THE WORLD BANK OPERATIONS EVALUATION DEPARTMENT Understanding Children's Work (UCW) Program Addressing Challenges of Globalization: An Independent Evaluation of the World Bank's Approach to Global Programs Case Study Manmohan Agarwal and Lauren Kelly Director-General, Operations Evaluation: Gregory K. Ingram Director: Ajay Chhibber Task Team Leader: Uma Lele 2004 This paper is available upon request from OED. The World Bank Washington, D.C. ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE IN EVALUATION The Operations Evaluation Department (OED) is an independent unit within the World Bank; it reports directly to the Bank's Board of Executive Directors. OED assesses what works, and what does not; how a borrower plans to run and maintain a project; and the lasting contribution of the Bank to a country's overall development. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of the Bank's work, and to provide accountability in the achievement of its objectives. It also improves Bank work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. OED Working Papers are an informal series to disseminate the findings of work in progress to encourage the exchange of ideas about development effectiveness through evaluation. The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Contact: Operations Evaluation Department Partnerships & Knowledge Programs (OEDPK) e-mail: eline@worldbank.org Telephone: 202-458-4497 Facsimile: 202-522-3125 http:/www.worldbank.org/oed Abbreviations and Acronyms CAS Country Assistance Strategy CRC Convention on the Rights of the Child DGF Development Grant Facility EAC Economically active children ESW Economic and sector work FY Fiscal year GDP Gross domestic product GP Global program GPG Global public good ILO International Labor Organization IPEC International Programme for the Elimination of Child Labour LDC Least developed country LSMS World Bank Living Standard Measurement Surveys MICS Multiple Indicator Cluster Surveys MDG Millennium Development Goals MVF Mammipuddi Ventakarangiya Foundation NGO Nongovernmental organization OED Operations Evaluation Department PRSP Poverty Reduction Strategy Paper SIMPOC ILO Statistical Information and Monitoring Programme on Child Labour MICS UNICEF Multiple Indicator Cluster Surveys SME Small and medium enterprise TA Technical assistance TOR Terms of reference UCW Understanding Children's Work program UNICEF United Nations Children's Fund WHO World Health Organization WTO World Trade Organization ii Table of Contents Acknowledgements .......................................................................................................... iv Preface.................................................................................................................................v List of Case Study Programs ......................................................................................... vii Executive Summary.......................................................................................................viii Genesis, Objectives and Activities....................................................................... viii Design and Implementation........................................................................................ix OED Findings............................................................................................................x 1. Introduction and Context: Global Challenges in the Sector....................................1 Global Challenges Facing the Sector and Program................................................1 An International Consensus for Collective Action...................................................3 2. Program Alignment With Global Challenges and Bank Practices ...........................4 Origin of UCW.........................................................................................................4 Overview of the Program.........................................................................................5 Goals, Objectives, Strategies, and Priority Activities..............................................6 Design and Implementation.....................................................................................7 Exit...........................................................................................................................8 Correspondence of Program With Bank Priorities .................................................8 3. Outcomes, Impacts, and Sustainability....................................................................9 Overview of Intended Program Outcomes and Impacts..........................................9 Monitoring and Evaluation....................................................................................10 A Review and Analysis of the UCW External Review............................................10 4. Governing and Managing the Program.....................................................................15 Overview of Program Governance........................................................................15 5. Fostering a Results-Based Partnership......................................................................18 Partnerships...........................................................................................................18 6. Financing of the Program ...........................................................................................20 7. Risks and Risk Management.......................................................................................21 iii 8. Bank Performance.......................................................................................................22 Bank Involvement in Child Issues..........................................................................22 Potential for Added Value From Bank Involvement in the Program.....................23 9. OED Findings...............................................................................................................23 Relevance: Are the Program's Objectives Right?........................................................23 Efficacy: Has the Program Achieved Its Stated Objectives?.................................24 Efficiency................................................................................................................26 Bank Performance .................................................................................................26 10. Lessons ........................................................................................................................27 References.........................................................................................................................30 Annex A. Evaluation Framework for Phase 2 Report and 26 Case Studies ..............33 Annex B. Program Expenditures....................................................................................41 Annex C. UCW Outputs, Research and Reports..........................................................42 iv Acknowledgements The report was prepared by Manmohan Agarwal and Lauren Kelly under the overall guidance of Task Team Leader Uma Lele. The OED evaluation team was able to visit one of the four target countries assessed by the UCW program. The team thanks World Bank Morocco Country Director Mr. Ferid Belhaj and his thoughtful staff for hosting and coordinating OED's country visit. In Morocco, the OED team interviewed several national officials concerned with child labor issues, including Mohamed Tadili, Minister of Work and Employment, Mme. Bouazzaoui, Director of Formal and Informal Education, Miss Sabah Benchekroun, Head of the Mission to the Prime Minister in charge of general economic affairs, S.E. Mme Yasmina Baddou, State Secretary in charge of Family, Solidarity and Social Action, and Mr. Abdelaziz Adnane, Minister of Finance and Privatization and Chief of the Social Sector Division. The OED team also met with the international program partners, including Mme Maie Ayoub, Representative of UNICEF, Mme Rajae Berrada, head of the Morocco child protection program, Mme Malak Benchekroun, ILO/IPEC Administrator for the Morocco program, and Mr. Berrada, Consultant UCW ­ UNICEF. The OED team interviewed several people integrally involved in the program, including Furio Rosati, Project Coordinator, Understanding Children's Work, Frans Röselaers, Director, IPEC, Peter Wichmand, Senior Evaluation Officer and Head of Design, Evaluation, and Database Unit, IPEC, and Frank Hagemann, Senior Policy Analyst and SIMPOC Coordinator, IPEC. In the Bank, the OED team met with Robert Holzmann, Sector Director, Human Development Network and Social Protection and Jean Fares, Lead Economist, Social Protection and Task Manager of the UCW Project, Jacques Baudouy, now Sector Director, Health, Nutrition & Population, Christine Allen Allison, Social Protection Specialist for the South Asia Region. Issues related to social funds and the Bank's Social Protection Strategy were discussed with Soniya Carvalho of OEDST. Thanks are extended to Lisa Block for editing this study and to Maisha Hyman for her editorial assistance. v Preface This report on Understanding Children's Work is one of 26 studies (see list on page vi) that have been prepared as source material for the second phase of OED's independent evaluation of the Bank's involvement in global programs. The Phase 1 Report, The World Bank's Approach to Global Programs, which was presented to CODE in June 2002, focused on the strategic and programmatic management of the Bank's global portfolio of 70 programs in five Bank Networks (a cluster of closely related sectors). The Phase 2 Report is based on case studies and background papers of 26 global programs and derives additional lessons for the Bank's strategic and programmatic management of global programs as well as lessons for the design and management of individual programs. The first and largest case study ­ for the Consultative Group on International Agricultural Research (CGIAR) ­ was completed in April 2003. The 25 remaining case studies and background papers have been completed in parallel with the Phase 2 Report. (All publicly disclosed reports are available on the study web site.) OED reports typically contain recommendations only in those reports presented to the Bank's Board or its committees such as the Committee on Development Effectiveness (CODE). While the case studies that underlie OED's Phase 2 Report were not presented to CODE individually, they were distributed in draft to program partners to obtain their feedback, which was taken into account in the final versions of each report before being disclosed to the public. Each case study follows a common outline and addresses four major evaluation issues, which correspond to the four major sections of each report: · The overarching global relevance of each global program · Outcomes and impacts of the program and their sustainability · Organization, management, and financing of the program · The World Bank's performance as a partner in the program These four issues correspond roughly to OED's evaluation criteria of relevance, efficacy, efficiency, and Bank performance, appropriately adapted for global programs. Each case study addresses 20 evaluation questions related to these four evaluation issues (Annex A, Table A.1) that have been derived from OED's standard evaluation criteria (Table A.2), the 14 eligibility and approval criteria for global programs that have been endorsed by the Development Committee and established by Bank Management (Table A.3), and the 8 eligibility criteria for grant support from the Bank's Development Grant Facility (Table A.4). Twenty out of the 26 case study programs and about two-thirds of the Bank's total portfolio of 70 global programs have received DGF grants. Global programs are defined as "partnerships and related initiatives whose benefits are intended to cut across more than one region of the world and in which the partners (1) reach explicit agreements on objectives, (2) agree to establish a new (formal or informal) organization, (3) generate new products or services, and (4) contribute dedicated resources to the program." (OED, The World Bank's Approach to Global Programs: Phase 1 Report, p. 3). vi Since November 2000, all new global and regional programs have had to be approved at the initial concept stage, based upon the six approval criteria in Table A.3, by the managing director responsible for the Network or Regional Vice Presidential Unit advocating the Bank's involvement. Such approval authorizes the respective VPU to enter into agreements with partners and to mobilize resources for the program ­ whether from the DGF, trust funds, or the Bank's administrative budget. Both before and after November 2000, the Bank's participation in some high-profile programs ­ such as the Global Environment Facility, the Multilateral Fund for the Implementation of the Montreal Protocol, the Prototype Carbon Fund, and the Global Fund to Fight AIDS, TB, and Malaria ­ has been considered and approved by the Bank's Executive Board. Once a program has been approved at the initial concept stage, the Network vice presidencies are responsible for oversight, management, and quality assurance of their respective portfolios of global programs. This includes establishing priorities among programs in their Networks, ensuring their coherence with the Bank's strategy for each sector, sponsoring applications for DGF grants, managing programs that are housed inside the Bank, fostering links to the Bank's country operations, and promoting synergy among programs within the Network, with the rest of the Bank, and externally with partners. Regional vice presidents oversee and manage the portfolio of regional programs and partnerships in their respective Regions. While regional programs are not covered in this OED evaluation, many global programs have strong regional dimensions, which are addressed in the case studies, in addition to their links to the Bank's country-level economic and sector work, policy advice, and lending. vii List of Case Study Programs Acronym/ Operational Size (US$ Short Form Full Name Start Date millions)1 Environnent & Agriculture 1. CGIAR Consultative Group on International Agricultural Research 1972 395.0 2. GEF Global Environment Facility 1991 387.53 1. MLF Multilateral Fund for the Implementation of the Montreal Protocol 1991 158.6 2. ProCarbFund Prototype Carbon Fund 2000 6.5 3. CEPF Critical Ecosystem Partnership Fund 2000 20.19 4. GWP Global Water Partnership 1997 10.25 5. GIF Global Integrated Pest Management Facility 1996 1.3 Health, Nutrition & Population 6. TDR Special Programme for Research and Training in Tropical Diseases Dec 1975 47.5 7. Global Forum Global Forum for Health Research Jan 1997 3.07 8. UNAIDS Joint United Nations Programme on HIV/AIDS Jan 1996 95.0 9. RBM Roll Back Malaria Nov 1998 11.4 10. Stop TB Stop TB July 1999 20.8 11. GAVI Global Alliance for Vaccines and Immunization Oct 1999 124.1 Infrastructure & Private Sector Development 12. WSP Water and Sanitation Program March 1978 12.4 13. ESMAP Energy Sector Management Assistance Programme Jan 1982 7.58 14. CGAP Consultative Group to Assist the Poorest August 1995 12.67 15. infoDev The Information for Development Program Sept 1995 6.07 16. PPIAF Public-Private Infrastructure Advisory Facility Dec 1999 15.61 17. CA Cities Alliance Dec 1999 13.25 Social Development & Protection 18. PostConFund Post-Conflict Fund 1998 10.6 19. UCW Understanding Children's Work 2000 0.56 Trade & Finance 20. IF Integrated Framework for Trade-Related Technical Assistance 1997 2.71 21. FSAP Financial Sector Assessment Program May 1999 10.46 22. FIRST Financial Sector Reform & Strengthening Initiative July 2002 4.64 Information & Knowledge 23. GDN Global Development Network Dec 1999 8.67 24. World Links World Links for Development 1998 6.52 /1 FY04/CY03 expenditures. For the following cases updated, audited data was not readily available so the previous fiscal or calendar year expenditures were used: Global Integrated Pest Management Facility, Water & Sanitation Program, Integrated Framework for Trade-related Technical Assistance. viii Executive Summary 1. Among the poorest segments of society, working children are unwitting participants in a vicious circle of poverty: Uneducated, poorly trained, and subject to illness, many of these children will continue in low-paying jobs into adulthood, perpetuating poverty into the next generation. Worldwide, the number of economically active children is estimated at about 350 million ­ about 23 percent of children ­ of whom 94 percent live in the developing world. Some 250 million of these are child workers, with nearly 50 percent engaged in hazardous work. GENESIS, OBJECTIVES AND ACTIVITIES Understanding Children's Work Established: December 2000 2. To address this critical issue, the World Objectives: The UCW Project aims to Bank, the International Labor Organization address the crucial need for more and better data on child (ILO), and the United Nations Children's Fund laborers. (UNICEF) were invited by donors in December Key Activities: Take stock of existing 2000 to jointly undertake an inter-agency information on child labor; research program, Understanding Children's Assess policies to eliminate Work (UCW). Financed by the governments of child labor & develop a Norway, Finland, and Sweden, the program is methodology for assessing the impact of various policies; based at the UNICEF-Innocenti Research Centre Build capacity at the country (IRC) in Florence, Italy. Implementation of the level to track and assess child program is a cooperative effort between the three labor, analyze policies and agencies, each to varying degrees involved in programs relating to child aspects of child development. The program's labor, and develop recommendations to support governance structure consists of a Steering elimination of child labor. Committee, a Secretariat, and country-level FY03 560,000.00 Working Groups. UCW is guided by the Oslo expenditures: Agenda for Action, unanimously adopted at the FY05 DGF 100,000.00 1997 International Conference on Child Labor. allocation: Governance Secretariat inside external The Agenda laid out priorities for the worldwide model: organization; Bank is member fight against child labor. The UCW Project aims of Steering Committee to enhance international cooperation and support Location: UNICEF-Innocenti Research broader efforts to develop effective and long- Centre (IRC) in Florence, term solutions to child labor. Italy Governing WB, ILO, UNICEF partners: 3. UCW has three main components: Core Donors: Norway, Finland and Sweden · Data review: Take stock of existing Latest program- Ronald Rodts, 2002. information on child labor; level evaluation · Policy Analysis: Assess policies to eliminate child labor & develop a methodology for assessing the impact of various policies; · Capacity building: Build capacity at the country level to track and assess child labor, analyze policies and programs relating to child labor, and develop recommendations to support elimination of child labor. ix DESIGN AND IMPLEMENTATION 4. The UCW is governed by a Steering Committee made up of senior managers from the World Bank, ILO and UNICEF that biannually provide strategic direction and approval of the annual budget and work program. (The Director of the Social Protection Unit is the Bank's representative on the UCW Steering Committee). In addition, semi- annual donor meetings of the UCW Project are held for all donors to participate and give feedback. A Project Coordinator, based at the Innocenti Research Centre in Florence, Italy, supervises and monitors all activities, but in some instances activities are supervised by a coordinator or agency designated by the Project Coordinator and the focal points. The Coordinator and his team serve as the Secretariat. The staff consists of 2 statisticians and an assistant. Consultants are hired as needed. 5. Regular working level agency support to the UCW Project is provided regularly through the agency focal points, appointed by the Steering Committee. The country activities are coordinated through regional/country offices and HQ staff. The Bank's Human Development Network Vice-President oversees the Social Protection Unit's coordination of this program. In the Bank, the Social Protection Unit has relied on the strong involvement and willingness of regional staff, which has materialized for the most part in MNA and SAR, due to fortunate individual staff commitment. 6. To date, no formal memorandum of understanding has been signed by the program partners. This key agreement was reportedly under negotiation at the time of this report's completion­ more than three years after the program's inception. The lack of early agreement on expectations, outputs and roles has resulted in a re-sequencing of the program's implementation stages (originally designed to deliver its outputs simultaneously, the program first re-sequenced its output across three stages and then subsequently scaled back its delivery to a two-stage process). This realignment has had the effect of pushing the program's key component of capacity building, for the most part, into the second phase. Efforts to expand the program's country level efforts to include additional countries should be reassessed against the need to deepen its ongoing country-level capacity building and policy recommendations in its five pilot countries. 7. Phase 1 of the UCW initial project phase ran from December 2000 to July 2003. Phase II is considering additional UN agencies as potential new partners, and it will end in 2006. While the first project phase was financed by the project's original bilateral donors, two out of the three original donors have opted not to finance the program any further. With an initial two-year budget of $2 million, the UCW had proposed a revised $3.4 million budget to finance activities developed since its establishment in 2000. The expansion of the program was not agreed to, though the implementation of the program was extended by a year ­ until the end of 2003. Meanwhile, a proposal for DGF funding for FY03 was rejected on the grounds that the program was more country-focused than global. DGF subsequently approved funding for FY04 and FY05 at a level of $100,000.00 in an effort to convey to donors the Bank's commitment as a partner engaged in the fight against child labor. 8. A major turning point in the project process has been the critical external review commissioned by the UCW Steering Committee mid-way through the program. A x generally negative review - one that highlights the very distinct cultures, mandates, and operating modalities of the agencies involved ­ it was rejected by both the Bank and ILO on the grounds of lack of independence (violated the arm's length criterion with UNICEF) and lack of technical merit. The disagreement that resulted caused some donors to rethink their support opting to redirect funds through the agencies' wider social protection efforts instead. Meanwhile, all three agencies reportedly remain strongly committed to the UCW Project, reflected by their support for a second, 41-month project phase designed to build on the achievements of the first phase. OED FINDINGS Relevance: Are the Program's Objectives Right? 9. The UCW Project responds to the need articulated by the Oslo Agenda for Action unanimously adopted at the 1997 International Conference on Child Labor, for strengthened co-operation and coordination among the three partner agencies to address the lack of reliable data in the field of child labor. There was a general recognition in Oslo that, despite a common policy framework in the form of ILO Conventions No. 138 and No. 182, the UN Convention on the Rights of the Child (CRC), and the Millennium Development Goals, action on child labor was poorly coordinated across the three agencies. 10. The UCW program provides global public goods in the form of aid coordination and knowledge sharing. However, as reported by the UCW planning documents for their second stage, little progress was made during the first implementation stage in terms of extending existing current knowledge on the impact of child labour program interventions both within and beyond the three agencies. That is not to say that interagency coordination is not vital to the effort of the elimination of child labor. The premise that underlies this program, that no one agency can go it alone, and its efforts to combine the particular but complementary analytical strengths of three separate agencies is a relevant approach in the face of a highly multi-faceted global concern. (The program cites UNICEF's expertise in child health, ILO's expertise in occupational health and the World Bank's experience with public health systems, for example.) Interagency coordination in policy planning is so vital because while child labor is often harmful, misguided strategies and knee-jerk policy direction can have the effect of deepening poverty for families and children. Interagency cooperation can also help ease the burden of national statistics officers who must coordinate and assimilate survey data separately for each agency. 11. Currently however, there is a lack of precise, agreed-on definitions for child work-related terms like child work, child labor, and economically active children. With child labor experts seemingly unable to reach agreement on a common set of definitions, in-country and international researchers are obliged to constantly redefine terminology and usage, complicating efforts to share data, compare collection and analytical methodologies, or conduct activities via partnerships. xi 12. The program meets two of the four criteria for Bank involvement in global programs. The program provides clear value added to meeting the Bank's development objectives, in addition to responding to an emerging international consensus that global action is required to address continuing high rates of child labor. The program also falls into one of the Bank's GPG Priority Areas ­ that of information and knowledge. The Bank does not have a comparative specialization in analyzing the incidence of child labor ­ it is for this reason that the Norwegian Government has funded the Bank's Global Child Labor Program (1998 ­ present) to identify and strengthen the comparative advantage of the Bank, to enable it to better address child labour issues in its lending and non-lending operations, and to encourage the Bank to facilitate greater cooperation among multilateral agencies. 13. The Bank's comparative advantage lies in its ability, through lending and non- lending activities in education, health and social development, to undertake the long-term investments needed to tackle the detrimental consequences of child work. Meanwhile, it recognizes that reducing the incidence of child work through broad-based poverty alleviation strategies will take time. Therefore, a targeted partnership like UCW is a relevant tool to assist the Bank in its own policy development. The Bank has the ability to work though a multi-faceted approach which focuses on increased access and better quality education, enhanced health facilities, and labor market reforms fitted with appropriate social safety measures. The Bank's efforts to address the long-term causes and consequences of child work will be essential to achieve the MDGs of halving the incidence of poverty, promoting gender equality and achieving universal primary education by 2015. Efficacy: Has the Program Achieved Its Stated Objectives? 14. Donors had been pushing for interagency cooperation in the field of child labor even before the need was announced officially at the Oslo Conference. The UCW Project has been operational for just over three years, yet it has substantially contributed to the international community's collective understanding of child labor issues. It has provided the agencies with a first-ever common overview of the total extent of available information on child labor. Data sets, country statistics, project interventions, and research publications ­ have been made available to the three agencies and the broader development community through the project website (www.ucw-project.org), and through seminars, conferences and other media. UCW has also played a key role in acting as a laboratory for innovative research on child labor, exploiting the flexibility to pursue new research ideas not enjoyed by the three agencies individually. A wide range of original research activities was undertaken, helping to fill key information gaps in agency efforts relating to child labor. 15. The participating agencies agreed on a series of joint country-level interventions, with activities originally planned in six countries ­ El Salvador, Guatemala, Morocco, Nepal, Tanzania, and Yemen. The effort culminated in five jointly prepared Country Child Labour Assessment Reports, which have generally been acknowledged by all three agencies to be very useful. An OED Global team mission to Morocco learned that while the UCW country study of the country's child labor issues raised the interest and enthusiasm of several social ministries, country officials regarded this exercise to be xii useful mainly for the agencies themselves ­ as an exercise to accommodate the new UNIECF mission ­ and emphasized that policy advice as opposed to data collection would be much more useful. Government linkages, at least in this one country case, appeared weak, as the country's Minister of Finance complained that he had not been consulted in the project's design. The Minister of Education suggested that stronger in- country links to regular Bank lending activities, particularly in the education sector, would increase the effectiveness of the exercise. 16. On the basis of the findings of data mapping, project mapping, survey comparison, and country comparisons, the UCW Project has committed to building national capacity in the field of child labor. The objective of this exercise is to assist countries to tackle the problem of child labor by enhancing their capacity to collect data on child labor, analyze the reasons underlying child labor and learn from international experience what works and what does not work in trying to reduce the incidence of child labor. The UCW aims to build a Pilot Capacity Building Exercise, which builds on a two- day pilot counterpart training event held in May 2003, which was attended by representatives from 18 countries. A review of this training event reveals that there was overwhelming participation from the LAC region (13 out of the 18 participants attending) while only a handful of participants attended from all other regions. Lessons learned from this exercise could greatly serve the Bank's LAC clients to formulate better strategies based on regional experience. Unfortunately, with minimal participation from such severely affected regions like South Asia and Sub-Saharan Africa, the global relevance of this exercise to disseminate international lessons learned is missed. 17. In light of the informal nature of the program's original agreement and the degree to which individual efforts have been involved, the current funding situation (or lack thereof), the underlying tension which has surfaced between the Bank and UNICEF, and the Bank's exit scheduled for 2006, it is not yet clear how UCW research outputs ­ such as tools development ­ and key UCW achievements will be fully integrated into the work of the three agencies by the end of the project's life. Nor is it clear how the current website, which has generated wide support across the agencies and among researchers, will be maintained and updated after 2006, unless it is "owned" by one of the three agencies. The upcoming evaluation scheduled for 2005 should explicitly address how the program can use its final year to deepen its dissemination and mainstreaming efforts. Efficiency 18. The UCW program demonstrates the difficulty of monitoring and evaluating impacts when program objectives are vaguely formulated or in constant flux. Unstated or implicit objectives pose a further challenge in terms of tracking and measuring program effectiveness. According to the external program review, UCW's monitoring track record has been "rather poor," due to a variety of gaps. The review noted that the program's start-up was not given adequate attention and that "the Steering Committee failed to ensure a constant reassessment of issues and progress towards long-term objectives." 19. The program, though small in scale, has run into both financial and reputational risks that do not appear to have been anticipated at the project's inception. UCW faces a financial risk due to the fact that its original donors have not contributed additional xiii resources for the program's second phase. Considering that the program has mapped activities across two distinct phases ­ phases which are highly dependent on one another for the overall success of the program ­ it is problematic that the partners must engage in fund-raising activities with new donors in order to sustain the core functions of the program. Meanwhile, divergent expectations of the program partners have led to an ineffective working relationship between the Bank and UNICEF. The seriousness of this risk does not seem to have been managed by the agencies' oversight mechanisms. Higher-level governance interventions may have corrected wayward tendencies based on personal differences. 20. The program also runs the risk of not being `owned' by the countries in which it is working. Developing countries are not represented in the governance body of this program, and beyond adjusting the terms of reference of the country study, developing country ministries' have had little voice in the design of the pilot exercises or the management of the activities. Adjusting the TORs at the country level to best fit the individual country needs is a relevant and efficient process. However. it is an exercise that appears to need at best regional coordination, as opposed to global organization. Bank Performance 21. Acknowledging that other organizations, such as ILO's International Program for the Elimination of Child Labour (IPEC), UNICEF, and some NGOs are at the forefront of global activities to directly attack the problem of child labor, the World Bank has committed itself to helping reduce child labor in the long run through its ongoing poverty reduction efforts. However, it has simultaneously recognized that while economic development and improved education opportunities are essential elements of effective strategies to combat child labor, additional faster acting measures are needed. A decade ago, the Bank advised that child labor can be used as a targeting device to help children through health, nutrition, schooling and other interventions, while the long term objective of the elimination of child labor should be approached through a package of legislative action and economic and social incentives, which take into account the types of country- specific child labor arrangements (Grootaert and Kanbur, 1995). The Bank's 1997 policy paper on child labor further recommended including the topic of child labor in training programs, addressing the issue in CAS reviews when appropriate, and highlighting the issue in project appraisal documents and in presentations to the Bank's Board of Executive Directors. More recent research published in the World Bank Economic Review confirmed the fact that a multiplicity of tools exist with the potential to effect reduction in child labor rates in countries. The volume provides insight into the impact of the Brazilian school subsidy program Bolsa Escola and similar cash transfer-based programs cutting across child labor, poverty, and education issues. 22. The Bank's social protection sector is one of its youngest sectors, yet is a growing sector: social protection lending has increased more than six-fold since 1994. The Bank's Social Protection Strategy Paper drafted in 2000 takes direction in part from the above mentioned analytical work on child labor. With its focus on "social risk management" the Bank's strategy strives to address the often expensive and inefficient coping strategies of the poor, such as pulling children out of school to help generate income when certain shocks occur. The strategy specifically states that " Removing children from school is a xiv common coping mechanism for poor households, but it endangers the long term potential of the children. Some areas of child labor are so clearly harmful that a major global effort should focus on their eradication. The World Bank Group (including the IFC) will build on its existing approach in this area, continuing to follow the lead of UNICEF and ILO. Furthermore, the regional Social Protection strategy papers in LAC, MENA, South Asia, East Asia and Africa have all highlighted the importance of child labor and protection issues. 23. Knowledge generated by the UCW Project has contributed to the Bank's development of its social risk management strategy, risk/vulnerability assessments and the development of strategies for vulnerable children in MNA & SAR, an the Bank-wide 'Children & Youth strategy.' It has played a key role in the development of the National Action Plan (NAP) for child labor in Yemen. Interviews with Bank staff have revealed that UCW research and country assessments have aided the work of the Bank's Global Child Protection Program and regional decision-making (the Nepal country assessment is considered to be of very high quality and utility for the Bank's South Asia Region). UCW has added value to the Bank by heightening its awareness of the multisectoral nature of child labor, education, and health, and the need for greater coordination among Bank units as well as enhanced dialogue with countries and other agencies. However, the Bank could have benefited from better strategic planning at the beginning of the project on ways to institutionalize the knowledge sharing, technical assistance, and inter-agency collaboration provided through the UCW Project. As the Social Protection Sector Strategy suggests, the Bank could help its UCW partners structure ex-ante/ex-post risk management instruments which incorporate strategies to eliminate harmful child labor. Lessons 1. Although UCW has a number of design weaknesses, including the failure to finalize its inter-agency Memorandum of Understanding as well as a lack of time-bound performance indicators to allow an objective assessment of progress and impacts of the program's mainly qualitative objectives, it is nevertheless a relevant program that responds to an international consensus that greater international coordination is needed to better address child labor issues. UCW has provided significant support for informal interagency technical coordination. It has not, however, been able to harmonize the main survey instruments of the three agencies involved. The coordination of data collection through these instruments at the country level could alleviate much administrative burden from national statistical offices. 2. The current tripartite arrangement will have to be broadened to mitigate the program partners' disparate expectations concerning key priority areas and policy outcomes. Suggestions to expand the governance structure to include possibly UNESCO and WHO are perhaps necessary, but not sufficient. A donor- led and agency-coordinated initiative, UCW's Steering Committee lacks developing country representation. There is also scant representation of developing country governments in the Country Working Groups overseeing activities in individual countries. Lack of participation limits the political xv commitment of these governments to the program and weakens their resolve to pursue the multi-pronged strategies needed to deliver a reduction in child labor. 3. The program identifies its main project beneficiaries to be poor children in developing countries, who are most vulnerable to joining the ranks of child laborers. The program to date has informed the work of international agencies involved and, to some extent, the greater academic community. A more targeted set of direct beneficiaries would have included developing country policymakers from the program's first phase onwards ­ as opposed to waiting to involve key country actors during the program's later stages. Involving more directly operational regional staff of the international agencies to demonstrate how policy and operational decisions could be better informed with improved data quality and analysis would have likewise increased the relevance of the program's first phase. Responding to developing countries' demand for policy support (as in the case of Yemen) can now broaden the reach and increase the potential impact of this program for working children. UCW is strategically positioned, but not well organized, to deliver a global public good by sharing its positive policy assistance experience across regions and within the institutions involved. Further mainstreaming of child labor considerations into the Bank's education, health and infrastructure sectors, however, will require a steadfast commitment on the part of the Bank, beyond the scope of the UCW Project, to advance innovative approaches through its multi-sectoral interventions to mitigate the underlying factors that cause children to work. 4. The UCW will require an external, independent review to examine the impact of the program's activities, the roles and responsibilities of the partners, the governance arrangement and functions, as well as its future transition strategy. While the UCW interagency Steering Committee collectively commissioned an external review, agreed on the terms of reference as well as the selection of the external consultant, the review, released in June 2003, has been rejected by two of the three agencies and has not been made public. This lack of transparency brings the program's accountability into question. 5. The project's disengagement strategy should define how project achievements will be integrated into the regular work of the agencies. Beyond 2006, it is not clear how the agencies will continue to work together given the informal nature of project correspondence; it is also not clear how the project website will be updated and maintained. While a `permanent inter-agency coordination body on child labor' has been suggested, the concept spurs a larger debate about the usefulness and effectiveness of financing and supporting short- term, catalytic programs if they require institutionalization eventually to sustain and augment their activities. How will a more permanent structure increase efforts within the Bank to mainstream knowledge related to child labor into its broader poverty alleviation strategies? 1 1. Introduction and Context: Global Challenges in the Sector GLOBAL CHALLENGES FACING THE SECTOR AND PROGRAM 1.1 Among the poorest segments of society, working children are unwitting participants in a vicious cycle of poverty: Uneducated, poorly trained, and subject to illness, many of these children will continue in low-paying jobs into adulthood, perpetuating poverty into the next generation.1 Worldwide, the number of economically active children between the ages of 5 to 14 is estimated at about 350 million ­ about 23 percent of children ­ of whom 94 percent live in the developing world. Some 250 million of these are child workers, with nearly 50 percent engaged in hazardous work. Most economically active children live in Asia and the Pacific (61 percent), with a lower number in Sub-Saharan Africa (19 percent). However, the highest incidence of child workers ­ that is, child workers as a percentage of all children in that age group ­ is in Sub-Saharan Africa. 1.2 A recent World Bank Economic Review focused on child labor and the myriad inter-related issues linked to the phenomenon.2 It focused on the close links between poverty and child labor ­ but these links are far from simple. In general, the poorer the family, the greater the labor force participation of children and the younger the age of labor force entry. While the existence of a `wealth paradox' has been observed in relation to child farm labor in data collected from rural Pakistan and Ghana (Bhalotra and Heady 2003), the overall proportion of family income derived from child labor is highest among poorest households. Though children are not well paid, children make a significant contribution to family income, often in the range of 20 to 25 percent. Sudden shocks, such as family illness, death of a parent, or an economic crisis often lead to coping mechanisms such as pulling children out of school ­ a decision which endangers the long term potential of the children. The HIV/AIDS crisis has resulted in large numbers of orphans and child-headed households, leading to a higher incidence of child labor in heavily affected areas through the disease's impact on family structure and reduction in the number of available teachers. 1.3 Lack of access to schools and/or a perceived low quality of the education contribute to child work. Schools in many developing areas suffer from problems such as overcrowding, inadequate sanitation and untrained teachers. Based on their 1Industrial studies in India have found that children usually carry out the most unskilled tasks at a workplace. Studies also indicate that children generally lack special skills or characteristics that could make them more productive in particular jobs. 2The volume is based on papers presented at a May 2002 international conference, "The Economics of Child Labour," held in Oslo, Norway, and sponsored by the World Bank, International Labour Organization (ILO), and the United Nations Children's Fund (UNICEF). Among other articles ­ including an overview of issues by Kaushik Basu and Zafiris Tzannatos and a historical perspective on child labor by Jane Humphries ­ the volume provides insight into the impact of the Brazilian school subsidy program Bolsa Escola and similar cash transfer-based programs cutting across child labor, poverty, and education issues. 2 dissatisfaction with the education available, parents opt to direct their children towards skills training (in the agricultural sector, for example) as opposed to more formal curricula. Cultural norms of acceptability also play an important role. Child work is often seen as part of a process of socialization and even as a way to acquire needed job-related skills. 1.4 Work by such NGOs as the MV Foundation in India has shown that such cultural norms can be altered, with poor people going on to make considerable sacrifices to send children to school.3 In short, as highlighted in the World Bank Economic Review volume, the study of child labor has benefited greatly from the experience and lessons gained from a number of school-based programs ­ including programs in Honduras, El Salvador, Mexico, and Brazil. These programs have shed significant light on decision-making and allocation mechanisms within households with an impact on child labor, and the interaction of such household-level phenomena with broader market forces and a variety of child labor-related incentives and interventions. 1.5 Findings from the volume underscore the fact that a multiplicity of tools exist with the potential to effect reduction in child labor rates in countries. As noted above, programs like the Brazilian program Bolsa Escola and other school subsidy efforts have shown that well-designed and -implemented cash-transfer schemes can result in significant reductions in child labor rates, while reducing poverty within families, enhancing educational outcomes, and supporting community development and broader economic growth. 1.6 That said, findings from the volume also highlight that globalization and economic development do not automatically trickle down to benefit the most vulnerable members of society ­ children. One conclusion to be drawn is the continuing need for collection and analysis of robust, reliable micro data on child labor across countries to allow country decision makers and international agencies to develop, test, and fine-tune crucial propositions and policies, and tailor broadly successful tools and interventions to fit local community needs and standards. 1.7 The role of data. Gathering and disseminating accurate, relevant, consistently defined data occupies a central role in the effort to understand and address child labor issues. For example, there is a need to further analyze the effect of work in various occupations on children's health and safety. Such an effort will require reliable data on hazards ­ an area in its infancy ­ to identify key health and work-intensity hazards affecting working children. During the past 10 years, a significant amount of information has been gathered on working children, primarily through household, workplace, and school-based surveys, supplemented by data from censuses and other sources. Three international organizations have led data-collection efforts, with each using a separate household survey methodology: the International Labor Organization (ILO), which uses its Statistical Information and Monitoring Programme (SIMPOC); UNICEF, through the 3For a discussion of the work of the MV Foundation, see Wazir 2002. The activities of the Foundation are partly funded by the World Bank. 3 organization's Multiple Indicator Cluster Surveys (MICS); and the World Bank, which uses the Bank's Living Standards Measurement Study (LSMS) survey instrument. 1.8 While the household survey has been the most commonly used methodology for collecting child labor data, it has its limitations. For one, homeless children are often missed when the survey focus is on households. Even more fundamental obstacles exist. On the most basic level, definitions of child labor can differ ­ for example, the age coverage of child labor vis-à-vis child work varies. Cross-country comparisons of data can be problematic as there is still no agreed-on set of indicators for representing a country's child labor situation. Instead, researchers must consult numerous data sources, define terminology, and select and identify indicators. 1.9 Defining child work and child labor. Although often used interchangeably, child work and child labor are terms that describe two distinct phenomena. Child work can be divided along three aspects: work in or outside the family; work in the formal or informal sector; and work for wages or not for wages. Child labor is more narrowly defined: child workers under the age of 12, children aged 12-14 engaged in more than light work, and children in the 15-17 age group performing hazardous work are categorized as child labor.4 Child laborers often work under hazardous conditions in agriculture, construction, and mining. While girls and boys are equally likely to be economically active, a slightly higher percentage of boys are child laborers and work in hazardous conditions. 1.10 Yet, although the concepts of child work, child labor, and economically active children are broadly understood and accepted on a general level, these and other child work-related terms lack precise, agreed-on definitions. As a result, researchers and others in the field must constantly redefine terminology and explain its usage, making it hard to compare data collection and analytical methodologies, share data, or conduct activities via partnerships. Many experts have expressed their sense of urgency regarding the need for adoption of a set of common definitions for basic child labor terminology ­ most fundamentally, the need for common acceptance of a definition for child labor, most likely based on the definition used by the ILO. AN INTERNATIONAL CONSENSUS FOR COLLECTIVE ACTION 1.11 There is broad agreement that children should be in school rather than at work; it is also widely acknowledged that many children labor under appalling conditions, with such work detrimental to their well-being. This consensus is reflected in the United Nations Convention on the Rights of the Child (CRC) adopted in 1989, Convention No. 138 on minimum age, and the designation of Convention No. 182 on the Worst Forms of Child Labor as one of the four principles of fundamental rights in the ILO Declaration on 4A total of 186.3 million child laborers are in the 5-14 age group, accounting for about 18 percent of economically active children in that age group. About 40 percent of economically active children in the 15- 17 age group are child laborers, who, by definition, perform hazardous jobs. 4 Fundamental Principles and Rights at Work. The speed of the ratification of Convention No. 182 was unprecedented in the history of the ILO.5 1.12 Lack of consensus on strategic approaches. While there is a substantial agreement on the need to tackle child labor, considerably less agreement exists regarding appropriate strategies and the proper role of the international community ­ for example, the relative merits of moral suasion to convince countries to accept ILO conventions versus increased reliance on regulation and enforcement. There are two broad approaches to analyzing child labor issues and developing solutions. One approach is based on improving children's income-earning opportunities, focusing on improving income- earning opportunities for people through higher growth, improved education and health, and stronger social safety nets. The other approach, reflected in the CRC, has a strong moral undertone, stressing the role of regulation and cultural norms. 1.13 In the final analysis, addressing child labor requires a coordinated, multifaceted approach, including labor market reforms, educational improvements, enhanced health services, stronger safety nets, and equity-based economic growth. Multidimensional actions to eliminate child labor require coordination between differently specialized international organizations, as well as coordination within organizations to ensure consistency across sectors. Another critical element is to identify and analyze the myriad interconnections linking child labor, environment, and trade issues ­ for example, the economic and social impact of the adoption of non-tariff barriers by industrialized countries to bar the import of child labor-manufactured products. 2. Program Alignment With Global Challenges and Bank Practices ORIGIN OF UCW 2.1 The Understanding Children's Work (UCW) program is guided by the Oslo Agenda for Action, unanimously adopted at the 1997 International Conference on Child Labor, which was organized at the initiative of the Norwegian Government and attended by governments, inter-government organizations, nongovernmental organizations (NGOs), employers' organizations, and trade unions. The conference focused on strategies and priorities of child labor programs of the three institutions and the potential for increasing inter-agency coordination and cooperation on child labor issues. It was agreed that an effective implementation of the Agenda for Action would require reliable data from a variety of sources to formulate policies, design programs, and evaluate impact, as well as improved inter-agency coordination. 5115 ratifications had been registered by February 2002. The process of ratification of Convention No. 182 has also encouraged ratification of the minimum age convention. Only 60 countries had ratified Convention No. 138 between 1975 and 1997, but this number doubled by 2002 when 120 countries ratified it. The acceleration of ratification was partly due to the greater effort made by the ILO, an effort substantially funded by the U.S. government. 5 2.2 The Conference was followed up by an Informal Consultation on Child Labour, convened by the Norwegian Government in September 1998 in Oslo and attended by participants from donor countries, research institutions, and international agencies, including UNICEF, the International Labour Organization (ILO), and the World Bank. One conclusion reached during the meeting was that little effort had been made to combine experience gained through data collection initiatives, to mainstream child labor considerations into regular data collection, or to develop synergies among initiatives. Another emerging theme was a need for the three agencies to improve uniformity in child work-related definitions, parameters, and research tools within the individual organizations. 2.3 To address these gaps, the ILO, UNICEF, and the Bank decided in December 2000 to jointly undertake an inter-agency research program, Understanding Children's Work and its Impact (UCW). Financed by the governments of Norway, Finland, and Sweden, the program is based at the UNICEF-Innocenti Research Centre (IRC) in Florence, Italy. Implementation of the program is a cooperative effort between the World Bank, UNICEF, and the ILO, each actively involved in various aspects of child development. The program's governance structure consists of a Steering Committee, a Secretariat, and country-level Working Groups. OVERVIEW OF THE PROGRAM 2.4 Although the mandates of ILO, UNICEF, and the World Bank in child labor issues vary, there are large areas of overlapping interests in such areas as data collection, analysis and use, and local capacity building. UCW is geared to serve these common interests by increasing understanding of the causes and effects of child labor, devising analytically useful categories for measuring the phenomenon, and developing strategies to address the problem through data collection, research, and assessment activities. 2.5 To meet these objectives, the UCW serves as an umbrella for a range of activities to strengthen and disseminate tools and methods to address child labor issues in countries. The program seeks to develop common methodologies for child labor data collection and analysis, including building a data bank of child labor statistics to facilitate analysis. A handbook reviewing the current state of knowledge and research on child labor is another UCW product. This handbook highlights issues related to defining child labor, collecting relevant data, and better understanding causes and possible remedies for the phenomenon. The program also works with local participants to carry out intensive studies on child labor in selected countries, resulting in specific recommendations on ways to tackle the problem. Training programs are also developed to enhance capability in developing countries in data collection and analysis, policy analysis, and activity assessment. 2.6 With the UCW program's ambitious mandate, no single actor can achieve all the required results. Simultaneous actions in health and education, safety net development and assessment, labor market reform, and investments to foster equitable growth are required. The UCW's broad field of action implies a strategy involving international agencies with specialized mandates working with multiple ministries and departments. 6 Given the program's anticipated scope, the Agenda encouraged the three international organizations that deal most directly with children ­ the World Bank, UNICEF, and the International Labor Organization (ILO) ­ to coordinate their efforts to tackle child labor to avoid duplication and foster synergies. Each organization can, within a common analytical framework, design projects in accordance with its individual mandate, working with governments to support development of local capacity in data and policy analysis. 2.7 The ultimate beneficiaries of UCW are designed to be poor children in developing countries, who are most vulnerable to joining the ranks of child laborers. As effectively addressing the problem of child labor requires adoption of appropriate policies at the country level, UCW's intermediate targets are developing country policy makers and analysts benefiting from enhanced skills and tools in data collection, research, and assessment. GOALS, OBJECTIVES, STRATEGIES, AND PRIORITY ACTIVITIES 2.8 During UCW program development, seven areas of common concern were identified as particularly well suited for future collaboration. These areas are the following: · Overview and presentation of existing data · Methodological issues and analysis of data · Labor force and household surveys with focus on child labor · Combination of household survey and experimental in-depth study · Relations between work and children's health, development, and education · Impact assessments · Capacity building at the local level 2.9 To address these areas of common concern, UCW has designed three main objective components: · Data review. The first component is directed toward taking stock of existing information on child labor. This includes analysis of existing survey methods, with the aim of developing common methodologies for collecting data on child labor and a survey and bibliography of current research on child labor. · Policy analysis. The second component involves assessing policies to eliminate child labor and developing a methodology for assessing the impact of various policies. Research is also being extended to areas where gaps in current information and analysis have been discovered. · Capacity building. The third component is directed toward building capacity at the country level to track and assess the child labor situation, analyze policies and programs relating to child labor, and develop recommendations to support elimination of child labor. 7 DESIGN AND IMPLEMENTATION 2.10 The implementation of the program has been broken up into two stages. The first project phase and the initial part of the second phase have been financed by the Government of Finland, the Government of Norway and the Government of Sweden. The first component ­ stocktaking of existing statistical information on child labor and existing research on child labor ­ has been completed. According to UCW's independent review, the program spent considerable time and effort on mapping existing child labor survey data sets and project interventions. The focus of the mapping exercise was on the review of the relevant surveys of the three agencies and, to a lesser extent, on other sources of data. The results of the mapping exercise are posted on the UCW Web site, launched in June 2001 (http://www.ucw-project.org). 2.11 As part of this effort, a database has been established, based on surveys and data from the three agencies. After reviewing various survey methods used by the agencies, agreement was reached on a core survey questionnaire on child labor; the survey is based on a standard definition of child labor used in the terms of ILO Conventions 138 and 182. But data in the existing data bank compiles previously collected data and does not reflect this agreement. Furthermore, differences on data collection methodologies persist among the three agencies. A bibliography of current research on child labor was also completed, with results distilled into a Child Labor Handbook. However, as the independent review noted, the bibliography does not give access to the actual articles. A substantial effort was made to go beyond mere listing of raw data and data sources by developing basic sets of statistics on child labor for nearly 50 countries. The data sets, currently available on the UCW website, were prepared by program staff with assistance from the ILO. Most were derived from SIMPOC and LSMS surveys, as MICS data were not accessible at the time.6 An additional activity ­ preparation of country briefs, aimed at providing a succinct overview of the various dimensions of child labor in specific national contexts ­ was initiated, but not pursued. 2.12 Research, impact assessments, and policy analysis, included in the second component began in April 2001 with a comparative analysis of the household survey instruments developed or used by ILO, UNICEF, and the Bank to collect child labor data. The program's research efforts are focusing on uncovering and analyzing new information on the determinants and consequences of child labor. The completion of these activities has, however, been delayed. The research does not seem to provide answers to many of the questions on policy options raised in the literature. The independent evaluation noted that the impact assessment studies fully take on board the experience of the organizations with such assessments. 2.13 The third component, which concentrates on activities at the country level, is building the capacity of national researchers to collect and analyze data and policy options. According to the independent review, UCW's performance record has been "modest" in this regard. The participating agencies agreed on a series of joint country- level interventions, with activities originally planned in six countries ­ El Salvador, 6The data sets were available at the time of UCW's independent review. 8 Guatemala, Morocco, Nepal, Tanzania, and Yemen. Interagency meetings, aimed at detailing the terms of reference for the country activities, eventually took place at the country level in five of the countries (El Salvador, Guatemala, Morocco, Nepal, and Yemen). The effort culminated in five jointly prepared Country Child Labour Assessment Reports, the product of a collaborative effort between the respective Central Statistical Offices, responsible government ministries, local consultants, and the three participating agencies. After considering several capacity-building proposals, agreement was reached on the organization of a Pilot Capacity Building Exercise focusing on a number of technical core modules on the economics of child labor that can be independently offered by each of the three agencies. The modules are being designed to give participants the required technical understanding and skills to develop UCW Country Assessment reports. However, capacity training has been concentrated in countries for which assessments have been completed. EXIT 2.14 HDN staff have set a Bank disengagement date of 2006 for this program, indicating that there will be no further DGF funding requests past FY05. CORRESPONDENCE OF PROGRAM WITH BANK PRIORITIES 2.15 The Bank is involved in the management of the program and in implementation of the activities of the program. The program meets two of the four criteria for Bank involvement in global programs. The program provides clear value added to meeting the Bank's development objectives, and is responding to an emerging international consensus that global action is required to address continuing high rates of child labor. The program also falls into one of the Bank's GPG Priority Areas ­ that of information and knowledge. The Bank does not have a comparative specialization in analyzing the incidence of child labor ­ it is for this reason that the Norwegian Government has funded the Bank's Global Child Labor Program (1998 ­ present) to identify and strengthen the comparative advantage of the Bank, to enable it to better address child labor issues in its lending and non-lending operations and to encourage the Bank to facilitate greater cooperation among multilateral agencies. 2.16 The Bank's comparative advantage lies in its ability, through lending and non- lending activities in education, health and social development, to undertake the long-term investments needed to tackle the detrimental consequences of child work. Meanwhile, it recognizes that reducing the incidence of child work through broad-based poverty alleviation strategies will take time. Therefore, a targeted partnership like UCW is a relevant tool to assist the Bank in its own policy development. The Bank has the ability to work though a multi-faceted approach which focuses on increased access and better quality education, enhanced health facilities, and labor market reforms fitted with appropriate social safety measures. The Bank's efforts to address the long-term causes and consequences of child work will be essential to achieve the MDGs of halving the incidence of poverty, promoting gender equality and achieving universal primary education. 9 3. Outcomes, Impacts, and Sustainability OVERVIEW OF INTENDED PROGRAM OUTCOMES AND IMPACTS 3.1 Assessing the impact of advocacy-oriented programs and their in-country activities is inherently difficult. In the case of UCW, program partners point to enhanced inter-agency working relationships at the agency as well as at the country level, exposure to research issues that lie outside the comparative interest of any one agency, and increased awareness and access to one another's statistical data related to child labor. Box 1. Overview of Intended Program Outcomes and Impacts Better overall understanding of child labor: · Agreement on a common definition of child labor, facilitating data collection and sharing; compilation and review of reliable data based on the agreed-upon definition · A clearer picture of the characteristics of child labor, including size and regional distribution of child labor force, age and gender characteristics, sector employment, and health and safety factors. · Stronger awareness of child labor issues and related work areas within international organizations. Insights into the causes of child labor, supported by improved data on economic and non-economic contributing factors: · Clarification of child labor and economic growth linkages; analysis of the impact of economic policy and the demand- and supply-side effects on child labor. · Information on the relation of poverty to children's labor force participation. · Data on the role played by educational access and quality, family status, and cultural norms. Strategies for tackling child labor identified and assessed: · Educational policies assessed. · Impact of regulatory approaches reviewed. Capacity developed in countries to address child labor issues: · Training for child labor researchers and policy analysts from developing countries. · On-the-job learning for local participants through implementation of country activities. · Broader in-country awareness of child labor issues through Child Labor Group activities. Development and dissemination of new tools, methodologies, and other outputs: · Establishment of a comprehensive child labor databank, including collection and vetting of large quantities of data for reliability and relevance. · Enhanced data collection methodologies and data analysis. · Compilation of a comprehensive handbook on the current knowledge of the causes of child labor and the effectiveness of policies to combat it. · Implementation of research surveys and country studies to evaluate the status and context of child labor within individual countries and develop recommendations on policies to combat it. · Development of training modules targeted to reach policy makers and other individuals with a potential impact on child labor. Performance indicators such as these, as well as the outputs and outcomes that are described in the box above (see Box 1), require the use of consistent reporting mechanisms based on a results-based framework which is agreed upon by all partners 10 involved. Performance indicators must not only be developed for those outputs which are quantifiable ­ such as the number of researchers or country analysts trained ­ but also for the `softer' outcomes, as designed in, for example the category entitled `better overall understanding of child labor' mentioned above. MONITORING AND EVALUATION 3.2 During a meeting in September 2002, UCW's partners agreed that the program would benefit from a review undertaken by an external consultant to assess progress, analyze challenges, clarify lessons, and identify potential future strategies. The evaluation carried out an inventory and document review of all relevant program and non-program documentation, including a desk review of strategy papers, funding proposals, donor agreements, project progress reports, other relevant working papers, and background documentation on the participating agency activities in child labor, as well as self- assessments by UCW program management and the three agencies. The inventory and desk study was used to reconstruct UCW's history and guided the second activity, a series of in-depth, semi-structured interviews at UCW's program headquarters and with relevant agency staff in the ILO, World Bank, and UNICEF. Further evaluations have been scheduled beginning in 2005.7 3.3 The 2002 external review was rejected by both the World Bank and ILO and has not been publicly posted on the program's or any of the agencies' websites. The Bank dismissed the review on the grounds that it failed to measure the progress, procedures and outputs of the program in relation to one of the program's main tenants: interagency coordination. In addition to what were perceived as `skewed benchmarks,' the Bank pointed to the use of an `opaque methodology' which relied on `unsubstantiated claims' used to support the review's findings. Despite the fact that the review's terms of reference were agreed upon by all three agencies, the Bank expressed its concern that the review misrepresented the Bank's performance in key areas such as ownership and management of the program. A REVIEW AND ANALYSIS OF THE UCW EXTERNAL REVIEW 3.4 Interagency Cooperation. The external review found that UCW, in a general way, responds to the need articulated at the Oslo Conference to strengthen cooperation and coordination among the three partner agencies in child labor. The program has facilitated the 7The UCW program has indicated in their 2003 Annual Report that two evaluations will be carried out during phase 2 of the project. The Report indicates that a mid-term evaluation of progress towards the overall phase 2 project objectives will be conducted in 2005 and that a final evaluation of the project's legacy in terms of how it influenced the individual and joint work of the three agencies will be conducted at the conclusion of phase 2 in 2006. This evaluation will be designed to identify key lessons learned regarding the mechanics, costs and benefits of interagency research collaboration in the child labor field. One of the evaluation units of the three agencies will manage the evaluation process. The nature of the evaluation, including participation, will be determined in discussions with the project focal points and will be reviewed by the steering committee. 11 creation of technical coordination mechanisms in child labor data collection and analysis: "By bringing together the different expertise and perspectives of the three agencies, the Project helped foster a process of cross-fertilization of experiences and ideas." 3.5 Yet, the review concluded that UCW's performance in facilitating actual coordination was "modest," largely as a result of the "poor quality of its design," a problem that has had a continued impact of the program throughout implementation. Thus, understanding and overcoming the challenges arising from different organizational structures and cultures has remained a major challenge. According to the review, "forging inter-agency technical co-operation mechanisms in a non-cooperative institutional environment is a time-consuming and daunting task." At first, according to the review, "there was a general lack of intellectual conviction, political will, and commitment ­ in other words, ownership ­ on behalf of the participating agencies." But, as time went on, the participating agencies showed increasing involvement and commitment to UCW and its objectives. This finding is substantiated by a recounting of the experience and lessons learned from the program's first phase as found in UCW's Work Project Document: Second Phase (August 2003) which concurred that `bringing three agencies with very distinct cultures, mandates, structures, and operating modalities together with a common research framework proved challenging during the first three year project period' (p.1). The theme is common among programs which are largely donor initiated, whereby the conceptual and operational framework must be overlaid across agencies who have themselves not yet conceived of the ways that the proposed program can complement or add value to their own strategic priorities. 3.6 Setting uniform standards for child labor data. UCW aimed at establishing a uniform definition for child labor data based on the ILO's past exercise in standard setting, which distinguishes child labor from child work by its detrimental effect on a child's development. But while an agreement in principle on the definition of child labor was reached, differences persist among the three institutions, complicating efforts to collect data according to the ILO definition. However, the program has not developed any method to collect data on children in the worst forms of exploitation, such as prostitution, child soldiers etc., data that is very difficult to collect. This is reflected in the lack of data on children in this condition in the country studies. Another program priority is to further analyze the effect of work in various occupations on children's health, including an existing list of hazardous occupations to better reflect the impact of work in various sectors on the health and safety of children. 3.7 Identifying the causes of child labor. The causes of child labor can be studied at several levels. First, there are immediate causes, which act at the individual level ­ for example, family poverty or a cash-flow crisis. These causes are easily identifiable. Another level involves underlying causes: the socio-cultural factors that lead a society to accept or even encourage child labor. Finally, there are structural causes that act on a broader economic and social level. Data collected and analyzed by the UCW supports the contention that economic growth leads to reductions in the extent of child labor. This finding is supported by the considerable reduction in child labor that has taken place in recent years in East Asia and, to a lesser extent, in South Asia. Child labor is most 12 extensive in Sub-Saharan Africa and South Asia, the two regions of the world with the greatest concentrations of poor people and most pronounced lag in social indicators. 3.8 The UCW handbook establishes the importance of non-economic factors, including good, affordable, relevant education in schools located nearby, family structure, and cultural norms. School curricula that are outdated, gender-biased, and irrelevant to local employment opportunities adversely affect school attendance.8 Over half of working children in Morocco lack any education; most of the others have attended only the first cycle of basic school.9 Distance of the school adversely affects enrollment, a factor confirmed by UCW country studies. School attendance in rural areas improves if school is not in session during the periods of peak work in the fields. Targeted subsidies geared to meeting specific educational costs, including indirect costs, improve school attendance and lower drop-out rates. The handbook, though, ignores the important role played by the cultural environment. NGOs in the field of child labor report that the social environment of poor parents can be altered so that they consider it normal for their children to attend school, similar to middle class parents. 3.9 The external review found that the program's impact was most visible at the level of the five countries where UCW initiated a "widely appreciated inter-agency collaborative process," which, according to the review, "enabled researchers to work together in new ways and new intellectual interactions and offered longer-term frameworks for further collaboration." According to the review, the UCW country studies were "perhaps the most successful activity of the Project and a collaborative effort of the participating agencies which is solely and uniquely attributable to the UCW Project." The studies have the potential to serve as useful tools for identifying gaps in a country's legislation and identifying ways to improve implementation, including improved regulation and development of a strong human rights environment. Yet, while country studies can aid in the drive for targeted legislation, they do not address the more complex issues of implementation, issues complicated by the lack of government resources and the need of poor families to send their children to work. For instance, the Yemen study found that laws on child labor were not in consonance with international conventions; in addition, the government indicated a lack of resources to implement existing laws. But the program is not designed or equipped to deal with the extremely sensitive issue of trade incentives or sanctions, a phenomena which was observed by the OED global team during its trip to Morocco in 2003.10 8In a survey, children in India mentioned lack of interest as the most significant factor in their not going to school. 9The report notes that this is caused by poor educational quality, the result of outdated curricula and teaching methods, and lack of adequate teaching materials. 10The use of trade sanctions to bar import into countries of goods produced by child laborers has been discussed both in international labor and trade conferences. Using trade sanctions is controversial. While many developed countries are in favor of the use of trade sanctions, developing countries unanimously reject the use of such sanctions. The issue is complicated. Developing countries would be the main beneficiaries from the elimination of child labor, but they have to bear the costs in terms of foregone output or expenditures for educating the children, and the benefits would accrue years after the costs have been incurred. 13 3.10 Building local capacity and awareness. According to the external review, the three agencies pointed to the need for capacity building within countries where child labor is a problem, particularly in the areas of data analysis and use. UCW has developed modules for training developing country researchers and policy analysts in child labor issues and analysis methodologies, with the modules implemented by training units within the agencies. Furthermore, the activities of the program's Child Labor Group have supported development of broader awareness of child labor issues, both in developing countries and within the Bank. 3.11 But the program has met with obstacles in its efforts to build local capacity and awareness. For example, the UCW country study of child labor issues in Morocco raised the interest and enthusiasm of several Moroccan government ministries. However, the study did not generate new data, merely compiling data already available from the government. Furthermore, a lack of coordination with the Ministry of Finance meant that the fiscal and financial implications of suggested policies and alternative solutions were never analyzed; although this shortcoming could be ameliorated over time, it raises the possibility that study recommendations with financial implications will not be acted on.11 Furthermore, the country-level working group for Nepal has no Nepalese members, which limits the potential for capacity building and reduces the likelihood of effective action on the study's recommendations. 3.12 Collecting data on child labor. Until the mid-1990s, ILO published child labor statistics based on surveys in which economically active children were used as a statistical proxy for child laborers, according to the UCW review. At the Sixteenth International Conference of Labour Statisticians in 1998, ILO discussed the concepts, definitions, measurements, and classifications of child labor in depth, opening the way for a new survey instrument built around the Statistical Information and Monitoring Programme on Child Labour (SIMPOC) program. Meanwhile, UNICEF had initiated its own process for improved data collection on child labor based on the definition in the Convention on the Rights of the Child. To address the lack of current data on key indicators for assessing progress in many countries, UNICEF developed the Multiple Indicator Cluster Survey (MICS) as a household survey tool for countries to adopt to fill data gaps. 3.13 The World Bank's Living Standards Measurement Study (LSMS) was developed in 1980 to improve household data collected by government statistical offices in developing countries. The aim of LSMS was to develop new methods to monitor progress in raising living standards, identify the consequences for households of government policies, and improve communications between survey statisticians, analysts, and policymakers. Although the first LSMS surveys followed a similar format, as time passed and countries with different circumstances were added, substantial variety arose in the surveys across countries. Although LSMS did not originally give special attention to working children, this changed after the Oslo conference in 1997, when the Bank became interested in developing better child labor data. A special project was set up under the 11Comments received from the Bank's Social Development Unit indicate that Morocco's Minister of Finance engaged in serious discussions with the agencies once the program's country report was finalized. 14 Social Protection Unit and a number of papers were commissioned on the child labor measurement challenge. 3.14 The SIMPOC, MICS, and LSMS initiatives were brought together in 2000 under UCW, which conducted a Web-based survey mapping exercise to identify, review, and present available agency survey information in an ordered, cohesive manner. 3.15 Identifying and assessing strategies to tackle child labor. Research leads to the conclusion that reducing child labor requires a multi-pronged approach: child labor legislation, improved schooling, compulsory education, improved health facilities, provision of a social safety net, labor market reform, technical change in industries with a predominance of child labor, and higher adult wages. Historically, growth and technical change have been important factors in reducing child labor. Regulation has limited impact and sometimes perverse effects. Most child labor is performed in the informal and household sectors, where it is difficult to implement regulations. The Yemen study notes that the government claims to lack the resources to implement appropriate labor legislation. Furthermore, premature or overly harsh regulation can drive prohibited activities from the organized to the unorganized sector ­ or drive children to other activities where conditions are worse. As a result of Nepal's banning of child labor within its carpet industry, many children were forced into prostitution; in India, many children were forced to work in small restaurants and similar establishments, toiling in worse working conditions for considerably lower wages. 3.16 One area of potential usefulness for the program is to identify and address the problem of conflicting interests within countries, as many parties expressing support for eliminating child labor and ensuring access to school are unwilling to pay the actual costs of these policies or accept the results that would flow from implementation. One important issue here is the extent to which the human rights approach ­ which provides the ethical underpinnings for social activist efforts to widen the ambit of human rights ­ can be used to set de facto or de jure standards to accelerate the reduction of child labor.12 The human rights approach favors regulation and more forceful government action to universalize primary education and to pass and implement laws against child labor. Some 12There is a growing trend to view economic development within a broader context of social development and establishment of rights, which includes children's rights, as reflected, for instance, in the adoption of the CRC. While some deny the proposition that social and economic rights are fundamental human rights and others argue that these become rights only after society has reached a certain level of development, the international community tends to increasingly regard basic social and economic rights as human rights. Emergence of civil society and the mass media have contributed to this trend as well. NGO concern is, thus, reinforced by globalization and by the globalization of information. The World Summit for Social Development held in Copenhagen in March 1995 had squarely identified the elimination of child labor as a key to sustainable social development and poverty reduction. After this summit, the minimum age convention was classified as a basic human rights convention. The UN makes no distinction between human rights and basic human rights. The ILO, however, makes a distinction. It defines a few of the conventions dealing with the right to form associations, the right to collective bargaining, and now treatment of child labor as basic human rights. Calling them basic human rights stresses the importance that the ILO membership attaches to these rights. Countries are expected to ratify the entire package of these basic human rights. This should also speed up the process of ratification. 15 believe in using trade sanctions against products of child labor or even generalized trade sanctions.13 3.17 Program Monitoring and Reporting Issues. According to the program review, UCW's monitoring track record has been "rather poor," due to a variety of gaps. The review noted that the program's start-up was not given adequate attention. An additional issue brought up in the review was that "the Committee failed to ensure, throughout the project cycle, a constant reassessment of issues and progress towards long-term objectives. Weak progress reporting was not denounced and no further action was taken to be better informed." The review concluded that there was recognition of the "intractability" of certain problems, but the Committee's approach was "too distant and weak on controls and too soft on the Projects failing to perform to expectations."14 3.18 Another issue raised by the review concerned program reporting. According to the review, both the quantity and quality of information in reports are "insufficient to permit rigorous analysis and make an objective determination of the Project's overall performance." Specifically, program management "confined itself to monitoring project outputs mainly and refrained from being unduly harsh or critical" toward program partners. However, the review also noted that UCW management made "a commendable effort" to improve the quality of the progress reporting following the critical assessment. By and large, it concluded, the program's Progress Report of April 2003 contained significantly higher-quality information, providing "a more solid basis for the needs of the Steering Committee to exert control and ensure accountability." 4. Governing and Managing the Program OVERVIEW OF PROGRAM GOVERNANCE 4.1 Implementation of UCW is a cooperative effort between the Bank (specifically, the Social Protection Sector of the Bank's Human Development Network), UNICEF, and the ILO. Its governance structure consists of a Steering Committee, a Secretariat, and country-level Working Groups, with periodic meetings between the Steering Committee and donors. 13The identification of exports with the problem of labor conditions is encouraged by highlighting working conditions in export processing zones or export industries in media reports, without showing general working conditions in those countries so that the impression is created that it is only in export industries that such conditions exist. This leads people to blame trade for the poor working conditions. The negative image is strengthened by the fact that usually a very large proportion of workers in export processing zones are women. 14In its last meeting of November 2003, the Steering Committee recognised that "UCW proposed activities needed to be better known and considered by the three agencies " and it requested the Project Coordinator to share the Project's plan of activities on a monthly basis. 16 Table 1: UCW Project Governance Level Role and Responsibility Membership Meeting frequency Steering Establishment and oversight of Senior management 2 per year Committee the project's overall strategic representative, one from each (SC) direction and goals. agency, plus project coordinator and representative of hosting institution as ex-officio members. Focal Develop and provide liaison Technical expert/manager, one Formal ­ 4 per year. Points (FP) with agency, approve 12 month from each agency, plus project Frequent informal work plans for submission to coordinator (PC). contacts among FPs SC, coordinate within agency and PC to carry out and provide substantial responsibilities technical input to project UCW Develop detailed ongoing 12 Project coordinator and support As required Secretariat month work plans for approval staff of FP; manage implementation of these work plans 4.2 Steering Committee. The program is managed by a Steering Committee, consisting of a program coordinator and focal points within the three implementing agencies. While the program coordinator was chosen through advertisement and open competition, focal points representing the agencies are chosen by internal processes within the agencies.15 Consensus on overall strategic direction and broad activity areas is reached at meetings of the Steering Committee, which are held roughly once every six months.16 Developing country participation is noticeably absent from this governance arrangement. 4.3 Secretariat. The UCW secretariat, located at the Research Center of UNICEF in Florence, implements the decisions of the Steering Committee. Within the secretariat special working teams undertake implementation of the core program tasks and activities. The program coordinator oversees the various working teams, ensures effective linkages between the various program activities, and provides regular and coordinated communication with the agencies. The program Focal Points from the three agencies act to help facilitate implementation of program activities. 4.4 The Bank's operational work (Country Assistance Strategies, economic and sector work, policy dialogue, and lending operations) is carried out by the Regional Vice Presidency's country departments. One challenge for all of the Bank's global programs is to ensure strong links between these different parts of the Bank to enhance the program's on-the-ground effectiveness ­ an issue explored as part of this review. For UCW, the lack of clear linkage between the Bank's focal point and operational departments has limited the impact of the Bank's considerable sector and macro expertise on policy. 15The program coordinator is Mr. Furio Rosati, and the focal points are Messers A. Fyfe ( UNICEF), F. Hagemann (ILO), and Jean Fares (World Bank). 16The first was in Geneva in November 2000, the second in Florence in April 2001, the third in Geneva in November 2001, and the fourth in Washington, D.C., in May 2002. 17 4.5 The original program called for a core budget of $2 million for an initial two-year period. A revised budget of $3.4 million was proposed to finance the specific activities developed since the program started in December 2000. Although the program was extended for a year to the end of 2003, no additional funding was provided. This could be because almost 25 percent of the original sum remained unspent. 4.6 Donors. The donors are informed about the operation of the program through an annual report as well as formal and informal annual meetings. These meetings serve to keep the donors abreast of developments in program implementation and provide a milieu for the exchange of views. Program management is left to the Steering Committee, which is appointed independently of the donors: the focal point members of the Steering Committee are appointed by the executing agencies and the project coordinator chosen through a process of advertisement and interview in which the donors were not involved. All in all, the donors have maintained an arm's length relation with the management of the project. 4.7 Other stakeholders. Developing countries are not involved in the governance or management of the program. At the same time, developing country governments can contribute to program implementation on a different level by influencing the development of terms of reference for the country studies.17 Indeed, the continued early involvement of country participation in the design of TORs should be encouraged as a way to speed the design-to-implementation process, promote local ownership of in- country activities, and increase program relevance and results. An example of where this has occurred is Morocco, where the government has taken a strong, active interest in child labor issues and was heavily involved in tailoring TORs to ensure their relevance to the country's needs. Furthermore, governments in developing countries can exert a strong influence on UCW results in other ways: for instance, Sri Lanka, China, and Indonesia have succeeded in raising school attendance rates, with a potentially strong impact on the incidence of child labor. 17 After agreement is reached to conduct a country study, a country-level Working Group ­ which consists of members from the organizations and one or more representatives of the government ­ is set up. The Working Group identifies and supports implementation of specific activities within the overall terms of reference, with discussions sometimes leading to extensive revisions of the terms of reference drawn up by the SC. Other stakeholders, such as civil society groups, are kept informed about the program mainly through a website maintained on the program. They are, however, not consulted about the program's operation. Furthermore, there is the danger that groups lacking access to Internet will be entirely excluded. The exclusion of civil society representatives may seriously lessen the program's ownership and impact. Given its small size, the program's main impact is likely to be through energizing other actors. In this regard, the program's interaction with organized country-level NGOs can be an important element in its eventual impact. 18 5. Fostering a Results-Based Partnership PARTNERSHIPS 5.1 Partnership in the UCW program operates at two levels. Countries can attend meetings of donors, in which the Steering Committee that manages the UCW reports on activities under the program. At the implementation level, the three international organizations together comprise a partnership; this partnership determines activities to be undertaken under the program and selects implementation modalities. There is no formal memorandum of understanding guiding the work. 5.2 With different mandates, comparative advantages, skills, and constituencies, the three implementing organizations of the UCW bring a unique perspective and skill set to program implementation. The differences in approach may explain the relatively limited number of donors involved in this effort. The limited number of donors may also be due to the program's relatively small size. Each of the three international organizations has activities that affect the extent of child labor: · ILO has been at the forefront of establishing international standards regarding workers' safety and their right to organize and bargain. It has also established standards regarding child work ­ minimum age standards for work and stricter standards for children working in hazardous occupations.18 ILO's International Programme for the Elimination of Child Labour (IPEC) commenced in 1992 with support from the Governments of Germany and Belgium. It has since grown into a major program supported by 30 donors with operations in 90 countries. · UNICEF has been involved in all aspects affecting the welfare of children and has provided technical assistance to governments and organizations involved with activities in child protection and children's and health and security. It has programs in 161 countries and works extensively with social welfare groups and civil society groups. · The World Bank affects children mainly through its loan operations in the areas of health, education, and social protection. It also implements a variety of programs that affect children, including a street children initiative and an early childhood program.19 World Bank involvement in child labor issues has increased 18Two child labor instruments were among the six adopted at the First Session of the International Labour Conference in 1919: the Minimum Age (Industry) Convention (No. 5) and the Night Work of Young Persons (Industry) Convention (No. 6). Early minimum age standards were linked to schooling. The conventions regarding child labor were emphasized by the adoption of the umbrella Minimum Age Convention No. 138 in 1973. Further impetus was given to the campaign against child labor by the unanimous adoption at the ILO in 1999 of convention No. 182 on The Worst Forms of Child Labour. The worst forms of child labor are defined as slavery, trafficking, debt bondage and other forms of forced labor, forced recruitment for use in armed conflict, prostitution and pornography, and illicit activities. 19In the last decade (FY 1990-2000) the World Bank financed about 635 Human Development projects in the areas of health, education, and social protection with lending of about $42.1 billion. Of these, about 302 19 since the preparation of a paper in January 1997 in response to a question at the Board. The creation of the Global Child Labor Program in 1998 supported the Bank's programs in child labor. For instance, the Bank has partnered with the Asian Development Bank and the Indonesian Government to implement the government's Back-To-School program, started in the wake of the 1997 financial crisis. 5.3 Even so, the three organizations have had little direct experience in tackling child labor at the ground level or in implementing the broad-based, multisectoral strategies needed to reduce child labor. An additional complicating factor is that the three organizations deal with different ministries and constituencies in countries. ILO works mainly with the ministries of labor but also has contacts with labor and employers' organizations. UNICEF not only works with health and other ministries involved in child-related activities, but also has extensive links with social welfare and civil society groups. The World Bank has extensive networks of in-country relations, including links with finance, planning, education, health and other related ministries. 5.4 A successful fight against child labor requires an effective partnership, not just among the three international organizations, but also strong internal coordination among Bank networks and regional operations to facilitate the Bank's necessary coordination of ministries at the country level and ensure the involvement of civil society groups for effective ground-level action. The program has raised awareness that cooperation among the three institutions is essential to eliminate child labor. However, steps need to be undertaken to increase cooperation in the field. Also, upstream work in the Bank ­ carried out, for instance, through economic sector work (ESW) ­ can have important downstream implications and affect the choice of instruments the Bank uses (for example, policy dialogue or lending) as well as the specific format of the instrument. For example, country lending operations can include a strong focus on formal primary education, or can take a broader approach that targets child labor issues through community-driven development programs, a strategy the Bank is increasingly undertaking. If the latter is selected, the Bank would need to address the need to mobilize and integrate non-formal educational expertise into its ESW and lending work to ensure operational support to countries in the area of child labor. PARTICIPATION 5.5 The activities of the UCW program are decided on by the Steering Committee. Once the Steering Committee has made a decision to undertake a work program in a country, a country-level Working Group is formed. The Working Group typically consists of representatives of the three organizations and one or more representatives of the government. The government representative acts as the focal point for interaction with the government. The government may choose to have a non-governmental person represent it or may independently carry out consultations with civil society members. operations with a lending of about $20 billion fully or partially supported child welfare, development, and protection (Kaur and Zafiris 2002). 20 5.6 The countries have not provided much input into the design or implementation of the overall UCW global program ­ nor is the broader program designed to ensure extensive participation by civil society stakeholders, including labor and employers' organizations and NGOs. But participation of stakeholders can be ­ and often has been ­ strong at the country level. For example, Morocco's Minister for Social Welfare and other high officials in the government are clearly interested in child labor issues and were strongly involved in reformulating UCW country study TORs to ensure relevance to Morocco's needs. However, the degree of country-level participation is unclear in the case of Nepal, which apparently does not have a government representative on the Country Working Group. 5.7 One country-level obstacle to full participation has been a lack of involvement thus far on the part of the finance and planning ministries ­ a fact with potentially serious consequences. This lack of coordination has meant that the fiscal and financial implications of suggested policies and alternative solutions have never been rigorously analyzed; although this shortcoming could be ameliorated over time, it does raise the possibility that study recommendations with financial implications will not be acted on. Without greater government involvement both of the Bank's regional operational colleagues and the finance ministries, development strategies that seek to provide broad- based growth to reduce poverty and child labor cannot be developed and implemented. 6. Financing of the Program 6.1 The original program proposal called for a budget of $2 million for an initial two- year period (see Annex Table 1). A revised budget of $3.4 million was proposed to finance specific activities developed since the program started in December 2000 as well as proposed new activities for the 2003-2004 period. But though the term of the program was extended to end in 2003, additional funding was not provided, perhaps because almost 25 percent of the original contribution remained unspent. Original funding for the program was mainly met by contributions from the governments of Finland ($590,408), Norway ($875,248), and Sweden ($499,812), along with a $40,000 contribution from the World Bank, later augmented by $100,000 in DGF funding for FY04.20 Italy, the Netherlands, and United Kingdom, have expressed interest in providing funds. 6.2 Of the original $2 million expenditure, $600,000 was for the planned six country studies, another $600,000 for project support, and the rest for data mapping and analysis, including household surveys and research on impact of work on children's health, education, and development. In the revised program, additional funding has been requested for analyzing the connection between HIV/AIDS and child labor, studying the role of the private sector, and quantifying the worst forms of labor. Undertaking studies and capacity building in an additional five countries is also proposed under the extended program. The money seems to have been spent judiciously, as almost 25 percent less has 20The UCW program applied for DGF funding in FY03. Its request was turned down on the grounds that most of the program's activities generated benefits at the national level. 21 been spent than budgeted. It is unclear whether this reflects some shortfall in reaching the objectives of the program, which might also be reflected in the non-acceptance of the revised program. Meanwhile, $2 million from the President's Contingency Fund was earmarked for addressing issues related to children, children's disabilities, and AIDS orphans; $250,000 of this sum is available for child labor-related activities, with some of this for the UCW. 6.3 Program funds do not cover the costs of staff from the organizations involved in program implementation. This can create two kinds of difficulties. The ILO and UNICEF, both fund-constrained, may find it difficult to continue playing an adequate role. Though this has not yet happened, the possibility that this could occur needs to be monitored. Even the participation of the World Bank might be affected by the availability of funds to finance its share of activities. The activities of the Bank unit have had to be funded from general funds available for child labor or labor market studies, not from the Bank's ESW budget for the countries. If staff in the agencies are obliged to raise funds to conduct activities, there is a risk of conflict of interest between raising funds for the UCW program and ensuring oversight of the program. 7. Risks and Risk Management 7.1 There is institutional risk to the Bank every time it engages in a global partnership. No matter how small the amount of funding allocated, global programs in general have high visibility and attract the attention of multiple stakeholders. In the case of UCW, expectations were high for the three agencies, with separate working cultures, to pool their disparate perspectives together in this collaborative program to combat child labor. 7.2 UCW, though small in scale, has run into both financial and reputational risks that do not appear to have been anticipated at the project's inception. UCW faces a financial risk due to the fact that its original donors have withdrawn (after the second fiscal cycle). Considering that the program has mapped activities across two distinct phases ­ phases that are highly dependent on one another for the overall success of the program ­ it is problematic that the partners must engage in fund-raising activities with new donors in order to sustain the core functions of the program. Meanwhile, divergent expectations of the program partners have led to an ineffective working relationship between the Bank and UNICEF. The seriousness of this risk does not seem to have been managed by the agencies' oversight mechanisms. Higher-level governance interventions may have corrected wayward tendencies based on personal differences. 7.3 The program also runs the risk of not being `owned' by the countries in which it is working. Developing countries are not represented in the governance body of this program, and beyond adjusting the terms of reference of the country study, developing country ministries' have had little voice in the design of the pilot exercises or the management of the activities. Adjusting the TORs at the country level to best fit the individual country needs is a relevant and efficient process, however it is an exercise that appears to need at best, regional coordination, as opposed to global organization. 22 8. Bank Performance BANK INVOLVEMENT IN CHILD ISSUES 8.1 The World Bank Group has been involved in addressing children's issues through its programs in education, health, and social protection. It has developed a more focused interest in child labor since the mid to late 1990s21 and the UCW program is a reflection of this interest. The level and nature of the Bank's commitment to the UCW program has been debated within the Bank. A proposal for DGF funding for the UCW program for FY03 was rejected, as the program was perceived to be more country-focused than global. But DGF funding was provided in FY04, apparently to demonstrate solidarity with the other international partners. 8.2 Yet linkages between the Bank's work at the level of anchor and its country-level work ­ i.e., ESW, CASs, policy dialogue, and lending ­ remains weak, partly due to competing country priorities. It is often unclear whether the anchor responsible for a global program has sufficiently engaged the Bank's country departments. Nor have all global programs succeeded in developing a real sense of agenda ownership on the borrower level or within country departments. Country departments preparing Country Assistance Strategies ­ which include Bank lending for education, health, and social safety nets ­ may not be involved in a global program's development. Not surprisingly, the link between the Bank's global programs and country-level work is often a weak one. 8.3 For example, although the Bank's MENA region has initiated work on child labor, the link between the UCW and MENA's work and strategy development remains unclear. To increase relevance and impact, sector-level interventions like the UCW should be tied in to a country's overall macro situation and development strategy to support broader country development priorities. 8.4 The Bank's LSMS has proved to be useful method for gathering information on child labor. Though not originally designed to gather detailed information on child labor, it has been recast to garner more such information. With greater geographical coverage than surveys conducted by the ILO and UNICEF, it has enlarged the data set available for research and analysis. 8.5 The World Bank was given responsibility to oversee work in three countries under the UCW program. The three countries were chosen on the basis of programs submitted after the regions were informed of the UCW's existence and the availability of funding for studies. As a consequence, two countries in MENA and one in LAC have programs. Other countries and regions have begun to evince interest in the program. MENA is preparing a regional strategy paper on child labor. This is not financed by the 21The Bank's involvement with child labor specifically started with the preparation of a background paper on the issue for the 1995 WDR and also an issues paper drafted by Peter Fallon and Zafiris Tzannatos: entitled "Child Labor: Issues and Directions for the World Bank,' Feb. 1998, prepared in response to an inquiry by the Board. 23 UCW nor covered under the UCW program. The purpose of the regional paper is to prepare the ground for development of programs for the CASs. POTENTIAL FOR ADDED VALUE FROM BANK INVOLVEMENT IN THE PROGRAM 8.6 The Bank could play an important role in raising the political visibility of the issue and providing access to key policy makers. Members of other agencies stressed that the Bank's involvement has provided access to key country decision makers in the ministries of finance and planning as well as often to the executive heads of governments, whereas their interactions had been with the ministries of education, labor, and social welfare. So their concerns about child labor could reach higher levels of government. This involvement of the Bank could lead to a larger allocation of resources to programs aimed at eliminating child labor. While much of this can be done by the Bank on its own, its lack of expertise in the area would limit its influence. 8.7 The Bank could also highlight the issue of the implementation of labor standards, including the elimination of child labor, by making them integral parts of its operations through inclusion in PRSPs, ESW, and operational strategies. Such a process would result in a debate within the Bank to define activity areas in which Bank involvement is both operationally feasible and desirable, and where the Bank can play an effective role. The elimination of child labor had been made a part of PRSPs for Nepal and Tanzania. 8.8 But child labor should not be addressed as simply a reporting issue: Resources need to actually be set aside for ESW development of strategies and projects that address the problem. Ultimately, reducing child labor will require coordinated, multisectoral investments. Like the Bank's efforts through its Children and Youth program in MENA, the Bank can move beyond the mapping of child labor statistics towards designing innovative mechanisms to help families keep their children in schools while working with national governments to develop initiatives that will support direct interventions for protecting those children who are already working in hazardous environments. It is widely acknowledged that the Bank has the reach to design and implement such interventions and the ability to mobilize the large investment resources needed to achieve substantial reductions in child labor. 9. OED Findings RELEVANCE: ARE THE PROGRAM'S OBJECTIVES RIGHT? 9.1 The UCW Project responds to the need articulated by the Oslo Agenda for Action unanimously adopted at the 1997 International Conference on Child Labor, for strengthened co-operation and coordination among the three partner agencies to address the lack of reliable data in the field of child labor. There was a general recognition in Oslo that, despite a common policy framework in the form of ILO Conventions No. 138 and No. 182, the UN Convention on the Rights of the Child (CRC), and the Millennium Development Goals, action on child labor was poorly coordinated across the three agencies. 24 9.2 The UCW program provides global public goods in the form of aid coordination and knowledge sharing. However, as reported by the UCW planning documents for their second stage, little progress was made during the first implementation stage in terms of extending existing knowledge on the impact of child labor program interventions both within and beyond the three agencies. That is not to say that interagency coordination is not vital to the effort of the elimination of child labor. The premise that underlies this program, that no one agency can go it alone, and its efforts to combine the particular but complementary analytical strengths of three separate agencies is a relevant approach in the face of a highly multi-faceted global concern. (The program cites UNICEF's expertise in child health, ILO's expertise in occupational health, and the World Bank's experience with public health systems, for example.) Interagency coordination in policy planning is so vital because, while child labor is often harmful, misguided strategies and knee-jerk policy direction can have the effect of deepening poverty for families and children. Interagency cooperation can also help ease the burden of national statistics officers who must coordinate and assimilate survey data separately for each agency. 9.3 Currently, there is a lack of precise, agreed-on definitions for child work-related terms like child work, child labor, and economically active children. With child labor experts seemingly unable to reach agreement on a common set of definitions, in-country and international researchers are obliged to constantly redefine terminology and usage, complicating efforts to share data, compare collection and analytical methodologies, or conduct activities via partnerships. 9.4 The program meets two of the four criteria for Bank involvement in global programs. The program provides clear value added to meeting the Bank's development objectives and is responding to an emerging international consensus that global action is required to address continuing high rates of child labor. The program also falls into one of the Bank's GPG Priority Areas ­ that of information and knowledge. The Bank does not have a comparative advantage in analyzing the incidence of child labor. However, the Bank through its investment and non-investment operations in education, health and social development is operationally oriented to undertake the long-term investments needed to tackle child labor, and more specifically the incidence of child work, through a multi-faceted approach which continues to work towards increased access and better quality education, enhanced health facilities, and labor market reforms coupled with appropriate social safety measures. The Bank's efforts to address the long-term causes and consequences of child labor will be essential in order to achieve the MDGs of halving the incidence of poverty, promoting gender equality and achieving universal primary education by 2015. EFFICACY: HAS THE PROGRAM ACHIEVED ITS STATED OBJECTIVES? 9.5 Donors had been pushing for interagency cooperation in the field of child labor even before the need was announced officially at the Oslo Conference. The UCW Project has been operational for just over three years, yet it has substantially contributed to the international community's collective understanding of child labor issues. It has provided the agencies with a first-ever common overview of the total extent of available information on child labor. Data sets, country statistics, project interventions, and 25 research publications ­ have been made available to the three agencies and the broader development community through the project website (www.ucw-project.org), and through seminars, conferences and other media. UCW has also played a key role in acting as a laboratory for innovative research on child labor, exploiting the flexibility to pursue new research ideas not enjoyed by the three agencies individually. A wide range of original research activities were undertaken, helping to fill key information gaps of relevance to agency efforts relating to child labor. 9.6 The participating agencies agreed on a series of joint country-level interventions, with activities originally planned in six countries ­ El Salvador, Guatemala, Morocco, Nepal, Tanzania, and Yemen. The effort culminated in five jointly prepared Country Child Labour Assessment Reports, which have generally been acknowledged by all three agencies to be very useful. An OED Global team mission to Morocco learned that country officials regarded this exercise to be useful mainly for the agencies themselves, however­ as an exercise to accommodate the new UNIECF mission ­ and emphasized that policy advice as opposed to data collection would be much more useful. Government linkages, at least in this one country case, appeared weak as the country's Minister of Finance complained that he had not been consulted in the project's design. The Minister of Education was concerned that the project lacked in-country links to regular Bank lending activities, particularly in the education sector. 9.7 On the basis of the findings of the data mapping, project mapping, survey comparison, and country comparisons, the UCW Project has committed to building national capacity in the field of child labor. The objective of this exercise is to assist countries to tackle the problem of child labor by enhancing their capacity to collect data on child labor, analyze the reasons underlying child labor and learn from international experience on what works and what does not work in trying to reduce this phenomenon. The UCW aims to build a Pilot Capacity Building Exercise that builds on a two day pilot counterpart training event held in May 2003, which was attended by representatives from 18 countries. A review of this training event reveals that there was overwhelming participation from the LAC region (13 out of the 18 participants attending) while only a handful of participants attended from all other regions. Lessons learned from this exercise could greatly serve the Bank's LAC clients to formulate better strategies based on regional experience. Unfortunately, with minimal participation from such severely affected regions like South Asia and Sub-Saharan Africa, the global relevance of this exercise to disseminate international lessons learned is missed. 9.8 Due to the informal nature of the original agreement and the degree to which individual efforts have been involved, the current funding situation (or lack thereof), the underlying tension which has surfaced between the Bank and UNICEF, and the Bank's exit scheduled for 2006, it is not yet clear how UCW research outputs ­ such as tools development ­ and key UCW achievements will be fully integrated into the work of the three agencies by the end of the project's life. Nor is it clear how the current website, which has generated wide support across the agencies and among researchers, will be maintained and updated after 2006 unless it is "owned" by one of the three agencies. 26 EFFICIENCY 9.9 The UCW program demonstrates the difficulty of monitoring and evaluating impacts when program objectives are vaguely formulated or in constant flux. Unstated or implicit objectives pose a further challenge in terms of tracking and measuring program effectiveness. According to the external program review, UCW's monitoring track record has been "rather poor" due to a variety of gaps. The review noted that the program's start-up was not given adequate attention and that "the Steering Committee failed to ensure a constant reassessment of issues and progress towards long-term objectives." 9.10 The program, though small in scale, has run into both financial and reputational risks that do not appear to have been anticipated at the project's inception. UCW faces a financial risk due to the fact that its original donors have withdrawn (after the second fiscal cycle). Divergent expectations of the project have led to an ineffective working relationship between the Bank and UNICEF. However, the seriousness of this risk does not seem to have been managed by the agencies' oversight mechanisms. Higher level governance interventions may have corrected wayward tendencies based on personal differences. 9.11 The program also runs the risk of not being `owned' by the countries in which it is working. Developing countries are not represented in the governance body of this program, and beyond adjusting the terms of reference of the country study, developing country ministries' have had little voice in the design of the pilot exercises or the management of the activities. Adjusting the TORs at the country level to best fit the individual country needs is a relevant and efficient process, however it is an exercise that appears to need at best, regional coordination, as opposed to global organization. BANK PERFORMANCE 9.12 Acknowledging that other organizations, such as ILO's International Program for the Elimination of Child Labour (IPEC), UNICEF, and some NGOs are at the forefront of global activities to directly attack the problem of child labor, the World Bank has committed itself to helping reduce child labor in the long run through its ongoing poverty reduction efforts. However, it has simultaneously recognized that while economic development and improved education opportunities are essential elements of effective strategies to combat child labor, additional faster acting measures are needed. A decade ago, the Bank advised that child labor can be used as a targeting device to help children through health, nutrition, schooling and other interventions while the long term objective of the elimination of child labor should be approached through a package of legislative action and economic and social incentives, which take into account the types of country- specific child labor arrangements (Grootaert and Kanbur, 1995). The Bank's 1997 policy paper on child labor further recommended including the topic of child labor in training programs, addressing the issue in CAS reviews when appropriate, and highlighting the issue in project appraisal documents and in presentations to the Bank's Board of Executive Directors (Fannon & Tzannatos). More recent research published in the World Bank Economic Review confirmed that the fact that a multiplicity of tools exist with the potential to effect reduction in child labor rates in countries. The volume provides insight 27 into the impact of the Brazilian school subsidy program Bolsa Escola and similar cash transfer-based programs cutting across child labor, poverty, and education issues. 9.13 The Bank's social protection sector is one of its youngest sectors, yet is a growing sector: social protection lending has increased more than six-fold since 1994. The Bank's Social Protection Strategy Paper drafted in 2000 takes direction in part from the above mentioned analytical work on child labor. With its focus on "social risk management" the Bank's strategy strives to address the often expensive and inefficient coping strategies of the poor, such as pulling children out of school to help generate income when certain shocks occur. The strategy specifically states that " Removing children from school is a common coping mechanism for poor households, but it endangers the long term potential of the children. Some areas of child labor are so clearly harmful that a major global effort should focus on their eradication. The World Bank Group (including the IFC) will build on its existing approach in this area, continuing to follow the lead of UNICEF and ILO" (p.ix). Furthermore, the regional Social Protection strategy papers in LAC, MENA, South Asia, East Asia and Africa have all highlighted the importance of child labor and protection issues. 9.14 Knowledge generated by the UCW Project has contributed to the Bank's development of its social risk management strategy, risk/vulnerability assessments and development of strategies for vulnerable children in MNA & SAR, the Bank-wide 'Children & Youth strategy' and it has played a key role in the development of the National Action Plan (NAP) for child labor in Yemen. Interviews with Bank staff have revealed that UCW research and country assessments have aided the work of the Bank's Global Child Protection Program and regional decision-making (the Nepal country assessment is considered to be of very high quality and utility for the Bank's South Asia Region). UCW program has added value to the Bank by heightening its awareness of the multisectoral nature of child labor, education, and health, and the need for greater coordination among Bank units as well as enhanced dialogue with countries and other agencies. However, the Bank could have benefited from better strategic planning at the beginning of the project on ways to institutionalize the knowledge sharing, technical assistance, and inter-agency collaboration provided through the UCW Project. As the Social Protection Sector Strategy suggests, the Bank could help its UCW partners structure ex-ante/ex-post risk management instruments which incorporate strategies to eliminate harmful child labor. 10. Lessons · Although UCW has a number of design weaknesses, including the failure to finalize its inter-agency Memorandum of Understanding as well as a lack of time- bound performance indicators to allow an objective assessment of progress and impacts of the program's mainly qualitative objectives, it is nevertheless a relevant program that responds to an international consensus that greater international coordination is needed to better address child labor issues. UCW has provided significant support for informal interagency technical coordination. It has not, however, been able to harmonize the main survey instruments of the three agencies involved. The coordination of data collection through these instruments 28 at the country level could alleviate much administrative burden from national statistical offices. · The current tripartite arrangement will have to be broadened to mitigate the program partners' disparate expectations concerning key priority areas and policy outcomes. Suggestions to expand the governance structure to include possibly UNESCO and WHO are perhaps necessary, but not sufficient. A donor-led and agency-coordinated initiative, UCW's Steering Committee lacks developing country representation. There is also scant representation of developing country governments in the Country Working Groups overseeing activities in individual countries. Lack of participation limits the political commitment of these governments to the program and weakens their resolve to pursue the multi- pronged strategies needed to deliver a reduction in child labor. · The program identifies its main project beneficiaries to be poor children in developing countries, who are most vulnerable to joining the ranks of child laborers. The program to date has informed the work of international agencies involved and, to some extent, the greater academic community. A more targeted set of direct beneficiaries would have included developing country policymakers from the program's first phase onwards ­ as opposed to waiting to involve key country actors during the program's later stages. Involving more directly operational regional staff of the international agencies to demonstrate how policy and operational decisions could be better informed with improved data quality and analysis would have likewise increased the relevance of the program's first phase. Responding to developing countries' demand for policy support (as in the case of Yemen) can now broaden the reach and increase the potential impact of this program for working children. UCW is strategically positioned, but not well organized, to deliver a global public good by sharing its positive policy assistance experience across regions and within the institutions involved. Further mainstreaming of child labor considerations into the Bank's education, health and infrastructure sectors, however, will require a steadfast commitment on the part of the Bank, beyond the scope of the UCW Project, to advance innovative approaches through its multi-sectoral interventions to mitigate the underlying factors that cause children to work. · The UCW will require an external, independent review to examine the impact of the program's activities, the roles and responsibilities of the partners, the governance arrangement and functions, as well as its future transition strategy. While the UCW interagency Steering Committee collectively commissioned an external review, agreed on the terms of reference as well as the selection of the external consultant, the review, released in June 2003, has been rejected by two of the three agencies and has not been made public. This lack of transparency brings the program's accountability into question. · The project's disengagement strategy should define how project achievements will be integrated into the regular work of the agencies. Beyond 2006, it is not clear how the agencies will continue to work together 29 given the informal nature of project correspondence; it is also not clear how the project website will be updated and maintained. While a `permanent inter-agency coordination body on child labor' has been suggested, the concept spurs a larger debate about the usefulness and effectiveness of financing and supporting short- term, catalytic programs if they require institutionalization eventually to sustain and augment their activities. How will a more permanent structure increase efforts within the Bank to mainstream knowledge related to child labor into its broader poverty alleviation strategies? 30 References Andvig, Jens Christopher. 2001. "Family-Controlled Child Labor in Sub-Saharan Africa ­A Survey of Research, Social Protection." Discussion Paper Series No. 122, World Bank, Washington, DC. Andvig, Jens Christopher, Sudarshan Canagrajah and Anne Kielland. 2001. "Issues in Child Labor in Africa." Africa Region Human Development, Working Paper Series, World Bank, Washington, DC. Bequele, A. and Boyden, J., eds. 1998. Combating Child Labour. ILO, Geneva. Bourguignon, F, and Francisco H.G. Ferreira, and Phillippe G. Leite. 2002. "Conditional Cash Transfers, Schooling, and Child Labor: Micro-Simulating Brazil's Bolsa Escola Program." The World Bank Economic Review, volume 17, number 2 229- 254. Deb, P. and Rosati. F. 2002. "Determinants of Child Labor and School Attendance : The Role of Household Unobservables." UNICEF Evaluation Office. 2003. "Developing Strategies for Understanding Children's Work and Its Impact: A Joint ILO-UNICEF-World Bank Project of Information Sharing and Research Collaboration." Review Report. Fallon, Peter & Zafiris Tzannatos. 1998. "Child Labor: Issues and Directions for the World Bank." Social Protection. Human Development Network. The World Bank, Washington DC. Fyfe, Alec, and Frans Röselaers, Zafiris Tzannatos, and Furio Rosati. "Understanding Children's Work: An Interagency Data and Research Cooperation Project." The World Bank Economic Review, volume 17, nr. 2 229-254. Grimsrud, Bjorne. 2001a. "Measuring and Analyzing Child Labor ; Methodological Issues, Social Protection." Discussion paper Series No. 123, October. World Bank, Washington, DC. ______. 2001b. "What Can Be Done about Child Labor? An Overview of Recent Research and its Implications for Designing Programs to Reduce Child Labor." Social Protection. Discussion Paper Series No. 124, World Bank, Washington, DC. Henschel, B. 2002. "Child Work Related Programmes: A Review of Impact Evaluation." UCW, Florence. < http://www.ucw- project.org/pdf/publications/childlabour_impactevaluation.pdf> ILO. 2002a. A Future Without Child Labor. Geneva. 31 ______.2002b. "Eliminating the Worst forms of Child Labor; A Practical Guide to ILO Convention 182." Geneva. ______. 2002c. "Every Child Counts: New Global Estimates on Child Labor." Geneva. ILO/IPEC. 2002. Action Against Child Labor: Progress and Future Priorities. Geneva. Kaur, Iqbal & Tzannatos, Zafiris. 2002. "The World Bank and Children: A Review of Activities." Working Paper, Nr. 25295. World Bank, Washington, DC. O'Donnell. O., Rosati, F, and E. van Doorslaer. 2002. "Child Labor and Health: Evidence and Research Issues." UCW Working Paper. UCW, Florence. Social Protection Unit. 2001. World Bank Spectrum, Winter 2001. World Bank, Washington, DC. Tzannatos, Zafiris. 1998. "Child Labor and School Enrollment in Thailand in the 1990s. Social Protection Working paper." December 1998. World Bank, Washington, DC. Understanding Children's Work. "2003 Project Fifth and Final Progress Report for Phase 1 (Reporting period: 1 December 2000 ­ 31 July 2003)." < http://www.ucw- project.org/> ______. 2002. Project 3rd Progress Report for Phase 1. << http://www.ucw- project.org/> Wazir, Rekha.2002. Getting Children Out of School and into Work; MV Foundation. MV Foundation, Secunderabad. 33 Annex A Annex A. Evaluation Framework for Phase 2 Report and 26 Case Studies 1. The Phase 2 Report and each case study follows a common outline and addresses 20 evaluation questions (Table A.1) that have been derived from OED's standard evaluation criteria (Table A.2), the 14 eligibility and approval criteria for global programs (Table A.3), and the 8 eligibility criteria for grant support from the Development Grant Facility (Table A.4). 2. The sheer number of these criteria, some of which overlap, can be daunting even to an evaluator. Hence the OED evaluation team has reorganized these criteria into four major evaluation issues, which correspond to the four major sections of each report (Table A.1): · The overarching global relevance of the program · Outcomes and impacts of the program and their sustainability · Governance, management, and financing of the program · The World Bank's performance as a partner in the program 3. These four issues correspond roughly to OED's evaluation criteria of relevance, efficacy, efficiency, and Bank performance, appropriately interpreted and expanded for the case of global programs. In the case of global programs, relevance must be measured not only against individual borrowing countries' priorities and Bank priorities, but also in terms of the interplay between global challenges and concerns on the one hand and country needs and priorities on the other. The former are typically articulated by the "global community" by a variety of different stakeholders and are reflected in a variety of ways such as formal international conventions to which developing countries are signatories; less formal international agreements reached at major international meetings and conferences; formal and informal international standards and protocols promoted by international organizations, NGOs, etc.; the Millennium Development Goals; and the Bank's and the Development Committee' eligibility criteria for global programs. While sponsorship of a program by significant international organizations may enhance "legitimacy" of a global program in the Bank's client countries, it is by no means a sufficient condition for developing country ownership, nor for ensuring its development effectiveness. "Relevance" and ownership by the Bank's client countries is more assured if the program is demanded by them. On other hand some "supply-led" programs may also acquire ownership over time by demonstrating substantial impacts, as in the case of the internet. Assessing relevance is by far the most challenging task in global programs since global and country resources, comparative advantages, benefit, costs, and priorities do not always coincide. Indeed the divergence of benefits and costs between the global level and the country level is often a fundamental reason for the provision of global public goods. Evaluating the relevance of global action to the Bank's client countries is however important because the global development agenda is becoming highly crowded and resources to finance it have remained relatively stagnant, therefore highlighting issues of selectivity. 4. For the global programs that have been operating for some time, efficacy can be assessed not only in terms of program outcomes but more crucially in terms of impacts on the ground in developing countries. Outcomes and impacts in turn depend on the clarity and evaluability of each program's objectives, the quality of the monitoring and evaluation of results and, where appropriate, the effectiveness of the links of global program activities to the country level. 5. Since global programs are partnerships, efficiency must include an assessment of the extent to which the benefit-cost calculus in collective organizational, management and financing arrangements is superior to achieving the same results by the individual partners acting alone. The 34 Annex A institutional development impact and the sustainability of the program itself (as opposed to that of the outcomes and impacts of the program's activities) are also addressed in this section of each report. 6. Finally, this being an OED evaluation, it focuses primarily on the Bank's strategic role and performance in playing up to its comparative advantage relative to other partners in each program. The Bank plays varied roles in global programs as a convener, trustee, donor to global programs, and lender to developing countries. The Bank's financial support to global programs ­ including oversight and liaison activities and linkages to the Bank's regional operations ­ comes from a combination of the Bank's net income (for DGF grants), the Bank's administrative budget, and Bank-administered trust funds. In the case of the Global Environmental Facility (GEF) the Bank is a trustee and in the case of the Global Fund to Fight HIV/AIDS, Tuberculosis, and Malaria (GFATM), a "limited" trustee. In the case of GEF and MLF the Bank is also an implementing agency. Thus, the assessment of Bank performance includes the use of the Bank's convening power, the Bank's trusteeship, Bank financing and implementation of global programs, and, where appropriate and necessary, linkages to the Bank's country operations. Bank oversight of this entire set of activities is an important aspect of the Bank's strategic and programmatic management of its portfolio of global programs. 7. The first column in Table A.1 indicates how the four sections and 20 evaluation questions addressed in the Phase 2 Report and case studies relates to the eight evaluation issues that were raised by the Bank's Executive Board in the various Board discussions of global programs during the design phase of OED's global evaluation and identified in the OED's Evaluation Strategy paper:1 · Selectivity · Monitoring and evaluation · Governance and management · Partnerships and participation · Financing · Risks and risk management · Linkages to country operations 8. The third column in Table A.1 indicates how the four sections and 20 evaluation questions relate to OED's standard evaluation criteria for investment projects (Table A.2), the 14 criteria endorsed by the Development Committee and established by Bank management for approving the Bank's involvement in global programs (Table A.3), and the 8 criteria for grant support from the Development Grant Facility (Table A.4). 9. The 14 eligibility and approval criteria for the Bank's involvement in global programs have evolved since April 2000 when Bank management first proposed a strategy to the Bank's Executive Board for the Bank's involvement in global programs and include the four overarching criteria endorsed by the Development Committee, and the four eligibility criteria and six approval criteria presented by Bank management to the Bank's Executive Board. Each global program must meet at least one of the four relatively more substantive eligibility criteria and all six of the relatively more process-oriented approval criteria. The first two eligibility criteria relate directly to the Bank's global public goods and corporate advocacy priorities (Table A.3). Although the six approval criteria resemble the topics covered in a project concept or appraisal document for Bank lending operations, unlike for Bank lending operations, there is currently only a one-step approval process for new global 1OED, The World Bank and Global Public Policies and Programs: An Evaluation Strategy, July 16, 2001, page 21. "Partnerships and participation" were originally listed as two separate evaluation issues in the evaluation strategy document. "Monitoring and evaluation" is now interpreted more broadly to include not only an assessment of the monitoring and evaluation procedures of each program but also the findings of previous evaluations with respect to the outcomes and impacts of each program, and their sustainability. 35 Annex A programs ­ at the concept stage and not at the appraisal stage. And new global programs only have to be approved by the Bank managing director responsible for the Network proposing a new program, not by the Bank's Executive Board. 10. While the approval of new global programs is logically separate from and prior to their financing (whether from the DGF, trust funds, or other sources), the eight DGF eligibility criteria for grant support from the DGF (Table A.4) were actually established in 1998. Twenty out of the 26 case study programs and about two-thirds of the Bank's total portfolio of 70 global programs have received DGF grants. Table A.1. Key Evaluation Issues and Questions Evaluation Evaluation Questions Reference Issues Section I. Overarching Global Relevance of the Program 1. Relevance. To what extent are the programs: A modification of OED's · Addressing global challenges and concerns in the relevance criterion (Table sector A.2) for the purpose of · Consistent with client countries' current development global programs. priorities The third bullet also relates · Consistent with the Bank's mission, corporate to managing director (MD) priorities, and sectoral and country assistance approval criterion #1 strategies? regarding a "clear linkage to the Bank's core institutional objectives" (Table A.3). 2. International consensus. To what extent did the Development Committee programs arise out of an international consensus, formal (DC) criterion #4 (Table or informal: A.3). 1. Selectivity · Concerning the main global challenges and concerns in the sector · That global collective action is required to address these challenges and concerns? 3. Strategic focus. To what extent are the programs: The four bullets · Providing global and regional public goods correspond to the four MD · Supporting international advocacy to improve eligibility criteria (Table policies at the national level A.3). · Producing and delivering cross-country lessons of relevance to client countries · Mobilizing substantial incremental resources? 4. Subsidiarity. To what extent do the activities of the DGF eligibility criterion #1 programs complement, substitute for, or compete with (Table A.4). regular Bank instruments? Section II. Outcomes, Impacts, and their Sustainability 5. Efficacy. To what extent have the programs achieved, OED's efficacy criterion or are expected to achieve, their stated objectives, (Table A.2). taking into account their relative importance? 36 Annex A Evaluation Evaluation Questions Reference Issues 6. Value added. To what extent are the programs adding The first bullet corresponds value to: to DC criterion #1 (Table · What the Bank is doing in the sector to achieve its A.3). core mission of poverty alleviation and sustainable development · What developing and transition countries are doing in the sector in accordance with their own priorities? 7. Monitoring and evaluation. To what extent do the MD approval criterion #6 programs have effective monitoring and evaluation: (Table A.3), since effective communications with key 2. Monitoring · Clear program and component objectives verifiable by indicators stakeholders, including the and Bank's Executive evaluation · A structured set of quantitative or qualitative Directors, requires good indicators monitoring and evaluation · Systematic and regular processes for data collection practices. and management · Independence of program-level evaluations · Effective feedback from monitoring and evaluation to program objectives, governance, management , and financing? 8. Sustainability of outcomes and impacts. To what OED's sustainability extent are the outcomes and impacts of the programs criterion (Table A.2). resilient to risk over time? Section III. Organization, Management, and Financing of the Program 9. Efficiency. To what extent have the programs A modification of OED's achieved, or are expected to achieve: efficacy criterion for the · Benefits more cost-effectively than providing the purpose of global same service on a country-by-country basis programs (Table A.2). · Benefits more cost-effectively than if the individual The first bullet also relates contributors to the program acted alone? to MD eligibility criterion #3 (Table A.3) and DGF eligibility criterion #3 (Table A.4). 10. Legitimacy. To what extent is the authorizing A modification of OED's environment for the programs effectively derived from evaluation criteria (Table 3. Governance those with a legitimate interest in the program (including A.2) for the purpose of and donors, developing and transition countries, clients, and global programs. management other stakeholders), taking into account their relative importance. 11. Governance and management. To what extent are the MD approval criterion #5 governance and management of the programs: (Tables B.3) and DGF · Transparent in providing information about the eligibility criterion #5 programs (Table A.4). · Clear with respect to roles & responsibilities · Fair to immediate clients · Accountable to donors, developing and transition countries, scientists/professionals, and other stakeholders? 37 Annex A Evaluation Evaluation Questions Reference Issues 12. Partnerships and participation. To what extent do DGF eligibility criterion #8 developing and transition country partners, clients, and (Table A.4). beneficiaries participate and exercise effective voice in 4. Partnerships the various aspects of the programs: and · Design participation · Governance · Implementation · Monitoring and evaluation? 13. Financing. To what extent are the sources of funding MD approval criterion #4. for the programs affecting, positively or negatively: (Table A.3). · The strategic focus of the program The third bullet also relates · The governance and management of the program to OED's sustainability · The sustainability of the program? criterion (Table A.2). 14. Bank action to catalyze. To what extent has the DC criterion #2 (Table A.3) 5. Financing Bank's presence as a partner in the programs and DGF eligibility criterion catalyzed, or is catalyzing non-Bank resources for the #4 (Table A.4). programs? 15. Institutional development impact. To what extent has A modification of OED's the program established effective institutional institutional development arrangements to make efficient, equitable, and impact criterion (Table A.2) sustainable use of the collective financial, human, and for the purpose of global other resources contributed to the program. programs. 6. Risks and risk 16. Risks and risk management. To what extent have the MD approval criterion #3 management risks associated with the programs been identified and (Table A.3). are being effectively managed? Section IV. World Bank's Performance 17. Comparative advantage. To what extent is the Bank DC criterion #3 (Table playing up to its comparative advantages in relation to A.3), MD approval criterion other partners in the programs: #2 (Table A.3), and DGF · At the global level (global mandate and reach, eligibility criterion #2 convening power, mobilizing resources) (Table A.4). · At the country level (multi-sector capacity, analytical expertise, country-level knowledge)? 18. Linkages to country operations. To what extent are MD approval criterion #1 there effective and complementary linkages, where (Table A.3) regarding needed, between global program activities and the "linkages to the Bank's 7. Linkages to Bank's country operations, to the mutual benefit of country operational work." country each? operations 19. Oversight. To what extent is the Bank exercising This relates to DGF effective and independent oversight of its involvement in eligibility criterion #6 on the programs, as appropriate, for in-house and "arm's length relationship" externally managed programs, respectively. (Table A.4). Both questions 17 and 18 together relate to OED's Bank performance criterion (Table A.2). 20. Disengagement strategy. To what extent is the Bank DGF eligibility criterion #7 facilitating effective, flexible, and transparent (Table A.4). disengagement strategies, as appropriate? 38 Annex A Table A.2. Standard OED Evaluation Criteria Criterion Standard Definitions for Lending Operations Possible Ratings The extent to which the project's objectives are consistent (1) with High, substantial, modest, the country's current development priorities and (2) with current negligible. Relevance Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, Operational Policies). The extent to which the project's objectives were achieved, or High, substantial, modest, Efficacy expected to be achieved, taking into account their relative negligible. importance. The extent to which the project achieved, or is expected to High, substantial, modest, Efficiency achieve, a return higher than the opportunity cost of capital and negligible. benefits at least cost compared to alternatives. The extent to which the authority exercised by the program is High, substantial, modest, effectively derived from those with a legitimate interest in the negligible. Legitimacy /1 program (including donors, developing and transition countries, clients, and other stakeholders), taking into account their relative importance. The extent to which a project improves the ability of a country or High, substantial, region to make more efficient, equitable and sustainable use of its negligible, modest. Institutional human, financial, and natural resources through: (a) better development definition, stability, transparency, enforceability, and predictability impact of institutional arrangements and/or (b) better alignment of the mission and capacity of an organization with its mandate, which derives from these institutional arrangements. IDI includes both intended and unintended effects of a project. Sustainability The resilience to risk of net benefits flows over time. Highly likely, likely, unlikely, highly unlikely. Highly satisfactory, satisfactory, moderately Outcome The extent to which the project's major relevant objectives were satisfactory, moderately achieved, or are expected to be achieved, efficiently. unsatisfactory, unsatisfactory, highly unsatisfactory The extent to which services provided by the Bank ensured quality Highly satisfactory, Bank at entry and supported implementation through appropriate satisfactory, unsatisfactory, performance supervision (including ensuring adequate transition arrangements highly unsatisfactory. for regular operation of the project). The extent to which the borrower assumed ownership and Highly satisfactory, Borrower responsibility to ensure quality of preparation and implementation, satisfactory, unsatisfactory, performance and complied with covenants and agreements, toward the highly unsatisfactory. achievement of development objectives and sustainability. /1 This represents an addition to OED's standard evaluation criteria in the case of global programs, since effective governance of global programs is concerned with legitimacy in the exercise of authority in addition to efficiency in the use of resources. 39 Annex A Table A.3. Selectivity and Oversight of Global Programs Selectivity Criteria for Bank Involvement in Global Public Goods: Endorsed by Development Committee (September 2000) /1 1. An emerging international consensus that global action is required 2. A clear value added to the Bank's development objectives 3. The need for Bank action to catalyze other resources and partnerships 4. A significant comparative advantage for the Bank. Approval Criteria for Bank Involvement in Partnership Initiatives Beyond the Country Level: Established by Bank Management (November 2000) /2 1. A clear linkage to the Bank's core institutional objectives and, above all, to the Bank's country operational work 2. A strong case for Bank participation based on comparative advantage 3. A clear assessment of the financial and reputational risks to the Bank and how these will be managed 4. A thorough analysis of the expected level of Bank resources required, both money and time, as well as the contribution of other partners 5. A clear delineation of how the new commitment will be implemented, managed, and assessed 6. A clear plan for communicating with and involving key stakeholders, and for informing and consulting the Executive Directors. Global Public Goods Priorities /3 Corporate Advocacy Priorities /3 Communicable diseases Strategic Focus for Oversight Empowerment, security, and social HIV/AIDS, tuberculosis, malaria and of Global Programs: inclusion childhood communicable diseases, Established by Bank Gender mainstreaming including the relevant link to education Management (March 2003) Civic engagement and participation Vaccines and drug development for Social risk management (including major communicable diseases in disaster mitigation) developing countries a. Provide global public Investment climate Environmental commons goods Support to both urban and rural Climate change development Water Infrastructure services to support Forests private sector development Biodiversity, ozone depletion and land b. Support international Regulatory reform and competition degradation advocacy for reform policy Promoting agricultural research agendas which in a Financial sector reform Information and knowledge significant way Public sector governance Redressing the Digital Divide and address policy framework Rule of law (including anti-corruption) equipping countries with the capacity conditions relevant for Public administration and civil service to access knowledge developing countries reform (incl. public expenditure Understanding development and accountability) poverty reduction Access to and administration of Trade and integration c. Are multi-country programs justice (judicial reform) Market access which crucially depend on Education Intellectual property rights and highly coordinated Education for all, with emphasis on standards approaches girls' education International financial architecture Building human capacity for the Development of international d. Mobilize substantial knowledge economy standards incremental resources that Health Financial stability (incl. sound public Access to potable water, clean air debt management) can be effectively used for and sanitation International accounting and legal development. Maternal and child health framework /1 From the Development Committee Communiqué issued on September 25, 2000. Both the Development Committee and Bank Management envisaged global programs as being the principal instrument for Bank involvement in providing global public goods. /2 The Initiating Concept Memorandum in the Partnership Approval and Tracking System (PATS) was initially organized according to these six criteria. /3 These are the five corporate advocacy priorities and the five global public goods priorities (and bulleted sub-categories) from the Strategic Directions Paper for FY02-04, March 28, 2001. Within the Partnership Approval and Tracking System (PATS), global programs are expected to identify, for tracking purposes, their alignment with at least one of these ten corporate priorities. 40 Annex A Table A.4. Eligibility Criteria for Grant Support from the Development Grant Facility 1. Subsidiarity The program contributes to furthering the Bank's development and resource mobilization objectives in fields basic to its operations, but it does not compete with or substitute for regular Bank instruments. Grants should address new or critical development problems, and should be clearly distinguishable from the Bank's regular programs. 2. Comparative The Bank has a distinct comparative advantage in being associated with the program; it does not advantage replicate the role of other donors. The relevant operational strengths of the Bank are in economic, policy, sector and project analysis, and management of development activities. In administering grants, the Bank has expertise in donor coordination, fund raising, and fund management. 3. Multi- The program encompasses multi-country benefits or activities which it would not be efficient, practical country or appropriate to undertake at the country level. For example, informational economies of scale are benefits important for research and technology work, and operations to control diseases or address environmental concerns (such as protect fragile ecosystems) might require a regional or global scope to be effective. In the case of grants directed to a single country, the program will encompass capacity-building activities where this is a significant part of the Country Assistance Strategy and cannot be supported by other Bank instruments or by other donors. This will include, in particular, programs funded under the Institutional Development Fund, and programs related to initial post- conflict reconstruction efforts (e.g., in countries or territories emerging from internal strife or instability). 4. Leverage The Bank's presence provides significant leverage for generating financial support from other donors. Bank involvement should provide assurance to other donors of program effectiveness, as well as sound financial management and administration. Grants should generally not exceed 15 percent of expected funding over the life of Bank funding to a given program, or over the rolling 3-year plan period, whichever is shorter. Where grant programs belong to new areas of activities (involving, e.g., innovations, pilot projects, or seed-capital) some flexibility is allowed for the Bank's financial leverage to build over time, and the target for the Bank grant not to exceed 15 percent of total expected funding will be pursued after allowing for an initial start-up phase (maximum 3 years). 5. Managerial The grant is normally given to an institution with a record of achievement in the program area and competence financial probity. A new institution may have to be created where no suitable institution exists. The quality of the activities implemented by the recipient institution (existing or new) and the competence of its management are important considerations. 6. Arm's length The management of the recipient institution is independent of the Bank Group. While quality an arm's relationship length relationship with the Bank's regular programs is essential, the Bank may have a role in the governance of the institution through membership in its governing board or oversight committee. In cases of highly innovative or experimental programs, Bank involvement in supporting the recipient to execute the program will be allowed. This will provide the Bank with an opportunity to benefit from the learning experience, and to build operational links to increase its capacity to deliver more efficient services to client countries. 7. Disengage- Programs are expected to have an explicit disengagement strategy. In the proposal, monitorable ment action steps should be outlined indicating milestones and targets for disengagement. The Bank's strategy withdrawal should cause minimal disruption to an ongoing program or activity. 8. Promoting Programs and activities should promote and reinforce partnerships with key players in the partnerships development arena, e.g., multilateral development banks, UN agencies, foundations, bilateral donors, professional associations, research institutions, private sector corporations, NGOs, and civil society organizations. Source: World Bank, Development Grant Facility documentation. 41 Annex B Annex B. Program Expenditures Source: Understanding Children's Work documentation. 42 Annex C Annex C. UCW Outputs, Research and Reports 1. Child Labour and Health: Evidence and Research Issues. O. O'Donnell, F. Rosati, E. van Doorslaer. January 2002. Abstract: The ILO definition of the worst forms of child labour includes work that is likely to jeopardize health and safety. Effective targeting of those child work activities most damaging to health requires both conceptual understanding and empirical evidence of the interactions between child labour and health. The aim of the paper is to review the current state of such knowledge, which is central to the design of policies that, whilst protecting children from work activities most damaging to their health, do not jeopardize the subsistence livelihood of their families. The relationships between child labour and health are complex. They can be direct and indirect, static and dynamic, positive and negative, causal and spurious. The diversity of potential relationships makes their empirical disentanglement a difficult exercise. A conceptual framework of analysis is required and important issues of measurement and of estimation must be given careful consideration. 2. Child Labour Related Programmes: A Review of Impact Evaluations. B. Henschel. November 2002. Abstract: This report is concerned with reviewing programme impact evaluations undertaken in the areas of child labour and education. The studies investigate the effects of promoted interventions on individuals, household and institutions exploring intended and unintended consequences, whether positive or negative. It is important for the evaluation system to be able to assess efficiency and short- to long-term outcomes. For a correct estimate of the programme impact, the type of evaluation methodology employed is fundamental. We present several case studies of programme evaluation, which may be classified in two major categories: 'Social Fund Programmes' and 'Targeted Human Development Programmes'. Our major concern is to highlight the evaluation of the effectiveness of the education programmes. 3. Children's Work in Zambia: A Comparative Study of Survey Instruments. N.H. Blunch, A. Dar. L. Guarcello, S. Lyon, A. Ritualo, F. Rosati. September 2002. Abstract: We analyze child work in Zambia applying two recent surveys, the LCMS 1998 (World Bank) and the SIMPOC 1999 (ILO). The analysis aims at contrasting and comparing findings on the incidence characteristics of the two surveys. The extent to which the findings are survey-dependent is assessed and implications for the design and implementation for future surveys for the analysis of child work is discussed. 4. A Comparison of Survey Instruments for Collecting Data on Child Labour. B. Grimsrud. November 2001. Abstract: This report contains a comparative analysis of the household survey instruments developed or used by the three agencies ILO, UNICEF and the World Bank for the purposes of collecting information on child labour. In addition, the usefulness of input from other household surveys, such as the USAID DHS, and qualitative instruments such as the ILO-UNICEF rapid assessment tool, on child labour measurement instruments is assessed. The report explains the similarities and differences in the various survey instruments by comparing sampling sizes and methodologies; analyzing the questionnaires by clarifying who is asked what, age limits, the definitions used etc.; and comparing and contrasting SIMPOC, MICS and LSMS data sets. 5. Children's Work in Côte d'Ivoire: An Overview. F. Francavilla, S. Lyon. March 2002. Abstract: The International Labour Organisation (ILO) estimates that in developing countries alone there are some 250 million children between the ages of five and 14 years who work. For 120 million of them, work is a full-time activity. Although child work occurs in all parts of the world, it is in Africa where a child is most likely to be involved in work and where child work is growing most rapidly. ILO estimates that the 80 million African child workers today could surge to 100 million by 2015. This paper looks at African child work in the context of Côte d'Ivoire. It aims at providing a brief overview of the various dimensions of the child work phenomenon in the country - its extent and nature, its causes and consequences, and national legislation and policies adopted to address it. 43 Annex C 6. Data Base on Children Labour in India: An Assessment with Respect to Nature of Data, Period and Uses. S. Thorat. June 2001. Abstract: The main purpose of this paper is to prepare a detailed account of the existing data-base on the various dimensions of child labour in India. The intention is to gain an insight into the nature of the data on child labour so that it may serve as a guide to possible users. India has a fairly well developed framework of data on child labour. However these data come from more than one source and therefore the administrative unit from which the data are generated, the period covered and the methodology used in terms of data collection varies from one source to another. Therefore there is a need to make a comparative assessment of the data on child labour and to provide a guideline for the proper use of the data from each of the various sources which clearly highlights each of the sources' strengths and limitations. 7. Information on Children's Work from Population Censuses and Labour Force Surveys: A Brief Review. S.K. Sinha. June 2002. Abstract: This paper reviews two important primary sources of data on child work - labour force surveys and population censuses - that researchers can turn to in countries where LSMS, SIMPOC and MICS data are not available, or that can be used to augment the data gained through these surveys. The paper looks first at data generated by the labour force surveys and population censuses, and then at the main secondary sources where these data can be found. 8. Determinants of Child Labour and School Attendance: the Role of Household Unobservables. P. Deb, F. Rosati. December 2002. Abstract: We develop a random effects multinomial logit model to distinguish between unobserved and observed household characteristics as determinants of child labor and school attendance. Using a semi- parametric approach, the random effect is drawn from a discrete distribution of latent classes of households. The results show that household-level unobserved heterogeneity is substantial. Household-level unobserved heterogeneity swamps observed income and wealth heterogeneity. Households that belong to the class with a high latent propensity to send their children to work are not influenced by marginal changes in the explanatory variables. Households most sensitive to changes in explanatory variables are those with a high propensity to have their children neither in school nor working. Policy interventions and changes in external conditions are likely to produce large changes in the behavior of this group of families. 9. A Review of Word Bank Lending for Children and its Bearing on Child Labour. I. Kaur, April 2002. Abstract: The purpose of this paper is to review World Bank interventions that have some bearing on child labor. This paper is neither meant to assess the impact of these interventions on reduction or elimination of child labor, nor does it make recommendations to address this issue. However, it does suggest linkages between human development intervention and child labor to justify the criteria for choosing only Human Development Projects as the sector projects with a bearing on child labor. The findings in this report are based upon assessing specific direct and indirect interventions as outlined below and in the sections on health, education and social protection interventions. This is the basic principle that is applied in identifying activities that have a direct or indirect impact on child labor within the World Bank projects. 10. Children's Working Hours, School Enrolment and Human Capital Accumulation: Evidence from Pakistan and Nicaragua. F. Rosati , M. Rossi. October 2001. Abstract: We analyse the determinants of school attendance and hours worked by children in Pakistan and Nicaragua. On the basis of a theoretical model of children's labour supply, we simultaneously estimate the school attendance decision and the hours worked by Full Model Maximum Likelihood. We analyse the marginal effects of explanatory variables conditioning on the "latent" status of children in terms of schooling and work. We show that these effects are rather different, and discuss the policy implication of this finding. Finally, we use our predicted hours of work to analyse the effects of work on children's school achievements. 11. Health Effects of Children's Work: Evidence from Vietnam. O. O'Donnell, E. Vandoorslaer, F. Rosati. March 2003. 44 Annex C Abstract: We test whether work in childhood impacts on health. We distinguish between urban and rural settings and focus on agricultural work, which is the dominant form of child work worldwide. We use a particularly rich two-wave panel data set ­ the 1993 and 1998 Vietnam Living Standards Surveys. The panel nature of the data and the availability of good instruments makes it possible to correct for potential endogeneity arising from both unobservable heterogeneity and simultaneity. Instruments take the form of small area labour market and education conditions obtained from community surveys matched to individual records. We use two indicators of health ­ reported illness/injury and the body mass index. Both contemporaneous and longer-term effects of child work on health are examined. A variety of estimators and identification strategies are employed to deal with endogeneity. There is strong evidence of unobservable heterogeneity bias but little indication of simultaneity bias. In rural areas, we find no evidence of a contemporaneous impact of child work on health. However, there is evidence that work undertaken in rural settings during childhood raises the risk of illness five years later. In urban areas, there is some evidence that child work has a contemporaneous negative impact on health. 12. Household Vulnerability and Child Labour: the Effect of Shocks, Credit Rationing and Insurance. L. Guarcello, F. Mealli, F. Rosati. November 2002. Abstract: Building upon the social-risk management approach, this paper examines dimensions of household behavior that are important for risk management and reduction of vulnerability, beyond issues of consumption. This paper attempts to assess to what extent risk and vulnerability factors are relevant for household decisions concerning children's school attendance and labor supply. Particular focus has been given to the evaluation of the effect of shocks, credit rationing and insurance on household decisions concerning children's activities. On the basis of a theoretical approach based on well known results relative to human capital investment decision and children's labor supply, the paper developed an estimation strategy that allows an assessment of the importance of a set of risk factors. Because of the potential endogeneity of the variable of interest, methodology based on propensity scores was applied. The analysis of the distribution of propensity scores for the "treated" and "not treated" population for the population of interest allow the conclusion that, given the maintained hypothesis of unconfoudness on observables, casual inference can be safely drawn from these estimates. Also computed is the ATE that confirms the main results obtained through the regression analysis: which indicates that credit rationing is extremely important in determining the household's decision to invest in the human capital of the children. 13. Child Labour Indicators used by the UCW Project: An Explanatory Note. A. Cigno, L. Guarcello, F. Rosati. April 2003. Abstract: This document briefly explains and illustrates the set of indicators developed for the Country Statistics found on the UCW Project website. The indicators are not the theoretical "best" of the many that could be constructed on child work, but rather those that are feasible given the available data. As such, the indicators are unable to fully capture the various dimensions of the child work phenomenon. They do, however, provide at least partial answers to the following key questions: 1) What is child work? 2) How widespread is it? 3) What are its characteristics? 4) How damaging is it for the child's health and future earning capacity? 5) Which household, local or national characteristics make it more likely that a child will work? The data used to develop the indicators are from household surveys conducted mainly by the World Bank, under its Living Standards Measurement Study (LSMS) programme, the ILO International Programme on the Elimination of Child Labour (IPEC), under its Statistical Information and Monitoring Programme on Child Labour (SIMPOC), and UNICEF, under its Multiple Indicator Cluster Survey (MICS) programme. These surveys yield a wide variety of data in areas such as education, employment, health, expenditure, and consumption that relate to child work. They do not, however, provide information on unconditional worst forms of child work, such as child prostitution and child slavery, for which different data collection methodologies are required. 14. Child Labour and Water Access in Yemen. L. Guarcello, S. Lyon. April 2003. Abstract: The strong link between water access and child health is well-documented. Much less is known about how water access affects children's activity patterns. Empirical evidence presented in this paper indicates that providing households with ready water access makes it much more likely that the children from these households attend school, and much less likely that they are reported as being idle. Improving 45 Annex C water access also reduces the likelihood of children being economically active, though the effect is smaller in magnitude. 15. Standard Core Questionnaire on Child Labour. Abstract: Consultations within ILO/IPEC and the framework of the joint UCW Project as well as country experiences in conducting national child labour surveys have been valuable in preparing the Core Questionnaire. This is the suggested minimum number of questions that can be administered to obtain basic information about child labour, thereby excluding some of the details like perceptions of parents, children living away from home, earnings from work, and housing characteristics. The criterion applied to included questions in the "Core" has been those that can generate information on magnitude, spread, nature of activities, hazardous (injuries) types of work, schooling, and impact of work on schooling. The Core questionnaire comprises seven (7) Sections, with a total of 77 questions. 16. Estimating the Effect of Fertility Decisions on Child Labour. P. Deb, F. Rosati. Abstract: We construct a joint distribution of fertility and children's activities treating the Poisson process generating the number of children as being endogenous in the multinomial logit process generating children's activities using a latent factor structure. Latent factors are incorporated into the equations for number of children and children's activities to allow for unobserved influences on fertility to affect children's activities, thus enabling us to make a distinction between selection on unobservables and selection on observables. We apply maximum simulated likelihood (MSL) techniques to estimate the parameters of our models. We find that the effect of fertility has a large downward bias in naive models. The marginal effect of fertility on child labor and schooling are twice as large when common unobserved heterogeneity in fertility and activity decisions are accounted for. COUNTRY-LEVEL STUDIES Morocco UCW Country Report Abstract: The current report was developed under the aegis of UCW Project activities in Morocco It provides an overview of the child work phenomenon in the Kingdom ­ its extent and nature, its determinants, its consequences on health and education, and national responses to it. The report serves two important UCW Project objectives in the country. First, it helps provide a common analytical understanding of child work, that can be used to inform the current activities of the three partner agencies and Government, and that can be used to develop joint interagency strategies for future cooperation with the Government in the field of child labour. Second, through close involvement of local counterparts in its development, the report contributes to a broader effort to build national capacity in analysing and using child work data for policy development. The statistical information presented in the review is drawn primarily from two recent household surveys - a national labour force survey conducted in 2000 (LFS 2000) and a national living standards measurement study (LSMS 1998-99), both conducted by the Statistics Directorate. The first involved a stratified sample of 48,000 households (32,000 urban and 16,000 rural) and the second a stratified sample of 5,184 households. An ILO/IPEC rapid assessment conducted in 1998 is the primary source of qualitative information on the nature and hazards of child labour. The review also draws on a number of smaller-scale studies, qualitative as well as quantitative, Government and NGO reports, agency documents and other information sources. 2. Socio-Economic Analysis Report. R. Mejjati Alami. September 2002. Abstract: A review and analysis from a sociological perspective of child work in selected sectors (e.g. traditional crafts, domestic service street vending), worst forms of child labour and street children based on available information both qualitative as well as quantitative. 46 Annex C 3. Economic-Normative Report and Interventions. G. Berrada. September 2002. Abstract: A comprehensive background review of institutional framework related to child labour, education, health and social protection, government policies and main economic and social indicators. A review of existing projects relating to child labour, education and welfare undertaken by central and local government, international agencies, NGOs and civil society. 4. Report on Child Labour Databases and Indicators. L. Baghagha. November 2002. Abstract: The report reviews all existing databases, which provide information on the conditions of children covering information on occupational status, education, health and nutrition. The report also includes information on the demographic structure and occupational status of children in Morocco and the development of indicators on child work. Guatemala UCW Country Report Abstract: The current report was developed under the aegis of UCW Project activities in Guatemala. It provides an overview of the child work phenomenon in Guatemala ­ its extent and nature, its determinants, its consequences on health and education, and national responses to it. The report serves two important UCW Project objectives in the country. First, it helps provide a common analytical understanding of child work, that can be used to inform the current activities of the three partner agencies and Government, and that can be used to develop joint interagency strategies for future co-operation with the Government in the field of child work. Second, through close involvement of local counterparts in its development, the paper contributes to a broader effort to build national capacity in using data for policy development in the child labour field. The report is the product of a collaborative effort of the National Statistical Institute, concerned Government ministries, local researcher institutes and ILO, UNICEF and the World Bank. The statistical information presented in the report is drawn primarily from a national living conditions survey conducted in 2000 (Encuesta de Condiciones de Vida (ENCOVI) 2000). The survey involved a stratified sample of 7,276 households and a total of 38,000 persons. The report also draws on a number of smaller- scale studies, qualitative as well as quantitative, Government and NGO reports, agency documents and other information sources. 2. Report on Municipality Pilot Initiatives to prevent Child Labour in San Raymundo. Guatemala. M.R. Camey, October 2002. Abstract: Report on the pilot initiatives carried out in San Raymundo to prevent or curb child labour. 3. Report on Municipality Pilot Initiatives to prevent Child Labour in Chiché, El Quiché. M.R. Camay. October 2002. Abstract: Report on the pilot initiatives carried out in Chiché to prevent or curb child labour. 4. Report on Municipality Pilot Initiatives to prevent Child Labour in Comitancillo, San Marcos. M.R. Camey. October 2002. Abstract: Report on the pilot initiatives carried out in Comitancillo to prevent or curb child labour. 5. Training tool (rotafolio) for capacity building for Municipal Committees. Abstract: A tested tool (rotafolio) developed to train municipal-level actors on child labour issues. 47 Annex C Yemen UCW Country Report Abstract: The current report was developed under the aegis of a joint ILO/World Bank/UNICEF project "Understanding Children's Work" in Yemen. It provides an overview of the child work phenomenon in Yemen ­ its extent and nature, its determinants, its consequences on health and education, and national responses to it. The report serves two important UCW Project objectives in the country. First, it helps provide a common analytical understanding of child work, that can be used to inform the current activities of the three partner agencies and Government, and that can be used to develop joint interagency strategies for future co-operation with the Government in the field of child labour. Second, through close involvement of local counterparts in its development, the paper contributes to a broader effort to build national capacity in collecting and using child work data for policy development. The statistical information presented in the paper is drawn primarily from two recent household surveys ­ the 1998 Yemen Household Budget Monitoring Survey (YHBS 1998) and the 1999 Yemen Poverty Monitoring Survey (YPMS 1999), both conducted by the Central Statistical Office. The first involved a stratified sample of 10,000 households and the second a stratified sample of 54,000 households. An ILO/IPEC rapid assessment conducted in 2000 is the primary source of qualitative information on the nature and hazards of child labour. The review also draws on a number of smaller-scale studies, qualitative as well as quantitative, Government and NGO reports, agency documents and other information sources. In the regression, the income data have been imputed to the National Poverty Survey 1999 using the information from the Household Budget Survey 1998, since the expenditure/income data in NPS are not satisfactory. As for the descriptive statistics on child work, they were cross-checked with results from labour force surveys and found consistent. 2. Qualitative Desk Review. M. Al Maitami. April 2002. Abstract: This review surveys the recent statistical data, official and non-official, studies, researches and other literature on child labor in Yemen. Many agencies, local and international have been involving in child labor issue, but with no coordination. Since, also, it has been raised up that the causes of child labor are poverty, outcomes of economic reform program, unemployment. Parent's socio-economic characteristic, demographic characteristic of population, weakness of educational system and so forth, this paper will highlight all these affect the child and will draw attention on what has been done and what doesn't. 3. Child Labour Initiative in Yemen: Review of Policies and Programmes. W. Keddeman. July 2002. Abstract: The main purpose of the paper is to identify and analyse interventions to reduce the incidence of child labour in Yemen. As such, the present review is part of a larger study on Child Labour in Yemen covering, inter alia, more detailed statistical analysis of several data sets and an assessment of the situation of child labour. These are undertaken in parallel to the present review; possible overlap and duplications are to be ironed out in the final stages which includes the preparation of a consolidated paper. Nepal UCW Country Report Abstract: The current report as part of UCW Project activities in Nepal. It provides an overview of the child work phenomenon in the Kingdom ­ its extent and nature, its determinants, its consequences on health and education, and national responses to it. The report serves two important UCW Project objectives in the country. First, it helps provide a common analytical understanding of child work, that can be used to inform the current activities of the three partner agencies and Government, and that can be used to develop joint interagency strategies for future cooperation with the Government in addressing child labour. Second, through close involvement of local counterparts in its development, the report contributes to a broader effort to build national capacity in analysing and using data on child work for policy development. The paper is the product of a collaborative effort involving local researchers, the three implementing partner 48 Annex C agencies and the UCW core team. The statistical information presented in the review is drawn primarily from the Nepal Labour Force Survey conducted in 1998/99 (NLFS 1998/99) and the Nepal Living Standards Monitoring Study conducted in 1995/96 (LSMS 1995/96). The first involved a stratified sample of 14,400 households and the second a stratified sample of 19,200 households. A series of five ILO-IPEC sponsored rapid assessments is the primary source of information on worst forms of child labour. The review also draws on a number of smaller-scale studies, qualitative as well as quantitative, Government and NGO reports, agency documents and other information sources. 2. Bibliographic Database and Report. B.Sharma. January 2003. Abstract: A comprehensive report listing all available research reports, data sets and statistical information concerning child labour, health and nutrition, education and poverty, migration and trafficking. An annotated list of research relating to Nepalese children and child labourers is also provided. 3. Child Labour in Nepal: An Analysis of the Determinants of Child Labour in Nepal, the Policy Environment and Response. B. Gilligan. January 2003. Abstract: This paper analyzes and summarizes the current understanding of child labour in Nepal, its determinants, and existing donor and civil society policy recommendations submitted to His Majesty's Government. It proposes, as a step towards implementing these recommendations, a series of low cost and shorter-term action recommendations. It is the wish of the cooperating agencies of the UCW Nepal Project, that in beginning a dialogue with His Majesty's Government on these possible actions, that a way can be found to quickly advance the common agenda to eliminate child labour. El Salvador Country Report Abstract: The report provides an overview of the child work phenomenon in El Salvador ­ its extent and nature, its determinants, its consequences on health and education, and national responses to it. The report serves two important UCW Project objectives in the country. First, it helps provide a common analytical understanding of child work, that can be used to inform the current activities of the three partner agencies and Government, and that can be used to develop joint interagency strategies for future co-operation with the Government in the field of child work. Second, through close involvement of local counterparts in its development, the paper contributes to a broader effort to build national capacity in using data for policy development in the child labour field. The report is the product of a collaborative effort of the National Statistical Institute (DIGESTYC), concerned Government ministries and ILO, UNICEF and the World Bank. The statistical information presented in the report is drawn primarily from a national household survey conducted in 2001 (Enquesta de Hogares de Propositos Multiples (EHPM 2001). The survey is National representative and involved a stratified sample of 11,953 households and a total of 16,253 children in the age group 5-17. REPORTS TO BE COMPLETED 1. Comparative Review of Household Survey Data Collected for over 45 Countries. S. Lyon, L. Guarcello. Abstract: Household surveys, conducted chiefly through ILO, UNICEF and World Bank survey programmes, are offering new insights into the phenomenon of child work in the developing world. This paper brings together and reviews household survey data from 45 countries, in an attempt to draw lessons concerning key dimensions of the child work phenomenon ­ its extent, its nature, its impact on health and education, and factors associated with it. These household survey data show that child work exists in all countries examined, but with large variation in levels of prevalence. Boys are generally slightly more likely to work than girls, though this does not hold for all countries. Very young children (5-9 years) constitute a substantial part of the overall working children population in most countries. The family is the most common work context; relatively few children, on the other hand, work for wages in most countries. Long 49 Annex C working hours appear to be the norm for many working children, including for many of those who are also attempting to attend school. There is no clear evidence of work having harmful health effects, but work does appear to interfere with schooling in many countries. Child work appears positively correlated with children's age, and negatively correlated with family incomes and the education level of parents. 2. The Assessment of Commercial Sexual Exploitation of Children: A Review of Methodologies. B. Henschel. Abstract: Commercial Sexual Exploitation of children and youth is a global phenomenon of growing proportion. The invisibility and defencelessness of the victims, as well as the lack of public debate and responses from the various States to commercial and non-commercial sexual exploitation of children, are some of the most serious features of this present form of exploitation. To combat and prevent Commercial Sexual Exploitation of Children, the nature, extension and causes of this phenomenon must be identified. Besides, the mixture of social, political and economic factors that make children more vulnerable to CSE must be understood as well as the motivation of adults to have sexual relations with children. The illegal, hidden or invisible property of CSEC makes it extremely difficult to gather reliable data on this phenomenon. Hence, following the literature a total absence of econometric work is confirmed. To overcome the lack of qualitative and quantitative information on the commercial sexual exploitation of children new research approach and non-traditional research methods must be elaborated and applied. The aim of this paper is to review all existing attempts to assess and quantify commercial sexual exploitation of children. 3. Long Term Effects of Children's Work on Health. R. Straub, F. Rosati. Abstract: The diversity of potential relationships between child labor and health makes the empirical disentanglement of the causal relationship a difficult exercise. This paper examines the long run impact of child labour on health with and without controls for unobserved household specific characteristics. In order to control for the unobserved households specific effect, we estimate a conditional fixed effect model using data on siblings constructed from the Guatemalan National Survey of Living Condition. The estimation results reinforce the relationship found in the standard logit set up that child labor is harmful for health in the long run. Our results can be interpreted as a lower bound of the true impact since healthier children are most likely to offer themselves for employment and to be appointed. 4. Key Findings of Project Mapping Exercise (ILO/IPEC). I. Kaur. Abstract: As part of UCW "Project Mapping and Analysis" activity, the review of ILO/IPEC program provides an overview of ILO/IPEC operations in more than 75 countries. The paper is guided primarily by UCW principle of promoting transparency and sharing information. Specifically, the paper reviews IPEC's interventions that address child labor while achieving IPEC's strategic and operational objectives. This paper also serves as reference guide to the contents of IPEC's Project Database that includes detailed information on IPEC project interventions by thematic groups, product lines and sectoral categories. To further substantiate the findings of this review, UCW Project will facilitate to find synergies among the approaches used by three agencies, and help identify gaps and overlaps in ongoing activities to minimize duplication of efforts in future activities on child labor. The Puzzle of 'Idle' Children: Neither in School nor performing Economic Activity: Evidence from six Countries. M. Biggeri, L. Guarcello, S. Lyon, F. Rosati. Abstract: This paper presents a set of descriptive statistics on the observed group of children that neither attends school nor performs economic activity. Drawing on datasets from six countries, evidence is provided suggesting that children can be absent from both school and economic activity because they are needed to perform household chores, because of their health, or because they are unable to find work after having left school. But a large proportion of children not in school or economic activity does not fall into any of these categories. A simple theoretical model of household decisions concerning children's time allocations is presented in an attempt to account for this 'unexplained' portion of the group of children absent from school and economic activity. 50 Annex C 5. Child Labour and Access to Basic Services: Evidence from five Countries. L. Guarcello, S. Lyon, F. Rosati. Abstract: Analyses of the determinants of child labour have largely neglected the role of access to basic services. The availability of these services can affect the value of children's time and, concomitantly, household decisions concerning how this time is allocated between school and work. This paper investigates the link between child labour and water and electricity access in five countries ­ El Salvador, Ghana, Guatemala, Morocco and Yemen. Employing an econometric methodology based on propensity scores for dealing with the potential endogeneity of access to water and electricity, average treatment effects for water and electricity access on children's activities are presented. The marginal effects of water and electricity access on children's activities obtained by estimating a bivariate probit model are also examined. Finally, a sensitivity analysis is presented designed to check the robustness of the conclusions concerning the causal relationship between water and electricity access and children's activities. 6. Child Work in Sub-Saharan Africa: An Overview. Abstract: The paper will aim to give an overview of the characteristics of child work in Sub-Saharan Africa using a variety of household surveys and definitions. It will consider both economic activities and household chores as part of children's work. It will also compare variation in children's activities and their characteristics across countries and relate them to differences economic and social characteristics. 7. Child Work in Latin America: An Overview. Abstract: The paper will aim to give an overview of the characteristics of child work in Latin American Countries using a variety of household surveys and definitions. It will consider both economic activities and household chores as part of children's work. It will also compare variation in children's activities and their characteristics across countries and relate them to differences economic and social characteristics. 8. Millennium Development Goals and Child Labour. B. Grimsrud. Abstract: Child labour is maybe one of the most striking indicators identifying vulnerable children and as such pointing to shortcomings in several of the millennium goals as poverty eradication, education for all, gender equality, combating HIV/AIDS and creation of a global partnership for development. Most working children do so after a decision in their parental household. To understand the household labour supply decisions, relations to the labour market and to public interventions is critical in designing programmes in order to achieve the MDG. The research on child labour represent in this respect a largely untapped resource of knowledge for policymakers in the fields of education programme and poverty reduction programmes. This paper focus particularly on the links between child labour and the labour market and argue that thought the links in the household between child and adult labour supply the Basu-Van model on dual equilibriums are relevant beyond labour markets where children participate directly. In regards to child labour and education the lack of education opportunities effect on child labour is well documented, but existence of widespread child labour also reduces the effectiveness of investment in education. This interlinks between the MDG and child labour could be the basis for win-win strategies for development. 9. Effects of Child Work on Children's Health: A review of the literature and of the available indicators. P. Roggero. Abstract: A comprehensive analysis of the existing public health and epidemiological literature which highlights and explores the effects of child work on children's health. Available survey instruments normally contain information on illnesses/injury incurred during the recent past, indications on the kind of illnesses/injury experienced and anthropometric indicators. The information on the kind of illness/injury is seldom if at all used in assessing the effects of child work on health. This report analyses the possibility of using such information to produce new indicators of the health status of the child and provides the necessary medical background information for a sound use of such indicators. 10. Vulnerability to Child Work. L. Guarcello, F. Mealli, S. Pudney, F. Rosati. Abstract: The concept of vulnerability to poverty has recently been developed and applied in the evaluation of the country situation and to formulate policies. Vulnerability comes from the notion that certain groups in society are more vulnerable and therefore more exposed to a variety of risks. In this 51 Annex C context, the concept of consumption has represented the main dimension of household and individual welfare but in terms of risk analysis, the investments in health and education as well as natural or household shocks become important factors which enable the identification of vulnerable groups. In fact, low investments in education and in health are likely to maintain the household in a situation of vulnerability. In this paper we will try to extend the concept and the measure of vulnerability to child work, particularly concerning the probability of children becoming workers and quitting school. Preliminary application of the concept and definition identified will be tested using a few suitable data sets. 11. The Impact of Conditional Cash Transfer Programmes on Indigenous People's Progress in School and Work Activities: The Case of Progresa/Oportunidades. L.F. López-Calva. Abstract: The indigenous people of Latin America live in extreme poverty and their children work more than do non-indigenous children. There are over 40 million indigenous people in Latin America, or about 8 percent of the region's population. The objective of the study is to assess whether the conditional cash transfer programmes Progresa/Oportunidades (that are widely adopted or considered for implementation in the region) have substantial effects on indigenous children's labour supply and school attendance. The determinants of schooling and work will be modeled and the differential impact of Progresa ­ conditional cash transfers tied to school attendance ­ on indigenous and non-indigenous children will be determined. Appropriate data sets ­ ENCASEH and/or ENCEL ­ will be used. 12. Does child labor reflect prior investments in children's pre-school human capital, IFPRI. Abstract: It is increasingly understood that children malnourished as pre-schoolers perform poorly when they start school. They have poorer psychomotor development and fine motor skills, lower activity levels, and have less interaction with their environments. We hypothesize that such children are less likely to be enrolled in school at an appropriate age or are more likely to repeat grades or drop out and become full- time farm laborers. If this is the case, then the policy objective of reducing child labor may be attained, in part, by increased attention to child development in the pre-school years. This analysis will exploit the fact that anthropometric data (heights and weights) of children were collected in the 1994/95 and 1997/98 survey rounds. These can be linked to subsequent observations on child schooling and child labor in the 1997/98 and 2000 surveys. The analysis will begin with descriptive data but will largely focus on an econometric analysis. There are two stages. First, we would examine the determinants of child anthropometric status, making use of data on transitory and covariant shocks (such as rainfall; loss of assets through theft or (in the case of animals) death and so on) safety nets, parental resources, and public investments such as schools and health care. Second, we examine the relationship between past measures of child anthropometry and measures of schooling and child labor. Note that in this analysis, child anthropometric status is itself an outcome and the estimation will need to take this into account. Fortunately, these data are rich in information on shocks and these events can be used as instrumental variables. 13. The roles played by shocks, private and public resources, wages and prices as determinants of future child labor and schooling, IFPRI. Abstract: It is increasingly understood that child labor is the outcome of a complex series of factors including household wealth, the ability to withstand shocks, public resources and market factors such as wages and prices. While there is a growing literature on this topic, it is dominated by studies that focus on the influence of current factors on current child labor outcomes. By contrast, this work focuses on whether current child labor outcomes may reflect past events including shocks, parental investments, public resources and safety nets. It would exploit the existence of multiple survey rounds in order to link past events to current outcomes. Further it focuses on an older age group ­ children of primary-school age. 14. Is child labor associated with subsequent early marriage, IFPRI. Abstract: Early marriage by girls is associated with pregnancy at an earlier age and higher rates of fertility. Both are problematic in that early pregnancies may result in children born with lower birth rates as well as complications at birth. If child labor is associated with subsequent early marriage, interventions that delay marriage carry benefits both for females (for example, in terms of lowered risks of maternal mortality) as well as for their children. In this sense, reducing child labor can be seen as part of a wider strategy to break the intergenerational transmission of poverty. 52 Annex C 15. Psychosocial impacts of child work: a framework for research, monitoring and intervention. M. Woodhead. Abstract: The paper outlines conceptual frameworks for assessing the multiple ways that work can impact (both positively and negatively) on children's well-being; and for identifying psychosocial indicators of impact. The paper draws attention to ways that the context of children's work mediates how far potential hazards constitute a risk to children and concludes by outlining theoretical models that link characteristics of work to evidence of impact on children's psychosocial well-being. 16. Household Chores and Child Health: Evidence from Six Countries. F. Francavilla, L. Guarcello, S. Lyon. Abstract: This paper uses survey datasets from six countries in an attempt to shed light on the relationship between involvement in household chores and child health. The descriptive statistics presented in the paper reveal no clear correlation between household chores and health in the six countries. Children spending at least four hours daily on household chores are not clearly worse-off health-wise than children without chores responsibilities, and children spending more time on chores actually appear better-off health-wise than children for whom household chores constitute only a relatively small time burden. Some types of chores appear to have more impact on health status than others, but the variation in reported illness by chore type is generally quite small. It is argued that these results are primarily a reflection of shortcomings in the measurement of the chores-health link. The simple measures of child health employed in the paper ­ reported illness and Body Mass Index ­ fail to account for the dynamic nature of the relationship between chores involvement and health, and for the potential endogeneity of chores involvement to health outcomes.