or the worse. Most situations of displacement begin with an assumption that they will be temporary. However, the vast majority of large-scale displacements in Europe and Central Asia (as well as in the world) have lasted several years or longer. Therefore long-term planning is always warranted, even in the earliest days of displacement. Such baselines should include as much as can be reasonably known about the predisplacement situation of affected populations. Again, such information will enrich a baseline for making later comparisons. Incorporate the assessment of DP conditions into core household poverty monitoring exercises: When possible, evaluation of the situa- tion of DPs should be incorporated into ongoing poverty monitoring, particularly quarterly household monitoring, conducted for the coun- try as a whole. Most of the surveys and assessments reviewed in this study were prepared as one-time projects. As a result, they can com- plicate comparisons with local populations (comparisons necessary when making policy decisions on resource allocation) and lead to sea- sonal distortions in data. Furthermore, in many situations, the dis- placed have shifted accommodation over time, as when they migrate to urban centers in search of work. This dynamic highlights the im- portance of periodic monitoring of DP situations instead of reliance on one-time survey projects. In addition, specific qualitative assessments or dedicated surveys may be necessary to capture aspects of DP vulnerability not ade- quately addressed in quarterly household surveys. When such full integration of diagnostics proves unfeasible, close contact between government ministries and humanitarian and development agencies is warranted to ensure that methodologies and approaches of work are similar enough to allow for meaningful comparisons. Design sampling strategies within demographic and poverty diagnos- tic instruments to reliably capture DP populations: Sampling strategies within quarterly household surveys, censuses, and other demographic and poverty diagnostic instruments need to reliably capture DP development programming 151 populations. Subnational sampling or integration of booster samples of the displaced is typically necessary because the displaced tend to be settled in specific regions and localities within the country. Without such sampling, undercounting of DPs will result. Special care must be taken with development of sampling lists. The following factors should be kept in mind: · Household address lists from preconflict censuses and tax records are not likely to capture IDPs who have moved to their current locations since the conflict began. Refugees not in the country at all during a preconflict period may be missed entirely. · In many cases, sampling lists do not incorporate DPs settled in "unconventional" housing such as tented camps, public build- ings, railway cars, and abandoned buildings (factories, schools, clinics, hotels, spas). When high proportions of the displaced are settled in these situations, ignoring such housing can result in significant undercounting of DPs. · Significant proportions of DPs may be living in shared housing with host families. Sampling strategies as well as training mod- ules for interviewers should take account of such joint house- holds and make decisions on how they will be treated in the sampling framework. · Sampling strategies need to be conscious of the range of poten- tial distortions of total numbers of DPs in official statistics (increases or decreases) due to political considerations and other factors. Local political dynamics, for example, may give various actors an incentive to undercount or overcount DPs. · Formal lists of DPs based upon relief distribution or formal national registration exercises need to be approached carefully. Frequently, such lists have proven inaccurate or incomplete for a variety of localized reasons. For legal or policy reasons, for instance, formal registration lists may be inflated or not capture all categories of displaced. In some cases, registration of new DPs may be closed during specific periods, leaving sections of the populations missing from official registration lists. · Diagnostic exercises must be carefully delinked from respon- dents' perceptions that such exercises may be utilized to increase or decrease assistance benefits. Such perceptions can influence survey responses. Delinking may not always be possible, so policymakers must, when interpreting the results of exercises, recognize the potential for this type of distortion. 152 living in limbo When designing questionnaires, take into account considerations specific to measurement of DP vulnerability: The design of question- naires should take into account any special considerations that may be specific to the measurement of DP vulnerability. Such questionnaires should · provide detailed information regarding the composition of household income; · incorporate specific questions that provide information on the actual delivery of public transfers and humanitarian aid; · include a module for DPs that provides information on their pre- displacement profile. Composition of household income data should include separate breakdowns of such sources as international humanitarian aid, public transfers intended as DP payments, and public transfers not intended as DP payments. Payments not intended specifically for DPs mean ones available to the local population in general (such as pensions, veteran's benefits, and social safety net payments based upon general vulnerability criteria). Data on delivery of public transfer and human- itarian aid should include information on delays or partial payments of entitlements. The predisplacement profile for DPs should include details on ownership of fixed and movable assets, previous profession, and household income as well as demographic information such as predisplacement household size. In poverty diagnostics for refugee countries of origin, consider incor- porating surrounding refugee populations: When poverty diagnostics are done in countries of origin for refugees, due consideration should be given to incorporation of refugee populations settled in surround- ing countries. Such populations are citizens of the country and may in the future be able to return there. Regardless of their geographic loca- tion, they constitute a segment of the overall population. If such in- corporation is not practical, researchers should attempt to ensure that parallel exercises are undertaken with similar methodologies. Hence when host governments or humanitarian agencies conduct surveys in such refugee settlements, efforts should be made to design exercises to maximize the potential for comparability of data. Take into account the wider circles of vulnerability when assessing DP vulnerability: Assessment of vulnerability stemming from displace- ment needs to take account of the wider circles of vulnerability created by displacement movements. The impacts of large-scale displacement, while primarily related to the condition of those displaced, also relate to the potential multiplier effect of displacement. Diagnostics should development programming 153 be designed to explore the socioeconomic pressures displaced popula- tions put on host areas. Such pressures often include overburdening of social infrastructure, wage deflation particularly in unskilled and casual labor markets, and environmental impacts at a national level and a subnational level. Impacts that appear insignificant at a national- level review may be severe in the context of subnational regions where concentrations of DPs are settled. When conflicts result in the dis- placement of only parts of communities, the impacts of the displace- ment must include the vulnerability of communities of origin as well as that of areas of asylum. Take into account the "Janus-faced" nature of displacement-induced vulnerability: Any diagnostic approach to assessing DP vulnerability typically focuses on the conditions in the present circumstances in the place of displacement. To the extent that such analysis reviews future trends and potential shocks, it tends to do so in the context of contin- ued displacement. This study focused on this period of limbo and the condition of DPs during this period. However, it is important to stress that if opportunities for return home emerge, the nature and roots of vulnerability will change. While the fragilities and strengths of house- hold economies during displacement retain a level of importance, equally important is the condition of the home areas to which DPs re- turn. The condition includes levels of destruction and capacity for rapid economic recovery of the return area in general and of the private assets of individual displaced households. Relatedly, in situations in which displacement is porous and DPs retain some access to home areas dur- ing displacement, analysis of vulnerability also must take into account the contribution of such access to household survival strategies, most notably through agricultural production. Such an analysis requires some assessment of the transaction costs for the access and the compli- cations and risks associated with shifts in security and other factors that may threaten portions of household income associated with the access. Continue to identify former DPs after a return home: Diagnostics should continue to identify former DPs after the return home and should do so for several years. We do not know if vulnerability ends with repatriation. During the course of this study, it has become clear that there is very little analysis, within this region, or in other regions, of whether the vulnerability created by displacement is mitigated or erased when DPs return home. Given the many economic and social transformations that occur within a population during displacement and the length of many displacements, it does not necessarily follow that returning home will dissolve the impacts of the displacement period. It is also possible that formerly displaced populations may 154 living in limbo actually perform better than surrounding populations due to a cat- alytic energy imparted by displacement. Collecting this kind of data may mean specialized surveys or may mean simply ensuring that ques- tionnaires on affected countries include a question to identify which groups had been displaced previously and for how long. Integration of DPs into Social Safety Nets Survey results consulted for this work were not sufficiently detailed to identify distinct patterns of vulnerability in many DP subgroupings. This being said, it seems certain that not all DPs are equally poor or vulnerable. When DP-based assistance programs survive declines in humanitarian assistance, more focused targeting of such programs is not only advisable but perhaps inevitable. This increases the impor- tance of developing diagnostic strategies that help to disaggregate DPs and identify subgroups with specific patterns of vulnerability. In practice, we have found that international assistance to DPs fre- quently has declined due not so much to a carefully evaluated demand- side approach--that is, planned reduction based upon vulnerability assessment--as to a supply-side approach--that is, the diminishment of available international humanitarian assistance. This dynamic is gradually changing, and there are several examples of vulnerability assessments driving shifts in assistance policy toward DPs. Nonethe- less, even such assessments are often driven by an understanding of an impending reduction in the stock of available assistance by donors rather than by an unbiased attempt to evaluate comparative need of DPs for continued assistance. Two factors are important to recognize in countries where such a process moves forward. First, individual targeting strategies for social welfare benefits need to ensure that specialized vulnerabilities of DPs are not overlooked. It is likely that any shift in DP assistance from an identity-based or "rights-based" approach to a needs-based social wel- fare approach would result in more accurate targeting of vulnerability. Yet it is also important to base such a transition on an assessment of the institutional capacity of public institutions to accurately capture DP vulnerability within the context of wider social safety nets. There is little doubt that an identity-based approach invariably represents an inefficient targeting mechanism and allows some DPs less vulnerable than other segments of the local population to receive extra benefits. But in the absence of a proven capability of public institutions to make relatively accurate assessments of individual household vulnerability, utilization of an identity-based approach through the targeting of development programming 155 "displacement" as a proxy for vulnerability is not unreasonable. Under such conditions, the removal of identity-based targeting without a con- currently effective process of vulnerability or needs-based targeting would result in a weakening of DP household survival strategies. This observation notwithstanding, the reality of pressures on inter- national and domestic public expenditure on DPs over time does lead to reduction in DP targeted assistance. This necessitates a shift in strategies toward addressing vulnerability of DPs. Some specific issues to be sensitive to in this regard include the following: 1. The possibility that some DPs may not possess national identifi- cation cards or may be blocked from access to particular types of safety net payments, such as unemployment insurance, should be reviewed. 2. The dynamics of capturing DPs who may not be citizens of the country in which they live and so may not be formally eligible for benefits despite real vulnerability should be considered. 3. Safety net provisions should ensure that assets such as property that may be formally owned by DPs but not immediately accessible due to their presence in a conflict zone is not counted as an asset in their possession. 4. Special attention should be paid to the specific situation of DPs living as guests in joint housing with host families and to the way that such situations should be taken into account in vulnerabil- ity assessment criteria. 5. Ensure that social safety net provisions are adaptable enough to integrate monitoring mechanisms that could reincorporate DPs who may become vulnerable in the future due to shocks to their survival strategies. Again, the large number of DPs living with host families in some countries represents an element of fragility that could suddenly change and alter the level of vulnerability of DP households. 6. Where governments wish to integrate DPs into existing social safety nets, a gradual approach might be warranted, one that first integrates vulnerability criteria into a dedicated DP scheme and eventually mainstreams the approach. This is the approach that WFP, for example, has taken in Serbia in regard to its food aid distributions. In the interim, cost savings could be channeled back into specialized support of DPs not covered under safety net payments--such as housing schemes. One seemingly obvious area in which such a transition could occur in some countries would be in the reduction of benefits to DPs who also receive salaries from host governments (see chapter 7). 156 living in limbo 7. It is essential that any shifts from prevailing practice in benefit entitlements for DPs be accompanied by public information campaigns clarifying the nature of any perceived changes by DPs in terms of their broader legal and political status. The uncertain future of DPs makes them highly sensitive to perceived changes in status that may impact their future entitlements to reclaim property or to return home. Sometimes such perceptions are accurate, but frequently they are not. 8. One area in which DPs may be increasingly at risk is in their access to public utilities such as electricity and water. As privati- zation of utilities and the establishment of market-based pricing for utilities increase in the region, DPs in collective centers and in private accommodation with illegal connections will be put under increased pressure. While some governments subsidize DP utilities in collective centers, the risk is mitigated in the short run. However, it is important to retain a consciousness of DPs as a particular subgroup with potentially high vulnerability, particu- larly in regard to heating, and to maintain a sensitivity to their situation in diagnostic exercises. Second, international agencies must be sensitive to the fact that many governments do not view DP assistance as purely a social safety net issue. They also see such assistance as part of a larger political strategy as well as, in some cases, a kind of limited compensation for lost assets. The World Bank should encourage governments to ensure that expenditures for DPs are transparent within the budget and not provided through off-budget measures such as payments through parastatal institutions or other channels. While not all governments equally provide assistance to DPs within their territory from public resources, many do. Under current practice, such government expen- ditures for assistance to DPs are often provided under separate categories, distinct from social safety net expenditures, and may not always be allocated in a manner allowing for accurate comparison of the cost-benefit of such assistance in relation to other safety net provi- sions for the local population. During consultation on this study, a critical question was posed re- garding social safety nets: Should assistance to DPs be needs tested and vulnerability based from the very onset of a conflict? There is no easy answer. Much of the above discussion of safety nets revolves around the current transition, in various countries of the region, from specialized, identity-based assistance to DPs to needs-tested, vulnerability-based assistance. The reality, however, is that much of this operational dis- course has emerged not from a demand-based careful review of DP development programming 157 vulnerability but from the supply-side reality that targeted DP assis- tance from international agencies in the region is being reduced. In the earliest days of conflict and displacement, the flow of displace- ment usually is too rapid and sudden for an adequate needs-based eval- uation of DPs to form the foundation of a sensible approach to assis- tance. In such situations, an identity-based assistance strategy that targets displacement as a proxy for vulnerability is seemingly the only course of action. Another factor dictating such an approach, as noted earlier, is the variable ability of host country administrative structures and international agencies to adequately evaluate individual vulnerabil- ity, particularly when displacement occurs in large numbers all at once. In fact, continuing identity-based targeting may remain the most effi- cient form of targeting for an extended period if such targeting achieves greater coverage of the vulnerable than an inefficient individual needs- based approach. This is particularly true when the specific nature of DP vulnerability has not been adequately incorporated into vulnerability criteria. In practice, international humanitarian agencies, particularly the WFP, have for many years been incorporating hybrid strategies into approaches for DP assistance. These approaches remain anchored in identity-based assistance but are designed to discourage better-off DPs from accessing benefits. Examples include distribution of lower qual- ity foodstuffs perceived to be undesirable to higher-income groups and distribution strategies requiring greater transaction costs in terms of time and effort on the part of beneficiaries. As discussed previously in this study, in the case of Bosnia, assistance strategies have, from the onset of the conflict, taken account of the wider vulnerability within the society in general and have not used displacement as a primary criterion for assistance. Nonetheless, it seems clear in the specific context of the long-term displaced populations within this region that a review of identity- based assistance and a transition to needs-based assistance for DPs should probably occur at an earlier stage than has been the case.1 Issues in Social Service Delivery in Displacement Situations in the Europe and Central Asia Region Enrollment rates in education among the displaced and local popula- tions appear similar in most cases. Yet given the multifaceted impacts of conflict, educational sector programs should undertake diagnostic work to explore whether attendance and performance indicators for displaced children are the same as those for local children. When 158 living in limbo obstacles to meaningful education of DPs are identified, educational programming should incorporate sensitivity to, and seek to address, these impediments. Some educational obstacles derive from poverty. Others more directly relate to the direct and indirect impacts of displacement. Obstacles blocking access to social services may be different for DPs than for local populations. Even when de jure access exists, issues such as lack of adequate documentation, language differences, and distance from facilities may represent special obstructions. In several of the postsocialist states, the displaced suffer from the same deterio- ration of public facilities as local populations. For DPs, the commit- ment of governments to DP support has less meaning given the de facto privatization of social services, particularly health care. While this reality does not make the displaced any more disadvantaged than others, it does emphasize a need to review critically the value of com- mitments made by governments. Diagnostic and social sector delivery strategies need to be informed by the typical subnational patterns of DP concentration. This relates to displaced and host populations in these areas. DP populations fre- quently occupy social sector infrastructure such as schools and clinics, a further pressure on remaining infrastructure. This may create spe- cific subnational impacts on social service delivery, which while signif- icant at the local level may not appear in a homogenized view at the national level. In regard to the health sector, when significant numbers of the dis- placed remain in collective centers, the possibility of deteriorating health among such populations must be well understood and periodi- cally evaluated. Unless some change occurs in the situation of these clusters of DPs, it seems likely that the continuing deterioration of facilities will contribute to increasing health problems, such as respi- ratory ailments. This is more important for some countries (Georgia, Azerbaijan) where higher proportions of DPs live in collective centers than others. Existing surveys may not have been detailed enough to identify such trends, which may become more significant with the pas- sage of time since internationally supported budgets for maintenance of such facilities have declined. When displaced populations have settled in areas outside the con- trol of the central government, either inside the country or in neigh- boring countries, efforts at passive coordination should be developed. In this coordination, curricula and standards of education for DP children should be similar to those for the local population to facilitate later adaptation to the home educational system. This has not typically development programming 159 been a major issue in this region, given the large numbers of internally displaced and the history of state succession and displacement even in relation to refugees. Yet when the environment in a return area differs from that of the current settlement area, constant reflection of the "Janus-faced" nature of displacement and the uncertain future direc- tion of settlement of DPs should be part of the planning process. This will maximize the possibility that education remains relevant for future disposition of DP settlement. Displaced Populations and CDD (Community-Based Programming, Municipal Development, Area-Based Development) For many years, humanitarian agencies have recognized that assis- tance to displaced populations should take into account the needs of surrounding host populations. Doing so dampens potential social ten- sion and recognizes the often significant degree of need among the poor within the host population. In the region's current policy envi- ronment, an important question arises: How do programs targeted at local populations take into account the presence of displaced popula- tions in their midst? Globally, the significance of this question varies greatly, depending upon the degree of isolation of DPs from local com- munities. In this region, given the general pattern of relatively close in- tegration of DP settlements with local communities, the question is of some importance, particularly for area-based programs. Work that follows this study will focus more intensively on issues of employment and self-reliance in specific DP situations in the region, and, in this context, many of the issues regarding area-based pro- gramming will be addressed in more detail. However, from the present analysis there emerge several observations that have operational im- plications in their own right. These observations center around two questions: Can DP settlements be treated as communities in terms of existing and future community-based initiatives? Can individual DP households access programs in wider initiatives directed toward the local population in general? Treating DP Settlements as Communities The question of whether individual settlements of DPs can be ap- proached in the same manner as settled communities is difficult to 160 living in limbo answer in generalities. The region's displaced face many obstacles in social mobilization and in maintenance of social networks. Their settle- ment patterns have rarely allowed for predisplacement communities to be settled in concentrated blocks. As noted, this differs from many other displacement situations in the world in which such continuity of social networks is a real aspect of settlement. In addition, as discussed in chap- ter 6, many factors have contributed to the undermining of the founda- tions of social networks among DPs. Unfortunately, it also must be recognized that years of displacement have probably served to "hollow out" DP settlements much as urban migration has many local rural communities. The passing years have drawn potential leaders and entre- preneurs away from the misery of DP settlements, leaving those settle- ments without much of the natural human capital that might otherwise have represented an invaluable asset in community organization. Yet a very cursory examination of the situation in some key countries reveals that a significant minority of DPs in most of these situations do remain in clusters corresponding to some degree to predisplacement communities. Some evidence of their ability to organize and work to- gether through collective action has been shown, at least anecdotally. Furthermore, there is anecdotal evidence in some situations that DPs living together in a locality, even when they come from different predisplacement communities, have to varying extents recognized their mutual identity as "DPs" as a basis for common purpose and organization. Though this situation needs to be explored further, there is at least a suggestion that such groups (for example those living in collective centers for several years) may have more capacity than out- siders think to become partners in community-based programming. If offered the chance, they might access existing channels of social funds and other mechanisms for group investment. One factor that could be important in both community-based approaches and other individually focused initiatives (such as micro- finance and small business support) is that DPs may respond better to initiatives offering relatively short-term gestation processes and invest- ment returns. While this logic is the basis for microfinance in general, DPs may be particularly sensitive to it given the inherent instability of their situations. This may create different discount rates for long-term investments. In other words, if you cannot be sure where you will be in a few years, you may not want to invest in an activity that will take several years to bring a return or requires commitments locking in even small sources of capital in ways that preclude quick liquidation without high penalties. There are in fact several microfinance pro- grams that target DPs, particularly in the South Caucasus. Part of the follow-up work to this study will review this sector in more detail and develop more focused lessons. development programming 161 DP Access to Local Community-Targeted Programs First, it must be reiterated that the subnational clustering patterns typ- ical of displacement mean DPs are a more significant characteristic of some communities or municipalities than others. In many cases, this dynamic means that up to 40 percent of the population of certain municipalities or districts may be DPs, while other populations have few DPs or none. Nationally designed community or municipality programs need to recognize this fact and explore the implications for municipal development programs, area development programs, social funds, and other programs in localities with significant DP numbers. Even within these larger patterns, DP settlements are very heteroge- neous. In some localities, the displaced may be living with local fami- lies or settled in individual apartments or small public buildings and be in close contact with local populations. In other areas, they may be concentrated in one or more large collective centers or separated into a specific section of town. The displaced may share water systems and other infrastructure and have common concerns with the local popu- lation, or they may have different concerns. Since it is difficult to define general patterns, such patterns must be examined on a case-by- case basis. In areas where DPs do represent a significant percentage of a locality population, they should be incorporated into stakeholder analysis of programming. There may be specific obstacles to participation of the displaced in local programming or, relatedly, to local public resource policy plan- ning. As noted in chapter 7, some DPs have voting rights in local elections; others do not. The inability to participate in local political discourse, to vote or stand for office, may indirectly act to exclude DPs from consideration as stakeholders in government or donor-financed community activities. Absence of documentation, lack of property ownership, and other factors may exclude DPs from the organization of particular stakeholder groups or complicate their access to public services provided through particular group-based or individual enti- tlement programs. Exclusion or access complication may not have much meaning where there is little significant government program- ming or donor-supported initiatives in an area. But where the quality of government programming improves or donor-supported initiatives do enter an area, new elements of DP exclusion may emerge. The frequent invisibility of the displaced in official statistics focused on local populations (census lists, property ownership, tax roles) may also mean that, without adequate reflection of the DP settlements in an area, comparative prioritization within programs may be skewed. In addition, initiatives may underestimate the amount of funding or administrative capacity required in a particular locality. 162 living in limbo Housing and Shelter The housing situation of the displaced can be seen as a derivative impact of income poverty. If DPs had better employment and higher incomes, they could afford to resolve their housing problems. However, this does not tell the full story. As noted in chapter 4, the disparity between high real estate prices and low wages means that even with increased incomes, it is very unlikely that most DPs, unless they return home, will ever be able to afford the purchase of shelter on the commercial market. Local populations would not have been able to afford their own residences on the local market with their own wages. For the most part, they only own property in the former Yugoslavia and Soviet Union because they owned it before or because they were given housing under the 1990s privatizations. The issue of housing for the region's long-term displaced popula- tions is perhaps the most difficult to address for three reasons: the view of private housing as private goods; the supposed high-opportunity costs for housing programs; and the perception of displacement as temporary. Private housing programs are viewed as private goods because they imply direct investment in housing construction and thus the creation of assets that will be distributed to individual families at subsidized costs. When this provides benefits to some households and not to others, a potential equity issue arises. Relatedly, for international agencies, such programs may require the purchase of land, something most international agencies are prohibited from financing. Many observers see housing programs, on a per capita basis, as expensive investments that, in the light of other competing demands for public funds, seem to have high opportunity costs. Due to the continuing perception that displacement is "temporary," many observers also have concluded, explicitly or implicitly, that housing construction programs or other durable solutions to the shel- ter problem are not practical or meaningful. In other words, if people are going to go home eventually, such investment seems wasteful and may result in double demands on international and domestic public funds. The latter would be the case in the event of a return home, when funds for the reconstruction or rehabilitation of existing housing stock in the home area will be eventually required as well. The World Bank, for example, has been willing to help support financing of hous- ing repair in natural disasters. The Bank has supported the repair of housing for returning IDPs in some places, such as in Bosnia. But for various reasons, including the perception of the "transient" nature of DPs, the Bank has not, to date, approved repair of existing housing or development programming 163 construction of new housing for DPs in their areas of displacement, even when such displacement has lasted for a generation or more. While from time to time some countries discuss matching DPs with empty houses or apartments in areas with high out-migration, this action would only provide a way forward when such shelter exists in places with at least a minimum of potential for employment or self-reliance. There has also been discussion of privatization of col- lective centers and investment of ownership in individuals or groups of IDPs. While there may certainly be scope for such initiatives, the situation varies considerably from one collective center to another. Some properties have sufficient space for families and can be en- hanced to represent acceptable living quarters. Others have signifi- cant commercial values and could be leveraged by DPs for investment or sold to finance purchase of other living space. However, many col- lective centers are crowded and beyond hope of rehabilitation, and many have little or no commercial value or meaning as an asset for a DP family. There is invariably a range of small-scale creative solutions to the shelter problem of the region's displaced. Perhaps a large part of the answer lies in recognizing the heterogeneity of DP shelter problems and building approaches that empower local authorities and commu- nities to develop individual solutions. Another part of the solution, at least with the minority of DPs who may have sufficient funds, would be making a concerted effort to in- form DPs that they can legally purchase housing. In several places in the region, national laws do allow the displaced to own land and pur- chase housing. Yet many DPs do not do so, at least openly, because they believe purchasing property would somehow legally bar them from taking back their property in their home areas should an oppor- tunity to return arise. This dynamic is particularly complicated in cer- tain cases because prior to displacement, the home area property was state owned and would need to be privatized. Sometimes this percep- tion is due to differing interpretations of national legal frameworks and government policies, and sometimes it is, effectively, mythology. However, explicit government information campaigns could help ease this situation. The shelter situation of the region's DPs is perhaps the aspect of their condition of most concern in the long run. There do not seem to be any obvious, low-cost answers, although some previously uncon- sidered ideas may yet emerge in an individual country situation. The colder climate and social mores of the societies in this region make the shelter problem for DPs here far severer than for those in most other global situations of displacement. 164 living in limbo As noted, housing programs, whatever their dimensions and de- signs, are costly. While a wide range of housing programs for DPs can be found around the region, in general they are relatively small-scale and deal with only a small percentage of DP requirements. Note 1. With the aforementioned caveat, that assistance, both domestic and inter- national, is often influenced by considerations other than vulnerability. ANNEX 2 Methodology A DESCRIPTION OF THE METHODOLOGICAL issues surrounding this study entails more than a simple formulaic statement. The study is not based upon the commissioning of a direct set of surveys but rather represents a kind of analytical stocktaking of a wide variety of existing literature on individual country situations. In a sense, the process of gathering and analyzing available material forms an intrinsic part of our objec- tives. By reviewing the extensive, albeit uneven, set of surveys and assessments, our intent is to explore the common impacts that occur in all situations of displacement as well as to see where gaps exist in our knowledge. Consequently, based upon this analysis, the conclusions of the study incorporate recommendations concerning the treatment of DPs in the context of poverty assessments and monitoring. During the course of the data collection, the study team collected more than 90 surveys of displaced populations and some 300 field re- ports, assessments, and other materials concerning 12 countries in the Europe and Central Asia region that currently have, or recently have had, significant numbers of conflict-displaced populations. These ma- terials were prepared by a wide range of actors, including national and international NGOs, UN and bilateral agencies, governments, and the World Bank itself. Surveys were often prepared for specific operational objectives. Some surveys are designed to compare the situation of the displaced population with that of the host population. Others focus specifically on DPs, without a parallel review of the situations of surrounding populations. Among the material there exists little standardization in the definitions of key terms, such as poverty and unemployment, and we have been unable to review most of the ques- tionnaires upon which the surveys were based. We have emphasized the use of data that are comparative because our objective is to under- stand DP vulnerability in the context of other groups in affected societies. To the greatest extent possible, we have also tried to make judgments about the rigor of specific surveys and, when there were choices, use the most reliable data. In addition to the inconsistency of the methodology of the surveys, two other issues complicate the gathering of accurate perspectives on the displaced. First, DPs have many reasons for inaccurately reporting 165 166 living in limbo their situations, above and beyond the reasons other populations have for doing so. DP status confers benefits, and, accurately or not, there may be perceptions that reporting of income or ownership of property could be used to withdraw such status. Second, DPs are difficult to find and consequently may be easily over- or undercounted by wide margins. Many reviews and censuses of numbers and locations of the displaced have found significant inaccuracies in official data. A recent census in Serbia, for example, reduced the number of DPs by over 20 percent. Part of this may represent a decision by some DPs to no longer view themselves as such and to avoid being characterized as a DP. Part of it is also likely to emerge from initial overcounting. Over- or undercounting the displaced are not uncommon phenomena. The frequent settlement of DPs in unconventional housing situations, such as in abandoned factories, spas, and other types of collective centers, has sometimes led to "missing" DPs in larger surveys of poverty based upon lists of residential addresses. Therefore, while we present various charts and figures within this study on numbers of DPs, we do so to illustrate magnitudes and to put a tangible face on a complex problem rather than to assert the absolute accuracy of the statistics utilized. The quality of existing data places certain limits on how definitive our observations can be. While we cite statistics on various issues, we do so to identify cross-country patterns and magnitudes of vulnerabil- ity and to signal areas of concern. There is often a question, as noted above, on the accuracy of numbers of displaced in a given situation, due to the potential motivation of different actors on inflating or re- ducing the numbers and the difficulty in keeping track of DPs. How- ever, given the paucity of cross-country analysis in this field, we feel this methodology nonetheless provides value, again for the purpose of identifying trends and key issues. This analysis represents the begin- ning of a process that, at the outset, we need to recognize may need further elaboration through direct investigation. Although this study is meant to identify common patterns and is- sues among a variety of situations, it is not designed as a set of case studies and so refrains from making recommendations on specific country situations. It is also important to note that the level of infor- mation on individual situations varies a great deal. In four countries, Azerbaijan, Bosnia and Herzegovina, Georgia, and Yugoslavia (Serbia and Montenegro), far more information is available than elsewhere.1 In other situations, such as Chechnya, there is useful information but virtually no comparative data on host populations are available. Three major gaps exist in the available data that have placed the most serious limitations on this study. The first concerns Turkey, host to what may be the largest conflict-displaced population in the region. methodology 167 The numbers of displaced reported differ greatly, ranging from 400,000 to up to 3 million. We are unable to locate reliable data com- paring the displaced there with other Kurdish migrants or with the local population in general. This limits our observations to aspects of their condition that do not require such comparative information. The second concerns time-series data. One of the areas we would have liked to understand better is the effect displacement has on affected populations over time. There are factors that would suggest a deterio- ration of conditions in DPs as time passes, such as the exhaustion of fragile survival strategies. There are also factors that would suggest that long-term displacement results in a gradual improvement of con- ditions as DPs adapt and find more sustainable coping strategies. We have something to say on this subject in chapter 8, but much of it is in the realm of hypothesis and this issue bears further exploration.2 The last area in which we found few data was on the fate of displaced pop- ulations who eventually return home. Specifically, we found few data on what has happened to former DPs in Tajikistan and Moldova, countries where major proportions of the population were displaced through conflict and have since returned home. This study deals primarily with large clusters of DPs who are settled in their countries of origins (IDPs) or in neighboring countries (refugees). There are other groups of refugees who have succeeded in leaving the region entirely, immigrating to, or acquiring political asy- lum status in, Europe, the United States, or other countries outside the region. The role of this diaspora is an important topic in its own right but is not addressed in any major way in this study. Notes 1. All available World Bank poverty assessments for affected countries have been reviewed. In particular, this study benefits a great deal from recent data from poverty assessments in Azerbaijan, Bosnia and Herzegovina, and Serbia and Montenegro, which are, as of this writing, in various stages of being finalized. The study team has also had the opportunity, to varying degrees, to help inform these processes. 2. The fact that, in many situations, DPs in the region continue to appear disproportionately among the poor after long periods of time does pro- vide some information for reflection. Bibliography Benjamin, Judy, and Khadija Fancy, The Gender Dimensions of Internal Displacement: Concept Paper and Annotated Bibliography. Office of Emergency Programmes Working Paper series, UNICEF (New York, 1998). Crisp, Jeff, Who Has Counted the Refugees? UNHCR and the Politics of Numbers. Working Paper 12, UNHCR (Geneva, 1999). Hampton, Janie, ed., Internally Displaced People: A Global Survey (London: Earthscan, 1998). Heleniak, Tim, The End of an Empire: Migration and the Changing National- ity Composition of the Soviet Successor States. Background Paper, World Bank (Washington, D.C., 1997). Kunder, James, The Needs of Internally Displaced Women and Children: Guiding Principles and Considerations. Office of Emergency Programmes Working Paper series, UNICEF (New York, September 1998). Machel, Grac'a, The Impact of War on Children (London: Hurst, 2001). , Promotion and Protection of the Rights of Children: Impact of Armed Conflict on Children. UNICEF (New York, August 1996.) Norwegian Refugee Council (NRC), Global IDP Project, http://www. idpproject.org. Refslund Sørensen, Birgitte, Self-Help Activities among Internally Displaced Persons, in Rights Have No Borders: Internal Displacement Worldwide, ed. Wendy Davies (Oslo: Norwegian Refugee Council/Global IDP Sur- vey 1998). http://www.nrc.no/global_idp_survey/rights_have_no_borders/ sorensen.htm. United Nations High Commissioner for Refugees (UNHCR), Women, Children, and Older Refugees: The Sex and Age Distribution of Refugee Populations with a Special Emphasis on UNHCR Policy Priorities. UNHCR (Geneva, July 2001). 169 170 bibliography UNHCR, Executive Committee of the High Commissioner's Programme, Standing Committee, Social and Economic Impact of Large Refugee Popu- lations on Host Developing Countries, sixth meeting, January 6, 1997 (EC/47/SC/CRP.7). United Nations Children's Fund (UNICEF), Poverty and Children: Lessons of the 90s for Least Developed Countries (New York: UNICEF, 2001). World Bank, The World Bank's Experience with Post-Conflict Reconstruction. Operations Evaluation Department (OED), World Bank (Washington, D.C., 1998). World Food Programme (WFP), Operational Expenditures for Operations with IDP/Refugee Component in Selected Countries, 1990­2001. , An Overview of Vulnerability Analysis Mapping (VAM). WFP (Rome, 2002). http://www.un.org/Depts/Cartographic/ungis/meeting/ march00/documentation/wfp_recalde2.pdf. , Situations of Displacement: Issues and Experiences. Discussion Paper, WFP (Rome, 2000). Albania Barjaba, Kosta, and Gloria LaCava, Social Impacts of the Kosovar Refugee Crisis in Albania. ECSSD Background Report, World Bank (Washington, D.C., 2000). Çabiri, Ylli, Arjan Starova, Milva Ekonomi, Piro Misha, Remzi Lami, Vassilaq Leno, and Ylli Vejsim, Albanian Human Development Report. UNDP Tirana and Institute of Statistics Albania (Tirana, 2000). De Soto, Hermine, Peter Gordon, Ilir Gedeshi, and Zamira Sinoimeri, Poverty in Albania: A Qualitative Assessment. Technical Paper Report 520, World Bank (Washington, D.C., 2002). World Bank, Albania: Interim Poverty Reduction Strategy Paper and Assessment. Europe and Central Asia Region Report 21178, World Bank. (Washington, D.C., 2000). , Albania: Poverty Reduction Strategy Paper and Joint Assessment. Europe and Central Asia Region Report 23469, World Bank (Washington, D.C., 2002). , A Qualitative Assessment of Poverty in Ten Areas of Albania. Europe and Central Asia Region Report 23521, World Bank (Washington, D.C., 2001). Armenia Braithwaite, Jeanine, The Social Safety Net in Armenia. Poverty and Social Policy Report, World Bank (Washington, D.C., 1994). UN Economic and Social Council (ECOSOC), Commission on Human Rights, Specific Groups and Individuals: Mass Exoduses and Displaced bibliography 171 Persons. Addendum: Profiles of Displacement--Armenia, report by Francis Deng, 57th sess., November 2001 (E/CN.4/2001/5/Add.3). United Nations Office in Armenia, Poverty of Vulnerable Groups in Armenia: Comparative Analysis of Refugees and Local Population. United Nations Department of Public Information (Yerevan, 1999). World Bank, Armenia: Confronting Poverty Issues. Europe and Central Asia, Report 15693, World Bank (Washington, D.C., 1996). , Armenia: A Poverty Profile. Europe and Central Asia Region, Report 19702, World Bank (Washington, D.C., 1995). WFP, Protracted Relief and Recovery Operation--Armenia 10053.0. WFP (Rome, April 2001). Azerbaijan Agency for Rehabilitation and Reconstruction of Areas (ARRA). Feasibility Study for the Resettlement and Reconstruction Program of the Liberated Areas. ARRA (Baku, 1998). ARRA, World Bank, and UNDP, Social Assessment of Internally Displaced People from the Rayon of Fizuli, Azerbaijan. Background Report, ARRA, World Bank, UNDP (Baku, March 1997). de Roos, A. C., and D. Venekamp. ECHO Food and Medical Assessment in Azerbaijan including Nakhichevan. European Community Humanitarian Office (ECHO) (Baku, August 1996). ECOSOC, Commission on Human Rights. Specific Groups and Individ- uals: Mass Exoduses and Displaced Persons. Addendum: Profiles of Displacement--Azerbaijan, report by Francis Deng, 55th sess., January 1999 (E/CN.4/1999/79/Add.1). Sardarov, Elchin, Refugees and Displaced Persons and Poverty. Case Study, SANIA Humanitarian Information Analytical Agency (Azerbaijan, 2001). SIGMA, Azerbaijan: Study of IDPs and Refugees: Social Matrix on Settling of the Communities of IDPs. SIGMA (Baku, January 2002). , Social-Economic State of Refugees and Internally Displaced People in Baku. SIGMA (Baku, 2001). Sovich, Nina, Azerbaijan's Forgotten Remnants of War. Christian Science Monitor, October 19, 2000. World Bank, Azerbaijan: Interim Poverty Reduction Strategy Paper and Joint Assessment. Europe and Central Asia Region Report 22336, World Bank (Washington, D.C., 2001). , Azerbaijan: Poverty Assessment Technical Papers. Europe and Cen- tral Asia Region Report 15601, World Bank (Washington, D.C., 1997). , Azerbaijan: Survey of Living Conditions (ASLC). Europe and Cen- tral Asia Region Basic Information Document, World Bank (Washington, D.C., 2000). 172 bibliography , Internal Displacement in Azerbaijan: Main Findings from the Azerbaijan Household Survey on Internally Displaced Persons, Refugees and the Resident Population." Europe and Central Asia Region Report, World Bank (Washington, D.C., August 2002). WFP, WFP Assistance to Internally Displaced Persons: Country Case Study on Internal Displacement, Azerbaijan. WFP (Rome, 1999). Balkans United Nations Sub-Committee on Nutrition, Report on the Nutrition Situa- tion of Refugees and Displaced Populations, Geneva, 2000 (RNIS 30). Bosnia CIET International, Unemployment, Employment and Wages in Bosnia- Herzegovina: 1997 Vulnerability Survey. CIET (New York, 1997). CRPC (Commission for Real Property Claims of Displaced Persons and Refugees), Return, Relocation, and Property: A Discussion Paper. CRPC (Sarajevo, December 1997). Dani, Anis, A Social Assessment of Bosnia. ECSSD Background Report, World Bank (Washington, D.C., April 1999). International Crisis Group (ICG), Going Nowhere Fast: Refugees and Inter- nally Displaced Persons in Bosnia and Herzegovina. Bosnia Report 23, ICG (Sarajevo, May 1997). Mareco Index Bosnia (MIB), Bosnia in Statistics. MIB (Sarajevo, December 1997). Ministry for Refugees and Displaced Persons, Republic Institute for Statistics, Census of Refugees and Displaced Persons in the Republic of Srpska. Ministry for Refugees and Displaced Persons (Sarajevo, March 1996). Pilegaard, Lisbeth, and Henrik Fosseldorff, Coming Home: A Survey of the Living Conditions of Refugees Repatriating from Denmark to Bosnia and Herzegovina, from August 1994 to August 1997. Danish Refugee Council (Copenhagen, 1997). Poggi, Patrizia, Bosnia and Herzegovina Social Assessment Qualitative Inter- view Report. ECSSD Background Report, World Bank (Washington, D.C., June 1998). United Nations Development Programme (UNDP), Early Warning System: Annual Report: Bosnia and Herzegovina (Sarajevo: UNDP, 2001). , Human Development Report: Bosnia and Herzegovina (Sarajevo: UNDP, 1998). UNHCR, Bosnia and Herzegovina: Repatriation and Return Operation 1998. Humanitarian Issues Working Group Report HIWG/97/7, UNHCR (New York, December 1997). bibliography 173 , Daunting Prospects--Minority Women: Obstacles to Their Return and Integration (Sarajevo: UNHCR, 2000). Vodopivec, Milan. Labor Market in the Postwar Bosnia and Herzegovina: How to Increase Worker Mobility and Job Creation. Europe and Central Asia Region Report 24889, World Bank (Washington, D.C., 2002). World Bank, Bosnia and Herzegovina: Consultations with the Poor. National Synthesis Report, Europe and Central Asia Region, World Bank (Washington, D.C., 1999). , Bosnia and Herzegovina: Diagnostic Surveys of Corruption. Europe and Central Asia Region Background Paper, World Bank (Washington, D.C., 2002). , Bosnia and Herzegovina: Local Level Institutions and Social Capital Study (Washington, D.C.: World Bank, 2002). , Bosnia and Herzegovina: Post-Conflict Reconstruction. Country Case Study series, OED, World Bank (Washington, D.C., 2000). WFP, WFP Assistance to Internally Displaced Persons: Country Case Study on Internal Displacement, Displacement in Bosnia-Herzegovina. WFP (Rome, 2000). Caucasus Abeles, Peter L., Assessment of the Shelter Sector for Refugees and Internally Displaced People in the Caucasus. Complex Emergency Support Unit, UN Department of Humanitarian Affairs (Geneva, April 1996). Bacon, Kenneth H., Lost on the Silk Road: Displaced People Looking for Home. Refugees International (Washington, D.C., September 5, 2000). http://www.reliefweb.int. Council of Europe (COE), Parliamentary Committee on Migration, Refugees, and Demography, Situation of Refugees and Displaced Persons in Armenia, Azerbaijan, and Georgia. Report 9480 (June 2002). Hansen, Lars Funch, Summary of Discussions during the Conference, in Con- flict and Forced Displacement in the Caucasus: Perspectives, Challenges, and Responses, ed. Tom Trier and Lars Funch Hansen (Copenhagen: Danish Refugee Council, 1999). Lattu, Kirsti, David Garner, and Dennis Culkin, Humanitarian Needs Evalua- tion for Victims of the Nagorno-Karabakh Conflict. U.S. Agency for International Development (USAID) (Washington, D.C., January 1998). http://www.geocities.com/fanthom_2000/usaid-rep.html. Women's Commission for Refugee Women and Children, Women Displaced in the Southern Caucasus: An Examination of Humanitarian Assistance Needs in Azerbaijan, Armenia, Nagorno-Karabakh, and Georgia. Field Report of the April 1998 Fact-Finding Delegation (1998). http://www. womenscommission.org. 174 bibliography Croatia Barakat, S., and S. Ellis, From Relief to Development: The Long-Term Effects of "Temporary" Accommodation on Refugees and Displaced Persons in the Republic of Croatia. Disasters 20 (2) (1996): 111­124. European Council on Refugees and Exiles (ECRE), Focus Group on South- Eastern Europe, Croatian Democracy's "Litmus Test": Refugee Return and Reintegration into Europe. ECRE (London, 2001). Gomart, Elizabeth, Social Assessment of Poverty in Croatia, December 1999­March 2000. Poverty Reduction and Economic Management in ECA Region Background Paper, World Bank. (Washington, D.C., April 2000). Martic-Biocina, S., S. M. Spoljar Vrzina, and V. Rudan, Anthropological and Psychodynamic Characteristics of Family Life of Displaced Persons and Refugees. Collegium Anthropology 20 (2) (1996): 301­308. Office for Displaced Persons and Refugees (ODPR), Government of the Republic of Croatia, Displaced Persons and Refugees in the Republic of Croatia. ODPR (Zagreb, July 1995). PRISM Research, Institutional Needs Assessment: Zadar and Sibenik-Knin Counties Refugee Return and Regional Development Project (Croatia). PRISM (Zagreb, 2002). Puls (Market, Media, and Public Opinion Research), Survey on Returnee Population in Croatia, Main Findings. Puls (Zagreb, January 2001). World Bank, Croatia: Economic Vulnerability and Welfare Study. Europe and Central Asia Region Report 22079, World Bank (Washington, D.C., 2001). Georgia Accion Contra el Hambre (ACH), Georgia, Whose Land? Overview in Land Reform and Privatization Process in Western Georgia Agriculture and Its Implications for Displaced People. ACH (Zugdidi, August 1998). Boutroue, Joel, and Stephen Jones, Prospect for the Return of Internally Displaced Persons and Refugees to Abkhazia: A UNHCR Review of the Situation in Georgia. Mission Report, UNHCR (Geneva, May 1997). Buck, Thomas, Women, War, and Displacement in Georgia. Research and Reference Services Project, USAID (July 1999). http://www.osgf.ge/wie/6 .html. Buck, Thomas, with Alice Morton, Susan Allen Nan, and Feride Zurikashvili, Aftermath: Effects of Conflict on Internally Displaced Women in Georgia. Center for Development Information and Evaluation Working Paper 310, USAID (Washington, D.C., September 2000). bibliography 175 Dershem, Larry, Food, Nutrition, Health, and Nonfood Vulnerability in Georgia, 1996: A Household Assessment. Save the Children, Assistance Georgia (Tbilisi, June 1996). Dershem, Larry, Nana Gurgenidze, and Steven Holtzman, Poverty and Vulnerability among Internally Displaced Persons in Georgia: An Update of Their Current Status and Circumstances. Save the Children, Assistance Georgia (Tbilisi, 2002). Dershem, Larry, and Irakli Sakandelidze, Various Characteristics of the Population in Western Georgia as of October 1999. Save the Children, Assistance Georgia (Tbilisi, 1999). Dudwick, Nora, Georgia: A Qualitative Study of Impoverishment and Coping Strategies. Europe and Central Asia Region Technical Paper 4, World Bank (Washington, D.C., April 1997). ECOSOC, Commission on Human Rights. Specific Groups and Indi- viduals: Mass Exoduses and Displaced Persons. Addendum: Profiles of Displacement--Georgia, report by Francis Deng, 57th sess., January 2001 (E/CN.4/2001/5/Add.4). ECHO, Food and Medical Assessment in Georgia. ECHO (Tbilisi, April/May 1996). Foundation for the Development of Human Resources (FDHR), Center for Psycho-Social Rehabilitation, Final Analysis of the Psycho-Social Rehabili- tation of Refugees (IDPs) in Georgia. FDHR (Tbilisi, 1996). Georgian Young Lawyers' Association (GYLA), Monitoring of Legal and Actual Status of Internally Displaced Persons in Georgia. GYLA (Tbilisi, 1999). Gilliland, Mary Kay, Sonja Spoljar-Vrzina, and Vlasta Rudan, Reclaiming Lives: Variable Effects of War on Gender and Ethnic Identities in the Nar- ratives of Bosnian and Croatian Refugees. Anthropology of East Europe Review 13 (1) (Spring 1995). Hunt, Kathleen, Repatriation in Georgia. Open Society Institute (New York, 1995). http://web.archive.org/web/20011031213656/www.soros.org/fmp2/ html/georgia.htm. International Federation of Red Cross and Red Crescent Societies (IFRC), Internally Displaced Persons: A Socio-Economic Survey, Georgia. IFRC (Tbilisi, 2000). , Physical Condition of Collective Centres Accommodating Internally Displaced Persons. IFRC (Tbilisi, October 1999). , Survey of the Living Conditions of the Population of Jvari, Samegrelo Region, Georgia. IFRC (Tbilisi, November 1999). IFRC and Georgian Red Cross Society, Beneficiary Survey 1995­1996. IFRC (Tbilisi, 1996). International Rescue Committee (IRC), A Socio-Economic Survey of the West Georgia and Shida Kartli Regions. IRC (Tbilisi, 1999). 176 bibliography Kharashvili, Julia, Georgia: Coping by Organising: Displaced Georgians from Abkhazia, in Caught between Borders: Response Strategies of the Internally Displaced, ed. Marc Vincent and Birgitte Refslund Sorensen (London: Pluto Press, 2001). , Internal Displacement in Georgia: A Personal Perspective. Presenta- tion at seminar on Response Strategies of the Internally Displaced: Changing the Humanitarian Lens, Oslo, November 9, 2001. National Centers for Disease Control and Save the Children, Nutritional Status of Children Less than Five Years of Age in Six Drought-Affected Regions of Georgia: 2000­2001. Final Report (Tbilisi, 2002). NRC, Survey on Internally Displaced People in Georgia: Living Conditions and Future Options of IDPs from Abkhazia. NRC (Tbilisi, November 1997). , Survey on Internally Displaced People in Georgia: Report and Recommendations. NRC (Tbilisi, April 1995). Rukhadze, Tinatin, Shelters and Social Infrastructure in Western Georgia. Report for Baseline Survey, CARE International (Kutaisi, 2000). Save the Children Fund, Assistance Georgia, Initiative Survey (February 2000), see http://www.assistancegeorgia.org.ge. Shakarishvili, George, Evaluation of Socio-Behavioral Patterns of Local and Internally Displaced Populations in Zugdidi Region of Georgia and Its Applicability for Local Policy Developments. Research Paper for Open Society Institute (New York, March 2002). United Nations Association of Georgia (UNAG), Summary of Cumulative Report on Education Among IDPs. UNAG (Tbilisi, 2000). UNHCR, Gali Population Poll in Georgia. UNHCR (Tbilisi, May 1997). , IDP [sic] in Georgia: Statistics Package as of March 2002. UNHCR BO Tbilisi (Tbilisi, 2002). , Committee on Economic, Social, and Cultural Rights, Consideration of the Initial Periodic Report of Georgia, 22nd sess., Geneva, May 2000. Vivero Pol, Jose Luis, Stable Instability of Displaced People in Western Georgia: A Food-Security and Gender Survey After Five Years. Journal of Refugee Studies 12 (4) (1999). WFP, Protracted Relief and Recovery Operation--Georgia 6122.01. WFP (Rome, April 2000). World Bank, Georgia: Interim Poverty Reduction Strategy Paper and Joint Assessment. Europe and Central Asia Region Report 21448, World Bank (Washington, D.C., 2000). , Georgia: Poverty and Income Distribution. Europe and Central Asia Region Report 19348, World Bank (Washington, D.C., 1999). , Non-Payment in the Electricity Sector in Eastern Europe and the Former Soviet Union. Energy Sector Unit, Europe and Central Asia Region Working Paper 423, World Bank (Washington, D.C., June 1999). bibliography 177 Kosovo Action against Hunger (AAH), Food Security Report: Kosovo. AAH (Prishtina, August 1999). , Food Security Assessment: Kosovo. AAH (Prishtina, January/ February 2000). Development Researchers' Network, Food Assessment in Minority Areas of Kosovo. Final Report. WFP and UNHCR (Prishtina, December 2001). Development Researchers' Network, Food Assessment: Kosovo 2002. Background Paper, WFP and UNHCR (Prishtina, March 2002). Inter-Agency Sub-Group on Poverty, Qualitative Poverty Assessment, Kosovo: Review of Secondary Materials. Care International (Prishtina, November 2000). International Organization for Migration (IOM), Basic Results: Kosovo Demographic, Socio-Economic, and Reproductive Health Survey. IOM (Prishtina, May 2000). Lawrence, Mark, Food Economy Assessment of Kosovo Province, July­September 1999. Final Report, WFP (Rome, November 1999). , Summary of Projected Food Aid Needs for Kosovo, April 2000­ March 2001. WFP (Prishtina, April 2000). Lawrence, Mark, and Raoul Balletto, Kosovo: Food Needs of the Albanian Majority Population, April 2001­March 2002. WFP/Food Security Surveillance Unit FSSU (Prishtina, June 2001). Lawrence, Mark, and Annalies Borrel, Joint WFP/UNHCR Food Needs Assessment of Minorities in Kosovo, November­December 1999. WFP/ UNHCR (Prishtina, 2000). Oxford Policy Management Group (OPMG) and CARE International UK, Kosovo: A Qualitative Poverty Assessment Data Analysis Report. OPMG (Oxford, January 2001). RIINVEST (Institute for Development Research), War Consequences on Fam- ily Economics and Businesses. Survey Outcomes. RIINVEST (Prishtina, December 1999). UNDP, Kosovo National Human Development Report 2002: Building Bridges to a Better Future. UNDP (Prishtina, June 2002). UNHCR and Organization for Security and Co-operation in Europe (OSCE), Ninth Assessment of the Situation of Ethnic Minorities in Kosovo, September 2001­April 2002. UNHCR/OSCE (Prishtina, 2002). Women's Commission for Refugee Women and Children, Internal Displace- ment in Kosovo: The Impact on Women and Children. Women's Commis- sion for Refugee Women and Children (New York, 1998). World Bank, Kosovo: Poverty Assessment Technical Papers. Europe and Central Asia Region Report 23390, World Bank (Washington, D.C., 2001). 178 bibliography , Kosovo: Transitional Support Strategy. Europe and Central Asia Region Report 21610, World Bank (Washington, D.C., 1999). , Kosovo: Transitional Support Strategy. Europe and Central Asia Region Report 21611, World Bank (Washington, D.C., 2000). Macedonia World Bank, Macedonia: Focusing on the Poor. Europe and Central Asia Region Report 19411, World Bank (Washington, D.C., 1999). , Macedonia: Interim Poverty Reduction Strategy Paper and Joint Assessment. Europe and Central Asia Region Report 21346, World Bank (Washington, D.C., 2000). Moldova De Soto, Hermine, and Nora Dudwick, Poverty in Moldova: The Social Dimensions of Transition, June 1996­May 1997. ECSSD Background Report, World Bank (Washington, D.C., 1997). Incredere Foundation, Evaluation of the Real Number of Internally Displaced Persons in the Republic of Moldova. UNHCR (Chisinau, 2000). Johnson, David, Moldova 2000 Food Security Program: Social Protection Sector Review. RESAL (Réseau Européen de Sécurité Alimentaire) (Brussels, April 2001). Nantoi, Oazu, Report on the Problem of Internally Displaced Persons in the Republic of Moldova. UNHCR (Chisinau, 1999). UNHCR Moldova, Report on the Republic of Moldova's Situation Related to the Refugees, Asylum Seekers, Stateless, Forcely [sic] Moved and Internally Displaced Persons as well as Persons Illegally Staying on the Territory of the State And Persons with Indeterminate Status. UNHCR (Chisinau, 1999), http://www.unhcr.md/artpdf/repwge_.htm. World Bank, Moldova: Interim Poverty Reduction Strategy Paper and Joint Assessment. Europe and Central Asia Region Report 21426, World Bank (Washington, D.C., 2000). , Moldova: Poverty Assessment Technical Papers. Europe and Central Asia Region Report 19846, World Bank (Washington, D.C., 1999). Russian Federation COE, Parliamentary Committee on Migration, Refugees and Demography. The Conflict in Chechnya. January 2000, Doc. 8632. DRC (Danish Refugee Council), Achkhoy-Martanovskiy District Household Survey: Survey of the Population of Achkhoy-Martanovskiy District, Republic of Chechnya, Russian Federation, as an Instrument for Providing Targeted Humanitarian Assistance. Interim Report, Draft 1, DRC (Nazran, Russia/Copenhagen, September 2001). bibliography 179 , Groznenskiy District Household Survey: Survey of the Population of Groznenskiy District, Republic of Chechnya, Russian Federation, as an Instrument for Providing Targeted Humanitarian Assistance. Interim Report, Draft 1, DRC (Nazran, Russia/Copenhagen, December 2001). , Grozny Household Survey: Survey of the Population of Grozny City, Republic of Chechnya, Russian Federation, as an Instrument for Providing Targeted Humanitarian Assistance. Interim Report, Final Version, DRC (Nazran, Russia/Copenhagen, September 2001). , Gudermesskiy District Household Survey: Survey of the Population of Gudermesskiy District, Republic of Chechnya, Russian Federation, as an Instrument for Providing Targeted Humanitarian Assistance. Interim Report, Draft 1, DRC (Nazran, Russia/Copenhagen, 2002). , Household Surveys in Chechen Republic: Survey of the Population of the Republic of Chechnya, Russian Federation, as an Instrument for Providing Targeted Humanitarian Assistance. Draft 1, DRC (Nazran, Russia/Copenhagen, January 2002). Hansen, Greg, and Robert Seely, War and Humanitarian Action in Chechnya. Occasional Paper 26, Brown University (Providence, R.I., April 1996). ICRC (International Committee of the Red Cross), Internally Displaced Persons in Ingushetia: Economic Security Survey. ICRC (Geneva, 2002). IOM, Resettlement of "Refugees" and "Forced Migrants" in the Russian Federation (Geneva: IOM, 1998). Médecins du Monde, Chechnya-Ingushetia: Testimonies and Sanitary Data. Press Kit, ReliefWeb (February 23, 2000), http://www.reliefweb.int. , Chechnya-Ingushetia: The Chechen Civilians' Despair. Press Kit, ReliefWeb (December 2000), http://www.reliefweb.int. Médecins Sans Frontières (MSF), Chechens Displaced in Ingushetia. MSF (Moscow, August 2000). , Chechnya-Ingushetia: A Deliberate Strategy of Non-Assistance to People in Crisis (Moscow, February 2002). , Chechnya-Ingushetia: Vulnerable Persons Denied Assistance. MSF (Moscow, January 2002). Trier, Tom, and Kharon Deniev, Chechnya in Figures: Survey of the Popula- tion of Chechnya, Russian Federation, as an Instrument for Providing Targeted Humanitarian Assistance. Interim Report. Final Version, DRC/ ASF­Danish People's Aid (Nazran, Stavropol, Russia, October 2000). United Nations in the Russian Federation, Northern Caucasus Humanitarian Action. Briefing Kits, ReliefWeb (2001), http://www.reliefweb.int. UNICEF, A Humanitarian Appeal for Children and Women, January­ December 2002: Northern Caucasus. UNICEF (New York, 2001). World Bank, Poverty in Russia: An Assessment. Europe and Central Asia Region Report 14110, World Bank (Washington, D.C., 1995). 180 bibliography , Russia: Consultations with the Poor. National Synthesis Report, Europe and Central Asia Region, World Bank (Washington, D.C., 1999). http://www.worldbank.org/poverty/voices/reports/national/russia.pdf. , Russia: Targeting and the Longer-Term Poor. Europe and Central Asia Region Report 19377, World Bank (Washington, D.C., 1999). WFP, WFP Emergency Operation 6197.01--Emergency Food Assistance to Displaced Persons and Vulnerable Groups in the Northern Caucasus. WFP (Rome, 2000). World Health Organization (WHO), UNICEF, and IOM, Health Assessment of Internally Displaced Population from Chechnya in Ingushetia. WHO/ UNICEF/IOM (Magas, Ingushetia, Russian Federation, January/February 2000). http://www.db.idpproject.org/Sites/IdpProjectDb/idpSurvey.nsf/ wViewCountries/7875EC64D5CC3B0DC12568C10052EACD. Serbia and Montenegro European Agency for Reconstruction (EAR), Study and Activity Identification for Refugee and IDP Local Economic Integration. EAR (Belgrade, 2002). Republic of Serbia, National Strategy for Resolving the Problems of Refugees: Summary for Donors. Government of Serbia (Belgrade, June 2002). Ilic, Vladimir, Yugoslavia: Displacement from Kosovo: From Patronage to Self-Help. In Caught between Borders: Response Strategies of the Internally Displaced, ed. Marc Vincent and Birgitte Refslund Sorensen (London: Pluto Press, 2001). ICRC, Beneficiary Monitoring: A Sample Survey of IDP Households in FRY. ICRC (Geneva, December 2001). International Council of Voluntary Agencies (ICVA), Survey--Vojvodina: New Aspects of Refugee Issues. ICVA (Belgrade, October 2001). IFRC, ICRC, and YRC (Yugoslav Red Cross), FRY: IDPs and Refugees Living Conditions. IFRC/ICRC/YRC (Belgrade, April­May 2000). Lowery, Kerry-Jane, A Study of the Internally Displaced People of Kosovo Living in Serbia and Montenegro. Final Report, ICRC Background Paper (Belgrade, February 2001). Matkovic, Gordana, and Borka Vujnovic, War Affected Persons in the FR of Yugoslavia. Stanovnistvo, 34 (3­4) (July­December 1996): 7­41. UN Office for the Coordination of Humanitarian Affairs (OCHA), Humani- tarian Situation, Protection and Assistance: Internally Displaced Persons in Serbia and Montenegro. OCHA (Belgrade, April 2002). , Income and Expenditures in Montenegro (Belgrade: OCHA, 2000). UNDP, Vulnerability Trends and Perception: Suspended Transition, 1990­2000. Report No. 0 (Serbia) of the Early Warning System for FRY, UNDP (Brussels, August 2000). UNHCR, Census of Refugees and Other War-Affected Persons in the Federal Republic of Yugoslavia (Belgrade: UNHCR, 1996). bibliography 181 UNHCR and Commissioner for Refugees of the Republic of Serbia, Refugee Registration in Serbia. Final Report. Draft Report, UNHCR (Belgrade, March­April, 2001). , Registration of Internally Displaced Persons from Kosovo and Metohija. UNHCR (Belgrade, December 2001). Women's Commission for Refugee Women and Children, Refugee and Inter- nally Displaced Women and Children in Serbia and Montenegro. Women's Commission for Refugee Women and Children (New York, September 2001). World Bank, Federal Republic of Yugoslavia: Transitional Support Strategy. Europe and Central Asia Region Report 22090, World Bank (Washington, D.C., 2001). WFP, Emergency Food Assistance in the Balkan Region--EMOP 6136.02. WFP (Belgrade, December 2000). , The Survey on Poverty and Living Conditions of Refugee Beneficia- ries of WFP Aid in Serbia. WFP (Belgrade, April 2001). Tajikistan Birkenes, Robert, Tajikistan: Survey of Household and Bazaar Economies. Save the Children (Dushanb, January 1996). Counterpart Consortium Tajikistan, NGO Poverty Assessment: Tajikistan. Final Report, Counterpart Consortium (Dushanbe, 2001). Falkingham, Jane, Welfare in Transition: Trends in Poverty and Well-Being in Central Asia. CASE paper 20, Center for Analysis of Social Exclusion (CASE), London School of Economics (London, 1999). , Women and Gender Relations in Tajikistan. Programs Department and Social Development Division, Office of Environment and Social Devel- opment, Asian Development Bank, Country Briefing Paper (London, 2000). Foroughi, Payam, Hope and Despair: An Assessment of Returnee Households in Khatlon. Background Paper, UNHCR (Dushanbe, 1998). , 1998 Socio-Economic Survey of Households, Farms, and Bazaars in Tajikistan. Save the Children Foundation (Dushanbe, January 1999). Freckleton, Ann, Who's Needy: An Assessment of Household Food Insecurity in Tajikistan. ECHO (Dushanbe, August 1997). Gomart, Elizabeth, Tajikistan Pilot Poverty Alleviation Program. Social As- sessment Field Work Report, World Bank (Washington, D.C., July 1996). McLean, Jennifer, Filling a Void: Responding to Internal Displacement in Tajikistan. Central Asia Monitor 7 (3) (1998). McLean, Jennifer, and Thomas Greene, Turmoil in Tajikistan: Addressing the Crisis of Internal Displacement. In The Forsaken People: Case Studies of the Internally Displaced, ed. Roberta Cohen and Francis Deng (Washington, D.C.: Brookings Institution Press, 1998). 182 bibliography Schumacher, Britta, Nutrition, Health, and Food Security Survey: In Selected Districts of Leninabad Region and the Regions of Republican Subordina- tion, Tajikistan. Background paper, Deutsche Welthungerhilfe (Dushanbe, December 1998). United Nations General Assembly, Human Rights Questions: Human Rights Situations and Reports of Special Rapporteurs and Representatives. Inter- nally Displaced Persons. Addendum: Profiles in Displacement--Tajikistan, report by Francis Deng, 51st sess., October 1996 (A/51/483/Add.1). UNHCR, Report on Tajikistan: January 1993­March 1996. UNHCR (Geneva, 1996). World Bank, Pilot Poverty Alleviation Project: Integrated Evaluation Report of Baseline Study. World Bank (Washington, D.C., July 1998). , Tajikistan: Interim Poverty Reduction Strategy Paper and Joint Assessment. Europe and Central Asia Region Report 21180, World Bank (Washington, D.C., 2000). , Tajikistan: Poverty Assessment Technical Papers. Europe and Central Asia Region Report 20285, World Bank (Washington, D.C., 2000). Turkey COE, Parliamentary Committee on Migration, Refugees, and Demography, The Humanitarian Situation of the Displaced Kurdish Population in Turkey, Report, February 18, 2002. Human Rights Watch (HRW), Turkey's Failed Policy to Aid the Forcibly Displaced in the Southeast. Human Rights Watch 8 (9) (D) (June 1996), http://hrw.org/reports/1996/Turkey2.htm. McDowall, David, Ilisu: The Economic and Political Context. Ilisu Dam Campaign (Oxford, UK), http://www.ilisu.org.uk/david.html. Turkish Daily News, Ankara, Online Edition, http://www.turkishdailynews. com. United States Committee for Refugees, The Wall of Denial: Internal Displacement in Turkey. Immigration and Refugee Services of America (Washington, D.C., 1999). World Bank, Turkey: Economic Reforms, Living Standards and Welfare Study. Report 20029, World Bank (Washington, D.C., 2000). , Turkey: Unemployment, Migration and Wages in Turkey, 1962­85. Working Paper series 230, World Bank (Washington, D.C., July 1989). Index A length of displacement, 3 Abkhazia, see Georgia material well-being and poverty access issues levels, 22, 35n1 community-targeted development numbers of DPs, 6 programs, DP access to, 161 public assistance, 147n2 denial of access to assets, xiv settlement patterns, 64 land, access to, see agricultural social capital and social employment and access to land networks, 113 public assistance and social Angola, 85, 100n11 services, 157­59 area-based development utilities, access to, 76­77, 156 programming, 159­61 Afghanistan, 50, 78n4, 85, 91, 94, Armenia, xi, 12, 13 100n11, 146 citizenship status for refugees, African DPs, 50, 78n4, 82, 85, 94, 124, 126­27, 145 100n11, 146, see also specific education, 83, 84, 86­87 countries employment/unemployment, 44, age groups, impacts of displacement 48, 54, 59n25 on, 82­83, 96, 97­98, 142 government interactions with agricultural employment and access DPs, 124, 126­29, to land, xiii, 17, 43­47 133n10, 145 access to agricultural assets in health, 87 home region, 44 housing, 68, 71­72, 78n6 landless laborers, 46 IDA funding eligibility, 127 legal inability to own land, 52­53 length of displacement, 3 moveable agricultural assets, 42 material well-being and poverty privatization of land, effect of, 43 levels, 23, 27, 37n24­25 Albania mortality rates of DPs, 82 housing, 68, 73 numbers of DPs, 6, 14 humanitarian assistance benefiting psychological impacts of host countries, 147n1 displacement, 89, 94­95 183 184 index Armenia (continued) Bosnia and Herzegovina, xi, 5, right to own land and IDP 11, 12 status, 126­27 conflict-related trauma, 90, settlement patterns, 65 101n15 Asian DPs, see specific countries employment/unemployment, 40, assets, see human capital; material 44, 45, 48, 51 impacts of displacement female-headed households, 95, asymmetric vs. symmetric 101n19 displacement, 72, 78n6, government interactions with DPs, 78n10 127, 128, 130, 132n2, Azerbaijan, xi, 1, 12, 13 133n10 agricultural land allocated to housing, 67, 70, 71, 74, 162 IDPs, 52 humanitarian assistance, 157 dependency syndrome, 91 IDA funding eligibility, 127 education, 83, 86­87, 100n6 length of displacement, 3 employment/unemployment, material well-being and poverty 40­46, 48, 51­52, 54, 57, levels, 23­25, 29, 31, 37n26 58n2, 59n32 mortality rates of DPs, 82 government interactions with "municipal governments in DPs, 124, 126­28, 130, exile," 128 132n3, 134n18 numbers of DPs, 6, 14 health issues, 87­88, 158 psychosocial factors, 95, 141 housing, 68, 71­72, 76, 78n6 return of DPs, 146 IDA funding eligibility, 127 settlement patterns, 64, 65 length of displacement, 3 social capital and social networks, material well-being and poverty 114, 117n2 levels, 23, 24, 26, 28­30, 32, statistics on DPs, reliability of, 34, 37n25, 38n35, 138 166, 167n1 mortality rates of DPs, 82 voting rights, 130 "municipal governments in Burundi, 85 exile," 128 numbers of DPs, 6, 14 C propiska system, 125 Cambodia, 78n4, 100n11 psychological impacts of capital, see human capital; material displacement, 89, 94­95 impacts of displacement right to own land and IDP CDD (community, municipal, and status, 126­27 area-based development), settlement patterns, 65 159­61 social capital and social networks, Central Asian DPs, see specific 106, 109, 111 countries statistics on DPs, reliability Cernea, Michael, 15 of, 166, 167n1 charts, see tables, charts, and voting rights, 130 figures Chechens, xi, xiii, 2, 9 B education, 100n8 Balkans, see specific countries employment/unemployment, 43 birthrates of DPs, 83 freedom of movement, 125 index 185 government interactions with numbers of DPs, xi, 1, 2, 5­9, 14 DPs, 122, 125, 130 objectives and scope of study, 4­5, housing, 68, 75, 79n21, 79n23 16­18 length of displacement, 3 organization of study, 16­18 material well-being and poverty return of DPs from, 5, 145­46, levels, 25, 28, 31 153­54, 167 mortality rates, 82 sources of information, xii­xiii, numbers of DPs, 7 17, 165 psychosocial factors, 141 specific situational public assistance, 147n4 dynamics, xiv­xv settlement patterns, 64­65 statistics on DPs, reliability social capital and social of, 150­51, 165­67 networks, 106, 118n4 conflict-related trauma, 90­91, 141 statistics on DPs, reliability of, 166 conflict's impact on social voting rights, 130 networks, 107­8 child labor, 58n3 Convention on Refugees, 1951, children and youth as DPs 11, 120 age groups, impacts of credit barriers to employment, 55­56 displacement on, 97­98 Croatia, 12 demographics, 82­83 employment/unemployment, 51 dependents in family groups, government interactions with ratios of, 82 DPs, 132n2, 133n10 education, 83­87 housing, 72, 74 single-parent households, 82, length of displacement, 3 95­97, 101n19 material well-being and poverty citizenship, see legal status levels, 23, 24, 25, 28 collective centers, xiii­xiv, numbers of DPs, 6, 14 68­69 return of DPs, 146 capacity problems, 132n2 settlement patterns, 64 differing situations by country, 67 social capital and social health issues, 88, 155 networks, 107 privatization, effect of, 132n3, cultural barriers 163 age groups differently affected rental accommodations vs., by, 98 74­75, 79n21 education, 85­86 vulnerabilities of, 140 employment/unemployment, community-based development 54­55 programming, 159­61 conflict-induced displacement, see D also refugees vs. IDPs Dayton Peace Accords, 25, characterization of DPs, xii, 95, 146 11­13 deconcentration of DPs, 63, 77, length of time displaced, xi, xiv, 108­10, 159­60 xv, 1­3, 94­95, 115, demographics of DPs, xiii, 18, 146, 167 81­83, 95­98, 150­51 methodological issues, 17, Deng, Francis M., 8 165­67 denial of access to assets, xiv 186 index dependency syndrome, 56­57, cultural barriers, 54­55 91­92, 101n16 dependency syndrome, 56­57, 91 dependents in family groups, ratios elderly, income contributions of, 82 of, 83 depoliticization of formal employment, 40­43 displacement, 144­47 housing, 63, 76, 162 deprivation, poverty as, 13 informal ("gray market") destruction of assets, xiv employment, 40, 58n3, development programming, 10, 60n39, 121­22, 133n11 127, 149 language barriers, 54­55 CDD (community, municipal, and legal issues area-based development), agricultural land, legal inability 159­61 to own, 52­53 housing, 151, 162­64 freedom of movement, 125 humanitarian assistance, professional and legal status 154­57, 159 difficulties, 53­54, "living in limbo," impact of, 153 133n12, 155 poverty diagnostics and unemployment registration, tools, 149­54 133n12, 155 public assistance, 154­59 networks, formal and informal, 55 vulnerability, 149, 152­53, 155 obstacles to employment of displaced persons (DPs), see conflict- DPs, 49­57 induced displacement psychological impacts of diverse/mixed communities, 107 displacement, 56­57, 89, 93 Doctors Without Borders, 88 public/government employment documentation problems, 124, 151, during displacement, 41, 42, 155, 158, 161, see also legal 128, 135, 138 status registration for unemployment DPs (displaced persons), see conflict- benefits, 133n12, 155 induced displacement self-employment, 55­56 duration of displacement, xi, xiv, xv, socioeconomic dislocation, 1­3, 94­95, 115, 146, 167 problems arising from, 53 subnational settlement environ- E ments, socioeconomic East Asian DPs, 78n4, 100n11, see weaknesses of, 51­52 also specific countries unemployment rates for DPs, education, xiii, 18, 83­87, 93, 106, 47­49, 138­39 157­58 environment, 145 elderly DPs, 82­83, 96, 97 ethnic groups displaced, see also employment/unemployment, xiii, specific ethnic groups, 17­18, 39­60 countries of origin, and host agricultural employment, see countries agricultural employment and length of displacement by ethnic access to land group and host country, 3 composition of DPs' household numbers of DPs by host country income, 134 and ethnic group, 6­7 credit barriers, 55­56 European DPs, see specific countries index 187 F numbers of DPs, 6, 14 family and kin psychological impacts of dependents in family groups, displacement, 89 ratios of, 82 public assistance, 147n2 housing shared with relatives and right to own land and IDP friends, 63, 70, 72­74, 136, status, 126 140, 147n3, 151, 155 settlement patterns, 64, 65 marriage and family alliances, 93, social capital and social networks, 116 107, 109, 111, 113, 142 single-parent households, 82, statistics on DPs, reliability 95­97, 101n19 of, 166 survival strategy, family as, 139 voting rights, 130 female DPs, 82, 95­97, 134n15 women entrepreneurs, 96­97 figures, see tables, charts, and figures government employment during Former Yugoslav Republic (FYR), displacement, 41, 42, 128, see Yugoslavia, dissolution of, 135, 138 and also specific succeeding government interactions with DPs, states 119­34, 143­44, see also FYR (Former Yugoslav Republic), host countries see specific succeeding states depoliticization of, 144­47 financial and institutional capacity G of state to care for DPs, 121 gender groups, impacts of displace- legal status, see legal status ment on, 82, 95­97, 141 local government institutions, geographic distribution of DPs, see 128­31 settlement patterns material impact of DPs on host Georgia, xi, xiv, 9 countries, 22, 121, 131, 135 dependency syndrome, 91 national institutions to education, 83­84, 100n6 handle, 127­28 employment/unemployment, political factors, 122­24, 144­47 40­44, 46, 48, 51­52, psychosocial factors, 120 54, 57 public assistance, 123, 132n3, government interactions with DPs, see also public assistance 122, 126­30, 132n3, refugees vs. IDPs, see refugees vs. 134n15­16 IDPs health issues, 87­88, 158 responsibility for DPs, government housing, 36n6, 68, 70­76, 79n24, sense of, 121­22 136 self-reliance and integration, humanitarian assistance benefiting effect on, 121, 123, 124, host countries, 147n1 128, 131 IDA funding eligibility, 127 social capital and social networks, length of displacement, 3 104, 118n13 material well-being and poverty voting rights, 129­30, 134n21, levels, 23, 25, 26, 29, 30, 161 32­34, 36n6 "gray market" informal employment, "municipal governments in 40, 58n3, 60n39, 121­22, exile," 128 133n11 188 index H private accommodations, 69­77 health, xiii, 18, 87­88, quality of housing and quality of 100­101n11, 158 life, 63, 75­76 Herzegovina, see Bosnia and refugee camps, 66­67, 69, 79n23 Herzegovina relatives and friends, living with, Holzmann, Robert, 15 63, 70, 72­74, 136, 140, host countries, 143­44, see also 147n3, 151, 155 government interactions with rental accommodations, 62, DPs; specific countries 74­75 housing policy of, 64, 65 space per person, DPs vs. locals, housing provided by governments 76, 79n24 of, 33­34 utilities, access to, 76­77, 156 humanitarian assistance variations in circumstances of benefiting, 147n1, 159 differences within one impact of DPs on, 8, 63, 66 country, 70 length of displacement by ethnic differing situations by country, group and host country, 3 67­68 material impact of DPs on, 22, vulnerability factors, 139­40 121, 131, 135, 159 human capital, xiii, xiv, 18, 81­101, numbers of DPs by host country 140­41 and ethnic group, 6­7 age groups, 82­83, 96­98, 142 private transfers to DPs from host demographics, xiii, 18, 81­83, households, 33 95­98, 150­51 social capital and social networks, education, xiii, 18, 83­87, 93, 112­15, 136 106, 157­58 subnational settlement environ- gender groups, 82, 95­97, 141 ments, socioeconomic health, xiii, 18, 87­88, weaknesses of, 51­52 100­101n11, 158 housing, xiii­xiv, 18, 61­80, 67­68, mobility of, 84 135, 139­40, see also psychosocial issues, see collective centers psychological impacts of development programming, 151, displacement 162­64 humanitarian assistance, xi­xii, xiii, employment/unemployment xv, 4, 9­11, 138­39 affected by, 63, 76, 162 decline in, 136, 138, 145 fragility of situations, 63 dependency syndrome, 91 health issues, 88 development planning, 154­57 home ownership, 62, 70­72, 163 development programming, 159 host government policy on, 64, 65 elderly, income contributions host governments providing, of, 83 33­34, 136 employment/unemployment, 39, initial shelter situations, 62­63 60n40 "living in limbo," impact of, 63, host countries aided by, 147n1, 162­63 159 material well-being and poverty material impacts of levels, 36n6 displacement, 30­31 poverty and, 162 mortality rates of DPs, 83 index 189 national institutions, 127 housing, 62, 68, 70, 73 settlement patterns and housing, length of displacement, 3 66, 67 material well-being and poverty levels, 23­25, 29­30, 35n2 I numbers of DPs, 7 ICRC (International Committee for private remittances, 32 the Red Cross), 88 Roma, 3, 7, 24, 54, 85, 114, IDA (International Development 118n4 Association), 127 settlement patterns, 64, 65 IDPs (internally displaced persons), social capital and social networks, see refugees vs. IDPs 113, 118n14 India, 78n10 Krajina, see Croatia informal ("gray-market") employ- Kurds, xiii, 3, 12 ment, 40, 58n3, 60n39, employment/unemployment, 45, 121­22, 133n11 47, 54 Ingushetia, see Chechens gender-based language barriers, integration and self-reliance, 145­47 96, 101n22 development programming, housing, 67, 73 154­57 length of displacement, 3 distinction between, 133n7, 134n17 mortality rates, 82 government interactions with DPs, moveable agricultural effect of, 121, 123, 124, 128, assets, 45 131 numbers of DPs, 7 internally displaced persons (IDPs), social capital and social see refugees vs. IDPs networks, 106 International Committee for the Red statistics on DPs, reliability Cross (ICRC), 88 of, 167 International Development Association (IDA), 127 L international humanitarian aid, see land, agricultural, see agricultural humanitarian assistance employment and access to Iran, 78n4 land land ownership rights J agricultural land, legal inability to "Janus-faced dilemma of DPs," own, 52­53 15­16, 153 community-targeted development Jorgensen, Steen Lau, 15 programs, DP access to, 161 housing and home ownership, 62, K 70­72, 163 Kosovo and Kosovars landmines, 100­101n11, 145 demographics of DPs, 83 language barriers education, 85 education, 85­86 employment/unemployment, 41, employment/unemployment, 42, 48, 51, 54, 58n5, 133n12 54­55 government interactions with DPs, gender-based, 96, 101n22 122, 128, 132n2, 132n4, social services, access to, 158 133n12 leadership, 111 190 index legal status, 124­27 numbers of DPs, 7, 14 documentation problems, 124, settlement patterns, 64, 65 151, 155, 158, 161 Malawi, 85 freedom of movement, 125­26 male DPs' identity and perceptions land, right to own, see land of self-worth, 97, 141 ownership rights marriage length of time displaced and legal family alliances and, 93, 116 status, 94­95 legal status affected by, 134n15 loss of IDP status, consequences material impacts of displacement, of, 126­27 xiii, xiv, 17, 21­38, 137­38, marriage affecting, 134n15 see also housing professional and legal status diffi- composition of DPs' household culties, employment problems income, 134 due to, 53­54, 133n12, 155 consumption expenditures and refugees and IDPs, 124, 126­27, poverty levels, 23­27, 134n15­16, 163 137­38 unemployment registration, host country problems, 22, 121, 133n12, 155 131, 135, 159 vulnerabilities due to problems humanitarian assistance, 30­31 with, 119 private remittances, 31­33 length of time displaced, xi, xiv, xv, psychosocial factors and, 89 1­3, 94­95, 115, 146, 167 public assistance, reliance on, linking capital, 104, see also govern- 28­30, 33­34 ment interactions with DPs Médecins Sans Frontiéres, 88 "living in limbo," impact of, xi, xiii, Meshketian Turk refugees, see xv, 3­4, 15, 92­95, 142­43 Azerbaijan development planning, 153 methodological issues, 17, 165­67 education, 93 microfinance programs, 31, 33, 45, employment/unemployment, 56, 55­56, 160 93 mixed/diverse communities, 107 essence of being DP, 100 Moldova, xiii housing, 63, 162­63 female-headed households, 95 length of time displaced, 94­95 statistics on DPs, reliability loss of IDP status, consequences of, 167 of, 126­27 Montenegro, see Serbia and material well-being, relationship Montenegro to, 89 mortality rates of DPs, 82­83 social capital and social Mozambique, 85 networks, 115­16 municipal development local government programming, 159­61 institutions, 128­31 "municipal governments in exile," 128­31 M Macedonia, FYR N government interactions with Nagorno-Karabakh, see Azerbaijan DPs, 122 networks, formal and informal length of displacement, 3 employment/unemployment, 55 index 191 social networks, see social capital psychological impacts of displace- and social networks ment, xiv, 18, 88­95, nongovernmental organizations 141­42, see also "living in (NGOs) limbo," impact of assistance programs, 33 cause and effect, 89 employment/unemployment, conflict-related trauma, 90­91 40n60 dependency syndrome, 56­57, new social networks, formation 91­92, 101n16 of, 111 employment/unemployment, settlement patterns and housing, 67 56­57, 89, 93 sources of data, xii, 165 government interactions with numbers of DPs, xi, 1, 2, 5­9, 14 DPs, 120 housing, 63 P "Janus-faced dilemma of DPs," Pakistan, 78n4, 78n10, 85, 91, 94 15­16, 153 political factors in government public assistance, xiii, 135­37 interactions with DPs, composition of DPs' household 122­24, 144­47 income, 134 poverty, xiii, 9­11, 13­17, 21­22, delivery of services to DPs, 167n2 157­59 development programming, diag- dependency syndrome, 91 nostics and tools for, 149­54 development programming, education, 87 154­59 host countries, problems faced elderly, income contributions by, 22 of, 83 housing problems and, 162 employment/unemployment, 39 human capital and, 81 government interactions with material well-being, consumption, DPs, 123, 132n3 and poverty levels, 23­27, housing provided by host govern- 137­38 ments, 33­34, 136 psychosocial factors, 89 integration of DPs into social social capital and social networks safety nets, 154­57 affected by, 111­12 material impacts of displacement, transitional poverty in successor 28­30, 33­34, 138 countries of Soviet Union and needs tested and vulnerability Yugoslavia, 22 based, 156­57 Poverty Reduction Strategy Papers registration for unemployment (PRSP) process, 1, 10, 25, 127 benefits, 133n12, 155 private remittances, DP reliance regularity and timeliness of on, 31­33 delivery, 147n2 privatization policies, 43, 71, 126, public authorities, see government 132n3, 162, 163 interactions with DPs; host professional status countries difficulties, 53­54 public employment during displace- PRSP (Poverty Reduction Strategy ment, 41, 42, 128, 135, 138 Papers) process, 1, 10, public information campaigns about 25, 127 DP programs, 156 192 index R health, 87­88 Red Cross, 88 housing, 67, 68, 70­75, 136, 140 refugee camps, 66­67, 69, 79n23, IDA funding eligibility, 127 see also housing length of displacement, 3 refugees vs. IDPs, xii, 4­5, 11­13, material well-being and poverty 119­20 levels, 23, 24, 28­31, 33, government sense of 37n29 responsibility, 121 "municipal governments in legal status, 124, 126­27, exile," 128 134n15­16 numbers of DPs, 6, 7, 14 political factors, 122­24 return of DPs, 146 trust issues, 110 right to own land and IDP relief agencies, see humanitarian status, 126 assistance settlement patterns, 64 return of DPs, 5, 145­46, 153­54, social capital and social networks, 167 107, 111, 113, 114, 118n14, Roma, 3, 7, 24, 54, 85, 114, 118n4 142 Russia, xi, 13, see also Chechens statistics on DPs, reliability of, numbers of DPs, 14 166, 167n1 Rwanda, 78n4, 82, 85, 94, 146 voting rights, 134n21 WFP, 155 S settlement patterns, 61­62, 63 self-employment, 55­56 communities, treating DP self-reliance and integration, settlements as, 159­60 145­47 conflict zones, clustering near, development programming, 64­65 154­57 deconcentration of DPs, 63, 77, distinction between, 133n7, 108­10, 159­60 134n17 government sense of responsibility government interactions with DPs, for DPs and, 121 effect of, 121, 123, 124, size of DP settlements, 114 128, 131 social capital and social networks Serbia and Montenegro, xi, xiv, 1, 5, affected by 8, 12, 18n3, see also Kosovo host community and Kosovars relationships, 114­15 citizenship status for refugees, social networks within DP 124, 133n10, 145 communities, 106­7 dependency syndrome, 91 socioeconomic environment of education, 84, 85 settlement area, 114­15 employment/unemployment, 40, urban centers, 65­66 42, 45, 47, 48, 51, 57, vulnerability of DPs, implications 133n12 for, 66­67 female-headed households, 95 sex, DPs considered by, 82, 95­97, government interactions with DPs, 141 124, 126­29, 132n2, shelter, see housing 133n9­10, 133n12, 134n21, single-parent households, 82, 95­97, 145 101n19 index 193 social assistance, see public material well-being and poverty assistance levels, 22 social capital and social networks, nostalgia for socialist past, 1, 12, 103­5, 116­17, 142­43, 92­93xii see also family and kin propiska system, 54, 125 choices regarding, 115­16 refugee/IDP distinction, 121 conflict's impact on, 107­8 social capital and social networks, defined, 103 85, 106, 112, 118n7 employment/unemployment, 53 space per person, DPs vs. locals, 76, factors undermining, 110 79n24 host countries, 112­15, 136 Srpska, Republic of housing, 63, 66, 76, 147n3 housing, 72 leadership, 111 material impacts of length of time displaced, 115 displacement, 25 levels of cohesion within social settlement patterns, 65 networks, 106 state responses to DPs, see "living in limbo," impact government interactions of, 115­16 with DPs mixed/diverse communities, 107 statistics on DPs, reliability of, new networks, formation of, 111 150­51, 165­67 postdisplacement social cohesion, survival strategies of DPs, 135­37, 108­10 139 poverty and, 111­12 symmetric vs. asymmetric predisplacement social capital displacement, 72, 78n6, host countries, links 78n10 with, 113­14 within DP communities, 105­6 T resilience of, 108 tables, charts, and figures settlement patterns affecting housing host community relationships, home ownership rates, DPs vs. 114­15 locals, by country, 71 social networks within DP private vs. collective housing, by communities, 106­7 country, 68 trust issues, 110­11 length of displacement, 3 within DP communities, numbers of DPs, 2, 6­7, 14 105­12 poverty levels, 23 socioeconomic environment, unemployment rates by country dislocation from, xiv, 53 and DP status, 49 Somalia, 85 Tajikistan, xiii South Ossetia, see Georgia length of displacement, 3 Soviet Union, dissolution of, xii, 1, numbers of DPs, 7 12, see also Russia and other statistics on DPs, reliability resultant countries of, 167 employment/unemployment, Tanzania, 85 47, 54 Thailand, 78n4 housing and privatization policies, trauma, conflict-related, 90­91 71, 78n9, 162 trust issues, 110 194 index Turkey, xi, xiii, xiv, 1, 8, see also housing, 61, 77n1, 139­40 Kurds human capital and, 81 child labor, 58n3 mitigating factors, 136­37 dependency syndrome, 91 multifaceted nature of, xi, 61, health, 87 135, 139 housing, 70, 72, 73, 136 multiplier effect of displacement, material well-being and poverty 152­53 levels, 22, 31, 33, 35n1 needs testing, 156­57 numbers of DPs, 7, 14 psychosocial factors, 89 public assistance, 147n4 settlement patterns, 66­67 settlement patterns, 65 social capital and social W networks, 106, 142 welfare, see public assistance statistics on DPs, reliability Western relief agencies, see humani- of, 166­67 tarian assistance WFP (World Food Programme), 15, U 155, 157 Uganda, 85, 94 women, 82, 95­97, 134n15 unemployment, see employment/ World Bank, 10, 15, 157, 162, 165 unemployment World Food Programme (WFP), 15, United Nations, 8, 10, 11, 165 155, 157 United States Agency for International Development Y (USAID), 76 Yugoslavia, dissolution of, xii, 1, 12, United States Committee for see also resultant states Refugees (USCR), 13 employment/unemployment, 47 U.S.S.R., see Soviet Union, dissolu- housing and privatization policies, tion of, and also Russia and 71, 78n9, 162 other resultant countries material well-being and poverty utilities, access to, 76­77, 156 levels, 22 nostalgia for socialist past, 1, 12, V 92­93xii voting rights, 129­30, 134n21, 161 refugee/IDP distinction, 121 Vukovar, see Croatia social capital and social vulnerability, xi­xv, 4, 9­11, 13­17, networks, 85, 106, 112 135­37, 139, see also poverty development programming, 149, Z 152­53, 155 Zambia, 85 host country populations, 22, 121, 131, 135, 159 T he conflicts in Europe and the former Soviet Union of the 1990s left in their wake nearly 10 million refugees and internally displaced persons. Even where peace treaties or ceasefires have brought an end to open conflict, about half of those originally displaced remain caught between two worlds.They cannot go home.At the same time, they face a range of difficulties in settling where they are. Long-term, conflict-induced displacement creates both conceptual and operational challenges for development agencies concerned with poverty reduction. Living in Limbo analyzes the special nature of displacement- induced vulnerability along several dimensions, including material well- being, employment, shelter, and human and social capital.The study draws on the authors' field work as well as extensive review of surveys, studies, and poverty assessments in 13 countries. A detailed analysis of the causes and characteristics of displacement- induced vulnerability, Living in Limbo provides pragmatic operational recommendations for policymakers and practitioners in both development and humanitarian agencies. ISBN: 0-8213-5850-2