GLOBAL 39473 MONITORING REPORT 2007 Millennium Development Goals Confronting the Challenges of Gender Equality and Fragile States GLOBAL MONITORING REPORT 2007 GLOBAL MONITORING REPORT 2007 Millennium Development Goals: Confronting the Challenges of Gender Equality and Fragile States © 2007 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 5 10 09 08 07 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not neces- sarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judge- ment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant per- mission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522- 2422; e-mail: pubrights@worldbank.org. ISBN-10: 0-8213-6975-X ISBN-13: 978-0-8213-6975-3 eISBN-10: 0-8213-6976-8 eISBN-13: 978-0-8213-6976-0 DOI: 10.1596/978-0-8213-6975-3 Cover art: "Three Faces," 2004, acrylic on paper, 4 1/8" x 11 5/8", by Eria Solomon Nsubuga, Uganda; World Bank Art Program PN: 646517. Courtesy of the artist and the World Bank Art Program. Cover design: Quantum Think, Philadelphia, Pennsylvania, United States Typesetting: Precision Graphics, Champaign, Illinois, United States Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Millennium Development Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviii Report Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Millennium Development Goals--Charting Progress. . . . . . . . . . . . . . . . . 21 1 Growth, Poverty Reduction, and Environmental Sustainability . . . . . . . . . 39 2 The Role of Quality in MDG Progress . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 3 Promoting Gender Equality and Women's Empowerment . . . . . . . . . . . . 105 4 Aid, Debt Relief, and Trade: Making Commitments Work . . . . . . . . . . . 149 5 Monitoring the Performance of International Financial Institutions . . . . 187 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 Statistical Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 1 Global Monitoring Report 2007: Five key messages . . . . . . . . . . . . . . . . . . . 3 1.1 Definition of fragile states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 1.2 Current issues in the environment debate . . . . . . . . . . . . . . . . . . . . . . . . . . 56 1.3 Gender and the environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 1.4 Adjusting saving rates to reflect a wider range of assets . . . . . . . . . . . . . . . 60 2.1 Early evidence that the EFA Fast Track Initiative is making a difference . . . 70 2.2 Fast progress on child mortality in Eritrea . . . . . . . . . . . . . . . . . . . . . . . . . 72 2.3 Preventing maternal mortality: Findings from three countries. . . . . . . . . . . 74 2.4 Measuring health progress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 2.5 Innovative new financing mechanisms for health are getting off the ground . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 2.6 Managing aid for health in Rwanda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 2.7 Rebuilding health services after conflict: Strategies from Timor-Leste and Afghanistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 2.8 Contracting for health services in Cambodia . . . . . . . . . . . . . . . . . . . . . . 102 3.1 "Good" parity levels may hide huge enrollment challenges. . . . . . . . . . . . 117 3.2 Sex ratios at birth and removing unwanted daughters in East Asia and South Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 3.3 Beyond participation: Self-employment, informality, and household work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 3.4 How Cambodia's Ministry of Women's Affairs addresses the MDG3 challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 3.5 Gender-informed public finance management . . . . . . . . . . . . . . . . . . . . . . 144 4.1 Accounting for debt forgiveness in ODA statistics . . . . . . . . . . . . . . . . . . 153 4.2 Country-based scaling up: The case of Ghana. . . . . . . . . . . . . . . . . . . . . . 158 4.3 Predictability of budget aid: Experience in eight African countries . . . . . . 167 4.4 Debt service savings and social expenditures: Is there a link? . . . . . . . . . . 172 4.5 Developing-country clothing exports in a postquota world. . . . . . . . . . . . 174 4.6 Economic Partnership Agreements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 5.1 The IMF's medium-term strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 5.2 The World Bank's Africa Action Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 5.3 The World Bank's framework on clean energy . . . . . . . . . . . . . . . . . . . . . 192 5.4 How well does the World Bank contribute to development effectiveness? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 5.5 Lending by the IMF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 5.6 The Marrakech Action Plan for Statistics and the Accelerated Data Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 5.7 Are the MDBs focusing on results? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 1 Learning levels of primary school­aged children . . . . . . . . . . . . . . . . . 7 2 Pathways from increased gender equality to poverty reduction and growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 3 Progress in official indicators of gender equality and women's empowerment, by region, 1990­2005 . . . . . . . . . . . . . . . . . . . . . . . . 11 1.1 Progress toward the poverty MDG target 1990­2004, and a forecast for 2015 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 1.2 Rates of extreme poverty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 1.3 Regulatory reforms can increase efficiency and reduce corruption. . . 50 1.4 Trajectories of Governance Improvements. . . . . . . . . . . . . . . . . . . . . 52 1.5 Adjusted net saving rates by region . . . . . . . . . . . . . . . . . . . . . . . . . . 61 1.6 Environment and overall CPIA score by region and income group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 2.1 Most out-of-school girls are "doubly disadvantaged": Female and from minority groups . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 2.2 Global HIV/AIDS epidemic, 1990­2006 . . . . . . . . . . . . . . . . . . . . . . 75 2.3 Development assistance for education and health, 2000­05 . . . . . . . 82 2.4 Developing countries are devoting more national resources to education and health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 2.5 Child mortality is higher and showing less progress in fragile states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 2.6 Primary completion rates are lower in fragile states, but improving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 2.7 Measles immunization in fragile states remains lower . . . . . . . . . . . . 89 2.8 A growing gap in access to improved water. . . . . . . . . . . . . . . . . . . . 89 2.9 Child mortality progress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 2.10 Measles vaccines are reaching the poor in many countries. . . . . . . . . 93 2.11 Primary completion progress is benefiting the poor in many countries, but not all . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 2.12 Reading and math performance on the OECD PISA Exams, 2000 and 2003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 2.13 Many children do not attain minimum learning levels. . . . . . . . . . . . 97 2.14 The quality of health care is not just a function of doctors' training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 3.1 Gender equality, domains of choice, and economic performance: A framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 3.2 Women's earnings, children's well-being, and aggregate poverty reduction and economic growth--The pathways . . . . . . . . 109 3.3 Progress in girls' enrollment rates between 1990 and 2005 . . . . . . . 112 3.4 Trends in gender parity in enrollment and literacy rates, 1990 and 2005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 3.5 Average youth literacy rates in Africa conceal rural-urban disparities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 3.6 Bolivia has a gender gap in schooling among indigenous children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 3.7 Progress in share of women in nonagricultural wage employment and proportion of seats in parliament held by women, by region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 3.8 Share of women in nonagricultural wage work by ethnicity . . . . . . 119 3.9 Female under-five mortality rate and female-to-male ratio, 2004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 3.10 Trend in adolescent motherhood . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 3.11 Female and male labor force participation rates by region, 1990­2005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 3.12 Comparison of country scores on official and expanded MDG3 indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 3.13 Changes in official MDG3 indicators for countries in the bottom and top quintiles, 1990­2005 . . . . . . . . . . . . . . . . . . . . . . . 135 3.14 Changes in two proposed indicators for countries in the bottom and top quintiles, 1990­2005 . . . . . . . . . . . . . . . . . . . . . . . 139 4.1 Evolution of aid: 1990­2006 and prospects . . . . . . . . . . . . . . . . . . 151 4.2 Expansion in ODA is concentrated in a few countries . . . . . . . . . . . 152 4.3 Evolution of Net ODA to SSA, 1990­2005 . . . . . . . . . . . . . . . . . . . 156 4.4 Gender equality focus of bilateral ODA by sector (2001­05) . . . . . 159 4.5 Quality of policy matters: Distribution of 2004­05 DAC bilateral ODA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 4.6 Sharper donor focus on policy and need . . . . . . . . . . . . . . . . . . . . . 161 4.7 Fragile states receive more of their aid in the form of debt relief and humanitarian assistance . . . . . . . . . . . . . . . . . . . . . . . . . . 163 4.8 Aid per capita to fragile states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 4.9 Quality of country public financial system and use of PFM system for aid to government sectors . . . . . . . . . . . . . . . . . . . 165 4.10 DAC members' and EC's ODA commitments for GPGs . . . . . . . . . 169 4.11 Reduction of debt stock (NPV terms) for the 30 decision-point countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 4.12 Overall trade restrictiveness has declined (2000­06) . . . . . . . . . . . . 177 4.13 Aid for trade is rising. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 5.1 Net private capital flows to developing countries . . . . . . . . . . . . . . 196 5.2 Concessional and nonconcessional lending by MDBs, 1999­2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 5.3 Nonconcessional lending by MDBs to different regions (gross disbursements), 1999­2006. . . . . . . . . . . . . . . . . . . . . . . . . . 198 5.4 Gross disbursements of concessional lending by MDBs, 1999­2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 5.5 Donor financing commitments to IDA under the Multilateral Debt Relief Initiative, as of December 31, 2006 . . . . . . . . . . . . . . . 199 5.6 Policy and poverty selectivity of concessional assistance by MDBs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 5.7 Grants and loans as shares of MBD concessional disbursements in 2006. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 1 Recommended additional indicators for MDG3 . . . . . . . . . . . . . . . . . . . . . 12 2 Fragile states face the largest deficit in most MDGs . . . . . . . . . . . . . . . . . . 13 1.1 Impact of growth of GDP per capita on poverty . . . . . . . . . . . . . . . . . . . . . 42 1.2 Per capita GDP growth for high-, middle- and low-income countries . . . . . 43 1.3 Real per capita growth and investment and savings rates of fragile and nonfragile states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 1.4 Macroeconomic indicators for low-income countries . . . . . . . . . . . . . . . . . 47 1.5 Quality of macroeconomic policies in low-income countries, 2006. . . . . . . 48 1.6 Status of "finalized" PEFA assessments (as of February 23, 2007) . . . . . . . 55 1.7 Key indicators of environmental sustainability . . . . . . . . . . . . . . . . . . . . . . 59 1A.1 Share of people living on less than $1.08 a day. . . . . . . . . . . . . . . . . . . . . . 65 1A.2 Share of people living on less than $2.15 a day. . . . . . . . . . . . . . . . . . . . . . 65 1A.3 Number of people living on less than $1.08 a day . . . . . . . . . . . . . . . . . . . 65 1A.4 Number of people living on less than $2.15 a day . . . . . . . . . . . . . . . . . . . 65 2.1 Several low-income countries are making strong progress on universal primary completion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 2.2 Progress on child mortality in a few countries. . . . . . . . . . . . . . . . . . . . . . . 71 2.3 Progress in assisted births . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 2.4 Use of insecticide-treated bednets by children under five. . . . . . . . . . . . . . . 78 2.5 TB incidence trends by region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 2.6 Changes in TB incidence, 1990­2004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 2.7 Access to improved water is growing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 2.8 Access to improved sanitation is growing . . . . . . . . . . . . . . . . . . . . . . . . . . 81 2.9 Fragile states lag most on MDGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 3.1 Official indicators for MDG3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 3.2 Regional performance in attaining the primary and secondary enrollment target by 2005. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 3.3 Recommended additional indicators for MDG3 . . . . . . . . . . . . . . . . . . . . 120 3.4 Girls lag behind boys in primary school completion rates in most regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 3.5 Sources of death and disability with largest gender differentials in disease burden for 15- to 29-year-olds, low- and middle-income countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 3.6 Trends in modern contraceptive use, selected countries. . . . . . . . . . . . . . . 126 3.7 Prospective indicators for which data are not currently available . . . . . . . 132 3.8 Countries in the top and bottom quintiles, according to scores on official MDG3 indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 3.9 Countries in the top and bottom quintiles, according to primary completion rates and under-5 mortality . . . . . . . . . . . . . . . . . . . . . . . . . . 137 3.10 Countries in the top and bottom quintiles, according to labor force participation rates and disability-adjusted life years . . . . . . . . . . . . . 138 4.1 Indicators pertaining to bilateral donors' implementation of the Paris Declaration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 4.2 Trade restrictiveness and its impact on welfare and trade flows, by country income group, 2004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 4.3 Market access (OTRI). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 5.1 Indicators pertaining to MDB implementation of the Paris Declaration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 5.2 IFI reforms to strengthen response in fragile states . . . . . . . . . . . . . . . . . . 208 Foreword T he 2007 Global Monitoring Report completion rate has increased from 78 per- takes stock of progress toward cent in 2000 to 83 percent in 2005 and the achieving the Millennium Develop- pace of increase has accelerated in all regions ment Goals and assesses the contributions (except Latin America and the Caribbean, of developing countries, donor nations, and where levels were already high). the international financial institutions as they Aid quality and effectiveness are improv- work toward meeting commitments under ing: signatories to the 2005 Paris Declaration the 2002 Monterrey consensus. This fourth on aid effectiveness are monitoring progress annual GMR finds both areas of progress and on harmonization, alignment, and managing gaps where far greater effort is required. This for development results. Still, many chal- year's special topics--achieving gender equal- lenges remain in accelerating the implemen- ity and addressing the problems of fragile tation of the Paris Agenda. states--highlight two particular areas where Strengthening future performance will serious challenges confront the international require greater attention in two important community. areas. The first relates to gender equality and The GMR presents striking evidence of real the empowerment of women. Second is the progress on the MDG agenda in several areas. condition of fragile states, where nearly 500 Globally, rapid growth is translating into fall- million people, or nearly one-fifth of all peo- ing levels of extreme poverty: in the five years ple in low-income countries, reside. between 1999 and 2004 global poverty fell Gender equality and the empowerment by nearly 4 percentage points, lifting an esti- of women are important for basic reasons-- mated 135 million people out of destitution. fairness, equality of opportunity, and eco- Sub-Saharan Africa's performance has also nomic well-being. Increasing efficiency and been encouraging over this period; the share achieving the full potential of men and women of extreme poor fell by nearly 5 percentage alike is a precursor to prosperity. Gender points, although the absolute number of poor equality is also vital to advancing the other has not fallen: Sub-Saharan Africa remains millennium goals--halving poverty, achiev- the poorest developing region in the world ing primary education for all, and lowering with about two-fifths of its people living on the under-five mortality rate. Achieving equal less than US$1 a day. opportunity for women will require greater Significant gains are occurring in human accountability among donors, developing development: globally the primary school countries, and international institutions such as our own. It will entail moving beyond in delivering on the promises of the Monter- a general institutional call for attention, rey Summit in 2002, or the 2005 Gleneagles toward a concrete strengthening of programs commitments to scale up aid to Africa. Cur- and project implementation. This would in rent examples of countries that have received turn allow a focus on outcomes as well as on significantly scaled-up aid to help finance a longer-term agenda. Such a shift requires sound programs to meet the MDGs are improving monitoring systems for tracking few and far between. This is not for lack of progress in gender equality, and evaluating opportunity, which abounds at the project-, the impact of interventions aimed at attain- sector-, and country-levels. Rather, the dearth ing equality of opportunity. of successful scaling-up efforts points to the Fragile states, with their limited capacity need for the greater "mutual accountability" to secure a better life for their citizens, merit called for under the Monterrey consensus. special attention because of the enormity of First, we need to identify and fund existing the problems they face. These countries by opportunities for scaling up based on cur- definition have weak governments and are rent knowledge and capacity, such as in the hard put to deliver basic services to their country and sector areas that the World Bank people. Over one-fourth of extremely poor and the UN have identified. Second, we must people in developing countries live in frag- work together to develop a dynamic strategy ile states. These nations face enormous chal- for country-based opportunities to sequen- lenges, regarding both how to take action to tially scale up, including with sufficient tech- meet human development needs, and how nical assistance from our two organizations, to stave off a potential downward spiral of working together with other development conflict, human abuse, and refugee flight. partners. This will require that donor coun- New instruments should be considered to tries fulfill their pledges to strengthen their help countries that have turned the corner to development strategies and that they put real quickly stabilize, restore both security and resources to work to enact these programs. basic services, and bring greater accountabil- Deadlines to deliver on promises in 2010, ity into public service. This will require better 2015, and 2030 are looming large and, col- coordination and more effective intervention lectively, we need to speed up investments in by the international community. projects and reform programs that will save To move both of these crucial agendas lives, create jobs, and promote growth. The forward and to secure faster progress toward responsibility for this lies with donors, our meeting the MDGs, international efforts to own and other institutions, and recipients scale up aid for deserving country programs alike. are vital. We have not made sufficient progress Paul Wolfowitz Rodrigo de Rato President Managing Director World Bank International Monetary Fund Acknowledgments T his report has been prepared jointly Carlos Primo Braga, Kirk Hamilton, Julia by the staff of the World Bank and Nielson, Bernard Hoekman, Sarah Cliffe, the International Monetary Fund. Sima Kanaan, Adam Ross, Giovanni Ruta, There was also consultation and collabora- and Juan Carlos Guzman. tion with other partner institutions including Many others have made valuable contri- the Organisation for Economic Co-operation butions, including the following from the and Development and its Development Assis- World Bank: tance Committee, several United Nations Aban Daruwala, Ajay Chhibber, Akihiko agencies, the African Development Bank, Nishio, Alessandro Nicita, Amie Batson, Ana the Asian Development Bank, the European Cristina Torres, Anand Rajaram, Arthur Kar- Bank for Reconstruction and Development, lin, Arunima Dhar, Bee Ean Gooi, Bonaventure the Inter-American Development Bank, and Mbida-Essama, Brian Levy, Brice Quesnel, the Islamic Development Bank. The coopera- Christopher Hall, Claudia Paz Sepulveda, tion and support of staff of these institutions Desmond Bermingham, Dhushyanth Raju, are gratefully acknowledged. Dianne Garama, Dorte Domeland-Narvaez, Mark Sundberg was the lead author and Doug Hostland, Eduard Bos, Elizabeth White, manager of the report. The work was carried Emi Suzuki, Eric Swanson, Robert Francis out under the general guidance of François Rowe, Gauresh Rajadhyaksha, Gary Milante, Bourguignon, Senior Vice-President, and the Gilles Bauche, Gisela Garcia, Hiau Looi Kee, overall supervision of Alan Gelb, Director, Jessica Lynn Ebbeler, Joseph Naimoli, Joy De World Bank. Beyer, Kavita Watsa, Louise Cord, Luc-Charles The core team for the report included Gacougnolle, Lucia Fort, Martin Ravillion, Peter Fallon (chapter 1); Barbara Bruns Mary Hallward-Driemeier, Maureen Lewis, (chapter 2); Mayra Buvinic, Elizabeth King, Meera Shekhar, Michael Koch, Monica Das Andrew Morrison, and Nistha Sinha (chap- Gupta, Nevin Fahmy, Nina Todorova Budina, ter 3); Punam Chuhan (chapter 4); and Ste- Olusoji Adeyi, Pablo Gottret, Patrick Grasso, fano Curto (chapter 5). The Bank core team Prem Sangrula, Rekha Mehra, Rene Bonnel, also included Brendan Fitzpatrick, Julien Reynaldo Martorell, Rifat Hasan, Robert Gourdon, and Sachin Shahria, who provided Watson, Robin Horn, Roula Yazigi, Shah- research, written inputs, and coordination to rokh Fardoust, Shaohua Chen, Shunalini the overall report. Other significant contri- Sarkar, Soe Lin, Sulekha Patel, Susan Stout, butions were made by Amar Bhattacharya, and Waafas Ofosu-Amaah. Contributors from the International Mon- received from the World Bank and Interna- etary Fund included: tional Monetary Fund management and staff Alun Thomas, Ben Umansky, Carlo Sdral- in the course of the preparation and review evich, Emmanuel Hife, Marie-Helen Le Man- of the report. chec, Mark Plant, and Scott Brown. The World Bank's Office of the Publisher Contributors from other institutions managed the editorial services, design, pro- included: duction, and printing of the book--in par- Kazu Sakai, Christopher MacCormac, ticular Susan Graham, Denise Bergeron, Manju Senapaty, Patrick Safran (ADB), Phi- Aziz Gökdemir, Nancy Lammers, Stephen libert Afrika, Ferdinand Bakoup, Maurice McGroarty, Randi Park, Santiago Pombo- Mubila, Penthesilea Lartey (AfDB), Ernesto Bejarano, and Janice Tuten, along with Castagnino, Marco Ferroni, Max Pulgar-Vidal, Kirsten Dennison and associates at Precision Afredo Garcia (IADB), Samuel Fankhauser, Graphics, Candace Roberts and associates at James Earwicker (EBRD), Brian Hammond, Quantum Think, Gerry Quinn, Bruce Ross- Patricia O'Neill (OECD-DAC), and Chan- Larson, and Michael Treadway, provided drika Bahadur (UNDP). excellent help with publishing this book on a Guidance received from the Executive very tight schedule. Directors of the World Bank and their staff, Finally, acknowledgments are due to and the International Monetary Fund during Matthew Burke and Maria Del Carmen Cosu discussions of the draft report is gratefully of the World Bank Art Program, who helped acknowledged. The report has also benefited us find the artwork for the cover. from many useful comments and suggestions Abbreviations AAP Africa Action Plan HIPC heavily indebted poor country/ ACP African, Caribbean, and Pacific countries countries HIV human immunodeficiency virus ADB Asian Development Bank IADB Inter-American Development AfDF African Development Fund Bank AsDF Asian Development Fund IBRD International Bank for AIDS acquired immunodeficiency Reconstruction and Development syndrome ICS Investment Climate Surveys AMC Advance Market Commitment IDA International Development ART antiretroviral treatment Association (of the World Bank COMPAS Common Performance Group) Assessment System IFAD International Fund for CPIA Country Policy and Agricultural Development Institutional Assessment IFC International Finance DAC Development Assistance Corporation (of the World Committee (of the OECD) Bank Group) DALY disability-adjusted life year IFFIm International Finance Facility DB Doing Business (surveys) for Immunization DHS Demographic and Health IFI international financial Survey institution EBRD European Bank for IMCI integrated management of Reconstruction and childhood illness Development IMF International Monetary Fund EC European Commission LIC low-income country EFA-FTI Education for All-Fast Track LICUS low-income countries under Initiative stress EPA Economic Partnership MDB multilateral development bank Agreement MDG Millennium Development Goal ES Enterprise Survey MDRI Multilateral Debt Relief FAO Food and Agriculture Initiative Organization (of the UN) NGO nongovernmental organization G-8 Group of Eight NTM nontariff measure GII Global Integrity Index ODA official development assistance GMR Global Monitoring Report OECD Organisation for Economic GNI gross national income Co-operation and Development OOF other official flow TRI trade restrictiveness index OTRI overall trade restrictiveness UNDP United Nations Development index Programme PEFA Public Expenditure and UNESCO United Nations Education, Financial Accountability Scientific and Cultural PFM public financial management Organization PIU project implementation unit UNFPA United Nations Population PRSP Poverty Reduction Strategy Fund Paper UNICEF United Nations Children's Fund R&R results and resources USAID United States Agency for SDR special drawing right International Development STD sexually transmitted disease WHO World Health Organization SWAp sectorwide approach WSS water supply and sanitation TB tuberculosis WTO World Trade Organization TARGET 1 Halve, between 1990 and 2015, the proportion of people whose income is less than $1 a day TARGET 2 Halve, between 1990 and 2015, the proportion of people who suffer from hunger TARGET 3 Ensure that by 2015, children everywhere, boys and girls alike, will be able to complete a full course of primary schooling TARGET 4 Eliminate gender disparity in primary and secondary education, preferably by 2005, and at all levels of education no later than 2015 TARGET 5 Reduce by two-thirds, between 1990 and 2015, the under-five mortality rate TARGET 6 Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio TARGET 7 Have halted by 2015 and begun to reverse the spread of HIV/AIDS TARGET 8 Have halted by 2015 and begun to reverse the incidence of malaria and other major diseases TARGET 9 Integrate the principles of sustainable development into country policies and programs and reverse the loss of environmental resources TARGET 10 Halve by 2015 the proportion of people without sustainable access to safe drinking water and basic sanitation TARGET 11 Have achieved a significant improvement by 2020 in the lives of at least 100 million slum dwellers TARGET 12 Develop further an open, rule-based, predictable, nondiscriminatory trading and financial system (including a commitment to good governance, development, and poverty reduction, nationally and internationally) TARGET 13 Address the special needs of the least developed countries (including tariff- and quota-free access for exports of the least developed countries; enhanced debt relief for heavily indebted poor countries and cancellation of official bilateral debt; and more generous official development assistance for countries committed to reducing poverty) TARGET 14 Address the special needs of landlocked countries and small island developing states (through the Programme of Action for the Sustainable Development of Small Island Developing States and the outcome of the 22nd special session of the General Assembly) TARGET 15 Deal comprehensively with the debt problems of developing countries through national and international measures to make debt sustainable in the long term TARGET 16 In cooperation with developing countries, develop and implement strategies for decent and productive work for youth TARGET 17 In cooperation with pharmaceutical companies, provide access to affordable, essential drugs in developing countries TARGET 18 In cooperation with the private sector, make available the benefits of new technologies, especially information and communication Source: United Nations. 2000 (September 18). Millennium Declaration. A/RES/55/2. New York. United Nations. 2001 (September 6). Road Map towards the Implementation of the United Nations Millennium Declaration. Report of the Secretary General. New York. Note: The Millennium Development Goals and targets come from the Millennium Declaration signed by 189 countries, including 147 heads of state, in September 2000. The goals and targets are related and should be seen as a whole. They represent a partnership of countries determined, as the Declaration states, "to create an environment--at the national and global levels alike--which is conducive to development and the elimi- nation of poverty." Report Overview Yet in spite of this optimistic outlook, the Broad-based global economic growth in international community faces a much more 2006, and more generally since 2000, pro- demanding agenda in advancing the MDGs as vides grounds for optimism about progress 2015 draws nearer. Despite progress, nearly 1 in advancing the Millennium Development billion people remain in extreme poverty. All Goals (MDGs). For low-income countries, regions are off track to meet the target for real per capita income growth in Sub-Saha- reducing child mortality; nutrition is a major ran Africa and South Asia has been stronger challenge, with one-third of all children in in the period since 2000 than at any time developing countries underweight or stunted; since the 1960s, and stronger than at any time half the people in developing countries lack since transition in Europe and Central Asian access to improved sanitation. countries. Based on this strong growth per- Action to scale up development efforts formance, the estimated number of extremely needs to accelerate, but steps forward still poor people (living on $1 per day) fell by 135 appear tentative. Nearly seven years after million between 1999 and 2004. the Millennium Summit and five years after Although still uneven, progress with pov- the Monterrey summit, there has yet to be a erty reduction is evident across all regions. country case where aid is being significantly Sub-Saharan Africa reduced the share of peo- scaled up to support a medium-term pro- ple living in extreme poverty by 4.7 percentage gram to reach the MDGs. While there has points over five years to 41 percent, although been modest progress in Paris or Brussels or high population growth left the same absolute London to address the well-recognized prob- number of poor, at nearly 300 million. South lems in designing and delivering international Asia, Latin America, and East Asia all appear aid--proliferation of aid channels, weak to be roughly on track to halve extreme pov- coordination, lack of resource predictability, erty by 2015 from 1990 levels. Europe, Central misalignment with country strategies, and so Asia, and the Middle East and North Africa on--viewed from the capitals of Ethiopia, have largely eliminated extreme poverty. There Madagascar, or Bolivia, this progress appears are also hopeful signs that international devel- to be slow. opment efforts may be gaining momentum, This Global Monitoring Report (GMR) and new innovations in resource mobilization highlights two areas that require greater inter- for development are taking shape. national attention if higher global growth trends are to translate into sustainable devel- about environmental policies that are beyond opment outcomes and if the gains are to be the scope of this report but may be tackled in shared more evenly: future GMRs. Risks from failure to advance multilateral Gender equality. The first of these arises trade liberalization and expand market access from gender inequality and lost opportu- are also highlighted in this year's report. The nities for all people to help generate and Doha Round of trade negotiations was effec- participate in the gains from economic tively suspended in July 2006, but early in growth. The choice to focus the 2007 2007 there was an informal agreement to report on the third MDG--the promotion resume talks. Failure to make progress means of gender equality and empowerment of depriving many countries of vital opportuni- women--reflects a recognition by the inter- ties for accelerating their growth through national community that more is needed trade. to support equality for the half of human- To address these risks and advance the ity disadvantaged through less access than MDG agenda there is a pressing need for bet- men to rights (equality under the law), to ter aid coordination to strengthen aid quality resources (equality of opportunity), and to and scale-up assistance. This requires efforts voice (political equality). by all parties--donors, international financial Fragile states. The second risk arises institutions (IFIs), and developing countries. from the especially difficult development Agreement needs to be forged at the global challenges and greater needs facing frag- level on practical mechanisms and instruments ile states. Fragile states--countries with to scale up aid and on measures to reduce the particularly weak governance, institu- costs of aid fragmentation. Progress with tions, and capacity--comprise 9 percent scaling-up will require more and better aid of the developing world's population but resources (donors); sound, sequenced devel- over one-fourth of the extreme poor. They opment strategies (developing partners); bet- represent an enormous challenge: how ter technical support for strong strategies (the can the international community provide IFIs); and a more coherent "aid architecture" resources to support efficient service deliv- to reduce the costs of fragmentation. ery, postconflict recovery, and reform? Without addressing these development challenges the fragile states pose risks that can cross borders--through civil conflicts, Growth and Poverty Reduction risks to public health, and humanitarian crises. The world economy is growing at a pace last seen at the beginning of the 1970s. This is Two additional risks pertain to environ- welcome news for developing countries in mental sustainability and securing the gains view of its implications for trade, aid, private from trade liberalization. Natural resource financial flows, and remittances. Both low- depletion and environmental degradation and middle-income countries have benefited pose risks to both the quality of growth, and from the trend. Performance varies widely the potential for sustaining future growth. across regions, but there is a favorable trend Growth based on the depletion of natural evident in East Asia, South Asia, Eastern wealth, rather than through increasing wealth Europe, and Central Asia, and particularly for current and future generations, is unsus- Sub-Saharan Africa, where the sustained tainable. The "adjusted net savings rate" and rising growth performance since the late measures national savings after accounting 1990s is in sharp contrast to the weak per- for resource depletion and damage to the formance evident over the last three decades. environment, raising broad policy questions Average per capita income growth in Sub- Growth is reducing poverty, but not everywhere or always sustainably. Continued strong growth is generating significant progress in poverty reduction globally. But many countries are failing to benefit, especially fragile states, and for some others the sources and quality of growth (unsustain- able resource extraction; accumulating pollutants) undermine environmental sustainability and future growth potential. Investing in gender equality and empowerment of women is smart economics. Greater gender equality helps to create a fair society, raises economic productivity, and helps advance other devel- opment goals. Major gains have been achieved, particularly in education, while in other areas progress is lagging. Better monitoring and mainstreaming of women's empowerment and equality into policy formulation and programs of international assistance are therefore vital to the develop- ment agenda. Fragile states are failing to keep up--speed and staffing by development agencies are critical. The largest "MDG deficit" is in states with weak institutions and governance, and often in conflict--the "fragile states." With 9 percent of the developing world's population, they account for over one- fourth of the extreme poor and nearly one-third of child deaths and 12-year olds who do not com- plete primary school. Efforts to support their transition from fragility must be deepened through improving response time to crises and opportunities, increasing field presence, better interagency collaboration, and building on lessons from successful state-building transitions. Quality lags quantity--children enroll in school but don't always learn. Advancement in pri- mary school completion has been rapid and encouraging in many countries. Yet cross-country evaluations suggest improvement in cognitive skills has often not kept pace. Quantity and quality in education and health need to proceed in tandem. More effort is needed to monitor outcomes (especially student learning). This provides an essential platform for tracking over time whether policies and incentives are truly producing more effective service delivery. Scaling up "quality" aid requires greater coherence among donors, developing countries, and international agencies. Donor commitments to scaling up aid have so far been unrealized as real aid flows have faltered and a more complex aid architecture--proliferation of donor channels, fragmentation of aid, ear-marking of funds--undermines aid quality and effectiveness. Scaling-up aid to meet the MDGs requires more and better aid resources (donors); sound, sequenced develop- ment strategies (developing partners); better technical support for strong strategies (the IFIs); and a more coherent "aid architecture" to reduce the costs of fragmentation. Saharan Africa has recently been at about along with marked improvements since the 3 percent and is forecast to continue at this late 1990s. The share of people in poverty fell level in 2007. By contrast, growth among by nearly 7 percentage points between 1996 low- and middle-income countries in Latin and 2004, although the absolute number of America, and the Middle East and North poor has stagnated. Africa, continues to be more modest. Preliminary estimates suggest that, on Evidence suggests that better growth is average, growth (in GDP) during the late translating into declining poverty levels. The 1990s through 2003/04 resulted in lower most recent data show that all regions except poverty incidence: for a sample of 19 low- for Sub-Saharan Africa are on track to reach income countries, 1 percent of GDP growth the MDG1 poverty target. In Sub-Saharan was associated with a 1.3 percent fall in Africa the share of people living in extreme the rate of extreme poverty and a 0.9 per- poverty has declined little from its 1980 level, cent fall in the $2-a-day poverty rate. For but this masks the protracted deterioration middle-income countries the impact of GDP during the 1980s and first half of the 1990s, per capita growth on poverty was much less, and average poverty has not declined with MDGs. The two regions that lag the most recent growth. Moreover, changes in income are South Asia and Sub-Saharan Africa. As distribution have not, on average, reduced regions they remain off track on all the goals; the impact of income growth on poverty however, there is considerable variation reduction in low-income countries, whereas within regions. MDG trends in fragile states income inequality widened on average in are also examined; while there is variance middle-income countries. within the group, fragile states have lower One factor behind this favorable perfor- absolute performance and slower improve- mance has been the continuing strength of ment than nonfragile ones. macroeconomic policies, as evident through It must also be recognized that there have continued moderate inflation rates and aver- been some significant successes. Since 2000, age fiscal balances that shifted from deficit over 34 million additional children in devel- into balance in low-income countries during oping countries have gained the opportunity 2006. The quality of macroeconomic poli- to attend and complete primary school--one cies, particularly fiscal policy, in low-income of the most massive expansions of schooling countries shows considerable improvement access in history. Over 550 million children over recent years. have been vaccinated against measles, reduc- The stronger growth performance in low- ing death from measles in Sub-Saharan Africa income countries is encouraging, particu- by 75 percent. By mid-2006 the number of larly in Sub-Saharan Africa where the higher AIDS (acquired immunodeficiency syndrome) growth may mark a potential turnaround patients with access to antiretroviral treat- from the region's protracted stagnation. ment had increased nearly sevenfold to over However, this has to be interpreted with cau- 1.6 million from 2001 levels. There is little tion. Concerns persist over the potential for a question that the MDG targets have helped growth slowdown resulting from a disorderly stimulate more rapid expansion of basic unwinding of global imbalances, protection- health and education services. ism, the future behavior of world oil prices, Nutrition (MDG1). Nearly one-third of or a possible global pandemic triggered by all children in developing countries are esti- avian influenza. mated to be underweight or stunted, and Optimism over the prospects for improved an estimated 30 percent of the total popu- growth and poverty reduction, however, does lation in the developing world suffers from not apply to the many fragile states. Extreme micronutrient deficiencies. Undernutrition poverty is increasingly concentrated in these is not only a threat to progress with poverty states: by 2015 it is estimated that given pro- reduction; it is the underlying cause of over jected growth performance, extreme poverty 55 percent of all child deaths, linking nutri- levels in nonfragile states will decline to 17 tion directly to reduction of child mortality percent, more than achieving the MDG1 tar- (MDG4). In striking contrast to the region's get, while levels of extreme poverty in fragile strong growth performance, the highest rates states will remain at over 50 percent, higher of malnutrition are found in South Asia: than the level in 1990. underweight prevalence is estimated between 38 and 51 percent in the large countries, none of which appears on track to meet the nutri- Progress with the Human Development tion goal. Sub-Saharan Africa is estimated to MDGs have a 26 percent prevalence of child mal- Broad MDG trends do not change appre- nutrition, and in some countries-- Burkina ciably year to year, and remain much as Faso, Cameroon, Zambia--trends are wors- described last year: all regions are off track ening. East Asia, Latin America, and Eastern on the child mortality goal, and some regions Europe show better performance although all are off track on at least some of the other have some countries that are off track. Universal primary completion (MDG2). 2005. This success appears in large measure Globally the primary school completion rate attributable to implementation of the inte- rose between 2000 and 2005 from 78 to 83 grated management of childhood illness and percent and the pace of progress in many points to the serious need to strengthen policy countries has accelerated. Gains are especially coherence and improve donor coordination strong in North Africa, Sub-Saharan Africa, in the health sector. and South Asia. But 38 percent of developing Maternal health (MDG5). Ninety-nine countries are unlikely to reach 100 percent percent of maternal deaths, about 500,000 primary completion by 2015 and another annually, occur in developing countries. Lack 22 percent of countries, which lack adequate of direct data on maternal mortality requires data to track progress, are also likely to be off the use of "skilled attendance at delivery" as a track. The most intractable groups to reach proxy measure. Survey evidence shows prog- with primary education are those that are ress in 27 of 32 countries but also suggests "doubly disadvantaged": girls from ethnic, that this is highly concentrated among richer religious, or caste minorities. About 75 per- households--equity gaps in access to skilled cent of the 55 million girls who remain out of attendance are larger than for any other health school are in this group. But recent data also or education service. Evidence on the main reveal countries that have made remarkable constraints to reducing maternal mortality progress in recent years; six of the seven top in three low-income countries reaffirms the countries in expanding primary completion importance of early recognition of the need for rates (all by over 10 percent per year between emergency medical attention, access to ade- 2000 and 2005) were in Sub-Saharan Africa quate medical facilities, and receiving appro- (Benin, Guinea, Madagascar, Mozambique, priate treatment. But it also underscores the Niger, and Rwanda). The weakest performers essential need for skilled attendance at birth. were also primarily in Africa, however, show- AIDS, malaria, and tuberculosis (MDG6). ing the sharp contrasts across countries in By end-2006 an estimated 39.5 million peo- the region. And in Asia, Cambodia has made ple were living with the human immuno- exceptional progress. deficiency virus (HIV), up 2.6 million since Child mortality (MDG4). Progress on 2004. An estimated 3 million people died child mortality lags other MDGs, despite the from AIDS in 2006. While the spread of this availability of simple, low-cost interventions disease has slowed in Sub-Saharan Africa, that could prevent millions of deaths each it is a rapidly growing epidemic in Eastern year. Oral rehydration therapy, insecticide- Europe and Central Asia. Recent experience treated bednets, breastfeeding, and common in combating the spread of AIDS has demon- antibiotics for respiratory diseases could pre- strated some important messages: reversing vent an estimated 63 percent of child deaths. its spread is possible, treatment is effective in Yet in 2005 only 32 of 147 countries were the developing world, but prevention efforts on track to achieve the child mortality MDG. need to be intensified. Moreover, 23 countries reveal stagnant or Annually there are an estimated 300 to worsening mortality rates. Problems in fragile 500 million cases of malaria, and 1.2 million states are particularly severe: nearly one-third deaths, mainly among children and mostly in (31 percent) of all child deaths in developing Sub-Saharan Africa. Several new initiatives countries are in fragile states, and only two hold promise for making inroads against of the 35 states currently considered fragile malaria: with support from the Dutch and are on track to meet MDG4. The experience the "Roll Back Malaria" initiative, the World of countries that have achieved rapid gains is Bank is leading efforts to implement a global also noteworthy, including in Eritrea which, subsidy for artemisinin-based combination despite per capita income of only $190, cut therapy, the most promising new treatment child mortality in half between 1990 and available because resistance to traditional drugs has grown. The Malaria Booster Pro- it rebounded somewhat after 2003, it still has gram, which supports country-led efforts to not returned to the 2000 level. Recent efforts deliver concrete and measurable results, such to ramp up financing for WSS, especially for as delivery of insecticide-treated bed nets and Africa, through such initiatives as the Africa malaria treatment for young children and Infrastructure Consortium and the Rural pregnant women, is currently operating in 10 Water Supply and Sanitation Initiative--even countries and aims to expand to 20 over the if successful--will take some time to have clear next five years. impact on the WSS target, given the long lead Tuberculosis (TB) is estimated to have led time for investments. to 2 million deaths in 2004, and 9 million new A continuing concern for all these aggre- cases. While incidence of TB is falling in five gate data is whether poor households partici- of six regions, global growth of 0.6 percent pate in the progress made. Demographic and annually is attributed to rapid increases in Health Survey data allow comparison across infections in Sub-Saharan Africa, linked to income quintiles on relative progress. While the greater likelihood of TB appearing from gaps in access between rich and poor house- latent infections in HIV carriers. The Directly holds remain significant, they are narrowing; Observed Treatment, Short-course (DOTS) the poor have had equal or faster rates of is the main strategy to combat TB, and has progress in child mortality reduction, immu- expanded rapidly, with high-burden countries nization coverage, and primary completion showing large decreases in TB incidence due in most countries. to DOTS (for instance, Cambodia and Indone- sia). In 2006 a new strain of TB--extensively Financing Trends and Alignment in drug-resistant TB--was discovered in South Health and Education MDGs Africa. International efforts to stop its spread are being led by the World Health Organiza- External financing for health and education tion (WHO) and the Stop TB Partnership. has nearly doubled in real terms since the Water supply and sanitation (MDG7). MDGs were adopted. Aid for health con- There has been significant progress on water tinued to rise from 2004 to 2005, whereas supply; globally access has increased from 73 education ODA commitments showed their percent in 1990 to 80 percent in 2004, but only first decline, reflecting lower commitments to Latin America and South Asia are considered China and India. Aid commitments for educa- on track to meet this part of the goal (although tion are expected to have increased again in more than one-quarter of developing countries 2006 and beyond, owing in part to a major lack data). However, within Africa there are initiative announced by the United Kingdom. some promising trends: 5 of the 10 countries Funding for health has grown even more making fastest progress are in Africa, and 17 strongly, from private sources such as the of the 36 countries for which data are available Gates Foundation; from global partnerships are on or almost on track. By contrast, global such as the Global Fund for AIDS, TB, and progress on sanitation has lagged, increasing Malaria; and from bilateral donors: France, only from 35 percent in 1990 to 50 percent Norway, Spain, and the United States have in 2004 and only three regions (East Asia and increased health funding between two- and the Pacific, Latin America, and the Middle East fourfold since 2000. Innovative financing and Northern Africa) are on track. Only 2 of mechanisms targeting the health sector are the 32 African countries for which data are also getting off the ground: the international available are on track. Despite its importance finance facility for immunization ($1 billion for achievement of multiple MDGs, official in 2006), advance market commitments for development assistance (ODA) for water sup- vaccines ($1.5 billion expected in 2007), ply and sanitation (WSS) declined significantly and the airline ticket tax implemented by 21 from the mid-1990s through 2002. Although countries ($300 million expected in 2007) are all mobilizing new funds for health interven- Figure 1 illustrates weak learning out- tions. Despite this influx of funds for health, comes and the gap across countries between there remains a large shortfall relative to education level and cognitive skills. By age financing needs to reach the health MDGs, nine reading skills in developing countries conservatively estimated at between $25 bil- can significantly lag those in developed coun- lion and $50 billion annually. tries. While over 96 percent of children in While increased external funding is crucial Sweden, Latvia, and the Netherlands can for progress on the health MDGs, there are read above the lowest--threshold--level of growing concerns about policy coherence, literacy on OECD-benchmarked tests by age aid alignment, and transactions costs in the nine, less than half the children in Argentina, sector, given the number of players and the Colombia, and Morocco can read at this absence of effective coordination mecha- level. Results from a regionally benchmarked nisms--a topic taken up below. assessment for Southern African countries are similarly distressing: in several countries, less than 50 percent of children are able to The Role of Quality in MDG Progress read by age 12. Evidence is emerging that in many countries It does not follow from this that there rapid progress in improving schooling enroll- exists an inherent trade-off between quantity ment and completion is not translating into and quality in education. In fact, cross-coun- better cognitive skills. New research suggests try data show a strong positive correlation that this may have a high cost for countries: between schooling coverage and cognitive returns to investment in education appear to skills, at least over the long term. There are accrue to the skills of the population and not also numerous countries that have increased to the quantity of schooling attained. learning outcomes at the same time as they Learning levels of primary school­aged children Source: Fourth-grade test: International Association for the Evaluation of Educational Achievement (IEA), Progress in International Reading Literacy (PIRLS) 2001, Sixth-grade test: Southern African Consortium for Monitoring Educational Quality (SACMEQ). Enrollment data: Demographic and Health Surveys. have expanded access. While this is not easy and create incentives to accelerate progress, to do, and there are many cases where qual- alongside efforts to expand school comple- ity has been strained as countries rapidly tion rates. scale up access, it is important to focus on The same concerns over quality arise in the strategies for managing expansion better. health care--and data are even harder to Many poor countries are far from achiev- collect. Creative efforts have been made to ing universal primary completion and must measure the quality of health care provid- accelerate service delivery to reach the MDG ers across countries and measure the overall by 2015. Slowing expansion would harm the quality of care. The extent of misdiagnosed poorest and most marginalized groups most. ailments, failure to complete basic check- The challenge must be to expand access while lists for major diseases, and mal-adherence enhancing learning outcomes. to recommended protocols is alarming. The Progress on this challenge requires stron- implication is that there are gaps between ger efforts to monitor student learning in the what health providers know is right and what developing world; most countries today lack they do. It suggests that greater attention to national assessment systems and extremely work incentives and institutional settings is few have engaged in any internationally needed rather than reliance on input-based benchmarked tests. Regular tracking of stu- approaches, such as raising training require- dent learning is essential for accountability ments or expanding medical schools. Perfor- in education--for equipping teachers to man- mance contracting is one promising approach age their class time better, for empowering for effectively improving health coverage and parents to hold schools accountable, and for quality. Greater attention is also needed to allowing administrators to evaluate the effec- bring greater coherence and donor coordina- tiveness of education spending. tion to health sector strategies, as discussed There is a strong case for donor support below. in developing benchmarked standards of competency linked to critical thinking skills Governance Indicators: An Update expected by the end of primary school--in other words, basic learning goals for primary Recently released aggregated governance education to complement the quantitative indicators (Kaufman-Kraay) suggest patterns goal of universal primary completion. An of performance that reinforce key messages internationally benchmarked test to mea- from the 2006 GMR. Governance is mul- sure end-of-primary learning levels could be tidimensional, and there is no unique path expensive and technically difficult to produce, from poor to good governance. Actionable but there is a clear public goods argument for indicators to track performance are being such an investment. Precisely at a time when developed in several areas, including contri- the global community is scaling up aid for butions from independent civil society orga- the education MDG, a globally benchmarked nizations: Global Integrity released 43 new assessment covering large numbers of devel- country reports, the Afrobarometer network oping countries would provide the strongest released the results for 18 African countries platform yet for generating knowledge on of its third round of surveys, and a new index "what works" to promote learning in differ- that monitors transparency in public bud- ent country contexts. gets--the Open Budget Index--was released Moving a proposal for basic learning after four years of development. The World goals for primary education forward will Bank Group also released publicly for the involve costs and face political and technical first time its Country Policy and Institutional obstacles. But an internationally supported Assessment (CPIA) scores--an important effort in this area could help countries build step in strengthening transparency and dis- national capacity to track learning outcomes closure of these scores, which play an impor- tant role in allocating concessional financing. women, families, and society. The disadvan- By contrast, Public Expenditure and Finan- tage of women in rights (equality under the cial Accountability (PEFA) assessments made law), resources (equality of opportunity), and less encouraging progress. While the use of voice (political equality) restricts basic free- PEFA indicators has greatly expanded and dom to choose and is unfair. This inequal- many new country assessments are planned, ity is reflected in the poorer performance by so far only 4 of 33 country reports have been women and girls across many of the MDGs. made public, limiting the potential benefits "Improving gender equality and empow- from this valuable tool for analysis. ering women" (MDG3) thus stands on its own merits as a development objective. In addition to this intrinsic importance, gender equality and women's empowerment are also important channels to attain other MDGs. Gender equality and women's empower- ment promote universal primary educa- The Importance of Promoting Gender tion (MDG2), reduce under-five mortality Equality (MDG4), improve maternal health (MDG5), The 2006 World Development Report and reduce the likelihood of contracting HIV/ on equity and development refers to gen- AIDS (MDG6). der inequality as the "archetypal inequal- Improving gender equality also influ- ity trap," pointing to the sharp differences ences poverty reduction and growth directly between men and women in access to assets through women's greater labor force partici- and opportunities in many countries, and the pation, productivity, and earnings as well as negative consequences for the well-being of indirectly through the beneficial effects of Pathways from increased gender equality to poverty reduction and growth Source: World Bank staff. women's empowerment on child well-being. Rwanda, and South Africa). Of the 14 fragile Figure 2 identifies the main pathways lead- states for which data are available, 9 are not ing from gender equality to both current expected to achieve the primary and second- and future growth and poverty reduction. ary enrollment targets. One path is through increasing the produc- The female tertiary enrollment rate lagged tive opportunities and higher incomes that behind the male rate in 63 countries (of 130 women have, raising consumption and sav- countries with data) and exceeded the male ings that help to raise investment rates. rate in 65 countries. The female disadvantage Another is through improving women's was evident mainly in Sub-Saharan Africa, control over decision making in the house- South Asia, and in fragile states. hold. Several studies have shown that the Progress in basic literacy skills and school greater the mothers' control over resources, enrollments over the years has resulted the more resources households allocate to in higher literacy rates among youth (age children's health, nutrition, and education. 15­24), but gender gaps remain: the United Better maternal education also benefits chil- Nations Educational, Scientific and Cultural dren through improved hygiene practices, Organization (UNESCO) estimates that of better nutrition, lower fertility rates, and the nearly 137 million illiterate youths in the hence higher per child expenditures. Taken world, 63 percent are female. The female-to- together, these contribute to future growth male literacy ratio is lowest in Sub-Saharan and poverty reduction. Africa, Middle East and North Africa, and South Asia--regions that also have female disadvantages in primary and secondary Progress toward Meeting MDG3 enrollment. The four official MDG3 indicators--measur- Progress is also evident in women's share ing gender equality in enrollments, literacy, of nonagricultural wage employment, which and the share of women in nonagricultural increased modestly in all regions during employment and national parliaments--pro- 1990­2005, and with significant variation vide an important, albeit incomplete, snap- across regions and countries (figure 3). In shot of progress toward gender equality. 2005 the share of women in nonagricultural Thanks to efforts to achieve universal pri- employment was highest in Europe and Cen- mary education (MDG2), girls' enrollments in tral Asia (47 percent), lowest in the Middle all levels of schooling have risen significantly East and North Africa (20 percent), and in- (figure 3). Most low-income countries made between in Latin America and the Caribbean substantial progress between 1990 and 2005. and East Asia and the Pacific (over 40 per- By 2005, 83 developing countries (of 106 cent). Trends and patterns in this indicator with data) had met the intermediate MDG3 are difficult to interpret without accounting target of parity in primary and secondary for country circumstances, such as the share enrollment rates. Most of these countries of nonagricultural employment as a percent- are in regions where enrollment has histori- age of total employment. A favorable score cally been high--East Asia and the Pacific, on this indicator might on the surface seem Eastern Europe and Central Asia, and Latin to indicate equitable conditions for women in America and the Caribbean. In the Middle labor markets, but it may capture conditions East and North Africa, most countries met for only a very small proportion of the total the target by 2005, but some still have a sig- labor force. nificant female disadvantage in enrollments. The fourth official MDG3 indicator is In Sub-Saharan Africa performance has been the proportion of seats held by women in varied; less than one-quarter of countries met national parliaments (with no set target). the enrollment targets for 2005, but some Between 1990 and 2005, all regions except have attained parity (for example, Botswana, Europe and Central Asia increased women's Progress in official indicators of gender equality and women's empowerment, by region, 1990­2005 Source: World Bank Indicators. The regional averages are calculated using the earliest value between 1990 and 1995 and the latest value between 2000 and 2005. The averages are weighted by the country population size in 2005. proportion of the seats in national parlia- and amenability to policy intervention. Indi- ment, but starting from a very low level (fig- cators that met all the three criteria but were ure 3). However, in no region did the average highly correlated with other indicators were proportion exceed 25 percent, at either the dropped from the list. beginning of the period or the end. This proposed list draws on the recommen- dations of the UN Millennium Project Task Force, but is more parsimonious. It takes into Strengthening Official Indicators account data availability, additionality (does The shortcomings of the official indicators it add new information), and the high costs for monitoring progress in attaining MDG3 associated with imposing additional moni- are widely recognized (see, for example, the toring burdens on already taxed national sta- report of the UN Millennium Project Task tistical offices. It also draws on a proposal to Force on Education and Gender Equality, refine the existing MDG indicators that was UN Millennium Project 2005). Five supple- put before the UN Secretary General's office mental indicators to better measure gender for consideration in March 2007. equality are proposed to address this (table Four of the five indicators monitor gen- 1). These indicators, complementary to the der equality in the household; the remain- official MDG3 indicators, meet three crite- ing indicator monitors gender equality in ria: data availability (wide country coverage), the economy. No additional indicators are strong link to poverty reduction and growth, recommended to monitor gender equality in Recommended additional indicators for MDG3 Household Economy and markets Modifications of official MDG indicators Additional indicators Additional indicators Primary completion rate Percentage of 15- to 19-year-old girls Labor force participation rates of girls and boys (MDG2)a who are mothers or pregnant among women and men aged with their first childb 20­24 and 25­49b Under five mortality rate for girls and boys (MDG4) Percentage of reproductive-age women, and their sexual partners, using modern contraceptives (MDG6) Source: World Bank staff. a. Recommended by the UN Millennium Project Task Force on Education and Gender Equality. b. Under consideration by the Inter-Agency and Expert Group for MDGs. the domain of society, because none of the However, in spite of strong donor policy indicators considered for inclusion met the commitments to gender equality objectives, criteria of data availability. Three of the rec- implementation has been disappointing. ommended indicators are modifications of Self-evaluations of nine donor agencies' per- official indicators already being monitored as formance reflect a gap between words and part of the MDGs, while two are indicators deeds. One of the reasons for this gap is the not currently part of the official set. diffusion of responsibility that resulted from the shared responsibility gender mainstream- ing called for: all staff were responsible for Strengthening International Support promoting it, yet no one group in particular for Gender Equality was held accountable for results. The success in increasing girls' enrollments These self-assessments have helped reen- in schooling shows that progress in gender ergize donors' commitments. Donors are equality is possible. This progress, however, revamping their approaches and setting requires political will (high-level leadership) more realistic targets to both strengthen and concerted effort from countries and inter- mainstreaming and introduce specific actions national development agencies. Donors and to advance gender equality. There is wide the multilateral development banks (MDBs) agreement that high-level leadership, techni- need to significantly improve the support cal expertise, and financial resources remain and coordination for gender equality issues key to implementing donor agencies' gender to accelerate progress toward MDG3; these policies. issues should become central in their dialogue The MDBs have made similar progress in with partner countries. Since the 1995 Beijing advancing their support for gender equal- Women's conference, which marked a mile- ity and women's empowerment. Systems to stone in international commitment to gen- monitor progress with mainstreaming gender der equality issues, donor support improved equality policies have been introduced, and somewhat, and more resources are devoted suggest there has been modest but steady to gender equality targets, particularly in the progress. Most MDBs have recently adopted social sectors. Overall, a quarter of bilateral Gender Action Plans to make their gender aid by sector--around $5 billion annually-- mainstreaming policies more strategic and is now focused on gender equality. operationally effective. Nonetheless, significant gaps remain. der equality and women's empowerment in Progress has been greater in the social sec- the results agenda, in leading international tors (especially health and education) than efforts to strengthen MDG3 monitoring, and in productive sectors (agriculture, infrastruc- in better assisting client countries in scaling ture, private sector development, and the up MDG3 interventions. The business case like). There is also evidence that attention to for MDBs' investments in MDG3 is strong-- gender issues is greater in project design than it is nothing more than smart economics. in implementation, and there has been little effort to monitor or evaluate outcomes. Insti- tutions have generally been slow to develop and adopt measurable indicators of progress in gender equality, and the rating systems Fragile states are generally characterized by primarily measure good intentions rather weak institutional capacities and governance, than results. Nor can the resources spent on and by political instability. These countries are gender mainstreaming be measured. Clearly the least likely to achieve the MDGs and they much more is needed to strategically realize contribute significantly to the MDG deficit. the comparative advantage that the MDBs They account for 9 percent of the population have in knowledge generation and analysis, of developing countries, but 27 percent of the in their convening and coordinating roles, in extreme poor (living on US$1 per day, see leading high-level dialogue, and in helping table 2), nearly one-third of all child deaths, formulate development policy strategies. The and 29 percent of 12-year olds who did not MDBs should utilize their comparative advan- complete primary school in 2005. Of all low- tage and take up a visible leadership role in income countries that are unlikely to achieve investing dedicated resources to include gen- gender parity in primary and secondary enroll- Fragile states face the largest deficit in most MDGs Total in developing Total in fragile states Indicator countries (millions) (in millions and % share) Extreme poverty 985 261 (27%) Malnourished children 143 22.7 (16%) Children of relevant age that did not complete primary school in 2005 13.8 4 (29%) Children born in 2005 not expected to survive to age five 10.5 3.3 (31%) Unattended births 48.7 8.9 (18%) TB deaths 1.7 0.34 (20%) HIV+ 29.8 7.2 (24%) Lack of access to improved water 1,083 209 (19%) Lack of access to improved sanitation 2,626 286 (11%) Source: World Bank staff estimates; for notes see table 2.9. ments, half are fragile states. Their weak per- from conflict, the sequencing and coherence formance is clearly linked to chronically weak of support for security, electoral efforts, and institutional capacity and governance and to aid-financing to boost growth and employ- internal conflict, all of which undermine the ment are critical for minimizing the risk of capacity of the state to deliver basic social and reversion to conflict. Donors need to con- infrastructure services and offer security to sider whether current instruments provide citizens. adequate continuity of support to minimize Conflicts are a major reason why coun- risks of renewed conflict. tries slide into fragility; they extract high Second, engaging in fragile states requires costs in terms of lives and physical damage, the IFIs and other donors to review their busi- they reduce growth and increase poverty. ness practices and procedures, to ensure that While there are fewer conflicts in low-income these are adapted to low-capacity and some- countries than before, conflicts have become times volatile environments. Taking advantage shorter and more intense, with an enormous of new peace-building or governance reform negative impact on GDP growth averaging opportunities, or adjusting programs in the about 12 percent decline per year of conflict. event of a crisis, requires a rapid response from Despite the enormous challenges of poverty all international partners engaged in these in fragile states, progress against the MDGs countries. Supporting reforms in low capacity is possible. A number of countries (Mozam- states also requires increased field presence. bique, Uganda) have made a successful transi- Third, fragile states are especially vulner- tion from weak institutions and/or the legacy able to donor fragmentation and its potential of conflict to sustained gains in growth and burden on government capacity. This makes poverty reduction. In countries that remain implementation of the Principles for Good fragile, successful progress against the MDGs International Engagement in Fragile States has been achieved: Timor-Leste, Eritrea, and and advancing principles of the Paris Declara- the Comoros, for example, decreased child tion on Aid Effectiveness particularly impor- mortality by 7.1 percent, 4.2 percent, and 3.5 tant. The IFIs need to work both between percent, respectively, between 2000 and 2005. themselves and with other international Aid is particularly important in fragile partners to develop common approaches and states because it constitutes the main source of operating principles in fragile states, in par- development finance. However, IFIs account ticular through efforts to improve coordina- for only about 8 percent of total Development tion and division of labor with organizations Assistance Committee (DAC) ODA flows to leading peace-building efforts, such as the fragile states, with the rest coming from bilat- United Nations and regional institutions. eral sources. The IFIs, nevertheless, have an important role to play in financing postcon- flict reconstruction, in aid coordination, and in policy dialogue and technical assistance. The MDBs have recently started to converge around four areas of specialized response to The expansion in global aid has stalled, and the development challenge in fragile states: two years after the Gleneagles summit the (1) strategy, policy, and procedural frame- trends in real aid flows suggest that DAC works; (2) exceptional financial instruments; donors' promises of higher aid to Sub-Saha- (3) customized organizational and staffing ran Africa appear increasingly unlikely to approaches; and (4) partnership work. materialize. Seven years after the Millennium Accelerating progress toward the MDGs Summit at which the MDGs were adopted, in fragile states requires attention to sev- there is yet to be a single country case where eral issues and lessons of recent experience. aid is being scaled up to support achieving First, since many fragile states are emerging the MDG agenda. Most "low hanging fruit" identified in the Millennium Report of 2005 that DAC donors have moved out of. have yet to be harvested. Progress with mul- Progress with scaling up aid to Africa has tilateral debt relief was rapid after the Gle- been disappointing. Five years after the Mon- neagles meetings in 2005, demonstrating how terrey Conference and two years since the G-8 quickly initiatives can advance when there is a pledges at Gleneagles, country examples of strong international commitment. The lack of programs to scale-up aid to support the MDG progress with multilateral trade reforms in the agenda are lacking. Beyond debt relief (impor- Doha Round similarly demonstrates just how tant to improving future growth opportunities), weak international commitment and consen- most countries in Sub-Saharan Africa are see- sus stymies change. Forging an international ing stagnant or declining aid inflows. Exclud- consensus beyond rhetoric is needed to accel- ing Nigeria (a recipient of exceptional debt erate progress. relief) real bilateral ODA from DAC members to the region fell in 2005 and was unchanged in 2006. Aid Volumes Trends: Bringing Actions There is evidence that aid allocation is in Line with Commitments becoming increasingly selective on the basis Although aid was on an upward trend through of need (poverty) and the quality of policies 2005 as DAC members, non-DAC donors, and (governance). Selectivity varies across differ- nontraditional donors expanded assistance to ent aid instruments. Flexible ODA--aid that developing countries, in 2006 the level of real can be used toward regular project and pro- aid from DAC members fell. After reaching a gram support as opposed to special-purpose record level in 2005, total DAC member aid grants such as technical or emergency assis- fell by about 5 percent to about just below tance--has been the most responsive to coun- $104 billion in 2006. These trends suggest that try improvements in governance and greater real aid delivery is falling well short of donor need. Technical assistance (much of which is commitments. Doubling of aid to Africa by for consultants and never leaves donor coun- 2010 looks increasingly unlikely. tries) is the least responsive. There has also been a continuing concen- Attention by donors to the needs of fragile tration of aid in a small number of coun- states is beginning to translate into increased tries, leaving the majority of countries with assistance. Overall aid to fragile states rose little or no real increase. Between 2001 and by more than two-thirds in 2005 to nearly 2005, real aid volumes grew by more than $20 billion (in 2004 dollars), of which about 50 percent, but nearly 60 percent of Inter- half was in debt relief and humanitarian assis- national Development Association (IDA) tance. Fragile states are seeing an improving countries saw a decline or little change in trend in aid received per capita, although aid over this period. Such heavy concentra- they receive somewhat less aid (excluding tion is not consistent with efforts to broadly humanitarian assistance and debt relief) than accelerate progress toward the MDGs. Even other low-income countries. Aggregate trends as assistance from DAC donors has declined mask the wide variation across different types in 2006, aid from nontraditional donors is on of fragile states: those emerging from violent an upward trend: Non-DAC OECD donors conflict typically receive much more aid than are expected to double their assistance to other fragile states, and more than other low- over $2 billion by 2010; Saudi Arabia and income countries. other Middle East countries provided nearly $2.5 billion in assistance in 2005; and other Progress with Harmonization emerging donors, China in particular, are and Aid Effectiveness also rapidly expanding aid and becoming significant foreign creditors. Much of this aid A critical agenda for improving aid effec- targets infrastructure and productive sectors tiveness is progress with harmonization and alignment of aid with country strategies, both aid for the health sector in Rwanda illustrates by donors and the international aid agencies. some key problems on the ground. First, the There is evidence of some progress in these government's ability to achieve policy coher- efforts. Two-thirds of donors place strategic ence is undermined by donors channeling the priority on implementation of the Paris Dec- majority (86 percent) of total reported aid for laration on aid effectiveness, and efforts to health outside the Ministry of Health through monitor its implementation are gaining trac- direct transfers to local NGOs, local govern- tion. However, translating this good intent to ments, and other providers. Second, most outcomes on the ground remains extremely on-budget donor funding is earmarked for challenging: the greatest need for better aid HIV/AIDS and malaria (85 percent in 2005), harmonization is often in countries least to the relative neglect of capacity building, capable of leading donor coordination them- human resource development, and other sec- selves. torwide needs. Only 1 percent was allocated A baseline survey for monitoring the Paris to child health. Third, aid is volatile as much Declaration was undertaken in mid-2006, is committed for only 1 to 2 years, constrain- yielding benchmarking data on the con- ing ability to scale up health services which straints facing donors and partner countries. require mainly stable recurrent expenditures On ownership by partnership countries it for salaries and facility maintenance. Finally, finds the story is mixed: while comprehen- there is a sharp disparity between donor fund- sive national strategies are being developed, ing for health, which has increased sharply, they lack well-specified prioritization and and infrastructure and agriculture, which sequencing of objectives and actions, leav- have been neglected. These factors point to ing them operationally weak. Less than one- the need for coordination among donors, fifth of countries had developed operational agencies, global programs, and developing strategies at the time of data collection. The countries, to develop an adequate coordinat- survey also finds that overall public financial ing mechanism and more coherent approach. management systems are weak in over one- Harmonization in the health sector is par- third of countries, and moderately strong or ticularly difficult: the number of donors is better in less than a third. large and includes numerous vertical pro- Regarding donor actions, it finds that grams; there is usually no critical mass of about 40 percent of aid is disbursed using a health financiers "on the ground" who can partner's public financial and procurement meet regularly to coordinate and harmonize. systems; about two-thirds of aid is disbursed There is also an inherent tension between on time; nearly half of technical cooperation the goals of harmonized aid through coun- is already coordinated--which is the 2010 try systems and the explicit mandates of ver- target, although different interpretations of tical funds--whose successful advocacy for "coordination" require caution. The survey specific global health issues depends criti- finds that donors are trying to harmonize. cally on their ability to show direct results. Forty-two percent of aid is provided through A viable harmonization strategy may be to program-based approaches such as direct move toward a country-led arrangement budget support or sectorwide approaches. whereby (1) all donor support is "on plan" One-third of missions and one-fifth of coun- and aligned with government priorities and try analytic work is joint. However, strategic initiatives; (2) funding is primarily through partnership "satisfaction" surveys in Africa the government budget, and where this is not suggest increased dissatisfaction over donor possible a share is specified for support to reporting requirements and coordination of system capacity building; and (3) reporting donor support. to donors is less frequent and done through Pledges of harmonization remain abstract multipurpose reports that meet multiple unless tested in the field. A recent review of donor needs. The Rwanda health sector example points completion point under the Heavily Indebted more broadly to challenges posed by the Poor Countries (HIPC) Initiative. Twenty- evolving and more complex aid architecture. two postcompletion-point HIPCs (and two The proliferation of new aid sources--donors, non-HIPCs) have benefited from the MDRI private foundations, global funds--increases to date, providing $38 billion, in nominal total resources, but also the difficulty of coor- terms, in debt relief. The ongoing HIPC ini- dination and coherence, and the costs posed tiative also saw substantial progress, and 30 by fragmentation and resource earmark- HIPCs had reached the decision point and ing. The average number of official donors were receiving debt relief as of end-2006. has tripled since the 1960s, and since 1990 the number of countries with over 40 active Developments in Global Trade bilateral and multilateral donors increased from zero to over 30. Emerging donors are World trade in 2006 continued the strong also expanding their presence rapidly, along growth trends of recent years. Merchandise with global funds, although these are dif- exports expanded by 16 percent in value, ficult to track due to insufficient data. The well above the average of 8 percent experi- problem of a large number of aid channels is enced during 1995­2004. Developing-coun- compounded by the trend towards small size try export growth continued to outpace the of funded activities, which declined on aver- global average, growing by 22 percent. In age from $1.5 million to $1 million between addition to cyclical factors, trade perfor- 1997 and 2004, while their number surged mance reflects continuing unilateral trade from 20,000 to 60,000. reforms. Average tariffs in developing coun- This places particular stress on countries tries have fallen from 16 percent in 1997 to with weak capacity. Countries with lower around 11 percent in 2006. As the pace of institutional capacity are found to have higher global integration accelerates, harnessing the aid fragmentation, with negative implications new opportunities and managing the risks for aid quality through higher transaction places a premium on a strategy of greater costs and a smaller donor stake in country openness, coupled with behind-the-border outcomes. Clearly excessive fragmentation is reforms. a serious problem and measures to address it, Owing to the steady reduction of tariffs, possibly through donors limiting their focus overall trade restrictiveness has declined in countries, providing larger funds, or adopt- recent years. With the exception of a num- ing more efficient vehicles (including through ber of African countries, most economies multilateral channels), and donors commit- are now less trade restrictive than they were ting to delegate authority to lead donors, in 2000. Much of this observed liberaliza- could help reduce transactions costs and tion pertains to manufacturing. Much less improve aid effectiveness. has been done in agriculture. For a number of countries (such as India) the agricultural sector is now more restrictive than six years Developments in Debt Relief ago; in the European Union there has been The past year saw important progress in no change, while Canada and the United deepening debt relief to the poorest countries. States have registered a small decline since The African Development Fund (AfDF), IDA, 2000. and the International Monetary Fund (IMF) Progress with the Doha Round. Despite all implemented the Multilateral Debt Relief intensive efforts to conclude the Doha nego- Initiative (MDRI), described in the 2006 tiations in 2006, they were effectively sus- GMR. This initiative provides 100 percent pended in July amid disagreement on the level debt relief on eligible claims to countries that of ambition in agricultural market access and have reached, or will eventually reach, the over reductions in domestic support. However, in early 2007 there was an informal agree- Evolving Roles ment by World Trade Organization members A number of commissioned reports or initia- to restart talks, providing a narrow window tives were completed in the last 12 months of opportunity to reach agreement in the first with implications for the changing demands half of 2007 on the key elements of a deal. on IFI resources and responsibilities. Discus- Failure to conclude the Doha Round sions have highlighted five key challenges: would send a strong negative signal to the support to the poorest countries; strength- world economy about the ability of coun- ened engagement in middle-income countries; tries to pursue multilateral solutions. It responding on critical global public goods; could weaken the multilateral trading sys- promoting coherence and collaboration; and tem, which provides developing countries strengthening the voice and representation with guaranteed nondiscriminatory market of developing countries. Reports released in access; the rules-based settlement of disputes; September 2006 included the IMF's Medium- and the transparency of trade regimes. But Term Strategy and the Report of the Interna- the biggest risk of failure is to countries' own tional Task Force on Global Public Goods; in economic growth, as trade reform is funda- the same month the Middle Income Country mentally about self-interest. Strategy Report was reviewed by the World Aid for trade. Progress was made on aid Bank's Board. The Review Committee on for trade in 2006. Donors indicated that they IMF­World Bank Collaboration released are prepared to offer large increases in aid its report in early 2007. In addition, initial for trade to help developing countries address measures were taken to address the need for the supply-side constraints to their increased changing voice and participation in the IMF participation in global markets and any tran- and the World Bank. sitional adjustment costs from liberaliza- These reports conclude that there is sig- tion. How much of this would be additional nificant progress in assisting poor countries to existing aid remains unclear. Also, more toward achieving the MDGs and in working remains to be done to operationalize this to promote country-led efforts in partner- agenda. ship with other donors. Connecting results and resources remains a major challenge, however. There is broad recognition of the importance of continuing to engage with Enhancing IFIs' effectiveness in advancing the middle-income countries, which are home to MDG agenda requires adapting their strate- some 70 percent of the world's poor, but also gies and developing capacity to be responsive of the need to improve the responsiveness of to (1) changing demands, including those the IFIs and tailor support to specific coun- related to globalization and global public try conditions. Critical public goods include goods; (2) growing differentiation among cli- international financial stability, a strong ents; (3) the availability of alternative finan- international trading system, preventing the cial resources; and (4) the growing number of emergence of infectious diseases, generating actors on the development landscape. Several knowledge, and tackling climate change. commissioned reports and events in 2006 Cooperation among MDBs is underpinned reflect on the evolving responsibilities of the by Memoranda of Understanding between IFIs and the need to strengthen performance them, and in 2006 the managing director of and collaboration. More coherent efforts the IMF and the president of the World Bank may be needed to strengthen the results man- commissioned an external review of collab- agement capacity of the IFIs, both to support oration between the two institutions. The capacity building in partner countries and to report noted many examples of good collabo- reflect on their own performance. ration, but also identified scope for improve- ment, including clarifying the role of the IMF These trends suggest that while demand in low-income countries. Concerning voice for MDB lending from middle-income coun- and representation, a program of revisiting tries has increased, the supply of concessional quotas and governance reforms in the IMF funds to low-income countries is now stag- was launched in 2006 to be completed by the nant. This has implications for the role of 2008. The first step was to revise quotas for MDBs in the future, particularly their ability a group of the most underrepresented coun- to respond to demands for scaling up multi- tries: China, the Republic of Korea, Mexico, lateral assistance. Viewed from the perspec- and Turkey. The changes approved in 2006 tive of overall ODA flows, the share of MDB increased these countries' total IMF quotas financing has fallen significantly since 1998; by 1.8 percent, raising their share to 7 per- if disbursements continue to stagnate while cent of total voting shares. Further steps are donors scale up bilateral ODA, the MDBs under way to develop a new formula for a will represent only about 6 percent of total second-round quota adjustment, and prepa- ODA flows by 2010. This poses important ration of a proposal to increase basic votes questions for the international community in order to enhance the voice of low-income over the implications of declining multilat- countries. Consultations on voice and rep- eralism, or of the shifting multilateralism to resentation are also under way in the World other agencies, primarily the UN system and Bank. the European Union. Debt relief under the MDRI has further potential repercussions for IFI financing, in Assessing Effectiveness: Financial Flows, particular for AfDF and IDA, which have Results, Harmonization, and Alignment provided debt relief extending out to 40 Assessing the effectiveness of the IFIs poses years. The MDRI commits donors to provid- difficult challenges. Development results ing additional resources, on a "dollar-for-dol- often lag policies and programs, and are lar" basis over four decades, to ensure that the hard to measure, but the bigger problem is cost of debt forgiveness does not undermine that of attribution of results. Each IFI has an these institutions' overall financial integrity independent evaluation agency that plays an or ability to provide future financing. Firm important evaluative role, but it remains dif- financing commitments cover 10 percent of ficult to address the results and attribution the total cost, and qualified commitments problems. Three aspects of international another 56 percent, leaving a gap of 34 per- financial institutions' performance--finan- cent between total costs and commitments for cial support, results-based management, and the MDRI. IDA 15 will be an important test progress toward harmonizing and aligning of donors' intentions regarding the MDRI aid through the Paris Declaration--are high- and future role of the MDBs. lighted. Results management. The Third Round- Financial flows. Despite the rapid growth table on Managing for Development, held in private capital flows to developing coun- in Hanoi in February 2007, provided a tries, the financing role of the IFIs remains an venue for many country delegations to com- important one. In 2006, the five MDBs dis- pare experiences and learn from them. The bursed $43 billion, up 20 percent over 2005 Roundtable included a meeting of the Asian levels. It is premature to assess whether this Community of Practice, and the launching increase is a temporary trend. Nonconces- of a similar Community of Practice in the sional gross disbursements increased by 29 Africa region. Five factors were highlighted percent to $32 billion. After strong growth in as important in building country capacity concessional gross disbursements since 2000, to manage for development results: leader- peaking at just over $11 billion in 2004, flows ship and political will, strong links between slightly declined in 2005 and 2006. results and planning practices, evaluation and monitoring tools to generate feedback tors, limiting performance comparisons over on programs, mutual accountability between the two years. A number of findings emerged, donors and country partners, and statistical including the need to better communicate the capacity (both to supply, and help generate results of COMPAS within each MDB. greater demand for, managing for results). IFIs and the Paris Declaration. Results of The need to scale up both financial and tech- the country-level monitoring of the imple- nical support for statistical capacity-build- mentation of the Paris Declaration's mutual ing was underscored as an essential element commitments, which took place for the first of the agenda--particularly as the financial time in 2006, will serve as a baseline to review costs of strengthening systems are relatively progress in 2008 and against the 2010 tar- modest. gets. They suggest that substantial actions are The Common Performance Assessment being taken by the MDBs in many areas of System (COMPAS) is an interagency effort to harmonization and alignment, including the develop a common system across the MDBs use of joint or collaborative country assis- for monitoring their results orientation, par- tance strategies, but that continued efforts ticularly with regard to their internal prac- will be needed to achieve the 2010 targets. tices. Its three-pillar structure was described Over half the country analytic work of the in detail in the 2006 GMR. A report for MDBs is joint with other donors and/or part- 2006, prepared under the leadership of the ner governments, relative to the target of 66 Inter-American Development Bank (COM- percent, but only 21 percent of MDB mis- PAS' chairmanship rotates), examines the sions are joint with other donors, relative to seven performance categories that were devel- a target of 40 percent, and there is an urgent oped for the 2005 report. In 2006, however, need to reduce the large number of parallel changes were made to improve the indica- implementation units. The following figures and commentary provide an ularly for poverty (over half), malnutrition, gender, overview of the main trends in country and regional and access to improved water. From the available progress toward achieving the Millennium Develop- data, those targets for which the greatest progress ment Goals. Owing to the limitations of both data has been made include gender equality (as measured and space, the coverage here is selective. An over- by gender parity in primary and secondary school view of performance can be seen from the figure enrollment), access to skilled care at birth (a proxy below showing the shares of all developing countries measure for maternal mortality), and reaching 100 globally that have achieved or are on track to meet percent primary school completion. Those targets the development goals, are off track and seriously lagging most severely include reducing child mor- off track to meet them, or countries for which there tality, halving extreme poverty, and improving child are insufficient data. It is immediately evident that nutrition. The global challenge of meeting the MDGs for these targets many countries simply do not have remains daunting. adequate data to measure their performance, partic- Global Progress toward the MDGs--Select indicators Source: World Bank staff. Note: Country on track means that the rate of progress is superior to the required growth rate to achieve the goal. Country off track means that the rate of progress is inferior to the required growth to achieve the goal, and superior to the required growth to achieve the half of the target. Country seriously off track means that the rate of progress is inferior to the required growth to achieve the half of the target. Extreme poverty--the proportion of the population Approximately 27 percent of the extreme poor in developing countries living on less than $1 a day-- in developing countries live in fragile states. Fragile fell from 29 percent in 1990 to 18 percent in 2004. states have consistently grown more slowly than in East Asia and Pacific experienced the most impressive other low-income countries. Clearly, this has been, reduction in poverty, and South Asia is now on track, and will likely continue to be, an obstacle to the but Sub-Saharan Africa lags behind. Over the same achievement of MDG1 in those countries. A typi- period, the number of people in developing countries cal fragile state had made negative progress toward grew by 20 percent to more than 5 billion, including MDG1 by 2005, at which point its poverty rate by 1 billion people in extreme poverty. Global poverty the $1 per day measure was about twice that of a is projected to fall to 12 percent by 2015--a striking typical nonfragile state. success. Share of people living on less than $1 or $2 a day in 2004, and projections for 2015 Source: World Bank staff estimates (weighted averages). While accelerating growth in India has put South East Asia has experienced a sustained period The transition economies of Europe and Central Asia on track to meet its goal, Sub-Saharan Africa of economic growth, led by China, while Latin Asia saw poverty rates rise in the 1990s and then lags behind. America and the Caribbean has stagnated, with fall. There and in the Middle East and North modest poverty reduction. Africa, consumption of $2 a day may be more relevant. TARGET 1: Halve, between 1990 and 2015, the proportion of people who live on less than $1 a day. TARGET 2: Halve, between 1990 and 2015, the proportion of people who suffer from hunger. Proportion of countries on track to Proportion of countries on track to achieve the poverty reduction target reduce under-5 malnutrition by half Source: World Bank staff estimates. Source: World Bank staff estimates. This chart shows the percentage of countries in each region that are on track More than half of countries in Sub-Saharan Africa are off track to reach the to achieve the poverty reduction target of the MDGs. Some have already 2015 target of cutting malnutrition rates by half. Half of the countries in achieved the target. Those shown as on track could reach the target by 2015, South Asia are on track to reach the target, but they also have the highest if they maintain current progress. But those shown as off track and seriously rates of malnutrition in the world and will continue to have the largest share off track are reducing poverty too slowly. Eighty percent of fragile states lack of malnourished children, even if the target is achieved. Malnutrition rates in the data needed to estimate their progress. fragile states are similar to those found in other developing countries. Number of people living on less than $1.08 a day (millions), 1981­2004 Between 1981 and 2004, the number of people in the world who lived in extreme poverty fell from nearly 1.5 billion to just under 1 billion. Both the Middle East and North Africa and Europe and Central Asia regions had essentially eliminated extreme poverty by 1981, but nearly 800 million poor people lived in East Asia--over half of the global total. By 2004 extraordinary progress had been made to lower poverty incidence in East Asia, lifting nearly 630 million people from extreme poverty in under a quarter of a century, which lowered the region's share of total poverty from 58 percent to just 9 percent. At the same time, poverty fell in South Asia from 52 percent of the population in 1981 to 32 percent in 2004, but absolute numbers have been persistent at around 470 million people. In contrast, in Sub-Saharan Africa poverty incidence hovered around 46 percent between the early 1980s and 1999, and declined to 41 percent in 2004. Despite this better trend, the absolute number of poor is still around 300 million. Source: World Bank staff estimates. Globally, the primary completion rate has increased Indonesia, and Kenya. Notwithstanding these very from 63 percent in 1990 to an estimated 83 percent in positive trends, the goal of universal primary comple- 2005, and the pace of annual improvement has accel- tion by 2015 will be difficult to reach: 57 of the 152 erated significantly since 2000 in the three regions fur- developing countries (38 percent) for which data are thest from the goal--North Africa, South Asia, and available are considered off track--meaning that they Sub-Saharan Africa--a sign of the increasing priority will not reach the goal on current trends (figure 2). given in these regions to universalizing primary educa- Most of the 33 countries that lack data are also likely tion (see figure 2.1). Latin America and the Caribbean, off track. Among African countries, 65 percent are which started from a higher base, has also sustained considered seriously off track, defined as unlikely to an exceptionally strong rate of progress. The num- reach the goal before 2040. Among fragile states, only ber of countries that have achieved universal primary 11 percent have achieved universal primary comple- completion increased from 37 in 2000 to 52 in 2005, tion or are on track to doing so, and 50 percent are and this includes some low-income countries: Bolivia, considered seriously off track. Primary school completion rate Source: World Bank staff estimates (weighted averages). Despite faster progress since 2000, Sub-Saharan East Asia, Latin America and the Caribbean, and Strong progress since 2000 in the Middle East Africa remains very far from the goal. In South Europe and Central Asia are all close to the goal. and North Africa has put that region on track to Asia, populous India's strong progress boosts the achieve universal primary completion, although regional picture, although some countries remain the regional average hides some variance across off track. countries. TARGET 3: Ensure that by 2015 children everywhere, boys and girls alike, will be able to complete a full course of primary schooling. Proportion of countries on track to Percent of 15- to 19-year-old cohort that achieve the primary education target has completed primary education by household wealth quintile and location Source: World Bank staff estimates. Source: DHS Surveys. In many developing countries children are able to complete a full course of Data from household surveys indicate that the largest gaps in primary primary education, but in all regions at least a few countries remain off track completion rates in virtually every developing country are between and unlikely to reach the primary education target. In Sub-Saharan Africa, wealthy and poor populations (figure 3). But gaps between urban and rural the poorest-performing region, 65 percent of countries are seriously off track populations can also be very large, especially in Africa. Completion rates and only 8 percent are on track. We observe a huge lag for fragile states, of for girls, which are discussed in the next section, also lag behind those of which 50 percent are seriously off track. boys in some countries, but in general--thanks to strong progress on gender equity in education over the past 15 years--these gaps are smaller than those linked to wealth or location. However, while expansion of primary education coverage tends to be pro-poor, pro-rural, and pro-girls in terms of equalizing access and completion, country experience also shows that specific actions to lower direct and opportunity costs or eliminate discrimination are often needed to keep vulnerable children in school, be they orphan, poor, rural, or female. When a country educates its girls, its mortality rates social and economic obstacles that keep parents from usually fall, fertility rates decline, and the health and sending their daughters to school must be overcome. education prospects of the next generation improve. For many poor families, the economic value of girls' Unequal treatment of women--by the state, in the work at home exceeds the perceived returns to school- market, and by their community and family--puts ing. Improving the accessibility of schools and their them at a disadvantage throughout their lives and sti- quality and affordability is a first step. Globally, 55 fles the development prospects of their societies. Illiter- percent of countries achieved the first target by 2005. ate and poorly educated mothers are less able to care Latin America and Europe and Central Asia can now for their children. Low education levels and responsi- focus on the second target. But huge improvement is bilities for household work prevent women from find- required in Sub-Saharan Africa and South Asia, where ing productive employment or participating in public only 20 percent and 35 percent, respectively, of coun- decision making. To improve girls' enrollments, the tries reached the 2005 target. Ratio of girls to boys enrolled in primary and secondary education Source: World Bank staff estimates (weighted averages). The differences between boys' and girls' East Asia and Pacific has almost achieved the In Europe and Central Asia a strong tradition of schooling are greatest in regions with the lowest 2005 target. In some Latin American countries, educating girls needs to be sustained. In Middle primary school completion rates and the lowest girls' enrollments exceed boys'. East and North Africa more girls are overcoming average incomes. the strong bias against them. TARGET 4: Eliminate gender disparity in primary and secondary education, preferably by 2005, and at all levels of education no later than 2015. Proportion of countries on track to achieve gender parity in Gender inequality in primary and secondary enrollment primary completion rate: The girl/boy gap Source: DHS surveys. Sub-Saharan Africa countries have some of the largest and smallest gender inequality gaps. Source: World Bank staff estimates. In Kenya, Madagascar, and Tanzania, girls' completion rates are over 5 percent higher than boys' completion rates, while boys' completion Even in regions that have achieved the target on average, such as Eastern Europe and Central Asia, some rates are over 10 percent higher in Chad, Ethiopia, countries still fall short. And in South Asia and Sub-Saharan Africa, where large numbers of children Nigeria, and Morocco. are out of school, girls are at a severe disadvantage. Fragile states lag behind in achieving gender parity in enrollment, and more than 50 percent of these countries do not have sufficient data to assess their progress. Share of men and women participating in the labor force, 1991­2004 Although the gender gap in school enrollments has declined in most regions, the gender gap in labor force participation remains. Age patterns of labor force participation show that compared to young men, fewer young women make the transition from school to work, and this gender gap tends to persist throughout the life cycle. However, the size of this gap varies considerably across regions. The gender gap is the largest in South Asia and the smallest in Europe and Central Asia. Source: Computed using household surveys (1991­2004, unweighted averages). Every year over 10 million children in developing annual decline of 4.3 percent over the entire period. countries die before the age of five. Most die from Only two regions, East Asia and Pacific and Latin causes that are readily preventable or curable with America and the Caribbean, are close to achieving existing interventions--such as acute respiratory the MDG target. But even in those two regions, more infections, diarrhea, measles, and malaria. Rapid than half the countries are off track. Progress has improvements prior to 1990 provided hope that mor- been particularly slow in Sub-Saharan Africa, where tality rates for infants and children under five could civil disturbances and the HIV/AIDS epidemic have be cut by two-thirds in the ensuing 25 years, but prog- driven up rates of infant and child mortality. As of ress slowed almost everywhere in the 1990s. Progress 2005, no Sub-Saharan Africa country was on track to on the child mortality MDG lags behind all other achieve the goal, and only 33 out of 147 (22 percent) goals. While the majority of countries have reduced of developing countries are making enough progress child mortality since 1990, progress has been insuf- to achieve the goal on current trends. ficient to reach the MDG target--which requires an Under-five mortality rate (deaths per 1,000) Source: World Bank staff estimates (weighted averages). The gap between goal and reality is greatest in East Asia and Pacific and Latin America and the More than half the countries in the Middle East Sub-Saharan Africa, but millions of children are Caribbean are nearly on track, but the regional and North Africa and Europe and Central Asia also at risk in South Asia. averages disguise wide variations between regions are off track to reach the target. countries. TARGET 5: Reduce by two-thirds, between 1990 and 2015, the under-five mortality rate. Proportion of countries on track to achieve the child mortality Under-five target mortality rate by quintile Source: World Bank staff estimates. Source: DHS surveys. A concerted effort to improve the measurement of infant and child mortality has filled many gaps in the Except in Chad, under-five child mortality is over international data set and reveals that many countries still fall short of achieving the target, even where 40 percent higher in the poorest quintile than regional averages have been improving. Based on estimates through 2005, only 33 countries are on track in the richest quintile. The greatest percentage to achieve a two-thirds reduction in the mortality rate. Every country in Sub-Saharan Africa is off track, disparity is in Egypt, where the number of deaths and in some countries mortality rates have increased since 1990. Some recent surveys have found rapidly per 1,000 live births is nearly 3 times higher for falling mortality rates. the poor than for the rich. Composition of under-five mortality in developing regions based on most recent data, 1995­2003 The gap in under-five mortality between Sub- Saharan Africa and other regions is due mostly to higher child mortality (between ages 1­4), yet neonatal and post neonatal mortality are also highest in the Sub-Saharan Africa Region. Source: WHO report, 2005 (unweighted averages). Death in childbirth is a rare event in rich countries, ity over time, other indicators are often used instead, where there are typically fewer than 10 maternal such as the skilled health personnel who are needed deaths for every 100,000 live births. But in the poor- to deal with the complications of childbirth that can est countries of Africa and Asia the ratio may be 100 claim mothers' lives. Survey evidence shows progress times higher. Ninety-nine percent of maternal deaths in 27 of 32 countries, but also suggests that this is occur in developing countries--around 500,000 highly concentrated among richer households. While annually. The MDG target--to reduce by 75 per- survey data also show progress in coverage for the cent the maternal mortality ratio between 1990 and poorest quintiles in many countries, differences in 2015--remains difficult to measure for almost all access to skilled delivery care between the poorest developing countries. No current direct estimates of and richest quintiles in most countries represent the maternal mortality ratio or trends exist. Because larger equity gaps than for any other health service. few countries are able to measure maternal mortal- Maternal mortality ratios in 2000 Source: World Bank staff estimates. Maternal deaths are still unacceptably high in many developing countries of Sub-Saharan Africa and South Asia as a result of high fertility rates and a high risk of dying each time a woman becomes pregnant. Some developing countries in East Asia and Pacific and Latin America and the Caribbean have substantially improved maternal health through better health services, including increased numbers of trained birth attendants and midwives. Still others, in Europe and Central Asia and Middle East and North Africa, have improved maternal health and significantly lowered fertility rates through the use of contraceptives and increased female education. TARGET 6: Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio. Proportion of countries on track to provide adequate coverage Access to delivery of births by skilled health personnel by medically trained personnel by household wealth quintile Source: World Bank staff estimates. Source: DHS surveys. This figure shows the proportion of countries in each region that provide skilled health personnel for 90 In countries with the lowest access to a medically percent of births or could do so by 2015 based on current trends. Countries that are off track may be able trained personnel for delivery, women in the to achieve 75 percent coverage by 2015, while seriously off-track countries will not reach even that level richest quintile are six times more likely to have unless they make rapid progress in the next decade. More fragile states are seriously off track compared access than women in the poorest quintile. to other developing countries. In most developing countries, the greater the access, the lower the inequality; however, inequality is still high in Bolivia, Cameroon, and Morocco. Adolescent (15­19) fertility rate by household wealth quintile Pregnancy at a young age puts the mother and child at higher risk of serious health consequences. In developing countries, maternal mortality in girls under 18 years of age is estimated to be two to five times higher than in women between 18 and 25. Adolescent fertility rates are higher among poorer people, often substantially so. Poor young women typically have less access to reproductive health services, but the choice to have children very early also reflects low-income girls' lack of access to schooling and limited economic prospects. Source: DHS surveys. Epidemic diseases exact a huge toll in human suffer- in the developing world, and prevention remains a ing and lost opportunities for development. Poverty, crucial challenge. More effective, evidence-based armed conflict, and natural disasters contribute to approaches to prevention are required--drawn from the spread of disease--and recovery, in turn, is often careful evaluation of what works in different con- endangered by disease. In Africa, AIDS has reversed texts and the continued tailoring of responses to the decades of improvements in life expectancy and left changing epidemic. millions of children orphaned. By end-2006 an esti- Increasing the awareness of the impact of malaria mated 39.5 million people globally were living with and tuberculosis on human development has been HIV, up 2.4 million since 2004, and an estimated matched with a commitment to fight these diseases, three million people had died from AIDS. While the and fight them more effectively. There are an esti- spread of AIDS has slowed in parts of Sub-Saharan mated 300­500 million new cases of malaria each Africa, it remains the center of the epidemic; home to year, leading to more than 1 million deaths. Nearly just over 10 percent of the world's people, 64 percent all the cases and more than 95 percent of the deaths of all HIV-positive people, and 90 percent of all HIV- occur in Sub-Saharan Africa. Tuberculosis (TB) positive children. About 60 percent of HIV-positive strikes 9 million people each year and kills 2 mil- adults in Africa are women. lion. But there has been clear progress in reducing The largest recent increases in the number of peo- TB prevalence and deaths in recent years. The only ple with HIV have been in Eastern Europe, and Cen- region where TB incidence is still growing is Africa, tral and East Asia (21 percent higher in 2006 than because of the emergence of drug-resistant strains in 2004). Recent experience in combating the spread and the greatly reduced resistance to TB among peo- of AIDS has demonstrated three important messages: ple with HIV. reversing its spread is possible, treatment is effective HIV prevalence and deaths in the developing world, 1990­2006 Source: UNAIDS/WHO, November 2006. TARGET 7: Have halted by 2015 and begun to reverse the spread of HIV/AIDS. TARGET 8: Have halted by 2015 and begun to reverse the incidence of malaria and other major diseases. Number of people receiving antiretroviral therapy by region Impressive progress has been made in extending antiretroviral coverage in Sub-Saharan Africa, where the number of people receiving treatment has more than doubled since 2004. While the number of people receiving antiretroviral therapy (ART) in East, South, and Southeast Asia has increased rapidly, progress in Europe and Central Asia and in North Africa and the Middle East has been less dramatic. Nonetheless, antiretroviral treatment in the developing world still reaches just 24 percent of those who need it. Source: WHO and UNAIDS report, 2006. Tuberculosis prevalence and number of TB deaths, 1990­2005 Source: UN statistics. Many developing countries have successfully reduced TB's prevalence since 1990. Sub-Saharan Africa is the only region where TB prevalence continues to increase; TB-related deaths reached 600,000 in 2004. Sustainable development can be ensured only by pro- the developing world and a major obstacle to contin- tecting the environment and using resources wisely. Less ued poverty reduction across many dimensions. First, than 20 percent of developing countries are on track developing regions are at a geographic disadvantage: or have achieved the 2015 target to increase access to they are already warmer, on average, than developed water, and less than 35 percent have increased access regions, and they suffer from high rainfall variabil- to sanitation, but Sub-Saharan African countries are ity. Second, developing countries--in particular the lagging behind other regions. And in the fragile states, poorest--are heavily dependent on agriculture, the the proportion of countries on track to achieve the most climate-sensitive of all economic sectors, and target for increased access to water and sanitation is 6 they suffer from inadequate health provision and low percent and 15 percent, respectively. quality public services. Third, low incomes and vul- Around the world, land is being degraded and car- nerabilities make adaptation to climate change par- bon dioxide (CO2) emissions are driving changes in ticularly difficult. Global emissions of CO2 rose by 4 global climate. Climate change is a grave threat to billion metric tons between 1990 and 2003. Population without access to an improved water source or sanitation facilities Source: World Bank staff estimates (weighted averages). In Sub-Saharan Africa, 300 million people lack In East Asia, rapid urbanization is posing a Many countries in Europe and Central Asia lacked access to improved water sources, and 450 challenge for the provision of water and other reliable benchmarks for measuring improved million lack adequate sanitation services. South public utilities. Latin America and the Caribbean, access to water and sanitation in the early 1990s. Asia has made excellent progress in providing the most urban developing region, has made In the Middle East and North Africa, Egypt, water, but progress has been slower in providing slow progress in providing sanitation. Morocco, and Tunisia have made the fastest sanitation. progress. TARGET 9: Integrate the principles of sustainable development into country policies and programs and reverse the loss of environmental resources. TARGET 10: Halve by 2015 the proportion of people without sustainable access to safe drinking water and basic sanitation. TARGET 11: Have achieved a significant improvement by 2020 in the lives of at least 100 million slum dwellers. Proportion of Adjusted net saving countries on track to achieve the target for access to improved water Source: World Development Indicators. Note: Numbers above bars show percent change from 1995 to 2005. Are countries saving enough for future growth? Adjusted net saving measures the rate of saving in an economy after taking into account investments in human capital, depreciation of produced capital, depletion of natural resources, and damage caused by pollution. A negative saving rate implies that current levels of welfare and growth may be threatened by resource depletion. The Middle East and North Africa, Sub-Saharan Africa, and Eastern Europe and Central Asia had negative saving rates in 2005 when depletion of natural resources was taken into account. The largest decline in saving between 1995 and 2005 occurred in these three regions, while the largest gain, 3.7 percent of GNI, was in South Asia. Source: World Bank staff estimates. Lack of clean water is the main reason that diseases transmitted by feces are so common in developing countries. Water is a daily need that must be met, but in some places people spend many hours to obtain water from sources that are not protected from contamination. Even the modest target of halving the number of people without access to an improved water source will not be met in many countries at the current rate of progress. Only 35 percent of countries are on track to achieve or have achieved the target. CO2 emissions, 1990­2003 Source: World Development Indicators. Note: Numbers above bars show percent change from 1990 to 2003. Carbon dioxide (CO2) is a greenhouse gas that contributes to global climate change. Global emissions of CO2 from burning fossil fuels and manufacturing cement rose by 4 billion metric tons between 1990 and 2003. Most of the increase in these emissions came from high-income countries (2.09 billion metric tons) and East Asia and the Pacific (2.07 billion metric tons). South Asia and the Middle East and North Africa have regions with the largest percentage increase in emissions, followed by East Asia and the Pacific. Conversely, owing to the economic recession and restructuring of the 1990s, the transition economies of Europe and Central Asia emitted less CO2 in 2003 than in 1990. Important steps toward global partnership were Total aid rose in recent years through 2005, and taken at the international meetings in 2001 in Doha, declined 5 percent in 2006. But much of the recent which launched a new "development round" of trade increase was due to debt relief, and this may provide negotiations, and in 2002 at the International Con- less than full additionality as measured by the current ference on Financing for Development in Monterrey, flow of new resources for development. Mexico, where high-income and developing coun- Owing to the steady reduction of tariffs, overall tries reached consensus on mutual responsibilities for trade restrictiveness has largely declined in recent achieving the Millennium Development Goals. The years. However, the poorest developing countries consensus calls for developing countries to improve faced the highest barriers, notably from developed governance and policies aimed at increasing eco- countries. South-South trade faces a high level of pro- nomic growth and reducing poverty and for high- tection. Most of this protection is in agriculture. income countries to provide more and better aid and greater access to their markets. Overall Trade Restrictiveness faced by countries in 2006 The effect of policies on exporters' access to markets differs by region. South Asian, Sub- Saharan African, and Latin American and Caribbean countries faced the highest barriers to their exports, since they export mainly agricultural products. For South Asia and Sub-Saharan Africa, restrictions by developed countries are especially high. East Asia and Pacific countries face less restrictions; the same is true for Europe and Central Asia and Middle East and North Africa. Source: OTRI (World Bank). Share of total aid toward fragile states Donor focus on fragile states is translating into substantial assistance to some of those countries and the group as a whole. On average, DAC countries allocated 20 percent of bilateral aid to fragile states in 2003­05. However, more than half of fragile states received less ODA in 2005 than in 2001. Aid flows were dominated by debt relief; several donors provided over 50 percent of their aid in debt relief. Humanitarian aid also accounted for a substantial share of assistance to fragile states. By contrast, "other ODA," which traditionally finances development projects and programs, was less than a quarter of aid. Source: OECD DAC Development Committee Report and DAC database. Note: Numbers above bars reflect average ODA volumes given to fragile states, 2003­05 (2004 US$ billions). TARGET 12: Develop further an open, rule-based, predictable, nondiscriminatory trading and financial system. TARGET 13: Address the special needs of the least developed countries. TARGET 14: Address the special needs of landlocked developing countries and small island developing states. TARGET 15: Deal comprehensively with the debt problems of developing countries through national and international measures in order to make debt sustainable in the long term. TARGET 16: In cooperation with developing countries, develop and implement strategies for decent and productive work for youth. TARGET 17: In cooperation with pharmaceutical companies, provide access to affordable essential drugs in developing countries. TARGET 18: In cooperation with the private sector, make available the benefits of new technologies, especially information and communications. Sectoral allocation of DAC members' Selected indicators on aid harmonization bilateral aid and alignment in 2006 Source: OECD DAC database. Source: OECD DAC database. In several areas, donors are coming quite close to reaching the 2010 goals for harmonization and alignment. The largest gap involved the use of The chart shows the breakdown of bilateral, sector-allocable aid, by social country procurement systems, an area slowed by concerns over the quality of services, economic infrastructure (roads . . .), sector production, and financial management. multisector (environment . . .). The share of aid devoted to government and civil society has increased. Also, the shares of aid for agriculture, industry, and economic infrastructure have declined. Growth, Poverty Reduction, and Environmental Sustainability U nder the first Millennium Develop- absorb external funds and to mobilize inter- ment Goal (MDG1), the interna- nal resources for sustained poverty reduction tional community aims to halve the and improved economic security. This chapter global rate of extreme income poverty--as focuses on the growth and macroeconomic measured by the share of the population liv- policies of fragile states, while later chapters ing on less than $1 per day--between 1990 deal with other aspects. and 2015. Current trends and growth fore- The chapter also reports on recent prog- casts indicate that this goal will be achieved, ress in further areas covered in last year's although not in Sub-Saharan Africa. High Global Monitoring Report (GMR) that are growth in China and India explains much central to achieving higher sustained growth, of the reduction in the global poverty rate, promotion of a better investment climate, although progress toward MDG1 has also and improvements in governance. A better quickened in many other developing coun- investment climate is key to attaining higher tries. High growth has continued in most of growth and employment creation, while, as the developing world in the past year as a noted in last year's GMR, governance is an result of better policies in developing coun- ongoing part of MDG monitoring, because it tries and a favorable global environment. is an important factor underpinning a coun- The outlook for growth and poverty reduc- try's development effectiveness and progress tion remains favorable, although some risks toward the MDGs. remain. In particular, low-income country per While higher economic growth is gener- capita growth is expected to remain above 5 ally desirable, one should also be aware of percent in 2007.1 its environmental costs. Although the recent Addressing the problems of fragile states boom in commodity prices has helped to (box 1.1) is central to the development underpin strong growth in many of the most agenda and to furthering progress toward natural resource­dependent economies, high the MDGs.2 Nine percent of the population, resource dependence can lead to high rates of and about 27 percent of the extreme poor resource depletion. Countries are liquidating in developing countries live in fragile states. assets when they extract minerals and energy, This situation will not improve unless frag- harvest forests and fish unsustainably, or ile states become less vulnerable to adverse deplete their agricultural soils, and this can shocks, and they increase their capacity to have consequences for future growth. Fragile states is the term generally used to refer to countries that are facing particularly severe development challenges such as weak governance, limited administrative capacity, violence, or the legacy of conflict. In defining policies and approaches toward fragile states, different organizations have used different criteria and terms. Despite methodological variations, however, development partners have been converging around an approach developed at the OECD, which recognizes com- mon characteristics of weak governance and vulnerability to conflict, together with differentiated constraints and opportunities in fragile situations of (1) prolonged crisis or impasse, (2) postconflict or political transition, (3) gradual improvement, and (4) deteriorating governance.a While important for the development of shared strategic and operational approaches, the OECD- DAC typology does not generate a country time series that can be used for research purposes. This year's GMR uses the World Bank definition of fragile states, which is based on a measure of the coun- tries' Country Policy and Institutional Assessment (CPIA) and governance scores.b The CPIA-based definition also has the advantages of (1) being a multidimensional concept; (2) being development- oriented, (3) stemming from a robust, review-based process; (4) giving weight to governance, a crucial variable that reflects the capacity of states; and (5) strongly correlating with conflict-related variables. a. "Fragile States: Policy Commitment and Principles for Good International Engagement in Fragile States and Situations," DAC Senior Level Meeting, December 5­6, 2006. b. The World Bank definition covers countries scoring 3.2 and below on the CPIA. This is similar to the bottom two quintiles of the CPIA, which the OECD-DAC has used for research purposes on fragile states, but has the advantage of being an absolute rather than a relative threshold, allowing the total number of countries covered to vary from year to year depending on changes in performance. This classification-- previously referred to as "Low Income Countries Under Stress" (LICUS)--has been in use in the Bank since 2003; CPIA scores over the years 1998 to 2005 are used to determine what states were fragile over this time period. For years before 1998, cutoff values were determined by comparing the distribution of the CPIA in each year with that for 1998­2001. Since it is determined for each year, fragility is a status, not a permanent classification. Countries may thus be intermittently fragile, although the data used throughout this report are smoothed to avoid excessive volatility in the classification of borderline cases. Gender equality--in the sense of equal- ity of opportunities, not outcomes--plays Progress on Poverty Reduction an important role in development. Cross- country data show an inverse relationship The prospects for achieving MDG1--halving between the incidence of poverty and the poverty by 2015--are largely unchanged from level of gender equality as measured by the last year's Global Monitoring Report. Overall, rate of female labor market participation. the world as a whole is on track to meet the Greater gender equality in access to educa- goal with the population share of the extreme tion, land, technology, and credit markets is poor in developing countries projected to also associated with lower poverty. While fall from 29 percent in 1990 to 12 percent the direction of causality of these relation- in 2015. By 2004, over halfway through the ships is unclear, it is evident that higher gen- goal period, this share had already dropped der equality is associated with better MDG to 18 percent. Preliminary estimates suggest outcomes, including higher nutritional sta- that the number of extremely poor people tus and lower poverty. These themes are in developing countries fell by 135 million explored in chapter 3. between 1999 and 2004. Progress toward the poverty MDG target 1990­2004, and a forecast for 2015 Source: World Bank staff. Note: The graphs show preliminary data with growth forecasts under review. This positive assessment overshadows sig- and Caribbean region is likely to come close. nificant regional differences (see figure 1.1). However, the main drivers of poverty reduc- Sub-Saharan Africa remains a long way off the tion globally continue to be countries in the path that would take it to MDG1, even assum- East Asia and Pacific and South Asia regions, ing projected growth rates higher than the his- which--thanks to spectacular rates of growth toric averages since 1990. Between 1999 and in the last decade--are both set to overshoot 2004, the share of people in extreme poverty in the poverty target. By 2015, extreme poverty the region fell to 41 percent, a decline of 4.7 per- rates are projected to be below 3 percent for centage points, but higher population growth EAP countries, and 18 percent for SA coun- left the same absolute number of poor at nearly tries, as compared to MDG1 targets of 15 and 300 million. The region now accounts for 30 22 percent respectively. percent of the world's extreme poor, compared For a number of countries,3 it is possible with 19 percent in 1990 and only 11 percent in to go beyond the regional estimates pre- 1981. The Europe and Central Asia region has sented in figure 1.1, and use poverty esti- lost ground since 1990, and may not meet the mates from household surveys to examine development goal. The Middle East and North whether, for a typical country, the upturn in Africa region is expected to achieve MDG1, growth since the late 1990s led to poverty albeit narrowly, while the Latin America reduction (table 1.1). The countries included Impact of growth of GDP per capita on povertya Averages (percent) Annual Annual Annual Percent Percent Percent Initial Change in Initial Change in Growth Poverty Poverty Poverty Poverty Initial Annual Region or Number of in GDP Rate Rate Level Rate Gini Diff. Income Grouping Countries per capita ($1/day) ($1/day) Elasticityb ($2/day) ($2/day) Elasticityb Index in Gini Low-income countries 19 3.7 23.5 ­4.85 ­1.32 54.0 ­3.12 ­0.85 40.5 ­0.12 Sub-Saharan Africa 5 1.7 48.8 ­2.50 ­1.46 78.5 ­1.05 ­0.61 47.1 ­0.14 East Asia and Pacific 4 3.2 16.9 ­9.36 ­2.91 60.8 ­4.04 ­1.26 34.5 ­0.02 Europe and Central Asia 7 6.0 8.6 ­26.05 ­4.31 33.7 ­9.81 ­1.62 36.3 ­0.41 Middle-income countries 26 2.1 5.7 ­1.51 ­0.72 19.7 ­1.06 ­0.51 44.7 0.20 Europe and Central Asia 8 5.2 0.8 ­2.32 ­0.45 9.9 ­1.27 ­0.25 33.7 0.45 Latin American and 14 ­0.1 7.7 0.02 ­0.21 21.5 0.20 ­2.78 50.5 0.07 the Caribbeanc Memorandum items India (1994­2005) 4.55 42.1 ­1.47 ­0.32 85.5 ­0.48 ­0.11 32.8 0.36 China (1999­2004) 8.11 17.8 ­11.73 ­1.45 81.1 ­7.23 ­0.89 41.6 0.44 Source: World Bank and IMF staff. a. Two surveys for each country were undertaken at intervals of three to eleven years. The last survey for each country was undertaken between 2002 and 2005. Estimates of GDP per capita growth and changes in poverty rates are annualized proportional changes in cross-country averages. All averages are unweighted. b. Percentage change in poverty rate divided by percentage growth in GDP per capita. c. Includes data for Argentina and Uruguay based on urban household surveys. are those with household surveys conducted ent for middle-income countries where the during both the middle/late 1990s and after impact of GDP percapita growth on poverty 2001. The results must be interpreted with was less. While a high negative elasticity caution in view of possible survey measure- was obtained for the Latin America and the ment and sampling errors, and, in view of the Caribbean sample by the $2-a-day poverty limited number of countries for which there defintion, this reflects increased poverty in are appropriate data, may not be representa- a context of near-zero negative growth. One tive of entire regions or country groups. Fur- hypothesis is that the poor in the middle- thermore, the relationship between growth income countries examined were drawn and poverty may be obscured by changes in relatively heavily from economically produc- relative prices, taxes and transfers, including tive groups, who did not enjoy the benefits worker remittances, and, as noted below, of growth given its sectoral and geographic changes in income distribution. composition, and from groups such as retir- In low-income countries the preliminary ees and the unemployed, who may depend estimates suggest that, on average, growth substantially on public transfers. has clearly resulted in lower poverty inci- There was also a somewhat different dence: for a sample of 19 low-income impact of growth on poverty incidence in countries, 1 percent of GDP growth was China and India. In China, high growth led associated with a 1.3 percent fall in the rate to very substantial decreases in poverty rates, of extreme poverty and a 0.9 percent fall while in India, the gains in poverty reduction in the $2-a-day poverty rate. Clear poverty were more modest. In both countries, poverty impacts are also evident in the three regions reduction took place despite a worsening of for which sufficient country-level data are the income distribution. Between 1981 and available. The picture is somewhat differ- 2004, there was an estimated decline in the absolute number of extreme poor in China of Improvements in Long-Term Growth over 500 million people, while in India, the It is reassuring that the pick-up in low-income- number of extreme poor remained roughly country per capita growth rates that started in constant (see annex table 1A.3). the 1990s continued in 2006 with an estimated Changes in income distribution have not, overall per capita GDP growth of 5.9 percent, on average, reduced the impact of income up from an average of 4.0 percent in 2001­05 growth on poverty reduction in low-income (table 1.2). As in previous years, most regions countries. Inequality in income as measured show strong growth performance, with a par- by the Gini index declined on average for the ticularly impressive rate of growth in the low- overall sample of low-income countries. In income countries of Europe and Central Asia, contrast, income inequality widened on aver- which are still experiencing a rebound after age in middle-income countries, thus hinder- the transition recession of the mid-1990s. The ing poverty reduction. region continues to benefit from strong com- Per capita GDP growth for high-, middle- and low-income countries 1986­90 1991­95 1996­2000 2001­05 2004 2005e 2006f 2007f Real per-capita GDP growtha 1.8 0.8 2.0 1.5 2.9 2.3 2.9 2.2 Memo item: World (PPP weights)b .. .. 3.9 4.0 5.2 4.7 5.3 4.5 2.9 1.4 2.4 1.4 2.6 2.0 2.6 1.9 2.9 1.8 2.2 4.0 5.0 5.6 5.9 5.4 4.7 5.4 0.4 3.8 4.1 4.9 4.7 5.1 Europe and Central Asia 6.6 ­11.3 3.8 6.8 7.1 10.2 11.5 9.3 Latin America and Caribbean ­1.1 ­0.3 1.4 0.7 1.8 1.7 1.8 1.8 Middle East and N. Africa 6.8 0.9 2.1 0.4 ­0.7 0.6 0.8 ­0.6 3.6 3.0 3.5 4.7 6.1 6.5 6.8 5.9 2.0 2.3 1.9 2.9 4.2 4.8 4.4 4.5 4.1 3.2 4.0 5.2 6.7 7.0 7.5 6.4 Sub-Saharan Africa 0.1 ­1.6 1.0 2.4 3.1 4.0 4.0 4.4 1.0 1.8 3.3 4.3 6.4 5.6 6.2 5.4 6.1 9.6 6.2 7.9 8.7 8.5 9.1 8.1 5.4 4.9 1.0 3.2 5.0 3.3 4.0 3.7 6.3 11.1 7.7 8.8 9.4 9.5 10.0 8.9 Europe and Central Asia 0.6 ­4.5 2.9 5.2 7.3 6.0 6.4 5.6 Latin America and Caribbean ­0.2 1.8 1.6 0.9 4.5 3.1 4.0 2.9 Middle East and N. Africa ­0.6 1.7 2.3 3.0 3.2 2.7 3.3 3.2 .. .. .. .. .. .. .. .. Sub-Saharan Africa ­0.2 ­1.3 1.4 3.0 3.8 4.4 4.2 3.3 3.4 ­2.6 0.9 2.5 3.4 4.7 4.6 4.7 Memorandum items Developing countries 1.1 1.5 2.8 3.9 5.8 5.3 5.9 5.2 excluding transition countries 1.6 3.4 2.9 3.8 5.8 5.3 5.9 5.1 excluding China and India 0.2 ­0.4 1.5 2.2 4.5 3.6 4.2 3.5 Source: World Bank. Note: PP = purchasing parity; e = estimate; f = forecast. a. GDP in 2000 constant dollars; 2000 prices and market and exchange rates. b. GDP measured at 2000 PPP weights. modity prices and export earnings. In South performer with an estimated per capita growth Asia, growth in India continues at a formida- of 10 percent in 2006. But other middle-income ble pace, but other countries in the region are countries in the region and elsewhere are also also doing well with the exception of Nepal, growing at sustained rates, thus improving which has been suffering from political unrest. prospects for the gradual reduction of the Most importantly, in view of the high poverty pockets of poverty that still exist in these in the region, Sub-Saharan African countries countries. Recent outcomes suggest that per are also experiencing sustained and rising capita growth rates in middle-income coun- growth rates. Oil-exporting countries have tries have increased, with average rates in the contributed significantly to this strong perfor- last few years significantly and consistently mance. Increased oil production and the large higher than pre-2000 values. terms-of-trade gains from the oil price hike have boosted domestic incomes and spend- Weak Growth and Less Poverty ing. Non-fuel-exporting African countries Reduction in Fragile States seem to have weathered the adverse shock of high oil prices well, thanks to a mixture of Fragile states have consistently grown more improved policies and strong non-fuel com- slowly than other low-income countries modity prices. In contrast with the high rates (table 1.3). Although the average per capita of per capita growth in other regions, growth growth of such states has picked up in recent among low-income countries in the Middle years, this is partly due to accelerated expan- East and North Africa and Latin America sion in a few fuel-producing countries and a and the Caribbean regions continues to be fall in the number of conflicts. Among non- much lower. fuel-producing fragile states, while growth Growth in middle-income countries also has increased since 2000, the outlook is for continues to be strong. China remains the star per capita growth to remain a full percentage Real per capita growth and investment and savings rates of fragile and nonfragile states (percent) Real per capita growth, investment and savings rates 1986­90 1991­95 1996­2000 2001­05 2004 2005 2006e Per capita GDP growth Fragile states 0.1 ­2.5 0.3 2.1 3.2 3.3 2.6 Fuel producers 1.7 ­5.5 2.4 2.9 2.2 5.9 4.0 Non­fuel producers ­0.1 ­2.0 ­0.1 2.0 3.4 2.8 2.3 Nonfragile states 1.3 ­0.1 2.9 3.0 3.5 4.1 4.5 Gross fixed capital formation/GDP Fragile states 16.1 17.7 17.7 18.3 18.3 19.2 20.4 Fuel producers 16.9 21.5 25.7 20.4 18.4 21.4 31.1 Non­fuel producers 16.0 17.0 16.4 18.0 18.3 18.8 18.6 Nonfragile states 21.4 24.6 22.6 24.1 25.5 25.2 25.6 Gross national savings/GDP Fragile states 11.3 11.1 9.8 12.6 15.7 15.8 18.8 Fuel producers 14.3 15.5 14.7 23.6 28.7 38.9 47.7 Non­fuel producers 10.9 10.3 8.9 10.7 13.6 11.0 12.8 Nonfragile states 14.6 15.9 14.4 16.6 18.0 17.2 18.6 Source: IMF staff. Note: Unweighted country averages; e = estimate. point lower than that experienced by low- Rates of extreme poverty (percent) income countries as a whole. Lower invest- ment relative to GDP in fragile states linked in part to lower national savings rates (domestic savings and net transfers from abroad, includ- ing official transfers and worker remittances) has been one cause of their slower growth. Clearly the inferior growth performance of fragile states has been, and is likely to con- tinue to be, an obstacle to the achievement of MDG1. Fragile states by the LICUS definition are home to 9 percent of the population of developing countries, and have nearly twice the incidence of extreme poverty of other low-income countries. About 27 percent of the extreme poor in developing countries live in fragile states. Moreover, fragile states can have adverse spillovers on neighboring coun- tries through conflict, refugee flows, orga- nized crime, spread of epidemic diseases, and barriers to trade and investment.4 Source: World Bank staff calculations. The rate of extreme poverty in the current set of fragile states is estimated to have risen somewhat in 1990­2004 from 49 percent to over 54 percent (figure 1.2). The projected ­12 percent per year of conflict. While in the poverty rate for this group of countries in past, the fall in growth was more gradual, 2015 is slightly higher than in 1990 under and was followed by a gradual and pro- current assumptions about future growth longed recovery within the conflict period, and income distribution, suggesting that no since 1990 the period of the growth collapse overall progress will be made toward MDG1 has largely coincided with that of the conflict, over the goal period as a whole. In contrast, leading to this higher annual GDP loss. It nonfragile states made significant progress in has also taken longer for countries to regain reducing poverty by 2004, and are projected their preconflict per capita income levels than to overachieve MDG1 by 2015. would have been the case before 1990. Conflicts have undermined growth perfor- Because conflict is both a major cause mance at various times in most fragile states. and consequence of poverty in fragile states, Conflicts are a major reason why countries the coherence and sequencing of interna- slide into fragility; they extract high costs in tional diplomatic, security, and development terms of lives and physical damage, but also engagement is more important in these envi- reduce growth and increase poverty. There ronments than elsewhere. Recent research is consensus in the relevant literature5 that (for example, Chauvet and Collier 2004) civil conflict reduces gross domestic product demonstrates that the risk of reversion to (GDP) growth, although estimates of the size conflict is significantly higher in the period of this impact vary. The impact of conflict on following postconflict elections than in the growth and poverty incidence seems to have period preceding elections. This increased worsened since the beginning of the 1990s risk does not diminish for the first postcon- (see Staines 2005). Conflicts have become flict decade. In discussions of these results shorter and more intense than before; their at the United Nations (UN) Peace-Building average impact on GDP growth is now about Commission, participants noted that this risk may have important implications for cal order. Subsequently in all three countries, the sequencing of electoral, peacekeeping, growth was enabled by the introduction of and development assistance, underlining the at least modest programs of market-oriented importance of efforts to ensure that electoral economic reform that were managed so as to assistance in fragile transitions is properly keep interested elites on board. sequenced with decisions to maintain or Limited capacity and willingness to under- draw down peace-keeping troops, and with take needed reforms in fragile states under- aid-financed efforts to support measures to mine the mainstream poverty reduction generate growth and employment and other approach based on partnership as exempli- initiatives that may mitigate the risks of fied by the Poverty Reduction Strategy Paper reversion to conflict. (PRSP). Difficulties donors experience when Conflict aside, all fragile states have working in these countries, particularly the weak institutions and governance, hinder- ones with limited geopolitical relevance, can ing growth.6 Some states may be willing to lead to excessively low or volatile aid flows promote growth and reduce poverty, but even after taking into account the coun- are unable to do so for a variety of reasons tries' low level of governance (see OECD/ such as a lack of territorial control, politi- DAC 2005). The international community cal cohesion, and administrative capacity. In is increasingly aware of issues particular other states, governments may be unwilling to fragile states, and has been considering to take necessary actions because they are alternative approaches tailored to the char- not substantively committed to overall pov- acteristics of specific countries, for example, erty reduction, or they may promote poverty emphasizing humanitarian assistance and reduction while excluding certain social or relying where possible on help from nonstate geographical groups. actors such as nongovernmental organiza- State fragility has proven to be a persistent tions (NGOs). In this context, the OECD/ condition. Of the 34 states judged as frag- DAC has recently issued a set of "Principles ile in 1980, 21 were still viewed as such in for good international engagement in fragile 2005, although of these, 6 had left and later states." resumed fragile status during the period. The average duration of fragility among the 2005 Macroeconomic Performance group of fragile states was 16.6 years. For the 20 countries that entered and permanently Continued good macroeconomic poli- left the fragile states list since 1980, the aver- cies--as shown by continued low inflation age duration of fragility was 7.8 years. Of and budget deficits--have helped under- these, Mozambique experienced the shortest pin improved growth performance in low- duration of fragility (3 years), and Niger the income countries (table 1.4). At about 6.2 longest (15 years). percent, median inflation in 2006 is esti- Nevertheless there are some success sto- mated to have decreased from a 2005 peak ries. Specifically, Vietnam, Mozambique, and of 7.2 percent associated with the sharp rise Uganda have graduated from fragile state sta- in oil prices, and is forecast to slow further tus. All three experienced severe violent conflict in 2007. Since 2000, inflation has been sub- but managed to achieve a durable cessation of stantially lower than 10 years earlier. The hostilities. Conflict ended either because there external indebtedness of low-income coun- was a change in geopolitical conditions that tries relative to GDP has also been declining, provided incentives for warring parties to in part reflecting the impact of the Heavily lay down their arms, or because there was a Indebted Poor Countries (HIPC) Initiative military victory by one party involved in the and the Multilateral Debt Relief Initiative conflict that eliminated opposition groups or (MDRI). In 2006, the average debt-to-GDP gave them a stake in the postconflict politi- ratio was 61 percent, compared to over 90 Macroeconomic indicators for low-income countries Annual averages, except where indicateda 1986-90 1991-95 1996-2000 2001-05 2006 est. 2007 proj. (median annual %)b Low-income countries 7.1 14.3 6.8 5.4 6.2 5.3 Fragile states 10.2 19.8 9.1 7.7 6.8 5.5 Nonfragile states 6.3 11.4 6.2 4.9 5.9 5.2 Middle-income countries 9.9 18.0 6.8 4.4 4.6 4.6 (% of GDP) Low-income countries 84.9 97.5 92.2 89.5 61.0 51.1 Fragile states 129.9 115.9 116.1 110.1 81.7 63.2 Nonfragile states 57.0 85.2 75.2 76.0 47.7 43.3 Middle-income countries 44.3 46.7 44.5 46.9 41.9 40.2 (% of GDP) Low-income countries ­6.5 ­6.9 ­4.9 ­3.5 0.4 2.2 Fuel producers ­8.7 ­11.2 ­5.6 6.7 23.9 58.7 Non­fuel producers ­6.4 ­6.6 ­4.9 ­4.3 ­1.6 ­2.7 Fragile states ­10.0 ­7.8 ­4.9 ­2.6 3.8 10.4 Of which: Fuel producers ­11.5 ­12.1 ­7.2 8.2 29.3 71.4 Non­fuel producers ­9.9 ­7.1 ­4.7 ­4.5 ­1.4 ­1.8 Nonfragile states ­4.2 ­6.3 ­4.9 ­4.1 ­1.8 ­3.2 Middle-income countries ­3.5 ­2.7 ­3.2 ­2.6 ­0.7 ­1.2 Source: IMF staff. a. Averages are calculated as unweighted means of country values. b. Median inflation is calculated from the annual medians and then averaged over five-year periods. percent throughout the 1990s. This decline debt indicators are also higher, reflecting in is particularly apparent for HIPCs that have some cases excessive past external borrow- passed the HIPC Initiative completion point, ing. In addition, fragile states have found it for which the external debt-to-GDP ratio difficult to satisfy the conditions for reach- in 2006 was half its average for 2001­05. ing the HIPC Initiative completion point The dramatic swing in average fiscal bal- and hence debt relief under the MDRI. Of ances from a deficit to a small surplus in the states classified as fragile in 2005, only low-income countries in 2006 is mainly three, Mauritania, São Tomé and Principe, explained by sharp increases in oil revenues and Sierra Leone, had reached the comple- in a few fuel producers. However, in non- tion point as of end-March 2007. Although fuel-producing countries there has been a fuel-producing fragile states have recently reduction in the size of fiscal deficits relative attained large fiscal surpluses through high to GDP since the late 1990s. oil export revenues or oil-related fees and Fragile states' macroeconomic indicators transfers, prior to the early 2000s, fiscal defi- have tended to be inferior to those of other cits relative to GDP among fuel-producing low-income countries. Until recently, inflation fragile states were consistently higher than rates were on average at least 2.9 percentage in non-fragile states, reflecting limited fiscal points higher than in nonfragile states, pos- discipline. Deficit ratios have, however, been sibly because of recourse in some countries similar in non-fuel producing fragile states to to monetary financing of the budget. External those of nonfragile states. Quality of Macroeconomic Policies favorable ratings regarding the consistency of their policy mix. For the fourth consecutive year, International Consistent with the evidence on macro- Monetary Fund (IMF) staff have carried out economic indicators, assessments of mac- assessments of the quality of macroeconomic roeconomic policies in fragile states are policies in each low-income country (table markedly more negative in some areas than 1.5). In addition to providing a snapshot of those for low-income countries as a whole. the quality of the main dimensions of macro- The composition of public spending receives economic policies for each year, these assess- a much worse assessment in fragile states, ments can be used to evaluate developments reflecting the inappropriateness of expendi- since 2003, the first year of the exercise. ture composition for poverty reduction in The assessment of fiscal policy continues these countries. The picture for monetary to be mixed: almost 50 percent of countries policy and the financial sector is more mixed. have earned a good rating, but 21 percent are The quality of monetary policy is consid- regarded unsatisfactory. A significant propor- ered good for a similarly large proportion of tion of countries have moved out of the unsat- fragile and nonfragile states, underlining the isfactory category--a marked improvement relative insulation of monetary authorities compared to 2003. However, the composition from weaknesses in administrative capacity. of expenditures continues to be rated unsat- However, the governance and transparency isfactory in almost half of low-income coun- of monetary and financial institutions is seen tries. In contrast, access to foreign exchange, as worse in fragile states. That said, there is the quality of monetary policies, and the gov- significant variance across the group: coun- ernance and transparency of monetary and tries such as Timor-Leste, and more recently, financial institutions have consistently rated Haiti and Liberia have made significant prog- relatively well, with a majority of countries ress in this regard. rated good and a relatively small percentage Although the assessments are not strictly rated unsatisfactory. In addition, more than comparable,7 the World Bank 2005 CPIA rat- half the countries surveyed in 2006 received ings of low-income-country macroeconomic Quality of macroeconomic policies in low-income countries, 2006 Share of countries falling into each category (percent) Governance in Access to Composition of Monetary Consistency of monetary and foreign Rating Fiscal policy public spending policy macro policies financial institutions exchange 2006 survey Unsatisfactory 20.5 48.7 10.3 15.4 15.4 3.8 Adequate 33.3 38.5 19.2 32.1 28.2 12.8 Good 46.2 12.8 70.5 52.6 56.4 83.3 2003 survey Unsatisfactory 33.8 49.4 11.7 22.4 17.1 9.2 Adequate 19.5 32.8 11.7 28.9 22.4 13.2 Good 46.8 18.2 76.6 48.7 60.5 77.6 Fragile states (2006) Unsatisfactory 46.7 70.0 20.0 30.0 30.0 10.0 Adequate 26.7 20.0 10.0 30.0 33.3 20.0 Good 26.7 10.0 70.0 40.0 36.7 70.0 Source: IMF staff assessments. policies are broadly similar to those of IMF they cannot be ruled out in view of possible staff. In particular, the CPIA ratings indicate stronger-than-expected demand and the less satisfaction with fiscal policies than with ongoing instability in the Middle East. There macroeconomic policies as a whole: 36 per- is also a danger that protectionism could rise cent of countries were given a score of 3.0 or in the years ahead, reversing some of the less for fiscal policy, compared with 21 percent gains from an increasingly integrated global for macroeconomic management. Roughly economy. Lastly, the chances of a global pan- consistent with the relatively low assessments demic derived from avian influenza remain. given by IMF staff for the composition of There are also some risks that could impinge public spending, the CPIA also shows that more directly on the growth prospects of devel- 37 percent of all low-income countries and oping countries. As noted above, the negative 78 percent of fragile states score 3.0 or less impact of high oil prices on many non-fuel- regarding the equity of public resource use. commodity-exporting developing countries has been limited by the improvement in their terms of trade arising from strong demand Prospects for the Global Economy growth. In the future, however, while a fall in The world economy is growing at a pace last oil prices is unlikely, a relative decline in non- seen at the beginning of the 1970s. This is fuel-commodity price could occur. In addi- welcome news for developing countries in tion, a rise in real interest rates in developed view of its implications for trade, aid, private countries could create turbulence in emerging- financial flows, and remittances. In 2006, market financial sectors, with possible adverse the United States continued to expand at a macroeconomic consequences. strong pace, but global activity was more bal- anced owing to an acceleration of growth in Need to Make Progress in Other Areas European countries. The exceptional growth performances of China and India also contin- To sustain and accelerate growth and poverty ued. In the coming years, growth is expected reduction, developing countries will not only to slow down slightly in most advanced coun- need to maintain and, in many cases, improve tries, on the back of a gradual resolution of the their macroeconomic frameworks, but also large global current account imbalances that make efforts in other areas. This chapter have been accumulating in the last decade. monitors progress in two such areas, the pri- However, the risks of growth slowdown vate investment climate and governance. remain, although the likelihood of these materializing has diminished recently. If the pace of economic activity were to translate into higher inflationary pressure in devel- oped countries, this might trigger more dra- The World Bank monitors the investment cli- matic rises in interest rates than experienced mate through two main vehicles: the Invest- so far, with the attendant danger of a sharp ment Climate Surveys (ICS) and the Doing slowdown in these countries' growth. The Business (DB) surveys. The former draws unwinding of global imbalances, and in par- data from firms, while the latter relies on the ticular of the exceptionally large U.S. trade views of experts. deficit, could also take place at a much faster In 2006, new firm-level ICS data became pace than expected, if the U.S. economy were available for 27 countries, bringing the total to slow down significantly, following, for to 73,000 firms in 104 countries. The year example, an acceleration in the fall of hous- marked the beginning of a shift to regional ing prices. The future behavior of world oil rollouts of the surveys, with 8 Latin American prices is another area of uncertainty. While countries and 17 Sub-Saharan countries cov- further sharp increases are not anticipated, ered in the latest round. Several fragile states (Burundi, Democratic Republic of Congo, Moreover, petty bribery associated with per- Côte d'Ivoire, Guinea, Guinea-Bissau, and mits and with customs clearance declined. Mauritania) were included. Key areas of inter- While corruption remains an issue, there is est in Latin America include innovation and evidence of significant progress. technology. The Africa report focuses on issues In Bangladesh, a set of firms were surveyed of competitiveness, gender, and the extent to every six months for three years. The impact which the investment climate can compensate policy change can be seen in international trade for geographic challenges such as being land- reforms, licensing reforms, and tax reforms. locked or natural resource intensive. With the end of the Multi-Fiber Agreement, Several countries are now collecting fol- international competition in garments and tex- low-on surveys, making it possible to evalu- tiles intensified as reflected in falling garment ate changes in the investment climate and export and textile input prices. The demand policy reforms. For example, a survey was for skilled workers also rose, with skilled fielded in Egypt in 2004 prior to a series of workers receiving higher wage increases, and reforms of the tax system, licensing, per- firms reporting greater delays in hiring new mits, and customs. The 2006 survey picked skilled workers. In addition, the automation up clear indications of the impact of these of municipal licensing through a new interface reforms. Whereas taxes had ranked as the top provided by the Dhaka Chamber of Commerce constraint in 2004, it became only the fifth- and Industry resulted in dramatic declines in highest constraint in 2006. Significantly, there the time taken to renew business licensing per- were fewer inspections by tax authorities and mits through Bangladesh's Municipal Corpora- fewer petty bribes associated with taxes. In tions. There was also a decline in the incidence addition, streamlining of licensing permits and value of bribes (figure 1.3). led to a decline in the time to get an operat- The 2007 doing business indicators mea- ing license from 3.3 months to 2.0 months. sure the status of de jure business environ- Regulatory reforms can increase efficiency and reduce corruption Source: Hallward-Driemeier (2006): Bangladesh Panel Survey. ment in which private firms operate, with commonly defined indicators across 175 countries. The goal is to benchmark and The 2006 GMR highlighted governance mon- monitor efforts to improve the business cli- itoring as a core ongoing part of the broader mate and provide policy makers with a set task of monitoring progress in reaching the of indicators as to how they compare with MDGs. The 2006 analysis underscored the other country practices. Donors and foreign following: investors can make use of the indicators to measure progress in them. The transparency Governance is multidimensional, with no and simplicity of the indicators also facilitate unique path from weaker to stronger gov- efforts to identify areas of inefficiency and ernance. The quality of bureaucracy and shortcomings in country regulatory frame- of checks-and-balances institutions com- works. While there are many other factors prise two broad dimensions along which affecting investors' decisions, improvement governance might change, with the pattern of the regulatory environment can have spill- of change varying from country to country. over benefits on other areas of public policy, Governance monitoring is an imperfect sci- further improving the attractiveness of good ence. All measures have margins of error. reformers in the eyes of investors. It would be a mistake to read significance The 2007 Doing Business report recog- into small differences across countries or nized the accomplishments of countries that modest changes over time. were able to improve their regulatory envi- Monitoring at aggregate levels, using ronments. Georgia was 2006's most impres- broad measures, can provide an overview sive reformer, making reforms in 6 of 10 areas of trends in governance change and cross- studied by Doing Business and improving its country patterns. But efforts at reform world ranking on the ease of doing business invariably focus on specific governance from 112 to 37 in the span of one year, point- subsystems, and (unless they can be dis- ing to the capacity of countries to quickly and aggregated) broad measures are too impre- significantly progress. Mexico and Romania cise to be useful for monitoring whether also improved their rankings through major specific interventions create progress. reforms. The African region, which had been There is strong potential for monitoring the slowest-reforming region in the previous at a disaggregated level, "using specific two reports, picked up pace in 2006, and, with measures of the quality of key gover- the exception of Europe and Central Asia made nance subsystems, and using the results as more progress than other developing-country `actionable indicators' to identify specific regions. Tanzania and Ghana were Africa's strengths and weaknesses in individual top reformers, but others also made significant countries." The 2006 GMR advocated progress. Many of the reforms in Africa were strongly for greater investment in devel- easy, stroke-of-the-pen reforms--one simple oping such measures. reform in Côte d'Ivoire cut the time it takes to register property from 397 days in 2005 to Broad Governance Trends in Low- and 32--although more difficult reforms will soon Middle-Income Countries, 1996­2005 be necessary. Other countries, including two fragile states, Zimbabwe and Timor-Leste, The 2006 GMR identified schematically three were identified as having deteriorating busi- distinct trajectories of governance improve- ness environments. Eritrea, another fragile ment: disproportionate gains in bureaucratic state, was noted as having the single worst capability, disproportionate gains in checks- reform of the year, which suspended all con- and-balances institutions, and balanced struction licenses and prohibited private busi- gains. The 2006 GMR also suggested ways nesses from entering the construction sector. to measure both bureaucratic capability8 and Trajectories of Governance Improvements Balanced--Significant Improvements in both Government More Improvement Effectiveness and Quality of Checks and More Improvement in Quality of Checks and Balances Balances Institutions in Government Effectiveness Institutions Sources: Kaufmann, Kraay, and Mastruzzi 2006; Polity IV Database 2005. the quality of checks and balances.9 Figure Eight countries--including three in Africa 1.4 uses aggregate governance indicators to and three in Eastern Europe--improved depict empirically these three trajectories of governance in a balanced manner over the improvement for 1996­2005. Forty-four of course of the decade. the 111 countries experienced governance Fifteen countries--including 10 from changes that were both relatively large and Eastern Europe or the former Soviet at least moderately significant.10 Union--improved mostly in bureaucratic The measures are sufficiently loosely capability/government effectiveness. defined,11 and the margins of error of the Ten countries saw disproportionate estimates sufficiently large, that the results improvement in the quality of their checks- are best viewed as heuristic. Even so, the and-balances institutions. In four of these systematically divergent patterns across the (Indonesia, Peru, Sierra Leone, and Syria) three sub-figures12 seems to underscore that the gains in accountability were offset by there is no unique path from poor to good declines in bureaucratic capability/govern- governance: ment effectiveness. Four of the six countries that improved checks and balances over the Independent civil society organizations decade without a corresponding decline made a variety of noteworthy contributions were in Sub-Saharan Africa (The Gambia, to the monitoring of the quality of checks- Ghana, Kenya, and Senegal). and-balances institutions, three of which Eleven countries, of which six are currently are illustrated here. First, in May 2006 the classified as fragile states, experienced gov- Afrobarometer network released the results ernance declines in at least one dimension for 18 African countries of its third round of without improvement in the other. For five surveys. Afrobarometer provides scientifically of these (Central African Republic, Côte reliable data, comparable across countries d'Ivoire, Eritrea, Nepal, and Zimbabwe) and over time, on citizen perceptions vis-à- the declines were both relatively large and vis a variety of governance issues including moderately significant across both dimen- popular understanding of, support for, and sions; for three (Argentina, Guinea-Bissau, satisfaction with democracy; the demand for, and the Lao People's Democratic Repub- and satisfaction with effective, accountable, lic) the declines were mostly in bureau- and clean government; satisfaction with edu- cratic capability; and for the remaining cation, health, and local government services; three (Ecuador, Guyana, and República and citizen participation in both democratic Bolivariana de Venezuela--all in Latin processes and development efforts. Second, America) mostly in the quality of checks- in January 2007, Global Integrity (whose and-balances institutions. GII index was introduced in the 2006 GMR) released 43 additional country reports. The data suggest broadly divergent pat- These included follow-up reports for 17 of terns among African countries relative to the 25 countries surveyed in the initial, 2004 countries in Eastern Europe and the former round. Soviet Union in the trajectories of governance The third contribution by civil society was reform. The most common improvements the release in October 2006 (after four years were, in the former group, a further consoli- of work) by the International Budget Proj- dation of the political openings of the early ect of a new index to monitor the transpar- 1990s, and in the latter group, gains mostly ency of public budgets. As with the GII, the in government effectiveness. Country-specific Open Budget Index is based on a combina- starting points thus surely matter in shaping tion of expert assessments and peer review at the agenda for governance change, although both country and global levels. Key findings understanding of these dynamics is still in its include the following: infancy. Tracking the impact of specific gov- ernance reforms requires more disaggregated Only 6 of 59 countries surveyed--France, "actionable" indicators. New Zealand, Slovenia, South Africa, the United Kingdom, and the United States-- were reported as consistently providing Growing Momentum for Actionable "extensive" budget information to citi- Governance Indicators zens in their budget documents. An addi- Over the past year, initiatives by independent tional 30 countries provide "significant" civil society organizations, work within the or "some" budget information. World Bank Group, and multidonor initia- Twenty-three countries were reported as tives all have contributed to progress in the providing "minimal" or "scant or no" development of specific governance indica- information--with 10 countries (Angola, tors that, given repeated measurement over Bolivia, Burkina Faso, Chad, the Arab time, can be used to monitor operationally, Republic of Egypt, Mongolia, Morocco, in a disaggregated way, the effectiveness of Nicaragua, Nigeria, and Vietnam) in the efforts to strengthen governance subsystems. latter, weaker category. In 32 of the countries surveyed, the govern- ability (PEFA) public financial management ment does not make available to the pub- indicators, and the OECD/DAC work on pro- lic information it is already producing for curement. Use of the PEFA indicators (which its own internal use or for donors. Thus, were described in depth in the 2006 GMR) many countries could sharply improve has expanded rapidly. As of October 2006, their transparency and accountability sim- public financial management assessments had ply by providing information they already been completed (to the point of final draft produce to the public. reports) in 33 countries and were under way in an additional 15, and 34 more assessments The World Bank Group has made three had been planned (but not yet begun). In July sets of contributions. First was the publi- 2006, following a long gestation period, the cation, for the first time (but only for low- OECD/DAC Joint Venture for Procurement income countries), of disaggregated CPIA published a revised tool--Methodology for scores. The 2006 GMR detailed the potential Assessment of National Procurement Systems for using some of the CPIA results--especially (version 4)--and an accompanying guidance those on public financial management, on note for scoring each of its 54 indicators on the quality of public administration, and on a four-point scale. The tool has successfully property rights and the rule of law--as gover- been piloted in five countries (Albania, Ban- nance measures. As these data have long been gladesh, Ghana, the Philippines, and Turkey) used in the allocation of International Devel- and is currently being used in 15­20 more. opment Association (IDA) resources, their So far, however, these multidonor initiatives release is an important contribution not only have been characterized by a notable discon- to the endeavor of governance monitoring, nect between scaled-up in-country efforts but also to the transparency of the interna- (that have made an important contribution tional financial institutions (IFIs). The sec- to harmonized monitoring at country-level ond contribution was the release of the 2005 of trends in the quality of public expenditure updates for both the Doing Business (DB) management), and transparent availability of indicators and the Kaufmann-Kraay/World the fruits of that effort. Bank Institute (aggregate) governance indica- For PEFA (table 1.6), only 8 of 45 "sub- tors; for the first time, the detailed indicators stantially completed" assessments have so far used to construct the KK measures have also found their way into the public domain, and been made available on the KK/WBI Web site. even those have been made available only as The third contribution was in systematizing individual reports, with no effort to consoli- and scaling up further its efforts on enter- date and contrast the results. The remaining prise surveys. Prior to 2006, only in Europe reports, although their drafting apparently and Central Asia were surveys systematically has been finalized, currently are in a consulta- done across all countries within a region, tion (or postconsultation) limbo; the problem rather than on a demand-driven country-by- is seemingly especially acute for assessments country basis. Regional rollouts currently are led by the World Bank or the European Com- under way worldwide, including for 30 coun- mission. The OECD/DAC-sponsored work tries in Africa and 15 in Latin America. As of on procurement assessments is less advanced February 2007, enterprise survey results were than PEFAs. But here too, there are no plans available on a new streamlined Web interface to make available in a consolidated way the for 100 countries--up from only 37 a year results of the ongoing country-level work. earlier. Research is under way that links the Underlying this caution is a concern among DB and Enterprise Survey (ES) results. donors as to how the findings will be used. In Multidonor initiatives have resulted in sig- particular, there is a fear that cross-country nificant progress on two sets of indicators: the comparisons will be used to construct new Public Expenditure and Financial Account- "red-lines"--absolute thresholds as to which Status of "finalized" PEFA assessments (as of February 23, 2007) WB and EC Other agency Total Number of assessments WB leading EC leading jointly leading leading assessments Substantially completed draft/final report 20 16 1 8 45 Of which final report completed 4 11 0 4 19 Of which final report in the public domain 4 2 0 2 8 Source: PEFA Secretariat. countries should receive aid, and the form in taken into account, the share of natural which that aid should be provided. This fear is resources in total wealth is substantially misplaced. Certainly, it is not relevant for the higher than produced capital in the poor- World Bank Group--where the performance- est regions--Sub-Saharan Africa and South based allocation system (using the CPIA) is Asia--and in the oil-producing countries of the basis for allocation of IDA resources. the Middle East and North Africa. In lower- More broadly, as highlighted in the 2006 middle-income countries the shares of pro- GMR, there is a growing consensus that scal- duced and natural capital in total wealth are ing up aid, and moving to country systems, roughly equal. Only in upper-middle-income principally should be based not on absolute countries is there a consistently higher share thresholds but on country-specific trends in of produced capital compared to natural the quality of these systems, as evidenced by capital in total wealth. For a broad spectrum improvement in actionable indicators. of developing countries, the effectiveness of natural resource management can therefore have a significant impact on development prospects and performance. As noted in the introduction to this chap- ter, countries are liquidating assets when The Environment as a Source of Growth they extract minerals and energy, harvest for- and Poverty Reduction ests and fish unsustainably, or deplete their agricultural soils. This liquidation of natu- Higher economic growth is clearly desirable, ral assets is obscured in traditional national but rather than a goal in and of itself, it should accounts measures, such as gross national be a process of increasing the wealth of pres- income (GNI), which treat depletion and ent and future generations. Defining wealth as depreciation as part of income. Careful anal- including not only physical and human capi- ysis of the net rate of wealth creation presents tal, but also natural assets, leads to concerns a very different picture of economic perfor- that current rates of depletion and degradation mance. In Sub-Saharan Africa, for example, of natural resources may be undermining the the net creation of wealth has been effectively sustainability of higher growth, particularly zero over the last three decades, a period in in developing countries. Such concerns have which total population more than doubled. In motivated four recent major reports on envi- countries such as Cameroon, Benin, Burkina ronmental issues (box 1.2). Faso, Mozambique, and Rwanda, low saving A distinguishing characteristic of devel- effort, resource depletion, and high popula- oping countries is their high dependence on tion growth combined to yield net reductions natural resources. When agricultural land, in wealth per capita of more than 10 percent minerals, energy resources, and forests are of GNI in 2000. Four recent reports have highlighted the urgency of many environmental and natural resource problems globally, and helped to link environmental factors to development outcomes. The Millennium Ecosystem Assessment: Ecosystems and Human Well-Being (MA) One of the central messages of the MA, a multiyear nongovernmental process involving nearly 1,400 experts, is that the unprecedented exploitation of ecosystems is rapidly destroying those eco- systems' abilities to continue providing services that are essential to our well-being. For example, in recent years, human activity has enhanced the ability of the ecosystem to provide crops but decreased the ability of marine fisheries to provide fish, a consequence of overfishing. The MA concludes that such unsustainable activity will prevent future generations from enjoying the ben- efits of certain ecosystem services. It also highlights the importance of valuing ecosystem services appropriately compared to the more common practice of valuing them primarily for the services that can be exploited for more private and immediate gains, including from revenues from har- vested timber and food. The content and lessons of the MA are directly relevant to the pursuit of sustainable poverty relief. The Stern Review on the Economics of Climate Change This review provides a thorough analysis of how climate change may impact the world economy and what can be done to minimize its costs. The review estimates that irreversible damages to the world from unabated climate change could entail a cost equivalent to a permanent drop of 5­20 percent of global per capita consumption depending on the climate scenario, with that cost being dispropor- tionately borne by the poorest people. This drop in GDP could cause as many as 220 million people in Africa and South Asia alone to remain below the $2-a-day poverty line at the century's end, with equally severe impacts on human development indicators. Changing precipitation patterns, extreme temperatures, increasingly violent storms, and rising sea levels could also lead to massive migration and increased conflict, compounding the misery of already suffering populations. The review argues that significant and immediate action can greatly reduce the likelihood of the direst scenarios. The cost of action to stabilize the climate at moderate levels of warming would be a permanent 1 percent drop in global per capita consumption. The review also advocates that the international community needs to invest more in adapting to climate change because the global temperature has already risen by 0.7° C, and will increase more because of the presence of past and projected emissions. Adaptation will be particularly difficult for people in low-income coun- tries, and "should be an extension of good development practice," including promotion of growth and economic diversification, and investing in education, health, water management, and disaster preparedness. While there has been an active debate over the assumptions and conclusions of the review, the importance of the issue calls for greater attention, possibly in future GMRs. Where Is the Wealth of Nations? This World Bank study aims to increase understanding of the role of natural resources and the envi- ronment in the development process. It provides a comprehensive analysis of the different sources of wealth in developed and developing countries, and reveals some strong tendencies in wealth composition: (1) in low-income countries natural resources are a much larger share of total wealth than produced capital--29 percent compared with 16 percent; (2) agricultural land makes up two- thirds of the natural capital of low-income countries; and (3) the largest share of wealth across all income classes consists of less tangible items such as human and institutional capital. The study shows that the majority of low-income countries are actually dissaving in per capita terms when resource depletion and population growth are taken into account. Policy responses for donors and developing countries include placing greater emphasis on improving natural resource continued management, such as efforts to preserve soil quality in agriculture; reducing incentives to over- exploit natural resources, particularly living resources; and balancing investment in the overall portfolio of natural, produced, human, and institutional capital. At Loggerheads? Agricultural Expansion, Poverty Reduction, and Environment in the Tropical Forests This World Bank Policy Research Report discusses the dual goals of preserving rapidly shrinking tropical forests and relieving the poverty of the hundreds of millions of people who live in and near them. The report emphasizes that the causes of deforestation are varied. Some forests are cleared to expand commodity production in order to meet the demands of wealthy urbanites. Other forests are cleared by poor people who rely on expanding low-productivity agriculture. Timber prices have a more ambiguous effect as they often encourage sustainable management of timber rather than clear-cutting. The consequences of continued high rates of deforestation include the annual emission of 3 billion tons of CO2, the disappearance of entire ecosystems and the species that inhabit them, and widespread changes in water flows, scenery, microclimates, pests, and pollinators. To reduce defor- estation the report discusses the pros and cons of different land management strategies--protected areas, regulated logging concessions, community forest management--and their appropriateness in different contexts. The policy recommendations in the report include building local institutions and social capital in forested areas, particularly among indigenous groups and communities that will collectively man- age forests. The report stresses the need to mobilize international resources, especially conservation and carbon finance. Two cross-cutting recommendations include equitably assigning property and land use rights where they are weak or absent and, as recommended in the Millennium Ecosystem Assessment, actively encouraging markets for environmental services at local, national, and inter- national levels. Natural wealth is a potential contributor ticular, the management of natural resources to growth and poverty reduction, but poli- has significant impacts on welfare (box 1.3). cies, institutions, and political economy can While natural resources can potentially all influence the strength of the contribution. make large contributions to growth and pov- Exports of commercial natural resources erty reduction, they present specific risks as (minerals, energy, forest products, and fish) well. Commodity boom and bust cycles can are a source of development finance, but many stress fiscal systems and increase the volatil- countries do not use this finance effectively, ity of exchange rates. "Easy money" in the and are consuming resource rents rather than form of resource rents can reduce the impetus investing them. Nature tourism is a growing for economic reforms. The evidence suggests source of exports in many countries, but gov- that a combination of sound macroeconomic ernment policies often hamper the expansion policies and strong sectoral policies and insti- of the sector. The productivity of agricultural tutions is required in order to parlay natural land--55 to 65 percent of the value of natu- resource wealth into successful development. ral resources in developing countries--has a profound impact on growth and poverty, par- Update of Key Indicators ticularly in low-income countries. For poor households, the environment and natural MDG7 calls for integrating the principles of resources contribute directly to health, liveli- sustainable development into country poli- hoods, and vulnerability. For women in par- cies and reversing the loss of environmental In most regions women are more commonly burdened with handling domestic work. Children, especially girls, also spend much of their time helping with these tasks. A major component of domestic work is retrieving water and firewood for the home. The amount of time and effort needed to complete these tasks is highly dependent on environmental conditions. For homes with access to piped water and modern fuels, the time burden can be minimal. However, in places where water and fuel are more difficult to access these tasks can take hours every day, reducing the amount of time women and girls can spend on other activities, including out-of-home employment or school. Deforestation and pollution of water resources exacerbate the problem, requiring people to con- tinually travel longer distances to fetch firewood or potable water (Barwell 1996). Malawi is a country where access to water and firewood is particularly critical and also precari- ous. More than 90 percent of people use fuel wood as their main source of cooking energy. During the 1990s, Malawi's deforestation rate was 3 times that of the rest of Sub-Saharan Africa's and 10 times that of the world as a whole, making this vital resource more difficult to access. Malawi is also expected to experience a water crisis by 2025 that will make this resource scarcer. As noted above, these developments are likely to decrease female school attendance and performance com- pared to those of boys. This result suggests that the gender disparity in schooling may not be the result of conscious discrimination but instead of traditional gender inequalities in the division of labor (Nankhuni 2004). Additionally, some environmental health hazards fall disproportionately on women. Exposure to indoor air pollution, especially particulates, is a major factor causing lower-respiratory infec- tions, the leading cause of death from infectious diseases. Women are at greater risk than men because they are more commonly responsible for household tasks that expose them to indoor air pollution, such as cooking with biomass fuels. A Kenya study shows that young and adult women are exposed to, respectively, 2.5 and 4.8 times the particulate matter that men are exposed to in their age groups (Ezzati and others 2000). Correspondingly, the acute respiratory infection rate for women was twice that of men (Ezzati and Kammen 2001). Projects that reduce indoor air pollution, promote reforestation, and improve water quality are often thought of as environmental projects that help serve the health and economic interests of local populations. However, these projects, if well targeted, can have disproportionate benefits for women, because they can ease burdens that have traditionally reduced women's ability to partici- pate in more empowering activities. resources. Given the high resource depen- est loss is crucial because of the environmen- dence of many developing countries, there tal goods and services provided by forests, is a strong link between this goal and that including CO2 sequestration. CO2 emissions of reducing poverty. Table 1.7 provides an contribute to global climate change, a long- update on key indicators in the context of run threat to development. Household reli- trends of the past 15 years. The indicators cho- ance on traditional biomass energy affects sen--adjusted net saving, rates of deforesta- both pressures on forest resources and dam- tion, CO2 emissions, and reliance on biomass ages to human health from indoor exposure fuels--aim to represent both a general view to smoke. of sustainability and the progress in specific Each of these indicators entails specific areas relevant to development. "Adjusted net policy goals: (1) for adjusted net saving saving" measures countries' net saving effort the aim is to achieve positive saving rates after accounting for depletion and damage to that are consistent with growth targets; the environment, thus providing an indicator (2) goals for CO2 emissions are driven by of the sustainability of development.13 For- the individual targets for the industrialized Key indicators of environmental sustainability Carbon dioxide Use of Adjusted net saving emissions Annual deforestation traditional fuels Combustible renewables Forest cover lost and waste Percent of GNI Metric tons per capita 1990­2005 (% of total energy) of which, Annual Annual global change change damages (percentage Percent Annual Annual (percentage caused by points, increase area lost percent points, Group 2005 CO2 emissions 1990­2005) 2003 1990­2003 (sq km) lost 2004 1990­2004) 7.4 0.4 ­0.19 4.0 ­0.2 83484 0.14 10.3 ­0.03 6.2 1.0 0.24 0.9 1.6 71694 0.59 44.9 ­0.53 Fragile states ­25.1 0.8 ­0.57 0.5 ­2.0 31799 0.56 78.1 0.07 Non-fragile states 11.0 1.0 0.31 1.0 2.5 39891 0.62 39.1 ­0.76 9.5 0.9 ­0.12 3.9 ­0.4 18288 0.03 9.0 ­0.08 8.0 1.0 ­0.01 2.4 ­0.6 90621 0.21 17.5 ­0.07 East Asia & Pacific 25.3 1.2 0.45 2.7 1.3 4939 ­0.22 16.1 ­0.61 Europe & Central Asia ­2.0 1.2 ­0.89a 6.9 ­3.1 ­1789 ­0.02 2.4 0.04c Latin America & Caribbean 3.7 0.4 ­0.11 2.4 0.4 45753 0.44 14.8 ­0.25 Middle East & North Africa ­13.0 1.2 ­0.92b 3.4 2.4 ­747 ­0.49 1.2 ­0.04 South Asia 16.4 1.1 0.64 1.0 3.0 ­831 ­0.18 38.0 ­0.79 Sub-Saharan Africa ­7.3 0.7 ­0.20 0.8 ­0.8 43296 0.58 55.7 -­0.01 7.7 0.3 ­0.21 12.8 0.7 ­7137 ­0.09 3.1 0.01 8.2 0.3 ­0.19 12.8 0.6 ­7041 ­0.09 3.3 0.01 Source: World Development Indicators. Note: Carbon dioxide figures refer to emissions from combustion of fossil fuels and cement manufacture. a. Annual change refers to the period 1995­2005. b. Annual change refers to the period 1993­2005. c. Annual change refers to the period 1992­2004. country signatories to the Kyoto Protocol countries cannot escape a state of low-level of the UN Framework Convention on Cli- subsistence. In an effort to comprehensively mate Change, which aims to reduce global assess a country's rate of saving, "adjusted net emissions by 5.2 percent from 1990 levels saving" modifies traditional saving measures by 2012; (3) bringing deforestation down to to take into account depreciation of produced zero is the appropriate policy goal for many capital, the depletion of natural resources, countries, preserving the environmental pollution damages, and investment in human services provided by forests and protecting capital (box 1.4). Negative saving rates are a the sustainable flow of timber and nontim- clear indication that an economy is not on a ber products derived from natural forests; sustainable path. Figure 1.5 shows trends in and (4) reducing and ultimately eliminating gross and adjusted net saving over time. household use of traditional biomass fuels In East Asia and the Pacific and in South through provision of affordable substitutes. Asia, adjusted net saving has been steady at about 20 percent and 10 percent, respectively, owing to strong saving efforts. In Sub-Saha- Saving Rates across the World ran Africa, it has been hovering around zero. Saving is a core aspect of development. With- Latin America and the Caribbean and Europe out the creation of a surplus for investment, and Central Asia have had modestly positive The following figure presents the calculation of adjusted net saving in Bolivia in 2005. Source: World Development Indicators. Gross saving in Bolivia in 2005 was roughly 20 percent of GNI. This falls to 10 percent when depreciation of fixed capital is deducted, but the drop is partially offset by investment in human capital (as measured by education expenditure). Deducting the depletion of natural resources (mostly natural gas in Bolivia's case) and damages from emissions of PM10 (particulate matter less than 10 microns) and CO2 leads to the bottom-line value of ­20 percent of GNI as the adjusted net saving rate of Bolivia. In net terms the country is consuming wealth, with negative consequences for potential growth. As just noted, while adjusted net saving focuses primarily on the net accumulation of wealth within a country's borders, it also accounts for damages inflicted on all countries when a unit of CO2 is emitted. This overall approach to accounting is based on two assumptions about property rights: (1) that countries own the natural assets lying within their borders, and (2) that countries have the right not to be polluted by their neighbors. The latter assumption is what underpins the Kyoto Protocol. If countries have the right not to be polluted by their neighbors, then the economic accounts of pol- lution emitters should show a charge for the damage inflicted--these figures are broken out in table 1.7. With the conservative carbon price used in the saving calculation ($24 per metric ton of carbon), these damages vary from 0.3 percent of GNI in high-income countries to 1.0­1.2 percent in most developing regions. This largely reflects the efficiency of energy use in the different regions. saving rates over time. However, in Europe equivalent increase in gross saving. Resource and Central Asia there has been a downward rents are clearly being consumed in many of trend in saving owing to an increasing extrac- these countries. While not shown in figure tion of oil, which has not been offset by an 1.5, adjusted net saving rates in high-income Adjusted net saving rates by region Source: World Development Indicators. countries have fallen steadily from nearly 20 dissaving, a common source of low gross sav- percent in the early 1970s to less than 10 per- ing rates; (2) investing more in human capi- cent in 2005--this is largely a reflection of tal; (3) reducing incentives to overexploit falling gross saving rates. natural resources, particularly forests and For countries with growing populations, fish; and (4) reducing excess pollution emis- there is an additional factor not included in sions through market-oriented policies. table 1.7--the reduction in wealth per capita associated with each new population cohort. Energy: From Global to Local Issues For a population growth rate of 2 percent per year this "wealth dilution" effect would Carbon dioxide emissions from fossil fuel imply a deduction from wealth per capita on combustion and cement manufacture world- the order of 10­12 percent of GNI in a typi- wide topped 27 billion metric tons in 2003, cal developing country. The change in wealth an increase of 19 percent compared to 1990 per capita is negative in the majority of low- levels. In the absence of policy interventions, income countries, often by significant pro- this trend will likely continue as economic portions of GNI. activity grows. China, which is already the Low or negative adjusted net saving places second-largest emitter, has increased its emis- growth at risk. The policy responses to insuf- sions per capita by 52 percent between 1990 ficient saving include (1) reducing government and 2003, while India's emissions per capita have grown 50 percent in the same period-- of ill-defined property rights and corruption. note, however, that the 2003 level of emis- Forests can be used wastefully if they are sions per person in each country is still a cleared for low-productivity ranches that are fraction of high-income-country levels. ultimately abandoned The net change in for- The major part of CO2 emissions from fos- est area during 1990­2005 is estimated to be sil fuel combustion and cement manufacture a loss of 8.3 million hectares a year (an area stems from rich countries, however, with the about the size of Panama or Sierra Leone). United States contributing 22 percent of total Deforestation is highest in Sub-Saharan emissions, the European Union 9 percent, Africa (0.6 percent per year between 1990 and Japan 5 percent in 2003. But the share of and 2005) and in Latin America and the developing-country contributions is rapidly Caribbean (0.4 percent per year). While frag- increasing. From 2000 to 2003, global CO2 ile states contained 8.2 percent of the world's emissions increased by 2.9 percent annually, forest area in 2005, they also accounted for and about 83 percent of this increase came 28.6 percent of world deforestation. from low- and middle-income countries. If CO2 emissions from deforestation and CO2- Update on Country Programs and equivalent emissions from agriculture are Policies included, the annual contribution of devel- oping countries to greenhouse gas concentra- Sustainable development requires that actions tions exceeds that of high-income countries. by the current generation not damage the The lowest level of CO2 emissions per development prospects of future generations. capita is in Sub-Saharan Africa. This is mainly This can be achieved by ensuring that wealth, driven by the lack of access to modern sources broadly conceived to include human-made of energy, which leads people in poor coun- and natural assets, does not decline from tries to depend on traditional biomass fuels for one generation to the next. Sustainability their energy needs. Solid biomass is associated presents a significant challenge, especially in with respiratory problems caused by indoor the presence of public goods and externali- smoke. Most of the victims are infants, chil- ties, because markets by themselves are not dren, and women from poor rural families. able to ensure efficient outcomes. In addi- Globally, 2 billion people rely on biomass tion to market failures, policy failure is also fuels for energy. The regions with the high- a distinct possibility. The "resource curse" est level of biomass fuel use are Sub-Saharan literature (see, for example, Auty and Gelb Africa and South Asia. The data show very 2001) argues that natural resource wealth little progress between 1990 and 2004 for may dampen economic growth owing to the Africa and more generally for low-income political economy of rent-seeking that occurs countries, where the use of biomass products in many resource-rich countries, while the and waste as a percentage of total energy use volatility of natural resource prices presents has gone from 55 percent in 1990 to 48 per- risks to macroeconomic stability. cent in 2004. Whether the problem is a market failure or a policy failure, sustainable development requires strong institutions that are able to Deforestation pursue a coherent economic policy and the Forests provide important ecosystem services objective of raising social welfare. Population- (CO2 sequestration and regulation of water weighted environment CPIA scores for regions flows, for example) and host most of the and income groups for 2005 (figure 1.6) show world's biodiversity. The causes of deforesta- that Sub-Saharan Africa scores lowest (3.2), tion and ecosystem loss include conversion while East Asia and Latin America have the to agriculture and unsustainable commercial highest regional average (3.8). The regional timber extraction, particularly in the presence average scores mask good performance in Environment and overall CPIA score by region and income group Source: CPIA database. Note: Scores are population weighted. many countries--Mauritius and South Africa opment, emphasizing the key roles played by are strong performers in Sub-Saharan Africa, agricultural land, forest and fisheries in gener- for example, while the Republic of Korea, ating income, and natural areas as a resource Malaysia, Thailand, Costa Rica, and Mexico for nature tourism. top the lists in East Asia and Latin America. As might be expected, there is a wide dif- ference between the environment scores of low-income countries (3.3) and upper-middle- 1. In this report, low-income countries are income countries (4.0). Figure 1.6 shows that those eligible for IDA assistance. Other develop- environment CPIA scores are generally lower ing countries are classified as middle income. than overall CPIA scores, indicating that the 2. As box 1.1 indicates, fragility is defined quality of environmental institutions in devel- according to cutoff values of the World Bank's oping countries is lagging in relative terms. country policy and institutional assessment (CPIA). In 2005 the list of countries and territories for which the CPIA rating (see World Development Indicators Looking Ahead 2007) was at 3.0 and below includes Afghanistan, This brief update on MDG7 has necessarily Angola, Burundi, Central African Republic, Chad, neglected many issues that could be taken up Comoros, Democratic Republic of Congo, Republic of Congo, Côte d'Ivoire, Eritrea, Guinea, Guinea- in future GMRs. Potential issues for consider- Bissau, Haiti, Kosovo, Lao PDR, Liberia, Myanmar, ation include (1) climate change and develop- Solomon Islands, Somalia, Sudan, Timor-Leste, ment; (2) poverty-environment links, including Togo, Tonga, Uzbekistan, West Bank and Gaza, evidence on the environmental contribution to and Zimbabwe. Marginal fragile states, for which the health and livelihoods of poor households; the CPIA rating is at 3.1 or 3.2 include Cambodia, and (3) natural resources as assets for devel- Djibouti, The Gambia, Mauritania, Nigeria, Papua New Guinea, São Tomé and Principe, Sierra Leone, variable for 2004 is used in the construction of the and Vanuatu. As CPIA ratings change, countries 2005 Quality of Checks and Balance Institutions move in and out of the list. dimension. 3. The low-income countries are Albania, Arme- 10. A two-step filtering process was used to nia, Bolivia, Burkina Faso, Côte d'Ivoire, Georgia, identify significant governance improvements Honduras, India, Indonesia, Kyrgyz Republic, Lao between 1996 and 2005. Under this process a PDR, Moldova, Mongolia, Mozambique, Nigeria, country should experience (1) improvement in at Tajikistan, Uzbekistan, Vietnam, Zambia. The mid- least one of its Government Effectiveness, Voice dle-income countries are Argentina (urban), Brazil, and Accountability, and Rule of Law indicators at Chile, China, Colombia, Costa Rica, Dominican the 75 percent confidence level; and (2) an increase Republic, El Salvador, Estonia, Guatemala, Jamaica, in its score on either the Government Effective- Jordan, Kazakhstan, Latvia, Lithuania, Macedonia, ness or Quality of Checks and Balance Institutions Mexico, Paraguay, Peru, Philippines, Poland, Roma- dimension by at least 0.15 points. nia, Russia, Thailand, Uruguay (urban), República 11. Each of the KK measures is a composite Bolivariana deVenezuela. that combines distinct but related concepts. Thus 4. Chauvet and Collier (2004) estimate that KK Government Effectiveness measures "the qual- when a fragile state is a neighbor, the result is a ity of public services, the quality of the civil service loss of 1.6 percent of GDP per year. and the degree of its independence from political 5. Those who performed studies of the nega- pressures, the quality of policy formulation and tive impact of conflict on GDP include Knight and implementation, and the credibility of the govern- others (1996); Collier (1999); and Caplan (2001). ment's commitment to such policies" (Kaufmann, Collier (1999) also found that a negative impact Kraay, and Mastruzzi 2006). persisted long after conflict. 12. A country is placed in Trajectory II if there 6. Chauvet and Collier (2004) estimate that is significant improvement along both dimensions state fragility as measured by LICUS status typi- according to the two filters, but the improvement cally reduces the annual growth rate of peace- in the Government Effectiveness dimension is two time economies by 2.3 percent relative to other times or greater than the improvement in Checks developing economies. and Balances. Similarly, if the improvement in the 7. Not only are the questions underlying the Checks and Balances dimension is two times or macroeconomic assessments different in the CPIA, greater than the improvement in the Government but the 2005 survey also omits four fragile states Effectiveness dimension, the country is placed in in the IMF staff assessments. Trajectory III. 8. The 2006 GMR suggested that bureaucratic 13. "Adjusted net saving" modifies traditional capability was best measured using the World gross savings measures to account for deprecia- Bank's CPIA measures of budget and financial tion of produced capital, the depletion of natural management, and administrative quality. However, resources, pollution damages, and investment in these measures are only available publicly for one human capital. The lack of comparable interna- year and for IDA recipients. The Kaufmann-Kraay tional data on many natural resources such as fish- (KK) government effectiveness indicator is closely ery depletion, diamond resources, and extraction correlated with these measures (the correlation of subsoil water means that the adjusted savings coefficients for 2005 data with the CPIA Budget figures published here and in the World Develop- and Financial Management, and Administrative ment Indicators will be incomplete for some coun- Quality measures are 0.71 and 0.81, respectively) tries. In addition, a portion of health expenditures and is used as an alternative. should be viewed as investment in human capital 9. The suggested checks-and-balances measure and captured in the adjusted savings measure, but is a composite of three indicators: KK Voice and data are again a problem. The divergence between Accountability, KK Rule of Law, and Polity IV local and international prices may distort both Executive Constraints. The Executive Constraints gross and adjusted net savings figures, because some investments (in education, or nontradables such as buildings) are valued at local prices, while natural resources and machinery and equipment are valued at world prices. 1A.1 Share of people living on less than $1.08 a day (%) Region 1981 1990 1993 1996 1999 2002 2004 Forecast 2015 EAP 57.7 29.8 25.2 16.1 15.5 12.3 9.1 2.4 China 63.8 33.0 28.4 17.4 17.8 13.8 9.9 2.6 ECA 0.7 0.5 3.6 4.4 3.8 1.3 0.9 0.5 LAC 10.8 10.2 8.4 8.9 9.7 9.1 8.6 6.0 MNA 5.1 2.3 1.9 1.7 2.1 1.7 1.5 0.8 SAR 51.6 43.0 37.1 36.6 35.8 34.7 31.9 18.0 India 54.3 44.3 42.1 40.6 38.8 37.5 35.8 22.1 SSA 42.3 46.7 45.5 47.7 45.8 42.6 41.1 35.4 Total 40.6 28.7 25.6 22.8 22.3 20.4 18.4 11.8 Fragile states 49.0 54.2 50.4 1A.2 Share of people living on less than $2.15 a day (%) Region 1981 1990 1993 1996 1999 2002 2004 2015 EAP 84.8 69.7 65.0 52.5 49.3 41.7 36.6 15.3 China 88.1 72.2 68.1 53.3 50.1 40.9 34.9 14.1 ECA 4.6 4.3 16.5 18.0 18.6 12.9 9.8 4.7 LAC 28.4 26.2 24.1 25.2 25.3 24.8 22.2 17.3 MNA 29.2 21.7 21.4 21.4 23.6 21.1 19.7 10.9 SAR 89.1 85.7 82.4 82.4 80.8 80.3 77.7 60.1 India 89.6 86.4 85.5 84.5 83.2 82.1 81.1 66.8 SSA 74.5 77.1 76.1 76.4 75.8 73.8 72.0 64.7 Total 67.1 60.8 59.4 55.5 54.4 50.8 47.7 34.2 Fragile states 73.4 75.8 72.7 1A.3 Number of people living on less than $1.08 a day (millions) Region 1981 1990 1993 1996 1999 2002 2004 2015 EAP 796 476 420 279 277 227 169 48 China 634 374 334 211 223 177 128 37 ECA 3 2 17 21 18 6 4 2 LAC 39 45 39 43 49 48 47 38 MNA 9 5 5 4 6 5 4 3 SAR 473 479 440 459 475 485 462 304 India 382 376 379 385 387 393 386 283 SSA 168 240 252 286 296 296 298 326 Total 1489 1247 1172 1093 1120 1067 986 721 Fragile states 172 261 306 1A.4 Number of people living on less than $2.15 a day (millions) Region 1981 1990 1993 1996 1999 2002 2004 2015 EAP 1170 1113 1083 908 883 766 684 312 China 876 819 803 650 628 524 452 196 ECA 20 20 78 85 88 61 46 23 LAC 104 115 111 122 128 131 121 109 MNA 51 49 52 55 65 61 59 40 SAR 818 954 976 1035 1073 1124 1124 1015 India 630 734 769 801 832 861 876 853 SSA 296 396 422 458 491 513 522 597 Total 2457 2647 2722 2664 2727 2655 2556 2095 Fragile states 257 365 441 Source: World Bank staff estimates. The Role of Quality in MDG Progress S ince 2000, over 34 million additional basic health care, and water and sanitation. children in the developing world have In reviewing overall progress toward the gained the chance to attend, and com- human development Millennium Develop- plete, primary school--one of the most mas- ment Goals (MDGs), this chapter examines sive expansions of schooling access in history. where the worst gaps persist and what poli- Over 550 million children have been vacci- cies can work to redress them. nated against measles--doubling the cover- The chapter also explores whether this age rates in some countries, and driving down strong push to expand coverage is eroding measles deaths in Sub-Saharan Africa by 75 service quality. Measuring quality in edu- percent. The number of developing-country cation and health poses very different chal- AIDS (acquired immunodeficiency syndrome) lenges, but in both cases limited data have patients with access to antiretroviral treat- inhibited comparisons across developing ment increased from 240,000 in 2001 to over countries. In education, while student learn- 1.6 million at mid-2006. Despite migration and ing outcomes offer a straightforward and resource constraints, health workers and clinic meaningful way to measure system quality, visits across the developing world are increas- few developing countries have tracked these ing significantly, as are the share of pregnant systematically. But new education research in women with access to health care when they 2006 has built directly comparable measures deliver, and the share of young children with of what children are learning in the develop- regular health and nutrition screening. There ing world, and the results have clear implica- is now little question that the "stretch" goals tions for MDG progress. In health, the latest adopted by the global community in 2000 to research exploits creative ways to measure the promote human development have helped quality of services that health workers actu- stimulate and support more rapid expansion ally deliver to patients, rather than what they of basic health and education services across are capable of delivering; the implications for the developing world. health policy are equally important. The progress in service delivery is not The first half of this chapter provides a even, of course--and it is not enough. Across brief overview of progress on each of the every region there are lagging countries, and human development MDGs, except the gen- within every country there are poor people, der goal (MDG3), which is the subject of rural areas, women, girls, and vulnerable chapter 3. The chapter also reviews global groups who lack fair access to schooling, trends in financing for human development sectors and the performance of major global weight prevalence rates are 38­51 percent programs. The second half of the chapter in the most populous countries--India, focuses on the role of quality in promoting Bangladesh, Afghanistan, and Pakistan. progress on the human development MDGs. Progress is being made in these countries, but none is on track to reach the MDG. Sub-Saharan Africa has an estimated 26 percent of all children suffering undernu- trition. Of concern are worsening trends Broad regional trends of MDG progress have in countries such as Cameroon, Burkina not changed since last year. All regions are Faso, and Zambia. off track on the child mortality goal, and at East Asia, Latin America, and Eastern least some of the others. The two regions lag- Europe all have some countries off track to ging most seriously behind--South Asia and reach the MDG. The highest levels of malnu- Sub-Saharan Africa--are off track on all of trition and micronutrient deficiencies are in the goals. This section highlights some of the Cambodia, Indonesia, Lao People's Demo- countries making exceptionally fast progress cratic Republic, the Philippines, Guatemala, toward different MDGs, countries where Haiti, Honduras, and Uzbekistan. Vietnam, outcomes are worsening, the performance on the other hand, has made impressive of major global programs, and new evidence progress in improving child nutrition. from research on what drives country prog- ress. The chapter gives special attention to the MDG2--Universal Primary Completion performance of fragile states--which lag far behind other developing countries on most Globally the primary completion rate has MDGs--and the special challenges they face increased from 78 percent in 2000 to an esti- in reaching the goals. mated 83 percent in 2005, and the pace of annual improvement has accelerated since 2000 in all regions except Latin America, MDG1--Nutrition Target where completion rates were already high. Access to adequate food is one of the most Progress has been especially strong in North basic conditions of survival and escape from Africa, South Asia, and Sub-Saharan Africa. poverty. Accordingly the first MDG links two Although some 38 percent of all develop- measures of poverty: income poverty, which is ing countries are considered off track, and discussed in chapter 1, and hunger. The target another 22 percent have inadequate data, the is to halve between 1990 and 2015 the pro- number of countries that have achieved uni- portion of people who suffer from hunger, as versal primary completion increased from 37 measured by the percentage of children under in 2000 to 52 in 2005. Nine of the 10 coun- five who are underweight. Undernutrition is tries making fastest progress globally are low- not only a threat to poverty reduction prog- income countries in Africa, but given the low ress, it is also the underlying cause of over 55 base they started from in 1990, most are still percent of all child deaths. In 2007 nearly a off track to reach the MDG (table 2.1). Of 38 third of all children in the developing world African countries for which there are data, 33 remain underweight or stunted, and an esti- are off track, with a few countries showing mated 30 percent of the overall population actual declines in primary completion. of the developing world suffers from micro- The two largest challenges for all coun- nutrient deficiencies. But the picture differs tries are ensuring that primary completion across regions: means completion with adequate learning-- discussed later in this chapter--and extend- South Asia has the highest rates and largest ing schooling access to the last 10 percent numbers of malnourished children. Under- of children. While survey data indicate that the largest gaps in schooling access in virtu- Several low-income countries are making strong progress ally every developing country are between on universal primary completion high- and low-income populations, an over- lay of issues can make particular groups of Primary poor students especially hard to reach. A completion rate Annual percentage increase recent study (Lewis and Lockheed 2006), Country 2000 2005 2000­2005 for example, showed that nearly 75 percent of the 55 million girls who remained out of school in the developing world in 2000 were "doubly disadvantaged"--female and from excluded ethnic, religious, or caste groups (figure 2.1). Achieving universal primary completion in these cases will require more than just building schools. It will also require actions to eliminate discriminatory policies, change teachers' attitudes, provide compen- satory preschool and in-school programs to help disadvantaged girls catch up, and tools Mauritania 51.6 44.5 ­ such as targeted stipends to overcome par- Namibia 85.4 75.3 ­ ents' reluctance to send girls to school and Malawi 67.2 60.7 ­ the direct and opportunity costs of doing so. The encouraging fact is that many policies and efforts to reach the "doubly disadvan- Source: UNESCO Institute of Statistics., 2007. taged" in fact will benefit all disadvantaged students, and bring countries closer to the goals of education for all. Most out-of-school girls are "doubly disadvantaged": Female and from minority groups Source: Lewis and Lockheed 2006. Donor aid for education fell in 2005, for FTI) (see box 2.1) appears to provide a useful the first time since 2000. Reasons for the framework to ensure that increased finance is decline appear to be short term in nature and used effectively to accelerate countries' prog- are discussed later in this chapter. With its ress toward MDG2 and harmonize donor announcement in April 2006 of a $15 bil- assistance. lion commitment over the next 10 years, the United Kingdom is now the leading bilateral MDG4--Child Mortality source of support for education. Other Euro- pean Union and Group of Eight (G-8) donor Over 10 million children under five in the countries have also pledged to increase sup- developing world die each year of diseases port for education within their rising overall that are preventable and curable with a levels of official development assistance. The handful of simple, low-cost interventions. Education for All-Fast Track Initiative (EFA- An estimated 63 percent of child deaths could The EFA Fast Track Initiative. The EFA Fast Track Initiative, launched in 2002, is the major global program promoting attainment of the education MDG, and Education for All Goals more broadly. The FTI is a "results-focused" partnership of all major donors for education and all low-income developing countries willing to commit to ongoing benchmarking of their education system perfor- mance, to ensure that spending produces results. The FTI is managed by a small secretariat staffed and supported by partner agencies and housed in the World Bank. Starting with 8 developing countries in 2002, the FTI had expanded to 28 countries by end-2006. At end-2008, it expects to involve 60 low-income countries, covering approximately 70 percent of the out-of-school children in the world. The FTI also has a special focus on supporting progress in fragile states. Donor financing for FTI countries is channeled both through expansion of existing donor programs and the multidonor FTI Catalytic Fund, administered by the Secretariat. Although the Catalytic Fund was originally conceived as a small and short-term source of bridge financing for countries entering the initiative without many active bilateral donors, the Fund has continued to attract donor pledges. At end-2006, the Fund totaled $1.1 billion in donor contributions and was disbursing approximately $150 million per year in grants to FTI countries. In FY 2008, Catalytic Fund disbursements are expected to be about 10­15 percent of total aid for basic education in low-income countries. In 2008, the Fund's initial limit on support of up to three years will also be relaxed. The FTI has sought to tighten the "compact" for results by monitoring the performance of both countries and donors. Among countries, the FTI's performance benchmarks promote sound policies and results focus; among donors, the FTI encourages more harmonized and efficient aid. A review of the program in 2006 concluded that the FTI is making progress on these goals (World Bank 2006). The FTI's performance benchmarks have acquired growing international con- sensus and, while there is less progress in some areas than others, FTI has helped some of the world's lowest enrollment countries substantially increase primary school coverage and girls' enrollments; reduce repetition; increase domestic spending on primary education; and boost spending on books, supplies, and maintenance. The FTI also seems to have had a large effect on donor harmonization and sector planning. Joint sector reviews, pooled funds, single reporting arrangements, and joint missions are becoming the norm in FTI countries, reducing transactions costs and increasing aid efficiency. By simultaneously promoting a scale-up of spending on primary education and policy reforms among both recipient countries and donors to improve the effectiveness of that spending, the Initiative appears to have created a useful framework for education MDG progress. be averted with oral rehydration therapy to nodeficiency Virus)/AIDS. Among the worst combat diarrhea, insecticide-treated bednets are Iraq (150 percent increase) and four coun- to prevent malaria, breastfeeding to improve tries in Southern Africa (Botswana, Zimba- nutritional status, and antibiotics to treat bwe, Swaziland, and Lesotho) in which child acute respiratory infections, if these were mortality increased because of HIV/AIDS. implemented universally (Jones and others The majority of countries have reduced 2003). But progress on the child mortality child mortality since 1990, but not at the MDG lags other goals. In 2005, only 32 out pace required to reach the MDG--an annual of 147 developing countries (22 percent) were decline of 4.3 percent over the entire period. making enough progress to achieve a two- Progress is possible, though, as shown by thirds reduction in child mortality between sharp declines even in some low-income 1990 and 2015. countries: between 1990 and 2005, under- Moreover, 23 low- and middle-income five mortality per 1,000 live births declined countries show worsening or stagnant child from 177 to 61 in Timor-Leste, from 53 to 19 survival trends. Many of these countries are in Vietnam, from 147 to 78 in Eritrea, and either in conflict, emerging from conflict, or from 166 to 75 in Bhutan (table 2.2). How- are heavily affected by HIV (Human Immu- ever, of the best performing Sub-Saharan Progress on child mortality in a few countries Under-five mortality rate per 1,000 births Annual percent change 1990 1995 2000 2005 1990­2005 Timor-Leste 177 154 107 61 ­ Vietnam 53 44 30 19 ­ Bhutan 166 133 100 75 ­ Mongolia 108 87 65 49 ­ Lao PDR 163 131 101 79 ­ Eritrea 147 122 97 78 ­ Malawi 221 193 155 125 ­ Cape Verde 60 50 42 35 ­ Comoros 120 101 84 71 ­ Mozambique 235 212 178 145 ­ Czech Republic 13 10 5 4 ­ Egypt, Arab Rep. of 104 71 49 33 ­ Peru 78 63 41 27 ­ Turkey 82 63 44 29 ­ Syrian Arab Rep. 39 28 20 15 ­ Source: World Bank What made it possible for Eritrea, a fragile, postconflict country with annual per capita income of just $190 and a primary completion rate for girls of only 33 percent to reduce under-five mortality by nearly 50 percent between 1990 and 2005? Based on assessments of child health services, the decline in mortality has been attributed in part to the implementation of the IMCI (integrated management of childhood illness) approach, including the training of over 500 health workers at different levels of the health care system in IMCI case management. Following the IMCI implementation, it was found that availability of drugs and equipment in health centers had improved, and that providers were doing a better job of following protocol for diagnosis and prevention of disease. Immunization coverage for fully vaccinated children increased from 41 percent in 1995 to 76 percent in 2002 in two Demographic and Health Survey (DHS) rounds. Encouraging and plausible as these factors are, programs such as Eritrea's should undergo well-designed evaluations before conclusions can be confidently drawn. The evidence used to explain changes in mortality often consists of after-the-fact rationalizations that link the implementation of specific activities with observed trends in mortality, an unscientific approach that does not take into account what might have happened without the intervention. Source: http://www.usaid.gov/stories/eritrea/cs_eritrea_mortality.html, retrieved in 2006. African countries, only Eritrea, described in Nations Children's Fund (UNICEF), the United box 2.2, is close to the MDG pace. Nations Population Fund (UNFPA), and the World Bank in 2007 will update the data with 2005 estimates. In the absence of direct esti- MDG5--Maternal Health mates, monitoring of progress toward the goal Each year an estimated 500,000 women in the has focused on one of the key determinants of developing world die in childbirth. While this maternal mortality: the presence of a medically number is far lower than annual child deaths, skilled attendant at the time of delivery. reducing maternal mortality is a global pre- The latest survey data show that in 27 out of occupation because deaths during delivery 32 developing countries (84 percent) the pro- are highly preventable. Maternal mortal- portion of deliveries with a skilled attendant has ity has been called a "tracer condition" for increased in recent years (table 2.3). Morocco, health systems (World Bank 1999), because Nicaragua, Indonesia, and Egypt registered if countries can ensure the three basic con- especially impressive progress. Survey data ditions of adequate access to antenatal care, also show that differences in access to skilled medical attendance at delivery, and a health delivery care between the poorest and richest referral system that ensures prompt treatment quintiles in most countries present larger equity of emergencies at adequately equipped clin- gaps than for any other health or education ser- ics, deaths during childbirth can be virtually vice. Even in some of the poorest low-income eliminated. Ninety-nine percent of maternal countries, such as Benin, Cameroon, Mali, deaths occur in the developing world. Mozambique, Zambia, or Zimbabwe, skilled The MDG target--to reduce the maternal attendance at delivery for the upper income mortality ratio by three-fourths between 1990 groups reaches levels near or above 90 percent, and 2015--remains difficult to measure; for several times the coverage for the poorest quin- almost all developing countries, no current tile. Finally, the surveys show that what con- direct estimates of the maternal mortality ratio stitutes "skilled attendance at delivery" varies or trends exist. A new joint effort by the World across countries. In Jordan, for example, where Health Organization (WHO), the United over 95 percent of women deliver with skilled Progress in assisted births Births attended by a medically trained person (percent of all births) Country Percent Year Percent Year Absolute change between surveys Haiti 46.3 1995 24.2 2000 ­ Peru 56.4 1996 46.9 2000 ­ Zambia 46.5 1996 43.4 2001 ­ Kenya 44.4 1998 41.6 2003 ­ Source: Demographic and Health Surveys. care, 63 percent of deliveries take place with four-pronged strategy to reduce its high mater- the assistance of a doctor, and 37 percent with nal mortality rate: (1) training traditional birth a midwife or nurse, but virtually all occur in attendants in how to recognize high-risk preg- hospitals. But in Benin, while a relatively high nancies and deal with obstetric emergencies; (2) 73 percent of deliveries are assisted, only 5 per- increasing health personnel and birthing centers cent of these are with a doctor, and 23 percent in remote areas; (3) strengthening emergency of births occur at home. obstetric care in rural health centers and district Maternal deaths most commonly result hospitals; and (4) improving emergency trans- from three critical sources of delay in accessing portation and communication systems. At the appropriate emergency care during pregnancy same time, the country introduced improved and delivery: (1) inadequate recognition of the surveillance to establish the cause of maternal need for care; (2) difficulty in getting to facili- deaths in all recorded cases. The successful ties (due to lack of transportation, poor roads implementation of this strategy is credited with or high costs); and/or (3) lack of adequate treat- reducing maternal mortality across Honduras ment once a facility has been reached (Thad- by more than 50 percent over the past decade, deus and Maine 1994). Box 2.3 summarizes including in the most remote and poorest areas new research from Ghana, Kenya, and India (Danel 1999; Ransom and Yinger 2002). which sheds light on the relative contributions of these "three delays" to poor maternal out- MDG6--AIDS, Tuberculosis, and Malaria comes, and how their underlying causes can differ in different country contexts. HIV/AIDS. No single MDG has galvanized One country whose progress in reduc- as much global attention and financial sup- ing maternal mortality has been researched is port since 2000 as the goal of "halting and Honduras. In the 1990s Honduras adopted a reversing the spread of HIV/AIDS." What To examine the relative contribution of the "three delays" to poor maternal care, the World Bank collaborated with research institutions in Ghana, India, and Kenya to carry out in-depth studies in very different settings: in northern Ghana (Kassena-Nankana District, a predominantly rural area, with population scattered and mainly dependent on subsistence farming); in Kenya, the Nairobi slums of Korogocho and Viwandani; and in Uttar Pradesh, India's most populous state, with a pre- dominantly rural population and low status of women. Household survey data, verbal autopsies, facility surveys, and in-depth interviews were used to analyze the three delays. In Ghana, although almost all women saw a midwife or nurse for antenatal care, the majority (59 percent) eventually delivered at home with a traditional birth attendant. Of the women who developed complications, about half recognized the problem as serious within a day, but 36 percent took three days or more. Sixty percent of these women reported the distance and travel time to a health facility as the major constraint to seeking service there. Three percent of women delivered while en route to a health facility, indicating the difficulty of getting to facilities in time. But women who did deliver in health facilities gave strongly positive opinions of the quality of obstetric care they received, notwithstanding reports from health workers at these facilities that they lacked adequate staff, supplies, and equipment. In the Nairobi slums, 70 percent of women delivered with the assistance of a health professional-- a substantially higher rate than in other parts of Kenya. Of the 62 percent who experienced obstet- ric complications, four-fifths made the decision to seek care within 24 hours, and distance and transportation were not major obstacles. In focus groups, women reported that the overriding factor against seeking care is cost. About 62 percent of women paid more than 1,000 KSh ($14) for delivery care in health facilities, and the requirement for a cash deposit prior to admission was an important obstacle. In Nairobi, women with obstetric complications frequently did not receive care promptly--owing to long queues, unavailability of health professionals, the demand for a deposit, or lack of equipment--and at times were sent to another health facility as a result. In contrast to Ghana, a large majority of Kenyan women reported poor treatment by health personnel, especially nurses and midwives at facilities. In Uttar Pradesh, India, antenatal care is nominally free of charge at government health facilities, but just 40 percent of women utilize it. Only 21 percent of women deliver at a health facility, and a higher share of these are at private facilities than government ones. The survey showed that the majority of women thought antenatal care or delivery with the assistance of skilled medical attendants was unnecessary; birth was viewed as something that should normally take place unassisted at home. Unlike rural Ghana, access factors such as long distance to health facilities and difficulty in obtain- ing transport were less frequently cited reasons for not delivering at the health facilities. However, 40 percent of the facilities indicated that they were not equipped to deliver basic emergency obstetric care, given limited staffing and equipment. While the study pointed to the need to improve the quality of government health facilities, it also suggests the need for outreach to women about the importance of professional assistance at births, since obstetric complications are unpredictable. Source: Mills and others 2007. has been achieved over the past six years-- known virus and radiates along myriad soci- especially the expansion of treatment across etal transmission fault lines, but also because the developing world--stands as a tribute to of the inherent challenges of rapidly scaling- the power of concerted global action. But up global action and funding in any area. what remains to be done is also substan- From just a few localized spots of infection tial--partly because of the unique nature of just 25 years ago, AIDS has spread to nearly AIDS, a disease that mutates faster than any every country in the world, and by end-2006 an estimated 39.5 million people were living prevalence at the national level have been with HIV, an increase of 2.6 million com- recorded in eight countries at the national pared with 2004 (figure 2.2). Behind these level (Kenya, Uganda, Thailand, Zimba- numbers is an increasingly heterogeneous bwe, Barbados, Bahamas, Cambodia, and epidemic ranging from Sub-Saharan Africa, Thailand) and in urban areas in six other where a mature, largely heterosexually trans- countries (Burkina Faso, Côte d'Ivoire, mitted epidemic has slowed its spread, to Ethiopia, Ghana, Malawi, and Rwanda). In rapidly growing epidemics in Eastern Europe India, declines have been recorded in four and Central Asia, linked to high-risk behav- southern states. In Latin America and the iors in segments of the population. Caribbean new infections in 2006 remained Globally, financial resources available at about the same level as in 2004. for implementing the AIDS response have The underlying drivers of the declines increased from $1.6 billion (about 20 percent are not fully understood, but it is likely that of total development assistance for health) in they reflect a different mix of factors in each 2000 to nearly $8 billion (60 percent of total context. It also seems likely that this prog- health support) in 2006. In numerous low- ress is easily reversible, so there is no cause income countries, external support earmarked for complacency. Recent surveillance data for AIDS is half or more of the entire public from Uganda--one of the first of the above health budget. However, external funding is countries to show signs of declining preva- concentrated in relatively few developing coun- lence--suggest that condom use has started tries, and large unmet needs remain. What has to decline, numbers of sexual partners have been achieved, and what have we learned to increased, and seroprevalence may again be guide future action? From the vantage point of increasing in some sites. 2007, five cautious conclusions can be drawn: In most of the countries, the natural course of the epidemic may be playing the Reversing the spread of HIV/AIDS is most important role in reducing the num- possible. The first signs of declining HIV ber of new infections; once epidemics reach Global HIV/AIDS epidemic, 1990­2006 Source: UNAIDS, AIDS epidemic update, 2006. a critical scale, "burnout" occurs since A recent analysis found that, on average, there are fewer uninfected individuals left within six months of starting treatment, to infect. But in the countries where preva- patients were able to increase their work- lence declined, there is also evidence of ing hours substantially (Thirumurthy, behavioral changes, including reductions Graff-Zivin and Goldstein 2005). Owing in the number of partners, increased use to increased work, even for low-income of condoms, and delays in the age of first day laborers, the study estimated that the sexual intercourse (UNAIDS 2006). incremental costs of treatment were fully Treatment is effective in the developing covered by the higher income generated, world. When the MDGs were adopted, and there were other social benefits accru- there was glaring global inequity in the ing to the household. chances an AIDS patient had of accessing Prevention efforts are inadequate. Preven- treatment depending on "which world" he tion will make the difference in the global or she lived in: in 2001, only 240,000 peo- trajectory of the AIDS epidemic. While 3 ple were on antiretroviral treatment (ART) million people died of AIDS in 2006, there of an estimated 5 million people in the were 4 million new cases of infection--over developing world with advanced HIV dis- 70,000 per week. Until progress is made ease who needed treatment. By mid-2006 in reducing this number substantially, the there were 1.6 million people on treat- battle against AIDS will continue uphill. ment.1 Botswana, Kenya, South Africa, Coverage of prevention efforts is too low Uganda, Rwanda, and Zambia have made almost everywhere, especially among strongest progress, and were among the 14 populations most at risk. A recent survey countries providing treatment to at least of more than 10,000 people living with 50 percent of those in clinical need by June HIV in 69 countries found that less than 1 2005 (UNAIDS and WHO 2005). percent of adults aged 15­29 had utilized It is now confirmed that the same gains voluntary counseling and testing and only in life expectancy achieved in high-income about 11 percent of pregnant women had countries are attainable in resource-scarce access to services for preventing mother- environments. Adherence to antiretroviral to-child transmission (Stover and Fahn- regimens in developing-country settings is enstock 2006). Prevention efforts aimed also as high as in high-income countries at high-risk populations only reached an (Mills and others 2006). Large cohort stud- estimated 33 percent of commercial sex ies have examined survival in over 27,000 workers, 34 percent of prisoners, 9 per- patients from developing and industrial- cent of homosexuals, and 8 percent of ized countries and found similar levels of injecting drug users, far short of the cov- viral suppression and declines in mortality erage of over 60 percent that is needed for (Stringer and others 2006). UNAIDS esti- effective impact. Finally, although some 5 mates that expanded provision of antiretro- billion condoms were distributed in 2005, viral treatment resulted in a gain of 2 million survey data suggest that this covered only life-years in low- and middle-income coun- an estimated 20 percent of risky sex acts. tries in 2005 (UNAIDS 2006). More evaluation of "what works" in dif- Globally, the WHO goal of putting 3 ferent contexts to prevent HIV is needed. million people with AIDS on ART by end- While many approaches to prevention 2005 was not reached, but numbers are have been tried, few approaches have been increasing rapidly. More important, access rigorously evaluated, and some evalua- to treatment has not only improved health tions have revealed unimpressive or mixed and saved the lives of numerous adults but results for programs previously believed to also helped restore the well-being of other be effective. While a WHO review found household members, especially children. that 13 of 23 evaluations of school-based HIV education showed some beneficial (Stover and others 2006). The experi- impact, some of the most rigorous evalu- ence of Brazil and Thailand is instructive; ations--randomized controlled trials in early and determined government action Western Kenya (Duflo and others 2006) focused both on preventing HIV in high and Mexico (Walker and others 2006)-- risk groups and making ART affordable found no evidence that HIV education has kept both countries on a trajectory of courses in secondary schools affected key very low prevalence. outcomes such as condom use. However, To meet the needs of all target popu- these and other studies have helped to lations and truly reverse the epidemic, identify other interventions that could be UNAIDS estimates that $22 billion per cost-effectively scaled up. Dupas (2005) year in external funding is required, found that informing young Kenyan girls almost a tripling of the current level. Yet about the higher risk of infection they face even the current levels of external financ- when engaging in sexual relationships ing for AIDS have had major effects on the with older men had a positive impact; one allocation of health resources in develop- year after the intervention, girls were 65 ing countries and placed strain on scarce percent less likely to have gotten pregnant factors of supply and costs. by adult partners. A randomized experi- In this environment, several areas ment conducted in rural Malawi found that stand as urgent priorities for action. The knowing one's serological status led to only first is better harmonization and align- small behavioral changes--suggesting that ment of donor efforts. The "three ones" universal HIV testing might not be the most platform--ensuring that in every country cost-effective way of preventing infections there is only one national AIDS leader- (Thornton 2005). However, offering an ship body, one national plan, and one sys- incentive equal to about one-tenth of a tem for monitoring progress--has made rural laborer's daily wage increased the some headway. Key agencies are working demand for testing, overcoming stigma on an explicit division of labor and better and offsetting opportunity costs of time. coordination in providing technical sup- There may also be scope for incentives port; there has been an increase in joint for other positive health behaviors, such donor reviews and supervision visits and as avoiding risky behaviors or adhering to agreement on a harmonized set of HIV ART. indicators. But there is still much more to The world has much more to do in fighting be done. The second priority is strength- HIV. Treatment access has expanded, but it ening health systems in developing coun- still reaches only one-quarter of all people tries to enable them to absorb additional with AIDS in the developing world. Pro- funding and deliver expanded services phylactic care for opportunistic infections efficiently. Third is the imperative of scal- also only reaches about one patient in four. ing up effective, evidence-based preven- Effective HIV prevention strategies need to tion strategies. be scaled up aggressively in all parts of the world--and these investments would yield Malaria. Malaria is both preventable and high returns. A 2006 study calculated that curable, but each year an estimated 300 to large-scale and effective prevention strate- 500 million cases of malaria result in an esti- gies implemented today in 125 low- and mated 1.2 million deaths. The majority of middle-income countries could avert more malaria deaths are among children, and an than half of the 28 million new infections estimated 80 percent occur in Sub-Saharan projected to occur between 2005 and Africa. Malaria also causes severe anemia 2015--and by 2015 could save $24 bil- and maternal illness, and contributes to low lion per year in associated treatment costs birth weight, a leading risk factor for child morbidity and mortality. In most countries Southeast Asia an artemisinin-based com- in Africa south of the Sahara, malaria cases bination therapy (ACT) has been successful are diagnosed and reported based on clinical in treating and reversing the spread of drug- grounds, such as fever, without laboratory resistant malaria, and WHO recommends testing. Since fever is common to many infec- use of ACT when new drugs are required. But tious diseases, misdiagnoses are common ACTs are 10­20 times as expensive as first- and the actual number of cases is unknown. line treatments, and there is a risk that malar- As a result, malaria incidence and mortality ia's toll could rise even higher if resistance to data by country generally do not accurately artemisinin were to spread. The challenge reflect the true scope of the disease, and are is thus to facilitate the use of artemisinins not reliable for monitoring trends. Instead, where appropriate while preserving their use of insecticide-treated bednets is tracked effectiveness for as long as possible. Arrow to monitor whether countries are addressing and colleagues (2004) and a separate study the disease through an effective preventive by the World Bank and the Roll Back Malaria strategy. Table 2.4 shows that coverage rates (RBM) Partnership concluded that actions to are still low in many countries. delay the development of resistance to ACTs Drug resistance and a global subsidy for create a benefit for all--"a global public antimalarial drugs. One reason for the resur- good." This would justify a sustained global gence and increased burden of malaria is the subsidy for ACTs, to ensure that artemisinins development of resistance to traditional first- are used with other antimalarials, and used line antimalarial treatments, such as chloro- judiciously. quine (CQ) and sulfadoxine pyrimethamine With support from the Netherlands and the (SP or Fansidar) by the parasite that causes RBM Partnership, the World Bank is leading a severe form of malaria. In some areas of an effort to translate this proposal for a high- level global subsidy into reality. The Bank is facilitating the analysis, consultations, and the design of possible management arrange- ments for the subsidy, which could be hosted Use of insecticide-treated bednets by children under five in an existing agency as appropriate. The Malaria Booster Program. In 2005 Poorest Richest Population the World Bank renewed its commitment to quintile quintile average the Roll Back Malaria Partnership, and the Colombia 2005 85.2 73.4 79.2 Africa Region launched the Malaria Booster Cambodia 2000 57.3 80.7 67.5 Program to support country-level efforts Chad 2004 67 71.2 55.8 to deliver concrete and measurable results. Mali 2001 34.4 48 38.3 Over the next five years, the Bank expects to Tanzania 2004 17.8 71.3 33.9 commit up to $500 million of International Benin 2001 24.7 57 33.8 Burkina Faso 2003 25 30.1 23.3 Development Association (IDA) resources Zambia 2001 6.4 29.8 17.7 to support the program in approximately Kenya 2003 8.1 36.3 16.7 20 countries. As of mid-December 2006, Ghana 2003 19.6 10.8 16.2 11 projects totaling $357 million had been Cameroon 2004 5 21.5 13.7 approved--funding malaria-control projects Mozambique 2003 4.7 20.9 10.1 Uganda 2000 6.3 23.3 9.4 in 10 countries and a subregional multisec- Nigeria 2003 11 3.3 7.1 tor project in the Senegal River Basin. The Namibia 2000 11.3 5.5 6.6 11 projects are expected to deliver at least 19 Rwanda 2000 1.8 24.7 5.8 million long-lasting insecticide-treated bed- Zimbabwe 1999 0.2 7.3 2.9 nets, primarily to young children and preg- nant women, and about 29 million doses of Source: Demographic and Health Surveys. ACT, primarily to children. Tuberculosis. There were 9 million new TB TB incidence trends by region cases and approximately 2 million TB deaths in 2004, making this the second deadliest Incidence of tuberculosis communicable disease, after AIDS. As table (per 100,000 people) 2.5 shows, while TB incidence was stable or Region or classification 1990 2004 falling in five out of six World Bank regions, it is growing at 0.6 percent per year globally, East Asia & Pacific 161 138 Europe & Central Asia 51 83 owing to rising incidence in Sub-Saharan Latin America & Caribbean 103 64 Africa. People latently infected with TB are at Middle East & North Africa 66 54 a much greater risk of developing active TB if South Asia 180 177 they are concurrently infected with HIV, and Sub-Saharan Africa 162 363 this has contributed to dramatically worsen- High income 28 17 Low income 177 224 ing TB incidence in southern African countries Lower middle income 134 115 and Kenya (table 2.6). In the Europe and Cen- Upper middle income 69 114 tral Asia Region, incidence per capita increased World 124 140 during the 1990s, but peaked about 2001, and has since fallen. The main strategy to combat Source: WHO. TB is careful treatment with a protocol called Directly Observed Treatment, Short-course (DOTS). In 2004, DOTS was being used in Changes in TB incidence, 1990­2004 183 countries, with 100 percent population coverage in 9 of 22 high-burden countries, Incidence of tuberculosis and almost complete in 5 others. By the end (per 100,000 people) Absolute difference of 2004, 83 percent of the world's population Country 1990 2004 1990 and 2004 lived in countries, or parts of countries, cov- ered by DOTS. High-burden countries with Swaziland 263 1,226 963 high levels of DOTS coverage, such as Indo- Zimbabwe 135 674 539 nesia and Peru, have shown large decreases in Lesotho 179 696 517 TB incidence in recent years (table 2.6). Kenya 108 619 511 Namibia 260 717 456 Tuberculosis can usually be treated with a South Africa 268 718 450 course of four standard, or first-line, anti-TB drugs, which cost $14­18 per patient (Stop TB Peru 394 178 ­215 Partnership 2006). If these drugs are misused Haiti 484 306 ­178 or mismanaged, however, multidrug-resis- Maldives 148 49 ­99 tant TB (MDR-TB) can develop, which must Indonesia 343 245 ­98 be treated with more expensive second-line drugs, which have more side-effects and take Source: WHO. longer to work. In September 2006, WHO detected a deadly new strain of the bacteria-- called extensively drug-resistant (XDR) TB-- half a million cases of MDR-TB worldwide in Kwazulu Natal Province, the epicenter of in 2004, and MDR-TB can presage XDR- South Africa's HIV/AIDS epidemic. XDR-TB TB. WHO is leading international efforts to is resistant to a number of first- and second- address the XDR-TB problem in collaboration line anti-TB drugs, and treatment options are with countries and the Stop TB Partnership. very limited. Because of constraints on equip- ment and skills for diagnosis in poor coun- MDG7--Water and Sanitation tries, the global distribution of XDR-TB is not known with accuracy at the moment. How- MDG7 "Ensure Environmental Sustain- ever, WHO estimates that there were almost ability," includes a target that is interlinked The health MDGs, and especially maternal mortality and AIDS, malaria, and tuberculosis (TB) goals, are difficult to monitor, owing to the absence of vital registration and disease surveillance systems in many countries. The Health Metrics Network (HMN) is a global network launched in 2005 to help developing countries improve the availability and quality of their health statistics. Partners include developing countries, multilateral agencies (including the World Bank), bilateral donors, the Gates Foundation, and the major global partnerships. While many donors have supported monitoring and evaluation units (M&E) to report on spe- cific project activities, or have supported the DHS and other household surveys, they have never before worked together to strengthen health information systems in the developing world. Improv- ing countries' capacity to establish vital registration, disease and risk factor surveillance, national health accounts, and regular household surveys promises over time to reduce the costs of generat- ing the information that both policy makers and donors need and avoiding uncoordinated M&E activities. The HMN has developed a common framework setting out the standards, policies, capacities, and processes needed at the country level. The framework also serves as a diagnostic tool to estab- lish a baseline of currently available health statistics, and provides a roadmap for development and implementation of health statistics strengthening. During 2006, some 40 countries received grants from the HMN to carry out baseline assessments using the common framework. The network is currently developing time-bound plans for strengthening health information systems in an initial batch of countries. Source: Health Metrics Network. to progress on most of the human develop- America and South Asia are considered on ment MDGs--the target of halving by 2015 track to reach the target (and about 26 per- the proportion of people without sustain- cent of developing countries lack adequate able access to safe drinking water and basic data to judge).2 Sub-Saharan Africa is the sanitation. Hygiene, sanitation, and water region most seriously off track, but there supply have important influences on child are also some promising trends: Malawi and health, schooling attendance, gender equity, Namibia have reached the MDG target; 17 and other human development outcomes. of 36 African countries for which data are Hand-washing initiatives have been shown available are on or almost on track; and 6 to reduce the probability of contracting diar- of the 10 countries making fastest progress rheal diseases--an important cause of child globally are low-income African countries morbidity and mortality--by 44 percent, (table 2.7). sanitation improvements produced a 32 per- There has been progress on sanitation too, cent decrease, and improved water supply but not enough. Globally, access to improved resulted in a 25 percent reduction (Fewtrell sanitation has increased from 37 percent in and others 2005). A recent international poll 1990 to 52 percent in 2004--which is not on by the British Medical Journal chose sanita- pace to the goal of 69 percent coverage by tion as the greatest medical breakthrough 2015. Only two regions (East Asia and the since the 1840s. Pacific and Latin America) are on track for Globally, there has been significant prog- basic sanitation. While the South Asia region ress on water supply; access to improved is not on track given its very low starting water sources has increased from 73 percent base, large gains in access have been made, in 1990 to 80 percent in 2004 but only Latin especially in India, where sanitation coverage more than doubled between 1990 and 2004. Access to improved water is growing About 46 percent of all developing countries are considered off-track to the sanitation tar- Percent of population get, while another 34 percent have inadequate with access to improved data. Progress in Africa has been slow. Only water sources Average annual percentage point one of the 32 African countries for which Countries 1990 2004 increase, 1990­2004 data are available is on track. This makes the performance of that country--Senegal--all the more noteworthy (table 2.8). The three largest challenges in achieving the water supply and sanitation MDG targets are: (1) ensuring that sanitation gets sufficient attention in national investment programs; (2) reducing rural-urban disparities in the access of water supply and sanitation ser- vices; and (3) ensuring the sustainability of investments already made. Access to sanitation tends to lag behind Maldives 96% 83% ­ water supply, which has historically been Uzbekistan 94% 82% ­ Algeria 94% 85% ­ accorded a higher priority both by govern- Comoros 93% 86% ­ ments and households. The lag is due to a number of factors, including limited demand from households, institutional fragmenta- Source: WHO-UNICEF Joint Monitoring Program, Meeting the Drinking Water and Sanitation Target: The Urban and Rural Challenge of the Decade, 2006. tion, poor coordination, and limited capac- ity to address the problem at scale. While the poorest quintiles in every region have the Access to improved sanitation is growing least access to water and sanitation, the big- gest gaps in access fall along the rural-urban Percent of population divide. In developing countries, 92 percent of with access to improved sanitation Average annual the urban population has access to improved percentage point water sources and more than 73 percent to Countries 1990 2004 increase, 1990­2004 sanitation, while coverage in rural areas is 70 percent for access to water and only 33 percent for basic sanitation. The rural-urban disparities are especially sharp in Africa and South and East Asia. A major lesson of the past decade is that water supply and sanitation investments are not sustainable unless adequate attention is paid to the institutional context and perfor- mance of service providers. Principles for a sound operating environment include the use of demand-responsive approaches in service provision, managing services at the lowest Liberia 39% appropriate level, adherence to cost recovery policies where necessary in combination with transparent subsidies targeted to the poorest users, appropriate technologies and standards Source: WHO-UNICEF Joint Monitoring Program, Meeting the Drinking Water and Sanitation to ensure cost effectiveness of investments, Target: The Urban and Rural Challenge of the Decade, 2006. and a shift from sewerage systems to on-site funding through these initiatives translates into sanitation and hygiene promotion programs. improved access to water sources and sanita- Different management models are appropriate tion, and faster progress on the WSS targets. for different country contexts and increasingly diverse management models are being suc- Financing MDG progress cessfully used in both the public and private sectors. Increasing use of public-private part- External financing. External financing for nerships is another promising trend. health and education has nearly doubled in Despite its importance for directly sup- real terms since the MDGs were adopted. porting achievement of the WSS MDG and While aid for health continued to rise from indirectly contributing to progress on health, 2004 to 2005, education-related official devel- education, and gender MDGs, ODA for water opment assistance commitments showed their and sanitation declined significantly from the first decline (figure 2.3). The decline reflects mid-1990s through 2002. Since 2003, assis- a drop in commitments for China and India, tance for WSS has begun to swing upwards both of which received large commitments again, but even in 2005 had not recovered to linked to new multiyear programs in 2004. the 2000 level. In the past two years, efforts On a sectoral basis, basic plus general educa- have been made to ramp up financing for WSS, tion funding rose to 53 percent of the total especially for Africa. This has resulted in the volume, and the share for postsecondary edu- recent establishment of the Africa Infrastruc- cation fell to 38 percent. Commitments for ture Consortium, and the Rural Water Supply Sub-Saharan Africa increased, to 29 percent and Sanitation Initiative led by the African of the total volume. Education commitments Development Bank. Although this is a posi- are expected to increase again in 2007. tive development, given the long lead time for In health, much attention has been given investments, it will be several years before new to the expansion of private funding sources Development assistance for education and health, 2000­05 Source: OECD/DAC data. over the past five years, with the Gates Foun- contrast, support for health from multilateral dation in particular increasing from about development banks has been flat in real terms $0.6 million to $1.2 billion per year in assis- since 2000, most likely reflecting recipient tance, two-thirds of which is channeled to countries' preference for grant funding over countries through global programs such as even highly concessional lending. the Global Fund for AIDS, TB and Malaria As large as the expansion of bilateral (GFATM); and the Global Alliance for Vac- and private assistance has been, current cines and Immunizations (GAVI). But as can support levels are still far short of the esti- be seen from figure 2.3, a number of bilateral mated financing needs to reach the health donors have increased their core assistance MDGs, the most conservative of which levels at least as strongly--the United States calls for $25­50 billion per year in external has more than doubled its health funding, support. There is also a growing imbalance reaching close to $4 billion in 2005; Spain between the volumes of funding mobilized has also doubled its assistance; and France for specific diseases and the core funding and Norway have quadrupled their official needs of health systems for scaling up basic development assistance for health. A number service delivery. Multilateral development of new financing modalities have also been banks (MDBs) could potentially play a key established with support from bilaterals. In role in the provision of "complementary" Along with mobilizing more funding, donors for health have worked since 2000 to improve aid quality and address market failures in the supply of global public goods for health through the development of innovative financing methods: The International Finance Facility for Immunization (IFFIm) was designed to increase the stream and predictability of funding for health and immunization programs. A pilot for the larger Interna- tional Finance Facility, the IFFIm mechanism converts donor pledges of off-budget commitments of future resources into funds available for near-term disbursement through bond markets. In 2006, $1 billion was raised in an initial bond offering and channeled to GAVI, which is now commit- ting these resources to country programs for the introduction of new vaccines and health system strengthening. Advance Market Commitments (AMCs) for vaccines are financial commitments from donors to subsidize future purchases of yet-to-be-developed vaccines. AMCs were designed to increase the incentives for global drug companies to invest in research, development, and production of vac- cines that would serve developing-country markets. The first AMC pilot is targeting pneumococcal vaccines, which could avert 1.6 million developing-country deaths a year. Donors will launch the AMC Pilot in February 2007, committing $1.5 billion to support the purchase of pneumococcal vaccines through roughly 2019. UNITAID, financed through a tax on airline tickets and other sources, was designed to provide a long-term, predictable funding stream for drugs and diagnostic kits to fight HIV/AIDS, TB, and malaria. In 2007, UNITAID is expected to receive $300 million from ticket levies implemented in France, Chile, and 19 other countries including Brazil. UNITAID will work with global funds and agencies such as the Clinton Foundation on the supply side of the international pharmaceutical market--pooling orders, stimulating competition, and negotiating lower prices for drugs--as well as providing support for programs on the ground, mostly through the Global Fund. In 2007, UNI- TAID expects to supply drugs for 100,000 people in 16 countries, and to reach 130,000 children with an "HIV pediatric package" of drugs, diagnostic kits, and nutrition. un-earmarked financing and technical sup- percentage of GDP rose by 26 percent between port aimed at overall strengthening of health 2000 and 2004.3 However, the region still systems. However, the trend in MDB financ- trails all other regions except South Asia in its ing for health over the past several years has spending share for health. been flat. Donor harmonization. Efforts to improve Developing-country spending. Govern- the "quality" of donor assistance are on ment spending on health and education has very different trajectories in education and grown as a share of the gross domestic prod- health. In education, an increasing number uct (GDP) in all regions (figure 2.4). The of developing countries are joining the EFA- largest increase has been for education in FTI, which the OECD/DAC Forum recently South Asia--from 3 to 3.8 percent, driven by ranked as one of the most effective among India--but the share of GDP devoted to edu- global programs in promoting donor harmo- cation in that region still remains the lowest nization and aid alignment with country-led in the world. In both the Middle East and priorities. Core principles of the FTI are to North Africa and Sub-Saharan Africa, edu- unify donors in reviewing and endorsing a cation shares were already a relatively high credible education sector strategy for each share of GDP and they have increased fur- FTI country and committing development ther. At 5.5 percent and 5 percent of GDP, partners to align aid with that framework. respectively, these two regions now trail only Off-plan and off-budget support are not East Asia and the Pacific in the national fiscal permitted. FTI donors have also been mak- priority given to education. ing progress on the ground in implementing In health, spending shares increased in all the Paris Harmonization accords: they have regions except Latin America and the Carib- reduced the number of separate donor mis- bean. The largest increase was in Sub-Saharan sions, increased their use of country systems, Africa, where government health spending as a and pooled their financing. The review paper Developing countries are devoting more national resources to education and health Sources: UNESCO Institute of Statistics; WHO. cited earlier concluded that harmonization ing needs (see box 2.6.) A recent analysis of progress under the FTI was real, significant, Ethiopia confirmed similar patterns. and a major achievement of the initiative to The global health community has acknowl- date (World Bank 2006). edged these issues at successive international In health, progress has been largely based fora. In 2004­05 WHO and the World Bank on country-specific efforts, and there is no convened a High-Level Forum on the Health comparable organizing entity for unifying MDGs (HLF) to examine aid effectiveness in donors around coherent national strategies. health. Major issues highlighted by the HLF Previous GMRs have documented the "ver- included poor alignment of aid with govern- ticalization" of global support for health ment priorities; the volatility and short-term over the past seven years and growing con- nature of many commitments; harmonization cerns about transactions costs, coordination and alignment issues created by the large num- failures, and poor alignment with recipient ber of different donors in health, particularly countries' national health priorities. A 2006 Global Health Partnerships; and the need for case study of Rwanda, whose strong national closer monitoring of the impact of increased commitment to achieving the health MDGs general budget support on funding available and innovative policies have attracted large for health. Perhaps the most important out- increases in donor funding, provided some come of the HLF is the development of a set graphic examples of the difficulties recipient of Best Practice Principles for Engagement of countries face in achieving policy coherence, Global Health Partnerships at the Country aligning aid to sectorwide financing needs, Level--based on the Paris Declaration--which and mapping volatile annual or biannual aid are now being implemented by major health commitments onto long-term recurrent fund- partnerships. Other contributions include Rwanda has made impressive progress in health over the past several years, with innovative reforms, sustained implementation and increased domestic spending supported by burgeoning donor sup- port: the health sector share of total government spending grew from 2.5 percent in 1998 to 10 percent in 2005. By 2004, donor grants represented about half of total government spending in Rwanda, but this figure actually underestimates the importance of foreign aid, because of large off-budget funding, especially in the health sector. There is no question that the current partnership between the government and donors is supporting the country's progress. But a recent review of Rwanda's development assistance for health documented some of the issues in aid delivery that the government must navigate in translating aid resources into results (Republic of Rwanda 2006). The first is the challenge of achieving policy coherence--and even basic fiscal monitoring-- given that only 14 percent of total donor support for health is channeled through the budget of the Rwandan Ministry of Health or through local governments and health districts (12 percent). The remaining 74 percent of aid is channeled by donors directly to NGOs or their own-managed projects. This aid may be effectively used: even the government has recognized the efficiency of contracting services to NGOs. The issue is that it is difficult for the Ministry of Health--which remains responsible for health outcomes in the country--to account for, or track, the total volume of health spending. A second issue is alignment: while Rwanda is trying to implement a major reform of its overall health delivery system, out of total on-budget official development assistance to the health sector continued in 2005 of $75 million, $46 million was earmarked for HIV/AIDS, $18 million for malaria, and only $1 million for child health. Although external funding for HIV/AIDS is still short of estimated needs, it far exceeds support available for other health priorities--although some HIV/AIDS fund- ing supports facilities upgrading that has broader benefits. A third issue is volatility: much of the assistance for Rwanda, like other countries, reflects com- mitments from bilateral donors that are for 1­2 years at most--although the United Kingdom is a notable exception. This leads to substantial variations in funding levels from year to year, and inhibits long-term planning. In two areas in particular--national decisions to scale-up health service provision by training and hiring more doctors and nurses, and the expansion of AIDS patients on antiretroviral treatment--Rwanda and other governments currently incur major risks of sustainability. Finally, the Rwanda case illustrates, from a country's viewpoint, the disparity between fund- ing available for vertical health programs and other development priorities. The report notes that "spending on health has increased markedly in recent years...(but) infrastructure and agriculture have been relatively neglected. Major investments in the road network are needed to support eco- nomic growth and poverty reduction. Major investments are needed in energy and in water and sanitation." While such multisectoral investments can have important effects on human develop- ment outcomes, the current aid architecture makes it difficult for countries to allocate resources efficiently across sectors to capture these complementarities. On-budget donor assistance to the health sector in Rwanda, 2005 Source: Republic of Rwanda, 2006. analyses of the challenges of scaling up ser- into coordinated sectorwide approaches vice delivery, including human resource con- (SWAps). The Rwandan government has also straints, and the "fiscal space" for sustainable recently developed an "Aid Policy and a Joint health financing. Agreement"--a compact to be signed with Tanzania and Uganda are good examples development partners--as part of the new of country leadership in pulling health donors Economic Development and Poverty Reduc- tion Strategy in 2007. Elements of the com- the MDGs: almost one-third (31 percent) of pact include donors' agreement that all aid all child deaths and 29 percent of all 12-year- will be included in the government budget, olds who fail to complete primary school in and a stipulation that aid projects not meet- developing countries are in fragile states. As ing this requirement will only be accepted table 2.9 shows, on virtually every MDG, if they are sustainable. Donors have also fragile states acccount for a disproportionate been asked to assess how well their practices share of the world's people who suffer from align with the draft policy, and to discuss poor outcomes and poor services. Only two all planned activities with the government. of the 35 states considered fragile in 2005 are Resource flows to the sector will be through on track to reach the child mortality MDG: a SWAp. Donors providing budget support Timor-Leste and Lao PDR. And 25 percent of will be guided by a new Partnership Frame- the countries where child mortality rates are work for Harmonization and Alignment of actually worsening are fragile states. Budget Support. Even when compared with other low- income countries, the performance of fragile states is considerably weaker; as figure 2.5 Fragile States shows, child mortality in fragile states over Fragile states or low-income countries under the past 15 years has remained higher and stress (LICUS) account for 9 percent of the progressed more slowly than in other low- population in developing countries but 27 income countries. percent of those living on less than US$1 per Primary completion rates in fragile states day. These countries are least likely to achieve also trail those of other low-income coun- Fragile states lag most on MDGs Millions of people Total developing Percent of total MDG indicator countries Total fragile states fragile states Underweight children 143 22.7 16% Children of relevant age that did not complete primary school in 2005 13.8 4 29% Children born in 2005 not expected to survive to age five 10.5 3.3 31% Unattended births 48.7 8.9 18% HIV+ 29.8 7.2 24% TB deaths 1.7 0.34 20% Lacking access to improved water 1,083 209 19% Lacking access to improved sanitation 2,626 286 11% Source: World Bank staff estimates. Child mortality is higher and showing less progress in sively quickly. In 7 out of 12 cases studied fragile states in a 2005 review, primary school enrollments were higher in the year after conflicts than in the year before they began--including in countries where 50 percent or more of schools were damaged or destroyed by hostilities. (Buckland 2005). The "surprising resilience" of primary education systems is attributed to a number of factors: the strong priority given to schooling by communities, which often strive on their own to keep schools open dur- ing conflicts; the high and immediate priority education typically receives from donors and countries once conflicts end; and the oppor- tunities for education reform that postcon- flict settings can present. Since 1995, buoyed by the exceptional progress in Cambodia, the fragile states as a group have increased pri- mary completion faster than nonfragile low- income countries. Source: World Bank. But primary education may be an excep- tion. The effects of conflict on child health Primary completion rates are lower in fragile states, but can be more severe and harder to reverse. improving More than 2 million children have died over the past decade as a direct result of armed conflict, and at least 6 million children have been seriously injured or permanently dis- abled (Bustreo and others 2005). A 2003 study found that during a typical five-year war, infant mortality increases by 13 percent and the effect is persistent: in the first 5 years of postconflict peace the infant mortality rate remains 11 percent higher than the pre- conflict baseline (Hoeffler and others 2003). In general, weak institutions--whether or not they are marked by conflict--constrain progress in expanding basic services in frag- ile states. As figures 2.7 and 2.8 show, while average access to improved water supply and measles immunizations have expanded in fragile states, they have not kept pace with the rates of improvement in nonfragile low- Source: UNESCO. income countries. Although the gap in average aid flows tries (figure 2.6), but the gap is not as great. between fragile states and other low-income Indeed, the data suggest that at least under countries has narrowed over the past five some circumstances, such as restoration of years, aid to fragile states is particularly primary school functioning after conflicts, it is volatile. When donors do engage, they often possible for fragile states to rebound impres- establish parallel systems because government systems are weak. This approach can further Measles immunization in fragile states remains lower undermine fragile states, and can make future capacity building difficult. A particular issue in the human development sectors is the potential for gaps in service delivery in the "transition phase" between the end of a humanitarian cri- sis and the beginning of longer-term recovery and reconstruction projects. Aware of these issues, donors assisting in the postconflict reconstruction of Afghanistan and Timor-Leste have worked carefully to imple- ment strategies to smooth the transition (box 2.7). The positive results of these experiences may hold lessons for donors working in other fragile states, and possibly in other sectors. Is MDG Progress Reaching the Poor? Equitable country progress toward the MDGs comes from reaching the poorest citizens-- Source: WHO. bringing up average national indicators by extending services and achieving outcomes A growing gap in access to improved for the lowest quintiles of the population. water Few sources of data allow national indica- tors to be "unpacked" to see which groups within countries have benefited most, but the international DHS does. As of end-2006, 21 developing countries had survey results that enable us to analyze MDG progress since the goals were adopted in 2000. While these countries are not a representative sample, they span all regions and different levels of per capita income. In every country, both access to services and outcomes for the poorest quintiles are lower than for other income groups, and the gaps are often disturbingly large. While child mortality rates across these 21 countries are 79 per 1,000 live births for the top quintile and 125 per 1,000 births for the population as a whole, they average 148 per 1,000 live births for the lowest quintile. While primary Source: UN Joint Monitoring Project. completion rates average 81 percent for the highest income quintile and 55 percent for the population as a whole, they average were immunized for measles in these coun- 36 for children from the poorest quintile. tries, but only 59 percent of children in the Similarly for the delivery of services such as bottom income quintile. immunizations, 82 percent of children in the It is sobering to keep these gaps in mind. top quintile and 69 percent of all children But the data also show some encouraging After the violent withdrawal of Indonesian troops in 1999, more than 70 percent of Timor-Leste's health facilities were destroyed or badly damaged and approximately 80 percent of the country's health managers had left the country. The government faced the immediate challenge of restoring health services and a longer-term challenge of rebuilding a sustainable health system. With support from IDA and other donors, the government implemented a two-tier strategy that addressed both. Under the first Health Sector Rehabilitation and Development Project (HSRDP I), the government addressed the short-term need to get services going again by contracting with local and interna- tional relief NGOs; for a time, NGOs became the main health service providers. At the same time, the government implemented strategic longer-term investments in a sustainable national health system--by reconstructing facilities, developing national health policies and regulations, redeveloping the organizational structure of the health system, and training new human resources. Afghanistan faced a similar challenge after the collapse of Taliban rule. The country had some of the worst health indicators in the world, with estimated under-five mortality of 256 per 1,000 births, compared to 92 for South Asia. To address urgent needs, the World Bank and the Ministry of Public Health initiated the Health Sector Emergency Reconstruction and Development Project, and similarly contracted with 10 local and international NGOs to deliver a priority basket of health services in 12 provinces. The results in both countries were impressive. In Timor-Leste, the use of health services increased from one visit per person per year on average to at least 2.5. Measles immunizations rose from 26 percent to 73 percent of cildren; skilled attendance at birth increased from 26 percent to 41 percent; and child mortality declined dramatically. In Afghanistan, even in provinces such as Helmand, where continuing violence cost the lives of several health workers, patient visits more than doubled, from 157,000 in 2004 to 338,000 in 2006. Across all 12 provinces, there was a fourfold increase in the number of people visiting rural health centers, a 60 percent increase in the number of functional health centers, and an increase from 5 percent to 63 percent of pregnant women receiving prenatal care. Afghanistan and Timor-Leste illustrate the potential of approaches that integrate the best features of the public and private sectors. In each case, the government led the strategy and oversaw imple- mentation, but delegated the role of principal service provider to NGOs. In Timor-Leste, subsequent projects have supported a progressive transition of service delivery from international NGOs to gov- ernment district management teams. In Afghanistan, owing to the perceived success of the current contracting arrangements, the possibility of a longer-term partnership is being considered. Prepared by: Fadia Saadah, EAHD and Benjamin Loevinsohn, SASHD. trends. In the countries where service delivery In 15 of the 17 countries making progress is expanding--for primary education, immu- in reducing child mortality, there is either nizations, and other health services--there little gap between the rates of improvement are many countries where the gains in service for the lowest quintile and the population access for the poorest children are larger than average or faster progress for the poor. for the population as a whole. Child mortal- In 14 of the 15 countries that have increased ity outcomes--which generally take longer measles immunization rates, coverage to produce and reflect many factors beyond increased faster for the poorest children direct service delivery in health--also show than for the population as a whole. some progress for the poor, although not as In 11 of the 13 countries which show strongly. Specifically, the data show the fol- increased primary completion rates, the lowing: poorest quintile improved more than the population as a whole. can countries, 2 in East Asia, 2 in North Africa, and 2 in Latin America (figure 2.9). Child mortality. The 21 countries with The continued progress in Egypt, which recent data include 14 Sub-Saharan Afri- trend data since 1990 show to be on Child mortality progress Source: Demographic and Health Surveys. Note: The boxed numbers show the numbers of child deaths per 1,000 live births in the most recent surveys. track to reach the child mortality MDG, for several decades and has a high level of is clear. But these data suggest that several overall coverage, including for the lowest other countries--Indonesia, the Philippines, quintile. Kenya, Malawi, Chad, and Nigeria Malawi, Madagascar, Mozambique, and again appear as troubling cases of decline Ethiopia--have accelerated their progress for all segments of the population, but with on under-five mortality since the MDGs the poorest harmed most. were adopted. While Egypt has maintained Primary completion. In the 18 countries a decline in under-five mortality of over 4.3 with comparable education data, the picture percent per year for the full period since that emerges is that primary completion prog- 1990, the four African countries and Indo- ress is strongly pro-poor. In all but two of the nesia and the Philippines have all begun reg- countries that registered increases in the share istering the necessary rate of progress over of youths who completed primary education, the past several years. This is encouraging. the poorest quintile improved more than the These data also show that in virtually all average, and in most of those countries the the countries where there is progress, it is differential was large. Burkina Faso, Mada- reaching the poor. In nine countries--Indo- gascar, Ethiopia, Bangladesh, and Morocco nesia, Mozambique, Bolivia, Bangladesh, stand out as making exceptional progress in Rwanda, Ghana, Guinea, Cameroon, and extending basic education to all segments of Senegal--children in the poorest quintiles are the population (figure 2.11). showing faster improvement in health than for As with the other indicators, Kenya and the population as a whole. In six other coun- Chad show no evidence of progress. More sur- tries (Malawi, Egypt, Madagascar, Burkina prising, perhaps, are the declines for Tanzania Faso, Morocco, and Tanzania) there is little and Mozambique--both of which have been gap. Only in two countries (Ethiopia and strongly committed to education for some Colombia) is the rate of improvement in child time and are considered high performers by health for the poorest groups seriously lagging donors. The explanation may be that the share behind the average. And in Chad and Kenya, of the cohort aged 15­19 that has completed overall deterioration is hurting the poor worst. primary education reflects changes in the edu- In Nigeria, there is a long gap in time between cation system and participation rates from the surveys, which could affect the compara- roughly a decade earlier; in a sense, these data bility of the results. But the general picture of "look in the rearview mirror." Policy changes very slow MDG progress in these countries is of the last few years--such as Kenya's 2002 corroborated by other data and is troubling. adoption of free primary education--will not Immunization against measles. Given be reflected in these results. But the impact of the aggressive global campaign since 1999 Mozambique's civil war of a decade ago on to increase measles immunizations, the DHS schooling participation at that time would be. data offer a measure of the success of this Neither of these explanations is fully satisfying effort. The data for 21 countries indicate in the case of Tanzania, however, which has that especially in Sub-Saharan Africa, mea- sustained educational progress for many years. sles vaccination rates have increased and While rates of improvement will necessarily children in the poorest income quintile have slow in countries with high levels of primary benefited substantially (figure 2.10). There school coverage, which applies to Tanzania as is only one case, Bangladesh, where measles well as several others of these countries, the coverage has grown without benefiting the poor should not necessarily show worsening lowest quintile relatively more. It should trends. be recalled, however, that Bangladesh has invested heavily in vaccination programs Measles vaccines are reaching the poor in many countries Source: Demographic and Health Surveys. Note: The boxed numbers show the immunization coverage (for children 12­24 months) in the most recent survey. Primary completion progress is benefiting the poor in many countries, but not all Source: Demographic and Health Surveys. Note: The boxed numbers show the share of 15- to 19-year-olds who have completed primary school in the most recent surveys. aged children across the country. Using a net- work of NGO volunteers, the simple test of reading and math skills was given to children Education: Are Developing-Country at home, whether enrolled in school or not. Students Learning? The good news was that over 90 percent of all In 2005, a respected NGO in India shocked Indian children reported being enrolled. The the nation by publishing the results of a test sobering result: 68 percent of primary school administered to 300,000 primary school­ students could not read a simple (second grade-level) paragraph and 54 percent of chil- the degree of earnings inequality in the labor dren could not solve a simple two-digit math force and the degree of dispersion in adult problem (Pratham 2005). The results var- literacy scores across the population (Nick- ied across states, but the mean performance ell 2004). While learning disparities do not was troubling. What was India's tremendous cause income inequality, the research does progress in expanding primary access over suggest that policies to lessen gaps in average the past decade really producing? learning levels across different segments of Other developing countries are confront- the population may have direct and positive ing the same question. An NGO-adminis- impacts on income distribution. tered test in a small set of low-income schools Finally, they find a connection between across Peru in 2005 found that 50 percent of education quality and growth that "dwarfs children at the end of second grade could not the association between quantity of educa- read a single word of a simple first-grade text tion and growth." A one standard-deviation (Cotlear 2006). A similar reading test in rural increase in a country's average performance Cameroon showed that 80 percent of the third on international tests is associated with a 1­2 grade children tested could not read a single percentage point higher annual per capita word of a first-grade text (Walter 2007). GDP growth--a huge effect. Although small New research shows that such learning subsamples make these effects more tenta- failures have high costs for countries, as well tive, Hanushek and Woessman also find that as the children involved. In a comprehen- effects for developing countries (an increase sive review this year, two leading education of 2.29 percentage points per year) are higher researchers show that most of the economic than for OECD countries (1.7 percentage returns to education are a return to the cog- points per year), and that countries' trade nitive skills of the population, and not to the openness and institutional quality signifi- average levels--or quantity--of education cantly enhance the impacts. attained (Hanushek and Woessman 2007). If Does it matter whether countries' average two countries have the same average years of scores reflect broad-based education systems schooling but in one country average learning of reasonable quality or systems with a pin- levels are higher, individual earnings, the dis- nacle of very high-scoring students? Both tribution of income, and the long-term rate seem to be important. Countries' mean scores of economic growth will all be higher in that are highly correlated with the share of stu- country. dents who reach a threshold level of skills on While it has long seemed intuitively obvi- international tests. In other words, countries' ous that a year of schooling in Mali is not achievement of "education for all" by bring- equal to a year of schooling in the Republic ing all students to basic levels of literacy and of Korea or Finland, it has never before been numeracy is key for capturing the economic possible to "unpack" the differential quality benefits of education. But the share of top per- element. Using a new data set that combines formers is also important, and "seems to exert the results of all major international tests separately identifiable effects on economic over the past 40 years, Hanushek and Woess- growth." Hanushek and Woessman postulate man draw several important conclusions for that top performers afford an economy the education policy. First, they demonstrate that capacity to innovate, and a large population a large part of the higher incomes that more with basic skills provides the ability to diffuse highly educated individuals earn, in both and apply new knowledge broadly. developed and developing countries, is a What does this mean for the education function of cognitive skill levels (as measured MDG? First, it provides powerful economic on international tests), rather than years of arguments to support the goals of Education schooling completed. Second, they document for All as universal primary completion with a tight correlation across countries between adequate levels of learning, and not simply completion of a target cycle of schooling. The directly compared, as only 7 developing coun- 2000 Dakar Education for All goals were tries--and no low-income countries--par- framed in these terms--universal coverage and ticipated in the latest OECD cross-country learning--and many observers have urged that assessment, and only about 20 developing targets for the primary education MDG should countries have participated in any major explicitly include learning goals as well. international assessment. However, figure Second, it points to the value of measuring 2.12 provides a graphic image of the large learning outcomes in relation to internation- gap in performance between OECD and ally benchmarked standards. The economic developing-country participants on recent benefits identified were associated with learn- exams to measure math, literacy, and think- ing levels that met an international threshold. ing skills among 15-year-old students. How many developing-country students Poor as these country results are, they rep- meet this threshold today? Very few can be resent the upper bound of student learning in Reading and math performance on the OECD PISA Exams, 2000 and 2003 Source: PISA 2003 and PISA 2000. Note: PISA = Program for International Student Assessment. the developing world, because only relatively Southern African countries cannot be directly high-income developing countries have partici- compared with that of the IEA test, the results pated in any international tests to date, and in are similarly distressing. In several countries these countries, the pool of students tested rep- in the region, 50 percent or less of children resents only the relatively privileged 15-year- are able to read by age 12. In Malawi, high olds still in school at that age. In the regions dropout rates in primary school combined where achievement of the education MDG is at with low learning results in only an estimated greatest risk--Sub-Saharan Africa and South 30 percent of children being able to read at Asia--very few countries have participated that age. in any OECD-benchmarked assessment. But Is there a quality-quantity trade-off in figure 2.13 shows how even by age nine read- education? The fact that rapid enrollment ing ability in developing countries can lag that expansion in developing countries is often in OECD countries by a significant margin. associated with strains on quality leads to By the fourth grade in Argentina, Colombia, speculation that there is an inherent trade- and Morocco, less than 50 percent of all chil- off between schooling coverage and learn- dren can read at the lowest-threshold-level of ing levels. But UNESCO's 2005 report on literacy, on an international test normed for education quality, which looked carefully at OECD countries. In contrast, 96 percent or the available evidence on learning outcomes, more of fourth graders in Sweden, Latvia, and and other recent cross-country studies do the Netherlands read at this level (Greaney and not support this (UNESCO 2005, Crouch Prouty 2007). and Fasih 2005). In fact, cross-country data Although the level of "minimum literacy" show a positive correlation between educa- in the regionally benchmarked assessment for tion coverage and average learning levels, at Many children do not attain minimum learning levels Source: Fourth-grade test: International Association for the Evaluation of Educational Achievement (IEA), Progress in International Reading Literacy (PIRLS) 2001; Sixth-grade test: Southern African Consortium for Monitoring Educational Quality (SACMEQ). Enrollment data: Demographic and Health Surveys. least over the long term. There is no case of developing countries because they tend to be an education system with high average learn- too difficult. It is important that developing ing levels that has not also achieved universal countries have access to evaluation and test- primary completion and virtually universal ing methodologies that allow them to link to secondary school completion. There are also performance in the developed world, but also cases of countries that have simultaneously provide good measures of the range of per- increased schooling access and raised learn- formance in their own populations. ing levels. There is no inherent tradeoff. One approach that has been proposed is That does not mean it is easy to maintain the development of a set of global learning school quality as enrollments swell, especially goals--a core set of literacy, numeracy, and when enrollments increase dramatically, as in critical thinking skills that children should response to policy changes such as the elimi- master by the end of primary school (Filmer, nation of school fees. But the essential fact Hasen, and Pritchett 2006). While in a world that there is no inherent trade-off is impor- of global economic competition, the most rel- tant, because a large number of countries are evant benchmark for learning is arguably a still far from universal primary completion global one, there are many alternative ways and many will need to find ways of scaling- to support a stronger focus on learning out- up service delivery even faster from now to comes. An internationally normed subset of 2015 to reach the MDG. Slowing expansion questions, for example, could be built into would harm the poorest and most margin- national or regionally benchmarked assess- alized groups most. The evidence suggests ments. The important thing is to generate a that most developing countries are striving valid measure of student learning levels that to universalize primary education as quickly can be tracked over time and directly com- as they can. The strategy must be to support pared with results for other countries. their efforts to give adequate attention to An important issue for cross-country quality and ensure that children learn. And comparisons in an MDG context is that the even many countries which have reached or average learning levels achieved by children are on track to the MDG in terms of coverage in school must be adjusted for the share of must do more to improve learning, to equip students not in the school system to gener- youth with the literacy, numeracy, and critical ate a true picture of the average literacy and thinking skills needed for full participation in numeracy skills of the population. To do oth- civic life and economic productivity. erwise creates adverse incentives for expan- Support for measuring learning out- sion, particularly for marginalized students. comes. The policies that can help countries Regular tracking of student learning is achieve improvements in quality and learning also essential for establishing accountabil- at the same time as they expand access and ity systems in education that focus teachers, completion have been analyzed comprehen- parents, and administrators on the right out- sively by UNESCO (GMR 2005), and other comes. When results are fed back to teachers, studies. A relevant question from the stand- they become a tool for classroom improve- point of MDG progress is whether donors are ment. When results are fed back to communi- currently giving adequate support to devel- ties, they can strengthen local voice in school oping countries' efforts to monitor student governance. And when results are tracked by learning. Today, a developing country that system administrators, they help evaluate the wishes to benchmark its students' learning effectiveness of education spending. Abun- performance against that of other countries dant research from developed countries and lacks good instruments for doing so. All of increasing evidence from developing coun- the currently available international tests tries shows a consistently weak correlation have been developed for advanced countries. between higher spending and improved stu- They do not provide useful information for dent learning (Hanushek and Kimko 2000; Woessman 2003; Pritchett 2004; Hanushek port is justified to underwrite its costs. The and Luque 2003; Mizala and Romaguera key is to develop new assessment instruments 2002). This does not mean that learning can that are suitable for developing countries, but be improved without more resources, and linked to existing international tests. How- there is some evidence that a minimum set of ever costly it might be to develop, pilot, and inputs (basic facilities, teacher presence, and sustain such assessments, the sums would be the availability of books) is a threshold con- small in relation to the billions of dollars in dition for education to occur (Duflo 2001). new aid being mobilized to support universal But it is evidence that many common and primary completion. Globally benchmarked costly "input based" strategies for strength- assessments covering large numbers of devel- ening education systems--such as upgrad- oping countries would provide the strongest ing teacher qualifications or lowering class platform yet for research on "what works" to size--do not work. Only by tracking student promote learning results in different country results over time can school systems gauge contexts. whether teacher quality is truly improving, Measuring learning outcomes per se does and which policies and investments aimed at not improve education systems; it does not making teachers perform better really pro- eliminate the political obstacles to key reforms duce results. or ensure that better policies are well imple- There are political and technical chal- mented. Some OECD countries that regularly lenges to standardized learning measurement. participate in internationally benchmarked Political challenges in part stem from the assessments, such as the United States, have power that exposing student learning results been notably unsuccessful in improving their can have in holding education stakeholders results. But other countries have done so: the accountable. But there are also legitimate United Kingdom, Australia, and Finland are concerns in developing countries that OECD good examples. Clearly, testing is not a pana- benchmarked tests do not measure their real- cea. But it is an essential tool for countries ity or their curriculum. that want their policy and program choices to Designing valid and reliable tests is costly, be guided by evidence that they work. Such takes time, and requires sophisticated skills. evidence holds crucial potential for more If a test is internationally benchmarked, effective developing-country policies, more sustained and intensive international coop- productive aid, and faster and more meaning- eration is also required. For this reason, the ful progress toward the education MDG. approach being piloted in some developing countries--to use simple tests of reading flu- Promoting Quality in Health ency at the end of second grade--is a wel- come development. Such tests are low cost, Quality in health can be measured in terms of relatively easy to administer, and provide the structure of supply, process, or outcomes. early feedback on literacy development that As in education, the most meaningful mea- school systems can act on. sures of a system's quality are the outcomes it But such nationally oriented tests are not produces. But in health, data on key outcomes a substitute for more systematic tracking of such as child or maternal mortality are diffi- learning across developing countries. Exist- cult to collect, slow to change, and heavily ing research points to the need for countries influenced by factors beyond health. Process and donors to shift from spending and aid data also present collection challenges. As a based on inputs to spending based on edu- result, as in education, the most commonly cation results--and specifically, measurable used quality measure is inputs--such as the improvement in student learning. There is distribution of health clinics, drug supplies, a genuine global public goods dimension in and the quantity and qualifications of health cross-country assessment, and donor sup- providers--and is the basis for most health sector planning (Collier, Dercon, and Mack- economic development and very different innon 2003; Lavy and Germain 1994). health systems--India, Indonesia, Tanzania, Much of the attention in relation to the Mexico, and Paraguay. There was substantial health MDGs has focused on the challenge variation across and within countries, but the of scaling up one key input: health provid- overall quality of care was low. In a disturb- ers. The Joint Learning Initiative of the WHO ing number of cases, clinicians routinely mis- and other health donors has estimated that diagnosed and mistreated common illnesses, large increases in doctors and nurses in the not because of lack of training or medicines, developing world will be needed to attain the but because they did not exert the effort nec- health MDGs. In Africa alone, an estimated essary to find the correct diagnosis (Leonard, 1 million new health workers are required Masatu, and Vialou 2005). by 2015, to increase the current ratio of 1 In Tanzania, 33 percent of clinicians mis- provider per 1,000 people to a target of 2.5. diagnosed a woman with pelvic inflamma- While cross-country studies provide some tory disease and 60 percent mistreated the evidence that provider density is correlated condition. This disease--which is caused by with services such as immunization rates untreated sexually transmitted diseases-- and assisted births, it is equally clear from makes a woman more susceptible to HIV/ research that weak incentives for perfor- AIDS and more likely to spread the illness mance can drive a large wedge between the to partners if untreated. While 86 percent theoretical availability of providers and the of clinicians correctly diagnosed a patient quantity of care they actually deliver. Chaud- suffering from classic symptoms of TB, 67 hury and others (2006), for example, found percent mistreated the disease. Less than 20 an average provider absence rate of 35 per- percent of clinicians informed TB patients cent in surprise visits to health facilities in six of the importance of taking medicine con- developing countries, with an absence rate sistently, even though they knew that effec- for doctors in some rural areas reaching as tive treatment of TB requires careful ongoing much as 75 percent. management (Leonard, Masatu, and Vialou Measuring the quality of clinical practice. 2005). Some of the most recent research in health In India, doctors completed only 26 per- is going a step further, analyzing the extent cent of the tasks medically required for a to which even when providers are present, patient presenting with TB--the number one they may deliver suboptimal care. Drawing killer among infectious diseases in India--and on creative strategies--doctor and patient only 18 percent of recommended tasks for a interviews, direct observation of doctor- child with diarrhea (Das and Gertler 2007). patient interactions, and vignettes (or the use Doctors in Tanzania completed less than 24 of actors to simulate sample patient cases)-- percent of the essential checklist for a patient researchers are gathering direct estimates of with malaria (Leonard, Masatu, and Vialou provider quality through key process mea- 2005). In the face of major global efforts to sures. The implications for attainment of curb HIV/AIDS, TB, and malaria, these data the health MDGs are substantial. If health are troubling: the impact of substantially provider availability does not guarantee increased funding will clearly be blunted if adequate care, strategies to achieve better clinicians in the target countries cannot diag- health outcomes by training and recruiting nose patients who suffer from these illnesses more workers will fail. But if effective strate- or effectively treat them. gies exist for getting more performance out of The important finding for policy makers is existing providers, availability could greatly that large gaps can exist between what pro- accelerate progress. viders know and what they do. On question- Recent studies have explored these issues naires about their standard practice, many in five countries with very different levels of doctors knew the correct diagnostic proto- cols to follow for different patient condi- The quality of health care is not just a function of tions. But in actual practice--particularly for doctors' training patients seen later in the day--adherence to (percent of recommended diagnostic protocol followed by protocols was low. In India and Tanzania, the different types of doctors in India) biggest gaps were found in the public sector. While competency levels were often higher among public sector doctors (all doctors in this sector are medical board qualified), their "effort" was sharply lower than in the private or NGO sector (Das and Hammer 2004). As figure 2.14 shows, in India the increased value of competence in the public sector was com- pletely offset by decreased effort. Although private doctors lacking the Indian medical board certification are significantly less well- trained than doctors in the public sector, the quality of care they delivered to patients was significantly better. These findings offer two potentially impor- tant policy implications. First, they suggest that trying to improve health care delivery by raising training requirements, expanding Source: Das and Hammer 2004. medical schools, or other input-based poli- cies countries often rely on may be an ineffi- cient route to better health care performance. Second, it suggests that countries concerned incentives" for performance into public about a "brain drain" of their trained phy- health systems. One approach is to con- sicians to OECD markets might be able to tract private providers for the delivery of reduce those risks by setting national train- public health services. While the first and ing requirements slightly lower than the rich most rigorously evaluated such experience countries' standards. These data suggest that is Cambodia (see box 2.8), contracting has there might be very little trade-off in terms of also been tried on a fairly large scale in nine delivered health quality. other countries. In a recent review of the Institutional setting--along with incen- experience, Loevinsohn and Harding (2005) tives--makes a large difference for the qual- conclude that contracting for primary health ity of service delivered. In Tanzania, clinicians care services has been effective and can pro- working in facilities with "high-powered duce rapid improvements. Positive results incentives"--measured by factors such as have been achieved in a wide range of coun- the ability of facility managers to hire or fire try settings and for a range of different ser- personnel, set salaries, determine the number vices, from nutrition in Africa to primary and types of staff who work for them, and the health care in Guatemala. In most places degree of financial independence at the facil- where it has been tried, governments have ity level--were much more likely to perform subsequently decided to scale-up contracting up to their ability than equivalently trained more broadly. The biggest impacts generally and paid doctors in less autonomous facilities have been for services that are relatively easy (Leonard, Masatu, and Vialou 2005). to deliver such as immunizations, vitamin A Contracting for results. An increasing supplementation, and prenatal care. Services number of developing countries are experi- such as family planning and assisted deliv- menting with ways to inject "high-powered eries have shown smaller changes, likely As part of a major effort to expand and improve health service delivery, Cambodia since 1999 has contracted out to private bidders the management of government health services in five rural dis- tricts. In each district, contractors were required to provide a "Minimum Package" of preventive care, health promotion, and simple curative services. Contractors were also responsible for services at district hospitals, subdistrict health centers, and rural health posts. Performance was measured against eight service delivery indicators. Contracts were for a fixed period and inadequate perfor- mance could lead to sanctions and nonrenewal.The five districts covered 1.2 million people, 11 percent of Cambodia's population. All of the contractors were international NGOs. Since the districts were randomly selected for the program, it has been possible to evaluate its impacts rigorously--in other words, to determine with confidence that the health delivery improve- ments seen in these districts were actually caused by the program.a Some of the effects have been large: immunization coverage increased by 21 percentage points, vitamin supplementation to chil- dren under age five increased by 42 percentage points, and antenatal care delivered to pregnant women increased by 36 percentage points. Contracting improved the management of government health centers, as measured by availability of 24-hour service, reduced staff absenteeism, increased supervisory visits, and continuous availability of supplies and equipment. It also led to lower out- of-pocket spending by patients, who shifted back to using public health facilities instead of private sector drug sellers and traditional healers. As the districts eligible for the experiment were in poor regions, this is an important equity effect. While measuring the impact on final health outcomes such as child mortality would require a larger-sized experiment, the researchers could conclude that health status in the five districts improved as a result of the program: both the average number of reported illnesses in a typical month and the incidence of diarrhea in children under five were reduced. A cautionary note from Cambodia's experience is that even though some of the program's impacts were genuinely large, they were less than would have been estimated through a simple "before and after" comparison of health service delivery in the districts implementing the reform. This is because over the period of the contracting experiment, there was a strong general improve- ment in health system performance. The careful manner in which the government rolled out this program allowed for accurate estimate of its causal impacts, and more confidence about what can be expected now that it is being scaled-up further. a. Bloom and others 2007. because these imply more behavioral change The cases to date provide suggestive evi- on the part of patients. dence that the most successful approaches base In all cases where contractor performance the contracts on specified outputs and out- was directly compared with government pro- comes, rather than inputs, and give contrac- vision of the same services, contractors were tors autonomy over how they use resources more effective in expanding coverage and to produce the contracted outcomes--includ- delivering quality care. In Hyderabad, India, ing the ability to offer differential pay to the for example, an NGO achieved a TB treat- public sector health workers they supervise ment completion rate that was 14 percent- and to hire and fire with greater flexibility. age points higher than the public clinic in a Contracting out services to private provid- nearby area, and at a lower cost (Murthy and ers is not the only formula for strengthening others 2001). performance incentives in health systems. Rwanda is experimenting with performance- growing concern about the quality of rap- based contracting for both public and NG0 idly expanding services, the "earmarking" of providers, using some of the same principles. development dollars to specific diseases and The system pays facilities (which in turn pay goals, and the transactions and coordination individual providers) bonus payments for costs associated with proliferating funding incremental improvements in basic health channels. But actions in each area could miti- services and HIV/AIDS testing and treatment. gate some of the problems and enhance the Facilities have autonomy over how funds are positive trends. Some of the most important used, including topping up staff salaries and are summarized below: freedom to raise outside sources of income, such as from user fees. Strong monitoring and Intensified Focus on Learning. The inter- auditing arrangements (including periodic national community could strengthen the surveys to track patient satisfaction) allow incentives for developing countries to keep for verifying the quality of care and making focused on student learning and school payments conditional on quality. Argentina's quality as they pursue universal primary innovative Plan Nacer program is similarly completion. There is a public goods argu- using bonus payments to create incentives for ment for donors to support developing better quality maternal and child health ser- countries in defining a relevant set of basic vices no matter where they are provided--in literacy, numeracy, and critical thinking public, semi-public, or private facilities. skills that should be attained by the end It is encouraging that countries such as Cam- of primary school. International support bodia, Rwanda, and Argentina are not only could also help build appropriate national developing new strategies to improve quality and regional assessment systems to track in health, but also rolling out reforms in a care- student learning in developing countries ful manner that permits rigorous evaluation of and permit them to benchmark their prog- their impact. The creative new research expos- ress. ing the large wedge between doctors' ability Health system strengthening. There is an and their practice makes it clear that attaining urgent need for scaled-up and stronger the health MDGs in most developing countries health systems in developing countries, to will depend at least as much on getting better stretch absorptive capacity and avoid the performance from providers as on scaling up risks that large amounts of needed aid are their numbers. lost to mal-coordination and corruption. There is also need for appropriate incen- tives to improve outcomes, and actions to Conclusions ensure that levels of support are sustain- From the vantage point of 2007, the stretch able and not distortionary at the macro goals to promote human development prog- level. The new World Bank Health, Nutri- ress set by the international community in tion, and Population (HNP) sector strat- 2000 have made an appreciable difference. egy commits the Bank to a central focus They have put a significant number of devel- on health system strengthening, but it will oping countries on a faster trajectory to require internal staffing changes and other universal provision of basic health and edu- actions to develop this potential fully. cation services. They have demonstrated the Donor harmonization in health. Action commitment of the rich world to mobilizing is needed to curb the transactions costs increased aid for specific human development and coordination failures increasingly goals. And they have given rise to a large num- associated with verticalized and prolifer- ber of new global institutions and programs ating health programs. The way forward to support these processes. Not all of these should be country-led processes that set developments are unalloyed goods; there is clear national priorities in health and insist on more rapid progress toward the Paris element in that they benefit all countries, Harmonization targets: donor alignment there is a strong case for increased donor with national plans, joint supervision mis- support. sions, use of national systems, and other goals. Monitoring results. For too many MDG targets and in too many countries, lack of data makes it impossible to track progress. 1. There is no official estimate for the end of There are several sources of support for sta- 2006, but Peter Piot of UNAIDS expected the tistical strengthening in developing coun- number would "probably approach 2 million" by tries, including new initiatives since the year end (remarks on November 27, 2006, Wash- MDGs were adopted, such as the Health ington DC). 2. Because the World Bank uses an exponen- Metrics Network, but collectively these tial, rather than linear, method to model progress efforts are not enough. Expanded donor to MDG targets, these estimates of the number of support is needed for building countries' countries and regions on track to the goal differ ability to monitor trends in key areas. slightly from those of the UN Joint Monitoring Evaluating impact. The key to faster MDG Program. progress is basing policies, programs, and 3. In previous years, this report cited IMF pub- donor support on evidence of what works. lic expenditure data for health and education in Too few innovative programs in the devel- developing countries. In 2005, the Fund stopped oping world are rigorously evaluated collecting sector-level expenditure data, so GMR today, and those that are often show that 2007 relies on data from UNESCO for education, program impacts in reality are lower than and on WHO for health spending. Both series show numerous inconsistencies with IMF data advocates predicted. This is not bad news; (which we have reported previously), with partic- it is the reality that explains why, in aggre- ularly large inconsistencies for education. Country gate, "aid effectiveness" is not higher. coverage of UNESCO's spending data is also much Only rigorous evaluations--which estab- weaker than WHO's. For 2004, UNESCO's devel- lish that the program caused the observed oping country coverage was extremely low, so the results--can build a solid base for policy terminal years presented above for education and and program design. Because such evalua- health are different. tions are expensive and have a public good Promoting Gender Equality and Women's Empowerment T he 2006 World Development Report indicators only partially capture the elements acknowledges the importance of of gender equality, the chapter introduces five ensuring equal opportunities across complementary indicators that provide a more population groups as an intrinsic aspect of complete and nuanced description of gender development and as an instrument for achiev- equality and women's empowerment. The ing poverty reduction and growth (World indicators are measurable, actionable, and par- Bank 2005). Noting that men and women have simonious; three of the five build on existing starkly different access to assets and opportu- measures of other MDGs, so the data require- nities in many countries around the world, the ments for monitoring them are not onerous. report refers to gender inequality as the arche- Further, some of these complementary indica- typal "inequality trap," reproducing further tors (or similar measures) are being considered inequalities with negative consequences for for inclusion in the MDGs as part of new tar- women's well-being, their families, and their gets for decent and productive work and for communities. MDG3 reflects the strong belief reproductive health services. Finally, the chap- by the development community that redress- ter extracts preliminary lessons from coun- ing gender disparities and empowering women tries that have achieved high levels of--or fast is an important development objective on progress toward--gender equality, but does grounds of both fairness and efficiency.1 not undertake a systematic analysis of poli- This chapter reviews the evidence on the cies. Countries that perform well on MDG3 relationship between gender equality, pov- illustrate that investments in equality in rights, erty reduction (MDG1), and growth. There resources, and voice can make a difference. is also compelling evidence that gender equal- Thanks to the push to achieve universal ity and women's empowerment are channels primary education with gender-informed to attaining other MDGs--universal primary education policies, girls' enrollments at all education (MDG2), lower under-five mor- levels of schooling have increased, and sev- tality (MDG4), improved maternal health eral countries have achieved gender parity in (MDG5), and lower likelihood of contract- primary enrollments. Their success story sug- ing HIV/AIDS (MDG6).2 gests that concerted action can foster progress The chapter also tracks progress of coun- in gender equality not only in education but tries toward meeting MDG3 since 1990, using also in the economy and the society, where the official MDG3 indicators. Because these advances have been more modest. Success in boosting girls' enrollment may ment Report 2006, gender equality means offer important lessons for the unfinished equal access to the "opportunities that allow agenda in education and the largely unad- people to pursue a life of their own choos- dressed agenda in the other domains of gen- ing and to avoid extreme deprivations in out- der equality: comes"--that is, gender equality in rights, resources, and voice (World Bank 2001; Closing the gaps in well-being (health and World Bank 2005). Equality of rights refers education) and opportunities for girls and to equality under the law, whether custom- women in disadvantaged subgroups within ary or statutory. Equality of resources refers nations who face multiple exclusions on to equality of opportunity, including equality the basis of their sex and their race, resi- of access to human capital investments and dence, ethnicity, caste, and disability. It is other productive resources and to markets. also essential to monitor progress in gen- Equality of voice captures the ability to influ- der equality and women's empowerment ence and contribute to the political discourse for these subgroups. and the development process. Giving priority to improving and monitor- Figure 3.1 presents a framework that ties ing gender equality and women's empow- together key elements of gender equality. erment in Sub-Saharan Africa, which Gender inequality in rights, resources, and consistently lags behind in most areas voice can surface in three domains: in the measured by MDG3. household, in the economy and markets, and Paying special attention to gender equal- in society. In the household, evidence sug- ity and women's empowerment issues in gests that increased gender equality between fragile states where progress on MDG3 is men and women changes the allocation of hampered both by slow economic advance- household expenditures, resulting in a larger ment and gender-specific consequences of share of resources devoted to children's edu- conflict. cation and health. Gender inequalities influ- Scaling up significantly the collection and ence the distribution of household tasks, analysis of sex-disaggregated data to mea- often limiting women's ability to work out- sure more accurately and fully the progress side the home, as well as women's control in achieving MDG3. Data on all six offi- over fertility decisions. In the market, gen- cial indicators of MDG3 are available for der inequality is reflected in unequal access only 59 out of 154 developing countries to land, credit, and labor markets, and in (for 2000­05), and even fewer countries significantly less access to new production have time series data that would allow technologies. In society, gender inequality is tracking over time. For both the official expressed as restrictions to women's partici- and expanded list of indicators recom- pation in civic and political life. Finally, as mended in this chapter, only 41 countries figure 3.1 shows, in addition to improving have current (2000­05) information. This individuals' lives, increased gender equality lack of data limits considerably the ability can contribute to better aggregate economic to monitor progress, learn from success, performance. and, ultimately, to make informed decisions These long-term benefits, of course, come regarding scaling up investments. with costs in the short run. Policies to achieve gender equality (for example, introducing quotas in representation in parliament or labor legislation prohibiting discriminatory practices) could have political costs for their Gender equality does not necessarily mean proponents when some groups win and some equality of outcomes for males and females. lose. Some policies may also have economic Using the definition in the World Develop- costs that come from unintentionally under- Gender equality, domains of choice, and economic performance: A framework Source: World Bank staff. cutting individual incentives in the name of (GEM).3 Economic growth also appears to be gender equality. These costs are additional to positively correlated with gender equality. This the budgetary expenditures associated with correlation is robust to changes in the length implementing the policies. It is important to of the period over which per capita GDP keep these short-term trade-offs well in mind growth rates are averaged and to two alterna- in assessing specific policies. tive measures of gender equality (the female- to-male ratio of sex-specific HDI indices and the GDI-HDI ratio).4 When gender equality is measured by the GEM, however, the relation- ship is not statistically significant. Poverty incidence tends to be lower in coun- Simple correlations across countries-- tries with more gender equality. This relation- while suggestive--do not imply a causal ship is quite robust to various measures of relationship between gender equality and poverty and of gender equality--in terms of poverty reduction or economic growth: gen- the latter, the female-to-male ratio of sex-spe- der equality could "cause" faster growth cific Human Development Indices, the ratio and accelerated poverty reduction, but faster of the gender-related development index to development could also spur improvements the human development index (GDI-HDI in gender equality. Alternatively, the causal ratio), and the gender empowerment measure arrows may point in both directions, or a third factor may be responsible for both faster ated with poverty reduction or faster growth. development and greater improvements in There are several pathways through which gender equality--perhaps better governance. gender equality in rights, resources, and voice Regression analyses that control for other stimulate productivity, earnings, and better (possibly confounding) factors have also child development outcomes, thus generating been employed, although the estimated coef- better development outcomes in an economy. ficients by themselves do not imply causality. Figure 3.2 depicts the pathways of women's Cross-country growth regressions, building labor force participation and earnings (iden- on widely accepted macroeconomic growth tified by dashed arrows) and children's well- models, have examined the link between being (identified by solid arrows). greater equality in educational opportuni- ties and growth rates or levels of per capita Women's Labor Force Participation, income.5 Productivity, and Earnings Recent studies using cross-country regres- sions typically find that female education has Whether engaged in self-employment or wage a larger impact on growth than male educa- employment, working women contribute to tion (Abu-Ghaida and Klasen 2004).6 Klasen household income and expenditure. In poor (2002), for example, finds that the direct and households, such contributions can be cru- indirect effects of gender inequality in educa- cial for keeping the household out of poverty; tional attainment account for 38 percent of this is a reason to increase access to educa- the 2.5 percentage point gap in growth rates tion, markets (labor, land, credit), and tech- between South Asia and East Asia, 17 percent nology. This increased access can contribute of the 3.3 percentage point gap between Sub- to current poverty reduction and economic Saharan Africa and East Asia, and 45 percent growth through higher consumption and to of the 1.9 percentage point gap between the future poverty reduction through the impact Middle East and North Africa and East Asia.7 on children's accumulation of human capital Growth regressions have serious limita- and the potential impact on aggregate saving tions, however, and those that use gender-dis- (the dashed arrows in figure 3.2).8 aggregated data are no exception. One serious Women face many constraints at home and limitation is the ad hoc nature of extensions in the marketplace when they decide to seek to the augmented Solow model, which under- paid employment. Numerous studies point to lies growth regressions. Variables have been women's reproductive role as affecting female added to capture economic openness, gov- labor force participation, in general, and work ernment spending, political instability, eth- for pay, in particular. In the Kyrgyz Repub- nic diversity, and a host of other potential lic, for example but not atypically, 24.8 per- determinants of growth--frequently with cent of women reported that "housekeeping, little or no justification in economic theory. taking care of children, sick persons, or the A second weakness is a simultaneity problem elderly" kept them from working outside the that results in biased results: gender equality home, but only 1.5 percent of men reported affects growth, but growth presumably also these reasons (Morrison and Lamana 2006). affects gender equality, because the economic Besides child care, women also face the time pressure in rapidly growing markets makes burden of domestic tasks, especially collecting gender discrimination much more costly. water and firewood. In rural areas of Burkina Finding appropriate identification factors to Faso, Uganda, and Zambia the potential time address this bias is extremely difficult, which savings from locating a potable water source leads to a search for other evidence. within 400 meters of all households range Cross-country correlations and growth from 125 hours per household per year to regressions can be suggestive, but they do not 664 (Barwell 1996)--time that could be used explain how gender equality might be associ- to work for pay. Women's earnings, children's well-being, and aggregate poverty reduction and economic growth--The pathways Source: World Bank staff. Wage gaps and discrimination against be lower than those of self-employed men. women in labor markets may lower labor These lower potential earnings may discour- force participation, both contemporaneously age women from entering self-employment. and for future generations. The contempo- For households dependent on agriculture, raneous effect occurs as the wage loss due land is the most important productive asset. to discrimination persuades some women The limited evidence available, however, indi- to stay at home rather than engage in paid cates that the distribution of land ownership is work. The wage loss due to discrimination heavily skewed toward men. For example, in a will also cause parents to systematically set of Latin American countries, roughly 70­90 underinvest in the education of girls relative percent of formal owners of farmland are men to boys (see Anderson and others 2003 for (Deere and Leon 2003). When women do own evidence on Malaysia). The segregation of farmland, their holdings are typically smaller women into low-paying occupations may be than men's.9 Similar evidence is found for another important driver of underinvestment Sub-Saharan Africa (Doss 2005; Udry 1996; in girls' education. But empirical documenta- Quisumbing and others 2004). The evidence tion of the impact of these two disincentives also clearly points to the importance of access to female labor for participation has been to land and land size for increasing income.10 limited, and further research is needed. When they do have access to land, women For self-employment, imperfections and frequently have less secure tenure rights. For discrimination in other markets constitute example, under customary law in much of barriers for women. If access to inputs such as Sub-Saharan Africa, permanent land rights land, credit, capital, and technology is limited are held by men, typically male household for noneconomic reasons, women's produc- heads. In contrast, women traditionally held tivity and earnings in self-employment will (strong) usufruct rights to individual plots offered by men (Lastarria-Cornheil 1997). significantly lower endowments of land, and Land redistribution reforms and land titling that land size is a significant positive determi- and registration programs have, in many nant of fertilizer use. cases, either maintained the rights status quo or weakened women's rights (Jacobs 2002; Children's Well-Being Agarwal 1994; Agarwal 1993; Lastarria- Cornheil 1997).11 Women's education, health, and greater Most studies find that women are not more control over household resource allocation likely to be rejected for loans or be subject improve children's well-being (figure 3.2, solid to higher interest rates by lenders, but they arrows). Studies from developing and devel- are often less likely to apply for loans than oped countries consistently show that when men, partly because they do not have land, mothers have greater control over resources, property, or other assets to offer as collateral more resources are allocated to food and to (Baydas and others 1994; Storey 2004; Ratusi children's health (including nutrition) and edu- and Swamy 1999; Buvinic and Berger 1990). cation. In Ghana, an increase in the share of As a result, in both Malawi and Bangladesh, women's assets raises household spending on women are more likely than men to face con- food and children's schooling (Doss 1996). straints to credit, as measured by credit lim- Similarly, in Côte d'Ivoire, the higher women's its (the maximum amount individuals report share of cash income, the higher is the house- they can borrow from various sources) and hold budget share allocated to food (Hoddinott unused credit lines (the difference between the and Haddad 1995). In Ghana, in years when credit limit and amounts borrowed) (Diagne the production of women's crops is higher, the and others 2000). But when women are the household spends a large share of its budget on direct beneficiaries of credit rather than men, food and on private goods for women; in years the impact of credit on various measures of when the production of men's crops is higher, household welfare is greater. however, the household spends more on goods Technological innovation and adop- consumed by men (Duflo and Udry 2004). tion have undoubtedly been key drivers of Better nutritional status of mothers is increases in productivity and household associated with better child health. In Brazil incomes. Most empirical studies of the deter- maternal height has a large impact on infants' minants of technology adoption and diffusion height (length) while paternal height has no in developing countries examine the adoption impact (Thomas and Strauss 1992). One way decision at the level of the household. They this happens is that mothers who are under- have not examined how female farmers fare weight or who suffer from micronutrient relative to male farmers in terms of school- deficiency before pregnancy are more likely ing and literacy (considered to be critical for to give birth to low-birth-weight infants (Gal- processing relevant new information); access loway and Anderson 1994). to information (through social networks and The benefits of mother's education for agricultural extension services); access to children are well known; they flow through credit, labor, and commodity markets; risk several pathways: exposure and risk aversion; and land size and land rights.12 Most of the evidence suggests Safer health and hygiene practices, which that many of the barriers to adoption are not improve children's health (Cebu Study related to the characteristics of the technol- Team 1991). ogy, but originate in other markets relevant More time and resources for children's for the adoption decision, such as land, health and education (Brown 2006). labor, credit, and information. For example, More exposure to information from a wider Croppenstedt and others (2003) find that range of sources, and higher ability to pro- female-headed households in Ethiopia have cess and act on the information received (Webb and Block 2004; Thomas, Strauss, allocated to children's schooling, clothing, and Henriques 1991; Caldwell 1979). and protein-rich foods.13 Better nutritional outcomes, in part Given the evidence linking increases in because of higher ability to process and act women's productivity and earnings to lower on information (Glewwe, Jacoby, and King household poverty and better outcomes for 2001; Alderman, Hoddinott, and Kinsey children, it is clear that barriers to women's 2006; Behrman and Rosenzweig 2004). labor force participation, productivity, and Fewer children, reducing household depen- earnings also constrain poverty reduction. dency ratios and increasing per-capita Given the weaker evidence linking increased consumption expenditure (Schultz 1997, women's productivity and incomes to faster 2002). growth, it is less certain that removing those Higher labor force participation and earn- barriers will generate a growth dividend. Nev- ings, which in turn increase household ertheless, it is difficult to imagine that higher consumption expenditures (see review in female labor force participation and earnings Schultz 2002). would not lead to higher levels of total output Greater bargaining power within the and per capita output. Similarly, increases in household and therefore a higher ability to productivity and earnings in health and educa- act on preferences for investing in children tion, and in control over resources, lead both to (World Bank 2001). better child development outcomes in the pres- ent and to an intergenerational transmission of Several studies have estimated the welfare earnings capability that improves the prospects effects of participation in programs where for future poverty reduction and growth. women are the main direct beneficiaries. In their study of microcredit programs in Bangladesh, and after controlling for self- selection in program participation, Pitt and Khandker (1998) find that female borrow- ing has a larger impact on children's school Progress toward attaining MDG3 is measured enrollment than male borrowing. Exploiting by the target and four indicators defined in a natural experiment in which black fami- the Millennium Declaration (table 3.1). The lies became eligible for large old-age pension target is "the elimination of gender disparities payments in South Africa, Duflo (2003) finds in primary and secondary education, prefer- that girls who live with a grandmother who is ably by 2005, and at all levels of education eligible to receive pension benefits are health- no later than 2015." The target and the first ier (measured by anthropometric measures) two indicators measure progress in gender than those who live with a grandmother who equality in the household, the third measures is not eligible to receive pension benefits; in progress in the economy generally, and the contrast, the effects were not statistically fourth measures progress in society. They significant for households in which the pen- provide important, albeit incomplete, mea- sion was received by a man. Comparing the sures of achievements (since 1990) in gender marginal effect on household expenditure equality for the three domains identified in patterns of transfers received by mothers the framework in figure 3.1. through Oportunidades--a conditional cash transfer program in Mexico initially imple- Progress in the Household Domain: mented as a randomized social experiment-- School Enrollment and Literacy with the marginal effect of other sources of household income, Rubalcava and others Through concerted efforts by government, (2004) find that cash transfers from the pro- civil society, and the development community, gram increased the household budget shares girls' enrollments in all levels of schooling rose significantly in the last decade. Indeed, Gains in girls' secondary school enrollment thanks to these efforts, most low-income were notable in East Asia and the Pacific, countries made substantial progress during Latin America and the Caribbean, and Middle the 1990s in reaching gender parity in pri- East and North Africa. This progress is quite mary school enrollments and literacy (Lewis remarkable and shows the responsiveness of and Lockheed 2006; EFA Global Monitoring girls' school enrollments to gender-informed Report 2007). Between 1990 and 2005, girls' policy interventions such as stipends, condi- enrollment in primary education increased in tional cash transfers, and vouchers. virtually all regions; the sole exception was These improvements in girls' enrollment East Asia and the Pacific, where girls' gross helped increase gender parity ratios. Figure enrollment rate already exceeded 100 percent 3.4 shows that between 1990 and 2005, there in the early 1990s (figure 3.3). Girls' enroll- were notable improvements in gender parity ment in secondary school increased as well. ratios in enrollments at all levels of schooling, Official indicators for MDG3 Household Economy and market Society Ratio of girls' to boys' enrollment in Share of women in wage Proportion of seats held by primary, secondary, and tertiary employment in the nonagricultural women in national parliaments educationa sector Ratio of literate females to males among 15­24-year-olds Source: United Nations 2003. a. Measured using gross enrollment rates. Progress in girls' enrollment rates between 1990 and 2005 Source: World Development Indicators 2006. Trends in gender parity in enrollment and literacy rates, 1990 and 2005 Source: World Bank Indicators (top); World Development Indicators 2006 (bottom). Note: The regional averages are calculated using the earliest value between 1990 and 1995 and the latest value between 2000 and 2005 for each country. The averages are weighted by the country population size in 2005. In the second figure, trend is shown for countries that were fragile states in the 2000­05 period. For fragile states, data are available for 25 countries for primary enrollment, 22 for secondary enrollment, 8 for tertiary enrollment, and 13 for literacy. For nonfragile comparator countries, corresponding sample sizes are 36, 31, 21, and 25. particularly in regions that had large dispari- in South Asia failed to meet the target because ties at the beginning of the period. of low gender parity in secondary school enroll- Combining performance in primary and ment. In South Asia, Bangladesh and Sri Lanka secondary enrollments, by 2005, 83 devel- are notable for achieving parity. oping countries (out of 106 with data) had In 2005 the female tertiary enrollment rate met the intermediate MDG3 target of parity lagged behind the male rate in 63 countries (of in primary and secondary enrollment rates 130 countries with data) and exceeded the male (table 3.2).14 Most of these countries are in rate in 65 countries. Female disadvantage was regions where enrollment has historically evident mainly in Sub-Saharan Africa, South been high--East Asia and the Pacific, Europe Asia, and in fragile states. Male disadvantage and Central Asia, and Latin America and the was notable in Middle East and North Africa Caribbean. In the Middle East and North (Algeria, Iran, Jordan, and Libya), East Asia Africa, most countries met the target by 2005, and the Pacific (the Philippines and Thailand), but this region also included 3 countries (out Latin America and the Caribbean (Honduras, of 11 with data) with significant female dis- Nicaragua, Panama), and Europe and Cen- advantages in enrollment. In Sub-Saharan tral Asia. Reflecting the legacy of the Soviet Africa, less than one-quarter of all countries Union and historically high enrollment rates met the target by 2005. in Europe and Central Asia, countries there Among 14 fragile states with data, five had high female tertiary enrollment rates that countries met the target by 2005. Poor data exceeded male enrollment rates. availability for this group of countries makes The education system's ability to deliver it difficult to accurately compare their prog- basic literacy skills and progress in school ress with that of nonfragile states. However, enrollments over the years has resulted in an analysis of averages for countries that have higher literacy rates and greater gender parity data for the two periods shows that, as com- among youth (ages 15­24). But gender gaps pared to nonfragile states, fragile states made remain: UNESCO estimates that of the nearly only modest progress in moving toward gen- 137 million illiterate youths in the world, 63 der parity in enrollments (figure 3.4). percent were female (UNESCO, EFA Global Despite significant improvements in girls' Monitoring Report, 2005). The female-to- primary school enrollment, half of all countries male literacy ratio was lowest in Sub-Saharan Regional performance in attaining the primary and secondary enrollment target by 2005 Off track or On track to unlikely to Achieved target achieve target achieve target by 2005 by 2015 by 2015 No data Total Sub-Saharan Africa 10 1 16 21 48 East Asia and the Pacific 13 0 0 11 24 Europe and Central Asia 22 0 1 4 27 Latin America and the Caribbean 27 0 0 4 31 Middle East and North Africa 8 0 3 3 14 South Asia 3 0 2 3 8 Total 83 1 22 46 152 of which: Fragile states 5 0 9 21 35 Source: World Bank estimates using data on enrollments between early 1990s and 2004/2005. Note: The column showing countries with no data indicates the number of countries with missing data either at the start of the period or at the end of the period or both. Of the 49 non-fragile low-income countries, 25 had met the target by 2005, 1 was on track to meet the target by 2015, 9 were unlikely to meet the target, and 14 had no data. Africa, Middle East and North Africa, and It has been estimated that of the 60 million South Asia--regions that also had female dis- girls not in primary school in 2002, 70 per- advantages in primary and secondary enroll- cent were from excluded groups (UNESCO ment (figure 3.4). In 25 countries in these Institute for Statistics 2005). In Bolivia, regions, there were fewer than 80 literate young household survey data show gender gaps in women for every 100 literate young men. The school attainment among indigenous chil- ratio was lowest in Yemen and Afghanistan, dren but not among non-indigenous children where only 36 young women were literate for (figure 3.6). Although boys and girls have every 100 literate young men. similar profiles at ages 7­13, attainment rates for indigenous girls already start to decline at age 9, with a faster decline after age 13 (Dur- The Unfinished Education Agenda yea, Galiani, Nopo, and Piras 2006). Even Despite the considerable success in increas- in countries that have attained the gender ing girls' enrollment and improving gender parity target in education, girls' enrollment parity ratios at all levels of schooling, several in some subgroups continues to lag behind challenges remain. that of boys' (for example, in Chile, Mexico, Fragile states and countries unlikely and Panama). Gender parity ratios in school to attain the enrollment target. For the 23 enrollments and literacy thus need to be dis- countries that did not meet the enrollment aggregated by characteristics related to disad- target by 2005, the World Bank estimates, vantage and exclusion, and targeted policies based on the rate of change in the ratios over are needed to increase the school enrollments the 1990s, that 22 countries are unlikely to of girls from disadvantaged groups. achieve the target by 2015; 16 of these coun- Levels not just ratios. A third challenge is tries are in Sub-Saharan Africa (table 3.2). that the ratio of girls to boys in enrollment This list includes 9 low-income countries such is silent on levels of enrollment. Although as Benin and Burkina Faso, where improve- all regions increased their secondary school ments in gender parity ratios in the 1990s enrollments since 1990, no region showed might not compensate for large pre-existing universal enrollment in secondary education gender disparities. The list also includes 9 in 2005 (figure 3.3). Only four regions had fragile states (all in Sub-Saharan Africa). two-thirds or more of their eligible popula- Disadvantaged and excluded groups. A tion in secondary school in 2005: Europe and second challenge is uncovered when average Central Asia with 86 percent, Latin America national gender parity ratios are disaggregated and the Caribbean with 70 percent, Middle by income, location, race, ethnicity, disability, East and North Africa with 66 percent, and or other features that identify disadvantaged East Asia and Pacific with 64 percent.15 In or socially excluded populations within a both South Asia and Sub-Saharan Africa, country. Large gender gaps in education and fewer than half of secondary-age students literacy in rural areas, among minority groups were enrolled (48 percent and 32 percent, or lower-income quintiles may explain why respectively). Efforts to increase school some countries have not reached the gender enrollments, especially in secondary educa- parity target (Bolivia, Cambodia, Ecuador, tion, need to be informed by gender equality Guatemala, Lao PDR, Morocco, and Paki- concerns and scaled up in regions with lag- stan). For education and literacy, the female ging enrollment rates (see box 3.1). disadvantage is always larger in rural areas Transitions from one level to the next and and among lower-income households (figure from school to work. Promoting equality in 3.5). This is further accentuated in countries education opportunities involves entering the that have not reached overall gender parity education system at primary level, progress- in school enrollments (largely in Sub-Saharan ing through to higher levels, and making the Africa and Middle East and North Africa). transition to the labor market. But the offi- Average youth literacy rates in Africa conceal rural-urban disparities Source: World Bank staff estimates using household survey data from 24 Sub-Saharan African countries. Four countries labeled for illustrative purposes. Bolivia has a gender gap in schooling among indigenous children Source: Duryea, Galiani, Nopo, and Piras 2006. cial MDG3 indicators measure only parity in get). Women's share of nonagricultural wage school participation at each level. They do not employment increased in all regions in 1990­ capture potential gender disparities in transi- 2005; this increase was modest, however, with tion from one level of schooling to the next, in significant variation across regions and coun- the quality of what is learned in school, and in tries (figure 3.7). In 2005 the share of women the transition from school to work. in nonagricultural employment was highest in Europe and Central Asia (47 percent) and lowest in Middle East and North Africa (20 Progress in the Economy and in Society: percent). In Latin America and the Caribbean Nonagricultural Wage Employment and and East Asia and Pacific, it exceeded 40 per- Political Participation cent. Women's share of nonagricultural wage The MDG3 indicator in the economy and employment was highest in the highly urban- market is the share of women in nonagri- ized upper-middle-income countries (43 per- cultural wage employment (with no set tar- cent) and lowest in the still predominantly Gender parity ratios say nothing about absolute levels of enrollment. A ratio of one (perfect equal- ity) may indicate "equality of deprivation" rather than equality of opportunity. In Haiti the parity ratio in secondary enrollment rates was 1.03 in 2003, but only 20 percent of both girls and boys in Haiti were enrolled in secondary school (box figure). In such an environment, the challenge is to boost enrollments while maintaining gender parity. Parity ratios can be at high and low secondary enrollments Source: World Development Indicators 2006. Progress in share of women in nonagricultural wage employment and proportion of seats in parliament held by women, by region Source: World Development Indicators 2006. The regional averages are calculated using the earliest value between 1990 and 1995 and the latest value between 2000 and 2005 for each country. The averages are weighted by the country population size in 2005. rural low-income countries (30 percent). In Like the education indicators, the average 15 countries, mostly in Europe and Central share in nonagricultural wage employment Asia, women dominated nonagricultural also conceals inequalities within countries. In wage work. Women also dominated this work several countries of Latin America, indigenous in Cambodia, Honduras, and Vietnam-- and Afro-descendent women, who have sig- countries where recent growth in export- nificantly fewer years of education than other oriented manufacturing industries increased women, are also less likely to be employed the demand for female workers. For 20 in nonagricultural paid employment (figure countries in Sub-Saharan Africa, South Asia, 3.8). For example, nearly 60 percent of all and Middle East and North Africa, women's women engaged in nonagricultural paid work share was below 20 percent. in Bolivia in 2002 were nonindigenous, a per- Trends and patterns in this indicator are dif- centage that far exceeds the population share ficult to interpret without taking into account of nonindigenous women. Duryea and Genoni the circumstances in each country--such as the (2004) find that in Bolivia, Brazil, Guatemala, share of nonagricultural employment as a per- and Peru, indigenous and Afro-descendant centage of total employment. A favorable score women are overrepresented in low-paying and on this indicator might, on the surface, seem informal jobs. to indicate equitable conditions for women in The fourth official MDG3 indicator is the labor markets, but it may capture conditions proportion of seats held by women in national for only a very small proportion of the total parliaments (with no set target). Between labor force (see the following section for a dis- 1990 and 2005 all regions except Europe and cussion of the limitations of this indicator).16 Central Asia saw an increase in the propor- tion of women's seats in national parliament, Share of women in nonagricultural wage work by but starting from low levels (figure 3.7). How- ethnicity ever, in no region did the average proportion exceed 25 percent, at either the beginning of the period or the end. Quotas to increase women's presence in parliament (candidate quotas and reserved seats) were adopted by a large number of countries during the 1990s. By 2005, more than 40 countries had introduced elec- toral quotas. Because of quotas, countries like Argentina, Costa Rica, Mozambique, Rwanda, and South Africa have reached levels of women's parliamentary representa- tion comparable to those in Nordic countries (Ballington and Karam 2005). However, quota rules are not sufficient by themselves to ensure increased participation by women; implementation and enforcement are key. In addition to quotas at the national level, increasing women's opportunities to partici- pate in local politics can increase the number Source: World Bank staff calculations using household survey data from Guatemala (2002), Bolivia (2002), and Brazil (2001). of women who are able to participate at the national level. regions); wage gaps (relative to men); limited access to complementary inputs such as credit, capital, and technology; and the impact of law and custom on women's ability to work outside the home (Morrison, Raju, and Sinha While the official indicators for MDG3 cover 2007). And political participation is captured the three domains of society, economy/mar- only at the national level, not at provincial or kets, and the household, they suffer from local levels where access to women's decision four serious shortcomings. First, the official making is also important. MDG3 indicators only partially capture gen- Second, the official indicators do not mon- der equality and empowerment in the areas itor key elements of gender equality such as they are designed to measure: education, health outcomes and disparities in access to employment, and political participation. productive resources such as land, credit, Education enrollment rates say nothing about and technology. Health outcomes are a par- equality in learning or educational outcomes. ticularly important determinant of well-being The share of women in nonagricultural wage and productivity. Although indicators for employment is of limited relevance for low- other MDGs measure performance on health income countries where wage employment is (MDGs 4, 5, and 6), they are not designed to not a main source of jobs. It does not capture monitor progress on gender equity in health many dimensions of job quality (Grown and status. others 2005), nor does it quantify the serious Third, while MDG3 refers to the promo- barriers that may inhibit women from par- tion of both gender equality and women's ticipating in labor markets: time burdens of empowerment, the official indicators are domestic tasks; limited availability of child far better at measuring gender equality than care; lower educational attainment (in some empowerment. Gender equality is a measure of the rights, resources, and voice enjoyed eracy are undoubtedly welcome, falling rates by women relative to those enjoyed by men. of male enrollment or literacy are not. Three of the official four indicators (ratio of Finally, national-level indicators--whether girls' to boys' enrollment rates, ratio of liter- parity ratios or absolute levels--can mask ate females to males among 24-year-olds, and inequalities between groups.20 Improve- share of women in wage employment in the ments in aggregate enrollment ratios, for nonagricultural sector) measure the status of example, may hide the fact that girls (or women relative to men, rather than whether in some cases, boys) belonging to socially women are empowered in an absolute excluded groups in the population fare much sense--that is, whether they have the abil- less well. Thus it is critical to disaggregate ity to exercise options, choice, control, and indicators by characteristics related to dis- power.17 Knowing, for example, that girls advantage and exclusion to monitor coun- are equally likely to be enrolled in second- tries' performance--and to develop targeted ary schools as boys indicates gender equal- interventions not just at the national level, ity but not necessarily empowerment if only but also for particular subgroups. a small percentage of girls are enrolled (box 3.1).18 Important elements of empowerment A Proposal for Strengthening the Official not captured by the official MDG3 indicators Indicators include the ability of women to work for pay (economic empowerment) and the ability to The shortcomings of the official indicators control their own fertility.19 for monitoring progress in attaining MDG3 In addition to being poor measures of are widely recognized (see, for example, the empowerment, changes in the indicators report of the UN Millennium Project Task based on parity ratios are difficult to interpret Force on Education and Gender Equality). (Grown 2006). Increases in female-to-male In response, this chapter recommends that ratios can result from a fall in male rates with countries consider monitoring five additional female rates remaining constant, or from a indicators complementary to the official MDG decline in both female and male rates with indicators, to better measure gender equality male rates declining faster, or from female (table 3.3). These indicators meet three crite- rates increasing faster than male rates. While ria: data availability (wide country coverage), rising female rates of school enrollment or lit- strong link to poverty reduction and growth, Recommended additional indicators for MDG3 Household Economy and markets Primary completion rate of girls and Percentage of 15- to 19-year-old girls Labor force participation rates boys (MDG 2)a who are mothers or pregnant with among women and men aged their first childb 20­24 and 25­49b Under-five mortality rate for girls and boys (MDG4) Percentage of reproductive-age women, and their sexual partners, using modern contraceptives (MDG6) Source: World Bank staff. a. Recommended by UN Millennium Project Task Force on Education and Gender Equality. b. Under consideration by Inter-Agency and Expert Group for MDGs. and amenability to policy intervention. Indi- than boys to complete the first schooling cycle, cators that met all three criteria but were particularly in South Asia, where the primary highly correlated with other indicators were completion rate is estimated at 90 percent dropped from the list.21 for boys and 83 percent for girls; and in Sub- This proposed list draws on the recom- Saharan Africa, the primary school comple- mendations of the UN Millennium Project tion rate is 67 percent for boys and 57 percent Task Force, but is more parsimonious in its for girls (table 3.4). Almost all regions made recommendations--both because it explicitly significant progress in raising girls' primary takes into account data availability and col- school completion rates between 1991 and linearity issues, and because of the high costs 2004. Sub-Saharan Africa, South Asia, and associated with imposing additional moni- Middle East and North Africa had the largest toring burdens on already taxed national sta- percentage increases (about 21 percent), fol- tistical offices.22 lowed by Latin America and the Caribbean Four of the five indicators monitor gender (14.4 percent). Gaps between girls' and boys' equality in the household; the remaining indi- completion rates, however, remain significant cator monitors gender equality in the economy. in Sub-Saharan Africa and South Asia. No additional indicators are recommended to monitor gender equality in society, because M O R T A L I T Y A N D M O R B I D I T Y none of the indicators considered for inclusion Under-five mortality. As mentioned above, meet the criteria of data availability. Three of health is one of the elements of gender equal- the recommended indicators are modifications ity that is not adequately covered by the offi- of official indicators already being monitored cial MDG3 indicators. A low-cost step toward as part of the MDGs. remedying this is for countries to disaggregate the MDG4 indicator of under-five mortality Additional Indicators for the Household by sex. Rates of under-five mortality are typi- Domain cally higher for boys than for girls (because of biological differences between the sexes) in E D U C A T I O N countries where there is no significant discrimi- Primary school completion rates. As men- nation against girls. For example, in four coun- tioned above, the official MDG3 indicator tries considered to be characterized by high of school enrollment ratios is a far better levels of gender equality (Denmark, Finland, measure of gender equality than of women's Norway, and Sweden) the girl-to-boy ratio is empowerment; it does not indicate whether enrolled students go on to complete primary school--the outcome that brings immense Girls lag behind boys in primary school completion rates benefits for development. Thus this chap- in most regions ter recommends supplementing the official MDG3 indicators with primary school com- Girls Boys pletion rates for girls and boys.23 This indica- tor (without sex disaggregation) is currently 1991 2004 1991 2004 monitored as part of MDG2 on attainment of Sub-Saharan Africa 47.1 56.9 62.3 67.3 universal primary education, and sex-disag- East Asia and Pacific 92.3 96.3 92.3 95.8 gregated data are reported annually by the Europe and Central Asia 92.9 92.6 94.3 96.5 Latin America and the Caribbean 88.4 101.1 83.0 99.4 World Bank and UNESCO. Middle East and North Africa 73.3 89.0 87.8 92.9 Primary school completion is measured as South Asia 68.3 83.0 90.4 90.2 the number of students in the last primary Total 78.6 84.0 93.4 89.4 grade minus repeaters in this grade as a pro- portion of the number of children at the Source: World Development Indicators 2006. expected graduation age.24 Girls are less likely Note: Population weighted regional averages. between 0.81 and 0.88. Where there is general in addition to Sub-Saharan Africa and South discrimination against girls, this is manifested Asia, stand out as regions where efforts must in higher under-5 mortality for girls than for be focused on reducing the health disadvan- boys. This comes about due to inequality in tages faced by girls (figure 3.9). nutrition and health care during childhood. Prenatal sex selection is one dimension of The data sources for sex-disaggregation discrimination against girls that is not cap- of under-five mortality are the same as those tured in under-five mortality. It affects the sex for the MDG4 indicator. The data issues are ratio at birth, and is especially prevalent in the same as well--although the best source the East Asia and Pacific region, and to much of data is a complete vital statistics registra- less extent in South Asia (box 3.2). Because it tion system, such systems are uncommon in is not a concern in other regions of the devel- developing countries, so estimates are also oping world, sex ratio at birth is not recom- obtained from sample surveys or derived by mended as an additional indicator. applying direct and indirect estimation tech- Mortality and morbidity beyond child- niques to other data sources. One source of hood. The use of sex-disaggregated data for internationally comparable data on global and monitoring under-five mortality is a first step regional trends in under-five mortality by sex is toward measuring gender equity in health. In the estimates published by the United Nations the area of health, the term "gender equity" using available national data. Using this data, is used instead of "gender equality" in order overall levels of under-five mortality indicate to emphasize that differences between men where efforts must focus on improving child and women in some health outcomes are due health outcomes--Sub-Saharan Africa and primarily to biological differences between South Asia (figure 3.9). When sex-disaggre- the sexes. Gender differences in average life gated data are analyzed, East Asia and Pacific, expectancy at birth are heavily influenced by Female under-five mortality rate and female-to-male ratio, 2004 Source: World Population Prospects 2004. gender disparities in child mortality rates and morbidity and mortality to healthy life. This so they do not reflect sufficiently the health measure, the disability-adjusted life years conditions of adolescents and adults. Over (DALYs), estimates potential years of life the life cycle, males and females face differ- lost due to premature death, poor health, ent risks and causes of morbidity and mortal- or disability for all age groups. Table 3.5, ity, and monitoring these differences should which summarizes the results of this study help inform health policy and programs. for the 15­29 age group, shows that there While MDG5 monitors maternal health, this are important differences in the distribu- indicator misses sources of illness and death tion of the burden of disability and death among women that are unrelated to mater- between males and females; it shows that nal causes and are not relevant for women young women are more likely to suffer from and girls not in the reproductive age group. mental health­related issues and HIV/AIDS, Monitoring the incidence of specific diseases, while young men are more likely to suffer as done by MDG6 (HIV/AIDS, malaria, and from the fallout of violence as well as inju- tuberculosis), while important, also does not ries and road traffic accidents. offer a full view of sex and gender differences The first DALY estimates were published in for the design of health policy priorities. World Development Report 1993. Two years An international attempt to measure adult of estimates are now available, one for 1990 morbidity and mortality was made by the and another for 2000. The 2000 estimates were Global Burden of Disease project, a world- subsequently revised, and the most recent esti- wide collaboration of over 100 researchers, mates are available for 2002. Due to changes sponsored by the World Health Organiza- in methodology and classification of mortality tion and the World Bank. The study used causes, the 1990 and 2002 estimates are not information from a number of countries to comparable; hence it is not possible to assess estimate the costs of individual causes of trends in DALYs. Sources of death and disability with largest gender differentials in disease burden for 15- to 29-year-olds, low- and middle-income countries Burden of disease Burden of disease Gender ratio Disease/condition (% of total) Females (% of total) Males (female/male) Fires 2.13 0.9 2.34 Source: WHO 2002. Note: The burden of disease has been calculated as the percent of DALYs lost due to a specific cause over the total DALYs lost (for men and women separately). For identifying priority diseases for gender equity, all diseases that primarily affect males (such as prostrate cancer) or females (such as maternal conditions) were removed from the list. The burden of disease for males and females were multiplied by the gender ratio. The diseases with the greatest gender differential are those that have a weighted differential above the statistical threshold of its distribution--mean plus one standard deviation. Sex ratios at birth have been excessively masculine in much of East Asia for decades, compared with the "normal" ratio of 104­106 males per 100 females in most populations. In China and the Republic of Korea, the sex ratio at birth was around 107 in 1982, rising sharply thereafter with the spread of sex-selective abortion in the mid-1980s. In China it increased to nearly 120 in 2005.a In the Republic of Korea, it peaked at around 116 in the early 1990s and has since declined to below 108 in 2005. Skewed sex ratios at birth reflect sex-selective abortion, but may also include some amount of female infanticide, where the child is not reported as a live birth and is therefore indistinguishable in the statistics from an abortion. Another route for removing unwanted daughters is selective neglect during early childhood, typically by giving girls less aggressive health care for illnesses than boys. As sex-selective abortion becomes more accessible, it becomes easier to remove daughters--as a result, sex ratios at birth rise, and selective neglect after birth declines. The net effect of sex selection before birth, at birth, and after birth is reflected in the sex ratios of children aged 0­4 (box figure). The data show that discrimination against girls is also increasing in India--by 2001 the 0­4 year sex ratio was high in the country as a whole.b It is concentrated in the northwestern states of Punjab and Haryana, where the 0­4 year sex ratios have been historically high, rising sharply with the spread of sex-selective abortion in the 1980s and 1990s. Cultural factors help explain why these parts of Asia exhibit child sex ratios so much higher than anywhere else in the world. These societies have similar lineage-based kinship systems, which effectively ensure that only boys can continue the household and lineage and care for their parents in their old age. In most other societies, daughters are not so sharply excluded from participating continued This chapter stops short of recommend- regulate fertility is strongly linked to labor ing that countries use DALYs as a monitoring force participation and earnings, it also is an indicator for adult morbidity and mortality indirect measure of the potential for women's to complement the indicator of child health, economic empowerment. The official indica- since its ultimate practicality and usefulness tor for MDG 6--contraceptive prevalence as an indicator will depend on how frequently rate--considers all methods of contraception, estimates will be available. (Given the avail- computed as the percentage of women who ability of recent DALY estimates for 2002, are practicing, or whose sexual partners are however, some of the analysis below incor- practicing, any form of contraception, tradi- porates DALYs in exploratory fashion.) One tional or modern (United Nations 2003). But significant shortcoming of DALYs is that the compared to traditional methods, modern methodology is costly for individual countries methods offer women and their partners a to apply. Countries that wish to use the DALYs more reliable way to control their fertility and for results-based monitoring can rely on the to prevent the spread of sexually transmitted WHO's production of country-level estimates. disease. This indicator also reveals the avail- ability of family planning services to women and their sexual partners. This chapter there- Reproductive Health fore recommends that countries also monitor Modern contraceptive use. This proposed the percentage of women of reproductive age indicator responds to the need to better mea- (and their sexual partners) who use modern sure women's ability to regulate fertility and contraceptives. Data on contraceptive use are choose desired family size. Since ability to typically available with a gap of three to four in the well-being of their parental households. Decades of urbanization and industrialization in the Republic of Korea have eased the grip of these traditional social structures, reflected in the trend toward normalizing child sex ratios. Child (0-4 year) sex ratios in China, India (Punjab and Haryana), and Korea (1950­2000) Source: Das Gupta and others (2003), computed from official national censuses for each country. Note: The data for India are for the age group 0­6. Sources: Korea National Statistical Office 2006; Goodkind 1996; Chung and Das Gupta 2007; and Das Gupta and others 2003. a. Derived from Chinese census and intercensal survey data: the 1982 estimate is from Zeng and others 1993; the 2000 estimate from Yuan and Tu 2005, and the 2005 estimate is reported by Xinhua (2005.08.24) and Shanghai's Business Weekly citing the Chinese Academy of Social Sciences (http://news.bbc.co.uk/2/hi/ asia-pacific/250557.stm). b. India does not officially estimate the sex ratio at birth, partly because of the absence of good vital registration data. years from Demographic and Health Surveys traception. Where available, unmet need can (DHS), Multiple Indicator Cluster Surveys, be used as an indicator of availability of fam- and contraceptive prevalence surveys. ily planning services. A related indicator, percentage of women The percentage of women reporting use of with unmet need for contraception, is being modern contraceptives has increased over the considered for inclusion in MDG5.25 It is last two decades, from 47 percent in 1990 to defined as the percentage of sexually active 56 percent in 2000 (United Nations, Depart- women who are not using any method of ment of Economic and Social Affairs, Popu- contraception and who either do not want to lation Division 2002). Table 3.6 reports data have any more children or want to postpone from countries that have had two or more their next birth for at least two more years DHS between 1985 and 2005. The percent- (Westoff 1978; Westoff and Pebley 1981). age of women aged 15­49 reporting use of Since computation of unmet need requires modern contraceptives has increased over the survey data on intentions for future births, 1990s in every country in this table. this indicator is available for a smaller set of Adolescent motherhood. Childbearing countries (mainly those with DHS surveys) among teenagers can bring disproportion- than is the indicator for use of modern con- ate health risks to the mother and the baby (maternal mortality, delivery complications, premature delivery, and low birth weight). In parts of Sub-Saharan Africa where female Trends in modern contraceptive use, selected countries genital mutilation is practiced, pregnancy can also heighten the health risks to teen mothers (Zabin and Kiragu 1997). Beyond Year % Using modern health outcomes for mother and baby, ado- contraceptive lescent motherhood is associated with early Cameroon 1991 4.3 departure from school, lower human capital 2004 12.5 accumulation, lower earnings, and a higher Ghana 1988 4.2 2003 18.7 probability of living in poverty (World Bank Kenya 1989 17.9 2006). Thus this chapter suggests that coun- 2003 31.5 tries monitor the percentage of women aged Mali 1987 1.3 15­19 who are mothers or are pregnant as an 2001 5.7 additional indicator of gender equality and Uganda 1988 2.5 2001 18.2 women's economic empowerment.26 Data for Egypt, Arab Rep. of 1988 35.5 this indicator are available from DHS and 2000 53.9 other reproductive health surveys. Morocco 1987 28.9 Births to teenage girls are common in many 2004 54.8 developing countries. Most recent data show Kazakhstan 1995 46.1 1999 52.7 that more than 10 percent of 15- to 19-year- Bangladesh 1994 36.6 olds are mothers in Sub-Saharan Africa, South 2004 47.6 Asia, and Latin America (figure 3.10). In Ban- Indonesia 1987 43.9 gladesh and Mozambique more than 30 per- 2003 56.7 cent of 15- to 19-year-olds are mothers or are Philippines 1993 24.9 2003 33.4 pregnant. In most developing countries, unlike Brazil 1986 56.5 in developed countries, teenage childbearing 1996 70.3 frequently takes place within marriage (World Colombia 1986 52.4 Bank 2006).27 The percentage of girls marry- 2005 68.2 ing before age 18 is high in a number of coun- tries and ranges from less than 20 percent in Source: Demographic and Health Surveys. Central Asia to more than 60 percent in Ban- Trend in adolescent motherhood Source: Demographic and Health Surveys. Note: Percentage of girls who are mothers or are pregnant. gladesh, Guinea, Mali, and Nicaragua.28 In women in wage employment in the nonag- Mali nearly 36 percent of young women were ricultural sector), this chapter recommends married by age 15. monitoring the labor force participation Between the early 1990s and 2000, the rates of men and women aged 20­24 and percentage of adolescent mothers declined in 25­49.29 The labor force participation rate is a number of countries (figure 3.10). This is, interpreted as indicating women's potential however, far from a universal trend: adoles- economic empowerment. cent motherhood increased over this period in Evidence from a number of developing Colombia, the Dominican Republic, Mozam- countries shows that girls are less likely than bique, Peru, the Philippines, Turkey, and the boys to make the transition from school to the Republic of Yemen. labor market (National Academy of Sciences 2005). Monitoring labor force participation among girls aged 20­24 thus indicates the Additional Indicators in the Economy extent to which education and skills acquired and Markets Domain in school are used in the labor market. Age Labor force participation. To strengthen the patterns of labor force participation show official MDG3 labor force indicator (share of that in almost all regions of the world the greatest gender gap in participation occurs mainly in unpaid agricultural work. House- between the ages of 25 and 49. This is not hold surveys asking respondents a detailed surprising since the gender division of tasks set of questions about their participation in typically results in women in this age group work activities offer a more accurate estimate contributing more of their time to child and of women's labor force activities. Thus, this home care while men increase in work out- chapter uses household surveys to calculate side the home. Monitoring employment indi- the indicators on labor force participation cators for this age group thus offers potential and employment by type.30 for policy interventions (such as child care When data on women's and men's labor services) to influence labor market behavior force participation rates are examined, three of men and women. patterns emerge. First, there are regions where Analysis of women's labor force participa- women's participation in the labor force itself tion in developing countries is sensitive to the is low. In these regions, there has been little data source. The UN Handbook on Indicators change in women's participation between for Monitoring Millennium Development 1990 and 2005 (figure 3.11). These are also Goals (United Nations 2003) recommends regions where the greatest gender difference using data from population censuses, labor in participation. In a sample of 96 developing force surveys, enterprise censuses and surveys, countries, female participation rates are the administrative records of social insurance lowest in countries of Middle East and North schemes, and official estimates. These sources Africa, South Asia, and Latin America and typically undercount women's participation, the Caribbean.31 For the 20­24 age group in especially where women workers participate these regions, the average female labor force Female and male labor force participation rates by region, 1990­2005 Source: Household and Labor Force surveys. The regional averages are calculated using the earliest value between 1990 and 1995 and the latest value between 2000 and 2005. The averages are weighted by the country population. Note: Computed from household surveys (1995­2005). Labor force participation rates for males and females aged 20­24 and 25­49. Population weighted regional averages for South Asia (5 countries), Latin America and the Caribbean (20 countries), Sub-Saharan Africa (10 countries), East Asia and Pacific (8 countries), and Europe and Central Asia (13 countries). For 5 countries in Middle East and North Africa, data are only available for 2000­05. participation rate ranges from 37 percent to Third, there are countries where female par- 49 percent--below the average of 55 percent ticipation rates are high for both age groups, or higher for the remaining regions. Similarly, gender gaps in participation rates are low, and for the 25­49 age group in these regions, the women's share in nonagricultural paid work average female participation rate is between is high. These countries are mainly in Europe 37 and 60 percent, again much lower than and Central Asia and in East Asia and Pacific that in other regions. where female participation rates are 60 per- Middle East and North Africa, South Asia, cent or higher (except in Turkey, where the and Latin America and the Caribbean are also participation rate is 38 percent). Despite this the regions with the greatest gender gaps in par- high participation, as well as high educational ticipation rates (figure 3.11). In these regions, attainment, women receive lower wages than for both age groups, male labor force partici- men (see World Bank 2001; World Bank 2002; pation rates are between 1.5 and 2 times the Pham and Reilly 2006).32 An analysis of gen- female labor force participation rates. For 20- der wage gaps in Russia and Poland during the to 24-year-olds in Latin America and Carib- mid-1990s found that only about 20 percent bean, this gender gap is paradoxical, given the of the gender gap in wages could be explained region's success in educating girls and elimi- by male-female differences in observed worker nating the gender gap in schooling. This gap or job characteristics. The remaining gap was suggests that, unlike their male counterparts, "unexplained," which is frequently interpreted young women there face barriers to reaping the as an indicator of labor market discrimination labor market returns to increased schooling. against women. Second, there are countries where female Of course labor force participation rates do participation rates for both age groups are not tell the whole story of women's economic high, the gender gap in participation is low, empowerment in labor markets. Additional but women are concentrated in low-paying information on quality of employment--in agricultural employment. These are mainly the addition to the official MDG3 indicator on countries of Sub-Saharan Africa. Countries in share of women in nonagricultural wage this region have among the highest female par- employment--is needed to give a context to ticipation rates. Of the 29 countries with data, participation rates (box 3.3). female labor force participation rates exceed 60 percent in 12 countries for the 20­24 Value-Added of the Proposed Indicators age group and in 21 countries for the 25­49 age group. In Benin, Burkina Faso, Burundi, The proposed complementary indicators deal Guinea, Mozambique, Rwanda, Tanzania, with some of the shortcomings identified in and Uganda, female participation rates are the official MDG3 indicators. They cap- close to 80 percent for both age groups. ture additional elements of gender equality In these countries, female participation in two of the three areas the official MDG3 appears to be concentrated in agricultural indicators measure: education (by examin- employment or self-employment in the non- ing sex-disaggregated completion rates) and agricultural sector. Of the 28 countries where employment (by adding sex-disaggregated the female share of agricultural employment labor force participation rates); they do not exceeds 40 percent (out of 71 countries for improve, however, the measurement of politi- which we have data on employment by type), cal participation. The indicators also incorpo- 17 are in Sub-Saharan Africa. For example, rate an important new area of gender equity in Burundi, Rwanda, and Uganda, close to 60 in health; due to data limitations, however, it percent of agricultural workers are female. was not possible to incorporate measures of In Ghana, where women workers dominate access to remunerated employment and dis- nonagricultural employment, most tend to be parities in access to productive assets. These self-employed. indicators, in conjunction with the official When women are employed, it is often claimed that, relative to men, they are more likely to: (1) be self-employed rather than work for wages; (2) work in the informal rather than the formal sector; and (3) work as own-account workers, domestic workers, and contributing family workers, while men are more likely to work as employers and wage and salaried workers (UNIFEM 2005). The evidence to support these contentions is mixed: Among 91 countries with recent data, male workers on average appear to be more likely to be self-employed than female workers (KILM, 4th edition). Whether women are overrepresented or underrepresented in the informal employment sector relative to total nonagricultural employment sector differs across developing countries. In Sub- Saharan Africa, Latin America, and India, the share of women in informal employment is larger than the share of men so engaged. In the Middle East and North Africa, the reverse seems to be the case (Grown and others 2005). Among 80 developing countries with data, male workers appear to be more likely than female workers to be own-account workers (KILM 4th edition). Female workers are indeed more likely to be domestic workers and unpaid workers in family enterprises than male workers. A key question is whether women prefer to work at home or in family-owned businesses because of the location or the flexibility of work hours, which allows them to more easily combine work, domestic chores, and care. Or, do prevailing gender norms condition women to assume this tri- ple workload or restrict their mobility? Or, is this pattern a result not of supply considerations, but rather of the gendered demand for labor, which presumably reflects existing societal gender norms? Sources: Grown and others 2005; KILM, 4th edition; Carr and Chen 2004. MDG3 indicators, do a better job of measur- five mortality rate).35 The figure shows that ing empowerment of women and of includ- the additional indicators provide significant ing sex-disaggregated levels of key indicators, information beyond that supplied by the thus overcoming some of the interpretation official indicators. Thirty-two of 54 coun- difficulties associated with parity ratios. tries change rankings. Some countries that The value added of the additional indi- score relatively high on the official indica- cators is in part an empirical question and tors score relatively low on the additional in part a policy question. On the empirical indicators (for example, the Baltic countries side, the additional indicators are only valu- and some other countries in Europe and able if they provide a ranking of countries Central Asia, as well as Cuba, Namibia, and on gender equality that is substantially dif- Maldives). On the other hand, some coun- ferent than that produced by the official tries with relatively low scores on the official MDG3 indicators--that is, they add infor- indicators have substantially higher relative mation.33 For each country, we calculate the scores on the additional ones (for example, rank according to the official MDG3 indica- Brazil, Indonesia, Mauritius, Morocco, and tors and the proposed indicators.34 Figure Swaziland). For both sets of countries, the 3.12 plots the scores of 54 countries on the additional indicators are capturing elements official MDG3 indicators against two of the of gender equality not captured by the offi- additional indicators (the female-to-male cial MDG3 indicators. For countries close ratio of primary completion and the under- to the diagonal in figure 3.12, the ranking on Comparison of country scores on official and expanded MDG3 indicators Source: Calculations using data from household surveys, World Development Indicators, and World Population Prospects 2004 for 54 countries 2000­05. Note: To calculate the aggregate rank of countries for each set of indicators, we rank all countries that have data on the official MDG3 and pro- posed indicators for the period 2000­05. For purposes of comparability, countries for which the most recent data are older than this period are not considered. Each country is ranked by the value of each component indicator, from lowest to highest, and assigned a consecutive number (starting with 1) in accordance with its rank. Countries with exactly the same value are given the same rank. The sum of these values for each country yields a composite measure for that country, which then becomes the basis for a final ranking of the countries. This composite measure or index takes a minimum value of 6 and a maximum value of 6 times 54. For each set of indicators--official and proposed--we group the coun- tries into quintiles according to this ranking. Because the official MDG3 indicators consist of four education indicators, a country's rank on the official indicators depends heavily on its performance in education. A country's ranking on the official indicators was not adjusted for this over- representation of education indicators, because it reflects the priority placed on education in the MDGs. the official and additional indicators is quite having a strong link to poverty reduction and similar; for these countries (for example, Alba- growth, but for which data are not currently nia, Argentina, Malaysia, Nepal, and Tunisia) widely available. Data collection for these the latter indicators provide little extra infor- indicators should be strongly considered. mation on overall levels of gender equality. Information is most needed for indica- On the policy side, the value added of the tors that measure gender equality in society. additional indicators hinges on their policy Data on the share of women in positions in relevance. One aspect of policy relevance is the the executive branch of government and in ability to identify countries that have scored local government are available only for some particularly high or particularly low on spe- regions and some countries. Increasing cov- cific elements of gender equality and draw erage to a larger number of countries and conclusions about how public policy has influ- regions should not be too onerous, especially enced these outcomes. Another related aspect because these data are straightforward and of policy relevance is whether scores on these easy to collect. Similarly, voting behavior by additional indicators are useful in identify- sex should be easily obtained from voting ing other areas for public policy intervention. records or from international opinion sur- These themes are explored in more detail in veys, such as the regional barometer surveys the section that follows. (Latin, African, and Asian barometers) and the World Values Surveys, which now include more questions about voting behavior. The Measurement of Gender Equality A potentially revealing indicator with direct and Empowerment: Data Needs implications for using services and enjoying cit- The proposed complementary indicators dis- izen rights, is the number and share of women cussed above do not remedy all the shortcom- and men with basic citizenship documents, ings of the official MDG3 indicators. Table starting with birth registrations (and ending 3.7 lists other prospective indicators that meet with death registries).36 Recent research by the the criteria of being modifiable by policy and Inter-American Development Bank showed Prospective indicators for which data are not currently available Household Economy and markets Society Test scores, male and female Gender gap in wagesa,b Percentage voting by male, female, and ratio Proportion of women who have Share of women in informal wage Proportion of seats held by ever been victims of physical and self-employment in nonagricultural women in local government violence by an intimate male employment partner Percent of employed women who Proportion of women in the have access to child care executive branch Businesses, by average size and sex Percentage of individuals of ownerb who possess basic citizenship documents, female, and ratio Access to credit for women and men Land ownership by female, male, and jointly helda,b Source: World Bank staff. a. Recommended by UN Millennium Project Task Force on Education and Gender Equality. b. Included in World Bank's Country Policy and Institutional Assessments (CPIA). underregistration of births in six Latin Ameri- sure the results of gender-informed educational can countries, varying from 8.4 percent in interventions. On the former, new international Peru to 25.8 percent in the Dominican Repub- tests, such as the TIMSS, PIRL, PISA, and SAC- lic. Characteristics associated with the risk of MEQ, measure achievement, but coverage of a child being undocumented from birth to age developing countries is still too limited. five included poverty, rural residence, and teen Reliable and comparable data with good motherhood (Duryea and others 2006). coverage are also needed on the prevalence of In the economy and market, comparable violence against girls and women in the family. and timely data with good country coverage Comparable victimization surveys conducted are urgently needed on the share of women by the World Health Organization in 10 coun- in informal and self-employment (as part of a tries and a recently published major study by more comprehensive package of information WHO (2005) are a promising start, as is the on women's and men's employment, cover- initiative by Macro International to include ing agricultural and nonagricultural activities questions about intimate partner violence formal and informal, wage, and self-employ- in Demographic and Health Surveys in nine ment). Data are also needed on wages and countries (Kishor and Johnson 2004). These earnings by sex and type of employment. This efforts need to be scaled up. WHO should lead recommendation, made first by the subgroup an international effort to collect data on an on gender indicators of the Inter-Agency and appropriate, measurable, and actionable indi- Expert Group on the MDG indicators, needs cator of violence against women: tentatively, to be implemented, with efforts to improve the proportion of women who in the past 12- and expand the collection and analysis of month period have been victims of physical such data by national statistical offices.37 violence by an intimate male partner.38 There also needs to be a significant inter- In describing data that are needed for an national effort to obtain even basic data on expanded set of indicators, one must not both productive and consumer assets--land, lose sight of the fact that data on all six offi- livestock, house ownership, other property, cial indicators of MDG3 are available for credit, business ownership--disaggregated only 59 out of 154 countries (for 2000­05). by sex, at the level of individuals, households, Many others have produced data but do not and firms. Some countries have information update the information regularly. This lim- on land tenure by the sex of the owner in agri- its the number of countries that can be used cultural censuses or surveys. But information for making valid cross-country comparisons; on access to credit (formal and informal) and of the 154 countries in the database for this business ownership by sex is almost nonex- report, only 41 have information for both the istent, except for micro studies. The interna- official and the expanded lists of MDG3 indi- tional development agencies that produce and cators for 2000­05. Collecting and publish- run large-scale specialized surveys--such as ing updated information for these indicators FAO's and IFAD's agricultural and rural sur- is a clear--and doable--priority. veys and the World Bank Group's household The UN Statistics Division, in collaboration and business surveys--need to make a special with the World Bank and UNFPA, recently effort to collect and analyze sex-disaggregated set up an Interagency and Expert Group for information on asset ownership and control. Gender Statistics, with broad representation The most complete existing coverage of of international organizations, national sta- reliable and actionable data is for indicators tistical offices, and nongovernmental institu- of gender equality in the household (where tions. At its inaugural meeting, in December there has been the greatest advance in gender 2006, the group launched a global gender equality). Additional data-gathering efforts statistics program to strengthen and comple- are needed to obtain measures of educational ment national, regional, and other interna- achievement (test scores) by gender and to mea- tional gender statistics programs. This and similar efforts need to be fully supported by Outlier Countries According to the the international development community-- Official MDG3 Indicators since without good data, little progress will be To identify outlier countries according to the made in national and international efforts to official MDG3 indicators, we use the ranking achieve MDG3. method described above. We group the coun- tries into quintiles according to this ranking. Of the 12 countries in the top 20 percent or quintile, 10 are in Europe and Central Asia This section identifies outliers--countries and Latin America and the Caribbean, indi- that have especially high or low performance cating a clear regional pattern (table 3.8). with respect to the official and expanded set Mongolia and Namibia deserve a closer look of MDG3 indicators in the most recent year because they differ significantly from the for which data are available (in the 2000­05 other countries in their regions. Of the 12 period).39 This exercise is undertaken in order countries in the bottom quintile, four are in to extract lessons learned about policies to Sub-Saharan Africa, three are in South Asia, promote gender equality from both high and and two are in North Africa. Note that four low performers, rather than to obtain a global out of the eight countries in Sub-Saharan ranking of countries for MDG3.40 For those Africa and three of the four in South Asia are countries identified as outliers (only with in this bottom quintile. Guatemala and Cam- respect to the relatively small subset of coun- bodia stand out because most other countries tries for which data are available), the section in their regions are in the top two quintiles. then examines the evolution of these indicators To understand better the levels of gender over the 1990­2005 period, both to understand equality achieved in about 2005, the perfor- how these countries ended up where they did mance of countries in the two quintiles on and to determine whether there has been con- the official MDG3 indicators is examined vergence among countries that started at very over a 10- to 15-year period. Three basic different levels of gender equality. Lastly, the patterns can be observed: section discusses changes in laws, institutions, and policies in outlier countries that may have For primary enrollment, secondary enroll- contributed to the improvement (or worsen- ment, and literacy rates, there has been ing) of indicators and the policy framework for significant convergence: countries that gender equality and women's empowerment. were in the bottom quintile in 1990 reg- Countries in the top and bottom quintiles, according to scores on official MDG3 indicators World region (number of countries with comparable data) Bottom quintile Top quintile Sub-Saharan Africa (8) Burkina Faso, Malawi, Kenya, Sudan Namibia East Asia & Pacific (8) Cambodia Mongolia Europe & Central Asia (20) Turkey Belarus, Lithuania, Estonia, Moldova, Latvia Latin America & the Caribbean (14) Guatemala Argentina, Honduras, Colombia, Suriname, Costa Rica Middle East & North Africa (5) Morocco, Algeria South Asia (4) India, Pakistan, Nepal Source: World Bank staff. Note: There are 59 countries with data for 2000­05. The number of countries with comparable data for these indicators in each region is given in parentheses. istered significantly more rapid progress several of the countries that were in the than those in the top quintile. The case bottom quintile in 1990 had caught up of primary enrollment (figure 3.13 shows with the top quintile. This reflects the the size of the change in the levels of one progress of most countries in basic educa- indicator, depicted by the length of the tion, a point made in the preceding section arrows) is especially striking.41 By 2005, and illustrated by the remarkable leaps for Changes in official MDG3 indicators for countries in the bottom and top quintiles, 1990­2005 continued Changes in official MDG3 indicators for countries in the bottom and top quintiles, 1990­2005 (continued) Source: World Development Indicators 2006. Note: A few countries do not have data around 1990, defined here as 1998­95. The countries for which initial data are missing appear with just one point. several bottom-quintile countries--espe- employment and political participation. cially in comparison with the countries For these two indicators, there is little dif- in the top quintile. Note the progress by ference in performance between top and Burkina Faso, Malawi, Morocco, and bottom quintile countries.42 For nonag- Nepal. Countries in the top quintile show ricultural wage employment (see figure little change in this ratio because they were 3.13), Kenya stands out because its rapid already close to gender equality in 1990. progress on this indicator puts its 2005 In the case of tertiary enrollment (also score at about the initial level for several pictured in figure 3.13), there has been top-quintile countries. Cambodia also a widening of the gaps between bottom stands out because it begins the period on and top quintile countries. By and large, par with the countries in the top quintile; top quintile countries had female-to-male in contrast, it is in the bottom quintile of enrollment ratios in excess of one in 1990, performers because of its record on the and these rates rose substantially over other MDG3 indicators. In several coun- the 1990­2005 period. This is indicative tries in the top quintile, notably Argentina of severe male disadvantage. At the same and Namibia, women's share in nonagri- time, bottom quintile countries, charac- cultural employment continues to rise. terized by severe female disadvantage, made significant progress in boosting the Outlier Countries Using the Expanded female-to-male enrollment ratio; the only Gender-Equality Indicators two exceptions to this pattern are Burkina Faso and Nepal. Now consider the proposed additional gen- A third pattern emerges with respect to der-equality indicators and how countries women's share in nonagricultural wage perform. Because comparable data across countries on these additional indicators are the exception of Burkina Faso, appear likely to even more scarce than for the official MDG3 achieve gender equality in this indicator. indicators, we focus on two subsets of indica- The gender ratio for under-five mortality tors: (1) the primary completion rates and the rates is a different story. There has been very under-five mortality rate, for which compa- little change in or among countries in both rable data are available for 54 countries, and quintiles. Remember that in four industrial (2) labor force participation rates and DALYs countries considered to be the most gender for which comparable data are available for equal (Denmark, Finland, Norway, and Swe- 41 countries. Countries are re-ranked for den), the female- to-male ratio is between 0.81 each subset, and the most gender-equal and and 0.88. The developing and transition coun- the least gender-equal countries are listed in tries in the top quintile were near this level in tables 3.9 and 3.10. about 1990. But for a few of these countries As with the official MDG3 indicators, the (Indonesia, Kazakhstan, Panama, the Philip- top quintile features several countries in Latin pines, and Suriname), the indicator suggests America, but even more with this new set of worsening mortality rates for boys relative to indicators. Countries in blue are common to girls. In contrast, the countries in the bottom both rankings, four in the top quintile (Argen- quintile are all above this range, with the three tina, Costa Rica, Honduras, and República South Asian countries having data showing Bolivariana de Venezuela) and seven in the the greatest disadvantage for girls. bottom (Algeria, Burkina Faso, Cambodia, On the second subset of proposed indicators India, Malawi, Nepal, and Sudan). (using data on 41 countries) the similarities and How have these countries performed since differences are striking. In the bottom quintile, 1990? Except for Honduras and República five countries are in the bottom quintile on the Bolivariana de Venezuela, there has been very official MDG3 indicators (table 3.8) and the little change in the gender ratio in primary first subset of proposed indicators (table 3.9). completion rates in the top-quintile countries; The top quintile consists only of the countries essentially, all these are roughly at parity (fig- in Europe and Central Asia (table 3.10), who ure 3.14). In the bottom quintile, however, five achieve their high ranking due to desirable countries achieved notable progress, and with outcomes for females in labor force participa- Countries in the top and bottom quintiles, according to primary completion rates and under-5 mortality World region (number of countries with comparable data) Bottom quintile Top quintile Sub-Saharan Africa (8) Burkina Faso, Cambodia, Mauritius, Indonesia, East Asia & Pacific (6) Malawi, Sudan Philippines Europe & Central Asia (18) Latin America & the Caribbean (14) Azerbaijan, Tajikistan Argentina, Brazil, Costa Rica, Dominican Republic, Honduras, Kazakhstan, Panama, República Bolivariana de Venezuela Middle East & North Africa (5) Algeria, India, Jordan, South Asia (3) Maldives, Nepal Source: World Bank staff. Note: The total number of countries with data during the period 2000­05 is 54. The number of countries with comparable data for these indi- cators in each region is given in parentheses. The countries in blue (Algeria, Argentina, Burkina Faso, Cambodia, Costa Rica, Honduras, India, Malawi, Nepal, and Sudan) also appear in the lists in table 3.8. Countries in the top and bottom quintiles, according to labor force participation rates and disability-adjusted life years World region (number of countries with comparable data) Bottom quintile Top quintile Sub-Saharan Africa (3) Burkina Faso, Malawi, Swaziland East Asia & Pacific (5) Indonesia Europe & Central Asia (18) Turkey Belarus, Estonia, Kazakhstan, Latvia, Lithuania, Moldova, Romania, Russian Federation, Ukraine Latin America & the Caribbean (12) Honduras Middle East & North Africa (1) Jordan South Asia (2) India, Pakistan Source: World Bank staff. Note: The total number of countries with data during the period 2000­2005 is 41. The number of countries with comparable data for these indi- cators in each region is given in parentheses. The countries in blue ( Burkina Faso, Malawi, Turkey, Pakistan, India, Latvia, Moldova, Estonia, Lithu- ania, and Belarus) appear also in the lists in tables 3.8 and 3.9; the countries in bold (Kazakhstan and Jordan) appear also in table 3.9. tion and an undesirable male disadvantage in women by prohibiting discrimination on the DALYs. Five of them are in the top quintile in basis of sex and by adopting special measures the previous lists. Due to a lack of comparable for women's advancement. The difference data for the beginning of the period, it is not between high- and low-performing countries possible to track the progress of countries on is not so much the laws themselves, however, this set of indicators. as in the mechanisms to implement them. High-performing countries tend to have more developed policy frameworks to enforce the Changes in Rights, Resources, and Voice laws. Without institutions for enforcement, in Outlier Countries including the information needed for it and the Sweeping changes in a country's institutional associated budget allocations, good laws can environment can affect gender equality. Large become mere statements of noble intention. social and economic transformations tend to A good illustration of a coherent package of change gender structures and relations, with legal reforms that includes provisions for their effects on gender indicators. For example, the enforcement is Moldova's Gender Equality fall of the Taliban in Afghanistan ushered in the Law, passed in February 2006.43 The law spec- highest increase in school enrollment rates in ifies the mandates and responsibilities of public its history for both boys and girls, but with the institutions with a role in enforcement, autho- previous restrictions harsher for girls, the rise rizes public budget funding for these agencies, in their enrollment was especially impressive. and establishes both penalties and repara- This change was brought about by a near-dou- tions for violations of the law. In Lithuania bling in the number of schools in the country the office of Equal Opportunity Ombudsman after 2003 and cessation of physical threats on gave "teeth" to a series of antidiscrimination girls who attend school. Even so, long-stand- laws passed between 2000 and 2004 by inves- ing institutional obstacles and investment tigating and penalizing offenders for violating shortfalls will keep Afghanistan's indicators of the Law of the Republic of Lithuania on Equal gender equality below those of other countries Opportunities for Men and Women. in the region for some time. High-performing countries do not have Equal rights. Both high and low perform- discriminatory laws condoning differential ers have enacted constitutional or legal reforms treatment between men and women, while to "level the playing field" between men and many low-performing countries do. Women Changes in two proposed indicators for countries in the bottom and top quintiles, 1990­2005 Sources: World Development Indicators (2006) for primary completion rates and World Population Prospects (2004) for under-five mortality rates. in low performing countries are often treated tion. And as a result of the commitments at as minors in family law--for instance, they the UN Beijing Women's Conference (1995), cannot pass on citizenship to a child, and they many governments elevated these offices to need their husband's permission to include ministerial status. Despite this high profile, a their children's names in a passport or obtain frequent commentary is that these offices or a national identity card. In addition, laws in ministries continue to be marginal or frag- these countries often directly or indirectly ile in their institutional capacity, budgets, constrain women's options for employment and influence (INSTRAW 2005; UNDAW and their ownership of productive assets. 2004). Their performance seems to be influ- Examples include supposedly "protective" enced by their location in the government labor laws, such as bans on women's night structure (the closer to the office of the presi- work in the agricultural sector (India's Planta- dent or prime minister, the better), by their tions Labor Act of 1951) and the requirement resources, and by their links with women's that employers bear all costs of maternity movements in civil society. Relatively well- benefits (Burkina Faso, India), that increase resourced women's ministries have played employers' costs of hiring women. Some of significant roles in promoting gender equal- these countries have no gender-specific pro- ity. Cambodia's Ministry of Women's and visions in labor laws that ban dismissal dur- Veteran's Affairs (established in 1998), for ing pregnancy (Burkina Faso, Kenya). Land example, was singled out as one of the key registration laws strengthen the land rights of ministries to execute the medium- term male heads of household and weaken wom- expenditure framework (2005­08) and to en's customary land rights (Kenya). implement Cambodia's National Strategic Many legal changes ensuring equal rights Development Plan (2006­10) (box 3.4). for men and women are quite recent, under- Equal resources. A notable feature that dif- scoring the fact that legal changes often fol- ferentiates the high-performing countries (by low and reflect social changes. Legal reforms definition) is their gender parity in education have often followed and may have benefited and health indicators--suggesting that gen- from improvements in gender equality in der-informed investments in human capital both high- and low-performing countries. are key to promoting gender equality, and that Pakistan, a low performer, suggests the close low-performing countries with aggressive edu- connection between social and legal changes. cation policies are on a good track. Malawi, Pakistan's national assembly passed the Pro- for instance, has achieved significant increases tection of Women Bill in November 2006, in gender parity ratios at all levels of schooling, after much debate and controversy. Remov- thanks both to universal free primary education ing rape from the jurisdiction of Islamic laws, (1994) and to a specific emphasis in increasing the bill makes rape a crime punishable under girls' attainment in basic education. Gender- Pakistan's penal code. Despite Pakistan's informed policy reforms included reducing the overall low scores on gender equality in the direct costs of girls' schooling, increasing access 1990­2003 period, it improved gender parity and retention of girls in school, and removing ratios considerably in secondary and tertiary gender bias in teaching (Semu 2003). schooling, and it increased women's represen- Various factors are associated with wom- tation in parliament from 10 to 22 percent. en's greater opportunities in the labor force These gains in gender equality and women's in the high-performing countries. For the empowerment quite likely set the stage for or countries of Europe and Central Asia, gender facilitated the bill's passage into law. equality is the legacy of explicit state policies Most countries now have separate gov- that emphasized employment as both a right ernment offices or agencies to promote gen- and a duty for both men and women. There is der equality and women's empowerment little evidence that the treatment of women in and to enforce equal opportunity legisla- the labor market has systematically deterio- Since peace was restored in 1997, Cambodia has made huge strides to recover from nearly 30 years of conflict. But the legacy of war still constrains growth and gender equality. Cambodia remains one of the poorest countries in East Asia, with 35 percent of households living below the national poverty line, and low levels of health and education in the population. A very unbalanced adult sex ratio dur- ing the war--the result of more men dying than women--is evening out, but has left a high percentage of poor households headed by women; it may have worsened gender relations and contributed to rising levels of domestic violence. In response, with backing from the prime minister, the Ministry of Women's Affairs (MWA) has had a major role in integrating gender equality concerns in government plans, including the poverty reduction strategy and the national development plan (2006­10). At the macro level, MWA is institutionally well positioned to influence laws and policies, and it benefits from having gender equality enshrined in the constitution. It has invested in statistics and has expanded the official MDG3 indicators to sharpen the government's focus on gender inequalities. It added the indicators of gender equality in: (1) literacy rates for 25- to 44-year-olds, to cover women in prime childbearing and working ages; (2) wage employment in agriculture, industry, and services, to monitor sex segregation within sectors (women are underrepresented in the service sector); and (3) all elected bodies (National Assembly, Senate, and commune councils) and government positions. In addi- tion, it added a new target focused on reducing all forms of violence against women and children. At the meso level, MWA provides technical assistance and training on mainstream gender issues in line ministries, including agriculture (the most important source of economic livelihood for women and men), education (to increase the number of females in secondary education), and labor (to draft bilateral agreements that will ensure safe international migration for women workers). At the micro level, MWA takes the lead in developing services that are not yet a priority for line min- istries and in piloting projects--for instance, assistance and business training to women garment workers in order to reduce their vulnerability to garment industry retrenchment. The MWA active engagement in mainstream policy formulation, collaboration with line min- istries, and ability to monitor progress have all contributed to its success. Constraints affecting its performance include insufficient technical and research capacity, limited allocation of resources, and poor understanding of gender equality and gender mainstreaming--on the latter, there is still a tendency to undertake isolated, women-specific activities, with little overall impact. Source: Phavi and Urashima 2006. rated with the transition to the market econ- nor economic growth, since women's partici- omy, though there is concern that women pation rose despite widespread economic stag- are not taking full advantage of the process nation in the 1990s. Instead, the rise is related of economic liberalization and privatiza- to secular changes in the role of women in tion. These countries thus need to strengthen households and in the labor market, associ- equal rights to access to resources, such as ated with their higher education, lower fertil- land, capital, and credit and other financial ity, and higher wages. While the gender gap services (World Bank 2002). in wages is still wide in some countries in the For the Latin American countries that region, it has narrowed significantly in oth- record high rates of women's participation in ers, and in high-performing Colombia it has the official nonagricultural wage employment already closed (Duryea and others 2004). indicator, the rise in women's participation in Cultural barriers often constrain the employ- the workforce in the 1990s seemed to be the ment options of women in low-performing result of neither specific government policy countries, especially in some countries in the Middle East and North Africa with com- women's rights, resources, and voice: laws, paratively high levels of female schooling. institutions, and policies. There has been In addition, structural changes in the econ- perhaps most progress in reforming constitu- omy--such as losses in the agricultural sector tional and legal frameworks to ensure equal and increased rural to urban migration--can rights for women under the law. CEDAW result in a decline in female labor force par- and other international and regional conven- ticipation. This was the case in Turkey, where tions have provided a general framework for men compensated for the steady fall in agri- national legislation that bans discrimination cultural employment by taking up nonagri- on the basis of sex and protects women's cultural work while women had to leave the rights. A first challenge is enforcement of market "voluntarily" (World Bank 2003). these laws. Second, and despite the advance- Equal voice. Women's representation in par- ments made, laws in many countries still con- liament is the only indicator currently available done differential treatment between the sexes to identify high- and low-performing countries or, more blatantly, treat women as minors or on the issue of equal voice, and it is the one second-class citizens. Laws without enforce- area in which changes can be more directly ment may not guarantee equal rights but dif- attributed to affirmative government action. ferential treatment under the law seems to be The two countries with the highest representa- a good predictor of gender inequality in soci- tion of women in parliament and the largest ety. Legal reforms, both to change the letter increases are Argentina and Costa Rica, which of the law and to strengthen the vehicles for adopted quota laws for women's representation enforcement, need to be paired with efforts in parliament in the early 1990s. In Argentina to improve the collection and analysis of sex- the current female membership in the National disaggregated statistics--the basic building Congress is the highest ever attained--42 per- blocks for enforcing rights, designing policy, cent in the Senate and 33 percent in the House. insuring government accountability, and Namibia, the only high-performing country in monitoring progress in MDG3. Sub-Saharan Africa, adopted quota laws for Institutions, even without the presence parliamentary and municipal elections in the of overtly discriminatory rules, can reflect mid-1990s. As result, women's representation and reinforce gender inequality by restrict- in parliament rose from single digits to 28 per- ing women's access to resources and ser- cent during 1990­2003. Quota laws appear to vices, or they can enforce equal rights and help solidify women's gains in parliamentary unlock opportunities for women. Gov- representation. Without them, women's gains ernments have most often promoted an in representation can be quite volatile. Mon- enabling institutional environment for gen- golia is a high-performing country with high der equality both by setting aside resources gender parity in education and health, grow- to "mainstream" promotion and enforce- ing women's participation in employment and ment functions in line ministries and other self-employment, and a new constitution that government agencies, and by establishing a guarantees equal rights and includes many separate office or ministry with promotion, provisions prohibiting gender discrimination oversight, and/or enforcement functions. (1996). Even so, women's representation in Mainstreaming should increase the positive parliament declined sharply in 1990-2003, impacts of government interventions and from the mid­1920s to the single digits. prevent unintended negative impacts of gov- ernment action on gender equality. Because mainstreaming makes it difficult to track the The Policy Framework for Gender amount of resources allocated to promote Equality and Women's Empowerment gender equality, public scrutiny of budgets These examples underscore three main instru- is an important tool for holding the govern- ments available to governments to advance ment accountable (box 3.5). Over time, effective mainstreaming roeconomic factors, demographic trends, should obviate the need for having a separate and the functioning of different markets. In agency or function for promotion. However, Bangladesh, for example, the opening of the more than two decades of experience with economy to trade significantly increased eco- "national women's machineries" (and with nomic opportunities for women in the garment the integration of gender concerns in donor export sector (Kabeer and Mahmud 2004). and international agencies--see chapters 4 Globalization and the opening of markets and 5) has shown that mainstreaming is a in many other countries, however, have ben- long-term process, and that technical sound- efited skilled over unskilled workers, which ness, instrumental rationales, and financial may have widened the differences in economic incentives all help with the mainstreaming opportunities between more educated and less task. The experience also suggests that gen- educated women, and between women and der mainstreaming does not reduce the con- men. A growing economy and well-function- tinuing need for a separate function or agency ing markets expand opportunities for all. If with vigilance functions. The challenge is to women's economic empowerment contrib- ensure that separate and mainstreaming func- utes to poverty reduction and growth, stable tions complement and reinforce each other, growth that generates quality employment rather than duplicate efforts and/or compete and provides an enabling environment for for scarce resources. entrepreneurship is necessary, if not sufficient, Advocacy organizations in civil society, to expand women's economic opportunities. including NGOs and grassroots groups, have Universal programs can work--and have been central in promoting gender equality worked--to reduce gender inequalities (as and women's rights. Effective action, espe- in the case of the push for universal primary cially in terms of protecting women's rights, education); nevertheless, reducing gender has often been the result of alliances between inequalities most often requires targeted them and government counterparts. action and (sometimes) specialized agencies. In terms of policies, there is considerable There is comparatively good knowledge, for knowledge on cost-effective ways to pro- example, on expanding labor market oppor- mote gender equality in the domain of the tunities for women through gender-informed household, especially in terms of increasing training and job intermediation programs. girls' and women's access to education and And there is substantial knowledge on the maternal and reproductive health services. design of microfinance institutions to increase Demand-side interventions that condition women's access to credit and other financial transfers and subsidies to gender equality services. More knowledge is needed in terms objectives are increasingly popular and prov- of what works to expand women's access to ing to be cost effective. Delivery of health and productive resources and productive infra- education services can be designed to promote structure. gender equality by putting in place measures Targeted action is especially needed in the that prevent discrimination by providers and case for those left behind because of the inter- encourage providers to be responsive to gen- action of gender and other forms of exclusion der differences in client needs. Such measures (such as ethnicity, race, location, or disabil- include the provision of single-sex facilities ity). Cumulative disadvantages present both and female providers which have been effec- institutional and legal challenges on how tive in increasing women's service utilization best to promote opportunities and protect in a variety of contexts (World Bank 2001). the rights of girls and women who belong to The promotion of gender equality in the excluded groups in the population. economy is less easily influenced, because it Much more can be done to promote gender depends not only on gender-targeted policy equality in the societal domain, a cornerstone and project interventions, but also on mac- for the promotion of overall gender equal- Public scrutiny of the budget from a gender equality perspective is important for both mainstreaming gender in government policies and empowering citizens to influence policy making and hold governments accountable for public finance management. In the last decade, more than 60 countries have undertaken analyses of public budgets to assess differential incidence and effect on men and women, as well as to measure men's and women's economic contributions. Different approaches to gender-informed budget analysis. Approaches have differed in terms of focus, coverage, and methodology: The Women's Budget Initiative (WBI) in South Africa expanded its initial broad focus on the national budget to analyses of specific budgets for domestic violence prevention, treatment, housing, and child support grant programs, among others. The Uganda Gender Budget Project analysis covered the national budget by sector. In Mexico, the analysis focused on antipoverty programs and public expenditure on health in several states. Korea and the Philippines analyzed women-targeted policies and activities at the local level. In Morocco, gender budget- ing is being introduced also at the local level. In general, the more specific or focused the gender-informed budget exercise, the easier its implementation. Most efforts have covered public expenditures, classified into: (1) women-specific expenditures; (2) gender equal- ity expenditures in sectors or line ministries; (3) mainstreamed government expenditures that provide goods or services to the whole community; and (4) expenditures to achieve equity in public sector staff rosters. Some also extended coverage to revenues: the South African WBI looked at taxation to reduce bias against women, and a review of the value-added tax (VAT) in Uganda recommended tax relief on items used by women in the care economy. The most commonly used method takes the government's policy framework and examines it sector by sector, both in terms of utilization of budget expenditures and longer-term impacts on men and women. The Uganda analysis compared administration expenditures with public services that citizens received and proposed realloca- tions within and between sectors. The Mexico analysis focused on how "gender-neutral" programs recognized and addressed the limitations women face and whether they covered women's needs and build their capacities. Ideally, these analyses should cover the four dimensions of government budgets and their interaction: expendi- ture, revenue, the macroeconomics of the budget, and participation in budget decision-making processes. Tools have included gender disaggregated beneficiary assessments, public-expenditure benefit incidence analysis, and tax incidence analysis, among others. Chile has included gender as a cross-cutting theme in a performance- based national budget, and is using incentives (salary bonuses) for public sector staff as a tool to achieve measur- able results. The World Bank has undertaken gender-disaggregated public expenditure reviews in a number of countries, including Cambodia, Ghana, Morocco, Paraguay, St. Vincent and the Grenadines, Vietnam, and Uganda. Most have combined the use of gender-disaggregated benefit incidence analysis with gender institutional analyses or gender impact assessments of public programs. The reviews have shown that undertaking gender analysis can contribute to better targeted, more efficient, and more equitable public expenditure. Lesson and challenge. The main lesson from the experience with gender-informed budget analysis is that changing public policy priorities is a more complex process than pointing out gender differences and disparities in budgets. The implementation of budget initiatives requires upgrading the technical skills of budget officials and gender experts; raising public awareness of gender issues to ensure the sustainability of the initiatives; and supporting well-informed coalitions of NGOs for advocacy. Most importantly, effective government agencies are central to their implementation. The key challenge for gender-informed budget analysis and policy making is moving beyond gender-targeted interventions to full and sustained gender mainstreaming in the budget process. Sources: Asesorias para el Desarrollo (2007); BRIDGE (2003); Budlender and Hewitt (2002, 2003); Elson (2006); UNIFEM (2002); World Bank (2007). ity. Women's voices in society--expressed to those that seek to facilitate the operation through leadership positions in politics and of microfinance institutions. But they also grassroots and other women's organizations include targeted interventions, especially for in civil society--should continue to be a subgroups of women in the population that main driver for gender equality and women's suffer multiple exclusions. A third policy les- empowerment. son is that civil society and the private sector This chapter has highlighted the intrinsic have key roles to play in promoting gender importance of MDG3. It has also documented equality--the former by forming alliances that progress toward attaining MDG3 should with government and promoting government have multiplier effects and spur progress in accountability, and the latter by expanding other MDGs. To monitor this advancement, economic opportunities for women. the chapter has recommended complement- In the short run, there may be policy-level ing the official indicators with selected addi- tradeoffs between equity and efficiency; in the tional ones; it has highlighted the need to long run, however, greater gender equality in strengthen the collection and analysis of sex- access to opportunities, rights, and voice can disaggregated data in all domains of gender lead to more efficient economic functioning equality. The additional investments needed and better institutions, with dynamic benefits to monitor MDG3 should go hand-in-hand for investment and growth. The business case with the scaling up of successful interven- for investing in MDG3 is strong--it is noth- tions to expand opportunities for girls and ing more than smart economics. women. Policy lessons from the review of the evi- dence in this chapter include, first and most importantly, that there can be significant 1. This belief is formalized in several interna- advancement in gender equality when there tional conventions: the Convention to Eliminate is the will--as shown by the remarkable All forms of Discrimination Against Women leaps that countries, even low-performing (CEDAW); the Inter-American Convention on ones, have made in improving girls' access to the Prevention, Punishment, and Eradication of schooling. Second is that laws, institutions, Violence against Women (Convention of Belem and policies matter. Laws provide an appro- do Para); and the Protocol to the African Charter priate framework for leveling the playing on Human and People's Rights on the Rights of Women in Africa. field between men and women, but have no 2. See UN Millennium Project Task Force on impact if they are not enforced; enforcement Gender Equality 2005; Germain 2004; Burkhalter requires institutions with budgets and with 2002; de Walque 2006; and van der Straten and reliable information to back enforcement. others 1998. The promotion of gender equality requires 3. For 73 countries, the correlation coeffi- distinct institutional arrangements (for vigi- cient between the poverty headcount ratio using lance and accountability), as well as actions 1997 data or the closest year to 1997 with avail- to mainstream gender issues across public able data (US$2/day; 1993 PPP dollars) and the sector agencies. Gender mainstreaming can female-to-male ratio in HDIs (1997) is ­0.67, with work, but it requires high-level leadership as an R square = 0.43. For the relationship between well as technical and budgetary resources--it poverty and other gender equality measures, see Klasen (2006). is not cost free. 4. For 103 countries, the correlation between the There are well-known policy tools avail- average annual GDP per capita growth rate 1997­ able to promote gender equality. They include 2004 (in percent) and the female-to-male ratio in interventions that, if designed properly, do HDIs (1997) is 0.35, with an R square of 0.14. not need to be specifically targeted to women. 5. Equality of opportunity in education has Examples range from government policies received particular attention for two simple to promote export-oriented manufacturing reasons. First, education--and, more broadly, human capital--are easily incorporated into two Chirwa 2003; Doss and Morris 2001). A much frequently used econometric models of economic smaller number of studies find that female-headed growth: the augmented Solow model and the households are more likely to adopt new technolo- endogenous growth models. Second, educational gies than male-headed households (Bandiera and inequalities are easily measurable, and these mea- Rasul 2005). sures are widely available. 13. Unlike the studies of the impact of women's 6. See Knowles and others (2002) and Lorgelly control over resources using household surveys, (2000) for careful reviews of this literature. the studies of the impact of transfer programs are 7. Klasen estimates the effect of the gender gap argued to be free from the potential simultaneity in years of total schooling in the adult population between unearned or earned income and control on per capita income growth, using cross-coun- over household resources. try and panel regressions for the 1960­92 period 14. Following UNESCO (2004), parity is for 109 developed and developing countries. He defined as a female-to-male ratio exceeding 0.97. uses a variety of techniques to deal with poten- A ratio below 0.97 indicates significant female tial simultaneity between economic growth rates disadvantage. In 35 countries (of the 83 that and educational attainment, including instrumen- achieved the 2005 target), there was significant tal variables and the use of only initial levels of male disadvantage, with boys' gross enrollment educational attainment, which are not affected by rate lagging behind girls' (the female-to-male ratio growth in the subsequent period. exceeded 1.03). In these countries, mostly coun- 8. For a cross-country analysis of the impact of tries of East Asia and the Pacific, Europe and Cen- increased gender equality in earnings on household tral Asia, and Latin America and the Caribbean, savings and gross domestic savings, see Seguino boys' enrollment exceeds 90 percent. Thus a male and Sagrario (2003). disadvantage tends to occur in education systems 9. The countries in Deere and Leon's sample with overall high participation in schooling. are Brazil, Chile, Colombia, Ecuador, Honduras, 15. Net enrollment rates. Mexico, Paraguay, and Peru. The gender differ- 16. In Cambodia the share of women in the ence in average farmland size is statistically signifi- nonagricultural sector is about 53 percent, but cant in only two countries: Chile and Paraguay. the sector as a whole represents only 30 percent 10. Direct evidence is largely lacking that wom- of total (male and female) employment--so only en's disadvantage relative to men's with respect to 16 percent of all employed women are in nonag- land ownership translates into inferior outcomes ricultural employment. Compare this with Latvia, for women in investment, productivity, and indi- where women's share in nonagricultural employ- vidual incomes. Much more research is needed to ment is also 53 percent, but the sector accounts for understand the efficiency and welfare effects of 86 percent of total employment, implying that 46 this disadvantage for women as farmers and as percent of all employed women are in this sector. household heads. 17. The fourth indicator, the proportion of 11. An important exception is Latin America, seats held by women in national parliaments, where recent episodes of agrarian reforms and land is expressed as a proportion, but it is actually a titling programs recognized dual-headed house- measure of empowerment. While there has been a holds, conferred joint titles, and explicitly targeted lot written on women's empowerment, there is no female-headed households (Deere and Leon 2001). single accepted definition of empowerment. How- 12. When studies do examine how gender ever, there is significant overlap in the words used affects technology adoption, they typically do to define the term: options, choice, control, and so by including a variable for female-headed power--most often in the context of the ability of households as an additional covariate in multiple women to make decisions and affect outcomes that regression analyses. The empirical evidence on the are important to them and their families (Malhotra conditional relationship between the gender of the and others 2002). Self-efficacy is also frequently an household head and technology adoption is decid- element of empowerment; women should be capa- edly mixed. Most studies find that, controlling for ble of defining self-interest and choice and be able differing sets of relevant characteristics, female- and entitled to make choices (Chen 1992; G. Sen headed households are either less likely than--or 1993; Rowlands 1995; A. Sen 1999; Nussbaum as likely as--male-headed households to adopt 2000; and Kabeer 2001; cited in Malhotra 2002). new technologies (Asfaw and Admassie 2002; 18. This should not be interpreted as an argu- Paolisso and others 2002; Wier and Knight 2000; ment for enrolling girls at the expense of boys, once gender equality in enrollments has been 24. A better measure would express the number achieved; rather, it simply notes that in order to of pupils graduating from the last grade of primary measure the economic empowerment of women, school as a proportion of the total number of chil- absolute values matter, not just parity ratios. dren at the typical graduation age. But countries 19. Although an MDG6 indicator does measure often do not report the number of primary gradu- contraceptive prevalence rate, this is problematic ates. Another shortcoming of this measure is that the because it includes all forms of contraception, not primary school cycle varies greatly across countries. just modern forms. Although primary school in most countries lasts five 20. This is clearly the case not just for the to six years, there is a large variation in the length of MDG3 indicators, but for all MDG indicators. the primary school (3 to 10 years) (UN Millennium 21. This list of proposed indicators was culled Project Task Force on Universal Primary Educa- from nearly one hundred indicators of gender tion). This affects the comparability of the indica- equality that are currently used or recommended tor across countries. It remains a useful indicator for use by the UN and sister agencies. They cover to measure gender equality in education, because it education, health, employment, violence against captures both access and quality of schooling. women, legal rights, and political voice. Their 25. To be discussed at the March 2007 meet- wide scope underscores the multidimensionality ing of the Inter-Agency and Expert Group for of gender equality. MDGs. 22. The UN Millennium Task Force recom- 26. The adolescent (15­19) fertility rate, a mends replacing the existing four MDG3 indica- closely related indicator, is being considered for tors with 12 indicators, of which only two are inclusion in MDG5. current indicators (gross enrollment rates in pri- 27. In countries of Sub-Saharan Africa and mary, secondary, and tertiary education and per- Latin America and the Caribbean, however, teen- centage of seats held by women in the national age childbearing before marriage or union is com- parliament). Of the remaining ten proposed mon. For example, data from Kenya and Colombia indicators, only four currently have enough data in 2003 show that close to 20 percent of teenage availability to be serious candidates as indicators: mothers were unmarried. (1) ratio of female-to-male completion rates in 28. Based on Demographic and Health Surveys primary, secondary, and tertiary education; (2) data. adolescent fertility rate; (3) proportion of contra- 29. The UN Millennium Project Task Force has ceptive demand satisfied; and (4) share of women recommended the indicator "share of women in in employment, both wage and self-employment, employment, both wage and self-employment, by by type. The remaining six (gender gaps in earn- type." This chapter does not recommend using this ings in wage and self-employment; hours per day, indicator, although it is a valuable descriptive tool, or year, women and men spend fetching water and because it is difficult or impossible to interpret as a collecting fuel; land ownership, by male, female, measure of job quality. First, the share of women or jointly held; housing title, by male, female, or in any particular sector or employment must be jointly held; percentage of seats held by women put in the context of the overall importance of the in local government bodies; and prevalence of sector to the economy as a whole. Second, there is domestic violence) do not currently have sufficient enormous heterogeneity of job quality in each of data availability. For more details on these recom- the categories of employment; some self-employ- mendations, see UN Millennium Project 2005a. ment is well remunerated and stable, while other 23. Survival to grade 5 is another measure of self-employment is low-paid, unstable, and with primary school completion rate. This indicator is no employment benefits. An alternative indicator not suitable for monitoring gender parity regard- of "percentage of women (as a share of female ing completion, because it is based on the popu- population) in remunerative employment" was lation of children enrolled in primary school and explored, but not chosen because it was highly thus potentially excludes a large group of girls correlated with the existing MDG3 indicator of who never enroll in school. Girls who do enroll share of women in wage employment in the non- in school are more likely to be from advantaged agricultural sector. backgrounds, especially in countries where dis- 30. The reference period of the survey and the crimination against females is prevalent. Indeed, depth of questions that are asked influence the survival rates tend to be higher for girls than for estimates of women's labor force participation. In boys, in all regions (UNESCO 2004). developing countries, activities related to agriculture predominate in rural areas, and large informal mar- 37. Analytical work in support of this recom- kets predominate in urban areas, where production mendation and indicator was carried out by Women often is home-based and mostly unregulated. The in Informal Employment: Globalizing and Organiz- standard mode of eliciting information appropriate ing and the ILO, and reported in UNIFEM (2005). to developed country settings, therefore, is likely to 38. Collecting data on intimate partner vio- yield much poorer estimates of labor force participa- lence presents both methodological and ethical tion, particularly for women. Sociocultural practices challenges. Methodologically, there is a tradeoff can also affect data gathering. In strongly sex-seg- between the higher cost and greater accuracy of regated societies like those of South Asia, surveys stand-alone surveys on intimate partner violence using female enumerators to elicit information from (such as the recent WHO multicountry study) women are generally better able to gather data on a and the lower cost and lower accuracy (such as range of topics, including data on work performed underestimation of prevalence rates) of modules by women. Female enumerators tend to have bet- incorporated in other surveys, such as Macro ter access to women in the households selected for International's DHS surveys (Ellsberg and others the survey. In a setting where female work--espe- 2001). The principal ethical challenge of collecting cially paid work--has negative connotations, a male data on intimate partner violence is to ensure that respondent such as the household head is likely to women respondents and interviewers are protected under-report female participation in labor. from potential retaliatory violence from the perpe- 31. Exceptions in these regions are Bolivia, Bra- trators of violence. WHO has recognized this chal- zil, Nepal, and Uruguay where the female labor lenge and has adopted a set of ethical guidelines force participation rate exceeds 60 percent. for conducting population-based surveys on inti- 32. During the 1990s a number of countries mate partner violence (Watts and others 2001). in East Asia (such as Cambodia, Mongolia, and 39. Outliers are defined as countries that fall in Vietnam) and Europe and Central Asia underwent the highest or lowest 20 percent of the distribution a transition from a centrally planned economy to of scores on gender equality. a market-based one. This transition was expected 40. A global ranking cannot be done because over to affect gender wage differentials, but there is no 100 countries lack comparable data for 2000­05. consistent evidence of a widening or narrowing of 41. For reasons of space, trends in second- the gap. ary enrollment and literacy rates are not pictured. 33. If the ranking were basically identical, the Although they are not pictured, progress was notable principle of parsimony would argue for retaining for Algeria, Malawi, Nepal, Pakistan, and Sudan re the existing indicators. literacy rates, and for Algeria, Cambodia, Malawi, 34. Empirical analysis was also carried out for and Nepal re secondary enrollment parity. the official MDG3 indicators against four rec- 42. For reasons of space, trends in women's ommended indicators (female-to-male primary political participation, as measured by women's completion rate, under-five mortality, female-to- share in the national parliament, are not pictured. male labor force participation rate, and DALYs). The top-quintile countries started off at levels simi- Because the sample size shrinks to 37 countries lar to those of some bottom-quintile countries--the when labor force participation and DALYs are shares of Argentina, Mongolia, and Namibia were incorporated, the text discusses the comparison about the same as Guatemala, India, and Nepal. between the official and two of the proposed Since then, however, the top-quintile countries indicators (the female-to-male ratio of primary achieved much larger improvements in this dimen- completion rate and the under-five mortality rate), sion of gender equality. Two exceptions are Hon- with a sample of 54 countries. duras and Mongolia, which lost ground after about 35. We include only two of the proposed 1990 and ended the period with levels below those indicators because of sample size issues; were all of countries in the bottom quintile. In the bottom the proposed recommended indicators included quintile, Morocco and Pakistan stand out because in one scatter plot, the sample size would shrink they made more significant gains than all other bot- significantly. tom-quintile countries. Indeed, Pakistan compares 36. Underregistrations of births and deaths are favorably to many top-quintile countries. perhaps some of the most telling indicators of soci- 43. Law on Ensuring Equal Opportunities for etal exclusion. Women and Men, No. 5-XVI of February 9, 2006. Aid, Debt Relief, and Trade: Making Commitments Work D eveloped countries can help devel- of DAC members' planned future aid flows oping countries' progress toward provides scant evidence of an intended scal- the MDGs by delivering on commit- ing up of aid to Africa. DAC donors need to ments of more (and more effective) assistance accelerate the provision of aid that they have and by improving market access for these promised, and provide reliable information countries. The chapter assesses donors' per- on resource availability. formance by monitoring recent trends in the For scaling up, action is required of recipi- overall volume, allocation, and delivery of ent countries as well. Although scale-up aid; implementation of debt relief; and prog- opportunities exist in a broad range of reform- ress on global trade reform. ing countries, these countries face difficulties The expansion in global aid has stalled. in developing sound, results-oriented develop- After climbing to a record high in 2005, offi- ment strategies. Scaling up requires strength- cial development assistance (ODA) by mem- ening development strategies; identifying and bers of the OECD's Development Assistance addressing absorptive capacity constraints; Committee (DAC) fell 5 percent in 2006. and establishing closer links between devel- Most of the increase in ODA in 2005 reflected opment strategies and policy making, plan- exceptional debt relief operations to two ning, and budgeting processes. For their part, countries; less than a quarter represented net donors and international financial institutions transfers of new resources. A winding down need to be ready to support country efforts. of debt relief operations and a decline in other The Paris Declaration of March 2005 forms of aid pulled ODA lower in 2006, and raised expectations and generated a momen- official assistance is projected to fall in 2007. tum for change in aid delivery practices. The At the Group of Eight summit in 2005, results of the 2006 Baseline Survey show that DAC donors pledged to scale up their aid to a number of challenges need to be addressed: Sub-Saharan Africa. Nearly two years later, ownership of the Declaration by operational there is little indication of actual increases. In staff; demonstration of tangible benefits from fact, official assistance to the region--exclud- doing business differently; and deepening the ing Nigeria, one of the two recipients of harmonization and alignment of aid efforts exceptional debt relief--edged lower in 2005 at the country level. The rise of nontradi- and stagnated in 2006. Moreover, a survey tional donors, including private foundations, and a profusion of global vertical funds also tional donors all expanded their assistance heighten the need for alignment and harmo- to developing countries. Total aid then edged nization among a wider donor community. lower in 2006 and is expected to decline The past year saw major progress in slightly in 2007. DAC members continue to extending and deepening debt relief to the be the largest source of official assistance, but poorest countries. The Multilateral Debt so far much of the increase in their aid flows Relief Initiative (MDRI) has provided debt has added little to total aid as measured by relief of about $38 billion (in nominal terms) net transfers or by the availability of new to 22 countries. The MDRI commits donors resources for development. Moreover, these to providing additional resources to Interna- donors have been slow to translate their aid tional Development Assocation (IDA) and commitments--in particular, a promised dou- the African Development Fund (AfDF) to bling of assistance to Africa--into increases cover the losses to these organizations stem- in aid volume and tangible action plans. ming from debt forgiveness; the upcoming After climbing to a record $106.8 billion 15th replenishment of IDA will be an impor- in 2005, DAC members' ODA pulled back tant test of donors' intentions regarding their to $103.9 billion (preliminary) in 2006 (fig- support of the MDRI and of IDA. ure 4.1). The 5.1 percent decline in real terms As the pace of global integration increases, was the first drop in ODA in real terms since harnessing the new opportunities and man- 1997. Over 70 percent of the $25 billion real aging the risks places a premium for all coun- increase in net ODA in 2005 was due to higher tries on a trade strategy of greater openness, amounts of debt relief ($18 billion) resulting coupled with behind-the-border reforms to from implementation of the Paris Club agree- dismantle remaining barriers to trade. Donors ments for Iraq and Nigeria. (See box 4.1 for need to honor their commitment to increasing a discussion of debt forgiveness in DAC sta- their support of trade liberalization in devel- tistics.) By contrast, the expansion in other oping countries, or "aid for trade." Aid for bilateral ODA--that is, ODA minus special- trade needs to be focused on bringing practi- purpose grants such as debt relief, humani- cal solutions to countries' trade needs. tarian aid, and technical cooperation--was a The Doha Round of multilateral trade moderate $5.6 billion, or less than a fourth of negotiations continued to struggle in 2006. the total increase. The reduction in ODA in A small window of opportunity remains open 2006 resulted from the winding down of these for a deal to be reached in 2007, and flex- debt relief operations; other forms of ODA also ibility will be required from all sides. Con- contracted by nearly 2 percent in real terms. cluding the round remains an important step The latest numbers show that 17 of the 22 in efforts to achieve the Millennium Devel- DAC members met their Monterrey commit- opment Goals (MDGs) by the target date of ments on 2006 ODA targets. At 0.43 percent, 2015. Even as multilateral efforts faltered, ODA relative to DAC-EU donors' average preferential trade agreements continued gross national income was above the 0.39 per- to proliferate, raising concerns about their cent target set in 2002. These countries now impact on countries left out. account for close to 60 percent of DAC assis- tance. Overall, the share of ODA in donor GNI was 0.3 percent, below the level of the early 1990s. 2007 could see a noticeable fall in ODA as Trends in Aid Volumes and Instruments debt relief continues to decline. Other forms of Higher aid flows amid a changing aid aid will have to expand very rapidly in 2008­ landscape. Aid worldwide continued on an 10 for donor promises of an additional $50 bil- upward trend in 2005 as DAC members, lion in annual aid (over 2004 levels) to be met non-DAC traditional donors, and nontradi- by 2010. Based on announced commitments, Evolution of aid: 1990­2006 and prospects Source: OECD DAC Development Co-operation Report 2006 and DAC database. Note: In the second panel, data for 2010 are shown only for DAC donors with announced ODA/GNI commitments. Prospects are for DAC donors only and are based on these donors' public announcements. nearly a third of donors face an expansion in IDA-eligible), Iraq saw the largest gain and ODA/GNI of 50 percent or more (figure 4.1). 70 percent (24 out of 34) saw a decline. This is prompting concerns that donors may Contributions to multilateral institutions fail to deliver on their commitments. dipped in 2005, both absolutely and as a The continuing concentration of aid share of total ODA as bilateral aid surged increases in a handful of recipient countries through exceptional debt relief. The share of meant that aid to most countries rose very these contributions in DAC members' ODA slowly, if at all. Despite a nearly 55 percent has averaged about 30 percent in recent increase in real aid volumes in 2001­05, years, but in 2005 this share dropped pre- only 16 out of 81 IDA-eligible countries saw cipitously to 23 percent. Within this group, their ODA expand by 50 percent or more in the share of IDA and regional development this period (figure 4.2). The largest increases banks was sharply lower as well (at 6.5 per- were in fragile states such as Afghanistan, cent), which could undermine these finan- Republic of Congo, Democratic Republic of cial institutions' role in supporting poverty Congo, Liberia, Nigeria, and Sudan. Indeed, reduction. Donors have promised to provide 10 fragile states saw an expansion in ODA of IDA and the African Development Fund with over 50 percent. Just over half of low-income additional resources to compensate them countries actually saw an outright decline for debt service forgone under the MDRI, in aid received during this period. Among and this should translate into higher future lower-middle-income countries (that are not contributions. However, it is important that Expansion in ODA is concentrated in a few countries Source: OECD DAC database. Note: Other LICs are IDA-eligible countries that are not fragile states; LMICs comprise lower-middle-income countries that are not IDA-eligible. the pace of increase of contributions to these provided nearly $2 billion in assistance in multilateral development banks not constrain 2005, and indications are that these amounts the relative importance of these institutions will continue to expand. in the future. With their growing global economic prom- A wide range of other donors are increasing inence, countries such as Brazil, China, India, their aid to poor countries, and the amounts the Russian Federation and South Africa are of this aid are set to rise. Non-DAC ODA also becoming more important providers was $5 billion in 2005, reflecting a threefold of official support to poor countries.1 For increase over 2001 amounts. Several non-DAC example, China was the third-largest food OECD countries--including the Republic of aid donor in the world in 2005, and it is fast Korea, Mexico, and Turkey--have ambitious becoming a leading foreign creditor to Africa. plans to scale up aid, as do some EU coun- The Export-Import Bank of China, now one tries that are non-DAC members. ODA from of the world's largest export credit agencies, is these countries is likely to double by 2010, playing an important role in facilitating trade to over $2 billion (OECD 2007a). Other EU between China and countries in that region, states that are not OECD members will also as well as providing economic support. It see their ODA increase because of EU com- has recently expanded operations in Africa: mitments. Saudi Arabia (with an estimated over the past two years the agency commit- $1 billion) and other Middle East countries ted around $8 billion in loans and credits to A surge in debt forgiveness grants beginning in 2002 has drawn attention to their treatment in ODA statistics. The table below shows the amount (in nominal terms) of debt forgiveness grants provided in recent years. These grants (measured in gross terms) have ballooned from a modest $2.5 billion in 2001 to $25 billion in 2005. Depetris Chauvin and Kraay (2006, 2005) argue that the standard data do not provide a reliable estimate of the value of debt relief--that is, in present value terms--and they have developed PV estimates of debt relief. Another problem with DAC debt relief statistics is that forgiveness of outstanding amounts, debt service flows, and arrears is treated in the same way, even though the cash flow implications for borrowers' budgets is quite different. Despite these method- ological issues, DAC debt forgiveness statistics are widely used. Debt relief from the donors' perspective (budget effort) can be quite different from that of the recipients' perspective (availability of resources). One question that arises is whether ODA debt forgiveness grants represent additional flows (cross-border flows) to recipients. ODA Bilateral Debt forgiveness grants: 2001­05 (in US$ billions) Debt forgiveness grants Offsetting entries for debt relief Net debt forgiveness grants (A) (B) (A)­(B) 2001 2.51 0.54 1.97 2002 5.33 0.81 4.52 2003 8.44 1.58 6.86 2004 7.11 2.92 4.19 2005 24.96 2.43 22.53 DAC statistical guidelines allow debt cancellation to be reported as debt forgiveness when the action on debt occurs within the "framework of a bilateral agreement and is implemented for the purpose of promoting the development or welfare of the recipient."b Thus, forgiveness of ODA, other official flows (OOF), and private claims--principal, interest, and arrears--is captured in DAC statistics under "Debt forgiveness grants."c Appro- priate offsetting items (or counter entries) for principal and interest of each type of claim are reported, but not all are ODA flows--only forgiven principal on ODA loans is included under "Offsetting entry for debt forgiveness" in ODA flows.d Most of the debt forgiveness grants in DAC statistics represent forgiveness of OOF and private claims typically under the framework of the Paris Club. The counter entries are not ODA flows, so there is concern that recent debt actions assign a large amount of flows to recipients, that do not represent any new transfer of resources. This point is well illustrated by the 2002 Paris Club debt relief agreement for the Democratic Republic of Congo. The country had an unbearable debt burden and under reasonable conditions was clearly unable to meet its obligations to exter- nal creditors. The Paris Club agreement restructured $8.98 billion of debt--$8.49 billion in principal and interest arrears and $490 million of future payments.e Approximately $1.4 billion in outstanding claims were ODA loans. The country received Naples Terms--67 percent of commercial credits were cancelled and the remaining 33 percent were rescheduled; and ODA credits were rescheduled.f The resulting DAC data for ODA disbursements in 2003 (when the bulk of relief granted under the Paris Club agreement was reported in the DAC statistics) show debt for- giveness grants of $4.441 billion and offsetting entries for debt relief of only $4.9 million. Together, these two items account for $4.44 billion of net ODA flows. Yet, the country did not receive additional resources anywhere near to this amount. However, the country's debt burden was substantially reduced and it was able to normalize relations with the international community, improving its prospects for growth. Although debt cancellation may not deliver additional flows to borrowers, it does reflect government budget effort. The extent of the current budget effort will depend upon the terms of government guarantees for export/ commercial credits and on the timing of write-offs for official loans--some may have been already written down.g continued Because of differences in practices across donors, the extent of the budget effort for a particular debt action varies across countries. a. A whole host of debt actions are presented in DAC statistics. The focus here is on debt forgiveness. b. OECD DAC "Handbook for Reporting Debt Reorganization on the DAC Questionnaire" and "DAC Statistical Reporting Directives." c. Reorganization of OOF and private claims within the framework of the Paris Club often involves concessionality in the form of debt reduction, debt service reduction, and capitalization of moratorium interest. The cancellation of part of the claims (or the amount equivalent to the reduction in net present value) is treated as debt forgiveness in ODA with no offsetting items in ODA flows. Amounts of OOF and private claims that are rescheduled are not part of ODA and are included as "Rescheduling" loans under OOF flows. d. Forgiven OOF principal is reported under "Offsetting entries for debt relief" in OOF flows and forgiven private principal is accounted in "Offsetting entry for debt relief" under private flows. There are no offsets to forgiven interest in ODA, OOF, or private flows. Instead, appropriate counter entries "Offsetting entry for forgiven interest" are to be noted in memo items--the data for which are usually incomplete. The result is that the treatment of debt cancellation in ODA statistics assigns a larger amount of net flows to recipients than amounts actually received. e. Paris Club Press Release of September 13, 2002. f. In November 2003 the country received Cologne Terms from Paris Club donors. g. Also see the OECD's Development Cooperation Report 2006. such countries as Angola, Ghana, Mozam- for borrowers to adhere to established inter- bique, and Nigeria (Moss and Rose 2006). national reporting standards. Creditor and India's export credit agency, Exim India, has borrowing countries alike are urged to use also issued lines of credit totaling $558 mil- the joint Bank-Fund framework for assessing lion to West African countries to enhance its debt sustainability in low-income countries, to commercial relations with the region. Little appropriately manage the associated risks. is known about the size and composition Private flows to developing countries have of flows from emerging donors, and better shown strong growth in recent years, but most information is needed to facilitate monitor- low-income countries remain heavily depen- ing and donor coordination. dent on grants and concessional finance (see Financing by emerging donors is targeting chapter 5). Private giving through founda- productive sectors and physical infrastruc- tions, charities, and other nongovernmental ture, areas that traditional donors have largely organizations is on the rise as well. Reported exited. Not all of this financing is in the form aid flows from private citizens more than dou- of aid; rather it represents a mix of conces- bled over 2001­05, reaching $14.7 billion. sional and nonconcessional funds. There is a Private sources have a significant role in mobi- concern that access to large amounts of funds lizing resources and setting policies for certain from these newer donors may strain recipients' activities, including humanitarian and disaster capacity to use additional resources effectively. relief and research into vaccines and tropical Low-income countries that have only recently diseases. For example, the Bill and Melinda received major debt reductions through the Gates Foundation has contributed over $6.6 Heavily Indebted Poor Countries (HIPC) Ini- billion for global health programs, $2 billion tiative and the MDRI may see yet another of which is for fighting malaria, tuberculosis, buildup of debt (see the section on debt relief and HIV/AIDS and other sexually transmitted below).2 This reinforces the need for good diseases. The growing role of private donors practices in accounting, reporting, and trans- calls for closer coordination and exploita- parency on the part of all donors and the need tion of possible synergies between official and nonofficial donors.3 [For a detailed discussion middle-income countries. The World Bank of private capital flows and remittances see is working with other MDBs to develop a Global Development Finance 2007.] menu of blending arrangements that would deploy part of current and future bilateral aid P R O G R E S S O N I N N O V A T I V E F I N A N C I N G to middle-income countries so as to leverage A number of innovative financing propos- the MDBs' nonconcessional lending. als for both the public and the private sector Various private entities, meanwhile, have are being developed, and some are already proposed a wide range of innovative financ- in the pilot stage. Some of the proposed new ing measures. These include electronic billing- mechanisms could contribute to expanding based fundraising; global development bonds, aid flows. which would use financial engineering tech- On the public sector side, new forms niques to reduce the risk of developing-country of taxation and securitized borrowing are investments; and results-based sequencing of being deployed. The International Finance funds, in which a country would receive addi- Facility for Immunization (IFFIm) issued tional funds only after specified targets have its first $1 billion bond last November, part been met (de Ferranti 2006). Together these of a plan to raise $4 billion over the next mechanisms are mobilizing new contributors 10 years. The bond is backed by six donor to development finance in both the private countries--France, Italy, Norway, Spain, and the public sector. Although this is surely Sweden, and the United Kingdom--and will a welcome development, it also heightens be repaid over 20 years. This new financ- the need for stronger cooperation and col- ing tool accelerates donor contributions laboration among all actors so as to deliver to programs under the Global Alliance for resources more efficiently and effectively. Vaccines and Immunization, and it helps to lock in associated aid flows, which usually Progress on Scaling Up Aid to Africa: depend on annual budgets. Actions Lag Commitments Air ticket levies could yield an estimated $1 billion to $1.5 billion a year, with France At the 2005 Gleneagles summit, the Group one of the largest contributors at $250 mil- of Eight leaders promised to double aid to lion a year. Views are mixed, however, on Sub-Saharan Africa by 2010. Nearly two whether these levies add resources to what years later, that promise is not translating would otherwise have been provided. Mean- into actual increases: Sub-Saharan Africa while (in February) donors launched the is seeing little new aid beyond debt relief first pilot Advance Market Commitment, to and certain special initiatives (figure 4.3). provide incentives for the development of Indeed, DAC members' ODA to Sub-Saha- vaccines of importance to developing coun- ran Africa--excluding Nigeria, the recipient tries.4 Canada, Italy, Norway, Russia, the of exceptional debt relief--actually declined United Kingdom, and the Gates Foundation by about 1 percent in real terms in 2005; pre- provided $1.5 billion in commitments to the liminary data show that these flows stagnated pilot. Although AMCs do not increase aid in 2006. The prominence of debt relief in aid flows to poor countries (especially not in the flows is evident during 2001­05: bilateral short run), by supporting the development of ODA to the region more than doubled over vaccines they could prove helpful in meeting this period and the share of ODA allocated the MDGs. to Sub-Saharan Africa increased, but close to Blending, or the use of a combination of 70 percent of the expansion represented debt financing mechanisms toward a common relief. The Bank's Africa Action Plan progress objective, is seen as a way to augment MDB report aptly notes that "A typical `well per- and donor resources aimed at accelerating forming' African country has seen little or no progress on poverty and on social goals in increase in the resources available to support Evolution of Net ODA to SSA, 1990­2005 Source: OECD DAC database. development projects and programs." ODA Monitoring Reports have addressed some of from several of the largest bilateral donors these issues and highlighted the importance to the region shows a similar pattern--the of sequencing interventions across the range exception is the United States where human- of constraints and of aligning public invest- itarian assistance and technical coopera- ments with these priorities.5 As well, there is tion account for the bulk of the expansion a need to establish closer links between the in ODA. Moreover, a partial survey of DAC PRS and decision making processes in order members' planned future aid flows by coun- to strengthen accountability to domestic try and region provides scant evidence of any stakeholders. One way of achieving this is by substantial scaling up of aid on the horizon. integrating the PRS and the budget process, Without concrete action to further boost aid, thereby strengthening results orientation and the Gleneagles promise risks going unful- domestic accountability.6 filled. A range of reforming countries is well Action is needed on both sides of the aid positioned to absorb scaled-up aid. The Bank relationship. Scaling up requires recipient teamed up with the OECD-DAC and UNDP countries to strengthen nationally owned pov- to focus on six well-performing low-income erty reduction strategies (PRSs), building upon countries in Africa which, because of their the solid progress of recent years. Particular improved economic performance and better- attention must be on enhancing analytic capac- developed aid alignment and harmonization ity to identify and assess absorptive capacity mechanisms, were readier for scaled-up donor constraints--macroeconomic, sectoral, and support. Within these countries, selected sec- institutional--and develop appropriate inter- tor programs were costed to demonstrate ventions to alleviate them. Previous Global that 15­30 percent of additional aid could be absorbed in these countries. Another recent The Africa Catalytic Growth Fund (ACGF) World Bank study (2007b) draws on 12 is also a possible vehicle for scaling up. The country case studies to identify a variety of ACGF is a new mechanism to provide targeted fiscal constraints to growth and to assess the support to help countries scale up to achieve financing options to achieve higher growth. economic growth and hard-to-reach MDGs. Among countries with high aid access, such Within a country-based framework, the ACGF as Madagascar, Rwanda, and Uganda, the identifies opportunities where resources from study finds that physical public infrastruc- multiple sources can fill funding gaps.7 ture is both a critical constraint to growth and to achievement of the MDGs. Because of Donor Support for Gender Equality: structural constraints (small size of the for- Toward a More Realistic Agenda mal sector, high dependency ratios, etc.) the study assesses the scope for generating addi- At the 1995 Beijing Women's Conference, tional domestic revenue to be modest. Given donors made commitments to focus actions the need to restrain new borrowing, and and resources on promoting gender equality relatively limited scope to capture efficiency in the developing world through mainstream- gains, these countries will have to rely sub- ing of gender issues and women's empower- stantially on scaled-up aid to finance growth- ment. Gender inequalities in the areas of enhancing expenditure. The study finds that education and health were of particular the situation is quite different in some other concern. Efforts to redress this situation are countries--Kenya could undertake limited apparent in the pattern of aid allocations for borrowing for key investments but will need 2001­05: nearly three-fourths of aid with a to address governance concerns to access gender equality focus is directed to the social greater aid flows; Tajikistan will need to sectors (figure 4.4). A quarter of all bilateral improve the regulatory and governance envi- ODA allocated by sector--$5 billion out of ronment to attract private investment and to $20 billion in average annual commitments-- access more aid. In other cases, such as India, is focused on gender equality. Because aid increased revenue effort and changes in the for activities with no explicit gender equality composition of expenditure will be needed to focus, such as infrastructure, can also have address constraints to growth. a beneficial impact on girls and women, it is The challenge is to establish an effective hard to say what the appropriate amount of mechanism for scaling up. Donors have come financing for gender equality should be. What to see results and resources (R&R) processes can be said is that the share allocated to gender as important to facilitating the scaling up of equality is increasing.8,9 aid within the country-based development Despite donors' strong policy commitments model. The R&R process builds and improves to gender equality, implementation has been on the consultative group meetings that are the disappointing. Self-evaluations of nine donor existing mechanism for donor coordination. It agencies' performance reflect a gap between proposes linking funds to ambitious country- words and deeds.10 One reason for the short- owned strategies and development results in a fall has been an unfortunate diffusion of framework of mutual accountability. Although responsibility, the result of initial decisions to the underlying principles are common to the spread responsibility for gender mainstream- countries involved, the specific modalities may ing across all staff, with little accountability. vary from country to country. Several African Agency staff found themselves suddenly tasked countries are in the process of preparing well- with mainstreaming gender issues in the proj- defined and well-costed-out programs for using ect portfolio, yet nobody was held specifically the additional aid. The example of Ghana illus- responsible for making it happen. A second trates how one country is implementing the reason was the broad, ambitious scope of the results and resources agenda (box 4.2). gender mainstreaming mandate itself, which Within the results and resources framework, work on the scaling-up agenda is now under way in the initial focus countries, with several R&R meetings planned in 2007. The first such meeting for Ghana took place in June 2006. Although the specific modalities of the R&R process will vary from country to country, the Ghana experience provides a good illustration of the key elements. Country ownership and leadership have been at the center of the process in Ghana. The R&R meeting was organized under the leadership of the Ghanaian government, closely assisted by the World Bank. An ambitious, country-led Growth and Poverty Reduction Strategy II (GPRS II), covering the period 2006­09, provided the strategic framework. The strategy is focused on raising economic growth from its recent 5 to 6 percent a year to between 8 and 8.5 percent, and it sets out clear development goals and identifies policies and programs to achieve them. GPRS II encompasses an initial costing of the country's whole development strategy, linking development goals to an investment plan and a medium-term expenditure framework. This is com- plemented by a resource assessment, evaluating the macroeconomic and fiscal space for the scaled- up development strategy and outlining the planned mobilization of domestic resources and the amount and composition of resources to be raised externally. GPRS II also included an assessment of, and articulated a plan for strengthening, the country's institutional absorptive capacity, with particular attention to public resource management and the monitoring and evaluation system, to ensure value-for-money and efficiency of expenditure. GPRS II was supplemented by several key documents prepared specifically for the R&R meeting: these included a synthesis paper setting out Ghana's enhanced growth strategy, the macroeconomic context, and specific scale-up opportuni- ties; a results matrix linking expected outcomes, resources, and indicators for monitoring progress; a financing matrix detailing recent and projected donor support; and a matrix on Ghana's Harmo- nization and Aid Effectiveness Action Plan. The World Bank worked closely with the government and with other development partners in preparing these documents. Ghana's strategy for scaled-up development, with its emphasis on stronger economic growth, was well received by donors at the R&R meeting. Donors confirmed significantly increased and more predictable financial assistance, including $5 billion in new disbursements ($800 million more than had been projected six months earlier at a consultative group meeting) and $1.3 billion in debt relief under the HIPC Initiative and the MDRI, for a total of $6.3 billion in identified partner support for the four-year GPRS II period. The government noted that additional financing of about $2 billion ($500 million a year, or about 4 percent of the country's GDP) would be needed to fully fund the scaled-up investment plan, especially its infrastructure components, to achieve the GPRS II targets for accelerated growth. Donors committed to a full review of the financing requirements of GPRS II, based on further work to flesh out and assess the government's enhanced growth strategy and the associated investment plan. This work is now in progress, and the World Bank is contributing through a country economic report focused on the agenda for accelerated growth. In keeping with the Aid Effectiveness Action Plan, Ghana's development partners, including the Bank, have prepared a Joint Assistance Strategy that establishes a framework for support of GPRS II. The next R&R meeting on Ghana, scheduled for June 2007, will assess program implementation over the past year against the agreed results and progress indicators. It will also consider the financing requirements that remain unmet. Future R&R meetings are envisaged on an annual basis. sought to encompass all actions throughout in tracking financial resources not specifically a given agency. This only reinforced inac- dedicated to or earmarked for gender equal- tion when the mandate was not translated ity objectives. into concrete priorities. Compounding these The evaluations did find some successes in problems were inconsistent or limited support mainstreaming gender issues in operations, from senior management levels, and difficulty although few projects systematically measured Gender equality focus of bilateral ODA by sector (2001­05) Source: OECD DAC. Note: Not all donors report the gender focus of their bilateral aid. the results and impacts. Examples include the ing and specific actions to reduce gender embedding of gender issues in country dia- inequalities and empower women. Proposed logues and Poverty Reduction Strategy Papers changes include the following: much greater (PRSPs) as well as in specific development selectivity in targeting efforts at gender main- projects, especially those where an enabling streaming; adoption of a results orientation environment supported the gender equality and strengthening of monitoring and account- agenda. In Tanzania, for instance, the United ability frameworks; stronger organizational Kingdom through its development agency the arrangements for gender mainstreaming; and Department for International Development exploitation of synergies with the aid effec- supported the development of gender indi- tiveness agenda. This last change links gender cators in the poverty monitoring system. As equality instrumentally to the effectiveness a result, progress in gender equality is now of aid. It builds gender dimensions into the systematically measured as part of Tanzania's results and country ownership frameworks overall poverty reduction performance assess- defined by the Paris Declaration, while recog- ment. In Nicaragua, the Swedish International nizing the challenge of mainstreaming gender Development Cooperation Agency and other issues into programs and budgets. donors were instrumental in helping main- Within this more realistic agenda, there stream gender issues in health sector projects. is wide agreement that high-level leadership, These evaluation exercises, together with a technical expertise, and financial resources more favorable political climate, have helped will be key to ensuring that donor agencies' reenergize the commitment of donors to gen- gender policies are implemented. Financial der equality and women's empowerment. Sev- resources, in particular, are needed up front, to eral donors are in the process of fine-tuning enhance the capacity of donor and implemen- or revamping their approaches, recognizing tation agencies to mainstream gender issues. the importance of both gender mainstream- Beyond these internal challenges, success in implementing gender equality policies will excludes Afghanistan because CPIA data are depend largely on recipient countries' inter- not available for the country and Iraq because est and institutional capacity. it is not in the group of 81 IDA-eligible coun- Despite the challenge of estimating the tries. Including these two countries could alter financial resources needed to achieve gen- the results. There is some variation in the dis- der equity goals, a few countries, such as tribution by type of aid: nearly half of flexible Ethiopia, Gabon, Kenya, Mauritania, Niger, ODA (ODA not in the form of special-purpose Senegal, Tajikistan, Togo, and Yemen have grants) is directed to the top third of recipients, drafted a variety of innovative gender inter- whereas debt relief is concentrated in the mid- vention proposals with corresponding cost dle third. Donors also focus on recipient-coun- estimates.11 These countries' proposed gen- try governance, and here the data exhibit a der mainstream interventions cover both somewhat similar pattern: flexible ODA tends traditional gender sectors such as health and to be allocated to countries with relatively bet- education and nontraditional sectors such as ter quality of governance, and debt relief to energy and infrastructure.12 those with relatively weaker governance. Gender mainstream intervention proposals Where the relationship between aid alloca- vary across countries contingent on national tion and policy performance and poverty has needs, priorities, and sectoral development of been studied empirically, bilateral donors are gender action. For example, Kenya calculates found to be increasingly focused on these crite- $2.41 per capita per year to provide energy ria. Both policy and poverty selectivity indexes subsidies to female-headed households to exhibit an improving trend over 2001­05 facilitate income generation through biomass (figure 4.6). This is true for each of the vari- and renewable energy, petroleum, and electric- ous types of aid as well. The policy selectiv- ity. In Niger, an intervention proposes provid- ity index in 2005 for non-humanitarian ODA ing vocational and skills training to facilitate shows that, on average, a 1 percent increase secondary school female graduate entry into in a country's policy and institutional quality the workforce with a projected annual cost (CPIA) score is associated with a more than 1.5 of $2.13 per capita. The Dominican Republic percent increase in ODA (with bilateral donors proposes transferring subsidies to mothers of showing less selectivity than multilateral ones; children in pre-primary, primary, and second- see chapter 5). Studies of the responsiveness ary education at an annual per capita cost of of aid flows to improved governance report $1.78. Outlining financial requirements is an fairly similar results: a 1 percent improvement important first step in targeting gender action in the quality of governance is associated with priorities and costs to which national govern- about a 1.4 percent increase in aid. The results ments and donor agencies may respond. [See for disaggregated aid show that flexible ODA box 3.5 for progress on gender budgeting.] is the most selective, and technical coopera- tion the least selective, with respect to policy Selectivity in Aid Allocations performance (and no significant relationship is found between CPIA scores and debt relief). A S H A R P E R F O C U S O N P O L I C Y The results for poverty selectivity are similar: Although aid is allocated on the basis of several flexible ODA is again the most selective with criteria, donors are becoming more focused respect to poverty, and technical cooperation on policy performance. For example, the dis- the least selective. tribution of DAC bilateral aid by the quality of countries' policies and institutions (CPIA) A I D T O F R A G I L E S T A T E S : shows that the best-performing third of recipi- A D D R E S S I N G T H E C H A L L E N G E S ents receive a modest 40 percent of all aid, while the worst-performing third receive only The world's fragile states with their multi- a little over a fifth (figure 4.5).13 The analysis tude of chronic problems pose a particular Quality of policy matters: Distribution of 2004­05 DAC bilateral ODA Source: ODA data from OECD DAC database; CPIA data from World Bank. Note: IDA-eligible countries are divided into three groups--bottom, middle, and top--using the 33.3 and 66.7 percentiles of overall CPIA or governance-CPIA (quality of public sector management and institutions). Afghanistan is not included in the above charts because CPIA data are not available for the country; Iraq is not in the group of IDA-eligible and is not included. Sharper donor focus on policy and need Source: Staff estimates, based on Dollar and Levin (2004). Note: The quality of policies and institutions is measured by the overall CPIA. Policy selectivity shows the policy selectivity index, which measures the elasticity of aid with respect to the CPIA. Poverty selectivity shows the poverty elasticity index, which measures the elasticity of aid with respect to recipients' per capita income. The selectivity results do not include Afghanistan and Iraq because data on CPIA are not available. challenge to the international community, is now under way to consider how to make the but progress has been made in understand- hoped-for collaboration a reality. The princi- ing and responding to these situations. In ples also have organizational implications for 2005 donors endorsed 12 principles for donor agencies; for example, donors will have engagement in fragile states. Since then to create internal capacity to respond quickly these principles have been piloted in nine to changing environments, build an appropri- countries, and their experiences have been ate local presence in fragile states, and attract fed back to allow the principles to be refined skilled staff to work in these countries. Finally, further (OECD DAC 2006b). The 12 prin- appropriate performance indicators are needed ciples emphasize the need to distinguish so that progress can be monitored against the between country situations and to custom- objective of building lasting peace among a ize the mix and sequence of aid instruments more robust family of nations. accordingly. The central focus of interna- The focus on fragile states is beginning to tional engagement in these countries should translate into increased assistance, although be on state building: supporting the legiti- aid flows to this group continue to be vola- macy and accountability of the state and its tile. Overall aid to fragile states rose by more capacity to foster development. The inter- than two-thirds in 2005 alone, to nearly $20 dependence among political, security, and billion (in 2004 dollars), and by 167 per- development objectives requires that donors cent over 2001­05, just over half of which devise coherent policies to deal with the consisted of debt relief (figure 4.7). When multidimensional challenges these countries debt relief and humanitarian assistance are face. Where possible, donors should align excluded, aid to fragile states was around their assistance with the recipient govern- $10 billion in 2005.14 Fragile states are also ment's own priorities, and avoid actions that seeing an improving trend in aid received undermine long-term capacity building. The per capita. Although this group continues principles call for coordination among donors to receive less aid (excluding debt relief and in making assessments, designing strategies, humanitarian assistance) than the group of and assigning tasks. They also call on donors other low-income countries, the gap is nar- to act flexibly, to be prepared to engage over rowing (figure 4.8). Large increases in aid a longer horizon than they do with other low- to Afghanistan, Sudan, and the Democratic income countries, and, specifically, to address Republic of Congo are behind this narrowing the problem of aid orphans--those countries trend. Meanwhile, more than half of fragile that for whatever reason get less assistance states actually saw a decline in aid from 2001 than their development indicators warrant. to 2005 (figure 4.2). Aggregate trends mask Much remains to be done to improve devel- the wide variation across different types of opment effectiveness of aid in fragile states. fragile states: those emerging from violent For one thing, application of the recently conflict typically receive much more aid than agreed principles needs to be extended other fragile states, and more than other low- beyond the few pilot programs to all fragile income countries. The DAC is therefore mon- states. The principles also need to be main- itoring annual resource flows to fragile states streamed with efforts to implement the Paris to help identify those where international Declaration. International actors need to engagement seems imbalanced. The monitor- adopt "whole of government" approaches, ing exercise conducted in June 2006 identified fostering close collaboration across the eco- eight countries that appear to be underfunded nomic, development, diplomatic, and security and another three where aid is volatile and, on fields. The support provided by the interna- balance, falling (OECD DAC 2006c). The aim tional community to the creation of the United of the monitoring exercise is to use the find- Nations Peace-Building Commission demon- ings on marginalized states to inform the con- strates the importance of this nexus, and work sultative group meetings for these countries Fragile states receive more of their aid in the form of debt relief and humanitarian assistance Source: OECD DAC database. Note: Other LICs are IDA-eligible countries that are not fragile states. Aid per capita to fragile states Source: OECD DAC database and staff estimates. Note: Data exclude India. Aid is net ODA less humanitarian aid and debt relief. and address the local bottlenecks to greater as a group continue to be underfunded com- engagement by donors. pared to other countries with similar policy Despite an overall increase in aid flows, and poverty characteristics. However, the empirical evidence shows that fragile states extent of underfunding (as measured by the coefficient on the fragile states dummy) of having an action plan. Along with the appears to be declining in recent periods.15 broad dissemination within donor agencies, Evidence of underfunding is stronger for cer- a majority of DAC members have provided tain types of aid, especially flexible ODA. training courses on the commitments under More important, the role of other factors the Paris Declaration to staff; some agencies such as regional and global spillovers and have advanced training in budget support, vulnerability has yet to be carefully exam- sectorwide approaches, and public financial ined in assessing the allocation of aid to this management. Donors have also collaborated group of countries (Amprou, Guilliaumont, with partner countries to disseminate Paris and Guilliaumont Jeanneney 2006). principles through regional workshops. But implementing the Declaration involves Focus on Aid Effectiveness and Results a broad and complex process of change. In terms of organizing to strengthen internal pro- P R O G R E S S O N H A R M O N I Z A T I O N , cesses, half of the donor agencies are focusing A L I G N M E N T , A N D R E S U L T S on mainstreaming responsibility for imple- This section provides an update on the imple- menting the aid effectiveness agenda while the mentation of the harmonization and align- other half are relying on specialized units to ment actions at the global and country level in promote implementation. Progress on decen- light of mutual commitments made by donors tralization is slower, however, with less than and partner countries in the Paris Declaration a third of DAC members reporting that field to improve aid effectiveness (actions by bilat- representatives are taking responsibility for eral donors are reviewed here, while chapter advancing the Paris agenda on the ground. 5 discusses implementation status of MDBs). At the country level, a growing num- Based on qualitative reviews and the Baseline ber of partner countries are collaborating Monitoring Survey of partner countries and with donors on harmonization and align- donors,16 it is evident that the international ment actions. Six countries--Burkina Faso, aid community is taking tangible actions Ghana, Mozambique, Tanzania, Uganda, toward meeting the Paris commitments but and Vietnam--are taking actions and mak- that results to date are modest. The Paris Dec- ing substantive progress across a broad range laration has raised expectations and generated of areas.18,19 An additional 13 countries are a momentum for change in aid practices, but a taking action and are making good progress. number of challenges still need to be addressed In the other countries, efforts are under way by partners and donors if the targets of the to implement harmonization and alignment Paris Declaration are to be met. actions but this progress is not as broad as in Following the adoption of the Paris frame- the more advanced countries. work in 2005, donors have taken a broad Although the findings are preliminary, the range of actions to disseminate the agreed qualitative and quantitative monitoring of the commitments. Two-thirds of DAC donors Paris Declaration20 provides a useful assess- have included the Paris Declaration as a stra- ment of the state of play on the basic trends tegic priority in official statements, indicat- in implementation of the commitments. The ing political ownership.17 Many donors have qualitative part of the 2006 baseline review also developed action plans for implement- finds that progress on ownership by part- ing the Paris framework. As of November nership countries is uneven. While countries 2006, 16 DAC members had adopted an increasingly have developed comprehensive action plan on implementing the Paris Dec- national strategies and improved the links to laration. An additional three members have operational frameworks, as well as enhanced adopted the principles of harmonization, the functioning of country systems, there is alignment, and results in their aid strategies significant room for improvement, even in and policies, and four others have intentions the better-performing countries. According to the review, only 17 percent of the surveyed utilization of country systems by donors. Sus- countries had developed operational national tained technical assistance from donors, when development strategies--strategies linked to coupled with adequate country ownership of fiscal policy and budget processes--that are PFM reforms, has led to improvements in considered "largely developed towards good PFM systems--for example, reform efforts practice," compared to a target of 75 percent in Mozambique and Tanzania have been well of partner countries with largely or substan- coordinated and owned by the countries, and tially developed strategies. Better use of the have resulted in significant improvements national budget to allocate resources in a in their PFM systems. The survey also finds more vigorous and consistent way to agreed that very few countries have a mechanism for policy priorities and activities is needed if mutual review of implementation of aid effec- countries are to close the gap on this target. tiveness commitments (those in the Paris Dec- Furthermore, to align aid with country sys- laration and in local agreements). As more tems requires reliable country systems--for countries develop and adopt harmonization example those for public financial manage- action plans, this is expected to change. ment (PFM) and public procurement sys- According to the 2006 Baseline Survey, tems. Nearly a third of countries have public the status of implementation against the financial systems that are moderately weak Paris indicators for which donors have pri- or worse--a CPIA-PFM quality score of 3 or mary responsibility is mixed; the breadth and less--and about a third have PFM systems depth of actions vary among donors. Donor that are moderately strong or better--a CPIA- effort to align aid with country policies and PFM quality score of 4 or higher (figure 4.9). processes is measured by several indicators The Paris target for this indicator calls for half including use of country systems, less reliance of the partner countries to move up at least on creating parallel project implementation half a point by 2010. Strengthening PFM and units, disbursing on time and on budget, and procurement systems is central to increasing coordinating support to strengthen capacity. Quality of country public financial system and use of PFM system for aid to government sectors Source: Preliminary results from the 2006 Survey on Monitoring the Paris Declaration, OECD/DAC (March 1, 2007); CPIA scores from World Bank. Table 4.1 shows the results from the survey that 48 percent of their capacity-building for bilateral donors and all donors for each of support is coordinated with other donors, these indicators. as compared to the Paris target for 2010 of Among bilateral donors, 39 percent of 50 percent. This is one area where the defini- their disbursements use the partner country's tion of "coordinated" needs to be reviewed public financial management system and 42 to ensure greater consistency or the target for percent of disbursements use the country's this indicator might be more ambitious. procurement system. Better quality of PFM The survey finds that the bilateral donors systems (as measured by the CPIA) do tend provide 40 percent of their aid through pro- to be associated with a higher use of these gram-based approaches such as budget sup- systems for aid that is provided to the pub- port and sectorwide approaches, relative to lic sector--correlation of quality with use is the Paris target of 66 percent; there appears just over 0.4 (figure 4.9). Nevertheless, there to be an increasing trend for sectorwide is wide variation among bilateral donors approaches and other similar programmatic (DAC members) on the use of PFM systems arrangements, although reliance on such pro- for delivering aid to government sectors-- grammatic approaches outside the social sec- from a low of 10 percent to a high of 90 tors is not widespread. An area where greater percent. attention is needed is with respect to project According to the survey, 67 percent of implementation units (PIUs); in the 34 coun- bilateral aid is reported to be disbursed dur- tries surveyed 1,005 parallel PIUs are relied ing the fiscal year in which it is scheduled, upon to implement projects, with a Paris tar- though with considerable variation at the get to reduce this number to 335 by 2010. country level in the shortfall between planned In addition, the extent to which bilateral and actual disbursements. (Also see box 4.3.) donors conduct joint missions is low--the For another indicator, bilateral donors report survey found that 24 percent of missions Indicators pertaining to bilateral donors' implementation of the Paris Declaration Indicator Bilateral Donors All Donors 2010 Target Strengthen capacity by coordinated support 48% 47% 50% Percentage of aid that is disbrused using 39% 39% Reduction of aid not country public financial management systems using country PFM systems by a third or more Percentage of aid that is disbursed using 42% 39% Target under development country procurement systems Number of Parallel Implementation Units (PIUs) 1,005 1,767 Reduction by 2/3 Percentage of aid that is disbursed on time 67% 66% Reduction by 50% of aid not disbursed on time Percentage of aid that is disbursed through 40% 43% 66% program-based approaches Percentage of missions that is done jointly 24% 24% 40% with other donors Percentage of country analytic work that is 48% 54% 66% done jointly with other donors and/or partner government Source: Preliminary results from the 2006 Survey on Monitoring the Paris Declaration, OECD/DAC (March 1, 2007). Note: The data reflect implementation as of 2005 for donors that provided a total of more than $100 million for the government sector. The data are undergoing final review by the OECD/DAC. Aid predictability has become a central issue in the quest for enhanced effectiveness, particularly in the discussion on how best to deliver untied budget aid. For countries whose budgetary spending depends heavily on disbursements of untied aid, volatile and unpredictable disbursements are seen as undermining the credibility and reliability of short- and medium-term budget planning, by ren- dering original allocations obsolete and forcing expenditure adjustments during execution.a These adjustments, in turn, can hamper the attainment of government objectives, most importantly by disrupting the implementation of poverty reduction strategies. Worse still, when significant adjust- ments are simply not feasible during a given budget year, the result can be deviations from macro- economic targets, with potentially significant consequences for macroeconomic stability. A recent study of the predictability of budget aid in eight African countries--Benin, Burkina Faso, Ghana, Mali, Mozambique, Senegal, Tanzania, and Uganda--finds that both negative and positive errors in the International Monetary Fund (IMF) aid projections are large, imposing burdens on bud- get management (Celasun and Walliser 2005). While weak donor reporting is an issue in aid predict- ability, another likely reason for aid shortfalls is the failure by recipients to meet disbursement triggers. Moreover, the delayed disbursements may be shifted to the following year, once the conditions are met, accounting for aid overruns. On average, the mean absolute error in projecting budget aid on a year-by-year basis was about 1 percent of recipient-country GDP during 1993­2004, or about 30 percent of actual aid received.b Errors have declined only slightly over time, from 1.13 percent of GDP during 1993­99 to 0.95 percent of GDP during 2000­04. Closer donor coordination and informa- tion sharing is needed to help improve the predictability of aid. Also, providing more donor financing through the budget would strengthen monitoring and forecasting. The bulk of adjustments to these unforeseen variations in aid take the form of domestic bank financ- ing of government, arrears, and changes in domestically financed investment expenditure. Little of the adjustment burden falls on recurrent expenditure, amortization, or other financing. It is also apparent that, for whatever reason, aid shortfalls tend to be accompanied by tax revenue shortfalls, and aid windfalls by tax windfalls (also see Bulir and Hamann 2003, 2006).c This, of course, only deepens the recipient country's predicament. Finally, the study also found that periods of excess aid and tax revenue are seldom used to accelerate domestically financed investment spending, to potentially catch up with previous shortfalls. This finding is important, since it signifies that aid volatility may have permanent costs in terms of lost output. The country examples point to potential gains from a greater emphasis on regular annual budget support and on strong donor coordination frameworks. Intrayear predict- ability remains an important issue for countries facing external financing constraints. It appears that coordinated budget support frameworks are likely to also improve intra-annual disbursement patterns and avoid the typical "year-end rush" of disbursements. Reaping these gains, however, is predicated on creating the environment for a reliable medium-term engagement of budget support donors. a. Gelb and Eifert (2005) argue that although predictability poses a special challenge for budget support, there are practical ways to address this issue. They show that performance-based allocation rules that have a flexible precommitment rule can allow for precommitment of aid in a multiyear framework, while avoiding drawn-out periods of misallocations. b. Although budget aid disbursements remain difficult to predict on a year-by-year basis and carry large prediction errors, the authors do not find evidence that, in the aggregate and over time, aid disbursements fall short of aid projections by large amounts. c. This procyclicality of aid hurts investment and public debt management. were undertaken jointly, while 48 percent of T O W A R D A M O R E E F F E C T I V E A I D country analytic work was prepared jointly A R C H I T E C T U R E ? with another donor, relative to the 2010 Paris Even as the Paris Declaration is beginning targets of 40 percent joint missions and 66 to change aid delivery practices among tra- percent joint analytical work. ditional donors, the rise of new aid sources, including new donors, private foundations, Proliferation is particularly pronounced in and a profusion of global vertical funds is the health sector, and the case of Rwanda, increasing the complexity of the global "aid which is discussed in chapter 2, highlights architecture." These changes increase total many of the challenges that are faced by resources, but they also present the inter- countries with weak capacity--the need for national community with the challenge of policy coherence at the sector level and for coherence--of forging an aid architecture complementarity between national, regional, with closer coordination of aid activities and global priorities and programs. The among a wider donor community, as well as need for building country capacity to effec- greater harmonization, less fragmentation, tively handle scaled-up aid is abundantly and earmarking of aid toward specific appli- clear as well. Indeed, countries with strong cations. Such coherence is vital if aid is to be capacity--Tanzania and Uganda--have had successfully scaled up to meet the MDGs. some success in pulling together a myriad of Donor proliferation has seen a marked health donors into coordinated sectorwide increase over time. The average number of approaches. Nevertheless, the profusion of official donors--bilateral and multilateral-- aid channels heightens the need for coordina- per country has increased threefold since the tion, alignment, and harmonization among a 1960s; the number of countries with over 40 wider donor community, particularly when active bilateral and multilateral donors has bal- engaging with poor countries that find it dif- looned from zero to over 30 since 1990 (World ficult to turn down offers of assistance. Bank 2007a). This has been accompanied The problem of a large number of aid chan- more recently by an expansion in the number nels is compounded by the trend toward the of emerging donors, many of which are gain- small average size of funded activities. Using ing a substantial presence in the aid commu- the OECD's Creditor Reporting Database, nity. Non-DAC donors are a heterogeneous World Bank (2007a) shows that between group; the degree to which DAC approaches 1997 and 2004 the average size of aid activi- and norms are applied varies from country to ties (measured in 2004 prices) dropped by $1 country. Insufficient data make it difficult to million to about $1.5 million and the number accurately assess aid volumes and prospects of activities surged from 20,000 to 60,000. from these sources. The number of private Moreover, countries with lower institutional foundations and charitable organizations has capacity had higher aid fragmentation. Previ- also mushroomed, as have global programs ous Global Monitoring Reports have noted (which are discussed in the next section) and the negative implications of higher frag- funds. Donor contributions to 20 major global mentation on aid quality, especially through programs have increased from almost zero in higher transaction costs for recipients and the mid-1970s to about $3 billion annually. donors and through a smaller donor stake The number of active trust funds administered in overall country outcomes. Excessive frag- by the World Bank alone in FY2006 was 929, mentation is a serious problem and measures up from 840 in the previous year (World Bank to reduce it, such as through donors limiting 2007f). At the same time, contributions to the number of countries in which they focus, multilaterals, including to IDA and the major providing a larger amount of funds through regional development banks, have dropped more efficient vehicles (including multilateral (see above). The proliferation of aid channels channels) and modalities, and committing to can overwhelm the often limited capacity of delegate authority to lead donors, could help recipient countries to implement the record- to reduce transactions costs and to improve ing, processing, auditing, monitoring, and the effectiveness of aid. assessment requirements of different donors. Recent trends in proliferation and frag- It can also complicate the management of aid, mentation are thus impacting the global aid undermining its effectiveness. architecture, and posing a challenge to the quality and effectiveness of aid. Although the total--is now allocated to such goods (figure Paris Principles address some of these chal- 4.10). The United States is the largest source of lenges, more is needed to achieve coherence funds for global public goods, financing nearly in the aid architecture. 40 percent of all DAC donors' ODA commit- ments of this type. Other important official M A K I N G G L O B A L P R O G R A M S M O R E contributors include the Germany, Japan, the E F F E C T I V E Netherlands, the United Kingdom, and the Programs and partnerships of global scope European Union as a group. have grown rapidly, reflecting the increasing Further evidence of the robust growth in influence of global issues over the develop- global programs is seen in the activities of the ment agenda. The international donor com- various trust funds administered by the World munity--including both official and private Bank, many of which are global and regional donors--is already channeling substantial in scope. Contributions to these trust funds resources through these vehicles. Global pro- have more than doubled in recent years, from grams typically focus on delivering either $2.2 billion in 2001 to $5.2 billion in 2006. targeted key services such as primary educa- Contributions by bilateral donors and the tion or HIV/AIDS treatment, or global public European Commission accounted for nearly goods such as peace and security, control of 90 percent of the 2006 figure, and those by infectious diseases, knowledge generation and private entities for 1 percent. Disbursements dissemination, protection of the global com- have shown remarkable growth as well, ris- mons, a free and open trading system, inter- ing to a record high of $4.4 billion in 2006. national financial stability, and protection Nearly half of all trust fund disbursements in from borderless crime. Global programs vary 2005 were for global programs. widely in terms of size, funding sources, financ- A key challenge for global programs is ing arrangements, governance structures, and how best to integrate their mandates and modalities. Their importance in delivering priorities with country-based programs into global public goods is revealed in the fact that coherent development strategies. Successful nearly $20 billion in ODA grant commitments integration depends on aligning the objec- by DAC members--over a quarter of the tives and design of the global program with DAC members' and EC's ODA commitments forGPGs Source: OECD DAC database and staff estimates. Note: GPGs include both global and regional public goods. conditions at the country level. A 2006 World costs of donor fragmentation that a welter of Bank report evaluates the alignment of global global programs might otherwise foster. programs at the country level for seven coun- tries. Focusing on the larger programs, such as the Global Fund to Fight AIDS, Tubercu- Progress on Implementing the MDRI losis, and Malaria, the Global Environment Facility, the Global Alliance for Vaccines and The past year saw major progress in extend- Immunization, the Education for All Fast ing and deepening debt relief to the poorest Track Initiative, and the Consultative Group countries. Following a 2005 proposal by the on International Agricultural Research, the G-8, three multilateral institutions--the AfDF, study finds considerable variation among IDA, and the IMF--implemented the new them. The Education for All Fast Track Ini- Multilateral Debt Relief Initiative (MDRI), tiative was found to be well aligned, and the and agreed to provide 100 percent debt relief Global Environment Facility reasonably well on their eligible claims to countries that have aligned, while other global programs, espe- reached, or will eventually reach, the comple- cially in health, were struggling to improve tion point under the HIPC Initiative. The their alignment.21 The study also finds that MDRI was implemented at the beginning of alignment is easier when the benefits of the 2006 by the IMF, in mid-2006 by IDA, and global public good to the individual country early 2007 by the AfDF (for the latter, delivery are perceived to be high. Shared objectives of debt relief will be provided retroactively to and strategies strengthen local ownership, the beginning of 2006). To date, 22 postcom- and this helps promote success. Establishing pletion-point HIPCs have benefited from debt ownership and alignment is more compli- relief under this new initiative, which amounts cated in countries with limited institutional to about $38 billion (in nominal terms)--two capacity and greater donor fragmentation. non-HIPCs, Cambodia and Tajikistan, have One way to improve the alignment of also received MDRI relief from the Fund. The global programs is to anchor them in the Paris remaining HIPCs that have not yet reached the Principles. Also central to success are predict- completion point will automatically qualify ability and sustainability of funds, especially for the MDRI once they do so. for recurrent expenditures, at the local level. The full cost of the MDRI for the three Evaluations suggest that only a few global institutions is expected to be around $50 bil- programs provide truly global public goods; lion. The MDRI commits donors to providing many so-called global programs actually pro- additional resources to ensure that the reflow vide national or local goods (see also chapter losses associated with debt forgiveness do not 2 of this report). This suggests the need for undermine these institutions' overall financial selectivity in establishing global programs integrity or ability to provide financial sup- to avoid their uncoordinated proliferation. port to low-income countries.22 Last year's Another concern is that the funds raised by GMR presented the baselines established by global programs may crowd out other funds IDA and the AfDF on which additionality of rather than provide net additional resources; donor financing is to be assessed. Monitoring they may also compete with country and donors' commitments on financing the MDRI local programs for scarce resources and staff. is important to ensuring actual additionality of Clearly a balance has to be struck between donor financing over time. Chapter 5 presents country-owned priorities and global program the progress on donor financing of the MDRI; objectives. The MDBs are well placed to help as of end-2006, IDA donors had provided firm reinforce the centrality of country strategies financing commitments of $3.8 billion over the and ensure that global programs complement four decades of MDRI implementation, against rather than compete with them. The conven- a volume of irrevocable debt relief provided by ing power of MDBs can also help reduce the IDA under the MDRI of currently $28.3 bil- lion.23 The upcoming IDA-15 replenishment pletion point--when creditors provide the full will be an important test of donors' intentions amount of debt relief committed at the deci- regarding their support of the MDRI. sion point on an irrevocable basis--contin- The proposal recently approved by the ued in early 2007; four additional countries Inter-American Development Bank (IADB) to reached the completion point, bringing the provide debt cancellation to postcompletion- total number of countries that have done so to point HIPCs (Guyana, Bolivia, Honduras, 22. Several of the eight countries in the interim and Nicaragua) and to Haiti once it reaches period between their decision point and com- completion point, in line with the MDRI, will pletion point are on track with respect to their provide debt relief of $4.4 billion--$3.4 bil- macroeconomic programs; others that experi- lion in principal and $1 billion in interest-- enced difficulties in program implementation to these countries. Debt cancellation will be are pursuing the necessary policy measures to financed by the IADB out of internal sources, bring their economic programs back on track more specifically, from existing Funds for (IMF-World Bank 2006b).24 Special Operations (FSO). However, there is Debt relief under the HIPC Initiative is pro- currently no consideration of an FSO replen- jected to substantially lower debt and debt ser- ishment linked to the above. vice ratios for most HIPCs that have reached the decision point. Net present value (NPV) of P R O G R E S S O N T H E H I P C I N I T I A T I V E debt stocks in the 30 HIPCs that reached the Overall, substantial progress has been made decision point by end-March 2007 are pro- in the implementation of the HIPC Initia- jected to decline by about two-thirds once they tive. As of end-March 2007, 30 HIPCs have reach their respective completion points, and reached the decision point and are receiving by about 90 percent after the application of debt relief. Progress toward reaching the com- the MDRI (figure 4.11). The ratio of debt ser- Reduction of debt stock (NPV terms) for the 30 decision-point countries Sources: HIPC Initiative documents; IDA and IMF staff estimates. Note: Based on decision-point debt stocks. (Updated compared to Progress Report to include Malawi, São Tomé and Principe, and Sierra Leone as completion-point countries and Haiti as interim country.) vice to exports for these countries is estimated $15 billion in 2005, and is projected to have to have declined from an average of about 17 increased to $18 billion in 2006. Because of percent in 1998­99 to about 7 percent in 2005. problems with cross-country consistency in These ratios are estimated to have declined fur- the definition of poverty-related expenditures, ther to about 4 percent in 2006. it might be useful instead to review the trend in For these 30 countries, poverty-reducing health and education expenditures (box 4.4). expenditures on average have risen from about 7 percent of GDP in 1999 to over 9 percent of T H E C H A L L E N G E O F K E E P I N G D E B T GDP in 2005, a level more than five times that B U R D E N S S U S T A I N A B L E spent on debt service. In absolute terms, pov- The challenge for countries receiving MDRI erty-reducing spending is estimated to have debt relief is to ensure that financial resources increased from about $6 billion in 1999 to freed up by debt reduction are used for Recent data suggest a relationship between cuts in debt service costs and increases in social expen- ditures for HIPCs. Using data on health and education expenditures for 110 countries for 1985­ 2003/04, Thomas (2006) finds that social expenditures in relation to output have risen gradually over time in both HIPC (an approximately 2 percentage point increase) and middle-income coun- tries, whereas the ratio has fluctuated more widely for other low-income countries.a Debt service Trends in social expenditures and debt service continued in relation to output has declined continuously among HIPC countries since the introduction of the HIPC initiative in 1996 while debt service costs have risen for middle-income and other low-income countries over this period. Econometric estimates confirm that the HIPC relationship is significant at the 10 percent level of significance. Moreover, social expenditures are protected from expenditure cuts among low-income countries but are boosted by expansionary budgets. Over 2000­02, the social expenditure ratio among HIPCs rose rapidly by almost 1 percentage point and has subsequently stabilized at this level. The main factors associated with these changes include an expansionary budget policy in 2000 and lower debt service payments since then. Exports also appear to have played a contributing role in 2000, possibly because strong export growth is an indicator of a strong-performing economy. Contribution to the change in social expenditures a. Some studies including Clements, Bhattacharya, and Nguyen (2003) find that the ratio of debt service to output is significantly negative relative to the public investment rate: with a coefficient of 0.2 they estimate that a decline in debt-service/GDP ratio from 10 percent to 5 percent would raise public investment/GDP by 1 percentage point. Kraay and Depetris Chauvin (2005) find that debt relief has not contributed to any significant change in health and education expenditures. One criticism is that they use imputed values of debt relief (PV terms). But one problem with using debt service is that debt service can change for reasons other than debt relief. reaching the MDGs. Post debt relief, these design a financing strategy that will mitigate countries are experiencing an increase in these risks without unnecessarily constrain- their perceived borrowing space as well as in ing access to resources for development. the availability of financing from emerging The Boards of the IMF and IDA reviewed official and private creditors and domestic in late 2006 the application of this frame- sources. While these developments are wel- work and have agreed to refinements aimed come, they could raise new risks, if debt is at improving the rigor and quality of the built up too rapidly and abruptly. The debt analyses in the post debt relief context. sustainability framework for low-income They concluded that ensuring long-term countries, endorsed by the Boards of the sustainability requires efforts by borrow- IMF and IDA in 2005, can help countries ers, lenders, and donors to promote prudent borrowing and a suitable mix of concessional percent, developing-country export growth and other finance. To this end, the broader continued to outpace the global average. Turn- use of the debt sustainability framework by ing to individual regions, higher energy prices creditors would help disseminate informa- contributed to export growth of 22 percent in tion between borrowers and creditors, as well Middle Eastern and North African countries. as among creditors, and better inform financ- China continued its strong recent export per- ing decisions.25 In addition, the capacity of formance, with a 27 percent increase in 2006. borrowing countries to manage and monitor Exports from Sub-Saharan Africa continued their debt must be strengthened so countries to benefit from the healthy global economy, can develop their own medium-term strategy recording a 23 percent increase overall, while to support national development objectives, least-developed countries experienced a while containing the associated risks of debt remarkable 25 percent growth. Industrialized distress and macroeconomic vulnerability. countries expanded their exports at a more Improving the quality and availability of debt modest 4 percent rate. data--external and domestic debt--will be In addition to favorable cyclical factors, important to support monitoring and assess- last year's global trade performance reflects ment of debt and associated risks. continuing unilateral trade reforms. Average tariffs in developing countries have fallen from 16.3 percent in 1997 to 11 percent in 2006. As the pace of global integration increases, the Major Developments in Global Trade challenge of harnessing the new opportunities while managing the risks places a premium on World trade in 2006 continued its strong a strategy of greater openness to trade, coupled growth trend of recent years. Worldwide with behind-the-border reforms. For example, exports of merchandise reached $12 trillion in a number of countries have proved able to 2006, growing 16 percent in value, well above compete with China following the phase-out the average growth of 8 percent recorded in of textile and clothing quotas, but central to 1995­2004. Reflecting increases in fuel prices, their success have been reform of their tariffs fuel-exporting countries experienced the high- and investment climate and support from the est export growth, at 23 percent, but global international community to overcome infra- non-oil exports also grew 15 percent. At 22 structure constraints (box 4.5). The clothing sector has been at the forefront of export diversification in many developing countries, and employment in the sector has been an important source of income for many women. But trade in this sector is undergoing a profound change, as the system of quantitative restrictions that man- aged the industrial countries' imports of textiles and clothing for 30 years was finally dismantled at the end of 2004 as part of WTO Uruguay Round agreement. This change engendered widespread fears that global markets for textiles and clothing would be swamped by Chinese products, with adverse implications for other developing countries. Does this sector still serve today's low-income countries as the first rung on the ladder to higher-value-added exports? Early signs suggest that adjustments in production patterns and trade flows following the removal of quotas have been less drastic than anticipated. Although China's total exports soared by 29 percent between 2004 and 2005, its share of global clothing exports has increased at a slower rate.a The substantial growth in the world market for clothing has allowed exports from many other countries, including Colombia, Egypt, India, Madagascar, Peru, Sri Lanka, and Turkey, to increase. continued In Bangladesh, where the loss of 1 million jobs had been predicted (Oxfam 2004), exports to the European Union and the United States instead increased strongly between 2004 and 2006. Nevertheless, some countries have seen substantial declines in exports of clothing. Exports to the European Union and the United States from Brazil, the Dominican Republic, and Swaziland have decreased by about 20 percent, and exports from Taiwan Province of China have fallen by nearly 30 percent. The clothing sectors in these economies face significant adjustment. Even in countries that have managed to increase exports, pressure for adjustment may be strong as the more efficient firms expand while those unable to compete in the global market decline. In the absence of other employment oppor- tunities, especially for women, workers laid off from the textile and clothing sectors may fall back into poverty. Minimizing the costs incurred by these workers and their families and facilitating their shift to other productive activities will be a major challenge in a number of developing countries. In Africa, where the end of the clothing sector had been predicted, exports have generally fallen (by 12.3 percent on a trade-weighted average), but some countries, such as Madagascar and Kenya, have managed to maintain or even increase exports. The slowdown in African clothing exports observed during 2005 has also started to reverse. Sub-Saharan African clothing exports to the European Union in the first eight months of 2006 were nearly 8 percent higher than in the same period in 2005 (but still 8 percent lower than in 2004). Exports to the United States, however, have continued to decline, falling 12.5 percent during 2006 after a 14 percent drop in 2005 (largely driven by Mauritius and South Africa). On the plus side, the downward trend in exports has reversed for 14 countries.b The prospects for African exports of clothing to the United States have been enhanced by the recent extension, to 2012, of the nonrestrictive rules of origin under the African Growth and Opportunities Act (AGOA). These rules allow African producers to globally source the fabrics used in their garments and still receive U.S. trade preferences. In contrast, the European Union offers tariff preferences with restrictive rules that deny producers the ability to use the best and lowest- cost inputs wherever they may be found. The AGOA rules of origin were critical in the substantial increase of African clothing exports to the United States between 2000 and 2004, a period during which exports to the European Union stagnated. The countries best able to expand clothing exports will be those that have a supportive business environment, low trade costs (efficient customs, ports, and other transport infrastructure), and com- petitive firms flexible enough to meet the changing demands of the global buyers that now dominate the industry. With these conditions in place, the clothing sector can still be a driver of industrial diver- sification in many poor countries, even in the face of unfettered competition from China. a. Europe and the United States have introduced a number of temporary restrictions on imports from China under the special safeguards agreement included as part of China's WTO accession. It has been suggested that the very rapid surge in exports from China to both the European Union and the United States before the safeguard was imposed was partly a response to its anticipated use by increasing the base for quota calculations. The re- imposition of limits on selected Chinese exports is likely also to have spillover effects on third parties. b. Based on figures for January through October 2006. The United States and the European Union accounted for a large but declining share of global clothing imports, 23 and 35 percent, respectively, in 2005. Madagascar accounts for nearly two-thirds of the net increase in exports to the European Union between 2005 and 2006. At the same time, however, its exports to the United States have fallen. U P D A T E O N C O U N T R Y P O L I C I E S and antidumping actions have the restriction Trade policies and domestic welfare. Govern- of imports as their explicit objective; regula- ments use a variety of instruments that have tory policies motivated by public health or the effect of restraining trade, whether inten- safety concerns do not, but may limit trade tionally or as a side effect. For example, tariffs nonetheless. In fact, such policies may affect trade in a very product-specific way, with dif- at more than 3 percent of GDP. By compari- ferent effects on different countries. When son, the average low-income country loses the quantifying the overall effects of a country's equivalent of about 0.4 percent of its GDP as national trade policy, then, it is important to a result of its tariffs. The total deadweight consider nontariff measures as well as tariffs. loss that tariffs inflict on the world as a whole This report does so using two separate mea- is on the order of $100 billion a year.28 sures of trade-related policies. The first is a Agricultural protection is the main source trade restrictiveness index (TRI), which is a of these losses (table 4.2), particularly in high- measure of the impact of tariffs imposed by income countries. Canada, Japan, Korea, a country on itself. The second, called the Norway, Switzerland, and the European Overall Trade Restrictiveness Index (OTRI), Union as a group all have TRIs for agricul- captures the impact of both tariffs and non- ture that exceed 40 percent (see the annex). tariff measures on a country's trading part- The worldwide average TRI on agricultural ners, that is, on the exporters that ship goods goods, at about 22 percent, is twice that for to the country.26 manufactured goods. Assuming a country cannot affect its terms Trade policies and trade flows. The last of trade, tariffs increase the relative price of two columns of table 4.2 summarize the effect imported goods in the country imposing them, of all observed trade policies on trade flows thereby reducing national welfare. One way of exporters, as measured by the OTRI, for to quantify this loss of welfare is to calculate high-, middle-, and low-income countries.29 the deadweight loss due to the existing tar- The OTRI includes the effects of regulatory iff structure. This can be done by comparing measures such as product standards, as well the welfare of the country given its TRI with as policies such as import licensing require- what would obtain under zero tariffs.27 The ments and antidumping actions. As previous average TRI for all countries is 13 percent, editions of Global Monitoring Report have with Bangladesh's the highest at 66 percent stressed, the OTRI is a measure of trade (see the annex to this chapter). The TRI is restrictiveness, not protection. It measures the generally higher in developing than in indus- effect of included policies in limiting trade; it trial countries (table 4.2). Tariffs in Japan, is not a measure of protectionist intent. As however, generate the largest absolute dead- the table shows, the average OTRI for devel- weight loss of any single country ($28 billion oping countries is 16 percent, and that for in 2004), with Korea ($25 billion) a close sec- high-income countries about 8 percent. The ond. Bangladesh and Korea suffer the highest effect of trade restrictions in the latter group losses in proportion to their economies, each of countries is to reduce annual imports by Trade restrictiveness and its impact on welfare and trade flows, by country income group, 2004 Tariffs and impact on importing country All trade restrictions and All goods Agricultural goods Manufactured goods impact on trade flows Welfare loss Welfare loss Welfare loss Loss Billions Percent (billions (billions (billions Country group TRI of dollars of GDP TRI of dollars) TRI of dollars) OTRI of dollars) High income 10 74.5 0.2 40 65.3 4 9.2 8 261.6 Middle income 15 13.3 0.2 30 3.8 12 9.5 16 180.3 Low incomea 16 5.2 0.4 18 1.4 16 3.9 16 29.2 Source: World Bank staff calculations. a. Data available for 22 countries only. about $260 billion. The comparable figure Changes in trade restrictiveness, 2000­06. for middle-income countries is about $180 Reflecting a steady reduction in tariffs, the billion, and that for the 22 low-income coun- global impact of policies in restricting trade tries in the sample about $29 billion. has declined in recent years.30 Except for a As with the TRI, the average OTRI for number of African countries, most economies agriculture is much higher than that for have lower OTRIs than in 2000 (figure 4.12). manufacturing. Many developing countries Developing countries that have seen substan- impose similar levels of restrictiveness on tial falls in their OTRIs include India, Egypt, agricultural imports as on manufactures, Nigeria, and Mauritius, as well as China and but for some developing nations the OTRI many Latin American nations. Among devel- for agriculture is far higher than for goods. oped countries, overall trade restrictiveness in In India, for example, the ratio of agricul- Japan and the United States has fallen some- tural to nonagricultural trade restrictiveness what, while it has remained largely unchanged increased significantly following the recent in Canada and in the European Union. liberalization of trade in industrial products. Much of the decline in OTRIs pertains to For developed countries this ratio tends to be manufacturing; much less has been achieved much higher than in most developing coun- in agriculture. In a number of countries the tries, rising to 10 or more. agricultural sector is now more restrictive Overall trade restrictiveness has declined (2000­06) Burkina Faso Philippines Albania Kenya Bangladesh Trinidad and Madagascar Sudan China Tunisia Algeria Malaysia Cameroon Tobago Malawi Uganda Côte d'Ivoire Uruguay Argentina Mali Canada Ukraine Uganda Egypt Australia Mexico Costa Rica Venezuela, South Africa Ghana Belarus Moldova Ethiopia R.B. de India Bolivia Nicaragua European Union Jordan Brazil Russian Fed. Gabon Kazakhstan Brunei Saudi Arabia Hong Kong, China oducts Lebanon Chile Senegal Iceland Pr Mauritius Colombia Singapore Korea, Rep. of All Morocco El Salvador Switzerland New Zealand Nigeria Guatemala Tanzania Norway Papua New Guinea Honduras Thailand Oman Paraguay Indonesia Turkey Romania Peru Japan United States Sri Lanka Zambia Decrease (>5%) Some Decrease (1­5%) Small­No Change Some Increase (1­5%) Increase (>5%) Argentina Ukraine Albania Sri Lanka Australia Switzerland Algeria India Bangladesh Zambia Canada Uruguay Brazil Tanzania Belarus Kenya Burkina Faso Colombia United States Brunei Trinidad and Bolivia Oman Chile El Salvador Venezuela, Costa Rica Tobago Cameroon Uganda China Ethiopia R.B. de Egypt Japan e Côte d'Ivoire Gabon European Union Madagascar Ghana Iceland Honduras Malawi ricultur Guatemala Indonesia Hong Kong, China Moldova Ag Lebanon Malaysia Jordan Morocco Mauritius Mexico Kazakhstan Norway Nicaragua Paraguay Korea, Rep. of Romania Papua New Guinea Russian Fed. Mali Rwanda Peru Saudi Arabia New Zealand South Africa Thailand Senegal Nigeria Sudan Turkey Singapore Philippines Tunisia Source: World Bank staff estimates. than it was six years ago; the European Union an average OTRI of 48 percent in the Middle has seen virtually no change, while Canada East and North Africa, while products origi- and the United States have registered a small nating in Sub-Saharan Africa confront a 38 decline in agricultural trade restrictiveness percent average OTRI in South Asia. since 2000. A small number of countries, Table 4.3 reports both the OTRI including including Argentina, China, and Chile have nontariff measures and a tariffs-only version of achieved substantial reductions in the OTRI the OTRI. Nontariff measures (NTMs) have a for agriculture since 2000. substantial impact on the level of the OTRI, Trade policies and market access. The especially in countries with low tariffs. As effect of policies on exporters' access to mar- mentioned previously, NTMs are not necessar- kets differs across exporting regions (table ily protectionist in intent, but they can repre- 4.3).31 These differences are partly the result sent a significant burden, especially for exports of discriminatory application of trade poli- originating in developing countries. In practice, cies (that is, trade preferences), but mostly developing countries often benefit from tariff they reflect differences in the product com- preferences in and enjoy duty- and quota-free position of exports. Agriculture generally access to many industrial-country markets, faces much more restrictive market access implying that NTMs, not tariffs, are the main conditions than manufacturing. Because of factor restricting their access to these markets. this, regions exporting mainly agricultural The incidence of nontariff measures is highly products generally face more restrictive product-specific. As the product composition markets than regions where manufacturing of exports varies widely across exporters, there dominates the export sector. This is one rea- is often a wide range of OTRIs confronting dif- son why trade among developing countries is ferent countries in the same market. affected by high levels of trade restrictiveness. Antidumping. With the steady decline in For example, Latin American exporters face tariff barriers, countries seeking to re-impose Market access (OTRI) Exporting Region Latin America Europe & East Asia Middle East & Importing Region & Caribbean Central Asia & Pacific North Africa South Asia Sub-Saharan Africa Latin America 31 11 15 17 24 17 5 5 6 5 8 11 Europe & Central Asia 39 21 18 6 40 18 10 8 7 4 9 14 East Asia & Pacific 10 12 20 8 15 18 4 3 4 3 5 8 Middle East & North Africa 48 25 26 24 28 23 20 11 12 4 7 6 South Asia 39 25 29 11 35 38 19 15 15 9 7 32 Sub-Saharan Africa 14 13 15 33 25 25 5 7 9 7 10 4 All developing countries 27 20 21 13 27 19 7 8 7 5 9 12 High-income countries 25 12 18 22 39 32 6 4 3 6 10 3 Source: World Bank staff estimates. Note: Values in italics are tariff-only OTRIs (i.e., excluding NTMs). Regions are developing country only. protection often resort to so-called contingent The remaining months of 2006 saw efforts protection measures, such as antidumping to relaunch the negotiations through inten- and safeguard actions. Major users of anti- sive bilateral consultations and a series of dumping now include developing as well as ministerial meetings. These efforts culmi- industrial countries; indeed, India is now the nated on February 7, 2007, in an agreement most frequent user of this instrument. World- by World Trade Organization (WTO) mem- wide, the number of antidumping investiga- bers to restart the talks. tions increased from about 1,200 in 1995­99 The general view is that there is now a nar- to almost 1,400 during 2000­04; the share of row window of opportunity to reach agree- investigations started by developing countries ment early in 2007 on the key elements of a rose from 43 percent to 48 percent. China is preliminary package that could pave the way the leading target of such actions. Notwith- for a final deal and be sufficient to assist the standing their much smaller share in world extension of U.S. Trade Promotion Author- trade, middle-income countries as a group ity (TPA) by July 2007. (TPA allows the have surpassed the high-income countries as U.S. president to submit a negotiated trade targets for investigations. agreement to Congress for an up-or-down Given the small share of total imports hit vote, without amendment.) Even with the by antidumping investigations, antidumping prospect of such a preliminary Doha deal, has a negligible effect on trade worldwide however, renewal of TPA is by no means cer- and little impact on overall trade restrictive- tain. Moreover, notwithstanding some indi- ness. However, the protectionist effect of cations of new flexibility among the major antidumping can be great for certain spe- players, it is unclear whether agreement on cific imports from targeted countries, and key elements of a preliminary package can the threat of antidumping can have a chill- be reached in the first half of this year. ing effect on the expansion and pricing of The World Bank and the IMF will continue exports. The average effect of antidumping to advocate a timely conclusion to the Doha investigations imposed by India is estimated negotiations and to argue against backsliding to be equal to a 44 percent ad-valorem tariff on progress already made. While a deal at any equivalent for the affected products. Anti- price is not supported, many of the proposals dumping investigations launched by China reportedly under discussion, when combined generate an ad valorem equivalent of 36 with progress already made (such as offers percent. In the industrial countries the effect to eliminate agricultural export subsidies of antidumping investigations tends to be by 2013 and to provide duty- and quota- smaller; in the case of the European Union free market access for at least 97 percent the average ad valorem equivalent is about of exports from least developed countries), 13 percent.32 are sufficiently substantive to make a deal worthwhile. Significant gains for developing countries would also flow from services liber- The Doha Round Negotiations alization and trade facilitation, and from aid Despite intensive consultations throughout for trade accompanying the Doha Round. June and July, negotiations on the Doha Failure to conclude the Doha Round Round were effectively suspended on July would send a strong negative signal to the 23, 2006, amid disagreement over how world economy about the ability of coun- ambitious an agreement to seek on agricul- tries to pursue multilateral solutions. It could tural market access and on greater reduc- weaken the multilateral trading system, tions in trade-distorting domestic support which provides developing countries with in agriculture. Major players expressed their guaranteed nondiscriminatory market access, disappointment while uniformly reaffirming rules-based settlement of disputes, and trans- their commitment to a successful outcome. parency of trade regimes. Trade disputes may also increase, feeding protectionist sentiment Aid for trade is rising and overstraining the WTO dispute settle- ment system. The risk of trade diversion from the growing number of preferential trade agreements (PTAs; see the discussion below) will also increase in the absence of progress on multilateral liberalization, and PTAs will, in any case, continue to leave unaddressed the high levels of trade-distorting domestic support in agriculture. But the biggest risk of failure in the Doha Round is to countries' own economic growth. Trade reform is fundamentally about self- interest: increasing global competition and domestic budgetary pressures both argue against delay in reforming costly domestic protection. Regardless of the outcome of the Doha talks, countries should continue to pur- sue unilateral trade reform, and the World Source: WTO/OECD-DAC Trade Capacity Building database. Bank and the IMF will continue to work closely with developing countries to sup- port efforts to integrate trade into national development strategies, and to advocate the budget support to competitiveness reforms in reform of developed-country trade practices Latin America and the Caribbean. From July that hamper development efforts. to December 2006 (i.e., the first half of fiscal 2007), lending reached $0.8 billion, and this A I D F O R T R A D E uptrend is expected to be sustained. Further progress was made on aid for trade The IMF has provided financial support in 2006. Support for trade-related assistance for trade-related adjustment, augmented by continued to grow, while efforts continued to the Trade Integration Mechanism (TIM) and establish the necessary international archi- the option to use floating tranches of IMF tecture to ensure improved coordination and lending to support trade reform in IMF- effectiveness of this aid. At 2004 constant supported programs.35 At present, three coun- prices and exchange rates, assistance for tries (Bangladesh, the Dominican Republic, trade policy and regulations in 2005 reached and Madagascar) are availing themselves of $905 million, while trade development activ- support under the TIM. Total financing of up ities peaked at $2.17 billion (figure 4.13).33 to $211 million is being made available to Additionally, an estimated $12.1 billion was these three countries under the TIM, includ- spent to support the economic infrastructure ing amounts that could be drawn (under the essential for international trade. TIM deviation feature) in the event the actual World Bank lending for trade also contin- BOP effect of qualifying trade policy events ued on an upward trend, with 49 trade-related turned out to be larger than anticipated.36 projects undertaken in fiscal 2005 valued at A focus of international activity on aid for $1.08 billion, and 51 projects in fiscal 2006, trade in the past year was on establishing the valued at $1.61 billion, representing a three- necessary architecture to improve the coordi- fold increase over 2003.34 Expansion to date nation of such aid. Important elements of this has been driven by trade-related infrastructure work have taken place under the auspices of in support of regional integration in Africa, the WTO, given the general consensus that trade-related infrastructure in East Asia, and aid for trade is an essential complement to, and not a substitute for, a successful out- an ad hoc consultative group to follow up on come to the Doha Round. The Task Force on its recommendations. Enhancing the Integrated Framework (IF)37 Aid for trade enjoys broad acceptance issued its recommendations on July 5, 2006, within the development community as part of which included the following: the creation the growth agenda, but a number of impor- of a new executive secretariat in Geneva; tant challenges remain. Donors have indicated strengthening of country capacity; a funding that they are prepared to offer large increases target for the IF trust fund of $400 million in aid for trade, but how much of this would over an initial five-year period; and creation be additional to existing aid remains unclear. of a monitoring and evaluation framework. Much also remains to be done to operation- The task force also agreed that an IF-type alize aid for trade, particularly with respect mechanism could be useful for low-income to supporting countries other than the least developing countries that are not among the developed in the articulation of trade inte- least developed. Transition teams were sub- gration strategies, addressing the needs of sequently established on institutional issues, the growing regional and cross-country in-country issues, and the trust fund, with a agendas, and establishing systems to monitor view to operationalizing the enhanced IF by and evaluate results. Establishing an effective January 2007. Although progress was made, enhanced IF will also be critical. Central to this deadline was not met. The World Bank meeting these challenges will be implementa- and the IMF continue to actively engage with tion of aid for trade within the parameters other parties in Geneva to develop workable of the Paris Principles, using existing devel- and effective administrative arrangements for opment institutions and mechanisms with the enhanced IF. proven effectiveness. A further international effort to improve the coordination of aid for trade took place P R E F E R E N T I A L T R A D E A G R E E M E N T S under the auspices of the WTO Task Force on PTAs continued to proliferate in 2006, both Aid for Trade, which also delivered its report among developing countries and between in July 2006.38 The report provided a welcome developed and developing countries. Approx- clarification of the definition of aid for trade. imately 170 still-active PTAs have been noti- Under that definition, aid for trade encom- fied to the WTO. However, given that many passes not only activities related to trade agreements have not been notified, the actual policy and regulation and trade development, figure is estimated at around 250, with 20 but also those involving trade-related infra- more awaiting ratification and another 70 structure, building productive capacity, and under negotiation. support for trade-related adjustment. The In the past year, interest in PTAs has report also contained a series of proposals to intensified in developing and developed improve monitoring and evaluation, includ- countries alike. Asia has seen a renewed ing establishment of a monitoring body in the interest in regional integration, prompted WTO, an annual discussion on aid for trade in part by China's rapid growth, and mean- in the WTO General Council, and inclu- while integration continues to deepen in sion of an assessment of aid for trade, for Latin America. The United States completed donors and recipients, in WTO Trade Policy negotiations toward a PTA with the Andean Reviews. The report put forward a range of countries and is continuing negotiations with proposals for strengthening both the supply Korea, Malaysia, Thailand, and the United and the demand sides of aid for trade (and for Arab Emirates. The European Union's new matching the two), although it stopped short trade strategy also places a greater empha- of providing concrete operational recommen- sis on PTAs (EU 2006). The Association of dations.39 The report proposed instead that Southeast Asian Nations, Korea, and Mer- the Director General of the WTO establish cosur are identified as priorities for PTAs, The end of the Cotonou agreement in 2008,a along with the need to negotiate another WTO waiver for any further unilateral preferences (given that they have been found to be inconsistent with WTO rules), prompted the EU and ACP countries to launch negotiations on EPAs in 2000. EPAs are to be reciprocally negotiated PTAs, providing for mutual market access; some services liberalization; a regulatory agenda on investment, competition, and intellectual property rights; institutional provi- sions to facilitate trade; and new technical assistance for trade negotiation and development. The EPAs could spur trade--and raise incomes--in the African, Caribbean, and Pacific countries if they catalyze progress toward integration within regions, with other EPA groupings, and with the global market. However, their development impact will depend on their design, and here there are several concerns. First, the EPA regional groupings do not conform to existing common markets and PTAs. Coupled with the uneven progress in implementing existing regional agreements and the reluctance of the European Union to date to accept variable geometry,b this makes it difficult for countries to leverage the negotiations to promote effective intraregional liberalization. Second, pref- erential access for the EU countries and their neighbors behind currently high most-favored-nation (MFN) tariffs may displace more efficient sources of supply, underlining the need for MFN tariff reductions to precede backloaded preferential access for EU firms in EPA markets. Third, access to the EU market may be restricted if rules of origin, as yet undefined, require local value added greater than (say) 10 percent. Fourth, trade reforms, especially preferences to EU firms, may cut into pub- lic sector revenue when they are enacted, requiring compensatory tax reforms phased in to match any losses in tariff revenues. Finally, a key issue is the level of additional aid for trade on the table. Although there are benefits to keeping discussion of appropriate levels of aid for trade separate from the negotiations, this has eliminated one major incentive for the least-developed countries to join EPAs, since they already enjoy duty- and quota-free access to the EU countries under the Everything But Arms initiative. Appropriately designed, pro-development EPAs can help the ACP countries move forward on reforms to promote their competitiveness and regional integration. Source: Hinkle, Hoppe, and Newfarmer 2005. a. The Cotonou agreement covers unilateral trade preferences and development cooperation between member states of the European Union and their former colonies in the African, Caribbean, and Pacific regions. b. Variable geometry refers to arrangements under which not all countries in a regional grouping need to have a common external tariff on a most-favored-nation basis or toward the European Union. with India, Russia, and the Gulf Coopera- which the design of agreements can influence tion Council members also of interest. The their development impact (box 4.6). Properly strategy also places emphasis on enhanced designed PTAs can benefit their members, espe- trade relations with Turkey, the Southern cially if combined with a nondiscriminatory Mediterranean countries, and some coun- reduction in external barriers. But if they are tries of the former Soviet Union (under the designed badly, the cost of such agreements, New Neighborhood Policy). in terms of trade diversion, high information The European Union is also due to finalize costs, and demands on limited institutional the ongoing negotiations with African, Carib- capacity, may well exceed the benefits. For bean, and Pacific (ACP) countries under the example, as many as half of all PTAs may Economic Partnership Agreements (EPAs).40 divert more trade than they create, and bilat- The EPAs offer an important opportunity for eral "hub and spoke" PTAs benefit the hub trade reform in many countries, notably in (the rich country) disproportionately more Africa, but have also highlighted the extent to than the spokes (developing countries).41 9. The financial cost of reducing gender inequal- ity is difficult to calculate, both because gender 1. The larger emerging economies are also play- inequality is multidimensional and multisectoral, ing a greater role in trade, investment, and private and because efforts to reduce it must necessarily financial flows to poor countries. For example, work through multiple channels, not just those both China and India doubled their annual growth focused on gender. The task of collecting adequate rates of imports from Africa between 1990­94 and data alone is expensive and difficult, which poses 1999­2004; these two countries account for 50 a challenge to countries in estimating the financial percent of Asia's exports to Africa (see Broadman resources needed to achieve gender equity goals. 2006 for an analysis of Africa-Asia trade.) Foreign 10. The nine agencies are the Australian Agency direct investment in Africa from emerging econo- for International Development, Finland's Minis- mies, particularly Brazil, China, India, Malaysia, try of Foreign Affairs, the Norwegian Agency for and South Africa, is also growing rapidly. Development Cooperation, the Swedish Interna- 2. Global Monitoring Report 2006 discussed the tional Development Agency, the United Kingdom's increased risk of situations where nonconcessional Department for International Development, the lenders may indirectly obtain financial gain from International Labour Organization, the United debt forgiveness, grants, and concessional financ- Nations Development Programme, UN Habi- ing by the international financial institutions. This tat, and the World Food Programme. See Gen- could lead to an excessive buildup of debt if non- der Mainstreaming Evaluations: An Assessment concessional borrowing is not carefully managed. report discussed at the DAC-WB sponsored OECD 3. In collaboration with Portugal and the Euro- DAC Gendernet meeting of November 2006. This pean Foundation Centre, the DAC will cosponsor section was completed with information from a an international conference on the developmental number of donor members' of DAC Gendernet, role of philanthropic foundations in early 2007. including from Canada and the United States. 4. Under an AMC, donors guarantee a set enve- 11. See individual country needs assessments lope of funding at a given price for a new vaccine and Grown and others (2006). that meets specified target requirements. 12. Health interventions include upgrading 5. Also see Heller and others (2006) for fiscal clinics with comprehensive prenatal, antenatal, policy issues with scaled-up aid; Gupta, Powell, and emergency obstetric care; providing vitamin, and Yang (2006) for macroeconomic management mineral, and nutrition supplements for mothers with scaling up aid; and Bourguignon and others and children; sensitizing men to risks of STDs and (2004, 2005) for a general dynamic equilibrium providing service; managing malaria and anemia model for analyzing MDG strategies. in pregnancy. Education interventions include 6. Based on a review of the PRS-budget link in literacy programs for women and provision of nine poor countries, Renzio, Wilhelm, and Wil- separate bathrooms for girls. Interventions in agri- liamson (2006) find that policy making, planning, culture, infrastructure, and urban development and budgeting are typically fragmented processes. include subsidizing home energy costs for female- To strengthen the PRS-budget link, they recom- headed households and building day care centers, mend accessing high-level support for policies, shelters, and community centers. targeting PRS and budget processes on the actual 13. The World Bank's Country Policy and Insti- decision making, and harmonizing existing plan- tutional Assessment rates countries against a set of ning and budgeting processes. 16 criteria grouped in four clusters: (1) economic 7. The ACGF is a multidonor trust fund with management; (2) structural policies; (3) policies initial capital of $379 million from the UK. The for social inclusion and equity; and (4) public sec- first tranche of $56.8 million was received in the tor management and institutions. Individual coun- fall of 2006, and five ACGF projects with total tries are scored on a scale of 1­6, with 6 being the projected disbursements of $148 million are under highest. preparation. The Bank's Africa region is actively 14. The special circumstances of fragile states seeking other contributors to the ACGF, and sev- could influence the composition of aid. eral donors have expressed interest in contributing 15. This follows the analysis in Levin and Dol- to the fund in fiscal 2007. lar (2005). The coefficient on the fragile state 8. See OECD (2005b) for an analysis of earlier dummy shows whether this group of countries trends in ODA for gender equality. receives more (i.e., the coefficient is positive and on tariffs (less timely and less comprehensive). significant) or less (i.e., the coefficient is negative Thus, underlying data weaknesses may affect the and significant) aid than would be predicted by accuracy of the OTRI for some countries. Greater the other explanatory variables in the regression investment by governments in support of the activ- equation. ities of the international trade agencies that com- 16. Data presented here are preliminary esti- pile the data on tariffs and, especially, nontariff mates (as of March 1, 2007) and are being reviewed measures, that feed into the OTRI estimates would by the OECD/DAC Joint Venture on Monitoring improve their quality. the Paris Declaration. The data reflect implementa- 27. A deadweight loss is the loss in economic tion as of 2005 for donors that provided a grand efficiency (consumer surplus) caused by policies total of more than $100 million for the government that prevent agents from equating marginal costs sector. A final report by that group is expected to to marginal benefits. The TRI is the uniform tariff be published by early April 2007. The broad trends that would provide the same level of welfare in are, however, expected to remain stable. the importing country as the existing tariff struc- 17. OECD DAC 2006a. ture. It is calculated on tariffs only, because certain 18. Last year's GMR detailed the following nontariff measures may be welfare enhancing. For areas: the harmonization road map; the joint/col- a detailed discussion of the methodology used to laborative assistance strategy; the common perfor- estimate this index, see Kee, Nicita, and Olarreaga mance assessment framework; coordinated budget (2006). support; sectorwide approaches; joint diagnostic/ 28. This figure is comparable to the results analytic work; use of common arrangement; and from general equilibrium models, which suggest the independent monitoring process. annual global welfare losses from protection rang- 19. World Bank 2006b. ing from $78 billion to $128 billion. 20. OECD DAC 2007b. 29. The OTRI is defined as the uniform tar- 21. Most aid for health programs is targeted to iff-equivalent that will yield the observed level of a particular program or disease, often reflecting aggregate imports, which is determined by the pre- the priorities and interests of foreign donors, as vailing set of tariffs and nontariff measures in a opposed to local residents, and these programs are country. In addition to tariffs and specific duties, the seldom integrated into general public health sys- OTRI includes all the nontariff measures reported tems. Moreover, by drawing resources away from in the UNCTAD TRAINS database as well as data other public health funding recipients, they often on product standards compiled by the World Bank. fail to improve the overall health of the population Only data from publicly available sources are used. (see Garrett 2007). It should be recognized that the quality of data on 22. Chapter 5 of the report presents donors' NTMs is below that for data on tariffs, in terms MDRI financing commitments against targets. of both comprehensiveness of coverage and timeli- 23. Multilateral Debt Relief Initiative (MDRI): ness (NTM data may not be updated annually). Update on debt relief by IDA and donor financing 30. Because the OTRI is designed to take into to date (February 2007). account the value of imports and the import 24. In September 2006, the two Boards decided demand elasticities at the six-digit level of the to let the sunset clause of the Initiative take effect at Harmonized System of commodity classification, end-2006 and to grandfather the countries that meet changes in the OTRI may reflect not only changes the income and indebtedness criteria at end-2004. in observed trade-related policies but also changes 25. IMF-World Bank 2006a. in import composition and import prices. 26. One of the major advantages of the OTRI 31. Calculations are based on the Market over the more commonly used indicators of trade Access version of the OTRI (MA-OTRI). This restrictiveness is that the OTRI takes into account measures the average level of restrictiveness in all the effect of nontariff policies on trade. While the importing countries that receive an export product OTRI estimates have been very robust to recent of a country, keeping the aggregate exports of that improvements to the methodology--which is country constant. For a detailed discussion of the well-established and directly related to generally methodology used to estimate this index, see Kee, accepted concepts in trade policy analysis--it is Nicita, and Olarreaga (2006). important to note that the existing data on non- 32. These estimates are based on the impacts tariff policy is generally of lower quality than data of antidumping investigations on bilateral trade in the affected products, controlling for other deter- 37. The Integrated Framework for Trade- minants of trade in a product between the export- Related Technical Assistance (IF) is a cooperative ing and the importing (imposing) country. This interagency effort (involving the IMF, the Interna- methodology focuses on the effects of investiga- tional Trade Centre, UNCTAD, the United Nations tions, not their outcomes, and therefore captures Development Programme, the World Bank, and the potential "chilling" effect of the mere launch- the WTO) supported by bilateral donors aimed at ing of investigations on trade flows. facilitating coordination of trade-related techni- 33. The definitions and figures used here come cal assistance to the least-developed countries and from the OECD-DAC/WTO database on trade- mainstreaming trade into national development related assistance. Figures refer to commitments. and poverty reduction strategies. Trade policy and regulations include support for 38. The task force included Barbados, Brazil, participation in multilateral trade negotiations, Canada, China, Colombia, the European Union, analysis and implementation of multilateral trade Japan, India, Thailand, the United States, and agreements, trade policy mainstreaming and tech- the coordinators of the ACP (Africa, Caribbean nical standards, trade facilitation including tariff and Pacific), African, and least-developed-country structures and customs regimes, support to PTAs, groups. and human resource development in trade. Trade 39. The task force also highlighted the need development activities include business develop- to address regional or cross-country aid for trade ment, activities aimed at improving the business issues, where cooperation on trade-related proj- climate, access to trade finance, and trade promo- ects could help promote the competitiveness of tion, including at the enterprise and institutional low-income countries in ways that purely national level. interventions could only do inefficiently or not at 34. For a broader overview of World Bank and all. This issue had also been highlighted by the IMF activities on aid for trade, see IMF-World World Bank and the IMF, which put forward a Bank (2006c). range of proposals to address these concerns at 35. The TIM seeks to assist IMF member coun- their 2006 annual meetings. The Development tries in meeting balance of payments shortfalls that Committee agreed on the need to improve exist- might result from multilateral trade liberalization. ing instruments to address cross-country and In addition to its regular policy dialogue on trade regional projects and to strengthen the monitor- with member countries in the context of Article ing of regional initiatives and funding. IV consultations and IMF-supported programs, 40. Particular attention is given to the EPA the IMF has provided diagnostic support and agreements becausee they concern many of the policy discussions to member countries, as well as world's poorest countries, which are facing par- an increased focus on trade and trade facilitation ticular challenges in meeting the MDGs. Further- issues in technical assistance for customs and tax more, EPAs have implications for implementation reform. of existing preferential trade arrangements in 36. Several regular, ongoing IMF arrangements Africa. Development of PTAs in other regions, support adjustment in the context of domestic notably Asia, is generally among the higher-growth trade reform. Thus far there has been no request middle-income countries. for the incorporation of floating tranches related 41. For a fuller discussion, see World Bank, to trade adjustment in Fund arrangements. Global Economic Prospects 2005. Monitoring the Performance of International Financial Institutions T he environment in which the interna- banks have devoted considerable attention tional financial institutions (IFIs)--the to clarifying roles and determining priorities. World Bank, the International Mone- Although deliberations are still under way, tary Fund (IMF), and the regional development and agreement has not yet been reached on banks--operate today is different from that of all the trade-offs, discussions to date have just a few years ago. Globalization, a growing highlighted five key challenges: differentiation among developing countries, the availability of alternative financial resources, How best to support progress toward the and the multiplication of actors on the devel- MDGs in the poorest countries opment landscape--all these have forced IFIs How to strengthen and adapt the IFIs' to adapt their strategies for supporting develop- engagement in middle-income countries ing countries' efforts to meet the Millennium How to respond to the challenges of glo- Development Goals (MDGs). Through closer balization and the need for global public collaboration with one another and with devel- goods opment partners, and through reform of their How to better promote coherence and col- own governance, these institutions are seeking laboration among IFIs and between them greater legitimacy and relevance in a world of and their development partners, and overlapping and increasingly complex devel- How to strengthen the voice and represen- opment mandates. This chapter examines the tation of developing countries in the gov- responsibilities of the IFIs within the Monterrey ernance of the IFIs. compact and their recent performance in carry- ing out those responsibilities. Accelerating Progress on MDGs in the Poorest Countries Increased financing, advice, and capacity All IFIs are constantly adapting their strat- support from the IFIs is critical if low-income egies to respond to new demands and the countries are to sustain recent progress in changing external environment. The IMF implementing the MDGs. Poverty reduction recently conducted a broad-based review of strategies provide a sound framework for its medium-term strategy (box 5.1).1 The countries to articulate development priori- World Bank and the regional development ties and for IFIs to support countries' efforts. The IMF's medium-term strategy, published in September 2005, considered the future direction of the IMF in areas key to its lead mandate on international monetary cooperation and global financial stability. Surveillance: The IMF is enhancing the effectiveness of surveillance through greater focus, can- dor, and even-handedness. The medium-term strategy is proceeding on two parallel tracks: imple- mentation of surveillance and development of its legal basis. Steps to enhance the implementation of surveillance have included initiation of the first multilateral consultation, a new modality for discussing common problems; deeper analytical work on exchange rates, including extension of the existing multicountry framework to emerging market economies; strengthening of the IMF's analytical and advisory capacity on financial sector and capital market issues, with better integra- tion of this work into surveillance; greater focus on cross-border spillovers, regional issues, and cross-country issues; and stronger outreach. The legal framework is being revisited through an ongoing review of the 1977 decision on surveillance over exchange rate policies and consideration of a possible remit-independence-accountability framework. Emerging market economies: The IMF is strengthening its advice on financial sector and capital market issues and considering the adequacy of instruments to support members, as well as the pos- sibility of a new contingent financing instrument for crisis prevention. Low-income countries: The IMF is enhancing support for efforts to achieve the MDGs by sharp- ening its focus on issues critical for growth within its macroeconomic and financial areas of respon- sibility, providing assistance for capacity building in these areas, helping meet challenges of effective use of increased aid inflows and debt relief, and supporting the development of debt strategies and improved debt management. Capacity building: The IMF is improving alignment with members' needs and its own strategic priorities, taking advantage of complementarities with other providers. Governance: Work in the area of governance is currently focused on reform of quotas and voice; other priority issues include the management selection process and the role of the IMF Executive Board. Efficient operations: The IMF is enhancing efficiency and prioritization in its operational work and support activities and strengthening its risk management systems. Efforts under way have resulted in a number of streamlining initiatives, as well as real reductions in the IMF's administrative budget in recent years. These efforts are being complemented by consideration of ways to strengthen the IMF's income base and the ongoing review of World Bank-IMF collaboration (see below). Source: IMF 2006. "Managing Director's Report on the IMF's Medium Term Strategy." September SM/05/332. Countries are now setting clear goals and global levels. Coordination has become para- targets linked to public actions, improving mount not only because of the increased num- their budgeting and monitoring systems, and ber of players, but also because many of the opening the public space to a more inclusive new providers deliver aid in a more fragmented discussion of national priorities and policies. fashion and outside the established domestic On the donor side, progress has been made process and framework. This includes ensur- to align and harmonize assistance with coun- ing that resources from vertical funds support tries' priorities, and filling country-specific country-specific development priorities, and analytical gaps. Yet connecting results with that the delivery of aid reinforces rather than resources remains a major challenge. undermines domestic processes (including bud- Increased attention is being devoted to get formulation and execution). As discussed coordinating aid at the country, regional, and in chapter 4, progress has been made in some of these areas, but scaling up has been lim- dynamic and adaptive in order to concentrate ited. The upcoming 15th replenishment of the on those areas that promise strong results and International Development Association (IDA) reflect the World Bank's evolving role in the will be an important test of donors' intentions development partnership. regarding not only their 2005 commitments IFIs are also paying greater attention to (to support the Multilateral Debt Relief Initia- the special needs of states with weak policies tive (MDRI) and scale up official development and institutions, and to tailoring support to assistance), but also the role they see for the different groups of low-income countries multilateral development banks (MDBs) in a with different needs. How they are strength- changing aid landscape. ening countries' capacity to promote growth The Africa Action Plan (AAP) illustrates and deliver basic services to the citizens of how the World Bank is working to promote fragile states is described in the final section country-led efforts in partnership with other of this chapter. donors (box 5.2).2 The AAP is based on an outcome-oriented framework to guide the Strengthening and Adapting Engagement work of the Bank's Africa region in four in Middle-Income Countries pillars: accelerating shared growth, build- ing capable states, sharpening the focus on The growing differentiation in development results, and strengthening the development conditions across countries and the increased partnership. The AAP was designed to be availability of alternative financial resources The first progress report on the AAP, to be presented to the World Bank's Board of Executive Directors in spring 2007, examines emerging regional and international trends and assesses progress against each of the four pillars on which the AAP is based. It also proposes changes to strengthen the World Bank's role in the development partnership in Africa. Specifically, the report recommends increasing efforts to accelerate economic growth by deploying resources in a concerted effort to overcome the most critical constraints to growth, supporting good governance and capacity development in resource-rich but slowly growing economies, using innovative instruments to mobilize development finance, and helping integrate vertical programs and new partners into sustainable country-based institutions. Implementation progress in fiscal years 2005/06 and 2006/07 was broadly satisfactory: Progress has been best in the shared growth pillar--supporting the drivers of growth and par- ticipating in growth. The AAP is on track to deliver the results committed to in all but two (agricultural productivity and gender) of the pillar's nine thematic areas. It is ahead of projected progress in four areas. Private sector development, closing the infrastructure gap, and addressing HIV/AIDS and malaria have shown significant progress, both in increased Bank Group support and from evidence that countries are closer to delivering development outcomes. Good progress has been achieved in establishing the preconditions for an export push, in regional integration, and in primary education, including addressing gender discrimination. Progress is on track in supporting skills development, and the IFC has played a leading role in pushing business educa- tion. Accelerated progress will be needed to increase agricultural productivity and to connect the poor to markets. Despite some promising initiatives, substantially more work is needed to increase the economic empowerment of women. The capable states pillar has supported African governments in improving the transparency, accountability, and provision of social services, but progress has been mixed. Good progress continued was made in improving public financial management and in rolling out the Extractive Industries Transparency Initiative. The Africa Region's Capacity Development Management Action Plan (CDMAP) supports countries in building capable states. More progress could have been made with an earlier launch of the program. The results pillar is assisting countries in developing operational strategies to deliver develop- ment outcomes. There has been good progress on the results framework, and the Bank Group is on track to deliver the priority actions. Some African countries, such as Burkina Faso, Ghana, Mozambique, and Tanzania, undertook substantial efforts to clarify their development goals and targets, based on a medium- to long-term vision, and to link these to public actions. They also developed action plans to improve monitoring and evaluation. Implementation of statisti- cal capacity building and accelerated data programs have picked up pace; however, progress in building statistical capacity remains low throughout Africa and lags other regions. Progress on the Bank's results agenda has been sufficient to mainstream these efforts into the day-to-day management of the AAP. The global development partnership pillar is leveraging IDA-14 for greater impact. There has been considerable progress on the partnership pillar. Countries have taken the lead in develop- ing baselines and action plans for the Paris Declaration with development partners. Progress on harmonization and alignment at the policy level--expressed through the working groups of the OECD Development Assistance Committee--has been encouraging. Progress includes work on selective scaling up of aid for Africa, the launch of resources and results processes, and improved alignment with development partners to the new generation of poverty reduction strategies. The Africa Catalytic Growth Fund received initial funding and has launched operations designed to crowd in substantial donor support. The IFC and the Multilateral Investment Guarantee Agency (MIGA) have made substantial contributions to implementing the AAP. Much of the progress in the shared growth pillar is due to the joint efforts of IDA, the IFC, and MIGA in private sector development, infrastructure, and skills development. The IFC began its Strategic Initiative for Africa based on three objectives: improving investment climates, improving support for small and medium-size enterprises, and supporting project development for potential IFC projects. Despite good progress, significant changes to the AAP are needed in light of the development picture in Africa, results achieved during early implementation, the World Bank Group's evolving role in the development partnership, and global priorities needing collective action. The AAP will focus more selectively on outcomes over the next three years, concentrating on areas that promise strong results and reflect the Bank's evolving role in the development partnership. It will also use a new country classification, based on economic performance and institutional capacity, to guide the implementation strategy. And it will strengthen the leverage of IDA by adapting its strategy for scaling up resources. Source: World Bank 2007. have important bearing on IFIs' support but many middle-income countries are also to middle-income countries. Traditional increasingly looking for more customized bundled lending and knowledge manage- financial and advisory services, unbundled ment products remain important for many from financing itself. Global public goods is middle-income countries, particularly those one area where the IFIs are expected to play with credit ratings below investment grade, a key role in supporting middle-income coun- tries' needs for customized financial and advi- more generally to the inadequate provision sory services (see below). of global public goods and to the increasing Despite real progress in many middle- role middle-income countries are expected income countries, the support of the IFIs to play in their provision. The recent report remains critical to these countries, which are of the International Task Force on Global home to 70 percent of the world's poor and Public Goods highlights the factors that con- still face major challenges in attaining the strain the provision of global public goods.4 MDGs. The IFIs retain comparative advan- Here the IFIs can play a central role, among tages in providing strategic policy advice, other things in enhancing international finan- development finance and financial services, cial stability, strengthening the international and technical assistance and knowledge trading system, addressing climate change, services, but this support needs to be bet- preventing the emergence and spread of infec- ter tailored to specific country conditions. tious disease, and generating knowledge. The dialogue between IFIs and their middle- One vital area where IFIs can have a income clients has also highlighted a number broader role is that of clean energy and cli- of impediments, such as the responsiveness mate change. The communiqué issued by the of the institutions to countries' concerns and leaders of the Group of Eight countries at the cost of doing business with them. These their July 2005 Gleneagles summit called on impediments must be tackled if the IFIs are to the World Bank to take a leadership role in provide the full measure of support of which creating a new framework for clean energy they are capable. and development. This request was reaf- The focus of discussion by the heads of firmed by the Development Committee in MDBs and the Development Committee September 2005, and the World Bank pre- last September in Singapore was on how to sented a framework for clean energy at the adapt strategies, priorities, and instruments Spring 2006 Meetings (box 5.3).5 Also the in an environment in which the IFIs provide Asian Development Bank (ADB) established a smaller share of financial flows to middle- the Asia Pacific Carbon Fund in November income countries than they once did.3 The 2006, and the Water Financing Partnership Development Committee strongly endorsed Facility in December 2006. the World Bank's corporate role and mission to eradicate poverty in its partnerships with Improving Coherence and Cooperation middle-income countries, reviewing the Bank's among Institutions proposals to strengthen the IBRD's value added and engagement in response to the evolving The Monterrey conference placed a high pri- and diverse needs of middle-income coun- ority on improving coherence and coopera- tries. The heads of the MDBs agreed to move tion among key multilateral players. This is ahead on three fronts in the middle-income especially important given the growing inter- countries' agenda: holding joint consultations connectedness of the development agenda. with middle-income countries; exploring the Cooperation among MDBs has been improv- possibilities for blending bilateral grant and ing in recent years, evolving from ad hoc multilateral lending resources; and expanding consultations to systemic cooperation across ongoing joint analytical work, technical assis- a broad range of issues. The heads of MDBs tance, and advisory and operational work. have by now articulated and published joint positions on most major global development challenges.6 Increasing the Provision of Global Thematic cooperation, often pursued Public Goods through technical working groups endorsed by Growing cross-country interactions and the heads of MDBs, has been central to these interdependence have also brought attention coherence-building efforts. It has encompassed "Clean Energy and Development: Towards an Investment Framework," the World Bank's frame- work on clean energy, is structured around three pillars: access, mitigation, and adaptation. The strategy supports widening access to energy services, efforts to control greenhouse gas emissions, and assistance to developing countries in adapting to climate risks. The approach was broadly endorsed by the Bank's Development Committee in April 2006. A second World Bank report was presented at the September 2006 Annual Meetings at the request of the Development Committee. That report, "An Investment Framework for Clean Energy and Devel- opment: A Progress Report," assessed the potential value of new and existing financial instruments for accelerating progress in each of these pillars. It argued that the current financing instruments for energy access and adaptation are adequate, but that financing needs far outstrip available funds. The Bank is currently working with donors to increase concessional financing for the access agenda in Sub-Saharan Africa and performing more detailed analyses of what financing is needed for the adaptation agenda. In contrast to instruments for energy access and adaptation, current financing instruments and resources for designing and scaling up mitigation efforts are inadequate. The report suggested that a clean energy financing vehicle, with an initial capitalization of $10 billion, is needed. Such an instru- ment would blend concessional and carbon financing to fund the upfront capital costs of low-carbon technologies; it would be recapitalized through loan repayments and the sale of carbon credits. The Development Committee did not endorse the proposed financing vehicle, arguing that better use should be made of existing instruments. A long-term, stable, global, and equitable regulatory framework is required to stimulate an international carbon market that could transfer tens of billions of dollars a year to developing countries in return for reducing their emissions. Without a significant increase in financing, progress on transitioning to a low-carbon economy is severely constrained, exacerbating climate changes and increasing the need for adaptation measures. The Bank and potential donors are discussing the concept of a carbon continuity fund that aims to ensure the carbon market does not collapse while governments negotiate a post-2012 regulatory framework. In addition to developing an investment framework, the Bank is responding to other climate change­related mandates that emerged from the Gleneagles summit of the Group of Eight. It is work- ing with the G8+5 countries to develop national action plans for a transition to a low-carbon economy. These plans would analyze which sectors and technologies provide the best opportunity for reducing greenhouse gas emissions, what policies would be needed, and what financing would be required. The lessons will be integrated in future country assistance strategies in order to promote growth options that are less intensive in greenhouse gas emissions but still meet developing-country priorities. The Bank is also developing a screening tool that will provide the core tools for World Bank staff and client countries to assess the exposure of investments to risks from climate change and develop steps for dealing with them. Pilot adaptation studies are being prepared in several countries. Finally, the IFC has responded to the mandate to develop local commercial capacity to develop and finance cost-effective energy efficiency and low-carbon energy projects by providing special- ized credit lines and credit enhancement vehicles to local banks and leasing companies to establish self-sustaining lending products for sustainable energy. Source: World Bank 2006. "Clean Energy and Development: Towards an Investment Framework." all key aspects of the current development and anticorruption. MDBs that operate con- agenda, including the harmonization of pro- cessional windows have made considerable curement, financial management, environment progress in harmonizing their performance- and safeguard policies, investment climate sur- based allocation approaches, including related veys, and capacity development, governance, country and institutional assessments. Memoranda of Understanding underpin share of the four countries from 5.4 percent to much of this cooperation at the country level. 7.1 percent (and their voting shares from 5.3 All the regional development banks have percent to 7.0 percent). such agreements with the World Bank, and Work has since begun on additional mea- the ADB has a similar agreement with the sures. One of these is a new formula to guide Inter-American Development Bank (IADB). the assessment of the adequacy of members' More explicit and accountable cooperation quotas in the IMF, in preparation for a second frameworks at the country level would be round of ad hoc quota increases. Another is a however useful for addressing duplication proposal to increase the "basic" votes allot- and allowing MDBs to undertake operations ted to each member, to ensure adequate voice on a larger scale. for low-income countries. Reform will also Collaboration between the World Bank involve steps to enhance the capacity of Afri- and the IMF and their division of responsi- can Executive Directors' offices to participate bilities have also received considerable atten- effectively in the governance of the IMF. These tion. In 2006 the managing director of the reforms will include an increase in staffing IMF and the president of the World Bank resources and possible amendment of the Arti- commissioned an external review on collab- cles of Agreement to enable Executive Direc- oration between the two institutions.7 The tors elected by a large number of members to report of the committee identifies scope for appoint more than one alternate director. improvement and makes several recommen- The Development Committee in Singapore dations. It calls on the institutions to develop "welcomed the [IMF] Managing Director's a new "understanding on collaboration"; report on progress made in the reform of IMF to strengthen cooperation on crisis manage- quotas and voice, acknowledging the mea- ment; to improve integration and harmoni- sures already taken by the Bank to enhance zation of work on fiscal issues; to clarify the capacity in Executive Directors' offices and IMF's role in low-income countries; and to capitals of developing and transition coun- continue to improve collaboration on finan- tries, [and] asked the Bank to work with its cial sector issues. shareholders to consider enhancement in voice and participation in the governance of the Bank." A series of consultations with the Increasing Voice and Participation World Bank executive directors is ongoing, by Developing Countries building on the issues identified by the Devel- At the Singapore meetings, the IMF's Board of opment Committee in previous meetings. Governors approved a package of reforms on quotas and voice aimed at better aligning the IMF's current governance regime with its mem- bers' relative positions in the world economy, and enhancing the voice and participation of Measuring the contribution of the IFIs to devel- developing countries within the IMF. These opment effectiveness is difficult, and the mea- reforms are designed as an integrated two- sures used are imperfect. This section reviews year program to be completed no later than the measurement of performance and suggests the 2008 annual meeting of the IMF Board ways of strengthening such evaluations. of Governors. The first step in the program-- increasing quotas for a small group of the most The Challenge of Assessing Effectiveness underrepresented countries, including China, the Republic of Korea, Mexico, and Turkey-- IFIs play a multitude of roles. Some of these, was also approved at Singapore. These ad hoc such as supporting macroeconomic stability, quota changes would increase total IMF quo- providing development finance, and reporting tas by 1.8 percent and would raise the quota on country performance, are more tangible than others and thus more amenable to assess- of national experiences, and through impact ment. Others, such as generating knowledge evaluations of specific interventions. Rigor- to inform policy makers or convening donor ous impact evaluation is indeed an important support around country strategies, are more tool for generating knowledge and insight to difficult. help guide policy formulation. Although traditional portfolio perfor- IFIs and other groups, including inde- mance indicators such as financial disburse- pendent evaluation agencies, international ments are relatively easily tracked, they say assessment initiatives, and interagency work- little about the IFIs' contributions to devel- ing groups, use various approaches to evalu- opment. As the ultimate objective of all the ate development effectiveness, as Global institutions' activities is to foster develop- Monitoring Report 2006 discussed. Each IFI ment outcomes on the ground, measurement has an independent evaluation agency that strategies must capture changes--or the lack conducts evaluations at the project, country, of changes--in the lives of households, the and sector or institutionwide levels (box 5.4). activities of firms, and organizations on the While such assessments play an important ground. evaluation role, some have asked that these The difficulty in assessing performance evaluations be complemented by external is compounded by the long time it takes for evaluations. many development initiatives to bear fruit, Three aspects of IFIs' performance--finan- and by weaknesses in statistics and moni- cial support, efforts to strengthen results- toring. But a central constraint is that of based management, and progress toward attribution: a reduction in child mortality, harmonizing and aligning aid through the for example, may be due in part to aid-sup- Paris Declaration--are discussed below. ported programs, but many other factors may also be at work. Clear attribution of credit Financial Resources for Development or blame is often not possible. Moreover, because money is fungible, it may be hard to IFIs provide a smaller share of financial know exactly what a given package of aid is flows to developing countries than they once financing, even if it is earmarked. did. Net official lending to developing coun- The challenge of attribution is made more tries has declined in recent years, while net difficult by the complex chain of causality private lending has significantly expanded. linking external financing to development Private flows are estimated to have reached outcomes. External assistance to policy mak- a record $643 billion in 2006, up from ers may influence the policy debate, sometimes $551 billion in 2005 and about equally split through conditionality that is attached while between foreign direct investment ($325 operating with imperfect knowledge and little billion), on the one hand, and private debt control over implementation. National and and portfolio equity ($318 billion), on the local policy makers manage the external dia- other (figure 5.1). Recorded private remit- logue with donors, but their ability to set and tances also continued their upward trend, implement policies often depends on politics reaching almost $200 billion in 2006. The and the quality and capacity of bureaucracies breakdown of private capital inflows under- and institutions. All this is compounded by lying figure 5.1 reveals the sharp difference the uncertainty over how policies themselves in their importance across country groups. influence final outcomes. The lion's share of flows is directed to mid- Although far from complete, a stock of dle-income and `blend' countries (those knowledge does exist on the development countries eligible for both concessional and impact of many policies. Economic research nonconcessional financing from IFIs, such as and evaluations can generate this knowl- India and Indonesia). By contrast, only 1.6 edge through ex ante and ex post analysis percent of private debt and portfolio equity Evaluations in 2006 by the World Bank's Independent Evaluation Group (IEG) shed light on how well the institution is reducing poverty, working with fragile states, implementing sectoral programs in education, infrastructure, and the environment, and responding to natural disasters. Reducing poverty: The Annual Review of Development Effectiveness 2006 identified three key areas where the Bank can strengthen its effectiveness in helping countries reduce poverty: Ensure that growth leads to jobs for the poor and productivity increases in poorer regions and sectors where the poor earn their incomes. Clearly articulate results chains so that Bank country assistance programs and individual proj- ects set realistic objectives, consider key cross-sectoral constraints to achieving them, and pay adequate attention to building capacity. Tailor efforts to increase the accountability of public sector institutions to local conditions. Working with fragile states: The Bank has contributed to macroeconomic stability and to the delivery of significant amounts of physical infrastructure, especially in postconflict situations. How- ever, reforms in some fragile states have lacked selectivity and prioritization, and while the Bank has generally been effective in the immediate postconflict phase, its effectiveness needs to be improved following this phase, when structural change is needed. Implementing sectoral programs: Bank projects in education have helped raise enrollment rates, but have paid less attention to learning outcomes. Best practice appears to be doing both simulta- neously. Infrastructure investments have contributed to growth and poverty alleviation, but often have imposed environmental burdens. Opportunities to mitigate these burdens require broader assessments of environmental impact, and sound national environmental strategies in designing and implementing infrastructure programs, setting environmental standards, and coordinating programs across sectors over a reasonably long time horizon. IEG also examined the Bank's work in looking at natural disasters. The Bank is the largest funding agency for disaster recovery and reconstruction in developing countries. Since 1984, the Bank has financed a total of $26 billion in disaster relief activities. The more than 500 projects involved represent almost 10 percent of all Bank loan commitments during this period. Over 80 percent of Bank disaster relief financing has addressed rapid onset disasters--floods, earthquakes, tropical storms, and fires. Within disaster relief projects, the Bank did better at reconstructing damaged infrastructure and housing than it did in reducing vulnerabilities and addressing their root causes. It is possible to anticipate where many natural disasters will strike, yet the Bank's disaster assistance efforts are underutilizing these vital lifesaving forecasts. inflows are received by the 63 low-income fluctuation or a permanent departure from countries, and only 4.7 percent of foreign recent trends, two occurrences stand out: a direct investment. Remittances, however, sustained increase in nonconcessional lend- are relatively important, with low-income ing by the International Finance Corporation countries receiving nearly 15 percent of total (IFC) and the European Bank for Reconstruc- estimated inflows. tion and Development (EBRD), mainly to the Despite the increasing share of private private sector; and a slight decline in conces- sources of financing, the IFIs' role remains sional lending and overall disbursements to important. In 2006, the five MDBs dis- Africa. bursed about $43 billion, a 20 percent Nonconcessional flows. Net official lending increase over 2005. Although it is too early declined in the first part of this decade, partic- to assess whether it represents a temporary ularly because of net repayments to the IMF of Net private capital flows to developing countries 1998 1999 2000 2001 2002 2003 2004 2005 2006 Private debt and portfolio equity (net inflows) 24.7 17.7 20.6 -6.5 12.1 113.7 194.6 270.6 318 IBRD and blend countries 26.2 18.0 21.5 ­4.2 13.5 114.4 191.2 266.2 IDA oil-exporters ­1.2 0.6 ­0.5 ­0.6 0.7 ­0.3 0.1 0.4 Other IDA countries ­0.4 -0.9 ­0.4 ­1.8 ­2.1 ­0.4 3.3 4.0 Foreign direct investment (net inflows) 170.0 178.0 166.5 171.0 157.1 160.0 217.8 280.8 325 IBRD and blend countries 161.1 167.1 157.9 161.2 145.4 144.8 203.9 267.6 IDA oil-exporters 4.7 5.4 4.6 4.3 4.9 6.5 7.4 7.7 Other IDA countries 4.2 5.5 4.0 5.5 6.7 8.8 6.6 5.5 Recorded Remittances (received) 70.7 76.6 83.8 95.3 116.2 143.8 163.7 189.5 200 IBRD and blend countries 61.3 66.7 73.1 82.1 100.0 125.1 140.9 161.9 IDA oil-exporters 5.9 6.8 7.4 8.0 10.0 12.4 14.6 17.9 Other IDA countries 3.4 3.2 3.3 5.2 6.2 6.3 8.2 9.7 Source: Work Bank Debt Reporting System and staff estimates. Note: e = estimate. large disbursements of emergency assistance which increased its lending in dollar terms by (box 5.5) as well as early repayments of World 80 percent,8 followed by the IBRD, and the Bank loans by some middle-income countries. IFC, which increased their lending by 31 per- However, gross disbursements--a proxy for cent and 38 percent respectively. Most of the new demand--suggest that nonconcessional recent increase went to Europe (figure 5.3). lending by MDBs to middle-income countries Concessional flows. Growth in conces- remains strong (figure 5.2). Non-conces- sional lending in previous years had been sional gross disbursements increased by 29 driven by IDA's contribution, with support for percent to about $32 billion, in 2006. This Sub-Saharan Africa growing sharply between represents the first significant increase in non- 2000 and 2004 and support for East and South concessional lending that was not preceded Asia rising in 2004. This support fell in 2005­ by a financial crisis. By far the greatest share 06 (figure 5.4), with lending by IDA declin- of the 2006 increase came from the EBRD, ing but lending by other MDBs continuing to The IMF's General Resources Account (GRA) provides nonconcessional financial support to mem- ber countries experiencing temporary balance of payments difficulties. The IMF also provides finan- cial support through special GRA facilities and policies (including emergency assistance for natural disasters and postconflict emergency assistance) and concessional loans to low-income countries under the Poverty Reduction and Growth Facility (PRGF). GRA net flows depend largely on the needs of large middle-income countries facing economic crises; they are consequently erratic on a year-to-year basis. Following early repayments of large loans by Argentina, Brazil, and Indonesia, the IMF's GRA credit outstanding declined to SDR 9.8 billion ($15 billion) at end-2006, its lowest level in 25 years, and well below its all-time peak of SDR 70 billion ($105 billion) in 2003. Net PRGF lending is less erratic but also substantially affected by the needs of larger low-income mem- ber countries and the provision of debt relief. Net flows from the IMF to developing countries, 2000­06 (in millions of U.S. dollars) 2000 2001 2002 2003 2004 2005 2006 Repayments b Of which: Sub-Saharan Africa ­22 ­178 165 ­394 ­318 ­739 ­ Source: IMF Finance Department. a. Includes disbursements and repayments of PRGF, SAF and Trust Fund loans. b. The sharp increase in repayments in 2006 reflects the provision of MDRI debt relief. c. Korea is classified as a high-income country and so their GRA repurchase of SDR 4,462.5 million in 2001 is excluded. increase. The dwindling relevance of the MDBs regional banks, peaked in the 1980s and have in leveraging external resources--due to a pos- declined thereafter. IDA's share in total mul- sible decline of contributions to IDA and the tilateral ODA declined from 42 percent in regional development banks (RDBs) as a share the 1970s to an average of 20 percent in the of total official development assistance--may 2001­05 period. further affect their role in supporting achieve- An additional factor that may affect the ment of the MDGs in low-income countries. future ability of IDA and the RDBs to sup- Among multilateral organizations, IDA's role port low-income countries could result from as main channel for multilateral ODA has the impact of debt relief provided under the been surpassed by the European Commis- Heavily Indebted Poor Countries (HIPC) and sion and the United Nations since the 1990s. Multilateral Debt Relief Initiative (MDRI). The amounts of core contributions channeled Through both of these initiatives, IFIs are through IDA and, on a smaller scale, through providing large amounts of debt relief to poor countries.9 Recognizing this, donors Although donors have agreed on the need have specified that additional contributions to have firm MDRI financing commitments are to be calculated relative to a baseline that backing the 10-year disbursement period of maintains current contribution levels in real each future IDA replenishment, their actions terms. do not yet reflect these commitments (figure 5.5). Monitoring donors' commitments on financing the MDRI is important to ensur- Concessional and nonconcessional lending by MDBs, ing the additionality of donor financing over 1999­2006 time. Of the 34 donor countries that have pledged to contribute to the MDRI replen- ishment of IDA, 28 had provided their Instruments of Commitment (IoC) as of end-December 2006.10 IoC provide firm or unqualified financing commitments of $3.8 billion (representing 10 percent of the origi- nal projected cost of the MDRI) and qualified financing commitments of $20.5 billion (56 percent of total MDRI costs).11 This leaves a gap between total costs and commitments of $12.4 billion (34 percent of total MDRI costs). Regarding forgone credit reflows resulting from the HIPC Initiative, donors have provided firm financing commitments to cover $1.4 billion in HIPC costs occurring under IDA's 14th replen- ishment (IDA-14). Beyond that, donor com- Source: Staff of the big five multilateral development banks. mitments will be needed to cover HIPC and Nonconcessional lending by MDBs to different regions (gross disbursements), 1999­2006 Source: Staff of the big five multilateral development banks. MDRI financing over the next two decades or ity. The upcoming replenishments of the MDBs' so. Donors need to be mindful that any short- concessional windows will be an important test fall between targeted and actual commitments of donor's intentions regarding their support undermines IDA's long-term financial capabil- for the MDRI and the future role they see for MDBs in a changing aid environment. Gross disbursements of concessional lending by MDBs, 1999­2006 Selectivity of Financial Resources in Support of the Development Agenda As part of the Monterrey compact, MDBs com- mitted themselves to using more transparent and incentive-improving resource allocation systems aimed at maximizing aid effectiveness and encouraging stronger policies and insti- tutions in recipient countries. At present, the foundation of each of these systems is a for- mula that calculates the share of the resources that will be allocated to individual countries on the basis of their financial need (proxied by population and income per capita) and per- formance. Each MDB combines these factors somewhat differently in its performance allo- cation formula and uses different methods to Source: Staff of the big five multilateral development banks. Donor financing commitments to IDA under the Multilateral Debt Relief Initiative, as of December 31, 2006 ($ million equivalent) Source: World Bank 2007. "MDRI: Update on Debt Relief by IDA and Donor Financing to Date." accommodate exceptional circumstances, such additional resources over and above their as countries in postconflict situations. In the performance-based allocation for a limited past, however, MDBs have taken significant period. steps to harmonize their performance-based allocation (PBA) systems and country perfor- Progress in Results Management mance and institutional assessment (CPIA) questionnaires. Because of their performance- The Third Roundtable on Managing for based allocation formula and the use of per Development, held in Hanoi in February 2007, capita income ceilings to determine countries' built on the findings of the 2004 Marrakech eligibility for MDBs' concessional resources, Roundtable. It provided a venue and format MDBs continue to exhibit higher policy and for each of the 43 country delegations to sum- poverty selectivity than bilateral aid agencies marize their experiences and to initiate a coun- (See figure 5.6.).12 try action planning process, with targets for As described in the final section of this steps to be completed in advance of the Ghana chapter, the MDBs recognize fragile states' High-Level Forum on Aid Effectiveness to be special needs and circumstances and their dif- held in September 2008. The Hanoi Round- ficulties in making investments that promise table provided compelling evidence that coun- sufficient returns to enable repayment even of try partners are keenly interested in improving concessional loans. In response, MDBs have the effectiveness of development assistance increasingly offered support in the form of and domestic resources by strengthening sys- grants, which now make up a much larger tems that enable information on expected and percentage of disbursements to them (31 per- actual results to be used in decision making. cent) than among other low-income countries The agenda for the Hanoi meeting was (9 percent; figure 5.7). Also, both the AfDB based on the recommendations of country and the World Bank have developed excep- practitioners and development partners made tional allocation frameworks for postconflict through an 18-month Mutual Learning Initia- countries to allow countries to benefit from tive supported by the Joint Venture on Manag- Policy and poverty selectivity of concessional assistance by MDBs Source: World Bank staff calculation based on Dollar and Levin (2004). ing for Results (one of four subgroups working Grants and loans as shares of MBD concessional on behalf of the Working Party on Aid Effec- disbursements in 2006 tiveness of the OECD-DAC). Representatives of 22 countries and bilateral and multilateral agencies met in a series of workshops (in Burkina Faso, Singapore, Uganda, and Uru- guay) leading up to the Roundtable.13 This process, as well as country experience with poverty monitoring, suggests that build- ing country capacity to manage for develop- Loans ment results needs to focus on five thematic Loans 69% areas:14 91% Leadership and political will Strong links from results to planning and budgeting processes, to strengthen incen- Source: Staff of four MDBs: World Bank, AfDB, IADB, and ADB. tives to use information on expected and actual results in decision making Evaluation and monitoring tools necessary to generate feedback on the performance tation is constrained by fragmentation at of policies and programs the line ministry and agency level. Country Mechanisms established by donors and partners are thus keen to develop results country partners that encourage mutual frameworks at the sectoral level. These accountability frameworks could then serve as a basis Statistical capacity in developing coun- for harmonizing donor efforts to moni- tries and systems for applied data use tor and evaluate specific interventions and in government, both to supply and help overcome the obstacles presented by the generate greater demand for managing presence of multiple, partial and donor- for development results in developing driven, monitoring systems. Countries also countries (box 5.6).15 expressed an interest in exploring the use of performance-based management tools, Several observations that emerged from including output- and outcome-based dis- the Roundtable relate to the progress that bursement principles, and they identified country and donor partners are making as the need to engage key stakeholders such they work to achieve greater development as legislators in both defining and moni- effectiveness through managing for results: toring the achievement of results. A par- ticularly strong common theme running Progress and opportunities at the country through the planning discussions was the level. A key achievement of the Roundta- importance of grounding results-based ble is that each of the country delegations management systems in stronger account- worked to initiate a country action plan- ability to citizens. Many country action ning process, to identify ways that they plans proposed to strengthen participatory could work to strengthen their own capac- approaches and to ensure that results were ity to manage for results, applying lessons communicated transparently to the public. learned along each of the five thematic Many countries were keen to explore better elements of capacity. Many countries rec- methods for assessing and tracking citizen ognized that, despite real progress toward satisfaction. All of the country delegations articulating a poverty-monitoring frame- stressed the need to strengthen statistical work at the national level, its implemen- capacity to ensure that the information nec- To address short-term data needs, a pilot Accelerated Data Program (ADP) was launched in 2006. Its goal is to produce relevant data for policy design, monitoring, and evaluation by implementing a coordinated program of data collection, analysis, and dissemination. The program focuses on sample household surveys, because they provide estimates of many indicators relating to the MDGs and other key outcomes, as well as data needed for research and impact evaluation. The pilot ADP is being implemented in selected countries in Africa, Asia, and Latin America by the World Bank and the PARIS21 Secretariat of the OECD, in cooperation with multiple partners (UN agencies and others). Two million dollars a year has been allocated for the pilot ADP through the Development Grant Facility for the Marrakech Action Plan for Statistics (MAPS) for 2006­08. The MAPS and the ADP provide robust frameworks, but a stronger commitment from donor agencies is still needed. The PARIS21 Secretariat estimates that development partners are spending about $70 million a year on statistical capacity improvement in Sub-Saharan Africa. Implementa- tion of the MAPS in IDA countries in Sub-Saharan Africa, including the scaling up of the pilot ADP, is estimated to require an additional $60 million a year. Source: World Bank. "Better Data for Better Results: An Action Plan for Improving Development Statistics." essary for an effective results management so as to think of results management as a system is made available. Countries put a country system, which, along with those high priority on finding ways to learn from for procurement and financial manage- countries that have done more to build up ment, permits greater accountability and their systems (Chile, China, Thailand, Viet- more credible feedback on performance. nam, and others). They recognized the value The issue of donor agency effectiveness of peer learning between countries and were was also prominent in the discussions. In keen to participate in the communities of particular, the question was raised of how practice that are developing to facilitate to strengthen the focus on managing for country-country learning. The Roundtable results, bearing in mind the Paris Decla- included a meeting of a community of prac- ration provisions on results-based frame- tice in the Asia region (supported initially by works and mutual accountability. Donors the ADB, in which practitioners among 11 will be pursuing ways to support coun- Asian countries are networking with each try-to-country learning and the further other to share practices and experience), development of communities of practice, and the launching of a similar community as well as finding ways to follow up on of practice in the Africa region. the action planning process in individ- Progress and opportunities at the donor ual countries. The strongest conclusion level. Donors recognized that manag- to emerge from the donor discussions, ing for results should not be seen mainly however, was the urgent need to scale up as a set of measuring and monitoring resources to support stronger statistical tools, although statistics and monitor- systems at the country level, through find- ing and evaluation are essential compo- ing ways to support, financially and with nents. Donors as well as country partners technical know-how, the further develop- agreed that it was useful to unpack the ment of statistical capacity. This should notion of capacity to manage for results serve the need for monitoring sectoral along the five themes of the Roundtable performance as well as that of central agencies, while accelerating the progress Global Monitoring Report provides a vehicle made through the Marrakech Action Plan for communicating these results to the broad on Statistics. development community, but greater effort is needed in communicating and sharing the results of this exercise within each institution. The Common Performance Assessment Review and discussion by both management System and staff are critical to ensure that the findings The Common Performance Assessment Sys- permeate the institutions and do not simply tem (COMPAS) is an interagency effort to gather dust in institutional files. develop a common system that all MDBs The 2006 COMPAS identifies two new can use to monitor their results orienta- opportunities. First, the similarities between tion. COMPAS focuses on processes and the private sector windows of the four MDBs results within the control of the institutions and the EBRD may militate in favor of their themselves. Its intent is not to compare per- merging their efforts under a more coherent formance across different MDBs--such com- performance assessment reporting format. parisons are exceedingly difficult, given that Second, other multilateral organizations have the institutions themselves are so diverse-- expressed an interest in joining the COM- but rather to provide baseline data against PAS effort. In particular, the MDB Working which each institution can measure its own Group on Managing for Results will be dis- progress over time. cussing proposals from International Fund The three pillars of COMPAS--actions to for Agricultural Development and the Islamic support country capacity for managing for Development Bank (IsDB) in the COMPAS in development results, actions to improve the the spring of 2007. results orientation of internal systems, and actions to improve interagency cooperation Improving Harmonization and for results--are described in Global Monitor- Alignment: MDBs and the Paris ing Report 2006. In 2006 a new COMPAS Declaration report was prepared under the leadership of the IADB (chairmanship of the COMPAS All the MDBs (together with the OECD DAC group rotates among members).16 The report and the UNDP) cosponsored the 2005 High- examines the seven performance categories Level Forum, which adopted the Paris Declara- developed for the 2005 report, adjusted tion on Aid Effectiveness. The cosponsorship to reflect the feedback received on the first by the MDBs reflects their commitment to COMPAS report. Broadly speaking, the improving the effectiveness with which aid is changes give greater specificity to the indi- planned, delivered, and managed. cators used, reduce the room for discretion Country-level monitoring of the imple- in the provision of answers, and increase the mentation of the mutual commitments in objectivity and credibility of the COMPAS that declaration, including through 12 quan- itself.17 As a result of these changes, few com- titative indicators of actions, took place for parisons are possible between this year's and the first time in 2006 (table 5.1). Along last year's COMPAS, but this year's COM- with bilateral and other donors, the ADB, PAS should provide a sound basis for track- the African Development Bank (AfDB), the ing future progress (Box 5.7). IADB, and the World Bank participated The 2006 COMPAS report illustrates the in exercises to measure performance in 34 MDBs' commitment to self-assessment. It also countries in which partner country and indicates their willingness to disclose informa- donor performance had been surveyed. tion about the way they conduct business and Results of this 2006 monitoring round will the way they organize themselves to meet their serve as a baseline for reviewing progress in strategic development objectives. The annual 2008 and against the agreed 2010 targets Several important findings emerge from the 2006 COMPAS report. Support for country capacity to manage for development results is increasing. MDBs use vari- ous approaches to assess country capacity to manage for results. The ADB and the AfDB produce diagnoses for a growing number of countries as part of their country strategy formulation. The IADB carries out capacity diagnostics through a specific program (PRODEV) that covers 69 percent of countries in the region. MDBs also support the strengthening of capacity through a variety of means, including World Bank­funded broad public sector management projects. Eighteen of 29 EBRD member countries received technical assistance through its Legal Transition program. Country Strategies are being put in place, but implementation weaknesses remain. Guidelines for preparing country strategies require sound results frameworks--clearly defined monitoring indica- tors, with baseline data and targets to be reached at the end of the strategy implementation period. There is significant room for improving the results focus of country strategies against these criteria. Concessional resources are being allocated on the basis of performance. All of the MDBs (except the EBRD, which does not provide concessional financing) allocate concessional resources on the basis of performance, as reflected in policies, institutions, and portfolio performance, among other criteria. Allocation criteria also typically include a "needs" factor. Project performance could be improved. More than half of all projects reviewed received overall quality-at-entry ratings of satisfactory or better. However, there is significant room for improve- ment in terms of monitoring frameworks and implementation: 3­25 percent of projects suffered from unsatisfactory implementation progress, were unlikely to achieve their development objec- tives, or both. Moreover, implementation delays affected 34­69 percent of operations. Completion reports prepared as a percentage of number due ranged from 57 to 100 percent across the MDBs. Some 51­94 percent of reports indicated satisfactory or better use of outcome indicators. Devel- opment objectives were achieved in 61­78 percent of projects. EBRD disbursed 55 percent of its commitments annually; the disbursement ratios at other MDBs were just 20­30 percent. Application of institutional learning from operational experience is not sufficiently systematic. All of the MDBs have formal devices for drawing lessons from operational experience and dissemi- nating them to staff members and member countries. It is not clear how well the lessons are applied, however. Independent evaluation offices in all MDBs help promote the learning of lessons and accountability through evaluations of individual operations, sectors, themes, and country strategies and programs. On the whole their recommendations appear to influence the way MDBs conduct their business, but only the World Bank has a formal mechanism to keep track of and measure management's adoption of independent evaluation recommendations. Salary increases are related to results. All of the MDBs have programs in place to strengthen the results-related skills of their operational staff; in recent years they have provided training on such topics as results-oriented planning, budgeting and monitoring, and evaluation. Although specific approaches vary across institutions, all MDBs link salary increases to the accomplishment of agreed upon objectives. Sources: AfDB, ADB, EBRD, IADB, and World Bank. 2006 COMPAS: Multilateral Development Banks Common Performance Assessment System: Steering for Results. January 26, 2007. for collective action. Preliminary results Paris Declaration in the Asia and the Pacific of the survey were presented in the 2006 region. Asian Regional Forum on Aid Effectiveness For the Paris Declaration to achieve results held at ADB to discuss best practices, and at the country level, MDBs and other donors measures to enhance implementation of the will need to bolster the capacity of partner Indicators pertaining to MDB implementation of the Paris Declaration (preliminary data based on 2006 Round of Monitoring) Indicator MDBs Other donors 2010 Target 4: Strengthen capacity by coordinated support 47% 47% 50% 5a: Percentage of aid that is disbursed using 44% 36% Reduction of aid not using country public financial management systems country PFM systems by a third or more 5b: Percentage of aid that is disbursed using country procurement systems 40% 38% Target under development 6: Number of Parallel Implementation Units 444 1,323 Reduction by 2/3 (PIUs) 7: Percentage of aid that is disbursed on time 72% 62% Reduction by 50% of aid not disbursed on time 9: Percentage of aid that is disbursed through 52% 40% 66% program-based approaches 10a: Percentage of missions that is done jointly 21% 26% 40% with other donors 10b: Percentage of country analytic work 52% 55% 66% donors that is done jointly with other donors and/or partner government Source: Preliminary results from the 2006 Survey on Monitoring the Paris Declaration, OECD/DAC (March 1, 2007). Note: The data reflect implementation as of 2005 for donors that provided a grand total of more than $100 million for the government sector. The data are undergoing final review by the OECD/DAC. More reliable data will be presented in OECD/DAC reporting in April 2007. countries to lead the effort, take action in their all key means of encouraging collaboration own programs, and push for greater collective among financing institutions and promoting implementation throughout the donor com- the alignment of programs with country pri- munity. Substantial actions are being taken. orities, strategies, and systems. The MDBs are continuing to harmonize their Joint analytic work can lever a more har- procedures, to strengthen country systems, monized delivery of aid. Not only does joint and to align their own activities with these analytic work lay a cost effective basis for systems where possible. Country financial forging a common understanding of issues management systems are currently being used and providing more consistent advice on for 44 percent of all lending by MDBs. strategy, it also provides the basis for future One key commitment in the Paris Declara- collaboration and donors on projects and tion is to increase the proportion of aid deliv- programs, drawing on common performance ered through program-based approaches that assessment frameworks and conditionality. are closely aligned with a country's sector or Data from the monitoring survey show that subsector priorities and strategies, using coun- the MDBs now undertake 52 percent of their try systems and procedures to the extent pos- analytic work jointly, and further attention sible, harmonized among donors, with close will be needed to meet the target of 66 percent attention paid to capacity building (indicator 9 for this indicator by 2010. of the Paris Declaration). Such program-based Decreasing reliance on use of project approaches are typically delivered through implementation units (PIUs) that are parallel sectorwide approaches, development policy to government administrative structures and operations, and projects with joint financing-- institutions and in many cases undermine capacity building is a common challenge fac- streaming gender policies and strengthening ing all donors. The monitoring data indicate institutional arrangements to achieve gender- that MDB-supported programs account for related objectives. In 1998 the ADB adopted one quarter of all these parallel PIUs. To meet a policy on gender and development that the ambitious 2010 Paris target of a two- marked a shift from targeted interventions in thirds reduction of these units will require the social sectors to identification of gender a substantial change in how MDBs organize equity as a cross-cutting issue in all areas of for project management and implementation, operation. The AfDB adopted a gender policy and work more closely with integrated PIUs. in 2002. The IADB expanded the scope of its MDBs are finding new ways to help achieve Women in Development Policy (1987) to pur- the objectives of the Paris Declaration. Of sue a dual strategy of mainstreaming gender particular note has been the increasing use of equality in its lending portfolio and address- joint or collaborative country assistance strat- ing critical themes of women's empowerment. egies to harmonize country diagnostics, align Acting on a commitment made in Beijing to efforts with country priorities, and prepare address domestic violence, the IADB main- a coordinated portfolio of activities. Such streamed its initiative to reduce domestic vio- exercises have recently been completed in lence against women into a broader initiative Bangladesh and Cambodia (by the ADB and to enhance citizen security throughout Latin the World Bank), Nigeria (the World Bank), America and the Caribbean.18 The World and Uganda (the AfDB and the World Bank) Bank had already adopted a gender policy and are virtually complete in Tanzania (the in 1990, but its emphasis on mainstreaming AfDB and the World Bank). Similar work is increased markedly after Beijing. Its Board of under way or planned in Ghana, the Kyrgyz Executive Directors adopted a gender main- Republic, Malawi, Mozambique, Nicaragua, streaming strategy in 2001.19 Vietnam, and Zambia and is being discussed Progress toward mainstreaming gender in a number of other countries. policies has been modest but steady. The MDBs also have a role to play in helping share of gender-responsive loans at the ADB ensure the integration of vertical programs increased from 15 percent in 1998 to 38 per- into sector strategies by drawing them into cent in 2004.20 At the IADB, lending opera- strategy development and planning their own tions that mainstreamed equal opportunities programs to ensure complementarity. They for women and men represented 37 percent have begun to discuss the coordination of of the total investment of the loan portfolio governance and anticorruption actions, and between 2002 and 2005, up from just 24 planning is under way to better harmonize percent in 1998­2001.21 The World Bank legal documentation. increased its share of projects incorporating gender issues in the design stage from 68 per- cent in 2001 to 87 percent in 2004­05.22 The AfDB, the IADB, and the World Bank This section describes actions by the IFIs all recently adopted gender action plans to in the two areas highlighted by this year's make their gender mainstreaming policies more Global Monitoring Report: gender equality strategic and operationally effective. The ADB, and fragile states. which adopted a focus on gender as an impor- tant cross-cutting theme in the 1990s, has been effectively using project-level gender action Promoting Gender Equality plans for some years.23 It is also developing Following the 1995 United Nations World an institutionwide three-year plan of action.24 Conference on Women in Beijing, the IFIs The AfDB adopted a gender plan of action in realigned their commitment to gender equal- 2004 and included gender among 14 key indi- ity and women's advancement by main- cators of development effectiveness.25 The gen- der mainstreaming action plan at the IADB, help them assess the advantages of invest- adopted in 2003, has helped target activities, ing in gender equality; translating gender develop sounder actions, and monitor these objectives into specific actions that can actions more effectively.26 "Gender Equality as have a measurable impact on women's Smart Economics," the World Bank's gender lives; budgeting adequate financial and action plan crafted in 2006, focuses on previ- technical resources to implement projects ously neglected economic sectors. It highlights at scale and measure results; and align- key upstream and downstream actions, linking ing investments on gender equality with them to outcomes and indicators of success.27 needed policy and institutional changes. Despite these improvements, significant gaps remain, particularly in the areas of eco- Supporting Fragile States nomic growth, agriculture, competitiveness, infrastructure, and private sector develop- IFIs have been working closely together ment, where progress has been slower than toward strengthening their support to fragile in the health and education sectors. Greater states by harmonizing their approaches along attention has been paid to gender in project four main areas of specialized response: design than in implementation; very little strategy, policy, and procedural frameworks; has been paid in monitoring and evalua- exceptional financial instruments; custom- tion. Institutions have generally been slow to ized organizational and staffing approaches; develop and adopt measurable indicators of and partnership work. Progress varies among progress in gender equality. The IFIs' internal international organizations, but all are com- rating systems have primarily measured good mitted to more effective and rapid responses intentions (whether gender has been incorpo- to fragility (table 5.2).29 At their meeting in rated into project design) rather than results London in March 2007, the heads of MDBs or financial commitments to gender issues.28 agreed to set up a working group on fragile Regarding the latter, an inherent problem states to identify common operating prin- arises in assessing the amount of resources ciples for engagement in fragile situations, allocated to an objective that by definition enhance partnerships, and coordinate the is fully mainstreamed. Since budget tracking division of labor within the MDBs and other is an important tool for accountability, the partner agencies. more the IFIs mainstream gender, the harder Strategies. The ADB's strategy for engag- it is to hold them accountable. ing weakly performing countries is designed The IFIs should use their comparative to increase the effectiveness of existing and advantages to significantly scale up the planned operations in countries character- MDG3 agenda. Specifically, they could: ized by weak governance, ineffective public administration, and civil unrest. Its frame- Invest dedicated resources in including work for guiding operational planning and gender equality and women's empow- implementation includes a methodology for erment in results frameworks and the classifying such countries and alternative results agenda, and the associated impact interventions that may be modified depending evaluation work, to both strengthen gen- on the country context. The ADB emphasizes der equality interventions and increase country ownership, bolstered by systematic accountability for their own performance capacity development. Play a leadership role in strengthening the The AfDB identifies 25 countries in its monitoring of MDG3 at the international region as fragile. Of those, 16 have been level designated "core fragile states." The AfDB Assist client countries in significantly scal- is in the process of enhancing its assistance ing up MDG3 interventions by using ana- to these countries by strengthening its oper- lytical, policy, and research instruments to ational response and enhancing resource IFI reforms to strengthen response in fragile states IMF WB ADB AfDB IADB Source: Staff of the of the IMF, WB, ADB, AfDB, and IADB. Note: Yes = Comprehensive specialized framework adopted and implemented; Under way = Specialized framework under development or some specialized response implemented; No = No specialized framework implemented. mobilization capacity. The AfDB's envisaged The IMF is actively engaged in assisting strategy focuses on the following categories almost all fragile states, although it also does of engagement: (1) catalytic role; (2) strategic not define them formally for such purposes. partnership; and (3) areas of minimal engage- This engagement focuses in the IMF's core ment. Where the AfDB undertakes a catalytic macroeconomic and financial areas of respon- role, it proposes to engage in rebuilding state sibility. Assistance takes the form of policy capacity and accountability and in rehabili- advice on fiscal, monetary, exchange rate, and tating and reconstructing basic infrastruc- financial issues; help in identifying gaps in the ture. Where it builds strategic partnerships, related institutional and legal frameworks; the AfDB intends to support economic and and technical cooperation to follow up much structural reforms and economic integration of this advice--all essential elements of state- and regional projects. The AfDB will also building. In cooperation with the World Bank, step up its efforts in generating knowledge the IMF assists countries seeking to qualify with respect to fragile states and situations in for debt relief under the enhanced HIPC Ini- Africa. The proposed strategy also identifies tiative and the MDRI, and also seeks to help a need to streamline and simplify the AfDB's them avoid the reemergence of debt problems procedures in these states. afterward. While the IMF's direct financial Although it has not formally defined assistance is generally not a major element fragile states for separate strategic engage- of financing packages, for some countries ment, the Islamic Development Bank (IsDB) its lending--most often through postconflict has developed policy notes on assistance to emergency assistance or the Poverty Reduc- regional member countries experiencing tion and Growth Facility (PRGF)--can also be fragility.30 It also utilizes simple and flex- important. IMF staff are currently preparing ible procurement and disbursement proce- a report that reviews support to postconflict dures for its work in fragile states, in line countries and other fragile states, and exam- with procedures proposed for emergency ines the adequacy of existing instruments in response. Similarly, the IADB utilized spe- terms of policy flexibility and their capacity- cial measures for its engagement in Haiti, building component. which included simplified start-up require- Since the inception of the Low-Income ments, broader eligible expenditure cat- Countries under Stress (LICUS) initiative, the egories, and elimination of counterpart World Bank has underlined the importance financing requirements. of supporting efforts that contribute to peace- building and state-building goals, highlighting program of nonreimbursable technical assis- the need for institutional flexibility and close tance and nonfinancial products to underpin international coordination. The World Bank program and policy implementation and has developed specific guidance on assistance increase country knowledge. IDA can provide strategies and transitional results frameworks pre-arrears grants to postconflict countries if in fragile states; they distinguish among coun- certain conditions are met. Under IDA-14 tries that are facing deteriorating governance, it can also provide exceptional support to those in postconflict or political transition, countries that are re-engaging with IDA after those currently in conflict or crisis, and those a prolonged period of disengagement. transiting from fragility. In February 2007, the Organizational capacity. All of the IFIs World Bank's Board also approved a "New recognize the importance of increasing their Framework for Rapid Bank Response to Cri- field presence in fragile states, where low ses and Emergencies," which provides quicker capacity and volatile conditions require sus- and more effective responses to emergencies tained assistance on the ground and empow- and crises through accelerated and streamlined erment of staff in the field. Until recently, the review and implementation procedures; it gives AfDB had limited field presence in African the World Bank the flexibility to respond to a fragile states, and two-thirds of the World wider range of fragile situations and clarifies Bank's field offices in fragile states had no or the objective of its engagement to include ade- just one international staff member in 2005. quate focus on the social aspects of recovery Both institutions are taking steps to increase and peace-building. their field presence. Under its decentraliza- Financing instruments and allocation. tion strategy, which is currently being imple- Both the AfDB and the World Bank have devel- mented, the AfDB is strengthening its field oped an exceptional allocation framework for presence in fragile states by opening field postconflict countries. Like IDA's special post- offices in Chad, the Democratic Republic of conflict assistance, the African Development Congo, Sierra Leone, and Sudan. The IADB Fund's postconflict enhancement factor allows has posted additional staff to Haiti, aligned countries to benefit from additional resources with areas of portfolio focus, and delegated over and above their performance-based allo- additional responsibilities to its representa- cation for a limited period after they are des- tive in Port-au-Prince. The World Bank has ignated postconflict countries. IDA extended proposed an increase of at least 50 percent the duration of exceptional assistance under in its field positions in fragile states over the the postconflict framework in IDA-14 to cor- next two years. respond with the results of research on the Partnerships. The IFIs have worked with pattern of aid and absorptive capacity for one another and participated in wider initia- countries emerging from conflict.31 tives to develop international consensus on Many fragile states face difficulties from shared approaches and tools in fragile states. the build-up of large and protracted arrears The World Bank co-chairs the Fragile States on their debt. The AfDB has established the Group within the OECD DAC; this group Post-Conflict Country Facility (PCCF) to includes the ADB and the AfDB. There is a help countries emerging from conflict to clear general shift toward joint assistance strate- these arrears. The IADB can grant limited gies and cofinancing with other donors: the grant financing to conflict-affected countries World Bank has four joint country assistance with large overdue debt payments, before strategies in place in fragile states and two arrears clearance. Recognizing the need to others under way; the ADB and the AfDB maintain positive financial flows, the IADB emphasize cofinancing with other partners. has introduced innovations in Haiti; ongoing Future priorities. All the IFIs recognize IADB interventions combining investment the need to strengthen approaches to frag- and policy loans are complemented with a ile states by focusing on strategy, financing instruments, organization and staffing, and partnerships. Specific priorities going for- ward include the following: 1. IMF (2006). 2. World Bank (2007). Support efforts under the Paris Declaration 3. World Bank (2006). to implement the Principles for Good Inter- 4. International Task Force on Global Public national Engagement in Fragile States. Goods (2006). 5. World Bank (2006). Strengthen exchanges among the IFIs on 6. MDBs Report (2006). strategic assistance models, strengthening 7. "Report of the External Review Committee and harmonizing business policy and pro- on Bank-Fund Collaboration" (2007). cedures, financing instruments, and orga- 8. The doubling of EBRD lending in USD nization and staffing issues. terms is a combination of actual growth and Strengthen common approaches with other exchange rate movements. In terms of Euros, international partners, in particular through growth has been about 15 percent--from 4.3bn efforts to improve both coordination and to 4.9bn. division of labor with organizations leading 9. The full cost to IDA, the AfDF, and the IMF peace-building efforts, such as the United of the MDRI was estimated to be around $50 bil- Nations and regional institutions. lion in July 2006. IDA (2006). 10. The other six donors--Hungary, the Repub- Review the types of financial assistance lic of Korea, Kuwait, Saudi Arabia, South Africa, provided to different kinds of fragile states and Switzerland--are in the process of securing along with the effectiveness of resource use required approvals to issue their instruments of in these countries. commitment. 11. Firm financing commitments for the MDRI The IFIs have been supportive of strength- are backed by necessary legislative and parlia- ening coherence across the diplomatic, secu- mentary approvals in the donor country. Quali- rity, and development spheres as they engage fied financing commitments are not backed by in fragile states, as demonstrated by their sup- such approvals and are subject to a donor's future port of the United Nations Peace-Building capacity and willingness to provide funding. Commission. A number of other international 12. Dollar and Levin (2004). 13. These workshops and reviews of experi- actors, including the above-mentioned Fragile ence are summarized in the OECD's Sourcebook States Group, now have work under way to on Emerging Good Practices in Managing for consider how to better integrate approaches Development Results (2006). among diplomatic, security, financial, and 14. Bedi, Coudouel, Cox, Goldstein, and development actors in fragile states. The Thornton (2006). World Bank also coordinates with the UNDG 15. World Bank (2004). in making postconflict needs assessments: 16. AfDB, AsDB, EBRD, IADB, and World these are joint planning tools that cover Bank (2007). the political, security, social, and economic 17. A good example corresponds to the sub- spheres; they are currently undergoing a revi- category relating to the timely implementation of sion to strengthen their focus on peace build- projects. The 2006 COMPAS offers a single metric for the disbursement ratio and for actual, versus ing, institution building, and the monitoring planned, execution period. of implementation and results. The regional 18. Buvinic (2004). development banks participate in these joint 19. World Bank (2001). assessment and planning missions for coun- 20. ADB (2006). tries in their regions. 21. IADB (2006). 22. Gender and Development Group, World regularly updated. EBRD does not have programs Bank (2006). Whereas the IADB and the ADB use targeted to fragile states, but does have two special as a denominator or comparator the totality of proj- programs for their least advanced members (the ects in the loan portfolio, the World Bank excludes Early Transition Countries and the Western Bal- from the exercise roughly 20 percent of projects that kans countries). potentially have no gender-relevant dimensions. 30. Unlike that from other MDBs, the IsDB's 23. ADB (2006). assistance to fragile states includes a substantial 24. ADB (2006). element of humanitarian assistance. The primary 25. Response to informal questionnaire, Janu- focus is on emergency relief, followed by basic ary 2007. social and economic infrastructure and long-term 26. IADB (2006). rehabilitation and reconstruction. 27. World Bank (2006). 31. The ADB invests in technical assistance and 28. How much is spent on gender issues is dif- capacity building in fragile states, usually in the ficult to determine, especially as these issues are form of grants. The AfDB continues to provide increasingly mainstreamed. emergency relief assistance in the form of grants to 29. Although the Bank has approved a new affected countries, many of them fragile states. The policy framework and business procedures for its IsDB has a special assistance account, the Waqf response in fragile situations and it provides special Fund, to provide humanitarian relief to member financing to fragile countries through the LICUS countries and Muslim communities affected by Trust Fund, it continues to work on strengthen- natural disasters and calamities. The World Bank ing its support to this fragile group of countries. established the LICUS Trust Fund from a series of The IADB experience is based on the single frag- grants from the IBRD surplus, to support peace ile country in the region (Haiti), for which interim building and capacity building in fragile states, country assistance strategies were formulated and with a focus on countries in nonaccrual status. References DFID (U.K. Department for International Abu-Ghaida, Dina, and Stephan Klasen. Development). 2005. "Why We Need 2004. "The Costs of Missing the Mil- to Work More Effectively with Fragile lennium Development Goal on Gender States." DFID, London. Equality." World Development 32 (7): Dollar, David, and Roberta Gatti. 1999. 1075­107. "Gender Inequality, Income and Auty, Richard M., and Alan H. Gelb. 2001. Growth: Are Good Times Good for "Political Economy of Resource-Abun- Women?" Policy Research Report on dant States." In Resource Abundance and Gender and Development, Working Economic Development, ed. R. M. Auty. Paper Series No. 1, World Bank, Wash- New York: Oxford University Press. ington, DC. Barwell, Ian. 1996. "Transport and the Ezzati, Saleh, and Daniel M. Kammen. 2000. Village: Findings from African Village- "The Contributions of Emissions and Level Travel and Transport Surveys and Spatial Microenvironments to Exposure to Related Studies." World Bank Discussion Indoor Air Pollution from Biomass Com- Paper No. 344, Africa Region Series, bustion in Kenya." Environmental Health World Bank, Washington, DC. Perspectives 108 (9): 833­40. Caplan, B. 2001. "How Does War Shock ------. 2001. "Indoor Air Pollution from the Economy?" Journal of International Biomass Combustion and Acute Respira- Money and Finance 21. tory Infections in Kenya: An Exposure- Chauvet, Lisa, and Paul Collier. 2004. Response Study." The Lancet 358. "Development Effectiveness in Fragile Hallward-Driemeier, Mary. 2006. "The Stated: Spillovers and Turnarounds." Impact of Improvements in the Invest- Centre for the Study of African Econo- ment Climate: Evidence from Panel mies, Oxford University. Surveys in Bangladesh." World Bank, Chen, Shaohua, and Martin Ravallion. Washington, DC. Forthcoming. "Absolute Poverty Mea- Kaufmann, Daniel, Aart Kraay, and Mas- sures for the Developing World." simo Mastruzzi. 2006. "Governance Collier, Paul. 1999. "On the Economic Con- Matters V: Governance Indicators for sequences of Civil War." Oxford Eco- 1996­2005." World Bank, Washington, nomic Papers 51 (1). DC. Klasen, Stephan. 2002. "Low Schooling World Bank. 2005. "Low-Income Countries for Girls, Slower Growth for All? Cross- Under Stress: Update." Operations Policy country Evidence on the Effect of Gender and Country Services, World Bank, Inequality in Education on Economic Washington, DC. Development." World Bank Economic ------. 2006a. Where Is the Wealth of Review 16 (3): 345­73. Nations: Measuring Capital for the XXI Knight, M., N. Loaza, and D. Villaneuva. Century. Washington, DC: World Bank. 1996. "The Peace Dividend: Military ------. 2006b. World Development Indica- Spending Cuts and Economic Growth." tors. Washington, DC: World Bank. IMF Staff Papers (43) 1. International Monetary Fund, Washington, DC. Marshall, Monty, and Keith Jaggers. 2002. "POLITY IV Project: Political Bloom, Erik, Indu Bushan, David Clinging- Regime Characteristics and Transitions, smith, Rathavuth Hong, Elizabeth King, 1800­2002." Dataset Users' Manual, Michael Kremer, Benjamin Loevinsohn, Center for International Development and Brad Schwartz. 2006. "Contracting and Conflict Management, University of for Health: Evidence from Cambodia." Maryland. Informal paper. Moreno Torres, Magui, and Michael Ander- Buckland, Peter. 2005. Reshaping the son. 2004. "Fragile States: Defining Future: Education and Post-Conflict Difficult Environment for Poverty Reduc- Reconstruction. Washington, DC: World tion." PRDE Working Paper 1, Poverty Bank. Reduction in Difficult Environments. Bustreo, Flavia, Eleonora Genovese, Elio DFID, London. Omobono, Henrik Axelsson, and Ian Nankhuni, F. 2004. "Environmental Degra- Bannon. 2005. "Improving Child Health dation, Resource Scarcity and Children's in Post-Conflict Countries: Can the Welfare in Malawi: School Attendance, World Bank Contribute?" Children and School Progress, and Children's Health." Youth, HDNCY No. 3. World Bank, PhD thesis, Agricultural Economics and Washington, DC. Demography, Pennsylvania State Univer- Chaudhury, Nazmul, Jeffrey Hammer, sity. Michael Kremer, Karthik Muralidharan, OECD/DAC (Organisation for Economic and F. Halsey Rogers. 2006. "Missing Development and Co-operation/Develop- in Action: Teacher and Health Worker ment Co-operation Directorate). 2005. Absence in Developing Countries." Journal "Principles for Good International of Economic Perspectives 20 (1): 91­116. Engagement in Fragile States." OECD, Clemens, M. 2004. "The Long Walk to Paris. School: International Education Goals ------. 2006. "Monitoring Resource Flows in Historical Perspective." Center for to Fragile States--2005 Report." OECD, Global Development, Washington, DC. Paris. Cohen, Joel E., David E. Bloom, and Martin Staines, Nicholas. 2004. "Economic Per- B. Malin. 2006. Educating All Children: formance over the Conflict Cycle." IMF A Global Agenda. American Academy of Working Paper WP/04/95, International Arts and Sciences, Cambridge, MA. Monetary Fund, Washington, DC. Collier, P., S. Dercon, and J. Mackinnon. Stotsky, Janet. 2006. "Gender and Its Rel- 2003. "Density versus Quality in Health evance to Macroeconomic Policy: A Care Provision: Using Household Data Survey." IMF Working Paper WP/06/233, to Make Budgetary Choices in Ethiopia." International Monetary Fund, Washing- World Bank Economic Review 16 (3): ton, DC. 425­48. Cotlear, Daniel. 2006. A New Social Contract of Student Achievement." Informal paper, for Peru. Washington, DC: World Bank. World Bank, Washington, DC. Crouch, Louis, and Tazeen Fasih. 2004. Hanushek, Eric A., and D. D. Kimko. 2000. "Patterns in Educational Development: "Schooling, Labor-Force Quality, and Implications for Further Efficiency Analy- the Growth of Nations." American Eco- sis." Informal paper, World Bank, Wash- nomic Review 90 (5): 1184­208. ington, DC. Hanushek, Eric A., and Javier A. Luque. Danel, I. 1999. "Maternal Mortality Reduc- 2003. "Efficiency and Equity in Schools tion, Honduras, 1990­1997: A Case around the World." Economics of Educa- Study." World Bank, Washington, DC. tion Review 22 (August): 481­502. Das, Jishnu, and Paul Gertler. 2007. Hanushek, Eric A., and Ludger Woessmann. "Practice-Quality Variation in Five 2007."The Role of School Improvement Low-Income Countries: A Conceptual in Economic Development." NBER Work- Overview." Informal paper, World Bank, ing Paper No. 12832, National Bureau of Washington, DC. Economic Research, Cambridge, MA. Das, Jishnu, and Jeffrey Hammer. 2004. Hoeffler, A., and M. Reynal-Querol. 2003. "Strained Mercy: The Quality of Medical "Measuring the Cost of Conflict." Care in Delhi." Economic and Political Oxford University Press, Oxford. Weekly 39 (9): 951­65. Jones, G., R. Steketee, R. Black, Z. Bhutta, Duflo, Esther. 2001. "Schooling and Labor S. Morris, and the Bellagio Child Survival Market Consequences of School Con- Study Group. 2003. "How Many Child struction in Indonesia: Evidence from Deaths Can We Prevent This Year?" The an Unusual Policy Experiment." Ameri- Lancet 362 (9377): 65­71. can Economic Review 91 (September): Lavy, Victor, and Jean-Marc Germain. 1994. 795­813. "Quality and Cost in Health Care Choice Duflo, Esther, Pascaline Dupas, Michael in Developing Countries." Working Paper Kremer, and Samuel Simei. 2006. "Edu- No. 105, Living Standards Measurement cation and HIV/AIDS Prevention: Evi- Study, World Bank, Washington, DC. dence from a Randomized Evaluation in Leonard, Kenneth, Melkiory C. Masatu, Western Kenya." Informal paper, Har- and A. Vialou. 2005. "Getting Doctors vard University, Cambridge, MA. To Do Their Best: Ability, Altruism and Dupas, P. 2005. "Relative Risks and Incentives." University of Maryland, Col- the Market for Sex: Teenagers, Sugar lege Park, MD. Daddies and HIV in Kenya." Informal Lewis, Maureen, and Marlaine Lockheed. paper, Dartmouth College, Hanover, NH. 2006. Inexcusable Absence. Washington, Fewtrell, L., R.B. Kaufmann, D. Kay, W. DC: The Center for Global Development. Enanoria, L. Haller, and J. M. Colford, Loevinsohn, B., and A. Harding. 2005. Jr. 2005. "Water, Sanitation, and Hygiene "Buying results? Contracting for Health Interventions to Reduce Diarrhea in Services Delivery in Developing Coun- Less Developed Countries: A Systematic tries." The Lancet 366 (9486): 676­81. Review and Meta-analysis." The Lancet Manor, James, ed. 2007. Aid That Works: Infectious Diseases 5 (1): 42­52. Successful Development in Fragile States. Filmer, Deon, Amer Hasan, and Lant Pritch- Washington, DC: World Bank. ett. 2006. "A Millennium Learning Goal: Mills, E.J., J.B. Nachega, I. Buchan, J. Measuring Real Progress in Education." Orbinski, and A. Attaran. 2006. "Adher- Center for Global Development, Wash- ence to Antiretroviral Therapy in Sub- ington, DC. Saharan Africa and North America: A Greaney, V., and Robert Prouty. 2007. Meta-analysis." Journal of the American "National and International Assessments Medical Association 296 (6): 679­90. Mills, Samuel, Eduard Bos, Elizabeth Lule, Middle-Income Countries in 2005." G.N.V. Ramana, and Rodolfo Bulatao. Funded by United Nations Programme 2007. "Preventable Maternal Deaths: on HIV/AIDS, U.S. Agency for Interna- Emergency Obstetric Care Is Lacking or tional Development, and United National Late in Ghana, Kenya, and India." Infor- Population Fund. Futures Group/POL- mal paper, World Bank, Washington, DC. ICY Project, Washington, DC. Mizala, A., and P. Romaguera. 2002. Stringer, J.S., I. Zulu, J. Levy J, et al. 2006. "Equity and Educational Performance," "Rapid Scale-up of Antiretroviral Ther- Economía, Journal of the Latin America apy at Primary Care Sites in Zambia: and the Caribbean Economic Association Feasibility and Early Outcomes." Journal 2 (2): 219­62. of the American Medical Association Murthy, K.J.R., T.R. Frieden, A. Yazdani, 296: 782­93. and P. Hreshikesh. 2001. "Public Private Thaddeus, S., and D. Maine. 1994. "Too Far Partnership in Tuberculosis Control: to Walk: Maternal Mortality in Context." Experience in Hyderabad, India." Inter- Social Science and Medicine 38: 1091­10. national Journal of Tuberculosis Lung Thirumurthy, H., J. Graff-Zivin, and M. Disease 5 (4): 354­59. Goldstein. 2005. "The Economic Impact Nickell, Stephen. 2004. "Poverty and Work- of AIDS Treatment: Labor Supply in lessness in Britain." Economic Journal Western Kenya." NBER Working Paper 114 (March): C1­C25. No. 11871, National Bureau of Eco- Pratham. 2005. "Annual Status of Educa- nomic Research, Cambridge, MA. tion Report (ASER)." Mumbai, India. Thornton, Rebecca. 2005. "The Demand Available at http://www.pratham.org/ for and Impact of Learning HIV Status: aserrep.php. Evidence from a Field Experiment." Pritchett, Lant. 2004. "Access to Educa- Informal paper, Harvard University, tion." In Global Crises, Global Solutions, Cambridge, MA. ed. Björn Lomborg, 175­234. Cambridge Umansky, Ilana, and Luis Crouch. 2006. U.K.: Cambridge University Press. Fast Track Initiative: Initial Evidence of Ransom, E., and N. Yinger. 2002. "Making Impact. Washington, DC: World Bank. Motherhood Safer: Overcoming Obsta- UNAIDS (United Nations Programme on cles on the Pathways to Care." Popula- HIV/AIDS). 2006. "Report on the Global tion Reference Bureau, Washington, DC. AIDS Epidemic." UNAIDS, Washington, Republic of Rwanda. 2006. "Scaling Up to DC. Achieve the Health MDGs in Rwanda." UNAIDS (United Nations Programme on A Background Study for the High-Level HIV/AIDS) and WHO (World Health Forum Meeting in Tunis, June 12­13, Organization). 2005. "AIDS Epidemic 2006. Update. Special Report on HIV Preven- Stover, John, Stefano Bertozzi, Juan-Pablo tion." UNAIDS, Washington, DC. Gutierrez, Neff Walker, Karen A. Stanecki, UNESCO (United Nations Educational, Sci- Robert Greener, Eleanor Gouws, Cath- entific and Cultural Organization). 2005. erine Hankins, Geoff P. Garnett, Joshua A. EFA Global Monitoring Report 2005-- Salomon, J. Ties Boerma, Paul De Lay, and The Quality Imperative. Paris: UNESCO. Peter D. Ghys. 2006. "The Global Impact ------. 2007. EFA Global Monitoring of Scaling Up HIV/AIDS Prevention Pro- Report 2007--Early Childhood Care and grams in Low- and Middle-Income Coun- Education. Paris: UNESCO. tries." Science 311 (5766): 1474­76. Vegas, Emiliana, and Jenny Petrow. 2007. Stover, John, and M. Fahnenstock. 2006. Raising Student Learning in Latin Amer- "Coverage of Selected Services for HIV/ ica: The Challenge for the 21st Century. AIDS Prevention and Care in Low- and Washington, DC: World Bank. Walker, Dilys, Juan Pablo Gutierrez, Pilar Transfers and Demand for Schooling in Torres, and Stefano M. Bertozzi. 2006. Malaysia, 1976­89." Journal of Develop- "HIV Prevention in Mexican Schools: ment Studies 39 (3): 1­28. Prospective Randomised Evaluation of Asfaw, Abay, and Assefa Admassie. 2004. Intervention." British Medical Journal "The Role of Education on the Adop- 332 (May 20): 1189­194. tion of Chemical Fertiliser under Dif- Walter, Steven. 2007. "Preliminary Report ferent Socioeconomic Environments in of Some Findings on Primary Education Ethiopia." Agricultural Economics 30: in Boyo Division." Informal paper, World 215­28. Bank, Washington, DC. Ballington, Judith, and Azza Karam, eds. WHO (World Health Organization). 2006. 2005. Women in Parliament: Beyond "Stop TB Partnership: The Global Plan Numbers. Stockholm: International Insti- to Stop TB, 2006­2015." WHO, Geneva, tute for Democracy and Electoral Assis- Switzerland. tance. Woessmann, Ludger. 2003. "Schooling Bandiera, Oriana, and Imran Rasul. 2006. Resources, Educational Institutions, and "Social Networks and Technology Adop- Student Performance: The International tion in Northern Mozambique." Eco- Evidence." Oxford Bulletin of Econom- nomic Journal 116: 862­902. ics and Statistics 65 (2): 117­70. Barwell, Ian. 1996. "Transport and the World Bank. 1999. "Safe Motherhood and Village: Findings from African Village- the World Bank: Lessons from 10 Years Level Travel and Transport Surveys and of Experience." World Bank, Washing- Related Studies." World Bank Discussion ton, DC. Paper No. 344, Africa Region Series, ------. 2006. "Progress Report for the World Bank, Washington, DC. Education for All­Fast Track Initiative." Baydas, Mayada M., Richard L. Meyer, and World Bank, Development Committee Nelson Aguilera-Alfred. 1994. "Discrimi- Meeting, Washington, DC. nation Against Women in Formal Credit ------. 2007. "Aid Architecture and the Markets: Reality or Rhetoric?" World Main Trends in Official Development Development 22 (7): 1073­082. Assistance Flows." World Bank, Wash- Behrman, Jere, and Mark Rosenzweig. 2004. ington, DC. "Returns to Birthweight." Review of Eco- nomics and Statistics 86 (2): 586­601. Brown, Philip H. 2006. "Parental Educa- tion and Investment in Children's Human Abu-Ghaida, Dina, and Stephan Klasen. 2004. Capital in Rural China." Economic "The Costs of Missing the Millennium Development and Cultural Change 54 Development Goal on Gender Equity." (4): 759­89. World Development 32 (7): 1075­107. Burkhalter, H.J. 2002. The Violent Trans- Agarwal, Bina. 1994. "Gender and Com- mission of HIV/AIDS. Washington, DC: mand over Property: A Critical Gap in CSIS HIV/AIDS Task Force. Economic Analysis and Policy in South Buvinic, Mayra, and Marguerite Berger. Asia." World Development 22 (10): 1990. "Sex Differences in Access to a 1455­478. Small Enterprise Development Fund Alderman, H., J. Hoddinott, and B. Kin- in Peru." World Development 18 (5): sey. 2002. "Long Term Consequences of 695­705. Early Childhood Malnutrition." Oxford Caldwell, J.C. 1979. "Education as a Factor Economic Papers 58 (3): 450­74. in Mortality Decline: An Examination of Anderson, Kathryn, Elizabeth M. King, Nigerian Data." Population Studies 33 and Yan Wang. 2003. "Market Returns, (3): 395­413. Carr, Marilyn, and Martha Chen. 2004. Countries: Methodological Issues and "Globalization, Social Exclusion and Evidence." IFPRI FCND Discussion Gender." International Labour Review Paper No. 90, International Food Policy 143 (1­2): 25­26. Research Institute, Washington, DC. Cebu Study Team. 1991. "Underlying and Doss, Cheryl. 1996. "Women's Bargaining Proximate Determinants of Child Health: Power in Household Economic Deci- The CEBU Longitudinal Health and sions: Evidence from Ghana." Staff Nutrition Study." American Journal of Paper Series P96-11. University of Min- Epidemiology 133 (2): 185­201. nesota, College of Agricultural, Food Chirwa, Ephraim. 2005. "Fertilizer and and Environmental Sciences, Depart- Hybrid Seeds Adoption among Small- ment of Applied Economics. holder Maize Farmers in Southern Doss, Cheryl. 2005. "The Effects of Intra- Malawi." Development Southern Africa household Property Ownership on 22 (1): 1­12. Expenditure Patterns in Ghana." Journal Chung, Woojin, and Monica Das Gupta. of African Economies 15: 149­80. 2007. "Why Is Son Preference Declining Doss, Cheryl, and Michael L. Morris. 2001. in Korea?" Paper prepared for the Popu- "How Does Gender Affect the Adoption lation Association of America meeting, of Agricultural Technologies? The Case of New York, March 29­31. Improved Maize Technology in Ghana." Croppenstedt, Andre, Mulat Demeke, and Agricultural Economics 25: 27­39. Meloria M. Meschi. 2003. "Technology Duflo, Esther. 2003. "Grandmothers and Adoption in the Presence of Constraints: Granddaughters: Old Age Pension and The Case of Fertilizer Demand in Ethio- Intra-Household Allocation in South pia." Review of Development Economics Africa." World Bank Economic Review 7: 58­70. 17 (1): 1­25. Das Gupta, Monica, Jiang Zhenghua, Xie Duflo, Esther, and Christopher Udry. 2004. Zhenming, Li Bohua, Woojin Chung, "Intrahousehold Resource Allocation in and Bae Hwa-Ok. 2003. "Why Is Son Côte d'Ivoire: Social Norms, Separate Preference so Persistent in East and South Accounts and Consumption Choices." Asia?" Journal of Development Studies NBER Working Paper No. 10498, 40 (2): 153­87. National Bureau of Economic Research, Deere, Carmen Diana, and Magdalena Cambridge, MA. Leon. 2001. Empowering Women: Land Duryea, Suzanne, Alejandra Cox Edwards, and Property Rights in Latin America. and Manuelita Ureta. 2004. "Women in Pittsburgh: University of Pittsburgh Press. the Latin America Labor Market: The Deere, Carmen Diana, and Magdalena Remarkable 1990s." In Women at Work: Leon. 2003. "The Gender Asset Gap: Challenges for Latin America, ed. Clau- Land in Latin America." World Develop- dia Piras, 27­60. Washington, DC: Inter- ment 31: 925­47. American Development Bank. De Walque, Damien. 2006. "Discordant Duryea, Suzanne, Sebastian Galiani, Hugo Couples: HIV Infection Among Couples Nopo, and Claudia Piras. 2006. "Educa- in Burkina Faso, Cameroon, Ghana, tion Gender Gap in Latin America and Kenya, and Tanzania." Policy Research the Caribbean." Research Department Working Paper No. 3956, World Bank, Working Paper, Inter-American Develop- Washington, DC. ment Bank, Washington, DC. Diagne, Aliou, Manfred Zeller, and Mano- Duryea, Suzanne, and Maria Eugenia har Sharma. 2000. "Empirical Measure- Genoni. 2004. "Ethnicity, Race and Gen- ments of Households' Access to Credit der in Latin American Labor Markets." and Credit Constraints in Developing In Social Inclusion and Economic Devel- opment in Latin America, ed. Mayra Kabeer, Naila, and Simeen Mahmud. 2004. Buvinic and Jacqueline Mazza, with "Rags, Riches and Women Workers: Ruthanne Deutsch, 247­60. Washington, Export-oriented Garment Manufacturing DC: Inter-American Development Bank. in Bangladesh." In Chains of Fortune: Galloway, Rae, and Mary Ann Anderson. Linking Women Producers and Workers 1994. "Prepregnancy Nutritional Status with Global Markets, ed. Marilyn Carr, and Its Impact on Birthweight." SCN 133­62. London: Commonwealth Secre- News 11: 6­10. tariat. Germain, Adrienne. 2004. "Reproductive Kishor, Sunita, and Kiersten Johnson. 2004. Health and Human Rights." The Lancet "Profiling Domestic Violence: A Multi- 363 (9402): 65­66. country Study." ORC Macro, Calverton, Glewwe, Paul, Hanan G. Jacoby, and Eliza- MD. beth M. King. 2001. "Early Childhood Klasen, Stephan. 2002. "Low Schooling Nutrition and Academic Achievement: A for Girls, Slower Growth for All? Cross- Longitudinal Analysis." Journal of Public country Evidence on the Effect of Gender Economics 81 (3): 345­68. Inequality in Education on Economic Goodkind, Daniel. 1996. "On Substituting Development." World Bank Economic Sex Preference Strategies in East Asia: Review 16: 345­73. Does Prenatal Sex Selection Reduce Post- ------. 2006. Guest editor's introduction, natal Discrimination?" Population and Special Issue on Revisiting the Gender- Development Review 22 (1): 111­25. related Development Index (GDI) and Grown, Karen. 2006. "Indicators and Indi- Gender Empowerment Measure (GEM). ces of Gender Inequality: What Do They Journal of Human Development 7 (2): Measure and What Do They Miss?" 145­59. Background paper for Global Monitoring Knowles, Stephen, A. K. Lorgelly, and Report 2007. World Bank, Washington, Dorian Owen. 2002. "Are Educational DC. Gender Gaps a Brake on Economic Hoddinott, John, and Lawrence Haddad. Development? Some Cross-country 1995. "Does Female Income Share Influ- Empirical Evidence." Oxford Economic ence Household Expenditures? Evidence Papers 54 (1): 118­49. from Côte d'Ivoire." Oxford Bulletin of Lastarria-Cornheil, Susana. 1997. "Impact Economics and Statistics 57 (1): 77­96. of Privatization on Gender and Property INSTRAW (United Nations International Rights in Africa." World Development Research and Training Institute for the 25 (8): 1317­333. Advancement of Women). 2005. "Institu- Lewis, Maureen, and Marlaine Lockheed. tional Mechanisms for the Advancement 2006. Inexcusable Absence: Why 60 Mil- of Women: New Challenges." INSTRAW lion Girls Still Aren't in School and What progress report. Available at www.un- to Do About It. Washington, DC: Center instraw.org/en/images/stories/Beijing/ for Global Development. institutionalmechanisms.pdf (accessed Lorgelly, Paula. 2000. "Are There Gen- January 2007). der-Separate Human Capital Effects on Jacobs, Susie. 2002. "Land Reform: Still a Growth? A Review of the Recent Empiri- Goal Worth Pursuing for Rural Women?" cal Evidence." University of Nottingham Journal of International Development 14: GREIT Research Paper 00/12, University 887­98. of Nottingham. Kabeer, Naila. 1999. "Resources, Agency, Malhotra, A., S. Schuler, and C. Boender. Achievements: Reflections on the Mea- 2002. "Measuring Women's Empow- surement of Women's Empowerment." erment as a Variable in International Development and Change 30 (3): 435­64. Development." World Bank Gender and Development Group Background paper. Credit Market Outcomes in Zimbabwe." World Bank, Washington, DC. Available Economic Development and Cultural at www.icrw.org/docs/MeasuringEmpow- Change 47: 585­604. erment_workingpaper_802.doc. Rubalcava, Luis, Graciela Teruel, and Dun- Morrison, Andrew, and Francesca Lamana. can Thomas. 2004. "Spending, Saving 2006. "Gender Issues in the Kyrgyz and Public Transfers Paid to Women." Labor Market." Background paper for Working Paper 024-04, California Center Kyrgyz Poverty Assessment, World Bank, for Population Research, University of Washington, DC. California, Los Angeles. Morrison, Andrew, Dhushyanth Raju, and Schultz, T.P. 1997. "The Demand for Children Nistha Sinha. 2007. "Gender Equality, Pov- in Low Income Countries." In Handbook erty, and Economic Growth." Background of Population and Family Economics, ed. paper for Global Monitoring Report M. R. Rosenzweig and O. Stark, 349­430. 2007. World Bank, Washington, DC. Amsterdam: North-Holland. Narayan, Deepa. 2006. Measuring Empow- ------. 2002. "Why Governments Should erment: Cross-disciplinary Perspectives. Invest More to Educate Girls." World Washington, DC: World Bank. Development 30 (2): 207­25. Paolisso, Michael J., Kelly Hallman, Law- Seguino, Stephanie, and Marla Sagrario rence Haddad, and Sibesh Regmi. 2002. Floro. 2003. "Does Gender Have Any "Does Cash Crop Adoption Detract from Effect on Aggregate Savings? An Empiri- Child Care Provision? Evidence from cal Analysis." International Review of Rural Nepal." Economic Development Applied Economics 17 (2): 147­66. and Cultural Change 50: 313­37. Semu, Linda. 2003. "Kamuzu's Mbumba: Pham, T. Hung, and Barry Reilly. 2006. Malawi Women's Embeddedness to Cul- "The Gender Pay Gap in Vietnam, 1993­ ture in the Face of International Political 2002: A Quantile Regression Approach." Pressure and Internal Legal Change." PRUS Working Paper Number 34, Pov- Africa Today 49 (2): 76­99. erty Research Unit at Sussex, University Storey, D. J. 2004. "Racial and Gender Dis- of Sussex. crimination in the Micro Firms Credit Phavi, Ing Kantha, and Cheryl Urashima. Market? Evidence from Trinidad and 2006. "Policies to Promote Women's Tobago." Small Business Economics 23 Economic Opportunities in Cambodia." (5): 401­22. Presented at High Level Consultation on Thomas, Duncan, and John Strauss. 1992. MDG3, "Gender Equality and Women's "Prices, Infrastructure, Household Char- Empowerment," World Bank, Washing- acteristics and Child Height." Journal of ton, DC, February 16. Development Economics 39 (2): 301­31. Pitt, Mark M., and Shahidur R. Khandker. Thomas, Duncan, John Strauss, and 1998. "The Impact of Group-Based Maria-Helena Henriques. 1991. "How Credit Programs on Poor Households Does Mother's Education Affect Child in Bangladesh: Does the Gender of Par- Height?" Journal of Human Resources ticipants Matter?" Journal of Political 26 (2): 183­211. Economy 106: 958­96. Udry, Christopher. 1996. "Gender, Agricul- Quisumbing, Agnes R., Jonna P. Estudillo, tural Production, and the Theory of the and Keijiro Otsuka. 2004. Land and Household." Journal of Political Econ- Schooling: Transferring Wealth Across omy 104: 1010­046. Generations. Baltimore, MD: John Hop- United Nations. 2003. The UN Handbook kins University Press. on Indicators for Monitoring Millennium Ratusi, Mayank, and Anand V. Swamy. Development Goals. New York: United 1999. "Explaining Ethnic Differentials in Nations. UNDAW (United Nations Division for the can Economies Working Paper 2000-5, Advancement of Women). 2004. "The Oxford University. Role of National Mechanisms in Promot- World Bank. 2001. World Bank Policy ing Gender Equality and the Empower- Research Report 2001: Engendering ment of Women." Report of the Expert Development: Through Gender Equal- Group Meeting. Rome, Italy, November ity in Rights, Resources, and Voice. New 29­December 2, 2004. York: Oxford University Press. UNESCO (United Nations Educational, Sci- ------. 2002. Gender in Transition. Wash- entific and Cultural Organization). 2004. ington, DC: World Bank. EFA Global Monitoring Report 2005. ------. 2003. "Country Gender Assessment Education for All: The Quality Impera- for Turkey." World Bank, Washington, tive. Paris: UNESCO. DC. UNESCO (United Nations Educational, ------. 2005. World Development Report Scientific and Cultural Organization) 2006. Washington, DC: World Bank. Institute for Statistics. 2005. Children out ------. 2006. World Development Report of School: Measuring Exclusion from Pri- 2007. Washington, DC: World Bank. mary Education. Montreal: UNESCO. Xinhua. 2005. "Imbalanced Sex Ratio in UNIFEM (United Nations Development China Not Owed to Family Planning." Fund for Women). 2005. Progress of China Daily. Available at http://www.chi- the World's Women 2005. New York: nadaily.com.cn/english/doc/2005-08/24/ UNIFEM. content_471890.htm (accessed January UN Millennium Project. 2005a. Taking 2007). Action: Achieving Gender Equality and Yuan, Xin, and Edward Jow-Ching Tu. Empowering Women. New York: Task 2005. "High Sex Ratio at Birth and its Force on Education and Gender Equality. Implications in China." Paper presented ------. 2005b. Toward Universal Primary at the IUSSP International Population Education: Investments, Incentives, and Conference, Tours, France, July 18­23. Institutions. New York: Task Force on Zabin, Laurie Schwab, and Karungari Education and Gender Equality. Kiragu. 1998. "The Health Consequences Van der Straten, Ariane, Rachel King, Olga of Adolescent Sexual and Fertility Behav- Grinstead, Eric Vittinghoff, Antoine Seru- ior in Sub-Saharan Africa." Studies in filira, and Susan Allen. 1998. "Sexual Family Planning 29 (2): 210­32. Coercion, Physical Violence, and HIV Zeng Yi, Tu Ping, Gu Baochang, Xu Yi, Li Infection Among Women in Steady Rela- Bohua, and Li Yongping. 1993. "Causes tionships in Kigali, Rwanda." AIDS and and Implications of the Recent Increase Behavior 2 (1): 61­73. in the Reported Sex Ratio at Birth in Webb, P., and S. Block. 2004. "Nutrition China." Population and Development Information and Formal Schooling as Review 19 (2): 283­302. Inputs to Child Nutrition." Economic Development and Cultural Change 52 (4): 801­20. WHO (World Health Organization). 2005. Alesina, Alberto, and David Dollar. 2000. WHO Multi-country Study on Women's "Who Gives Aid to Whom and Why? Health and Domestic Violence Against Journal of Economic Growth 5 (1): 33­64. Women. Geneva, Switzerland: WHO. Amprou, Jacky, Patrick Guillaumont, and Wier, Sharada, and John Knight. 2000. Sylviane Guillaumont Jeanneney. 2006. "Adoption and Diffusion of Agricultural "Aid Selectivity According to Augmented Innovations in Ethiopia: The Role of Criteria." Working Paper No. 9. Agence Education." Centre for the Study of Afri- Française de Dévelopment, Paris. Arslanalp, Serkan, and Peter Blair Henry. Dollar, David, and Victoria Levin. 2004. 2003. "The World's Poorest Countries: "The Increasing Selectivity of Foreign Debt Relief or Aid?" Stanford Research Aid." Policy Research Working Paper Paper Series 1809, Stanford University. 3299, World Bank, Washington, DC. Berthelemy, J.C., and A. Tichit. 2004. "Bilat- EU (European Union). 2006. "Global eral Donors' Aid Allocation Decisions. A Europe: Competing in the World. A Con- Three Dimensional Panel Analysis." tribution to the EU's Growth and Jobs Bourguignon, François, Mark Sundberg, and Strategy." Communication of the EU Hans Lofgren. 2005. "Building Absorptive Commission, October. Capacity to Meet the MDGs." Manu- Garrett, Laurie. 2007. "The Challenge of script, presented at the Seminar on Foreign Global Health." Foreign Affairs Jan./Feb. Aid and Macroeconomic Management, Gelb, Alan, and Ben Eifert. 2005. "Improv- Maputo, Mozambique, March 14­15. ing the Dynamics of Aid." Policy Broadman, Harry G. 2006. Africa's Silk Research Working Paper No. 3732, Road: China and India's New Economic World Bank, Washington, DC. Frontier. Washington, DC: World Bank. Grown, Caren, Chadrika Bahadur, Jes- Bulir, Ales, and A. Javier Hamann. 2003. sie Handbury, and Diane Elson. 2006. "Aid Volatility: An Empirical Assessment." The Financial Requirements of Achiev- IMF Working Paper 01/50, International ing Gender Equality and Women's Monetary Fund, Washington, DC. Empowerment. Levy Economics Institute ------. 2006. "Volatility of Development Working Paper No. 467, Bard College, Aid: From the Frying Pan onto the Fire?" Blithewood, NY. IMF Working Paper 06/65, International Heller, Peter S., Menachem Katz, Xavier Monetary Fund, Washington, DC. Debrun, Theo Thomas, Taline Koranche- Celasun, Oya, and Jan Walliser. 2005. "Pre- lian, and Isabelle Adenauer. 2006. "Mak- dictability of Budget Aid: Experiences ing Fiscal Space Happen: Managing Fiscal in Eight Africa Countries." Paper pre- Policy in a World of Scaled-Up Aid." IMF sented at World Bank practitioners forum Working Paper 06/270, International on budget support, Cape Town, South Monetary Fund, Washington, DC. Africa, May 5­6. IMF (International Monetary Fund) and Clements, Benedict, Rina Bhattacharya, and World Bank. 2006a. "Applying the Debt Toan Quoc Nguyen. 2003. "External Sustainability Framework for Low- Debt, Public Investment, and Growth in Income Countries Post Debt Relief." IMF Low-Income Countries." IMF Working and World Bank, Washington, DC. Paper 03/249, International Monetary ------. 2006b. "Heavily Indebted Poor Fund, Washington, DC. Countries (HIPC)-Status of Implementa- de Ferranti, David. 2006. "Innovative tion." IMF and World Bank, Washington, Financing Options and the Fight Against DC. Global Poverty: What's New and What's ------. 2006c. "Doha Development Agenda Next." In Transforming the Development and Aid for Trade." IMF and World Landscape, ed. Lael Brainard, Washing- Bank, Washington, DC. ton, DC: Brookings Institution Press. Kee, Hiau Looi, Alessandro Nicita, and Depetris Chauvin, Nicolas, and Aart Kraay. Marcelo Olarreaga. 2006. "Estimating 2005. "What Has 100 Billion Dollars Trade Restrictiveness Indices." Available Worth of Debt Relief Done for Low- at http://econ.worldbank.org/WBSITE/ Income Countries." Manuscript. EXTERNAL/EXTDEC/EXTRESEARCH/ ------. 2006. "Who Gets Debt Relief?" 0,,contentMDK:21085342~pagePK: Policy Research Working Paper No. 64214825~piPK:64214943~theSitePK:46 4000, World Bank, Washington, DC. 9382,00.html. Lele, Uma, Nafis Sadik, and Adele Simmons. Oxfam. 2004. "Stitched Up: How Rich- 2006. "The Changing Aid Architecture: Country Protectionism in Textiles and Can Global Initiatives Eradicate Pov- Clothing Trade Prevents Poverty Alle- erty?" Draft. viation." Briefing Paper 60, Oxford, Levin, Victoria, and David Dollar. 2005. London. "The Forgotten States: Aid Volumes and Reisen, Helmut, Marcelo Soto, and Thomas Volatility in Difficult `Partnership Coun- Weithoner. 2004. "Financing Global and tries (1992­2002)." Draft. Regional Public Goods Through ODA: McGillivray, Mark. 1989. "The Allocation Analysis and Evidence from the OECD of Aid Among Developing Countries: A Creditor Reporting System." OECD Multi-Donor Analysis Using a Per Capita Development Centre Working Paper 232, Aid Index." World Development 17 (4): OECD, Paris. 561­68. Renzio, Paolo, Vera Wilhelm, and Tim Wil- Moss, Todd, and Sarah Rose. 2006. "China's liamson. 2006. "Minding the Gaps: Inte- Export-Import Bank and Africa: New grating Reporting on Poverty Reduction Lending, New Challenges." Center for Strategy and Budget Implementation for Global Development, Washington, DC. Domestic Accountability." Draft. OECD (Organisation for Economic Devel- Strategic Partnership with Africa. 2006. opment and Co-operation). 2000. "Strategic Partnership with Africa: Sur- "Handbook for Reporting Debt Reor- vey of Budget Support, 2006." Draft. ganization on the DAC Questionnaire." Report by the Budget Support Working OECD, Paris. Group, Strategic Partnership with Africa, ------. 2005a. "Directives for Reporting to Washington, DC. the Creditor Reporting System." OECD, Thomas, Alun. 2006. "Do Debt-Service Paris. Savings and Grants Boost Social Expendi- ------. 2005b. Aid Activities in Support tures?" IMF Working Paper 06/180, Inter- of Gender Equality: 1999­2003. Paris: national Monetary Fund, Washington, DC. OECD. World Bank. 2005a. "A Strategic Frame- ------. 2006a. "Donor Reporting on Dis- work for the World Bank's Global Pro- semination of the Paris Declaration." grams and Partnerships." World Bank, DAC Meeting, Development Co-operation Washington, DC. Directorate, Paris, November 15. ------. 2005b. Global Economic Prospects ------. 2006b. "Fragile States: Policy Com- 2005. Washington, DC: World Bank. mitment and Principles for Good Interna- ------. 2006a. Global Development Finance tional Engagement in Fragile States and 2006. Washington, DC: World Bank. Situations." Development Co-operation ------. 2006b. "Harmonization and Align- Directorate, Paris. ment for Greater Aid Effectiveness: An ------. 2006c. "Monitoring Resource Flows Update on Global Implementation and to Fragile States 2005 Report." Develop- the Bank's Commitments." World Bank, ment Co-operation Directorate, Paris. Washington, DC. ------. 2006d. "Whole of Government ------. 2006c. "Integrating Global Part- Approaches to Fragile States." OECD, nership Programs with Country-Led Paris. National Programs: Synthesis of findings ------. 2007a. "Development Co-operation and Recommendations." World Bank, Report 2006." OECD, Paris. Washington, DC. ------. 2007b. "Progress Report on the ------. 2007a. "Aid Architecture and the 2006 Survey on Monitoring the Paris Main Trends in Official Development Declaration." Development Co-operation Assistance Flows." World Bank, Wash- Directorate, Paris. ington, DC. ------. 2007b. "Fiscal Space Policy for American Bank Success." Washington, Growth and Development." Draft. World DC. Bank, Washington, DC. Dollar, David, and Victoria Levin. 2004. ------. 2007c. "IDA's Long-Term Financial "The Increasing Selectivity of Foreign Capacity." Draft. World Bank, Washing- Aid, 1984­2002." Policy Research Work- ton, DC. ing Paper No. 3299, World Bank, Wash- ------. 2007d. "Multilateral Debt Relief ington, DC. Initiative (MDRI): Debt Relief by IDA IADB (Inter-American Development Bank). and Donor Financing to Date." Draft. 2006. "A Report to the Board of Execu- World Bank, Washington, DC. tive Directors." IADB, Washington, DC. ------. 2007e. "The World Bank's Africa IMF (International Monetary Fund). Action Plan: Progress in Implementation." 2006. "Managing Director's Report Draft. World Bank, Washington, DC. on the IMF's Medium-Term Strategy." ------. 2007f. "The World Bank Group SM/05/332, IMF, Washington, DC. 2006 Trust Funds Annual Report." World International Task Force on Global Public Bank, Washington, DC. Goods. 2006. "Meeting Global Chal- World Food Program. 2006. Avail- lenges: International Cooperation in the able at http://www.wfp.org/english/ National Interest." ITF, Stockholm. ?ModuleID=137&Key=2166). Malan, Pedro, Michael Callaghan, Caio Koch-Weser, William McDonough, Sri Mulyani Indrawati, and Ngozi Okonjo- Iweala. 2007. "Report of the External ADB (Asian Development Bank). 2006a. Review Committee on Bank-Fund Col- "Enhancing ADB Support to Middle- laboration." Washington, DC. Income Countries and Borrowers from World Bank. 2001. "Integrating Gender into Ordinary Capital Resources (OCR). the World Bank's Work: A Strategy for ------. 2006b. "Implementation Review of Action." World Bank, Washington, DC. the Policy on Gender and Development." ------. 2004. "Better Data for Better ADB, Manila. Results: An Action Plan for Improving AfDB (African Development Bank), ADB Development Statistics." World Bank, (Asian Development Bank), EBRD Washington, DC. (European Bank for Reconstruction and ------. 2006a. "Clean Energy and Devel- Development), IADB (Inter-American opment: Towards an Investment Frame- Development Bank), and World Bank. work." World Bank, Washington, DC. 2006. "Cooperation among Multilateral ------. 2006b. "Gender Equality as Smart Development Banks." Update. Washing- Economics: A World Bank Group Gender ton, DC. Action Plan (Fiscal Years 2007­10)." ------. 2007. "The 2006 Multilateral World Bank, Washington, DC. Development Bank Common Perfor- ------. 2006c. "Implementing the World mance Assessment System (COMPAS) Bank's Gender Mainstreaming Strategy." Report." Washington, DC. World Bank, Washington, DC. Bedi, Tara, A. Coudouel, M. Cox, M. ------. 2006d. "Strengthening the World Goldstein, and Nigel Thornton. 2006. Bank's Engagement with IBRD Partner "Beyond the Numbers: Understanding Countries." World Bank, Washington, the Institutions for Monitoring Poverty DC. Reduction Strategies." Washington, DC. ------. 2007. "Accelerating Development Buvinic, Mayra. 2004. "Mainstreaming Outcome in Africa: Progress and Change Attention to Domestic Violence in Lend- in the Africa Action Plan." World Bank, ing Operations: Six Elements of Inter- Washington, DC. Statistical Appendix Table A.1. Millennium Development Goals Goal 1: Eradicate extreme poverty Goal 2: Achieve universal primary education Goal 3: Promote gender equality Goal 4: Reduce child mortality Goal 5: Improve maternal health Goal 6: Combat HIV/AIDS and other diseases Goal 7: Ensure environmental sustainability Goal 8: Develop a global partnership for development Table A.2. Measures of Governance Performance Overall governance performance Control of Corruption (TI, ICS) Bureaucratic capability Budget and financial management (CPIA 13) Public administration (CPIA 15) Checks-and-balances institutions Voice and accountability (KK) Justice and rule of law (KK) Table A.3a. Trade Restrictiveness Index (TRI) and deadweight loss Table A.3b. Overall Trade Restrictiveness Index (OTRI) and import loss Table A.3c. Market Access­Overall Trade Restrictiveness Index (MA-OTRI) Table A.4a. Net Official Development Assistance (ODA) by DAC and non-DAC Countries Table A.4b. Net Official Development Assistance (ODA) Receipts Table A.5 Measures of Gender Performance Primary Education Female primary completion rate Female-to-male ratio: Primary completion rate Child Mortality Female under-5 mortality rate Female-to-male ratio: Under-5 mortality rate Labor Force Participation Female labor force participation (Ages 20­24) Female-to-male ratio: Labor force participation (Ages 20­24) Female labor force participation (Ages 25­49) Female-to-male ratio: Labor force participation (Ages 25­49) CPIA: Country Policy and Institutional Assessment; ICS: Investment Climate Surveys; KK: Kaufmann and Kraay; OTRI: Overall Trade Restrictiveness Index; TI: Transparency International; TRI: Trade Restric- tiveness Index. Millennium Development Goals Goal 2 Goal 1 Achieve universal Goal 3 Goal 4 Eradicate extreme poverty primary education Promote gender equality Reduce child mortality Primary Women in Child Measles education Ratio of girls nonagricultural mortality immunization Poverty Share of completion to boys in sector (under-5 (% of (US$1 a day revenue (gross intake to Secondary primary and (% of total mortality children headcount to poorest final primary enrollment secondary nonagricultural rate per age 12­13 ratio,%) quintile (%) grade,%) (gross,%) school(%) employment) 1,000) months) 1998­2005a 1998­2005a 2001­2005a 2005 2005 2004 2005 2005 Afghanistan .. .. 32 16 55 .. 257 64 Albania <2 8.2 97 78 99 31.7 18 97 Algeria <2 7.0 96 83 102 17.0 39 83 Angola .. 0.0 .. 17 .. .. 260 45 Argentina 6.6 .. 101 86 111 45.5 18 99 Armenia <2 .. 91 88 108 46.5 29 94 Australia .. 5.9 .. 149 102 48.6 6 94 Austria .. 8.6 .. 101 102 46.2 5 75 Azerbaijan 3.7 7.4 94 83 98 48.8 89 98 Bangladesh 36.0 9.1 77 46 101 23.1 73 81 Belarus <2 .. 100 95 105 56.0 12 99 Belgium .. 8.5 .. 109 103 44.8 5 88 Benin 30.9 7.4 65 33 73 .. 150 85 Bhutan .. .. .. .. .. .. 75 93 Bolivia 23.2 1.5 101 89 93 36.5 65 64 Bosnia and Herzegovina .. 9.5 .. .. .. .. 15 90 Botswana 28.1 3.2 92 75 102 43.0 120 90 Brazil 7.5 2.8 108 102 105 46.7 33 99 Bulgaria <2 .. 98 102 100 53.0 15 96 Burkina Faso 27.2 6.9 31 14 77 14.6 191 84 Burundi 54.6 5.1 36 14 83 .. 190 75 Cambodia 34.1 6.8 92 29 87 51.3 87 79 Cameroon 17.1 5.6 62 44 83 21.6 149 68 Canada .. 7.2 .. 109 106 49.4 6 94 Central African Republic 66.6 2.0 23 12 65 .. 193 35 Chad .. .. 32 16 60 12.8 208 23 Chile <2 3.8 95 89 98 38.1 10 90 China 9.9 4.3 98 73 98 40.9 27 86 Hong Kong, China .. .. 110 87 93 47.3 .. 81 Colombia 7.0 2.5 98 79 104 48.3 21 89 Comoros .. .. 51 35 84 .. 71 80 Congo, Dem. Rep. of .. .. 39 22 73 20.1 205 70 Congo, Rep. of .. .. 58 39 89 .. 108 56 Costa Rica 3.3 3.5 92 79 104 38.5 12 89 Côte d'Ivoire .. .. 43 25 67 .. 195 51 Croatia <2 8.3 91 88 104 46.2 7 96 Cuba .. .. 94 94 110 37.7 7 98 Czech Republic <2 10.3 104 96 101 47.1 4 97 Denmark .. 8.3 99 124 109 48.8 5 95 Djibouti .. .. 32 24 75 .. 133 65 Dominican Republic 2.8 4.0 92 71 111 38.2 31 99 Ecuador 17.7 3.3 101 61 .. 42.7 25 93 Egypt, Arab Rep. of 3.1 8.6 95 87 .. 20.6 33 98 El Salvador 19.0 2.7 87 63 100 34.8 27 99 Goal 6 Goal 7 Goal 8 Goal 5 Combat HIV/AIDS Ensure environmental Develop a global partnership for Improve maternal health and other diseases sustainability development Maternal Births Access to Fixed-line and mortality ratio attended Incidence of Access to an improved mobile phone (modeled estimate by skilled HIV prevalence tuberculosis improved sanitation suscribers Internet users per 100,000 health staff (% of population (per 100,000 water source facilities (per 1,000 (per 1,000 live births) (% of total) ages 15­49) people) (% of population) (% of population) people) people) 2000 2000­2005a 2005 2005 2004 2004 2005 2005 .. 14 0.1 168 39 34 44 1 55 98 .. 20 96 91 493 60 140 96 0.1 55 85 92 494 58 1700 45 3.7 269 53 31 75 11 82 95 0.6 41 96 91 798 177 55 98 0.1 71 92 83 260 53 8 99 0.1 6 100 100 1470 698 4 .. 0.3 12 100 100 1441 486 94 88 0.1 76 77 54 397 81 380 13 0.1 227 74 39 71 3 35 100 0.3 62 100 84 755 347 10 .. 0.3 13 100 100 1337 458 850 75 1.8 88 67 33 98 50 420 51 0.1 103 62 70 111 39 420 67 0.1 211 85 46 334 52 31 100 0.1 52 97 95 656 206 100 94 24.1 655 95 42 541 34 260 97 0.5 60 90 75 587 195 32 99 0.1 39 99 99 1128 206 1000 38 2.0 223 61 13 51 5 1000 25 3.3 334 79 36 18 5 450 44 1.6 506 41 17 40 3 730 62 5.4 174 66 51 102 15 6 98 0.3 5 100 100 1080 520 1100 44 10.7 314 75 27 27 3 1100 14 3.5 272 42 9 14 4 31 100 0.3 15 95 91 860 172 56 97 0.1 100 77 44 570 85 .. 100 .. 75 .. .. 1799 508 130 96 0.6 45 93 86 648 104 480 62 0.1 45 86 33 55 33 990 61 3.2 356 46 30 48 2 510 86 5.3 367 58 27 102 13 43 99 0.3 14 97 92 575 254 690 68 7.1 382 84 37 108 11 8 100 0.1 41 100 100 1097 327 33 100 0.1 9 91 98 87 17 9 100 0.1 10 100 98 1465 270 5 .. 0.2 8 100 100 1628 527 730 61 3.1 762 73 82 69 13 150 99 1.1 91 95 78 508 169 130 75 0.3 131 94 89 601 47 84 74 0.1 25 98 70 325 68 150 92 0.9 51 84 62 492 93 (continued) Millennium Development Goals (continued) Goal 2 Goal 1 Achieve universal Goal 3 Goal 4 Eradicate extreme poverty primary education Promote gender equality Reduce child mortality Primary Women in Child Measles education Ratio of girls nonagricultural mortality immunization Poverty Share of completion to boys in sector (under-5 (% of (US$1 a day revenue (gross intake to Secondary primary and (% of total mortality children headcount to poorest final primary enrollment secondary nonagricultural rate per age 12­13 ratio,%) quintile (%) grade,%) (gross,%) school(%) employment) 1,000) months) 1998­2005a 1998­2005a 2001­2005a 2005 2005 2004 2005 2005 Eritrea .. .. 51 31 70 .. 78 84 Estonia <2 6.7 102 98 114 52.2 7 96 Ethiopia 23.0 9.1 55 31 76 40.6 127 59 Finland .. 9.6 100 109 107 50.7 4 97 France .. 7.2 .. 111 105 47.2 5 87 Gabon .. .. 66 50 94 .. 91 55 Gambia, The 59.3 4.8 .. 47 97 .. 137 84 Georgia 6.5 .. 87 83 103 50.3 45 92 Germany .. 8.5 96 100 .. 46.6 5 93 Ghana 44.8 5.6 72 44 91 .. 112 83 Greece .. 6.7 102 96 105 40.7 5 88 Guatemala 13.5 2.9 74 51 91 38.8 43 77 Guinea .. 7.0 55 31 74 .. 160 59 Guinea-Bissau .. 5.2 27 18 65 .. 200 80 Guyana <2 4.5 96 102 100 39.9 63 92 Haiti 53.9 2.4 .. .. .. .. 120 54 Honduras 14.9 3.4 79 66 109 46.8 40 92 Hungary <2 .. 95 97 107 47.0 8 99 India 33.5 8.1 89 54 87 17.3 74 58 Indonesia 7.5 8.4 102 64 97 31.1 36 72 Iran, Islamic Rep. of <2 5.1 96 81 99 13.7 36 94 Iraq .. .. 74 45 76 .. 125 90 Ireland .. 7.4 101 112 103 47.6 6 84 Israel .. 5.7 105 93 105 49.6 6 95 Italy .. 6.5 101 99 106 41.3 4 87 Jamaica <2 5.3 84 88 104 47.0 20 84 Japan .. 10.6 .. 102 98 41.2 4 99 Jordan <2 6.7 97 87 102 25.0 26 95 Kazakhstan <2 7.4 114 99 106 49.4 73 99 Kenya 22.8 6.0 95 49 94 38.7 120 69 Korea, Dem. Rep. of .. .. .. .. .. .. 55 96 Korea, Rep. of <2 7.9 104 93 87 41.6 5 99 Kuwait .. .. 101 95 110 25.2 11 99 Kyrgyz Republic <2 .. 98 86 105 43.8 67 99 Lao PDR 27.0 8.1 76 47 84 .. 79 41 Latvia <2 6.6 92 97 115 53.2 11 95 Lebanon .. .. 90 89 104 .. 30 96 Lesotho 36.4 1.5 67 39 103 .. 132 85 Liberia .. .. .. 32 73 .. 235 94 Libya .. .. .. 104 106 .. 19 97 Lithuania <2 .. 98 102 110 52.2 9 97 Macedonia, FYR <2 .. 96 84 103 42.3 17 96 Madagascar 61.0 4.9 58 .. 96 .. 119 59 Malawi 20.8 7.0 61 28 98 12.4 125 82 Goal 6 Goal 7 Goal 8 Goal 5 Combat HIV/AIDS Ensure environmental Develop a global partnership for Improve maternal health and other diseases sustainability development Maternal Births Access to Fixed-line and mortality ratio attended Incidence of Access to an improved mobile phone (modeled estimate by skilled HIV prevalence tuberculosis improved sanitation suscribers Internet users per 100,000 health staff (% of population (per 100,000 water source facilities (per 1,000 (per 1,000 live births) (% of total) ages 15­49) people) (% of population) (% of population) people) people) 2000 2000­2005a 2005 2005 2004 2004 2005 2005 630 28 2.4 282 60 9 18 16 63 100 1.3 43 100 97 1402 513 850 6 1.4 344 22 13 14 2 6 100 0.1 6 100 100 1401 534 17 .. 0.4 13 100 .. 1376 430 420 86 7.9 308 88 36 498 48 540 55 2.4 242 82 53 192 33 32 92 0.2 83 82 94 337 39 8 .. 0.1 7 100 100 1628 455 540 47 2.3 205 75 18 143 18 9 .. 0.2 17 .. .. 1472 180 240 41 0.9 78 95 86 457 79 740 56 1.5 236 50 18 20 5 1100 35 3.8 206 59 35 8 20 170 86 2.4 149 83 70 521 213 680 24 3.8 306 54 30 64 70 110 56 1.5 78 87 69 247 36 16 100 0.1 22 99 95 1257 297 540 43 0.9 168 86 33 128 55 230 72 0.1 239 77 55 271 73 76 90 0.2 24 94 .. 384 103 .. 72 .. 56 81 79 57 1 5 100 0.2 12 .. .. 1501 276 17 .. .. 8 100 .. 1545 470 5 .. 0.5 7 .. .. 1659 478 87 97 1.5 7 93 80 1146 404 10 .. 0.1 28 100 100 1202 668 41 100 .. 5 97 93 423 118 210 .. 0.1 144 86 72 350 27 1000 42 6.1 641 61 43 143 32 67 97 .. 178 100 59 41 0 20 100 0.1 96 92 .. 1286 684 5 100 .. 24 .. .. 1140 276 110 99 0.1 121 77 59 191 54 650 19 0.1 155 51 30 120 4 42 100 0.8 63 99 78 1131 448 150 93 0.1 11 100 98 554 196 550 55 23.2 696 79 37 163 24 760 51 .. 301 61 27 3 0 97 .. .. 18 .. 97 156 36 13 100 0.2 63 .. .. 1510 358 23 99 0.1 30 .. .. 882 79 550 51 0.5 234 46 32 31 5 1800 56 14.1 409 73 61 41 4 (continued) Millennium Development Goals (continued) Goal 2 Goal 1 Achieve universal Goal 3 Goal 4 Eradicate extreme poverty primary education Promote gender equality Reduce child mortality Primary Women in Child Measles education Ratio of girls nonagricultural mortality immunization Poverty Share of completion to boys in sector (under-5 (% of (US$1 a day revenue (gross intake to Secondary primary and (% of total mortality children headcount to poorest final primary enrollment secondary nonagricultural rate per age 12­13 ratio,%) quintile (%) grade,%) (gross,%) school(%) employment) 1,000) months) 1998­2005a 1998­2005a 2001­2005a 2005 2005 2004 2005 2005 Malaysia <2 4.4 94 76 109 36.9 12 90 Mali 36.1 6.1 38 24 75 .. 218 86 Mauritania 25.9 6.2 45 21 96 .. 125 61 Mauritius .. .. 98 89 98 37.5 15 98 Mexico 3.0 4.3 99 80 101 37.4 27 96 Moldova <2 .. 92 82 109 54.6 16 97 Mongolia 10.8 7.5 97 94 116 50.3 49 99 Morocco <2 6.5 80 50 88 21.8 40 97 Mozambique 36.2 5.4 42 14 82 .. 145 77 Myanmar .. .. 79 40 105 .. 105 72 Namibia 34.9 1.4 75 61 101 48.8 62 73 Nepal 24.1 6.0 75 46 88 17.4 74 74 Netherlands .. 7.6 100 119 99 45.4 5 96 New Zealand .. 6.4 .. 118 113 50.5 6 82 Nicaragua 45.1 5.6 76 66 103 .. 37 96 Niger 60.6 2.6 28 9 72 7.8 256 83 Nigeria 70.8 5.1 82 34 82 .. 194 35 Norway .. 9.6 101 116 109 49.2 4 90 Oman .. .. 93 87 99 25.7 12 98 Pakistan 17.0 9.3 63 27 76 8.6 99 78 Panama 7.4 2.5 97 70 110 43.5 24 99 Papua New Guinea .. 4.5 54 26 87 35.4 74 60 Paraguay 13.6 2.4 91 63 101 43.9 23 90 Peru 10.5 3.7 100 92 103 34.6 27 80 Philippines 14.8 5.4 97 86 106 40.4 33 80 Poland <2 .. 100 97 109 47.2 7 98 Portugal <2 5.8 104 97 108 46.6 5 93 Puerto Rico .. .. .. .. .. 39.3 .. .. Romania <2 8.1 93 85 105 46.5 19 97 Russian Federation <2 .. 94 93 110 50.9 18 99 Rwanda 60.3 5.3 39 14 99 .. 203 89 São Tomé and Principe .. .. 77 45 99 .. 118 88 Saudi Arabia .. .. 85 88 101 13.5 26 96 Senegal 17.0 6.6 52 26 90 .. 119 74 Serbia and Montenegro .. 8.3 96 89 103 45.4 15 96 Sierra Leone 57.0 1.1 .. 30 71 .. 282 67 Singapore .. 5.0 .. .. .. 47.0 3 96 Slovak Republic <2 8.8 99 94 104 52.0 8 98 Slovenia <2 9.1 102 100 109 47.6 4 94 Solomon Islands .. .. .. 30 91 .. 29 72 Somalia .. .. .. .. .. .. 225 35 South Africa 10.7 3.5 99 93 101 45.9 68 82 Spain .. 7.0 109 119 107 42.0 5 97 Sri Lanka 5.6 7.0 .. 83 102 43.2 14 99 Goal 6 Goal 7 Goal 8 Goal 5 Combat HIV/AIDS Ensure environmental Develop a global partnership for Improve maternal health and other diseases sustainability development Maternal Births Access to Fixed-line and mortality ratio attended Incidence of Access to an improved mobile phone (modeled estimate by skilled HIV prevalence tuberculosis improved sanitation suscribers Internet users per 100,000 health staff (% of population (per 100,000 water source facilities (per 1,000 (per 1,000 live births) (% of total) ages 15­49) people) (% of population) (% of population) people) people) 2000 2000­2005a 2005 2005 2004 2004 2005 2005 41 97 0.5 102 99 94 943 435 1200 41 1.7 278 50 46 70 4 1000 57 0.7 298 53 34 256 7 24 99 0.6 62 100 94 863 146 83 83 0.3 23 97 79 650 181 36 100 1.1 138 92 68 480 96 110 97 0.1 191 62 59 279 105 220 63 0.1 89 81 73 455 153 1000 48 16.1 447 43 32 40 7 360 57 1.3 171 78 77 13 2 300 76 19.6 697 87 25 206 37 740 15 0.5 180 90 35 26 4 16 .. 0.2 7 100 100 1436 739 7 .. 0.1 9 .. .. 1283 672 230 67 0.2 58 79 47 260 27 1600 16 1.1 164 46 13 23 2 800 35 3.9 283 48 44 151 38 16 .. 0.1 5 100 100 1489 735 87 95 .. 11 .. .. 623 111 500 31 0.1 181 91 59 116 67 160 93 0.9 45 90 73 555 64 300 41 1.8 250 39 44 15 23 170 77 0.4 68 86 80 374 34 410 73 0.6 172 83 63 280 165 200 60 0.1 291 85 72 459 54 13 100 0.1 26 .. .. 1074 262 5 100 0.4 33 .. .. 1487 279 25 100 .. 5 .. .. 974 221 49 99 0.1 134 57 .. 820 208 67 99 1.1 119 97 87 1119 152 1400 39 3.0 361 74 42 18 6 .. 76 .. 105 79 25 97 131 23 93 .. 41 .. .. 740 70 690 58 0.9 255 76 57 171 46 11 92 0.2 34 93 87 917 148 2000 42 1.6 475 57 39 19 2 30 100 0.3 29 100 100 1435 571 3 99 0.1 17 100 99 1065 464 17 100 0.1 15 .. .. 1288 545 130 .. .. 142 70 31 28 8 1100 25 0.9 224 29 26 73 11 230 92 18.8 600 88 65 825 109 4 .. 0.6 28 100 100 1375 348 92 96 0.1 61 79 91 235 14 (continued) Millennium Development Goals (continued) Goal 2 Goal 1 Achieve universal Goal 3 Goal 4 Eradicate extreme poverty primary education Promote gender equality Reduce child mortality Primary Women in Child Measles education Ratio of girls nonagricultural mortality immunization Poverty Share of completion to boys in sector (under-5 (% of (US$1 a day revenue (gross intake to Secondary primary and (% of total mortality children headcount to poorest final primary enrollment secondary nonagricultural rate per age 12­13 ratio,%) quintile (%) grade,%) (gross,%) school(%) employment) 1,000) months) 1998­2005a 1998­2005a 2001­2005a 2005 2005 2004 2005 2005 Sudan .. .. 50 34 89 16.8 90 60 Swaziland .. 4.3 64 45 94 29.9 160 60 Sweden .. 9.1 .. 103 112 50.9 4 94 Switzerland .. 7.6 97 93 94 47.1 5 82 Syrian Arab Republic .. .. 111 68 94 18.2 15 98 Tajikistan 7.4 7.9 102 82 84 53.3 71 84 Tanzania 57.8 7.3 54 .. 95 45.4 122 91 Thailand <2 6.3 82 73 101 46.4 21 96 Togo .. .. 65 40 72 .. 139 70 Trinidad and Tobago 12.4 5.9 99 88 104 41.1 19 93 Tunisia <2 6.0 97 81 105 25.0 24 96 Turkey 3.4 .. 88 79 84 19.9 29 91 Turkmenistan .. 6.1 .. .. .. .. 104 99 Uganda .. 5.7 57 16 96 .. 136 86 Ukraine <2 9.2 114 89 102 55.1 17 96 United Arab Emirates .. .. 77 64 126 14.5 9 92 United Kingdom .. 6.1 .. 105 107 49.4 6 82 United States .. 5.4 .. 95 109 48.5 7 93 Uruguay <2 .. 91 108 114 46.8 15 95 Uzbekistan <2 7.2 97 95 96 39.5 68 99 Venezuela, R. B. de 18.5 3.3 92 75 104 41.5 21 76 Vietnam .. 9.0 94 76 94 49.1 19 95 West Bank and Gaza .. .. 98 99 104 17.9 23 99 Yemen, Rep. of 15.7 7.4 62 48 61 .. 102 76 Zambia 63.8 3.6 78 28 92 .. 182 84 Zimbabwe 56.1 4.6 80 36 95 21.8 132 85 World 18.3 .. 85 65 94 38.1 75 77 Fragile States .. .. 65 38 82 .. 178 56 Low income .. .. 74 45 87 23.4 114 65 Middle income .. .. 96 77 99 40.9 37 87 Lower middle income .. .. 97 76 99 40.2 39 86 Upper middle income .. .. 95 86 99 44.2 27 93 Low & middle income .. .. 84 61 93 36.2 82 75 East Asia & Pacific 9.1 .. 98 71 99 40.6 33 83 Europe & Central Asia 0.9 .. 92 91 96 47.6 32 96 Latin America & Caribbean 8.6 .. 98 86 102 43.3 31 92 Middle East & North Africa 1.5 .. 89 73 90 17.7 53 92 South Asia 31.7 .. 82 50 88 17.8 83 64 Sub-Saharan Africa 41.1 .. 58 31 86 .. 163 64 High income .. .. 97 100 100 46.0 7 93 European Monetary Union .. .. 101 106 105 45.1 5 90 Source: 2007 World Development Indicators database. Figures in italics refer to periods other than those specified. a. Data are for the most recent year available. .. Not available Goal 6 Goal 7 Goal 8 Goal 5 Combat HIV/AIDS Ensure environmental Develop a global partnership for Improve maternal health and other diseases sustainability development Maternal Births Access to Fixed-line and mortality ratio attended Incidence of Access to an improved mobile phone (modeled estimate by skilled HIV prevalence tuberculosis improved sanitation suscribers Internet users per 100,000 health staff (% of population (per 100,000 water source facilities (per 1,000 (per 1,000 live births) (% of total) ages 15­49) people) (% of population) (% of population) people) people) 2000 2000­2005a 2005 2005 2004 2004 2005 2005 590 87 1.6 228 70 34 69 77 370 74 33.4 1262 62 48 208 32 2 .. 0.2 6 100 100 1804 764 7 .. 0.4 7 100 100 1610 498 160 70 .. 37 93 90 307 58 100 71 0.1 198 59 51 46 1 1500 43 6.5 342 62 47 56 9 44 99 1.4 142 99 99 537 110 570 61 3.2 373 52 35 82 49 160 96 2.6 9 91 100 861 123 120 90 0.1 24 93 85 692 95 70 83 .. 29 96 88 869 222 31 97 0.1 70 72 62 82 8 880 39 6.4 369 60 43 56 17 35 100 1.4 99 96 96 546 97 54 100 .. 16 100 98 1273 308 13 .. .. 14 100 .. 1616 474 17 99 0.6 5 100 100 1227 630 27 99 0.5 28 100 100 624 193 24 96 0.2 114 82 67 80 34 96 95 0.7 42 83 68 606 125 130 90 0.5 175 85 61 306 129 .. 97 .. 21 92 73 398 67 570 27 .. 82 67 43 92 9 750 43 17.0 600 58 55 89 20 1100 .. 20.1 601 81 53 79 77 410 63 1.0 136 83 57 523 137 886 .. 3.1 278 58 42 85 26 684 41 1.7 220 75 38 114 44 150 88 0.6 111 84 62 590 115 163 87 0.3 113 82 57 511 95 91 92 2.2 104 94 84 901 196 450 61 1.1 158 80 52 382 85 117 87 0.2 137 79 51 496 89 58 94 0.7 84 92 85 898 190 194 87 0.6 61 91 77 496 156 183 74 0.1 43 90 76 389 89 564 37 0.7 174 84 37 119 49 921 45 5.8 348 56 37 142 29 14 .. 0.4 17 100 100 1338 527 10 .. 0.3 13 100 .. 1511 439 Measures of Governance Performance Overall governance performancea Bureaucratic capabilitya Checks-and-balances institutionsa Budget and financial Public Voice and Justice and Control of corruption management administration accountability rule of law ICS--unofficial payments for TI Corruption time to get Perceptions things done KK Voice and KK Rule Indexb (% of sales)c CPIA 13d CPIA 15d Accountabilitye of Lawe Est. 2006 S.E. 2006 2005 2005 Est. 2005 S.E. Est. 2005 S.E. Afghanistan .. .. .. .. .. ­1.28 0.15 ­1.68 0.21 Albania 2.6 0.3 1.6 4 3 0.08 0.12 ­0.84 0.16 Algeria 3.1 0.9 6.0 .. .. ­0.92 0.12 ­0.71 0.14 Angola 2.2 0.5 .. 2.5 2.5 ­1.15 0.12 ­1.28 0.15 Argentina 2.9 0.5 0.7 .. .. 0.43 0.14 ­0.56 0.13 Armenia 2.9 0.3 0.7 4 4 ­0.64 0.12 ­0.46 0.14 Australia 8.7 0.7 .. .. .. 1.32 0.16 1.80 0.14 Austria 8.6 0.7 .. .. .. 1.24 0.16 1.87 0.14 Azerbaijan 2.4 0.4 2.7 4 3 ­1.16 0.11 ­0.84 0.13 Bangladesh 2 0.5 2.1 3 3 ­0.50 0.12 ­0.87 0.14 Belarus 2.1 0.3 0.5 .. .. ­1.68 0.12 ­1.04 0.16 Belgium 7.3 1.3 .. .. .. 1.31 0.16 1.47 0.14 Benin 2.5 0.8 4.6 4 3 0.34 0.15 ­0.59 0.17 Bhutan 6 3.2 .. 3.5 4 ­1.05 0.15 0.52 0.24 Bolivia 2.7 0.6 2.1 3.5 3.5 ­0.09 0.12 ­0.78 0.14 Bosnia and Herzegovina 2.9 0.4 0.3 3.5 3 ­0.11 0.12 ­0.74 0.15 Botswana 5.6 1.8 1.0 .. .. 0.68 0.14 0.70 0.14 Brazil 3.3 0.5 .. .. .. 0.36 0.14 ­0.41 0.13 Bulgaria 4 1.4 1.0 .. .. 0.59 0.11 ­0.19 0.13 Burkina Faso 3.2 0.8 5.7 4 3.5 ­0.37 0.13 ­0.54 0.19 Burundi 2.4 0.4 2.9 2.5 2.5 ­1.15 0.16 ­1.17 0.19 Cambodia 2.1 0.5 4.6 2.5 2.5 ­0.94 0.16 ­1.13 0.16 Cameroon 2.3 0.4 2.6 3.5 3 ­1.19 0.14 ­1.02 0.15 Canada 8.5 0.9 .. .. .. 1.32 0.16 1.81 0.14 Cape Verde .. .. 0.0 3.5 4 0.83 0.18 0.21 0.21 Central African Republic 2.4 0.3 .. 2 2 ­1.15 0.16 ­1.29 0.19 Chad 2 0.5 .. 3 2.5 ­1.25 0.16 ­1.23 0.18 Chile 7.3 1 0.3 .. .. 1.04 0.14 1.20 0.13 China 3.3 0.6 1.6 .. .. ­1.66 0.12 ­0.47 0.13 Colombia 3.9 1.2 0.7 .. .. ­0.32 0.12 ­0.71 0.13 Comoros .. .. .. 2 2 ­0.28 0.19 ­0.96 0.26 Congo, Dem. Rep. of 2.2 0.1 .. 2.5 2.5 ­1.64 0.14 ­1.76 0.16 Congo, Rep. of 2 0.4 .. 3 2.5 ­0.71 0.18 ­1.42 0.17 Costa Rica 4.1 1.5 2.3 .. .. 0.99 0.14 0.54 0.14 Côte d'Ivoire 2.1 0.2 .. 2.5 2 ­1.50 0.14 ­1.47 0.16 Croatia 3.4 0.6 0.3 .. .. 0.51 0.11 0.00 0.13 Cuba 3.5 2.9 .. .. .. ­1.87 0.14 ­1.14 0.16 Czech Republic 4.8 0.8 0.4 .. .. 1.01 0.11 0.70 0.12 Denmark 9.5 0.2 .. .. .. 1.51 0.16 1.99 0.14 Djibouti .. .. .. 3 2.5 ­0.84 0.19 ­0.87 0.21 Dominican Republic 2.8 0.8 .. .. .. 0.20 0.14 ­0.66 0.14 Ecuador 2.3 0.3 2.8 .. .. ­0.16 0.12 ­0.84 0.14 Egypt, Arab Rep. of 3.3 0.7 1.3 .. .. ­1.15 0.12 0.02 0.13 (continued) Measures of Governance Performance (continued) Overall governance performancea Bureaucratic capabilitya Checks-and-balances institutionsa Budget and financial Public Voice and Justice and Control of corruption management administration accountability rule of law ICS--unofficial payments for TI Corruption time to get Perceptions things done KK Voice and KK Rule Indexb (% of sales)c CPIA 13d CPIA 15d Accountabilitye of Lawe Est. 2006 S.E. 2006 2005 2005 Est. 2005 S.E. Est. 2005 S.E. El Salvador 4 1.6 1.1 .. .. 0.26 0.14 ­0.37 0.15 Equatorial Guinea 2.1 0.5 .. .. .. ­1.71 0.19 ­1.33 0.20 Eritrea 2.9 1.3 0.2 2.5 3 ­1.83 0.13 ­0.81 0.22 Estonia 6.7 1.3 0.2 .. .. 1.05 0.12 0.82 0.12 Ethiopia 2.4 0.4 .. 3.5 3 ­1.10 0.12 ­0.77 0.15 Finland 9.6 0.3 .. .. .. 1.49 0.16 1.96 0.14 France 7.4 1.1 .. .. .. 1.28 0.16 1.35 0.14 Gabon 3 0.9 .. .. .. ­0.71 0.16 ­0.48 0.15 Gambia, The 2.5 0.5 .. 2.5 3 ­0.72 0.18 ­0.29 0.18 Georgia 2.8 0.5 0.2 3.5 3.5 ­0.27 0.12 ­0.82 0.14 Germany 8 0.6 0.3 .. .. 1.31 0.16 1.76 0.14 Ghana 3.3 0.6 .. 3.5 3.5 0.41 0.14 ­0.23 0.14 Greece 4.4 1.1 0.2 .. .. 0.95 0.16 0.66 0.14 Guatemala 2.6 0.7 2.6 .. .. ­0.37 0.14 ­1.04 0.14 Guinea 1.9 0.4 .. 3 3 ­1.18 0.15 ­1.11 0.19 Guinea­Bissau .. .. .. 2.5 2.5 ­0.31 0.18 ­1.33 0.21 Guyana 2.5 0.4 0.4 3.5 2.5 0.49 0.18 ­0.80 0.18 Haiti 1.8 0.1 .. 2.5 2.5 ­1.41 0.15 ­1.62 0.20 Honduras 2.5 0.3 1.7 4 3 ­0.14 0.12 ­0.78 0.14 Hong Kong, China 8.3 1.1 .. .. .. 0.26 0.17 1.50 0.14 Hungary 5.2 0.4 0.5 .. .. 1.01 0.12 0.70 0.12 India 3.3 0.5 4.2 4 3.5 0.35 0.14 0.09 0.13 Indonesia 2.4 0.4 1.1 3.5 3.5 ­0.21 0.14 ­0.87 0.13 Iran, Islamic Rep. of 2.7 0.8 .. .. .. ­1.43 0.12 ­0.76 0.14 Iraq 1.9 0.5 .. .. .. ­1.47 0.14 ­1.81 0.17 Ireland 7.4 1.2 0.1 .. .. 1.41 0.16 1.63 0.14 Isle of Man .. .. .. .. .. .. .. .. .. Israel 5.9 1.3 .. .. .. 0.61 0.16 0.76 0.14 Italy 4.9 1 .. .. .. 1.00 0.16 0.51 0.14 Jamaica 3.7 0.6 0.3 .. .. 0.57 0.14 ­0.55 0.14 Japan 7.6 1.1 .. .. .. 0.94 0.16 1.33 0.14 Jordan 5.3 1.2 .. .. .. ­0.74 0.14 0.43 0.13 Kazakhstan 2.6 0.5 0.7 .. .. ­1.19 0.11 ­0.79 0.12 Kenya 2.2 0.4 2.9 3.5 3 ­0.12 0.14 ­0.94 0.14 Korea, Dem. Rep. of .. .. .. .. .. ­2.06 0.14 ­1.15 0.18 Korea, Rep. of 5.1 0.8 0.0 .. .. 0.74 0.14 0.73 0.13 Kuwait 4.8 1.4 .. .. .. ­0.47 0.15 0.67 0.15 Kyrgyz Republic 2.2 0.6 2.4 3 2.5 ­1.03 0.12 ­1.07 0.14 Lao PDR 2.6 1.1 .. 2.5 2.5 ­1.54 0.13 ­1.12 0.17 Latvia 4.7 1.5 0.5 .. .. 0.89 0.12 0.43 0.13 Lebanon 3.6 0.6 2.5 .. .. ­0.72 0.14 ­0.36 0.15 Lesotho 3.2 0.7 0.2 3 3 0.28 0.18 ­0.19 0.19 Liberia .. .. .. .. .. ­0.92 0.15 ­1.60 0.26 (continued) Measures of Governance Performance (continued) Overall governance performancea Bureaucratic capabilitya Checks-and-balances institutionsa Budget and financial Public Voice and Justice and Control of corruption management administration accountability rule of law ICS--unofficial payments for TI Corruption time to get Perceptions things done KK Voice and KK Rule Indexb (% of sales)c CPIA 13d CPIA 15d Accountabilitye of Lawe Est. 2006 S.E. 2006 2005 2005 Est. 2005 S.E. Est. 2005 S.E. Libya 2.7 0.8 .. .. .. ­1.93 0.12 ­0.73 0.16 Lithuania 4.8 1.4 0.8 .. .. 0.90 0.11 0.46 0.13 Macedonia, FYR 2.7 0.3 0.4 .. .. 0.03 0.11 ­0.38 0.14 Madagascar 3.1 1.4 0.9 3 3.5 ­0.01 0.15 ­0.15 0.17 Malawi 2.7 0.5 1.2 3 3.5 ­0.45 0.14 ­0.35 0.14 Malaysia 5 1 .. .. .. ­0.41 0.14 0.58 0.13 Mali 2.8 0.8 2.9 4 3 0.47 0.15 ­0.12 0.17 Mauritania 3.1 1.6 4.6 2 3 ­1.09 0.15 ­0.54 0.20 Mauritius 5.1 2.2 0.6 .. .. 0.92 0.15 0.79 0.15 Mexico 3.3 0.3 0.5 .. .. 0.29 0.14 ­0.48 0.13 Moldova 3.2 1.1 0.8 3.5 3 ­0.49 0.11 ­0.59 0.13 Mongolia 2.8 1.1 .. 4 3.5 0.36 0.15 ­0.26 0.17 Morocco 3.2 0.7 .. .. .. ­0.76 0.14 ­0.10 0.14 Mozambique 2.8 0.5 .. 3.5 3 ­0.06 0.13 ­0.72 0.15 Myanmar 1.9 0.5 .. .. .. ­2.16 0.14 ­1.56 0.17 Namibia 4.1 1.3 0.8 .. .. 0.36 0.14 ­0.01 0.14 Nepal 2.5 0.6 .. 3.5 3 ­1.19 0.15 ­0.81 0.15 Netherlands 8.7 0.7 .. .. .. 1.45 0.16 1.78 0.14 New Zealand 9.6 0.2 .. .. .. 1.39 0.16 1.95 0.14 Nicaragua 2.6 0.5 1.8 3.5 3.5 ­0.01 0.14 ­0.70 0.15 Niger 2.3 0.5 4.7 3.5 3 ­0.06 0.15 ­0.82 0.19 Nigeria 2.2 0.3 .. 3 2.5 ­0.69 0.14 ­1.38 0.14 Norway 8.8 0.7 .. .. .. 1.45 0.16 1.99 0.14 Oman 5.4 2.1 1.0 .. .. ­0.94 0.17 0.72 0.16 Pakistan 2.2 0.4 1.6 3.5 3.5 ­1.23 0.14 ­0.81 0.13 Panama 3.1 0.5 2.6 .. .. 0.52 0.14 ­0.11 0.14 Papua New Guinea 2.4 0.3 .. 3.5 3 ­0.05 0.15 ­0.92 0.16 Paraguay 2.6 1.1 5.3 .. .. ­0.19 0.12 ­1.00 0.14 Peru 3.3 1 0.1 .. .. 0.04 0.12 ­0.77 0.13 Philippines 2.5 0.5 1.2 .. .. 0.01 0.12 ­0.52 0.13 Poland 3.7 1.2 0.4 .. .. 1.04 0.11 0.32 0.12 Portugal 6.6 1.4 0.1 .. .. 1.32 0.16 1.01 0.14 Puerto Rico .. .. .. .. .. 1.03 0.21 0.62 0.28 Romania 3.1 0.2 0.6 .. .. 0.36 0.11 ­0.29 0.12 Russian Federation 2.5 0.4 1.0 .. .. ­0.85 0.10 ­0.84 0.12 Rwanda 2.5 0.3 .. 3.5 3.5 ­1.32 0.13 ­1.00 0.20 Saudi Arabia 3.3 1.5 .. .. .. ­1.72 0.14 0.20 0.15 Senegal 3.3 0.9 0.2 3.5 3.5 0.30 0.14 ­0.26 0.15 Serbia and Montenegro 3 0.6 0.6 3.5 4 0.12 0.11 ­0.81 0.15 Sierra Leone 2.2 0.1 .. 3.5 3 ­0.38 0.15 ­1.12 0.20 Singapore 9.4 0.3 .. .. .. ­0.29 0.14 1.83 0.13 Slovak Republic 4.7 0.9 0.4 .. .. 1.04 0.12 0.41 0.12 Slovenia 6.4 1.3 0.1 .. .. 1.08 0.12 0.79 0.13 (continued) Measures of Governance Performance (continued) Overall governance performancea Bureaucratic capabilitya Checks-and-balances institutionsa Budget and financial Public Voice and Justice and Control of corruption management administration accountability rule of law ICS--unofficial payments for TI Corruption time to get Perceptions things done KK Voice and KK Rule Indexb (% of sales)c CPIA 13d CPIA 15d Accountabilitye of Lawe Est. 2006 S.E. 2006 2005 2005 Est. 2005 S.E. Est. 2005 S.E. Somalia .. .. .. .. .. ­1.89 0.15 ­2.36 0.26 South Africa 4.6 1 0.1 .. .. 0.82 0.14 0.19 0.12 Spain 6.8 0.9 0.0 .. .. 1.12 0.16 1.13 0.14 Sri Lanka 3.1 0.8 0.1 4 3 ­0.26 0.14 0.00 0.14 Sudan 2 0.4 .. 2.5 2.5 ­1.84 0.14 ­1.48 0.17 Swaziland 2.5 0.5 0.5 .. .. ­1.28 0.15 ­0.75 0.19 Sweden 9.2 0.3 .. .. .. 1.41 0.16 1.84 0.14 Switzerland 9.1 0.3 .. .. .. 1.43 0.16 2.02 0.14 Syrian Arab Republic 2.9 0.9 .. .. .. ­1.67 0.12 ­0.42 0.16 Taiwan, China 5.9 0.6 .. .. .. 0.79 0.14 0.83 0.13 Tajikistan 2.2 0.4 1.0 3 2.5 ­1.17 0.11 ­0.99 0.15 Tanzania 2.9 0.4 0.2 4.5 3.5 ­0.31 0.14 ­0.47 0.14 Thailand 3.6 0.7 .. .. .. 0.07 0.12 0.10 0.13 Timor­Leste 2.6 0.7 .. .. .. 0.18 0.18 ­0.55 0.26 Togo 2.4 0.7 .. 2 2 ­1.23 0.15 ­1.07 0.19 Trinidad and Tobago 3.2 0.8 .. .. .. 0.44 0.16 ­0.07 0.15 Tunisia 4.6 1.7 .. .. .. ­1.13 0.12 0.21 0.14 Turkey 3.8 0.9 0.4 .. .. ­0.04 0.12 0.07 0.13 Turkmenistan 2.2 0.6 .. .. .. ­1.95 0.12 ­1.41 0.15 Uganda 2.7 0.6 1.3 4 3 ­0.59 0.14 ­0.74 0.14 Ukraine 2.8 0.5 1.4 .. .. ­0.26 0.11 ­0.60 0.12 United Arab Emirates 6.2 1.3 .. .. .. ­1.08 0.14 0.58 0.15 United Kingdom 8.6 0.7 .. .. .. 1.30 0.16 1.69 0.14 United States 7.3 1.2 .. .. .. 1.19 0.16 1.59 0.14 Uruguay 6.4 1.1 0.1 .. .. 0.99 0.14 0.43 0.14 Uzbekistan 2.1 0.4 0.6 3 2.5 ­1.76 0.11 ­1.31 0.14 Venezuela, R. B. de 2.3 0.2 .. .. .. ­0.50 0.14 ­1.22 0.13 Vietnam 2.6 0.5 0.5 4 3.5 ­1.60 0.14 ­0.45 0.13 West Bank and Gaza .. .. .. .. .. ­1.22 0.20 ­0.52 0.31 Yemen, Rep. of 2.6 0.3 .. 3 3 ­1.07 0.14 ­1.01 0.15 Zambia 2.6 0.9 1.1 3 3 ­0.35 0.12 ­0.62 0.14 Zimbabwe 2.4 0.8 .. 2.5 2 ­1.65 0.14 ­1.47 0.14 Sources: Various indicators as labeled for individual columns. a. Though shown only for KK and TI, all indicators have margins of error b. Transparency International's Corruption Perceptions Index (CPI) score relates to perceptions of the degree of corruption as seen by businesspeople and country analysts and ranges between 10 (highly clean) and 0 (highly corrupt) (http://www.transparency.org/policy_research/surveys_indices/cpi/2006) c.http://www.enterprisesurveys.org d. The CPIA 2005 data are grouped from strong (1) to weak (5) e. KK Governance indicators lie between ­2.5 amd 2.5, with higher scores corresponding to better ourcomes (www.worldbank.org/wbi/governance/data) .. Not available Trade Restrictiveness Index (TRI) and deadweight loss (US$, millions) TRI DWL DWL/GDP TRI Agric. DWL Agric. TRI Mfg. DWL Mfg. Albania 9 9 0.1% 8 1 9 8 Algeria 15 219 0.2% 14 42 15 176 Argentina 16 432 0.2% 11 6 16 426 Australia 7 284 0.0% 5 8 7 275 Bangladesh 66 2,103 3.5% 11 11 74 2,091 Belarus 10 71 0.2% 11 9 9 62 Bolivia 8 7 0.1% 10 1 8 6 Brazil 11 496 0.1% 9 24 11 472 Brunei 8 4 0.1% 0 0 8 4 Burkina Faso 13 10 0.2% 14 1 12 9 Cameroon 16 34 0.2% 18 8 15 27 Canada 14 2,867 0.3% 54 2,555 5 312 Chile 6 40 0.0% 6 3 6 36 China 9 1,991 0.1% 14 258 8 1,733 Colombia 14 186 0.2% 16 32 14 154 Costa Rica 8 20 0.1% 13 6 7 14 Côte d'Ivoire 11 22 0.1% 12 4 11 18 Egypt, Arab Rep. of 34 677 0.8% 41 235 32 442 El Salvador 9 19 0.1% 17 9 7 10 Ethiopia 16 28 0.3% 14 4 17 24 European Union 14 13,672 0.1% 49 12,045 5 1,626 Gabon 18 16 0.2% 21 4 18 12 Ghana 17 61 0.6% 13 8 18 54 Guatemala 8 32 0.1% 12 8 7 24 Honduras 7 9 0.1% 11 4 6 5 Hong Kong, China 0 0 0.0% 0 0 0 0 Iceland 6 7 0.0% 13 3 4 4 India 21 2,321 0.3% 78 1,196 15 1,125 Indonesia 8 198 0.1% 7 18 8 181 Japan 33 27,567 0.6% 85 24,321 12 3,246 Jordan 15 95 0.7% 16 17 14 78 Kazakhstan 3 3 0.0% 3 0 3 2 Kenya 15 49 0.3% 35 31 10 18 Korea, Rep. of 48 24,873 3.2% 183 24,374 7 500 Lebanon 9 38 0.2% 17 19 7 19 Madagascar 11 8 0.2% 11 1 11 7 Malawi 14 10 0.5% 12 1 14 9 Malaysia 10 432 0.3% 5 7 10 425 Mali 11 7 0.1% 14 1 10 5 Mauritius 6 6 0.1% 8 2 6 4 Mexico 21 4,428 0.6% 52 2,111 16 2,317 Morocco 25 594 1.1% 43 194 21 400 Moldova 6 3 0.1% 11 1 5 2 New Zealand 8 85 0.0% 13 15 8 69 Nicaragua 8 7 0.1% 13 2 7 5 Nigeria 13 118 0.1% 24 49 10 69 Norway 14 561 0.2% 45 462 6 99 Oman 5 10 0.0% 4 1 5 9 Papua N. Guinea 7 3 0.1% 12 2 6 2 Paraguay 11 16 0.2% 12 1 11 15 Peru 10 58 0.1% 11 11 9 47 continued Trade Restrictiveness Index (TRI) and deadweight loss (US$, millions) (continued) TRI DWL DWL/GDP TRI Agric. DWL Agric. TRI Mfg. DWL Mfg. Philippines 6 93 0.1% 17 41 5 52 Romania 18 605 0.6% 26 75 17 530 Russian Fed. 12 575 0.1% 17 169 10 406 Rwanda 23 5 0.3% 22 1 23 4 Saudi Arabia 5 70 0.0% 7 15 5 55 Senegal 11 17 0.2% 12 6 10 11 Singapore 0 0 0.0% 0 0 0 0 South Africa 12 408 0.2% 9 10 13 398 Sri Lanka 10 41 0.2% 17 14 9 28 Sudan 20 74 0.3% 20 9 20 65 Switzerland 10 559 0.2% 39 513 3 46 Tanzania 15 32 0.3% 30 17 11 15 Thailand 11 628 0.4% 15 62 11 567 Trinidad and Tobago 10 30 0.2% 21 10 9 20 Tunisia 26 387 1.3% 53 110 23 276 Turkey 9 452 0.1% 37 265 6 187 Uganda 19 36 0.4% 32 20 14 16 Ukraine 7 85 0.1% 7 6 7 80 Uruguay 10 15 0.1% 13 2 10 13 USA 7 3,770 0.0% 13 714 6 3,055 Venezuela, R. B. de 14 110 0.1% 16 20 13 90 Zambia 12 19 0.3% 17 3 12 16 High income 10 74,537 0.2% 40 65,299 4 9,237 Middle income 15 13,320 0.2% 30 3,800 12 9,520 Low income 16 5,249 0.4% 18 1,398 16 3,852 Source: World Bank staff estimates. Note: TRI is estimated using the most recent available tariff schedules (2005­2006). These data originates from the UNCTAD TRAINS and World Bank WITS databases. TRI is the uniform tariffs that would provide the same level of welfare in the importing country as the existing tariff struc- ture. Deadweight loss (DWL) is measured by comparing the existing TRI to zero tariff. Deadweight loss is in million USD. For a detailed discussion of the methodology used to estimate the TRI see Kee, Nicita, and Olarreaga (2006). Overall Trade Restrictiveness Index (OTRI) and import loss (US$, millions) All Products Agriculture Manufacturing Import Loss Import Loss (Percent of Import Loss Import Loss OTRI (US$, millions) Total Imports) OTRI (US$, millions) OTRI (US$, millions) Albania 7 ­165 ­6% 9 ­37 7 ­129 Algeria 38 ­4,567 ­25% 54 ­1,235 34 ­3,332 Argentina 16 ­4,237 ­15% 19 ­155 16 ­4,083 Australia 10 ­8,239 ­7% 41 ­1,454 8 ­6,785 Bangladesh 28 ­1,327 ­12% 30 ­313 27 ­1,014 Belarus 16 ­1,721 ­10% 34 ­337 13 ­1,384 Bolivia 15 ­209 ­9% 35 ­51 12 ­158 Brazil 22 ­11,185 ­15% 35 ­1,045 21 ­10,140 Brunei 9 ­77 ­6% 19 ­20 7 ­57 Burkina Faso 13 ­126 ­10% 36 ­34 10 ­93 Cameroon 16 ­341 ­12% 22 ­75 14 ­266 Canada 7 ­14,302 ­5% 34 ­3,306 5 ­10,996 Chile 9 ­1,746 ­6% 29 ­355 7 ­1,392 China 11 ­43,411 ­7% 19 ­3,377 11 ­40,034 Colombia 22 ­2,936 ­14% 46 ­709 19 ­2,227 Costa Rica 5 ­301 ­3% 11 ­67 4 ­233 Côte d'Ivoire 29 ­700 ­12% 53 ­309 20 ­391 Egypt, Arab Rep. of 32 ­2,211 ­17% 39 ­614 30 ­1,597 El Salvador 12 ­369 ­8% 16 ­89 11 ­281 Ethiopia 14 ­228 ­8% 11 ­40 14 ­189 European Union 9 ­73,084 ­2% 56 ­26,901 5 ­46,183 Gabon 16 ­127 ­13% 19 ­36 15 ­92 Ghana 14 ­468 ­11% 23 ­156 12 ­312 Guatemala 13 ­900 ­9% 48 ­328 9 ­572 Honduras 5 ­159 ­3% 19 ­81 3 ­79 Hong Kong, China 2 ­2,104 ­1% 17 ­749 1 ­1,355 Iceland 5 ­171 ­3% 19 ­45 4 ­126 India 20 ­15,092 ­10% 99 ­1,849 17 ­13,243 Indonesia 6 ­2,706 ­5% 16 ­652 4 ­2,054 Japan 16 ­38,760 ­8% 75 ­24,447 5 ­14,313 Jordan 23 ­1,327 ­13% 27 ­228 22 ­1,099 Kazakhstan 10 ­474 ­3% 43 ­147 7 ­327 Kenya 9 ­277 ­6% 29 ­99 5 ­178 Korea, Rep. of 10 ­12,078 ­5% 86 ­3,741 4 ­8,337 Lebanon 17 ­939 ­10% 64 ­483 8 ­456 Madagascar 9 ­104 ­9% 9 ­13 9 ­91 Malawi 16 ­141 ­12% 31 ­44 13 ­97 Malaysia 22 ­12,191 ­11% 47 ­1,704 20 ­10,487 Mali 12 ­108 ­11% 27 ­29 10 ­79 Mauritius 14 ­235 ­7% 33 ­94 10 ­141 Mexico 27 ­34,405 ­15% 63 ­5,051 24 ­29,354 Moldova 5 ­74 ­3% 7 ­15 5 ­59 Morocco 39 ­4,484 ­22% 80 ­910 33 ­3,574 New Zealand 14 ­2,249 ­8% 36 ­432 12 ­1,817 Nicaragua 12 ­195 ­8% 43 ­86 7 ­109 Nigeria 26 ­2,295 ­15% 63 ­598 20 ­1,697 Norway 6 ­1,513 ­3% 46 ­1,023 2 ­490 Oman 10 ­657 ­7% 47 ­291 5 ­366 continued Overall Trade Restrictiveness Index (OTRI) and import loss (US$, millions) (continued) All Products Agriculture Manufacturing Import Loss Import Loss (Percent of Import Loss Import Loss OTRI (US$, millions) Total Imports) OTRI (US$, millions) OTRI (US$, millions) Papua N. Guinea 9 ­66 ­5% 31 ­39 4 ­27 Paraguay 14 ­272 ­9% 48 ­65 11 ­207 Peru 18 ­1,483 ­12% 56 ­540 11 ­942 Philippines 14 ­3,891 ­8% 48 ­800 12 ­3,091 Romania 17 ­4,984 ­12% 42 ­638 16 ­4,346 Russian Fed. 26 ­14,221 ­14% 44 ­3,472 22 ­10,749 Rwanda 24 ­39 ­15% 20 ­7 25 ­31 Saudi Arabia 8 ­3,078 ­5% 19 ­621 6 ­2,458 Senegal 33 ­610 ­17% 47 ­242 28 ­368 Singapore 14 ­12,906 ­6% 54 ­1,570 13 ­11,336 South Africa 7 ­3,174 ­6% 12 ­289 7 ­2,885 Sri Lanka 6 ­419 ­5% 16 ­126 5 ­293 Sudan 47 ­998 ­14% 47 ­126 47 ­873 Switzerland 6 ­4,160 ­3% 42 ­1,968 3 ­2,192 Tanzania 40 ­640 ­23% 50 ­107 38 ­533 Thailand 8 ­6,112 ­5% 44 ­1,407 6 ­4,705 Trinidad and Tobago 7 ­306 ­5% 25 ­82 5 ­224 Tunisia 30 ­2,155 ­16% 81 ­338 26 ­1,818 Turkey 11 ­9,022 ­8% 23 ­639 11 ­8,383 Uganda 10 ­159 ­8% 19 ­51 8 ­109 Ukraine 10 ­2,381 ­7% 10 ­175 10 ­2,206 Uruguay 16 ­310 ­8% 50 ­66 13 ­244 USA 9 ­91,807 ­5% 26 ­14,745 8 ­77,062 Venezuela, R. B. de 23 ­1,822 ­8% 58 ­541 16 ­1,281 Zambia 10 ­205 ­8% 31 ­36 8 ­169 High income 8 261,605 46 80,557 6 181,048 Middle income 16 180,270 37 27,083 13 153,186 Low income 16 29,214 30 5,053 13 24,161 Source: World Bank staff estimates. Note: OTRI is estimated using the most recent available tariff schedules (2005­2006) and nontariff measures (about 2001). These data originate from the UNCTAD TRAINS and World Bank WITS databases. The OTRI measures the restrictiveness of a country's own trade policies. It is defined as the uniform tariff that would keep aggregate imports at their observed level. Import loss is calculated by comparing the existing OTRI to zero tariff. Import Loss is in million USD. For a detailed methodology on the estimation of the OTRI see Kee, Nicita, and Olarreaga (2006). The OTRI published here is not directly comparable with the OTRI published in previous Global Monitoring Reports, as the underlining data have been improved. There are 21 fewer countries due to consolidation in the European Union (8) and elimination of 13 countries estimated out of sample (basically without their own data). Market Access-Overall Trade Restrictiveness Index (MA-OTRI) MA-OTRI MA-OTRI Ag. MA-OTRI Mfg. MA-OTRI MA-OTRI Ag. MA-OTRI Mfg. Albania 40 34 41 Mali 10 19 10 Algeria 1 1 Mauritius 51 101 34 Argentina 35 54 7 Mexico 13 35 10 Australia 21 57 6 Moldova 32 45 24 Bangladesh 44 16 46 Morocco 28 49 20 Belarus 10 34 9 New Zealand 45 66 7 Bolivia 12 34 4 Nicaragua 29 30 29 Brazil 18 44 6 Nigeria 12 61 0 Brunei 16 16 Norway 8 36 4 Burkina Faso 23 55 14 Oman 2 12 2 Cameroon 18 54 2 Papua N. Guinea 21 46 4 Canada 6 33 2 Paraguay 13 18 5 Chile 20 35 7 Peru 14 28 9 China 15 23 14 Philippines 13 15 13 Colombia 15 29 5 Romania 22 41 22 Costa Rica 21 26 14 Russian Fed. 8 30 5 Côte d'Ivoire 26 30 3 Rwanda 37 65 4 Egypt, Arab Rep. of 14 30 11 Saudi Arabia 4 4 El Salvador 47 37 48 Senegal 31 37 12 Ethiopia 46 56 10 Singapore 9 9 European Union 25 40 18 South Africa 10 48 3 Gabon 2 0 2 Sri Lanka 34 32 35 Ghana 32 45 2 Sudan 38 72 4 Guatemala 34 25 42 Switzerland 9 91 6 Honduras 41 27 45 Tanzania 29 43 5 Hong Kong, China 24 38 24 Thailand 26 71 15 Iceland 28 40 5 Trinidad and Tobago 3 80 2 India 23 32 21 Tunisia 24 27 24 Indonesia 18 30 16 Turkey 23 38 22 Japan 10 10 Uganda 35 39 7 Jordan 28 38 28 Ukraine 11 34 7 Kazakhstan 9 23 8 Uruguay 36 66 8 Kenya 39 50 13 USA 11 42 8 Korea, Rep. of 14 14 Venezuela, R. B. de 3 9 3 Lebanon 12 26 9 Zambia 23 43 18 Madagascar 22 14 36 High income 15 53 9 Malawi 34 36 11 Middle income 21 39 16 Malaysia 14 24 12 Low income 30 39 19 Source: World Bank staff estimates. Note: MA-OTRI is estimated using the most recent available tariff schedules (2005­2006) and non tariff measures (about 2001). These data originate from the UNCTAD TRAINS and World Bank WITS databases. The MA-OTRI is calculated accounting for tariff preferences. The MA-OTRI measures the restrictiveness of other countries' trade policies on the export bundle of each country. For a detailed methodology on the estimation of the MA-OTRI see Kee, Nicita, and Olarreaga (2006). The MA-OTRI pub- lished here is not directly comparable with the MA-OTRI published in previous Global Monitoring Reports, as the underlining data have been improved. There are 21 fewer countries due to consolidation in the European Union (8) and elimination of 13 countries estimated out of sample (basically without their own data). MA-OTRI Ag. was not estimated for Algeria, Brunei, Indonesia, Rep. of Korea, Saudia Arabia, and Singapore owing to their limited agricultural exports. Net Official Development Assistance (ODA) by DAC and non-DAC Countries 2001 2005 of which: ODA ODA ODA Technical Humanitarian Debt Other Contributions ODA (current (percent (current Co-operation and Food Aid Forgiveness Bilateral to (percent US$ millions) of GNI) US$ millions) Grants Grants Grantsa ODAb Multilaterals of GNI) DAC Donors Austria 633 0.34 1573 150 89 904 57 341 0.52 Belgium 867 0.37 1963 500 124 472 166 655 0.53 Denmark 1634 1.03 2109 115 225 0 901 751 0.81 Finland 389 0.32 902 98 91 150 224 305 0.46 France 4198 0.31 10026 2364 652 3211 678 2787 0.47 Germany 4990 0.27 10082 2865 357 3441 578 2635 0.36 Greece 202 0.17 384 77 27 0 73 178 0.17 Ireland 287 0.33 719 13 85 0 354 237 0.42 Italy 1627 0.15 5091 121 79 1670 360 2821 0.29 Luxembourg 139 0.76 256 4 25 0 147 69 0.82 Netherlands 3173 0.82 5115 609 503 324 2002 1432 0.82 Portugal 268 0.25 377 115 13 3 72 159 0.21 Spain 1737 0.3 3018 483 145 473 658 1155 0.27 Sweden 1666 0.77 3362 140 405 53 1532 1106 0.94 United Kingdom 4579 0.32 10767 845 628 3506 2758 2603 0.47 DAC EU Members, Total 26388 0.33 55745 8498 3448 14207 10559 17236 0.45 Australia 873 0.25 1680 740 324 19 290 231 0.25 Canada 1533 0.22 3756 335 344 455 1450 924 0.34 Japan 9847 0.23 13147 1873 574 3553 3704 2740 0.28 New Zealand 112 0.25 274 41 66 0 102 50 0.27 Norway 1346 0.8 2786 320 412 0 1164 754 0.94 Switzerland 908 0.34 1767 154 329 224 664 367 0.44 United States 11429 0.11 27623 8966 4111 4076 7071 2343 0.22 DAC Members, Total 52435 0.22 106777 20926 9607 22533 25003 24644 0.33 Non-DAC Donors Czech Republic 27 0.05 135 15 19 10 17 71 0.11 Hungary 100 40 61 0.11 Iceland 10 0.13 27 20 7 0.18 Korea, Rep. of 265 0.06 752 80 27 337 289 0.1 Poland 36 0.02 205 48 157 0.07 Saudi Arabia* 490 1700 Other Arab Countries 200 689 633 56 Slovak Republic 8 0.04 56 31 25 0.12 Turkey 64 0.04 601 163 179 134 69 0.17 Other Bilateral Donors 95 665 568 98 Non-DAC Countries, Total 1194 4931 258 225 10 1827 832 a. Debt forgiveness grants are offset in order to avoid double-counting of Debt forgiveness of loans previously counted as ODA. b. Other Bilateral ODA is Bilateral ODA - special purpose grants (technical cooperation, debt forgiveness, food and emergency aid) and administrative costs (not shown). * Saudi Arabia has not yet reported to the DAC. No breakdown by instrument is available. $1700 is an estimate from the DAC. Net Official Development Assistance (ODA) Receipts 2001 2005 of which: ODA ODA Technical Debt Food and ODA ODA (current US$ (current Co­operation Forgiveness Emergency Other per capita (percent millions) US$ millions) Grants Grants Aid Grants ODA (in current US$) of GNI) Afghanistan 405 2775 1132 0 296 1348 38.5 Albania 270 319 98 0 8 214 101.82 3.7 Algeria 224 371 199 38 41 93 11.28 0.4 Angola 283 442 80 0 122 240 27.72 1.7 Argentina 146 100 61 0 1 38 2.57 0.1 Armenia 198 193 85 0 46 62 64.07 3.9 Azerbaijan 232 223 76 0 38 110 26.64 2.0 Bangladesh 1025 1321 191 5 86 1039 9.31 2.1 Belarus 39 54 28 0 5 21 5.50 0.0 Benin 272 349 94 2 10 244 41.36 8.2 Bhutan 61 90 23 0 0 67 98.06 11.0 Bolivia 734 583 193 4 36 351 63.48 6.5 Bosnia and Herzegovina 639 546 135 4 48 360 139.77 5.7 Botswana 29 71 53 1 0 17 40.17 0.8 Brazil 229 192 192 0 2 ­2 1.03 0.0 Burkina Faso 390 660 97 4 25 534 49.86 12.8 Burundi 137 365 40 4 151 170 48.36 46.8 Cambodia 418 538 181 0 20 336 38.22 10.4 Cameroon 486 414 152 148 13 100 25.35 2.5 Cape Verde 77 161 44 0 10 107 316.88 17.1 Central African Republic 66 95 39 2 6 50 23.60 7.0 Chad 185 380 47 2 111 220 38.96 8.6 Chile 57 152 60 0 2 89 9.31 0.1 China 1473 1757 859 0 37 861 1.35 0.1 Colombia 380 511 448 0 53 11 11.21 0.4 Comoros 27 25 13 0 3 9 42.02 6.6 Congo, Dem. Rep. of 243 1828 189 143 332 1164 31.76 27.5 Congo, Rep. of 74 1449 32 1234 26 156 362.31 36.8 Costa Rica 2 30 29 0 1 ­0 6.82 0.2 Côte d'Ivoire 169 119 77 11 58 ­27 6.56 0.8 Croatia 113 125 52 0 8 66 28.22 0.3 Cuba 54 88 33 0 9 46 7.79 Djibouti 58 79 32 0 5 41 99.11 10.1 Dominican Republic 107 77 65 0 2 11 8.66 0.3 Ecuador 173 210 139 ­42 8 104 15.84 0.6 Egypt, Arab Rep. of 1256 926 386 148 2 390 12.51 1.0 El Salvador 237 199 86 0 52 61 28.98 1.2 Equatorial Guinea 13 39 17 3 0 20 77.45 0.0 Eritrea 281 355 26 0 171 158 80.69 36.3 Ethiopia 1104 1937 217 34 678 1008 27.19 17.4 Fiji 26 64 29 0 0 35 75.45 2.3 Gabon 9 54 46 0 1 8 38.93 0.7 Gambia, The 54 58 10 0 3 45 38.33 13.1 Georgia 300 310 132 0 49 129 69.23 4.7 Ghana 641 1120 129 66 22 903 50.65 10.6 Guatemala 227 254 97 0 62 95 20.13 0.8 Guinea 281 182 63 5 35 79 19.37 6.9 Guinea-Bissau 59 79 16 1 17 45 49.88 27.3 (continued) Net Official Development Assistance (ODA) Receipts (continued) 2001 2005 of which: ODA ODA Technical Debt Food and ODA ODA (current US$ (current Co­operation Forgiveness Emergency Other per capita (percent millions) US$ millions) Grants Grants Aid Grants ODA (in current US$) of GNI) Guyana 97 137 28 1 3 105 182.08 18.6 Haiti 171 515 143 0 152 220 60.39 12.1 Honduras 679 681 108 158 48 367 94.50 8.2 India 1701 1724 363 0 104 1258 1.58 0.2 Indonesia 1467 2524 476 10 667 1371 11.44 0.9 Iran, Islamic Rep. of 114 104 75 0 24 5 1.54 0.1 Iraq 121 21654 1646 13920 484 5603 Jamaica 54 36 44 9 10 ­27 13.45 0.4 Jordan 449 622 126 0 2 494 114.95 4.7 Kazakhstan 148 229 95 0 7 128 15.13 0.5 Kenya 462 768 206 3 76 483 22.43 4.3 Korea, Dem. Rep. of 118 81 10 0 46 26 3.61 Kyrgyz Republic 189 268 93 4 15 156 52.07 11.4 Lao PDR 245 296 82 0 7 207 49.92 11.2 Lebanon 243 243 105 0 35 103 67.93 1.2 Lesotho 56 69 14 0 6 49 38.34 3.8 Liberia 38 236 55 0 130 52 71.93 54.1 Libya 7 24 11 0 1 13 4.18 0.0 Macedonia, FYR 247 230 101 0 3 127 113.23 4.0 Madagascar 374 929 96 294 41 498 49.94 18.7 Malawi 404 575 101 5 68 401 44.66 28.4 Malaysia 27 32 53 0 2 ­23 1.25 0.0 Mali 351 692 130 5 28 529 51.15 14.1 Mauritania 267 190 44 1 45 101 62.04 10.4 Mauritius 21 32 24 0 0 8 25.58 0.5 Mexico 73 189 149 0 1 39 1.84 0.0 Moldova 122 192 60 0 57 75 45.59 5.8 Mongolia 211 212 86 0 15 111 82.95 11.6 Morocco 518 652 386 0 6 260 21.61 1.3 Mozambique 931 1286 205 4 39 1038 64.97 20.8 Myanmar 126 145 41 1 33 70 2.86 Namibia 109 123 66 0 0 57 60.73 2.0 Nepal 391 428 128 5 34 261 15.77 5.8 Nicaragua 930 740 111 131 44 454 134.88 15.2 Niger 256 515 65 23 78 350 36.93 15.2 Nigeria 168 6437 234 5548 25 631 48.94 7.4 Oman 1 31 11 0 0 20 11.95 0.0 Pakistan 1942 1667 201 0 698 767 10.70 1.5 Panama 28 20 27 0 0 ­7 6.05 0.1 Papua New Guinea 203 266 161 0 4 102 45.21 6.6 Paraguay 61 51 46 0 0 5 8.30 0.6 Peru 449 398 228 7 12 152 14.22 0.5 Philippines 572 562 247 0 36 279 6.76 0.5 Rwanda 299 576 114 11 51 400 63.73 27.4 São Tomé and Principe 38 32 11 0 1 20 203.81 58.6 Senegal 413 689 224 85 11 370 59.12 8.4 Serbia and Montenegro 1306 1132 326 201 81 524 138.54 4.3 (continued) Net Official Development Assistance (ODA) Receipts (continued) 2001 2005 of which: ODA ODA Technical Debt Food and ODA ODA (current US$ (current Co­operation Forgiveness Emergency Other per capita (percent millions) US$ millions) Grants Grants Aid Grants ODA (in current US$) of GNI) Sierra Leone 343 343 31 4 46 262 62.15 29.6 Solomon Islands 59 198 143 0 1 54 414.95 70.5 Somalia 148 236 9 0 165 62 28.73 South Africa 428 700 270 0 2 428 15.49 0.3 Sri Lanka 313 1189 97 0 400 692 60.73 5.1 Sudan 181 1829 131 0 1267 431 50.47 7.1 Swaziland 29 46 5 0 4 37 40.70 1.7 Syrian Arab Republic 153 78 77 0 7 ­7 4.09 0.3 Tajikistan 169 241 57 0 45 139 37.09 10.8 Tanzania 1269 1505 196 1 47 1261 39.27 12.5 Thailand 281 ­171 149 0 45 ­364 ­2.66 ­0.1 Timor-Leste 194 185 74 0 5 105 189.37 26.7 Togo 43 87 34 2 5 47 14.11 4.0 Tunisia 377 377 146 0 2 229 37.57 1.4 Turkey 169 464 177 0 76 212 6.39 0.1 Turkmenistan 72 28 24 0 0 4 5.84 0.0 Uganda 790 1198 235 1 187 775 41.58 14.0 Ukraine 519 410 228 0 12 170 8.69 0.0 Uruguay 15 15 15 ­9 0 8 4.22 0.1 Uzbekistan 153 172 76 0 3 94 6.48 1.3 Venezuela, R. B. de 44 49 36 0 14 ­0 1.83 0.0 Vietnam 1450 1905 316 1 13 1575 22.96 3.7 West Bank and Gaza 869 1102 208 0 153 740 303.81 Yemen, Rep. of 458 336 59 1 15 261 16.02 2.6 Zambia 349 945 181 340 37 388 80.99 14.2 Zimbabwe 162 368 61 0 84 223 28.26 11.6 East Asia & Pacific 7390 9497 3198 12 1147 5140 Europe & Central Asia 5230 5732 2036 209 584 2902 Latin America & Caribbean 5868 6309 2783 268 632 2626 Middle East & North Africa 4905 26947 3550 14108 817 8472 South Asia 5861 9261 2144 9 1655 5452 Sub-Saharan Africa 13981 32620 4863 7985 4731 15041 Unspecified by Region 8758 16008 4140 0 1848 10021 Fragile States 6198 21477 3555 6957 3483 7482 Low-income Countries 20878 40353 6979 6615 5643 21115 Lower-Middle-income Countries 18322 43146 8879 15971 2823 15473 Upper-Middle-income Countries 1746 2776 1353 ­5 150 1278 Middle-income Countries 20580 46913 10612 15966 3082 17254 Unallocated 10535 19106 5123 10 2690 11284 Developing Countries, Total 51993 106372 22714 22591 11415 49653 Source: OECD DAC Database Regional totals do not include ODA that is unspecufued by region. The total for developing countries includes ODA that is unallocated by country or income group. Income group totals reflect the classifications used in the World Development Indicators which are different than the classifications used elsewhere in this report. Measures of Gender Performance Primary Educationa Child Mortalityb Labor Force Participationc Between 20 to 24 years old Between 25 to 49 years old Female-to- Female-to- Female-to- Female male ratio: Female-to- Female male ratio: Female male ratio: primary Primary Female male ratio: labor Labor labor Labor completion completion under-5 under-5 force force force force rate rate mortality rate mortality rate participation participation participation participation 2001­2004d 2001­2004d 2000­2005d 2000­2005d 2000­2005d 2000­2005a 2000­2005d 2000­2005d Afghanistan .. .. 255 1.02 21.0 0.27 24.5 0.27 Albania 99.2 1.00 31.3 0.86 47.4 0.99 64.7 0.83 Algeria 94.5 1.01 39.6 0.95 .. .. .. .. Angola .. .. 230.8 0.89 61.4 .. 81.7 .. Argentina 103.1 1.05 15.1 0.77 44.7 0.63 60.9 0.64 Armenia 107.6 1.02 32.5 0.87 40.1 0.71 57.3 0.71 Azerbaijan 95.2 0.98 86.5 0.92 44.3 0.66 62.4 0.69 Bangladesh 78.8 1.06 79.5 1.01 6.9 0.01 10.4 0.11 Belarus 98.7 0.96 14.6 0.71 62.3 1.03 90.3 0.97 Benin 38.3 0.65 158.8 0.97 73.6 1.19 88.8 0.93 Bhutan .. .. 81.7 0.96 69.3 0.96 77.3 0.83 Bolivia 97.9 0.96 67.4 0.88 59.2 0.76 77.7 0.80 Bosnia and Herzegovina .. .. 14 0.81 54.0 0.65 52.5 0.56 Botswana 94.1 1.05 100.2 0.90 .. .. .. .. Brazil 111.4 1.01 30.5 0.76 63.0 0.72 66.7 0.71 Bulgaria 97.4 0.98 14.9 0.81 63.8 0.82 83.0 0.93 Burkina Faso 25.3 0.76 190.7 0.95 79.0 0.89 85.9 0.88 Burundi 26.9 0.68 175.8 0.89 80.9 .. 96.3 .. Cambodia 77.8 0.91 134.4 0.92 79.9 0.92 81.9 0.86 Cameroon 57.6 0.84 155.3 0.91 57.5 0.82 78.1 0.83 Cape Verde 95.3 1.00 25.2 0.54 62.8 0.76 70.3 0.75 Central African Republic .. .. 159.3 0.83 .. .. .. .. Chad 18.4 0.45 191.3 0.89 .. .. .. .. Chile 94.6 0.98 8.5 0.78 47.2 0.69 56.5 0.60 China .. .. 47.1 1.36 .. .. .. .. Colombia 96.4 1.04 30.1 0.85 64.2 0.75 68.3 0.71 Comoros 49.0 0.95 68.3 0.80 .. .. .. .. Congo, Dem. Rep. of .. .. 200.7 0.90 .. .. .. .. Congo, Rep. of 63.2 0.91 96.5 0.81 .. .. .. .. Costa Rica 93.5 1.03 10.7 0.79 53.8 0.63 54.0 0.56 Côte d'Ivoire 33.7 0.65 180.1 0.91 50.7 0.76 65.0 0.70 Croatia 90.8 0.99 7.6 0.88 51.3 0.84 78.2 0.90 Cuba 92.1 0.99 6.9 0.81 .. .. .. .. Czech Republic 104.0 1.00 5.7 0.93 .. .. .. .. Djibouti 25.0 0.75 131.2 0.88 61.3 .. 42.2 .. Dominican Republic 93.4 1.06 45.5 0.82 49.5 0.57 60.3 0.64 Ecuador 101.1 1.01 25.1 0.73 48.5 0.61 59.0 0.62 Egypt, Arab Rep. of 92.8 0.96 38.9 0.83 52.2 .. 53.9 .. El Salvador 86.2 1.01 31.2 0.82 45.3 0.57 59.8 0.64 Equatorial Guinea 47.0 0.87 172.6 0.91 .. .. .. .. Eritrea 34.4 0.68 89.5 0.91 .. .. .. .. Estonia 100.2 0.98 9.2 0.65 56.6 0.73 84.9 0.89 Ethiopia 48.8 0.80 163.7 0.91 56.5 0.64 59.9 0.62 Gabon 67.6 1.04 90.1 0.90 .. .. .. .. Gambia, The .. .. 121.8 0.90 47.0 .. 66.0 .. (continued) Measures of Gender Performance (continued) Primary Educationa Child Mortalityb Labor Force Participationc Between 20 to 24 years old Between 25 to 49 years old Female-to- Female-to- Female-to- Female male ratio: Female-to- Female male ratio: Female male ratio: primary Primary Female male ratio: labor Labor labor Labor completion completion under-5 under-5 force force force force rate rate mortality rate mortality rate participation participation participation participation 2001­2004d 2001­2004d 2000­2005d 2000­2005d 2000­2005d 2000­2005a 2000­2005d 2000­2005d Georgia 87.5 1.04 38.6 0.81 35.9 .. 70.1 .. Germany .. .. .. .. .. .. .. .. Ghana 68.7 0.91 100.2 0.97 67.2 .. 93.0 .. Guatemala 65.4 0.87 45.5 0.79 34.2 0.40 36.9 0.39 Guinea 38.8 0.67 167.2 1.02 79.5 .. 89.3 .. Guinea-Bissau 19.3 0.56 199.5 0.90 .. .. .. .. Guyana 91.5 0.92 57.2 0.73 42.8 .. 51.2 .. Haiti .. .. 102 0.87 48.7 0.81 70.6 0.82 Honduras 81.8 1.06 43 0.81 42.4 0.48 51.0 0.54 Hungary 95.9 1.01 9.4 0.80 61.8 0.94 86.2 0.98 India 83.9 0.90 102 1.07 28.1 0.34 37.1 0.38 Indonesia 102.1 1.01 47.2 0.78 49.3 0.60 51.0 0.52 Iran, Islamic Rep. of 96.9 1.05 39.4 1.01 .. .. .. .. Iraq 62.7 0.74 118.4 0.92 .. .. .. .. Jamaica 85.7 1.03 19.5 0.89 39.4 0.69 50.4 0.76 Jordan 96.3 0.99 25.6 0.93 28.4 0.37 24.3 0.26 Kazakhstan 109.5 0.99 62.8 0.69 65.6 0.86 88.2 0.93 Kenya 90.3 0.97 110 0.88 48.8 .. 57.0 .. Korea, Dem. Rep. of .. .. 54.9 0.87 .. .. .. .. Kyrgyz Republic 93.4 1.01 60.3 0.84 55.5 0.81 84.3 0.92 Lao PDR 69.6 0.89 137.1 0.95 .. .. .. .. Latvia 92.0 0.99 12.9 0.90 48.6 0.72 85.5 0.96 Lebanon 96.5 1.05 21.2 0.67 .. .. .. .. Lesotho 82.0 1.37 116.2 0.90 64.6 0.94 73.0 0.81 Liberia .. .. 214.8 0.92 .. .. .. .. Libya .. .. 21.1 1.00 .. .. .. .. Lithuania 97.2 0.99 9.5 0.69 94.2 0.94 97.1 0.97 Macedonia, FYR 97.3 1.02 17.1 0.88 .. .. .. .. Madagascar 46.0 1.03 125.8 0.92 54.3 0.81 54.0 0.90 Malawi 57.2 0.96 179.1 0.94 16.2 0.67 18.0 0.52 Malaysia 91.1 1.00 11.4 0.78 .. .. .. .. Mali 29.6 0.51 217.2 0.97 .. .. .. .. Mauritania 41.0 0.91 149.6 0.92 22.4 0.52 27.8 0.44 Mauritius 99.4 1.04 15.3 0.76 .. .. .. .. Mexico 99.6 1.02 21.8 0.80 42.4 0.53 53.0 0.56 Moldova 91.5 1.01 28.5 0.83 75.7 0.98 79.2 1.01 Mongolia 96.3 1.02 83.2 0.95 47.7 0.84 75.9 0.95 Morocco 72.0 0.91 38.6 0.72 30.3 .. 35.3 .. Mozambique 23.4 0.68 173 0.91 77.8 .. 86.6 .. Myanmar 79.1 1.03 102.3 0.84 .. .. .. .. Namibia 85.3 1.12 74.5 0.92 48.6 .. 66.2 .. Nepal 69.5 0.87 91 1.07 70.9 .. 82.8 .. Nicaragua 77.0 1.01 35.2 0.79 38.6 0.47 53.2 0.58 Niger 20.0 0.67 266.5 1.02 30.9 0.51 47.4 0.54 Nigeria 68.2 0.83 196.6 0.96 33.0 1.03 63.7 0.72 (continued) Measures of Gender Performance (continued) Primary Educationa Child Mortalityb Labor Force Participationc Between 20 to 24 years old Between 25 to 49 years old Female-to- Female-to- Female-to- Female male ratio: Female-to- Female male ratio: Female male ratio: primary Primary Female male ratio: labor Labor labor Labor completion completion under-5 under-5 force force force force rate rate mortality rate mortality rate participation participation participation participation 2001­2004d 2001­2004d 2000­2005d 2000­2005d 2000­2005d 2000­2005a 2000­2005d 2000­2005d Oman 89.6 0.97 16.5 0.87 .. .. .. .. Pakistan .. .. 118.6 1.08 26.6 0.33 30.4 0.33 Panama 96.9 1.01 22.9 0.75 46.4 0.55 57.3 0.60 Papua New Guinea 49.5 0.86 103 1.01 .. .. .. .. Paraguay 91.1 1.01 39.5 0.78 58.2 0.68 64.1 0.67 Peru 98.9 0.99 46.9 0.82 62.1 0.78 74.1 0.78 Philippines 99.9 1.07 28 0.71 .. .. .. .. Poland .. .. 9.7 0.87 60.2 0.95 65.2 0.99 Romania 92.8 0.99 18.7 0.73 65.8 0.99 71.3 0.98 Russian Federation .. .. 18.7 0.77 72.2 0.94 100.0 1.00 Rwanda 36.8 0.97 178.4 0.89 85.3 .. 95.1 .. Senegal 41.7 0.85 129.7 0.96 28.2 .. 52.6 .. Serbia and Montenegro 95.9 0.99 13.8 0.82 49.2 0.81 74.5 0.82 Sierra Leone .. .. 277.4 0.92 57.4 1.19 78.4 0.95 Slovak Republic 99.6 1.00 9.4 0.90 .. .. .. .. Somalia .. .. 205.8 0.95 .. .. .. .. South Africa 97.5 1.04 69.3 0.89 46.1 0.87 67.9 0.80 Spain .. .. .. .. .. .. .. .. Sri Lanka .. .. 14.8 0.60 55.3 0.64 47.6 0.50 Sudan 44.5 0.84 112.7 0.90 .. .. .. .. Swaziland 64.2 1.10 134.1 0.88 47.1 0.75 55.2 0.67 Syrian Arab Republic 104.3 0.96 17.8 0.74 .. .. .. .. Tajikistan 89.6 0.95 109.7 0.90 44.1 0.77 54.3 0.71 Tanzania 53.1 0.96 155.7 0.91 74.2 0.85 79.6 0.82 Thailand .. .. 18.7 0.62 70.6 0.87 86.2 0.88 Timor-Leste .. .. 130.2 0.94 42.0 0.55 40.8 0.43 Togo 55.0 0.71 128.7 0.88 .. .. .. .. Trinidad and Tobago 95.3 1.02 15.9 0.74 41.3 .. 46.8 .. Tunisia 97.5 1.01 23 0.85 .. .. .. .. Turkey 82.3 0.88 43.7 0.80 40.7 0.50 38.8 0.41 Turkmenistan .. .. 88.7 0.81 .. .. .. .. Uganda 53.3 0.88 133 0.92 82.5 1.06 92.8 0.96 Ukraine .. .. 14.8 0.72 52.3 0.74 81.8 0.88 Uruguay 92.9 1.04 12.6 0.70 66.6 0.79 77.8 0.81 Uzbekistan 96.5 1.00 63.5 0.84 .. .. .. .. Venezuela, R. B. de 92.0 1.06 26.8 0.85 46.9 0.58 61.7 0.65 Vietnam 97.6 0.94 33 0.75 79.8 0.97 94.0 0.96 West Bank and Gaza 98.7 1.01 21.8 0.81 .. .. .. .. Yemen, Rep. of 45.6 0.58 91 0.92 13.5 .. 19.3 .. Zambia 61.5 0.87 164.9 0.91 54.3 0.97 68.6 0.75 Zimbabwe 78.6 0.96 110.1 0.89 .. .. .. .. Sources: a. 2006 World Development Indicators database. b. 2004 World Population Prospects. c. Household surveys various year. d. Data are for the most recent year available. Figures in italics refer to periods other than those specified. .. Not available Environmental Benefits Statement The 2007 Global Monitoring Report examines the responsibilities and accountability This year's report focuses on gender equality and the empowerment of women, both of donor countries, developing countries, and the international financial institutions central development issues. Gender equality is intrinsically fair, and empowering to support attainment of the eight Millennium Development Goals (MDGs), as women improves both economic performance and progress in other development agreed by 189 countries in 2000, and monitors recent performance against the goals--including education, nutrition, and reducing child mortality. Some areas have seen rapid progress, such as achieving educational parity for girls in school. MDG targets. But in other dimensions--including political representation and nonagricultural The report examines progress toward the MDGs: while halving of extreme poverty employment--performance falls short. Strengthening performance will require is on track for 2015 globally, there is less progress in the human development realistic goals, strong leadership, technical expertise, and financing. MDGs (education, health, access to sanitation, etc.), and regional differences are To advance the MDG agenda, the international community needs to do more: sharp--both Sub-Saharan Africa and South Asia tend to lag further behind. The donors need to provide more and better quality assistance; developing countries largest gap in meeting development goals is in fragile states--countries with weak need to adopt sound, sequenced development strategies; international institutions governance and capacity--posing major developmental challenges highlighted in should provide more technical support to strengthen strategies; and all need to work the report. toward a more coherent and efficient "aid architecture." ISBN 0-8213-6975-X