UNEQUAL 37966 CITIZENS Gender, Caste and Ethnic Exclusion in Nepal S U M M A R Y A Kathmandu businessman gets his shoes shined by a Sarki. The Sarkis belong to the leatherworker subcaste of Nepal's Dalit or "low caste" community. Although caste distinctions and the age-old practices of "untouchability" are less rigid in urban areas, the deeply entrenched caste hierarchy still limits the life chances of the 13 percent of Nepal's population who belong to the Dalit caste group. UNEQUAL CITIZENS Gender, Caste and Ethnic Exclusion in Nepal S U M M A R Y DFID THE Department For International WORLD Development BANK DFID THE Department For WORLD International Development BANK The World Bank DFID Nepal Nepal Office P.O. Box 106 P.O. Box 798 Kathmandu,Nepal Yak and Yeti Hotel Complex Tel.: 5542980 Durbar Marg Fax: 5542979 Kathmandu,Nepal Tel.: 4226792, 4226793 E-mail Fax: 4225112 enquiry@dfid.gov.uk Websites www.worldbank.org.np, Website www.bishwabank.org.np www.dfid.gov.uk A copublication of The World Bank and the Department For International Development, U.K. The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of DFID, the Board of Executive Directors of the World Bank or the governments they represent. Rights and permissions This material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The World Bank and DFID encourage dissemination of this report and will normally grant permission to reproduce portions of its work promptly. For permission to photocopy or reprint any part of this work please send a request to The World Bank, Kathmandu, Nepal. ISBN 99946-890-0-2 Photo credits: Min Bajracharya: pages 25, 41, 64, 94, 97, 99; Umesh Basnet: pages 48, 92; Rishikesh Dahal: page 85; Chandra Shakher Karki: page 3; Kishor Kayastha: front cover; Mani Lama page xii; DB Maharjan: page 11; Mohan Mainali: page 54; Naresh Shrestha: pages 7, 62, 67, 83, and back cover. FEDO photo page 58. UNICEF Library photos by Kiran Panday: pages 12, 15, 37, 39, 51, 74, 76, 81; by Mani Lama: pages 1, 8, 22, 45, 52, 78, 91; by Hugues Laurenge: pages 34, 95; by Mukunda Bogati: pages 69, 87, and by Martin Chamberlain: page 72. Designed and processed by WordScape (www.wordscapeonline.com). Printed in Nepal. The Gender and Social Exclusion Assessment Team dedicates this book to our friend and colleague, DR. KISHOR PRASAD GAJUREL March 5,1957 ­ April 14,2006. A great teacher, a true scholar and a Nepali citizen whose research has helped us take this step towards the shared goal of building an inclusive Nepal. Contents Foreword viii Preface ix Acronyms and abbreviations xiii Executive summary xv I. Background and framework 1 The GSEA study 4 Social exclusion as a concept 4 Historical and cultural context of exclusion in Nepal 5 GSEA conceptual framework 8 II. Poverty outcomes 15 Defining poverty 17 Economic poverty 18 Human development poverty 25 Political poverty 30 Local power relations and poverty 32 III. Legal exclusion 39 IV. Public discourse and actions 45 Government policy and institutional framework 48 Responses to gender discrimination 50 Responses to caste discrimination 54 Responses to ethnic discrimination 62 V. Inclusive service delivery 67 Improving access to health 69 Improving access to education 75 VI. Inclusive governance 83 Local development groups and coalitions for influence from below 85 Affirmative action 91 VII. Summary and priorities for action 97 Key action points 102 Selected bibliography 110 List of Figures FIGURE 1 Dimensions of exclusion in Nepal 5 FIGURE 2 The Nepal caste pyramid 6 FIGURE 3 The complementary roles of empowerment and social inclusion 9 FIGURE 4 GSEA conceptual framework 9 FIGURE 5 Sites of disempowerment and exclusion are different for the different excluded groups 13 FIGURE 6 Percent composition of the population by caste/ethnicity 18 FIGURE 7 Trend in the incidence of poverty by caste/ethnicity`95/96 and`03/'04 18 FIGURE 8 Household income from remittances as percent of total household income among recipients 23 FIGURE 9 Household and female ownership of house,land and livestock 24 FIGURE 10 Sex ratio by caste/ethnicity 27 FIGURE 11 Assistance during delivery among rural women 27 FIGURE 12 Literacy rate of the population by age and sex 29 FIGURE 13 Net primary and secondary enrolment rates for boys and girls in Nepal 1995-96 and 2003-04 29 FIGURE 14 Percentage of school participation of age 6-10 year olds by gender,caste and ethnicity, 2003-04 29 FIGURE 15 Women in local elected bodies 1997-2002 30 FIGURE 16 Ethnic/caste and gender representation in parliament,1959-1999 31 FIGURE 17 Gender/ethnic composition of elected members (1999 parliamentary elections) 31 FIGURE 18 Wealth ranking based on consumer items by caste/ethnicity 32 FIGURE 19 Composite empowerment and inclusion index (CEI) by gender/ caste/ ethnicity 36 FIGURE 20 Caste/ethnic differences in levels of female empowerment and inclusion 50 FIGURE 21 Distribution of the Dalit population (in percentage) to total Dalit population in Nepal 57 FIGURE 22 Determinants of health outcomes in rural Nepal 70 FIGURE 23 Average age of females at marriage (in rural areas) by caste/ethnicity 73 FIGURE 24 Correlation between mothers' education and the under-5 mortality rates of their children 73 FIGURE 25 Immunisation status of children under five years by caste/ethnicity 73 FIGURE 26 Total number of male and female teachers (2001) 79 List of Tables TABLE 1 Caste/ ethnic groupings ­ simplified analysis 17 TABLE 2 Unstandardisedregressioncoefficientspercapitaconsumption(inNRs.) 21 TABLE 3 Disparities in health outcome indicators by caste/ethnic groups,1996 26 TABLE 4 Infant and child mortality rates by sex,1996-2001 26 TABLE 5 Educational attainment (percent) by caste/ethnicity 28 TABLE 6 Representation in government,1999 60 TABLE 7 Janajati groups classified by NEFIN based on socio-economic status 63 TABLE 8 Indicative typology of group-based institutions 86 List of Boxes BOX 1 An inclusive state 4 BOX 2 Barriers to citizenship 5 BOX 3 Social inclusion and equity 10 BOX 4 Caste and ethnic dimensions of poverty: A summary of NLSS II findings 20 BOX 5 The school as a site of social inclusion 28 BOX 6 Measuring empowerment and inclusion 33 BOX 7 Listening to rural people during the MESI study 35 BOX 8 Non-discrimination and equality 42 BOX 9 Discriminatory laws 42 BOX 10 Examples of discriminatory laws based on religion,caste and ethnicity 43 BOX 11 Examples of discriminatory laws based on sex 44 BOX 12 Different definitions of Dalit sub-castes 56 BOX 13 Key recommendations to improve the understanding of the Dalit sub-castes 59 BOX 14 Fishing for guaranteed livelihoods 89 BOX 15 Mindset needed for effective affirmative action 90 BOX 16 Alternative approaches to proportional representation 93 List of Maps MAP 1 Ethnographic map of Nepal 19 MAP 2 Distribution of the Dalit population 55 MAP 3 Distribution of the Janajati population 61 viii U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal Foreword Social inclusion is one of the four pillars of the Nepal Poverty Reduction Strategy Paper (PRSP)/Tenth Plan. Exclusion remains an important hurdle that Nepal has to overcome in order to be able to attain the development objectives of both the PRSP and the Millennium Development Goals. All major development partners have now incorporated social inclusion as a core pillar in their assistance strategies,vindicating the long-term development vision of His Majesty's Government and the National Planning Commission (NPC). The Nepal Gender and Social Exclusion Assessment (GSEA) brings together the main findings of a multi-year study on social exclusion and how it has affected development outcomes in Nepal. It examines gender, caste and ethnicity as three intertwined institutions or"rules of the game"that determine individual and group access to assets, capabilities and voice based on socially defined identity.The GSEA study, entitled Unequal Citizens, attempts to understand how these institutions affect the people and suggests ways to overcome the obstacles.I am confident that the GSEA will be very useful in translating Nepal's development vision into actions. Some of the answers lie in the interventions recommended by the report,which also provide a sound basis for development partners to continue supporting Nepal's on- going efforts towards building a fair,equitable and inclusive society. The study is the outcome of a collaborative effort by the Department for In- ternational Development (DFID) of the Government of the United Kingdom and the World Bank in close collaboration with the National Planning Commission.The NPC assisted the process by appointing an Advisory Group of key government and civil society stakeholders to provide guidance to the study team. In addition, DFID, the World Bank and the Danish government supported a wide range of background studies and consultations whose findings have been incorporated in the report.The research was undertaken by a team of Nepali and international scholars and develop- ment workers and adequately reflects the reality on the ground. Finally, I congratulate all the GSEA team members and all the men and women of Nepal who took part in the multi-level consultations for producing what is a very comprehensive assessment of Nepali society. The NPC will continuously advocate for greater social inclusion and will continuously monitor progress as part of the implementation of the Poverty Reduction Strategy Paper,the Tenth Plan. Singha Durbar Dr. Shankar Sharma Kathmandu,Nepal Vice Chairman January 2006 NationalPlanningCommission ix Preface The partnership between DFID and the World Bank that made the Gender and Social Exclusion Assessment possible was first discussed on a very mo- mentous day in 2001. I was actually at the DFID office in Palace Street, London talking about possible collaboration on social development issues in Nepal when the news of the September 11 bombings first came out. In the elevator on the way down after the meeting, some DFID staff told me that an airplane had flown into the World Trade Centre. Despite this inaus- picious beginning, the partnership between our two institutions on social exclusion issues in Nepal has been a fruitful one that has grown stronger over the years. In a Memorandum of Understanding that was signed in 2002 with the Bank, DFID agreed to support two thirds of my time as a World Bank staff person to work on a jointly agreed social development work programme around the issue of social exclusion. The GSEA was to be the major product of this partnership. I came out to Nepal to take up the as- signment in October of that year ­ just as discussions on the draft PRSP were taking place. Those discussions were critical and led to a third key partner in the GSEA collaboration, the National Planning Commission. NPC itself decided to bring exclusion issues to the forefront of its analysis and to make social inclusion one of the four pillars of the PRSP. Dr. Shankar Sharma, as Vice Chair of the NPC,invited a group of distinguished scholars and activ- ists working on gender,caste and ethnic issues to form an informal Advisory Group for the GSEA study.This group consisted of Professor Santa Bahadur Gurung, Director of the National Foundation for the Development of Indig- enous Nationalities (NFDIN); Dr. Om Gurung,President of the Nepal Federa- tion of Indigenous Nationalities (NEFIN); Dr. Pushpa Shrestha, Member, NPC; Durga Sob, President of the Feminist Dalit Organization (FEDO) and former Member-Secretary of the National Dalit Commission; Hira Bishwakarma, Dalit Empowerment and Inclusion Project (DEIP); Dr.Durga Pokhrel,former Chair- person of the NepalWomen's Commission and currently Minister forWomen, Children and Social Welfare; Dr.Bina Pradhan and Dr.Meena Acharya. From the beginning,the collaboration with DFID has been very sub- stantive. Frances Winters was DFID's Social Development Advisor during the first few months of the work. She was succeeded by Dr. Rebecca Calder,who guided the GSEA team through the necessary bureaucratic processes on the DFID side and more importantly, infused her DFID colleagues with an x U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal understanding and excitement about the social exclusion agenda ­ thereby helping to embed it as a central element in the DFID Country Assistance Plan. Since Rebecca Calder's transfer to Pakistan in the fall of 2005, Jasmine Rajbhandary, DFID's present Social Development Advisor, has taken on the responsibility for leading the follow-on work on social exclusion at DFID. The whole GSEA team also owes a special thanks to Mark Mallalieu,Head of Office, DFID Nepal for his strong support to the social inclusion agenda. With additional support from DFID and also the Danish Government (on the health and education chapters), we put together a team to carry out the assessment. Members of the GSEA team and their contributions were as follows:1) Mukta Lama Tamang, Dr. Pratyoush Onta and Dr. Seira Tamang on Janajati issues; 2) Dharma Swarnakar and Manjushree Thapa on Dalit Issues; 3) Seira Tamang and Manjushree Thapa on gender issues; 4) Sapana Malla and Sabin Shrestha on legal issues; 5) Dr. Meena Acharya, Chaitanya Subba, Harihar Regmi, Shankar Aryal and Dr. Kishor Gajurel on the statistical profile; 6) Kiran Bhatia,Dr. MarkTurin and Chhaya Jha on education and health; 7) Dr. Stephen Biggs, Dr. Sumitra Gurung and Dr. Don Messerschmidt on group- based approaches, which was worked upon further by Dr. Saubhagya Shah; 8) Dharma Swarnakar and Dr. Mallika Shakya for the budget analysis and 9) Dr. Aruna Rao and Dr. David Kelleher on affirmative action. I served as team leader and contributed the conceptual framework and the chapter on social- cultural and historical foundations of exclusion as well as chapters on macro and micro level poverty outcomes. Dr. Isabella Bassignana Khadka, Binod Bhattarai,Judith Amtzis,Zamila Bunglawala and Bela Malik served as editors, assisted by the core writing team of Lynn Bennett, Pratyoush Onta, Seira TamangandManjushreeThapa. TeamsupportwasprovidedbyKrishnaThapa, Wangmu Sherpa, Sanjiv Shrestha and Tara Shrestha from the World Bank. Thanks also go to FEDO and to the UNICEF office in Kathmandu for sharing many of the photographs reproduced in this volume and to Dr. Harka Gurung for permission to reproduce three of his most recent ethnic and caste maps. It was particularly fortunate that just before I came out to begin work in Nepal, I was granted funds by the Poverty Window of the Trust Fund for Environmentally and Socially Sustainable Development (TFESSD) to develop an instrument and carry out research to track changes in empowerment and social inclusion levels in the context of the Bank-assisted Rural Water Supply xi and Sanitation (RWSS II) project. This research,entitled Measuring Empower- ment and Social Inclusion (MESI),combined qualitative in-depth case studies with data from a survey administered to one man and one woman in 1000 households from 60 villages. Additional funding was received from the GENFUND and a second instalment from TFESSD. This support has enabled the GSEA team to supplement the macro-level analysis we carried out on the Census,Nepal National Living Standard Survey (NLSS) and Demographic and Health Survey (DHS) data sets with analysis of primary data reflecting the ground realities of exclusion in rural Nepal. Dr.Kishor Gajurel led the statisti- cal analysis and Dr.Sondra Hausner and Kim Armstrong oversaw the qualita- tive research. An important part of the GSEA study was the consultation process organized by the team to share the conceptual framework, and later the findings and preliminary conclusions, with key stakeholder groups. In ad- dition to presentations along the way to DFID and the World Bank col- leagues, we also informally shared our framework and findings with a range of other donors including ADB, CARE, GTZ, USAID, SDC and the Gender Do- nor Coordination Group. Most helpful to us, however, were the series of six consultation workshops, facilitated by Bikram Subba and Chhaya Jha, that were held with Dalit, Janajati and women's organizations and with key gov- ernment policy makers during the winter and spring of 2005. An important part of the GSEA has been its close connection with the country assistance strategies and the on-going sectoral and policy re- form work of both DFID and the World Bank. Without the strong support from Ken-ichi Ohashi,World Bank Country Director for Nepal, and the whole Nepal Country Team, the critical link between social inclusion and Nepal's long term reform agenda could never have been so forcefully articulated. UNEQUALCITIZENS:Gender,CasteandEthnicExclusioninNepal is now available. In addition to this summary, a 30 page executive summary has been published, and a shortened version of the full report is also being pre- pared. A Nepali version is also in the works.The entire GSEA report (which in its draft version is over 700 pages) will be placed on the Internet and key chapters will be published separately for those with a special interest on spe- cific issues.But what we hope will be the most important aspect of the GSEA is not its publication as much as the on-going process of introspection and xii U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal debate on social exclusion in Nepal,which the GSEA has tried to capture and advance. Like all social transformation, this will be a long and contentious process that cannot be "projectised" and by its very nature, can only be di- rected and carried through by Nepali citizens. DFID and the World Bank are committed to continuing their support to this process at this critical juncture in Nepal's history. Over the next several years, continued support for the so- cial inclusion agenda will be provided in part by the Bank and by DFID/Nepal's Social Exclusion Action Programme (SEAP), currently in the planning stage. We look forward to the continued leadership of the NPC as Nepal works through its historical legacy of exclusion and forges a polity in which the play- ing field is truly level for all its diverse citizens. Kathmandu, Nepal Dr.Lynn Bennett January 2006 Lead Social Scientist The World Bank, Nepal xiii Acronyms and abbreviations BA Bachelor of Arts B/C Brahman/Chhetri BCN Brahman,Chhetri,Newar CBS Central Bureau of Statistics CDO Chief District Officer CEDAW Convention on Eliminating all forms of Discrimination Against Women CEI Composite Empowerment and Inclusion Index CPN (M) Communist Party of Nepal (Maoist) CPN-UML Communist Party of Nepal (Unified Marxist Leninist) CRC Convention on the Rights of the Child CWC Central Working Committee DDC District Development Committee DEIP Dalit Empowerment and Improvement Plan DFID Department for International Development DHS Demographic and Health Survey DNF Dalit NGO Federation DWO Dalit Welfare Organization EHCS Essential Health Care Services EMI Empowerment Index EOC Emergency Obstetric Care FECOFUN Federation of Community Forest Users Nepal FEDO Feminist Dalit Organization FRC Fisheries Research Centre GSEA Gender and Social Exclusion Assessment HMG/N His Majesty's Government/Nepal HOR House of Representatives HSS Health Sector Strategy ILO International Labour Organization IMR Infant Mortality Rate INGO International Non Government Organization IPM Integrated Pest Management JMC Jagaran Media Centre JUP Jana Uttham Pratisthan LHMC Local Health Management Committees LSGA Local Self-Governance Act MCH Maternal and Child Health MDG Millennium Development Goal MESI Measuring Empowerment and Social Inclusion xiv U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal MLD Ministry of Local Development MOES Ministry of Education and Sports MOF Ministry of Finance MOGA Ministry of General Administration MOH Ministry of Health MOU Memorandum of Understanding MR Mortality Rate MWCSW Ministry of Women,Children and Social Welfare NC Nepali Congress NCDN National Committee for the Development of Nationalities NDC National Dalit Commission NDHS Nepal Demographic Heath Survey NEFA Nepal Education for All NEFIN Nepal Federation of Indigenous Nationalities NEFSCUN National Federation of Savings and Credit Cooperative Unions,Ltd. NFDIN National Foundation for Development of Indigenous Nationalities NGO Non Government Organization NHDR Nepal Human Development Report NHSP Nepal Health Sector Programme NLSS Nepal Living Standard Survey NNDSWO Nepal National Depressed Social Welfare Organization NPC National Planning Commission NWC National Women's Commission PCRW Production Credit for Rural Women PMAS Poverty Monitoring Analysis System PR Proportional Representation PRSP Poverty Reduction Strategy Paper RPP Rastriya Prajatantra Party RWSS Rural Water Supply and Sanitation SEAP Social Exclusion Action Programme SII Social Inclusion Index SLC School Leaving Certificate SMC School Management Committee SPOSH Society for Preservation of Shelters and Habitations in Nepal SWAp Sector-Wide Approach SWC Social Welfare Council TFESSD Trust Fund for Environmentally and Socially Sustainable Development UNDP United Nations Development Programme UNICEF United Nations Childrens' Fund VCDP Vulnerable Community Development Plan VDC Village Development Committee WB World Bank WDO Women Development Officer WDS Women Development Section xv EXECUTIVE SUMMARY xvi U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal EXECUTIVE SUMMARY Background and framework The Gender and Social Exclusion Assessment (GSEA) examines old hierarchies that continue to structure access to political influence and economic oppor- tunities. Democracy was established in Nepal in 1990. Even in the demo- cratic polity, however, women, the formerly "untouchable" castes who now callthemselvesDalits,theethnicgroupsorJanajatis,theMuslimsandtheplains dwellers or Madhesis remain on the margins. The GSEA has examined gender, caste and ethnicity as three inter- locking institutions that determine individual and group access to assets, capabilities and voice based on socially-defined identity. Inclusion is one of the four pillars of Nepal's Poverty Reduction Strat- egy Paper (2003). However, attaining its inclusion goal will require funda- mental shifts not only in the structure of governance and access to economic opportunity but also in the underlying hierarchical norms, values and behaviours that govern social interaction. In Nepal political and economic power was consolidated by interlinking it with the Hindu caste system. The priestly Brahmans were at the top of the ritual order, with the Kshatriya (kings and warriors) just beneath them and in command of the political order; next came the Vaishya (mer- The transforma- chants) and the Sudra (peasants and labourers). Beneath everyone were oc- tion from subjects cupational groups,considered "impure", and "untouchable" or acchut. In the to citizens remains Hills, in-migrating Hindus of Caucasoid stock made up the priests and war- incomplete. riors and the lowest "untouchable"groups. The middle rank was accorded to E X E C U T I V E S U M M A R Y xvii indigenous groups,the Janajatis,generally of Mongoloid racial stock.Officially abolished in 1963, caste-based discrimination, while diluted, remains even today. During the Panchayat period (1962-1990) ­ although directly ruled by a king ­ Nepalis for the first time began to think of themselves as citizens rather than subjects. The transformation from subjects to citizens remains incomplete. Nepal's new Constitution (1990) established a more inclusive state. It describes Nepal as "multi-ethnic, multi-lingual and democratic" and de- clares that all citizens are "equal irrespective of religion, race, gender, caste, tribe or ideology". However,it also retained some ambiguities ­ by declaring Nepal a Hindu Kingdom,denying women the right to pass their citizenship to their children and explicitly protecting "traditional practices". On February 1, 2005 the King suspended democracy and began di- rect rule. The parliamentary parties have continued to protest against the new order and demand the restoration of democracy. The suspension of democratic rule could delay the advancement of the rights of all Nepalis, especially the most excluded populations, mainly women, Dalits and Janajatis. The GSEA framework The GSEA analyses relationships between people and the institutions or "rules of the game" that shape the opportunity structure of their social,political and economic world. Empowerment and social inclusion are means to shift these relationships and the institutions that embody them towards greater equity. There are three domains of change where the state, civil society and donor organizationscanintervenetoensurethefollowingforthepoorandexcluded: n access to livelihood assets and services; n the ability to exercise voice,influence and agency; and Empowerment n amoreequitableopportunitystructurewith"rulesofthegame" thatallow all citizens to participate on the same terms in the life of the state and and social larger society as well as in their access to livelihood opportunities and inclusion play political influence. complementary "Access to assets and services"and"voice,influence and agency"are roles in promoting part of the empowerment process. The other domain of change,the "rules of equity of agency the game", is where social inclusion does, or does not, take place. Empower- and sustainable ment and social inclusion play complementary roles in promoting equity of prosperity for all. agency and sustainable prosperity for all. xviii U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal Poverty outcomes The GSEA examines poverty outcomes using indicators of economic well be- ing,human development levels,and voice and political influence. Nepal began generating data on caste and ethnicity only in 1991. The 2001 census listed 103 social groups. Numerically no single group is pre- dominant and the population can be broadly divided into Hindu caste groups, Janajatis and the religious minorities (mostly Muslim). In 2001, caste groups constituted 57.5 percent of the population, Janajatis 37.2 percent and the religious minorities 4.3 percent. The Nepal Living Standards Survey, 2003/04 estimated that 31 percent of Nepalis were living below the poverty line. The Brahman/Chhetri group and the Newars have the fewest households in poverty and the Tarai Middle Castes also have low proportions under the poverty line. In contrast, almost half of all Dalits live in poverty, and poverty incidence among Hill Janajatis and Muslims is significantly higher than the national average. How- ever,this data must be approached with some caution because Janajati pov- erty aggregates mask intra group differentials. The analysis also reveals that certain groups pay a "penalty" in terms of lower household per capita con- sumption because of their caste,ethnic or religious identity. A gender dimension of poverty affects health and education out- comes and leads to greater economic insecurity for women. Political poverty is manifested in the main political parties'failure to increase participation of women, Dalits and Janajatis in governance institutions. Dalits were almost totally absent from parliament during the entire multiparty period. The GSEA carried out a separate study to measure and analyse the rela- tive empowerment and social inclusion levels of a sample of one man and one woman from 1000 households in 60 villages. Members of the Brahman/Chhetri/ Newar(BCN)groupsscoredthehighestonbothempowermentandsocialinclu- sionandtheDalitswereatthebottom. Janajatiswere intermediatebetweenthe two groups ­ closer to the BCN group in some measures of livelihood empower- mentbutclosertoDalitswithrespecttoothermeasures. Inallgroupsmenscored Certain groups pay higher than women, but BCN and Middle Caste and Janajati women all scored a"penalty"in terms higher than Dalit men. Statistical analysis revealed the following: of lower household n Caste and gender together account for a third of the variation in per capita empowerment and inclusion levels. consumption n Caste is a more powerful predictor of empowerment/inclusion than because of their gender. caste,ethnic or n Membership in local groups was associated with higher empowerment religiousidentity. and inclusion. E X E C U T I V E S U M M A R Y xix Legal exclusion The lack of laws is not the main issue in Nepal. The Constitution and the Civil Rights Act of 1955 prohibit discrimination on the grounds of "religion, race, sex,caste,tribe,ideological conviction or any of these". The laws also prohibit untouchability, denial of access to any public place or depriving citizens of the use of public utilities. Enforcement is lax,however. Discriminatory provi- sions also exist, such as the declaration of Nepal as a Hindu kingdom and of Nepali as the only official language ­ and the protection of "traditional prac- tices", which has been used to bar Dalits from temples and to permit contin- ued caste discrimination. The law denies women equal inheritance rights and the right to pass citizenship to their children. Existing laws are inadequate to deal with sexual offences and Nepal has no law to deal with sexual harassment. Public discourse and actions This section examines how the "rules of the game" have influenced Nepal's The three major excluded groups in terms of government policies and institutional structures. social movements Until April 1990, Nepal's movements for women, Dalit and Janajati rights re- remain mained subsumed within the larger struggle for democracy. independent of The women's movement has succeeded in placing questions of each other,despite gender equality and justice on the national agenda,and the Dalit movement their many has begun to challenge Nepal's caste society. The Janajati movement, once commondemands. described by many Brahmans and Chhetris as a "divisive" phenomenon, has now succeeded in bringing fundamental issues of fair ethnic representation to thefore. Exclusionandhierarchywithinexcludedgroupsisalsobeingquestioned. Thethreemajorsocialmovementsremainindependentofeachother, despite their many common demands. Because little dialogue has taken place between them, the demands of some groups contradict those of others. This has given the state space to delay fulfilment, and in turn has resulted in the growth of radical or revolutionary offshoots. Government policy and institutional framework Nepal's Eighth Plan (1992-1997),the first formulated by a democratic govern- ment,introduced poverty alleviation as one of its three objectives. It was also the first public document to address the caste/ethnic issue, albeit indirectly and incompletely. The Ninth Plan (1997-2002) addressed Dalits and Janajatis by name ­forthefirsttime­andhadaseparatechaptersubsectiondealingwithAdivasi xx U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal Janajatis in development. The government began allotting some public funds to programmes for Janajatis. Planned efforts to improve the situation of women began in the Sixth Plan(1981-1985)buttheapproachwaswelfaredriven. TheEighthPlanraisedthe issue of women's representation in decision-making and acknowledged the ex- istence of gender-based discrimination ­ but failed to define either term. The Tenth Plan (2002-2007), the PRSP, recognises that lack of voice, political representation and empowerment are as important dimensions of poverty as are the economic and human development dimensions, and pro- poses "affirmative action" to level the playing field. However, it too fails to present a realistic strategy and concrete mechanisms to mainstream inclusion. Responses to gender discrimination Before 1990 women's issues were cast in the framework of development and welfare ­ not rights. Nepal's Constitution does not permit discrimination on the basis of sex and advocates special legal provisions to protect and advance the inter- ests of women. The Local Self Governance Act (LSGA),1999 introduced man- datory representation of women in local government. However,similar inter- ventions are lacking at higher levels. Nepal has ratified the Convention on Elimination of all forms of Dis- crimination Against Women (CEDAW). The CEDAW requires Nepal to change about 85 laws and 137 legal provisions that are discriminatory, a task which remains to be done. The Ministry of Women, Children and Social Welfare (MWCSW) lacks adequate financial and human resources to carry out its numerous responsi- bilities effectively. It has also largely failed to consider the priorities and needs of women from traditionally excluded castes and ethnic groups. Nepal set up the NationalWomen's Commission (NWC) in 2002. How- ever, it lacked a legal basis and its mandate remained unclear. Its members retired in March 2004 and replacements had not been appointed by Febru- ary 2006. Brahman and Chhetri women ­ appointed mainly on the basis of Nepal needs to their political affiliation ­ dominated the NWC membership. change about 85 Despite various efforts,the kind of structural change implied by the laws and 137 legal term"gender mainstreaming"has not occurred. Tension also exists between provisions that are technocratic "fixes"and those advocating longer-term socio-political change. discriminatory,a The latter is more likely to occur, as a process of democratic trial and error ­ task which remains often led by ordinary people ­ tends to be "messier" and less amenable to to be done. donor timetables and budget cycles. E X E C U T I V E S U M M A R Y xxi Representation of women in political parties is low,especially at the higher echelons of power. Non-representation remains a major obstacle to the mainstreaming of policies and programmes that focus on women and other excluded groups. Responses to caste discrimination DalitsremainattheverybottomofNepal'scastehierarchy. Evennow,thegovern- ment and many development/aid organizations use euphemisms such as"occu- pational castes", "backward classes", "marginalised", and"disadvantaged groups", insteadofreferringtothemasDalits. ThehesitationtousethetermDalitdeflects attention from the everyday reality of caste-based discrimination in Nepal. Over 200 forms of caste-based discrimination have been identified in Nepal. Discrimination is more entrenched in the country's less-developed areas, especially in the Mid- and Far-western regions, but caste continues to influence inter-personal behaviours throughout the country. TheTenth Plan No consensus has been reached on exactly which communities fall (2002-2007),the into the category of Dalit or on the actual population size. According to the PRSP,recognises 2001 Census,Dalits comprised 13 percent of the population but the figure is that lack of voice, contested. The Dalits can broadly be categorised as either Hill Dalits (who political make up 61 percent of the Dalit population) or Tarai Dalits. Ironically,among representation and themselves the Dalits have traditionally practiced Hindu type stratification. empowerment are Unlike many Janajatis, the Dalits have no geographical centre or"traditional as important homeland"where they are numerically predominant. dimensions of NepalestablishedtheNationalDalitCommission(NDC)inMarch2002 poverty as are with an all-Dalit membership. Its members were chosen based on party affili- economic and ations;its functions were not legally mandated and funding was inadequate. human The NDC did draft a bill for itself but it had not yet become law by early 2006. development Dalit representation in the executive bodies of political parties re- dimensions,and mains very low. The only Dalit member of the House of Representatives was proposes elected in 1991. "affirmative The Dalits have essentially been left to fend for themselves. With a action"to level the few exceptions, Nepal's non-Dalit actors have left it to Dalit leaders, activists playingfield. and organizations to"fight their own battle", which has not helped the Dalit movement. Responses to ethnic discrimination The demands of Nepal's Adivasi Janajati movement centre mainly on issues of governance and political representation. One is the need for constitutional reform to remove discriminatory provisions. Another is for equitable repre- xxii U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal sentation. The Janajatis also seek greater equality in linguistic rights,and guar- anteed access to common properties/resources. Nepal originally prepared a schedule listing 61 Janajati groups,which was later reduced to 59 in the law. Various complexities are involved in com- piling a definitive list. Among the groups in the current list 18 are from Moun- tain regions,24 from the Hills,7 from the InnerTarai and 10 from theTarai. The 2001 Census enumerated only 43 of 59 Janajati groups and reported a popu- lation of 8.27 million or 37 percent of Nepal's population. Members of 16 "missing"groups were apparently not counted. Many disparities are found among the different Janajati groups. The Nepal Federation of Indigenous Nationalities (NEFIN) reports that 10 of the 59 Janajati groups are "endangered", 12 "highly marginalised", 20 "marginalised", 15 "disadvantaged" and two "advanced" or better off. The GSEA recommends a fresh classification to identify disadvantaged Janajati groups based on poverty incidence, education levels and key health indica- tors to serve as a basis for eligibility to special state initiatives, including reser- vations and scholarships for those most disadvantaged. Overcoming Nepal's Constitution explicitly uses the term Janajatis and acknowl- the legacy of edges both their presence and their relative social and economic deprivation. past inequality The use of Nepali as the only official language is discriminatory, however. involves more Constitutional reform is both an overarching demand related to many other than allotting issues,and an affirmation that the Janajatis want a wholly reformed contract some reserved with the state. The movement wants Nepal to be declared a secular state,and seatsinelected, all Janajati languages recognised for use in state affairs alongside Nepali. administrative Equitable representation through different methods including "re- government,or structuring the Nepali state" by changing the electoral system and affirma- in university tive action measures are other key demands, as is access to common prop- admissions. erty resources once communally owned by certain Janajati groups. Inclusive service delivery Improving access to health Nepal has started to put a greater emphasis on preventing diseases that afflict the poor and has begun reaching out to those with the greatest health burden. However, the effort to reorient policy and health services along a rights-based approach remains ad-hoc and immature. Many interrelated factors ­ cultural, religious and social beliefs and norms (especially those that reflect the entrenched gender,caste and ethnic hierarchies) as well as economic, institutional and location-related specifici- E X E C U T I V E S U M M A R Y xxiii ties ­ lie behind these differential health outcomes. Because of their repro- ductive role and their low social status,outcomes for women are the worst. Many determinants of health outcomes operate indirectly by reduc- ing certain people's access to healthcare and influencing the kind of care they receive. Institutional and political factors are important and are a major focus for policy intervention. These include government budgetary allocation and policy attention to rural healthcare and the diseases of the poor and women. Although many professionals in the government health care ser- vice are dedicated to caring for the poor and disadvantaged, others have little motivation to serve those who are beneath them in the socio-economic hierarchy. Generally, most healthcare facilities, including trained person- nel, are concentrated in urban areas. Men mostly staff the higher service positions ­ a major obstacle to proper healthcare for women. Women's health outcomes are directly affected by their subordinate status vis-à-vis the men and the senior women in the family. The preference for male children varies from group to group and is reflected in poorer female performance on all indicators,especially education and health. When healthcare usage and outcomes are better for women, they are better for children as well. Brahmans,Chhetris and Newars have the best health indicators for women and also the lowest infant mortality rates. The government has acknowledged and tried to address the prob- lem of gender discrimination as a barrier to healthcare. But very little atten- tion has been given to how the legacy of caste and ethnicity ­ and particu- larly the practice of untouchability ­ affects the interface between health service providers and patients of both sexes. For Janajatis and members of linguistic minorities in the Tarai,language is also an inhibiting factor. Some of the determinants of high morbidity and mortality among excluded groups require actions beyond the health system. Improved trans- portation and sanitation infrastructure, reduced income and consumption poverty and increased education levels are all associated with better health outcomes. Meeting the objectives of the current health sector reform programmewillrequirepatientdevelopmentofdetailedformalmodalitiesand mechanisms to overcome the barriers to inclusion. Instead of Improving access to education patronage,(the The state assumed responsibility for the education system in the 1970s; pre- excluded) want viously locally run schools were turned over to a centralised educational ad- rights ­ the same ministration. Public education expanded rapidly thereafter. Quality did not rights accorded to keep up with the expansion in numbers,however. every citizen by law. xxiv U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal To help poor and socially excluded children access the kind of educa- tion that will open opportunities for them, simply getting them into Nepal's public schools as they currently operate will not be enough. The Nepal Educa- tion for All (NEFA) programme sets out three primary objectives: (i) ensuring access and equity in primary education,(ii) enhancing quality and relevance of primary education,(iii) improving efficiency and institutional capacity. For the first objective the government has specifically committed to provide equal accesstoeducationalresourcesforallexcludedgroups­girls,linguisticminori- ties,Dalits and Janajatis. In an effort to reform the system and shift the incentives,in 2001 the parliament passed the Seventh Amendment of the Education Act, allowing management of local public schools to be handed over to School Manage- ment Committees (SMCs). The rules require at least one woman member but do not mandate Dalit or Janajati representation. Participation of both Dalits and women in the SMCs is low. Preventing elite capture and undue politicisation of SMCs is vital for their success, and is only possible through proper representation. Schools with female teachers tend to attract more female students. For that reason the policy of having at least one female teacher per school in multi-teacher schools was established over a decade ago, and the NEFA re- quiresatleasttwofemaleteachersinsuchschools. However, neitherpolicyhas yet been fully implemented. Just as having a woman teacher tends to attract girlstudents,havingJanajatiorDalitstaffhasapositiveimpactonthosegroups. For most Janajati children Nepali is not their mother tongue so they Brahmans, are introduced to school and to a new language at the same time. Success in Chhetris and Nepali medium primary schools is also difficult for many people from theTarai Newars have who speak Maithili,Bhojpuri or Hindi as their mother tongues. the best health Primary education is the foundation for ensuring educational parity indicators for among various groups, the first step towards effective social inclusion. The women and excluded groups are under represented in higher education ­ with Dalits be- also the lowest ing less than one percent of those with BA and above ­ and this is largely due infant to exclusion at the lower levels. Reforming education from below must be mortalityrates. matched with affirmative action initiatives from above to support the higher education of members of excluded groups. Inclusive governance Governance is at the core of the GSEA ­ focusing as it does on equal citizen- ship. It also proposes two promising approaches for realising the equal citi- zenship goal ­ group-based development and affirmative action. E X E C U T I V E S U M M A R Y xxv Local development groups and coalitions Some grassroots groups have begun to replicate themselves and have or- ganized into larger federations. These higher-level associations give voice and added political representation to their constituents. Local level groups are an important mechanism through which bottom-up empow- erment has been taking place in Nepal. This is especially important in the current situation where the elected local bodies that were to be the pillars of grassroots democracy and the institutional anchors for decentralisation have remained inoperative since July 2002. However, the roles and responsibilities of grassroots community groups vis-à-vis local elected government remain to be clarified. Some GSEA findings relating to groups include the following: n Nepal has about 400,000 local-level sponsored groups that are being monitored by development agencies. The idealised n The idealised notion of "community" fails to recognise factional in- notion of terests within communities: class, caste and gender-related conflicts "community" can and do occur even within community forestry groups, which are fails to recognise said to be the most successful of the local groups. Stratification and factional elite capture occur in women's groups as well. interests within n Although women are fairly well-represented as group members, they communities: continue to play a less prominent role on the executive committees. Data class,caste and on group membership and leadership disaggregated by caste and gender-related ethnicity is almost non-existent. conflicts can and n Often homogenous groups ­ in terms of gender,caste and ethnicity ­ are do occur even best suited for serving the interests of disadvantaged groups. Studies within show that "elite capture"is more likely to occur in mixed groups. community The quiet revolution underway in Nepal is the expansion of the im- forestrygroups, petus for group-based collective action from the village level to district, na- which are said to tional (and sometimes international) arenas through the formation of federa- be the most tions and associations of grassroots groups. These actions can only succeed successful of the in an environment that promotes freedom to form associations to pursue localgroups. collective goals as well as government support ­ not restrictions. Affirmative action Affirmative action seeks to correct historical disadvantages and unfair dis- crimination by enabling access to full opportunity and benefits to groups that have been excluded. Overcoming the legacy of past inequality involves more than allotting some reserved seats in elected,administrative government,or in university admissions,etc. xxvi U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal Affirmative action as debated in Nepal relates not only to the civil service, but also to elected government and to the education, employment and health sectors. Affirmative action can also encompass changes in the electoral system in order to ensure proportional representation of different groups. This may be part of the answer in Nepal as well. The government's views on affirmative action are unclear. While the need for some sort of affirmative action for excluded groups is not disputed, the modality has been the source of some contention ­ as has the issue of which groups should be included. A major challenge to the affirmative action agenda is the low num- An effective ber of qualified candidates in certain groups such as the Dalits. An effective "road map"to "road map"to affirmative action needs to address this dimension ­ perhaps affirmative through a special programme to develop a"pipeline"of qualified candidates. action needs to Probably the most contentious sphere for affirmative action is in address this elected government,and this is an area where the political parties have failed. dimension ­ The power structures of the main political parties have never been represen- perhaps tative in terms of the gender, caste or ethnicity of the diverse citizens they through a claim to represent. special Affirmative action as a lever for social inclusion is necessary but not programme to sufficient to bring about significant and sustainable positive outcomes for develop a socially excluded groups. In order to be truly effective and sustainable, affir- "pipeline"of mative action requires the broad social and political commitment to equality qualified and human rights articulated by Nepal's Constitution,laws and policies. Ulti- candidates. mately,it requires changes in people's beliefs and values. Conclusions After centuries of thinking about themselves as subjects of feudal rulers,more and more Nepalis are beginning to see themselves as citizens of a democratic state. Although the pace of this fundamental change in self-perception is uneven among groups at different levels on the social hierarchy, it is now being embraced even by those traditionally at the lowest echelons ­ espe- cially women,Dalits and Janajatis. This change in self-perception has also al- tered expectations:people do not want favours from the powerful. Instead of patronage, they want rights ­ the same rights accorded to every citizen by law. They want uniform"rules of the game"to apply to all social players across the board. Social inclusion and empowerment are the interrelated processes that can bring this about. The GSEA ends with a set of recommendations on the long overdue policy and actions for addressing the various dimensions of social exclusion in Nepal. C H A P T E R 1 BACKGROUND AND FRAMEWORK I BACKGROUND AND FRAMEWORK The GSEA study Social exclusion as a concept Historical and cultural context of exclusion in Nepal GSEA conceptual framework B A C K G R O U N D A N D F R A M E W O R K 3 Democracy was established in Nepal in April 1990. It provided diverse groups space to express their opinions openly and to assert their identities and rights as citizens. However, the dominant order has remained largely confined to male Brahmans (Bahuns) and Kshatriyas (Thakuris and Chhetris) from the traditionally influential Parbatiya or Hill Hindu group, and the urban-based and generally well-educated Newars. The democratic transition failed to de- liver on the promise of an inclusive polity mainly because, like most institu- tions in Nepal,the political parties continued to operate on the basis of deeply embedded and mutually reinforcing feudal,caste and patriarchal norms and networks ­ and were thus unable to represent and articulate the demands of all Nepalis. Those left at the margins were women; the "tribal" indigenous ethnic groups, the Adivasi Janajatis or "indigenous nationalities"; and the formerly "untouchable" castes now calling themselves Dalits ("oppressed", "broken" or "crushed"). Muslims, who have high poverty levels, and the plains dwellers or Madhesis, who have substantial numbers but are largely excluded from political influence,are also on the margins. The fact that even after the advent of multi-party democracy old hi- Women, Janajatis and erarchies continued to structure access to political influence and economic Dalits have not been opportunity led to the radicalisation of the demands of those who felt ne- elected in numbers that match their share glected by the new political order. The Communist Party of Nepal (Maoist), of the population. 4 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal CPN (M),launched its "People'sWar" in February 1996 and included the per- sisting caste,ethnic and gender-based disparities in its political agenda. The Maoists were quick to capitalise on the growing discontent and sense of in- justice,and even though their controlled state model has little space for indi- vidual or group freedoms or effective social change, they have been able to provide important symbolic recognition to disaffected women, Dalits and Janajatis and to bring their demands into public debate. The GSEA study The Nepal Gender and Social Exclusion Assessment (GSEA) has been a col- laborative effort of the UK's Department for International Development (DFID) and the World Bank. It is based on a series of studies, including primary re- search, by a number of Nepali and international scholars and analysts that have been submitted for discussion and review by stakeholders at different levels. The GSEA report has examined gender, caste and ethnicity as three interlocking institutions that determine individual and group access to assets, capabilities and voice based on socially- BOX 1 An inclusive state defined identity. It reports on how these insti- tutions are changing and how state and civil In a socially inclusive state, the individual's identity as society actors are responding to changes a citizen supersedes all other identities (e.g. gender, taking place in Nepal. It examines the linkages caste, ethnicity, language or religion) as a basis for claims for state services and commitments on citizen's between exclusion and poverty and recom- rights (e.g. justice, social service provision, mends strategies and actions to promote investment in public infrastructure, police protection) progresstowardsamoreinclusiveandequitable given in the constitution and the legal system. society. The GSEA was envisioned, researched and written between late 2002 and mid-2005. Social exclusion as a concept Social exclusion gained prominence in public discourse after inclusion was incorporated as one of four pillars of the 2003 Poverty Reduction Strategy Paper (PRSP),which is also Nepal'sTenth Plan. As a result, there is now greater understanding that social exclusion is a structural problem and that solving it requires the state to move beyond welfare handouts to define and assure citizenship rights to all. In return,Nepal's citizens need to shift from a mindset of dependency and patronage to one of individual and collective responsi- bility ­ the"price"for their rights. Balanced social and economic development can be attained only when rights and responsibilities complement each other. Therefore,attaining the PRSP's inclusion goal will require fundamental shifts B A C K G R O U N D A N D F R A M E W O R K 5 FIGURE 1 Dimensions of exclusion in Nepal Social Category Gender Caste Ethnicity/Race Language Religion Geo-political Status Dominant Men/Boys Tagadhari: Caucasoid Nepali Hindu Parbatiya (Hill dweller) Brahman, Chhetri Subordinate Women/Girls Dalit Janajati/ Other Non-Hindu Madhesi (Plains dweller) Mongoloid not only in the structure of governance and access to economic opportunity but also in the underlying hierarchical norms,values and behaviours that gov- ern social interaction. It is interesting to note that the seven party alliance is lobbying for a return not just to democracy, but to inclusive democracy ­ perhaps signalling a realisation that the first decade of multi-party democ- racy in Nepal came up short in that dimension. Nepal's geography also contributes to exclusion. There are urban/ rural differences in access to markets, services, information and political in- fluence. Age and disability can also be serious barriers to inclusion,but they are not a specific focus of the report. Another exclusion is both geographic andhistorical­thedistinctionbetweentheParbatiya(hilldweller)andMadhesi (plains dweller). (See Figure 1.) Historical and cultural context of exclusion in Nepal The priestly Brahmans were at the top of the caste hierarchy with the Kshatriya (kings and warriors) just beneath them;next came theVaishya (merchants) and BOX 2 Barriers to citizenship At present, citizenship itself is problematic for many members of excluded groups, particularly for women who cannot pass citizenship rights on to their children and for many Dalits and Janajatis - especially those in the Tarai because of their high levels of landlessness. Although nothing in the legal code requires proof of land ownership as a basis for citizenship, officials continue to demand it. This requirement has excluded many landless Tarai Dalits and Janajatis from the basic rights and protection of citizenship. Included among these is the right to migrate for employment, which could, ironically, put them in a position to be able to buy land. Nepalis who are identified by their language, dress and customs as being "of Indian origin" face particular difficulty in government offices staffed predominantly with Nepali speakers of Hill origin. These people face many informal barriers to getting their citizenship papers - no matter how many generations their families may have lived in Nepal. Since the number of people born and residing in Nepal, but lacking citizenship papers is estimated to be around 3.4 million (Dhanapati Upadhyaya Commission, HMG/N, 1995) this is not an insignificant issue. The right to citizenship need not pose such a problem. In fact, the Interim Constitution of 1953 granted the right to citizenship to "every person who had been permanently residing within the territory of Nepal with their family". The 1990 Constitution greatly restricts this right. 6 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal FIGURE 2 The Nepal caste pyramid the Sudra (peasants and labourers). (See Figure 2.) Be- neath everyone were occupa- tional groups, considered "im- pure", and "untouchable" or acchut. Theynowcallthemselves the Dalits. In the Hills the top two ranks (priest and warrior) and the lowest ("untouchable") rank were filled by the in-migrating Hindus of Caucasoid stock who spokeanIndo-Aryanlanguageon which modern Nepali is based. The middle rank was accorded to indigenous groups, generally of Mongoloid racial stock. These groups - classified by the Hindus as Matwali or liquor drinkers - generally spoke Tibeto-Burman languages and followed Buddhism or various shamanist/ani- mist religions. The Matwalis comprise the Adivasi Janajatis (indigenous nation- alities). The MulukiAin or Country Code (1854) brought all these diverse groups together under a single legal system, but accorded differential privileges and obligations to each caste and sub-caste. For many groups, therefore, the con- quest by the rulers of Gorkha and their subsequent unification of Nepal was an "exclusionaryinclusion". During the Shah-Rana era (1768-1951), Nepal had no alternative "institutions" or ideologies backed by any economic and political power equivalent to the feudal regime. Especially during the rule of the Rana oligar- chy (1847-1951), the caste system and the patriarchal gender system of the dominant group were reinforced by the state. It was an era of consolidation of power and entrenchment of social inequity that can occur in the absence of competing world views. During the Panchayat period (1962-1990), the state attempted to builda"modern" and "unified"nation. Althoughdirectlyruledbyaking,Nepalis were for the first time beginning to think of themselves as citizens rather than subjects. Nepal abolished caste-based discrimination in 1963. However, the diversity of languages, gender, kinship systems and spiritual outlooks of Nepal's many different social groups were framed as barriers to development that "had" to be merged to conform to a single common "modern" Nepali culture. Cultural"unity"was projected as essential to nation-building and the maintenance of independence. B A C K G R O U N D A N D F R A M E W O R K 7 The Constitution of 1990, drafted after the Jana Andolan or People's Democratic Movement against the Panchayat regime, established Nepal as a more inclusive state. It describes the country as "multi-ethnic, multi-lingual and democratic" and states that all citizens are "equal irrespective of religion, race, gender, caste, tribe or ideology". The Constitution also gives all com- munities the right to preserve and promote their language,script and culture, to educate children in their mother tongue,and to practise their own religion. Nevertheless, itretainssomecontradictionsandambiguities­declaringNepal For women the home and family is a key site a Hindu Kingdom, denying women the right to pass their citizenship to their where norms, beliefs and children and explicitly protecting "traditional practices". These have left room behaviours have to be changed to enable them for the continuation of gender and caste-based discrimination. to exercise their agency. The new Constitution al- lowed space for another major de- velopment ­ the growth of civil so- cietyorganizations, especiallythose based on ethnic and caste identity. The post-1990 period witnessed the dismantling of the old projection of a "single Nepali culture" based on that of upper-caste Parbatiyas. Self- chosen terms like Dalit and Janajati emergedtoreplacetermslike"tribal", Matwali and "sano jat" ("small caste") that had been used to de- scribeethnicand "lowcaste" groups. However, in many hierarchical institutions, especially the powerful informal networks, behavioural norms and expectations remained unchanged. Therefore the unitary, centralised and non-inclusive state structure is still largely unchal- lenged. The political parties failed to adequately integrate issues of ex- clusion into their action plans, and even aid agencies, focused on their political need to disburse aid, did not for the most part insist on fun- damental changes in the rules of the game. 8 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal On February 1, 2005,the King began direct rule,as chair of the Council of Ministers. Some new institutions reporting directly to the King have been created,controlshavebeenplacedonthemediaandcivilsocietyorganizations and there is a widespread sense that constitutionally guaranteed freedoms are under threat. The parliamentary parties have continued to protest against directruleanddemandrestorationofdemocracy. Nepal'seffortstochangethe lives of the poor and excluded remain caught up in uncertainty resulting from theunresolvedpoliticaltusslebetweentheKing, politicalpartiesandtheMaoists. The suspension of democratic government in February 2005 (for three years) by the King could delay the advancement of rights of all Nepalis,especially the most excluded populations: women,Dalits and Janajatis. Women's daily wages in the nonskilled, nonagricultural sector GSEA conceptual framework are NRs. 54 compared The GSEA analyses relationships between people, institutions and organiza- to NRs. 104 for men. tions. Institutions are defined as the "rules of the game". Organizations are groups of individuals, bound by a common purpose, involving a defined set of authority relations and dedicated to achieving objectives within particular "rules of the game". The interrelatedprocessesofempowermentand social inclusion are means to shift these re- lationships and the institutions and organi- zations that embody them, towards greater equity and overall prosperity as shown in Figure 3. At the core of this conceptual framework are three domains of change where the state, civil society and donor or- ganizations can intervene to improve access to the following for the poor and excluded: n livelihood assets and services; n the ability to exercise voice,influence and agency; and n a say in framing "the rules of the game" that mediate and regulate people's participationinthelifeofthestateandlarger society as well as their access to livelihood opportunities and political influence. B A C K G R O U N D A N D F R A M E W O R K 9 The first two do- FIGURE 3 The complementary roles of empowerment and social inclusion mains of change ("access to Social Inclusion: changes - at the system level - in the external institutional assets and services" and environment or the rules that determine distribution of the assets, "voice,influenceandagency") capabilities and voice necessary to exercise agency are part of the empowerment process. The last domain of Negotiation for more inclusive Equality of Agency & & equitable institutions Sustainable Prosperity change ("rules of the game") is where social inclusion does, Empowerment: changes - from below - in the internal self-perception and or does not,take place. sense of agency of marginalised groups and their access to assets, The definitions of capabilities and voice empowerment and social in- clusion used by the World Bank (2002) are as follows: n Empowerment is the enhancement of assets and capabilities of diverse individuals and groups to function and to engage, influence and hold accountabletheinstitutionsthataffectthem. n Socialinclusionistheremovalofinstitutionalbarriersandtheenhancement of incentives to increase the access of diverse individuals and groups to development opportunities. Empowerment is seen as occurring at the individual and group level and, to an important extent, has to do with changes in the internal self-per- ceptions of those who have been in some way negatively defined and ex- cluded by the dominant society. Empowerment builds their sense of agency or their capacity to act on their own behalf, and helps them realise the FIGURE 4 GSEA conceptual framework power they gain from collective ac- tion. This domain of change is pre- sented as a triangle in Figure 4. Em- powerment also has to do with in- creasing their access to assets, ser- vices, andlivelihoodopportunities, as is represented by the pentagon in Fig- ure 4. Social inclusion seeks to bring about system-level institutional reform and policy change to remove inequities in the external environ- ment. Social inclusion requires a shift from an institutional environment that gives some individuals and groups more opportunity to realise 10 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal their agency than others to one where the political system and the rule of law support equal agency for all (Rao andWalton 2004). Social inclusion changes the opportunity structure within which individuals and groups seek to exer- cise their agency. It requires change in incentives and also improved capacity within state and community organizations to ensure that organizations can and will respond equitably to the legitimate demands of all individuals - re- gardless of their social identity. This process, as it leads to greater equality of agency, will bring about sustainable prosperity for all through improved access to the assets and capabilities required for achieving a secure livelihood and broad-based economic growth. Empowerment and social inclusion play complementary roles in pro- moting equity of agency and sustainable prosperity for all. These two concepts are the basis of the GSEA. More often than not, reforms that promote social inclusion are re- luctantly conceded by entrenched power holders who are forced to do so by economic and political events they can no longer control. Occasionally re- forms are also actively championed by change agents who are allies of the poor and who may have come to power BOX 3 Social inclusion and equity within the current ruling group or from the " .... equity is intrinsically important as a development opposition. In other words, while the inclu- goal in its own right... a broad sharing of economic and sion dimension of the social change process political opportunities is instrumental for economic may be a response to pressure from below growth and development. Broadening opportunities created through empowerment, it can also strongly supports the first pillar of the Bank's beinstigatedfrompositionsofrelativepower development strategy namely, improving the within the existing institutional framework. investment climate for everyone. The interdependence of the economic and political dimensions of development (See Figure 3.) also reinforces the importance of the second strategic The Conceptual Framework in Fig- pillar, empowerment.... these two pillars are not ure 4 is useful in linking the theory of social independent from each other in supporting development change underpinning the GSEA to the kinds but instead reinforce each other." of concrete policy choices and programme actions that government, donors and civil Paul Wolfowitz, President, The World Bank, from the Foreward to the 'World Development Report 2006' society actors can carry out to support that change. It is premised on the understanding that change that does not happen in all three domains will have less impact and will be less sustainable than change that does occur in all three. For example, assets may be increased temporarily by providing food or cash for work, but unless poor people can claim their rights to education there will be no sustainable improvement in livelihoods, people will not be empowered and social exclusion will remain. Processes to increase assets and access to services; to increase voice,influence and agency; and to change the rules of B A C K G R O U N D A N D F R A M E W O R K 11 the game are all interlocking, and the arrows in Figure 4 represent relatively unidirectional change. Meaningful and sustainable change in each domain is unlikely to happen without change in the other domains,and change in one domain generally follows from change in the others. The implications of this are the following. In order for the discriminatory and exclusionary "rules of the game" thatnowexisttochangeinawaythatwillincreasetheaccessofdiversegroups to development opportunities, the poor and excluded who constitute these groups must have greater voice and agency, either through their own repre- sentative organizations or as part of coalitions for change. Furthermore,they must be able to use this new-found voice and agency to influence existing institutions towards greater openness and equity. Rules do not change on their own; discriminatory and exclusionary rules are created and perpetu- ated because they benefit those who hold power. Those who hold power do not change these rules unless they are compelled to do so. Removing barriers To a large extent the and creating and enforcing incentives for change is usually a long and arduous rules/norms/beliefs and behaviours laid down by process that can take years of advocacy and lobbying. the surrounding institutions define who For poor and excluded people to gain greater access to assets and the individual is, how services, the rules of the game must change in their favour. The reason that individuals are valued by society and what they can or cannot do. 12 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal For poor and excluded people to exclusion causes and perpetuates poverty is because exclusionary and dis- gain greater access to assets and services, criminatory rules of the game deny certain groups of people access to those the rules of the game things that will help them rise out of poverty. The livelihood status of poor must change in their favour. and excluded people can be improved in a sustainable manner only when the rules change to be more equitable and just. To meaningfully engage, influence and hold accountable the insti- tutions that affect them, poor and excluded people must have the capabili- ties necessary to voice their rights, to form effective representative organiza- tions, and to forge coalitions for change. Illiterate, ill, starving, isolated and poverty-stricken people cannot do this; they need assets and access to ser- vices and opportunities. Theinter-relatedprocessesofexclusionanddisempowermenttakeplace at various levels. One is within the individual and involves internalisation of a negative definition of his or her own identity. To a large extent the rules/norms/ B A C K G R O U N D A N D F R A M E W O R K 13 beliefsandbehaviourslaiddownbythesurroundinginstitutionsdefinewhothe individual is,how individuals are valued by society,and what they can or cannot do. In Nepal, since males from the dominant privileged caste group have pri- marily defined these institutions,it is not surprising that these institutions are very disempowering for an individual woman, Dalit, Janajati or member of a linguisticorreligiousminority. Critical sites of disempowerment and social exclusion may vary for different categories of excluded people. (See Figure 5.) For example, for women,the home and family is a key site where norms,beliefs and behaviours have to be changed to enable them to exercise their agency. Community norms and formal laws must also be changed, but change in the domestic site is fundamental. In contrast,for Dalits,the local community is where caste- based discrimination is likely to be most strongly enforced and harshly expe- rienced. A senior Dalit man is still dominant within his family despite the re- strictions he faces in the community. However,a Dalit woman who is subordi- nate in both the gender and the caste domains encounters discrimination in thehomeaswellasinthecommunity. Communityleveldiscriminationagainst FIGURE 5 Sites of disempowerment and exclusion are different for the different excluded groups International/Global l International conventions l Internet, media l Trans-national corporations l Bilateral and multilateral development organisations Social Most Most problematic Solidarity problematic Most for Women Groups problematic l Gender for Dalits for Janajatis Individual l Caste l Psychological: self-worth, l Religion sense of efficacy, etc. l Ethnicity l Socialised concepts of Community personal behavioural Household/ Economic/ Ward/Village District Nation Presented as norms, obligations and Family Livelihood & Lowest tiers l Laws harmonious, but entitlements Service User Groups of elected l Policies contains many l Inter-personal & intra- l Credit groups government. l Institutions different interest household relationships l Water user groups l Public resource groups and often l Personal endowments l Forest user groups allocation dominated by l Political local elites. representation 14 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal the Janajatis is much more muted and has in most cases been effectively coun- tered by the pride Janajatis take in their ethnic identity and in the cultural traditions of their own group. For Janajatis,the most problematic site in terms of empowerment and inclusion is at the level of the state - in terms of laws, policies,resource allocation and representation. P O V E R T Y O U T C O M E S 15 C H A P T E R 2 POVERTY OUTCOMES II POVERTY OUTCOMES Defining poverty Economic poverty Human development poverty Political poverty Local power relations and poverty P O V E R T Y O U T C O M E S 17 Defining poverty The GSEA examines poverty outcomes using indicators of economic well be- ing, human development levels and voice and political influence. Data from national surveys were reanalysed and members of the team also carried out primary field research and analysis on empowerment and social inclusion lev- els of men and women and different social groups. Poverty outcomes were examined by caste and ethnic group ­ when data were available, further analysis was done by sub-group. The National Living Standards Survey (NLSS), the major source of data on economic poverty outcomes, is collected at the household rather than the individual level; thus gender disaggregated analysis of this dimension is more limited. Nevertheless, the analysis has drawn on a number of indicators (e.g. asset ownership, labour force participation and wage rates) that are available separately for male and female. Nepalbegangeneratingdataoncasteandethnicityonlyin1991when 60 caste and Janajati groups were listed. The 2001 census listed 103 social groups, based on caste, ethnicity, religion and language and unidentified groups, some numbering less than 0.1 percent of the population. For ease of analysistheGSEAhasorganisedthe103groupsinto10majorcategories (Table 1). Numerically no single group is predominant. Broadly, the population can be divided between the Hindu caste groups and Janajatis, and a third group, the religious minorities (mostly Muslim). In 2001, caste groups constituted TABLE 1 Caste/ethnic groupings ­ simplified for analysis % Total GSEA / NLSS II 2001 Census population (10 groups) (103 groups) Hindu 1. BC (Hill) Brahman, Chhetri, Thakuri, Sanyasi, caste 2. BC (Tarai) Kayashta, Rajput, Baniya, Marwadi, Jaine, Nurang, Bengali groups 3. Tarai Middle Castes Yadev,Teli, Kalwar, Sudi, Sonar, Lohar, Koiri, Kurmi, Kanu, Haluwai, (57.5%) Hajam/Thakur, Badhe, Rajbhar, Kewat Mallah, Numhar, Kahar, Lodha, Bing/Banda, Bhediyar, Mali, Kamar Dhunia 4. Dalits (Hill) Kami, Damai, Sarki, Gaine, Badi 5. Dalits (Tarai) Chamar, Musahar, Tatma, Bantar, Dhusadadh/Paswan, Khatway, Dom, Chidimar, Dhobi, Halkhor, Unidentified Dalit Janajatis 6. Newar All Newari Castes (37.2%) 7. Janajatis (Hill) Magar,Tamang, Rai, Gurung, Limbu, Sherpa, Bhote, Walung, Buansi, Hyolmo, Gharti/Bhujel, Kumal, Sunuwar, Baramu, Pahari, Adivasi Janajati, Yakkha, Shantal, Jirel, Darai, Dura, Majhi, Dunuwar, Thami, Lepcha, Chepang, Bote, Raji, Hayu,Raute, Kasunda 8. Janajatis (Tarai) Tharu, Dhanuk, Rajbanshi, Tajpuriya, Gangai, Dhimal, Meche, Kisan, Munda, Santhal/Satar/Dhangad/Jhangad, Koche, Pattarkatta/Kusbadiya Muslims 9. Muslims Muslim, Churoute (4.3%) Others 10. Others (1%) 18 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal FIGURE 6 Percent composition of the population 57.5 percent of the population, Janajatis by caste/ethnicity 37.2percentandthereligiousminorities4.3 percent. As shown in Figure 6, among the Religious Minorities Unidentified 10 major social groups, the Hill Brahmans (Muslims) 1% andChhetris(B/C)hadthehighestnumbers Tarai Janajatis 4.3% 8.7% (30.9% of the population), Hill Janajatis (in- Hill B/C cluding the Newars) 28.5 percent and the 30.9% Tarai Middle Castes 12.9 percent. The population density of Nepal Other Hill varieswidely. ApproximatelyhalfofNepal's Janajatis 23% population lives in the Mountain and Hill areasandhalflivesinthemuchsmallerarea Tarai B/C 1.9% of the Tarai plains. Nepal is still largely a Newars rural country,with 86 percent of its popula- 5.5% Tarai Middle Tarai Dalits Castes tion living in rural areas and the remaining 4.7% Hill Dalits 7.1% 12.9% 14 percent living mainly in Kathmandu and other emerging urban areas. The geo- Source: 2001 Census, CBS, Acharya 2004 graphicareaswheredifferentethnicgroups are concentrated are shown in Map 1. FIGURE 7 Trend in the incidence of poverty by caste/ethnicity '95/'96 and '03/'04 Economic poverty Income and Consumption poverty The Nepal Living Standards Survey, 2003/4 (NLSS-II) estimated that 31 percent of Nepalis were living below the poverty line. This means they lived in households where per capita expenditure for food and non- food items was beneath that required to purchasetheminimumcaloricrequirement and other "basic needs". This was a signifi- cant drop in poverty incidence from the NLSS I in 1995/6, which showed that 42 percent of the population were below the poverty line. Figure 7 shows that poverty incidence has fallen for all groups, but that theBrahman/ChhetrigroupandtheNewars Source: NLSS I,II, 2004, Gajurel. The comparison of poverty incidence for havethefewesthouseholdsbelowthepov- Tarai Middle Castes and Tarai Janajatis between NLSS I and II should be treated with caution since for NLSS I the Tarai Middle Castes group were erty line, and that the Tarai Middle Castes represented only by the Yadavs and the Tarai Janajatis only by the Tharus. (Figures have been rounded off.) also have relatively low proportions in pov- P O V E R T Y O U T C O M E S 19 Nepal of map Ethnographic Gurung a 1 Hark ce: MAP Sour 20 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal BOX 4 Caste amd ethnic dimensions of poverty: A summary of NLSS II findings The All Nepal Poverty Incidence in 2004 was 31% but was greater amongst selected groups: n Hill Dalits: 48% n Tarai Dalits: 46% n Hill Janajatis: 44% - but there is wide variation between Hill Janajati groups n Muslims: 41% Poverty incidence dropped 11 points between 1996 and 2004 - from 42% to 31%; it declined across all regions, quintiles, rural/urban and caste and ethnic groups. However: n For Brahman/Chhetri groups, it decreased most significantly, by 46% n For Dalits, the decrease was close to the national average: 21% n For Hill Janajatis, it decreased by about half of the national average: 10%, and n For Muslims, it decreased least of all: 6%. For certain groups, their share of the total population under the poverty line has increased in 2004 compared to 1996: n Hill Janajatis from 19.7% of the poor to 27.8% n Muslims from 5.7% to 8.7% Caste/ethnicity are powerful predictors of per capita consumption. Even after the confounding factors are controlled, the per capita consumption levels in Dalit, Janajati and Muslim households are still between 15 and 13 percent lower than in Brahman/Chhetri households. This difference in consumption levels can be called the "penalty" that certain groups pay because of their caste, ethnic or religious identity. For Dalit households this "penalty" translates into NRs. 4,853 less annual per capita consumption than that enjoyed by similar Brahman/Chhetri households. erty. In contrast,almost half of all Dalits live below the poverty line,and poverty incidence among Hill Janajatis and Muslims is significantly higher than the na- tional average. The data must be approached with some caution both because Janajati poverty aggregates mask intra group differentials and because the sample size for NLSS data was small when compared to Census data. For example, the Gurungs in the NLSS II sample have only 20 percent living in poverty compared to the Tamangs, whose proportion below the poverty line (61%) is almost double the national average. These differences, how- ever,are less extreme when small area estimation statistical techniques are used to project the relatively small NLSS sample onto the Census data. Nevertheless, the results of this exercise show that Tamang poverty inci- dence (at 50%) is still considerably more than that for Gurungs (35%) (Parajuli, forthcoming). Caste and ethnic "penalties" According to NLSS II, the average per capita consumption for Brahman/Chhetri households is NRs.19,105 whereas the per capita consumption for Dalit, Janajati and Muslim house- P O V E R T Y O U T C O M E S 21 holds is NRs.10,207,NRs.12,331, and NRs.10,909 respectively. However, some assert that these welfare differentials may be due to factors other than the effect of caste/ethnicity per se. Proponents of this line of thinking point out that the lower standards of living and higher incidence of poverty of certain caste/ethnic groups may be due to large family size or a higher proportion of dependent children, or to the fact that they live in disadvantaged remote areas - rural Hill areas or the Far-Western region. Others in political and devel- opment circles have frequently suggested that the higher poverty levels among Dalits,Janajatis and Muslims result from their lower levels of resource endowment (including educational attainment) and consequent lack of ac- cess to more productive occupations compared to those groups who have been historically privileged. To explore the strength of these various factors and answer the basic question of whether caste/ethnicity differentials in the standards of living still persist when these important confounders are taken into account, a mul- tiple regression analysis was carried out. For the purpose of this exercise the dependent variable is the nominal per capita consumption at current prices, TABLE 2 Unstandardised regression coefficients per capita consumption (in NRs.) Regression Coefficients Variables Model I Model II Caste/Ethnicity (omitted group: B/C + other high caste groups) Tarai Middle Caste -5911 -3724 Janajati -6774 -4099 Newar 8056 2772 Dalit -8898 -4853 Minorities -8196 -4449 Female Household Head 4388 Household Size -46 Proportion of Household Members (omitted group: adults) Proportion of Children up to 6 Yrs. -20184 Proportion of Children 7 ­ 15 Yrs. -13616 Occupation (omitted group: manual labour) Professional 2564 Technical -36 Farm/forestry 1520 Education (omitted group: illiterate) Primary 2404 Above Primary 7618 Remittance 4873 Rural -11348 Hill and Mountain -1980 Midwest and Far West -3179 Constant 19105 32231 Adjusted R2 0.083 0.291 Source: Calculated by Gajurel 22 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal which has been considered as a better index of welfare measure than income in developing regions (Van de Walle and Gunewardenar, 2001). Table 2 compares two models. In Model I the Brahman/Chhetri group is the constant in the regression analysis, which includes only the caste/ethnic variables. The results show that the average per capita con- sumption for Dalit households is NRs. 8898 (or 46%) lower than that of Brahman/Chhetri households. Similarly, the per capita consumption in Janajati households is NRs. 6774 (or 35%) less and for Muslim households it is NRs. 8196 rupees (or 43%) less than the reference. In Model II, which takes into account the above background variables,household per capita Almost half of all Dalits fall below the consumption among Dalits, Janajatis and Muslims is still considerably lower poverty line and poverty among Hill than that of Brahman and Chhetri households. This unexplained gap in Janajatis and Muslims consumption levels can be seen as a"penalty"attached to social identity as a is significantly higher than the national Dalit, Janajati or religious minority and it translates into NRs. 4853, NRs. 4099 average. and NRs.4449 less respectively in per capita consumption. P O V E R T Y O U T C O M E S 23 Other poverty measures FIGURE 8 Household income from remittances and as percent In addition to the poverty incidence of the total household income among recipients (what percentage of a given group arebelowthepovertyline), analysts also consider consumption quintiles and calculate both the poverty gap (which measures the depth of poverty by calculating the amount needed to bring the con- sumption level of a particular group up to the poverty line) and the squared poverty gap (which mea- sures the severity of poverty by giv- ing more weight to the poorest of Source: NLSS II, 2004, tabulated by Gajurel the poor). Nevertheless, when the data are analysed by caste and ethnic group a common pattern emerges re- gardless of the indicator: Hill Janajatis, Dalits and Muslims always fare the worst; at the other end of the spectrum, the Newars and the B/C group al- ways fare the best. Remittances On average, in 2004/05 Nepal was receiving about one billion rupees in remittances every day. Between 1996 and 2004, the percentage of households receiving remittances rose from 23 to 32 percent and the average amount received increased from NRs. 625 to NRs. 2,100 per person per year. While in 1996 only 22 percent of the remittance income came from household members working outside Nepal in countries other than India, by 2004 remittances from countries other than India accounted for 53 percent of the remittance flow (CBS 2004). The group with the largest share of migrant members who are work- ing abroad in places other than India are the Hill Janajatis (29%),followed by the Muslims (24%), who are in high demand in the Middle East. For other groups the proportion of their total migrants in these high-wage countries ranges from 14 percent for theTarai Janajatis andTarai Dalits to 20 percent for the Newars. Remittances make up about one third of the annual household in- come for families who receive remittances. Not surprisingly, since they have the highest proportion migrating to the countries offering the best wages, the Hill Janajati group has the highest average remittance income - followed by the total B/C group and the Newars. But this is a very dynamic sector and new opportunities,which could shift this balance,open up almost daily. The Muslims,who on average have among the lowest average per capita incomes, 24 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal receive a slightly higher percent of their household in- FIGURE 9 Household and female ownership of house, comes from remittances.(See Figure 8.) Women make up land and livestock only 10.8 percent of the migrants sending funds back 100 home, but they constitute 48.5 percent of the recipients 80 60 (NLSS II, CBS, 2004). Percent 40 Gender dimensions of poverty Household 20 level data such as the NLSS are not very useful for docu- 0 House Land Livestock None menting and understanding the gender dimensions of Household ownership poverty. Average household per capita income and con- Female ownership sumption data do not show the intra-household dispari- No female ownership of ties in access and control over household resources and house, land, or livestock may mask significant gender-based differentials in con- Source: Census 2001, Acharya 2004 sumption levels and certainly in economic security. As a window into these disparities, it is useful to look at the data gathered in the 2001 Census on ownership of land, livestock and real estate by gender. In spite of the cultural differences between the caste Hindus and the Janajati groups (and even within each of these groups) in terms of the social norms governing gender relations, land is inherited universally in all communities from the father to the son. Women therefore face much greater economic insecurity than men since their access to what has traditionally been the pri- mary means of production has always been indirect and dependent on their relation as daughter, wife or mother of a land owning male (Acharya and Bennett,1981; Gurung,1999). Only about 11 percent of households reported any land in female legal ownership;six percent reported that women had"some"ownership of a house. (See Figure 9.) Surprisingly,only seven percent reported female owner- ship of livestock,even though for many groups livestock rearing is traditionally a femaletask,andinspiteofthefactthatmanycreditinstitutionsandmicrofinance programmes have targeted women and made loans to them for this activity. Overall,fewer than one percent of households reported female own- ership of all three types of assets (house, land and livestock). While on the whole,77 to 92 percent of households reported owning a house and between 42 to 80 percent reported owning some livestock,only a miniscule proportion of the women in these households had such tangible property in their names. Traditional gender roles continue in force. If households can afford it, women are confined to non-market (unpaid) work in the care economy and family enterprises. When they do work for pay, women are largely confined to less productive jobs. The National Labour Force Survey (1998) shows that: n 72 percent of women are working in agriculture versus 48 percent of men - and this overall pattern holds across all ethnic/caste groups. P O V E R T Y O U T C O M E S 25 n Theproportionofmencurrentlyemployedintheformalnon-agriculturalor "modern"occupationalsectorismuchhigher(21%)thanthatofwomen(6%). n Women continue to be confined primarily to unpaid family labour. Nearly 60 percent of currently employed women fall in this category, compared to 21 percent of men. Due in part to their lower education levels women also earn less than men. Women's daily wages in the agricultural sector are NRs. 47 per day com- pared to the average male wage rate of NRs. 63; women's daily wages in the nonskilled, nonagricultural sector are NRs. 54 compared to NRs.104 for men; and for skilled non-agricultural labour women get an average of NRs.126 com- pared to NRs.315 for men (World Bank 2005a). Human development poverty Among rural women, Newars and Brahman/Chhetris have the highest Health access to trained Health outcomes are affected by many interrelated factors. These include assistance during delivery and cultural, religious and social beliefs and norms - especially those that reflect Janajatis and Dalits have the least. 26 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal TABLE 3: Disparities in mortality rates and life expectancy by caste/ethnic groups, 1996 Caste / Ethnicity Under 5 MR (per '000) IMR (per `000) Life Expectancy* Brahman 69.0 52.5 61.4 Chhetri 109.1 77.8 58.4 Yadav/Ahir 142.0 98.5 54.2 Dalit 171.2 116.5 50.8 Newar 74.9 56.0 63.2 Tamang 141.2 98.0 54.2 Magar 135.9 94.7 54.9 Limbu 133.3 93.2 55.2 Rai 133.0 92.9 55.3 Gurung 126.3 88.6 56.1 Tharu 106.4 76.0 58.7 Muslim 158.3 108.6 52.2 Source: UNDP 2001: Nepal Human Development Report (NHDR) * Disaggregated data by gender and caste is unavailable the entrenched gender, caste and ethnic hierarchies - as well as economic, institutional and location-related specificities. Caste and ethnic disparities Caste and ethnic disaggregated data presented inTable 3 highlights the glaring disparities in the health outcomes of different groups. The Brahman/Chhetri group as a whole and Newars have higher health indicators than other groups, and have the longest life expect- ancy. On average,Brahmans and Newars live 11 to 12 years longer than Dalits andMuslims (UNDP2001). TheBrahman/ChhetrigroupandNewarsalsohave the lowest infant mortality rates - 52.5/77.8 and 56, respectively - compared to the national average of 79 per thousand. A Dalit child is twice as likely to die in its first year than either a Newar or a Brahman child. Under-five mortal- ity is also much lower for Brahmans and Newars (69 and 75 respectively) than for Dalits (171),Muslims (158) or Tamangs (141). This is probably at least partly because women from the first two groups have the highest literacy rates and a strong linkage is found TABLE 4: Infant and child mortality rates by sex, 1996-2001 between mother's education and IMR (per '000) Under 5 MR (per '000) child survival (NDHS 2001). 1996, total 83.7 135.5 Gender disparities Until Male 78.5 118.3 the 2001 Census reported for the Female 101.9 142.8 2001, total 75.2 104.8 first time that female life expectancy Male 64.4 91.2 in Nepal was slightly higher than Female 79.2 112.4 male life expectancy (60.7 years for Source: UNDP 2004: Nepal Human Development Report (NHDR) women, 60.1 years for men), Nepal was one of few countries in the world where a woman's life expectancy was lower than that of a man. Another indicator of female vulnerability in Nepal is the persistence of higher infant and under-five mortality rates for girls than boys. (See Table 4.) P O V E R T Y O U T C O M E S 27 The sex ratio is an important indicator of FIGURE 10 Sex ratio* by caste/ethnicity (* males per 100 females) the relative balance of policy attention given to male and female welfare and the socio-cultural gender val- ues that underpin it. Figure 10 shows the sex ratio by Hill B/C Tarai B/C caste/ethnicity; note that for Tarai based groups this Tarai Middle Castes ratio is high whereas for the Hill Dalits it is low. The Hill Dalits Tarai Dalits "missingwomen"intheconservativeTaraibeltisindica- Newars** tive of marked gender discrimination. Other Hill Janajatis There are significant variations in access to Tarai Janajatis Muslims healthcare between women from different caste and Nepal ethnic groups. Data from the 2001 NDHS show that 85 90 95 100 105 110 access to and use of a range of health and family plan- Ratio ning services for rural women is lowest among Dalit Source: Census 2001, Acharya 2004 (** includes Thakalis, and Tarai Middle Caste women. Knowledge levels of who comprise approximately 0.1% of the population) Dalit women are also very low compared to Newars, Brahman/Chhetris and Hill Janajatis. Contraceptive use among married rural women is lowest for Dalits (28%) and Muslims (15%), while Newars and Tarai Janajatis have the highest contraceptive use. There is still significant unmet contraceptive need, most notably among Muslims (45%),but also among many other groups. The national average stands at 30 percent. Muslim women and the Tarai Middle Caste groups (both practicing female seclusion) have the lowest awareness of HIV/AIDS prevention. Knowl- edge levels of Dalit women are also very low compared to Newars,Brahman/ Chhetris and Hill Janajatis. Tharu women, despite high contraceptive use, have the lowest HIV/AIDS awareness levels among the Janajatis. FIGURE 11 Assistance during delivery among rural women The use of professional help for deliveries is also low - over six of ten births are unassisted (World Bank 2005a). Almost 95 percent of Nepal's Emergency Obstetric Care (EOC) needs remain unmet. The com- bined effect of these factors is at least 12 daily deaths from pregnancy related complications (MoH/DFID 2004). Among rural women, Newars and Brahman/ Chhetris have the highest access to trained assistance during child delivery (Figure 11), and Janajatis and Dalits have the least. The same trend is seen for ante- natal care. Maternal mortality also includes deaths caused by unsafe abortions. According to the Minis- try of Health (2002) at least 20 percent of maternal deaths are caused by unsafe abortions. The demand Source: DHS 2001, Gajurel 28 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal TABLE 5 Educational attainment (percent) by caste/ethnicity Never Attended Grades 1-4 Grades 5-10 SLC and Above Hill B/C 31.6 24.6 33.7 10.1 Tarai B/C 26.6 25.6 29.6 18.2 Tarai Middle Castes 56.8 22.0 17.4 3.8 Hill Dalits 43.3 31.9 23.1 1.7 Tarai Dalits 76.4 14.9 7.5 1.2 Newars 28.5 20.8 27.6 23.0 Hill Janajatis 44.6 27.2 25.2 3.0 Tarai Janajatis 47.2 24.1 24.2 4.5 Muslims 62.4 24.9 11.0 1.7 Nepal 44.0 24.7 24.6 6.7 Source: NLSS II, tabulated by Gajurel. for safe abortion rose after abortion was legalised by the 11th amendment of the Muluki Ain in 2002. The Total Fertility Rate for uneducated women is 4.8 whereas for women with some secondary education it is 2.3 (NDHS 2001). The under-five mortality rate for children of uneducated mothers in Nepal is 121 per 1,000 births: this is 64 percent higher than that for children of mothers with some primary education and nearly double that of children whose mothers have some secondary education (NDHS 2001). The risk of death among children of uneducated mothers is eight times higher than the risk for children of moth- ers with SLC and above education. (See Figure 24,on page 73.) Similarly, the prevalence of underweight children is 78 percent higher,and the prevalence of stunting is 62 percent higher among children whose mothers do not have secondary education when compared with those whose mothers do (World Bank 2005a). Further analysis by the World Bank found that even when con- trolling for income and other confounding variables, Dalit,Tarai Middle Caste and Muslim children had a significantly higher prevalence of both stunting and underweight (World Bank 2005a). Education BOX 5 The school as a site of social inclusion Education is the key to building a more inclu- One Kumal girl reported that the two Dalit sive Nepal since it offers people an outlet from children in her class were treated just like traditional systems of economic dependency other children. "We all eat and play together." and opens opportunities for better jobs. The in- When asked whether their parents minded, creased awareness that comes with more she said, "They do, but we don't tell them!" schooling also provides the foundation needed Bennett, Gajurel et al., 2006 to build a critical awareness of how existing sys- tems have used social differences to maintain P O V E R T Y O U T C O M E S 29 differential power and opportunity struc- FIGURE 12 Literacy rate of the population tures,and the confidence to question them. by age and sex Table5 showsthegreatrangeinthe 80 Age 6+ Age 15+ percentage of different groups who have 60 ever been to school. Fewer than 30 percent 40 of Brahmans and Chhetris and fewer than 29 noitalupopehtfo% 20 percent of Newars have not been to school, 0 Male Female Male Female compared to 43 percent of the Hill Dalits, 76 1991 2003/04 percent of the Tarai Dalits, 62 percent of the Source: CBS 1991, NLSS 2004 Muslims and 45 percent of the Hill Janajatis. Dalits have the lowest completion rates for FIGURE 13 Net primary and secondary enrollment primary education, followed by Muslims. rates for boys and girls in Nepal 1995- 96 and 2003-04 For SLC level education the completion rates 100 are even lower for Dalits. While Dalits have increased their share in the graduate or 80 Boys higher education level,they still account for less than one percent of the graduate 60 Girls population (Census 2001,CBS). 40 Nepal has made rapid progress to- noitalupopehtfo% wards its MDGs of universal primary educa- 20 tion and gender parity by 2015; however, 0 there are persistent gender, caste and 1995-96 2003-04 1995-96 2003-04 ethnic disparities. The overall literacy rates Primary Lower secondary Secondary for the population age six years and above Source: NLSS I, II, Nepal Poverty Assessment 2005 have improved significantly,from 23 percent in 1981 to 54 percent in 2001. Figure 12 and FIGURE 14 Percentage of school participation of Figure 13 show that the same pattern is seen age 6-10 year olds by gender, caste and ethnicity 2003-04 both for literacy and school enrolment. Lit- eracy rates are improving for both males and females,faster for females so the gender gap is narrowing. It is noteworthy that over the past ten years the net enrolment for girls in the 6-10 age group rose by 44 percent (to 67%) and is quickly approaching the enrol- ment rate for boys (78%). Assuming that the gender gap continues to narrow at this rate, Nepal will achieve gender parity in enrol- ment by 2010. The national data mask stark inter- group differentials and Figure 14 also shows Source: NLSS II 30 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal that gender disparities are particularly high in the Tarai Middle Caste group, where 94 percent of the 6 -10 year old boys are in school compared to only 58 percent of the girls. It is also notable that participation is low for both boys and girls in the Muslim community (the percentage of Muslim girls going to schoolinthe11-15agegroupremainedstagnantat23percentbetween1995/ 6 and 2003/4). Disparities are particularly strong among the Tarai and Hill Janajatis. Many of these groups, most notably the Chepang and Bote (with literacy rates of 14% and 21% respectively), lag seriously behind the national average. FIGURE 15 Women in local elected bodies 1997- 2002 Many Janajati groups are at a disadvantage District Councils 150 because they live in remote areas and be- District Development cause Nepali is not their mother tongue. On Committees 75 Municipalities 808 the other hand, it is encouraging to note a Village Development sharpincreaseintheenrolmentofDalits, and Committees 3,916 Village Councils 3,861 that now the six to ten year olds are only a few points below the Brahman/Chhetri Ward Committees 35,206 group. Among children in poor households, 0 5 10 15 20 however, the percentage of 6-10 year olds Share of women (%) and 11-15 year olds out of school climbs to Source: Election Commission 1991, 1994, 1999 36 and 42 percent respectively (World Bank 2005a),and more than two-thirds of these are girls. These children are most at risk if Nepal does not achieve the education MDGs. They will almost certainly inherit and perpetuate a life at the margins of economic and political life. Political poverty Women's participation in governance Efforts to increase women's participation in elected government after 1990 have largely failed. Women have never gained more than six percent of the seats in the lower house and even in the upper house,where parties can place women if they desire or where women can even be appointed by the King, their proportion has mostly hovered at five percent. They account for only small percentages of the Central Executive Committee membership of the main political parties:9.6 percent of the central committee of the Nepali Con- gress (NC), 7.3 percent of the central committee of the Rastriya Prajatantra Party (RPP) and 7.1 percent of the central committee of the Communist Party of Nepal (Unified Marxist-Leninist) (CPN-UML). Figure 15 shows that in local government,where it is mandated that one of the fiveWard Committee members and 20 percent of the Municipality P O V E R T Y O U T C O M E S 31 FIGURE 16 Ethnic/caste and gender representation in parliament, 1959-1999 70 60 50 40 30 20 10 0 1959 1967 1978 1981 1986 1991 1994 1999 Brahman-Chhetri Dalit Janajati Newar Madhesi Women Source: Lawoti 2002 members be women, representation by women is good. However, in the more powerful VDC and DDC Committees, women's representation after the 1997 local election was only around seven percent and in the District and Village Councils ­ where local policy and budget allocation decisions are made ­ women's representation fell to around three percent. Women make up only about seven percent of the civil service and fewer than one percent of the officers at the First Class level and above. Their representation in the executive and judiciary is even lower. Dalit, Janajati and Madhesi participation in governance During the Panchayat period and the first 10 years of multiparty democracy Brahmans and Chhetris were able to maintain around 60 percent presence in the legislature, and Newars just below 10 percent. (See Figure 16.) Janajati and Madhesi presence is limited, and does not accord with their proportion in the FIGURE 17 Gender/ethnic composition of elected population. Dalits, moreover, were almost members (1999 parliamentary elections) entirely absent from parliament and only Women candid.contested 143 had one representative during the entire Women winners 12 multiparty period. Given their dominance inthelegislature,itisnotsurprisingthatmen Dalit candidates contested 82 from the Brahman/Chhetri group also held Dalit winners = 0 the lion's share of cabinet appointments. Janajati candid. contested 1,167 Janajati winners Their dominance in the civil service also in- 78 creasedfrom70to90percentbetween1985 Total contested 2,238 and 2002. The leadership of the civil service Total winners 205 has been dominated by Hill Brahmans, 0% 20% 40% 60% 80% 100% Chhetris and Newars: all 19 of the Cabinet UML NC RPP Others Secretaries since 1951 have been from this group. Similarly, out of the 21 people Source: HDR 2004, UNDP 32 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal appointed asVice-chairpersons of the National Planning Commission during this period, there has been only one Janajati and all the rest have been either Brahmans, Chhetris or Newars of Hill origin (Yadav 2005). Perhaps more telling is to consider who actually ran for public office. Figure 17 shows data from the election held in 1999. About 75 percent of candidates who contested were not affiliated with any of the three major par- ties. While the three main political parties (dominated by B/C males) may have chosen not to field many women, Janajati or Dalit candidates, these groups themselves expressed their need to exercise their democratic rights and felt empowered to do so. These candidates either ran as "independents" or as members of small, locally-based or special interest parties. Among women candidates this number is about 70 percent, among Janajati candi- dates it is over 80 percent and among Dalits, over 95 percent. Women and caste and ethnic minorities obviously want to be part of the democratic politi- cal process. While their success in the past elections was limited,as democracy takes root the experience gained during these early forays will be valuable. Applicants to civil service positions are also overwhelmingly (83%) Brahman/Chhetri. Candidates from this group are more than twice as likely to be chosen as Newars,nearly three times more likely than a Janajati candidate and over four times more likely than a Dalit candidate. The pattern extends to the judiciary where the B/C and Newar groups hold virtually all positions. Local power relations and poverty In addition to its analysis of the national level poverty data, the GSEA com- missioned a separate in-depth study on Measuring Empowerment and Social Inclusion (MESI) (Bennett and Gajurel et al,2006) to analyse the material,so- cial and political status of the relatively privileged Brahman,Chhetri and Newar (BCN) groups and the Tarai Middle Castes compared to the Janajati groups and the Dalits in rural Nepal. Using both qualitative and quantitative data,the study documents how the various types and lev- FIGURE 18 Wealth ranking based on consumer els of exclusion shown in Figure 5 play out in real items by caste/ethnicity life. (See Box 6.) BCN Tarai Middle Castes Dalits Who are the poor? Janajatis Overall,the patterns on economic indicators that All Groups emerged from the MESI field data ground truth 0 20 40 60 80 100 those reported in the national level data sets re- Percent Low Lower Medium viewed above. The Janajatis and the Tarai Middle Medium High Caste groups tied for second place on a compos- ite wealth ranking score. The Brahman/Chhetri/ Source: Bennett and Gajurel et al. 2006 P O V E R T Y O U T C O M E S 33 BOX 6 Measuring empowerment and inclusion The Measuring Empowerment and Social Inclusion (MESI) study attempted to understand the experiences of people living within the existing social institutions of caste, ethnicity and gender, and the effects of these institutions on their self- perceptions, day-to-day social interactions and their ability to exercise agency. It used both quantitative and qualitative techniques to try to understand, measure and analyse empowerment and social inclusion ­ what blocks it and what enhances it. In other words, it was an effort to explore what the rather abstract concepts of empowerment and social inclusion meant to people in real life. The MESI sample comprised of one man and one woman from 1,000 households in 60 villages. Twenty were villages where the Rural Water Supply and Sanitation (RWSS) interventions had been completed, 20 were where the RWSS was just beginning and 20 were control villages. These 60 villages were neither the most prosperous nor the poorest and therefore represented communities that were generally representative of the majority of the rural population. Research included a six-week qualitative study in four of the sample villages. The indicators used to measure levels of empowerment (EMI) and social inclusion (SII) are shown below. The EMI and SII indices were combined to form the Composite Empowerment and Inclusion Index (CEI). Indicators used in the Empowerment Index (EMI) Indicators used in the Social Inclusion Index (SII) 1. Knowledge/awareness of rights and 1. Self-perceived status of own caste/ethnic group procedures Relative economic status/success of own group Understanding of police procedures Relative contentment/comfort with social status of own group Understanding of court procedures Respectful treatment Knowledge of the Country Code and rights of Dalits Relative access to opportunity Knowledge of local services Cooperation from other groups Respect in the community 2. Participation in local development services Seeking local services 2. Restricted access & public intimidation Participation in programmes of child's school Whether the respondent is restricted from entry into certain public areas (like temples or peoples' homes) and/ 3. Confidence/comfort level in or prevented from using public facilities (like water taps). accessing services/exercising rights Whether the respondent faces verbal or physical Approaching the police intimidation/humiliation/violence in public spaces such as Approaching the courts the village and/or in the nearest bazaar. Approaching children's school 3. Effectiveness of local political influence 4. Social networks (economic and political) Result of complaints/suggestions they have made at Connections for getting a job for oneself ward/village/DDC level. Ability to help others get a job Connections at ward level 4. Effectiveness in getting services and opportunities Connections to local service agencies Invited by agencies to participate Connections at VDC level Promptness of service Connections at DDC level Consulted for opinion Access to training opportunities 5. Efforts to Influence Suggestions/complaints at ward level Suggestions/complaints at VDC level Suggestions/complaints at DDC level Advice to school officials 34 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal Newar households are well on the top (by more than 1.3 times) and the Dalits are at the bottom. The Janajatis are intermediate between the two groups and are closer to the BCN group in some measures of livelihood empower- ment (such as land size), but closer to Dalits with respect to other measures (such as household consumption goods). However, for policy making purposes it is also important to note a pattern that appeared in several of our wealth ranking measures. For example, Figure 18 shows the distribution between four different levels of consump- tion goods ownership for each group. The BCN group has the largest concen- tration in the top category while Dalits have the lowest. But the same figure shows that there is not a great deal of difference between the caste and eth- nic groups when it comes to the proportion of each group who are in the "lower medium" or second poorest category. For all groups this "lower me- dium or middle poor"group is the largest cohort,containing over 45 percent The lack of wage labour of the whole population. The Janajatis had the highest proportion of their and other kinds of employment is a major members in this group (about 48%), but the BCN group also had about 42 concern for many Nepalis. P O V E R T Y O U T C O M E S 35 BOX 7 Listening to rural people during the MESI study On restrictions for Dalits to enter certain spaces: When asked if she had entered the homes of high caste people in her village, a Dalit woman in Tanahu district laughed, "Not once! I've watched functions of high-caste families from outside their windows." On self-perception of own caste: A Dalit man commented, "Sometimes I hate my caste, because people hate us because of our caste. My caste is a barrier to progress. Once in Class Nine, my teacher asked, `What do you have in your tiffin [lunchbox]. Is it the flesh of a dead animal?' I complained to my headmaster, but he too, harassed me." On the aspirations for the future: "To be listened to" commented a Dalit woman in the Dibya Nagar scheme, in Nawal Parasi, "we need more education or more money and power. If I ever had a chance to lead, I could not, as I am not educated and have never performed such duties. I think people would never listen to me due to my caste." On changing attitudes towards Dalits: In Jamune one old Brahman woman stated that there was no discrimination between men and women in her village and no touchability or untouchability. "I am old, so I do not want to eat food touched by Dalits. But I have never restricted my son or his children from taking food from Dalits or going into their houses," she noted. percent. These findings are similar to the national level data sets and caution once again against any easy equation between caste/ethnic identity and eco- nomic status. It is important to keep in mind the reality that there are many poor Brahmans and Chhetris in rural Nepal. Our qualitative research confirmed the quantitative findings about the economic dominance of the BCN group. However, it also revealed some different perspectives on poverty and alerted us to the economic issues that people are most concerned about. None of the four communities where the qualitative research was conducted seem to experience severe seasonal short- ages of food, although one community reported that food shortages were a regular occurrence before an irrigation project came to the area. With the pos- sible exception of a few of the very poorest households, the concern of most people is not subsistence,but rather better opportunities to earn cash income. Despite the importance of land ownership for security and status (Box 7),many of our informants felt that agricultural work (on their own land) is inferior to wage work (agricultural or otherwise) because it is physically demanding and brings minimal rewards. This was a surprise since agricultural wage work carries little prestige and has usually been considered a last resort. Both men and women spoke about the lack of wage labour and other kinds of employ- ment in their areas as a major concern for themselves and for their children. For many Dalits and Janajatis,particularly lower status Janajatis,such as the Kumal - who formed a large proportion of one of our case study vil- 36 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal FIGURE 19 Composite empowerment and inclusion index lages - lack of income earning oppor- (CEI) by gender/caste/ethnicity tunities was a serious issue. More and BCN more poor people in Nepal perceive Tarai Middle the lack of opportunities to earn cash Castes incomeasthegreatestconstraint. This Dalits Male Female was cited as a reason for removing Janajatis children from school and for not be- ing able to send a family member All Groups overseas for employment. Lack of in- 0.0 0.1 0.2 0.3 0.4 0.5 come prevented some from seeking Index health care and from benefiting from Source: Bennett and Gajurel et al. 2006 developmentinterventions, including the opportunity to attend training. Lack of income also prevented some from taking out loans because,in spite of the ubiquity of savings and credit groups, membership generally depends on being able to meet the group's agreed level of weekly cash contributions to the saving pool. "Time"as an asset is seen as closely related with material well-being. "Time poverty" is a common problem, particularly for women, but also for men in asset-poor households. Women often complained that they could not participate in training opportunities or attend various meetings because of a shortage of time (usually a result of their many household responsibili- ties). Household wealth was found to be an important determinant for group membership (for both men and women). Respondents from wealthy house- holds were over seven times more likely to belong to a group than those from poor households. In poor households,the struggle to bring in enough food and toearnenoughincomenotonlymeantthathouseholdmemberswereunableto become involved in community activities,but in many cases also that children were needed to assist in the daily chores and could not attend school. Empowerment and inclusion results The criteria used by the MESI study for the Empowerment Index (EMI) to measure empowerment, and for the Social Inclusion Index (SII) to measure social inclusion,are shown in Box 6. The results of the Composite Empowerment and Social Inclusion Index (CEI), which combines the EMI and the SII,are shown in Figure 19. According to the findings of the MESI study: n The ranking of social groups in the CEI broadly reflects the traditional caste hierarchy: The CEI levels of the Brahman/Chhetri/Newar group (46) were significantly higher than those for Janajatis (36),who in turn scored higher than Dalits (25). n For separate indicators such as knowledge of rights and procedures, confidence/comfortlevelinaccessingservicesandexercisingrights, social P O V E R T Y O U T C O M E S 37 networks,localpoliticalinfluenceandeffortstoinfluence,BCNscoreswere consistently around twice as high as those of Dalits. n For the indicator on restrictions and intimidation in public space,the study found that 90 percent of BCN group never experienced any restriction or intimidation. Those from this group who had encountered spatial restrictions faced them only on a temporary basis during ritual pollution due to death in the family or (for women) menstruation or childbirth. For Dalits,ritual pollution and spatial restriction is not a temporary state,but a permanent part of their social identity. All Dalit respondents reported having experienced some degree of restriction on entering certain public spaces and public intimidation/harassment, and about 20 percent experienced high levels of restriction and intimidation/harassment. n In all social groups,men consistently have higher CEI scores than women. Looking at just a few of the specific indicators we find: l Participation in local development services: men participate in/take advantage of local development services 1.6 times more often than women. Inclusion alone can break the transfer of l Efforts to influence: Men try to influence the institutions that are deprivation from supposed to deliver services to them 2.7 times more than women do. generation to generation. 38 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal l Effective local political influence: Men are 4.8 times more able to actually influence their institutional environment than are women. n Whenonlyfemalescoresareexamined,thecastehierarchyre-appearswith Brahman/Chhetri and Newar women scoring much higher than Dalit or Janajati women. This is a surprising reversal from only 25 years ago when the opposite was found to be true (Acharya and Bennett 1981). Caste/ ethnic identity plays a role,but education and membership in local groups can tip the balance and can affect the degree to which rural women have beenempoweredandincludedincommunity-leveldevelopmentactivities. l Many poor Dalit and disadvantaged Janajati women have little time to spare for group activities that benefit other women. l Even when they are able to join various types of groups, their voices are often muted by the more confident and highly educated BCN women unless special efforts are made to ensure that they participate in the governance of the group. n The relationship between caste and gender is complex. Women in all caste and ethnic groups score lower on both empowerment and inclusion thanmenintheirowngroups. However, Brahmanwomenaresignificantly more empowered than Dalit men ­ and in terms of inclusion, women from all other groups score higher than Dalit men. Dalit women who experience both gender and caste discrimination have the lowest levels of empowerment and inclusion scores of any group. The mean CEI score for BCN men is more than double the mean CEI index for Dalit women. Thestudyalsotriedtodeterminethemajorfactorsinfluencinglevelsof empowerment and social inclusion and the regression results showed that: n Caste/ethnic identity and gender together explained 33 percent of the variation in the CEI index. n Caste/ethnic identity is a more powerful predictor of empowerment/ inclusionthangender; itexplained26percentofthevariationinCEIscores while gender explained only 7 percent. n Being a member of a local development group was associated with a five percent increase in CEI levels. n Tenyearsofeducationwasassociatedwitha19percentincreaseinCEIlevels. n Contact with the local Women Development Office, holding office in a group and exposure to media were also significantly positively associated with higher CEI scores. C H A P T E R 3 LEGAL EXCLUSION III LEGAL EXCLUSION L E G A L E X C L U S I O N 41 The current international understanding of the human right to equal treat- ment does not allow discrimination based on gender,caste or ethnicity. Un- fortunately, cultural practices and even some laws in Nepal still discriminate on the basis of sex,caste,ethnicity and religion. Other laws aimed at protect- ing people from discrimination have been weakly implemented. It is clear that even the best-designed legal provisions cannot on their own end the exclusion some citizens face because of their gender,caste or ethnicity. Deep- rooted values and discriminatory attitudes often lead to poor implementa- tion of laws. The challenges, therefore, are not only amending laws but also changing the mindset of people,and formal and informal institutional mecha- nisms that perpetuate discrimination. The Muluki Ain, the Country Code (1854), first formalised the caste system into law. Most of the penal provisions in the Country Code reflected the caste hierarchy, i.e. the lower the caste,the higher the degree of punish- The law discriminates against women in the ment for the same offence. The law also reproduced the patriarchal view of areas of citizenship, inheritance rights, womenasproperlysubordinatetomenandeconomicallydependentonthem. education, employment, A new provision prohibiting discrimination on the basis of caste and ethnicity health, sexual offences, marriage and family was inserted in the 1963 Country Code but it came with ambiguity in the relations, court proceedings and identity. form of a provision protecting "traditional practices". Moreover, Nepal has not yet signed the International Labour Organization (ILO) Convention No.169,the international instrument that deals specifically with the rights of indigenous and tribal people. The lack of laws is not the issue in Nepal. The Con- stitution and the Civil Rights Act of 1955 prohibit discrimi- nation against any citizen on the grounds of "religion, race, sex, caste, tribe, ideological conviction or any of these". The laws also prohibit untouchablility, denial of access to any public place or depriving someone of the use of pub- lic utilities. But enforcement is lax. For example, even as recentlyas2005, severalDalitsinSaptariDistrictweremade to pay a heavy fine for entering certain temples ­ purport- edly to cover the cost of "purifying" the areas they had "defiled". Those who could not pay were forced to leave the district. Although this incident was reported by the national press, those who levied the fine were not pros- ecuted. In addition, and particularly as noted earlier with regard to obtaining citizenship papers (see Box 2), many ad hoc discriminatory administrative practices still prevail against certain communities ­ such as Madhesis, people who live in the flat Tarai belt adjacent to India and share 42 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal manylinguisticandculturalcharacter- BOX 8 Non-discrimination and equality istics with them. Many laws enforc- ing outright discrimination also re- Existing legislation in Nepal has failed to distinguish between non-discrimination and equality. Non- main to be amended. discrimination prohibits discrimination, but may not Thelawdenieswomenequal be able to bring equality as a result. Gender neutral inheritance rights. The 11th Amend- law promotes a "sameness" approach but does not ment to the Country Code recognises take socially constructed gender gaps into daughters as joint-heirs (ansiyars) to consideration. In addition, the present legal family property and partly secures framework has not been able to prohibit discrimination on the basis of origin or language. women's right to ancestral property. However, daughters must return their share of the family property to the family when they marry. In the case of intestate property, daughters fall below sons in the line of succession and married daughters fall even farther below. Sexual exploitation is also not addressed properly. While the Consti- tution prohibits"traffic in human beings, slavery, and serfdom or forced labour", the law on trafficking is neither comprehensive nor sensitive to human rights issues. Moreover,it exempts buyers from legal jurisdiction as the "purchase" of a human being is not considered an offence. In addition,the present laws are inadequate to deal with sexual offences. Several problems are found with the laws on rape, and Nepal has no law to deal with sexual harassment. Discriminatoryprovisionsbasedoncaste,ethnicityandreligioninclude the Constitutional declaration of Nepal as a Hindu kingdom, and of Nepali as the only official language. Box 10 below sets out some of the other laws that support continued discrimination on the basis of religion,caste and ethnicity. Overall,the laws discriminate against women in the areas of citizen- ship,property,education,employment,health,sexual offences,marriage and family relations,court proceedings and identity. Nepali women are unable to confer citizenship on their children or husbands; and their identity as a mother and/or wife is also not legally recognised for official purposes,since BOX 9 Discriminatory laws official forms require the name of the While the CEDAW Committee recommended that Nepal change 85 laws and 137 legal provisions that father. Box 11 lists some of the dis- discriminated against women, an analysis of Nepali criminatorylegislation. Progressisbe- laws, including the Constitution and Country Code, ing made, however. Some of this leg- conducted for the GSEA found 32 provisions that islation is under challenge, and, as discriminate on the basis of religion, caste and ethnicity, noted in the box, recent Supreme and 176 provisions in 83 pieces of legislation that Court rulings have declared certain discriminate against women (Malla and Shrestha 2005). provisions unconstitutional. L E G A L E X C L U S I O N 43 BOX 10 Examples of discriminatory laws based on religion, caste and ethnicity Provision Discrimination against Nepal is a multiethnic, multilingual, democratic, independent, indivisible, Non-Hindus sovereign, Hindu and Constitutional Monarchical Kingdom. By declaring Nepal a Hindu Kingdom, the Constitution -Article 4 (1) of the Constitution of Kingdom of Nepal (1990) itself privileges Hinduism over religions, which account for at least 20% of the population. ...punishments shall not be more or less merely based on people's All but "High Caste" males higher or lower status..... The Country Code recognises people's higher or lower status, - Preamble of the Country Code (1963) even though the intention is for equality in punishments. The Nepali language in the Devnagari script is the language of the Janajatis & linguistic minorities nation of Nepal. The Nepali language shall be the official language. Ethnic languages are not recognised as "official -Article 6(1) of the Constitution of Kingdom of Nepal (1990) languages of the nation". Each community shall have the right to operate schools up to the primary Janajatis & linguistic minorities level in its own mother tongue for imparting education to its children. The Constitution itself permits education in the mother - Article 18(2) of the Constitution of Kingdom of Nepal (1990) and tongue only up to the primary level. Moreover, it does Section 7(1) of Education Act (1972) not commit the state to fund such schools. (No person) shall deliberately slaughter cows or bullocks, or instigate Janajatis, Dalits Muslims and other religious minorities others to slaughter, or attempt to slaughter, or take cows and bullocks to who do not perceive cow slaughter to be a crime foreign countries with the intention of slaughtering them, or take them to Cow slaughter is an offence punishable by up to 12 foreign countries and kill them. years of imprisonment. - No 1 of Chapter on Quadrupeds of the Country Code (1963) In case any person is found to be raising a weapon or doing anything else Janajatis, Dalits, Muslims and other religious with an intention to slaughter cows or bullocks, he/she shall be prohibited minorities who do not perceive cow slaughter to be from slaughtering them. In case he/she refuses and uses weapons a crime against the person prohibiting him/her, the latter shall not be deemed to Killing a human being is justified for the protection of have committed an offence if he/she attacks the slaughterer and if he/she cows and bullocks. acts or strikes at (the would be slaughterer) and kills him/her... - No 4 of Chapter on Quadrupeds of the Country Code (1963) Whoever takes cows, bulls, bullocks or calves of any category to a foreign Janajatis, Dalits, Muslims & other religious territory from Nepal and slaughters them, or causes them to be slaughtered, minorities who do not perceive cow slaughter to be he/she shall be liable to a punishment of imprisonment for 6 years. a crime - No 10 of Chapter on Quadrupeds of the Country Code (1963) Extra-territorial jurisdiction is created for cow slaughter. Kipat (common communal property) lands which lack official documents, Janajatis are equivalent to Raikar, lands on which taxes can be levied. This has led to the loss of ethnic based communal - No 1 of Chapter on Land Evictions of the Country Code (1963) (Kipat) ownership as Raiker can be used, transferred, and disposed of by anyone. Kipat land can be registered as tax levied land (Raikar) by means of Janajatis deed. Same as above as the restriction on selling such land is - Section 3(1) of the Land Reformation Act, (1964) withdrawn. Anyone who desires to enter an ancient historic, artistic or important religious Dalits place, whether owned by the government or by a private person, may do so if this Dalits are excluded on the basis of this provision, as it does not disrupt a traditional custom that has been practiced for a long period. states one may enter religious places only if long- - Section 10 of the Ancient Monument Protection Act, (1956) standing custom is not disturbed. No one shall disrupt social customs fraudulently or coercively or commit or Dalits cause any such act to be committed. Whoever commits, or causes Dalits are excluded from temples and other public places another to commit such an act is liable to a punishment of imprisonment on the basis of this provision as it protects social up to one year or a fine up to one thousand rupees. customs that are being practiced. - No 10 of Chapter on Miscellaneous Provisions of the Country Code (1963) 44 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal BOX 11 Examples of discriminatory laws based on sex Name of the legislation Discrimination Article 9 (1) and 9(2) of the Constitution of the Kingdom of Women cannot provide citizenship to their children. Nepal, 1990 Article 9 (5) of the Constitution of the Kingdom of Nepal, 1990 Women cannot provide citizenship to their spouse. Annex/Schedule (2)&(4) of the Nepal Citizenship Rules, 1992 Citizenship certificate only states father's or husband's name. Application forms of the Passport Rules (1970) Women require their guardian's or husband's approval to THIS PROVISION WAS RECENTLY STRUCK DOWN obtain a passport. No. 1A of the Country Code, 1963 - Partition Married daughters are not considered as coparceners in the ancesteral estates. No. 16 of the Country Code, 1963 - Partition Unmarried daughters must return their share of parental property upon marriage. No.19 (1) of the Country Code, 1963 - Partition Consent of married daughters is not required to dispose of more than half of the immovable family property. No. 2 of the Country Code, 1963 - Women's Exclusive Women require consent to dispose of more than half of the Property immovable family property. THIS PROVISION WAS RECENTLY STRUCK DOWN No. 7 of the Country Code, 1963 - Women's Exclusive Women are restricted from freely using their own share of Property inherited property. No. 1 of the Country Code, 1963 - Intestate Property Definition of Hakwala (owner) includes seven generations on the male side. No. 2, 3, 6, 7, 9, 10, 12 of the Country Code, 1963 - Married daughters fall behind in the line of succession of Intestate Property intestate property. Section 26(1) of the Land Act, 1964 Tenancy right is transferable only to an unmarried daughter after she attains the age of 35 years. Section 38 of the Insurance Act, 1992 Married daughters are excluded from the line of succession. Section 15 (a) (1) of the Employees Provident Fund Act, Married daughters fall behind in the line of succession to 1962 receive/claim provident fund. Section 23(1) of the Pension Act, 1986 Married daughters fall behind in the line of succession to receive/claim bank deposit. No. 10 of the Country Code, 1963 - Rape The rape of a married woman technically creates an end of the family relation, because the husband is considered as an ex-husband by law. Section 10 of the Army Act, 1959 Women in the Army can serve only in non-combatant roles. Section 12 of the Foreign Employment Act 1985 Permission of the guardian and the government is a prerequisite for women to go abroad for employment. No.28, 28(a), 32 of the Country Code, 1963 - Homicide Higher punishment is provided for a woman undertaking an abortion than for third parties whose actions cause an abortion. No. 3 (1), 3(2), 3(3), 3(4) of the Country Code, 1963 - Paupers Deprives mothers of care and guardianship of their children. No. 1(1) of the Country Code, 1963 - Husband and Wife Grounds for divorce are not the same for men and women. 2 & 2(a)Country Code, 1963 - Adoption Women face additional restrictions in adopting a child. No. 9 of the Country Code, 1963 - Marriage The law appears to encourage bigamy as a second marriage is permitted under certain conditions without divorcing the first wife. No. 3(1) of the Children's Act 1991 Father has priority over mother in naming a child. Section 31 of the Revenue Tribunal Rules, 1974 Court dress is specified for male judges only. P O V E R T Y O U T C O M E S 45 C H A P T E R 4 PUBLIC DISCOURSE AND ACTIONS IV PUBLIC DISCOURSE AND ACTIONS Government policy and institutional framework Responses to gender discrimination Responses to caste discrimination Responses to ethnic discrimination P U B L I C D I S C O U R S E A N D A C T I O N S 47 This section examines how the "rules of the game" have influenced the ex- cludedgroupsinNepalintermsofgovernmentpoliciesandinstitutionalstruc- tures for women, Dalits and Janajatis. Until April 1990, Nepal's movements for women, Dalit and Janajati rights remained subsumed within the larger struggle for democracy. The People's Movement that resulted in an end to absolute monarchy and direct rule under the Panchayat system marked the beginning of a new era in Nepal ­ that of citizens'rights. Various social move- ments have emerged after April 1990 and collectively these movements have critiqued the three cornerstones of pre-1990 national identity ­ the monar- chy,Nepali language and Hinduism. They have also continuously challenged the traditional exclusionary definition of what constitutes a"real Nepali". The women's movement has succeeded in placing questions of gender equality and justice on the national agenda,and the Dalit movement has begun to challenge Nepal's caste society. The Janajati movement, once described by many Brahmans and Chhetris as a"divisive"phenomenon, has now succeeded in bringing fundamental issues of fair ethnic representation to the fore. It has also challenged Nepal's identity as a Hindu kingdom and placed on the national agenda issues related to rights of language (other than Nepali), culture (other than that of Hill Parbatiyas), and religion (other than Hinduism). Post 1990 guarantees of political and civil liberties have significantly altered the consciousness of Nepalis who ­ even in rural areas ­ increasingly see themselves as sovereign rights holders. Nevertheless, many formal and informal institutions and policies remain stubbornly exclusionary in terms of gender, caste and ethnicity, and most demands of those excluded remain unmet. Therefore, the shift of Nepalis from being subjects of an absolute monarch to becoming sovereign citizens (regarding their political destinies) remains incomplete. Exclusion and hierarchy within excluded groups is also coming under scrutiny. For example,Dalit activists are questioning the dominance of their own Biswakarma and Pariyar castes,and the Janajatis are examining the dominance of wealthier groups within their movements. Both groups also face issues of male dominance. Within the women's rights movement,domi- nation by the historically privileged Brahman/Chhetri/Newar groups is also being questioned. Another divisive element in each of these movements has been the alignment of different civil society groups with political parties. Conflicting party loyalties have sometimes blocked cooperation between dif- ferent Dalit organizations or caused women's organizations to refuse to work together on a common agenda. 48 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal These three major social movements remain independent of each other,despite their many common demands. Because little dialogue has taken placeamongthesemovements,thedemandsofsomegroupscontradictthose of others ­ giving the state space to delay fulfilment. The delay in the state's response has also resulted in the growth of radical or revolutionary offshoots thatsometimesovershadowthemainstreamreformistideas. Theradicalstrains are most notable amongst the CPN (M), even though smaller leftist organiza- tions and even some mainstream parties espouse similar ideals. Government policy and institutional framework Nepal has completed almost half a century of planned development, which Change will require a was instituted through a series of five-year plans. The Eighth Plan (1992-1997), complex mixture of political and senior management the first formulated by a democratic government, introduced poverty alle- commitments, the viation as one of three objectives. It was also the first public document to communication of that commitment, and pressure address the caste/ethnic issue, albeit indirectly and incompletely. The plan for results, as well as negotiation and dialogue. disaggregated the "poor" and attempted to target particular ethnic and caste P U B L I C D I S C O U R S E A N D A C T I O N S 49 groups under the somewhat vague and unspecified category of "disadvan- taged groups". However, its approach to these groups was uneven; for example,while it introduced primary school scholarships for Dalits,it did not have similar schemes for other excluded groups. The Ninth Plan (1997-2002) addressed Dalits and Janajatis by name ­ for the first time ­ and had a separate section dealing with Adivasi Janajatis indevelopment. ItalsobeganallottingaportionofVillageDevelopmentCom- mittee (VDC) and District Development Committee (DDC) grants for income- raising and skill development programmes for Janajatis and envisaged found- ing an autonomous Janajati Council at the district level. While DDC and VDC level commitments were not fully implemented, in 1997 the government did form a National Committee for the Development of Nationalities (NCDN) and the National Foundation for the Development of Indigenous Nationalities (NFDIN) Act was enacted in 2002. Planned efforts to improve the situation of women began as early as the Sixth Plan (1981-1985) but the approach was welfare driven. The Eighth Plan raised the issue of increasing women's representation in decision-making and also acknowledged the existence of "gender"-based "discrimination"­ but did not define either term. The Ninth Plan again used the term gender but did not progress much beyond the women-in-development mindset in its formu- lation of the problem and, like earlier plans, lacked operational guidelines. The Tenth Plan (2002-2007) is also Nepal's Poverty Reduction Strat- egy Paper (PRSP), and is the most serious and comprehensive government statement about inclusion to date. It identifies social exclusion as one of the three main aspects of poverty and acknowledges that exclusion is the main reasonforthedeprivationsufferedbywomen, certaincasteandethnicgroups, and people living in remote areas. It recognises that lack of voice, political representation and empowerment are as important dimensions of poverty as are economic and human development dimensions. It also understands that exclusion is one of the factors behind the current conflict and supports this thesis by presenting a detailed caste, ethnicity and gender-disaggregated analysis showing Dalits at the bottom of almost all Human Development In- dicators. The PRSP also proposes the need for "affirmative action" to level the playing field. The PRSP, however, fails to present a realistic strategy and concrete mechanisms to mainstream inclusion. TheTargeted Programmes it proposes are narrowly based and even though the government ranks programmes by priority for funding under the Medium Term Expenditure Framework, inclu- sion is not one of the criteria for prioritisation. The document does mention caste straightforwardly when it discusses st rategies for Dalit inclusion, but 50 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal elsewhere it reverts to the term"deprived communities", sending mixed sig- nals about the commitment. The PRSP also lacks quantitative targets in im- portant sectors such as education,health and Targeted Programmes. Another major weakness of the PRSP is its ambiguity in assigning roles and accountability for implementing the proposed actions. Neverthe- less, a positive first step in this direction on the part of the government has been to develop a Poverty Monitoring and Analysis System (PMAS) to sup- port the implementation of the PRSP. PMAS requires monitoring at three levels: implementation, performance and outcome and has a number of out- come/intermediate indicators. More importantly, the PMAS requires the use of data disaggregated by caste and social groups for monitoring. Overall,the identification of social exclusion as a development prob- lem, and the accompanying commitment to social inclusion are significant steps. Nevertheless, the operational modality fails to address gender, caste and ethnicity directly as rights and tends to be driven by the welfare approach, which is not inclusive and is at best, partial. The GSEA study goes on to review public discourse and actions in relation to the three main social movements in some detail and to assess their achievements and shortcomings. Responses to gender discrimination Before 1990,women's issues were cast in the framework of development and welfare ­ not rights. With some exceptions, this generally remains the norm. This approach is not empowering to women since it characterises them as uniformly"backward, illiterate and tradition-bound". The assumption under- lyingthisviewisthatauniform "Hindupatriarchy" constrainsallNepaliwomen in the same way and that a single policy towards women is therefore appro- priate, regardless of their class, caste, FIGURE 20 Caste/ethnic differences in levels of female ethnicity, religion and age. In other empowerment and inclusion words, the understanding of gender BCN hasignoredtheimportantspecificities TaraiCastes Middle of class, caste, ethnicity, age and Dalits other cross-cutting divides. The MESI Janajatis All Study,however,showssharpdifferences intheempowermentandinclusionlev- 0.0 0.1 0.2 0.3 0.4 els of women from different caste and Index ethnic groups. (See Figure 20.) Source: Bennett and Gajurel 2006 P U B L I C D I S C O U R S E A N D A C T I O N S 51 In the eyes of the Nepali state,women have a much longer history as a recognised socially excluded group than do either Janajatis or Dalits. The earliest women's organizations were founded in the late 1940s. Some pre- Panchayat protests by women included the demand for the right to vote. Under the 1976 Class Organizations Act, the Panchayat rulers recognised women as a social group. More recently, women's organizations have been petitioning for conditions of economic equality by demanding equal prop- erty rights,quotas in education and jobs and greater voice in political parties and government. The development response to women's claim for equal rights,however,still hinges largely on the welfare model. Nepal's Constitution does not permit discrimination on the basis of sex and advocates special legal provisions to protect and advance the inter- ests of women. The Local Self Governance Act (LSGA),1999 introduced man- datory representation of women in local elected government. However, at progressively higher levels of decision-making,where there has been no such The development response to intervention, the data show that women are vastly outnumbered. When women's claim for elected local bodies were suspended in July 2002, the representation require- equal rights still hinges largely on ments of the LSGA were also suspended. the welfare model. 52 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal In 1991 Nepal ratified the Convention on Elimination of all forms of Discrimination Against Women (CEDAW). Although the convention requires Nepal to change about 85 laws and 137 legal provisions that are discrimina- tory, this still remains to be done. Nor has the government signed the Op- tional Protocol to CEDAW,which would give Nepali women the right to chal- lenge these discriminatory laws internationally. The Ministry of Women and Social Welfare,established in 1995,was re- named the Ministry of Women, Children and Social Welfare (MWCSW) in 2000. The MWCSW lacks adequate financial and human resources to carry out its nu- merous responsibilities effectively,and has also largely failed to consider the pri- oritiesandneedsofwomenfromtraditionallyexcludedcastesandethnicgroups. The Ministry of Local Development (MLD) was the first to incorpo- Only seven percent of rate gender issues in development programmes when it set up the Women's women own livestock, Development Section (WDS) in the early 1980s. Its major achievements to even though for many groups livestock rearing date include the institution of the Production Credit for RuralWomen (PCRW) is tradtionally a female task and in spite of the programme and other follow on group-based microfinance programmes for fact that many credit rural women,the promotion of reservations for women and the requirement institutions and microfinance that User Groups must have at least 30 percent women members. Women programmes have targeted women and Development Officers (WDOs) ­ who were first hired under the PCRW made loans to them for programme in 1983 and who have always had a field-based focus on rural this activity. P U B L I C D I S C O U R S E A N D A C T I O N S 53 women ­ have now been shifted to MWCSW, which has no real presence in rural areas. This has to some degree sidelined the WDOs, who continue to head Women Development Offices in the districts but are no longer seen by MLD as part of its district development machinery. The WDOs are now being trained to serve as district Gender Focal Points mandated to integrate gender and children's rights in the decentralised planning and review processes. The focal points will also be responsible for collecting disaggregated data and conductinggenderauditsofsectoralanddistrict-levelprogrammes. Although most of the WDOs have served more than 20 years, they have not yet been made permanent government employees. Nepal set up the National Women's Commission (NWC) in 2002. However,the commission lacked a legal basis and its mandate remained un- clear. Its members retired in March 2004 and replacements had not been appointed by January 2006. Brahman and Chhetri women ­ appointed mainly on the basis of their political affiliation ­ dominated the NWC membership. Generally,inclusion of women in development continues to fall into very specific gendered roles that often reinforce unequal access to resources and institutions, particularly for poor and socially excluded women. Despite a series of gender audits carried out by the Ministries of Agriculture, Educa- tionandHealthoverrecentyearswithsupportfromtheUNDP'sMainstreaming Gender Equity Programme,the kind of structural change implied by the term "gender mainstreaming"has not occurred. Tension also exists between tech- nocratic "fixes" ­ often preferred by donors ­ and those advocating more long-term socio-political change. The latter is more likely to occur as a pro- cess of democratic trial and error, often led by ordinary people (including politicians),and tends to be"messier"and less amenable to donor timetables and budget cycles. There is also an apparent donor bias favouring "profes- sional" NGOs ­ those whose leaders are familiar with the current develop- ment trends and able to converse in English ­ over smaller local NGOs or mass organizations such as the women's wings of political parties, even though the latter could potentially be more effective at mobilising women and influ- encing policy change. As a result, programming has tended to remain nar- rowly focused,without the necessary policy foundations and without the ap- propriate linkages needed to achieve gender mainstreaming. The representation of women in political parties continues to be low, especially at the higher echelons of power, and this non-representation re- mains a major obstacle to the mainstreaming of policies and programmes that focus on women and other excluded groups. Women's organizations within the parties remain in a subordinate relationship within these typically male-dominated institutions. 54 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal Responses to caste discrimination Dalits ­ as people at the very bottom of Nepal's caste hierarchy ­ continue to be a source of discomfort to educated bureaucrats, who in their formal roles regard the caste system as outdated and inconsistent with the goal of devel- oping Nepal into a "modern" state, but who informally are still very much products of their caste-conscious upbringings. Even now, the government and many development/aid organizations use euphemisms such as"occupa- tional castes", "oppressed castes", "backward classes", "depressed castes", "deprived castes", "marginalised", and "disadvantaged groups", in- stead of speaking of them as Dalits. The hesitation to use the term Dalit de- flects attention from the everyday reality of caste-based discrimination that must be tackled head-on. Over 200 forms of commonly practiced types of caste-based discrimi- nation have been identified in Nepal (Bhattachan et al 2004). These include limiting the so-called lower castes to socially-sanctioned roles, forcing them to carry out demeaning caste-based tasks such as removing the carcasses of dead cattle, refusing to share water sources with them and behaviours in- tended to avoid any direct bodily contact ­ the literal practices of untouch- Ad hoc discriminatory administrative practices still prevail against certain communities. Many laws enforcing outright discrimination also remain to be amended. P U B L I C D I S C O U R S E A N D A C T I O N S 55 population Dalit the of Distribution 2 Gurung a Hark MAP ce: Sour 56 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal BOX 12 Different definitions of Dalit sub-castes Dalit castes defined by Dalit castes classified by Dalit castes defined by the Dalit caste classified Ministry of Local Dahal et al, 2002 National Dalit by Jha, 2004 Development, 1997 Commission, 2003 1. Lohar Hill 1. Biswakarma (Kami, Hill 2. Sunar 1. Kami Sunar, Lohar, Tatma, 1. Gandharva (Gaine) 3. Kami 2. Damai Chunara, Aod) 2. Pariyar (Damai, Darji, Suchikar, 4. Damai 3. Sarki 2. Darji (Damai, Pariyar, Nagarchi, Dholi, Huduke) 5. Sarki 4. Badi Hudke, Dholi) 3. Badi 6. Badi 5. Gaine 3. Sarki (Mijar, 4. Biswakarma (Kami, Lohar, Sunar, Oda, 7. Gaine Charmakar) Chunara, Parki, Tatma) 8. Kasai Tarai 4. Badi 5. Sarki (Mijar, Charmakar, Bhool)Tarai 9. Kusule 6. Tatma 5. Gaine (Gandharva) 10. Kuche 7. Khatwe 6. Kapali* Tarai 11. Chyame 8. Chamar 7. Khadki* 6. Kalar 12. Pode 9. Dusadh 8. Deula* 7. Kakihaya 13. Chamar 10. Musahar 9. Kuche* 8. Kori 14. Dhobi 11. Batar 10. Tatma 9. Khatik 15. Paswan (Dusadh) 12. Dhobi 11. Dusadh (Paswan, 10. Khatwe (Mandal, Khang) 16. Tatma 13. Dom Hajara) 11. Chamar (Ram, Mochi, Harijan, Rabidas) 17. Dom 14. Halkor 12. Dhobi* (Rajak) 12. Chidimar 18. Batar 13. Dom 13. Dom (Marik) 19. Khatwe Newar 14. Batar 14. Tatma (Tanti, Das) 20. Mushar 15. Kusule 15. Khatwe 15. Dusadh (Paswan, Hajra) 21. Santhal 16. Khadki (Kasai) 16. Musahar 16. Dhobi (Rajak) 22. Satar 17. Pode 17. Halkhor 17. Pattharkatta 23. Halkhor 18. Chyame/ 18. Patharkatta 18. Pasi Chyamkhala 19. Chamar 19. Bantar 19. Halahulu 20. Musahar * Castes that have asked not 21. Mestar (Halkhor) to be defined as Dalit. 22. Sarbhang (Sarbariya) Newar 23. Chyame 24. Kasai 25. Kuche 26. Kusule 27. Pode ability ­ which in most cases still go unpunished. Generally, discrimination is more entrenched in the less-developed regions of the country, especially in the Mid- and Far-western regions. However, caste rank continues to influ- ence inter-personal behaviours throughout Nepal,with variations only in the degree and in the nuances. There are districts in the eastern Tarai where the privileged castes have even resorted to economic threats and physical vio- lence to enforce traditional caste-based practices. No consensus yet exists on which communities fall into the category of Dalit or on the actual size of their population. According to the 2001 Census Dalits comprised 13 percent of the population but the figure is con- tested. Box 12 shows how the different Dalit groups have recently been clas- sified by different agencies and authors. While many commonalities are found P U B L I C D I S C O U R S E A N D A C T I O N S 57 between these classifications, there are also significant differences. One de- mand of Dalits is to have an accurate,acceptable classification and a database on the Dalit sub-castes. Nepal's Dalit population can broadly be categorised as either Hill Dalits (who make up 61% of the Dalit population), or Tarai Dalits (who make up the rest). The largest sub group is that of the metal workers, the Biswakarmas, including Kamis (blacksmiths) and Sunars (gold workers) and according to some classifications,the Lohar,Tatma and Chunara from theTarai. Many smaller groups collectively comprise the Tarai Dalits. (See Figure 21.) One irony of the situation of Dalits is that they have traditionally practiced Hindu type stratification among themselves. The Dalit movement rejects this hierarchy and is working to remove the barriers between its constituent groups. Unlike many Janajatis,the Dalits have no geographical centre or"tra- ditional homeland" where they are numerically predominant. Map 2 shows that the Hill Dalit group is mainly concentrated in the Mid-western and West- ern Development regions (containing over 50 percent of their population), FIGURE 21 Distribution of the Dalit population (in percentage) to total Dalit population in Nepal Chamar/ Badi Harijan Ram Mushahar Dusad 0.15 % 9.12 % 5.84 % Paswan Gaine 5.37 % Tatma 0.20 % 2.60 % 1.21 %Bantar Khatwe Sarki 0.42 % Chidimar 10.79 % 2.53 % Dhobi 0.28 % Dom 2.49Jhagar % 0.10 % Halkhor 1.42 % 0.04 % Koche 0.02 % Pathharkara 5.86 % Unstated 7.92 % Dalits Dholi Damai 13.49% Kami, Sunar, Lohar (B.K) 38.10% Source: Census 2001 58 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal whereas 85 percent of Tarai Dalits live in the Central and Eastern regions (Gurung et al. 2005). One of the few pro-Dalit moves on the part of the government has been the ratification of several international conventions whose compliance is monitored by a group of human rights organizations and NGOs, including Dalit NGOs. However,the government does not appear to be taking the moni- toring reports seriously. NepalestablishedtheNationalDalitCommission(NDC)inMarch2002 with an all-Dalit membership. Like the Women's Commission, its members were chosen based on party affiliations; similarly, its functions were not le- gally mandated and funding was inadequate. Despite the odds,the NDC was Ram Krishna Dhakal, a well- known"upper"caste singer, able to draft a bill for itself, which, however, had still not been enacted as law drinking water offered by a by early 2006. Dalit rights activists have not been notably successful in lob- tailor, an "untouchable" in the caste hierarchy. The bying for either amending or repealing existing laws through public interest photograph is part of the Feminist Dalit litigation. In spite of this, and perhaps in response to growing public aware- Organization's campaign ness, there are occasional signs of progress. For example, in 2005 an NGO against untouchability. The caption on the photograph that had taken a case to court won an important judgement against the gov- read: "The only relation between two human beings ernment policy of building separate water taps for Dalits. is thirst and water". P U B L I C D I S C O U R S E A N D A C T I O N S 59 BOX 13 Key recommendations to improve the understanding of the Dalit sub-castes n A lingering hesitation to use the term "Dalit" or to name caste-based discrimination head-on and a preference for euphemisms only serves to confuse issues pertaining to Dalit rights. The term "Dalit" should be accepted universally. n No national exercise has been conducted for scheduling all castes and Janajati groups to produce a universally accepted list of who is and who is not Dalit. Several ministries use their own lists, and interna- tional agencies and national NGOs remain confused about which castes qualify as "Dalit" and which do not. A consensus list that could be adopted by all government offices should be prepared in collaboration with the National Dalit Commission. n Since no reliable, accurate database exists on the population and situation of Dalits, the government needs to support work aimed at gathering better caste-disaggregated data. This work can be part of the Central Bureau of Statistics' census and the NPC's Poverty Monitoring and Analysis System (PMAS). Disaggregated data is essential to the PMAS set up to monitor performance under the PRSP. Qualitative ethnographic data on the situation of different Dalit sub-castes such as the preliminary work done by the Dalit Empowerment and Improvement Plan (DEIP) on Tarai Dalits is also needed. The scholarship scheme for Dalits, which began in 1996, is con- strained by over-politicisation,procedural flaws and inadequate funding. Al- though the recently initiated Nepal Education For All (NEFA) primary educa- tion programme has set aside additional funds to provide scholarships for all Dalit children, the November 2005 Technical Review report documents that only 65 percent of Dalit students attending primary school in the sampled districts received scholarships. The report concludes that not only are gov- ernment allocated funds simply insufficient to cover all Dalit students as prom- ised,but the governance procedures in place for distributing these funds are not adequate (Bajracharya 2005). Donors have supported Dalit rights but have not done enough. NGO/ INGO critiques of donors include sporadic funding, non-interest in capacity enhancement,emphasis on political/social rights without sufficient attention to livelihood issues, and excessive reliance on"high"caste staff in the design and implementation of their interventions. Two kinds of Dalit NGOs are found. The first are large national NGOs, including the Dalit Welfare Organization (DWO), the Feminist Dalit Organiza- tion(FEDO),NepalNationalDepressedSocialWelfareOrganization(NNDSWO), Jana Utthan Pratisthan (JUP) and the Jagaran Media Centre (JMC). The second are the many small Dalit organizations working at the community level. All DalitNGOsconductadvocacy/activistfunctionsalongsidetraditionaldevelop- ment activities. The Dalit NGO Federation (DNF), with a membership of over 200DalitNGOs, isemergingasapowerfulconvergingpointforthemovement. 60 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal Dalit representation in executive bodies of political parties remains very low. The only Dalit member of the House of Representatives was elected on a NC ticket in 1991. There has been no Dalit representation in parliament since 1994, and Dalit representation has remained low in both elected dis- trict and village-level offices. Much blame for this falls on the political parties, which failed to field Dalit candidates or ghettoised them in losing constituen- The stamp says, "Untouchability is cies. With the exception of four Dalit members in the National Assembly, a social crime." there was in 1999 not a single person from the Dalit castes in the executive, legislative or judicial branches of government (Neupane 2000). TheNepaliCongressappointednoDalitministerbetween1990-2002. Only after that did the NC (Democratic) appoint one Dalit as assistant minis- ter. Following the takeover of February 2005, the King has included Dalits in the cabinet. Because Nepal's democracy has remained suspended, however, their presence in government has not been much different from that during the Panchayat period, when there were Dalit ministers. As noted earlier, the Brahmans, Chhetris and Newars have a monopoly over public service jobs and resources, and there is a near-total absence of Dalits in the public service. (See Table 6.) Moreover, very few Dalits are found either in the media or in other civil society organizations. In conclusion, with a few exceptions, Nepal's non-Dalit actors have leftittoDalitleaders,activistsandorganizationsto"fighttheirownbattle". This has not helped the Dalit movement, which will succeed only when it can build coalitions with reform-minded non-Dalits to add voice to Dalit demands. Among the exceptions,a one-rupee postage stamp (shown above) issued by the government recently is a good beginning. In addition, the Lawyers Na- tional Campaign Against Untouchability (LANCAU Nepal) is a group of prima- rily non-Dalits who have been doing vigorous advocacy and public litigation against untouchability. On the positive side, civil society in Nepal is experiencing an undeni- able mass surge towards democratic values and practices and a growing TABLE 6: Representation in government, 1999 Brahman/Chhetri Janajati Madhesi Dalit Newar Others House of Representatives 159 36 46 4* 20 0 & National Assembly Cabinet 20 4 5 0 3 0 Public administration 190 3 9 0 43 0 Constitutional bodies 14 2 3 0 6 0 Court judges & officials 181 4 18 0 32 0 DDC/ municipality 106 23 31 0 30 0 chairpersons Source: Neupane 2000: 82-83 *National Assembly only P U B L I C D I S C O U R S E A N D A C T I O N S 61 population Janajati the of Distribution 3 Gurung a Hark MAP ce: Sour 62 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal Even after democracy was established the Janajatis, religious minorities (mainly Muslims) and the Madhesi were left on the margins alongside the women and the Dalits. momentum for social reform. The question most hotly debated vis-à-vis Dalit rights is not whether the Dalit castes should enjoy equal rights,but what the best means are to achieve this. The post-1990 shift in values creates unprec- edented opportunities to push through measures and policies for Dalit rights. Responses to ethnic discrimination Most of the demands of Nepal's Adivasi Janajati (indigenous nationalities) movement centre around issues of governance and political representation. One such demand is the need for constitutional reform to remove discrimina- toryprovisions. Anotherisforequitablerepresentation, whichcanbeachieved either by changing the electoral system or through reservations ­ or both. The Janajatis also seek greater equality in linguistic rights and guaranteed access to common properties/resources which they claim were in the past unlawfully expropriated by privileged caste groups. The government originally prepared a schedule listing 61 Janajati groups,which was later reduced to 59 in the law. (SeeTable 7.)This number is still fluid however,and subject to change after negotiation. Various complexi- ties are involved in compiling a definitive list. For example, the Rai commu- nityencompassesanumberofdifferentgroupsincludingtheLohrung, Kulung, Chemling, Tachhing and Bahing who speak distinct languages but over the P U B L I C D I S C O U R S E A N D A C T I O N S 63 past two centuries have used the title"Rai"in presenting themselves to out- siders. These groups are now demanding recognition as separate ethnic groups in the government's list. Among the groups in the current official list 18 are from Mountain regions,24 from the Hills,7 from the Inner Tarai and 10 from the Tarai. Overall the concentration of Janajati groups is heaviest in the eastern part of the coun- try and in the northern Hill and Mountain areas. (See Map 3.) The Census (2001) enumerated only 43 of 59 Janajati groups and reported a population of 8.27 million or 37 percent of Nepal's population. After incorporating the 16 "missing" indigenous nationalities, their numbers might be as high as 40 per- cent of the entire population. Four Janajati groups have populations exceed- ing one million, six have numbers between 100,000 and one million, nine groups have populations of less than 100,000 and some have less than 1,000 people. Several other groups do not exceed 10,000. As noted in the discussion on poverty, many disparities are found among the different Janajati groups. The Nepal Federation of Indigenous Nationalities (NEFIN) has highlighted this, defining 10 of the 59 Janajati groups as"endangered", 12 as "highly marginalised", 20 as "marginalised",15 as "dis- advantaged" and two as "advantaged" or better off. Thus,the Janajati move- ment has recognised its own heterogeneity and expects different levels and TABLE 7: Janajati groups classified by NEFIN based on socio-economic status Classification of indigenous nationalities Region Endangered Highly Marginalised Marginalised Disadvantaged Advantaged Mountain (18) Shiyar Bhote Bara Gaule Thakali Shingsawa (Lhomi) Dolpo Byansi (Sauka) Thudam Larke Chhairotan Lhopa Marphali Thakali Mugali Sherpa Topkegola Tangbe Walung Tingaule Thakali Hill (24) Bankariya Baramu Bhujel Chhantyal Newar Hayu Thami (Thangmi) Dura Gurung (Tamu) Kusbadiya Chepang Pahari Jirel Kusunda Phree Limbu (Yakthung) Lepcha Sunuwar Magar Surel Tamang Rai Yakkha Hyolmo Inner Tarai (7) Raji Bote Darai Raute Danuwar Kumal Majhi Tarai (10) Kisan Dhanuk (Rajbansi) Dhimal Meche (Bodo) Jhangad Gangai Santhal (Satar) Rajbanshi (Koch) Tajpuriya Tharu Total 10 12 20 15 2 Source: NEFIN 64 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal forms of policy and affirmative action for its constituents. The GSEA report recommends a fresh classification to identify all Janajati groups based on poverty incidence, educational levels and key health indicators to serve as basis for eligibility to special state initiatives, including reservations and schol- arships, for those most disadvantaged. As noted above, the use of periodi- cally up-dated national statistics as the basis of affirmative action by the state would help ensure that 1) such special assistance does not become an iden- tity-based "entitlement" (as seems to have happened in India), and that 2) only the groups revealed by recent Census,NLSS and DHS data as being most in need,receive such assistance. The Janajati community Nepal's Constitution (1990) explicitly uses the term Janajatis in itself has also been ambivalent about receiving Article 26 (10) acknowledging both their presence and their relative social donor aid ­ the opponents and economic deprivation. The use of Nepali as the only official language to argue such support would mean an end to the exclude languages spoken by Janajati groups and by other linguistic minori- independent spirit of the Janajati movement. ties like Maithili, Bhojpuri, etc., is discriminatory. However, modest efforts havebeenmadetouseminoritylan- guages in newscasts on state-run radio. Also being piloted is a programme to train and use bilin- gual teachers in primary schools in areas with significant numbers of non-Nepali speakers. (For further discussion see section on educa- tion.) Nepal began planning for the Janajatis only in the Ninth Plan (1997-2002). TheTenthPlandevotes an entire chapter to issues related to the indigenous nationalities. Whilethisisapositivedevelopment, it is noteworthy that the plan does not have quantitative targets. The GSEA estimates the presence of at least 150 Janajati or- ganizations (but claims of numbers ashighas300havealsobeenmade). Itissignificantthat,unlikemostDalit andWomen's organizations,Janajati organizations have not relied on do- nor or government funding. They P U B L I C D I S C O U R S E A N D A C T I O N S 65 have generally been financed largely through personal donations and vari- ous types of fees, including membership. Some individuals have donated land and buildings to specific organizations and others have created trusts or endowments to fund their activities. NEFIN,with 48 Janajati organizations (June 2005), is at the forefront of theJanajatimovement. Initiallyitfocusedonreligiousfreedom, linguisticequal- ity and rights, and the promotion and preservation of culture. More recently it has also raised issues of governance, human rights, biological diversity, indig- enous knowledge systems, conflict and peace building, constitutional reform, restructuringNepal'spoliticalinstitutionsincludingtheelectoralsystem, feder- alism, affirmative action and social inclusion. Manydonorshave beenunabletoclearlydifferentiatebetweenDalits and Janajatis. Until ethno-politics was legalised in 1997, donors feared that the dominant caste groups could interpret their support to Janajati-oriented projects as assistance to those who wanted to"tear Nepal apart." The Janajati community itself has also been ambivalent about receiving donor aid ­ the opponents argue such support would mean an end to the independent spirit of the Janajati movement. The major demands of the Janajati movement include: n Constitutional reform. This is both an overarching demand related to many other issues, and an affirmation that the Janajatis want a wholly reformed contract with the state. The movement wants Nepal to be declared a secular state, and all Janajati languages recognised for use in state affairs alongside Nepali. n State assistance in implementing the acknowledged right of every community to run schools in which primary-level classes are taught in themothertongueofthestudents. Theywanttheabolitionofcompulsory Sanskrit study. They also propose a three-language policy in education and administration: mother tongue, a second Nepali language (in most cases,Nepali itself) and an international language. n Equitable representation through different measures including "restructuring the Nepali state" by changing the electoral system, and by affirmative action measures to increase the ethnic diversity of the civil service. n Accesstocommonpropertyresourcesthatwereoncecommunallyowned as kipat land by certain Janajati groups. The movement demands Janajati rights to resources based on recognition of their "traditional right of ownership and usage" especially for forests and pastures. While return to kipat land ownership is probably neither logistically nor politically feasible, some recognition of the special rights of Janajati groups to use 66 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal forest areas and high pastures that were traditional sources of livelihood may be possible within the scope of existing community forestry regulations and decentralisation policies. n Full self-determination. On this issue,the general consensus seems to be that the formation of "mini" ethnic states in areas where major Janajati groups are concentrated is neither politically nor economically realistic. (See Map 1 and Map 3.) Instead, there are demands to establish self- governing ethnic autonomous regions within the current unitary state or a newly organized federal polity. Census data analysed by Gurung et al. (2005) show that although some ethnic groups remain concentrated in certain regions of the country,Nepal's diverse caste and ethnic groups have increasingly intermingled across the landscape ­ as they have done for thousands of years. Finding a governance structure and electoral system that permits greater self determination and political influence for Nepal'sdiversegroupsremainsachallenge­butgenuineimplementation of decentralisation is likely to help. C H A P T E R 5 INCLUSIVE SERVICE DELIVERY V INCLUSIVE SERVICE DELIVERY Improving access to health Improving access to education I N C L U S I V E S E R V I C E D E L I V E R Y 69 Improving access to health Nepal has started to put a greater emphasis on preventing diseases that af- flict the poor and has begun reaching out to those with the greatest health burden. It has also started decentralising health care delivery by transferring funds and responsibility for managing health facilities (including sub-health posts and selected health posts and primary health care clinics) to locally constituted Local Health Management Committees (LHMCs) in 28 districts. The logic behind this devolution is that by making health care providers ac- countable to a local committee,the local residents will have more say in how public resources for health are used, and that consequently the quality of care will improve for the whole community (including the poorest, who are The need to touch patients is central to most dependent on public health services). Just how much voice poor and the delivery of care, yet the socio-cultural excluded groups will have in the decisions of these local committees,however, obstacle of is not yet clear. With only 700 of the 3179 sub health posts turned over to date, untouchability has not been discussed or it is too early to determine whether the decentralisation process is working. addressed by Nepal's health care system. 70 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal FIGURE 22 Determinants of health outcomes in rural Nepal In addition, as a signatory to the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW,1993) and the UN Con- vention on the Rights of the Child (CRC, 1989), Nepal implicitly recognises healthcare as a basic citizenship right, encompassing racial and gender equity and equality, sexual and reproductive rights and client-centred ser- vices. However, the effort to reorient policy and health services along the rights-based approach remains ad-hoc and immature. The Health Sector Re- form Strategy (2004),the basis for the Nepal Health Sector Programme (NHSP), defines the new decentralised healthcare approach. The NHSP is also the framework for the on-going Sector Wide Approach (SWAp) in health. Nepal's child mortality declined during the 1990s, putting it on track for achieving the MDG of reducing child mortality by two thirds by 2015. Infant mortality has also decreased: between 1996 and 2001, the death of children under age one dropped from 84 to 75 per 1,000 live births. Nevertheless this is still extremely high, as is Nepal's maternal mortality, and there are sharp dis- parities between different groups that are related not only to economic status, rural/urban residence and region but also to gender, caste and ethnic identity. I N C L U S I V E S E R V I C E D E L I V E R Y 71 Determinants of health outcomes There are many interrelated factors ­ cultural, religious and social beliefs and norms (especially those that reflect the entrenched gender, caste and ethnic hierarchies) as well as economic, institutional and location-related specifici- ties ­ behind these differential health outcomes. (See Figure 22.) Supply-side factors Many determinants of health outcomes operate indirectly by reducing the person's access to healthcare and influencing the kind of care they get. (See Figure 22.) Here institutional/political factors are important and are a major focus for policy intervention. These factors include government budgetary allocation and policy attention to rural healthcare and the diseases of the poor and women. Although many professionals in the government health care service are dedicated to caring for the poor and disadvantaged, too many others have little motivation to serve those who are beneath them in the socio-eco- nomic hierarchy. The idea that to be touched by a member of the Dalit caste puts one in a state of ritual pollution or religious impurity is still deeply em- bedded in people's thinking and integrated into the norms of social interac- tion in rural Nepal. A World Bank study found that caste discrimination by health service providers was a major barrier to health seeking behaviour among Dalit women (World Bank 2001). Similarly, the DFID supported Safer Motherhood Project found that disrespectful attitudes of service providers towards those beneath them in the caste hierarchy was a major reason that many women did not seek either pre or post natal care or emergency obstet- ric care (Clapham et al. 2005). As trained scientists, Nepali doctors may not be influenced by ideas of untouchability,but the same enlightened behaviour is not assured for all health service workers. The need to touch patients is central to the delivery of care, yet this potential socio-cultural obstacle has not been discussed or addressed by the health care system. In addition to location-related service delivery problems,other sup- ply side issues that influence public healthcare include the type of services offered, flexibility in timing for care provision, the public's perception of the quality of care and attitudes of providers. The number of both public and private health facilities has increased over the past decade but the services remain easily accessible only to the richest groups or those closest to urban centres. Generally, most healthcare facilities, including trained personnel, are concentrated in urban areas while rural facilities remain under-staffed and under-supplied. 72 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal Another aspect is that whereas women mostly staff the lower health service positions, men mostly staff the higher service positions and this re- mains a major obstacle to proper healthcare for women,who are often reluc- tant to be seen by a male physician. The government's recognition of the need for female service providers to reach women patients prompted the hiring of female Mother and Child HealthWorkers. To overcome the problem of frequent transfers of health staff, these women are purposefully recruited and trained locally and,unlike other government workers,they are expected to remain in their community. At the very bottom of the system are the un- paid Female Community Health Volunteers, roughly one for every ward in the country. These women have been extremely successful in carrying out the biannual vitamin A supplementation and de-worming programmes. With community based training, these female volunteers have also learned to di- agnose and treat pneumonia as part of the government's programme on In- tegrated Management of Childhood Illnesses. Ascertaining the degree to which the conflict in Nepal has compro- Health service delivery is weak due mised health service delivery is difficult. Anecdotal reports suggest that in to a lack of trained some areas healthcare providers have been threatened and forced to aban- staff, inadequate supply of drugs and don their posts ­ or they have used this as a justification for leaving remote medicines and misallocation of and difficult postings. resources. Demand-side factors Economic status affects access to and utilisation of health services. In the richest households, per capita expenditure on health is 10 times what it is in poor households (World Bank 2005a). Certain caste, ethnic and minority religious groups have higher poverty incidence; thus they are more likelytofaceeconomicconstraintstoseekingqualityhealthcare. Factors related to gender, caste and ethnicity Women's health outcomes are directly affected by their sub- ordinate status vis-à-vis the men and the senior women in the family. Although the preference for male children varies from group to group, overall females tend to be less valued thanmalesandthisisreflectedinpoorerfemaleperformance on all indicators, especially education and health. Health outcomes are the worst for women because of biology (specifically their reproductive roles). The biologi- cal risks associated with childbirth can be compounded by cultural practices including early marriage and child bear- ing, and food restrictions during pregnancy and during I N C L U S I V E S E R V I C E D E L I V E R Y 73 menstruation. In terms of age at marriage in rural areas, FIGURE 23 Average age of females Newars and Janajatis tend to marry the latest and Mus- at marriage (in rural lims, Dalits and Tarai Middle Castes the earliest. (See Fig- areas) by caste/ethnicity ure 23.) Girls aged 15 to 19 are twice as likely to die in Total B/C Tarai Middle Castes childbirth as are women in their 20s. Childbirth is seen as Total Dalits a normal occurrence not ordinarily requiring medical at- Newars tention,so pre- and postnatal care may be considered un- Total Janajatis necessary in many families. Compounding all of this is Muslims the sense of laaj or"shame"about their own body and re- Nepal productive functions that make it difficult for women to 14 15 16 17 18 communicate their symptoms ­ even within the family. Age When healthcare usage and outcomes are bet- Source: DHS 2001, K. Gajurel ter for women, they are better for children as well. Brah- mans, Chhetris and Newars, who have the best health in- FIGURE 24 Correlation between dicators for women,also have the lowest infant mortality mothers' education and the under-5 mortality rate (52.5 and 56 per thousand respectively) compared to rates of their children a national average of 84 (CBS,NLSS I). 140 'ooo)120 As well as differences in poverty incidence be- (per 100 tween these various groups, the higher average educa- rates 80 tional status of Brahman, Chhetri and Newar women is 60 40 another contributing factor in their relatively better out- mortality 20 comes. In addition to strong linkages between education Child 0 No Some SLC and and reduced fertility, a similar relationship between education Primary secondary above mother's education and child survival has also been docu- Mothers' educational background mented. (See Figure 24.) Source: NDHS, 2001 Note: SLC = School Leaving Certificate The government has acknowledged and tried to address the problem of gender discrimination as a barrier to health care. But very little attention has been given to FIGURE 25 Immunisation status of children under five years howthelegacyofcasteandethnicity­andparticularlythe by caste/ethnicity practice of untouchability ­ affects the interface between Total B/C health service providers and patients of both sexes. The Tarai Middle Castes under-five mortality rate among excluded groups such as Total Dalits the Gurung, Rai, Limbu, Magar and Tamang is about 133, Newars and among Dalits it is 171, significantly higher than the na- Janajatis Muslims tional average of 105 (UNDP: NHDR 2001 and 2004, IIDS 2004). Figure 25 shows that immunisation campaigns in Nepal Nepal have been very successful and that fewer than 10 0 20 40 60 80 100 percentofNepal's childrenhavenothadanyimmunisation Percent at all. The average gap between the percentage of boys full partial none and girls being immunised is also very small (2.4%). Source: NLSS II 74 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal However, even these very successful campaigns have been less effective in reaching the Tarai Middle Castes,the Muslims and the Dalits. For Janajatis and members of linguistic minorities in the Tarai, lan- guage is also a factor. For groups who do not speak Nepali as a mother tongue ­ and who may not speak it at all ­ the absence of health care providers able to converse in their own language makes it difficult to convey their symp- toms or to understand instructions for treatment or medicines. National responses to address social exclusion in health Clearly some of the determinants of high morbidity and mortality among ex- cluded groups depicted in Figure 22 require actions beyond the health sys- Less than 10 percent of tem. Improved transportation and sanitation infrastructure, reduced income Nepal's children have not had any and consumption poverty, and increased education levels are all associated immunisation at all. But even the very with improved health outcomes. successful Nepal's PRSP/Tenth Plan accepts that health service delivery is weak immunisation campaigns have been due to a lack of trained staff, inadequate supply of drugs and medicines and less effective in reaching the Tarai misallocation of resources. The Health Sector Strategy (HSS) acknowledges Middle Castes, the the need to reorient the health delivery system towards the poor, but does Muslims and the Dalits. I N C L U S I V E S E R V I C E D E L I V E R Y 75 not discuss how this is to be done ­ and does not deal directly with the kind of trade offs that would have to be made between equity and efficiency if the most vulnerable are to be reached. Nor does it deal directly with the issue of social exclusion. The way ahead The Nepal Health Sector Programme (NHSP) based on the HSS is currently being supported by pooled donor funds through a Sector­Wide Approach (SWAp). The NHSP goal is to establish equitable access to quality healthcare for all. The formal policies and funding put in place by the NHSP are only the beginning. Without passionate attention to the details of implementation, these policies could easily be circumvented, leading to continued exclusion of women, Dalits and Janajatis. Successful health sector reform will require patient development of detailed formal modalities and mechanisms to overcome the barriers to inclusion; it means recognising the supply-side constraints and putting in- centives and disincentives in place to reinforce changes in the behaviour of service providers. This is where good monitoring and evaluation, and social accountability mechanisms are essential to develop greater accountability at all levels of the health care system. Fortunately,numerousexamplescanbefoundinNepalwherethiscom- plex change process has been initiated and where improved service is already evident. Most of the interventions are relatively small-scale,however,bringing changeonlytoafewvillagesoratmosttoclustersofdistrictswheretheprojects are implemented. Some essential steps for improving healthcare include: n Tapping the potential and enhancing the skills of local women health providers, n Paying attention and ensuring inclusive governance and accountability in the LHMCs, and n Developing procedures for community social audit of health care delivery and collecting institution-level data on service utilisation disaggregated by gender, caste and ethnicity. Improving access to education Reforming public education in Nepal The state assumed responsibility for the education system in the 1970s and previously locally run schools were turned over to a centralised educational administration. Public education expanded rapidly thereafter, from 7,256 76 U N E Q U A L C I T I Z E N S Gender, Caste and Ethnic Exclusion in Nepal primary schools catering to 102,704 students in 1970 to 23,746 primary schools serving 4,030,045 students in 2004 (Bhatta 2005). Quality, however, did not keep up with the expansion in numbers; average examination scores at grade three and grade five reached only about 51 and 50 percent respectively in 2001. A centralised bureaucracy trying to oversee this vast network of schools Social inclusion is used to describe the and increasing political interference in the hiring of teachers resulted in schools complementary having only 17 percent of trained primary-level teachers, text books arriving approach that seeks to bring about late in the school year and teachers often not showing up in the classrooms. system-level institutional reform The involvement of teachers in party politics during the Panchayat period and policy change to continued even after 1990 and the politicisation of teachers often affected remove inequities. I N C L U S I V E S E R V I C E D E L I V E R Y 77 government moves to introduce more discipline and quality into the class- rooms. In general,public school education has come to be perceived as poor in quality and parents who can place their children in private schools ­ where the pass rates for the all important SLC examinations are 85 percent as com- pared to 15 percent in public schools. Major efforts aimed at making education inclusive Main elements of the NEFA SWAp To help poor and socially excluded chil- dren receive the kind of education that will open opportunities for them, simply getting them into Nepal's public schools as they are will not be enough. Efforts to improve their access must also be closely linked to overall efforts to improve the quality and accountability of Nepal's education system. The Nepal Education for All (NEFA) Core Document (2003) sets out a five-year plan with three primary objectives: (i) ensuring access and equity in primary education,(ii) enhancing quality and relevance of primary education, (iii) improving efficiency and institutional capacity. Donors have come to- gether to support the NEFA as a SWAp rather than embarking on their own separate initiatives. Some have chosen to support this through project fund- ing within the overall NEFA framework whereas others have chosen to pool their funding and channel it through the government's education budget on the basis of progress on key indicators. Through the NEFA, HMG/N has committed to provide equal access to educational resources for all communities in Nepali society, including girls and women, linguistic minorities, Dalits and Janajatis. The overall objective of the NEFA is to strengthen Nepal's institutional capacity at national, district and school levels in order to deliver more efficient and better quality educa- tion. From the perspective of social justice, its objective is to foster a genu- inely inclusive educational environment in Nepal. The articulation of this com- mitment came in the form of a Vulnerable Community Development Plan (VCDP) prepared by government. The main focus of the following discussion is on the elements of this plan. Decentralisation In an effort to reform the system and shift the incentives, in 2001 parliament passed the Seventh Amendment of the Edu- cation Act ­ allowing management of local public schools to be handed over to School Management Committees (SMCs) composed of elected parents and guardians as well as the head teacher. The parents are the main stakeholders in the SMCs and the lawful managers, and also have the right to hire and fire teachers. The head teacher is supposed to present monthly expense state- ments as well as annual audits to the SMCs, and to prepare a social audit of how block grants are spent. However, these responsibilities are new and the 78 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal SMC membershave notbeentrainedfortheseroles. Only38percentofschools in a sample survey carried out for the Technical Review (Bajracharya 2005) completed the social audit ­ and only 16 percent actually made their reports available by public notice. At present, the SMC requires only a single woman member on the committee. No requirement is made for SMCs to have representation of Dalits or Janajatis. The New Era Baseline Study on community-managed and gov- ernment-managed schools (June 2005) found Dalits were represented in only one third of the SMCs,despite their considerable population size in the com- munities served by the schools. The Technical Review found Dalit participa- tion in the SMCs very low,merely four percent of the sample. Female partici- To help poor and socially excluded children receive pation was even more alarming, despite the requirement of having one the kind of education that woman member on the SMC. In the sample schools women constituted only will open opportunities for them, simply getting them two percent of the SMC members. This suggests that greater attention needs into Nepal's public schools as they are will not be enough. to be paid to setting up mechanisms to ensure the creation of more repre- Efforts to increase their sentative parent bodies ­ broadly reflecting the caste,ethnic and gender pro- access must also be closely linked to overall efforts to file of the communities they serve. improve the quality and accountability of Nepal's education system. I N C L U S I V E S E R V I C E D E L I V E R Y 79 The involvement of local FIGURE 26 Total number of male and female teachers (2001) communitygroups UndertheVCDP the SMCsaresupposedtosignagree- ments (MOUs) with community Secondary groups to jointly carry out social 1,488 17,358 mapping to help trace out-of-school Lower Secondary children and to carry out a social au- 3,740 22,938 dit to ensure that scholarships go to Primary those most in need. Preventing elite 24,427 72,232 capture and undue politicisation of SMCsisvitalfortheirsuccess. Involve- Female Male ment of community groups is a criti- cal mechanism to provide some Source: School Level Educational Statistics of Nepal, MOES 2001 checksandbalancesontheSMCsand to increase their accountability. Block grant funding with incentives for inclusion The Commu- nity School Support Project began in 2003 and has already transferred 8,000 of the approximately 23,000 schools to community management with block grant financing. The project makes progressive block grants tied to perfor- mance in support of improved student educational outcomes and inclusion of girls and Dalit and Janajati students. From a basic block grant of NRs.10,000, the SMCs can receive as much as NRs. 50,000 per year if they meet all the performance requirements. This mechanism will create incentives for com- munities to reach out to children from socially excluded communities. Increasing the diversity of teachers Schools with female teach- ers tend to attract more female students. For that reason the policy of having at least one female teacher per school in multi-teacher schools was estab- lished over a decade ago. However, the policy has still not been fully imple- mented. Currently only 25 percent of teachers are women, an increase of only four percent since 1996 ­ and these are heavily concentrated in urban areas. While the number of women teachers is insufficient at the primary level,their proportion declines even further at lower secondary and second- ary school levels. (See Figure 26.) Not surprisingly, the districts with the low- est proportion of women primary school teachers in 2001 (e.g. Bajhang, Achham,and Baitadi) had very low female net enrolment rates. The Technical Review (Bajracharya 2005) noted that just as having a womanteachertendstoattractgirlstudents,havingJanajatiorDalitstaffmem- bers has a positive impact that manifests in greater enrolment of students from these social groups. Unfortunately, however, the review noted that as of now, the majority of teachers in the sample of 1,000 schools surveyed were 80 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal from the Brahman/Chhetri group; 23 percent were from Janajati groups and only two percent were Dalits. Incorporating local languages and cultural diversity in the pri- mary school curriculum:overcoming language-based exclusion Overall, 52 percent of Nepalis do not speak Nepali as their mother tongue and this puts non-Nepali speakers at a disadvantage in terms of access to education. For most Janajati children Nepali is not their mother tongue so they are intro- duced to school and to a new language at the same time. Success in Nepali medium primary schools is also difficult for many people from the Tarai who speak Maithili,Bhojpuri or Hindi as their mother tongues. The main causes of dropout among both these groups of children are: 1) the absence of a cur- riculum representative of Nepal's ethno-linguistic composition and 2) few teachers who can speak local languages to assist children to understand the lessons taught in Nepali. Implementation of the constitutional provision for primary educa- tion in the mother tongue faces institutional inertia for two reasons: 1) the complex and emotive issues of mother tongue education when the govern- ment sees universal fluency in the Nepali language as an important source of social cohesion and 2) the very real financial and technical difficulties of pro- viding such education in a nation with some 127 languages ­ though only a few have a literate tradition. Indeed, many rural indigenous people would prefer to have their children learn Nepali ­ along with English ­ since they see these skills as clearly related to future economic success. For increasingly vo- cal ethnic activists, however ­ and for many belonging to linguistic minorities orindigenousgroups­thefailuretoimplementthepromiseofmother-tongue education is part of what they see as a larger failure of the state to ensure equal access to public services, economic opportunity and political voice to all social groups. Two key goals of education are:1) to impart the basic skills and knowl- edge a child needs to be economically successful in life and 2) to give each child a sense of social identity. The state is taking on the responsibility of ensuring that non-Nepali speaking children achieve the first goal through developing and implementing a bilingual education module for schools with large numbers of non-Nepali speaking students. This module aims to facili- tate language transition for non-Nepali speaking children by employing bi- lingual women teachers from the local community. To achieve the second goal of building social identity, communities themselves can take responsi- bility by collecting and documenting information on their own history and culture for use in the curriculum, since up to 20 percent of the overall curricu- lum is now permitted to be decided locally. A pilot project is being imple- I N C L U S I V E S E R V I C E D E L I V E R Y 81 mented to develop the Bilingual Mod- ule. While the approach is still only being tested in a small number of schools, it will be important to care- fully monitor this pilot and learn from it in order to eventually integrate this approach into the national education system wherever significant non- Nepali speaking populations reside. Monitoringprogressonso- cial inclusion For the NEFA, measur- ing results in terms of enrolment, re- tention and completion rates is core to the design and funding mecha- nisms of the SWAp. The Vulnerable Community Development Plan also commits to track gender parity and the inclusion ofDalitsandJanajatichil- Literacy rates are dren. The Flash Report which has now completed three rounds is a major improving for both males and females, faster for vehicle for tracking inclusion. It is an impressive effort that asks every school females so the gender gap is narrowing. Nepal should in the country to report on a number of parameters and actually processes achieve gender parity in the data and makes the results available rapidly enough so that managers at enrolment by 2010. all levels in the education system (including the SMCs) should be able to as- sess their situation and make course corrections. However, the system is facing a number of challenges that need to be addressed if the data are to be meaningful. Getting consistent and accurate data from the field continues to be a challenge. For example, some 20 percent of schools did not respond on the first Flash Report and it is not clear, when schools do report, that they have clear and consistent definitions of the different social categories being tracked ­ especially for the Janajati group. As noted earlier, the broad Janajati category con- tains many groups (like the Newar and the Thakali and others) who have high human development indicators and it is important to distinguish between better off and disadvantaged Janajatis ­ especially when try- ing to track performance on social inclusion in education. The high par- ticipation of the more advantaged Janajati groups could easily mask poor outreach to disadvantaged groups. Also,the reporting formats in the Flash Report do not indicate how the reported numbers on children from dif- ferent groups relate to the total number of children in the appropriate age cohort in the population of a particular district ­ information that is 82 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal important to give a sense of whether out-of-school children in a particu- lar group are being reached. Beyond primary education Primary education is the foundation for ensuring educational parity among various groups, which is also the first step towards effective social inclusion. The excluded groups in Nepal are virtually unrepresented in higher education, and this is largely due to exclu- sion at the lower levels. Reforming education from below must be matched with affirmative action initiatives from above to support higher education of members of excluded groups who have managed to overcome the odds. (See discussion of affirmative action.) C H A P T E R 6 INCLUSIVE GOVERNANCE VI INCLUSIVE GOVERNANCE Local development groups and coalitions for influence from below Affirmative action I N C L U S I V E G O V E R N A N C E 85 The issue of governance is at the core of this report ­ as suggested by its title, which focuses on equal citizenship. In this section the report presents two promising approaches for realising the goal: one ­ group-based develop- ment ­ that draws its strength from the grassroots but needs to be nurtured by the state, and another ­ affirmative action ­ where the state must take the lead. Local groups such as the community forestry group shown below are becoming a powerful Local development groups and coalitions for social force that can change the rules of the influence from below game. The placards shown below read,"Establish community forestry rules", Nepal has a long tradition of local civic organizations. Many of the tradi- "Stop insurgents from destroying our community tional groups were organized on the basis of religion but their functions also forests". encompassed secular management of common resources. Over the last 30-40 years, the customary groups have been supplemented by"spon- sored groups" ­ most formed by state agencies,donors and NGOs for specific development objectives such as service delivery, livelihood improvement, infrastructure build- ing, resource management, credit extension and empowerment. Table 8 gives an indicative typology of group-based institutions. Some grassroots groups have begun to replicate themselves and have organized themselves into larger federations ­ some at the dis- trict level,some at the national level and a few that even articulate loosely with international advocacy groups. These higher-level associa- tions give voice and added political representation to their constituents through lobbying for policy change, networking and publicity cam- paigns. In terms of the GSEA con- ceptual framework, local level groupsareanimportantmechanism 86 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal through which bottom up empowerment ­ however partial and contested ­ has been taking place in Nepal. When such groups federate, some like FECOFUN, SPOSH, NEFSCUN and many others ­ have demonstrated the poten- tial for influencing "the rules of the game" at the district and national level to make the overall opportunity structure in Nepal more inclusive. A major challenge for the stalled decentralisation agenda, despite the provisions in the LSGA, is the clarification of roles and responsibilities ­ not only between the central line ministries and locally elected bodies, but also between the latter two and local civil society groups like the school man- agement committees, forest user groups etc., many of which have been given significant control over state resources. Attention needs to be given to deter- mining the comparative advantages of each of the three actors, coordinating their efforts and ensuring that there is adequate representation of excluded groups and their interests at all levels. TABLE 8: Indicative typology of group-based institutions Trusts Associations/Networks* Committees Cooperatives* Federations n Farmer n District IPM n Chairman/ n Society of nSociety for Managed Farmer Field Manager Women's Preservation of Irrigation School (FSS) Committee Unity, Nepal Shelters and (in VDCs) System Association (NMES) Habitations Phewa Tal Promotion n n Association of Fishers' n Mushroom in Nepal Trust IPM Trainers Enterprise Cooperative (SPOSH) (FMIS) (TITAN) Committee (of Pokhara) nFederation of n Seed Sector (of Pokhara) n Production Credit Community Support Group n School for Rural Women Forest Users, Network Management (PCRW) Nepal Committees n Himalayan Cooperatives (FECOFUN) (SMC) Grassroots (individual names nNational Women's Natural vary) Federation of Resource n Horticulture Irrigation Management Cooperative Water Users Association (of Bardiya) Association (HIMAWANTI) n Credit of Nepal n NTFP (Non- Cooperative (NFIWUAN) Timber Forest (national) nNepal Products) n Milk Cooperative Federation of Association (national) Forest n CDG n Savings and Resource User (Community Credit Groups Development Cooperative (NEFUG) Groups) Society nNational Network# (SACCOS) Federation of n Small Farmers Savings and * Some formal (registered) federations call themselves Cooperative, Ltd. Credit associations; while some associations and networks are informal (SFCL) Cooperative (unregistered). n Kaski District Unions, Ltd. # CDGs are formed by the Department of Soil and Water Fish Growers (NEFSCUN) Conservation. They are similar to the forestry user groups Association nFederation of (FUGs) of the Department of Forests. The CDGs have formed (KDFGA) Water and a network to obtain information and to share activities. n Microfinance Sanitation Association of Users in Nepal Nepal (MAN) (FEDWASUN) (The lists in this table are illustrative and by no means complete; they serve as examples only.) I N C L U S I V E G O V E R N A N C E 87 Groups remain an Nepal's Constitution incorporates decentralisation as one of the di- effective modality for rective principles and policies of the state. Decentralised governance ­ the empowering and facilitating greater devolution of initiative,authority and resources to local bodies and organiza- inclusion for women and for producing longer- tions ­ has been conceived within an overarching state apparatus. The Maoist term positive conflict, however, has reduced the effective reach of both elected governing development outcomes. The MESI study found bodies and administrative agencies in the countryside. The elected local bod- that membership in a ies that were to be the pillars of grassroots democracy and the institutional group was associated with an increase in anchors for decentralisation have remained inoperative since July 2002,after empowerment levels. which several non-elected alternatives have been tried out with limited suc- cess. How development groups used to operating without coordination with or accountability to local authority will relate with the elected bodies when they are reinstated remains to be seen. Another issue concerns the barriers that many groups face in ob- taining legal registration as a federation or association at the district or na- tional level. These organizations can register either as NGOs (with the CDO at the district level) or as cooperatives (under the Cooperative Law). Registra- tion as an NGO has become more problematic recently due to current restric- 88 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal tions imposed by the new NGO code. There are also certain persistent difficul- ties in the Cooperative Law, namely the outmoded restrictions prohibiting the formation of more than one of a particular type of cooperative per dis- trict. In some districts this restriction has allowed certain powerful groups to establish a monopoly and has kept grassroots based cooperatives from regis- tering. Hesitation on the part of the government to register new coopera- tives may also be due to the concern of financial regulators at the national level over the growing number of scandals involving urban savings and loan cooperatives, some of which have disappeared or gone bankrupt and ab- sconded with members'savings. Recognising that the latter situation is quite rare in rural cooperatives,where membership participation is active,the gov- ernment is considering the establishment of a new regulatory institution for second tier financial service providers (such as microfinance NGOs, the Grameen replicators and savings and credit cooperatives). Better regulatory supervision may remove the hesitation to register rural primary cooperatives. However,the issue of allowing only one of each type of cooperative to regis- ter in each district remains. Some GSEA findings relating to groups include: n A background study carried out by the GSEA attempted to roughly estimate the total number of local-level groups, based on statistics available and separate studies done by 17 agencies. Based on data on programmes in nine sectors, about 400,000 local groups are operating in Nepal (Biggs et al. 2004a; Biggs et al. 2004b). n Hillareasaremorelikelytohavegroup-baseddevelopmentactivitiesthan the Terai and Mountain regions. n The idealised notion of "community" fails to recognise factional interests within communities: class, caste and gender-related conflicts can and do occur even within the community forestry groups, which are said to be the most successful of the local development groups. n Groups remain an effective modality for empowering and facilitating greater inclusion for women and for producing longer-term positive development outcomes. As noted earlier, the MESI study found that membership in a group was associated with a five point increase in empowerment levels. However, disparities are found in terms of empowerment, with Brahman, Chhetri and Newar members of groups tending to benefit the most. n Although women are fairly well-represented as group members, in mixed gender groups they continue to play a less prominent role on the executive committees. Data on group membership and leadership disaggregated by caste and ethnicity is almost non-existent. I N C L U S I V E G O V E R N A N C E 89 BOX 14 Fishing for guaranteed livelihoods There are some examples of groups that have successfully developed sustainable livelihood activities in collaboration with government agencies. Two government agencies, the Directorate of Fisheries Development and the Fisheries Research Centre (FRC) working with local Fishers' Enterprise Committees in the Pokhara valley have helped improve fishing practices and protect the fishing rights of the Jalari, the occupational fisher caste. The livelihood of the low-caste Jalari was threatened after the catch in the lakes of Phewa, Begnas and Rupa began to decline in the early 1970s. The FRC then stepped in to help the nomadic fisher-folk by introducing Subsistence Cage Aquaculture as a substitute to traditional fishing. By the mid-1980s the Jalari fishers in the Pokhara lakes had organized themselves loosely into groups that were formally structured only in the 1990s. A women's sub-group -- Machhapuchare Mothers' Group -- was also formed. This sub-group undertook anti-drinking and anti-gambling campaigns and other social activities within the community. A few years ago the Kaski District Development Committee (DDC) was planning to call for bids for fishing rights in the Phewa Lake and a real possibility existed that other parties could have out- bid the Jalari fishers. In collaboration with the FRC, the Phewa Tal Fishers' Enterprise Committee successfully lobbied the government to withdraw the tender. In return for cancelling the tender notice the Committee promised to pay the DDC a tax per fish harvested, and to maintain the lake environment by regular clean-ups and annual re-stocking of fingerlings. Collective action helped the Jalaris retain their exclusive rights and continue making a living from the lake. All three big lakes in the Pokhara area now have Fishers' Enterprise Committees and these have federated to form a higher-level Kaski District Fish Growers Association, which has been registered at the Kaski District Administration Office. The Association has an advisory board with a representative from the District Agriculture Development Office and the FRC. The Association's constitution does not specify a fixed gender quota for the executive committee, but it currently has two elected women members. n The opportunity cost of group membership and activities remains high for Dalits and other poor and excluded groups whose daily livelihood strugglesleaveverylittletimeorenergyforsuchactivities. Asnotedearlier, the MESI study found that better off villagers were seven times more likely to belong to a group than poor villagers. n Often homogenous groups ­ in terms of gender,caste and ethnicity ­ are best suited for serving the interests of disadvantaged groups. Clearly, "elite capture" is more likely in mixed groups. The full potential of local organizations to empower the poor and socially marginalised thus remains to be fully realised. Not enough attention has been given to the governance structure of groups, especially in terms of building in checks and mechanisms to prevent elite capture and to ensure wide representation. 90 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal BOX 15 Mindset needed for effective affirmative action Within the bureaucracy, affirmative action is a long, hard process that only begins once the policy and institutional frameworks are in place. Affirmative action is both a political and an organizational change problem. Political action imposing an affirmative action programme and a set of targets without an organizational change process will result in a policy, but little action. Organizational change without political involvement maintains the interests of privileged groups and uncouples affirmative action from the larger question of social exclusion. The opposition to reservations within the bureaucracy is pervasive and high. Those who genuinely believe that reservations are good for the civil service are in a distinct minority and even they are more likely to support the greater inclusion of women in the bureaucracy than increased representation by either Janajatis or Dalits. For Janajati and Dalit groups, the concessions that are being hotly debated among those in power are irrelevant ­ far less than either what they expect or what they are demanding. Change will require a complex mixture of political and senior management commitments, the communication of that commitment, and pressure for results, as well as negotiation and dialogue. It will require support for learning and capacity building ­ both on the part of the new entrants and on the part of those already in the system (who may need to be sensitised to the new realities). It will also need to put into place human resources systems such as recruitment, testing, hiring, training, mentoring, and promotions and cultural change. The process of change is dynamic and iterative and needs continuous fine-tuning. Implementing affirmative action sets into motion various parts of the organization and as each part changes in response to events, the cycle is repeated and each is able to secure greater resources, information and legitimacy. At the end of the day, affirmative action must benefit those who most need help. Paraphrasing Gandhi, we have to ask if the steps we are contemplating will restore to them control over their life and destiny. Selected case studies show that the livelihood aspects of develop- ment can be addressed through group-based programmes when they are appropriatelyimplemented. (SeeBox14.) Theincomeofdisadvantagedpopu- lations can be raised by fostering savings-based microfinance organizations and through organizations that promote new employment and economic activities. However, implementing such activities is problematic because programmes that can offer the kind of comprehensive technical support needed are rare. The quiet revolution underway in Nepal is the expansion of the im- petus for group-based collective action from the village level to district, na- tional (and sometimes international) arenas through group-based federations and associations. These movements seek new platforms and aim to influence policyat higher levels. Groups, therefore, have the potential to support mem- ber empowerment by forging and practicing new, more egalitarian rules of the game for social and economic interactions ­ and by putting pressure on the existing power structure to do the same. I N C L U S I V E G O V E R N A N C E 91 Affirmative action Affirmative action seeks to correct historical disadvantages and unfair dis- crimination by enabling access to full opportunity and benefits to groups that have been excluded. Overcoming the legacy of past inequality involves more than allotting some reserved seats in elected or administrative government, or in university admissions, etc. To be effective, affirmative action must be based on a holistic approach that addresses not only formal governance struc- tures and electoral systems but also informal and entrenched behaviours, In order to be truly attitudes and networks of preference and patronage that are all part of the effective and sustainable, affirmative existing system. action requires a Based on the assumption that people are the same and that they broader social and political committment start from the same point, many well-intended government policies practice to equality and human rights, as articulated by formal equality in their treatment of citizens. In contrast, governments that the Constitution, laws practice substantive equality recognise that treating everyone equally with- and policies. out recognising the legacy of histori- cal discrimination ends up perpetuat- ing inequality. So substantive equal- ityapproachestoaffirmativeactionfo- cus on remedying past discrimination. This is naturally contentious because itisessentiallyaboutshiftingpowerre- lations and the space for even mar- ginal change is highly contested. Affirmative action does not necessarily overrule the "merit" prin- ciple ­ though this has often been an issue raised by those who resist change ­ as long as the basic qualifi- cationsneededforperformingspecific functions remain as essential criteria. However, those who define merit of- ten represent a select and privileged minority rather than the broad citi- zenry. Contemporary approaches to affirmative action in both the private and public sectors are based on the management discovery that a diverse workforce is a better workforce. For 92 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal the private sector, this means that the firm will be better able to understand and respond to the needs of a diverse customer base. For the civil service it means that the bureaucracy will be more representative of, more responsive to, and hence seen as more legitimate by a diverse citizenry. Affirmative action as currently debated in Nepal relates not only to the civil service, but also to elected government and to the education, em- ployment and health sectors. In many countries affirmative action also en- compasses changes in the electoral system in ways that ensure proportional representation of different groups. This may be part of the answer in Nepal as well. (See Box 16.) Affirmative action and education are helping to The earlier discussion on human development and political poverty level the playing field documented the disparities in the health and education levels of women, and bring greater prosperity to all groups. Janajatis and especially Dalits compared to other groups,as well as their low levels of representation in the nation's gover- nance institutions. The government's current views on affirmative action are not clear,how- ever. The need for some sort of affirmative ac- tion for these groups is not disputed but the modality has been the source of some conten- tion ­ as well as the issue of which groups should be included. In December 2004, a High Level Res- ervations Committee was established under the chairmanship of the then Finance Minis- ter with the mandate to prepare a report with recommendations for affirmative action mea- sures for women, Dalits and Janajatis. At that time MOGA was also preparing a "road map" for affirmative action in the civil service, and one formula for reservations circulating in the halls of the bureaucracy was 20 percent for women; 10 percent for Janajatis and five per- cent for Dalits or 35 percent overall. A change of government caused the High Level Reser- vation Committee to be disbanded before it could present its report and no follow up has taken place. The government's hesitation to rec- ommendspecificquotasforreservationsinthe civil service stems from the concern that such I N C L U S I V E G O V E R N A N C E 93 a move might undermine the civil service's reputation as a meritocracy. How- ever, the second amendment of the Civil Service Act passed by the cabinet in July 2005, while it avoids setting specific percentages of reserved posts, does (for a period of five years) permit the government to "recruit candidates from among women,Dalits and Janajatis and disabled people by organizing sepa- rate open-competitive examinations for a stated number of positions." MOGA is currently developing the details of the affirmative action process to be fol- lowed for the next civil service recruitment in 2006 and is also putting in place a longer term affirmative action plan that goes beyond quotas to lay out a more comprehensive change in the management process that will not only increase the number of women, Dalits and Janajatis in the civil service, but will also lead to a greater diversity of skills and perspectives,with the ultimate goal of staffing a civil service that is better able to represent and respond to Nepal's diverse citizenry. One of the biggest challenges of the affirmative action agenda is the low number of qualified candidates in certain groups such as the Dalits,who as noted earlier, make up less than one percent of those holding BA or higher degrees. An effective "road map" to affirmative action needs to address this dimension ­ perhaps through a special programme to develop a "pipeline" BOX 16 Alternative approaches to proportional representation Lawoti (2005) has proposed two possible proportional representation ( PR) systems geared to guarantee the representation of smaller parties ­ including those based on ethnic identities ­ in the House of Representatives (HOR), based on the popular votes they receive in elections. One system would be based on a simple PR system in which parties announce their national candidate lists and receive an appropriate number of seats in the HOR according to their share of the popular vote. Another system would be a mixed system in which half of the HOR is filled through the "first past the post" method and the remaining half through the PR method. Lawoti believes that either of these options would have given the smaller parties more seats in the HOR than they received after the 1999 national elections. For that reason alone, he argues that the PR electoral system is more inclusive and desirable for a multi-ethnic society like Nepal. Political scientist Krishna Khanal (2004b) has offered another model for adopting the PR electoral system, one that would result in a legislature even more representative of the Nepali population in terms of ethnicity/caste than Lawoti's proposed models. Khanal argues that the electoral appeal of smaller parties, including the ethnically based ones, is weak. His model is thus geared more toward representing social- cultural-ethnic formations in proportion to their shares in the total population. He proposes a 14- constituency system based on Nepal's existing districts. Both Khanal and Lawoti caution that the PR system is likely to result in a HOR where no political party will have a majority and hence the Nepali electorate will have to be prepared to face a culture of coalition governments. See Lawoti (2005) and Khanal (2004a; 2004b) for further details of the respective PR electoral models they advocate. 94 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal of qualified candidates by offering scholarships for 10 plus 2 and university education to girls, Dalits and disadvantaged Janajati students with the top SLC scores in the public school system in each development region or,if possible, in each dis- trict. Fostering genuine diversity in Nepal's civil service will require a complex mixture of political and senior management commitment, communication of that commitment, and pres- sure for results,as well as negotiation and dialogue. It will also requiresupportforlearningandcapacitybuildingofbothnew entrants and those already in the system. Affirmative action in the education and health sec- tors has ­ at least formally ­ been built into the primary edu- cation SWAp and the Health Sector SWAp through the mecha- nisms mandated in theVulnerable Community Development Plans (VCDP) for each of these national programmes. In addi- tion, criteria for access to scholarships under a newWork Study Programme in higher education now consider gender, caste and ethnicity in addition to economic need for eligibility for With a few exceptions (such as that shown the subsidy portion of the programme. This should help increase the pool of above) men from the qualified women, Dalits and disadvantaged Janajatis. Brahman/Chhetri group have dominated Within the bureaucracy, affirmative action is a long, hard process cabinet appoinments and their presence in which only begins once the policy and institutional frameworks are in place. the civil service has Affirmative action is both a political and an organizational change problem. increased from 70 to 90 percent between 1985 Political action imposing an affirmative action programme and a set of tar- and 2002. gets without an organizational change process will result in a policy but little action. Organizational change without political involvement maintains the interests of privileged groups and uncouples affirmative action from the larger question of social exclusion Probably the most contentious sphere for affirmative action is in elected government. In the existing system, in addition to the mandatory inclusion of a certain proportion of women in various tiers of local govern- ment, the political parties are also required to put up women candidates for at least five percent of the constituencies they contest. As we saw in the sec- tion on "political poverty" however, none of these provisions seems to have brought much change in this male controlled arena. This is one area where the political parties have failed. The internal power structures of main politi- cal parties have never been very representative in terms of gender, caste or ethnicity of the diverse citizens they claim to represent. Women have made I N C L U S I V E G O V E R N A N C E 95 up less than 10 percent of the central committee membership of the three main parties and, while the RPP includes about 25 percent Janajatis in its central leadership, the two major parties (the Nepali Congress and UML) had only ten and three percent respectively ­ even though Janajatis represent over a third of Nepal's population. None of the parties has had a single Dalit on their Central Committees. The lack of women and other excluded minorities in the leadership of the major political parties persists even though all the political party mani- festos commit them to promoting gender,caste and ethnic equality. For years Nepal's politicians have been able to say one thing and do something else with apparent impunity. A functioning democracy requires credible political parties. One urgent step the parties need to take to restore their credibility Even though the BCN and regain their rightful place as the legitimate leaders of democratic Nepal group ranks highest in all well-being indicators, it is is internal reform to bring greater transparency,accountability and inclusive- so large that one must ness to their own organizations. As long as the mindset of the party leaders keep in mind that there are many poor Brahmans and and the internal power dynamics of their organizations continue to be struc- Chhetris in rural Nepal. tured on the basis of caste, ethnicity, gender and age ­ hierarchies left over from feudal times ­ the parties will lack the legitimacy they need to guide Nepal out of its current governance crisis. One of the major parties appears to have begun the pro- cess of internal reform by setting aside positions for women, Dalits, Janajatis and Madhesi candi- dates as well as for a candidate from the long ne- glected Karnali zone in its January 2006 elections to the CentralWorking Committee (CWC). If,when democracy is restored, these new CWC members are able to effectively speak out for the groups they represent, perhaps more attention will be paid to actually implementing some of the long- standing promises of social inclusion in the party manifestos. The parties, however, have not been unique in their failure to match their words with action or to align their informal behaviour with formally espoused policies regarding caste, eth- nic and gender discrimination. The point is, that like all policy reforms,affirmative action as a lever for social inclusion is necessary but not sufficient 96 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal to bring about significant and sustainable positive outcomes for socially ex- cluded groups. In order to be truly effective and sustainable, affirmative ac- tion requires broad social and political commitment to equality and human rights, articulated by the Constitution,laws and policies. In other words, trans- lating the formal commitments into reality will require a change in the inter- nal values and behaviour of Nepali citizens. I N C L U S I V E G O V E R N A N C E 97 C H A P T E R 7 SUMMARY AND PRIORITIES FOR ACTION VII SUMMARY AND PRIORITIES FOR ACTION Key action points S U M M A R Y A N D P R I O R I T I E S F O R A C T I O N 99 After centuries of thinking about themselves as subjects of feudal rulers, more and more Nepalis are now beginning to see themselves as citizens of a demo- cratic state. Although the pace of this fundamental change in self-perception is uneven among groups at different levels on the social hierarchy, it is now being embraced even by those traditionally at the lowest echelons ­ espe- cially women, Dalits and Janajatis. This change in self-perception has also altered expectations:people do not want to plead for favours from the pow- erful. Instead of patronage, they want rights ­ the same rights that are ac- corded every citizen by law. All segments of society want to be included and they want uniform"rules of the game"to apply to all social players across the board. Social inclusion and empowerment are the interrelated processes that can bring this about. Exclusion is one of the The GSEA study provides insights into the various dimensions of factors behind the current conflict.Lack of voice, social exclusion in Nepal. Overall,the main findings of this study are: political representation n Democracy has ushered in numerous new organizations and has created and empowerment are as important dimensions of the space needed for debate and freedom of expression. This in turn has poverty as the economic and human development led to the emergence of genuine social movements in response to dimensions. exclusion and the obvious inconsistency between exclusion and democracy. n Society has progressed from feudal patronage through a period of state-dispensed welfare to an era where rights are the legitimate basis of citizens' demands and the state's responses. The shift from subjects to citizens has progressed but remains incomplete. n Labels for many diverse identities have changed. Groups that had been traditionally excluded are reasserting their identities or constructing new identities in an attempt to reflect a rediscovered pride in being part of their own group. The discourse is shifting from sano jat to Dalit, from tribal or matwali to Adivasi Janajati. Women are redefining themselves inthe Shakti dimensiontobalancethepreviousdocile SitaandSavitriimagesoftheunquestioning, obedient and dependent wife. n Mainstreaming: excludedgroupsarenotcontentwith piecemeal schemes that allocate small amounts of public funds exclusively for them as special interest groups. Instead,they want the structural barriers that 100 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal have kept some groups from gaining full access to mainstream programmes and services systematically diagnosed and removed. They want specific mechanisms (governance rules, incentive regimes and monitoring systems) in place to help overcome these barriers. n There is widespread recognition that in order for democracy to function properly the political parties need internal reform. They need to lead by example ­ restructuring themselves along lines that are more democratic andinclusive. Thewomen, DalitandJanajatiwingswithinpoliticalparties, with their deep grassroots links, have the potential to become influential allies in the move towards social inclusion but first these important constituencies need to have greater voice and influence within their own parities. n The exclusions overlap. Gender, caste and ethnicity have cross-cutting dimensions; therefore inclusion efforts need to be advocated within the many different hierarchies, sectors and institutions that make up Nepali society. n Many of the social transitions that are brought about by inclusion and affirmativeactionareinitiallypainfulandunsettlingbecausetheythreaten the entrenched existing power structures ­ as well as some of the deeply held values and meaning systems through which individuals and groups define their very identity. n Nepali citizens (in government and civil society) who are pushing for reforms in support of social inclusion have already begun the process of re-defining themselves in terms that emphasise the egalitarian elements in their own tradition and blending these elements with generally accepted democratic norms. n There is a pattern of persistent gaps between promises made in periodic plans and policy statements and outcomes on the ground. This well known"implementation gap", frequently cited as the reason for project or policy failure,is not just lack of "capacity". It also reflects the fact that many in positions of power do not welcome change and continue to be able to call upon informal networks to thwart the intentions of officially sanctioned policy change. Elite resistance remains a continuous challenge,and the possibility of reversal of progress is a constant threat. Nevertheless, consensus is growing that in the long run full inclusion is crucial to broad-based poverty reduction and lasting peace. n The absence of parliament since 2002 and the resulting legislation block has retarded the legal and policy changes needed to eliminate exclusion. Among these is implementation of the framework set out in the LSGA. S U M M A R Y A N D P R I O R I T I E S F O R A C T I O N 101 Decentralisation is central to the demand by many excluded groups for a greater voice in local governance. An effective decentralisation process can set off inclusion from the grassroots which, when matched by appropriate policy responses from the top, can create an environment where the basis for settling differences is through dialogue and negotiation rather than intimidation and violence. Decentralisation is thus an important foundation for lasting peace. n Disaggregated data and analysis along the lines of gender and caste/ ethnicity is essential to change. Only by consistently and accurately tracking exclusion can accountability and incentives for its elimination be created. Reliable data are especially important for inclusive budgeting ­ a key tool to eradicating exclusion. Many recommendations have been made throughout this summary of the Gender and Social Exclusion Assessment. Some are implied; others are more directly stated; some are directed at HMG/N, while NGOs, academicians and other members of civil society or donors can implement others. Some can be acted on immediately to produce results quickly while others, that involve deep structural and cognitive changes, will have to be implemented over the long term and are aimed at producing fundamental societal changes. The following chart presents twelve points that, from our many consultations, the GSEA team considers to be the most important policy actions to be un- dertaken. It lists the action,the justification for that action,the lead (in upper case), and the supporting actors who need to take responsibility for imple- mentation (in lower case). 102 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal KEY ACTION POINTS RESPONSIBLE ACTION BASIS FOR ACTION ACTORS * Incorporate an inclusion lens into the government To date the government has only proposed piecemeal 1. n GOVERNMENT planning, budget allocation and monitoring solutions to the problem of social inclusion in the form ndonors process to ensure full access for women, Dalits of "targeted programmes". Currently funds earmarked and Janajatis in all core government services and under the inclusion pillar of the PRSP amount to just development programmes. This approach to public over 6% of the budget. This is clearly inadequate to expenditure would expand the concept of "gender create a level playing field for excluded groups and to budgeting/auditing" that has been increasingly adopted meet the PRSP inclusion goals. In order to be by HMG/N. effective, inclusion needs to be adequately funded and fully embraced by all government programmes. Steps towards inclusive budgeting would entail: n Conducting a systematic analysis of all mainstream Structural change towards social inclusion as programmes to identify barriers to access for envisaged by the PRSP is possible only with 1) women, Dalits and Janajatis; increased, focused investment by both government n Developing specific mechanisms and incentives to and donors to assure inclusion across all core services overcome the barriers; and development programmes and 2) systematic n Assigning clear accountability for achieving the monitoring of results. inclusion objectives in all sectors; n Developing clear outcome indicators disaggregated by caste, ethnicity and gender; and n Tracking indicators in real time sectoral monitoring and evaluation systems linked to the PMAS to ensure effective corrective policy actions. 2. Make organizational changes for effective The government's efforts at inclusion have not been n GOVERNMENT implementation of the inclusion pillar: translated into coordinated action at the sectoral n donors n Establish a national inclusion task force in the ministry level, where both formal and informal barriers n civil society National Planning Commission to coordinate and still seem to be entrenched. Effective coordination of monitor inclusion initiatives by government policies and actions under the inclusion pillar can ministries, with appropriate linkages to the central greatly increase impact, reduce duplication and lead to PRSP monitoring system. greater impact at all levels. n This task force could lead a review of inclusion results in the 10th Plan and propose revisions for The concept of inclusive programming and budgeting the 11th Plan. recommended in (1) above necessitates that trained n Empower the Gender Focal Points in all line professionals in each major sectoral ministry examine ministries by making them part of a sectoral social the ministry's major policies and programmes for the inclusion unit responsible for vetting all ministry impact that they are likely to have on excluded groups. programmes and policies from an inclusion Furthermore, to fully ensure inclusive programming, perspective. A senior government official empowered each ministry needs to recommend specific mechanisms to hire experts on gender, Dalit and Janajati issues to ensure that its sectoral policies are inclusive and that should head this unit. they deliver equal benefits to excluded groups. * Actors depicted in upper case are to take the lead; actors depicted in lower case are to support. S U M M A R Y A N D P R I O R I T I E S F O R A C T I O N 103 ACTION BASIS FOR ACTION RESPONSIBLE ACTORS 2. n Build on the strength of the existing district-level Commitments to ensure that DDC and VDC budget Contd. Women Development Offices to establish District allocations and programmes are responsive to women, Gender and Social Inclusion Offices and link Dalits and Janajatis through a "watchdog" committee them with the sectoral ministries and representative have not been implemented. One possibility that has national organizations - as well as with been suggested by many groups is to enlarge the decentralised representatives of the National mandate of the WDO to encompass all dimensions of Women's Commission (NWC), the National Dalit social inclusion by adding staff whose responsibility it Commission (NDC) and the National Foundation would be to ensure that Dalit and Janajati groups also for the Development of Indigenous Nationalities benefit from local government spending. This office (NFDIN) to ensure coordination. would have increased accountability to the various national commissions (such as NWC, NDC and NFDIN) and to the DDCs and VDCs. As more and more resources are devolved to the elected DDC and VDC governments they would be expected to allocate matching funds to the work of the District Gender and Social Inclusion Offices. Both the increased responsibility of this office and the devolution of funding would help lay the foundation for the realisation of the LSGA's commitment to social inclusion. 3. n GOVERNMENT Improve the governance structure of the national The commissions set up under an executive order do n donors commissions for women and Dalits: not have the legal authority to function independently n civil society n Re-establish the National Women's Commission of government and political influence. Legal and the National Dalit Commission through recognition and autonomy would enable them to legislation. function effectively and independently, using n Enable the commissions to function as semi- professional help where needed. autonomous constitutional bodies, with authority to receive a regular budget directly from the MOF and support from donors. n Ensure that these commissions are aware of the changing situation on the ground for excluded groups by encouraging them to have a "listening relationship" with civil society organizations as well as with the proposed gender and social inclusion units at the district level (see recommendation 2 above). 104 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal RESPONSIBLE ACTION BASIS FOR ACTION ACTORS 4. Revive the stalled decentralisation process with While decentralisation on its own does not guarantee n GOVERNMENT safeguards and incentives to promote inclusion at all social inclusion, it can provide more inclusive and n donors levels. accountable governance by delegating funds and n civil societ decision-making authority closer to the local level, where ordinary people are more likely to be able to take part in and influence decisions ­ and monitor outcomes. In particular, it provides a possible governance framework within which diverse ethnic and language groups can have greater autonomy in certain key areas and still remain citizens of a unified Nepali nation. Nepal has the necessary framework for decentralisation in the LSGA, which even includes a number of provisions to ensure voice for women, Dalits and Janajatis (e.g. through the district "Watchdog Committees" and other provisions). But implementa- tion on overall decentralisation ­ and on the provisions to promote inclusion ­ has been slow for lack of political commitment. Decentralisation has particularly suffered after July 2002, when the government allowed the terms of elected local governments to lapse. 5. Enact critical legal changes to ensure equal rights The definition of who has the right to citizenship was n GOVERNMENT for all citizens ­ and equal access to citizenship: greatly curtailed in the 1990 Constitution, and those ndonors most notably affected are the most excluded. ncivil society On Citizenship n Reinstate the language of the Interim Constitution of Existing informal practices effectively deny citizenship 1953 that guarantees citizenship to "every person to many segments of excluded society, especially to who had been permanently residing within the those who do not own land and to those whose territory of Nepal with their family". Carry out a language and social customs mark them as being social audit of informal government practices and historically "of Indian origin". requirements for obtaining citizenship papers. For women: n Ensure equal citizenship rights to women and An analysis of Nepali laws, including the Constitution permit them to transmit citizenship to their children and Country Code, conducted for the GSEA found: and their spouses. n 83 pieces of legislation that discriminate against n Ensure equal rights to ancestral property for women women and and married daughters. n Repeal the provision that allows a man to enter a bigamous marriage under certain conditions. S U M M A R Y A N D P R I O R I T I E S F O R A C T I O N 105 RESPONSIBLE ACTION BASIS FOR ACTION ACTORS 5. For Janajatis: Contd. n Remove the word "Hindu" from Article 4 of the n 32 provisions that discriminate on the basis of Constitution. religion, caste and ethnicity. n Amend Article 19 (1) to permit the right to religion. n Amend Article 6 (1) to permit alternate official languages in addition to Nepali. For Dalits: Allowing discrimination as a social custom reinforces n Remove the ambiguity about the right to practice the traditional institutions based on inequality. untouchability/caste-based discrimination as a Increasing punishment for caste-based discrimination social custom. can serve as an effective deterrent, but long-term n Ensure enforcement of punishment for caste-based social change can come only through changes in discrimination in the public and private spheres. values and practices. 6. Enhance Poverty Monitoring and Analysis System Previously, the large number of social groups (103) n GOVERNMENT (PMAS) by standardising social categories and covered by the Census made meaningful analysis of n donors improving monitoring capacity of key sectoral outcomes by social groups difficult and limited the n civil society ministries to track social inclusion: extent to which HMG/N could track progress in poverty n At the national level, a common classification of reduction along social dimensions. In NLSS I (1995/6) the main social groups has been used by the GSEA more than 20 percent of the population belonging to and the CBS for data collection and analysis in the smaller groups remained unaccounted for in the "other" NLSS II and other national surveys that contribute to category. The adoption by the Central Bureau of the PMAS. There are six major social categories Statistics of the broad categories developed by the (BC, Tarai Middle Castes, Dalits, Newars, Janajatis GSEA for NLSS II has overcome this limitation and and Muslims) but when separated by Hill and Tarai paved the way for better PMAS tracking of progress on there are the following 10 categories: the social inclusion pillar of the PRSP through national 1. Hill Brahman/Chhetris Census and Survey data. The "other" category in the 2. Tarai Brahman/Chhetris 2003/4 NLSS II now accounts for only about one 3. Tarai Middle Castes percent of the population. 4. Hill Dalits 5. Tarai Dalits 6. Newars 7. Hill Janajatis 8. Tarai Janajatis 9. Religious Minorities (Muslim) 10. Other 106 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal ACTION RESPONSIBLE BASIS FOR ACTION ACTORS 6. At sectoral level the PMAS tracks performance Despite the progress made for national datasets, the Contd. n n GOVERNMENT using the Management Information Systems of the existing monitoring and information systems of the ndonors various sectoral ministries. Sectoral performance various sectoral ministries still do not permit the ncivil society data also need to be disaggregated by gender, government to track progress on the social inclusion caste and ethnicity for all monitoring purposes ­ pillar. MOES Flash Reports have made a start at this, preferably using the 10 categories listed above. but are still not reliable. Disaggregated data are essential to learning about which policies and n When it is not possible to acquire such detailed data programmes work to improve inclusion and which do for sectoral monitoring, then the interested parties not. It is also an important element in HMG/N's move should be encouraged to at least adopt four main towards results-based budgeting and part of the categories into which all caste/ethnic groups could budget release conditions for the pooled donor support be sorted ­ plus gender, since women are to the government's health and education disadvantaged across all groups. The four programmes. categories would be: 1. Dalits There are large differences even among Janajati and 2. Disadvantaged Janajatis, including those Tarai Middle Caste groups that could mean that the Janajati groups who fall a set percentage (to most disadvantaged might not be reached by be determined) below the national average on supportive programming. To ensure transparency and consumption poverty, health and education accountability as well as effective targeting, the NPC's indicators based on the Census, NLSS and Poverty Monitoring Unit needs to lead an exercise DHS. where the NLSS and other national data sets can be 3. Other excluded/disadvantaged groups, used to accurately identify the truely disadvantaged on including Muslims and certainTarai Middle a scientific bases. This effort would need to involve Caste groups based on the data mentioned representative women, Dalit and Janajati groups (such above. as the Women's Commission, the Dalit Commission, 4. Non-excluded groups such as Brahmans the Dalit NGO Federation, the National Foundation for and Chhetris, Newars, Thakalis, Gurungs and Development of Indigenous Nationalities and the those Tarai Middle Caste groups whose National Federation for Indigenous Nationalities) along poverty indicators are a certain percentage (to with CBS. be determined) above the national average. n Identify the disadvantaged: Effective annual monitoring of inclusion outcomes in l Ensure that this categorisation is done on a each sector is a potentially powerful tool in support of scientific basis and ask the NPC Poverty affirmative action in the areas of health, education and Monitoring Unit to lead it with support from CBS other critical services. (and participation from NFDIN and other concerned groups). It would be based on statistical analysis of NLSS, DHS and Census data to identify the truly disadvantaged among the Janajati and other groups. S U M M A R Y A N D P R I O R I T I E S F O R A C T I O N 107 RESPONSIBLE ACTION BASIS FOR ACTION ACTORS 6. l Develop a system to periodically update the Contd. status of different groups as new data become available in order to help keep targeted programmes and affirmative action policies from becoming identity-based entitlements and to ensure that government resources go to those most in need. n Continue income-based tracking and targeting to ensure that the poor within the privileged caste/ ethnic groups are not missed. n Include social accountability mechanisms in the sectoral monitoring processes to create incentives for inclusion. 7. Develop a holistic strategy for reservation and A more diverse civil service can improve service n GOVERNMENT affirmative action: delivery based on better understanding of the needs n donors and perspectives of diverse clients. Similarly, greater n civil society n Appoint a broad-based task force to develop a road representation of excluded groups in elected map for increasing diversity and representation of government at all levels will increase the legitimacy disadvantaged groups in politics, civil society and and accountability of Nepali democracy. academia. Nepali society and government now accept n Build a pipeline of qualified women, Dalits and reservation/affirmative action as a means to level Janajatis by establishing a fast-track scholarship/ the playing field and increase diversity. However, internship programme for the most promising girls, modalities to achieve this objective have not been Dalits and Janajatis completing school level finalised, despite efforts to do so. education in the public system. This will help ensure that "meritocracy" is not It is important to ensure there is a "pipeline" of compromised while reserving positions for women, qualified candidates from under represented groups Dalit and Janajatis candidates in the civil service. who can compete for reserved positions. n Explore alternative electoral systems as part of the Full implementation of decentralisation can also be affirmative action policy to help ensure greater an important mechanism for affirmative action. representation and voice for Nepal's diverse groups. 108 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal RESPONSIBLE ACTION BASIS FOR ACTION ACTORS 8. Make donor agencies and NGOs more inclusive: Recent reports have shown that donors and NGOs n DONORS n Both sets of actors should undertake inclusion have alarmingly poor inclusion levels of women, n CIVIL SOCIETY audits and inclusion reviews of their organizations Dalits and Janajatis and that current donor programming still tends to be largely based on and portfolios to identify exclusion and make information from the traditional "elite" sources. corrections. Greater internal diversity could help deliver more n Donors should require the NGOs they support to effective programmes. conduct similar audits and share findings with government. n Donor agencies should be encouraged to expand their "circle" of contacts and deepen their under- standing of inclusion by seeking information from and interacting with diverse groups that ordinarily do not have access to donor ears. 9. Strengthen the power of local development groups: A level playing field within local development groups n GOVERNMENT is necessary to ensure that members from n CIVIL SOCIETY n Establish governance rules for local develop- disadvantaged backgrounds benefit equally from n DONORS ment groups to help them better deliver inclusion shared group activities and that group-based and prevent elite capture by implementing effective approaches live up to their potential for delivering governance rules, transparent monitoring and inclusion, sustainability and empowerment for all. evaluation mechanisms. Many local groups that form federations to increase their economic efficiency and their political voice n Create a supportive environment for federations of face bureaucratic delays and barriers when they local level groups: seek to register either as a cooperative under the l Review and amend existing cooperative laws to Cooperative Laws or as an NGO under the CDO or permit more than one of a given type of the SWC. One such barrier for cooperative cooperative to be registered in a single district federations is the outmoded rule that there can be and to remove other regulations and procedural only one of any particular type of cooperative (e.g. barriers to the registration and operation of dairy, credit, multipurpose cooperative) per district. This is particularly counterproductive for women's cooperative federations. groups whose members practice a variety of l Encourage wide consultations, between the livelihoods and need the multipurpose designation to government and NGO and donor stakeholders, allow this. Another emerging barrier for any aimed at revising the new NGO code to make it federated groups seeking to become an NGO is the less restrictive. restrictive new NGO code. S U M M A R Y A N D P R I O R I T I E S F O R A C T I O N 109 ACTION BASIS FOR ACTION RESPONSIBLE ACTORS 10. Develop a knowledge base to inform policy debate Practical proposals for inclusion that are n CIVIL SOCIETY on inclusion: economically and politically feasible require n DONORS Encourage civil society groups to consistently conceptual clarity, and collective thinking and n government generate and share knowledge and understanding on debate at all levels. Thinking through policy diversity and related issues, by encouraging choices and developing the mechanisms through participation from members of traditionally excluded which these policies will be implemented on the groups. ground requires representative participation. 11. Support strategic coalitions between women, The women, Dalit and Janajati movements are n CIVIL SOCIETY Dalits and Janajatis: currently fractured and almost independent of each Build alliances for collective equal citizenship goals other, even though they often seek to attain similar between the individual social movements. citizenship goals. There is strength in numbers, and alliances can help them forcefully advocate and achieve their collective goals. 12. Encourage internal reform of the main political The success of democracy hinges largely on the n CIVIL SOCIETY parties to make them more democratic and extent to which the political parties, once in power, n government broadly representative: can articulate and respond to the demands of n donors The parties need to be encouraged to adopt more every segment of society. In Nepal's emerging democratic and transparent procedures and to be held democracy the parties have been less than accountable to implement the many inclusive promises successful in implementing democratic norms and made in their manifestos. procedures within the context of their own organizations. They have also delayed implemen- tation of their own pledges to be inclusive to women, Janajatis and Dalits. 110 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal SELECTED BIBLIOGRAPHY Acharya, Meena. 2004. Analysis of Caste, Ethnicity and Gender Data from 2001 Population Census in Preparation for Poverty Mapping and Wider PRSP Monitoring, World Bank/DFID, Kathmandu, 6 July 2004 Acharya, Meena and Lynn Bennett. 1981. The Rural Women of Nepal: An Aggregate Analysis and Summary of Eight Village Studies. Kathmandu: Centre for Economic Development and Administration Bajracharya, B.B. 2005. Technical Review of School Education in Nepal. MoES/ESAT Bhattachan, Krishna B., Tek Tamrakar, Yam Bahadur Kisan, Ratna Bahadur Bagchand, Pavitra Sunar, Sadananda Paswan, Bishnu Pathak and Chandrakala Sonar. 2004. Dalits' Empowerment and Inclusion for Integration in the National Mainstream: Strategy for Influencing Policy and Institutions. Kathmandu: Dalit Empowerment and Inclusion Project Bennett, Lynn, Kishor Gajurel et al. 2006. "Negotiating Social Change: Gender Caste and Ethnic Dimensions of Empowerment and Social Inclusion in Rural Nepal" in Empowerment in Practice from Analysis to Implementation. Ruth Alsop, Mette Bertelsen, and Jeremy Holland, eds. Washington D.C.: The World Bank Bhatta, Pramod. 2005. "Decentralisation of Primary Education in Nepal", Studies in Nepali History and Society 10(1):3-38 June 2005 Biggs, Stephen D., Sumitra M. Gurung and Don Messerschmidt. 2004a. An Exploratory Study of Gender, Social Inclusion and Empowerment in Development Groups and Group-Based Organizations in Nepal: Working from the Positive. Report submitted to the Gender and Social Exclusion Assessment (GSEA) Study, National Planning Commission, World Bank and DFID, Kathmandu (Version 2, November 2004) ­­­­. 2004b. An Exploratory Study of Gender, Social Inclusion and Empowerment through Development Groups in Nepal: Building on the Positive. Report submitted to the Gender and Social Exclusion Assessment (GSEA) Study, National Planning Commission, World Bank and DFID, Kathmandu (Version 1, August 2004) CBS. 2004. Nepal National Living Standards Survey-II (NLSS-II). CBS, HMG/N ­­­­. 2003. NLSS/CBS 2003/04. Population Monograph of Nepal. CBS, HMG/N Clapham, Susan et al. 2005. Increasing Access to Emergency Obstetric Care: Addressing the Attitudes of Service Providers, forthcoming Dahal, Dilli Ram, Bidhan Acharya, Yogendra Bahadur Gurung, Kamala Hemchuri and Dharma Swarnakar. 2002. National Dalit Strategy Report, Part 1: Situation Analysis of Dalits in Nepal. Kathmandu: National Planning Commission S E L E C T E D B I B L I O G R A P H Y 111 DFID/Department of Health Services. 2004. Increasing Access to Essential Obstetric Care: A Review of Progress and Process. Family Health Division, Department of Health Services. Options. DFID. October 2004 Gurung, Harka, Yogendra Gurung and Chhabi Lal Chidi. 2005. Nepal Atlas of Ethnic and Caste Groups. NFDIN, Lalitpur, Nepal Gurung, Jeanette D. ed. 1999. Searching for Women's Voices in the Hindu Kush-Himalayas. Kathmandu: ICIMOD HMG/N. 1995. Dhanapati Upadhyaya Commission ­­­­. 1991, 1994, 1999 Election Commission Reports ­­­­. Family Health Division, Department of Health Services, MoH 1998. Maternal Mortality and Morbidity Study ­­­­. Ministry of Local Development. 1997. Definition of Dalit Castes HMG/N, UNICEF and Columbia University. 2000. Needs Assessment on the Availability of Emergency Obstetric Care Services in Panthar, Saptari, Dang and Kapilvastu District of Nepal IIDS. 2002. Adivasis/Janajatis in National Development: Major Issues, Constraints and Opportunities. Plan of Action Proposed for the Tenth Plan 2003­2007, Nov. 2002. Kathmandu: IIDS Jha, Haribansh. 2004. "Dalits and Their Participation in National and Local Governance." Presented at the seminar on Raising Dalit Participation in Government. Kathmandu: Centre for Economic and Technical Studies and Friedrich Ebert Stiftung Khanal, Krishna. 2004a. Rajyako Punarsamracana: Eak Prastav. Contemporary Paper Series, No. 9. Kathmandu: Nepal Centre for Contemporary Studies ­­­­. 2004b. "Consideration on possible model of Proportional Representation for Nepal". Paper presented at a national workshop on Affirmative Action and the Electoral System in Nepal organized by SAGUN, ESP/DFID and IIEDA in Kathmandu, 26-27 September Lawoti, Mahendra. 2005. Towards a Democratic Nepal: Inclusive Political Institutions for a Multicultural Society. New Delhi: Sage ­­­­. 2002. "Defining Minorities in Nepal". Nepali Journal of Contemporary Studies 2(1): 20-52 Malla, Sapana and Sabin Shrestha. 2005. Legal Exclusion in Nepal. Report submitted to the Gender and Social Exclusion Assessment (GSEA) Study, National Planning Commission, World Bank and DFID, Kathmandu MOES. 2001. School Level Educational Statistics of Nepal. Kathmandu: MOES, HMG/N 112 U N E Q U A L C I T I Z E N S Gender,Caste and Ethnic Exclusion in Nepal MOH, Family Health Division, Department of Health Services and DFID. 2004. Increasing Access to Essential Obstetric Care: A Review of Progress and Process. October 2004. Kathmandu: MOH, HMG/N MOH. 2002. Health Sector Strategy: An Agenda for Change (Reform). August 2002. Kathmandu: MOH, HMG/N National Dalit Commission. 2003. Biswakarma, Deepak J., and Min Biswakarma, eds. 2060 v.s. Nepalma Dalit Samudayako Sthiti. Kathmandu: National Dalit Commission NDHS. 2001. Nepal Demographic and Health Survey 2001. Kathmandu: FHD, DoHS, MOH, HMG/N New Era. June 2005. Baseline Study on Community-Managed and Government-Managed Schools. Kathmandu: New Era Neupane, Govinda. 2000. Nepalko Jaatiya Prashna. Kathmandu: Centre for Development Studies Parajuli, Dilip. forthcoming. Ethnicity-based Poverty Mapping in Nepal: Preliminary estimates Rao, Vijendra and Michael Walton. 2004. "From Equality of Opportunity to Equality of Agency: Grounding a Cultural Lens for Public Policy in an Unequal World." in Culture and Public Action. A Cross Disciplinary Dialogue on Development Policy. Vijendra Rao and Michael Walton, eds. Palo Alto: Stanford University Press UNDP. 2004. Nepal Human Development Report (NHRD): Empowerment and Poverty Reduction ­­­­. 2001. Nepal Human Development Report (NHRD) Van de Walle, Dominique and Dileni Gunewardenar. 2001. Sources of inequality in Viet Nam. Journal of Development Economics. Vol. 65 (2001) 177-207 World Bank. 2005a. Resilience Amidst Conflict: An Assessment of Poverty in Nepal, 1995-96 and 2003-04. Washington, D.C.: Poverty Reduction and Economic Management Sector Unit, South Asia Region ­­­­. 2005b. Understanding the Access, Demand and Utilisation of Health Services by Rural Women in Nepal and their Constraints. Human Development Sector, South Asia Region ­­­­. 2002. Social Analysis Sourcebook: Incorporating Social Dimensions into Bank-supported Projects (working draft). Washington, D.C.: World Bank Social Development Department ­­­­. 2001. World Development Report 2000/2001: Attacking Poverty. OUP, New York: Oxford University Press Yadav, R.P. April 2005. Caste/Ethnic Representation in His Majesty's Government at Policy Level in Nepal. Unpublished manuscript DFID THE Department For International WORLD Development BANK TheWorld Bank DFID Nepal Nepal Office P.O.Box 106 P.O.Box 798 Kathmandu, Nepal Yak andYeti Hotel Tel.: 5542980 Complex Fax:5542979 Durbar Marg Kathmandu, Nepal Tel.: 4226792, 4226793 E-mail Fax:4225112 enquiry@dfid.gov.uk Websites www.worldbank.org.np, Website www.bishwabank.org.np www.dfid.gov.uk I S B N 9 9 9 4 6 - 8 9 0 - 0 - 2 9 7 9 9 9 9 4 6 8 9 0 0 1 >