A W O R L D B A N K S T U D Y Roma I N A N E X P A N D I N G EUROPE B r e a k i n g t h e P o v e r t y C y c l e Dena Ringold Mitchell A. Orenstein Erika Wilkens EDITION CONFERENCE C O N F E R E N C E E D I T I O N Roma in an Expanding Europe: Breaking the Poverty Cycle By Dena Ringold Mitchell A. Orenstein and Erika Wilkens A World Bank Study Copyright © 2003 The International Bank for Reconstruction and Development /THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, USA All rights reserved The findings, interpretations, and conclusions expressed here are those of the authors and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. Cover design by Joyce Petruzzelli Typography by Don Rau, Word Design, Inc. Cover photos by Mathieu Mazza and Dena Ringold ii TABLE OF CONTENTS FOREWORD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .viii OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Context and Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 The Environment for Policy Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 Policy Implications and Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Roma in the Transition Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 Roma Diversity, Culture, and Social Exclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16 Policy Approaches and Debates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24 2 ROMA POVERTY AND WELFARE: AN OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26 Measurement Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26 An Analysis of Roma Poverty in Three Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 Housing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32 Labor Market Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 Education Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37 Health Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 3 POVERTY AND EXCLUSION: ROMA SETTLEMENTS IN THE SLOVAK REPUBLIC . . . . . . . . . .47 Roma in Slovakia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47 Poverty in Roma Settlements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50 Economic Activities and Coping Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57 Access to Social Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61 Addressing Poverty in Settlements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66 iii 4 THE DIVERSITY OF ROMA IN ROMANIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68 From Slavery to Ceaucescu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68 Diversity of Roma Settlements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70 Income Sources and Economic Opportunities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72 Access to Social Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75 Social and Ethnic Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86 Current Policy Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .88 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90 5 PROJECT EXPERIENCE IN HUNGARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .92 A Regional Frontrunner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .92 Government Policies and Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .94 Project Experience: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98 Project Case Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102 Project Experience: Lessons Learned . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .107 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .109 6 ROMA IN SPAIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112 Roma in Spain: A Western European Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112 Living Conditions of Spanish Roma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .118 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123 7 THE ROAD AHEAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125 The Nature of Roma Poverty and the Policy Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125 Policy Implications and Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .127 Policy Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .133 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .139 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .141 LIST OF TABLES Table 1.1: A Typology of Roma Policy Approaches in Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Table 2.1: Poverty Rates among Roma and Non-Roma Households, 2000 . . . . . . . . . . . . . . . . . .28 Table 2.2: Roma Population Sizes by Type of Identification, 2000 . . . . . . . . . . . . . . . . . . . . . . . . .29 Table 2.3: Main Poverty Correlates, 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30 Table 2.4: Housing Characteristics by Ethnicity, 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34 Table 2.5: Educational Attainment by Ethnicity, 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38 Table 4.1: Main Features of the Case Study Sites, 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71 Table 4.2: Roma Population in Case Study Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72 Table 4.3: Income Sources and Economic Stratification, 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . .74 Table 4.4: Roma School Attendance Trends, National, 1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76 Table 4.5: Share of Roma Students in Valcele, 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76 Table 4.6: Access to Education in the Case Study Communities, 1999 . . . . . . . . . . . . . . . . . . . . .77 Table 4.7: Roma Access to Local Medical Services in the Case Study Communities, 1999 . . . . .81 Table 4.8: Access to Social Assistance in Case Study Communities, 1999 . . . . . . . . . . . . . . . . . . .84 iv Table 5.1: Hungarian Minorities in Other Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93 Table 5.2: Government Budgetary Support for Minorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 Table 5.3: Roma MSG Local Development Initiatives, 1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97 Table 5.4: Project Inventory, 1990­1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .99 Table 5.5: Projects by Region, 1990­1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .99 Table 5.6: Distribution of Projects by Sector, 1990­1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101 Table 5.7: Projects by Donors, 1990­1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101 Table 5.8: Program and Activities Supported under the Social Land Program . . . . . . . . . . . . . .105 Table 6.1: Roma Population Estimates, Selected Western European Countries . . . . . . . . . . . . . .114 Table 6.2: Roma Population per Autonomous Community (estimates, 1993­1999) . . . . . . . . . .120 Table 7.1: PHARE-Funded Programs for Roma in Central and Eastern Europe, 1993­2001 . . .127 LIST OF FIGURES Figure 1: Poverty Rates, 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 Figure 1.1: Estimated Roma Population in Central and Eastern Europe . . . . . . . . . . . . . . . . . . . .12 Figure 1.2: Poverty Rates in Selected Transition Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Figure 2.1: Household Size in Bulgaria, Hungary, and Romania, 2000 . . . . . . . . . . . . . . . . . . . . . .34 Figure 2.2: Duration of Unemployment, 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36 Figure 2.3: Enrollments in Education, 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38 Figure 2.4: Age Structure of Roma and the Total Population in the Czech Republic, 1991 . . . . .42 Figure 3.1: Registered Unemployment by Education and Ethnicity, 1999 . . . . . . . . . . . . . . . . . . .58 Figure 3.2: Unemployment by Duration, 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .58 LIST OF BOXES Box 1: In Their Own Words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Box 2.1: Who Are Roma? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29 Box 2.2: Spatial Segregation within Roma Settlements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Box 2.3: Measuring Unemployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36 Box 2.4: School Drop-Outs in Bulgaria: The Case of the Missing Children . . . . . . . . . . . . . . . . . .39 Box 2.5: Entrance to Remedial Special Schools in the Czech Republic . . . . . . . . . . . . . . . . . . . . . .40 Box 2.6: Heroin Addiction in Varna, Bulgaria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 Box 3.1: The Study of Roma Settlements in Slovakia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48 Box 3.2: Typology of Perceptions of Social Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .52 Box 3.3: Housing Conditions in a Village in Stará ŞubovŔa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56 Box 4.1: The Qualitative Study of Roma Communities in Romania . . . . . . . . . . . . . . . . . . . . . . . .70 Box 4.2: Government Education Initiatives Targeting Roma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78 Box 4.3: The Partnership Fund for Roma: Pilot Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89 Box 5.1: Public Foundations in Hungary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .100 Box 6.1: The History of Roma in Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113 Box 6.2: The Andalusian Plan for the Roma Community . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115 Box 6.3: The Acceder Program: Training and Employment Services . . . . . . . . . . . . . . . . . . . . . . .117 Box 6.4: The School Monitoring Program, Municipality of Madrid . . . . . . . . . . . . . . . . . . . . . . . .119 Box 6.5: Roma Housing Program in Madrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .121 v Roma in an Expanding Europe: Breaking the Poverty Cycle Box 7.1: Monitoring and Evaluating School Success for Roma Children . . . . . . . . . . . . . . . . . . .131 Box 7.2: Promoting Roma Employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .133 Box 7.3: Alternative Secondary Schools in Hungary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .135 Box 7.4: Desegregation of Roma Schools in Bulgaria: The Vidin Model . . . . . . . . . . . . . . . . . . . .136 Box 7.5: Lessons from US Welfare Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .137 vi FOREWORD R oma have suffered from severe poverty and There is reason for optimism. The process of exclusion throughout European history. For EU accession has focused attention on the need to many Roma in Central and Eastern Europe, address Roma exclusion at the national level and the period of transition from communism has has highlighted common European challenges. been especially dire. Low education and skill lev- Most importantly, a small but growing core of els, compounded by discrimination, have led to experienced and dedicated young Roma leaders widespread long-term unemployment and deteri- now can work both within their communities and orating living conditions. Even in countries on the with governments to advocate change. brink of accession to the European Union, Roma This volume was prepared for the conference are likely to live in poverty and lack access to edu- "Roma in an Expanding Europe: Challenges for cation, health care, housing, and other services. the Future" in Budapest, Hungary, June 30­-July Their plight has not gone unnoticed. Over the 1, 2003. I hope that this conference will catalyze past decade, governments, civil society, and the an ongoing dialogue between the new Roma international community have actively supported leadership and the wider policy community that initiatives to keep Roma children in school, will improve the living conditions and future expand access to jobs, and overcome discrimina- opportunities of Roma over the long term. tion. Many of these interventions have helped, and the time is right to scale up. Lessons from these projects can make policies more inclusive and can expand their reach. This volume calls for an inclusive approach to overcoming Roma poverty, based on increased involvement and James D. Wolfensohn participation of Roma in society, and respect for President their diversity. The World Bank vii ACKNOWLEDGMENTS T his volume brings together analysis done SPACE Foundation along with Michal Vaeăka of by teams of researchers in different coun- the Institute of for Public Affairs (IVO), and tries. The authors benefited from the guid- Michal ·ebesta of the Department of Political Sci- ance and inputs of peer reviewers Christine Jones ence, Faculty of Arts, Comenius University. Helen and Ana Revenga. Useful comments and contri- Shahriari and Dena Ringold led the work at the butions at various stages were received from Ziad World Bank and wrote the final report. Imrich Alahdad, Asad Alam, Warren Bass, Tünde Buzet- Vaeăka was a consultant to the team, and Marián zky, Mukesh Chawla, Armin Fidler, Richard Flo- Babitz of the SPACE Foundation was the Project rescu, Clare Gillsater, Boryana Gotcheva, Daniela Assistant. The listing of the field research team Gressani, Richard Hirschler, Franz Kaps, Alexan- can be found in the full report: "Poverty and Wel- dre Marc, Veronica Nyhan, Alexey Proskuryakov, fare of Roma in the Slovak Republic," available Michal Rutkowski, Ana Maria Sandi, Merrell at: www.worldbank.org/eca/roma. Tuck-Primdahl, and Julius Varallyay. Maureen Chapter Four: The qualitative study of Roma Lewis and Annette Dixon provided feedback and communities in Romania was conducted by the overall guidance throughout. Ian Conachy pre- Romanian researchers Cosima Rughinis and Mar- pared the document for publication. ian Preda. Comments were provided by Liliana The main data sources and contributors are as Proteasa of the Ministry of Education and follows: Research. Chapter Two: The second chapter draws on an Chapter Five: The project inventory and case analysis by Ana Revenga, Dena Ringold, and studies were carried out by a team of Hungarian William Martin Tracy of a three-country house- researchers led by János Zolnay and included: hold survey of poverty and ethnicity. The dataset Gábor Bernáth, Angéla Kóczé, József Kolompár, was made available by Iván Szelényi and his Katalin Kovács, and Zsolt Zádori. Richard team at the Center for Comparative Research in Hirschler edited the project case studies. the Sociology Department of Yale University. Chapter Six: The case study of Spain draws Chapter Three: The qualitative study of Roma upon information compiled by Francisco Alvira settlements in Slovakia was undertaken by Slo- Martin in Madrid, and Bronwyn Alsop in vak researchers led by Iveta Radicova of the Washington. viii OVERVIEW I ncreasingly severe poverty among Roma in Roma has grown to between 6 and 9 percent of the Central and Eastern Europe has been one of the population in Bulgaria, FYR Macedonia, Roma- most striking developments in the region since nia, and the Slovak Republic. These shares are the transition from socialism began in 1989. While likely to increase in the near future because of Roma have historically been among the poorest high population growth among Roma and people in Europe, the extent of the collapse of their decreasing fertility among the majority popula- living conditions in the former socialist countries is tions. Romania has the highest absolute number unprecedented. While most Roma had jobs during of Roma in Europe--estimated at between 1 and 2 the socialist era, formal unemployment and pover- million. Large populations of between 400,000 ty among Roma communities is now widespread. and 1 million also live in Hungary, Bulgaria, the The problem is a critical one. Because of higher Slovak Republic, Turkey, and Serbia and Mon- birth rates, the relative size of the Roma popula- tenegro. Western Europe's largest Roma popula- tion is increasing across the region. A minister of tions are found in Spain (estimated at 630,000), education in a leading European Union (EU) France (310,000), Germany (70,000), and Italy accession country recently noted that every third (130,000). In total, about 7 to 9 million Roma live child entering school in his country is Roma. Poli- in Europe--a population equal to that of Sweden cies to address Roma poverty therefore need to be or Austria. an integral component of countries' economic and Why has attention to Roma issues increased so social development strategies. sharply over the past decade? Political liberaliza- tion following the collapse of the iron curtain in Who Are the Roma? 1989 allowed for increased international and Roma, or "gypsies," are a unique minority in domestic awareness of the situation of Roma, Europe. Unlike other groups, they have no histor- including emerging human rights violations and ical homeland and live in nearly all countries in humanitarian concerns related to deteriorating Europe and Central Asia. The origins of Roma in socioeconomic conditions. National governments Europe are widely debated. Historical records have a large stake in the welfare of Roma, for indicate that they migrated in waves from north- human rights and social justice concerns, but also ern India into Europe between the ninth and four- for reasons of growth and competitiveness. In teenth centuries. Roma are extremely diverse, countries where Roma constitute a large and with multiple subgroups based on language, his- growing share of the working-age population, tory, religion, and occupations. While Roma in increasing marginalization of Roma in poverty some countries are nomadic, most in Central and and long-term unemployment threatens econom- Eastern Europe have settled over time--some ic stability and social cohesion. Understanding the under Ottoman rule and others more recently nature and determinants of Roma poverty, and under socialism. taking policy action are thus important priorities. Estimates of the size of the Roma population differ widely. Census data are intensely disputed, Roma Poverty as many Roma do not identify themselves on such Roma are the most prominent poverty risk questionnaires. By most estimates the share of group in many of the countries of Central and 1 Roma in an Expanding Europe: Breaking the Poverty Cycle Eastern Europe. They are poorer than other ble that without policy interventions, the next groups, more likely to fall into poverty, and more generation of Roma will remain in poverty. More- likely to remain poor. In some cases poverty rates over, very few Roma are active in local or nation- for Roma are more than 10 times that of non- al politics, which mutes their political voice. Roma. A recent survey found that nearly 80 per- Access to social services in Eastern Europe's cent of Roma in Romania and Bulgaria were liv- transition period has been threatened by growing ing on less than $4.30 per day (Figure 1). Even in needs and tight fiscal constraints. These condi- Hungary, one of the most prosperous accession tions have brought formal and informal charges countries, 40 percent of Roma live below the for previously free services and eroding service poverty line. quality. Roma are particularly hurt by increasing barriers to access because they are at a higher risk Why Are Roma Poor? of poverty and are often geographically isolated. For several interwoven reasons Roma pover- Similarly, because Roma frequently live in set- ty is rooted in their unfavorable starting point at tlements where property ownership is unclear, or the outset of the transition from planned to mar- in remote areas, they may lack the documentation ket economies. Low education levels and over- necessary for enrolling in school and claiming representation among low-skilled jobs led to dis- social assistance or health benefits. The high advantages on the labor market, which are prevalence of Roma in informal sector employ- compounded by discrimination and low expecta- ment--such as petty trade and construction-- tions of employers. Roma have thus had more also limits their access to benefits based on social difficulty re-entering the job market than other insurance contributions, including health care groups, and have become caught in a vicious cir- and unemployment benefits. cle of impoverishment. Additional barriers Social and cultural factors also affect access include a lack of access to credit and clear prop- and interactions with service providers. Because erty ownership. These factors, combined with an of language barriers, Roma may have difficulty overdependence on welfare, create a poverty trap communicating with teachers, understanding that precludes many Roma from improving their doctors, and maneuvering through local welfare living conditions or starting their own businesses. offices. Poor communication and stubborn stereo- Persistent disadvantages in education, including types of both Roma and non-Roma breed mis- low school attendance and overrepresentation in trust and reinforce preconceptions on both sides. "special schools" intended for physically and Moreover, the overall absence of Roma personnel mentally disabled children, make it highly proba- involved in policy design and delivery of public services means that few individuals can bridge between cultures. Figure 1: Poverty Rates, 2000 (% of population below $4.30 per day)* Regional Context 100 Roma issues have gained increasing interna- 80 tional attention over the past decade because of 60 emerging evidence of human rights violations 40 and seriously deteriorating socioeconomic condi- 20 tions within many Roma communities. These 0 developments have caught the attention of inter- Roma Non- Roma Non- Roma Non- national organizations such as the UNDP, the Roma Roma Roma Council of Europe, and the OSCE, as well as Bulgaria Hungary Romania NGOs including the Open Society Institute, Save * Adjusted to purchasing power parity (PPP) to allow for the Children, and UNICEF. Perhaps most signifi- price comparisons across countries. cantly, Roma issues are now an integral part of Sources: Yale dataset; Revenga et al. 2002. the European Union accession process; in 1993 2 Overview attention to Roma issues was adopted as part of vakia and drawing on examples from other coun- the Copenhagen criteria for accession. At the tries. The first chapter provides background on international level, such Roma NGOs as the Inter- Roma, their characteristics, and origins. It also national Romani Union and the Roma National discusses contrasting policy approaches that have Congress have become increasingly active. shaped the position of Roma in Europe over time. Chapter Two looks at the nature and charac- CONTEXT AND CONTENTS teristics of Roma poverty using quantitative data from household surveys--including a new cross- The Role of the World Bank country household data set on Roma. It examines In 2000 the World Bank published the first the correlates of Roma poverty including poor cross-country report on the poverty and human housing conditions, education, and health. development challenges facing Roma in Central Chapters Three and Four report the results of and Eastern Europe (Ringold 2000). Unlike prior detailed field studies by Central and East Euro- analyses that had largely focused on questions of pean sociologists on diverse Roma communities human rights, the Bank report addressed Roma in Slovakia and Romania. These studies draw issues from the perspective of economic and directly from interviews with Roma and non- social development. This volume updates and Roma to provide a better understanding of the expands that work, incorporating the findings of interrelated challenges Roma face in accessing new surveys and publishing, for the first time, markets and services. One of the strongest find- some of the background studies which were ings was that lack of access to public services and included in the 2000 report. Policy makers, the labor markets is compounded by the geographic Roma and NGO community, and a wider audi- isolation of some Roma settlements. Often, these ence interested in Roma issues to the 2000 study isolated settlements originated from exclusionary showed a strong interest in more detailed infor- policies of the past. Today the geographic isola- mation on the conditions in Roma communities tion of Roma settlements limits access to educa- and policy responses. tion, health care, and waste collection, and thus This volume responds to that demand, but increases poverty over the long run. does not quench it. Surveys and case studies pre- Other causes of Roma poverty are interrelat- sented here are still incomplete. Further work is ed as well. For instance, Roma parents' choice to needed to examine the particular circumstances enroll their children in "special schools" intend- of Roma living in the countries of the former ed for the mentally and physically disabled is Yugoslavia, Albania, Moldova, Russia, and sometimes driven by discrimination experienced Ukraine, among others. Issues related to health, by Roma in regular schools. Roma parents some- housing, and the situation of Roma women also times feel they are protecting their children by need further attention. These gaps stem from the sending them to special needs schools with other lack of information and measurement challenges. Roma children, but the education they receive Despite the severity of Roma poverty, informa- there ill prepares them for life, again exacerbat- tion on their living conditions and challenges is ing the risks of poverty and exclusion over the scarce, often unreliable, and frequently anecdotal. long term. The analyses presented here are intended to fill Finally, Chapters Five and Six look at the these gaps--and to stimulate further action. experience of projects in Hungary to compare the Central and East European experience with that Contents of Spain, a West European country with a large The chapters draw on both quantitative analy- Roma population. Chapter Five reports the ses of household surveys and qualitative, socio- results of a survey of Roma projects in Hungary logical case studies that document the experiences and shows that, despite the proliferation of such of Roma communities in different countries, projects after 1989, it remains difficult to evaluate focusing on Romania, Bulgaria, Hungary, and Slo- their impact. Case studies of several projects 3 Roma in an Expanding Europe: Breaking the Poverty Cycle show several factors to be important, including tive analysis provides a complementary set of the quality of project leadership, local economic perspectives and a better starting point for analy- conditions, and monitoring and evaluation. sis and policymaking. Methods and Approach THE ENVIRONMENT FOR POLICY This volume draws from both quantitative DEVELOPMENT and qualitative methods in order to paint a fuller Policies to ease Roma poverty need to be picture of the living conditions of Roma. Both designed with three key factors in mind: (i) the approaches have distinctive benefits and draw- multidimensional nature of Roma poverty and its backs. Quantitative methods are useful in illus- interconnected roots; (ii) the diversity of Roma in trating where Roma stand relative to non-Roma Central and Eastern Europe; and (iii) the process populations in individual countries and Roma of European integration. populations elsewhere. On the other hand, data on Roma are notoriously unreliable and difficult Multidimensional and Interconnected Roots to attain. Even basic population figures are sub- of Roma Poverty ject to dispute. Since Roma often do not identify The central insight of this study is that Roma themselves as such, survey-based research has poverty has multiple and interrelated causes. serious limitations. Still, quantitative data offers These tend to reinforce one another in a vicious useful comparisons of welfare measures that can cycle of poverty and exclusion, and require a improve policy analysis and responses. multifaceted approach. Roma often have poor While quantitative research shows that Roma access to labor markets because of low education poverty is distinctive, it does not provide an ade- levels, geographic isolation, and discrimination. quate basis for understanding the particular Low education levels result from constraints on dynamics that underlie Roma poverty. Here, both the supply and the demand side. Roma qualitative research has the greatest impact. often face discrimination at school and feel that Qualitative research can identify social processes, schools ignore Roma culture and language. In mechanisms, and relations between variables that addition, Roma sometimes lack sufficient food or are difficult to discern by looking at numbers clothing to support school attendance. Thus, atti- alone. For example, the empirical analysis pre- tudes, experiences, and social conditions conspire sented in Chapter Two shows that much of the to reduce Roma education levels and labor mar- gap between Roma and non-Roma welfare is like- ket performance. Because of these interconnected ly due to factors such as discrimination and roots, one cannot adequately address Roma exclusion, which cannot be assessed empirically. poverty by focusing on a single aspect. Rather, a Therefore qualitative research provides a sharper comprehensive approach is needed. picture of Roma living conditions in different An example from the case study material is communities, to emphasize the diversity of Roma instructive here. In several countries, researchers populations and better understand interconnec- found that Roma poverty was caused in part by tions between causes of poverty. Carefully con- poor housing conditions. In many cases, this is structed qualitative surveys conducted by Cen- because Roma were left out of the property and tral and Eastern European researchers formed the land privatization processes that took place during primary source for this analysis. This qualitative the early 1990s. Information was scarce about how research lets us hear how Roma perceive their to navigate the bureaucratic procedures for prop- poverty situation in their own words. erty ownership, and Roma were less likely than Qualitative research has drawbacks as well. It others to do it successfully. Hence, Roma today tends to provide a snapshot of a single area, disproportionately live in unregistered dwellings, emphasizing certain factors over others and with contributing to poverty in complex ways. Accord- biases that may reflect observer concerns. How- ing to one man interviewed in Kyjov, a segregated ever, a combination of quantitative and qualita- Roma settlement in the town of Stará ŞubovŔa, 4 Overview Slovakia, "We built our house with a building per- by relatively well-off Caldarari Roma, who speak mit, but there are still problems with the site, the traditional Roma language, and work prima- although it was officially given to us during social- rily in trade, after being laid off from a large state- ism. But today the land is not ours, therefore we owned enterprise. Such diversity complicates any cannot install any water, gas, or sewage pipes." approach to addressing Roma poverty, since the This example shows that Roma poverty is rooted root causes may also differ dramatically. in incompatibilities between Roma social practices, dominant state behaviors and norms, limited The European Dimension political representation, and geographic exclusion. Policies for addressing Roma poverty also It also shows how Roma may have missed out on must be framed in the context of the Central and the benefits of economic reforms. Eastern European countries' drive for member- ship in the European Union. The timing of the Diversity publication of this volume and other reports on While demonstrating the distinctive nature of Roma are not coincidental. The accelerating Roma poverty, this volume also emphasizes the process of European integration has focused diversity of Roma populations in Central and attention on the Roma issue through the adoption Eastern Europe--ethnic, occupational, religious, and monitoring of the Copenhagen criteria for EU and economic. The proportion speaking dialects accession. Based on these criteria, the countries of of the Roma language differs greatly from coun- Central and Eastern Europe have built institutions try to country, as does the proportion living in and legislative mechanisms to address Roma cities, integrated neighborhoods, or segregated issues. However, this is only the beginning. Even rural settlements. These differences have a major once accession happens--as early as 2004 for impact on welfare status. Efforts to create, define, some countries--addressing Roma poverty will or represent a single Roma community will simi- require a long-term approach, as part of each larly founder on the rocks of internal cultural country's overall economic and social develop- diversity. Roma tend to have distinctive problems ment program. of integration and access, but the situation of dif- Interactions between Roma policy and the ferent communities and individuals varies European accession process can be seen most immensely and cannot be reduced to a single, vividly in Hungary--the first Central and East simple set of answers or policy responses. European country to apply for EU membership, A study of nine Roma communities in Roma- and also the first to make a substantial policy nia, included in the main report, illustrates this effort to address Roma issues. Hungary passed a diversity. Each of the nine communities consists Minorities Act in 1993 that granted considerable of different combinations of Roma subgroups, cultural, educational, and linguistic rights to with different languages, religions, and occupa- Hungary's thirteen recognized minorities, includ- tions. The Zabrauti neighborhood of Bucharest ing Roma. This Act created a system of national contains a mosaic of Roma ethnic groups, varying and local minority self-governments that let from the quite traditional Sporitori, who speak minorities initiate social, educational, and devel- the Roma language, to more integrated Roma opment projects. Approximately half of these are who speak primarily, or only, Romanian. The Roma self-governments. urban community of Babadag has three main Hungary has also established a national Roma groups, the largest of which is Muslim. Office for National and Ethnic Minorities, an However, in the rural community of Iana, most independent Minorities Ombudsman to oversee Roma are active Orthodox Christians. Other com- minority rights and protections, and a Roma munities are relatively homogenous. One urban Office under the Office of the Prime Minister to and one rural community studied in Romania coordinate Roma policy across the government. consisted primarily of Hungarian-speaking Together, these offices enable Hungary to comply Roma. Another rural community was populated with EU norms, in part through the implementa- 5 Roma in an Expanding Europe: Breaking the Poverty Cycle tion of a "medium-term package" of measures and sustain growth, while improving social wel- aimed toward social inclusion of Roma. Hun- fare outcomes and the inclusiveness of policies for gary's extensive experience with Roma institu- all populations. However, macro-level policies will tions and projects provides an important example not be sufficient to reach all Roma, so targeted for other EU aspirants. interventions are needed to tackle unique prob- lems of exclusion and ensure that Roma are able to POLICY IMPLICATIONS AND APPROACHES work and participate fully in public services. While the plight of Roma in Central and East- Related to this, better access to quality social ern Europe has not gone unnoticed, many lessons services for Roma is linked to the overall effec- still need to be drawn and new policy approach- tiveness of the education, health, and social pro- es pursued. During the past decade, numerous tection systems in each country. In many ways, initiatives by governments, NGOs, and interna- the inherited systems were ill-suited to the reality tional organizations have been launched to of a market economy, and one way in which they address various aspects of the Roma issue, from have proven ineffective is their inability to reach combating human rights violations, to addressing all vulnerable groups, including Roma. Through- racial stereotyping in the media, and promoting out the region, countries have embarked upon education and employment. The level of activity systemic reforms to improve the efficiency, equi- varies significantly across countries. As many of ty, and relevance of public services. These meas- the countries of Central and Eastern Europe ures are making a difference. Addressing sys- move toward becoming EU members, a more sys- temic issues and improving access and quality of temic policy-oriented approach is needed to social services will improve conditions for the address gaps in Roma economic and social devel- entire population. Again, these systemwide opment. Project lessons from the 1990s can be measures need to be accompanied by interven- used to inform policy interventions in key areas tions designed to reach Roma. such as education, health, social assistance, and the labor market. Toward an Inclusive Approach Together, the multidimensional and inter- As Roma poverty is rooted in broad-based related roots of Roma poverty, the diversity of social exclusion--economic, social, and geo- Roma communities, and the European back- graphic--addressing it calls for an inclusive ground constitute a unique context for policy. approach which would aim to expand and pro- This report makes several recommendations. mote Roma involvement and participation in First, a comprehensive policy approach is mainstream society, while maintaining cultural required to address multiple and interrelated and social autonomy. Only policies that allow causes of Roma poverty simultaneously. Second, Roma to take advantage of opportunities in primary emphasis needs to be placed on further- national and European labor and housing mar- ing the social inclusion of Roma in European soci- kets, education and health systems, and social eties. In identifying policy approaches, useful les- and political networks have a chance of reducing sons can be drawn from other countries with poverty over the long term. Policy mechanisms similar experience. And finally, greater attention would include those which make existing policies needs to be paid to policy implementation and more accessible to Roma, and identifying areas evaluation and the central role of Roma them- where targeted initiatives which specifically selves in these processes. reach Roma are needed. An emphasis on policies of inclusion would complement rights-based Links with Systemic Reform approaches by tackling the economic and social Improving conditions for Roma are inherently barriers which Roma face. linked to the overall success of each country's eco- A central policy goal should be the multifac- nomic and social development strategies. So each eted inclusion of Roma into institutions and country must implement policies that promote mechanisms that create economic and social 6 Overview opportunities. Emphasis should be placed on school programs and provision of food, providing incentives, rather than forcing compli- transportation, and clothing to enable ance. Interventions which reduce the isolation attendance; and exclusion of Roma can help improve living · Increasing outreach to Roma communities conditions over the longer term. An inclusive by social service providers, including approach also needs to rely on greater participa- health and social workers; tion of Roma in the projects and programs which · Involving Roma as liaisons between com- affect them. A number of successful projects use munities and public services; Roma mentors as liaisons between Roma and · Providing relevant job training and pro- non-Roma communities. For example, Roma grams that increase Roma participation in teachers' assistants who work with parents, or formal labor markets. peer advisors who assist with job placement, can facilitate integration while strengthening the Learning from Example Roma community itself. When considering future policy directions, a Addressing exclusion and the wounds of seg- key source of ideas and experiences may be regation also involves overcoming divisions found in the policy experiences of other countries between Roma and non-Roma communities. This and regions in minority policies, particularly in would build trust and help develop social capital the West. North and South America provide within communities. Such measures need to interesting counterpoints to Europe's experience, involve both Roma and their non-Roma neigh- in part because the histories of African and bors. In most cases, policies should target com- indigenous peoples in the Americas offer more munities at large, rather than Roma in particu- parallels to that of Roma than other national lar--although there may be exceptions where minorities in Europe. While all ethnic groups explicit attention to ethnicity would be appropri- have distinct features, minority­majority rela- ate, such as overcoming language barriers. Multi- tions share important similarities everywhere, cultural education and a curriculum which and much can be learned from the policy experi- includes the history and culture of Roma and ence of other countries which have confronted other minorities are critical vehicles for overcom- these issues over centuries. ing cultural barriers. Training teachers, local gov- What is distinctive about Roma in Europe is ernment officials, and other personnel working in that they have endured centuries of exclusionary social services can address discrimination within and assimilationist policies without being public services. Finally, public information cam- absorbed into majority societies. They remain paigns can promote multiculturalism and raise stateless and have founded no movement for awareness about discrimination. statehood because they lack a historic homeland. Policies need to balance three related sets of These general characteristics underline the chal- objectives: first, increasing economic opportuni- lenges facing an integration-oriented approach to ties by expanding employment participation; sec- Roma poverty. However, they also focus attention ond, building human capital through better edu- on the stakes involved in getting policy right. Pol- cation and health; and third, strengthening social icymakers need to approach issues of Roma capital and community development through poverty from a long-term perspective, with a increased empowerment and participation of clear idea of objectives and tradeoffs. Roma. In this vein, options include: · Reducing segregation in housing, particu- Learning from Evaluation and larly by alleviating the problems associat- Implementation ed with, or providing alternatives to, iso- Development of a comprehensive national lated rural settlements; policy response to Roma poverty entails attention · Integrating Roma students into main- to monitoring and evaluation. The wealth of stream educational systems through pre- Roma projects in Central and Eastern Europe has 7 Roma in an Expanding Europe: Breaking the Poverty Cycle provided a great deal of experience in implemen- to assess project impacts and outcomes. These tation. But very few initiatives have been evaluat- require collection of baseline data at the outset of ed or monitored, making it extremely difficult to projects for comparison once they have been com- identify lessons for future interventions. It is pleted. For example, an intervention designed to important to examine this body of experience to improve school enrollments should measure distill lessons for future work. Mechanisms for enrollments prior to the project and assess whether monitoring and evaluation should be built into participants stay in school during the project, as new and ongoing initiatives. So should opportu- well as afterwards. The time horizon for outcome nities for exchanging information within and evaluation should also be enough to assess longer- across countries. term impacts. Again, in the case of education, the A first step is increasing the availability and evaluation should assess not just whether children quality of information on Roma. To do this, coun- are in school at the end of the project, but what tries need to examine their statistical instru- they have learned, whether they graduate and ments--for example, censuses and household continue their education, and how the project surveys--and administrative data, to assess how affects their chances in higher education and the they can better capture policy-relevant informa- labor market. tion on Roma and other minorities. Multilateral coordination, advice, and guidance can be impor- Ensuring Participation tant for ensuring comparability of data. More Regardless of whether programs and policies information on international practices, particular- are explicitly designated for Roma, Roma partici- ly in handling the privacy issue on ethnic identi- pation is essential. The success of the inclusive fication, is needed. The outcomes of targeted pub- approach outlined earlier rests on the ability of lic policies and NGO initiatives also require close Roma to contribute to the development processes monitoring, and program evaluations should be which affect them. The experience of policies and used for ongoing policy development. Mecha- programs directed at Roma during both the nisms should be in place for disseminating les- socialist and transition periods showed that sons across regions and countries. involvement of Roma in the design, implementa- Privacy concerns about data collection must tion, and evaluation of programs is essential. The be respected. But up-to-date information is criti- recent past is littered with projects and programs cal for policy makers to make decisions about that, however well-intentioned, failed because program design and to monitor outcomes. Such they were designed and implemented without data collection should benefit Roma in the long the involvement of the future beneficiaries. run through better designed and targeted inter- Ensuring Roma involvement in policy and ventions. Privacy concerns can be respected by project development rests on the existence of making declarations of ethnicity voluntary and effective mechanisms for participation. While using periodic sample surveys, rather than Roma have been increasingly involved in civil national administrative data, to collect informa- society and various aspects of policymaking, sig- tion on specific topics. Involvement of Roma nificant challenges to ensuring effective commu- groups in the development and implementation nication and involvement remain. Some of these of surveys, as well as the analysis, is also extreme- have been discussed in this volume, including ly important. This was an emphasis of recent cen- low education levels and illiteracy which dimin- suses in Slovakia and Bulgaria. Qualitative ish the potential pool of Roma leaders and vot- assessments can also provide valuable informa- ers, and mistrust and prejudices between Roma tion for project design. and non-Roma. Continued expansion of oppor- Building monitoring and evaluation mecha- tunities for Roma to participate in civil society at nisms into projects and policies is vital. Monitoring the local and national levels is essential. So is should be an integral part of all projects to ensure non-Roma involvement. The example of Slova- accountability. Equally important are evaluations kia presented in Chapter Three, in particular, 8 Overview Box 1: In Their Own Words Interviews with Roma throughout the region highlight the range of experiences and living conditions, across and within countries. These snapshots illustrate this diversity. The challenges they face are explored further throughout this volume. Education: Many Roma children do not attend school. Some parents are unable to send their children to school because they lack basic supplies, or even clothes. Other children are excluded because of social and cultural factors, such as language. We can't afford to send them to school in the winter. We have no sneakers, no proper shoes for snow. They can't go to school in slippers. They don't have jackets or warm clothes either. We can't afford anything--copybooks, pens... Chil- dren have no money for meals. That's why they don't go to school. --Parent, Bulgaria Children from segregated Roma settlements do not master the Slovak language and do not understand their teachers. The teachers do not speak the Roma language, so they communicate by using gestures. --School Director, Slovakia. While demand for education is low in some Roma communities, other parents expressed a strong inter- est in their children's education and recognized its importance for their future success. My grandson is a first grade student. We sent him to kindergarten and hope in the future that he will put more impor- tance on education than we did. --Grandparent, Slovakia. I waited for my daughter to return from school every day and asked her what happened at school. I sat beside her when she was writing up her homework. I would not let her go out until I saw that she had finished. I would not allow any- one at home to touch her and make her do some other housework...I do not know what will happen to her after she com- pletes her education, but whatever that is, it will be better. She can become a doctor, a teacher. She will go higher than us. --Parent, Bulgaria. Employment: Formal unemployment in some Roma settlements can reach 100 percent. Many Roma face severe obsta- cles in finding a job because of their low education and skill levels, as well as discrimination on the labor market. Who is going to give me a job? I have no education, no skills and am Roma. Even in my neighboring village nobody wants to give us any work. --35-year-old father of five, Slovakia. If his Bulgarian name is Angel or Ivan or Stoyan or Dragan, he'll get all the application forms and be asked to come in. As soon as they realize he's Gypsy, Roma, he's turned down, they lower their voices and tell him to come some other time. When your name is Bulgarian and they see you are a Gypsy, they throw you out! --Roma, Bulgaria. Roma Identity: In Hungary, experiments with alternative education for Roma high school students, which include Roma language, culture and history in the curriculum, have sparked interest in Roma identity among young people. My grandmothers spoke the Roma language, and my parents can understand it. I do not speak the language, but I would very much like to learn it. --Student, Hungary I would like to know more about the origin of my people and our values. --Student, Hungary highlights the perils of separation and segrega- society. Increasing contacts and partnerships tion. Roma who lack opportunities for interac- between non-Roma and Roma will facilitate tion with wider society, including other Roma inclusion and address the mistrust and miscom- communities and non-Roma, are cut off from 9 Roma in an Expanding Europe: Breaking the Poverty Cycle munication that limit the progress of local and ence through a comprehensive change of direc- community development. tion is significant and bright. The current level of activity and interest in CONCLUSIONS Roma issues in Central and Eastern Europe pro- Poverty among Roma remains one of the vides a promising start. The next step is to inte- most pressing issues for Central and Eastern grate the lessons of this experience into policy. European states as they move toward EU integra- The mechanisms to facilitate this have been put in tion and sustained economic development. Using place. Most Central and East European countries a variety of sources and approaches, this report have formulated strategies for improving the examines the nature of Roma poverty--a multi- conditions of Roma and established institutions faceted challenge that can only be addressed by a to develop, coordinate, and administer policies policy approach that attends to all dimensions of and projects. However, the agenda is complex Roma social exclusion and focuses on the poten- and improvements will not come overnight. tial contributions Roma can make to social and Indeed, poverty among Roma communities in economic development. Since the dominant poli- some West European countries highlights the cy approach in the years after socialism has tend- scope of the challenge. Effective policy responses ed to rely on a fragmented set of projects, often will require a multilayered approach involving delivered by local NGOs with limited assistance cross-country partnerships among Roma and from the state, the opportunity to make a differ- international organizations, national and local governments, NGOs and communities. 10 Chapter One: INTRODUCTION I ncreasingly severe poverty and exclusion of address Roma poverty during the transition? Roma in Central and Eastern Europe has been What lessons have been learned, and how can one of the most striking developments in the these be applied in the future? region since the transition from socialism began In answering these questions, this volume in 1989. While Roma have historically been draws on quantitative analyses of household sur- among the poorest people in Europe, the extent veys and qualitative, sociological case studies of the collapse of their living conditions in the which document the experiences of Roma com- former socialist countries is unprecedented. munities in Romania, Bulgaria, Hungary, and Slo- While most Roma had jobs during the socialist vakia, and drawing examples from other coun- era, formal unemployment is now widespread. tries. This approach is intended to provide a more Even in countries which are set to join the Euro- nuanced picture of Roma poverty and its deter- pean Union (EU) poverty is striking. Poverty minants, as well as of policy experience. Identify- rates for Roma range between four and ten times ing the unique factors underlying Roma poverty that of non-Roma in Bulgaria, Hungary and helps to explain why the transition has been Romania. Because of higher birth rates, the rela- harder on Roma than others, and what interven- tive size of the Roma population is increasing tions are needed to expand opportunities for across the region. Roma, within the context of economic and social As a result, addressing poverty among Roma development for the population as a whole. is one of the most pressing development issues in This first chapter provides background on Central and Eastern Europe, particularly as the Roma in Central and Eastern Europe, their char- countries move toward membership in the EU. acteristics and origins. It also discusses contrast- While living standards have declined for many ing policy approaches that have shaped the posi- during the transition from socialism to market tion of Roma in Europe over time. Chapter Two economies, conditions for Roma have deteriorat- looks at the nature and characteristics of Roma ed more severely than for others, and Roma have poverty using quantitative data from household been poorly positioned to take advantage of surveys--including a new cross-country dataset. emerging opportunities in the economy and soci- It examines the correlates of Roma poverty, ety. Poverty among Roma is a complex and mul- including housing conditions, educational, and tidimensional phenomenon related to poor health health status. and education status, limited chances in the labor Later country chapters on Slovakia, Romania, market, discrimination, and unique aspects of Hungary, and Spain explore aspects of Roma Roma social organization, which together con- poverty through qualitative methods. Sociologi- tribute to their social exclusion. cal field studies enrich the picture of living condi- The multidimensional nature of Roma pover- tions in Roma communities. Case studies of ty and social exclusion raises three interrelated Roma settlements in Slovakia highlight the rela- questions: What distinguishes Roma poverty tionship of Roma poverty to social exclusion. from poverty among other groups in the transi- Chapter Four, on Romania, examines conditions tion countries of Central and Eastern Europe? in nine Roma communities and reveals substan- How have countries in the region attempted to tial diversity in access to social services. Chapter 11 Roma in an Expanding Europe: Breaking the Poverty Cycle Five focuses on policy lessons, drawing from the the ninth and fourteenth centuries. Roma consti- case of Hungary, where more projects to address tute an extremely diverse minority, with multiple Roma issues have been undertaken than in any subgroups based on linguistic, historical, and other country in Central and Eastern Europe. occupational distinctions. While some Roma Chapter Six, on Spain, provides a counter- groups are nomadic, the vast majority of Roma in point to the case studies from Central and Eastern Central and Eastern Europe have settled--some Europe, illustrating the commonalities and differ- during the Austro-Hungarian and Ottoman ences between Roma in East and West, while empires, others under socialism. drawing lessons from policy. These lessons form Estimates of the size of the Roma population the basis of the discussion of policy recommenda- in Europe range from 7 to 9 million, similar to the tions in the final chapter. Examples of programs total population of many smaller European states. and policies from other countries are included Approximately 70 percent of Roma in Europe live where possible. Together, these multiple in the countries of Central and Eastern Europe approaches provide a striking picture of Roma and the former Soviet Union, and nearly 80 per- poverty with policy implications for the future. cent of this population live in countries that have been invited to join the European Union in 2004 or Who Are the Roma? are in accession negotiations.1 Roma are estimated Roma, or "gypsies," are a unique minority in to make up between 6 and 11 percent of the popu- Europe. They have no historical homeland but lations of Bulgaria, FYR Macedonia, Romania, live in nearly all countries of Europe and Central and the Slovak Republic (Figure 1.1). Asia. The roots of the Roma are widely debated. Why has attention to Roma issues increased so Historical records indicate that they migrated in sharply over the past decade? The fall of the iron waves from northern India into Europe between curtain in 1989 allowed for increased internation- Figure 1.1: Estimated Roma Populations in Central and Eastern Europe 14.0 2500 12.0 2000 10.0 % of total population 1500 Roma population population 8.0 '000s total 6.0 1000 of % 4.0 500 2.0 0.0 0 SpainFrance MacedoniaRomaniaBulgariaRepublicHungary AlbaniaRepublic Montenegro FYR Slovak and Czech Serbia Sources: Wheeler (1999); Liegeois (1994). Total population data: World Bank Atlas (1995). 12 Introduction al awareness. Subsequently, concern over human Figure 1.2: Poverty Rates in Selected rights violations and seriously deteriorating Transition Countries socioeconomic conditions for Roma gained atten- (% living under $4.30 PPP* per day) tion from international organizations and interna- tional NGOs, such as the Council of Europe, the 70 OSCE, various UN agencies, the Soros Founda- 60 tion, and the Western news media. Many interna- 50 tional organizations have issued major reports on 40 Roma issues in recent years, including a recent 30 Human Development Report on Roma by the 20 UNDP.2 Most significantly, attention to the rights 10 and living conditions of Roma have become part 0 of the EU accession process, as Roma issues are Albania Romania Bulgaria Slovak Estonia Hungary Czech included in the political criteria for accession. (96) (98) (95) (96) (98) (97) (96) Paying attention to Roma issues is squarely in Notes: Poverty line adjusted using purchasing power parity the interest of national governments. The severe (PPP) for comparisons across countries. Household surveys differ across countries, refer to source for details. deterioration of living standards has raised * PPP= purchasing power parity humanitarian concerns and called attention to Source: World Bank 2000b. human rights issues. Countries also cannot ignore the growth of Roma long-term unemployment and poverty, which will undermine competitive- ness over the longer term. In countries where countries. Even in the more prosperous countries, Roma constitute a large and growing share of the significant poverty persists within some seg- working-age population, their increasing margin- ments of the population. In the transition coun- alization threatens stability and social cohesion. It tries of Central and Eastern Europe, the unem- has become a priority to understand how Roma ployed, the poorly educated, those living in rural poverty differs from poverty generally in the areas, and children are more likely to be poor. In transition countries, in order to overcome it. Slovakia in 1996, while the national poverty rate was 10 percent--low by regional standards--the Poverty in Transition poverty rate for those with primary education, or Changes in the socioeconomic status of Roma lower, was 40 percent higher than the national in Central and Eastern Europe over the past average, at 14 percent (World Bank 2001b). Even decade are closely linked to the effects of eco- worse, poverty rates for households headed by an nomic transition. The shift from planned to mar- unemployed person were nearly four-and-a-half ket economies has led to an increase in poverty times the national average. Roma represent one and lower living standards across the region.3 of the main poverty groups. They are both poorer However, regional figures mask considerable than other population groups and more likely to diversity across and within countries. Poverty in fall into poverty and remain poor. Poverty there- the leading EU accession countries, including the fore has a substantial ethnic dimension.4 Czech and Slovak Republics and Hungary-- remains substantially lower than in the poorer HISTORY countries of the region--such as Romania and The roots of Roma in Europe have long been Bulgaria (Figure 1.2). This due to many factors, a subject of mystery and controversy. According including the slower pace of economic reforms in to records, Roma arrived in Europe from northern the latter two countries in the early 1990s (World India, although the reasons for their migration Bank 2000b). are unknown. Linguistic evidence and the limited Deep pockets of poverty distinguish the pro- documentation suggest that Roma came first file of poverty in many of the leading accession through Persia and the Caucasus, through the 13 Roma in an Expanding Europe: Breaking the Poverty Cycle Byzantine Empire, into southern Europe (Fraser half a million Roma from across Europe were 1995), although some Macedonian legends place executed or killed in concentration camps. The Roma in Europe at the time of Alexander the largest population losses were among Roma from Great as early as the fourth century B.C. The first Yugoslavia, Romania, Poland, the USSR, and detailed references to Roma in Central and East- Hungary (Fraser 1995; Lewy 2000). ern Europe are found in twelfth century records from the Dalmatian Coast and Hungary--now The Socialist Period the Slovak Republic (Crowe 1994). Soviet bloc policies adopted toward Roma in The subsequent history of Roma in Europe is Central and Eastern Europe left a legacy that as varied as the countries to which they migrated. affects the socioeconomic status of Roma today. However, marginalization and discrimination Although the extent varied, socialist govern- have been common (Bárány 2002). During their ments made a concerted effort to assimilate Roma first centuries in Europe, Roma were valued for and minimize ethnic differences. Communist par- their skills in metalworking, armaments and ties issued decrees and adopted policies that music. They were also subject to prejudice and aimed at socioeconomic integration by providing persecution. As early as the fifteenth century, housing and jobs for Roma. Roma were traded as slaves in the principalities These measures were frequently culturally of Moldavia and Wallachia (currently Romania). repressive, though their stringency varied. Among Draconian anti-Roma policies were adopted the most repressive campaigns were movements throughout Europe. A scholar on Roma notes that in Czechoslovakia and Bulgaria that sought to "[h]ad all the anti-Gypsy laws which sprang up erase ethnic divisions completely. In contrast, in been enforced uncompromisingly, even for a few socialist Yugoslavia, Roma were granted official months, the Gypsies would have been eradicated nationality status in 1981 (Poulton 1991). from most of Christian Europe well before the In Czechoslovakia in 1958, the government middle of the sixteenth century" (Fraser 1995). proclaimed that Roma were not a separate ethnic In Central and Eastern Europe, the policies of group and embarked upon a violent campaign the Austrian Empire, the Hungarian Kingdom, against nomadism. The regime planned a "dis- and the Ottoman Empire had a significant role in persal and transfer" scheme to resettle Roma shaping Roma communities. Under Empress from areas with large Roma communities in east- Maria Theresa, in the latter half of the eighteenth ern Slovakia to the Czech lands. But this program century, Habsburg policies aimed to eliminate the was never fully implemented, and conditions Roma's nomadic lifestyle and encourage assimi- were relaxed somewhat during the period of lation. While these restrictions were loosened "Prague Spring" reforms of 1968. During this with the end of Maria Theresa's reign, they were time, Roma language teaching was introduced in the first step toward the settling of Roma--a fea- schools. However, assimilation programs were ture that still distinguishes Roma in Central and imposed with new vigor following the Soviet Eastern Europe from those living in Western crackdown on the reformists (Fraser 1995). Europe. Policies toward Roma under the In Bulgaria, all ethnic minorities--including Ottoman Empire were on the whole more Bulgarian Turks and Roma--were targeted with relaxed, and mostly allowed for free movement "Bulgarization," as the regime attempted to sup- across borders, despite occasional attempts at press cultural identities through forced assimila- forced settlement, including an initiative against tion. Minorities were forced to change their Serbian Roma in the 1630s (Fraser 1995). names to Bulgarian names and could lose access The Nazi era marked the darkest period of to social services for not complying. In Romania, Roma history. Like Jews, Roma were targeted President Ceaucescu mounted an aggressive with discriminatory legislation and subsequently "systematization" program across the country in extermination. During the course of the "Devour- the 1980s, resettling entire villages and urban ing," as Roma call the Holocaust, approximately neighborhoods. While the campaign was not 14 Introduction explicitly targeted at Roma, both Roma and non- garian labor force found that half of Roma work- Roma settlements were destroyed (Crowe 1994). ers were unskilled, in comparison with 12 percent Assimilation efforts under socialism trans- of the Hungarian population (Crowe 1994). formed the Roma. Policies forced Roma into the As a whole, socialist policies did improve mainstream economy by providing employment, conditions for Roma by increasing access to edu- housing, and education. The impact of these cation, employment, and housing. However, efforts was mixed. In education major strides these initiatives also created new divisions were made in enrolling children in schools. In between Roma and the state. The forced and Czechoslovakia, a campaign increased kinder- often repressive assimilation campaigns foment- garten enrollment rates for Roma--from 10 per- ed mistrust and tensions between Roma and serv- cent in the early 1970s to 59 percent by 1980. At ice providers. This strain was further reinforced the same time, the share of Roma finishing com- by the absence of participatory processes, authen- pulsory education rose from 17 to 26 percent, and tic self-government, and Roma involvement in literacy rates rose to 90 percent among adults. In policy development and implementation. Pater- Poland, an education initiative enrolled 80 per- nalistic state provision of "cradle to grave" jobs, cent of Roma children in the late 1960s. Some housing, and other benefits also created a culture school-promotion initiatives, such as a Hungari- of dependency. The transition, employment loss- an effort in the late 1980s, attempted to increase es, and growing poverty have left many Roma, as Roma school attendance by experimenting with well as others, feeling abandoned and alienated. Roma-language teaching (Fraser 1995). However, these gains were tempered. In ROMA IN THE TRANSITION PERIOD many cases, socialist education policies helped The transition to democracy and market only to perpetuate earlier inequities. In the push economies has presented new challenges to Roma to increase enrollments, Roma were often chan- in Central and Eastern Europe. On the one hand, neled into segregated schools intended for chil- Roma have greater opportunities to organize dren with mental and physical disabilities. For politically and express themselves culturally; on example, the education campaign initiated in the other they have also proven more vulnerable Hungary in the 1960s focused on creating "spe- than other groups. There are four broad sets of cial classes...within the national school system reasons for this. for retarded or difficult children" (Crowe 1994). A First, as Roma generally have less education disproportionate number of Roma were enrolled and skills than others, they have had difficulty in special classes and schools. Similar practices competing for jobs in the new market economies. were followed in other countries, including Roma were often the first laid off from state- Czechoslovakia and Bulgaria. The practice of owned industrial factories, mines, and agricultur- pushing Roma into special schools has continued al cooperatives. As a result, they face significant following the transition. hurdles to labor market reentry, and have Employment programs were also a mixed depended instead on poorly funded public assis- blessing. Some attempted to formalize traditional tance, insecure jobs in the informal sector, or Roma trades. For example, the Polish govern- work abroad. ment set up cooperative workshops to support Second, the transformation exacerbated nu- traditional crafts, such as coppersmithing. How- merous social problems facing Roma, including ever, these low-paying and physically difficult low educational and health status. Third, the jobs were in less demand and did not attract transition has had a profound impact on Roma Roma workers (Fraser 1995). Because of their low housing. Roma were historically not landowners. education levels and skills, Roma were often As a result, they have generally not benefited employed instead in state-owned enterprises and from land restitution and privatization policies. on collective farms, frequently in the most oner- Fiscal constraints during transition have meant ous, unskilled positions. A 1995 study of the Hun- fewer state resources for maintaining the public 15 Roma in an Expanding Europe: Breaking the Poverty Cycle housing in which many Roma live. Finally, polit- for public expression of hatred against Roma. Anti- ical transformation has been accompanied by ris- Roma violence has been documented in the 1990s ing discrimination and violence against ethnic in all of the countries of Central and Eastern minorities, including Roma. Europe. Hence, designing and implementing pro- Addressing Roma poverty is therefore a mul- grams to address the exclusion of Roma requires tifaceted problem, related to a complex mix of attention to the unique issues of diversity, culture, historical, economic, and social factors. Although and social exclusion. other vulnerable groups in the region face simi- lar circumstances, given the lack of social capital ROMA DIVERSITY, CULTURE, AND among some Roma communities and the added SOCIAL EXCLUSION barrier of discrimination, the challenges loom large. Aspects of Roma culture and living condi- Diversity tions also reinforce stereotypes by limiting com- A defining characteristic of Roma is their munication between Roma and non-Roma and diversity. Researchers refer to a "kaleidoscope" contributing to a vicious circle of isolation and and "mosaic" of Roma groups (Liegeois 1994; marginalization. Fraser 1995), with numerous cross-cutting sub- Moreover, access to social services has been groups, including family clans and religion. Many threatened by an increasing need for services and Roma groups have little or no contact with each tight budgets. Formal and informal charges now other. Because of their varied history in Asia and accompany previously free services, as does erod- Europe, Roma also participate in many different ing quality. Roma are particularly affected by religions. There are Roma of different Christian increasing barriers to access because they are at a denominations, as well as Muslim Roma. In Bul- higher risk of poverty and face unique circum- garia, Roma have traditionally been Eastern stances that limit their access to services. Geo- Orthodox or Muslim, although in recent decades graphically isolated Roma communities may lie many have begun to attend Protestant and Pente- far from social service facilities and personnel. costal churches (Iliev 1999). There are also geo- Similarly, because Roma frequently live in remote graphic and historical groups, such as the Sloven- areas or illegal housing, they may lack the docu- sko Roma from Slovenia, and subgroups based mentation necessary for enrolling in school and upon occupational categories, including former claiming social assistance or health benefits. The cauldron makers (Kalderashi) in Bulgaria and prevalence of Roma in informal sector employ- Romania, bear trainers (Ursari) and basket makers ment also limits their benefits based on social (Kosnicari) in Bulgaria. insurance contributions, including health care Roma may have multiple affiliations, such as and unemployment benefits. with an extended family group, as well as a geo- Political liberalization has also proven a mixed graphic and occupational subgroup (Liegeois blessing. The transition brought new opportunities 1994). The densest concentration of different for ethnic minorities to express their identity and Roma communities is found in Southeastern participate in society. In most countries, minorities Europe, where there is greater variation in reli- were once again recognized as distinct ethnic gious affiliation, dialect, and occupation (Fraser groups. For example, in Czechoslovakia, the new 1995). The degree of assimilation also varies Declaration of Basic Human Rights and Freedoms notably across subgroups, depending on "the adopted by the Federal Assembly in January 1991 amount of time they have lived...in the proximi- allowed for free determination of ethnic identity. ty of the dominant population, the size of the Roma political parties emerged in some countries, Gypsy community, familiarity with the majority as did a range of Roma NGOs. However, the tran- language, the presence of (an)other...strong sition also brought new civic challenges and hard- minority, and the history of interethnic relations" ships. Political liberalization let extremist parties (Bárány 2002). In Hungary, the most integrated onto the political scene and opened other avenues are the Romungro Roma, who speak Hungarian. 16 Introduction Use of the Roma language still prevails among way, Indians; their identity, their culture some Roma communities, and there are numerous would, however--regardless of all the dialects. In Bulgaria half of the Roma speak the transformations--remain sharply distinct Roma language at home.5 In the Slovak Republic from that of the gadze [sic] who sur- and Hungary, much less of the population does so. rounded them, and on whom their eco- However, in both countries, language barriers nomic existence depended (Fraser 1995). have been found to limit the school participation and performance of some children (Ministry of This distinction continues to influence Roma Labor 1997; Radó 1997). integration, participation in civil society and use The diversity of Roma creates significant chal- of public services. To varying degrees, Roma lenges for researchers. Information on Roma liv- communities have remained insular and sepa- ing conditions and poverty is scarce, fragmented, rate. While some Roma communities have inte- and often anecdotal. In addition to the difficulty grated, more traditional Roma communities and of drawing generalizations about such a diverse extended families are close knit, providing both group, measurement challenges include under- security and protection from the outside world sampling in censuses and household surveys; pri- (Wheeler 1999). This division between the Roma vacy legislation in many countries which prohib- and gadje worlds has reinforced stereotypes and it data collection by ethnicity; and the reluctance mistrust on both sides. Roma may be reluctant to of many Roma to identify themselves as such. send their children to state schools because of fear of losing their cultural identity. This concern like- Culture ly influences other aspects of life, including Given the striking diversity of Roma commu- employment preferences and use of health servic- nities, generalizing about the nature and charac- es. The distance between Roma and non-Roma teristics of Roma culture is extremely difficult. The communities breeds mistrust and misunder- literature paints a fragmented and sometimes con- standing among non-Roma and reinforces nega- tradictory picture. However, it is clear that aspects tive stereotypes and discrimination. of Roma social organization and values affect the The socially heterogeneous nature of Roma interactions of Roma and non-Roma, the dynam- society also influences the level of integration of ics among Roma subgroups, and many aspects of various Roma communities, their political par- their welfare. Cultural factors can influence the ticipation, and relations among different Roma level of integration of communities, participation groups. For example, traditional Roma groups in civil society and political institutions, demand may distrust or reject more integrated Roma. In for public services, and household behavior. Hungary, the more traditional Vlach Roma have Despite the complexity of the topic, there is few interactions with the Romungros Roma, and consensus concerning the importance of the rela- in Bulgaria the Kalderashi relate little to the poor- tionship between Roma and the "gadje," the er Ierlii, whom they believe have abandoned their Roma word for non-Roma. Roma define them- traditions (Stewart 1997; Iliev 1999). Little is selves as distinct and different from gadje. This known about the complex hierarchy among helps explain how Roma have maintained a sep- Roma groups, which is based not only on their arate identity across centuries, despite repeated adherence to Roma traditions but also to the pres- pressures for integration: tige of clans and occupational groups, religion, and other divisions. These factors may correlate Their ethnicity was to be fashioned and strongly with poverty and social exclusion. remoulded by a multitude of influences, internal and external, they would assimi- Social Exclusion and Discrimination late innumerable elements which had A defining aspect of poverty among Roma is nothing to do with India, and they would its relationship to social exclusion. Social exclu- eventually cease to be, in any meaningful sion and discrimination severely affects Roma 17 Roma in an Expanding Europe: Breaking the Poverty Cycle access to employment opportunities, education, The roots of such sentiments are difficult to and public services. Social exclusion refers to a trace but undoubtedly stem from a combination process of social separation between individuals of factors, including history, difficult economic and society (Rodgers et al. 1995; Silver 1994). conditions, and feelings of social insecurity. As Exclusion can have multiple dimensions, includ- mentioned earlier, aspects of Roma culture and ing economic, political, sociocultural, and geo- living conditions have reinforced stereotypes and graphic (Baker 2001). In economic exclusion, spurred marginalization. Self-exclusion of some individuals cannot participate in market activity, Roma can breed misunderstanding and mistrust including employment, access to credit, and among non-Roma. Similarly, the poverty of many land. Political exclusion refers to limitations on Roma communities contributes to resentment as participation in democratic processes, such as Roma are perceived as dependent on welfare ben- voting, participation in political parties and efits and burdens on the state. other associations within civil society. Sociocul- tural exclusion encompasses separation based POLICY APPROACHES AND DEBATES upon linguistic, religious, and ethnic grounds. European states' policies toward Roma histor- Geographic exclusion involves various types of ically have either aimed to further exclude Roma spatial differentiation. Moreover, different facets from majority societies--through expulsion, of exclusion often reinforce each other. For exam- forced ghettoization, and denial of services--or to ple, geographic exclusion in housing can lead to fully assimilate Roma into the majority society, economic exclusion if people are unable to find often through coercive measures. Policies of jobs where they live or attend mainstream public exclusion and forced assimilation, though differ- schools there. ent in many ways, share one important goal: both For Roma, social exclusion from majority seek to reduce the visibility of Roma communi- societies in Europe has mainly taken the form of ties--on the one hand by forcing them to the mar- ethnic discrimination. Roma have been shunned gins of society, on the other by forcing them to throughout European history, and ethnic ten- assimilate. Both deny Roma communities and sions have intensified in the transition period individuals the right their own culture. with the revival of nationalism in some coun- While the legacy of exclusionary and assimi- tries. Discrimination, both explicit and implicit, lationist policies lives on in Europe, current poli- permeates many aspects of life, including educa- cy approaches to Roma are built on different tion, employment, and housing. Roma have been foundations, emphasizing individual and group barred from restaurants and hotels in Central rights for ethnic minorities. This section discusses and Eastern Europe. Documented racial vio- historic and current policy approaches toward lence, including skinhead attacks and police vio- Roma within a conceptual framework that helps lence, has also been on the rise during the transi- to understand the influences and trends that tion period.6 shape current policy development. Stereotypes of Roma continue to be wide- spread throughout Central and Eastern Europe. Roma Policy: Four Approaches UNDP Human Development Reports for Bulgar- Policy approaches taken by European govern- ia and the Slovak Republic quoted opinion sur- ments in modern times fall into four broad veys that found deeply negative perceptions of groups: policies of exclusion, assimilation, inte- Roma to be pervasive. In Bulgaria, nearly 80 per- gration, and minority rights (see Marko 2000 for a cent of the population surveyed in 1999 said that similar typology). These approaches reflect differ- they would not want to have Roma as neighbors, ent responses to two basic questions about Roma a figure far higher than for any other ethnic or policy: whether Roma should be treated as a dis- social group, including former prisoners (UNDP tinct group or as individual members of a broad- 1999). Similar results have been reported from er society and whether Roma policy should be surveys in other countries in the region. pursued through coercive measures or with 18 Introduction Table 1.1: A Typology of Roma Policy Approaches in Europe Coercive Rights-Based Roma Treated as a Separate Group Exclusion Minority Rights Roma Treated as Individual Members of Broader Society Assimilation Integration respect for Roma rights. Table 1.1 shows that ed to protect the majority from perceived threats these policy approaches reflect different answers and are often pursued coercively. Lack of regard to these fundamental questions. for Roma rights and interests is justified by the In this discussion, Roma policy refers to both view that Roma are not members of the majority explicit governmental policies toward Roma, as community but rather dangerous parasites. well as other state policies that affect Roma Therefore, the majority community has no obliga- together with other social groups but may have a tion to concern itself with the welfare of Roma different effect on Roma. In addition, this discus- individuals or communities. While policies that sion also considers how official state policies set reinforce social exclusion are widely rejected in the tone for the unofficial attitudes of non-state international law today, their legacies persist. organizations, enterprises, and associations One of the most important of these legacies is whose practices toward Roma also relate to their housing segregation--a form of geographic social status and poverty. As with any typology, exclusion. Sociocultural exclusion of Roma in these definitions are ideal types; some policies Europe has long been underpinned by housing will not fit neatly into one or another of these cat- policies shunting many Roma into separate set- egories, and some may be explicitly geared at tlements or ghettos. Under the Ottoman empire, blurring lines of distinction. Nonetheless, this typ- urban neighborhoods, or mahalas, were organized ology captures the broad logic of policies toward along religious and ethnic lines. As a result, many Roma in Europe over time and reflects enduring Roma neighborhoods in the Balkans--such as the differences in how societies address Roma issues. large Roma enclave of Suto Ozari, in Skopje, FYR Macedonia--have their roots in longstanding Policies of Exclusion policy legacies. In Slovakia, policies enacted dur- As noted above, exclusion of Roma from ing and after World War II forced Roma to settle majority societies in Europe results partly from on the outskirts of towns, leading to the creation beliefs, attitudes, and behaviors of Roma them- of a large number of Roma settlements. Roma selves. However, this self-marginalization is relat- also live in ghettos on the outskirts of cities in ed to a long legacy of European policies that Western European countries such as Italy (Euro- sought to reinforce Roma exclusion. Policies of pean Roma Rights Center 2000). exclusion seek to exclude Roma from the majori- More recent policies in Central and East Euro- ty society along economic, political, sociocultural, pean countries have--both directly and indirect- and geographic dimensions. ly--led to continued geographic marginalization. Why have European governments often In one notorious case, Czech authorities erected a sought to exclude Roma? The rationale is usually wall around a Roma settlement in the town of based on a racial and nationalist perspective that Usti nad Labem. The wall was later torn down holds Roma to be inferior and separate from the after protests from Roma, the international com- majority. Contact and intermarriage between munity, and Czech political leaders. As the study Roma and the majority community is seen as of Slovakia shows in Chapter Three, geographic harmful. Exclusionary policies are usually enact- exclusion of Roma powerfully reinforces social 19 Roma in an Expanding Europe: Breaking the Poverty Cycle exclusion of other kinds, including access to found, "There are few, if any, other population employment and state services. groups in Europe against which regular racist pro- Beyond geographic exclusion, current policies nouncements and actions still pass largely unre- at the national and local levels continue to marked" (Save the Children 2001a; OSI 2001). exclude Roma from public services, such as Such outbursts rarely cost the officials their jobs. health and education. Such policies may have a Reinforcement of exclusionary norms by public critical impact on Roma poverty. The OSCE has officials are an unofficial policy of exclusion. documented extensive evidence of continued dis- crimination in the justice system, housing, educa- Policies of Forced Assimilation tion, and other areas (OSCE 2000; OSCE ODIHR Unlike exclusionary policies, policies of 1997). Roma children often are excluded from forced assimilation aim to eradicate differences education in mainstream public schools in Cen- between Roma and non-Roma, by making Roma tral and Eastern Europe and instead relegated to adopt the norms, values, and behaviors of the schools for the mentally handicapped. mainstream. Roma in Central and Eastern Europe also Like policies of exclusion, assimilationist poli- have been stripped of fundamental political cies are by definition coercive. However, assimi- rights, including citizenship. A notorious law lationist policies tend to be undertaken not to enacted after the division of Czechoslovakia harm Roma but to help them. Assimilationist forced non-Czech citizens to reapply for Czech logic asserts the benefits of belonging to the citizenship including provisions that prevented majority culture and participating in economic many Roma originally from Slovakia from win- life, and takes the view that all individuals would ning Czech citizenship (Orentlicher 1998). Lack of be better off if they were elevated to full member- citizenship can prevent people from acquiring ship of this culture. Assimilation is often con- property, voting, working, and receiving educa- ceived as a "civilizing mission," helping margin- tion, health care, and social assistance. Political al or outside groups win greater prosperity and rights are also important for allowing Roma to culture. Opponents of assimilation argue that assert their economic interests. assimilation often entails repression, losses to Economic exclusion of Roma in Central and minority groups and cultures, and disproportion- Eastern Europe often results not from official ate benefits to the majority group. However, state policy but from the actions of other actors, assimilated individuals are often granted nearly particularly businesses and social associations. full rights in the majority society. Many firms in the transition countries do not hire Assimilationist policies have been common in Roma, compounding the labor market woes of a Europe for centuries. An assimilationist policy population with low skills and education levels. approach was adopted by the Austrian Empire Governments may foster employment discrimi- and Hungarian Kingdom under the modernizing nation by not acting effectively to prevent it. rule of Empress Maria-Theresa (Bárány 2002). Other acts of exclusion toward Roma are sim- Maria-Theresa issued four "Gypsy decrees" ilarly outside the direct control of the state--such between 1758 and 1773 that "ordered all Gypsies as barring Roma from restaurants and clubs, skin- to settle, pay taxes, and do mandatory service to head attacks, and the portrayal of Roma by the churches and landowners... prohibited their press as "the most problematic section of the pop- leaving the villages to which they were assigned ulation, disturbers of the social order" (PER without permission... mandated compulsory 1997b). In these areas too, governments' failure to military service... eliminated the authority of take firm action can reinforce exclusionary social Romani leaders over their communities, banned practices by signaling their acceptability. traditional Gypsy dress and the usage of Romani Such signaling almost undoubtedly occurs language... forbade marriages between Gypsies through the expression of anti-Roma sentiment by and ordered Roma children over age five to be state officials in public. As Save the Children taken away to state schools and foster homes" 20 Introduction (Bárány 2002). Maria-Theresa did not shy away members. In these approaches, Roma individuals from coercive measures to promote assimilation. retain their cultural identity while adopting much Assimilation was also the predominant Roma of the lifestyle and practices of the dominant policy of socialist regimes in Europe after the Sec- society. ond World War. Following the lead of Karl Marx, Critics of integration warn that it shares the socialist regimes believed in advancing "common flaws of assimilation, since both approaches aim interests of the entire proletariat, independently to subsume Roma in the broader society and to of all nationality" (Marx 1985). In practice, this downplay the importance of ethnicity. However, meant promoting cooperation between different policies of integration genuinely differ from those ethnic groups and nationalities, with the goal of of assimilation. They are inspired by modern, lib- forging an undivided, classless socialist society. eral values that "favor broad political participa- Policy toward Roma was therefore guided by an tion of all those within the geographic boundaries effort to merge the population into the proletari- of a given state, regardless of their ethnic identi- an mainstream. ty" (Wippman 1998). Integration policies also Thus in the 1950s and 1960s, most socialist typically respect individual rights and individual regimes in Europe engaged in a strong, multi- choices about how to integrate, leaving room for pronged policy initiative to assimilate Roma (Ulă continued ethnic identification. Assimilation poli- 1991). "The fundamental goal was to assimilate cies do not. them and transform them into productive, coop- Integration policies seek to integrate Roma, erative, and supportive socialist citizens" (Bárány without coercion, into the majority society while 2002). This was to be achieved through improved protecting their individual rights. As Pace housing, higher educational enrollment, and expresses it, "[a]ssimilation refers to the absorp- guaranteed employment. Many of these policies, tion of a minority group into the host or majority however, were pursued with a heavy hand. Settle- society, with consequent dissolution of the cultur- ments were broken up, housing was assigned, and al features of the group...Integration means that work was made mandatory under threat of an ethnic group tries to maintain some or all of its imprisonment. Roma generally were not given the cultural characteristics, while seeking to mini- opportunity to participate in decisionmaking or in mize the practical problems inherent in adapting the administration of these policies (Bárány 2002). to the dominant society" (Pace 1993). Neither policies of exclusion nor policies of Philosophically, integration policies are forced assimilation allow room for individual based on a belief in progress, individual rights, choice, or individual rights. They are often pur- and equal opportunity. Proponents of integration sued, at least in part, through official coercion. tend to believe that modern society is better than However, with the rise of a liberal democratic traditional society, providing forms of human international order during the latter half of the development unavailable in the past. Members twentieth century, both of these models of minor- of more traditional groups, such as Roma, can ity policy began to be discredited, at least in the benefit from integration if it facilitates individual eyes of international law and organizations such growth and well-being. Proponents of integra- as the UN, OSCE, and EU (Wippman 1998). These tion also argue that no individual should be dis- trends opened the way for two rights-based poli- criminated against, and that all individuals cy approaches to emerge: policies of integration should be allowed to progress in society to the and minority rights. best of their abilities. Integration has been the dominant European Policies of Integration policy paradigm toward ethnic minorities since Policies of integration and minority rights dif- the 1970s (PER 1997a), except in the former fer on whether rights are accorded primarily to socialist states. It has also been the dominant par- individuals or groups. Policies of integration adigm in international law (Wippman 1998). focus on bringing individuals into society as full Some examples include integrating Roma into 21 Roma in an Expanding Europe: Breaking the Poverty Cycle regular school systems, banning labor market On the basis of these emerging European discrimination, increasing access to social servic- standards, the 1993 Copenhagen Summit of the es, addressing housing discrimination, and European Commission included "respect for reducing ghettoization. All these policies seek to minorities" as one political criterion for the acces- provide individuals with equal rights and the sion of new member states.11 This has shaped pol- same opportunities as members of the dominant icy toward Roma in the accession states. society. Both the OSCE (formerly the CSCE) and the Council of Europe (CE) have been actively Minority Rights Policies engaged in constructing a framework for policy Starting in the 1990s, European and interna- on minorities, including Roma.12 Over 40 years tional policies toward minorities have increasing- ago, the CSCE led the way in taking on the issue ly emphasized group rights (Wippman 1998; Pejic of minorities in Europe. The Roma question was 1997; PER 1997a; Save the Children 2001b). This explicitly addressed a series of Human Dimen- reflects "a growing acceptance of the legitimacy sion meetings held in the 1990s. In 1995, the of group consciousness" in Europe--and, indeed, OSCE created the Contact Point on Roma Issues the world (Basurto 1995). This minority rights within the Office of Democratic Institutions and approach differs from the integration approach, Human Rights (ODIHR), to focus on Roma rights since it advocates the establishment and protec- and protections in general. In 1998, the Contact tion of group, rather than individual, rights as the Point's mandate was extended to "oversee, coor- basis of minority policy. dinate and advise on legislative and policy devel- The minority rights approach stresses the opments affecting Roma (and Sinti) both at the importance of cultural preservation as a means of European and state levels" (Kováts 2001a). improving the condition of minority groups. The Council of Europe has demonstrated a Minority rights advocates suggest that the situa- concern for minority issues for many years, tion of socially marginalized groups, such as including the development of a convention on lin- Roma, will not be improved simply by integrat- guistic rights and protections for Roma.13 In 1993, ing individuals into the majority society. Instead, a Council resolution declared Roma to be "a true their welfare will be secured best by enhancing European minority," and established a Specialist opportunities for group empowerment and cul- Group on Roma/Gypsies. Together with the tural self-determination. OSCE High Commissioner for National Minori- The last two decades have seen a growing ties, the Specialist Group produced the Guiding international concern for the rights of minorities Principles for Improving the Situation of Roma in can- in Europe. Intergovernmental organizations such didate countries. Adopted by the EU in 1999, this as the OSCE, EU, and the Council of Europe have document has been influential in shaping EU rela- taken a particularly active role in establishing tions with post-communist countries regarding minority rights. The result is an emerging "com- the Roma issue, as well as marking a convergence mon European standard" for minority policy, in CE, OSCE, and EU approaches to Roma policy. grounded primarily in international commit- Over the years, the Council has undertaken vari- ments undertaken by European states (De Witte ous initiatives and have had an indirect influence 2002). These include the European Convention on on Roma through their work in the field of minor- Human Rights,7 the Copenhagen Document (1990),8 ity and linguistic rights. the Framework Convention for the Protection of The European Union--founded to build eco- National Minorities (1995),9 and the new EC Treaty nomic cooperation in Europe ­historically has not article 13, established within the Amsterdam engaged directly in minority policies. As a result, Treaty (1997, the first treaty provision to explicit- through the principle of subsidiarity, education, ly include antidiscrimination measures relating to culture, and language have predominantly ethnic minorities), and the EU Charter on Funda- remained the policy concerns of member states, mental Rights (2000).10 not the Community. Nevertheless, the Treaty of 22 Introduction Maastricht, signed February 1992, established the head violence. Roma communities are also divid- EU as an economic and political union. It also ed between those who advocate more integration opened the door for the EU to include within its with majority societies in Europe and traditional- scope some actions pertaining to culture, provid- ists who want to maintain a distinct identity. Such ing that the EC shall "contribute to the flowering divisions may be reflected in debates over of the cultures of the Member States, while whether to emphasize teaching of Roma lan- respecting their national and regional diversity." guage and culture in schools, in an effort to pre- In the context of minority rights, this article rec- serve and promote Roma culture, or to emphasize ognized the existence of diversity within and early education programs that train Roma stu- between its member states, as well as the impor- dents in the majority language and culture. tance of EU and member state support for pre- The Roma leaders Nicholae Gheorghe and serving this diversity. Andrej Mirga hold that there is no fundamental Most Central and East European countries contradiction "between integration and maintain- have, at some level, accepted the importance of ing a Romani identity. It is rather a question of a protecting national minority rights. Those aspir- conscious attempt to modernize the Romani iden- ing to join the EU are beginning to undertake tity without necessarily implying its abandon- reforms. As of March 2001, 33 states had ratified ment. Thus, integration or even partial assimila- and entered into force the Council of Europe tion, which would lead to an undifferentiated Framework Convention for the Protection of National incorporation of the Roma into mainstream soci- Minorities (Council of Europe 1995), the first legal- ety, can be regarded as a worthy ideology by ly binding multilateral instrument devoted to the Romani elites. The fear of losing their identity, protection of national minorities in general. All strongly endorsed by the traditionalists, should EU accession candidate countries have signed, be overcome by a serious reassertion and redefin- ratified, and entered into force the convention, ition of the Romani identity" (PER 1997a). How- with the exception of Latvia and Turkey. Six EU ever, a movement to create a modern Roma iden- member states have not ratified the Framework: tity more compatible with modern economic France, Finland, Belgium, Luxembourg, Nether- development and integration into European soci- lands, Portugal (Goldston and Guglielmo 2001). eties, would have to come from the Roma com- In Central and Eastern Europe, some have munity itself. And one unique aspect of Roma complained of a gap between this broad political culture in Europe is that such a movement has agreement and effective action, including legal not occurred. enforcement by the European Court of Human Both individual and group rights approaches Rights.14 Existing domestic institutions address- complement economic and social development ing minority issues also may lack the resources, approaches designed to address the roots of Roma or the mandate, to coordinate and enforce policy poverty and exclusion, but they are not substi- implementation (OSI 2001). Still, minority rights tutes for them. Rights based approaches alone as a distinct approach to Roma issues has been cannot reduce poverty and promote economic gaining ground in Central and Eastern Europe and social development. Roma need not just for- (Pogany 1999). mal rights but real economic opportunities. Creat- ing these opportunities goes beyond the usual Tensions Between Policy Approaches legal protections that rights-based approaches Current policy toward Roma in Europe is normally encompass. Furthermore, instruments shaped by the tensions between various policy to enforce economic and social rights are weak approaches, as well as the legacies of past poli- and in most cases cannot be addressed directly cies. Legacies of exclusion, for instance, live on through legal standards and mechanisms. Only a and conflict with newer policies of integration ­as few of the constraints faced by Roma can be seen in disputes about banning Roma from pub- addressed through legislative measures. The final lic establishments or about ways to reduce skin- chapter of this volume proposes addressing Roma 23 Roma in an Expanding Europe: Breaking the Poverty Cycle poverty through inclusive policies that comple- 2. OSCE 2000; OSI 2001; Save the Children ment rights-based measures and tackling the eco- 2001; UNDP 2003. nomic and social issues facing Roma. 3. Europe and Central Asia (ECA) refers to the former socialist countries of Central and Eastern CONCLUSIONS Europe and the Former Soviet Union. After a difficult transition to market econo- 4. There is substantial international evidence mies and democratic political regimes, the coun- that welfare and socioeconomic status can have tries of Central and Eastern Europe face serious an ethnic dimension, including analysis on the challenges in addressing poverty and social disparities in welfare between blacks, whites, exclusion. Nowhere are these problems more and native Americans in the United States, the acute than for Roma. Transition has had a worse conditions of indigenous peoples in Latin Amer- impact on Roma than on other groups for a vari- ica, and the status of ethnic minorities in other ety of interconnected reasons: legacies of past parts of the world. For a review of the literature, policies, low skill levels and educational attain- see Psacharopoulous and Patrinos (1994). ment of Roma themselves, a tendency toward 5. This differs significantly across subgroups, cultural separation, a history of poor relations ranging from 14 to 85 percent (Tomova 1998). with the mainstream societies and states of 6. The European Roma Rights Center has ex- Europe, poor policy responses, and a reduction in tensively documented discrimination and human social spending caused in large part by macro- rights violations of Roma. Regular updates and economic decline. country reports can be found at: www.errc.org. Addressing Roma poverty requires, first of 7. Article 6 of the Treaty on European Union all, understanding it. Therefore, the following refers to the European Convention for the Protec- chapters will set the stage for a deeper policy dis- tion of Human Rights and Fundamental Free- cussion by asking: What distinguishes Roma doms, which has been ratified by all European poverty from that of other groups in the region? states. The protection of individuals belonging to Chapter Two presents the results of surveys that minorities are considered to be "an inherent part" seek to answer this question, while country chap- of the EU policy on human rights. The Conven- ters on Slovakia, Romania, and Hungary provide tion's Article 14 states that the rights and free- a more in-depth look at Roma poverty in selected doms laid down in the Convention should "be communities, using interviews with Roma them- secured without discrimination on any ground selves. Country chapters also explore the ques- such as sex, race, colour, language, religion, polit- tion: How have the transition countries of Central ical or other opinion, national or social origin, and Eastern Europe addressed the Roma issue to association with a national minority, property, date? In particular, the chapter on Hungary birth or other status." See http://europa.eu.int/ reviews a number of Roma programs and policies comm/external_relations/human_rights/rm/. and the chapter on Spain provides an example of 8. The Copenhagen Document is sometimes how west European countries have addressed referred to as the "European Constitution of Roma issues. Finally, the book concludes with the Human Rights." It was adopted in 1990 by the lessons of this experience, and new strategies for Conference on the Human Dimension of the Con- the future. ference on Security and Cooperation. While legal- ly nonbonding, it explicitly recognizes the impor- NOTES tance of national minorities. 1. The countries invited to join the Copen- 9. The Framework Convention, developed by hagen Summit of the EU on December 12, 2002 the Council of Europe in 1995, entered into force include Cyprus, the Czech Republic, Estonia, in February 1998; it is legally binding under inter- Hungary, Latvia, Lithuania, Malta, Poland, the national law and contains principles that each Slovak Republic, and Slovenia. Bulgaria and Contracting Party must implement through Romania are in accession negotiations. national legislation and government policies. 24 Introduction 10. The EU Charter on Fundamental Rights 13. The Council of Europe's European Charter lays down the equality before the law of all people for Regional or Minority Languages (1992) con- (Article 20), prohibits discrimination (Article 21), tains provisions which may be applied to "non- and requests the Union to protect cultural, territorial" languages such as the Roma language. religious and linguistic diversity. See http:// 14. The European Court of Human Rights europa.eu.int/comm/external_relations/ recently noted that while there was an "emerging human_rights/rm/ . international consensus... recognizing the special 11. These political conditions were deter- needs of minorities and an obligation to protect mined during the European Council meeting of their security, identity and lifestyle... [the divided June 1993. According to the concluding docu- Court itself is] not persuaded that the consensus is ment, "membership requires that the candidate sufficiently concrete for it to derive any guidance country has achieved stability of institutions as to the conduct or standards which Contracting guaranteeing democracy, the rule of law, human States consider desirable in any particular situa- rights and respect for and protection of minori- tion" (from Chapman v. United Kingdom, UCHR, ties" (Conclusions of the Copenhagen European Judgment of 18 January 2001 (No.27238/95), in Council 1993). Open Society Institute 2001). 12. See also Kováts 2001a and 2001b. 25 Chapter Two: ROMA POVERTY AND WELFARE: AN OVERVIEW P overty is a multidimensional phenomenon order to pull together a comprehensive view of that goes well beyond low income or lack of Roma welfare and living conditions. material consumption.1 According to the World Bank's 2000­2001 World Development MEASUREMENT CHALLENGES Report Attacking Poverty, poverty encompasses Data on social welfare in Central and Eastern such things as the psychological pain of being Europe are plagued with problems due to weak poor, a sense of vulnerability to external events, and sometimes biased statistical systems inherit- and powerlessness toward the institutions of ed from the socialist era and the use of definitions state and society (World Bank 2001a). The Coun- and methodologies that are often outdated, cil of Europe (1995) has defined poverty as affect- inconsistent with international standards, or not ing those "persons, families or groups of persons comparable across countries. These issues, how- whose resources (material, cultural, and social) ever, pale in comparison with the challenge of are limited to the extent that they exclude them measuring socioeconomic conditions of Roma. from the minimally accepted lifestyle of the coun- Seemingly straightforward questions, such how tries where they live." many Roma live in a particular country, prove In the case of Roma, poverty is particularly extremely challenging. multifaceted. Many Roma are deprived of the Different approaches among surveys fre- resources necessary for adequate living condi- quently yield contrasting results and impede tions, as well as access to opportunities and chan- comparability of data. For example, some house- nels for participation. These problems are often hold surveys ask respondents to identify their interconnected. This chapter synthesizes evi- ethnicity, while others ask the interviewer to indi- dence from primary and secondary sources to cate the respondent's ethnicity, or to determine illustrate the interrelated challenges facing Roma ethnicity by asking about the respondent's native in social welfare, housing, education, and health language. The latter approach may underestimate status. This sets the stage for further analysis of the number of Roma, many of whom do not poverty and welfare in the following country speak Roma dialects. Other obstacles exist in the chapters on Slovakia and Romania, which look analysis of administrative data, such as education further at geographic and social exclusion, and and labor market statistics. Several countries the diversity of living conditions among Roma have stopped collecting data by ethnicity because communities. of privacy legislation. Czechoslovakia stopped The chapter begins with a discussion of some collecting data on students by ethnicity in 1990, of the particular issues that arise in the analysis of and Hungary followed suit in 1993 (ERRC 1999; data regarding Roma, and identifies the caveats Radó 1997). Government officials are also fre- that should be considered in interpreting the quently reluctant to inquire about ethnicity in information. While the gaps and limitations of surveys, for fear of raising ethnic tensions. the information base on Roma are real, this does More fundamental questions about ethnicity not invalidate the entire body of analysis. and identity complicate the assessment of wel- Throughout this volume information from multi- fare. Some Roma do not consider themselves ple sources and perspectives are presented, in Roma or affiliate with a different ethnic group. 26 Roma Poverty and Welfare: An Overview An ethnic Roma living in Hungary may feel sions of poverty. The following chapters use qual- more Hungarian than Roma, or vice versa. For itative analysis to identify some non-measurable the purpose of analysis in this volume, Roma are aspects of welfare and exclusion. defined broadly to include both those who iden- Poverty is usually measured using a national- tify themselves as Roma and those identified by ly representative household survey that assesses others as Roma. This stems both from the data the welfare of the population. Welfare indicators, sources used and from the policy focus of this including poverty rates, are constructed using analysis. After all, if policies affect ethnic minori- either consumption--measured by household ties, they will do so regardless of how people expenditures on food and non-food items--or identify themselves. household income. Consumption data are gener- Another unique challenge of research on ally considered more reliable; there are substantial Roma is the legacy of biased research. Early stud- problems with measuring income, including the ies on Roma in the late nineteenth century in difficulty of capturing in-kind income. Individu- Western Europe sought to confirm theories about als may also be reluctant to report income from genetic inferiority (Fraser 1995). Recent works informal activities for fear of having to pay taxes. reviewed in the Czech and Slovak Republics were The disincentives to reporting consumption are found to have a social Darwinist slant (ECO- less problematic, but methodological questions HOST 2000). More recent scholarship on Roma also remain here, including what to include as may suffer from political biases. Roma leaders consumption and the difficulties that respondents and activists have an interest in portraying the have in recalling household expenditures. situation as worse that it may actually be, while Once the welfare measure is constructed, government reports may gloss over failings to poverty rates are usually defined as that share of present a more favorable picture (Bárány 2000). the population living below a designated poverty A further caveat is warranted. The diversity line. There are many possible poverty lines. The of Roma impedes generalizations at the regional most commonly used lines for analysis are and country level. In addition to notable ethnic absolute lines, related to basic nutritional and differences, there is significant diversity among social needs, or relative poverty lines, which are Roma settlements: rural/urban, assimilated/ related to prevailing income levels such as one- non-assimilated, homogenous/heterogeneous, half, or two-thirds of mean income per capita (per as well as religious affiliations. Some groups person). Relative lines are useful for measuring speak variations of the Roma language, while poverty at the country level and for international others do not. For analytical purposes, this report comparisons of the characteristics of the most assumes some commonalities across countries deprived individuals in a country. Many interna- and groups, but its conclusions are necessarily tional comparisons of poverty rates are based on tentative. The qualitative case studies presented relative lines.3 in subsequent chapters illuminate some of these In addition to these measures, the World Bank differences. uses two absolute poverty lines to compare poverty across countries: $2.15 purchasing power How is Poverty Measured? parity (PPP) per capita per day and $4.30 PPP per Measuring poverty is an inherently subjective capita per day.4 The adjustment to purchasing task fraught with methodological complexities.2 power parity accounts for differences in price lev- There is no correct or scientific method. Empirical els across countries. These standard poverty lines analyses of poverty generally focus on measuring allow comparisons of real values between coun- income poverty, and therefore provide only a par- tries. This chapter, uses quantitative, income- tial picture. In this chapter, quantitative measures based definitions of poverty and shows how of poverty are complemented with other data these connect with other dimensions of social sources--for example on education and health exclusion in housing, labor markets, education, status--to fill in some of the non-income dimen- and health services. 27 Roma in an Expanding Europe: Breaking the Poverty Cycle AN ANALYSIS OF ROMA POVERTY Poverty rates for Roma in all three countries IN THREE COUNTRIES are strikingly high--in all cases several times The following section looks at poverty among higher than among non-Roma.6 Table 2.1 summa- Roma in three countries: Bulgaria, Hungary, and rizes the poverty rates for all three countries Romania. Together these three countries com- under the three different poverty lines--a relative prise a significant share of the Roma population line amounting to half of median per capita and in the region. It relies on the household survey per equivalent adult expenditures, and then the mentioned in Box 2.1 which was conducted by two international poverty lines, $2.15 and $4.30 the Center for Comparative Research in the Soci- per person, per day, adjusted for purchasing ology Department of Yale University in 2000.5 power parity. The survey was the first of its kind to address the The highest level of absolute poverty among ethnic dimension of poverty across countries and Roma households lies in Bulgaria, followed allows for a comparative quantitative assessment closely by Romania. Even at the lower $2.15 line, of the living conditions of Roma in the region. In 41 percent of all Roma households in Bulgaria each of the three countries, Roma were oversam- and 38 percent in Romania are found to be pled to allow for a more statistically robust pic- poor--a strikingly high proportion. At the high- ture of their living conditions. er line of $4.30 PPP per capita, 80 percent of Annual household expenditures are used as Roma households in Bulgaria and almost 70 per- the main measure of household welfare. Because cent of those in Romania are poor. Poverty measures of poverty are very sensitive to the among non-Roma households at the $4.30 line in composition of the household, two sets of results both of these countries is also high, but less than were calculated based on (i) per capita expendi- the levels among Roma. Although absolute ture (obtained by dividing total household poverty among Roma households is lower in expenditure by the number of household mem- Hungary, the difference between the situation of bers); and (ii) per equivalent adult expenditures, Roma and non-Roma households is equally (where expenditures are adjusted for both the stark. About 7 percent of Roma households in size and composition of the household). In gener- Hungary are poor based on the $2.15 line, as al, this adjustment for household size (per capita compared to only 0.5 percent of non-Roma or equivalent adult) tends to yield much larger households. At the higher $4.30 absolute poverty differences in poverty risks between Roma and line--arguably a more appropriate one for pros- non-Roma than using unadjusted (per capita) perous Hungary--as many as 40 percent of household expenditures, because Roma house- Roma households are poor, compared to 6.9 per- holds tend to be much larger. cent of non-Roma households. Table 2.1: Poverty Rates among Roma and Non-Roma Households, 2000 50% of median $2.15PPP $4.30PPP Country Per equiv. adult Per capita Per capita Per capita Bulgaria Roma 36.1 37.2 41.4 80.1 Non-Roma 3.8 3.4 4.1 36.8 Hungary Roma 24.5 26.3 6.6 40.3 Non-Roma 4.5 3.6 0.5 6.9 Romania Roma 39.5 43.1 37.6 68.8 Non-Roma 10.9 11.1 7.3 29.5 Sources: Yale dataset; Revenga et al. 2002. 28 Roma Poverty and Welfare: An Overview Box 2.1: Who Are Roma? Estimating the number of Roma in a country is both difficult and controversial. Household surveys and census data rarely include questions on ethnicity beyond asking individuals to report their ethnicity. A household survey conducted by a team of researchers from Yale University in 2000 experimented with dif- ferent approaches to asking about ethnicity. The results provide lessons for the design of future surveys. The dataset takes a multifaceted approach, including questions on self-identification (asking the interviewee to report their ethnicity), interviewer identification (asking the interviewer to identify the ethnicity of the inter- viewee), language, parents' language, appearance, and family name. This approach allows for analysis based upon differing definitions of ethnicity. The Roma population can be estimated in different ways using the survey data (Table 2.2). After identi- fication by the interviewer, self-identification yields the largest populations. Very few individuals who report being Roma were not identified by the interviewer as Roma--two in the case of Bulgaria, and none in Hun- gary and Romania. On the other hand, the interviewers identified many people as Roma who did not iden- tify themselves as Roma. In Romania, 61 percent of those identified as Roma by the interviewer did not self- identify. The corresponding shares for the other two countries are 38 percent in Hungary, and 24 percent in Bulgaria. It is difficult to know how to interpret these results. It may be that Roma in Romania are more inte- grated and feel more Romanian than Roma. Conversely Roma in Romania could be more afraid of identify- ing as Roma than Roma in the other countries. Table 2.2: Roma Population Sizes by Type of Identification, 2000 (% of random sample) Bulgaria Hungary Romania Self-Identification 6.6 3.1 1.2 Language 6.2 1.7 1.6 Mother's Ethnicity 7.2 3.1 1.1 Father's Ethnicity 6.9 3.5 1.2 Spouse's Ethnicity 6.6 1.1 0.8 Interviewer Identification 8.7 5.0 3.1 Sources: Yale Dataset; Revenga et al. 2002. It could also be that only certain groups of Roma self-identify as Roma. The share of the population self- identifying as Roma is relatively close to the share of the population who report speaking the Roma language at home. For example, in Hungary, only the Wallach Roma speak the Roma language. The other two main groups of Roma in the country--the Beash and Rumungro Roma--generally speak Romanian, and Hungar- ian respectively. This could mean that self-identification is more likely to capture the Wallach, while other Roma are less likely to consider themselves Roma. The data also suggest that ethnic identity may be weak- ening over time. In Bulgaria and Hungary, the share of respondents who identify their parents as Roma is higher than the share who identify themselves as Roma. The differences in poverty rates between among Roma households, it is close to 25 percent Roma and non-Roma when using the relative in Hungary and about 37 percent in Bulgaria. In poverty line are also very large. On an equivalent Romania, the differences between relative pover- adult basis, Hungary and Bulgaria look fairly ty rates for Roma and non-Roma are equally similar: relative poverty among non-Roma large, but poverty among the non-Roma is notice- households oscillates around 4 percent, while ably higher than in Bulgaria or Hungary, indicat- 29 Roma in an Expanding Europe: Breaking the Poverty Cycle ing a more skewed distribution of expenditure for varies significantly across countries and between all households. Roma and non-Roma families (Table 2.3). As expected, poverty looks worse among The risk of poverty is highest among families Roma households when using the per capita line, where the household head has little education or which basically reflects the fact that Roma house- is unemployed, as well as among families with holds have a larger number of children. The per three or more children. But the association capita figures treats every household member as between poverty and these correlates appears having the same consumption needs, whereas the stronger for non-Roma families than for Roma. figures based on per equivalent adult measures For example, among non-Roma families where assume children have lesser consumption needs. the household head has no education at all, the poverty rate is several times that of families Correlates of Poverty where the head has secondary education. Among Why are poverty rates so different between Roma families, poverty tends to be relatively high Roma and non-Roma households? In large part, irrespective of educational attainment (with the this is due to differences in the underlying corre- possible exception of Hungary). Similar results lates of poverty, especially educational achieve- occur with respect to employment status: among ment, employment status, and household size. non-Roma families, the risk of poverty in house- The main correlates of poverty for Roma and holds where the head is unemployed is many non-Roma alike are the employment status of the times that of households where the head is head of the household, educational achievement employed, but among Roma families headed by of the household head, and the number of chil- an employed person, the risk of poverty remains dren, although the nature of the relationship high. Taken together, the evidence suggests a Table 2.3: Main Poverty Correlates, 2000 (poverty rate in %) Bulgaria Hungary Romania Roma Non-Roma Roma Non-Roma Roma Non-Roma Education: No school 31.2 0.0 39.6 47.4 84.3 16.5 Primary 39.8 5.3 27.2 5.07 42.6 16.0 Secondary 31.3 2.9 6.2 1.47 31.8 7.5 Higher 0.0 0.8 0.0 0.0 0.0 0.6 Employment status of household head: Employed 20.5 3.3 15.3 2.5 15.6 8.5 Unemployed 48.5 9.1 34.5 15.1 45.3 26.1 Out of the labor force 46.7 2.2 50.3 6.9 69.4 20.5 Retired/disabled 17.4 1.2 19.8 2.7 32.9 7.1 Number of children Zero 25.2 2.5 14.2 2.8 27.3 7.3 One 34.7 6.1 23.3 3.7 38.8 13.5 Two 49.1 5.6 29.0 9.9 52.7 26.4 Three 59.2 15.8 42.0 11.2 73.4 50.3 Four 65.5 0.0 82.8 44.2 59.9 64.9 National Poverty Rate 37.2 3.4 26.3 3.6 43.1 11.1 Note: Poverty line is equal to 50 percent of median of per capita expenditure. Sources: Yale dataset; Revenga et al. 2002. 30 Roma Poverty and Welfare: An Overview strong association between Roma poverty and even when controlling for other characteristics. In education, employment, and household size. other words, if the other household characteris- However, for Roma, the probability of being poor tics are held constant, per adult equivalent expen- is higher than that of non-Roma, irrespective of diture of Roma households is between 20 and 40 educational achievement and employment status. percent lower in the three countries than that of Although these poverty correlates (education, non-Roma households--a striking difference. The employment status, number of children) are asso- other household characteristics also affect wel- ciated with a high risk of poverty, households fare. The number of children, for example, is with these characteristics do not necessarily con- strongly negatively associated with per adult stitute the bulk of the poor. In fact, the composi- equivalent consumption in all three countries. tion of the poor largely reflects the weight of each Employment is positively associated with welfare demographic group in the overall population. in all cases, while unemployment shows a nega- Among non-Roma families, a sizeable fraction of tive association (although not always a strongly the poor are the so-called working poor--in other significant one). The relationship between educa- words, the head of household is employed. tion of the household head and household wel- Among Roma, the fraction of household heads fare is positive, as expected, but there are notice- who are working is much lower, and their weight able differences in the returns to education across in the composition of the poor is correspondingly countries. Returns to higher education--in terms lower. In the large majority of poor Roma fami- of higher household consumption--are high in lies, the head of the household is unemployed. all three countries but highest in Romania. Similar differences exist by educational Additional analysis looked at factors influ- attainment. While among non-Roma a sizeable encing welfare for Roma only. There is no reason fraction of poor heads of households have pri- why returns to education or other characteristics mary or secondary education, the bulk of poor should be the same for both Roma and non- Roma households are headed by someone with Roma. If Roma families live in different areas, primary or less than primary education. engage in different activities, or make different decisions regarding household investment and Multivariate Analysis of Roma Poverty consumption, then the returns to household char- The previous discussion focused on a one- acteristics--in terms of welfare--may be quite dimensional analysis of poverty, examining how different.7 Such differential behaviors, while ben- poverty rates differ across households based on a eficial in the short run, can reduce long term single characteristic such as education or employ- prospects for escaping poverty. For this reason, ment status. But many household characteristics additional analysis was undertaken for the Roma are often correlated among themselves. For exam- households only, including variables of little rele- ple, households where the head has a low level of vance to the majority population, but important education are more likely to be poor; household to Roma welfare. heads with low education may also face a higher One factor shown elsewhere to influence probability of being unemployed; and being behavioral patterns is location. Residential differ- unemployed is also correlated with a higher entiation or segregation can lower returns to pro- probability of being poor. Does low education ductive endowments for minority groups relative increase the risk of poverty directly? Or does it to the returns on the same endowments for the increase poverty through its impact on employ- overall population (van der Walle and Gunewar- ment status? Or both? To answer these questions, dena 2001; Nord 1998). For Roma, the effect of multivariate regression analysis is needed to con- location is probably best captured by the differ- trol for the differential influences of diverse fac- ence between those living in a Roma settlement tors. The following highlights these findings. versus those living in a more integrated neigh- The results underscore the strong negative borhood. Another factor that may be important is association between Roma ethnicity and welfare, whether the individual or the interviewer identi- 31 Roma in an Expanding Europe: Breaking the Poverty Cycle fies itself as Roma. Households which self-identi- Many socialist initiatives to integrate Roma fy as Roma are likely to be from less integrated provided housing along with employment. Cur- and more traditional Roma communities--and rent Roma neighborhoods in some areas have hence may be poorer than other Roma.8 their roots in these settlements, although it is The Roma-only analysis does yield some dif- unclear how many (Macura and Petrovic 1999). ferent results, suggesting that using the same Findings from a government housing survey in model for Roma and non-Roma samples may be Hungary indicate that 60,000 Roma--approxi- inappropriate. Most strikingly, adult equivalent mately 13 percent of Roma in Hungary--live in expenditures are lower for Roma households liv- settlement-type environments isolated from the ing in Roma-only settlements than for those liv- majority population (Puporka and Zádori 1999). ing in other locations, suggesting a connection This was confirmed in another 1994 survey, between living in a geographically segregated which found that 14 percent of Roma lived in set- area and welfare. Additional analysis found that tlements (Kémeny, et al. 1994). This spatial segre- much of the difference between the welfare of gation results from such reasons as the historical Roma and non-Roma is due to differences in location of Roma neighborhoods, municipal plan- opportunities and characteristics--such as educa- ning, and housing preferences (Box 2.2). Some tion levels and employment status. But an impor- Roma communities have chosen to live separate- tant component is structural, reflecting differ- ly; others who had hoped to move hit barriers of ences in the communities. This likely reflects discrimination. discrimination, exclusion, and cultural factors. In the countries of South East Europe which While the quantitative data cannot provide more were formerly part of the Ottoman Empire, insights into these issues, the qualitative analysis Roma neighborhoods--called mahalas--are presented in the following chapters examines common in cities. Towns under the Ottomans these unmeasurable dimensions of exclusion were organized into administrative units based more closely. on the ethnicity and religion of the inhabitants. While these divisions--themselves known as HOUSING mahalas, giving rise to the name--have largely The multivariate analysis highlighted a link disappeared, Roma settlements based them still between geographic location and Roma poverty. exist. In the countries of the former Yugoslavia, This is closely related to housing conditions. Roma mahalas range from several hundred to Because of the diversity of Roma communities several thousand inhabitants; in Bulgaria, some and contrasting conditions across countries, it is are as large as 15­20,000. In some cases Roma difficult to draw conclusions about the character- mahalas were originally built on the outskirts of istics of Roma settlements and housing. Many towns, but as urbanization has proceeded and issues in housing are similar to those faced by the towns have grown, these settlements may non-Roma populations, particularly for commu- now lie close to the center of some cities. nities and households that have integrated into Another common type of settlement rooted in non-Roma areas. But Roma confront unique the socialist era are neighborhoods near state- problems. The housing policies of successive owned enterprises, often in one-company towns. empires, socialist regimes, and recent govern- As part of their integration or assimilation cam- ments have often led to regional and geographic paigns, socialist governments provided housing isolation and segregation of Roma neighbor- for Roma along with employment. Rents were hoods. This has, in turn, limited access to public either free to employees or heavily subsidized. In services and raised questions about land and the transition period, as many state enterprises property ownership. Compounded by discrimi- have been closed or restructured and collective nation from some surrounding communities and farms have been broken up, the inhabitants lost municipal governments, conditions in many their jobs. Many of these areas have become Roma settlements have deteriorated significantly. impoverished. 32 Roma Poverty and Welfare: An Overview Box 2.2: Spatial Segregation within Roma Settlements There are common patterns to the internal geography of some urban mahala neighborhoods. The Niko- la Kochev district in Sliven, Bulgaria provides a typical example. Approximately 4­6,000 Bulgarian Roma live in Nikola Kochev, a settlement traced to the fifteenth century. Most of the inhabitants are textile workers, descendants of some of the first workers in a textile industry that dates back to the mid-1800s. The organization of the district reflects class distinctions within Roma society. The best-off members of the Roma community live in direct contact with Bulgarians on the periphery of the settlement, a large share of the adults are employed, and most of their children attend school regularly and continue on through sec- ondary school. There are a large share of elderly inhabitants in this part of the settlement, as many of the young people have moved to apartments in more ethnically mixed parts of town. Poverty increases further into the settlement, in an area nicknamed "the Jungle." The inhabitants here are poorer, less educated, and less integrated. Most are unemployed. Conditions in the Jungle are extremely bleak, with houses often constructed from scavenged materials and lacking water and electricity. Source: Tomova 2000. Roma in cities are highly segregated. tion. In addition, many Roma communities have Research in Hungary traced the growth of these tapped into public services illegally, channeling areas to the migration of Roma from the country- water or electricity into their settlements. side during the economic crisis at the end of the Housing options for Roma have also been 1980s. Faced with growing unemployment, many limited by discrimination by municipal officials Roma moved to Budapest in search of better and landlords. In some cases, local governments opportunities. Over time, due to declining living have attempted to reduce illegal tenancy by mov- conditions and poor access to municipal services, ing settlements to the outskirts of towns.9 In other conditions in these neighborhoods severely dete- cases, municipal officials have overtly banned riorated. Common side effects of slums appeared, Roma--as was the case in 1997, when two Slovak including drug addiction and rising crime villages prohibited Roma from entering and set- (Ladányi 1993). The further deterioration of liv- tling. These bans were challenged in the Euro- ing conditions and employment opportunities pean Court of Human Rights and lifted. Other has likely led to continued rural­urban migration. municipal governments have reportedly bought The transition process has created problems land and apartments to ensure that Roma will not with the legal status of housing for Roma, in part be able to settle in them (OSCE 2000). because property rights were often not clearly Questions about the legality of property own- defined under communism. Some Roma were ership have arisen with land as well. The post- evicted from state-owned apartments when hous- communist process of land restitution has had a ing subsidies were withdrawn, properties priva- varied impact on Roma. Because Roma were not tized, or returned to prior owners. Many Roma traditionally landowners, few were eligible to file now find themselves living illegally in dwellings, claims. In some cases, Roma who worked on col- either because they had no choice but to squat or lective farms were entitled to receive land after because the property rights on their building the cooperatives dissolved. were transferred following the transition (OSCE 2000). In other cases, poor Roma have intention- Housing Conditions ally become squatters. These developments have Roma neighborhoods are frequently extreme- seriously limited access to social services, as resi- ly overcrowded and destitute. Some Roma slums dency and ID papers are frequently required for have evocative nicknames; for example, "Abys- social assistance benefits, health care, and educa- sinia" and "Cambodia," are extremely impover- 33 Roma in an Expanding Europe: Breaking the Poverty Cycle Figure 2.1: Household Size in Bulgaria, Lack of water, gas, electricity, and public serv- Hungary, and Romania, 2000 ices such as waste collection bedevils many Roma neighborhoods. According to the Yale survey 6 data, Roma are less likely to have access to water and sewage than other groups. Access to utili- 5 ties--including electricity, heating, and water--is 4 persons significantly lower for Roma households (Table Non-Roma of 3 2.4). Only 9 percent of Roma houses in Bulgaria, Roma 2 and 10 percent of them in Romania had hot water. number Access to bathroom facilities and indoor toilets is 1 similarly low. Few Roma households have tele- 0 phones: only 12 percent in Bulgaria, 41 percent in B ulgaria H ungary R omania Hungary, and 26 percent in Romania (in contrast with between 58 and 81 percent for non-Roma Sources: Yale Dataset; Revenga et al. 2002. households). Over half of Roma households in Bulgaria reported wet walls and leaky roofs, sig- nificantly more than in the other countries.11 ished areas within Roma ghettos in Bulgaria. The Cultural preferences of Roma communities also household survey data show that Roma living affect conditions within Roma settlements, quarters are smaller than others, have larger although it is difficult to generalize here. Non- households and are consequently more crowded Roma across countries sometimes complain that (Figure 2.1). According to the Yale dataset, Roma Roma do not take care of their surroundings, and households are nearly twice the size of non- that they destroy property and public spaces. Some Roma. In Romania, based upon a 1998 household of these perceptions may stem from cultural differ- survey, Roma dwellings were, on average, 20 ences. For example, some Roma groups reportedly percent smaller than those for Romanians, have taboos against adjoining kitchen and toilet although their household size was significantly facilities. As public housing initiatives did not larger.10 incorporate the views and culture of Roma into Table 2.4: Housing Characteristics by Ethnicity, 2000 (% of households) Bulgaria Hungary Romania Non-Roma Roma Non-Roma Roma Non-Roma Roma Households with: Electricity 99.6 94.5 99.0 98.1 99.1 94.5 Central or gas heating 16.1 4.1 78.6 35.3 51.2 25.6 Cold running water 96.8 67.6 92.0 65.3 67.4 41.4 Hot running water 39.1 9.4 83.2 45.1 35.3 10.7 Sewer or cesspool 90.3 52.3 58.3 33.4 53.6 30.0 Telephone 80.6 12.1 76.0 41.4 58.2 26.4 Bathroom/shower 82.5 23.5 88.8 50.2 54.3 18.9 Indoor toilet 65.2 15.0 86.4 49.9 52.6 18.3 Wet walls 20.6 50.4 16.6 40.1 21.0 44.9 Leaky roofs 19.2 54.2 9.6 33.0 14.8 40.2 Earthen floor used for sleeping 7.4 36.7 5.8 13.2 19.3 39.0 Sources: Yale Dataset; Revenga et al. 2002. 34 Roma Poverty and Welfare: An Overview their design, inhabitants have had little interest in considered illegal in some countries and sanc- the maintenance and upkeep of the buildings. tions could be imposed for part-time work, self- Above all, poverty makes it difficult for households employment or not working. For example, the to maintain their housing conditions. In effect, right and obligation to work was enshrined in the some of these complaints reflect a dual prejudice: Czechoslovak Constitution (Ministry of Labor exclusion leads to Roma poverty, and then fellow 1997). In 1970, the Romanian government citizens castigate Roma for living in squalid condi- decreed that "social parasitism" and other tions--as if anyone chooses to be poor. "deviant behaviors" were punishable with prison and forced labor (Rughinis 2000). LABOR MARKET STATUS In this context, employment of Roma was Perhaps the most dramatic changes for Roma actively promoted through recruitment and following the transition from socialism took place assimilation campaigns. Along with the rest of in the labor market. In Central and Eastern the population, Roma were brought to work in Europe, employment levels fell significantly dur- the process of industrialization and collectiviza- ing the early years of the transition, as restructur- tion of agriculture. Because of their low education ing began and subsidies for large state-owned levels, Roma were most frequently employed in enterprises were slashed.12 Because of their low low-skilled manufacturing industries. During the skill levels, as well as discrimination in the labor socialist period, employment rates for Roma in market, Roma were frequently among the first to some countries did not differ greatly from those be laid off. This has directly influenced Roma of non-Roma (Box 2.3). In the Slovak Republic in welfare. Roma have limited opportunities to the 1980s, 70 percent of working-age Roma were reenter the workforce, so unemployment rates-- employed (Ministry of Labor 1997).13 A survey of and particularly long-term unemployment--are Roma in Hungary in 1971 found that employ- often exceptionally high. Reports of unemploy- ment levels of working-age Roma men were ment rates of up to 100 percent in Roma settle- slightly higher than those of non-Roma, with ments are not uncommon. employment rates of 88 and 85 percent respec- Roma have historically had connections to tively (Kertesi 1994). traditional occupations. Indeed, many of the names of Roma subgroups derive from associa- Developments in Transition tions with particular crafts dating back to the Large-scale restructuring in the early years of Middle Ages. But few of these connections still the transition period had an immediate impact on exist. Roma were traditionally not landowners the labor market status of Roma. By 1993, and had scant involvement in agriculture. In the employment levels of Roma in Hungary had fall- early twentieth century, many of the traditional en to 26 percent of the labor force and 63 percent occupations declined with industrialization. for the population at large (Kémeny, et al. 1994). Crafts such as metal and woodworking faced These trends have worsened during the transition competition from manufactured goods and Roma period, as Roma found it difficult to reenter the began to shift into other areas of economic labor force, and the gap in unemployment activity. between Roma and non-Roma widened. In the With socialism, Roma were compelled to Czech Republic, government estimates for 1999 move from self-employment and informal sector suggested that 70 percent of the Roma were activity into full-time public sector jobs. Full unemployed, in contrast with 10 percent of the employment and job security were defining char- total population (OSCE 2000). acteristics of the socialist regimes. Employment Because Roma were among the first laid off in was encouraged through guaranteed jobs, low the early 1990s, the duration of their unemploy- wages, and a wide range of associated benefits ment is exceptionally high (Figure 2.2). The gap is and services, including housing subsidies, child particularly bad in Bulgaria, where the duration care, and health services. Unemployment was of unemployment lasted 27 months on average 35 Roma in an Expanding Europe: Breaking the Poverty Cycle Box 2.3: Measuring Unemployment Reports of exceptionally high unemployment rates for Roma settlements--between 70-100 percent--are common but difficult to fathom, particularly in countries with active informal sectors. In these cases, it is important to note how unemployment is measured and defined. In general, there are two main instruments for measuring unemployment. First are registration statistics based on the administrative records of the labor offices. But registration data capture only those individuals who report to labor offices and do not reflect any kind of informal labor market activity. These data may sig- nificantly underestimate the long-term unemployed. Many countries limit the duration of their unemploy- ment benefits, and once these have expired, people have no incentive to report to the labor offices. Registra- tion data also generally do not capture ethnicity. The second important source are labor force and household surveys. These surveys ask about economic activity in general and can reflect both informal and formal employment. However, as was discussed in the first chapter of this report, survey data are limited in their ability to differentiate by ethnicity. Data included in this report are mainly from household surveys and other targeted surveys of the Roma population. Unless otherwise indicated, employment includes the share of the working-age population (defined differently depending on the country and source), that has worked for in-cash or in-kind payment during a set period (either the previous week or month). In this case, informal employment is included. In contrast, unemployment refers to the share of the working-age population that has not worked for payment. but soared to 51 months for Roma. Long-term of unemployment for Roma and non-Roma is not unemployment has been consistently high in Bul- significant in Romania. garia during the transition period, indicating the High rates of unemployment among Roma persistence of a stagnant pool of long-term unem- only tell part of the labor market story. Informal ployed who are unable to reenter the labor mar- sector activity is also an important source of ket. Among them is a sizeable fraction of Roma. income. The types of activities vary widely, from On the other hand, the difference in the duration lucrative trade and work in neighboring coun- tries, to more marginal subsistence occupations ranging from seasonal farming to gathering herbs and recycling used materials. Some Roma Figure 2.2: Duration of Unemployment, 2000 may prefer more flexible and entrepreneurial (months) informal sector activities and self-employment to wage labor. Popular stereotypes characterize Roma as 60.0 lazy. However, survey data indicate that Roma 50.0 actively seek employment. In Bulgaria in 1997, 46 40.0 percent of unemployed Roma reported that they Non-Roma 30.0 were looking for a job, compared to 19 percent of months Roma 20.0 the total unemployed population. In Romania, 35 10.0 percent of unemployed Roma had looked for employment during the previous week, in com- 0.0 parison with 15 percent of the total population. B ulgaria H ungary R omania Similar results were found for Hungary (Kertesi 1994). However, more information on Roma val- Sources: Yale dataset; Revenga et al. 2002. ues and attitudes toward work is required to 36 Roma Poverty and Welfare: An Overview understand these data fully. For example, as Despite these achievements, the educational Roma are more frequently engaged in short-term attainment of Roma lagged significantly behind informal sector activities and may have more the non-Roma population, with Roma much less than one job, they may inevitably spend more likely to continue on to secondary and post-sec- time looking for work. ondary education. The Yale dataset also illustrates lower educa- EDUCATION STATUS tional attainment among Roma. Most Roma have The education status of Roma has historical- primary education or below, while most non- ly been low across Europe. While significant Roma in the three countries have some secondary, gains were made in enrolling Roma children in post-secondary or university education (Table school during the socialist era, the gap in the 2.5). Bulgaria provides the most dramatic exam- educational attainment of Roma and the rest of ple; 89 percent of Roma had primary education or the population was not bridged in any of the less, while only 10 percent had some secondary countries for which data are available. The evi- education. In contrast, 33 percent of non-Roma dence suggests that access has eroded during the had primary education or below, while 54 percent transition period, and Roma children of basic of the population continued on to secondary school age are increasingly not starting or finish- school and 14 percent to tertiary. Results are sim- ing school. These trends are consistent with ilar for Hungary and Romania. Less than 1 per- national developments in enrollments, although cent of Roma in all countries continued past sec- data suggest that the decline in access among ondary school. Roma has been deeper than for the rest of the It is not surprising that education levels vary population. notably within countries, between urban and Gaps in access to education among the Roma rural areas, and across different types of Roma are not new. Not until the socialist regimes came communities. In Hungary, for example, the 1993 to power in Central and Eastern Europe follow- survey mentioned above found that the share of ing World War II were large numbers of Roma Roma who had not completed primary education compelled to participate in public education. was 16 percent in Budapest, 24 percent in towns Education was a key element of socialist assimi- and 27 percent in villages, reflecting different lation campaigns. It was viewed as an instrument constraints to access (Puporka and Zádori 1999). of political and economic socialization which Differences between types of Roma are also would facilitate the integration of the Roma into important. For example, the same survey found the full employment society. Despite the achieve- that the share of Roma with less than basic edu- ments in reducing literacy and increasing school cation was 23 percent for the Romungro Roma participation, the efforts undertaken during the (whose native language is Hungarian), 42 percent socialist era laid the foundation for inequities in for the Bayash (native language Romanian), and education quality, as many Roma were channeled 48 percent for the Wallach Roma (native language into separate or segregated schools outside the is Roma) (Puporka and Zádori 1999). Similar mainstream system. finding were noted in Bulgaria. Education in the Transition Period Enrollments and Attendance Gaps in education persist in the transition Disparities in enrollments between Roma and period and are most evident in analysis of the non-Roma suggest that the gaps in educational educational levels of the population. Comparable attainment will persist into the next generation. surveys conducted in Hungary in 1971 and 1993 In Bulgaria and Romania, the Yale data show a illustrate these trends. In 1971, about 26 percent significant difference in enrollment levels for chil- of Hungarian Roma aged 20­29 had finished 8 dren of basic school age. In Bulgaria, enrollment years of primary school. This had increased to rates for Roma were 33 percent lower, while in over 77 percent by 1993 (Kémeny et al. 1994). Romania, the difference is 20 percent (Figure 2.3). 37 Roma in an Expanding Europe: Breaking the Poverty Cycle Table 2.5: Educational Attainment by Ethnicity, 2000 (age 18+) Bulgaria Hungary Romania Non-Roma Roma Non-Roma Roma Non-Roma Roma Primary or Below 32.7 89.6 35.0 76.4 33.1 66.5 No Education 1.3 15.0 0.3 4.3 1.9 13.4 Incomplete Primary 9.4 39.6 10.7 22.1 15.0 27.0 Complete Primary 22.0 35.0 24.0 49.9 16.2 25.2 Some secondary 53.8 9.6 53.0 23.4 56.3 32.4 Completed Primary and Apprenticeship 2.2 1.8 25.5 19.0 18.9 13.1 Incomplete General Secondary 2.3 1.6 6.1 1.5 6.6 9.0 Completed General Secondary 19.4 3.0 17.5 2.7 23.9 8.9 Secondary and Vocational 29.8 3.2 3.8 0.2 6.9 1.3 Higher education (Complete and Incomplete) 13.5 0.5 12.0 0.2 10.4 0.3 Incl. post-secondary and university No Answer 0 0.4 0.1 0.1 0.1 0.8 Sources: Yale Dataset; Revenga et al. 2002. In Hungary, the gap in enrollments was not sig- attended preschool. This dropped by 60 percent nificant, at less than 2 percentage points. in the 1991 school year, and by 1997 less than 20 Enrollment rates tell only part of the story. In percent of Roma children were thought to attend some cases, students may enroll at the beginning (Slovak Ministry of Labor 1997). In Hungary, of the year but not actually attend school. Quali- where preschool is compulsory for all children at tative studies show that this often happens in age five, 11 percent of Roma did not attend poor Roma communities where the costs of edu- school in 1997 (Radó 1997). This is a serious cation for families are high (Box 2.4). It is also development; children who do not start pre- important to note that enrollment rates calculated school are less likely to attend primary school from the Yale survey data indicate only whether children are enrolled in school--and not whether they are enrolled in the appropriate level. In con- Figure 2.3: Enrollments in Education, 2000 trast with conventional enrollment rates, the rates (% of children aged 6-14) presented above indicate whether children between the ages of 6 and 14 were enrolled at all, 100 which may be misleading if many children are repeating grades. 80 Preprimary attendance may have been most 60 damaged during the transition period. In gener- Non-Roma al, preschool and kindergarten enrollment rates enrolled) 40 Roma have fallen across the region, as subsidies for (% 20 schools connected to state enterprises were with- drawn and fees were introduced (UNICEF 1997). 0 Growing costs have discouraged parents from Bulgaria Hungary Romania sending children to school. In the Slovak Repub- lic in 1990, 80 percent of Roma children aged 3­6 Sources: Yale dataset; Revenga et al. 2002. 38 Roma Poverty and Welfare: An Overview Box 2.4: School Drop-Outs in Bulgaria: The Case of the Missing Children National administrative data paint a rosy picture of access to education in Bulgaria. Gross enrollment rates are nearly universal, and very few children are identified as being out of school. But a qualitative sur- vey found that the reality is much more grim. In fact many children fall through the cracks, never attend school, and do not show up in the official administrative data. These children are frequently those from the poorest households. In the Nadezhda district, a Roma neighborhood in Sliven, a town in Eastern Bulgaria, the researchers found 273 children who had never been to school. Why is this the case? The study identified several reasons: · There are no records of children from households which lack residence requirements--a serious issue for poor households, particularly Roma families who live in unregistered settlements, or in properties with illegal status. · Monitoring of children has weakened. Children are no longer required to enroll in the school in the district in which they live. There is no coordination between district schools to ensure that all children are enrolled and no system to monitor whether children who have left one school enroll in another. · There are no mechanisms for following up on children who have been expelled--to find out what happens to them and whether they reenroll in school. Similarly, there is no follow up for children who leave school voluntarily and are not officially considered drop-outs. · School and local officials face incentives not to report drop-outs to maintain class sizes to avoid school closures. Source: Kabachieva and Iliev 2002. and may have more difficulty remaining in aspects of education systems in the region which school. For Roma children, these issues are com- influence quality including the prevalence of pounded by the fact that many do not speak the "special schools," the segregation of Roma stu- national language at home and thus begin pri- dents within the mainstream system, and inade- mary school at a disadvantage. quate teacher training and curriculum (Box 2.5). As illustrated in the breakdown of the educa- One of the most damaging legacies of the tional status of the population, the gulf between socialist era is the tendency to channel children education levels is wider for Roma than non- into special schools for the mentally and physi- Roma, indicating the challenges of moving from cally handicapped. This policy had its roots in the one level of education to the other. Limited evi- socialist legacy of "defectology" which assumed dence suggests that dropout rates have risen dur- that differences among students were due to dis- ing the transition period--disproportionately so ability rather than environmental conditions and, for Roma children (UNICEF 1998). Informal esti- as a result, should be addressed as medical prob- mates for Bulgaria suggest that 45,000 students lems in institutions separated from the rest of drop out of school each year, most of them Roma. society (Ainscow and Memmenasha 1998). The legacy has been the persistence of a parallel sys- Education Quality tem of schools which provide lower quality edu- Access to education is also directly affected by cation and fewer opportunities in post-basic edu- the quality of schooling, as students may be cation and the labor market than mainstream deterred from attending school if the quality is schools. low. Uneven quality of education also affects Evidence of this practice is widespread. Data equity. There is evidence that the quality of edu- for the Czech Republic are striking. Estimates for cation for Roma students is lower than for the rest 1997 indicate that 64 percent of Roma children in of the population. The following discusses primary school are in special schools, in compar- 39 Roma in an Expanding Europe: Breaking the Poverty Cycle Box 2.5: Entrance to Remedial Special Schools in the Czech Republic Roma children end up in special schools for many reasons. A study in the Czech Republic found that because of discrimination and the highly discretionary nature of the entrance process, many more Roma chil- dren end up in special schools than the regulations should allow. Children can be enrolled directly into special schools, or transferred from a mainstream basic school. By law, placement is based upon the recommendation of the school director in consultation with the parent and an educational psychologist. In some cases parental consent is not obtained, or is abused. Parents may not realize that they are authorizing their children to be shifted into a special school: "My daughter is in the second year of basic school. She is doing alright. One day in November 1997 her teacher came to see me saying, "We want to move her to another class which will be better for her." He gave me a piece of paper to sign. I should have read it but it was long and I didn't think a teacher would try to cheat us, so I just signed it...The next day I got a letter saying that my daughter had been moved to a reme- dial special school." Roma parent, Prague. Educational psychologists play a pivotal role in determining whether children will be sent to special schools. They recommend students for examination and administer the exams. These procedures were found to be highly discretionary. In some cases children were even transferred without the required psychological exam. The tests themselves are problematic, psychologists may use a number of different instruments, many of which are culturally biased. Because of the widespread abuses that have been documented, parents of 18 Roma children from the Czech town of Ostrava initiated legal proceedings against the government last year. The Czech Constitutional Court ruled in favor of the government. An appeals process opened in April 2000 in front of the European Court of Human Rights in Strasbourg. Source: ERRC 1999. ison with 4 percent for the total population. In are only allowed to enter technical secondary other words, Roma are fifteen times more likely schools, which offer limited training in narrowly to end up in special schools than the national defined fields. Students are then dually chal- average (ERRC 1999). Similarly, in Hungary lenged on the labor market, as employers look about half the number of students enrolled in unfavorably upon graduates of special schools, special schools are Roma (Radó 1997; 2001). and technical training fails to adequately prepare Regardless of the quality of teaching in spe- young people for the labor market. cial schools, students enrolled in these institu- There is growing recognition that the exis- tions are at a disadvantage. The curriculum is less tence of special schools adversely affects the inte- rigorous and expectations are lower. A detailed gration and educational development of Roma report on the Czech schools notes that students in children. However, the obstacles to change are special schools receive fewer Czech language les- notable. Not only does resistance to integration sons per week, and are not expected to read for come from non-Roma parents and education offi- comprehension until the fourth grade--while the cials who fear that increasing the share of Roma expectation is first grade for students in main- children in a classroom will lower the quality of stream schools (ERRC 1999). education for non-Roma students but opposition Opportunities for graduates of special schools comes from Roma parents as well. Special schools are also limited. Even if children are able to over- can be attractive to poor Roma families for eco- come low expectations, they are not allowed nomic reasons, in that school meals and--for res- equal access to school-leaving exams. In the idential institutions, housing--are provided. Spe- Czech Republic, graduates from special schools cial schools are also viewed by some parents as 40 Roma Poverty and Welfare: An Overview safe havens--free from discrimination that is dren out of the dining room and insult them" more pervasive in mainstream schools. (ERRC 1999). Even where Roma children are kept within the mainstream school system, they are often sep- HEALTH STATUS arated into separate classes, or schools. This is fre- Data on the health status of Roma is scarce quently related to geography if Roma families and fragmented. However, the information that live together in a neighborhood. However, there does exist paints a bleak picture, pointing to sig- is also evidence of further separation of Roma. In nificant gaps in health status between the Roma Bulgaria "Roma schools" are schools in which the and non-Roma populations. Because of the share of Roma is over 50 percent. The overrepre- absence of data, it is difficult to discuss trends in sentation of Roma in these schools is due to geo- health during the transition period. On aggre- graphic concentration, and attempts by some gate, Roma are estimated to live around ten years municipal and education officials to place Roma less than the majority populations in Central and students together into separate schools. Eastern Europe (Braham 1993). Because of sub- A recent survey conducted by the Open Soci- standard living conditions, Roma communities ety Institute in Sofia found more than 60 elemen- are particularly susceptible to communicable dis- tary, 350 primary and 9 secondary schools in the eases, including hepatitis and tuberculosis. Very country in which Roma comprise between 50 and little is know about the incidence of non-commu- 100 percent of the student body. In general, qual- nicable diseases among Roma. There are increas- ity and conditions in Roma schools are poorer ing indications that Roma have a higher inci- than in mainstream schools, infrastructure has dence of health problems associated with deteriorated and materials are lacking (Denkov, unhealthy life styles, including drug and alcohol et al. 2001). There are also serious problems with addiction, and HIV/AIDS. attendance in Roma schools. Teachers from Data on life expectancy and mortality for Haskovo noted some Roma students had not Roma indicate significantly worse health condi- attended class for an entire year. Similarly, field- tions than for the rest of the population. Estimates work in Romania found situations in which non- derived from the Czechoslovak census data for Roma parents would request that their children the 1990s found that life expectancy for the total be taught in classes without Roma students, and population was 67 years for men and 74 for teachers would divide up classes to keep Roma women, while for the Roma the figures were 55 separate (World Bank 2000d). and 60 (ECOHOST 2000). In Hungary the life Discrimination against Roma by non-Roma expectancy gap is estimated at 10 to 15 years. A parents, children, and teachers contributes to low study conducted in Pest County documented that attendance and can both discourage children Roma men lived 13 years less and women 12 from attending school and affect the quality of years less than non-Roma inhabitants. Estimates education in the classroom. Stereotypes about of infant mortality rates show a similar gulf. In Roma and their attitudes toward education lower the Czech and Slovak Republics infant mortality teachers' expectations about the potential of their for Roma was double that of non-Roma. Howev- students. Discrimination can be both explicit--as er, in Hungary, infant mortality for Roma has in the case of schools creating separate classes-- declined faster than that of the total population, or more subtle if parents discourage their chil- and the gap between Roma and non-Roma has dren from interacting with Roma classmates. A narrowed. While infant mortality was 38 per study of the Czech system documented cases in thousand births for the total population and near- which Roma children had been abused by educa- ly 118 for Roma in 1970, this decreased to 17 for tion staff. One parent from Prague noted that the whole population and 21 for Roma by 1990 "The teachers who teach Gypsy children are fine, (Puporka and Zádori 1999). but the others are terrible. They chase our chil- 41 Roma in an Expanding Europe: Breaking the Poverty Cycle Demographic Trends Figure 2.4: Age Structure of Roma and the Roma have historically had significantly Total Population in the Czech Republic, 1991 higher population growth than other groups. This has been--and continues to be--a sensitive age Male Female political issue because across the region the size 80+ of the Roma population is growing much faster 70 than the non-Roma population. In 1958, the 60 Czechoslovak government issued a decree stating 50 that Roma were not a distinct ethnic group, but Total rather were a people "maintaining a markedly 40 Roma different demographic structure" (Fraser 1995). 30 Roma women marry at a younger age and begin 20 having children earlier than other groups. This 10 has serious consequences for women's reproduc- tive health. The precise roots of high fertility among Roma are unknown, but likely result from 150 100 50 0 50 100 150 per 1000 socioeconomic factors, including poverty, low education levels, and cultural preferences. Source: Census data, from ECOHOST 2000. Because of higher birth rates, the Roma com- munity is significantly younger than other groups. Data from two representative surveys of health. Due to inadequate access to care, unhealthy Roma conducted in Hungary illustrate this phe- lifestyles--including poor living conditions and nomenon (Puporka and Zádori 1999).14 In 1993, nutrition--and high birth and abortion rates, 39 percent of the Roma population was under 14 Roma women are at a higher risk of complications years old, while only 19 percent of the total pop- during pregnancy than non-Roma women. A ulation fell into this age group. In contrast, 19 per- study conducted in Szablocs-Szatmar County in cent of the total population was over 60, while Hungary in the 1980s found that Roma women only 5 percent of Roma fell into this category. were twice as likely to have difficulties during Birth rates among Roma are much higher than pregnancy, as well as premature births and low those of other groups. Age pyramids from the birth weight babies, than non-Roma women 1991 Czechoslovak census illustrate a similar (Puporka and Zádori 1999). Similarly, a study con- phenomenon (Figure 2.4). ducted in a district in the Slovak Republic in 1995­ Evidence on demographic trends for Roma 1997 found low birth weights for Roma to be more during the transition period is mixed. While over- than double that of non-Roma (ECOHOST 2000). all fertility has declined significantly in Central Women's health is a serious issue. Because of and Eastern Europe, it is not clear whether this low levels of awareness about health issues and also holds true for Roma. Fertility has dropped in impoverishment among many communities, some Hungarian Roma communities (Puporka Roma women face other health challenges--also and Zádori 1999), while a study in Bulgaria found common to the general population--including that birth rates were increasing among the poorer inadequate nutrition and high levels of smoking subgroups of Roma (Tomova 2000). Regardless of during pregnancy (OSCE 2000). The Hungarian these contrasting messages, the available data survey mentioned above found that 63 percent of suggest that Roma families remain larger than pregnant women were smokers (Puporka and those of other ethnic groups. Zádori 1999). Awareness about contraception varies across Reproductive Health Roma communities. The qualitative study for High infant mortality and perinatal death rates Romania found that better off Roma women were for Roma are linked to women's reproductive more likely to use contraception (Rughinis 2000). 42 Roma Poverty and Welfare: An Overview Another study, also in Romania, by Médicins Sans Tuberculosis is on the rise throughout the Frontičres indicated that many Roma women pre- region. Hungary recorded a 20 percent increase ferred IUDs because they gave them more inde- between 1990 and 1995. Tuberculosis risk is asso- pendence (OSCE 2000). As is the case throughout ciated with poor living conditions, putting some the former socialist countries, abortion is much Roma communities at higher risk. In the 1960s, a more common than in the West, and is used as study in the western part of the Slovak Republic contraception. In 1997 abortion rates ranged from found that the prevalence of tuberculosis among 63 abortions per 100 live births in the Czech Roma was higher than for the majority population Republic, to 135 in Bulgaria and 147 in Romania (ECOHOST 2000). However, there are no indica- (UNICEF 1999). Small-scale studies in the Slovak tions currently that incidence is higher among Republic and Bulgaria suggest that abortion rates Roma. Reports from physicians working in one of are higher for Roma than non-Roma women the main tuberculosis hospitals in Hungary found (ECOHOST 2000; Tomova 1998). The issue of that Roma women were more susceptible to forced sterilization of Roma women has gained tuberculosis than men (Puporka and Zádori 1999). international attention. A recent study document- Another worrying trend has been outbreaks ed cases of women who had been sterilized of measles among Roma in the Slovak Republic against their will by doctors in Eastern Slovakia and Hungary, which may have been due to laps- (Center for Reproductive Rights 2003), The gov- es in immunization coverage. Aggregate immu- ernment is currently investigating this issue. nization rates throughout the region are high, reaching nearly full coverage. However, gaps in Communicable Diseases immunizations in Roma communities have been Poor living conditions, such as overcrowding documented. In a study of Roma in Bulgaria, 11 and lack of adequate sanitation facilities make percent of households reported that their children Roma communities more susceptible to infectious had not been vaccinated, with the rate nearly 20 diseases than other groups. Reports of epidemics percent in the poorest sites (Tomova 2000). of hepatitis, tuberculosis, and parasitic diseases were common, during and after the socialist peri- Congenital Disorders od. Skin diseases, such as eczema are also com- Research on congenital disorders among mon. The last reported cases of poliomyelitis in Roma is sparse and frequently problematic. A Bulgaria, Romania, and FYR Macedonia all were review of literature on health among Roma in the in Roma communities (OSCE 2000). In Bulgaria in Czech and the Slovak Republics noted that some 1992, 90 Roma children in the regions of Sliven research was tainted by concepts of contagion and Sotirya caught poliomyelitis. There were no and "social Darwinist" motivations. The studies cases among Bulgarians. In 1993 a diphtheria out- focused on identifying race-based inferiorities break occurred in the same areas (Tomova 2000). among the Roma (ECOHOST 2000). Neverthe- In the 1990s a number of hepatitis outbreaks less, the prevalence of genetic diseases among have been documented in Roma settlements in Roma is a valid concern, particularly since some the Czech Republic and Hungary. In 1990, an out- groups of Roma have remained relatively isolated break was recorded in Brno, in the Czech Repub- from the majority populations and a high degree lic and 1999 in central Moravia 40 children in a of intermarriage has been documented in some Roma settlement were hospitalized with the dis- communities. The extent of this is not known.15 ease (ECOHOST 2000). Hepatitis B, a more dan- gerous form of hepatitis, has been found to have Non-Communicable Diseases an even higher incidence in Hungary among Very little information on non-communicable Roma. Among pregnant women routinely diseases among Roma is available. Across Central screened for hepatitis B in Hungary, approxi- and Eastern Europe, mortality from non-commu- mately half tested positive and the majority were nicable diseases is high--particularly conditions Roma (Puporka and Zádori 1999). associated with unhealthy lifestyles such as poor 43 Roma in an Expanding Europe: Breaking the Poverty Cycle nutrition, smoking, and alcoholism (Staines tional intake of Roma was worse than that of 1999). Some Roma communities may be particu- Czech children. Roma had inadequate consump- larly susceptible to these conditions because of tion of vegetables, dairy products, grains, and lifestyles. Prevalence of smoking, alcoholism, and meats. On the other hand, Roma children were poor diets are reported to be higher among some found to consume four-and-a-half times the rec- Roma communities. Another study in Hungary ommended daily allowance of snack foods con- found that smoking was exceptionally high taining fat and sugar (ECOHOST 2000). Improp- among Roma, and particularly among Roma er nutrition for children can adversely effect women. A survey of students in a Roma school in growth and future development. Some evidence Hungary found that 85 percent of students of stunting among Roma has already been docu- between 15 and 22 had tried cigarettes, and 45 mented. A study of the growth of children in the percent smoked an average of a package of ciga- eastern part of the Slovak Republic found that rettes per day (Puporka and Zádori 1999). Roma children developed more slowly than Slo- Although little information is available, occu- vak children of the same age (ECOHOST 2000). pational injuries and environmental conditions are also likely sources of ill health among Roma. Sexually Transmitted Diseases (STDs) As the lowest skilled jobs were also most likely to and Drug Abuse be the most hazardous, many Roma were There is very little information on the preva- employed in dangerous professions during the lence of STDs and drug abuse among Roma com- socialist period, including mining and other munities. Prostitution and trafficking of women aspects of heavy industry, such as working with with western Europe has been on the rise during toxic substances. The incidence of disability from the transition in Central and Eastern Europe workplace injury is thought to be disproportion- (UNICEF 1999). Women have resorted to employ- ately high among Roma. Similarly, exposure to ment in the sex industry as result of the dearth of hazardous materials and highly polluted regions employment opportunities elsewhere. Estimates is also an issue for Roma, as many live or work in reported for the Czech Republic suggested that areas of dumpsites, mines and abandoned facto- out of the nearly 40,000 prostitutes in the country, ries. A report described conditions in the eastern some 25,000 are Roma women (ECOHOST 2000). Slovak town of Rudnan, where 500 Roma are Prostitution increases the risk of STDs, including living in an abandoned iron and mercury mine. HIV/AIDS for the Roma community at large. The area is known to be highly contaminated However, to date there is no information on dis- (Erlanger 2000). Many Roma engage in recycling ease incidence. A study of Roma in Miskolc, Hun- activities, including trading in scrap materials. In gary found that Roma were uninformed about a highly publicized case in Hungary, Roma sup- the risk of AIDS and the options for prevention ported themselves by melting down batteries. (Puporka and Zádori 1999). This created serious pollution which was blamed Drug abuse is on the rise among some Roma for a child's death from lead poisoning (Puporka groups. Information is scarce because--although and Zádori 1999). the number of addicts is thought to be high-- Roma are generally less likely to seek help at test- Nutrition ing and counseling clinics and are not counted Unhealthy diets are an important contributor (ECOHOST 2000). The head of the Drug Preven- to poor health status across Central and Eastern tion Center in Budapest estimated that 20 percent Europe (Galloway, Rokx, and Brown 2000). of patients treated in his clinic are Roma. He cate- Because of low socioeconomic status Roma are gorized Roma drug users into two groups, young more susceptible to unhealthy dietary habits children between 9 and 12, who are addicted to associated with poverty and low public health sniffing glue, and older addicts, usually over age awareness. A 1997 study of nutrition among chil- 19, who use "hard drugs" including heroin, dren in the Czech Republic found that the nutri- cocaine, speed, and LSD (Puporka and Zádori 44 Roma Poverty and Welfare: An Overview 1999). Drug usage and trade may be most preva- try level through case studies. Persistent disad- lent among communities in border areas, as is the vantages in education, including low school case in the Black Sea region of Bulgaria (Box 2.6). attendance and overrepresentation in special schools, which limit future opportunities, create a CONCLUSIONS high probability that without policy interventions The evidence suggests that the roots of perva- the next generation of Roma will continue to sive Roma poverty are closely linked to low edu- remain in poverty. cation levels, limited employment opportunities, and more unfavourable health status. The NOTES unfavourable starting point of Roma at the outset 1. The analysis of the Yale survey included in of the transition period--with low education lev- this chapter draws from Revenga, Ringold and els and overrepresentation among low-skilled Tracy, 2002. jobs--has led to disadvantages on the labor mar- 2. For more on measuring poverty, see Raval- ket. Compounded by discrimination and low lion 1993. expectations of employers, Roma have had more 3. Such as the Luxembourg Income Study. difficulty re-entering the job market than other 4. The official World Bank poverty rates are groups, and have consequently become caught in US$1 and US$2 per day, however because of a vicious circle of impoverishment. The next two higher heating costs in the Europe Central Asia chapters examine these issues further at the coun- region, the higher rates of $2.15 and $4.30 are Box 2.6: Heroin Addiction in Varna, Bulgaria The Black Sea port city of Varna is the third largest city in Bulgaria. According to the 1992 Census, 1.3 per- cent of the population identified as Roma, but the actual number is thought to be significantly higher. Because of its location on the Black Sea, informal trading opportunities with neighboring countries are rife and recent evidence from the Maksouda Quarter, a Roma mahala on the western outskirts of the city, indicates a flour- ishing drug trade, particularly in heroin. The Maksouda Quarter dates back at least 100 years to Ottoman times. Formerly a camp for nomadic Roma, the Quarter grew rapidly with the establishment of a textile factory at the turn of the century, and additional employment opportunities provided by the Varna ship-yards during the socialist period. The pop- ulation reached 15,000 by the 1970s. Informal sector activity has been prevalent, even under the socialist regime because of the large numbers of foreign tourists in Varna who are attracted by popular beach resorts, and opportunities for travel to other Black Sea border states. Among other ventures, currency trading, "trad- er-tourism" in clothing and other goods, and prostitution are common. According to estimates by the police and doctors at the Varna Medical University, there were approxi- mately 750 Roma heroin users in Maksouda in 1999. The users are predominately young between 13 and 35 years old, with two-thirds between 15 and 25. While a few began using heroin before 1989, serious trade and usage of heroin took off after the transition with the increase in travel opportunities. Drugs, including hero- in, marijuana, and cocaine are either bought abroad, or brought in by traders from countries. "There wasn't such a thing before. But when this democracy came, it began all of a sudden. It is mainly people from poor families that became addicts. There are also some from rich families, but not so many." Milko, 40 years old. More recently, addicts have shifted from smoking and inhaling heroin to intravenous injections. While no cases of HIV have been reported yet, there have been hepatitis outbreaks among users. The university hos- pital in Varna has a clinic for substance abuse, and many users interviewed identified it as an important source of help and hope for breaking the cycle of addiction. Source: Konstantinov 1999. 45 Roma in an Expanding Europe: Breaking the Poverty Cycle more appropriate for the countries analyzed in 9. A highly publicized attempt was made by this report. the city of Kosice in the Slovak Republic, which 5. Further information on the methodology sought to move people who were not paying rent used in this analysis can be found in Revenga, (largely Roma) to the Lunik IX neighborhood, a Ringold, and Tracy 2002. housing development on the outskirts of the 6. The dataset allows for multiple definitions town (OSCE 2000). of Roma ethnicity. For the analysis, the broadest 10. Results from the Romania Integrated definition of Roma is used. If either the individ- Household Survey, 1998. ual, or the interviewer indicated that the individ- 11. This reflects the fact that more Roma in the ual was Roma using any of the criteria included Bulgarian sample live in Roma settlements, in the survey, all members of the household are where housing conditions are generally poorer assumed to be Roma for the purposes of the than in more integrated neighborhoods. analysis. 12. For a discussion of labor market dynamics 7. The US and European literature on poverty in the early transition see Allison and Ringold and social exclusion finds that socially or eco- 1996; and Commander and Coricelli 1995. nomically excluded groups may often adopt 13. Employment rates are not comparable behavior patterns that differ from the majority across countries because of differences in defini- population, and which affect the return to pro- tion of the working-age population. ductive endowments and the overall welfare of 14. Birth rates are not ideal measures, as they the excluded population (Loury, 1999; Silver do not account for the age distribution, however 1994). fertility rates were not available. 8. This issue is discussed further in the next 15. The Romanian case study of Babadag chapter on Slovakia. found a high rate of intermarriage (Rughinis 2000). 46 Chapter Three: POVERTY AND EXCLUSION: ROMA SETTLEMENTS IN THE SLOVAK REPUBLIC T he situation of Roma in Slovakia is unique The chapter aims to address information gaps in a number of respects.1 More Roma in by bringing together findings from qualitative Slovakia live in settlements, on the out- case studies of Roma settlements with existing skirts of villages and towns, than in other coun- surveys. Sociological fieldwork was undertaken tries in the region. Many of these settlements are to supplement the incomplete picture given by rooted in exclusionary policies adopted during the quantitative data. Indeed, there is currently the Second World War and early socialist period no quantitative survey which allows for an which curbed the rights of Slovak Roma in many assessment of Roma living conditions in Slovakia. ways, including housing. Regulations allowed This is the first of the chapters in this volume Roma to enter towns and villages only on certain which draws upon country level qualitative days and at specific times, and ordered them to analysis. The chapter first provides historical move their homes a minimum distance of two background and current data on the population kilometers from all public roads. This policy of Roma in Slovakia. It then describes the nature formed the basis for the establishment of many of poverty in Roma settlements, their labor mar- Roma settlements which still exist in Slovakia ket status and coping strategies. Finally, the chap- today (Box 3.1). ter addresses access to public services, including The geographic and ethnic characteristics of education and social assistance. settlements vary significantly. An estimated one- quarter of Roma in Slovakia live in settlements, ROMA IN SLOVAKIA many of which are in the poorer, eastern regions of the country. The actual number is difficult to Historical Background gauge, because of the difficulties in measuring The oldest references to Roma living on the the Roma population and defining a "settle- territory of the Slovak Republic date back to 1322. ment." Living conditions for Roma in settlements Roma came to the area as settlers and nomadic are generally worse than for the rest of the Roma groups with travel permits issued by the Holy population. In this chapter, a settlement refers to Roman Emperor and the Pope. Roma who settled a group of people living together in a distinct in Slovakia worked as castle musicians, metal- geographic area, either within or outside of a workers, and served in the Hungarian royal town or village. armies. Anti-Roma policies began to emerge in As highlighted in the previous chapter, Roma the fifteenth century in Europe and intensified in poverty is multidimensional, encompassing the Hungarian kingdom in the sixteenth century, many aspects beyond low income. This chapter after the Turkish occupation of central Hungary, explores interrelated aspects of Roma poverty when Roma were thought to be Turkish spies. As and vulnerability further, including the material a result, Roma settlers were restricted to living on dimensions of poverty--nutrition, clothing, and the outskirts of towns and villages, and metal- housing--access to opportunities in the labor workers were allowed to sell only a limited quan- market, and social services. It discusses the par- tity of goods.2 ticular nature of exclusion experienced by Roma Restrictive policies continued during the in settlements in Slovakia. early Austro-Hungarian Empire in the eighteenth 47 Roma in an Expanding Europe: Breaking the Poverty Cycle Box 3.1: The Study of Roma Settlements in Slovakia A qualitative study by Slovak sociologists of conditions in Roma settlements in three contrasting districts in Slovakia form the basis of this chapter. In each district, fieldwork involved in-depth interviews with indi- viduals, households, and local public figures--including teachers, doctors, social assistance workers, reli- gious leaders and local government officials. The research was conducted between December 2000 and Jan- uary 2001. The study examined the characteristics and correlates of poverty, conditions in the settlements, and the experience of Roma in these areas. Although the survey is not representative, the results provide a snapshot of the conditions of Roma in geographically and socio-economically diverse locations. The districts were ranked based upon unemployment levels and the share of the population receiving social assistance. The dis- tricts were: Malacky: A better off district with below average unemployment (13.5 percent in 1999) and share of pop- ulation receiving social assistance benefits. Malacky is in the Bratislava Region near the capital city. There are very few segregated settlements in Malacky. Stará ŞubovŔa: An average region in terms of unemployment, social assistance beneficiaries and compo- sition of Roma settlements. The district is located in eastern Slovakia in the Preov Region where the con- centration of Roma is high. Rimavská Sobota: A relatively poor district in the Banská Bystrica Region, with a high level of unemploy- ment (35 percent in 1999) and a high share of the population receiving social assistance. The study looks at poverty including the lack of access to education and employment, income insecuri- ty, social exclusion, and the lack of opportunities for participation in civil society. Poverty is defined in dif- ferent ways, based upon self-assessment, and the interviewers' assessment of material conditions--including housing conditions, nutrition, health care, and access to public services. These measures are inherently sub- jective and the interviewers' assessments of poverty did not always coincide with those of the households being interviewed. Source: World Bank, Foundation SPACE, INEKO, and the Open Society Institute 2002. century. Leopold I declared Roma to be outlaws measures were used as models for other Euro- and ordered all Roma men to be hanged. Policies pean countries, which aimed to assimilate Roma changed under Empress Maria Theresa, and in the nineteenth and the twentieth centuries. Joseph II, her son and successor. Both sought to assimilate Roma as citizens within the Empire. WORLD WAR II. The first Czechoslovak Republic Legislative measures required Roma to settle, pay (1918­1938) passed legislation which limited the taxes, and provide compulsory service to local mobility and civil rights of Roma, particularly landowners. Other edicts included mandated nomadic and homeless groups. Laws mandated school and church attendance and improvement identification cards and fingerprinting. Condi- of housing infrastructure. tions deteriorated substantially during World These policies were the first step toward set- War II. Like Jews, Roma throughout Europe were tling the Roma, a feature that still distinguishes targeted with discriminatory legislation, and sub- Roma in Central and Eastern Europe from those sequently extermination under the "Final Solu- living in Western Europe. Although these policies tion." During the course of the "Devouring," as aimed, sometimes aggressively, at assimilation, Roma called the Holocaust, approximately one- they also represented the first time that Roma half million Roma from across Europe were were treated as state citizens. Austro-Hungarian killed. 48 Poverty and Exclusion: Roma Settlements in the Slovak Republic The experience of Roma in the Czech and Slo- nomadism, and drew up plans for a "dispersal vak Republics during the Holocaust differed sig- and transfer" scheme which aimed to resettle nificantly. The majority of Czech Roma were Roma from areas in eastern Slovakia to the Czech killed in concentration camps.3 In contrast, fewer lands. This program was never fully implement- Roma from Slovakia were deported to camps, ed, although many Roma families were trans- although many were sent to forced labor camps. ported to the Czech Republic against their will. In 1941, several labor camps were established for The program was coordinated by a Commission for Roma, where workers lived under extremely the Problems of the Gypsy Population in Slovakia, poor conditions. which was established in 1966 under the auspices After Slovakia was invaded by the German of the Presidium of the Slovak National Council. army in September 1944, the situation for Slovak In 1967 alone, 3,178 Roma were resettled from Roma became increasingly dire. Mass executions Slovakia. Of that number, a total of 1,034 Roma were carried out in several towns and villages, returned to Slovakia within the same year. and Roma living in the south and southeastern To combat nomadism, state officials broke up parts of Slovakia, annexed to Hungary during the caravans, sometimes slaughtering horses in the war, were transported to the concentration camp middle of the night (Fraser 1995). Policies relaxed at Dachau. somewhat during the period of the Prague Spring After World War II, large numbers of Roma reforms in 1968. Roma began to form official migrated from Slovakia into the Czech lands in organizations for the first time, and approximate- search of better living conditions and employ- ly 200 Roma musical groups and 30 football clubs ment. In many cases, migration was driven by were established.4 Forced migration and resettle- state policies which forced Roma out of certain ment policies resumed following the Soviet areas. Over several years, more than 15,000 Roma crackdown in 1969. Between 1972 and 1980, 4,000 migrated westward. As a result, the majority of Roma dwellings were destroyed and 4,850 Roma Roma living in the Czech Republic today are orig- were resettled. inally from Slovakia. Efforts to improve school attendance were similarly forced. Regulations were issued to THE SOCIALIST PERIOD. The Czechoslovak socialist implement compulsory schooling. Since the regime, which came to power after the war, objectives were not communicated to parents, adopted policies aimed at assimilating Roma and they tended to view school attendance as an eliminating ethnic differences. These measures externally imposed obligation. School attendance left behind a legacy that has affected the socioe- did increase dramatically. In 1971 only 17 percent conomic status of Roma into the transition peri- of Roma finished compulsory education; by 1980 od. The government refused to officially recog- this number increased to 26 percent. However, nize Roma as an ethnic minority, but rather many were enrolled in "special schools" intended identified them as "citizens of a gypsy origin." for the mentally and physically disabled. These Without the rank of ethnic minority, Roma lacked practices have persisted, and large numbers of certain legal and cultural rights. Among other Roma children in both the Czech and Slovak constraints, this implied that Roma cultural activ- Republics still study in special schools. ities were banned. Roma were not allowed to establish their own music ensembles, youth or ROMA IN SLOVAKIA AFTER 1989. With the Velvet Rev- sports clubs. Roma folk songs were not allowed olution in November 1989, came new opportuni- to be sung at schools, and Roma books and mag- ties for minorities to express their ethnic identity azines were banned. and participate in civil society. In January 1991 Assimilation policies in the areas of housing, the new Declaration of Basic Human Rights and employment, and school attendance were strin- Freedoms adopted by the Czechoslovak Federal gent and aggressive. In 1959, the government Assembly allowed for the free determination of embarked upon a violent campaign against ethnic identity. Subsequently, in April, the gov- 49 Roma in an Expanding Europe: Breaking the Poverty Cycle ernment of the Slovak Republic adopted "The Population Principles of Government Policy Regarding The Slovak Republic has one of the largest Roma." For the first time in history, Roma were shares of Roma to population in Europe. Accord- recognized as an independent ethnic minority, ing to the 2001 census, Roma represent 9.7 percent with equal status to that of other minorities living of the population, making them the second largest in the Slovak Republic. minority in the country after Hungarians. As The first Roma political party, the Romany many likely do not report their ethnicity in the Civic Initiative (ROI), was established after the census, the actual size of the population is thought transition in November 1989. Other parties and to be between 10 and 11 percent of the population, cultural associations soon followed. In the 1990 or between 420,000 and 500,000 people. The share parliamentary elections, Roma were elected to of Roma in the population is likely to rise in com- parliamentary posts for the first time and other ing years because of higher birth rates. Demo- Roma representatives were appointed to positions graphic projections have indicated that Roma within the Office of the Government, the Ministry could become a majority of the population in Slo- of Culture, and the Ministry of Education. vakia by 2060 (Economist 2001). Government activity related to Roma issues A survey of district officials estimated that accelerated in the late 1990s, with increased local there were 591 Roma settlements in Slovakia in and international attention. In November 1997, 1998, in comparison with 278 in 1988.5 The total the Slovak cabinet adopted the "Conceptual number of people living in settlements also has intents of the Slovak Republic for the solution of grown dramatically. In 1988 there were approxi- the problems of Romany population under cur- mately 14,988 people living in settlements, and rent social and economic conditions." The docu- by 1997 this figure had grown to 123,034. During ment outlined the issues facing Roma and institu- the past decade, some Roma have returned to tional responsibilities for addressing them. settlements because of a lack of affordable hous- One of the most significant developments was ing. This, in addition to the high birth rate among the establishment of the Office of the Plenipoten- Roma living in settlements, largely explains the tiary for Roma Communities after elections in increase.6 1998. The Office falls under the jurisdiction of the The demographic profile of Roma in Slovakia Deputy Prime Minister for Human Rights, Minori- is considerably different from that of the total pop- ties and Regional Development, and has been ulation. The Roma population is significantly headed by a Roma since 1999. The Office is younger and has been growing more rapidly than charged with implementing government policy other ethnic groups. The national birth rate for Slo- regarding Roma. vakia has declined steadily during the transition In 1999, the new government adopted the period from 15.2 live births per 1,000 population in "Strategy of the Slovak Government to Solve 1990 to 10.7 in 1998 (UNICEF 2000). In contrast, Problems of the Romany Ethnic Minority and the birth rates among Roma have been increasing, Set of Implementation Measures." The new docu- especially in the most isolated, segregated settle- ment formulated a more detailed action plan for ments. The life expectancy of Roma is considerably policy measures related to Roma issues. A second lower than the national average, although recent phase of this strategy was adopted in May 2000, data are not available. Estimates derived from the which further detailed measures to be undertak- 1970 and 1980 censuses put life expectancy for en. The document charged ministers and heads of Roma at 55 for men and 59 for women, in compar- regional public administration offices with specif- ison with 67 for men in the total population and 74 ic responsibilities. The areas of focus within the for women (ECOHOST 2000). strategy were: human rights, education, unem- ployment, housing, social security, and health. POVERTY IN ROMA SETTLEMENTS The strategy is limited in that it fails to specify In general, there are three types of settle- levels and sources of financing for activities. ments, based upon living arrangements between 50 Poverty and Exclusion: Roma Settlements in the Slovak Republic Roma and non-Roma. First are completely inte- leads to a vicious cycle: the more isolated and seg- grated towns and villages. This was the case of regated the settlement, the more severe and deep Nová ŞubovŔa in the district of Stará ŞubovŔa, a the poverty, the fewer opportunities residents have district of average development in the east of the to leave and work outside of the settlement, and country. Second are separated areas, where Roma consequently the higher the chances are that Roma live together within a town or village, either on will continue to live in isolated settlements and, the outskirts, or within a particular street or consequently, remain in poverty. neighborhood, as in Studienka in Malacky. Final- This level of spatial separation is positively ly, segregated settlements are situated outside of correlated to the level of poverty. The social status a village or town, such as Kyjatice in the district of Roma living in segregated settlements is con- of Rimavská Sobota, a settlement three kilometers siderably lower than that of those who are inte- from the nearest town. These definitions are sub- grated among the majority population. Roma liv- jective and were used to document general pat- ing in segregated settlements in marginalized terns. In particular, the distinction between sepa- regions are significantly worse off than those who rated and segregated settlements is frequently live in segregated settlements in more developed blurred. and economically better off regions. This study found that poverty has different The concentration of Roma also matters. The characteristics in the Roma and non-Roma popu- level of poverty in areas with a higher share of lations in Slovakia. Poverty among Roma is close- Roma in the population is higher than in areas ly linked to four main factors: (i) regional eco- where the population density of Roma is lower. nomic conditions; (ii) the size and concentration Poverty among Roma in districts where at least of the Roma population in a settlement; (iii) the five percent of the population was "officially" share of Roma in a settlement; and (iv) the degree classified as Roma (which likely underestimates of geographic integration or segregation of the the true population), was consistently worse than settlement and its proximity to a neighboring vil- those for the region as a whole. lage or town. With the exception of Roma in completely The situation of Roma in more economically integrated areas and some in separated settle- developed regions is generally more favorable ments in better off regions, high unemployment than that of Roma in poorer areas. For example, in and dependence on social assistance were com- 1999 the living conditions of Roma in Malacky, a mon in Roma settlements. While the national district with a lower unemployment rate (14 per- unemployment rate was 18 percent in 2000, in the cent) than the national average of 17 percent and qualitative sample, it was approximately 85 per- close to Bratislava (less than 50 km), were better cent. This was due to the inclusion of segregated than conditions in Rimavská Sobota, a district with settlements in the sample where unemployment 35 percent unemployment. Roma houses in segre- often reaches almost 100 percent. gated settlements in Malacky resembled those of The theme of contrasts between Roma living the majority population. They were generally in segregated and integrated areas cuts through made of solid materials such as bricks, and had this chapter. In general, Roma in integrated areas access to electricity. In contrast, housing conditions are less poor than Roma in settlements and have in settlements in Rimavská Sobota were poorer, greater access to opportunities in the labor mar- lacked access to basic services, and had worse ket and education. Conversely, Roma living in health and education status worse. isolated and marginalized settlements have limit- Within regions, the level of poverty in a Roma ed chances for upward mobility and interactions settlement appears to be closely connected to its with the rest of society. As conditions within set- geographic location, and the level of ethnic inte- tlements appear to be worsening over time, and gration and segregation. Conditions in settlements the population living in settlements is growing, which consisted only of Roma were significantly some observers have noted the emergence of an worse than in more integrated communities. This "underclass" of Roma in Slovakia, who are being 51 Roma in an Expanding Europe: Breaking the Poverty Cycle left behind in the processes of economic and Roma associate the previous regime with an political transition. The following sections abundance of job opportunities and benefits explore the extent of poverty among Roma, its including subsidized consumer goods, utilities, roots and correlates. and animals for breeding. Roma also recall hav- ing more housing options and better relation- Perceptions of Poverty ships between citizens. A Roma respondent Roma in urban and rural areas define poverty reflected, "People are not as willing to help each in both relative and concrete terms (Box 3.2). For other as they used to be because everyone has most Roma poverty is a recent phenomenon, and troubles today." Another noted: "During commu- they describe their living conditions mainly in nism we were better off because everyone had to relation to the past. Although none of those inter- work, even if it was pointless or unskilled work." viewed described themselves as well off before Many Roma also related their descriptions of 1989, most felt that they had lived decently rela- poverty to their current circumstances. Roma liv- tive to prevailing living standards. A minority ing in segregated, as well as many in separated said that they had always been "poor." The most areas, explained that the worst aspects of their salient difference with the communist period for present situation were: poor housing conditions, older Roma was that they all had jobs. overcrowding, lack of infrastructure, poor health, Box 3.2: Typology of Perceptions of Social Status Roma in the Slovak study can be categorized into four groups based upon their perceptions of their social status: Non-Poor: These Roma do not consider themselves poor, but rather view themselves as average Slovaks. They believe that there are many people who are worse off, and that the transition has not led to dramatic changes in their lives: "The only difference between Christmas today and Christmas during the Communist period is that today there are fewer presents." The problems, in their view, are national, including inflation, unemployment and subsequent constraints on living standards. This group constitutes a small share of the respondents living in integrated areas (e.g. approximately 25 percent in Malacky), who are either employed or engaged in the informal economy. Subjectively poor: This group can be characterized by the statement: "We are not rich but we are able to support ourselves." This is typical of integrated Roma who believe that the demographic groups hit most severely by poverty include elderly people, young families, and Roma from Eastern Slovakia. Their views about more segregated Roma are similar to the non-Roma view of Roma: "People there are worse off than dogs, but they are to blame. They should take better care of themselves. When they don't have a job, they should at least keep themselves and their house clean." The majority of people expressing these views had a better starting point after 1989; they lived in integrated localities in better off regions, and mostly own their homes. Relatively poor: These Roma perceive themselves to be poor and are generally unemployed, living on social benefits: "It is bad without a job, we live from one day to another." Most lived in integrated and par- tially separated types of settlements, and face difficulties in re-entering the labor market because of low edu- cation levels: "I have no clue what could help us out. If we could turn back time we would get a proper voca- tional training or move to another country. People on TV say that everybody is doing better there and that everybody has a job." Absolutely poor: This group of poor live mostly in segregated settlements in marginalized regions. They express a strong sense of apathy and helplessness and feel totally excluded from mainstream society: "We have nothing here, no roads, no electricity, no running water, no job. Nobody helps us either, not the mayor, or even the priest in the village." Some receive social assistance benefits, however, in certain cases some have lost eligibility because they lack documentation and official residency status. 52 Poverty and Exclusion: Roma Settlements in the Slovak Republic lack of adequate food and clothing, lack of a reli- and related institutions, mainly social assistance able social network, unemployment, and social offices, and to a lesser extent schools and health exclusion. A Roma respondent from a separated care centers. The majority of Roma living in segre- settlement explained: "We are poor because we gated and separated settlements describe a loss of don't have a proper house, we don't have any hope for the future, and a pervasive sense of money and have no one to borrow from." uncertainty and insecurity. It was common for households living in poor segregated settlements to identify food insecurity GENERATIONAL AND GENDER DIFFERENCES. Although as a main element of their poverty. One woman young Roma are less likely to compare their situ- explained that it was difficult for her to feed her ation to the past than their parents, the experience children properly all the time: "We have no cash and interpretation of "poverty" does not vary most of the month to buy food on a regular basis much across generations. Most young people and nobody will give us anything. Here we all identify the same problems and constraints in have the same condition." Generally, Roma from their lives as their parents: lack of jobs, adequate segregated settlements in marginalized regions education, and a sense of exclusion. associated poverty with material insecurity, while For young married couples, poverty means Roma in more developed and integrated regions the inability to live independently from their par- perceived poverty in relation to secondary needs ents, to start life on their own, and to enjoy priva- such as employment, quality education, and a cy and independence. In many settlements, more inclusive society. young couples live with their parents, or their Many Roma also compared their situations to inlaws, in one or two rooms with three or four of those of fellow citizens. Unemployed Roma liv- their siblings. Due to the low availability of hous- ing close to non-Roma felt much worse off in ing and high costs, many young Slovaks live with comparison with others. Roma in villages or their parents, however circumstances are espe- towns with non-Roma believed that it was more cially difficult for Roma in poor areas where the difficult for them to find work than their non- size and quality of housing is extremely low. Roma neighbors. As one said: "nowadays all the Perceived poverty among young Roma also work is for gadje."7 In contrast, Roma in segre- has a gender dimension. Some young girls noted gated settlements were less likely to compare that they were worse off than the young men in themselves to non-Roma. their communities and had access to fewer oppor- tunities in employment and education. They felt INSECURITY AND SHAME. Poverty has important social that their only option was to start having children and psychological components. Respondents liv- at an early age. A number of young women said ing in segregated settlements describe poverty as that they could not even get unskilled work, while associated with feelings of defenselessness, and young men in their communities could at least exclusion from the larger community. Poverty for participate in public works or unskilled jobs. many is also associated with shame. Even those These patterns reflect barriers to employment for respondents who appeared extremely poor to the young women, as well as traditional gender roles interviewers often preferred to define themselves for women in closed communities. as "close to" but not completely "poor." For the The poorest respondents identified common very poorest, however, "not completely poor" elements of poverty, including: inadequate nutri- means little more than "not dying of hunger." tion (e.g. sufficient food and nutritional composi- These responses have their roots in communism tion), poor housing, and ill health. The ability to which stigmatized poverty as a consequence of provide a good education for their children and personal failure and laziness. lead a better life--for example, having opportuni- Many Roma feel that existing institutions are ties to travel--were also identified by some Roma hostile, or at best indifferent, to their predicament. as important, but this took second place to the In particular, they lack trust in local governments more immediate issues of hunger and shelter. 53 Roma in an Expanding Europe: Breaking the Poverty Cycle Material Dimensions of Poverty It doesn't matter what comes tomorrow." A doctor in the town of Podolínec in Stará ŞubovŔa, who HUNGER AND NUTRITION. Prior to 1989, very few sees patients from a number of nearby settlements households had difficulty obtaining basic food- reported that she sometimes sees dehydrated stuffs, because of near-full employment and sub- babies. Mothers explain that they have no money sidized consumer goods. These circumstances for milk after their social assistance benefits run have changed. Roma households in the poorest out. Many Roma from marginalized settlements, settlements reported difficulties in affording suf- including some poorer integrated settlements, ficient food and maintaining adequate nutrition. admitted that during the week before social assis- Child malnutrition, in particular, was a frequent tance benefits were paid their family often had one problem. Researchers observed evidence of stunt- simple meal for the entire week. Many also said ing among some children. Some teachers report- that they had to buy cheap food items to make it ed that Roma children do not receive school through the month. A woman explained: lunches because their parents are unable to pay. The director of a school in Stará ŞubovŔa noted We have to buy the cheapest food and pre- that "in the entire primary school only one child pare it so that the whole family will not goes to lunch at school." feel hungry. I use fatty meat and potatoes Some elderly Roma also reported problems in to feed my family. maintaining adequate food intake and explained that they were unable to afford necessary foods Very few residents in segregated settlements because of the low levels of welfare benefits. An had access to land to grow food for their own elderly Roma man from the village of Rimavská consumption. Some pick mushrooms or berries Píla related that he had to maintain a high protein from the forests. Non-Roma living in nearby vil- diet for medical reasons, but could not afford it. lages reported that Roma steal potatoes and other Roma in integrated and segregated communi- food items from their fields. ties have contrasting strategies for ensuring ade- quate nutrition. Roma in integrated, as well as HOUSING POLICIES. Most Roma in segregated settle- many in separated areas, prefer to plan ahead and ments do not own their homes or land. In some economize in order to secure enough food for the settlements, property ownership is unclear. This rest of the month, regardless of their employment prevents the improvement of housing condi- status. Those who live in rural areas and own tions--since individuals and local governments land are able to grow vegetables during the sum- are unable to maintain or invest in buildings or mer months, and some do so. The wife of an local infrastructure. Roma were more likely to unemployed Roma man in a community in Stará have been left out of the property and land priva- ŞubovŔa reported that they tend to buy things tization processes that took place during the early which last, such as potatoes and beans in bulk. As 1990s than non-Roma. During the communist much as possible, she makes sure that her chil- period, houses were mostly privately held, while dren have sufficient food, despite the fact that her the land belonged to the state. The "tenants" husband is unemployed, and they live mainly on would rent their house or flat for 99 years from social assistance. "Sometimes I buy on credit, but the state. After 1989, the government privatized usually I make sure that we have enough to feed land, or gave it to municipal governments. The our family during the month." land was given to the tenants for free if the house In contrast, Roma in segregated settlements had a valid building permit, or appropriate legal focus more on their immediate survival and are status; and if the property was registered with the less able to plan ahead. Consumption tends to land-registry office and there were no pending increase after social assistance payments are made. applications for restitution. If these conditions A resident of Lipovec in the Rimavská Sobota dis- were met, the tenant could apply for the transfer trict noted, "why not eat now that we have money? of property to his or her name. 54 Poverty and Exclusion: Roma Settlements in the Slovak Republic Public communication regarding the process provide funds for investment in infrastructure, was limited, and many people were unaware of such as roads, and public services. their options and the steps needed to initiate the Roma are also poorly positioned to borrow transfer of land. In theory the mayor was respon- money, because of their economic status and lack sible for informing residents of their rights. How- of access to information on processes and proce- ever, in practice, few mayors did so. None of the dures. Loan criteria have become more demand- mayors in the settlements included in this study, ing since 1989, and the process for obtaining a with the exception of one mayor in Stará building permit has become extremely complex. ŞubovŔa, provided information to their con- Current requirements include 32 individual per- stituents without being explicitly asked. Roma in mits and approvals from different government integrated areas were more likely to learn about bodies. The research team encountered many the process from their neighbors, while Roma in unfinished homes that consisted of one or two segregated areas had more limited access to infor- rooms and a kitchen. Many of the occupants mation. As a result, a larger share of integrated began building before 1989 and were unable to Roma were able to secure property ownership. finish construction because they lacked financial Those who do not own their land are limited in resources or building permits. A Roma man in their ability to make needed improvements to Stará ŞubovŔa explained, "I started to build this their homes. A man from Kyjov, a segregated house before 1989, but could not finish it because Roma settlement in Stará ŞubovŔa, explained: I have no chance to put together enough money "We built our house with a building permit, but and cannot get a loan." there are still problems with the site, although it was officially given to us during socialism. But HOUSING CONDITIONS. Housing conditions vary sub- today the land is not ours, therefore we can not stantially between integrated and segregated install any water, gas, or sewage pipes." areas. Conditions are the poorest in the most iso- Roma in segregated areas face substantial lated and segregated settlements (Box 3.3). The challenges with legalizing their homes. The vast homes of Roma living in more integrated areas, majority of houses in segregated settlements were and those separated within a village, are similar. built illegally, mostly on land with unclear own- It is frequently not possible to identify the ethnic- ership. In some of these settlements, such as the ity of the owner from the outside of the house. village of JabloŔové in Malacky, Roma moved In segregated settlements, with the exception into the area in the early 1990s and began to build of Malacky, Roma houses are typically made of houses on municipal land at the edge of the vil- wood or scrap metal, plaster, tin, and tree branch- lage. As a result, they lack legal access to electric- es. However, the construction type varies within ity and water. In the case of electricity, they tap regions depending on the kind of building mate- into homes of neighbors who have legal connec- rials available in the area. In the village of tions, and pay them directly. Kolaăkov in Stará ŞubovŔa, there was only one Houses in settlements which are constructed stone house, while the others were constructed with makeshift materials often do not comply from wood and clay. In the same district, in the with basic construction standards and were built village of Kyjov, houses and shacks are made of a without the required permits. Some Roma mixture of stone and other materials, while in explained that the only way that they could ·ariské Jastrabie stone houses are the norm. In afford to build shelter for themselves was to use Lomniăka, a settlement of 1,200 people with only materials that they found around their settle- 100 houses, the majority (over 90 percent) of the ments, in forests or in garbage dumps. One houses were built from stones and bricks. explained, "we can never have legalized housing The extent of overcrowding within Roma and obtain a permit, so why ask." This creates a houses is closely related to the degree of segrega- vicious circle in which buildings do not have tion and geographic isolation of the community. legal status, and as a result, municipalities cannot In general, in both Roma and non-Roma houses 55 Roma in an Expanding Europe: Breaking the Poverty Cycle ELECTRICITY. In some of the most isolated settle- Box 3.3: Housing Conditions in a Village in ments, electricity was unavailable. In Stará Stará L'ubovna ŞubovŔa, two settlements lacked coverage and in four settlements households were receiving elec- Kolaăkov is a segregated settlement of 220 tricity through illegal connections. The situation inhabitants in the Stará ŞubovŔa district. None was similar in Rimavská Sobota, where seven of of the houses in the settlement are legally regis- the thirteen settlements either lacked electricity, tered. Unemployment is nearly 100 percent. In or relied on illegal sources. Residents of the village, a family of 7 people (the parents, their oldest daughter of 17, newly wed and preg- Rimavská Sobota explained that the lack of elec- nant, her husband, and three other children) tricity was particularly problematic in the winter, lives in a two room shack constructed from as it is difficult for them to afford candles or fuel. wood and tin. The house lacks access to water and sewage and there is no garbage collection in WASTE COLLECTION. Lack of garbage collection also the settlement. The family has a wood burning seriously affects living conditions and creates stove which is used for heating and cooking. health problems. In the majority of segregated settlements, garbage collection was either nonex- istent or sporadic because residents were unable in integrated areas, the qualitative study found to afford the service. Even in three segregated set- that there were approximately 1.5 people per tlements in Malacky (Lozorno, Malé Leváre, and room, while in segregated settlements there were Plaveck ·tvrtok), where nearly all homes had an average of 3 to 4 persons per room. Estimates access to electricity, residents complained about by district officials put the number of people per the lack of garbage collection. They noted that dwelling in Roma settlements at 9 in 1997. garbage dumps were located near to their settle- ments, but there were not enough bins, and col- ACCESS TO UTILITIES. Access to utilities and public lection was irregular. services is nonexistent, or limited, in most mar- The situation was even worse in the poorer ginalized settlements. The most serious problems districts of Stará ŞubovŔa and Rimavská Sobota. include lack of access to electricity, water, sewage, In most settlements in these districts, even if and garbage collection. Integrated settlements, garbage collection facilities did exist, residents and separated settlements within a town or vil- often complained that the municipalities only col- lage, were more likely to be connected to services. lected the garbage a couple of times a year (twice In the better-off district of Malacky, all settle- a year in Jakubany, or once in Şubotín in Stará ments, with one exception, had access to electric- ŞubovŔa). As a result, some residents throw their ity and roads. In the other districts, more isolated garbage into a nearby stream, or in the area settlements did not have access to utilities. around the containers. Roma also complained that garbage dumps WATER. Many settlements lack access to running were too close to their settlements, leading to con- water. Five of the seven segregated settlements in tamination of land and water, and in some cases, the study, and four out of ten separated settle- attracting rats, stray dogs and cats. Many local ments, had no access to running water. In some authorities blamed residents for not paying local areas, residents linked poor health conditions to fees for garbage collections. Mayors explained the inadequacy of the water supply. Residents of that some non-Roma communities purchased Rimavská Píla in Rimavská Sobota complained their own waste bins, while this was not the case that their drinking water was contaminated and in Roma settlements. Some mayors provided set- caused diarrhea, parasites, and trachoma among tlements with containers free of charge--but were children. In other areas, parents blamed epi- unwilling to pay for waste removal, despite the demics of scabies and lice on the lack of running fact that the charge for garbage collection was and hot water for washing. nominal.8 56 Poverty and Exclusion: Roma Settlements in the Slovak Republic Lack of garbage collection perpetuates nega- ECONOMIC ACTIVITIES AND tive stereotypes about poor hygiene among COPING STRATEGIES Roma. Some non-Roma blame Roma for the situ- The emergence of unemployment has been ation of poor waste collection in settlements. An one of the most serious social problems of eco- educated non-Roma commented: nomic transition in Slovakia.10 By 2000, unem- ployment had reached nearly 19 percent of the Gypsies are themselves responsible for labor force--the highest rate in the OECD. Unem- the terrible situation around their commu- ployment is closely linked with poverty. House- nities. I know of a situation where there is holds headed by an unemployed member are a garbage bin close to a building occupied more than six times more likely to be poor than by gypsies, but since it is 20 meters from households headed by an employed individual.11 the building and they are too lazy to walk Roma were more immediately affected by enter- there, they just throw their garbage out of prise downsizing at the outset of transition than their windows. other groups, and now comprise a disproportion- ate share of the unemployed. HEATING. Most Roma households rely on wood, Education levels are closely related to labor the cheapest form of fuel, for heating. Gas was market status in Slovakia. Unemployment rates for available to some households in integrated areas. workers with basic education, or less, were close to In Stará ŞubovŔa, gas was used by households in 40 percent in 2000 (World Bank 2001b). Workers three integrated settlements. In one of the segre- with vocational and apprenticeship education gated settlements, only one household had access have higher unemployment rates than workers to gas. In Malacky, a few households in three set- who have completed general secondary education. tlements used gas, while the rest relied on wood. This reflects changes in labor market demand Roma generally expressed little interest in having which have favored workers with more flexible gas pipes installed, because of the significantly academic backgrounds, rather than narrow techni- higher costs. In the majority of houses in rural cal training. As discussed further below, very few areas, wood burning fireplaces were used for Roma complete secondary education, and those both heating and cooking. Residents argued that that do are more likely to have participated in they could not afford gas since it was extremely vocational and apprenticeship schools, than aca- expensive to install a connection.9 demic secondary schools. The composition of reg- istered unemployment by ethnicity reflects the SEWAGE. Only households in integrated areas education status of Roma (Figure 3.1). have access to standardized plumbing. Most segregated and separated communities used Unemployment septic tanks or nothing at all. A few households The labor market status of Roma has changed in each district have toilets, but the majority use dramatically during the transition period. Under latrines. In Stará ŞubovŔa and Rimavská Sobota socialism, many Roma held formal public sector toilets were available in three settlements jobs, most commonly in agricultural cooperatives, included in the studies. In Malacky, all settle- factories, public construction, and mines. Many of ments, with the exception of Plaveck ·tvrtok these enterprises have closed or have been sub- had access to toilets. stantially restructured over the last decade. A 1997 survey by the Ministry of Labor, Social Affairs and HOUSEHOLD ASSETS. Ownership of cars was quite Family, estimated that Roma comprised between unusual. A few Roma in integrated and separated 17 to 18 percent of the total unemployed in 1996, areas had cars. Only a limited number of house- with this figure as high as 40 to 42 percent in east- holds had telephones. In segregated settlements ern districts with large Roma populations (e.g. only a few residents owned cellular phones and Koice, Spiská Nová Ves). Similarly, the registries cars; in many cases these were local moneylenders. from the National Labor Office (which contained 57 Roma in an Expanding Europe: Breaking the Poverty Cycle Figure 3.1: Registered Unemployment Figure 3.2: Unemployment by Duration, 1999 by Education and Ethnicity, 1999 (% of total unemployed) 100 90 100 80 90 70 Other registered 80 60 group) 50 unemployed 70 + 48 months of 40 Roma 60 (% 30 < 24 months group) 50 20 of < 12 months 40 10 (% 6 months or less 0 30 No education Elementary Technical General 20 secondary secondary 10 and 0 university Total Roma Source: Ministry of Labor, Social Affairs and Family, Slovak Source: Ministry of Labor, Social Affairs and Family, Slovak Republic. Republic. information on ethnicity until 1997) suggest that, aspects of their lives. A resident of Klenovec in for the country as a whole, Roma represented as Rimavská Sobota who had found employment much as one-quarter of all the registered unem- explained: "we were happy that we found a ployed in the Slovak Republic through 1999.12 meaningful way of spending a day. In two or three Furthermore, the share of Roma receiving unem- years a man gets used to doing nothing and then ployment benefits was lower than the share it gets really tough." Another respondent noted, among the total population. This was mostly due "when I had a job, it had a positive impact on the to the long duration of unemployment for Roma. family because everybody felt more secure." The majority of unemployed Roma have been Roma also expressed discouragement with the out of work for over one year. According to Min- lack of employment opportunities. Roma in segre- istry of Labor, Social Affairs and Family data for gated settlements are particularly disadvantaged, the first half of 1999, 92 percent of Roma regis- as job prospects are generally limited to seasonal tered as unemployed had been out of work for employment in neighboring towns and villages. A over one year, in comparison with 63 percent of 35-year-old father of five in a marginalized settle- the total population; 17 percent of Roma had been ment in Stará ŞubovŔa explained: "Who is going unemployed for over four years (Figure 3.2). to give me a job? I have no education, no skills, Most of the Roma interviewed for the qualitative and am Roma, even in my neighboring village study had been out of work for over two years. nobody wants to give us any work." Even though unemployment is a problem Unemployment among young people, and faced by Roma across Slovakia, to a large degree especially women, is high. Most young Roma its extent is linked to regional economic condi- interviewed from the settlements had never been tions. In Malacky, where the overall district un- formally employed. Young women generally do employment rate was 13.5 percent in 1999, unem- not enter the labor force, because of early preg- ployment among Roma ranged from 60 percent in nancies. Many get married and begin having chil- integrated settlements, to nearly 100 percent in the dren soon after completing primary school. Near- most segregated settlements. In Stará ŞubovŔa ly all of the girls over 18 interviewed for this and Rimavská Sobota, where total unemployment study, with the exception of those from more inte- rates were higher, unemployment among Roma grated villages in Malacky, or those in completely was between 80 and 100 percent. integrated areas in other districts, were already Many Roma identified ongoing unemploy- married with children, or pregnant. Women in ment and insecurity as the most demoralizing more integrated areas were more likely to be 58 Poverty and Exclusion: Roma Settlements in the Slovak Republic employed in traditionally female jobs as teachers, rated communities are also discouraged from cleaning ladies, or public administrators. migration because of high costs and insecurity about finding work. It is more common for Roma Employment families to move from towns and villages to set- The employment status of Roma included in tlements, rather than the other way around. the survey differed according to the degree of segregation. The majority of Roma from highly Discrimination integrated settlements had finished secondary Many Roma cited ethnic discrimination as a vocational education and had regular jobs, significant barrier to employment, and as a regardless of gender. In contrast, of Roma who rationale for not seeking work outside of their were employed in the settlements, most were communities and villages. Although Slovakia has engaged in unskilled labor, frequently in season- adopted antidiscrimination legislation, consistent al agricultural work, or construction. In many set- with ILO conventions, Roma described experi- tlements, public works are the only source of ences of discrimination. A number of Roma relat- employment. A few Roma were employed in ed anecdotes about friends or relatives who had more skilled labor, including construction and applied for a job, and although they were accept- stone masonry, some of them had vocational ed over the phone, were subsequently rejected as training. However, not all Roma with vocational soon as the employer realized that they were education had jobs. Roma. While none of the Roma in the study had experienced this directly, it undoubtedly had an Labor Mobility and Migration effect on their readiness to apply for jobs. Labor mobility among Roma and non-Roma A school director explained that a Roma in Slovakia is low. Of those Roma in the study woman had applied for a teaching position in his areas who were employed, most worked in the school. He had a difficult time deciding whether immediate surroundings of their settlements, to hire her, since he suspected that non-Roma par- because of transportation costs. Very few Roma ents might protest his decision. In the end she sought employment in neighboring districts or was not hired. A director of a vocational school in countries, such as Hungary and the Czech Podolínec for cooks and waiters reported that he Republic. Those that did commute to the Czech had difficulty finding restaurants which would Republic complained that their wages were too accept his Roma students for practical training. low to make it worthwhile, that employers were Roma also explained that they were denied often late in paying wages, or did not pay at all. employment because of low education levels: Roma were more likely to work abroad if some- "Even trained people have no chance to find a one else in their family or settlement had gone job, so how could I find one?" Women noted this first and had a successful experience. Roma from problem even more than men, "Men are allowed Malacky and Stará ŞubovŔa were working in the to take jobs for which they are not trained, but Czech cities of Hradec Králové and Ostrava. from a woman, they always require that she be However, Roma noted that commuting had been trained." Labor market discrimination was a more common during the socialist period, "hard- source of stress for many Roma, and in many ly anyone from our village goes to the Czech cases led people to give up their job search. A Republic these days, as it was in the past." young Roma in Rimavská Sobota expressed a Other Roma work in construction or seasonal common sentiment: "No one will employ a agriculture in nearby towns or villages, where Gypsy anyway, why try?" transport expenses are lower. Moving perma- nently--or for extended periods of time--to other Public Works Programs districts or towns was not an option for most Many Roma participate in public works pro- Roma. Roma from segregated communities are grams run by the Ministry of Labor, Social too poor to afford to move, and those from sepa- Affairs, and Family through local municipalities. 59 Roma in an Expanding Europe: Breaking the Poverty Cycle This program was initiated in 2000. Jobs general- construction tasks. Some men painted houses and ly last three months and most commonly involve women worked as cleaning ladies. unskilled work, such as cleaning streets and Roma in geographically isolated and segre- parks, and garbage collection. Jobs do not include gated areas have fewer opportunities for informal training or preparation for future employment. A employment because their communities are significant share of Roma, especially those in sep- closed off from broader society. Moreover, they arated settlements in all of the three districts, par- have limited connections outside of the settle- ticipated in these projects. However, these pro- ment to help them find work. A number of Roma grams may not always reach Roma. In two admitted to resorting to theft as a coping strategy, localities Roma explained that they were denied including stealing potatoes, firewood, and con- participation in the local public works program struction materials. because the mayor preferred to hire a non-Roma applicant. ACCESS TO CREDIT. Roma lack opportunities to bor- Some Roma respondents complained about row money, and therefore have limited capacity the quality of work in the public works program to establish small businesses. Credit is scarce and and observed that, in some cases, work was costly for all small borrowers in the Slovak focused almost exclusively on cleaning around Republic but Roma may face additional hurdles. non-Roma houses, and ignored Roma neighbor- In many cases Roma lack collateral to borrow hoods and settlements. On the other hand, many because of unclear property ownership. Access to Roma interviewed explained that public works loans from commercial institutions is virtually were a better alternative to unemployment: zero. Some Roma do borrow small sums from "When a man has a job, it is easier to live, he is neighbors, friends, and relatives, as well as healthier, he has more energy and life is more through local Roma usurers. In some communi- fun." ties the Roma leader, or vajda, lends money, how- ever interest rates were reportedly extortionate-- Coping Strategies at 40 percent or higher, while the interest rate for consumer credit was around 14 percent. INFORMAL SECTOR EMPLOYMENT. Due to limited formal employment opportunities, many Roma work in SUBSISTENCE FARMING. Growing food was not the informal sector. Because of the absence of reported to be an important coping strategy for taxes and official and unofficial fees, informal the majority of Roma, including those who actu- employment is frequently more attractive than ally own land. Many Slovaks cultivate land, formal jobs for both employers and employees. including small plots and gardens, to support Common activities include salvaging and selling their consumption. This practice was common scrap metal, petty trade, and part-time work in during the socialist period, although never for agriculture and construction. Roma. Nearly all Roma households in integrated One of most widespread informal economic settlements, and some living on the margin of vil- activities for Roma in the study settlements was lages, own at least a small amount of land. Most working as musicians. This was particularly the household plots are small, ranging from 8­10 by case for Roma in Jesenské, Hodejov, and an urban 3­4 meter plots in back of their houses, or larger ghetto in Rimavská Sobota. A few Roma had if not adjacent to the house. Some more affluent small workshops where they produce tools for households did cultivate land. Crops vary construction workers, such as in Kaloa in according to region and include potatoes, wheat, Rimavská Sobota. Another common activity, grapes, and vegetables. mainly among those from segregated localities, The majority of Roma in segregated settle- was to salvage scrap material for resale. Other ments do not own land. In two settlements in occasional and informal employment, especially Stará ŞubovŔa, families owned their homes and for men, included helping non-Roma with minor land, and have been involved in agricultural 60 Poverty and Exclusion: Roma Settlements in the Slovak Republic activities for three generations. In Studienka and patterns of Roma in the settlements were consis- Malé Leváre in Malacky, all of the households tent with this pattern. The majority of adults owned land, but only half grew crops. Roma interviewed in the settlements had some primary explained that they did not make use of land for education, although not all of them had complet- a number of reasons. In some cases the plot was ed all grades. too small to be viable, in other cases the soil was Almost all Roma from segregated, as well as poor, there was no convenient source of water, or some from separated areas, had not completed the household could not afford the necessary secondary school. In many cases, students inputs. Others explained that cultivation of land dropped out after completing 10 years of compul- was not traditionally a Roma occupation. sory education. Secondary education in Slovakia Very few families raised animals. Some fami- includes three main types of schools: gymnasia (or lies in the settlements, including those without grammar schools); vocational schools; and spe- land, kept chickens or pigs. However, raising cialized secondary schools. Gymnasia provide livestock for household or commercial use was general academic training and prepare students to not reported. This was mainly due to the lack of continue on to university. In 1998, 21 percent of land. Only five families included in the study cul- Slovak secondary students were enrolled in gym- tivated land and raised animals. Some non-Roma nasia. None of the Roma interviewed for the study explained that the breeding of animals for home were enrolled in, or had attended, gymnasia. use had declined during the transition period. Vocational schools include apprentice Prior to 1989 it was common for agricultural schools, which prepare students for specific occu- cooperatives to give employees animals for pations through two-year programs, secondary domestic use, but now "[Roma] do not breed vocational schools, which offer two to three year them since no one hands out small pigs for free programs, and secondary specialized schools, anymore." which prepare students for the labor market in specialized fields through professional programs. ACCESS TO SOCIAL SERVICES Most secondary school students are enrolled in Roma in settlements are more likely to be geo- vocational and apprentice schools, 46 percent in graphically isolated and out of the range of cov- 1998, and 33 percent in secondary specialized erage of health services and education--particu- schools. Graduates from secondary vocational larly preschool. Communication problems schools are not eligible to enter higher education between non-Roma service providers and Roma institutions unless they complete an additional also affect access and quality of services. Some two years of education and pass an examination, Roma who are not fully proficient in the Slovak while graduates from secondary specialized language are unable to communicate effectively schools may continue on to university. The major- with teachers, doctors, social workers, and other ity of Roma who had continued on to secondary service providers. Social isolation and mistrust school were enrolled in apprentice schools or sec- between Roma and non-Roma also influences ondary vocational schools. Roma from integrated relationships and access to services. areas, and some better-off Roma from separated areas, were more likely to attend secondary spe- Education cialized schools. Most of the respondents who According to the 1991 census, 77 percent of had graduated from these schools had jobs. Roma had completed primary education, 8 per- Many Roma do not see a direct relationship cent had completed vocational training, and less between education and employment, partly than 2 percent had completed academic second- because of widespread unemployment. The ary or university education.13 An earlier survey majority of Roma in separated and segregated from 1990 found that 56 percent of Roma men communities have only primary or unfinished and 59 percent of Roma women had not complet- secondary education. In general, unskilled work- ed primary education (Vaeăka 2000a). Education ers have found it increasingly difficult to partici- 61 Roma in an Expanding Europe: Breaking the Poverty Cycle pate in the labor market. This may reflect the lack absence of electricity in isolated settlements of demand for labor with low skills; it may also be makes it difficult for children to study and do due to high payroll taxes and other non-wage homework. Some Roma children need to stay costs which--given differences in productivity-- home to help with housework and take care of make unskilled labor relatively costly compared siblings. As a result, they have difficulty keeping to hiring workers with higher skills. up with the curriculum. In the poorest areas, such as segregated settlements in Stará ŞubovŔa and SCHOOL ATTENDANCE. Teachers and school directors Rimavská Sobota, there were reports that chil- in the study districts reported that the attendance dren were unable to attend school because they of Roma children had been declining since 1989. lacked clothing and shoes. Particularly in the poorest settlements, many chil- dren were observed playing in the streets during PRESCHOOL ATTENDANCE. Few Roma children from the school day. Some doctors reported that Roma segregated settlements attend preschools. children came to them to ask to be excused from Preschool in Slovakia is not compulsory and gen- school. Very few Roma children in the areas visit- erally includes children between three and six ed for the study continued beyond compulsory years of age. Most segregated settlements lack education. preschool facilities. An exception was the settle- Under socialism, penalties for truancy were ment in Plaveck ·tvrtok in Malacky, where the more stringent and frequently enforced through church had opened a preschool mainly for the mechanisms including interrogation by the children of the settlement. Many parents inter- police, placement of children in institutions, and viewed did not recognize the value of preschool, reduction of social benefits. Some examples of and felt that mothers could adequately prepare these types of penalties were found in the study their children. A Roma mother explained, "all of sites. In Rimavská Sobota, teachers reported my children are at home, together with me, I am absent students to the police and cut welfare ben- at home, so why send them to kindergarten?" efits to motivate attendance. As a result, many Parents also cited costs related to attending parents understood education more as an obliga- kindergarten such as fees and clothing as a deter- tion to the state than to their children. One parent rent. "Kindergarten is not free of charge, we explained: "They must go to school, this is the would need to pay and we cannot afford that."14 law. The teacher was here and told us, if we do Because Roma children begin primary school not send our children to school, we will lose our unprepared, they face additional difficulties in financial support." adapting to the school environment. These circum- Children from the most segregated and isolat- stances exacerbate preconceptions of non-Roma ed settlements face the greatest challenges in students and teachers that Roma are not capable of accessing education. Some settlements are simply learning, and lead to further exclusion. In many too small to be able to have their own school. In cases, Roma are placed in separate classes or spe- Malacky and Stará ŞubovŔa, all separated settle- cial schools because of their lack of preparation. ments either had primary schools, or there was a school close by. In Rimavská Sobota, five settle- LANGUAGE. Roma in Slovakia also differ linguisti- ments included in the sample had fewer than 500 cally. Over half of Roma in Slovakia are thought inhabitants and no primary school, so children to speak some of the Roma language, but it is not commuted to neighboring villages. Roma moth- known how many speak it at home. Roma from ers from Kyjov, a settlement in Stará ŞubovŔa, isolated and segregated settlements may be intro- asked school officials not to let their children go duced to the Slovak language only once they on to the fifth grade because they were unable to enter primary school. pay for transportation to the school. Teachers are poorly equipped to handle this Poverty and a lack of basic infrastructure are gap in the children's knowledge, and in some also notable barriers to school attendance. The cases send Roma children to separate classrooms 62 Poverty and Exclusion: Roma Settlements in the Slovak Republic or special schools if they cannot keep up with the future he will put more importance on education rest of the students. School directors explained: than we did." A resident of Rimavská Sobota con- curred, "I think that Roma should change. For Children from segregated [Roma] settle- example we need to make sure that our children ments do not master the Slovak language go to better schools, because their future depends and do not understand their teachers. The on that." teachers do not speak the Roma language, A significant share of Roma view education as so they communicate by using gestures. a system representative of gadje society, which is of limited relevance for them. Parents explained: In a school where teachers do not speak "From the beginning, since the first grade, our the Roma language at all, or only some, children have difficulties understanding what is the first grade is not enough for these chil- going on: other children are singing the songs we dren to eliminate the gap [with other chil- do not know." And, "all poetry, literature, history dren]. is not about and from our life." It is easier to remove the language barrier PARENTAL INVOLVEMENT. As demand for education in mixed classes, but many Roma kids are among Roma in isolated and segregated commu- in separate classes. nities is low, Roma parents are less likely to be involved in their children's education. Many The issue is even more complex in ethnically Roma students lack effective role models. Roma diverse areas, such as Slovak-Hungarian areas in parents are frequently poorly positioned to help the south. In Rimavská Sobota, some children their children with school work at home because speak Hungarian in addition to the Roma lan- of their own limited educational backgrounds. In guage, but are not proficient in Slovak. Others are more integrated areas some parents were neither fully proficient in Slovak nor Hungarian, involved in schools. A parent in Malacky noted yet attend Hungarian schools. The situation is the importance of being involved: "I help my chil- similar in some villages in Stará ŞubovŔa, where dren learn every day, if I miss out on one day of most non-Roma speak Ruthenian. reading with my son, the very next day he has a problem. Therefore I help them study every day." DEMAND FOR EDUCATION. Low demand for education However, most Roma students lack the advan- among some Roma families discourages children tages of other students whose parents assist their from attending school. This has its roots in chron- children with school work, and/or hire private ic unemployment, which is common in many tutors. Roma settlements due to the lack of job opportu- nities. The dismal labor market situation leads SPECIAL SCHOOLS AND CLASSES. As discussed in Chap- parents to undervalue the importance of educa- ter Two, Roma are at a higher risk of receiving tion. A Roma parent noted, "my daughter com- lower quality education because of institutional pleted secondary school, now she is sitting at factors and incentives which lead to separate home without work." Another asked, "why force education for Roma and non-Roma. Special our children to study when there aren't jobs for schools are a legacy from the socialist era, and the educated ones?" were designed to provide special education for In some cases, parents, especially those from children with mental and physical disabilities. A integrated and separated localities where disproportionate share of Roma are enrolled in employment opportunities are greater, acknowl- special schools. A majority of Roma students edged the importance of education for their chil- from the segregated settlements in the qualitative dren's future. A grandparent in Malacky study attend special schools.15 Students enrolled explained, "My grandson is a first grade student. in special schools are at a dual disadvantage, first We sent him to kindergarten and hope in the because the curriculum is less rigorous and 63 Roma in an Expanding Europe: Breaking the Poverty Cycle expectations of teachers are lower than in main- want to put their first child here, but as they have stream schools, and second, because opportuni- more children they find out that here the children ties for graduates of special schools are limited. achieve better results than in a `normal' primary Even when Roma children are educated with- school." Roma parents also indicated that they in the mainstream Slovak school system, they preferred special schools because there are more may be placed in separate Roma classes. The Roma children and their children are "protected" majority of primary schools in segregated and from discrimination and hostility from non-Roma separated settlements have separate classes for students. In some cases, special schools provide Roma students. Maximum class sizes are low, and housing, making them more financially attractive provide teachers with a rationale for separating to parents. Roma children.16 According to teachers, non- Roma parents favor this separation by arguing TEACHERS. Teachers are central to the quality of that Roma students slow down the educational education and play an important role in motivat- process. These dynamics create an environment ing student attendance and performance. In that can be hostile. A Roma mother in a village in many settlements, teachers were poorly prepared Stará ŞubovŔa observed that "children are not to work with Roma children. Many teachers racist, it is their parents that tell them to keep sep- interviewed expressed an interest in training and arate, and that is why they tease our kids and call teaching materials in Roma culture and history, as them names." very few of them had any knowledge of Roma Some Slovak teachers argued that Roma issues. Prejudices and low expectations of Roma should attend special schools and classes because students by teachers can adversely affect student they need special care and assistance which can- performance. This phenomenon manifests itself not be provided in a regular classroom. Others in different ways. Some parents complained that took an opposite view. A third grade teacher at a teachers did not let their students bring textbooks primary school in ·ariské Jastrabie; in Stará home because they believed Roma children ŞubovŔa explained: "It is simplistic to consider would destroy them. As a result, students lacked these children mentally disturbed--and there the opportunity to do homework and adequately should be even more reasons to step up the effort. prepare for classes.17 If you can do it, they catch on." The study also found a number of examples Despite the disadvantages of special schools in which teachers and school directors took the and classes, some interviewed parents believed initiative to reach out to Roma communities and that their children receive more attention at spe- support Roma children at school, but these exam- cial schools and are not singled out. A Roma ples were sporadic and stemmed from individual mother said: "The youngest son does not go to a initiative. Educational advisors also played an kindergarten, since I am at home. My son and important positive role in some schools. In daughter go to a special school. At the beginning ·ariské Jastrabie, advisors worked with Roma my son went to a normal primary school, but he parents to encourage them to send their children was not good in reading, so the teacher suggest- to school and continue on to secondary educa- ed he go to a special one. We are satisfied with tion. In some communities, such as Jarovnice, him, he gets only A's. We put our daughter into a Tepl Vrch and JabloŔové in Malacky and special school ourselves." Most Roma parents Rimavská Sobota, teachers and school officials expressed a preference for mixed classes, so that maintain close relations with Roma parents and their children would be exposed to the Slovak children. They make frequent visits to Roma set- language. tlements and work to mitigate conflicts between The director of a special school noted, children. "approximately 30­40 percent of children attend Some teachers visit Roma settlements on their special primary schools on the basis of their par- own initiative to persuade parents to send their ents decision. Sometimes, the parents do not children to school. Because Roma from segregat- 64 Poverty and Exclusion: Roma Settlements in the Slovak Republic ed and some from separated settlements often do by the distorted incentives arising from the not have officially registered residences, local and design of the social safety net. Social assistance in school officials would not know about some Slovakia lacks mechanisms for benefits to taper Roma children without the assistance of teachers. off gradually as workers become employed, thus A teacher explained the challenge of convincing building pro-work incentives. Consequently, the parents to send their children to school: "One boy system penalizes those who find employment told me that his father did not want to enroll him and sets up a dependency trap. The relationship in a secondary school. So I invited his father to between the design of the safety net and these school and tried to convince him that it was a distorted work incentives is not in any sense good idea. I think now [the boy's] chances are unique to Roma families, but the demographic about fifty­fifty." characteristics of the Roma, with relatively low In areas where teachers and school directors levels of educational attainment, and a large were more available and involved in their com- number of children, makes them particularly vul- munities, Roma parents expressed satisfaction nerable to falling into this dependency trap. with the schools, and children were happy to Many Roma complained that the reforms to attend school. Roma mothers explained: the Act on Social Assistance, which cut benefits for those who had been unemployed for two years or The teacher visits our settlements on a more, made it impossible for them to survive on regular basis. She has bought some books social assistance. Although this change was for my children and also organized after- intended to promote work incentives, Roma in noon activities for them. isolated settlements were particularly disadvan- taged because of the absence of job opportunities. We go to school meetings, but that is not Non-Roma social workers and local government the only meeting with the teacher. He officials also felt that the current system of child comes here, to the settlement, and borrows allowances and the subsistence minimum provid- tapes with Romany music. Children then ed incentives for Roma to have large families. learn Romany songs with the teacher. While there is no empirical evidence to confirm this, the importance of these benefits for the sur- The interviews indicate that such initiatives vival of many poor Roma families contributes to have increased communication and understand- the impression among non-Roma that Roma are ing between Roma parents and schools in these overly dependent on the state. communities. Relations between social workers and Roma were reportedly more contentious than relations Social Assistance with other public service providers. Roma view Social assistance benefits provide an important social workers as representatives of the state, and source of income for many Roma households. they are frequently the only contact Roma have Nearly all of the long-term unemployed Roma with government authorities. Social workers are interviewed for this study, and especially those liv- responsible for conveying "bad news" on eligibil- ing in poorer segregated settlements, are depend- ity for benefits, and as a result, are often the tar- ent upon social assistance benefits for income sup- get of frustration with decisions that are not nec- port. Many noted that these benefits were essarily under their control. indispensable, but felt that they were not adequate Social workers are poorly prepared to work to secure basic living conditions. For many out- with Roma communities. This lack of preparation siders, the dependency of Roma on benefits rein- is linked to systemic problems within the welfare forced stereotypes of Roma as social parasites who system itself. Social workers in Slovakia rarely do would rather receive income support than work. field visits and are not trained to work directly Reintegration of unemployed Roma workers with clients. Instead, their jobs are largely admin- into the labor force may be made more difficult istrative, focused on disbursing cash benefits. 65 Roma in an Expanding Europe: Breaking the Poverty Cycle Social workers explained that they had no time need to be sensitive to Roma culture and the desire left for field visits and complained about the of communities to maintain their cultural identity. administrative burden of their work: "Every time This objective can be ensured through partici- the law is amended, we have to check and review pation of Roma. A number of successful projects all files. We often work late in the evening and do use Roma mentors as liasons between Roma and not have time for fieldwork." Only two of the non-Roma communities. For example, Roma assis- social workers interviewed for the study actually tant teachers who work with parents, or peer advi- visited Roma settlements. The lack of contact sors who assist with job placement can facilitate between Roma and social workers contributes to integration, while strengthening the Roma com- poor communication on both sides. munity itself. Many Roma complained that social workers Addressing exclusion and the negative were not responsive to their needs. Some social impacts of segregation also involves overcoming workers were not effective at communicating divisions between Roma and non-Roma commu- with Roma, as many Roma lacked basic informa- nities. Measures in this regard need to involve tion on social assistance programs and eligibility Roma and non-Roma alike. Education is an criteria. Some Roma asked the interviewers for important vehicle for overcoming cultural barri- information on various benefits. In other cases, ers by including the history and culture of Roma Roma appeared well versed in the eligibility cri- and other minorities in the curriculum. Training teria of benefits. of teachers, local government officials and other personnel working in social services can be ADDRESSING POVERTY IN SETTLEMENTS important mechanisms for addressing discrimi- An important finding of the field work in Slo- nation within public services. Finally public vakia is that the degree of segregation and mar- information campaigns can promote multicultur- ginalization of a Roma settlement is correlated alism and raise awareness about discrimination. with the level of poverty in the settlement. While Addressing Roma poverty in Slovakia is a com- these linkages need to be validated through fur- plex challenge, which will take time and greater ther research, the basic findings are clear. Roma understanding of the opportunities and chal- living in more remote and segregated settlements lenges ahead. have fewer opportunities to participate in the mainstream economy, access social services, and NOTES tap into social networks and information about 1. This chapter is based on a study by the jobs. In other words, geographic and social exclu- World Bank, Foundation SPACE, INEKO and the sion are important correlates of poverty. In con- Open Society Institute, Poverty and Welfare of trast, Roma in integrated areas are more likely to Roma in the Slovak Republic. Bratislava, 2002. The interact with non-Roma and are better informed work was led by Iveta Radicova, Helen Shahriari, and positioned to identify and take advantage of and Dena Ringold. opportunities. 2. For a more detailed discussion of the histo- These results have important policy implica- ry of Roma in Slovakia see Crowe (1994). tions. In the first place, they highlight the diverse 3. Estimates differ, however approximately nature of Roma in Slovakia and the need for varied 6,000­8,000 Czech Roma are thought to have been approaches to different circumstances. Second, killed. they indicate that interventions which reduce iso- 4. Roma were officially allowed to form lation and exclusion of Roma through integration organizations during the first Czechoslovak can facilitate the improvement of living conditions Republic (1918­38), but none did. The first Roma over the longer term. This does not imply that pro- organization was established in 1948 and was grams and policies should revert to the type of banned soon after by the communists. forced assimilation which was prevalent under the 5. This figure is based on a loose definition of socialist period. Rather, policy and project design settlements, including integrated areas in towns 66 Poverty and Exclusion: Roma Settlements in the Slovak Republic and villages. It is unclear to what extent the high- judged by labor office staff, which was inconsis- er number of settlements in 1998 reflects an actu- tent with Slovak legislation aimed at protecting al increase, or whether it is due to changes in the basic individual rights. way in which settlements were counted. 13. Primary education in Slovakia includes 6. The database, housed in the Office of the grades 1­9 and generally covers children from 6 Plenipotentiary for Roma Communities, provides to 16 years old. only a rough estimate of the number of settle- 14. Fees are set regionally and vary based ments and their conditions. The fieldwork con- upon the economic situation of the region. For ducted for this study found significant errors in example, in 2000 fees ranged from 600 Slovak the database regarding the number and location crowns in Bratislava (approximately US$13), to 20 of settlements. crowns per month in Rimavská Sobota (approxi- 7. "Gadje" is a Roma word for non-Roma. mately US$0.50). 8. The amount differs from one municipality 15. There are approximately 380 special to another, depending on the wealth of the schools throughout the Slovak Republic for men- municipality. For instance, in a better-off neigh- tally and physically disabled children. A total of borhood close to Bratislava the annual collection 30,583 students study in special schools, which fee is 1000 Slovak crowns (about US$21). In other amounts to about 3 percent of the total number of areas it is much less. students at kindergartens, primary schools and 9. According to residents of one village it can secondary schools altogether. cost up to 47,000 Slovak crowns for a gas connec- 16. The minimum class size is 4. The maxi- tion (close to US$1000). mum number of students is 8 for grade 1; 10 for 10. For an in-depth discussion of the Slovak grades 2­5; and 12 for grades 6­9. labor market, refer to World Bank 2001b. 17. In Slovakia, textbooks are free of charge. 11. Based on 1996 Microcensus data (World Children keep their books at home. However in Bank, 2001b). some--especially segregated--settlements, teach- 12. The practice of collecting information ers keep the books in the classroom, explaining based on ethnicity was discontinued in 1998 after that children do not have a place to keep the books protest from Roma and Hungarians. One reason at home and as a result they are damaged for these criticisms was that ethnicity was being throughout the course of the year. 67 Chapter Four: THE DIVERSITY OF ROMA IN ROMANIA R omania has the largest Roma population in lation under the socialist Ceaucescu regime. The Central and Eastern Europe, and also one legacies of these different regimes have had of the most diverse.1 This variety reflects important implications for the overall status of historical, religious, linguistic, and occupational Roma in the country. While the socialist period characteristics, which are often overlapping. brought some improvements to Roma in social Roma communities also vary in terms of regional and economic terms (see Chapter One for a gen- settlement patterns, levels of integration, and eco- eral discussion), the assimilationist policies of nomic and social development. However, Roma this era were accompanied by considerable in Romania face common issues, related to access political repression and created a gulf of mis- to education, health care, social assistance, and trust between Roma and the state which contin- housing, that underpin widespread poverty. This ues to this day. chapter looks at these common challenges, draw- By most accounts, Roma first arrived in ing from case studies which provide a more Romania's historical provinces of Wallachia and detailed understanding of the interlocking mech- Moldavia toward the end of the eleventh century anisms of Roma poverty. (Crowe 1991). Initially free to pursue their crafts Poverty in Romania, and that of Roma in par- and trades, by the fifteenth century Roma slavery ticular, is related to interconnected factors, was institutionalized in the Romanian provinces, including inherited policies from past regimes, lasting well into the nineteenth century (Panaites- the fiscal constraints associated with the transi- cu 1941; Gheorghe 1983; Beck 1989; Crowe 1991 tion process, policy design, and aspects of exclu- 1994). Romanian rulers brought large numbers of sion within society. Romania faces these issues Roma slaves back from various military cam- within an international environment concerned paigns.2 At the turn of the sixteenth century, the with human rights and minority protection, par- Romanian provinces fell under the Ottoman ticularly in the context of Romania's candidacy Empire. During this period, the condition of both for EU accession. This chapter examines the situ- slaves and the Romanian peasantry deteriorated ation of Roma in Romania at the nexus of these further. Because Roma had unique skills as arti- converging factors. It begins with a discussion of sans, craftsmen, and metallurgists, laws were the historical setting. The second section explores enacted to ensure that they would remain slaves.3 the diversity of nine Roma communities analyzed Slaves were generally treated poorly, and cases of as case studies. The third section examines access torture and death were not uncommon (CED- to social services for Roma and the final section IME-SE 2001). discusses social and ethnic relations between Throughout Europe, the Enlightenment peri- Roma and others in Romania. od of the nineteenth century brought about a change in attitude toward Roma and minorities FROM SLAVERY TO CEAUCESCU in general. By the middle of the century, a num- The history of Roma in Romania is particu- ber of prominent owners freed their slaves. Slav- larly dark and difficult, characterized by ery was finally abolished in the 1850s and 1860s. enslavement until 1856, repression and extermi- However, the situation of Roma did not improve nation during the Holocaust, and forced assimi- appreciably after the abolition of slavery, and 68 The Diversity of Roma in Romania many fled. This exodus was initially stimulated tional Roma occupations were declared illegal by fears of re-enslavement, and subsequently (Gilberg 1974; Beck 1985); many Roma were relo- continued due to deteriorating socioeconomic cated;5 and cultural expression was suppressed conditions (Crowe 1994). Of those Roma who through bans on folk music and the use of the stayed, few were given land, and those who did Roma language (CEDIME-SE 2001). Roma were receive land often lacked the skills to cultivate it also often subject to persecution by police and effectively. During these times, Roma were local officials (Zang and Levy 1991). While poli- engaged in occupations ranging from metalwork- cies aimed at settling Roma by providing them ing and carpeting to bottle-collecting and begging with housing, education, and jobs did lead to (Zamfir and Zamfir 1993b). Others, unable to find overall improvements in their living standards, any other means of survival, offered themselves deteriorating economic conditions during the for resale to their old masters (Hancock 1997). final years of the communist regime led to the The redrawing of boundaries following emergence of widespread unemployment and World War I brought a large new, mostly Hun- poverty. On the margins of a rapidly changing garian, minority population to Romania. The society, some Roma began to turn to illegal means share of minorities in Romania's total population for survival, perpetuating societal stereotypes increased from 8 percent to nearly 30 percent after and hostility (Zamfir and Zamfir 1993b). the war, significantly altering the ethnic composi- tion of the state (Livezeanu 1995). Of this number, The Transition Period and Beyond less than 1 percent were estimated to be Roma For Romania's minorities, the overthrow of (Crowe 1991). Although agreements signed by the Ceaucescu regime in 1989 brought the poten- Romania following the war included measures tial for new economic and political opportuni- for the protection of minority rights, these were ties. Over the past decade, however, very few not implemented, largely because of the assimila- Roma have been able to take advantage of them. tion policies of the new government. The Depres- The particularly acute economic decline in sion of 1929 were followed by an increasingly Romania led to rapidly falling living standards nationalist and oppressive period, accompanied for the entire population. Roma have been dis- by increased prejudice against Roma. proportionately affected by trends of rising As elsewhere in Europe, conditions for Roma unemployment, growing poverty, shrinking in Romania deteriorated significantly with the rise social assistance, as well as limited access to of fascism and the onset of World War II. Between housing, education, and health care. As dis- 1941 and 1942, under the fascist Antonescu cussed in Chapter Two, the share of Roma who regime, an estimated 25,000 to 36,000 Roma were are poor is more than twice as high as that of expelled and transported to camps in Transdneis- non-Roma.6 The deterioration of Roma living ter. At least half died of cold, starvation, and dis- conditions has been exacerbated by entrenched ease (Crowe 1991).4 From 1944 to 1947, under the patterns of discrimination, prejudice, and inci- increasing influence of the Soviet Union, many dences of ethnic violence (Cartner 1994; ERRC minorities were promised improved rights as a 1996; OSI 2001).7 part of Stalin's efforts to use "the national minori- The situation of Roma in Romania has attract- ties as a means for undermining anticommunism ed particular attention in part because they con- in Romania" (Crowe 1991). Initially, many Roma stitute the largest absolute population Europe. and members of other minorities joined the According to the 1992 census, less than one-half a Romanian Communist Party. million Roma live in Romania. Unofficial esti- Policies toward Roma during the socialist era mates are much higher. For example, Zamfir and were largely assimilationist, many Roma farmers Zamfir estimated that in 1993 the Roma popula- and nomadic Roma were forced into employment tion was just over 1 million (or 4.6 percent of the in agricultural collectives and heavy industry. total population), a figure subsequently revised These efforts continued through the 1970s. Tradi- to 1.5 million in 1999 (Bárány 2002). 69 Roma in an Expanding Europe: Breaking the Poverty Cycle DIVERSITY OF ROMA SETTLEMENTS area in southwestern Romania which was hard A qualitative study of nine case studies of hit by restructuring in the 1990s. Valcele, Ciopeia, contrasting communities was undertaken to doc- Iana, and Nadrag are all located in rural areas. ument the diversity of Roma social and economic Rural Roma communities frequently lack conditions in Romania (Box 4.1). The sites were basic infrastructure and utilities such as paved selected for their diversity along a number of roads, running water, electricity, and telephone dimensions, including urban and rural locations, lines. In urban areas, communities are frequently ethnic and religious composition, income sources, ghetto-like, located in distinct neighborhoods, and economic opportunities, socioeconomic sta- and situated on the periphery of cities or towns. tus, and levels of political participation (Table In Sf. Gheorghe, Roma live in small houses scat- 4.1). The sites are located in six different counties, tered on hillsides on the outskirts of towns. Other or judets, in Romania: Bucharest, Tulcea, Vaslui, Roma in the area live in two dilapidated blocks of Covasna, Hunedoara, and Timis. Field research in flats which are separated from the other houses these communities was undertaken in 1999. by a concrete wall, nicknamed the "Berlin Wall". As highlighted in the discussion of Slovakia Each of these communities reflects different in the previous chapter, the physical locations of combinations of Roma subgroups. Some localities the Roma communities in rural and urban areas are quite heterogeneous, such as the Zabrauti reflect different degrees of geographic exclusion, community in Bucharest, which contains multi- which in turn are related to other types of exclu- ple ethnic groups. These groups range from quite sion within society. Urban localities in the study traditional, speaking primarily the Roma lan- include the Zabrauti neighborhood in the capital guage (Sporitori), to more integrated, speaking city of Bucharest and a Roma community in primarily, or only, Romanian. The Babadag com- Timisoara, one of the largest cities in Romania. munity has three main Roma groups, the largest The Roma communities of Babadag and the Örko of which are Muslim Roma. In Iana, the majority quarter of Saint Gheorghe (Sf. Gheorghe), are of Roma actively participate in the Orthodox located in smaller, provincial towns, while the church. Nadrag and Sf. Gheorghe are more Iscroni quarter in the small town of Aninoasa is homogeneous communities which consist prima- located in the industrialized Jiu Valley, a mining rily of Hungarian-speaking Roma. The Ciopeia Box 4.1: The Qualitative Study of Roma Communities in Romania In 1999, qualitative fieldwork was carried out in nine different sites across Romania to get a more com- plete picture of living conditions and access to social services. The sites are located in six different districts in Romania: Zabrauti (a neighborhood within Bucharest), Babadag, Iana, Sf. Gheorghe, Valcele, Ciopeia (a vil- lage within the Santamaria Orela commune), Iscroni (a quarter of Aninoasa), Timisoara, and Nadrag. Table 4.1 provides the summary. In an attempt to reflect the diversity in Roma settlements in Romania, the selection of the case study com- munities was based on considerations including geographic diversity; historical factors; variety of Roma sub- groups; income sources and living standards; the degree of integration of Roma in their respective commu- nities; family and social structures; and degrees of political participation and access to information channels in their respective communities. Information for the study was gathered from over 65 in-depth interviews between June and November 1999. Key informants included: educational personnel, such as teachers, administrators and staff; medical staff, including doctors, nurses and hospital administrators; local government authorities; representatives of NGOs; and religious officials. Over 165 interviews were also conducted with individuals (155) and groups (10) of Roma in these localities. 70 The Diversity of Roma in Romania Table 4.1: Main Features of the Case Study Sites, 1999 Case Study Type of Community County/Judet Rural/Urban Roma Subgroups Origins of the Community Zabrauti Bucharest Urban Mixed (including Squatters occupied seven Spoitori, Ursari, abandoned apartment Turkish Roma, and buildings after 1989. The Vatrasi) majority are Roma. Babadag Tulcea Urban Muslim (Turkish) In the 1950s, Roma families Roma were settled in Babadag as part of the housing policies of the socialist government. Sf. Gheorghe Covasna Urban Hungarian-speaking Roma settled in Örko after (Örko quarter) Roma World War II. Timisoara Timis Urban Mixed community Roma settled in this neighbor- of Rudari and hood in the 1950s, from Caldarari Roma neighboring villages, but also from more distant regions. Aninoasa Hunedoara Urban Lingurari and Most Roma migrated to the (Iscroni quarter) Rudari Roma Jiu Valley during the socialist era, where they were employed in the mining industry. Iana Vaslui Rural Lingurari and Roma first came to Iana as Rudari Roma freed slaves following the 1864 rural reforms, and later as veterans of World War I. Valcele (Araci, Covasna Rural Lingurari and Unknown origins. Ariusd, Hetea Rudari Roma and Valcele) Ciopeia Hunedoara Rural Caldarari Roma, Unknown origins. relatively wealthy Nadrag Timis Rural Small community of Roma arrived in Nadrag in Hungarian-speaking the late 1970s, from the north- Roma ern town of Satu-Mare, follow- ing a powerful earthquake. village in Hunedoara is populated primarily by Iana has existed in Vaslui since 1864, when Roma relatively well-off Caldarari Roma. slaves were freed and granted land under rural The communities vary significantly in their land reforms. Roma in Babadag arrived at the end origins and histories. The most recent, Zabrauti, of World War II, while Roma settled in Nadrag emerged after 1989 when Roma occupied desert- and Iscroni, in the Jiu Valley, during the socialist ed apartment buildings in Bucharest as squatters, period, when low-skilled labor was in demand due to housing shortages and deteriorating eco- for the mining industry. nomic conditions. In contrast, the Roma village in 71 Roma in an Expanding Europe: Breaking the Poverty Cycle The nine communities differ substantially in these trades for work in state-run agricultural size, and there is considerable disagreement cooperatives, forestry, and industries such as con- between the official and unofficial population fig- struction, manufacturing, and food processing. ures. According to official data, there are no Roma Although the production and trade of traditional in Iana. However, estimates made by local goods was considered illegal economic activity authorities and service providers in 1999 suggest under the socialist regime, some Roma continued that between 1,200 and 1,500 Roma live in the to work in these trades, either full- or part-time, commune. Similarly, 1992 census data for in order to supplement their income from official Babadag report that Roma account for nearly 10 employment in this way. percent of the population, while unofficial esti- While income sources in the Roma communi- mates put the figure closer to 16 percent. Local ties vary widely, there are some common features. officials in Valcele believe that Roma make up In a 1993 study, Zamfir and Zamfir found that nearly 60 percent of the population, while the income derived from formal wage employment 1992 data indicate only 9 percent. According to constitutes a significantly lower proportion of local sources, Sf. Gheorghe has the largest Roma average income for Roma than for the majority community, between 2,500 and 5,000 Roma, while population. Roma still employed in traditional Nadrag has the smallest at 70 people (Table 4.2). trades tended to have income levels which were higher than the national average.8 A substantial INCOME SOURCES AND ECONOMIC percentage of Roma income derived from a com- OPPORTUNITIES bination of part-time, casual, and self-employed Prior to the socialist period, many Roma work, much of which was conducted in the infor- worked in traditional trades. During the socialist mal economy and at times on the margins of period, and particularly under the Ceaucescu legality. High levels of Roma unemployment in regime, many Roma were forced to abandon the formal sector reflected low qualifications for Table 4.2: Roma Populations in Case Study Communities Roma Population Roma Population (1992 Census) (1999 Unofficial Estimates) Percentage Percentage Case Study Total Number of Total Number of Total Community Population of Roma Population of Roma Population Zabrauti -- 1,000 -- 800 -- Babadag 10,435 969 9.3% 1,700 16.3% Iana 3,850 0 0% 1,200­1,500 30­40% Sf. Gheorghe (Örko quarter) 68,359 886 1.3% 2,500­5,000 3.6­7.3% Valcele (villages of Araci, Ariusd, Hetea and Valcele) 3,500 300 8.6% 2,018 57.7 % Ciopeia 4,000 289 7.3% 290 7.25 % Aninoasa (Iscroni quarter) 5,985 29 0.5% 500 8.4% Timisoara 334,115 2,668 0.8% -- -- Nadrag 3,250 0 0% 65­70 2.1% -- Not available. Sources: National Commission of Statistics for the Census; estimates of local officials and service providers. 72 The Diversity of Roma in Romania jobs. The 1993 study found that 60 percent of Although Roma living conditions and eco- employed Roma were unskilled, and only 2 per- nomic opportunities depend substantially on cent reported having middle or higher-level qual- regional economic conditions, with few excep- ifications. Exclusion and discrimination also limit tions, nearly all Roma in the study sites were poor labor market opportunities. Roma report that and worse off than non-Roma in the area.9 they are generally the last hired and first hired. Bucharest and Timisoara are among the most More recent survey findings suggest relatively prosperous counties in Romania. Sf. Gheorghe is low Roma unemployment rates in Romania (24 also a relatively prosperous town. Babadag, on percent) compared to those in Bulgaria, Hungary, the other hand, has been severely affected by the the Czech Republic, and Slovakia (UNDP 2003). collapse of heavy industry, as has the Jiu Valley, According to this same survey, these lower rates where Hunedoara is located. Consequently, are due in large part to high levels of Romanian unemployment is high among the Roma commu- Roma participation in the informal sector. nities in Hunedoara, including Ciopeia and Around 16 percent of Roma are estimated to be Iscroni. The economy of rural Nadrag depended reliant predominantly on state support for their in large part on a local mechanical factory. After survival. successive waves of layoffs, the factory closed in The situation of Roma in the village of 1998, causing Nadrag's economy to collapse. Val- Ciopeia illustrates employment patterns--shift- cele, Iana, and Ciopeia are largely agricultural ing from traditional trades to formal sector economies. With the exception of Iana, Roma employment during the socialist period, and sub- generally do not own land and depend either on sequently to informal employment during the day-labor or other non-agricultural occupations 1990s. Ciopea is located in Hunedoara judet, a such as small trades or work abroad. former center of heavy industry and mining. Cio- For the communities in the more prosperous pean Roma are largely Caldarari, a traditional counties, such as Zabrauti, Babadag, Sf. Gheo- subgroup which speaks the Romani language. rghe, and Timisoara, the income levels of Roma Relative to other Roma and Romanian communi- vary from moderate to extremely low (Table 4.3). ties in the municipality, Roma in Ciopea have In other communities, such as Valcele and been well off. Prior to the socialist regime, Cio- Nadrag, the interviewers identified nearly all of pean Roma were engaged in traditional occupa- the families as extremely poor. Overall, urban tions including manufacturing bricks and buckets communities have higher and more mixed and selling hand-made soap. During the income levels, while rural communities, with the Ceaucescu regime, many became employed in a exception of Ciopeia, range from low to extreme- large metallurgical factory in the area, or at the ly low. In Ciopeia incomes were higher, and local butcher's shop. Although it was illegal, Roma living standards were found to be largely some workers continued their traditional roles, equivalent to non-Roma. and others began trading merchandise with near- In the study sites, Roma employment was by Serbia. categorized into four main types of occupations: Following the revolution in 1989, many Cio- work abroad, day labor, trade, and subsistence pean Roma were laid off when restructuring occupations. Few Roma in the communities are began at the factory. Since then, involvement in employed in the formal economy, either as sala- trade, employment abroad, and other informal ried workers, or as owners of small businesses or sector activities have intensified. Many Cio- farms. Rather, the most lucrative and steady peans sell and barter secondhand clothing at flea sources of income come from trade and work markets in Hateg and Petrosani (towns 40 kilo- abroad, including day labor such as selling news- meters away), and with neighboring villages. papers. Trade in secondhand clothing, itinerant Still others have emigrated, or have taken on trade in villages, and agricultural day labor, also short-term work in Western Europe, most com- help prevent families from living in extreme monly in Germany. poverty. 73 Roma in an Expanding Europe: Breaking the Poverty Cycle Table 4.3: Income Sources and Economic Stratification, 1999 Primary Income Sources Case Study Rural/ Community Urban Income Levels Moderate-Income Low-Income Extremely Low-Income Zabrauti Urban Mixed Employment1 Day laborers3 Day laborers Small trade4 Waste recycling Begging Babadag Urban Mixed Itinerant trade Itinerant trade Day laborers in villages in villages Sf.Gheorghe Urban Mixed Work abroad Work abroad Day laborers (Örko (Hungary)2 Day laborers Waste recycling quarter) Timisoara Urban Mixed Work abroad Employment Recycling (bottles and (Western Europe) scrap metal) Trade Aninoasa Urban Low- to Mining Trade Recycling (scrap metal) (Iscroni Extremely-Low (second-hand quarter) Income clothing) Iana Rural Low-Income Agriculture Day laborers Retirement pensions Valcele Rural Extremely Low- Day laborers (local (Araci, Income and itinerant) Ariusd, Informal manufacturing Hetea and Gathering and trading Valcele) Ciopeia Rural Moderate to Work abroad Trade (second-hand Low-Income (Western Europe) clothing) Trade Nadrag Rural Extremely Low- Employment Day laborers Income Day laborers Gathering and trading 1. Employment: wage labor in the formal economy. 2. Work Abroad: employment sought in Western European or neighboring countries such as Hungary. 3. Day Labor: employment for predominantly low-income Roma as day laborers in agriculture or other sectors such as con- struction. 4. Trade: employment gained through itinerant trade or more formal trade in local and regional markets. 5. Subsistence Occupations: work including gathering and trading natural commodities such as fern leaves, forest fruits, or mushrooms, or recycling used materials such as clothing or scrap metal. 74 The Diversity of Roma in Romania The poorest families survive on day labor and cent of Roma children between the ages of three informal activities such as recycling waste, used and six participated in preschool, in comparison iron, and other scrap metal. Two families inter- with 60 percent of the whole population.11 The viewed for the study, one in Zabrauti and one in share of Roma who continue beyond compulsory Timisoara, lived exclusively from scrap dealing basic education is also dramatically lower than and both lived in extreme poverty. This type of for the rest of the population. One study found employment can have negative long-term conse- only 7 percent of Roma men and 3 percent of quences. Because of the itinerant nature of the Roma women completed secondary school, com- work, older children were needed to help care for pared to 73 percent of men and 63 percent of younger children while their parents were work- women in the general population (OSI 2002). This ing, and thus were unable to attend school. Child study also reported that there was some evidence labor was also evident in Valcele. Older children, that the proportion of Roma completing second- over 14 years old, worked as day laborers with ary school has increased over the last two their parents. decades. Considerable variation may exist among ACCESS TO SOCIAL SERVICES Roma groups in terms of participation in educa- Roma poverty in Romania and other coun- tion. Zamfir and Zamfir (1993a) found that tries of Central and Eastern Europe is intertwined Roma children are proportionally more likely to with numerous factors, including relatively low regularly attend school if their fathers are educational attainment and access to health care, employed, if they live in mixed rather than pre- social assistance, and housing. This section dominantly Roma communities (60 percent explores access to social services in the case study compared to 33 percent), and if their mothers communities to illustrate general trends and the have had more than eight years of schooling (73 diversity of situations. percent) compared with mothers with no school- ing (21 percent). Education Further data suggests that the proportion of Roma in Romania tend to have higher levels Roma pupils who drop out of school increases of illiteracy and lower levels of educational with age (Table 4.4). According to these data, at attainment than the total population. A 1992 the age of seven, over half of all Roma children study found that, compared to the estimated attend school, either regularly (49 percent) or national illiteracy rate of 2 to 4 percent (Ministry occasionally (7 percent). By the age of nine, school of Education, Romania 1998), 44 percent of Roma attendance becomes the norm, with over 66 per- men, and 59 percent of Roma women were illiter- cent of Roma children enrolled full or part-time. ate in 1992, and an estimated 27 percent of Roma However, between the ages of nine and thirteen, never attended school, or if they had, it was only an increasing number of Roma students drop out for a few years (Zamfir and Zamfir 1993a; 1996). of school. By age fifteen, the proportion of chil- There is also evidence of worsening trends in dren attending regularly decreased by one third recent years. Data from two nationally represen- compared to rates of attendance at age 14. Over tative household surveys found that for Roma, 15 percent of Roma 16 year olds reported that the share of the population that had not complet- they never attended school. ed basic education grew from 36 percent in 1994 The case studies confirmed the sharp to 44 percent in 1998.10 decline in Roma school attendance after the Low preschool attendance is a serious issue in fourth grade. The number of Roma students Romania, which has implications for children's repeating the fourth grade was also higher than future participation in school. According to the for other grades. In the schools in Valcele, for 1992 census, 40 percent of children under the age example, the proportion of Roma students of eight did not attend kindergarten or school. A enrolled in school steadily decreases with age more recent study from 1998 reported that 17 per- (Table 4.5). In the first grade, Roma students 75 Roma in an Expanding Europe: Breaking the Poverty Cycle Table 4.4: Roma School Attendance Trends, National, 1993 Rate of Attendance % Number Age of Students (in years) in Sample Regularly Occasionally Dropped Out Never 7 139 48.4 6.6 8.8 36.3 8 348 44.6 14.9 7.0 33.5 9 318 52.4 13.9 10.4 23.4 10 404 51.4 15.5 14.2 18.9 11 318 48.4 15.1 19.6 16.9 12 458 42.4 15.6 20.7 21.3 13 377 43.3 17.2 21.8 17.6 14 402 41.0 13.2 37.6 8.3 15 442 29.4 9.4 41.5 19.6 16 111 30.5 8.5 45.8 15.3 Sources: Zamfir and Zamfir 1993; 1994. make up 93 percent of all students. By the Local Education Provision: Challenges eighth grade, the percentage of Roma students and Opportunities declines to 53 percent. Many Roma students The organization of education differed across repeat grades. As a result, they tend to be older the study sites. Roma children were educated in than average. The decline in the number of stu- separate classrooms in Zabrauti, Sf. Gheorghe, dents in fourth grade may be related to the and Timisoara, while schools in Timisoara and Sf. greater exposure to more teachers and subjects Gheorghe offered some Roma language training. and the larger number of academic require- Four of the nine communities, Zabrauti, Babadag, ments. The stigma of repeating grades or receiv- Ciopeia, and Timisoara had at least one staff ing remedial instruction may also contribute to member who was able to speak the Roma lan- the high dropout rate among Roma. guage. Basic characteristics of education in the study sites are summarized in Table 4.6. Table 4.5: Share of Roma Students in Valcele, 1999 Total Number Roma Students Over-aged Grade of Students % Roma Repeating the Grade Roma Students 1st grade 123 94 61 13 2nd grade 99 95 30 30 3rd grade 83 73 11 17 4th grade 72 82 3 11 5th grade 58 79 18 20 6th grade 28 64 4 6 7th grade 24 54 3 7 8th grade 15 53 n.a. n.a. Total 502 83 130 104 n.a. = not applicable Source: Rughinis 2000. 76 The Diversity of Roma in Romania Table 4.6: Access to Education in the Case Study Communities, 1999 Special Hours of Roma-speaking Concerns of Teachers Case Study Classrooms Roma Teachers and School Concerns of Roma Community For Roma Language and Staff Administrators Parents and Students Zabrauti Within the No1 Yes Low parental support Discriminatory attitudes local school both material and from teachers and educational non-Roma pupils. Babadag No No Yes Absenteeism; Limited access to the high drop-out rates "Step-by-Step" educational program due to prohibi- tive costs and adminis- trative obstacles to enrollment. Iana No No No None reported Insufficient capacity (for kindergarten). Sf. Gheorghe Within a Yes No Low parental Corruption and inefficiency (Orko quarter) distinct support in allocating support to school students. Valcele (villages No No No Inadequate curriculum; Discriminatory treatment of Araci, Ariusd, Low parental support; and violence against Roma Hetea and Poor attendance; children. Valcele) High drop-out rates Ciopeia No No Yes Aninoasa No No No Low parental support; High costs of education; (Iscroni High drop-out rates limited attention from quarter) teachers. Timisoara Within a Yes Yes Absenteeism; Corruption and incompe- private Within a High drop-out rates tence (in the private educational private educational center). center educational center Nadrag No No No Poor results; Repeating grades 1. One teacher on staff speaks the Roma language but does not offer Roma language instruction. While educational opportunities for Roma dif- recently, however, the government has imple- fer throughout Romania, the Romanian Education mented a number of special educational initia- Law itself does not stipulate the organization of tives targeted to the needs of Roma children (Box general educational practices on the basis of eth- 4.2). Areas with majority or relatively high Roma nic criteria. An exception are schools organized populations may request tailored educational pro- for minority groups, in which all classes are con- grams for their children. For example, in Zabrauti, ducted in the students' mother tongue. More special Roma language classes and a daily lunch 77 Roma in an Expanding Europe: Breaking the Poverty Cycle Box 4.2: Government Education Initiatives Targeting Roma The Romanian Ministry of Education and Research has organized a number of educational initiatives specifically for Roma students. The "Second Chance" project is an experimental project which was organized in 1999 in cooperation with the Center for Education 2000+.1 The program provides the opportunity for young Roma school drop-outs (aged 14­24 years) to complete basic primary and secondary school, together with additional vocational training. Graduation certificates are provided to those who successfully complete a 3.5 year program, giving students the possibility to register for the primary school graduation examination and facilitating access to the labor market. As of 2001, the project had been implemented in eleven schools in six counties of Romania, with 300 stu- dents enrolled in the programs. In addition, 120 teachers have been trained in remedial education and stu- dent counseling, 16 monitors are responsible for monitoring the implementation process, and 10­14 Roma mediators facilitate cooperation between school and Roma communities. Some of the challenges faced by this project include a relatively high turnover of teachers, the need for more vocational apprenticeship opportu- nities in the community, and student drop-outs due to financial constraints. As an extension of this project, the Ministry of Education and Research, in partnership with PHARE, launched the program: "Access to Education of All Disadvantaged Groups, with a Special Focus on the Roma Communities."1 The main objectives include increasing access to quality pre-school education, reducing the number of Roma children who drop out of school early, and providing drop-outs a second chance to com- plete basic education. Financing for the 2002­2004 implementation totals 8.3 million Euro, with 7 million Euro provided by PHARE, and the remainder from national co-financing. Additionally, in 2000 the Ministry of Education reserved a limited number of places for Roma in high schools, vocational schools, teacher training colleges, and universities. Finally, to address the economic con- straints to education, a new initiative is underway which provides free school supplies and a school snack to specific categories of children in need, including many Roma children. Notes: 1. The Center for Education 2000+ was initiated in 1999 as a partnership between the Open Society Foundation Romania, and the Ministry of Education and Research. It aims to provide professional and financial assistance for the elaboration, implementation, and evaluation of educational programs in Romania. 2. See http://europa.eu.int/comm/enlargement/pas/phare/programmes/national/romania/2001. Sources: PHARE 2001; Center for Education 2000+ 2002. program were organized for students in coopera- to buy clothing second-hand, affordable shoes are tion with the Step-by-Step program. While this more scarce. Schools provide free textbooks; how- program is no longer active in Zabrauti, it was ever, most students were required to provide reportedly well received by Roma in the commu- their own notebooks and school supplies, as well nity. Differences in educational practices are also as a daily lunch. A group of Roma women inter- the result of local conditions and attitudes. viewed in Valcele discussed the costs of sending children to school. ECONOMIC CONSTRAINTS. Both Roma and education personnel cite poverty and economic constraints Now, before it is cold and before the earth as significant obstacles to education. Poor parents freezes, they go more. After the winter often cannot afford the necessary school supplies, comes, we won't send them any more-- shoes, clothing, and food. Teachers have we don't have clothing and shoes... There observed that many Roma students report to is no food also. And the children won't school without proper food and clothing. One stay: if we take them, they stay one hour student in Iscroni explained that she had to stay and then they come running home, home from school so that she could wash her because they are hungry. They see the only set of clothing. While many parents are able other children eating, and they aren't. 78 The Diversity of Roma in Romania Another obstacle is the pressure for children DEMAND FOR EDUCATION. The motivation of parents to work and support the family income. Students and their attitudes and expectations toward edu- who continue to attend school and stop working cation also have an important influence on may be scorned by others in the community. school attendance. Many parents resisted send- Young Roma are often expected to work. A moth- ing their children to school, citing reasons such er of seven children in Timisoara explained that as the need for their children to work, fears of older children, particularly girls, are often forced discrimination and maltreatment of their chil- to stay home to care for younger siblings while dren, and a general skepticism about the utility their parents work. of education. Roma from Timisoara indicated that the most common occupations for Roma, I cannot send my children to school such as trade and work abroad, did not require because I have nobody else to help me education. Others did not see how education with the bottles. Look, I have small chil- would lead to higher employment. A young dren--now if I go away, wandering on Roma mother from Iana was skeptical about the roads for three months...with another whether her children's education would improve baby coming... who will take care of their chances of becoming anything more than them? agricultural workers, though she still hoped that they would benefit from education. This pressure to work is particularly intense in the higher grades. A Roma student from What can my children become? It is now Timisoara with an exceptional talent for math as it was before--when could they ever graduated from high school and was admitted to become something? Never. They can pull the Architecture Department of Timisoara Uni- the hoe but what else? May they learn versity. While studying, he also taught math at well... yet they will work the land. the educational center for Roma students. Yet in his second year of university his father forced Some teachers blamed the absenteeism of him to leave school to support the family. Roma students on parents, either because they Although at the time of the interviews he was were not interested in their children's education, selling newspapers in Italy, his teachers reported or because they felt that parents misused their that he was determined to continue his studies. resources and deprived their children of adequate Absenteeism is not limited to poor families. food and clothing, which kept their children out More affluent families who rely on work abroad of school. Lack of support at home for completing as their main source of income often require their homework was noted by a number of teachers in school-aged children to accompany them during Ciopeia and Iscroni. their travels, forcing the children to drop out of On the other hand, many Roma emphasized school. In Timisoara, many of these children have the importance of education for their children. difficulty restarting school upon their return. This In addition to general feelings about the positive is not always the case. In communities such as value of education, many cited practical reasons Iscroni and Babadag, where more well-off fami- for sending their children to school, ranging lies earn most of their income from local business from literacy, which is a basic qualification for or trade, parents were more inclined to send their many jobs, to gaining the required number of children to school and often carefully monitored years in school for driver certification. Teachers their performance. For example, in Iscroni, teach- in Zabrauti, Babadag, Iscroni, and Ciopeia noted ers reported that wealthy Roma families hired that many Roma parents demonstrated an private tutors. These parents considered private interest in their children getting at least a mini- tutoring necessary for their children to achieve a mal education to improve their employment higher level of performance in school. prospects. 79 Roma in an Expanding Europe: Breaking the Poverty Cycle DISCRIMINATION AND EXCLUSION. Another category of Access to Health Care constraints to Roma education relates to discrim- Reliable information about the health status ination and exclusion. Many Roma were reluctant of Roma in Romania and their access to care is to send their children to school due to fear of scarce or nonexistent. Yet there are identifiable prejudice and lack of acceptance in the regular trends. Roma life expectancy is significantly schools. A father of a Roma student from Zabrauti lower than that of the majority population, while described the kind of stereotypes directed against child mortality and fertility rates are higher (Kali- his son by other non-Roma students. bova 2000). Compared with non-Roma, Roma are more likely to suffer from health conditions The other children look at my children directly associated with poor nutrition and living differently: `Look at the Gypsies from conditions. A significant number of Roma have Zabrauti!' Others are calling them `Ghetto limited, or no access to medical care. A recent sur- Boys!' ... So I sent my boy to the school vey indicates that health insurance coverage rates with special classes, because it is closer among Roma are relatively low, with 37 percent and I have heard there is some assistance, of respondents indicating they did not have some free notebooks... I have money now, health insurance (UNDP 2003). but maybe next year I will not have any The case study communities have varying more, so I thought that a notebook and a access to health care, based on their geographic pen would do good... But the boy didn't locations (Table 4.7). Services are necessarily want to go: `I won't go there, Daddy!' more limited in the more remote, rural areas. This `Why?' `Why should they call me "Ghetto is the case throughout the country, for both Roma Boy" and mock me? Am I a "Ghetto Boy"? and non-Roma. In Valcele, the two Roma com- So I sent him to the General School in [the munities are situated at the periphery of the com- neighboring] Sebastian quarter, to the nor- mune on communal grazing land, and in an iso- mal school. lated village. The nearest medical center, in Araci, is two kilometers away. In Iana, some Roma Other parents complained of discrimination reported that they travel over half an hour by cart against Roma and favoritism toward non-Roma to get to the nearest medical services. In both students. Discrimination ranged from teachers cases, lack of telephones in settlements was a bar- ignoring their needs, pejoratively calling them rier to emergency health services. Roma in rural "gypsies," and even violent treatment. In Valcele, areas also noted that physicians are not regularly some students complained that teachers either available. One doctor worked in Valcele, and in disregarded them or physically abused them. Stu- Iana a part-time doctor consults only on Satur- dents also complained some teachers only offered days. help to non-Roma students. One young mother Family planning services are often not avail- from Iscroni complained that her son was held able in local health clinics and women sometimes back from the second grade and said the blame must seek more expensive treatment from gyne- should be placed on the teachers rather than the cologists in the nearest town. These women were parents. reportedly more likely to have unwanted preg- nancies, resort to abortions, or rely on informal They didn't allow my child to graduate information as their main source of information from the first grade... They wouldn't let on family planning. him go. They don't take care of him, but they say it is my fault that I don't take care ECONOMIC CONSTRAINTS. Roma explained that the of him at home... But this is why I send costs of medical services limited their access to care. him to school, to learn there. These costs generally include payments for med- ication, and costs associated with hospital care. In all cases, except for Nadrag, outpatient visits were 80 The Diversity of Roma in Romania Table 4.7: Roma Access to Local Medical Services in the Case Study Communities, 1999 Geographic Availability Use of Case Study Accessibility of of Medical Family Need Expressed by Concerns of Communities Medical Services1 Personnel Planning2 Medical Personnel Roma Respondents Zabrauti High High High Immunizations Infant care Lack of ID papers Babadag High Low High Immunizations, infant care, pulmonary and skin diseases for children Iana Low Low Low Sf. Gheorghe High High Low Immunizations, child Discriminatory (Örko quarter) care, affordable treatment medications Valcele Low Low Low Immunizations, Long distance to (villages of infant care, scabies health services Araci, Hetea and Valcele) Ciopeia High High High Immunizations, tuberculosis Aninoasa High High Low Immunizations (Iscroni quarter) Timisoara High High High Nadrag High High Low High costs (Only private medical care is available) 1. Availability of a health clinic within a reasonable distance (including availability of transportation). 2. Access was considered low when respondents had systematic complaints about reproductive services; the use of family planning was considered high when the researchers encountered cases and/or reports of such practices ­ especially concern- ing contraception instead of abortions. officially free of charge. In Nadrag, where the only Sf. Gheorghe recalled that she had gone to the hos- medical center has been privatized, consultation pital for an emergency appendectomy, but the doc- costs, for adults as well as children, were as much tors refused to treat her if she did not pay 50,000 lei as 75000 lei (US$5). These costs made seeking med- (US$3). A mother from Valcele explained the diffi- ical care prohibitively expensive for many Roma culties her daughter faced obtaining an abortion: families. Despite the absence of official fees, infor- mal charges, particularly for hospital care and spe- My daughter didn't want to have the cialized services, are widespread. A woman from baby... She even went to have an abor- 81 Roma in an Expanding Europe: Breaking the Poverty Cycle tion, but the doctor refused. She was in I do not register gypsies as a family physi- the third month, and the doctor could cian. I do not accept gypsies... They come have done it, if he wanted. And I gave her here and ask for money, ask for medi- money... I did the impossible almost, and cines... They have a lot of nerve. You have I gave her 160,000 lei (around US$10) but to keep an eye on them when they enter the doctor wouldn't do it. Maybe if I had here. I do not think they are poorer than four or five hundred, a million, maybe he other people. They go to Hungary with would have done it. She went once and it business, they probably have more money was in vain. I sent her again, I forced her, than we have. They go by car to ask for she went to another doctor, but this one social aid. refused too. What should I have done-- should I have killed her? May the child live--if it comes, it comes. COMMUNICATION BETWEEN PATIENTS AND SERVICE PROVIDERS. One of the key factors influencing Compounding the prohibitive costs of health access to health care for Roma is the quality of care was a general confusion over the rules of the communication between health care personnel health insurance system, which was introduced and patients. While there are positive examples, in 2000, at the time of the interviews. Under the this relationship is more often characterized by insurance system, all Romanians need to be reg- miscommunication, mistrust, and, in some cases, istered with a family physician in order to be eli- discrimination on behalf of the practitioner. The gible for care. For those working in the formal attitudes and perceptions of both patients and sector, contributions are made through payroll medical staff can have a significant impact on tax deductions, while the self-employed, includ- how health needs are conceptualized and the ing farmers, are supposed to make contributions quality of service delivery. Roma noted that they on their own. The uninsured are to be covered by were treated poorly by health personnel. A young the state budget, with eligibility determined by mother from Sf. Gheorghe complained: registration for social assistance. However, because social assistance covers such a small I don't know why, but some doctors speak share of the poor in Romania, many uninsured so rudely with the people... when I went were not registered and fell through the cracks. there and I saw this, I felt offended and I Health care personnel and patients alike have left. I went and I bought the medicines been confused by the process, and many Roma myself... They said, "Where are you hur- have gone uninsured. rying? Wait", and other people went In Babadag, a doctor reportedly stopped pro- ahead of me, and they offended me. I felt viding subsidized medications to Roma, because like crying... I was waiting there for 2 or 3 she had received written instructions to provide hours, and afterwards they cursed me. So them only to insured patients. The doctor noted I went and I bought the medicines that only one Roma household--a relatively well- myself... it is better to go to the pharmacy. off family engaged in cross-border informal I tell them what hurts me and I ask for the trade--had paid their health insurance contribu- proper medicine. tion (approximately US$20 per month). Some Roma have also been left uninsured due to the From their side, medical personnel frequently increasing choice of family physicians to select perceive Roma as irresponsible patients. Some only low risk patients. A physician in Tirgu Secui- providers felt that Roma parents often give their- esc, a town in Covasna County in the Transylva- infants inappropriate food. Other doctors noted nia region explained: that children were dressed too warmly in the summer and not warmly enough in the winter, leading to pulmonary infections. Some expressed 82 The Diversity of Roma in Romania concern about poor hygienic practices that can will have to buy it. We threatened some lead to skin diseases. illiterate parents with false papers, telling Medical staff also complained about the reli- them that they will have to pay penalties if ability of Roma patients in following their they do not have the children to come for advice. Doctors noted that parents do not follow vaccination. [She displays the false penalty prescribed treatments for their children. A gener- certificate, smiling]. We put a lot of stamps al practitioner in Babadag explained that Roma on it... Sometimes we brought policemen patients did not heed her advice, and took medi- with us in the area to be more convincing. cines irregularly. This doctor no longer trusts And it worked many times. Roma parents to dispense expensive medicines at home, asserting that, "It would be a waste." A doctor in Iscroni relied on the same method Now she believes that Roma should be given as the nurse in Araci, often sending local police treatments only if they are hospitalized and officers with nurses on vaccination campaigns. supervised by medical staff. These doctors indi- Some doctors reported that they threatened par- cated that they prefer to dispense medications ents by telling them that they would restrict their through injections, so that they can control the eligibility for medicines unless their children treatment. were immunized. Until 1993, doctors in Sf. Gheo- Medical staff claimed that Roma refuse to rghe made the distribution of powdered milk for immunize their children. In Zabrauti, numerous infants conditional on the child being immu- Roma children were kept out of school because nized. Some teachers in Örko even admitted lock- they did not have the appropriate immunization ing Roma children in a classroom so that the records. According to physicians, Roma parents nurses could immunize them. refuse immunizations because of a general mis- It is difficult to gauge the degree to which trust of formal medicine. Many Roma mothers these problems are related to distrust of the fear immunizations that induce fever as a side health system by Roma, or other factors, includ- effect. In Valcele, the doctor reported that nurses ing low education levels and economic incen- make repeated home visits to Roma families to tives. A doctor in Zabrauti, for example, claimed immunize children, and are refused for various that many Roma patients come for free medica- reasons. Some parents claim that their children tion. If the medication was not free, she predict- are sick, while others say that their husbands ed, they would resort to more traditional reme- would beat them if their child were immunized. dies. Roma resented these assumptions. In In one particular polio vaccination campaign in Babadag, for example, Roma protested the sug- Araci (a village in Valcele), the nurse recruited the gestion that they sought out "old women's police to accompany her on home visits. After a advice" rather than modern medical treatments. few days, a rumor spread in the village that the "We don't go to old women," they remarked, vaccine was really a sterilization device and the "we go to the doctor." campaign was quickly abandoned. Some health care officials have resorted to Access to Social Assistance tricking their patients in order to persuade them Social assistance cash benefits are an impor- to get immunized. A doctor from Babadag tant source of income for many poor Roma fami- explained: lies. Because of the deep fiscal crisis of the transi- tion period, the availability of social assistance [Roma] have been never willing to have benefits has been severely restricted throughout their children be vaccinated... We have to Romania. Alongside the budgetary constraints, motivate them with methods adequate to the transfer of responsibility for delivering social their values. For instance we told them that assistance benefits to local governments has left a vaccine is very expensive, 60,000 lei, and benefits unpaid in many of the poorest areas. we administer it for free now, later they Local governments have become caught in a 83 Roma in an Expanding Europe: Breaking the Poverty Cycle vicious circle of impoverishment. The poorest ties, benefits were either paid irregularly or municipalities which have the greatest need for stopped altogether. Benefits were available on a social assistance--are least able to pay them. In regular basis in Zabrauti, Iscroni, Timisoara, and response, local officials have used their own dis- Nadrag, and only intermittently in Sf. Gheorghe cretion to adopt various coping strategies, such as and Ciopeia. MTSAB benefits were discontinued limiting coverage of benefits by creating addi- for over three years in Iana, two years in tional eligibility criteria, decreasing the level of Babadag, and one year in Valcele (Table 4.8). benefits, or ceasing payments altogether (World In Babadag, the number of households receiv- Bank 2000d). ing social assistance dropped dramatically from As a whole, coverage of social assistance has 1,207 families in 1995 to 75 in 1998, due to the dropped to extremely low levels in Romania.12 budget constraints. The mayor estimated that The country's main monthly cash benefit pro- approximately 75 percent of those who lost bene- gram for the poor, the Means Tested Social Assis- fits were Roma. Even though the households may tance Benefit Program (MTSAB) reaches very few have been eligible according to the national legis- households. In some of the case study communi- lation, local officials limited eligibility based Table 4.8: Access to Social Assistance in Case Study Communities, 1999 Means-Tested Case Study Community Social Problems Income Support Eligibility Criteria Zabrauti Sub-standard housing Available None Extreme poverty Babadag Extreme poverty Discontinued For school scholarships, recipients must submit a passport as proof they have not traveled abroad. Iana Extreme poverty Discontinued Sf. Gheorghe Sub-standard housing Intermittent None (Örko quarter) Extreme poverty Valcele (villages of Araci, Sub-standard housing Discontinued Ariusd, Hetea and Valcele) Extreme poverty Ciopeia Intermittent Social worker discretion Aninoasa (Iscroni quarter) Sub-standard housing Available None Extreme poverty Timisoara Available School attendance; proof of land ownership; active job search. For educational grants: proof of unemployment. Nadrag Substandard housing Available None Extreme poverty 84 The Diversity of Roma in Romania upon additional criteria, because they could not Roma woman from Babadag explained: "I afford to finance all eligible beneficiaries. One received benefits twice... In Cernavoda they pay official believed that many Roma applications it every month, why is it that we don't receive this were rejected on illegal grounds. "Many of these money? In Medgidia they give it every month, families have been denied benefits on the why don't we get it?" Her comments also reflect grounds that the head of the household is able to a common perception among Roma of the dispar- work," he remarked, "or because they have a tele- ity in the availability and levels of social assis- vision, or a pig, or because they live with their tance between municipalities. In Timisoara, a parents... or because somebody has some infor- Roma woman explained the difficulty of her situ- mation that they are involved in trade." ation: she no longer received benefits because she In some cases, additional eligibility criteria did not send her children to school. However, she may create positive incentives for beneficiaries, if could not afford to send her daughter to school they are--for example--required to work or send due to reductions in her benefits. While she col- their children to school. In other cases, additional lects bottles in order to support the family, her rules may exclude beneficiaries who are unable to older daughter must take care of the younger work, or lack the necessary documentation for children and subsequently cannot attend school. receiving benefits. In Timosoara, the city council and mayor restricted access to benefits by adding Access to Housing a mandatory work requirement of five days per An increasing number of Romanians, both month. Children were also required to attend Roma and non-Roma, risk exclusion from hous- school, and benefits were only paid to those who ing as a result of privatization and housing resti- could prove permanent residence in Timosoara. tution, changes in the legal status of land, and Additional restrictions included asset tests, which declining incomes. Many Roma in Romania live excluded households which owned land and ani- in integrated areas, while others live in urban mals. The number of aid recipients dropped from slums, such as Zabrauti in Budapest, or rural set- 3,000 in 1995 to 306 in January 1998. tlements, such as Iana and Nadrag. Many urban Similar strategies were adopted for the pay- and rural Roma communities suffer from a lack of ment of school scholarships for low income fami- infrastructure and services. Poor infrastructure-- lies. These education benefits are paid to poor bad roads, lack of water or sewage systems, and households to cover school-related expenses. In absence of telephone lines are more pronounced Babadag the school director devised a method to in rural areas. Internal migration from rural to exclude some Roma families which had incomes urban communities has contributed to the expan- from informal trade. He asked the Roma parents sion of urban settlements and to the growth of an to bring their passports to school to prove that informal system of supplementary social security they had no visas, and consequently no incomes where rents, gas, and electricity bills are unpaid from trade. The director reported that this but evictions and stoppages do not immediately reduced the number of Roma applicants substan- occur (Save the Children 2001a). tially. Other Roma children were excluded from Conditions in Roma neighborhoods are fre- receiving benefits because they had not passed quently poor, with problems of extreme over- certain exams. In one of the Timosoara schools, crowding and a lack of social services. According the principal decided to ask the parents to submit to 1998 household data, Roma living quarters in a formal proof of unemployment and other docu- Romania are--on average--20 percent smaller ments in order to apply for benefits. than those of non-Roma, although Roma house- Roma expressed dissatisfaction with the hold size is significantly larger.13 Lack of access to reductions in social assistance, and particularly utilities, such as water, gas, electricity, and public with the disparities across localities. Roma in services such as waste collection, is a significant areas where benefits were paid irregularly were problem in many neighborhoods. In 1998, only 24 aware that benefits were being paid elsewhere. A percent of Roma in Romania had access to public 85 Roma in an Expanding Europe: Breaking the Poverty Cycle water facilities within their housing units, com- opposed by the city council. Roma from Valcele pared to 46 percent of the total population. Simi- were disappointed that the mayor had broken her larly, 34 percent of Roma households have no toi- electoral promise to provide them with land. In let facilities (indoor or outdoor), in comparison Zabrauti, UNDP, in cooperation with the local with 28 percent of total households. mayor's office, initiated the legal transformation The absence of quality and affordable hous- of four buildings, which residents had occupied ing was an issue flagged by Roma in almost all of illegally as squatters, into official public housing. the case site communities. In Zabrauti, Nadrag, While the project was legally approved, imple- and Iscroni, where most residents live in public mentation has proven difficult. In addition to housing owned by the city, people mentioned these bureaucratic obstacles, some tenants have poor maintenance and the lack of investment in had difficulty meeting rental requirements due to housing. In Nadrag, residents reported that they lack of identification papers or criminal records, were unable to pay their rents or electricity costs, and faced evictions. and houses are often in extremely poor repair. As In Sf. Gheorghe, where the community center one young mother reported: functions as a homeless shelter and a temporary housing facility, the mayor proposed a compre- The toilets are broken... Yesterday our hensive urban renewal strategy for the Roma administrator called us to clean every- neighborhood. The mayor of Aninoasa intends to thing, because, he said, an inspector is move all of the inhabitants of the Roma quarter of going to come... He is a very kind man, he Iscroni to another quarter situated at the periph- helps us. He provided us with a hose to ery of the town in order to build a new civic cen- clean here, because it was such a misery, ter in a central location. While the move would you couldn't count the dirt piles here... entail an improvement in housing conditions, Because we have only two toilets, but they Iscroni residents oppose the initiative, most likely are broken, and the misery spills out. because of the undesirable location. In Aninoasa, the Roma neighborhood consists SOCIAL AND ETHNIC RELATIONS of dilapidated barracks with outside water taps and nonfunctioning public toilets. In the Örko Relations between Roma and the neighborhood of Sf. Gheorghe, most people do Majority Population not own the land on which their houses are built The frequency and quality of contact between and the neighborhood has outdoor plumbing that Roma and non-Roma varies, depending in part poses a serious health risk. In the rural communi- on the geographic location of Roma communities, ties of Valcele and Iana, young couples often but also on the socioeconomic status and the age build homes illegally, due to the scarcity of of the settlement. In general, the relationship affordable land. Illegal housing is also a problem between Roma and non-Roma has been charac- in Zabrauti, where houses often have outside terized by miscommunication and mistrust. water taps, improvised electrical installation, and Equally, aspects of Roma society may contribute lack central heating or gas connections. to their isolation, as well as to popular stereo- Because of the legacy of state-provided public types and myths. Indeed, Roma social exclusion housing during the socialist period, expectations may be traced in part to the nature of their inter- among the population are high. Most Roma action with non-Roma, and the mutual construc- expect that local governments will address hous- tion and negotiation of boundaries between com- ing shortages and improve the quality of existing munities. houses. Local government responses in the sites The strongest evidence of continuing mistrust have varied. In Valcele, the mayor's office pro- is the lack of geographic integration of Roma and posed granting land to Roma to build new hous- non-Roma communities. With the exception of ing on communal grazing land, but the plan was Iana, Roma and non-Roma in the study sites gen- 86 The Diversity of Roma in Romania erally do not live together. Rather, Roma tend to uals and circumstances. In some cases, Roma constitute relatively segregated pockets located reported encountering sympathetic officials who next to non-Roma neighborhoods. In Iana, which recognize and attempt to accommodate the par- was first settled in 1864, Roma are more integrat- ticular needs of Roma, but more often they ed, particularly in comparison to more recent and reported encountering indifference, hostility, heterogeneous communities, such as Zabrauti in intolerance, and corruption of officials who are Bucharest. Nevertheless, despite the high level of already strained by inadequate resources. integration between these groups, Roma are still One of the dominant stereotypes about pejoratively called "Gypsies." In Iana, there were Roma is that they are the "undeserving poor." a number of mixed marriages, while they were The pervasiveness of this view was evident in found to be rare elsewhere. discussions with local officials. In Hetea, a Roma In other communities, relations between village in Valcele, the Romanian administrator Roma and non-Roma are reportedly much more of a Dutch aid program described Roma as strained, and reflect a high degree of social dis- "thieves" and "lazy." In Babadag, local officials tance based on a lack of contact between ethnic were reluctant to pay social assistance to Roma groups. Such isolation starts young. Roma chil- citing similar reasons. The Mayor himself dren in Zabrauti are not fully integrated into claimed that Roma do not work but "stay in the mainstream classes and are still required to pubs all day long," grow nothing on their land, attend "experimental classes" which are not only and are overly reliant on trade. Not everyone segregated, but are also relatively isolated within ascribes to such beliefs. The social assistance the school building. The teacher reported that coordinator in Babadag denounced the widely Roma students often do not mingle with other held conception that "Everything bad that hap- children during breaks. pens is the gypsies' fault," arguing that special The socioeconomic status of Roma also has a programs are needed in order to improve the sit- bearing on interethnic relations, as well as relations uation of the Roma. between Roma groups. Wealthier Roma are more Relations with local police were frequently integrated, often adopting some of the cultural described as strained. In Zabrauti, Roma reported practices of Romanians. For example, in Babadag, frequent police raids and fines for squatting. Over despite the high proportion of Muslim Roma, and time, relations with the police have evolved from lingering adherence to traditional dress and cus- what the Roma perceived as arbitrary, punitive, toms, most Roma names are Romanian rather than and often violent interventions, to the total Muslim in origin. Furthermore, most Roma absence of a police presence. women, particularly wealthier ones, get married in Finally, the success of a number of outside traditional white bride's dresses customary in development initiatives was threatened by perva- Romanian wedding ceremonies. Some Roma in sive mistrust and suspicion between program Ciopeia are relatively well off and have two-story administrators and the community. This was houses and more expensive cars. In general, Roma most evident in Sf. Gheorghe where Roma inhab- in Ciopeia reported fewer tensions between ethnic itants accused a priest and teachers who were groups. However, difficulties remain. As one involved in a project of stealing donations. On the Romanian respondent remarked, "We have no ene- other hand, program administrators accused mies, but also no friendships with the gypsies." Roma of misusing aid and failing to abide by the Examples of tensions and discrimination were goals of the programs. The examples of these pro- reported involving access to running water, land grams highlight the importance of cooperative distribution, and to community celebrations. relationships between the state and local service providers. For example, in Zabrauti coordination Relations with Public Officials and NGOs between the assistance program and the local Relationships between Roma and public offi- administration was effective and facilitated the cials are mixed, depending largely upon individ- project. The Mayor of the fifth district of 87 Roma in an Expanding Europe: Breaking the Poverty Cycle Bucharest, which governs Zabrauti, was a partner eration of the Roma, the Roma Party, and the in the program and allocated resources. Local Roma Union. Despite these improvements, Roma officials also managed to secure the doctor's remain underrepresented in local political institu- cooperation to improvise medical records in tions. While there are currently no Roma Sena- order to allow Roma children to attend the local tors, during the November 2000 elections, a mem- kindergarten. ber of the Roma Party was elected to the Chamber of Deputies. CURRENT POLICY DEVELOPMENTS Both Roma and non-Roma organizations Over the last decade, significant efforts have have played an important role in discussions been made by government, political parties, and and agenda setting with respect to Roma affairs, nongovernmental organizations toward improv- particularly concerning the development of a ing the condition of Roma in Romania. While national strategy for Roma. Approximately 150 many of these efforts have been improvised and NGOs in Romania are devoted to promoting the uncoordinated, there is evidence that efforts are rights and interests of Roma and pressing for beginning to draw on more than a decade of proj- change in minority policies.16 However, the ect learning and experience in order to implement small size and marginal resources of many of a more coherent, systematic approach. The adop- these NGOs limit their influence. tion of a national strategy to improve the condi- tion of Roma in 2001 reflects such efforts. The National Strategy for Improving the Since 1989, Romania has ratified the main Condition of Roma international documents addressing racial and In April 2001, the Government adopted the ethnic discrimination. In November 2000, Roma- "National Strategy of the Government for Improv- nia became the first EU candidate country to enact ing the Condition of Roma," which aims at stimu- general antidiscrimination legislation. In April lating the participation of Roma in the economic, 2001, a law on public administration was enacted social, educational, and political life of society allowing for the use of minority languages in through their involvement in sectoral assistance areas where minorities constitute 20 percent of the and community development programs, as well as population.14 The use of non-Romanian lan- through programs for the prevention of institu- guages in criminal and civil proceedings is also tional and societal discrimination. constitutionally guaranteed, although this does The duration of the strategy is intended to be not always happen in practice. 10 years (2001­2010), with the first four-year Specialized institutions dealing with minori- action plan focusing on the following areas: ties have also been set up. In 1993, the Council for administration and community development, National Minorities was established as a consulta- housing, social security, health, economy, justice, tive body of the Romanian Government. A Depart- child welfare, education, culture, communication, ment for the Protection of National Minorities was and civic involvement. The Strategy is organized, established in 1997 within the Prime Minister's coordinated, and implemented through a series Office, including an Office for Social Integration of of structures including the establishment of a Roma.15 Following the 2000 elections, these offices Joint Implementation and Monitoring Commit- were relocated to the Ministry of Public Informa- tee, Ministerial Commissions for Roma, County tion and renamed the Department of Inter-Ethnic Offices for Roma, and local experts for Roma Relations, and the National Office for Roma. issues. At the parliamentary level, there are standing commissions on minorities and human rights Progress and Challenges which function both within the Senate and the The adoption of the National Strategy marks Chamber of Deputies. After 1989, a number of a significant milestone in the official policy Roma political parties were established, includ- approach toward improving the situation of ing the Democratic Roma Union, the Ethnic Fed- Roma. One of its greatest strengths is that the 88 The Diversity of Roma in Romania Box 4.3: The Partnership Fund for Roma: Pilot Projects Between January 2001 and April 2002, the Partnership Fund for the Roma,1 a grant-fund made available by the European Commission's PHARE programme, provided 900,000 Euro to support 40 pilot projects for improving the situation of Roma communities in Romania. In keeping with the principles and goals of the National Strategy, the main aims of these projects were to test policy initiatives of the Romanian government, and to support partnerships between local authorities and Roma organizations. These projects tested many of the action items included in the National Strategy, such as new schools for Roma children, vocational and pre-school projects, the renovation of apartment blocks, income-generating projects, Roma-managed farms, an environmental project, and health projects. All projects were subject to on-going monitoring and evalua- tion and were offered technical assistance by the local management organization (The Roma Communities Resource Center--the RCRC--in Cluj Napoca) as well as by the PHARE team. A recent evaluation found these pilot projects to be generally successful. Many provide useful examples of positive partnering between a range of public institutions (schools, town halls, regional inspectorates, and prefectures) and Roma organizations. These experiences also grant valuable insights into the specific kinds of challenges and misunderstandings that arise in such collaborative attempts. Some PHARE staff and Roma NGO leaders have raised concerns about the longer-term sustainability of individual projects, as well as the degree to which the lessons learned will be incorporated into policy affecting the situation of Roma. The orig- inal Fund has been amalgamated with the RCRC and in 2002, they received a new grant fund from PHARE Romania (The Civil Society Development, Improvement of Roma Situation Fund). Note: 1. The Partnership Fund for the Roma was a component of a larger, PHARE funded project called The Improvement of the Roma Situation in Romania. The two main aims of the project, developed by leading Roma and government representatives as well as the EC Delegation were to provide technical assistance to the Government of Romania for the development and implementation of a strat- egy for the improvement of the Roma situation and the implementation of the Partnership Fund. Source: Murray 2000. strategy was elaborated with the participation of currently able to contribute to furthering the delegates from the Ministries as well as Roma Strategy's goals. The initiation of 40 pilot projects leaders and representatives from a variety of non- has provided valuable experience and lessons governmental organizations. The strategy's prior- about the opportunities for collaborative efforts ities are widely seen by Roma and non-Roma between local officials and Roma representatives leaders as reflecting those articulated by Roma (Box 4.3). representatives. While these steps represent important The National Strategy for Roma is to be car- progress, much work remains in further elaborat- ried out through a series of institutions estab- ing and strengthening the institutional infrastruc- lished at multiple levels of government, each of ture and implementation of the Strategy. While a which is intended to include both government number of structures have been put in place, the and Roma representatives. This includes a Roma degree to which they are active varies consider- Office within the Ministry of Public Information, ably. There is also a general lack of clarity about which will have judet-level representation, com- the specific roles and responsibilities of represen- missions within ministries, to address sectoral tatives at different levels. Further, questions have policy, and a government-level Joint Implemen- been raised about the criteria used for appointing tation and Monitoring Committee to oversee county and local representatives. The Roma Party organization and implementation of the strategy. has developed a close relationship with the Social While significant progress has been made in Democratic government elected in 2000. While establishing these institutions, there is consider- this has granted the Roma Party greater influence able variation in the degree to which they are over Roma affairs it has also prompted criticisms 89 Roma in an Expanding Europe: Breaking the Poverty Cycle about the politicization of Roma appointments in NOTES the public administration. Finally, n ongoing con- 1. This chapter is based on a qualitative study cern is the lack of systematic monitoring, evalua- of nine Roma communities in Romania in 2000 tion, and enforcement of the strategy. conducted by Cosima Rughinis and Marian Preda. CONCLUSIONS 2. For example, in 1461­2, the Wallachian This study of Roma in nine communities in ruler Vlad IV Tepes (the Impaler) brought Romania demonstrates that Roma face a number 11,000--12,000 Roma from Bulgaria, while in of interlinking challenges. Each of these commu- 1471, the Moldavian ruler Stephen the Great nities has faced varying degrees of labor market reportedly brought 17,000 Roma to use as slave exclusion, limited access to education, health, labor (Crowe 1994). social assistance, and housing. The case studies 3. Such laws were passed to restrict the free- demonstrate that the nature and extent of this dom of movement of Roma slaves, to forbid inter- poverty and exclusion reflect both localized geo- ethnic marriage, and to discourage escape at- graphic and economic conditions, and the consid- tempts, and the illegal trade in slaves (Crowe erable diversity among Roma populations. These 1994). challenges point to the need for integrative poli- 4. By some accounts, official policy was not to cies that can be adapted to local circumstances. annihilate the Roma per se, but to ensure that The cases showed variation in relations they were removed from tainting the Romanian between Roma and non-Roma, from integration nation. However, according to the War Crimes to exclusion. They also suggested a relationship Commission established by the Romanian Peo- between geographic and economic exclusion. ple's Court, 36,000 Roma died during the war, Rural communities lack basic infrastructure and constituting the highest absolute number of utilities, and have more limited economic oppor- Roma deaths of any European country. Those tunities and access to education and health care who survived are reported to have lived in rela- than urban communities. Moreover, while region- tive freedom, with some even serving in the al economic conditions were found to be influen- Romanian national army (CEDIME-SE 2001). tial on Roma living conditions and economic 5. Ceaucescu's "systemization" program up- opportunities in general, Roma in all localities rooted many Roma and non-Roma communities. tended to be worse off than their non-Roma coun- 6. Based on a poverty line of US$4.30 per capi- terparts. Few Roma were employed in the formal ta per day. economy, rather, the majority relied on other 7. For example, according to an "Ethno- sources, including trade and day labor. barometer" survey taken in 2000, it was reported Access to social services is hampered by a that 38­40 percent of non-Roma would prohibit variety of interrelated factors. Persistently low Roma from settling in their country; 23 percent of levels of educational attainment reflect difficul- ethnic Romanians and 31 percent of ethnic Hun- ties in accessing education due to economic con- garians would refuse to accept Roma in their city, straints, discrimination by educators, as well as town, or village. Another recent poll (2000) found Roma attitudes toward education. Relations that 67 percent of the population feel resentment between Roma and non-Roma were found to play toward the Roma (OSI 2001). an important role in perpetuating patterns of 8. According to the 1993 study, only a small exclusion. Miscommunication and distrust on proportion of the Roma population (7 percent of both sides compound other forms of exclusion. adult men) practiced traditional trades, while a Efforts such as the training of Roma mediators to larger proportion of the population acquired facilitate "back to school" programs represent a "modern" skills (35 percent of adult men). The positive step toward improving the communica- large majority of the sample--58 percent of men tion between Roma communities and service and 85 percent of women--reported having no providers. 90 The Diversity of Roma in Romania trade (either traditional or modern) (Zamfir and 13. Data are from the Romanian Integrated Zamfir 1993b). Household Survey, 1998. 9. The descriptions of income levels are based 14. Estimates indicate that the Roma popula- upon the observations of the field researchers and tion is unlikely to reach 20 percent in most, or all, provide only a rough indication of relative wel- territorial administrative units (OSI 2001). fare in the different communities. 15. Under the supervision of the Department, 10. Data from the 1998 Romania Integrated a limited number of initiatives were undertaken Household Survey. within the Framework of the Strategy. The Min- 11. This study, cited in Save the Children istry of Labor and Social Solidarity formulated a (2001), was conducted by the Open Society Insti- special program for Roma involvement in local tute and the Central European University Centre departments for labor and social protection. In for Policy Studies for their country report on parallel, the General Police Inspectorate imple- Romania. mented programs for preventing violence in 12. Since this study was conducted, social localities and communities with the participation assistance reforms have improved its effectiveness of Roma organizations and associations (UN and coverage. Positive effects for Roma have been 2001). noted including: (i) greater access to ID cards; (ii) 16. A list of Romanian Roma NGOs, compiled improved targeting to the unemployed through a by the Resource Center for Roma Communities, workfare requirement; and (iii) increased social is available at: . capital generated by the participation of Roma and non-Roma in workfare activities. 91 Chapter Five: PROJECT EXPERIENCE IN HUNGARY S ince 1989, more policy and project activity been both faster and more successful than most. related to Roma has taken place in Hungary Hungary has been among the leading countries than in any other country in Central and in the EU accession process. Second, Hungary has Eastern Europe. Considerable research has been historically had a greater involvement in minori- conducted; a wide range of Roma-related NGOs ty issues than its neighbors, because of the large have been set up; and numerous projects and number of Hungarians who live as minorities in pilot projects have been implemented. Successive other countries. Third, the growth of civil society governments have played an active role in policy has been more rapid in Hungary than in other setting. Nevertheless, Roma remain among the countries. most marginalized groups in Hungary. As Chap- ter Two illustrated, their socioeconomic condi- EU Accession tions remain well below the national average. The EU accession process has accelerated the There is still room for improvement in the devel- adoption of policies related to minorities in Hun- opment of effective policies for Roma and inte- gary. While the process has influenced develop- gration into Hungarian society. ments in neighboring countries, the impact came This chapter marks a departure from the earlier in Hungary. Integration into the EU has country studies of the two previous chapters, been a key goal of Hungary since 1990. In Decem- focusing on the experience of projects and poli- ber 1991, Hungary and Poland were the first coun- cies. It explores reasons why Hungary has seen a tries in Central and Eastern Europe to sign associ- generally higher level of activity on Roma issues ation agreements with the EU. In March 1994, and minorities policies than other countries in the Hungary became the first of the transition coun- region, and examines some of the project experi- tries to submit an official application for member- ence close up. It concludes with an assessment of ship. Hungary has long been considered a fron- some of the lessons learned from the experiences trunner for accession, given its relatively high of selected projects, and points to future direc- level of development. Formal negotiations com- tions in national policy. Looking to the future is menced in 1998 and were concluded in December important because of the large and growing share 2002.1 Following the EU's decision to offer admis- of the Roma population in the country--estimat- sion to ten accession countries, including Hun- ed at between 4 and 6 percent. The significant size gary, at the Copenhagen summit in December of the Roma population in Hungary, and the 2002, these countries are expected to be able to marked deterioration of its living standards dur- take part in European Parliament elections in June ing the transition are important factors which 2004 as members, following ratification of the continue to motivate government attention. accession treaty by the European and national parliaments.2 A REGIONAL FRONTRUNNER EU policy relating to ethnic minorities, and There has been a far greater proliferation of Roma in particular, informs the political criteria Roma policies and programs in Hungary than in for accession under the subchapter on "human other countries. Why? In the first place, Hun- rights and the protection of minorities" that was gary's post-transition development process has adopted at the 1993 Copenhagen European Coun- 92 Project Experience in Hungary cil. In its 2002 Regular Report on Hungary's wake of World War I led to the relocation of mil- progress toward accession, the European Com- lions of former citizens, mostly Hungarians, to mission concluded that Hungary has adopted other countries. Following the division of the most of the major international legislation on Austro-Hungarian Empire and the 1920 Treaty of human rights and has developed a wide-ranging Trianon, Hungary lost nearly 70 percent of its pre- institutional framework for the protection of vious territory, and 60 percent of its total popula- minorities. However, it lacks a unified law against tion, including 28 percent of the Hungarian discrimination. Current antidiscrimination provi- speakers and the large bulk of its minorities. sions are fragmented and are included in laws Hungarians continue to make up substantial regulating different fields--such as employment shares of the population in a number of neigh- and education (Kádár et al. 2001). Since the 2001 boring countries: nearly 10 percent in Slovakia Regular Report, Hungary has continued to make (2001), 7 percent in Romania (1992), and 4 percent progress on the short term Accession Partnership in Austria and Yugoslavia (1991). priority: the implementation of the government's Policies toward Hungarian minorities abroad Medium-Term program for the integration of do not necessarily translate into the full realiza- Roma. According to the 2002 Regular Report: tion of domestic minority policies. For example, Roma, who are classified as an "ethnic minority," The institutional framework [of the Medi- rather than a "national minority," were not origi- um-Term program] has been further nally covered under early drafts of the Minorities strengthened and a new monitoring sys- Act (Cahn 2001). tem introduced. Still, Roma policy is not well integrated into general social devel- The Growth of Civil Society opment strategies and exists as a separate Hungary's attention to minority concerns is and parallel project. Roma continue to also a function of the significant level of develop- suffer discrimination. The Government is ment of civil society in the country. NGO activity currently revising its Roma policy. The has been greater in Hungary than in many other envisaged adoption of a comprehensive countries of the region, in part because of the less long-term strategy and comprehensive restrictive nature of Hungarian communist rule antidiscrimination legislation would be major steps forward in this regard (Com- mission 2002). Table 5.1: Hungarian Minorities in Other Countries Over the past decade, the EU has provided support to Hungary for Roma projects and pro- Percentage grams to meet the objectives of the Copenhagen of Total criteria. Between 1992 and 2001, the PHARE Pro- Country Year Number Population gram allocated 1,259 million Euro to Hungary, and another 120.7 million Euro in 2002.3 PHARE Austria 1991 census 33,459 4.3 support on Roma issues has been earmarked for (based on projects in education, community development, language) Croatia 1991 census 22,355 0.5 policy formulation, and monitoring. Romania 1992 census 1,624,956 7.1 Slovakia 2001 census 520,528 9.7 Hungarian Minorities and Minority Policies Yugoslavia 1991 census 345,376 3.9 Hungary's approach toward its ethnic (taken while minorities has been influenced by a concern for still united) the rights of Hungarians living as minorities in Source: Government Office for Hungarian Minorities other countries (Crowe 1991; 1994) (Table 5.1). Abroad, http://www.htmh.hu/english.htm; Slovak The reorganization of Hungarian territory in the Statistical Office. 93 Roma in an Expanding Europe: Breaking the Poverty Cycle and earlier adoption of legislation regulating the cussed further below) and self-organized groups sector (Bárány 2002). While the socialist era was (Kováts 2001a). characterized by a state monopoly of all spheres Despite this growth in activity, the influence of political, social, and economic life, civil organ- of many groups is limited by inadequate access to izations, including Roma organizations, were financing. Only a small number have survived able to secure gradually more autonomy than for more than a few years because of legislative, those under more totalitarian regimes, which did financial, and organizational problems (PER not even allow the formation of such groups 1998). Most NGOs, including those concerned (Bárány 2002). with Roma issues, are small and donor driven, The Hungarian Civil Code, adopted in 1959, "their existence tied to the implementation of spe- provided for the establishment of civil society cific projects and their activities defined at least in organizations (Jenkins 1999). This code explicitly part by the agendas of the organizations that fund recognized, although under strict administrative them" (OSI 2001). control, "social organizations," including political groups, trade unions, and organizations of GOVERNMENT POLICIES AND APPROACHES women, youth, and other groups. Many of the Successive Hungarian governments have early organizations formed in the final years of played an active role in policymaking and estab- socialism survived the transition in 1989, chang- lishing institutions to address minority policies, ing their names and transforming themselves and Roma issues in particular. Perhaps most into new legal entities (Jenkins 1999). For exam- notably, in 1993, Hungary adopted the Minorities ple, many informal political associations became Act which granted considerable cultural, educa- political parties. In 1993 a unique type of govern- tional, and linguistic rights to Hungary's thirteen mentally supported NGO, the "public founda- recognized minorities through a system of tion," was formed (Box 5.1). national and local minority self-governments This legacy has contributed to the remarkable (MSGs).6 This system is unique to Hungary. The post-transition growth in civil society organiza- country has also established an Office for Nation- tions. The NGO sector grew fivefold between al and Ethnic Minorities and the independent 1989 (with just under 8,800 organizations) and Minorities Ombudsman to oversee minority 1995 (with more than 43,000 registered organiza- rights and protections. Most recently, following tions) (Jenkins 1999).4 At the same time, the num- elections in 2002, the government established a ber of organizations involved in social policy has new Roma Office under the Office of the Prime increased significantly, from virtually no presence Minister, to coordinate Roma policy across the in the early 1980s, to between one-fifth and one- government.7 fourth of NGO activity in 1995. There are also Government funding for minorities is distrib- numerous associations active in the areas of cul- uted through multiple channels (Table 5.2). ture, religion, and business. Roma populations often benefit from general Many of these organizations have focused on programs for minorities. For example, the largest the expansion of services and rights for ethnic budget allocation goes to the education of minorities in general, and for Roma in particu- minorities. In 1999, a total of 4.6 billion forints lar. In the last decade, the greater availability of was allocated for preschools, schools, and dormi- state and NGO funding for Roma issues has led tories for national and ethnic minorities, and for to a rapid proliferation in Roma organizations additional subsidies for non-minority bilingual and events. By the end of 1991, 96 civic bodies educational institutions (Implementation Report concerned with such issues were officially regis- 1999). In 1999/2000, the Public Foundation for tered (Kállai and Törzsök 2000).5 By 1994­1995 National and Ethnic Minorities provided sup- there were nearly 500 organizations, and by port for 586 secondary school Roma students and 1998, there were over 1,000 registered Roma 111 Roma university students.8 Budgetary alloca- organizations, including self-governments (dis- tions specifically targeted for Roma include those 94 Project Experience in Hungary Table 5.2: Government Budgetary Support for Minorities 1997 1998 1999 (million HUF) (million HUF) (million HUF) National Self-Governments for Minorities 306.0 398.7 506.0 Roma National Self-Government 96.0 120.0 138.0 Local Self-Governments for Minorities 300.0 350.0 730.0 Minority civil organizations 70.0 79.1 87.8 Public Foundation for National and Ethnic Minorities in Hungary 395.0 474.0 530.0 Public Foundation for Hungarian Gypsies 170.0 250.0 280.0 Minority Coordination and Intervention Budget1 50.0 55.0 44.0 Ministry of Education, for minority tasks 274.9 290.0 250.0 Gandhi Foundation 325.0 230.0 210.0 Minority theaters 62.0 67.0 74.0 Ministry of National Cultural Heritage (support for cultural programs and minority literature) -- -- 100.0 -- Not applicable. 1. The Minority Coordination and Intervention Budget is used for solving crisis situations concerning minorities that require urgent resolution. Source: Implementation Report, 1999. for the Roma national and local minority self- been raised about its ability to effectively perform governments, the Public Foundation for Hungar- its mandate, due to a perceived lack of authority ian Gypsies, and the Gandhi Foundation. The and difficulty in coordinating across government Gandhi Foundation has also received a signifi- agencies. In 2002 many of NEKH's responsibilities cant proportion of government financing, begin- related to the social integration of Roma and coor- ning with 325 million forints in 1997 and falling dination on sectoral policies were transferred to to 210 million forints in 1999.9 the new Roma Office. This change is intended to strengthen coordination and monitoring of Roma Office for National and Ethnic policies at the interministerial level. Responsibili- Minorities (NEKH) ty for Roma culture and minority rights remain The Office for National and Ethnic Minorities with NEKH. (NEKH) was one of the first new institutions established in 1990. Its mandate has been to assist Medium-Term Package in the development of government minority poli- The first version of the Medium-Term Pack- cies and to review and monitor the situation of age was adopted in 1997 and was aimed at fur- minority communities. Its activities in supporting thering the social integration of Roma in Hun- the Roma minority include financial support to a gary. It outlines measures to be undertaken in number of Roma organizations, such as the education, culture, employment, housing, health, National Roma Information and Cultural Center, antidiscrimination, and communications. Imple- the network of Roma Minority Community mentation efforts were to be coordinated by the Houses, as well as various Roma magazines, Council for Gypsy Affairs, established in 1995 to radio, and TV programs. Since the mid­1990s, harmonize the efforts of government ministries NEKH has taken a leading role in developing and and other institutions in addressing Roma issues. overseeing the implementation of the govern- A review in 1999 shifted the primary emphasis to ment's "Medium-Term Package" for Roma. education and culture, and replaced the Council While there has been consensus about the for Gypsy Affairs with the Inter-Ministerial Com- necessity of NEKH's activities, concerns have mittee for Gypsy Affairs.10 Implementation itself 95 Roma in an Expanding Europe: Breaking the Poverty Cycle is assigned to different government ministries both local and national governments. The Act and other bodies, in cooperation with the Nation- guarantees all recognized minorities individual al Roma minority self-government. In 2000, and collective rights. The Act explicitly estab- budgetary resources allocated for the implemen- lished Roma as one of the thirteen recognized tation of the Medium-Term Package amounted to national and ethnic minorities in Hungary for the around 4.86 billion forints (Kállai and Törzsök first time. 2000).11 The National Minority Self-Government The adoption of the Medium-Term Package (NMSG) system for Roma and other minorities was an important step in addressing the concerns was established in 1995. Representatives and of Roma in Hungary. However, its effectiveness spokespersons of local MSGs vote for the to date has been limited (OSI 2001). Main cri- NMSG.13 The first Roma NMSG was a coalition tiques revolve around weak coordination across formed by the Lungo Drom Gypsy Association the government and a general lack of competence which won all 53 seats. The government is and authority, both of which significantly hamper required to provide funding for NMSG head- implementation. This lack of coordination, com- quarters, infrastructure, and operating costs. bined with a lack of clarity, transparency, and The scope of authority and duties of the financial resources has limited the package's NMSG outlined in the Minorities Act fall into two effectiveness. general categories: independent decision making in specific areas14 and consultation and oversight Parliamentary Commissioner on sectoral policies and administration (Walsh for Ethnic and National Minorities 2000). In this latter role, the NMSGs act as media- The Parliamentary Commissioner for Ethnic tors between local MSGs and the government, and National Minorities (Minorities Ombuds- and as consultants in the drafting of legislation man) is an independent institution established to that affects the minority at all levels of govern- monitor the implementation of minority rights, to ment. They are also expected to take part in the investigate complaints, and to pursue remedia- supervision of minority education. Despite these tion for the infringements of the rights of nation- guidelines, the Roma NMSG and those of other al and ethnic minorities.12 According to the minorities have been challenged by the lack of Ombudsman's office, Roma have lodged the vast precedent and clarity on the nature of the rela- majority of complaints. Moreover, unlike other tionships between the NMSGs and local MSGs, complaints, their concerns are disproportionately and their corresponding governmental authori- related to acts of discrimination (OSI 2001). The ties (Kováts 2001b).15 While the Roma NMSG had Ombudsman reports that prejudice and discrimi- an important role in shaping the Medium-Term nation against Roma is widespread in areas such Package, and has undertaken a number of high as law enforcement, employment, education, profile initiatives, insufficient finances have also housing, and access to public and private institu- limited its capacity (Kováts 2001b). tions (Commission 2000). Further, the Ombuds- Local MSGs are elected bodies at the local man reported that approximately 48 percent of level. They do not have a vote in the majority complaints submitted by Roma in 2000 were filed local governments, but they can veto any local against local governments (OSI 2001). government decision over matters that may affect them, particularly those concerned with educa- The Minorities Act and the Role tion, culture, local media, efforts to sustain cul- of Minority Self-Governments tural traditions, and the use of minority lan- The Minorities Act of 1993 expanded minori- guages (Commission 2000; NEKH 2000). The first ty rights in Hungary and established Hungary's MSG elections were held in 1994­95, and resulted unique system of minority self-government in a total of 738 MSGs, of which 477 were Roma. (MSG), which allows minorities to form their Following the second round of elections in 1998, own elected bodies to work in partnership with the number of local and Roma MSGs nearly dou- 96 Project Experience in Hungary bled to 1,367 and 753, respectively (NEKH 2000). to exhibit higher levels of social cohesion among By June 30, 2000, there were 738 Roma local MSGs Roma themselves. They successfully built rela- out of a total of 1339 local MSGs, compared with tions between Roma and non-Roma based on 271 German, 75 Croatian, and 75 Slovak.16 norms of trust and cooperation and could create The MSG system has garnered international effective institutional linkages outside of the attention and has raised the profile of Roma locality (Schafft and Brown 2000). issues, the status of Roma communities and their On the other hand, this study and others have representatives, as well as those of other minority identified a number of factors which limit the groups in Hungary. However, its effectiveness effectiveness of MSGs including their relatively has been mixed. A national survey of 232 Roma narrow focus on "cultural" issues, financial con- political leaders in 1998 showed that some Roma straints, limited capacities and influence, and MSGs had been more active in political and social their perceived lack of legitimacy. The following areas within their communities than others summarizes some of the main challenges. (Schafft and Brown 2000). Many Roma MSGs have been able to success- DEPENDENCE ON LOCAL GOVERNMENTS FOR FINANCING. fully initiate a variety of projects to the benefit of Minority self-governments receive a small their communities. The same survey identified amount of funding annually from the state budg- the frequency with which MSGs engaged in dif- et, and are therefore required to seek funding ferent kinds of development initiatives, as well as from multiple sources, including county and local which of these were perceived to be most impor- governments and outside bodies. The Minorities tant (Table 5.3) (Schafft and Brown 2000). Over 75 Act does not provide for any explicit financing to percent indicated that their MSG was involved in MSGs. The amount specified by the Budget Act the provision of social welfare, and just over 60 for MSGs is allocated in equal proportions among percent identified cultural and education pro- municipal governments, irrespective of their size, grams. The provision of local media program- or the size of the minority population in the area. ming and entrepreneurial activities were less fre- These funding considerations have three impor- quent. Agricultural support was cited as the most tant consequences. First, in practice, MSGs are important activity, followed by social welfare increasingly dependent on local municipal gov- programs and education/job training. Least ernments for funding, which can compromise important were opportunities for local media their independence. Second, financial constraints programming. in many cases prevent MSGs from implementing The survey suggests that some Roma MSGs even their short-term mandates, let alone meeting do function as a valuable resource to their com- the expectations of the electorate (Kállai 2000). munities. It also found that the more successful Third, funding uncertainties limit medium- and MSGs with higher institutional capacity tended long-term strategic planning. Table 5.3: Roma MSG Local Development Initiatives, 1998 Cited Among "Most Important" Development Initiatives Involvement (percent) MSG Activities (percent) Social welfare programs 78.8 21.4 Cultural programs/festivals 61.1 18.1 Education/job training 60.7 30.2 Agricultural support 58.4 32.0 Local media programming 45.5 9.0 Economic enterprises/business start-ups 42.4 12.7 Sources: Schafft 1999; Schafft and Brown 2000. 97 Roma in an Expanding Europe: Breaking the Poverty Cycle LACK OF CAPACITY AND INFLUENCE. In addition to the impoverishment of much of the Roma popula- general lack of financing, some of the most signif- tion--issues which are only indirectly addressed icant problems reported about the MSG system the current system. are related to their lack of capacity (Kállai 2000). For example, most MSGs do not have sufficient PROJECT EXPERIENCE: AN OVERVIEW information about the various legal, tendering, This section overviews a selection of Roma lobbying and cooperating opportunities, and lit- projects initiated in Hungary during the first tle of this information is readily available, even decade following transition, while the following from the National Roma Self-Government. Sec- section draws some general lessons. In the spring ond, minority representatives tend to lack politi- of 2000, the World Bank collaborated with a team cal experience and professional training. These of Hungarian researchers to compile a database problems are compounded in the case of Roma of Roma projects in Hungary.17 At that time, no MSGs, given the smaller pool of professionally aggregate information was available on the types trained, educated, and politically experienced of projects which had been implemented, their Roma candidates. size, coverage, geographic concentration or sec- toral focus. Policymakers, the NGO community LACK OF LEGITIMACY/REPRESENTATIVENESS. Roma MSGs and others involved in the Roma issue had no are also criticized for being unrepresentative. information on which to base their project devel- One reason is that there is no requirement that opment and partnership. representatives who are elected to the MSG need The database aimed to review the landscape to be a member of that minority. While the of Roma projects which had been implemented in Minorities Act addresses the difficult question of Hungary between 1990 and 1999, and to provide who is a minority, relying solely on self-identifi- a basis for an assessment of Roma policy in this cation implies a degree of flexibility that has period. The review focused on selected projects in caused difficulties, particularly in terms of the areas of employment, education, housing, minority representation (Kállai and Törzsök and health, and was designed to create a resource 2000). For example, in the 1998 round of local which would be useful for policymakers, NGO elections, a number of individuals were elected officials, Roma leaders, and others involved in to MSGs, even though they did not belong to that Roma projects. The inventory identified 1,396 minority. This happened in the town of projects with a total cost of 3.6 billion forints, Hajdúhadház, where two non-Roma were elect- implemented and financed by central and local ed to the Roma MSG. The role of non-Roma in governments, NGOs, and the private sector. MSGs is controversial. On the one hand, a mix of A broad definition of "Roma projects" was Roma and non-Roma can facilitate cooperation, used. Some projects included in the inventory particularly with the local government. On the had both Roma and non-Roma beneficiaries, such other, it can undermine the local Roma commu- as the social land project discussed later in this nity's confidence in the MSG. chapter. The aggregate data from the inventory illustrate a steady increase in project activity FOCUS ON CULTURAL ISSUES. As outlined in the 1993 related to Roma during the 1990s, as well as in the Minorities Act, the most important task of MSGs amount of resources spent (Table 5.4). is to build cultural autonomy for minorities. The spike in expenditures in 1993 represents While this issue in all its forms--educational, a grant of 215 million forints which was provid- linguistic, maintenance of traditions--is an ed by the Soros Foundation for the establishment important aspect of improving the status of of the Gandhi secondary school for Roma in Pécs Roma, it may not always be the most urgent in southern Hungary. In 1996, the increase in issue for local Roma communities. More press- expenditures represents the government's initia- ing are concerns related to the disproportionate- tive to establish the Public Foundation for Roma ly high rates of unemployment and the mass (Box 5.1). Of this amount, approximately 30 per- 98 Project Experience in Hungary Table 5.4: Project Inventory, 1990­1999 Year Number of projects Total expenditures (HUF '000s) % of Government expenditures 1990 1 150 0.00 1991 19 6,218 0.00 1992 29 70,657 0.00 1993* 47 413,726 0.02 1994 63 179,486 0.01 1995 116 279,332 0.01 1996 241 569,910 0.02 1997 288 555,877 0.02 1998 253 643,731 0.02 1999 339 922,240 0.02 Total 1,396 3,641,327 0.07 *The increase in resources in 1993 reflects the investment of the Soros Foundation in the Gandhi School. Source: World Bank project database. cent was allocated to income generating pro- Regional Distribution of Projects grams, 20 percent to small business develop- The geographic distribution of projects across ment, and 15 percent to student scholarships. counties in Hungary reflects the ethnic diversity A fund established in 1998 by the National of the country. Table 5.5 illustrates the regional Roma Self-Government and the central govern- breakdown for all projects which could be ment to help local governments cofinance regional mapped to a specific county.18 The regions with development programs through grants significant- the largest shares of Roma, Northern Hungary, ly increased the total resources allocated for Roma the North Great Plain, and Southern Danubia, projects. The fund started with 100 million forints have the greatest share of projects. These are also for the upgrading of local infrastructure. The first the regions which have consistently had the programs were launched in 1999 but as the flow of highest unemployment rates, indicating that information from Regional Development Councils project activity may also reflect greater need in to the relevant government ministries is limited, those areas. only some of these projects were included in the Per capita expenditures on projects vary sig- project inventory. nificantly across regions but were the highest in Table 5.5: Projects by Region, 1990­1999 % of % of Total expenditures Per capita County total projects total Roma* (HUF `000s) expenditures (HUF) Budapest 4.5 8.2 93,590 2,472 Central Region 2.4 6.0 63,403 2,302 Western Danubia 3.5 5.0 111,096 4,803 Northern Danubia 2.1 5.5 22,799 905 Southern Danubia 16.5 14.2 784,492 11,993 Northern Hungary 34.3 27.9 865,739 6,722 North Great Plain 29.0 25.4 817,098 6,965 South Great Plain 7.7 8.0 205,703 5,571 Total 100 100 2,963,920 6,413 *1992 estimates by G. Kertesi and G. Kézdi. Source: World Bank project database. 99 Roma in an Expanding Europe: Breaking the Poverty Cycle Box 5.1: Public Foundations in Hungary There are two main types of NGOs in Hungary: associations and foundations.1 In 1993, a separate, inter- mediate category of "public foundations," unique to Hungary was created. These public foundations are established by the government to fulfill a specific public policy objective. While supported primarily through government financing, they are independent bodies, intended to be both "state" and "civic" in character. The identification, development, and implementation of their goals are overseen by an advisory or trustee board, which can consist of members of government (national, county, or municipal), representatives of civil society, business interests, research and academic communities, as appropriate to the goals of the foundation. In some cases, a representative from a relevant ministry has a position on the board, and the min- istry may provide additional direction and oversight. Public foundations are found in all sectors (e.g. educa- tion, health, labor, environment, arts and culture), and operate at the national, county and municipal levels. By 1995, there were 458 public foundations, or 3 percent of all foundations representing just over 1 percent of the total NGO sector (Jenkins 1999). The Public Foundation for Hungarian Gypsies In 1990, the Office for National and Ethnic Minorities (NEKH) was established to develop and oversee minority policy. In conjunction with this Office, two public foundations were established in 1995 to manage the government funds allocated to for improving the situation of Roma. In the same year, the Public Foun- dation for National and Ethnic Minorities was set up, primarily to promote the self-identity and to preserve the cultures of minority communities. Later in 1995, the Public Foundation for Hungarian Gypsies was estab- lished. This Foundation's objectives are to promote social integration, mitigate unemployment, increase Roma school attendance, and protect civil rights. While by-laws do not exclude outside funding, to date financing has come almost exclusively from the state budget. Total government financing was 350 million forints for 2001 and 2002 combined. The majority of this funding is allocated to a scholarship program, enter- prise promotion, and support for income-generating projects. The Foundation also supports a wide variety of programs, including small business development and training of Roma officials in public administration. The Foundation works closely with government and civil society organizations with similar interests. Applications for funding can come from Roma minority self-gov- ernments (MSGs), communities, foundations, municipalities, and individuals. The decision-making authority of the Foundation rests with a 21 member Board of Trustees which is responsible for the mandate of the Foundation, and for approving all applications for funding. The Public Foundation Office consists of ten employees, over half of whom are Roma. They assist the board, administer the projects and funds, and oversee the completion and processing of the applications. In addition, 5 inde- pendent external monitors help prepare applications, and monitor and evaluate projects. Note: 1. These organizations are defined in the Hungarian Civil Code on Associations (Sections 61­64), and Foundations (Section 74). Both of these organizational forms existed under communism, subject to tight administrative control. Such controls were relaxed and inde- pendence from government supervision was instituted through the Law on Association (Law 1990/II, January 1989) and an amendment to the Civil Code (Law 1990/I, January 1990). Source: Office of National and Ethnic Minorities, 2002. Southern Danubia. This reflects higher spending nok County in the North Great Plain region, the in two counties--Baranya and Tolna. In Baranya, situation is different, as a large number of Roma a significant amount was spent on the Gandhi organizations have been involved in implement- School in Pécs. In Tolna 280 million forints were ing projects. spent on infrastructure for utilities. In Zala Coun- ty in Western Danubia and Jász-Nagykun-Szol- 100 Project Experience in Hungary Table 5.6: Distribution of Projects by Sector, 1990­1999 Number % of Total expenditures % of total Average project size Sector of projects total projects (HUF `000s) expenditures (HUF `000s) Education 200 14.3 1,024,233 28.1 5,121 Employment 1,121 80.2 2,174,272 59.7 1,940 Health 36 2.6 32,795 0.9 911 Housing 3 0.2 2,700 0.1 900 Community centers 9 0.6 11,877 0.3 1,320 Miscellaneous 29 2.1 395,451 10.9 13,636 Total 1,398 100 3,641,328 100 23,828 Source: World Bank project database. Sectoral Distribution of Projects period covered by the inventory, 3 percent of There is a wide divergence in the level of projects were devoted to health issues and less activity (measured as percentage of total projects than 1 percent to housing. Projects categorized as and percentage of total expenditures) in each of "miscellaneous" are multisectoral, generally the sectors examined in the inventory. The high- addressing housing and employment issues, and est levels of activity took place in education and are largely financed by the Regional Develop- employment, and the lowest levels in health and ment Councils. Community development proj- housing (Table 5.6).19 In education, 21 percent of ects in Borsod-Abaúj-Zemplén county (57 million the total expenditures represent the investment in forints) and the installation of gas pipes in Tolna the Gandhi school. The Soros Foundation's schol- county (280 million forints) comprise a large arship program also comprises a significant share share of this category. of the resources allocated to the sector. During the Table 5.7: Projects by Donors, 1990­1999 Number % of Total expenditures % of total Donor of projects total projects (HUF `000s) expenditures (%) Private foundations Autonómia Foundation 474 34.0 274,409 7.5 Network for Democracy (DemNet) 4 0.3 8,379 0.2 Soros Foundation 52 3.7 824,902 22.7 Government financed Public Foundation for Modernizing Public Education 4 0.3 5,500 0.2 Ministries 318 22.8 1,364,313 37.5 National Foundation for Employment (OFA) 12 0.9 117,784 3.2 National Institution for Health Prevention (NEVI) 29 2.1 21,602 0.6 Office for National and Ethnic Minorities (NEKH) 37 2.6 63,891 1.8 Public Foundations 431 30.9 369,349 10.2 Regional Development Councils 34 2.4 586,615 16.1 Total 1,395 100 3,636,744 100 Source: World Bank project database. 101 Roma in an Expanding Europe: Breaking the Poverty Cycle Sources of Funding the second largest city in the country. Approxi- Between 1990 and 1999, most projects were mately 2,400 residents are thought to be Roma, implemented with government financing (62 per- and the share of the Roma population in the town cent), although a significant portion (38 percent) is growing rapidly. According to local estimates, were financed by private foundations (Table 5.7). about half of the children who registered for Government ministries financed 23 percent of all school during the past few years are Roma. The projects, representing the largest percentage of local economy has deteriorated since 1989. In total expenditures (38 percent).20 Government 2000, the unemployment rate was estimated at 40 sponsorship of projects also included public percent for the whole population, and 95 percent foundations (30 percent of all projects, and 10 per- for Roma. According to local leaders, the large cent of total expenditures), and the Regional majority of educated residents leave the area for Development Councils (with only 2 percent of all opportunities elsewhere. Local relations between projects accounting for 16 percent of total expen- Roma and non-Roma are generally characterized ditures). Public foundations supported 31 percent by segregation, hostility, and tension. of all projects. Two private, nonprofit founda- There is a high degree of segregation within tions, Autonómia and Soros accounted for 30 the Hajdúhadház schools. The majority of Roma percent of the total expenditures for this period. children attend separate remedial classes in the two primary schools in the town, the Földi János PROJECT CASE STUDIES and Bocskai schools. They study separately from In addition to the project inventory, more in non-Roma children, with different teachers, text- depth case studies, based on interviews with books, poorer conditions, and fewer resources project administrators and beneficiaries, were than their peers. A 1999 report by the Debrecen undertaken by the Hungarian research team on a public health service "strongly objected" to con- set of selected projects. As noted throughout this ditions in the Roma section of the Bocskai volume, there has been limited evaluation of school. The classroom walls were rotting, the Roma projects. While these assessments do not floor was unstable, toilets broken, and lighting substitute for rigorous project evaluation mecha- insufficient. Since there was no gymnasium, nisms built in ex ante, they provide insight into physical education classes were held in the hall- issues and lessons. The case studies were ways and classrooms. designed to validate the information collected during the inventory, to see if projects included in BRIDGING CLASSES. Both schools receive state subsi- the database were actually implemented on the dies for the education of Roma children. Accord- ground. Based on that experience, general lessons ing to law, these resources are intended for reme- can be drawn to inform future projects. dial "bridging" classes and courses on Roma culture and education. Bridging classes are Case Study #1: Segregated Education intended to overcome the educational disadvan- in Hajdúhadház tages of Roma children--many of whom do not The case of the education of Roma children in attend preschool--and to integrate them into the Hajdúhadház illustrates how poorly-designed mainstream education system. In practice, these incentives can undermine the objectives of a pol- programs tend to perpetuate segregated educa- icy.21 In Hajdúhadház, as is the case throughout tion and are constrained by the lack of qualified Hungary, local governments receive subsidies to staff and resources. In 1998, 67 percent of the support education for Roma children. However, Roma students in the Földi János school studied as illustrated in this case, these subsidies work to in segregated Roma classes. reinforce segregation and compromise the quality Students in the bridging programs learn the of education for Roma students. same material as those in regular classes, but go Hajdúhadház is a town of 13,000 in eastern through less material at a slower rate. As a result, Hungary. It is situated 12 miles from Debrecen, while the aim may be to bring Roma students up 102 Project Experience in Hungary to standard levels, their chances of returning to in Hajdúhadház. From the Földi school, thirteen mainstream classes are reduced each year that families reported that they had protested against they stay in the bridging classes. In addition to the the poor school conditions and teaching quality, bridging classes, an estimated one-fourth of Roma but their children were never admitted to the cen- students in Hajdúhadház study in special classes tral building. Nearly 70 percent of Roma families for the mentally disabled which are run by the who were interviewed indicated that they felt Bocskai school. In 2000, 132 out of 156 students in that their children should be allowed to study in the special education classes were Roma. It is very mixed classes. Similar sentiments were expressed rare for children who attend these classes to con- at the Bocskai school. tinue their education at the secondary level.22 In Hajdúhadház, students are placed in the FINANCING SEPARATE EDUCATION. Hungary's system of bridging classes based on the evaluation of teach- intergovernmental transfers reinforces the sepa- ers and whether they have attended preschool. rate education of Roma students. As the subsidies While preschool in Hungary lasts three years, are earmarked for bridging and special classes, only the final year is compulsory. As a result, schools have a financial incentive to maintain Roma students are frequently placed in bridging these programs, regardless of their usefulness. classes, because they generally attend only one Both schools have expanded their Roma sections. year of preschool. The principal of the Földi Janos At the Földi school, a storeroom was recently con- school explained: "The only selection criteria [for verted into another Roma classroom. bridging classes] is preschool education. To place While schools have an incentive to retain their someone in the normal class without complete subsidies, local governments use the provision of preschool education would be similar to a com- additional resources to the schools as an excuse to petition between a Trabant and a Mercedes." decrease their contributions to the schools' budg- The educational subsidies are also earmarked ets. In other words, the local governments reduce to support the inclusion of Roma culture and his- their support to the schools in proportion to the tory into the curriculum. While the intentions of amount of the subsidy. This squeezes the schools this policy are positive, the schools and teachers fiscally, as the bridging and special schools cost were unprepared to provide this type of instruc- more than the regular classes. The principal of the tion. While more than half of the 160 teachers in Földi school estimated that the Roma program Hajdúhadház teach Roma children, few have any cost three times as much per student as the sub- training in multicultural education, or access to sidy provided by the state. appropriate teaching materials. The principals of There is little monitoring of the use of the both schools also noted prejudices among their subsidies. However, under pressure, the Ministry teachers. In 1992 the principal of the Földi school of Education undertook a national survey of their asked his colleagues to list the three best features use in 2001. The Ministry, through a research cen- of their school. The third most common answer ter, contacted more than 900 mayors, of whom was the segregation of Roma pupils into separate 370 did not respond. Those mayors who did buildings. Some teachers also noted that they respond acknowledged that there were no bridg- viewed having to teach bridging and special ing classes in their schools, although they did classes as punishment. Even when teachers have receive the subsidies. good intentions, their lack of background and The challenges of addressing Roma education understanding of Roma issues limit their effec- in Hajdúhadház are evident elsewhere in Hun- tiveness. In the Földi school, one of the teachers gary. Recent studies indicate that the implementa- learned the Roma language, but teaches a dialect tion of "catch-up" classes in Hungary is wide- which is not spoken in the settlement. spread. As of 1995, catch-up programs were in 433 The majority of Roma parents interviewed in schools (Radó 1997). While initially envisaged as a both the Földi and Bocskai schools expressed dis- temporary solution, in many cases bridging class- satisfaction with the segregated schooling system es have resulted in long-term institutional segrega- 103 Roma in an Expanding Europe: Breaking the Poverty Cycle tion, in part due to inadequate facilities and quali- socialist period, some private farms alongside ty of education in the segregated classes, and the agricultural cooperatives specialized in cultivat- growing resistance by teachers and parents in the ing horseradish. As a result, at the outset of the mainstream schools to register Roma students at transition, a market existed with a network of all. Analyses of catch-up programs have revealed producers who processed and sold the product that they are generally of low quality, sometimes on domestic and foreign markets. Although existing "in name only"--not following the speci- Roma participated in horseradish cultivation as fied curriculum (Radó 1997; Havas et al. 2001). seasonal workers, they were left out of the priva- Further, in many areas exclusionary practices per- tization process because they did not own land sist, including the continued practice of placing or were not members of the local cooperative. So Roma in special schools for the disabled. they were not eligible to become landowners and independent horseradish farmers. Case Study #2: Roma Employment Project In 1996 Miklós Rózsás, an energetic and in Bagamér prominent member of the local Roma community A common sentiment among policymakers and Chairman of the Local Association of Roma and others interested in expanding opportunities Leaders, and Sándor Zsákai, another leader of the for Roma in Hungary and in other countries is same association, came up with the idea to help that agricultural programs can provide opportu- Roma become horseradish farmers. They tried to nities for self-sufficiency, particularly for Roma in raise money and sent a proposal to the Autonó- rural areas. In reality there has been very little mia Foundation and the Public Foundation for experience with agricultural programs, and even the Gypsies in Hungary. Their initiative was less evaluation of whether such projects yield rejected at first, but in 1997 they received 1.5 mil- their intended results, and mitigate rural Roma lion forints (about US$7,000) from the Autonómia unemployment. The case of the horseradish proj- Foundation under the condition that half of the ect in Bagamér provides a glimpse of how such a sum would be repaid to the Foundation after the project can play out in practice.23 harvest. After that they received support from the The village of Bagamér is situated near the Foundation every year for their horseradish- Romanian border, 30 km from Debrecen. In 1999, growing program, and in 2000 and 2001 the total it had a population of 2,580. There are 186 Roma subsidy was provided as a grant. families in Bagamér, or approximately one-third During the first phase of the project, between of the population. Between 1989 and 1992, the 1997 and 1999, resources were requested for plow- majority of Roma employed in state-owned enter- ing, fertilizer, pesticides, spraying, irrigation, har- prises lost their jobs. In 1999, 80 percent of the vesting, and transportation. The Association also heads of Roma families were without legal and requested money for leasing land, while pledging permanent work. Restructuring and unemploy- that part of the profit would be spent on future ment affected the entire labor force. The agricul- land purchases. The funds from Autonómia were tural cooperative in the area, which primarily transferred in several installments, linked to employed non-Roma, was privatized. These progress in the project. The Association paid developments led to the emergence of a number providers directly for services such as plowing, of private farms, which rely on more temporary, while other services were paid by the individual seasonal labor, rather than on permanent employ- producers. The contract between Autonómia and ees. This increased competition for employment the Association defined the upper limit of what and heightened ethnic tensions in the village. could be paid to each household and for each Horseradish cultivation has a long tradition phase of work, but the beneficiaries themselves in Bagamér. The plant is processed for use in the could decide when to withdraw the money. food industry, and as an ingredient in some During this first phase of the project, all partic- pharmaceuticals. Growing horseradish is labor- ipating households but one repaid the loans after intensive, and requires expertise. During the the harvest. In 2000 financing conditions changed 104 Project Experience in Hungary significantly. The project cycle was extended to ronmental conditions, and a donor which is ready two years from one, and the subsidy became a to take risks. nonrefundable grant. From 2000 onwards the Autonómia Foundation focused its efforts on proj- Case Study #3: The Social Land Program ects that could become self-sustaining over time. in Zsadány The aim was to support entrepreneurial initiatives The case of Zsadány provides another exam- and Roma who could become primary producers. ple of an agricultural project.24 Zsadány is a vil- The majority of the participants in the Bagmér lage in Békés County. Of its 1,882 inhabitants, project in 2000 and 2001 already owned land and between 100 to 150 are Roma. The village has were ready to continue farming. been struck by widespread unemployment, agri- A weakness of the Bagamér project is its cultural crisis, and rural poverty. It has an excep- small scale. While the nominal value of the finan- tionally high rate of unemployment. Out of the cial support from Autonómia has essentially re- 670 working-age inhabitants, 300 are registered as mained the same since 1997, the cost of cultiva- unemployed. Despite this, the population has tion has significantly increased, and that has been stable for many years, with amicable rela- deterred many households from participating in tions between Roma and the majority population. the program. In 2000 13 families were included in Roma in Zsadány are relatively well integrated the program, down from 19 in 1997. Another crit- into the larger community; they work and live icism of the project was its lack of targeting and together with non-Roma, and mixed marriages transparency in selecting beneficiaries. The Asso- are not uncommon. Rather than succumbing to ciation's main concern was to repay the grant to economic decline, the mayor and the local gov- Autonómia, therefore it sought families which ernment have actively sought to rejuvenate the were most likely to succeed in the project, and village, including applying successfully for pub- targeting based upon need was not a primary cri- lic work programs from the central government, teria. This practice has led to charges of elitism and initiating the social land program. from some households left out of the project. Despite these criticisms, the project remained THE SOCIAL LAND PROGRAM. Since 1990, social land viable. The project demonstrates that even given programs have been supported across Hungary, favorable market conditions, success requires a financed by relevant ministries, mostly by the fortuitous combination of circumstances, includ- Ministry of Health, Social and Family Affairs. The ing enthusiastic leadership, a profound knowl- main objective of these programs is to alleviate edge of the production process, conducive envi- rural poverty by providing financial assistance, Table 5.8: Program and Activities Supported under the Social Land Program Program types Activities 1. Assistance in production and services (means-tested) Use of land; leasing of land, seeds, machinery and chemicals; support for animal breeding. 2. Organization of production Organization of production and marketing; assistance in processing, storing, and transportation of crops; securing tools. 3. Services Technical assistance; training courses, events; community development; self-help groups; setting up organisations for more effective production. Source: Ministry of Health, Social and Family Affairs of Hungary. 105 Roma in an Expanding Europe: Breaking the Poverty Cycle services, and support to poor households which By most accounts, the program is considered lack the means and capacity to engage in small- to be important in its attempts to address prob- scale farming or animal breeding projects (Table lems such as poverty, unemployment, and social 5.8). Nearly 75 percent of the programs offer assis- exclusion. The profitability of the program is tance in production and services and are aimed at modest at best. According to the rough estimates increasing self-sufficiency and income levels. The of the local government in 1998, every forint of program is open to Roma and non-Roma families, assistance generated an income of 1.3 forints. A and is means-tested to reach the poorest house- benefit of the program is that the long-term holds. Roma comprise 51 percent of the benefici- unemployed become eligible for unemployment aries of the program throughout the country, benefits after six months of participating in the while regionally the rates vary from 29 percent program. (Békés County) to 70 percent (Jász-Nagykun-Szol- Given the kinds of environmental and market nok County). conditions that plague agricultural production generally, as well as local challenges related to the THE PROGRAM IN ZSADÁNY. Organized by the local lack of arable land and the small scale of produc- government, the social land program started in tion, small projects supported under the social Zsadány in 1995. While the primary goal of the land program can rarely become sustainable. program is to improve the living conditions and Nevertheless, the program has demonstrated prospects of the poor, other aims include stimu- adaptability and flexibility. Investments in assets lating community involvement in local develop- have improved the efficiency and effectiveness of ment, providing public works, promoting inclu- the program which contributed by providing rel- sion of poor and excluded groups, and reducing atively stable sources of legal income to benefici- prejudices against Roma and other vulnerable aries, easing the poverty of rural families, and groups. increasing community acceptance and inclusion The Ministry of Health, Social and Family of both Roma and non-Roma families. Affairs has supported the Zsadány initiative every year since 1995, with the exception of 1997, when Case Study #4: Is Nyíregyháza Building the Autonómia Foundation filled the gap. The a "Roma Town"? Public Foundation for Gypsies in Hungary also Between 1998 and 2000 the local government provided support. Ministry assistance resumed of Nyíregyháza worked hard to develop one of between 1998 and 2000. While the amount of the largest Roma settlements in Hungary.25 The financial support remained relatively stable, the city invested significant resources into the devel- content of the program has changed considerably opment of the settlement, named "Gusev," out of over the years. The project initially focussed on its own resources, as well as funds received from growing tomatoes; however because of unfavor- the central government and donors. City officials able environmental and market conditions, in organized public works programs, developed the 1998 cultivation shifted to corn and mixed vegeta- sewage system, replaced the water pipes, and bles, along with rabbit-breeding. Over the years, opened a Roma community center. The results of significant investments have been made into agri- these investments are significant. However, if the cultural assets (machines and land). entire program is implemented, it will further Of the 40 families participating, as of 1999, 20 separate Roma in Nyíregyháza from the rest of were Roma. The turnover of participants is rela- the population. tively high, with only half continuing in the pro- Nyíregyháza is located in Szabolcs county. It gram for a second year. There are various reasons has two large Roma settlements: the Orosi street for this high turnover rate, including better settlement which is located at the eastern gate of employment opportunities elsewhere. As of 1999, the city in a prosperous area, and the Gusev settle- only three persons were excluded from partici- ment, which is one of the largest Roma settlements pating in the program. in Hungary. Gusev is situated on the outskirts of 106 Project Experience in Hungary the city, separated from the majority population by of Orosi Roma into other neighborhoods in Nyír- a railway station, military barracks, and an indus- egyháza was deemed too politically risky. trial zone. The city's plans are to remove the Orosi In 1998, the local government established a settlement and rehabilitate the Gusev settlement to corporation to oversee local housing develop- accommodate both populations. ment, including new housing and rehabilitation, The Orosi settlement was built in the 1960s as the management of the meager public rental a temporary housing settlement. By the 1980s, housing stock, collection of debts, and reloca- urban sprawl around the impoverished settle- tions. The city council approved 60 million forints ment had grown substantially, prompting the for the program and received an additional grant municipality to demolish half of the houses, and of 25 million forints from the central government relocate the families to Gusev. In the 1990s, pres- for public works. Staircases of the apartments sures to remove the remainder of the settlement were repainted, basements cleaned, and side- increased, prompted in part by increasing prop- walks constructed in the narrow streets. Water erty values and dynamic development in the sur- pipes were replaced throughout the settlement rounding area, including several new shopping and water meters were installed in the single- malls, and plans to build a foreign-funded exhibi- room units. The Roma Community Hall was tion center adjacent to the Roma settlement. refurbished. Special programs for children, job According to a 1993 survey, there were approxi- clubs, art clubs, and various competitions were mately 510 Roma still living in the settlement. introduced to reduce the exclusion in the settle- In contrast, the Gusev settlement was created ment and reinforce local trust in municipal insti- in the late nineteenth century and served as bar- tutions, programs, and resources. A wide range of racks of the Monarchy's cavalry regiment. In the further improvements are planned, including a 1950s, it was used to house Soviet officers, and homeless shelter, additional sewage, and the later it became the residence area of the local installation of district heating in all apartments. party and administration elite. In the 1960s, the Providing the Gusev settlement with im- appeal of the settlement declined as more affluent proved living conditions and better access to families moved to new high-rise housing estates. services seems to be a move in the right direction. By the 1970s, the settlement became a "penal Yet, rehabilitation of the settlement, and the relo- colony" within the public housing system. Fami- cation of Roma families from Orosi, will further lies in debt, evicted families, and many Roma isolate the Roma population by increasing their who had moved into the city were provided with geographic and educational segregation from the housing in Gusev. A survey conducted in the rest of the population in Nyíregyháza. Further, it early 1990s indicated that there were around 830 ignores the strong potential for increased inter- registered residents living in Gusev. Current esti- ethnic tensions among Roma. The two Roma mates place the population at over one thousand. communities are opposed to the idea of living Over the years, the infrastructure and reputation together. This has further heightened tensions of the settlement have deteriorated and the popu- within the Roma community and increased sus- lation has become predominantly Roma. picion of the local government among Roma. In the 1990s, the city of Nyíregyháza under- took a wide range of urban development projects PROJECT EXPERIENCE: LESSONS LEARNED aimed at cleaning up the inner city, and fostering Despite the diversity of the projects reviewed investments in industry and services. Investors in the case studies, some general issues and con- interested in the development potential of the city clusions emerged which have implications for have pressed for the removal of the Orosi street future projects in Hungary and elsewhere settlement. In response, the municipality agreed that Gusev be rehabilitated and more housing be TRADEOFFS IN MANAGING PROJECT OBJECTIVES. The suc- built to accommodate new residents, including cess of many projects depends on how project those from Orosi street. Moving the large number objectives are interpreted and managed. In many 107 Roma in an Expanding Europe: Breaking the Poverty Cycle cases, project objectives entail difficult tradeoffs some--but not all--interests. Significant efforts and the negotiation of multiple--often conflict- have been made to improve the living condi- ing--interests of donors, implementing organiza- tions of the Gusev Roma settlement, yet in the tions, beneficiaries, and majority communities. long run, these efforts, and the relocation of Roma families from Orosi into this community, TARGETING BENEFICIARIES. Beneficiary selection in- will exacerbate the social exclusion of Roma volves tradeoffs in objectives. For example, the through explicit geographic and educational Bagamér horseradish project selected participants segregation. Alternative programs aimed at based upon their capacity for success in farming, facilitating the greater integration of Roma and and did not necessarily seek the families in great- non-Roma communities were either not consid- est need. This approach can be controversial, ered or were deemed politically too risky. These because of the high level of poverty among Roma issues echo the challenges described in the pre- communities. However, in the long run this vious chapter on Roma settlements in Slovakia. approach may improve the welfare of the com- While it is urgent to improve living conditions munity. Targeting households with the greatest in settlements, such investments are invest- potential can help ensure that the project gets suf- ments in the future separation of the settle- ficiently established and that it can be expanded ments from the majority population. to those in greater need. However, perceptions of Desegregation efforts in Hajdúhadház have inequality, a lack of transparency, and increased come up against similar barriers. Both primary ethnic tensions (through the inclusion of non- schools in the town have made significant efforts Roma families) may also jeopardize project objec- to improve the conditions for Roma students. tives. In the Bagamér case, further information is However, both are concerned about becoming needed to assess whether the households that did known as the local "gypsy-school" and risk losing participate in the project benefited from the proj- the children of the local elite to the other school. ect, and whether they would have succeeded There is intense competition between these two without project assistance. schools for resources, better students, and reputa- tions. While it is in the interest of both schools to RISKS OF DECENTRALIZATION. The Roma resettlement retain the state subsidies for special education program in Nyíregyháza demonstrates some of classes, they have been reluctant to challenge the the potential risks of decentralized projects which strong opposition to desegregation expressed by are overseen by local governments. In Hungary, non-Roma parents and education officials. housing policies and programs are determined exclusively at the local level. While this will allow ORGANIZATIONAL LEADERSHIP AND EXPERIENCE. A num- projects to be tailored to local conditions, it raises ber of the case studies demonstrate the impor- the risk of their being "captured" by local inter- tance of linkages with established and respected ests if they are designed and implemented with- organizations, as well as the benefits of capable out incorporating the needs and concerns of local and committed leadership. The experience and minority self-governments, other civil society reputations of leaders and implementing organi- groups, or Roma themselves. National moni- zations affect their abilities to secure support toring and evaluation would allow for more from donors and manage projects. inclusive criteria to be applied, and could help A key factor in the success of the Bagamér ensure that beneficiaries are included in decision- horseradish project was its leadership. The head making processes, as well as in project imple- of the Association is a charismatic leader who mentation. was formerly the head of the local minority self- government, and has had significant leadership IMPROVED CONDITIONS: SEGREGATION OR INTEGRATION? experience in Roma civil society. He received The Nyíregyháza case study shows how the training in project management and was effective interpretation of project goals may serve in raising resources for the horseradish project 108 Project Experience in Hungary and other activities from a wide range of public In Bagamér, the collapse of the market in 1998 cre- and private sources. He is widely accepted by the ated significant difficulties for many producers in community, and his staff observed that he the area. Agricultural crises and market vagaries "speaks the language of the donors." In addition also significantly affected the profitability and to funds from the Autonómia Foundation, the self-sufficiency of the social land programs in Association received resources from the Public Zsadány. In neither case would the projects have Foundation for National and Ethnic Minorities in survived without significant outside support. Hungary, the Soros Foundation, the Ministries of Youth and Sports, and Social and Family Affairs LACK OF MONITORING AND EVALUATION. Finally, most of and PHARE. Over time, the experience and cred- these projects highlight the need for clear guide- ibility of the Association have grown, in part due lines and rigorous monitoring and evaluation to the personality of its leader, as well as the Asso- mechanisms. Programs financed from budgetary ciation's involvement in a number of other com- and private sources are not coordinated and often munity programs in social welfare, education, appear to be randomly selected. The majority of health promotion, and crime prevention. projects have no mechanisms for monitoring or The dominance of personality in project lead- evaluation. The Public Foundation for the Gyp- ership also has its risks. The Bagamér case illus- sies in Hungary has been the only public sponsor trates that while a strong leader can motivate and to set up a monitoring system in addition to move a project forward, such leadership can also strictly collecting loan installments. The Autonó- limit transparency if the leader relies on inside mia Foundation is the only non-governmental connections and networks in securing resources sponsor that regularly monitors its programs. and selecting project participants and staff. In While Autonómia's monitors are prohibited from Bagamér this has led to resentment and tensions giving advice or practical assistance to beneficiar- within the community. ies, they follow the progress of the projects to Leadership also played an important role in completion, and have at least one clear criterion the Zsadány case. The local government, headed for success, namely the proportion of loans by the mayor Árpád Dudás, has worked hard to repaid. In most other cases, supervision of the secure a variety of public works programs, as well programs and the utilization of funds is at best, as the social land program, for reducing rural irregular. These issues are discussed further in the poverty. Dudás is widely perceived as the engine final chapter. of the social land program. His combination of rel- evant experience and commitment has con- CONCLUSIONS tributed to the relative success of the program, the The proliferation of Roma policies and proj- ongoing support of the Ministry of Social and ects in Hungary is impressive. The government Family Affairs,26 and general social cohesion has made significant strides in creating and estab- between Roma and non-Roma participants. lishing the institutional framework for the protec- tion of minorities generally, and Roma in particu- LOCAL ECONOMIC ENVIRONMENT. The broader eco- lar. This includes the establishment of the nomic context within which projects are under- Minority Self-Government system, a Hungarian taken also has important implications for project initiative unique in Europe. These efforts have success. For example, the horseradish project in been supported and supplemented by a large and Bagamér was able to draw on existing growing amount of project activity undertaken resources--a previously developed network and by a wide variety of community-based organiza- market, cultivation experience, and expertise of tions. Nevertheless, barriers remain to the more workers--which contributed to its relative suc- effective functioning of this growing network of cess. However, agricultural projects tend to be governmental and nongovernmental entities. particularly vulnerable to outside shocks, and as Responsibility for policy development on a result may be more risky than other initiatives. Roma issues, coordination, and implementation 109 Roma in an Expanding Europe: Breaking the Poverty Cycle have been distributed among a number of govern- NOTES ment bodies, leading to challenges in transparency, 1. See http://europa.eu.int/comm/ accountability, and coordination. For example, the enlargement/hungary. implementation of the Medium-Term Package for 2. The ten EU candidate states in this category the integration of Roma has been hampered by a are: Cyprus, Estonia, Hungary, Latvia, Lithuania, lack of clarity over institutional mandates, ongoing Malta, Poland, the Slovak Republic, the Czech difficulties in coordination across government Republic, and Slovenia. agencies, as well as insufficient funding for all of 3. See http://europa.eu.int/comm/ the programs included. Further, while many gen- enlargement/hungary. eral policies aimed at assisting marginalized and 4. More recently, it has been estimated that minority groups do benefit Roma, much of Roma there are some 50,000 registered NGOs in Hun- policy itself remains poorly integrated into broad- gary (http://www.autonomia.hu/english/ er social policy in Hungary. Finally, because of the indexen.html). high degree of decentralization in Hungary, signif- 5. For a list of Roma and other NGOs dealing icant challenges remain in translating national pol- with issues related to sustainable development and icy into local implementation, in large part due to advocacy, see the Non-Profit Information and a lack of effective monitoring, evaluation, and Training Center http://www.niok.hu/indexe.htm. enforcement. 6. The other recognized minorities are: Arme- The Minority Self Government (MSG) system nians, Bulgarians, Croats, Germans, Greeks, has had mixed results. It has helped to raise the Poles, Romanians, Ruthenians, Serbs, Slovaks, profile of Roma issues in Hungary and has in- Slovenes and Ukrainians (PER 1998). creased access to national and local policymaking 7. As this chapter was prepared as the Roma in areas concerned with minority education and Office was being set up, it focuses on the previous culture. Moreover, many Roma MSGs have be- structure. come active politically and socially in important 8. See the "Summary of measures taken by ways within their communities. Despite this, the the Government affecting the Roma minority capacities of both the national and local Roma over the past two years", http://www.meh.hu/ MSGs have been limited by a combination of nekh/Angol/roma_summary.htm. insufficient finances, weak political competencies 9. In 1994 the Gandhi Foundation--a joint and influence, and a general lack of authority and government and private initiative--established a legitimacy. Some observers cite their relatively high school and dormitory, the Gandhi Gymnasi- limited mandate to "cultural issues" as insuffi- um, at Pécs, primarily aimed at educating talent- cient in addressing the real needs and concerns of ed Roma youth. The school is managed by the Roma communities. Foundation but financed through the Ministry of Despite these challenges, the substantial poli- Education. cy and project experience in Hungary provides a 10. The new Inter-Ministerial Committee for rich foundation from which considerable learn- Gypsy Affairs as the new coordinating body for ing and lessons have been derived. Efforts to the Medium-Term Package was given greater improve monitoring and evaluation will further power to appoint subcommittees. In addition, it enhance the ability to translate this learning into provides greater consultative access to Roma policy and project development and implementa- social organizations who, by invitation, may tion better able to meet Roma needs and facilitate attend up to four of the Committee's sessions per integration. To these ends, the Hungarian gov- year. The Parliamentary Commissioner for ernment remains committed to improving and Minority Rights and the Director of the Gandhi expanding its efforts as indicated by its plans for Foundation have standing invitations to all Com- the future adoption of a long-term strategy for the mittee deliberations (Kállai and Törzsök 2000). integration of Roma, accompanied by compre- 11. See also the "Summary of measures taken hensive antidiscrimination legislation. by the Government affecting the Roma minority 110 Project Experience in Hungary over the past two years," http://www.meh.hu/ 19. Housing expenditures do not include sub- nekh/Angol/roma_summary.htm. sidies for home construction. 12. See the Minorities Ombudsman home page 20. These figures are considered to be under- at http://www.obh.hu/nekh/en/index.htm. representative of the total activity of ministries on 13. Concerns about the accountability of Roma projects; however, more detailed and com- NMSG representatives have been raised due to prehensive information on these projects was dif- the lack of formal mechanisms and the electoral ficult to obtain. college style of representation by which neither 21. Drawn from original case study by Gábor members of the minority, or the general Hungar- Bernáth (2000). ian population have any direct say in the compo- 22. Throughout Hungary, a disproportionate sition of the NMSG (Kováts 2001b). number of students are designated as mentally 14. According to Article 27 of the Minorities disabled. According to a 1996 OECD report, 35 Act of 1993, by law, the NMSG independently children in one thousand were labeled mentally may take responsibility for the establishment and disabled. This was in comparison with two in maintenance of institutions to support the devel- Turkey, four in Finland and nine in Italy. In the opment of national identity and culture including small villages in Borsod-Abaúj-Zemplén County, a theater, museums, an institute for the arts/sci- Roma students are automatically sent to special ences, and a minority library. They also may take classes for the disabled. This has been the practice responsibility for the maintenance of secondary for years. Ninety percent of Roma students in that and higher educational institutions with country- county attend special classes. wide coverage, and the establishment of legal 23. Drawn from an original case study by advisory services. Zsolt Zádori (2000). 15. For example, the first Roma NMSG set up 24. Drawn from an original case study by its own form of intermediary representation (23 Zsolt Zádori (2000). regional offices as of 1997) from its own resources 25. Drawn from an original case study by in order to facilitate the link between the NMSG János Zolnay (2000). and the 477 Roma MSGs. Research from 1998 26. The Ministry has supported the Zsadány indicated that these actions were of mixed suc- initiative every year since 1995 except for 1997 cess, in part due to the unofficial status of these when, for reportedly politically motivated rea- offices (Kováts 2001b). sons, the subsidy was halted. In 1997, the 16. See www.meh.hu/nekh/Angol/data2_ Autonómia Foundation stepped in with financ- 2000.htm ing of 1.7 million forints, which allowed the agri- 17. The team was led by János Zolnay, and cultural initiative to continue. Ministry financial included: Gábor Bernáth, Angéla Kóczé, József aid was resumed again for 1998, 1999, and 2000. Kolompár, Katalin Kovács, and Zsolt Zádori. 18. Of the total projects in the inventory, 93 percent could be mapped. 111 Chapter Six: ROMA IN SPAIN T he situation of Roma in Spain provides a tained traditional niche occupations as craftsmen, useful counterpoint to the analysis of Roma traders, or seasonal farm laborers. Similarly, in the countries of Central and Eastern although most Roma in Western Europe are set- Europe discussed so far, with both important sim- tled, there are more nomadic Roma in Western ilarities and differences. Sizable Roma communi- Europe than in the countries of Central and East- ties live in many of the existing member countries ern Europe. of the European Union, but the largest population Roma in Western Europe have also not expe- lives in Spain. Thus, Spain has a wide range of rienced the widespread upheaval in their eco- project experience, both positive and negative, to nomic circumstances brought about by transition draw upon. This chapter explores these policy in the East. The economic conditions for Roma in approaches in order to place developments in Western Europe, including access to social servic- Central and Eastern Europe in an emerging Euro- es and employment opportunities, have been rel- pean context. atively stable. At the same time, rising xenopho- To frame this discussion, it is important to rec- bia and anti-immigration sentiments are causes ognize some significant differences in the experi- for concern across Europe. While the majority of ence of Roma on both halves of the European con- Roma are not immigrants or foreigners in the tinent. Exclusion from the labor market and countries where they live, they are often consid- economic opportunities have been a long-term ered as such and bear the brunt of prejudice and phenomenon for many Roma in Western Europe. discrimination. An overview of Roma living con- In contrast, Roma in Central and Eastern Europe ditions and policies in Spain provides insight into had jobs during the socialist period. As a result, the conditions for addressing Roma poverty in an many have high expectations that the govern- expanding Europe. ment will step in to provide jobs and services. This sentiment lies behind much of the frustra- ROMA IN SPAIN: A WESTERN tion expressed by Roma in Central and Eastern EUROPEAN EXAMPLE Europe. Spanish Roma face many similar issues to Levels of integration and relations with non- their eastern counterparts, particularly in access Roma also differ between Western and Central to opportunities on the labor market and educa- and Eastern Europe. Socialism required a large, tion, housing, and living conditions. Because of settled labor force without a high level of skills or Spain's higher level of economic development, education. Assimilation efforts thus focused on levels of poverty and social exclusion among erasing specific national, ethnic, and cultural Spanish Roma (gitanos) are relatively lower than identities, while drawing Roma into the formal those faced by many in Central and Eastern labor force. In contrast, in Western Europe, with Europe. Nevertheless, Roma in Spain have had a its more diversified labor markets, the integration long and tumultuous history, and currently face process has generally been less systematic and many of the same challenges, if to a lesser degree, sustained. Without the concerted employment as Roma in Central and Eastern Europe in terms campaigns associated with socialist industrializa- of social exclusion, poverty, and discrimination tion, many Roma in Western Europe have main- (Box 6.1). 112 Roma in Spain Box 6.1: The History of Roma in Spain Little is known about the origins of the Spanish Roma, due to their early migrations and the absence of a written history. The historical experience of Roma in Spain is marked by five distinct periods in the evolution of Spanish government policy.1 Until 1499: Acceptance The first Roma to reach Spain are thought to have arrived between 1415 and 1425. Between their arrival and 1499, Roma were generally accepted by the Spanish population. They were thought to be Christian pilgrims and were valued for their trades and skills. 1499­1633: Expulsion Persecution of Roma began with the reign of Ferdinand and Isabella of Aragon and Castile in the late 1400s with their efforts to create a homogeneous Catholic state. Ethnic and religious minorities, including Roma, were ordered to either assimilate or leave the country. Non-integrated Roma were branded as highway robbers, thieves, and sorcerers. Although faced with the prospect of expulsion and the loss of their language, many Roma decided to stay in Spain, while at the same time attempting to preserve their traditional way of life. 1633­1783: Forced Assimilation With the economic growth in Spain of the early 1600s, policies toward Roma shifted from expulsion to forced assimilation. Various laws were passed in an attempt to end the nomadic lifestyle of Roma, and settle them. The government hoped that Roma would simply seek formal employment and assimilate into the larger population. Again, however, Roma overwhelmingly managed to maintain their traditional way of life outside of mainstream society. 1783­1939: Incorporation and Legal Equity Following the late 1700s, Spanish Roma experienced a period of formal legal parity, accompanied by con- siderable discrimination and exclusion in practice. In 1783, Charles III signed a decree which formalized legal equality between Roma and non-Roma citizens. The establishment of anti-Roma laws after that date was for- bidden, and Roma were not to be singled out as a distinct ethnic group in official texts.2 These actions were followed by a period of relative incorporation, when further attempts were made by the government to extend the rights of Roma and to reduce anti-Roma sentiments. For example, the Constitution of 1812 stressed the recognition of legal equality for the Roma, granting Roma the full rights and responsibilities of citizenship. At the same time, the government gave little attention to improving their social and economic status within Spain. During this period, there were no government initiatives to assist the Roma. 1939­present: From Dictatorship to Democracy These general trends continued through the Spanish Civil War and the onset of the Franco dictatorship in 1939. Under Franco, Roma were openly discriminated against and prohibited from speaking cálo in public. The Spanish National Guard classified Roma as a "dangerous group of people" to be dealt with cautiously. After Franco's death in 1975, King Juan Carlos assumed the throne and began the democratic transition. This marked a shift in government policy toward addressing Roma issues more openly. The transition was a time of general change and re-incorporation in Spain, with an emphasis on democratic and human rights for all Spanish citizens. Article 14 of the Constitution guarantees equality and full citizenship, and prohibits discrimi- nation on grounds of racial origin, religion, or gender. Formally, the post-1978 policy was one of "assisting in the development of the Gypsy people and the recognition of the fact that the Gypsies have their own cul- ture"(Gamella 1996). Notes: 1. Unless otherwise noted, the historical background is drawn from Gamella (1996), Martín (2000). 2. This law has made the collection of data on the Roma population extremely difficult, as the 1783 action strongly discouraged the dis- tinction of the Roma community in data collection and lawmaking. However, the collection of data based purely on ethnicity was techni- cally not made illegal until the Constitution of 1978.pendence from government supervision was instituted through the Law on Association (Law 1990/II, January 1989) and an amendment to the Civil Code (Law 1990/I, January 1990). 113 Roma in an Expanding Europe: Breaking the Poverty Cycle The most recent government estimate of the The National Program for the Development number of Roma in Spain, from 1999, is just over of Roma 630,000. However, as the Spanish Constitution of The situation for Spanish Roma has changed 1978 prohibits the collection of data on the basis substantially in the post-Franco era. Significant of ethnicity, these numbers are disputed. Govern- gains have been achieved through the overall ment officials, NGOs, and academics generally improvement in economic conditions throughout agree that the population ranges between 400,000 Spain. These developments have had a significant and 600,000.1 Spain thus has the largest popula- positive impact on the advancement of Roma, tion of Roma in Western Europe, close to the pop- through improved access to public housing, edu- ulation in Hungary. Following Spain, the largest cation, health services, and social assistance populations of Roma in Western Europe are in (ASGG 2001). Greece, Italy, France, the United Kingdom, and In 1988, the government began the implemen- Germany (Table 6.1). tation of the National Program for the Develop- The majority of Roma in Spain speak Spanish, ment of Roma (NPDR), which marked an impor- however, a significant number speak the Roma tant turning point in recognizing the exclusion of language, caló.2 It is not known how many Roma Roma and formulating policy strategies (Villareal speak caló, although estimates range widely 2001). The main goals of the NPDR are to between 40,000 to 140,000.3 The primary distinc- improve the quality of life for Roma, foster equal tion between groups of Roma is made between opportunities, promote the inclusion of Roma in Spanish and Portuguese Roma. Portuguese Roma Spanish society, and improve relations between mostly reside in the western part of the country, Roma and non-Roma. Despite significant im- and speak a slightly different dialect of caló. provements and government policy efforts, the exclusion and poverty of Roma in Spain persist in Table 6.1: Roma Population Estimates, Selected Western European Countries Government Estimate1 Council of Europe Estimate2 Minority Rights Group Estimate3 Austria -- 20,000­25,000 20,000­25,000 Finland 10,000 (1998) 10,000 7,000­9,000 France -- -- 280,000­340,000 Germany 50,000­70,000 (1996) 70,000 100,000­130,000 Greece 150,000­300,0004 80,000­150,000 160,000­200,000 Italy 130,000 120,000 90,000­110,000 Portugal 40,000 (1997) -- 40,000­50,000 Spain 630,000 (1999)5 -- 700,000­800,000 Sweden 20,000(1996) 40,000­50,000 15,000­20,000 Switzerland -- 35,000 30,000­35,000 United Kingdom 90,000 300,000 90,000­120,000 -- Not available. Sources: 1. Estimates submitted to the UN Committee on the Elimination of all Forms of Racial Discrimination (except Greece, Italy, Spain and the United Kingdom); see http://errc.org/publications/factsheets/numbers. 2. Council of Europe, 2002 (Questionnaire on the Legal Situation of Roma/Gypsies/Travellers in Europe), http://assem- bly.coe.int/documents/workingdocs/doc02/EDOC9397.htm 3. Liegeois and Gheorghe (1995). 4. In 1997, the General Secretariat for Adult Education estimated the number of Roma in Greece to be between 150,000 to 200,000; the year before they were estimated at around 300,000. 5. Estimate by a Subcommittee of the Spanish Parliament (1999). In 1995, the government estimate submitted to the UN Committee on the Elimination of all Forms of Racial Discrimination was between 325,000 and 400,000. 114 Roma in Spain many areas, suggesting that continued and spe- cific actions are still needed to further improve mission responsible for program oversight,5 the their welfare. Inter-Ministerial Working Group responsible for The NPDR was endowed with an annual coordinating sectoral initiatives among govern- budget of around 500 million pesetas (approxi- ment ministries; and the Consultative Commis- mately US$4 million), with matching funds prom- sion, comprised of Roma and non-Roma represen- ised from regional and local governments. Since tatives,6 whose aim is to ensure cooperation 1989, roughly one billion pesetas (US$8 million) between government and NGOs in the implemen- have been spent annually on projects targeted at tation of the NPDR and to represent the main Roma. Additionally, starting in 1989, 0.52 percent issues affecting Roma to the other commissions. of the net personal income tax collected has been Responsibility for implementing the program allocated to supporting the Catholic Church and rests at the regional level. Projects are chosen by various NGOs.4 This program has channeled an regional governments (Box 6.2). Once chosen, additional 200 to 500 million pesetas to NGOs they are submitted to the Ministry of Labor and which work with the Roma community. Social Affairs at the federal level, where they A central administrative body, the Roma must fulfill certain requirements to be considered Development Program Service Unit, was estab- for funding. Decisions on the selection of projects lished to support and coordinate the NPDR with- and funding are made jointly by both federal and in the public administration, and to provide tech- regional governments. National funding must be nical and financial assistance to NGOs. This cofinanced by regional and local authorities assistance includes facilitating participation of which are required to contribute at least 40 per- Roma in official institutions, organizing training cent of the total project cost (Villareal 2001). programs for professionals working with Roma, and promoting greater awareness about Roma. Project Activity Further coordination of the program is carried According to the annual report of the Service out by three commissions: the Follow-up Com- Unit of the NPDR, an average of 100 projects Box 6.2: The Andalusian Plan for the Roma Community Approximately 43 percent of Roma in Spain live in Andalucia. The region of Andalucia has made addi- tional efforts to improve the living conditions of Roma. In 1996 the Andalusian Government approved a "Comprehensive Plan for the Gypsy Community," which became operational in 1997. The primary task of the Plan is to coordinate activities concerning Roma. This function is considered particularly important because of the large number of programs and projects implemented in the region. Andalusia receives the largest share of money from the National Program for Roma Development (almost half of the total budget). Andalusia is also the largest beneficiary of the European Social Fund, because of its relatively lower levels of development than other regions in the country. The majority of initiatives targeted at Roma in Andalusia are small scale and highly localized. The programs are financed by a combination of: · Transfers to the NPDR from the Ministry of Labor and Social Affairs (in 1997, close to 220 million pese- tas, or 60 percent of estimated project costs) plus matching funds from the Andalusian government for 40 percent of total project costs; · Transfers from the 0.52 percent personal income tax for non-profit organizations and/or associations, the majority of which go to the Federacion de Asociaciones Romanies Andaluzas. · Contributions from various European Social Fund programs. While these programs are open to the broader community, in some cases 80 to 90 percent of the participants are Roma. Source: Martín 2000. 115 Roma in an Expanding Europe: Breaking the Poverty Cycle have been implemented annually since 1995, LOCAL GOVERNMENT INVOLVEMENT. For all NPDR inter- with a peak of 120 projects in 1998. Of the more ventions, central government funds must be than 500 employees responsible for the imple- cofinanced by regional autonomous communities mentation of the projects each year, approxi- or local administrations. National authorities mately 22 percent are Roma. There are an esti- coordinate, finance, and carry out followup activ- mated 50,000 direct beneficiaries per year, or ities, but the allocation of resources and project approximately 12,000 families. implementation are done by regional and local There is considerable diversity in the kinds of governments. This decentralized system has projects being implemented.7 The majority are helped to place the Roma issue on the agenda of carried out in the fields of education (including regional and local governments. prevention of school absenteeism, extracurricu- lar activities, and adult education), social assis- ROMA INVOLVEMENT. Another strength of the NPDR tance, housing (including renovations and reset- is its emphasis on fostering Roma participation. tlement support), health education (including This is achieved in two ways. First, where possi- courses for young mothers and drug abuse pre- ble, the NPDR recruits Roma personnel to work vention programs), and vocational training on the projects, and to participate in the training courses. A few projects have focused on cultural and development of Roma mediators, teachers, activities, including Roma language classes or and social workers. Second, a portion of the funds cultural exhibitions. have been spent on supporting Roma associa- In addition to sectoral projects, in 1992 the tions, which have played an active and important NPDR Unit began sensitivity training programs role in project implementation.8 for regional and local civil servants, aimed at improving the ability of regional and local FOCUS ON ACCESS TO SOCIAL SERVICES. In the 1980s, administrators to address Roma issues. Diversity social welfare services became universally avail- awareness has also been promoted through infre- able to all citizens of Spain, including access to quent roundtables bringing together Roma repre- education, health, general social services, and spe- sentatives and civil servants. The NPDR Unit also cialized social services (e.g. for disadvantaged chil- lodges frequent complaints against negative por- dren, and the elderly). The Program works to inte- trayals of Roma in the media. grate Roma more effectively into these mainstream For Spain's Roma population, the NPDR social service and social assistance networks marked a significant milestone because it repre- through outreach and specialized programs. For sented the first time the national government offi- example, in Andalusia, children's vaccination and cially recognized the specific issues faced by family planning programs for Roma are part of the Roma and established concrete, nationwide mainstream public programs. measures to address them. The NPDR exhibits a number of strengths and weaknesses. Weaknesses Strengths WEAK LEGISLATIVE STATUS. There are also a number of ongoing concerns related to the ability of the CENTRALIZED CONTACT POINT. The NPDR Unit pro- NPDR to effectively carry out its mandate. One vides an important contact point for organiza- concern is that the NPDR does not enjoy the sta- tions, individuals, policymakers, and members of tus of a legislated plan. While the NPDR was ini- Parliament working in the field of Roma issues. tially introduced as a bill to Parliament, it was The Unit provides a focal point for information never passed. This lesser status may threaten the sharing, and facilitates meeting a wide range of long-term financial sustainability of the NPDR. interests over program goals, project, and imple- mentation strategies. LACK OF MONITORING AND EVALUATION. A significant concern is the lack of systematic monitoring and 116 Roma in Spain evaluation of projects and programs. In the major- al collaboration, the FSGG is governed by a Board ity of cases, the only documentation available are of Trustees, half of whom are Roma. In 2001, expenditures, or the project implementer's own roughly 40 percent of the 647 members of the subjective evaluations of the project's success or total staff were Roma; and 67 percent of the total failure to meet its expected objectives. were women.11 The activities of the FSGG have been growing steadily over the last 38 years, with A Spanish NGO to Watch: the Fundación significant expansion in the last couple of years. Secretariado General Gitano Between 2000 and 2001, the number of projects The case of the Fundación Secretariado Gene- increased from 30 to 38. Over this same time peri- ral Gitano (FSGG)9,10 provides a useful example od, the number of direct beneficiaries grew from of the type of project activity which has resulted 29,000 to 64,000, with a corresponding increase in from NPDR financing. The FSGG is the largest financing for projects from around 4.6 million to and most prominent Spanish NGO working 8.4 million Euro. toward the advancement and integration of The majority of financing for the FSGG comes Roma. It is an example of a strong NGO with from the Spanish central government (roughly 36 experience working with the Roma community percent) and European sources (approximately 27 which has successfully promoted the develop- percent) and in particular, the European Social ment of collaborative relationships with a range Fund (see Box 6.3 on the Acceder Project). Signif- of government, private, and international entities. icant financial support also comes from Autono- The organization began operating during the mous Community and local governments mid­1960s, but did not become a legal entity until (around 36 percent). In recent years, the FSGG 1982. In keeping with its emphasis on intercultur- has pursued more collaborative initiatives, Box 6.3: The Acceder Program: Training and Employment Services The Acceder program began in 1998 as a two-year pilot project in Madrid, and has subsequently expand- ed to a national program (ASGG 2000; ASGG 2001). The national program is currently being implemented throughout the principal municipalities in Spain (a total of 34) in 13 different Autonomous Regions. The pro- gram is administered by the ASGG, a national, non-profit organization working for the advancement of Roma, which receives financial support from the National Program for the Development of Roma, via per- sonal income tax contributions. The main objectives of the program are to: (i) provide Roma with professional qualifications and access to work contacts by addressing their needs and those of employers; (ii) increase the accessibility of general vocational training and employment services to unemployed Roma; and (iii) raise awareness of discrimina- tion against Roma and work to improve society's view of the community. The program provides individualized support to participants in identifying and preparing for employ- ment. While the program is open to all interested applicants, 79 percent were Roma in 1999. Roma mediators work closely with job-seekers and employers to identify their skills, training needs, and employment oppor- tunities. The mediators provide support to applicants throughout the training and job search process. In 1999 there were 304 active job seekers enrolled in Acceder and 63 percent found employment. How- ever, the job retention rate is not known, and cost-benefit analysis of the program is not available. Staff of the Asociación and participants noted that the strengths of the program are its individualized approach in assess- ing and matching skills and jobs and the use of mediators who can bridge the gap between gitanos and non- Roma. Challenges include the difficulty of providing adequate and appropriate training for individuals with low education levels, persistent discrimination on the labor market, and work incentives. Participants may be reluctant to accept low paying jobs and risk losing access to social assistance benefits. Sources: ASGG 2000; 2001; Martín 2000. 117 Roma in an Expanding Europe: Breaking the Poverty Cycle including cofinancing, with close to 60 public and coordination with other NGOs and local govern- private organizations. ments. For example, the FSGG has developed The FSGG is engaged in a wide variety of ini- three territorially-based, integrated programs tiatives in vocational training and employment, through agreements with the governments of education, health, youth, women, and regionally Madrid, Aranjuez, and Castilla. Also, under the based social interventions (integrated action pro- auspices of the Acceder Program, the FSGG has grams). In general, nearly half of the budget goes been active in joint efforts with 13 autonomous to employment programs, nearly 20 percent to communities. The organization also has worked education initiatives, and just under 12 percent to closely with the European Commission on a health, youth, and women's programs. The most series of multicultural pilot projects focusing on prominent initiative to improve Roma inclusion integration (identifying good practices in combat- into the labor market is FSGG's involvement in ing discrimination against Roma) and identifying the Acceder Project (Box 6.3). measures to combat social exclusion. In 1999, FSGG (then ASGG) started working in several EDUCATION. Among the education programs that Central and Eastern European countries, and has FSGG has initiated are a series related to educa- technical assistance in the Czech and Slovak tional mainstreaming, focused on improving Republics and Hungary. Roma access and integration into the compulsory education system, reducing absenteeism, improv- LIVING CONDITIONS OF SPANISH ROMA ing performance, and encouraging positive rela- While Roma live throughout Spain, they are tions between Roma and non-Roma (Box 6.4). geographically concentrated in four regions, or Extracurricular activities are also offered, as well "autonomous communities," of the country (Table as economic and tutorial support for Roma stu- 6.2). Almost half of the total Roma population (43 dents interested in university education. The percent), are found in the Southern province of FSGG supports a variety of training programs, Andalusia. Madrid has the second highest concen- including teacher training, and vocational train- tration of Roma with nearly 10 percent, followed by ing for the socially disadvantaged. Catalonia and Valencia at close to 9 percent each. Despite perceptions to the contrary, Spanish HEALTH. To promote the improvement of Roma Roma generally live in permanently settled com- health, the FSGG works to improve Roma access to munities. In some autonomous communities, 87 health services through mediation and information percent of Roma have lived in the same munici- services. Projects have also been implemented palities for 15 years or more (Gamella 1996). Nev- which offer technical assistance to organizations on ertheless, for many, the concept of mobility is still specific Roma health issues, including HIV/AIDS an important element of social organization and prevention. Additional actions have focused on the culture (Gamella 1996). There also has been a prevention of drug abuse among Roma youth, and trend toward greater urbanization. Many Roma public drug-abuse health services. The FSGG sup- have moved from rural to urban areas in recent ports the European Community funded project decades (Fresno 1994). entitled "Health and the Roma Community." Roma in Spain share a similar demographic profile to that of Roma in Central and Eastern WOMEN. In partnership with eight Roma associa- Europe. Historically, birth rates among Roma are tions, the FSGG has a number of programs focused higher than for the majority population. Over the on advancement and support for the development past five centuries, the population has grown to of Roma women, focusing on health education, lit- over 30 times its original size. In comparison, eracy, and integration into the labor market. over the same period the Spanish population increased 10 times from 4 million to 40 million. COLLABORATIVE EFFORTS. One of the most important The Roma population is much younger than features and strengths of the FSGG is its active the majority population. Approximately 40­50 118 Roma in Spain Box 6.4: The School Monitoring Program, Municipality of Madrid The School Monitoring Program1 began in three districts in the Municipality of Madrid as a part of a collab- orative Plan of Action, launched in January 1999, by the Madrid City Council and the Asociacion Secretariado General Gitano (ASGG, now FSGG) (see Box 6.3).2 The general aim of the Program is to facilitate the integration of Roma children into the regular school system. The main objectives of the program are: · To promote the increased participation of Roma children in pre-primary education (0-6 years). · To promote the continuation in and completion of compulsory education (6-16 years). · To develop greater skills in terms of school habits, constructive relationships among classmates, and class- room learning techniques. · To promote the involvement of the Roma families in the educational process. · To carry out an ongoing diagnosis of the school situation of Roma pupils. Four complementary sets of interventions are included in the project. The first involves efforts by Roma medi- ators and trainers to make contacts with teachers and social workers to identify the main problems encountered by Roma pupils and the school (e.g., absenteeism, school conflicts, under-performance, etc.), as well as to increase teacher and staff awareness of Roma culture. The second involves the monitoring of Roma pupils' attendance and performance, including home visits to families to encourage greater support and involvement of parents in their children's education. To develop additional skills, motivation, and cultural confidence, children are recruited to participate in a series of complementary extra-curricular activities developed by the Centers (e.g., dance classes, sports activities, field trips, visits to museums, training workshops). Finally, efforts are undertaken to increase the number of Roma children in preschool education. Earlier exposure to the education system is expected to increase children's overall skill levels, and to improve their familiarity and comfort with formal education. Results and Challenges As of 2000, the program has been carried out in 16 state schools, selected from three districts involved in the Plan.3 In 1999, the Program followed the progress and assisted 314 Roma pupils, including with 220 visits to fam- ily homes. 174 pupils participated in the formation of 14 workshop groups focusing on traditional Andalusian song and dance. In 2000, increased attention was given to the problem of continued school attendance of children aged 12 to 16 years with particular emphasis on extra-curricular courses/workshops focused on practical vocational skills (e.g., carpentry, bricklaying). In the first 3 months, the program increased the number of interventions, undertak- ing the monitoring of 136 pupils, 150 visits to families, and assisting 568 pupils through extra-curricular activities. Results as of 2000 are difficult to assess in light of the short time since implementation. It was too early to assess increases in educational outcomes. However, there were signs of progress. In addition to the inclusion of an increasing number of Roma pupils in education support initiatives, progress is evident in terms of noticeably lower rates of absenteeism and high levels of participation in and motivation for extra-curricular activities. In particular, Roma participation in activities related to Roma culture (e.g., workshops on traditional Andalusian song and dance) is reportedly high. Moreover, this program has increased constructive contact between Roma and non-Roma, as well as improved awareness of Roma issues within the educational system. The Program reported a number of on-going challenges including an inability to conduct home visits to all those families in need. Initially, interventions with families were conducted in an unstructured, ad hoc manner by the trainers which in some cases were ineffective at generating greater parental understanding and involve- ment. As a last resort for children not attending school until the legal age of 16, education authorities may open a file on the pupil and impose fines on the families. However, in the Municipality of Madrid, families reported- ly rarely pay, and collection is rarely enforced. Program efforts have been taken to improve on these aspects of this process. In addition, proposals have been put forth suggesting the need for individual tutorials for children with greater learning needs, special training for teaching staff in Roma culture, and the production of education- al materials that better reflect Roma culture and interests. Notes: 1. This Program is also know as the "Program of Support and School Monitoring of Infants and Gypsy Youth." 2. Centers to carry out the implementation of the Plan were set up in the districts of Carabanchel (Pan Bendito), Villaverde/Usera (El Espinillo) and Puente de Vallecas (Adali Cali), with an additional enter responsible for overall coordination and management located in the ASGG headquarters in Carabanchel. 3. The schools were chosen on the basis of a set of criteria including: a minimum percentage of Roma pupils, school proximity to the Plan of Action centers, interest of the teachers and administrators. Source: Martín 2000. 119 Roma in an Expanding Europe: Breaking the Poverty Cycle Table 6.2: Roma Population per Autonomous Community (estimates, 1993­1999) % of total Roma Autonomous Community Estimate for 1993 population (Spain) Estimate for 1999 Andalusia 157,097 42.8 286,110 Aragon 10,961 2.7 18,209 Asturias 2,877 0.8 4,780 Balearic Islands 6,877 1.9 5,423 Canary Islands 515 0.1 854 Cantabria 2,320 0.6 4,021 Castile-Leon 20,198 5.5 28,339 Castile-La Mancha 17,072 4.7 33,552 Catalonia 31,881 8.7 52,937 Extremadura 6,811 1.9 11,318 Galicia 7,374 2.0 13,741 Madrid 35,588 9.7 59,082 Murcia 19,877 5.4 33,006 La Rioja 4,433 1.2 7,361 Valencia 31,585 8.6 52,455 Navarra 3,593 0.9 5,954 Basque Country 7,028 1.9 11,675 Ceuta and Melilla 1,222 0.3 2,030 TOTAL 367,039 99.6 630,847 Source: Ministry of Employment and Social Affairs of Spain (El Ministerio del Empleo y Asuntos Sociales de Espańa). percent of Spanish Roma are below the age of 16 dered many traditional Roma occupations irrele- (Giménez Adelantado 1999; ASGG 2001). This vant or obsolete (e.g., blacksmiths, horse dealers, can be attributed in part to high birth rates. While farm hands, and peddlers). Many rural Roma the birth rate for Roma in Spain is unknown, in have been compelled to move to the cities in Andalusia, the birth rate is estimated to be 23.8 search of employment. per thousand, compared to 13 per thousand for The labor market characteristics of Roma in non-Roma Andalusians, and 10 per thousand for Spain differ substantially from those of the rest of the total population of Spain (Gamella 1996). the population. Few hold salaried full-time jobs. Roma women marry at a young age, often as Most are engaged in independent, part-time or early as 13 to 14 years old, and have children casual labor. Recent data from the Subcommittee between then and age 30. The average size of a of the Parliament (Congress of Deputies 1999b) Roma family is 5.4 members, in comparison with show that the employment standing of Roma in 3.7 in the average Spanish family (Congress of Spain is characterized by jobs that are low paid Deputies 1999a). Roma also have a lower life and largely in the informal sector. It was estimat- expectancy than the general population, estimat- ed that 50 to 80 percent of Roma work in "tradi- ed at 65 years (Vásquez 1980), compared with the tional professions" of peddling, collecting solid much higher national average of 78. urban waste, and performing seasonal work. Another 5 to 15 percent work as antique dealers, Labor Market Status shop owners, and in the arts, while 10 to 15 per- As in Central and Eastern Europe, Roma in cent work in "new professions" of construction, Spain were historically employed in traditional public works, and as civil servants. trades. Since the 1970s more rapid economic A variety of government and NGO initiatives development and technological advances have have been undertaken to improve access to displaced these jobs. New technologies have ren- employment for Roma. In particular, job training 120 Roma in Spain and related employment services have been pro- However, because these settlements were gener- vided in conjunction with the European Social ally located on the outskirts of cities and towns Fund job training initiatives (see Box 6.3).12 where they were more easily neglected by munic- ipal authorities, the condition of many of these Housing settlements rapidly deteriorated into slums. During the 1970s, government housing policy Roma inhabit approximately 95 percent of the was aimed at eliminating shantytowns and infor- chabolas (makeshift housing and slums) around mal settlements. A state housing program was larger cities in Spain (Congress of Deputies developed to address the housing needs of the 1999b). Around 80 percent of these houses are dislocated, treating Roma and non-Roma equally. smaller than 50 square meters and house more Many Roma were relocated to high-rise apart- than 4 people (Congress of Deputies 1999a). The ment buildings. However, these relocation pro- lack of sanitation and running water in these grams did not take into consideration Roma pref- areas threatens the health of the inhabitants. In erences. For example, the new buildings did not some areas, particularly those on the outskirts of allow for large families to live together and did larger cities, it is reported that this ghettoization not allow Roma to continue with certain occupa- and a lack of police presence has contributed to tions, such as the collection and storage of scrap increases in Roma drug use and trafficking. metals. As a result, a large number soon left their Since the early 1990s, government policies new homes to return to more traditional living have evolved to address the specific needs of conditions (Gamella 1996). Roma families (Box 6.5). These policies in part In the 1980s, shantytowns continued to grow, reflect the acknowledgement that the segregation populated mostly by Roma. Government policy of Roma into isolated communities has inhibited toward Roma shifted toward the creation of small their integration into society, and a recognition towns and housing settlements exclusively for that the deterioration of rapidly built, low quali- Roma. In these towns, more open, one- and two- ty, state-constructed housing has contributed to level houses with courtyard areas were created. social deterioration and illegal activities within Box 6.5: Roma Housing Program in Madrid The Institute for New Homes and Social Integration (IRIS) was created in 1998, and is run by the Madrid Regional Community.1 Funding for IRIS is provided by the National Development Program for Roma through the Regional Autonomous Community of Madrid, with some support from the national government. IRIS has two main objectives: (i) to move slum and ghetto dwellers to improved housing; and (ii) to pro- vide follow-up services for those re-accommodated to facilitate social integration into their new communi- ties. IRIS pursues its objectives by acquiring apartments for Roma families. Along with apartments, IRIS pro- vides follow-up support services. An estimated 1,550 slum dwellings exist in the city of Madrid, with an additional 305 in the region's municipalities. In 1998, 272 families were re-housed with a similar number in 1999. Subsidies secured in 1998 for these re-accommodations totaled 450 million pesetas. To date, the program is generally perceived to be a success, in part because of the rapid pace of re-accom- modation and the low proportion of program drop-outs (less than 2 percent). These successes are attributed to the process of allocating apartments based on consensus. Additionally, the program makes a significant effort to further social integration through the provision of complementary social programs for children's education and inclusion at school, and employment support. Note: 1. IRIS was created after the Consortium for Re-accommodation of Slum Dwellers was dissolved. Part of the Consortium's compe- tence was absorbed by the Municipal Housing Enterprise and partly by IRIS. Source: Martín 2000. 121 Roma in an Expanding Europe: Breaking the Poverty Cycle these communities. Currently, housing issues are percent and 45 percent for men and women evaluated on a case-by-case basis, with the aim of respectively (CIDE 1999). One important factor integrating Roma families into more diverse contributing to the lower illiteracy rates for neighborhoods and Roma children into the main- Roma young people is that the law on compul- stream schools (Box 6.4). A number of new asso- sory education, requiring children between the ciations and NGOs working with Roma housing ages of 6 and 15 to attend school, began to be issues have also been formed. enforced in 1990. Efforts in the housing area over the last 30 Despite gains in literacy, the Spanish school years have yielded mixed results. While the over- system is still not adequately reaching or retain- all success of programs and Roma participation ing many Roma children. In a 1993 report, an esti- has remained relatively low, an increasing num- mated 25 percent of Roma children of school age ber of Roma are taking advantage of better hous- were not enrolled in school (Jiménez González ing opportunities and very few exclusively Roma 1993). According to the same source, of those 75 neighborhoods remain. percent enrolled, 36 percent did not attend school regularly. Other sources report truancy rates that Health Status are sometimes as high as 70 percent (Congress of Reliable data on the health status of Roma in Deputies 1999b). Additionally, the school dropout Spain, as in other countries, is scarce and limited rate is very high, at close to 60 percent for boys to scattered surveys. The information that is and 80 percent for girls (Jiménez González 1993). available paints a worrisome picture. For exam- Most dropouts leave school after age 11, although ple, one study reported a high incidence of birth most boys spend more years in school than girls. defects among some groups of Roma (Martinez- A very small number of Roma finish the Frais and Bermejo 1992). Another study from required 10 years of education in Spain, known 1995 reported a nine times higher prevalence of as the Educación General Básica (Basic General hepatitis A in Spanish Roma children, than in the Education). In 1993, it was estimated that only non-Roma population (Cilla et al 1995). The most about 5 percent of Roma pupils completed, and serious health problems facing Roma in Spain only 1 percent of Roma students succeeded in include: inadequate nutrition, congenital dis- reaching secondary education (university prepa- eases, gaps in vaccination coverage, and drug ration). In 1993, the total number of Roma stu- addiction. HIV/AIDS has also become a concern, dents attending university in Spain was 200. however there is no published data on incidence Another study found that up to 80 percent of or trends among Roma. Roma pupils do not complete basic education and many pupils are two or more years behind Education the average (Santos 1999). While the Spanish education system has taken additional measures over the last decade Barriers to Roma Education to reach Roma students, access to adequate edu- Roma children in Spain face barriers to edu- cation remains a challenge. Literacy, enrollment, cation which are similar to many of the issues dis- attendance, and completion rates are all very cussed in earlier chapters for Roma in Central low among Roma. Illiteracy levels for adult and Eastern Europe, including discrimination, Roma are high, with rates approaching 70 per- cultural perceptions about the role and value of cent (Congress of Deputies 1999b). For the pop- education, and systemic constraints inherent in ulation over the age of 55, illiteracy rates for men the educational system. and women are around 75 percent and 90 per- Schools can be a hostile environment for cent, respectively (CIDE 1999). However, data Roma children. Roma may face discrimination at on younger Roma indicate that illiteracy rates, school, both from non-Roma (payo) parents and while still high, are dropping. For the popula- teachers, as well as school administrators and tion under the age of 25, illiteracy rates were 20 local authorities (Jiménez González 1993). Low 122 Roma in Spain school attendance and completion rates are often thought to contribute to lower levels of atten- attributed to low demand for education among dance and academic performance. Roma families (Roma 2000). Low demand may be due in part to the opportunity costs of education, INSUFFICIENT TEACHER TRAINING. Similarly, there contin- and the need for children, and particularly girls, ues to be a widespread lack of teacher training on to work at home. As basic education does not issues of cultural diversity, such as multicultural guarantee Roma students a job upon completion, education and social and cultural anthropology of many Roma students see few incentives to stay in the minority groups within Spain. Although some school. Concerns also exist about the negative attempts have been made to provide courses on impact of majority values that are transmitted Roma schooling and multicultural education, through the education system on traditional there has not been a concerted and organized Roma culture (Santos 1999). effort to educate teachers on these issues. Finally, a recent study identified a number of shortcomings with the government's current edu- CONCLUSIONS cation policy in their ability to reach Roma stu- The Spanish experience provides a useful dents (Roma 2000). These include deficiencies in example for the countries of Central and Eastern the remedial education system, the lack of multi- Europe. In post-Franco Spain, improving econom- cultural education, and insufficient attention to ic conditions, better social services, European inte- teacher training. gration, and a democratic system have opened opportunities for tackling poverty across the coun- REMEDIAL EDUCATION. Under the Spanish education try, and for Roma in particular. The National Pro- system, disadvantaged students are provided sup- gram for the Development of Roma provides a port through "remedial education." To a large framework for the involvement of regional and degree, remedial programs have evolved into tech- local governments and NGOs in Roma issues-- nical and language training courses to prepare stu- including many Roma themselves. The European dents for (often low-wage) employment. Further, Social Fund--an instrument that will soon be under the program, disadvantaged students are available to the accession countries--has also been provided with school books, meals, hygiene pro- involved in project development and finance. grams, and vaccinations. This system has been crit- This context has promoted innovative proj- icized as perpetuating the segregation of Roma ects which aim to overcome exclusion in educa- children from their non-Roma peers, as well as tion, housing, employment, and other areas. limiting their ability to pursue higher education. While further evaluation is needed, projects such as the Acceder employment project, which pro- LACK OF MULTICULTURAL EDUCATION. School curricula vides Roma with support for entering the main- typically do not include materials on Roma. stream labor market are useful project experience Although efforts are being made to increase edu- for the Central and Eastern European countries. cational materials in schools that teach students In fact, the NGO which runs the Acceder project about Roma in a positive manner, there is still a has consulted in Slovakia and Hungary. The very long way to go. As of 1993, Roma culture experience of Roma projects in Spain has not been was still largely absent from textbooks. One study wholly positive. Lessons from failed housing examined close to 49,000 pages from texts used in projects can also provide cautionary examples. primary and middle school, secondary school, Indeed, the situation of Roma living in Spain is and technical training, and found that only 50 far from ideal. Projects and policies suffer many of lines made any mention of Roma (Calvo Buezas the same weaknesses as other countries, including 1989). The majority of these references to Roma lack of sustainability and an absence of monitoring "were either foolish or negative representations and evaluation. However, the robust NGO commu- of them." The inability of Roma children to iden- nity, the high level of Roma participation in projects, tify with their own history and values in school is 123 Roma in an Expanding Europe: Breaking the Poverty Cycle and a positive track record of initiatives in key social 8. Some of the more prominent Roma associa- areas are a promising base for further progress. tions and nonprofit organizations with programs for the Roma community funded in part by NOTES NPDR are: the Asociacion Nacional Presencia 1. For example, Fresno (1994) estimates that Gitana (ANPG), the Fundacion Secretariado Gen- there are between 400,000 and 450,000 Roma in eral Gitano (FSGG), Union Romani, Federacion Spain. In 1997, the Unión Romani suggested there de Asociaciones Romanies Andaluzas (FARA), were between 500,000 and 600,000, while in 1999, Federacion de Asociaciones Gitanas de Aragon the Subcommittee of the Spanish Parliament esti- (FAGA), Federacion de Asociaciones Gitanas de mated that the size of the population was 630,000. Castilla y Leon, Federacion de Asociaciones 2. Caló (also called Gitano, Iberian Romani, Gitanas de Cataluna (FAGIC), Federacion de Aso- Hispanoromani) is considered a version of Span- ciaciones Gitanas de Extremenos (FAGEX), Fed- ish which is based on a combination of Romani, eracion Autonomica de Asociaciones Gitanas de regional dialects, and Spanish and Portuguese. It la Comunidad Valenciana, Caritas Espanola, and is also spoken in Brazil, France, and Portugal by the Spanish Red Cross. small numbers of people. 9. Information drawn primarily from ASGG 3. Summer Institute of Language (SIL) Ethno- 2001 and www.fsgg.org. logue http://www.ethnologue.com/show_lan- 10. Prior to 2001, the FSGG was known as the guage.asp?code=RMR Asociacion Secretariado General Gitano (ASGG). 4. Taxpayers are given a choice as to which While this chapter refers only to the FSGG, any they wish their money to be allocated to. actions and programs prior to 2001 were under- 5. Established in 1989, this body consists of taken through the ASGG. delegates from the General Directorate of Social 11. In 2001, of the 657 staff, 457 were salaried Action within the Ministry of Labor and Social workers and 190 were volunteers or interns. Affairs, and from the Spanish Federation of Community mediators and educators comprise Municipalities and Provinces. 42 percent of the staff; administrators responsible 6. Formed in 1990, this group is comprised of for the coordination and management of pro- representatives from the General Directorate of grams and teams account for 39 percent of the Social Action and from various national Roma staff. organizations. Roma NGOs joined the commis- 12. Since 1994, The European Social Fund's sion in 1993. INTEGRA program has promoted measures to 7. Due to a lack of systematic monitoring and improve access to the labor market and the evaluation of projects, data on the specific kinds employability of marginalized groups (e.g. long- of projects which are being implemented, the term unemployed, Roma, homeless). allocation of resources and relative weight of spending on different sectors was not available. 124 Chapter Seven: THE ROAD AHEAD T he plight of Roma in Central and Eastern poverty; (ii) the diversity of Roma populations; Europe has not gone unnoticed. During the and (iii) the context of European integration. 1990s, initiatives by governments, NGOs, and international organizations addressed vari- Aspects of Roma Poverty ous issues related to Roma, from human rights to Roma poverty is strikingly high in Central racial stereotyping in the media to education and and Eastern Europe. Poverty rates for Roma in employment. This volume was designed to Bulgaria, Hungary, and Romania are as much as advance these efforts by providing detailed ten times that of non-Roma. As with non-Roma information about the nature of Roma poverty, citizens, poverty among Roma is highest among the course of project experience thus far, and families where the household head has little edu- avenues for future policy. This chapter suggests cation or is unemployed, and among families some lessons learned--first, about the nature of with three or more children. These characteristics Roma poverty and the policy context in Europe; are also found among the non-Roma poor. But for second, about general policy approaches for Roma, the chances of being poor are higher than addressing Roma poverty; and finally, about spe- for their non-Roma neighbors, irrespective of cific policies. education level and employment. The conclusion Improving conditions for Roma is closely is clear: Roma poverty is partly related to low linked to the overall success of each country's educational attainment, limited labor market par- economic and social development strategies. In ticipation, and larger family sizes, but it also this context, policymakers need to make it a pri- stems from factors associated with being Roma, ority to implement policies that promote and sus- including the multiple dimensions of exclusion. tain growth while trying to boost social welfare Qualitative case studies of Roma poverty and ensure the overall inclusiveness of govern- showed that many of the causes of Roma poverty ment policies. But the extent and characteristics are interrelated. For instance, access to health care of Roma poverty indicate that these sectorwide and waste collection is limited in remote Roma policies will not be sufficient. Some areas will settlements. Roma parents sometimes enroll their require targeted interventions to ensure that children in special schools for the mentally hand- Roma are able to participate fully in the labor icapped after suffering discrimination in regular market, public services, and society in general. schools. The interconnections between the differ- ent aspects of Roma social exclusion uncovered in THE NATURE OF ROMA POVERTY AND this study suggest that Roma poverty cannot be THE POLICY CONTEXT addressed by projects that focus on a single issue. The unique characteristics of Roma poverty Instead, poor Roma need comprehensive policy mean that certain issues must be addressed coun- approaches that address all sides of their plight. try by country. But some common lessons and Another important finding of this study-- implications cut across national borders. In par- highlighted in the case of Slovakia--is that the ticular, policies to address Roma poverty must marginalization of a Roma settlement correlates respond to three main aspects of the policy envi- to its level of poverty. Roma living in more ronment: (i) the multidimensional roots of Roma remote and segregated neighborhoods have 125 Roma in an Expanding Europe: Breaking the Poverty Cycle fewer chances to participate in the mainstream A LEGACY OF DISCRIMINATION. Chapter Two showed economy, access social services (including educa- that, correcting for factors such as educational tion and health care), and tap into social networks attainment and age, there was still an undefined and information about economic opportunities "Roma factor" in poverty rates. All other consid- such as jobs. In other words, geographic and erations and explanations aside, Roma were sim- social exclusion are important correlates of pover- ply more likely to be poor. This probably reflects ty. In contrast, Roma living in integrated areas are both discrimination and the aftermath of poor more likely to interact with non-Roma, leaving relations between Roma and the majority com- them better positioned to spot and seize econom- munities in Central and Eastern Europe--a her- ic opportunities. itage of intolerance that itself results in part from past state policies and deep societal prejudices. Multidimensionality of Roma Poverty Roma poverty extends far beyond relative Attention to Diversity income deprivation. Instead, it relates to a com- While demonstrating the distinctive nature of plex set of phenomena including poor labor mar- Roma poverty, this volume also emphasizes the ket and education status, inadequate housing, the diversity of the Roma themselves. Roma are not legacies of past policies, and a long history of all alike; neither are their social conditions. troubled relations between Roma and majority Indeed, the ethnic, occupational, religious, and populations in Central and Eastern Europe. All of economic diversity among Roma populations is these factors combine to make it hard to address tremendous. The proportion of Roma-language individual problems in isolation. speakers differs greatly from country to country, For instance, as the country case studies show, as does the proportion living in cities, integrated deep-seated mistrust and poor communication neighborhoods, or segregated rural settlements. between Roma and public officials make even a These differences deeply affect welfare. Efforts to seemingly simple immunization program diffi- create, define, or represent a single Roma com- cult to implement. Roma parents sometimes munity will founder on the rocks of internal refuse immunizations, distrusting the intentions diversity. Roma tend to have distinctive problems of doctors. Indeed, health officials in Romania of integration and access, but the situations of resorted to intimidation to press Roma women to vastly different communities and individuals immunize their children. But such coercion was, cannot be shoehorned into a single, simple set of at the very best, a partial, stop-gap solution that answers. helped a few children's health even as it deep- ened underlying social divisions. Key, interrelat- The European Dimension ed features of Roma social exclusion include: Policies for addressing Roma poverty also must be framed in the context of Central and East- POOR LABOR MARKET STATUS. As detailed in Chapter ern European countries' drive for European Two one of the primary reasons Roma have been Union membership. The timing of the publication slower to benefit from the transition to market of this volume and other reports on Roma is hard- economies has been their difficulty in securing ly coincidental. Roma poverty has gained atten- employment. tion because of the accelerating process of Euro- pean integration. To meet the EU's accession GEOGRAPHIC EXCLUSION. As Chapter Three high- criteria, Central and East European countries have lights, Roma poverty is often closely related to the built institutions and passed legislation to address geographic separation of Roma settlements. In Roma issues. However, this marks only the begin- Slovakia, such remote towns were legacies of ning of the process. Even after accession--for World War II­era discrimination. Roma living in some countries, as early as 2004--tackling Roma such far-flung communities were poorer and poverty will require a long-term approach that more cut off from basic social services. 126 The Road Ahead remains part of each country's overall economic issues, even as they develop long-term strategies and socialdevelopment program. for reducing Roma poverty. However, this marks The main channel for EU support for Roma- only the beginning of the process. Even once related activities in candidate countries is the accession happens--as early as 2004 for some PHARE program.1 Between 1993 and 1999, 20 countries--addressing Roma poverty will take a million Euro were allocated to Roma-linked proj- long-term approach. ects across six candidate countries (European The European dimension of the Roma poverty Commission 1999). The total amount of PHARE issue provides a useful framework for policy. First, funding allocated for financing Roma projects in Roma are not poor only in Central and Eastern candidate countries has risen from 11.7 million Europe. Chapter Six examined the situation in Euro in 1999 to 31.4 million in 2001 (European Spain, which has also faced issues of integration Commission 2002, Table 7.1).2 The European Ini- and Roma poverty. Second, the process of Euro- tiative for Democracy and Human Rights has also pean integration offers a unique opportunity for provided EU financing.3 In the four years after addressing Roma poverty at a cross-national level. the initiative's establishment in 1994, approxi- It also lets countries learn from one another mately 4.5 million Euro were allocated to Roma throughout the accession process. Third, since the projects. The Directorate General for Education ongoing project of creating an integrated Europe and Culture also manages programs to encourage will not be completed when the latest accession cooperation between EU member states and can- treaties are ratified, the accession process offers didate countries in the fields of education, train- both an opportunity to institutionalize a long-term ing, and youth. Projects for Roma are also sup- approach to reducing Roma deprivation in Central ported through the Socrates and the Youth for and Eastern Europe, and a chance to reflect on the Europe Programmes.4 shortcomings of Roma policy further west. Beyond direct funding, the European Union has dramatically shaped the policy context POLICY IMPLICATIONS AND APPROACHES through the accession criteria that have led all The multidimensional roots of Roma poverty, candidate countries to put in place institutions the diversity of Roma communities, and the Euro- and legislative mechanisms to address Roma pean context suggest several policy implications. Table 7.1: PHARE-Funded Programs for Roma in Central and Eastern Europe, 1993­2001 EC Grants, in thousands of Euro 1993-19971,2 1999 2000 2001 Total Bulgaria 1,565 500 3,500 6,350 11,915 Czech Republic 1,778 500 2,850 3,000 8,128 Hungary 1,919 6,900 2,500 5,000 16,319 Romania 2,661 0 1,000 7,000 10,661 Slovakia 1,935 3,800 3,800 10,000 19,535 Total 9,858 11,700 13,650 31,350 66,558 1. Includes funds in support of Roma communities channeled through the Civil Society Development Foundations (funded under the PHARE National Programme), the Democracy Programme, the Lien Programme, and the Access Programme. 2. Includes both macro and micro projects: macro-projects are large partnership projects intended to promote sustained activities for up to 24 months and which may continue after the EU grant has ended; micro-projects are intended to con- tribute to citizens' initiatives and locally inspired activities. Source: European Commission, Directorate General for Enlargement, 2002. 127 Roma in an Expanding Europe: Breaking the Poverty Cycle Only a comprehensive policy approach can simul- the worst by such non-wage labor costs (World taneously address multiple causes of poverty. Bank 2001b). Moreover, with full respect for their heritage and Education reform, too, is particularly relevant deep involvement by their leaders, Roma must be for Roma. Comprehensive reforms of both gener- better integrated into European societies. Here, al and vocational education are needed to better some useful lessons can be drawn from other prepare workers for the labor market. Secondary countries with similar experiences. Finally, any school programs and curricula must be reviewed policies that are tried must be carefully imple- to ensure that they properly position young peo- mented, meticulously evaluated, and anchored in ple for the labor market by shifting away from participation by Roma themselves. The following narrow vocational and technical training, to more section addresses these policy lessons, before dis- general, rigorous, and academic programs. Im- cussing more specific interventions. proved vocational education, which expands ele- ments of the general education curriculum, could Links with Systemic Reform attract young Roma and help them secure mar- Better access to quality social services for ketable skills. Roma is linked to the overall effectiveness of each Reforms of social assistance can improve country's education, health, and social protection work incentives and reduce the risk of dependen- systems. Throughout the region, countries have cy on cash benefits. Many countries have worked embarked upon complex systemic reforms to to ensure that social assistance benefits provide a improve the efficiency, equity, and relevance of meaningful safety net for the poor. Benefits must public services. In many ways, the socialist sys- not inadvertently discourage able-bodied people tems were ill-suited to the realities of a market from working even as they help low-income economy. One way in which they have proven working families. ineffective is in their inability to reach vulnerable In addition to improving the effectiveness of groups, including the Roma. But this is hardly cash benefits, reforms of social assistance should just a minority issue. Systemic reform, improved also enhance the roles of social workers working access, and higher quality social services will with poor communities. Social workers in most improve conditions for the entire population. countries in the region function largely as admin- Reducing unemployment is a critical step istrators, instead of fully utilizing their capacities toward reducing poverty and improving living to work with individuals and households. For standards. This requires a multi-pronged ap- many Roma in the most isolated settlements, proach. It is necessary but not sufficient to main- social workers are the main contact point with the tain macroeconomic and political stability, and outside world. These workers should refer their advance financial sector reform. Increasing em- clients to other social services, provide informa- ployment opportunities hinge on a better envi- tion about employment opportunities, and coun- ronment for job creation--including measures to sel and support households in a variety of ways. support small, and medium-sized enterprises-- and easier credit for small entrepreneurs. Many An Inclusive Approach of these measures can encourage self-employ- Since Roma poverty is rooted in broad-based ment and entrepreneurship. social exclusion--economic, social, and geo- Another national-level issue that would help graphic--ameliorating it will require an inclusive unskilled Roma workers is lowering the non- approach designed to expand and promote Roma wage costs of labor. High payroll taxes and non- involvement and participation in mainstream wage labor costs in many countries discourage society, while maintaining their cultural and employers from hiring unskilled laborers, who social autonomy. Only policies that let Roma take are proportionately more expensive than workers advantage of national and European labor and with higher skills. Studies in numerous OECD housing markets, education, and health systems, countries show that the unskilled are often hurt and social and political networks have a chance 128 The Road Ahead of reducing poverty over the long term. There- · Increasing outreach to Roma communities fore, existing policies should be made more by social service providers, including accessible to Roma, and new initiatives should health and social workers; specifically reach Roma. Policies of inclusion · Involving Roma as liaisons between com- would complement rights-based approaches by munities and public services; tackling the economic and social barriers which · Providing job training and programs that Roma face. increase Roma participation in formal A central policy goal should be the multifac- labor markets. eted inclusion of Roma into institutions and An inclusive approach rejects the coercion mechanisms that create economic and social implicit in assimilationist and exclusionary policy opportunities. The emphasis here should be approaches towards Roma while remaining com- placed on incentives, not coercion. Interventions patible with the rights-based approaches dis- that reduce Roma isolation and exclusion can cussed in Chapter One. Nevertheless, a policy help improve their living conditions over the approach based on social inclusion centers on longer term. An inclusive approach must also improving opportunities and social and econom- include Roma in the projects and programs that ic welfare. Often, rights are necessary but not suf- affect them. Several successful projects use Roma ficient to create opportunities. One reason for this mentors to bridge between Roma and non-Roma is that rights are often exercised vis-ŕ-vis the communities. Roma teachers' assistants who state, while economic opportunities arise from work with parents, or Roma peer advisors who the market. Participation in market activities help with job placement, can facilitate social inte- often cannot be mandated. Thus an inclusive pol- gration while strengthening the Roma communi- icy must be comprehensive, creating incentives ty itself. for inclusion across a range of market, state, and An inclusive approach should also overcome social networks and institutions. divisions between Roma and non-Roma. Such policies build trust and help develop social capital. Learning from Example In most cases, inclusive policies should target When considering future policy directions, everyone in a community, rather than just Roma, ideas may be found in the policy experiences of although there may be exceptions where explicit other countries' and regions' minority policies, attention to ethnicity is necessary, as in overcom- particularly in the West. North and South Ameri- ing language barriers. Multicultural education and can countries provide interesting counterpoints to curricula that include the history and culture of European experiences, in part because the histo- Roma and other minorities are also critical for ries of African and indigenous peoples in the overcoming cultural barriers. Training teachers, Americas offer more parallels to Roma than to local government officials, and other social service other national minorities in Europe. While all eth- personnel can reduce discrimination by public nic groups have distinct features, minority­major- service providers. Finally, public information cam- ity relations share important similarities every- paigns can promote multiculturalism and raise where, and much can be learned from the policy general awareness about discrimination. In this experience of countries that have confronted these vein, policies that expand opportunities include: issues in the past and still face them today. These · Reducing segregation in housing, particu- issues deserve further exploration. larly by alleviating problems associated To be sure, Roma in Europe have endured with isolated rural settlements; centuries of exclusionary and assimilationist poli- · Integrating Roma students into main- cies without being absorbed into majority soci- stream educational systems by establish- eties. They remain stateless and have founded no ing preschool programs and providing movement for statehood. In this regard, their food, clothing transportation subsidies to closest parallel may be with native Americans, a make it easier for poor students to attend; separate ethnolinguistic community that has 129 Roma in an Expanding Europe: Breaking the Poverty Cycle often preferred preserving its own traditions and the countries of the former Yugoslavia. The way of life to integration. These general charac- review of the western literature on Roma under- teristics underline both the challenges facing an taken for this report found little data on Roma in inclusive approach to Roma poverty and the these countries, despite the large estimated Roma long-term nature of the policy responses population in countries such as FYR Macedonia. required. They also underscore the stakes. From a sectoral perspective, regular and compa- rable information on Roma household welfare Attention to Evaluation and Implementation and living conditions--in addition to data on The development of a comprehensive nation- education and health status--are needed across al policy response to Roma poverty must be com- countries to identify community needs and bined with attention to monitoring, evaluation, develop policy strategies. Of the main policy and implementation. The range of Roma projects areas, health (particularly reproductive health) in Central and Eastern Europe has provided has perhaps been the most neglected to date, and much experience in implementation. Still, despite instruments for monitoring health status and the high level of activity, very few initiatives have communicable diseases are sorely needed. been evaluated or monitored, making it extreme- While privacy concerns about data collection ly difficult to identify lessons for the future. As must be respected, policy makers need up-to-date countries move forward, they must examine this information to design programs and monitor out- growing body of experience. A related priority is comes. Such data collection should benefit Roma the need to build mechanisms for monitoring and in the long run through better-designed and tar- evaluation into new and ongoing initiatives and geted interventions. To protect privacy, declara- to provide opportunities for exchanging informa- tions of ethnicity should be voluntary, and peri- tion within and across countries. odic sample surveys, rather than national administrative data, should be used to collect FILLING INFORMATION GAPS. The first step toward information on specific topics. Roma groups must increasing monitoring and evaluation capacity-- also be involved in the development, implemen- and, hence, improving project design--is making tation, and analysis of surveys, as happened dur- more and better information available. This vol- ing the 2001 census in Slovakia. Finally, qualita- ume has highlighted the critical lack of basic tive assessments can also provide valuable information about Roma. To remedy this, coun- information for project design. tries need to examine their statistical instruments and administrative data to find out how they can MONITORING AND EVALUATION. The importance of better capture policy-relevant information on building monitoring and evaluation mechanisms Roma and other minorities. Multilateral coordi- into projects and policies cannot be overstressed nation, advice, and guidance can help ensure the (Box 7.1). To ensure accountability, monitoring comparability of data. Still, more information on should be an integral part of all projects. Evalua- international practices is needed, particularly in tions to assess a project's impact and outcomes addressing privacy issues about ethnic identifica- are equally important. This entails collecting tion. On a related note, the outcomes of targeted baseline data at the start of a project to use for public policies and NGO initiatives require close comparison once the project has been completed. monitoring, and the results of program evalua- For example, an intervention designed to tions should be used for ongoing policy develop- improve school enrollment should measure ment. The lessons should be disseminated across enrollment before the project began and then regions and countries. assess whether participants stay in school longer Gaps in information about poverty and wel- and perform better with the new program in fare persist at both the country level and about place. The time horizon for outcome evaluation particular subject areas. In particular, more infor- should also be enough to assess the longer-term mation is needed on the conditions of Roma in impact. Again, in the case of education, the eval- 130 The Road Ahead Box 7.1: Monitoring and Evaluating School Success for Roma Children The Step-by-Step Special Schools Initiative, supported by the Open Society Institute, provides a useful example of how project evaluation can improve the success of a project and contribute to policy development. This project aimed to address a particularly troubling problem: the shunting of Roma children into "special schools" intended for the mentally and physically handicapped. It also sought to formulate policy recom- mendations that would improve the chances for Roma children in mainstream schools. The project operated in Bulgaria, the Czech Republic, Hungary, and Slovakia. Roma students in special schools in each of the countries were taught the mainstream curriculum instead of the slower special school curriculum. Teachers and administrators were trained in anti-bias education, second-language learning, and the Step-by-Step methodology. Additional support was provided in the form of classroom materials. A Roma assistant teacher was assigned to each site to help in the classroom and work with students and their fami- lies outside of school. Evaluation was built into the project from the start. Local researchers were hired to collect data in each project site, as well as several control sites; their efforts were coordinated by an international researcher who ensured the data's comparability. Data were collected on process indicators such as student attendance and parental involvement, as well as on educational outcomes. Students were also tested at the end of each aca- demic year, and interviews were conducted with students and parents. Monitoring was integral to the project to keep it on track. Master teacher-trainers at the national and international levels worked closely with proj- ect staff to provide support and ensure consistency across project sites. The results after the second year of the project were heartening. On aggregate, the project found that 64 percent of Roma second-graders in the project sites did not belong in special schools. In other words, these Roma pupils were able to meet national standards for the mainstream curriculum with the support of the project. These powerful results make the case for interventions to get Roma children out of special schools and into mainstream classes. The empirical analysis makes a compelling case that investments in education for Roma students--including teacher training, language support, and parental involvement--can pay off over the longer term; graduates of mainstream schools have far more employment and education opportu- nities than graduates of the special schools. Similarly, the ongoing monitoring let project managers make course corrections and distill lessons for follow-on projects. Source: Rona and Lee 2001. uation should consider not just whether children well-intentioned, that failed because they were are in school at the end of the project, but what designed and implemented without the partici- they have learned, whether they graduate, pation of the future beneficiaries--such as hous- whether they continue their education, and how ing projects that built apartments that were the project affected their chances for higher edu- unsuitable for Roma or social assistance pro- cation and employment. grams that gave Roma goods they would rather have sold. Ensuring Participation Roma involvement in policy and project Regardless of whether policies are explicitly development rests on the existence of effective designed for Roma, Roma must be involved. The mechanisms for participation. While Roma have track record of programs directed at Roma--dur- been increasingly involved in civil society and ing both the socialist and transition periods-- policymaking, significant challenges remain. clearly showed that including Roma in the Some of these have been discussed earlier, includ- design, implementation, and evaluation of pro- ing lingering prejudices, mistrust between Roma grams is essential for success. The recent past is and non-Roma, and the low education levels and littered with projects and programs, however widespread illiteracy that shrink the potential 131 Roma in an Expanding Europe: Breaking the Poverty Cycle pool of Roma leaders and voters. Policymakers nate consequences for society's most vulnerable, must continue to expand opportunities for Roma including Roma. Roma participate in local gov- to participate in civil society and public service at ernments by running for elective office, using the local and national levels. Non-Roma involve- public services, and interacting with local offi- ment is also crucial. The example of Slovakia in cials. Local governments could also be important Chapter Three highlighted the perils of segrega- sources of support for Roma communities, indi- tion. Roma who do not interact with wider socie- viduals, and associations. But, as the Nyíregy- ty, including other Roma communities and non- háza case study from Hungary in Chapter Five Roma, are cut off from social services, the labor illustrated, local governments can also further market, education, and--all too often--prosperi- marginalize Roma. Even where national policies ty. More contacts and partnerships between non- do not discriminate, biased implementation at the Roma and Roma will ease the mistrust and mis- local level can derail original intentions. communication that limit local and community In recent years, all countries in the region development. have introduced institutions for integrating Across the region, post-Soviet political liber- Roma into policymaking at the national and local alization created a proliferation of civil society levels. Perhaps the most ambitious approach was organizations, including NGOs, political parties, taken in Hungary, which in 1993 introduced a religious organizations, and community associa- system of minority self-governments, as dis- tions. Many groups have been formed to address cussed in Chapter Five. Other countries, includ- particular issues related to ethnic minorities, ing the Czech and Slovak Republics and Roma- including a wide range of Roma organizations, nia, have established national consultative bodies many of them financed by external sources. to shape policymaking related to minorities. Chapter Five discussed the range of NGOs In the Czech case, a new Inter-Ministerial around Roma issues in Hungary. Similarly, a 1996 Commission comprised of Roma and non-Roma survey carried out by the Union of Bulgarian representatives of government agencies now Foundations and Associations identified more advises the parliament. In Romania, the Council than 1,300 organizations that addressed ethnic of National Minorities, tied to the parliament, issues and put Roma among their priorities (Iliev includes representatives from minority organiza- 1999). tions. The strength of these bodies varies, howev- Roma NGOs, like Roma communities them- er, and some even lack budgets. In most cases, it selves, are diverse and often fragmented. In some is too early to gauge how representative and cases, this limits the effectiveness of Roma in their effective these institutions are. dialogue with government officials and other Many countries have also recently adopted potential partners. Roma organizations disagree national policy strategies on Roma issues. Such a frequently and struggle to reach consensus. This plan is currently under discussion in Serbia and may reflect several issues, including the groups' Montenegro. In Bulgaria, the government adopt- relative political inexperience, divisions between ed "The Framework Program for Equal Integra- Roma subgroups, and some characteristics of tion of Roma in Bulgarian Society" in March Roma social organization such as the rather com- 1999--the culmination of an unprecedented mon absence of hierarchical structures within process of consultation and consensus-building Roma groups. both between the government and Bulgaria's Local governments are also important. Roma community and among Roma NGOs them- Throughout the region, the role of local govern- selves. The program, which was endorsed by 75 ments has changed substantially during the tran- Roma NGOs, offers strategic guidelines in the sition as decentralization replaces communist areas of antidiscrimination policy, economic centralization. The process of building effective, development, and social services (OSCE 2000). accountable, responsive local governments has While it is too soon to judge the impact of these not been an easy one--with particularly unfortu- strategies and action plans, they have helped ele- 132 The Road Ahead vate dialogue between the Roma community and additional barriers that Roma face, including national governments and have raised core policy lower education status, geographic isolation, and issues. Examples from Western European coun- discrimination. Experiences from Spain and Hun- tries--such as Spain, a case discussed in Chapter gary provide examples of promising employment Six--can provide useful insights here. projects (Box 7.2). Initiatives that increase oppor- tunities for Roma in the labor market start with POLICY DIRECTIONS improvements in education status. Addressing Roma issues will take experimen- Improving access to credit makes it easier for tation, patience, and close collaboration between Roma and other low income groups to start their Roma communities, the international community, own businesses. NGOs can play an important NGOs, and national governments. Initiatives role in helping communities initiate projects. need to be designed and adapted to local country Partnerships between these organizations and circumstances, as well as to the varying needs of banks are needed to establish credit mechanisms. Roma groups. Policies need to balance three relat- On a related note, expanding the availability of ed sets of objectives: first, increasing economic microcredit could weaken the grasp of local opportunities by expanding employment partici- usurers who currently lend funds at extortionate pation; second, building human capital through rates in some Roma settlements. better education and health; and third, strength- Another important element is more effective ening social capital and community development public works programs. Many current programs through increased Roma empowerment and par- focus on short-term, low-skilled employment and ticipation. Implementing these measures will provide participants with neither enhanced skills involve collaboration between central govern- nor better long-term employment prospects. ment ministries, local governments, Roma com- Governments should improve the quality and munities, NGOs, and international partners. training content of public works jobs so that par- ticipants gain transferable skills. EMPLOYMENT. Expanding labor market opportuni- Training programs can also facilitate labor ties is a priority throughout the region. Opportu- market reentry for low-skilled and unskilled nities must be widely shared, and the poorest workers. However, because international experi- must have the means to take advantage of new ence with such programs is mixed, programs jobs. Without this, a core poor "underclass" will must be carefully tailored to fit labor-market con- persist. Specific attention is needed to address the ditions--a point particularly relevant to Roma Box 7.2: Promoting Roma Employment One of the most established programs to promote employment and income-generating opportunities for Roma is Hungary's Autonómia Foundation, which provides grants and interest-free loans to develop employment programs for Roma. Its income generating initiatives include livestock breeding, agricultural programs, and small enterprise development. The success of Autonómia's projects, as measured by the repayment rate of its loans, has soared since the foundation was established in 1990. In 1998, repayment rates reached nearly 80 percent, compared to 10 per- cent during Autonómia's first year. Autonómia attributes this improvement to the involvement of trained monitors, some of whom are Roma, who work closely with project teams throughout the implementation process. Autonómia is now expanding its programs to other countries in the region. In 2000, the first group of Roma began training to start small grant-and-loan programs for Roma in four CEE countries, including Slovakia. Further evaluation of the project should examine the impact of the project on participants' welfare. Sources: Autonómia Foundation; Tanaka et al. 1998. 133 Roma in an Expanding Europe: Breaking the Poverty Cycle (Dar and Tzannatos 1999). Some initiatives have and compulsory. For its part, the Slovak govern- sought to train Roma in traditional trades that are ment has supported the Zero Grade Program, not in much demand. which expands preschool attendance for Roma At the policy level, antidiscrimination legisla- children. tion must be in place, complete with effective and NGOs can also play important roles. The accessible mechanisms for appeals. Beyond leg- Open Society Foundation initiated the "Step-by- islative measures, project interventions can over- Step" program, modeled on the US "Head Start" come barriers between non-Roma and Roma by initiative, in both Roma and non-Roma commu- building confidence through on-the-job training nities. In 2000, over 8,000 Roma students in 17 and employment experience. A successful public countries in Central and Eastern Europe and the works project in Bulgaria showed non-Roma con- former Soviet Union enrolled in Step-by-Step tractors that Roma could be reliable, effective programs. Step-by-Step takes an integrated employees--a standing rebuke to deeply held approach that provides training and support to stereotypes about Roma laziness. Another possi- teachers while involving parents in the class- ble approach is offering tax incentives to employ- room. Parental involvement at all levels of educa- ers who employ Roma. tion should be explored and fostered, including bringing parents into the classroom as teachers' EDUCATION. Because education is so central to aides, parent­teacher associations, and regular improved welfare and economic status, it has been parent­teacher interactions. a priority focus for both governments and NGOs. Initiatives that reduce the dropout rate and More project activity has taken place in the educa- smooth the way to secondary and tertiary educa- tion area over the past decade than in any other tion are also critical, but there is less experience sector. The review of social sector projects in Hun- here. Still, mentoring programs and extracurricu- gary presented in Chapter Five found that nearly lar activities that provide tutoring and supple- 30 percent of resources allocated to Roma projects mentary educational events have been intro- during the past decade were for education. duced in some countries. Schools like the Gandhi Initiatives in education take various forms School in Pécs, Hungary, and the Romani High and intervene at different points within the edu- School for Social Affairs in Kolin in the Czech cation cycle. One key priority is lowering the bar- Republic integrate Romani studies, including lan- riers that prevent Roma children from starting guage, history, and culture into the curriculum. school. Many children are discouraged from Successful elements from these schools--includ- attending school because of deprivation at home ing multicultural curricula, teacher training, and and cultural differences, including language. Eco- parental involvement--can be incorporated into nomic constraints can be loosened by coordinat- all public schools (Box 7.3). ing social assistance and education policies to Better education for Roma students can boost ease the cost of education for poor families-- school attendance and educational outcomes. including such tactics as school feeding programs This will require fighting discrimination within (which boost both nutrition and attendance), school systems and diminishing the role of spe- linkages between child allowances and school cial schools and institutions for Roma. The prac- enrollments, and scholarships for low income tice of unnecessarily channeling Roma students students. Social workers can also identify house- into special schools in the first place must be holds in need of assistance. reviewed, as should policies that limit the future Preschool programs can prepare children for opportunities of special school graduates. Special the classroom and surmount language barriers. education should be reformed to address true Several countries have tried targeted preprimary learning disabilities and the special needs of at- initiatives to facilitate school attendance and per- risk children. formance. In 2002, the Bulgarian government Limiting the use of separate classrooms and announced its intention to make preschool free schools for Roma can improve education quality 134 The Road Ahead Box 7.3: Alternative Secondary Schools in Hungary Hungary has experimented with alternative approaches to secondary-school education that aim to help Roma children bridge the gap between basic and secondary school, improve their academic performance, and create future opportunities. Roma are much less likely to start and complete secondary school than other children. A 1993 survey of Roma in Hungary found that only 1 percent of Roma took the final examination for secondary schools and only 13 percent received training as skilled workers. A review of these alternative approaches commissioned by the World Bank looked at six different schools, most of which had been established during the previous five years. All of the schools were private and received support from a range of local and international foundations and NGOs, as well as state budget subsidies. While most students in each of these schools were Roma, not all the institutions explicitly target- ed Roma children. The type of education provided by the different schools varies greatly. In some cases, the schools provide vocational training, such as the "Roma Chance" Alternative Vocational Foundation School in Szolnok, the Don Bosco Vocational Training Center and Primary School in Kazincbarcika, and Budapest's Kalyi Jag School. Others, such as the Jószefváros School and the Collegium Martineum in Mánfa, support secondary school students through extracurricular activities and classes and (in the case of the Collegium Martineum) dormi- tory accommodations in a supportive home environment. Finally, the Gandhi School and Students' Hostel in Pécs is a six-year secondary school (or gymnasium) that prepares students for university. The schools differ in the extent to which they emphasize the Roma background of their students in their curricula and approach. In most of the schools, strengthening Roma identity and preserving Roma tradition are explicit and integral components of school mission. These schools offer classes in Roma language, histo- ry, and art. Others, such as Don Bosco, focus on building the self-confidence of students through profession- al training. The schools also take different approaches to the underlying socioeconomic disadvantages of students. Some, such as the Collegium Martineum, target disadvantaged students and provide housing and other sup- port to boost attendance. Most of the schools also involve parents, although this often proves difficult because of low education levels. Source: Orsós et al. 2001. and reduce divisions between Roma and non- ronment and link Roma communities and Roma communities. Within schools, separate schools. In Romania, the Ministry of Education classrooms for Roma should be abolished. For has appointed Roma education inspectors in each geographically remote settlements, other options of its 41 counties to monitor the quality of Roma could be considered, such as the pilot project in education. Bulgaria that transports Roma children from a Roma settlement to an integrated school (Box HEALTH CARE. Relative to the other policy areas, 7.4). much less is known about the health issues facing Teachers define the quality of education and Roma communities. This calls out for better mon- must be trained to meet the challenges of a multi- itoring. In particular, more effective observation cultural environment. Ongoing support mecha- of communicable diseases such as tuberculosis, nisms that help teachers on the job are also criti- hepatitis, and HIV/AIDS is critical. Health serv- cal. Particular training should include Roma ices must also be available in isolated Roma com- history and culture, conflict resolution, and class- munities. Policies that can expand such access to room management. Some countries have also remote rural areas and segregated urban commu- experimented with Roma teachers' assistants and nities include providing incentives for physi- mediators who can assist in the classroom envi- cians, community health workers, and social 135 Roma in an Expanding Europe: Breaking the Poverty Cycle Box 7.4: Desegregation of Roma Schools in Bulgaria: The Vidin Model In Vidin, the Open Society Institute and the Roma NGO known by the acronym DROM have been collab- orating on a innovative program to integrate Roma students into the mainstream school system. Vidin is a town of 85,000 in northwest Bulgaria where 6 percent of the population identified as Roma in the 2001 census. In the 2000-01 school year, 460 students, or half of the school-age students, were integrated into the mainstream school system; more followed in the next school year. Under the project, students are bused from the settle- ment to school and back and receive supplementary classes at school to help them catch up with their peers. The project involves Roma supervisors who interact with parents and the school to encourage attendance. Low-income students also receive shoes and school lunches; students are given their lunch on the bus to reduce the stigma of receiving it at school. While preparing the program, DROM went door-to-door in the Roma settlement explaining the project. DROM also sought the support of the schools, the mayor, and the media. The project eventually gained sup- port of all the stakeholders except the mayor, who did agree not to block it. With the agreement of several Roma parents, DROM invited the six mainstream schools in Vidin to participate in a TV program at which each school presented its program, philosophy, and teachers. Roma parents then selected a school for their children. This lessened their concerns and marked the first time that their views had been solicited by the authorities. At the end of the first semester, the project was a dramatic success, as seen in 100 percent attendance, first-term final-grade averages identical to those of non-Roma pupils, parental and teacher satisfaction, the absence of reported incidents of anti-Roma prejudice, full support from the Regional Directorate of the Min- istry of Education, and encouragement to scale-up in other cities. In addition, 35 Roma parents of the bused children themselves returned to school in adult-education programs, and three teenagers who had dropped out in the third grade asked to join the program, prompting teachers to work extra hours with them. On the negative side, 24 pupils received failing grades in one or more subjects, and three left the program. (One returned to the Roma school, and two functionally illiterate eighth graders dropped out.) The success to date of the program is attributable to three major factors. First, parents feel that their chil- dren are protected from prejudice because they are bused and monitored throughout the day by adult Roma; parents also feel that their children can meet the higher scholastic standards. Second, the schools have accept- ed young adult Roma monitors in the schools who assure that the children are not mistreated. The monitors also follow parental engagement and student participation in extracurricular activities. Moreover, the moni- tors help the teachers and ease cultural differences. Third, the children are happy to be in schools where real learning takes place. Ongoing assessment of project outcomes will be essential to understand the longer-term implications of the highly encouraging Vidin project. Since the project started in Vidin it has been expanded to seven more cities in Bulgaria. Source: Open Society Institute. workers to work with communities to address campaigns are also critical for addressing many problems and teach prevention. emerging health risks, including substance abuse, Public health interventions can be designed to sexually transmitted diseases, and conditions overcome cultural barriers to care. Some coun- associated with poor nutrition and housing. tries have experimented with using Roma media- Other initiatives can include better dissemination tors to promote health activities within Roma of public-health information through the media communities and facilitate interactions between and schools, as well as better coordination with Roma and health care professionals--especially organizations such as churches and Roma NGOs. around overcoming Roma resistance to such basic Health services need particular attention. Out- care measures as immunizations. Information reach can raise awareness about a range of issues, 136 The Road Ahead Box 7.5: Lessons from US Welfare Reform During the 1990s, as concerns grew about the increasing number of welfare caseloads, the US govern- ment introduced substantial legislative changes in programs designed to assist low-income families. In par- ticular, the federal government granted a growing number of waivers early in the decade, allowing states to experiment with alternative rules for the Aid to Families with Dependent Children (AFDC) and Food Stamps programs. These changes were followed in 1996 by the Personal Responsibility and Work Opportunity Act, which fundamentally changed the public assistance system in the United States. The Act abolished AFDC, which required states to match federal welfare funds, and replaced it with Temporary Assistance to Needy Families (TANF), which granted unconditional, fixed amounts of funding to states and allowed them to set their own rules for eligibility and benefits. In the light of these changes, several states started using "diversions" (one-time assistance, rather than enrollment in ongoing TANF-funded programs) and benefit programs that let recipients keep more public- assistance benefits after returning to work, thus increasing both work incentives and income among low- income families. Some states also worked to transform public assistance offices into employment assistance offices where applicants were given constant incentives to seek and find work. Moreover, several states imposed more penalties on those who did not respond to these work incentives. Finally, individual states spent more money on work-related programs, relative to cash benefits. To what extent was welfare reform responsible for these trends? To be sure, the US economy enjoyed tremendous prosperity during the 1990s. As a consequence, employment growth was high, unemployment was low, and wages grew significantly among workers of all skill levels since 1996. These factors influenced the welfare of less skilled workers and are therefore important in explaining the trends described above. In fact, between one-third and two-thirds of the caseload change can be attributed to the overall performance of the economy.1 Unfortunately, a strong economy affects not only poverty but economic policy, which makes it hard to measure the effect of welfare policy changes independently of the business-cycle effect.2 But while the over- all effect of welfare reform is difficult to pin down, both Canada and the United States have experimented with particularly innovative types of welfare reform programs in ways that permit some form of evaluation. These programs combined financial incentives with work mandates. In particular, the Minnesota Family Investment Program (MFIP) substantially decreased the benefit-reduction rate for public assistance recipients (thus allowing them to keep more public assistance income when they went to work) while also mandating participation in work/welfare programs. Striking a similar note from north of the US border, Canada's Self- Sufficient Program (SSP) provided substantial financial support to long-term public assistance recipients who worked 30 hours or more per week. These programs' results showed that employment, earnings, and family income increased for program participants even as poverty fell. Although these programs are not money-savers in the short run--indeed, they actually provide more assistance to low-income families than did traditional welfare programs--it is important to consider their long-run effects, particularly since studies of people leaving welfare suggest that most of them (55 to 85 per- cent) become employed at a future date and about one-half to two-thirds report higher incomes after they get off welfare (Brauner and Loprest 1999). Continued on next page including women's health. Attention to reproduc- ways. In the Slovak Republic, for instance, a team tive and family health care issues can help over- of NGOs organized a course on hygiene and child come cultural taboos, such as the fear of screening development for Roma mothers. Countries in for cervical cancer. Some of the PHARE projects other regions have launched successful initiatives are addressing women's health issues in different 137 Roma in an Expanding Europe: Breaking the Poverty Cycle Box 7.5 (continued) Moreover, such programs can be improved through good design. For instance, employment is associated with extra expenses in the form of child care, transportation, and more. So in some states, public support for those items was included as part of their welfare policy, together with health insurance coverage through the Medicaid system.3 At the federal level, the Earned-Income Tax Credit program served a similar function. In sum, in the US case, a confluence of events seemingly came together--a strong expanding economy, substantial revisions of public-assistance programs that emphasized work and reduced benefit eligibility, and major policy changes that increased the numbers of people returning to work and the subsidies to support work, particularly among vulnerable groups. This seems to have created the right environment for the decline in poverty rates and welfare caseload observed in the data. Moreover, because many of the programs described above rely strongly on the availability of jobs, it is not clear how sustainable these welfare policy changes are in the long run--or how dependent their success has been on a booming US economy. Still, the fact that the SSP managed to succeed despite a Canadian economy that did not do as well as the US in the 1990s shows that programs can work in less favorable environments with high unemployment, if they are designed correctly. Notes: 1. Different studies provide different measures. See Figlio and Ziliak (1999) and Schoeni and Blank (2000). 2. There is some crude evidence that such changes had a substantial effect on caseloads, but there has been significantly less research relating TANF changes to work behavior or poverty rates. In this respect, the best evidence comes from the fact that participation rates are increasing among vulnerable groups (e.g. single mothers with young children). 3. Most low-skilled jobs do not offer health insurance and this could act as a deterrent for employment. Sources: World Bank, 2001b; Blank 2000; Peterson 2000; Schoeni and Blank 2000. for improving women's health through communi- Adequate mechanisms for community in- ty groups and education. volvement and choice are equally important. The legacy of failed housing programs and projects HOUSING. Because Roma live in such different con- during the socialist era has made this particularly ditions, housing is a complex sector that requires critical, but there are still few experiences from close coordination between governments and which to draw. Some promising facilities have communities. Effective legislation and enforce- emerged recently that let communities and ment mechanisms are needed to prevent housing households apply for resources for local develop- discrimination and clarify property ownership. In ment projects and better housing, including many slum areas and settlements, unresolved microcredit arrangements and social funds. Find- questions about building ownership and residen- ing out whether these instruments can reach cy rights have blurred the responsibilities for Roma communities will take careful considera- upgrading and maintenance to the point where tion and monitoring. no one is responsible. Similar dynamics block Measures to alleviate poor conditions in incentives for residents to invest in and maintain some of the most disadvantaged Roma settle- properties. A UNDP program in Bucharest ments include: (i) clarifying property rights; (ii) worked towards legalizing apartments for house- resolving disputes over the ownership of land holds in a neighborhood where ownership was and buildings that are stopping residents and not clear. The municipality assumed the manage- local governments from investing in and main- ment of the properties and let residents apply for taining rundown properties; (iii) simplifying rental contracts. procedures for obtaining building permits to allow residents to upgrade their property; and 138 The Road Ahead (iv) providing clear information to the public on CONCLUSIONS procedures for applying for construction permits Poverty among Roma remains one of the fore- and acquiring property. most policy issues for Central and Eastern Europe states as they move towards EU integration and UTILITIES. Outlying Roma settlements need sustained economic development. By going deep- expanded coverage of utilities and public servic- er into the nature of Roma communities them- es. One option would be bringing isolated settle- selves and providing a more complete picture ments into mainstream service networks. While through both quantitative and qualitative data, inhabitants should still be charged for utilities, this report finds that Roma poverty is a multifac- subsidies may be needed for low-income house- eted problem that can only be addressed by an holds, particularly to cover the cost of public inclusive approach--involving government, civil goods such as sanitation. Local governments and society, and other partners--that addresses all communities can be given incentives to provide dimensions of Roma social exclusion simultane- services in settlements, possibly through a central ously. The dominant policy approach since 1989 fund. Finally, opportunities within public works tended to be the opposite, relying on a fragment- programs can improve basic infrastructure and ed set of projects, often delivered by local NGOs services in Roma communities. with limited assistance from the state. So the potential to make a difference through a compre- SOCIAL ASSISTANCE. Safety-net programs that pro- hensive change of direction is large and bright. vide cash assistance to the poor are an important The current level of activity and interest in source of income for many Roma families. Many Roma issues in Central and Eastern Europe pro- countries in the region are reforming cash bene- vides a promising start. The next step is to inte- fits to make them more effective and capable of grate the lessons of these experiences into policy. reaching the poor. Such programs need to meet The mechanisms to facilitate this have been put in the needs of poor households without discourag- place. Most countries have now formulated ing those who can from working--which would strategies for improving the conditions of Roma leave them in a "poverty trap," dependent on and have built institutions to develop, coordinate, social benefits. The Slovak case, in particular, and administer policies and projects. But the road highlighted the perils of this reliance on social ahead is long and winding. Improvements will benefits. not come overnight. Indeed, the debilitating Work incentives can be built into social assis- poverty among Roma communities in some West tance programs through time limits, work European countries highlights the scope of the requirements, and other mechanisms. Benefits challenge for their neighbors to the east. Effective should be phased out so that low wage work- policy responses will require a multilayered ers--the working poor--will still be entitled to approach, involving cross-country partnerships. benefits but at a level that will not discourage With sustained leadership, both by Roma them- them from working. This would improve work selves and by those who recognize how much incentives for those at the margins and increase Roma can contribute to an enlarging Europe, income among low-income working families. Roma can look forward with real hope. Social workers should also shift their role to act as employment facilitators who can help the unem- NOTES ployed find work. Work-related programs, such 1. Since 1989, the EU has provided support as support for child care and transportation sub- for Central and Eastern European countries. The sidies for low-income workers, can also make it main instrument through which this assistance is easier to find jobs and break the dependency provided is the PHARE Programme, under the cycle. Lessons from the US welfare reform experi- responsibility of the Directorate General for ence of the 1990s are illustrative (Box 7.5). Enlargement. In 1993, PHARE support was reori- ented to focus more on the needs of countries 139 Roma in an Expanding Europe: Breaking the Poverty Cycle applying for EU membership, including an ex- Republic, Hungary, Poland, Romania, Slovakia pansion in support to infrastructure investment. and Slovenia, see European Commission 2002. In 1997, PHARE funds were again reoriented to 3. The EIDHR is managed by the Directorate focus entirely on the pre-accession priorities General for External Relations. For more infor- highlighted in each country's Accession Partner- mation on the EIDHR and its projects, see http:// ship agreements. PHARE funding is distributed europa.eu.int/comm./europeaid/projects/ddh_ as grants. en.htm. 2. While having a special focus on Roma 4. The Socrates and Youth Programmes are issues, some of the projects are not targeted sole- managed by the Directorate General for Educa- ly at Roma, and may include other ethnic minori- tion and Culture. For more information on the ties or disadvantaged groups. As a result, these Socrates Programme http://europa.eu.int/ figures do not represent the amount spent exclu- comm/education/socrates.html and for the Youth sively in support of Roma. For a more detailed Programmes see http://europa.eu.int/comm/ breakdown of PHARE-funding for Roma by sec- education/youth.html. tor and project title, for Bulgaria, the Czech 140 REFERENCES Ainscow, M. and Memmenasha, H.-G., 1998. ety, North American Chapter. New York: "The Education of Children with Special Gypsy Lore Society. Needs: Barriers and Opportunities in Central Beck, S. 1989. "The Origins of Gypsy Slavery in and Eastern Europe," Innocenti Occasional Romania." Dialectical Anthropology 14 (April Paper No. 67, UNICEF-ICDC: Florence, Italy. 1989): 53­61. Allison, C. and Ringold, D. 1996. "Labor Markets Blank, R. M. 2000. "Fighting Poverty: Lessons in Transition in Central and Eastern Europe: from Recent US History." Journal of Economic 1989­1995." World Bank Technical Paper No. Perspectives, vol. 12, No. 2, Spring 2000. 352. Washington, D.C. Brauner, S. and Loprest, P. 1999. "Where are they Asociación Secretariado General Gitano (ASGG). Now? What State's Studies of People Who 2000. Proyecto ACCEDER: Informe Final, Left Welfare Tell Us." Washington, DC: The Acceder integra. Madrid, Spain: ASGG. Urban Institute. Asociación Secretariado General Gitano (ASGG). Cahn, C. 2001. "Smoke and Mirrors: Roma and 2001. The Roma Community in Spain and Slova- Minority Policy in Hungary." Roma Rights 4. kia: A Guide to Action Strategies. Madrid, Spain: European Roma Rights Center. http:// ASGG and Ministerio de Asuntos Exteriores. www.errc.org/rr_nr4_2001/noteb6.shtml Bárány, Z. 2000. "The Poverty of Gypsy Studies." Calvo B, T. 1989. Los racistas son los otros: Roma, NewsNet: The Newsletter of the American Associ- minorías y derechos humanos en los textos esco- ation for the Advancement of Slavic Studies 40(3). lares. Madrid: Editorial Popular. Bárány, Z. 2002. The East European Gypsies: Regime Cartner, H. 1994. "Romanian Lynch Law: Vio- Change, Marginality and Ethnopolitics. Cam- lence Against Roma in Romania." Human bridge: Cambridge University Press. Rights Watch Newsletter, November 1. New Blanchard, O., et al. 1995. Spanish Unemployment: York: Human Rights Watch. Is There a Solution? London: CEPR. CEDIME-SE. 2001. Minorities in Southeast Europe: Basurto, P. 1995. Children: Victims and Symbols. Roma of Romania. Report prepared by the Cen- In Children of Minorities: Deprivation and Dis- ter for Documentation and Information on crimination. Innocenti Insights Series. Flo- Minorities in Eastern Europe--Southeast rence, Italy: UNICEF International Child Europe (CEDIME-SE) in cooperation with the Development Centre. Ethnocultural Diversity Resource Center Beck, S. 1984. "Ethnicity, Class and Public Policy: (EDRC). Cluj, Romania: CEDIME-SE/EDRC. Tsiganii/Gypsies in Socialist Romania." In CIDE 1999. Las desigualdades en la educación en Kot Shangriladze and Erica Townsend, eds., Espańa. Madrid, Spain: Ministerio de Educa- Papers for the Fifth Congress of Southeast Euro- ción y Cultura, Centro de Investigación y pean Studies (Belgrade, September 1984). Documentación Educativa. Columbus, OH: Slavica Publishers. Center for Reproductive Rights, 2003. "Body and Beck, S. 1985. "The Romanian Gypsy Problem." Soul: Forced Sterilization and other Assaults In Joanne Grumet, ed., Papers From the Fourth on Roma Reproductive Freedom in Slovakia." and Fifth Annual Meeting of the Gypsy Lore Soci- New York. 141 Roma in an Expanding Europe: Breaking the Poverty Cycle Cilla G., Perez-Trallero E., Marimon J.M., Holsti, eds., The Gypsies of Eastern Europe. Erdozain S., and Gutierrez C. 1995. "Preva- Armonk, NY: M.E. Sharpe, Inc. lence of Hepatitis A Antibody among Disad- Crowe, D. 1994. A History of the Gypsies of Eastern vantaged Gypsy People in Northern Spain." Europe and Russia. New York: St. Martin's Epidemiology and Infection 115: 157­61. Press. Commander, S. and Coricelli, F., eds. 1995. Unem- Dar, A. and Tzannatos, Z. 1998. "Active Labor ployment, Restructuring and the Labor Market in Market Programs: A Review of the Evidence Eastern Europe and Russia. Washington, D.C.: from Evaluations." Washington, D.C.: The The World Bank. World Bank. Commission of the European Communities. 2000. Denkov, D., Stanoeva, E., Vidinsky, V. 2001. 2000 Regular Report from the Commission on "Roma Schools," Open Society Institute: Hungary's Progress. Brussels, Belgium. http:// Sofia, Bulgaria. europa.eu.int/comm/enlargement/report_11 De Witte, B. 2002. "Politics versus Law in the EU's _00/pdf/en/hu_en.pdf. Approach to Ethnic Minorities." In J. Zielon- Commission of the European Communities. 2001. ka, ed., Europe Unbound: Enlarging and Reshap- 2001 Regular Report from the Commission on ing the Boundaries of the European Union. New Hungary's Progress Toward Accession. SEC York, NY: Routledge. (2001) 1748, Brussels, Belgium: Commission. Dunai, M. 2002. "Political Battles Heat Up in http://europa.eu.int/comm/enlargement/ Hungary." Time Europe Daily (February 15). report2001/hu_en.pdf. http://www.time.com/time/europe/eu/ Commission of the European Communities. 2002. daily/0,13716,203284,00.html 2002 Regular Report from the Commission on The Economist. 2001. "Europe's Spectral Nation." Hungary's Progress Toward Accession, SEC The Economist (May 10, 2001). (2002) 1404, Brussels, Belgium: Commission. Erlanger, S. 2000. "The Gypsies of the Slovak Conclusions of the Copenhagen European Coun- Republic: Despised and Despairing." The New cil. 1993. Bulletin of the European Communities York Times (April 3): A10. 6, point I.13. Esman, M. 2001. "Policy Dimensions: What Can Congress of Deputies. 1999a. Informe de la Sub- Development Assistance Do?" In Ronald J. comisión, creada en el seno de la Comisión de Herring and Milton J. Esman, eds. Carrots, Política Social y empleo, para el estudio de la prob- Sticks, and Ethnic Conflict: Rethinking Develop- lemática del pueblo gitano. Madrid, Spain: Con- ment Assistance. Ann Arbor: University of greso de los Diputados, Comisión de Política Michigan Press. Social y Empleo (December). European Centre on Health of Societies in Transi- Congress of Deputies. 1999b. Informe de la Sub- tion (ECOHOST). 2000. "Health Needs of the comisión para el estudio de la problemática del Roma Population in the Czech and the Slovak pueblo Gitano. Madrid, Spain: Congreso de los Republics: Literature Review." Draft. Diputados, Comisión de Política Social y European Commission, Directorate General for Empleo. Enlargement. 1999. EU Support for Roma Com- Council of Europe. 1995. Framework Convention for munities in Central and Eastern Europe-- the Protection of National Minorities. European Enlargement Briefing. European Commission Treaty Series 157 (1.II.1995). Strasbourg: DG Enlargement, Enlargement Information Council of Europe. Unit. Council of Europe. 2002. Legal Situation of the Roma European Commission, Directorate General for in Europe. Document 9397 (26 March). http:// Enlargement. 2002. EU Support for Roma Com- assembly.coe.int/Documents/WorkingDocs/ munities in Central and Eastern Europe-- doc02/EDOC9397.htm. Enlargement Briefing, May 2002. European Crowe, D. 1991. "The Gypsy Historical Experi- Commission DG Enlargement, Enlargement ence in Romania." In David Crowe and John Information Unit. http://europa.eu.int/ 142 References comm/enlargement/docs/pdf/brochure_ 1 (June). Biannual magazine of the ASGG roma_may2002.pdf (Asociación Secretariado General Gitano). European Committee on Romani Emancipation Goldston, J. and R. Guglielmo. 2001. "Shared (ECRE). 2001. "The Social and Economic Standards." Transitions On-Line (October 11). Inclusion of the Roma--Annual Progress Government of Romania, Ministry of Public Report." Document ECRE (2001) 003, Brussels, Information. 2001. Strategy of the Government 20 October. http://www.romaniworld.com/ of Romania for Improving the Condition of Roma, ec0103a.htm 430/2001. Bucharest, Romania: Ministry of European Roma Rights Center (ERRC). 1996. Sud- Public Information. den Rage at Dawn: Violence Against Roma in Hancock, I. 1997. "The Struggle for the Control of Romania. Budapest: European Roma Rights Identity." Transitions 4 (4/September). Center. Havas, G., Kémeny I., and Lisko, I. 2001. Segrega- European Roma Rights Center (ERRC). 1999. "A tion in the Education of Roma Children. Final Special Remedy: Roma and Schools for the Research Study, 2nd draft. Budapest: MNEKK, Mentally Handicapped in the Czech Repub- OM, Soros Foundation, OTKA. lic." Country Reports Series 8. Budapest: Iliev, I. 1999. "Some Approaches at Measuring European Roma Rights Center. Social Capital among Roma Communities in European Roma Rights Center (ERRC). 2000. Bulgaria: Preliminary Notes." Draft. "Campland: Racial Segregation of Roma in Implementation Report. 1999. Implementation of the Italy." Country Reports Series 9. Budapest: Council of Europe Framework Convention for the European Roma Rights Center. Protection of National Minorities, Government Figlio, D. and Ziliak, J. 1999. "Welfare Reform, the Resolution 2023/1999 (II. 12.). Budapest, Hun- Business Cycle, and the Decline in AFDC gary: Office for Ethnic and National Minori- Caseloads," in Economic Conditions and Welfare ties. http://www.meh.hu/nekh/Angol/ Reform, ed. S. Danziger, Kalamazoo, MI: 4­1.htm Upjohn Institute for Employment Research, Jenkins, R. 1999. "The Role of the Hungarian 17­48. NGO Sector in Poscommunist Social Policy." Fraser, A. 1995. The Gypsies. Oxford: Blackwell. In Linda J. Cook, Mitchell A. Orenstein, and Fresno, J.M. 1994. Análisis socioantropológico sobre Marilyn Rueschemeyer, eds., Left Parties and la situación actual de la Comunidad Gitana en Social Policy in Postcommunist Europe. Boulder, Espana. Documentos Técnicos 2, Madrid: Aso- CO: Westview Press. ciación Secretariado General Gitano. Jiménez González, N. 1993. "Schooling of Gypsy Galloway, R., Rokx, C. and Brown, L. 2000. Infancy in Spain." Paper presented at the "Nutrition Status and Issues in ECA." The Seminar on "Integration and Education of World Bank: Washington D.C. Draft. Gypsy Children," Snekkersten and Elsinore, Gamella, J. F. 1996. La Población Gitana en Andalu- Denmark. cia: Un Estudio Exploratorio de Sus Condiciones Kabachieva, P. and Iliev, I. 2002. "Background de Vida. Sevilla, Spain: Consejería de Trabajo y Paper for the World Bank Bulgaria Poverty Asuntos Sociales, Junta de Andalucía. Assessment." Unpublished. Sofia, Bulgaria: Gheorghe, Nicolae.1983. "The origin of Roma's The World Bank. slavery in the Romanian principalities," Kádár, A., Farkas, L. and Pardavi, M. 2002. Hun- Roma, 7(l):12-27. gary: A Comparison of the EU Racial Equality Gilberg, T. 1974. "Ethnic Minorities in Romania Directive and Protocol 12 with National Anti- Under Socialism." East European Quarterly 7 Discrimination Legislation. Brussels: ERRC, (January): 435­464. Interights & MPG. Giménez Adelantado, A. 1999. "Contexto socio- Kalibova, K. 2000. "The Demographic Character- político y cultural: los espanoles Roma." Roma istics of Roma/Gypsies in Selected Countries in Central and Eastern Europe." In W. Haug, 143 Roma in an Expanding Europe: Breaking the Poverty Cycle P. Compton, and Y. Courgage, The Demograph- Romania and Hungary During Market Tran- ic Characteristics of National Minorities in Cer- sition." Review of Sociology 7 (2): 79­89. tain European States. Population Study 31, Vol. Lewy, G. 2000. The Nazi Persecution of the Gypsies. 2. Council of Europe Publishing. Oxford: Oxford University Press. Kállai, E. 2000. "The Roma and Research on the Liegeois, Jean-Pierre. 1994. Roma, Gypsies, Trav- Roma." In Caught in the Trap of Integration: ellers. Strasbourg : Council of Europe Press,. Roma Problems and Prospects in Hungary. Inter- Liegeois, J-P. and N. Gheorghe. 1995. Roma/Gyp- national Conference, Budapest 22­23 June sies: A European Minority. London: Minority 1999, Budapest: Bureau for European Com- Rights Group. parative Minority Research. Livezeanu, I. 1995. Cultural Politics in Greater Kállai, E. and Törzsök, E. eds. 2000. A Roma's Life Romania: Regionalism, Nation-Building, Ethnic in Hungary. Budapest: Bureau for European Struggle 1918­1930. Ithaca, NY: Cornell Uni- Comparative Minority Research versity Press. Keating, M. and McGarry, J. 2001. Minority Loury, G.C., 1999. "Social Exclusion and Ethnic Nationalism and the Changing International Groups: The Challenge to Economics." Paper Order. Oxford: Oxford University Press. presented to the Annual Bank Conference on Kémeny, I., Havas, G., and Kertesi, G. 1994. "The Development Economics, World Bank, Wash- Education and Employment Situation of the ington D.C., mimeo. Gypsy Community: Report of the 1993/4 Lovatt, C. and Lovatt, D. 2001. "News from National Sample Survey." Working Paper 17. Romania: Hungarian Status Bill." Central Budapest, Hungary: ILO/Japan Project on Europe Review 3(18/21 May). http://www.ce- Employment Policies for Transition in Hun- review.org gary. Macura, V. and Petrovic, M. 1999. "Housing, Kertesi, G. 1994. "The Labor Market Situation of Urban Planning and Poverty: Problems Faced the Gypsy Minority in Hungary." Working by Roma/Gypsy Communities with Particu- Paper 14. Budapest, Hungary: ILO/Japan lar Reference to Central and Eastern Europe." Project on Employment Policies for Transition Document of the Council of Europe, MG-S- in Hungary. Rom (99) 1. Konstantinov, Y. 1999. "Case Study of Roma Marko, J. 2000. "Equality and Difference: Political Heroin Users: Maksouda Quarter, Varna." and Legal Aspects of Ethnic Group Relations" Background Paper for the Bulgaria Consulta- In F. Matscher, ed., Vienna International tions with the Poor Study. Unpublished . Encounter on Some Current Issues Regarding the Kováts, M. 2001a. "The Emergence of European Situation of National Minorities. Kehl: Engel Roma Policy." In Will Guy, ed. Between Past Publishers: Kehl 1997: 67­97. Web version and Future: the Roma of Central and Eastern published 2000. Europe. Hertfordshire: University of Hertford- Martín, F.A. 2000. "Roma in Spain," Background shire Press. paper prepared for the World Bank. Unpub- Kováts, M,. 2001b. "The Political Significance of lished. the First National Gypsy Self-Government." Martinez-Frais, M.L. and E. Bermejo. 1992. Journal on Ethnopolitics and Minority Issues in "Prevalence of Congenital Anomaly Syn- Europe (Autumn). European Center for dromes in the Spanish Gypsy Population." Minority Issues. http://www.ecmi.de Journal of Medical Genetics 29: 483­6. Ládanyi, J. 1993. "Patterns of Residential Segrega- Marx, K. 1985 The Communist Manifesto. Har- tion and the Gypsy Minority in Budapest." mondsworth, England: Penguin Books. International Journal of Urban and Regional Ministerio de Trabajo y Asuntos Sociales. 1999. Research 17(1): 30­41. Secretaría General de Asuntos Sociales, Direc- Ládanyi, J. and Szelényi, I. 2001. "The Social Con- ción General de Acción Social, del Menor y de struction of Roma Ethnicity in Bulgaria, la Familia, "Programa Para el Desarollo del 144 References Pueblo Gitano." March. Madrid, Spain: Min- Organization for Security and Co-operation in isterio de Trabajo y Asuntos Sociales. Europe, Office for Democratic Institutions Ministerio de Trabajo y Asuntos Sociales. 2000. and Human Rights (OSCE ODIHR). 1996. Sit- Servicio Programa de Desarrollo Gitano, uation of Roma and Sinti in the OSCE Region: Memoria del Programa de Desarrollo Gitano Background Material for the Review Confer- 1988. Madrid, Spain: Ministerio de Trabajo y ence. Warsaw: ODIHR. Asuntos Sociales. Organization for Security and Co-operation in Ministry of Labor, Social Affairs and Family of the Europe, Office for Democratic Institutions Slovak Republic. 1997. "Conceptual Intents of and Human Rights (OSCE ODIHR). 1997. the Government of the Slovak Republic for ODIHR Background Report, OSCE Human Solution of the Problems of Romany Popula- Dimension Implementation Meeting, Novem- tion under Current Social and Economic Con- ber 12­28. ditions." Bratislava. Organization for Security and Co-operation in Minorities Act. 1993. Act LXXVII of 1993 on the Europe, High Commissioner on National Rights of National and Ethnic Minorities. Minorities (OSCE). 2000. "Report on the Situ- Budapest: National Assembly, Republic of ation of Roma and Sinti in the OSCE Area." Hungary. http://www.riga.lv/minelres/ The Hague: OSCE. NationalLegislation/Hungary/Hungary_ Orsós, Hegyesi, E., Bóhn, K., Fleck, G., and A. Minorities_English.htm Imre. 2000. "Alternative Schools and Roma Murray, R., ed. 2002. Improving the Roma Situation: Education: A Review of Alternative Sec- Successful Projects from Romania...and Lessons ondary School Models for the Education of Learned. Bucharest, Romania: European Com- Roma Children in Hungary." World Bank mission, EU PHARE Programme RO9803.01. Regional Office Hungary, NGO Studies No. 3. Nord, M. 1988. "Poor People on the Move: Coun- Pace, E. 1993. "The Making of Minorities." In San- try-to-Country Migration and the Spatial dro Costarelli, ed., Children of Minorities: Gyp- Concentration of Poverty." Journal of Regional sies. Innocenti Insights Series. Florence, Italy: Science 38: 329­352. UNICEF International Child Development Open Society Institute (OSI). 2001. Monitoring the Centre. EU Accession Process: Minority Protection-- Panaitescu, P.N. 1941. "The Gypsies in Wallachia Country Reports on Bulgaria, Czech Republic, and Moldovia: A Chapter of Economic Histo- Estonia, Hungary, Latvia, Lithuania, Poland, ry." Translated by Doris Hardman. Journal of Romania, Slovakia, Slovenia. Budapest: Central the Gypsy Lore Society 20 (2/April/Third European University Press. Series): 58­72. Open Society Institute (OSI). 2002. Monitoring the Pejic, J. 1997. "Minority Rights in International EU Accession Process: Minority Protection. Vols. Law." Human Rights Quarterly 19 (3): 666­685. I and II. OSI EU Accession Monitoring Pro- Peterson, J. 2000. "Welfare Reform and Inequali- gram. Budapest, Hungary. ty: The TANF and UI Programs." Journal of Open Society Institute (OSI). 2002. Research on Economic Issues, vol. 34, No. 2, June 2000. Selected Roma Education Programs in Central Pogany, I. 1999. "Minority Rights in Central and and Eastern Europe: Final Report. Hungary, Eastern Europe: Old Dilemmas, New Solu- Budapest: Education Sub-Board of the Open tions?" In Deirdre Fottrell and Bill Bowring, Society Institute. http://www.osi.hu/iep/ eds., Minority and Group Rights in the New equity/roma_report_part1.pdf Millenium. The Hague: Martinus Nijhoff Pub- Orentlicher, D. F. 1998. "Citizenship and National lishers. Identity." In David Wippman, ed. Internation- Poulton, H. 1991. The Balkans: Minorities and States al Law and Ethnic Conflict. Ithaca, NY: Cornell in Conflict. London: Minority Rights Group. University Press: 296­325. Project on Ethnic Relations (PER). 1997a. "The Roma in the Twenty-First Century: A Policy 145 Roma in an Expanding Europe: Breaking the Poverty Cycle Paper." Princeton, NJ: Project on Ethnic Rela- Rona, S. and Lee, L. 2001. "School Success for tions. Roma Children: Step by Step Special Schools Project on Ethnic Relations (PER). 1997b. "The Initiative Interim Report." Open Society Media and the Roma in Contemporary Institute. Europe: Facts and Fictions." Princeton, NJ: Rughinis, C. 2000. "Romania: Local Service Deliv- Project on Ethnic Relations. ery Module on Roma Communities." Back- Project on Ethnic Relations (PER). 1998. Self-Gov- ground paper prepared for the World Bank. ernment in Hungary: The Gypsy/Romani Experi- Sanchez-Paramo, C. 2001. "Unemployment, ence and Prospects for the Future (May 9­11 Skills, and Incentives. An Overview of the 1997). Budapest, Hungary: Project on Ethnic Safety Net System in the Slovak Republic." Relations. Washington, DC: The World Bank. Psacharopoulous G. and Patrinos, H. A. 1994. Santos, Montserrat. 1999. "Cultural Diversity: "Indigenous People and Poverty in Latin Equal Opportunities?" European Journal of America: An Empirical Analysis." Washing- Education 34 (4/December): 437­448. ton, DC: The World Bank. Save the Children. 2001a. Denied a Future: the Puporka L. and Z. Zádori. 1999. "The Health Sta- Right to Education of Roma/Gypsy and Traveller tus of Roma in Hungary." Budapest, Hun- Children in Europe. Volume 1: South-Eastern gary: World Bank Regional Office Hungary, Europe. London, UK: Save the Children NGO Studies, No 2. Fund. Radó, P. 1997. Report on the Education of Roma Stu- Save the Children. 2001b. Denied a Future? The dents in Hungary. Expert Study for the Office Right to Education of Roma/Gypsy & Traveller of National and Ethnic Minorities (ONEM), Children in Europe. Vol. 2: Western & Central Budapest, Hungary: ONEM and the OSI Insti- Europe. London, UK: Save the Children tute for Educational Policy. Fund. Radó, Peter. 2001. "Roma Education Policies in Schafft, K. 1999. "Local Minority Self-Governance Hungary", in Christina McDonald, Judit and Hungary's Roma." The Hungarian Quar- Kovacs, Csaba Fenyes eds., The Roma Educa- terly XL (155/Autumn). http://www.hun- tion Resource Book, Budapest: Open Society gary.com/hungq/no155/ Institute. Schafft, K. and D. Brown. 2000. "Social Capital Ravallion, M. 1993. Poverty Comparisons. and Grassroots Development: The Case of Philadelphia: Harvard Academic Publishers. Roma Self-Governance in Hungary." Social Revenga, A., Ringold, D. and Tracy, W.M. 2002. Problems 47 (2): 201­219. "Poverty and Ethnicity: A Cross-Country Schafft, K. and D. Brown. 2002. "Social Capital, Study of Roma Poverty in Central Europe". Social Networks, and Social Power." Paper Washington, DC: The World Bank. prepared for the Cornell University Work- Ringold, D. 2000. "Roma and the Transition in shop on Social Capital and Civic Involve- Central and Eastern Europe." Washington, ment, Ithaca, NY, September 13­14. DC: The World Bank. Schoeni, R. F. and Blank, R. M. 2000. "What Has Rodgers, G, Gore, C. and Figueiredo, J. 1995. Welfare Reform Accomplished? Impacts on Social Exclusion: Rhetoric, Reality, Responses. Welfare Participation, Employment, Income, Geneva, Switzerland: International Institute Poverty and Family Structure." NBER Work- for Labor Studies, United Nations Develop- ing Paper, No. 7627. ment Programme. Shuinéar, S. 1993. Growing Up as a Gypsy: Roma. 2000. "El pueblo Gitano y la educación." Insights from the October 1992 UNICEF Roma number 7/8, December 2000, Biannual ICDC Workshop. In Sandro Costarelli, ed., Magazine of the General Gypsy Secretariat Children of Minorities: Gypsies. Innocenti Association (ASGG). Insights Series. Florence, Italy: UNICEF Inter- national Child Development Centre. 146 References Silver, H., 1994. "Social Exclusion and Social Soli- UNICEF. 1998. Education for All? Regional Moni- darity: Three Paradigms." International Labor toring Reports 5. Florence, Italy: UNICEF Review 133 (5­6): 531­578. International Child Development Centre. Staines, V. 1999. "A Health Sector Strategy for the UNICEF. 1999. Women in Transition, Regional Europe and Central Asia Region." Human Monitoring Reports 6. Florence, Italy. Development Network. Washington, DC: The UNICEF International Child Development World Bank. Centre. Stewart, M. 1997. The Time of the Gypsies. Boulder, Van der Walle, D. And Gunewardena, D. 2001. CO: Westview Press. "Source of Ethnic Inequality in Viet Nam." Tanaka, J. Bíró, A. Gheorghe, N., Heuss, H., 1998. Journal of Development Economics 65: 177­207. "Toward a Pakiv European Roma Fund: Vaeăka, M. 1999. "The Roma." Bratislava: Insti- Income Generating Programmes for Roma in tute for Public Affairs. Central and Eastern Europe." Report commis- Vaeăka, M. 2000a. "Analysis of the Situation of sioned by the Council of Europe and Roma in Slovakia." Draft prepared for the Freudenberg Foundation. MG-S-ROM (98) 10. World Bank. Council of Europe: Strasbourg. Vaeăka, M. 2000b. "Roma--The Greatest Chal- Tomova, I. 1998. "Ethnic Dimensions of the lenge for Slovakia on its Way into the Euro- Poverty in Bulgaria." Report Commissioned pean Union." Bratislava: Institute for Public for the Bulgaria Social Assessment. Sofia, Bul- Affairs. garia: The World Bank. Vásquez, J.M. 1980. Estudio sociológico sobre los Tomova, I. 2000. "Fiscal Decentralization: Ethnic- Roma espańoles. Madrid: Instituto de Soci- ity Module." Unpublished draft. Sofia, Bul- ología Aplicada. garia: The World Bank. Villareal, F. 2001. "Spanish Policy and Roma." Ulă, O. 1991. "Integration of the Gypsies in ERRC Notebook 2. European Roma Rights Czechoslovakia." Ethnic Groups 9: 107­117. Center. http://www.errc.org/rr_nr2-3_2001/ United Nations (UN). 2001. Joint UN/Romanian noteb2.shtml Government Seminar on the Improvement of the Walsh, N. 2000. "Minority Self Government in Situation of the Roma in Romania. Summary of Hungary: Legislation and Practice." Journal on Seminar Proceedings, 2­3 November. Ethnopolitics and Minority Issues in Europe Bucharest, Romania: The Senate of Romania. (Summer). European Center for Minority http://www.un.ro/English.pdf. Issues, Germany. http://www.ecmi.de United Nations Development Program (UNDP). Wheeler, A. 1999. "Gypsies in Eastern Europe-- 1999. National Human Development Report: Bul- Issues and Possible Actions." Draft Discus- garia 1999. Sofia, Bulgaria: UNDP. sion Paper. Washington, DC: The World Bank. United Nations Development Program (UNDP). Wippman, D. 1998. "Introduction." In David 2003. The Roma in Central and Eastern Europe: Wippman, ed. International Law and Ethnic Avoiding the Dependency Trap. Bratislava, Slo- Conflict. Ithaca, NY: Cornell University Press: vakia: UNDP. 86­111. UNICEF. 1995. Children of Minorities: Deprivation World Bank. 2000a "Balancing Protection and and Discrimination. Innocenti Insights Series. Opportunity: ECA Social Protection Strategy Florence, Italy: UNICEF International Child Paper," Washington, D.C.: The World Bank. Development Centre. World Bank, 2000b. Making Transition Work for UNICEF. 1997. Children at Risk in Central and Everyone: Poverty and Inequality in Europe and Eastern Europe: Perils and Promises. Region- Central Asia. Washington, DC: The World al Monitoring Reports 4. Florence, Italy: Bank. UNICEF International Child Development World Bank. 2000c. "The Road to Stability and Centre. Prosperity in South Eastern Europe: A 147 Roma in an Expanding Europe: Breaking the Poverty Cycle Regional Strategy Paper." Washington, DC: Welfare of Roma in the Slovak Republic." The World Bank. Bratislava. World Bank. 2000d. Romania Local Social Services World Bank and National Commission for Statis- Delivery Study. Volumes I and II. Washington, tics. 1998. From Poverty to Rural Development. DC: World Bank Human Development Sector Zamfir E. and Zamfir C. 1993a. Gypsies: Between Unit, Europe and Asia Region. Report No. Ignoring them and Worrying About Them. 23492-RO. Bucharest, Romania: Editura Alternative. World Bank. 2001a. Attacking Poverty: World Zamfir, E. and Zamfir, C. 1993b. "The Romany Development Report 2000/2001. Washington, Population." Social Policy Series 8. Bucharest, DC: The World Bank. Romania: Bucharest University and the Insti- World Bank. 2001b. "Slovak Republic: Living tute for the Quality of Life. Standards, Employment and Labour Market Zamfir and Zamfir. 1996. "Children at Risk in Study." Report 22351-SK. The World Bank: Romania: Problems Old and New." Florence, Washington, D.C. Italy: Unicef Child Development Center. World Bank, 2002. "Poverty and Welfare of Roma Zang, T. and R. Levy. 1991. Destroying Ethnic Iden- in the Slovak Republic." Washington, DC: The tity: The Persecution of Gypsies in Romania. New World Bank. York: Helsinki Watch. World Bank, Foundation SPACE, INEKO, the Open Society Institute. 2002. "Poverty and 148 R THE WORLD BANK GROUP 1818 H Street, NW oma have suffered from severe poverty and Washington, DC 20433 USA Telephone: 202 473-1000 exclusion throughout European history. For many Roma Internet: www.worldbank.org in Central and Eastern Europe, the period of transition E-mail: roma@worldbank.org from communism has been especially dire. Low education and skill levels, compounded by discrimination, have led to widespread long-term unemployment and deteriorating living conditions. Their plight has not gone unnoticed. Over the past decade, governments, civil society and the international community have actively supported initiatives to keep Roma children in school, expand access to jobs, and overcome discrimination. Lessons from these projects can make policies more inclusive and can expand their reach. This volume was prepared for the conference "Roma in an Expanding Europe: Challenges for the Future" in Budapest, Hungary, June 30­July 1, 2003. I hope that this conference will catalyze an ongoing dialogue between the new Roma leadership and the wider policy community that will improve the living conditions and future opportunities of Roma over the long term. From the foreword by James D. Wolfensohn, President, The World Bank