79489 AN EVALUATION OF THE 2007 STRATEGY AND IMPLEMENTATION PLAN World Bank Country-Level Engagen1ent on Govern'ance and Anticorruotion The Worlc! Bank Group WORKING FOR A WORLD FREE OF POVERTY he World Bank Group consists of five institutions- T the International Bank for Reconstruction and De- velopment (IBRD), the Internationa l Finance Corporation (IFC), the International Development Association (IDA), the Multilateral Investment Guarantee Agency (MIGA). and the International Centre for the Settlement of Invest- ment Disputes (ICSID). Its mission is to fight poverty for lasting results and to help people help themse lves and their environment by providing resources, sharing knowl- edge, building capacity, and forging partnerships in the public and private sectors. The Independent Eval\,.lation Group IMPROVING DEVELOPMENT RESULTS THROUGH EXCELLENCE IN EVALUATION he Independent Evaluation Group (lEG) is an T independent unit w1th1n the World Bank Group. It reports directly to the Board of Executive Directors, which oversees lEG's work through its Committee on Develop- ment Effectiveness. lEG is charged with evaluating the activities of the World Bank (the Internationa l Bank for Reconstruction and Development and the International Development Association), the work of the International Finance Corporation in private sector development, and the guarantee projects and services of the Multilateral In vestment Guarantee Agency. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the resu lts of the Bank Group's work, and to provide accountability in the achievement of its objectives. It also improves Bank Group work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings . .. l IEG WORLD BANK INDEPENDENT EVAlUATION GROUP IFC MIGA I AN EVALUATION OF THE 2007 STRATEGY AND IMPLEMENTATION PLAN World Bank Country-Level Engagement on Governance and Anticorruption ©2011 The Independent Evaluation Group The World Bank Group 1818 H Street NW Washington DC 20433 Telephone: 202-458-4497 Internet: http:/ /ieg.worldbankgroup.org E-mail: ieg@worldbank.org All rights reserved 1 2 3 4 5 14 13 12 11 This volume, except for the "Management Response;' "Statement of the External Expert Panel;' and "The Chairper- son's Comments;' is a product of the staff of the Independent Evaluation Group (lEG) of the World Bank Group. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. This volume does not support any general inferences beyond the scope of the evaluation, including any inferences about the World Bank Group's past, current, or prospective overall performance. lEG does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank Group or lEG 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. lEG encourages dissemination of its work and permits this document to be copied or otherwise transmitted, with appropriate credit given to lEG as the authoring body. How to cite this report: lEG (Independent Evaluation Group). 2011. World Bank Country -Level Engagement on Governance and Anticorrup- tion: An Evaluation of the 2007 Strategy and Implem entation Plan. Washington, DC: Independent Evaluation Group, The World Bank Group. Cover: World flags at the Opening Plenary Session of the World Bank/International Monetary Fund Annual Meet- ings 2009. Photo courtesy of Simone D. McCourtie /The World Bank. ISBN 13:978-1 -60244-195-8 ISBN 10: 1-60244-195-2 Independent Evaluation Group Strategy, Communication, and Learning E-mail: ieg@worldbank.org Telephone: 202-458-4497 Facsimile: 202-522-3125 ( ) Printed on Recycled Paper Contents Evaluation Managers • Vinod Thomas Director-General, Evaluation • Ali Khadr Abbreviations .................................................................................................................... vi Senior Manager, Evaluation, Country, Corporate and Global Acknowledgments ......................................................................................................... viii Evaluations Foreword ..............................................................................................................................ix • Navin Girishankar Task Manager Overview ...............................................................................................................................x Management Response ................................................................................................................................................... xx Chairperson's Comments: Committee on Development Effectiveness (CODE) .....................................xxviii Statement of the External Advisory Panel .............................................................................................................. xxix 1. World Bank Engagement on Governance and Anticorruption: A Historical Summary ..................... 1 Objectives of the Evaluation .... . .. . ...... . .. .. 1 Organization of the Report ....... . .. .. . 1 Evolution of the Bank 's Engagement on Governance and Anticorruption ...... . .. .. .. . 2 Salient Features of 2007 GAC Strategy Implementation Plan .... 5 2. Design of the Evaluation .......................................................................................................................................... 9 Framework ........... . .. . .. .. .. . .. .. ...... .. ....... 9 Scope of the Evaluation ....... . ..... . ... . ... .. . .... . ...... . ... . . . .. . . . . .. . . . ... ... .... .. ..... . . .. 11 Main Questions . .. .. ..... .. .... 11 Methods of Analysis........ . .. .... 12 3. Relevance of 2007 GAC Strategy and Implementation Plan ..................................................................... 15 Why and How Governance Matters-Recent Currents in the Literature . . .... . .. . . .. . .. .. 15 Relevance of 2007 GAC Strategy Objectives . . . . . . . . . . . . . . . . . . . . . ... . ... .. ... . . . . .. . .. .. . . . .. .. 17 Appropriateness of Implementation Plan Design ..... 21 4. Incentives and Institutional Arrangements .....................................................................................................25 Financing GAC Implementation ..... 25 Strategic Staffing ... . . .30 Coordination and Accountability .............. . . ..... ... ..... . . . .. ...... . . . ................. 32 5. Implementation of the 2007 GAC Strategy ......................................................................................................33 Adherence to Key Operational Controls ...... .. . . . .. .. .. ......... .. .... .33 Guidance and Tools ... 34 Delivery of Support .. .. .. 38 6. Effectiveness: Is the Bank More Responsive to GAC lssues? .......................................................................41 Achievement of Implementation Plan Objectives: A Summary . .. .... . .. ................ 41 GAC Elements in Country Strategies and Programs .. .. 44 GAC Elements in Sectors and Themes .............. .47 GAC Elements in Projects ..................... .. ....... .. . . ... . .... . ..... .. .... .. .. .. ........ .. .. .. ......... .. .. . .. . ............. . . . . .47 Contents iii Bank Analytical Work .......... . . .. . .......... . ...... . ... . ...... .. .. . .. . . ... . .. . ....... . .. .. . . . .......... .. . . . .... . . . . .. .. .. •.. . . . . . . .53 Staff Motivation and Attitudes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 56 7. Early Outcomes ....................................................................................................................................,..................... 59 Governance at the Country Level ............. . . . . . .59 GAC Entry Points in Sectors and Themes. . .. . .. 60 Service Delivery Sectors .................. .. ..... ... .... .. . ..... .. ...... . .......... . . . . . ...... .. .. . . .. . ..... .. . . ....... . ... . . . . .. 61 Investment Cl imate . . .... . ... . .. . ... . ... . ... . .. 63 Project Fiduciary Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . • . . . . . . . .. 63 8. Conclusions and Recommendations ..................................................................................................................65 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 65 Recommendations ........ . ... . . . . ... . ... . ... . .. . ... . ... . .. . ... . ......... . ................. . ......... . ... . . ...... 66 Appendixes Appendix A Methods and Data Set Description ...............................................................................................69 Appendix B Resourcing of GAC Strategy-Funding and Staffing .............................................................. 81 Appendix C Summary Findings on GAC Responsiveness .............................................................................. 85 Appendix D Detailed Desk Review Findings for Country Programs and Projects .............................. 103 Appendix E Econometric Analysis ...................................................................................................................... 113 Appendix F lEG GAC Staff Survey-Summary Results ................................................................................ 137 Appendix G Persons Interviewed ......................................................................................................................... 143 References . ... . ............ . ............... . .... . ..... . ....... ... . ............. ............ ... 153 Endnotes ............... .. ... .. ... .. ... .. ....... . ....... ..... .... . ...... . .. ... ... .. ......... .. 157 Boxes Box 1.1 The Multiple Objectives and Guiding Prin ciples of th e 2007 GAC Strategy .. ... .. . .. .. .. . . . . . . . .. 6 Box 3. 1 Governan ce and Anticorruption According to Bank Presidents Wolfowitz (2006) and Wolfen so hn (1996) .. . .... . ... . .... . . .. .. . .. .... .... .. ................ ... ... ... . . . ....... 19 Box4.1 Earlier Reviews of the Bank's Orga nizationa l Setup to do Governance Work . . . .... .. ......... . . 31 Box 5.1 Gu idance Material s and Too ls Developed und er the GAC Im plementation Plan, 2007- 10 ... . . .. 35 Box 5.2 Select Donor Methods to Appl ied Po litica l Analysis, Mid-2000s ...... . .......... . ... .. ..... . . . .. 37 Box6.1 GAC -Responsiveness of the Bank in CGAC and GPF Window Countries .... .. .. . ... . . . .. . ... .. .. 43 Box 6.2 Bank and Donor Support for Institutiona l Ca pa city Bu ilding ........... .... . ................ . . .. 45 Box 6.3 Risk Manag em ent Intensity and Implicatio ns fo r Operation al Design .. . . ... . . . ... . ...... ... ... . 52 Box 6.4 Some Th emes from Po litica l Eco nomic Analyses .... ... ............ ... . .. .... . . . .. .. . . .. . .. .. .. 55 Box 7. 1 Co mbining the Demand and Supply Side in Loca l Governance Initiatives .. ... ......... .. ...... 62 Box 7.2 Using an Independent Procurement Agent on Bank Projects in Ca mbodia . ... .... . . . . .. ....... 64 iv World Bank Country-Level Engagement on Governance and Anticorruption Tables Table 3.1 Relevance of GAC Issues to Client Countries (Percent of Countries) ........... . . . . . . .... .. . .. ... 20 Table 4. 1 Ba nk Spending on Governance Wo rk-FY04-1 0 ($ M ill ion) .................................... 25 Tab le 4.2 Incremental Bank Budget Fund ing of Operational Un its by Purpose, FY08- 09 ($ Million ........ 26 Tab le 4.3 Incremental Bank Budget Fund ing for GAC - FY08- 11 ($ Mill ion) ... .. .... . .... .. .. ... .. . ...... 26 Table 4.4 Incremental Bank Budget Fund ing by Region-FY08-11 ($ Mil lion) ............. . .............. 27 Table 6.1 Use of Pol itica l Analysis in Projects (Pre- and Post-GAC Averages) .. . .. . . . ......... . ....... . . ... 48 Table 6.2 Qua lity of Political Economic Analysis in Bank Economic and Sector Work .......... . .......... 53 Table 6.3 Staff Assessment of GAC Competencies and Achievement of Resu lts ...... . .... . . .. ...... . . ... 56 Figures Figure 1.1 Bank Support for Governance-Related Themes, FY 1990-201 0: Number of Projects and Lend ing Volumes .. ............. .. ............................ . ................... .... . .... 4 Figure 2.1 GAC Resu lts Cha in ... . ... . ... . .... ... .. ... ... . . .. ... ....... . .. .. .... .. ... .. ..... . .. ..... ... ..... 9 Figure 3. 1 Strengthening Domestic Accountabi lity Systems .. ... . . .. .... . . . ... .. ... . ...... .. ..... .... . . . . 17 Figure 3.2 Were Alternative Approaches to Designing the GAC Considered? .............................. 23 Figure 4.1 Changes in Overall Regional Bank Budget Expenditures on Governance, FY08- 10 ... . ..... .. .. 27 Figure 4.2 All ocation of GPF Grants by World Bank Sector Units . ..... . ......... .. ............. . ........... 28 Figure 4.3 Nu mber of GPF Grants by Expected Outcomes, Operational Activities, and GAC Elements .. . ... 29 Fig ure 5. 1 Staff Perceptions of Relevance and Use of Gu idance Materials and Too ls "to a Great Extent"..... 39 Figure 6. 1 Country Programs and Projects Addressing GAC Issues "to a Great Extent;' Pre- and Post-GAC Periods .. ... .. ... .. . . . ... .. ... .. ............ . .... . .... .. .. . ....... . ......... .. . .... 42 Figure 6.2 Institutional Ana lysis and Project "Fit" to Governance Real ities .. . ............................... 49 Figure 6.3 Demand-Side Measures in Pre- and Post-GAC Projects ......................................... 51 Figure 7.1 Achievement of Project Objectives by GAC Entry Point in lEG Sample Projects, FY04- 10 ........ 60 Contents v Abbreviations AML/CFT Anti-money laundering and combating the financing of terrorism BB Bank budget BETF Bank-executed trust fund CAE Country Assistance Evaluation CAS Country Assistance Strategy CASCR Country Assistance Strategy Completion Report CEM Country Economic Memorandum (World Bank) CFAA Country Financial Accountability Assessment (World Bank} CGAC Country governance and anticorruption CODE Committee on Development Effectiveness CPAR Country Procurement Assessment Report CPE Country Program Evaluation CPIA Country Policy and Institutional Assessment (World Bank) cso Civil society organization CoST Construction Sector Transparency Initiative DANIDA Danish International Development Agency DEC Development Economics Vice Presidency (World Bank} DIR Detailed Implementation Review DFID Department for International Development (UK) DPL Development policy loan EAP East Asia and the Pacific ECA Europe and Central Asia EITI Extractive Industries Transparency Initiative ESW Economic and sector work FY Fiscal year GAC Governance and anticorruption GFR Grant Funding Request GNI Gross national income GPF Governance Partnership Facility GTZ German Agency for Technical Cooperation (Deutsche Gesellschaft fur Technische Zusammenarbeit) IBRD International Bank for Reconstruction and Development ICR Implementation Completion Report IDA International Development Association lEG Independent Evaluation Group IP Implementation plan IPA Independent Procurement Agency ISR Implementation Status and Results Report MeTA Medicines Transparency Alliance NCB National competitive bidding NGO Nongovernmental organization OECD-DAC Development Assistance Committee of the Organisation for Economic Co-operation and Development OPCS Operations Policy and Country Services (World Bank) ORAF Operational Risk Assessment Framework PEA Political economy analysis PECoP Political Economy Community of Practice vi World Bank Country-Level Engagement on Governance and Anticorruption PER Public Expenditure Review PFM Public financial management PPAR Project Performance Assessment Report PREM Poverty Reduction and Economic Management (World Bank) PRS Poverty Reduction Strategy PRSP Poverty reduction strategy paper PSIA Poverty and Social Impact Assessment QAG Quality Assurance Group RETF Recipient-executed trust fund SAR South Asia SIDA Swedish International Development Cooperation Agency StAR Stolen Asset Recovery SWAp Sectorwide approach TA Technical assistance ucs Use of Country Systems US AID United States Agency for International Development VPU Vice presidential unit WBI World Bank Institute Overview Well-governed countries are better able to formulate In many countries, it has sustained a medium-term GAC growth-enhancing policies, deliver essential services to dialogue on issues such as public financial management, the poor, and regulate financial and product markets. service delivery, and the investment climate. Borrowers, The appeal of governance reform and the fight against development partners, and civil society organizations corruption can resonate widely across diverse countries continue to value the Bank's analytical capacity and its and social groups, as demonstrated by recent events in long-term experience in this area. There have been signs the Middle East and North Africa. Building on more than of progress since the strategy was launched. Notably, the two decades of experience, the World Bank's 2007 gover- Bank made plans to support institutional strengthening in nance and anticorruption (GAC) strategy reaffirmed its three times as many countries in the fiscal 2008-10 period continuing commitment to the crucial and challenging as it did in the fiscal 2004-07 period. Its use of governance agenda of helping countries develop accountable and and political analysis in project design has increased sig- effective states. nificantly, as has its use of some country systems in proj - ects in Mrica and in countries with weaker institutions. Focusing on the country operational aspects of the overall GAC agenda, the Independent Evaluation Group At the same time, important opportunities have yet to be (lEG) assessed the relevance and effectiveness of the seized. Project-level solutions to the challenge of institu- strategy and its first phase of implementation efforts over tion-building need to keep pace with increased commit- fiscal years 2008-10. A key feature of the evaluation is its ments in country strategies. The quality and coverage of benchmarking of the content and quality of the Bank's political economy analysis in policy dialogue needs to country-level engagement on GAC issues, before and af- show the systematic improvement evident in projects. ter the 2007 strategy. It did not review the organizational There is still room for country programs and projects to restructuring of the Integrity Vice Presidency and related improve measurement of governance results, to expand reforms to strengthen the Bank's corporate investiga- overall use of measures to foster the demand for good tions and sanctions regimes, following the Volcker Panel governance, and to deploy-more systematically-en- report. It also did not cover the organizational renewal of hanced GAC measures to manage fiduciary and gover- the World Bank Institute or individual global partnership nance-related risks. Also, the Bank's operational response programs with GAC themes (such as the Stolen Assets in countries experiencing governance downturns needs Recovery Program). to be more consistent. Many stakeholders inside and outside the Bank hold the view that lending goals conflict The findings of this evaluation are, by design, intended with pursuing GAC objectives. to inform a planned second phase of the GAC imple- mentation-a learning process that signals one of the Key design elements of the 2007 strategy and the imple- Bank's strengths. They can be viewed in the context of mentation plan are to be addressed if the Bank is to more an ongoing change management and internal reforms effectively help countries overcome deep-seated gover- agenda, which includes the strengthening of the Integrity nance challenges such as civil service dysfunction, cap- Vice Presidency, a new World Bank Institute strategy, ture of natural resource rents, and political-institutional an institutionwide transparency initiative, and efforts barriers to market entry and improved service delivery. to modernize investment lending. Taken together, these To date, GAC operational efforts have focused more on reforms have sought to create an enabling environment the Bank's own capacities, resources, and reputation as a for the Bank to pursue the GAC agenda on multiple development partner, than on strategic issues facing part- fronts-including its operational engagement in partner ner countries. Guidance to operational teams emphasized countries, the subject of this review. managing transaction-level fiduciary risks in investment projects rather than updating the Bank's approach to As it did in the years preceding the strategy, the World managing systems-level risks, including in policy-based Bank has continued to .support good governance objec- lending. Implementation arrangements within the Bank tives in virtually every country where it has operations. were fragmented and needed to be more oriented to front-line concerns and results. And internal and donor GAC issues-their GAC responsiveness. The framework resources needed to be more strategically deployed to defined GAC responsiveness as the Bank's selectivity in meet assessed GAC needs. identifying entry points and instruments; its support for institutional strengthening; its identification, signaling, and Drawing on these lessons, the planned second phase of mitigation of risks; and its use of smarter project design (for GAC can more fully deliver on its potential by focusing example, improved "fit" to country contexts, demand-side on operational solutions that help build country gover- and preventive measures against fraud and corruption, nance capacities. In particular, the Bank needs to update use of country systems, and results orientation). It did not its approach to institutional strengthening in the core seek to quantify the actual levels of fraud and corruption public sector, the social and infrastructure areas, the in projects. The evaluation also reviewed GAC-in-sectors investment climate, and the demand side of governance. issues in the roads and primary education sectors, as well as Such an approach would take advantage of innovations accountability institutions. The evaluation looked at support in financial instruments, improved analytics, and more provided to country-level GAC efforts, for example, the pre- systematic measurement of results. The Bank can also ventive services work of the Integrity Vice Presidency, key encourage innovation in these areas by clarifying its "zero multi-stakeholder engagements by the World Bank Institute, tolerance" stance on corruption and, in particular, by and some global efforts. It did not review the organizational focusing more on systems-level risks across all opera- restructuring of the Integrity Vice Presidency and related tions rather than only transaction-level risks in invest- reforms to strengthen the Bank's corporate investigations ment projects. In parallel, internal budget and trust fund and sanctions regimes, the organizational renewal of the resources can be more strategically deployed in ways that World Bank Institute, or individual global partnership pro- empower sector and field-based units. grams (such as the Stolen Asset Recovery Program). Evaluation evidence comes from reviews of the GAC re- Evaluation Background sponsiveness of 50 country programs, 200 lending and trust- The Bank is currently implementing its 2007 strategy funded operations, and relevant economic and sector work Strengthening World Bank Group Engagement on Gover- over the FY04-07 or pre-GAC period and the FYOS-1 0 or nance and Anticorruption. The strategy and its implemen- post-GAC period. Its findings are also based on economet- tation plan sought to increase significantly the number of ric analysis, detailed case studies of six country programs, countries and projects in which the Bank helped system- and reviews of relevant Bank budgets, trust funds, staffing, atically address GAC issues, although it did not indi- and institutional arrangements. In undertaking this empiri- cate targets or a timeframe within which this objective cal work, the evaluation recognized that new metrics and would be achieved. Bank teams were provided guidance, data on GAC issues are under continuous development. toolkits, and operational support as part of initiatives to These methods were complemented by structured feedback enhance country engagement (GAC-in-countries) and to from stakeholders inside the Bank and those outside such strengthen incentive and risk management frameworks as government, donor, and civil society representatives. in sector dialogue and project management (GAG-in- sectors and GAG-in-projects). Global GAC efforts tried to Before finalizing the report, lEG presented the findings increase Bank involvement in international peer learn- to the Bank's GAC Council, shared the full country- and ing networks and collaborative governance initiatives. project-level data set for review, and held meetings with This evaluation assessed the relevance and effectiveness Bank management to solicit written and verbal feedback. of the strategy and FYOS-1 0 or Phase 1 implementation This engagement will be amplified by dissemination of the efforts in enhancing the Bank's country-level response to final report. Through this process, the evaluation aims to GAC issues. Through this evaluation, lEG aims to help inform a planned second phase of GAC implementation. strengthen the Bank's overall support to countries in developing effective and accountable states, and to fulfill Is the Bank More GAC Responsive? the institution's commitment to independent evaluation of large corporate strategies and initiatives. Building on two decades of engagement on governance issues, the 2007 GAC strategy acknowledged that it "im- The evaluation developed and applied a framework for as- plied a change in the way the Bank does business:' Stra- sessing how Bank country programs and projects address tegic communications and engagement by Bank senior Overview xi management externally and internally signaled this goal. of enhanced GAC measures to manage fiduciary and The implementation plan and annual progress reports governance-related risks. Even with some improvements, viewed successful implementation as systematic improve- less than a third of CASs and projects scored highly for ment in the GAC responsiveness of country operations, quality of enhanced fiduciary aspects in either the pre- or although they did not set targets or a timeframe for the post-GAC periods. A minority of pre- and post-GAC achievement of this goal. Over the FY08-l 0 period, the country programs aimed to use portfolio processes to Bank's response to GAC issues in its country programs identify, signal, and mitigate GAC-related risks. These and projects has demonstrated continuity without sys- measures could have been better tailored to the risk pro- tematic improvement as yet. file of projects and programs. GAC in Country Programs and Projects. The Bank Additionally, the Bank's operations would have benefitted has continued to support GAC-related objectives in its from a clearer definition and more consistent application strategies in every country where it has operations. Pre- of risk tolerances across countries. To date, operational and post-GAC country assistance strategies (CASs) were responses varied considerably in countries experiencing similarly selective in identifying entry points for a GAC governance downturns (for example, incidence of grand dialogue. In many countries, the Bank has sustained a corruption, periods of political instability, outbreaks of medium-term dialogue on GAC issues and provided civil conflict)-a finding that raises concerns about con- a program of support in areas such as public financial sistency of treatment, a key GAC principle. The modest management, sector service delivery, and the investment performance of projects in the measurement of country climate. Sustained engagement on these issues, even in GAC results continued post-GAC without statistically challenging settings, remains one of the Bank's strengths. significant improvements. While some form of benefi- ciary involvement in Bank operations was prevalent, less Since the launch of the strategy, there have been signs than a third of pre- and post-GAC projects scored highly of progress. Figure A shows the percentage of country for the overall use of demand-side measures. programs and projects that incorporated GAC ele- ments "to a great extent:' Notably, the Bank committed Political Economy Analysis. Efforts to strengthen the to support institutional strengthening in three times as Bank's approach on political economy analysis have many countries in FY08-10 as it did in FY04-07. The received greater attention and support post-GAC. But the number of Bank projects that relied on governance and quality and coverage of political economy issues in the political economic analysis upstream of design increased Bank's economic sector work did not show the systematic significantly. In countries with weaker institutional improvements evident in projects. The operational ben- capacities (as measured by Country Policy and Institu- efits of free-standing political economy analysis reports tional Assessment governance scores lower than 3.5), 41 were often limited by an overly academic orientation, percent of post-GAC projects used at least some country uneven methodological rigor, and a lack of consistency systems-that is, public financial management, procure- between recommended actions and prevailing interpreta- ment, or personnel systems-compared to a quarter of tions of the Bank's Articles of Agreement. pre-GAC projects. In Africa, projects using at least some country systems increased from 11 percent pre-GAC to The Bank's Standing on GAC Issues. The implementa- 40 percent post-GAC. In principle, the Bank's increased tion plan's goal of improving the Bank's reputation on GAC use of these systems in weaker settings allowed countries issues was achieved partially. Government and donor to more efficiently deploy limited capacities on pressing officials consulted during the evaluation were appreciative domestic priorities rather than sui generis donor project of the Bank's capacity to advise on governance issues and management and reporting requirements. to provide operational support for institutional develop- ment. Some felt that the wider development community At the same time, important opportunities have yet to be would benefit from the Bank's efforts to develop political seized. CAS commitments to significantly scale up insti- economy analytics. Others recognized the Bank's high tutional strengthening efforts post-GAC have yet to be fiduciary standards relative to other development agencies. matched by an expansion in institutional strengthening components in projects. In both country programs and At the same time, stakeholders inside and outside the projects, there is still a need for systematic deployment Bank observed a potential conflict between its lending xii World Ba nk Country-Level Engagement on Governance and Anticorruption Country Programs and Projects Addressing GAC Issues" To a Great Extent" Percent 0 10 20 30 40 so Ill Portfoliowide fiduciary risk mitigation -~ c ;, Selectivity ... 0 .5 Signaling u <1!1 Cou ntry institutiona l strengthening ~ D PreGAC 0 10 20 30 40 so D Post GAC Qua lity of governance and PEA ~ ~ Quality of enhanced fiduciary aspects ... cu ·a ... Demand-side of governance CL c Use of country systems 0 < ~ 1!1 Qua lity of institutional strengthen ing Resu Its orientation Source: lEG desk review. Note: Arrows indicate statistically significant changes. goals and its pursuit of GAC objectives, particularly in effective, the Bank was usually focused on specific GAC poorly governed settings. Opinion leaders polled in a entry points and realistic in its aims. Further, it balanced 2008 Gallup Survey recommended that the Bank not commitments to support long-term institutional develop- lend to countries unless they took serious actions to fight ment with accountability for interim results. Also impor- corruption. Civil society organizations consulted for this tant was the choice of financial instruments (for instance, evaluation agreed that, in such settings, the Bank should development policy loans or investment loans), which was reduce lending, impose stricter GAC-related loan condi- associated with the achievement of certain GAC objec- tions, and channel funds outside of government. Nearly tives. These operational design issues were not new and half of Bank operational staff surveyed also believed that often pre-dated- although did not adequately inform- "the Bank's lending imperative conflicts with its ability to the 2007 GAC strategy and implementation plan. implement the GAC strategY:' Public Sector Reform . Progress in supporting public financial management reforms was uneven. Standardiza- Early Outcomes and Lessons tion of assessments and operational support for public In many countries, the Bank has supported efforts to financial management systems improved, but front-line address deep-seated governance challenges such as civil service delivery concerns were not adequately prioritized. service dysfunction, capture of natural resource rents, Greater engagement with citizens and better coordina- and political-institutional barriers to market entry and tion with sector initiatives enhanced the credibility of improved service delivery. As part of this evaluation, reforms in some cases (Azerbaijan and Moldova). Low detailed case studies were conducted in six countries civil service pay was a pervasive problem and imposed (Azerbaijan, Bangladesh, Cambodia, Guatemala, Liberia, major constraints on development efforts. Given the and Moldova) where the Bank sustained a medium-term complex political economy of civil service reform, the dialogue on these GAC issues over the FY04-l 0 period. Bank adopted opportunistic and selective approaches that produced modest results (Cambodia and Liberia). The evaluation's desk reviews and case studies showed that The achievement of public sector reform objectives was the Bank's record in helping to achieve countrywide gov- positively associated with the use of development policy ernance improvements was limited. Where its support was loans. Overview xiii Efforts to strengthen the demand and supply side of governance, particularly at the local level, can be mutually reinforcing. Building on lessons from earlier community-driven development and decentralization efforts, local governance initiatives in several countries have combined fiscal and capacity-building support for local executives and their constituents with financial transparency measures. These were intended to develop local institutions that can effectively and accountably meet local service delivery needs. • Bangladesh's Local Government Support Program has sought to empower its lowest tier of government through a nationwide program of district-based support. The program provides discretionary transfers and capacity-building support to 4,500 Union Parishads. It employs an accountability framework based on district- level progress reporting and monitoring, transparency measures, and audits. To date, it has helped complete over 12,000 annual audits, train nearly 50,000 personnel, and support 500 local-level peer learning sessions. • Cambodia's Rural Infrastructure and Local Governance Project supported decentralized and participatory processes, as well as financing of priority public goods at the commune or sangkat level. The project uses an arm's length arrangement to reimburse the costs of commune-level investments, and thereby allows the Bank to channel funds through Cambodia's basic intergovernmental system while shielding it from fiduciary risks. To date, it has contributed to the development of 1,800 irrigation schemes, a few rural roads and bridges, and some social services. • As part of Guatemala's public financial management reforms, a new framework for municipal financial management (SIAFMUNI) was implemented in more than 200 municipalities to improve both efficiency and transparency. In parallel, a citizen-oriented portal, Consulta Ciudadana, was established to offer user- friendly applications to facilitate access and interpretation of complex financial reports. Taken together, these measures have enabled citizens to access information about local government financial and procurement processes. Additional demand-side training efforts have been launched to empower citizens, some of whom expressed discomfort with the quality, accessibility, comprehensiveness, accuracy, and consistency of fiscal information. Source: lEG desk review and country case studies. Accountability and the Demand Side. Bank support be done by the Bank and countries to ensure that ef- focused primarily on supreme audit institutions and forts to strengthen cross-cutting systems (for example, anticorruption bodies (Azerbaijan). Achievement of ob- public financial management and personnel systems) jectives was heavily dependent on the independence and are better coordinated with sector initiatives to improve political composition oflegislatures. The Bank primarily service delivery (for example, the development of sector supported the demand side through community-driven workforces). Regression analysis showed that projects and local governance initiatives in a number of countries that included public sector capacity building and public (Bangladesh, Cambodia, and Guatemala among them, see disclosure measures were more likely to achieve sector Box A) . Direct financing of non-state actors-as op- objectives. Yet, the emphasis ofGAC efforts in FYOS- 10 posed to contracting nongovernmental organizations on was on mitigating project fiduciary risks rather than on Bank projects-was rare and, in some cases, stretched promoting service performance more broadly. the limits of the Bank's role as a multilateral develop- ment agency. Importantly, the use of development policy Investment Climate. The investment climate in several lending was positively associated with the achievement case study countries was constrained by public sector of accountability and rules-based governance objectives, bottlenecks, which the Bank sought to help remove. but it was negatively associated with the achievement Support to improve the operations and management of of demand-side objectives-a finding confirmed by the these public agencies included customs modernization in Bank's own reviews. Cambodia, and streamlining of licensing and registration procedures and strengthening supreme audit agencies GAC in Sectors. A central concern in the social and in Moldova. Given the importance of transparency for infrastructure sectors was the alleviation of public market entrants, the Bank's advocacy of greater informa- management constraints on service delivery. More could tion disclosure proved important to the private sector. xiv I World Bank Country-Level Engagement on Governance and Anticorruption However, support for consultative mechanisms between ment projects). It was based on the premise that a lack of the private sector and the government needed to be bet- commitment and capacity of Bank staff posed binding ter calibrated to risks of capture. constraints on the achievement of GAC objectives. The plan lacked a results chain and clearly communicated Project Fiduciary Measures. During GAC implemen- implementation targets. As a result, by the end of its tation, the ring-fencing of fiduciary controls on Bank third year, its original goal of making systematic and projects was given particular attention. These methods time-bound improvements in the GAC responsiveness of sought to limit exposure to fraud and corruption risks operations was no longer widely recognized by key staff. and also manage reputational risks to the Bank and bor- rower governments. The focus on ring-fencing methods Phase 1 efforts needed to more concretely focus on press- in some countries (such as the use of an independent ing strategic and substantive issues facing GAC reformers procurement agent in Cambodia) but not in others in countries. For instance, what lessons could countries reflected the Bank's lack of consistency in setting risk draw from the 2008-09 global financial crisis for cor- tolerances. Generally, initiatives designed to manage the porate governance and the integrity of their financial Bank's reputational risks relating to GAC were not neces- systems? How could public sector reforms be tailored to sarily the same as those that would help countries take on meet the particular needs of conflict-affected states? How calculated development risks. could development partners help reforms address deep- seated problems of systemic corruption? The GAC One- Year and Second-Year Progress Reports acknowledged What Difference is Phase 1 Making and Why? that the Bank had intended to focus on such issues. GAC implementation has involved considerable efforts by Bank operational staff, who reported almost universal Implementation commitment to the strategy's objectives. While some projects and country programs benefitted directly from GAC implementation involved the provision of guidance Phase 1 support (for example, through country gover- and the delivery of support to Bank teams as well as the im- nance and anticorruption or CGAC processes and Gover- plementation of risk management measures and controls. nance Partnership Facility financing), many did not. Bank teams continue to face operational hurdles in helping GAC Guidance. Guidance materials issued over the countries systematically address the types of deep-seated FYOS-10 period concentrated on GAC-in-projects is- governance challenges noted above. To help overcome sues. More emphasis was given to managing fiduciary these hurdles, the Bank needs to revisit key elements of risks on transactions on investment projects than to the 2007 strategy and implementation plan. proposing practical solutions for deepening the use of country systems. GAC guidance in the roads sector focused on managing procurement risks rather than Relevance and Appropriateness of Design strengthening sector institutions overall. GAC-in-edu- The 2007 GAC strategy represented an important step cation efforts appropriately highlighted the importance in reaffirming the Bank's longstanding commitment to of measuring sector incentives, but were less concerned helping develop effective and accountable states. The with operational solutions. Considerable attention strategy's objectives were highly relevant to the needs was given to multistakeholder engagement, although and goals of countries, and it benefitted from sustained more clarity was needed on the trade-offs for the Bank support from the Bank's top management. Even so, the between helping to "create space" for non-state actors strategy could have addressed the mixed record of the (for example, through transparency measures) and Bank's public sector reform and related business lines, actively motivating demand-side pressures. The Bank's which needed strengthening. Also, while it promoted framework for political economy analysis appropriately GAC as "everybody's business;' the strategy defined the emphasized formal and informal institutions. Yet, the agenda too loosely to allow for priority-setting. guidance in this area would have been more relevant had it more clearly defined what constitutes "good insti- The Phase 1 implementation plan was focused on the tutional fit" to country realities, addressed the political Bank's own capacities and resources, its reputation as a economy of aid, and recommended rules for disclosure development partner, and its fiduciary risks (in invest- of sensitive analyses. Overview xv Delivery of Support. Emphasis on internal communi - remained the most selective development agency. Over cations and training workshops paid off: 63 percent of the entire period of review, improvements in governance respondents to lEG's staff survey were familiar with the scores on the Country Policy and Institutional Assessment 2007 GAC strategy. While learning activities focused on were associated with increases in International Develop- GAC-in-projects and country accountability institutions, ment Association (IDA) commitments and disbursements. GAC-in-sectors was not given adequate attention. The Yet, Bank flows to International Bank for Reconstruction generally low staff ratings on the relevance of specific and Development (IBRD) countries were relatively less se- guidance materials and tools were also reflected in low lective in terms of country governance performance. Also, utilization rates. The GAC implementation plan could the relationship between governance performance and have been better exploited as an opportunity to prioritize IDA disbursements was affected-sometimes negatively- and coordinate learning activities. by the mix of financial instruments in country portfo- lios, including the use of development policy lending. Beyond access to guidance information, relatively few staff This finding-taken together with the above-mentioned reported receiving tangible support. For those that did, strengths of development policy lending-points to the Phase 1 support varied in terms of its value added to GAC need for clear guidance on instrument choice. responsiveness of operations. Staff felt that Bank manage- ment could have done more to help them address imple- This point is further amplified by the evaluation's finding mentation challenges, and sought more clarity on risk tol- that the risk management intensity of Bank operations erances for Bank engagement in different settings as well was associated more with the choice of instrument than as streamlining of risk reviews on investment projects. For the risk profile of an individual operation. Relevant risk example, the perceived risk of complaints to the Integrity management measures used in a given project included Vice Presidency and ensuing investigations discouraged links to economic and sector work, governance and an- the use of country systems-a key GAC principle-and ticorruption plans, supplemental supervision, grievance encouraged the ring-fencing of investment projects. mechanisms, and disclosure measures. Regression analy- sis found that risk management intensity-defined as the Bank Controls on Aid Allocation and Operational total number of relevant risk management measures used Risks. The Bank's aid allocation procedures represented in a single project-differed significantly by the type of a critical element of its controls framework. Since the lending instrument used (Figure B). When controlling launch of the GAC strategy, the Bank has continued to use for other factors , the level of financial management and governance performance as a key criterion for allocat- procurement risk was not significantly associated with ing concessional resources across countries, and it has risk management intensity. The result was in part explained by the distinct opera- Factors Associated with Risk Management Intensity of tional controls used for investment loans and develop- Bank Operations ment policy loans, even though distinctions between the two instruments have become less pronounced. - 2.0 I Development Policy Loan Reinforced by the transaction-level emphasis of GAC- in-projects guidance, the layers and complexity of risk - 1.6 I FPDNetwork reviews for investment lending continued to differ mark- - 1.31 PREM Network edly from those for development policy lending. These Risk Exposure po.7 efforts strengthened controls on fraud and corruption risks in Bank investment projects, but did not emphasize An a lysis of Informal Institutions 12.0 systems-level risks that affect all instruments (including Analysis of Forma/Institutions 12.2 development policy loans). The recently introduced Op- South Asia Region erational Risk Assessment Framework (ORAF) does not 12.4 address this issue. Looking forward, an updated method- Eastern and Central Asia Region 12.6 ology for review of systems-level risks could be usefully applied across financial instruments (including the - 3.0 -2.0 - 1.0 0 1.0 2.0 3.0 anticipated Program for Results instrument) and would Source: lEG desk review. Figure shows marginal effects (Appendi x E, ensure a more consistent risk management approach. Table E.l6b). xvi World Bank Country-Level Engagement on Governance and Anticorruption F.IGURE <: Allocation of Governance Partnership Facility Grants by World Bank Sector Units Public Sector $28.6 (N = 40) Country Services Panel $8.1 (N = 7) Social Development $6.9 (N =11) Econ Policy & Public Sector $3.4 (N = 7) Social Protection $3.1 (N = 3) Others Operational Services $2.0 (N = 3) Water $1.8 (N = 3) Financial Management Private Sector Development Economic Policy $0.9 (N = 3) Agriculture & Rural Dev. $0.9 (N = 2) Procurement Poverty Reduction Urban Development Infrastructure Human Development $0.5 (N = 1) Education Communications Health, Nutrition & Population Econ Policy & Poverty Reduction $0.2 (N = 1) ~~--~~----.------.------.------.-----. 0 5 10 15 20 25 30 D Amount in $ million (no. of grants) Sources: Operations Portal; Governance Par tnership Facility Secretariat, as of December 2010. Incentives Partnership Facility trust fund. Incremental budget al- locations gave priority to the Integrity Vice Presidency An important aspect of GAC implementation was its in- and Regional vice presidential units (VPUs), in particu- centive framework for change management. Three factors lar Africa. Designed to jump-start innovation within that affected incentives during Phase 1 warrant continued the Bank and promote GAC as "everybody's business;' attention: the Governance Partnership Facility had approved 94 grants, totaling $65 million as of December 20IO. The Financing of GAC Implementation. The 2007 strategy overwhelming majority of these grants was administered did not specify what it was adding to the Bank's consid- erable body of work on governance. Instead, the Phase through public sector management units in the Bank's I implementation plan identified a set of GAC change Poverty Reduction and Economic Management Network, initiatives for which it sought additional funding. It did and largely supported country-level efforts, mostly in not seek to first align the Bank's existing and already Africa (Figure C). growing base budget funding for governance work with new GAC priorities. Rather, from fiscal 2008 onward, At a time when Bank spending on governance work was the GAC strategy was resourced at the margin through already large and growing (increasing 2I percent from incremental Bank budget and donor funds . $I40 million in FY04 to $169 million in FYIO), these parallel arrangements did not achieve their incentive The Phase I plan earmarked a total of $II9 million for objectives. First, the intended effect of Bank budget GAC implementation, consisting of$54 million in Bank increments-to increase Regional VPU spending on budget increments allocated for the FY08-II period and governance work-was muted. Due to the fungibility of $65 million in donor funds allocated for the FY09-I2 resources, Regional VPU spending on governance work period under the largely Bank-executed Governance over FY08- I 0 increased by $9.6 million less than antici- Overview xvii pated. The remainder of these increments was deployed solutions to meet the challenge of helping build country away from the governance priorities identified in the governance capacities. The findings of this evaluation implementation plan. Second, the distinct Governance point to five sets of actions: Partnership FaCility allocation procedures-competitive selection by a Bank-donor committee-were outside the Focus on helping countries make tangible and time- Bank's budget process and did not systematically iden- bound governance improvements, while acknowledging tify innovative efforts. This caused frustration for some and seeking to resolve trade-offs between: operational units due to the perceived failure to clearly justifiy Governance Partnership Facility decisions and to • Committing Bank support for institution building consider the priorities of the country and theVPU. Third, over the long term and ensuring accountability for incremental financing was not systematically linked to results (for example, in service delivery) in the short incremental GAC activities. Even though Governance term. Partnership Facility reporting on the use of funds was • Supporting system-wide public sector reforms and systematic, overall reporting on GAC results lacked supporting selective public management improve- information on activities financed through incremental ments in priority service delivery sectors. Bank budget resources. Generally, corporate reporting on GAC implementation focused more on Bank inputs • Helping governments respond to demand-side pres- than on the quality of operations and country governance sures and directly engaging non-state actors in order performance. to motivate demand-side pressures. • Upgrading of country systems through their deliber- Dedicated GAC Staffing. Phase 1 financing enabled the ate use and safeguarding Bank funds from abuse. recruitment of about 64 new and redeployed positions, including several dedicated GAC personnel. Efforts to Update the Bank's approach to institutional strength- formally establish competencies for this GAC stream risk ening by: creating overlaps with well-established competencies for public sector specialists and, in some cases, fiduciary and • Leveraging innovations in financial instruments and social development staff. A separate GAC cadre is not building on lessons learned to strengthen business likely to be sustainable without a realignment of some lines that warrant more immediate attention- civil network and central units. service pay reform (particularly in fragile states); pub- lic management support for basic service delivery and Coordination and Accountability. The GAC Council the investment climate; public financial management served more as an information-sharing forum than a of natural resource rents; and civil society capacity decision-making body. Council meetings, which were building. regularly attended by a large number of nonmembers, • Strengthening Bank-country dialogue, primar- usually involved presentations by Bank units show- ily through better integration of political economy casing their efforts. These presentations would have analysis into standard Bank economic and sector benefited from critical review of what was working and work (and less through the creation of confidential, what was not. Other specialized GAC arrangements, free-standing political economy analysis products). such as program secretariats housed in the Poverty Reduction Economic Management Network (PREM) • Adapting actionable governance indicators more Anchor and GAC focal points in Regions and networks, systematically to project results frameworks. also ensured that GAC received continuous manage- Clarify the Bank's "zero tolerance" stance on corrup- ment attention. However, operational staff reported tion and improve operational controls by: that these arrangements could have been more relevant to their work. • Developing a harmonized approach to reviewing and managing systems-level fiduciary and GAC risks across instruments-and not simply transaction-level Recommendations risks in investment projects. The approach should The planned second phase of GAC can more fully deliver provide for additional due diligence on operations on its potential by focusing on developing operational with specialized risks. xviii World Bank Country-Level Engagement on Governance and Anticorruption • Providing guidance to operational teams on the ap- • At the corporate level, the GAC Council should propriate use of different Bank financial instruments focus on institutionwide issues and risks, and on in different governance settings, consistent with the benchmarking the GAC responsiveness of Bank op- institution's overall risk appetite. erations. • Consistently defining risk tolerances for the levels and Align GAC implementation arrangements with Bank composition oflending as well as the use of country administrative and operational processes by: systems in different settings (for example, through lending scenarios) so that expectations of governance • Consolidating current fragmented financing arrange- performance are clearly understood by country stake- ments (that is, separate Bank budget and trust fund holders and the Bank's shareholders. allocations) while improving monitoring of GAC activities. Clarify roles and accountabilities for setting GAC strategic priorities: • Rather than creating a separate cadre of GAC special- ists, applying GAC competencies across existing Bank • At the country level, Bank country strategies should networks and career streams, and allow transferability continue to serve as the primary mechanisms for of GAC-competent staff across networks. reflecting the priorities and needs of clients on GAC • Streamlining specialized GAC institutional arrange- issues. Donor-funded initiatives need to be appropri- ments with a view to empowering line managers in ately aligned. VPUs to achieve GAC objectives. • At the VPU level, GAC work plans should be informed • Supporting increased applied research on what works by demand in partner countries and should set pri- in various GAC areas. orities for overall resources use-both Bank budgets and trust funds. In keeping with Bank policies on the • Developing a results framework that includes baseline integration of trust fund allocations with the budget indicators of Bank and country-level performance, sets process, decisions on allocations of trust funds to GAC targets, and integrates monitoring of GAC responsive- activities should involve line management in VPUs. ness into standard portfolio monitoring. Overview xix Management Response Putting the Report in Context. Helping countries in their many competing pressures. As the 2011 World Develop- efforts to make governance improvements and strengthen ment Report points out, institution building requires a 20 institutions is an important element of the Bank's gover- year time horizon, but citizens have needs for government nance and anticorruption (GAC) strategy, and Management services now. This evaluation covers the first three years of welcomes the opportunity to comment on the Indepen- GAC implementation (the average operation or country dent Evaluation Group (IE G) evaluation of this work. The strategy examined in the evaluation is less than two years report's endorsement of the high, and continuing, strategic in) , and there is much still to do in GAC and much to build relevance of GAC goals and objectives, and its recognition on from the evaluation. There are also obvious limits to of strong top-level management support, resonate, as do what can be achieved in such a short time in what amounts early findings such as that the Bank is supporting country to a major institutional culture change. That said, Manage- systems strengthening in three times as many countries in ment agrees with many of the lEG findings. While it does FY 08- 10 as it did in FY 04- 07. Management also agrees have areas of disagreement, set out below, it will use the with the evaluation that while the World Bank continues lEG evaluation, along with other lessons of experience, to to support good governance objectives in virtually every help update the GAC strategy. In particular, Management country where it has operations, there are still important will implement lEG's recommendations in three areas: opportunities that can be seized. stronger support for country institutional strengthening; an improved risk framework; and a stronger results frame- However it is also important to underscore that these work, where lEG's work provides valuable baseline data. country actions are only part of a broader GAC strategy that the lEG report does not cover. This broader GAC work Structure of the Management Response. The Manage- includes key institutionwide and partnership initiatives, ment Response first reviews significant progress in GAC such as strengthening the Bank's corporate investigations strategy areas not covered in the lEG evaluation. Next it and sanctions regimes, revitalizing the World Bank Institute sets out the outlines of where we want to go with the next (WBI), supporting global programs such as that on Stolen phase of the GAC strategy, taking on board the learn- Assets Recovery (StAR), and undertaking path breaking ing to date, including from lEG. It then sets out some reforms to make the World Bank's operations and research important areas of agreement with the evaluation before more open, transparent and accountable. In reading the discussing areas for clarification and issues on which lEG report on country work it is important not to lose sight Management does not agree. It ends with comments of this larger governance and anticorruption work program, on lEG's recommendations, the overall thrust of which or to sell short the very real progress the World Bank has Management supports. The full draft Management Ac- made on this broader agenda. It is particularly important, tion Record is attached as an annex. as the lEG Report acknowledges, to emphasize that the World Bank has taken great strides, including implement- ing all recommendations of the Volcker Report, reforming The Bank's Broader Strategic Engagement on sanctions and debarment proceedings, strengthening pre- Governance and Anticorruption vention work and scaling up the activities of the Integrity In addition to the Bank's country level engagement- Vice Presidency, to secure its own investment funding from the subject of this evaluation, Management would like fraud and corruption. to highlight notable achievements in the Bank's broader strategic engagement on GAC that the evaluation does not The Focus of the Report: The Country Component of address and that are highly relevant not only to the role GAC. The focused topic of the lEG evaluation-how the the Bank has played on the global stage to drive initiatives Bank engages with countries to assist them in addressing on governance and anticorruption, but to its own fiduciary their own governance issues- is perhaps the most complex due diligence. It is ironic that the Report finds fault with and challenging sphere of development and one with which the Bank for spending too much time focusing on ring- bilateral agencies and international institutions are all fencing and protecting its own projects from corruption. coming to grips. There are few quick fixes in this work and This is not something we should apologize for. xx World Bank Country-Level Engagement on Governance and Anticorru ption Broader GAC efforts not covered by the lEG Report since 2008; generating close to 200 combined sanc- have ranged from a strengthening of the Bank's cor- tions applications and debarments, and making porate investigations and sanctions regimes, following over 150 referrals to national authorities. the Volcker panel report; renewal of the WBI; support • Implementing the Volcker recommendations and to global programs such as that on StAR; to ground- scaling up the Integrity Vice Presidency's preven- breaking work to make the World Bank's operations tative work-including building precautions into 75 and research more open, transparent and accountable. high -risk operations at the design and implementa- These efforts have brought real achievements such as: tion phases during FYl0-11 through direct work with task teams; inputting Integrity Vice Presidency • Mobilizing collective action on global governance information on country-level risks into Country issues-including supporting the Extractive Indus- Assistance Strategy (CAS) discussions; undertaking tries Transparency Initiative (EITI), the Medicines thematic assessments of risks and lessons learned Transparency Alliance (MeTA), the Construction on preventive measures at the country, instrument Sector Transparency Initiative (CoST), the Program or sector level such as the Global Roads Review; on Forests and the Forest Law Enforcement and and building country capacity on the ground by Governance partnership (PROFOR-FLEG), the providing hands-on practical application training to Global Program on Fisheries (PRO FISH), and the over 2700 officials representing PIUs, national audit Global Roads Integrity Initiative. institutions, anticorruption authorities, and Bank • Bringing together new partners to tackle global staff. By largely excluding Integrity Vice Presidency's threats to good governance and influencing the preventive work from its evaluation, lEG has missed global policy agenda-including launching the an important aspect of GAC implementation. International Corruption Hunters Alliance (ICHA); partnering with INTERPOL, the United Nations Moving Forward on GAC Support- Office on Drugs and Crime (UNO DC), the World Updating the GAC Strategy Customs Organization, and the Secretariat of the Con- vention on International Trade in Endangered Species The GAC strategy and implementation plan (IP) were (CITES), to establish the International Consortium for designed to be refined over time on the basis of experi- Combating Wildlife Crime (ICCWC); and launching ence. Indeed given the disappointing trends in gover- the Stolen Assets Recovery (StAR) initiative, endorsed nance over the decade leading up to the strategy and by the G-20 and the United States Senate Permanent the need for new approaches, the IP was deliberately Subcommittee on Investigations in its report, Keeping set out as a change management strategy, requiring Foreign Corruption Out of the United States. continuous adaptation and learning by doing. The lEG evaluation is useful in that regard and in the next phase • Developing and implementing policies to apply the of the Bank's GAC efforts, Management will place par- transparency principle to its own activities and make ticular emphasis on several areas highlighted by lEG. the World Bank Group a global transparency leader- These include: refining how the Bank engages with including through a new Access to Information policy; countries to strengthen institutions, while simultane- an Open Data initiative enabling free access to develop- ously ensuring that International Bank for Reconstruc- ment data for researchers, students, development practi- tion and Development (IBRD) and International De- tioners and others across the globe; and the publication velopment Association (IDA) resources are adequately oflmplementation Status and Results (ISR) reports, a protected, including through strong fiduciary support; key tool for reporting on the implementation perfor- further developing a framework for assessing risks and mance and results in all Bank-supported operations. rewards to help guide operational decisions; and set- • Enabling concerted anticorruption efforts across ting out an overall results framework so that over the international institutions-including signing next five years, Management and the Board will have a with other leading multilateral development banks clear idea of what has and has not been achieved. (MDBs) an agreement to cross-debar firms and individuals found to have engaged in wrongdoing Management will remain ambitious in its support to in MDB-financed development projects. countries. It must also, however, be realistic about its capability to influence in this challenging reform arena, • Expanding investigations and sanctions activity- given the long-term nature of institutional change in the concluding 553 external and internal investigations GAC context and the centrality of country ownership. Management Response xxi Areas of Agreement projects over a three-year period. But this premise is faulty. Management's goals were much more modest. IP The evaluation's endorsement of the high (and continu- implementation was designed as a change management ing) strategic relevance of GAC goals and objectives, strategy that sought to strengthen staff awareness and and its recognition of strong top level management capacity, build tools, and generate good practice in se- support, is encouraging. Experience has shown that lected areas-learn what works, iterate, and scale up over strong and continuing senior management support is time. Management has been explicit and consistent over essential to any change management effort. Manage- time in describing GAC implementation as a learning- ment is pleased that the evaluation finds that the Bank oriented, participatory, change management process. is supporting country systems strengthening in three Thus in 2007 the GAC Implementation Plan noted: "The times as many countries in FY 08-10 as it did in FY implementation of the GAC agenda will be a long-term 04- 07. This finding is important, as the GAC strategy effort. As the (initial] year-long learning-by-doing pro- and IP stress that stronger country systems are central cess unfolds, the medium-term challenges and actions to helping countries develop into fully accountable will become clearer-and will be detailed in implementa- states. Management also welcomes the finding that tion progress report[(s) to the Board.]" country teams have sustained country dialogue on governance and anticorruption issues. It is this combi- Given this approach, Management believes that, the nation of sustained dialogue and support for country evaluation missed an opportunity for learning about systems that is likely to result in long-term gains in how Management could strengthen its change manage- governance and anticorruption. Management also ment approach-Box 1 provides select details on one notes that important improvements were recorded, too, aspect of this concern-namely the missed opportunity in countries with weaker institutional capacities: the to evaluate the Bank's flagship GAC in country-pro- proportion using country systems (financial manage- gram efforts. It is Management's continued view that ment, procurement, personnel) increased from 27 when there are substantial disagreements between lEG percent "pre-GAC" to 41 percent "post-GAC:' and Management at the stage of the approach paper, more effort should be made to resolve those differences Management agrees with lEG that GAC progress has prior to embarking on the evaluation. fallen short in sectors, relative to overall GAC in proj- ects or in country strategies. Management also agrees Empirical methods. Besides the sampling issue above, that, rather than promoting a significant separate cadre Management believes some of the evaluation's find- of GAC specialists, there is a need for more general ings are based on limitations to the analytical frame- GAC competency across sectors and is now working in work and econometric work that are not sufficiently that direction. Management also agrees that more work acknowledged in the evaluation. One concern worth is needed on demand-side measures, but would have noting is that while a theme of the report is the lack appreciated more analysis by lEG on what works, where, of systematic progress on GAC-related inputs and and why. As noted above, Management agrees on the outputs, in comparing "pre-GAC" to "post-GAC" need to introduce a stronger results framework for GAC CASs and projects, in general, the "post-GAC" sample and to work more comprehensively on risk issues. seems to be of too early a vintage (on average in opera- tion only a year and a half after the IP was finalized) to Some Areas of Management Concern make a significant determination as to GAC respon- siveness. Indeed, much of the guidance developed in While there are some broad areas of agreement, Man- the IP years is just now being rolled out and tested in agement notes its concerns about the methodological operations. Furthermore, due to the short implemen- approach and the quality of some of the analysis. Three tation time-period covered, data capturing the extent of these areas of concern include: (i) the framing of the to which projects and CASs are implementing GAC goals Management set out in the GAC IP; (ii} empiri- measures are much more frequently censored (sta- cal methods; and (iii) and the possible interpretation of tistically, data left out because the results are not yet findings related to Development Policy Lending (DPLs). known) in the post-GAC period, which could poten- tially bias results. Finally, Management is concerned Management's Intentions. lEG evaluates the IP against that various passages in the evaluation discuss specific an objective of systematic improvement in the way the empirical findings in terms that imply causality (for ex- Bank engaged on governance in all countries and all ample, "The use of fast-disbursing DPLs increased the xxii World Ba nk Country-Level Engagement on Governance and Anticorruption Management's change management approach to strengthening GAC in country programs has been to identify and empower teams most committed to GAC mainstreaming. This effort proceeded along these steps: • In 2007, even prior to the IP, Management sought to create an innovation fund for committed country teams that put forward quality programs for GAC mainstreaming. The resulting multi-donor Governance Partnership Facility (GPF)-supported by the United Kingdom, the Netherlands and Norway-came on stream in September 2008. • In early 2008, to 'prime the pump' in advance of the GPF, Management invited Regional Vice Presidencies (RVPs) to identify country programs for the scaled-up GAC effort; 27 programs were identified, and each was provided with modest seed money (total budget cost $2.7 million, or an average of $1 00,000). • When the GPF came on stream, it was opened to all country teams (not only the 27 initially identified by RVPs). The GPF allocated resources competitively, with its decisions made on the basis of the quality of proposals and the demonstrated commitment of the applicant country teams. • Eighteen teams were selected to receive truly significant support, an average of $2 million, for implementation of their country programs. Consistent with its bottom-up approach, Management views the eighteen teams selected by the GPF to be the flagships of its GAC -in-country programs effort (total budget cost $36 million). Unfortunately, less than one-third of lEG's treatment sample was drawn from the flagship eighteen. Over two- thirds was drawn from the initial set of programs that received very modest seed money, but subsequently did not meet the GPF's tests of quality and commitment, and thus did not have access to significantly scaled-up resources. likelihood that countries-even when they had poorer management and control processes over the Bank's man- governance-would receive a flow ofiDA funds" agement ofDPOs were satisfactory. These processes were paragraph 5.4). To avoid possibly misinterpretation, adequate and effective to provide reasonable assurance Management welcomes lEG's disclaimer that "[t]his regarding the operation of key controls:' evaluation does not assert causality between variables; hence, it is not the purpose . .. to analyze the "impact" of any specific variables on GAC responsiveness:' Recommendations While Management has some concerns about the evalua- Clarifications on DPLs. First, Management would note tion's framework and the interpretation of some findings, that IDA flows are significantly and positively related Management welcomes the Report's recommendations, to governance performance including in countries with which focus on developing operational solutions to meet DPLs. This point does not come out clearly in the evalu- the challenge of helping countries build country gover- ation. Second, Management strongly believes that the evi- nance capabilities. While noting that achievements to date dence of experience shows that DPLs (or Development Policy Operations-DPOs-including credits and grants represent a substantial strengthening in the Bank's GAC to IDA countries) have proven their usefulness in weaker efforts, Management recognizes that there is scope for governance environments. The DPO process includes continued improvement, including in the way the Bank the relevant tests of country commitment, institutional engages with countries on GAC issues. As noted above, capacity, and fiduciary environment. Policy changes Management will draw on lEG's recommendations in its supported by DPOs are often important elements in update of the GAC strategy. improving governance. Third, Management notes that the number of risk reviews is not a good indication of In broad terms, Management endorses the recommen- the quality of risk review. All DPOs have a high level of dations, and they will usefully inform Management as it Management review and control mechanisms are robust. refines and adapts its strategic approach to supporting In June 2010, the Internal Audit Department completed countries on governance and anticorruption. Specifi- an audit of Bank processes for managing DPOs and gave cally, the updated strategy will draw on lEG recommen- those processes their highest rating (Satisfactory). The dations with regard to supporting countries in strength- audit states: "Our overall opinion is that governance, risk ening country systems and institutions, guidance to staff Management Response xxiii on risks, and a more robust results framework consis- for improvement are within the Bank's direct control tent with the next phase of GAC support. As noted in (e.g., stronger results frameworks in projects), whereas the attached Management Action Record, though, there others are more complex (e.g., attempting to influence a are some differences in view on specific issues and ap- range of critical and contested variables at the country proaches. Management also highlights that some areas level to effect governance reforms) . Management Action Record Major Monitorable lEG Recommendation Management Response Requiring a Response Focus on helping countries make tangible and time-bound Agreed : Management notes that the first recommendation is governance improvements, while acknowledging and seeking fundamentally cha llenging; the 2011 WDR stresses a 20 year time to reso lve trade-offs between: horizon for institution building, whi le recognizing t hat citizens Com mitting Bank support for institution building over the have immediate needs. Decisions regarding trade-offs ca n be long term and ensuring accountabi lity for results (for example, made on ly at country level and congruent w ith Bank comparative in service delivery, public financia l management, regu lation) in advantage. The final three recommendations may not entail trade- the short term; offs; the alternatives in each case may be complementary. Supporting systemwide public sector reforms and supporting selective public management improvements in priority service Management therefore commits to: delivery sectors; Articulate its approach t o upstream public sector reforms Helping governments respond to demand-side pressures (those focused on core govern ment functions and systems) and directly engaging non-state actors in order to motivate and downstream reforms (which focus on service delivery) in demand-side pressures; t he context of GACI I, to be discussed with Executive Directors Upgrading of country systems through t heir deliberate use in the second Quarter of FY 2012. and safeguarding Bank funds from abuse. A strengthened approach to supporting demand-side mea- sures will be set out in an annex to the GACII strategy. Continue and deepen work on supporting strengthening of country systems and continue to monitor and report peri- odica lly to Executive Directors on progress in use of country systems in procurement, financial management, and project management implementation. Management cannot commit to time -bound actions by coun- tries. Management does commit to support countries that have country-owned time-bound strategies to improve governance and will report on that support as part of GAC monitoring. Update the Bank's approach to institutional strengthening by: Partially Agreed. Management is developing a new resu lts-based Leveraging innovations in financial instruments and bu ilding lending instrument that will finance the delivery of results in many on lessons learned to strengthen business line.s that warrant of the critical areas listed by lEG. Management also endorses the call more immediate attention-civil service pay reform (particu- for strengthening country dialogue through the more widespread larly in fragi le states); public management support for basic use of polit ical economy ana lysis at cou ntry, sector and project level. service delivery and the investment climate; public financia l However, Management believes t hat country context is crit ical, and management of natural resource rents; and civi l society capac- whether or not to 'integrate' political economy analysis into econom- ity bui lding; ic and sector work is a decision that shou ld be taken at the country Strengthening Bank-country dialogue primarily through better level. Management agrees that actionable governance indicators integration of political economy analysis into standard Bank should be used more systematically in resu lts frameworks. economic and sector work (and less through the creation of confi- dential, free -standing political economy analysis products); and Management commits to: Adapting actionable governance indicators more systemati- Seek Board approval for new results-based lending instrument cally to project results frameworks. (the 'Program for Results') in FY 2012. xxiv I World Bank Country-Level Engagement on Governance and Anticorruption Major Monitorable lEG Recommendation Management Response l Requiring a Response Building on guidelines contained in "Problem Driven Governance and Political Economy Analysis;' published in September 2009, Management will develop further tools as necessary but will leave the decision on whether or how to use the tools and guidance to Regional staff, who have the relevant country knowledge. The guidance on using political economy analysis in fragile and conflict-affected states will be issued in the first half of GACII. Develop and disseminate guidance on using actionable gov- ernance indicators in project results frameworks, and monitor their usefulness and modify as necessary in the context of regular GAC reporting. Management will disseminate guidance in the first half of GACII. Clarify the Bank's "zero tolerance" stance on corruption and Partially agreed. Management has a clear position vis a vis 'zero improve operational controls by: tolerance' and will make efforts to ensure it is widely understood. Developing a harmonized approach to reviewing and manag- In summary, Management's position is that while we have no ing systems-level fiduciary and GAC risks across instruments- appetite for corruption, we have an ex-ante tolerance for risk (in and not simply transaction-level risks in investment projects. that it is recognized that such efforts in developing countries are The approach should provide for due diligence on operations more likely to encounter such challenges which the Bank seeks to with specialized risks. manage to as close to zero as possible), combined with an ex-post Providing guidance to operational teams on the appropriate zero tolerance when it is shown that fraud, corruption or other use of different Bank financial instruments in different gov- malfeasance has occurred. In such circumstances the Bank will ernance settings, consistent with the institution's overall risk always and everywhere take action to address the problem. appetite. Consistently defining risk tolerances for the levels and composi- Management also agrees that the attention to systemic risk be tion of lending as well as the use of country systems in different increased while continuing to pay attention to 'transactions' risk. settings (for example, through lending scenarios) so that expec- However, Management notes that a ' harmonized approach' to as- tations of governance performance are clearly understood by sessing and managing risk should not mean that responses must country stakeholders and the Bank's shareholders. always and everywhere be the same. Country context matters fundamentally and explicit decisions must be based on the specif- ics of country circumstances. What is missing from lEG's recommendation is the concept of reward versus risk. Management analyzes risk against expected operational development outcomes and has already differentiated across coun- try contexts in this regard by setting a target of 70 percent Marginal Satisfactory (MS) or better average lEG ratings for operations in Fragile States (where risks are, of course, high but the returns to suc- cessful operations tend to be especially high), 75 percent or better in other IDA countries, and 80 percent in IDA countries. Management commits to: Clarify its position on "zero tolerance;' explaining again to staff that development support is a risky business and there is no way, other than not lending at all, to guarantee the absence of fraud and corruption in Bank-supported operations; the Bank supports borrowers in providing reasonable assurance that funds are used as intended; but the Bank has zero tolerance Management Response xxv Management Action Record (continued} Major Monitorable lEG Recommendation Management Response Requiring a Response once fraud or corruption is found. Management will set out its internal communications plan, involving the World Bank Group's Chief Risk Officer, in the context of the GACII discus- sions. Review the experience with ORAF after two years (notably its use as intended in differentiating the management of opera- tions by risk) and make adjustments as needed. Continue to develop its comprehensive approach to risk management related to operational support to client countries, reporting progress regularly to the Board, via the Audit Com- mittee, on the ongoing efforts to define risk appetites and the tolerances via which Management uses to ensure that risk is kept within permitted levels, and through the annual Integrated Risk Monitoring Report from the World Bank Group's Chief Risk Officer. Clarify roles and accountabilities for setting GAC strategic Partially agreed. Management concurs with the view that CASs priorities: and ISNs be the primary means by which client governance issues At the country level, Bank country strategies should continue are raised and addressed. The Bank is committed to the Paris and to serve as the primary mechanisms for reflecting the priorities Accra agendas on alignment. Management agrees that the GAC and needs of clients on GAC issues. Donor-funded initiatives Council should focus on strategic and institutionwide issues. need to be appropriately aligned. At the VPU level, GAC work plans should be informed by de- Management is unconvinced of the need to create add itional GAC mand in partner countries and should set priorities for overall work plans at the VPU level. There is a danger that these would resources use-both Bank budgets and trust funds. In keeping create an additional task that would be of limited addit ional value. with Bank policies on the integration of trust fund allocations with the budget process, decisions on allocations of trust Management commits to restructure the GAC Council, with a funds to GAC activities should involve line management in further increased focus on strategic and institutionwide issues, in Vice Presidency Units (VPUs). the context of GACII and will report in the context of a m id-term At the corporate level, the GAC Council should focus on GAC update. institution-wide issues and risks, and on benchmarking the GAC responsiveness of Bank operations. Align GAC implementation arrangements with Bank adminis- Partially agreed. Management agrees that articulating a clear trative and operational processes by: results framework for GACII is critical. GAC Phase II in particular Consolidating current fragmented financing arrangements will lay out a clear Bank-wide results framework, with realistic yet (that is, separate Bank budget and trust fund allocations) while stretching targets which will include baseline indicators, and a improving monitoring of GAC activities. functioning monitoring system. Management also agrees on the Rather than creating a separate cadre of GAC specialists, ap- importance of applying GAC competencies across sectors, and plying GAC competencies across existing Bank networks and on the importance of applied research to support learning and career streams, and allow transferability of GAC -competent knowledge management. staff across networks. Streamlining specialized GAC institutional arrangements with However, Management is of the view that line managers are a view to empowering line managers in VPUs to achieve GAC already empowered to achieve GAC objectives, and that the objectives. competit ive process for allocating GAC Trust Fund monies was Supporting increased applied research on what works in vari- effective and appropriate. ous GAC areas. xxvi I World Bank Country-Level Engagement on Governance and Anticorruption Management Action Record (continued) Major Monitorable lEG Recommendation Management Response Requiring a Response Developing a results framework that includes baseline indicators Management commits itself to: of Bank and country-level performance, sets targets, and inte- Set out as noted above a GAC results framework and monitor- grates monitoring of GAC responsiveness into standard portfolio ing system in the context of GACII. monitoring. Finish ongoing work on "GAC competencies" and an associated training program and put them into use by FY13. As noted above, the oversight and management arrangements for GACII will be revised. Set out priorities for research in the context of GACII and moni- tor progress in the context of regular GAC reporting. Chairperson's Comments xxvii Chairperson's Comments: Committee on Development Effectiveness (CODE) The Committee on Development Effectiveness (CODE) Members agreed that the evaluation provides a compre- considered the report entitled World Bank Country- hensive review of the Bank's operational responsiveness Level Engagement on Governance and Anticorruption: An to GAC issues in country operational aspects, sector Evaluation of the 2007 Strategy and Implementation Plan programs and projects before and after the 2007 strat- (CODE2011-0044), prepared by the Independent Evalu- egy. Notably, they were pleased with lEG's findings that ation Group (lEG) , together with the Draft Manage- the Bank supported institutional strengthening in three ment Response (CODE20 11-0046). The Statement by the times as many countries in the fiscal 2008-10 period as External Advisory Panel (COD E20 11-0047I 1) on the lEG it did in the fiscal2004- 07 period. But they also agreed report was distributed as a background document. with lEG that the Bank has to significantly strengthen its efforts to work with clients to improve their governance The Committee welcomed the timely discussion of the systems going forward. As the Bank prepares for the lEG evaluation assessing the 2007 Governance and next GAC phase, members encouraged Management to Anticorruption (GAC) strategy and the first phase of do more to further support country-owned reforms, in- its implementation (FY08-10). It appreciated Manage- novate approaches to institutional capacity building, and ment's draft response and commitment to consider lEG's foster the demand for good governance while strengthen- findings and recommendations as inputs to the develop- ing country dialogue. There were also comments on the ment of GAC Phase II, including in the areas of country importance of focusing more on strengthening country institutional strengthening, risk management, and results systems, monitoring GAC results, harmonizing risk man- measurement. The Committee noted the complexity agement, and strengthening the understanding of the of GAC issues and the relevance of this evaluation in political economy context taking into account the Bank's connection with ongoing work on the design of the new own analytical work as well as analysis prepared by financial instrument-Program-for-Results; the need to others. With regard to resources to implement the GAC look at implementation of the GAC strategy within the strategy, comments were raised on integrating the alloca- Bank and in partner countries; and the alignment of the tion of trust funds to the Bank's budget process, and to GAC agenda with the Bank's modernization and internal strengthen staff skills building internal GAC expertise reform agendas. across the institution. xxviii 1 World Bank Country-Level Engagement on Governance and Anticorruption Statement of the External Advisory Panel Governance is a defiJ;ting challenge of our era. The of public services. Third, the processes through which Independent Evaluation Group's report, focusing on the transparent mechanisms of oversight by the legislature national dimensions of the World Bank 2007 Governance and civil society organizations have been established and and Anticorruption (GAC) strategy, is therefore of global public participation has become systemic. importance. We welcome the serious and detailed com- ments of management in response to the draft report's Understanding of corruption as a system rather than findings and recommendations, and appreciate the pro- mere moral failure is critical to bringing a political fessionalism of lEG staff for defending and refining their economy perspective to the analysis of good and bad main conclusions. The exchange is a sign of organization- governance. Elites in such systems are highly vested allearning and a tribute to lEG's independence. in perpetuation of corruption and often do not want reforms to succeed. Additionally, the elites that champion Despite the centrality of GAC to the Bank's poverty a particular set of reforms for mobilization of resources reduction mandate, the design of the 2007 strategy and and consolidation of their power may become the largest its implementation did not match the ambitious vision constraint to reforms in a subsequent phase. Declaration of the organization. The focus was largely internal rather of early victory, betting on individuals as champions, and than on strategic issues facing reformers in countries. awarding the title of reformer to individuals and coun- Strong incentives and accountability frameworks were tries without having clear criteria for depth and breadth not created and funding arrangements did not allow for of reforms that would result in a systemic tipping point achievement of stated objectives. Analytic and opera- are, therefore, to be avoided. Tolstoy observed that all tional work remained misaligned, gaining traction only families are dysfunctional but each in their own distinc- sporadically. Especially important are the findings that: tive way-and the same could be said for countries. "overall the Bank's operational response to GAC issues There is, therefore, the need for tailored interventions demonstrated continuity without systematic improve- within given contexts and efforts to continue support for ment as yet"; and "important opportunities to managing reforms for the duration required to consolidate systemic risks and developing innovative operational solutions change. This also implies that when corruption and have yet to be seized:' abuse has become so institutionalized and systemic on such a grand scale in a government- and where there is As the Bank is committed to a second phase of GAC, lEG no government leadership to reform-the Bank should makes a series of sensible recommendations for achiev- be prepared to stop lending to counteract the lending ing operational effectiveness. We associate ourselves with imperative. these recommendations but would like to raise a broader series of issues. The Bank as an organization, however, has some bind- ing constraints in dealing with GAC at the strategic level. Whether corruption is an exception or the norm can pro- First, in relation to the analysis above, as an institution vide the first step in terms of classification. Useful lessons it is not always able to conduct accurate and timely can then be drawn from the cases of outliers as to how political-economy evaluations of given country contexts. the organizational culture of the state can change from High staff turnover and technical approaches to analytic corruption to accountability. Three areas would be of work can prevent deep understanding of dynamics that particular importance here. First, the processes through can affect positive and negative change. Second, there is which a critical mass of public support can be galvanized a tension, as lEG documents, between the Bank's lending against systemic corruption to drive the demand for imperative and its rhetoric on governance. As a result, accountability past a tipping point. Second, the process the informal but operating rules of the game in the orga- through which control systems can be established to nization remain misaligned with the formal emphasis on transform hierarchy into effective and efficient delivery good governance and accountability. Statement of t he External Ad visory Panel xxix Second, the project-based model oflending is an ap- experience of outliers, ranging from Singapore and Malay- plication of a Fordist model of mass production to messy sia to Qatar and the United Arab Emirates. social realities of the 21" century. There is an implicit but questionable assumption that micro-level interven- Fourth, the Bank's development policy has a mixed tions through ring-fenced mechanisms can produce record. Despite the consensus on the political economy macro-level changes. Given Bank efforts in ring-fencing underpinnings of reform and elite interests, Bank poli- its projects, it would be revealing for lEG to undertake cybased lending has been based on "stroke of the pen" an analysis of how these projects fare in terms of effi- reforms that prove reversible. ciency, cost effectiveness, transparency, and sustainability compared to those managed directly by the government Fifth, the Country Assistance Strategy (CAS) is a misno- or through the private sector or the nongovernemental mer, for it is not a strategic, but rather a lending program organizations. hammered together through compromises among sectors justifying their bureaucratic turfs. Neither management Third, the Bank has not had the skills base to deal with nor the Board has ensured adherence to the discipline of governance on a systematic basis. Its public sector port- selectivity. folio has lacked distinction and, as lEG states, it missed the opportunity to use the GAC strategy to bring order to These constraints raise a fundamental question. Can the this portfolio. Economists, engineers, and other profes- Bank-designed with the assumptions of the mid-20'h sionals have consequently been assuming leadership roles century, and with its own distinctive political economy for programs and projects on governance, improvising and organizational culture and lending imperatives-be to the best of their abilities rather than being guided by a turned into the premier instrument for promotion of coherent and focused approach. governance and the fight against corruption? Answering this question requires a deeper examination of change This problem is compounded by the Bank's reliance on management in the Bank and the pivotal role of Bank large-scale technical assistance. Over-reliance on techni- presidents as tone setters. The GAC strategy was the cal assistance alone as a remedy for governance failure is signature program of Paul Wolfowitz, the embattled bound to fail, but that technical assistance used judiciously president who became the first leader of the organiza- in the right circumstances can be very useful, as Paul Col- tion to resign under staff pressure. lEG has offered a set lier has argued in The Bottom Billion. Lacking a coherent of recommendations for launching the second phase of approach to the basic building blocks of finance minis- the GAC strategy. We are arguing that the Management tries, the Bank relies on vendors to provide such services, and the Board have a larger challenge. If the Bank is to which can lead to inefficiencies and mismanagement. It is become a catalyst of good governance it must first change little wonder, therefore, that the Bank lends for the same its own inherited governance of the twentieth century reforms over and over again. A number of governments and embrace a twenty-first century form of governance around the world have transformed auditing and account- and organizational culture that can make it lead by ing in profound ways as tools of democratic accountability. example. This is the challenge that the Board and the Moreover, there are major lessons to be learned from the management must confront in earnest. Dr. Ashraf Ghani Chairman, Institute of State Effectiveness Ms. Monica Macovei Member of the European Parliament and former Minister of Justice of Romania Mr. Andrew Natsios Professor in the Practice of Diplomacy, Georgetown University, Walsh School of Foreign Service xxx I World Bank Country-Level Engagement on Governance and Anticorruption 1 World Bank Engagement on Governance and Anticorruption: A Historical Summary For more than two decades, the World Bank has sought to make governance and anticor- ruption integral to its work on economic growth and poverty reduction in developing countries. Governance and anticorruption refers to an objective of Bank assistance, that is, to develop capable and accountable public institutions that formulate and implement sound policies, provide public services, set rules governing markets, and combat corrup- tion. It also refers to an approach to development assistance-one that enlists countries and their partners in ensuring that development resources are channeled to their most ef- fective use and protected from fiduciary risks relating to weak governance and corruption. The Bank is currently implementing its 2007 strat- The evaluation attempts to respond to the concerns of egy, Strengthening World Bank Group Engagement on diverse stakeholders inside and outside the Bank. In Governance and Anticorruption (henceforth the GAC preparing the evaluation, lEG consulted with a wide ar- strategy). That GAC strategy and its implementation ray of actors, including former and current World Bank plan are the focus of this evaluation. Management has senior management; headquarters and country-based staff reported extensively on GAC implementation through involved in GAC implementation; borrowing governments annual progress reports as well as several discus- and non-state actors in selected countries; academics and sions with the Bank's Board of Executive Directors. In policy experts; civil society organizations, including some response to a request from the Board's Committee on of those originally consulted in the preparation of the Development Effectiveness (CODE), the Independent strategy; donors involved in supporting the GAC strategy; Evaluation Group (lEG) has undertaken this evaluation and the GAC Group of External Advisers. of the first phase of GAC implementation, covering the 2007-10 period, to inform a planned second phase of Stakeholders identified four areas where the evaluation the GAC strategy. should add value to the Bank's GAC work. First, they suggested that the evaluation help develop a framework for assessing progress on GAC implementation at the coun- Objectives of the Evaluation try, sector, and project levels. Second, it should identify The evaluation aims to help enhance the Bank's ap- implementation challenges and their root causes, so that proach to governance and anticorruption and to they can be addressed in the second phase of the GAC. improve its effectiveness in helping countries develop Third, it should seek to identify good-or at least better- capable and accountable states that create opportunities practice in various aspects of GAC work. Finally, it should for the poor. Pursuant to this objective, the evaluation as- ensure that lessons learned are extensively disseminated to sessed the relevance of the 2007 GAC strategy and imple- the diverse stakeholders that will continue to be involved mentation plan, as well as the efficiency and effectiveness in GAC implementation. of implementation efforts in making Bank engagement with countries and other development partners more re- Organization of the Report sponsive to GAC concerns. It also sought to identify early lessons about what works and what does not in helping to This report is organized into eight chapters. This promote good governance and reduce corruption. chapter puts the 2007 GAC strategy in the context of the Bank's long history of involvement in governance issues. varying degrees of success-influenced the direction of Chapter 2 presents the design of the evaluation, includ- Bank work. 1 ing its logical framework, scope, main questions, and analytical methods. Chapter 3 evaluates the relevance The Bank's governance agenda also reflects its attempts of GAC strategy objectives and the appropriateness of to satisfy multiple, and, at times, competing interests GAC strategy and implementation plan design. Chapter 4 (Weaver 2008). The Bank's diverse shareholders, as well reviews the incentive and implementation arrangements as diverse groups of external stakeholders in civil society, that supported GAC rollout, including financing, stra- have shaped the governance agenda. Some shareholders tegic staffing, and accountability and oversight. Chapter and civil society organizations have increasingly voiced 5 discusses the fulfillment of commitments under the strongly-held views that the Bank should do more, not implementation plan, including the strengthening of key less, in support of good governance. Others have called Bank controls, the provision of guidance and tools, and for restraint, particularly in light of the limited success the delivery of support to Bank teams. Chapter 6 pro- of governance-related efforts. The Bank's approach also vides an overview of the responsiveness of Bank opera- sought to respond to partner country concerns that how tions as well as the commitment of Bank operational it worked (that is, its choice of aid modalities) could itself staff to GAC issues. Chapter 7 provides a snapshot of help or hinder governance prospects in aid-dependent early outcomes of Bank engagement on GAC issues (for countries. example, in areas such as core public management, roads, and education). Chapter 8 makes recommendations for From Fiscal Adjustment to Public Sector Management, future Bank efforts. 1983-1996. The World Bank's interest in the quality of government began in the mid-1980s with a relatively narrow focus on the fiscal impact of unrestrained Evolution of the Bank's Engagement on public sector wage bills. 2 Over this period, Bank-sup- Governance and Anticorruption ported structural adjustment programs, particularly in To be credible, any review of the 2007 GAC strategy Africa, drove pay and employment reforms that sought should be based on an overview of the Bank's engage- to reduce wage bills, decompress wage ratios, downsize ment on governance and anticorruption over two bloated civil services, and rationalize ministries and decades. That engagement evolved through three stages: agencies. However, the "short time horizon, narrow (i) a focus on the quality of government between the prism, and supply-driven nature of adjustment lending" mid-1980s and mid-1990s; (ii) the emergence of gover- were ill-suited to the goals of sustainable performance nance as a key pillar of poverty reduction from the mid- improvement in the public sector (lEG 1999 and Jenkins 1990s through the mid-2000s; and most recently, (iii) the and Plowden 2006). In some cases, investment lend- establishment ofGAC as a corporate strategy in 2007. ing was also used to support institutional development. These efforts were hampered by the limited flexibility of At each stage, the Bank's approach to governance the traditional project instrument. Bank support to the issues evolved as it sought to respond to numerous Africa Capacity Development Foundation, starting in factors. Over the past two decades, major historical 1991, attempted to bolster regional initiatives. Overall, events-the collapse of communism in Eastern Europe in these early efforts produced mixed results. the early 1990s, democratic transitions in Africa during the 1990s, the East Asian financial crisis in 1997, and the Evaluations of these initial actions stressed the im- emergence of several countries from prolonged conflict- portance of addressing the root causes of poor public underscored the importance of developing inclusive, sector performance-that is, poor governance. Inde- transparent, and durable state institutions. These events pendent and self-evaluations pointed to the political costs converged with advances in knowledge on the role of and long gestation period of bureaucratic reforms, as well institutions in economic development, which was seen as the need for better country knowledge and more flex- as a compelling reason for the Bank and other donors ible aid instruments to support institutional change (lEG to deepen their engagement on governance issues (lEG 1999 and 2008). These reviews also pointed to governance 2008 and World Bank 1997a, 2002, 2004, 2005b, and or the exercise of public authority-not only institutional 2005d). In addition, a growing body of evaluative work capacity or formal structures-as a key determinant of by lEG and others helped identify lessons, which-with public sector performance. During this period, by clari- 2 World Bank Country-Level Engagement on Governance and Anticorruption fying the legal basis for its involvement in governance and reached 35 percent in FY02. This trend was sus- issues, the Bank set the stage for an expansion of support tained for more than a decade until it was eclipsed by for institutional capacity building (IEG 2008, World Bank significant crisis-response lending in 2008. A similar 2000a, Levy and Kpundeh 2004, Thomas 2007). trend was observed in the share of governance-related prior actions for development policy lending. They grew Poverty Reduction through Good Governance, 1996- in prominence as an instrument of supporting improve- 2006. In the mid-1990s, the Bank committed itself to ments in public sector governance, and public financial tackling the "cancer of corruption" in its own projects management became a mainstay of Bank support in and in its support for country efforts to promote good both International Development Association (IDA) and governance (Wolfensohn 1996 and World Bank 1997b International Bank for Reconstruction and Develop- and 2000b). It launched a strategy to help countries ment (IBRD) countries (World Bank 2009a). Investment combat corruption and announced a "zero tolerance" in public sector capacity, including in the infrastructure policy with regard to fraud and corruption in its projects. and social sectors, continued to account for a sizeable The Bank also embarked on a significant policy research share of Bank assistance to IBRD, and to a lesser extent, agenda. For instance, a series of World Development Re- IDA countries. Bank assistance for capacity building, ports noted the importance of institutions, capacity, and in particular, focused mainly on the core pubHc sector, governance for public service delivery, the investment sector administration, and workforce development. More climate, and poverty reduction (World Bank 2004 and modest commitments were made in support of private 2005b). The Asian financial crisis and greater attention to sector and demand-side capacity building. the problems of conflict-affected states further validated the Bank's interest in helping develop market and state The Bank also became more selective in its allocation institutions. of scarce aid resources. Over its Twelfth and Thirteenth Replenishments, IDA's Performance-Based Allocation The Bank followed through on this new thinking with system increased the effective weight given to governance a 2000 strategy, Reforming Public Institutions and performance in countries. Regression analysis under- Strengthening Governance. The strategy set a course taken for this evaluation showed that, by the mid-2000s, to help strengthen core public institutions such as civil IDA was the most selective of donors. According to the service and public financial management systems, regula- Bank, IDA's allocation rules ensured that aid resources tory bodies, the judiciary, and local governments. It also were directed to settings where fiduciary and develop- proposed a broader menu of products to support public mental risks could be most effectively managed. They sector reforms, for instance, more programmatic lending also signaled the importance of governance as a devel- instruments, new tools for measuring institutional qual- opment goal: a growing number of country assistance ity, and participatory processes to help reform constitu- strategies (CASs) in IDA countries were including gover- encies (World Bank 2000b). In parallel, the Bank intro- nance pillars. Convinced by the logic of the Performance- duced the "governance cluster" to its established Country Based Allocation, other donors, such as the Netherlands PoHcy and Institutional Assessment (CPIA), launched the and the United Kingdom, aligned their own aid aJloca- widely-cited World Bank Institute (WBI) indicators (now tion decisions accordingly (IDA Deputies' Reports from the Worldwide Governance Indicators), and instituted IDA13-14; and IDA 13, IDAI3 Midterm Review, IDA 14 standardized assessments of country fiduciary con- papers; Hout 2004). trols (through the multiagency Public Expenditure and Financial Accountability partnership). The strategy also The Bank was also attempting to adopt new aid intended to strengthen the Bank's own internal organiza- modalities in order to support country ownership tion and skills base. and domestic accountability, in line with the Paris Declaration. The launch of country-led Poverty Reduc- What followed was a watershed for the Bank's work on tion Strategies (PRSs) in the early 2000s provided an governance. Support for governance-related themes impetus for the Bank to innovate "how" it worked in grew significantly as a share of total Bank lending IDA countries (lEG 2010). Efforts to update operational (Figure 1.1). Bank assistance for governance (inclusive approaches involved analytical and advisory activities to of public sector reform) grew to more than 25 percent develop country-led governance strategies within PRSs; of total Bank lending in volume terms starting in FY99, a shift to programmatic budget support and multisector World Bank Engagement on Governance and Anticorruption: A Historical Summary 3 Bank Support for Governance-Related Themes, FY1990-201 0: Number of Projects and Lending Volumes (US$ Millions) 10,000 40 ... ~ c Ql 9,000 35 E § ~ c Ql 8,000 30 E ·e ~ E 7,000 8 ·e ~ E 6,000 25 < 0 u 0 5,000 20 a: a < 0 4,000 !e 15 ;; a ~ a: !e 3,000 2,000 10 ~ 5 - 0 ~ c Ql 1,000 ~ 0 0 ~ FY90 FY92 FY94 FY96 FY98 FYOO FY02 FY04 FY06 FYOS FY10 0 DPL PSG 0 IL PSG - Share in totai iBRD/IDA commitments Source: World Bank database. Note: Includes investment projects and development policy operations with prior actions covering the following themes: (25) Administrative and civil service reform, (26) Decentralization, (27) Public expenditure, financial management, and procurement, (28) Tax policy and administration, (29) Other accountability/anticorruption, (30) Other public sector governance, (90) Managing for development results, (31) Access to law and justice, (32) Judicial and other dispute resolution mechanisms, (33) Law reform, (34) Legal institutions for a market economy, (36) Personal and property rights, (40) Regu- lation and competition policy, (57) Participation and civic engagement, and (73) Municipal governance and institution building investment operations; harmonization efforts with other countries. Global poll respondents agreed that efforts donors (for example through sectorwide approaches, to reduce corruption remained one of the Bank's least or SWAps). In addition, the Bank committed itself to effective "mission areas:' increasingly use-rather than bypass-country systems (World Bank 2003a). In public sector reform, the Bank However, new efforts were creating operational chal- attempted innovations in its financial and nonfinancial lenges. Chief among these was how to support institu- product lines to provide added incentives for institutional tion building, particularly in fragile states. For instance, change. a 2005 lEG evaluation found that capacity building efforts in Africa were fragmented. The Bank often lacked By the mid-2000s, these efforts produced some tan- the knowledge base and programmatic tools required gible results, for example, in public financial manage- to make a lasting impact, particularly in sectors. Soon ment and revenue administration. According to a 2008 after the evaluation, a 2005 World Bank Africa Capacity lEG review, Bank support in those two areas gained Development Taskforce echoed these concerns. It defined traction. So did Bank support for merit-based recruit- capacity development as a "governance challenge"-one ment and promotion. Outside observers began to take that required a balance between state building and social notice of these improvements in the Bank's approach, accountability. The taskforce recommended that the Bank even by the early 2000s. 3 A 2002 global poll of opin- consolidate its existing capacity-building business lines ion leaders across regions-in particular in Africa- and expand new ones (for instance, in areas relating to the noted that "improvements outweighed setbacks" in demand side). It also identified the need for more flexible the Bank's efforts to help strengthen governance. Both lending instruments, a more proactive stance on the use lEG and outside observers agreed that certain aspects of country systems, and reform of donor approaches to of Bank support did not produce the desired results technical cooperation. (for instance, some aspects of civil service reforms and anticorruption efforts). According to lEG, direct mea- Notwithstanding some progress, the implementation sures, such as the promulgation of anticorruption laws of Paris Declaration principles was proving more diffi- and the establishment of anticorruption commissions, cult than anticipated. In a growing number of countries, did not reduce the perceived incidence of corruption in the Bank and other development partners were jointly 4 World Bank Country-Level Engagement on Governance and Anticorruption supporting policy and institutional reforms through themes. They recognized the potential for states to act budget support programs. However, efforts to harmonize not only as "facilitator[ s] of networks [of organizations donor approaches and use country systems on invest- inside and outside the public sector]" but also as ve- ment projects tended to lag. In addition, harmonized hicles for elites to safeguard their interests and preserve donor programs faced special challenges during political power (OECD 2009). They also accepted that governance governance crises. Donors that attached explicitly politi- reforms required the right political incentives, credible cal conditions to budget support programs felt compelled champions, and appropriate demand-side pressures. to respond differently from those that focused on eco- nomic issues. More generally, by the mid-2000s, donors Governance and Anticorruption as "Everybody's Busi- were becoming increasingly aware of the potential unin- ness," 2007-10. Beginning with the arrival of Paul tended consequences of their programs-budget support Wolfowitz as World Bank President in 2005, gover- in particular-on the political economy of governance nance and anticorruption issues gained an unprec- reforms in partner countries (Barkan 2009; Thomas 2007; edented level of attention. A very public and sometimes Langbein and Knack 2010). contentious discussion surrounding the Bank's approach to governance culminated in a highly negotiated 2007 The Bank's use of governance indicators was coming GAC strategy document (Weaver 2008). Now considered under increasing scrutiny and critique from experts, "everybody's business;' the 2007 strategy was not simply academics, and partner countries. Composite indices, a sector strategy but a corporate strategy that sought to such as the Worldwide Governance Indicators, Doing change the way the Bank did business. Business, and others, were based on empirical research on the links between institutional quality and development The 2007 strategy set forth several objectives relat- outcomes. Through its support for these and other indi- ing to the development of capable and accountable ces, the Bank demonstrated the power of evidence-based states and committed the Bank to seven principles of dialogue on governance issues, as well as the potential for engagement on GAC issues (Box 1.1). In response to benchmarking governance performance across coun- shareholder concerns about the perceived arbitrariness tries. By the mid-2000s, several reviews highlighted the of senior management decisions to cut off lending to limitations of these indicators. For instance, composite certain countries, the strategy reiterated the Bank's use of scores such as the Worldwide Governance Indicators rules-based criteria for allocating resources, as well as its were relative rankings of countries within a given period aim to stay engaged even in poorly governed countries and not meant for time-series analysis. Similarly, some to ensure that the "poor do not pay twice:' At the same indices sought to integrate multiple source indicators, time, the strategy placed considerable emphasis-more each of which measured distinct governance phenomena. than earlier strategies-on safeguarding Bank funds Correlations between measurement errors across source from fiduciary risks. Early on, it was acknowledged that, indicators were also a problem (Arndt and Oman 2006 to achieve its "vision of success;' the strategy required a and Thomas 2010). For their part, partner countries also more detailed implementation plan. started to voice concerns that changes in rank ordering of governance performance did not necessarily reflect the Salient Features of 2007 GAC Strategy achievement of reform efforts. Implementation Plan These challenges implied the need for an updated Bank The implementation plan (IP) sought to define con- approach. In fact, several other donors were already crete steps for "what the World Bank itself will do to rethinking their strategies on governance. For instance, support the GAC agenda, and how it will work with by 2005-06, several bilateral agencies-among them the governments, domestic stakeholders, and develop- U.S. Agency for International Development (USAID) , ment partners to support country-level governance U.K. Department for International Development (DFID) , improvements and regional and global initiatives?' The Danish International Development Agency (DANIDA), plan's success was to be measured by (i) a significant and French Agency for Development (AFD) , Swedish Inter- growing number of countries seriously addressing key national Development Cooperation Agency (SIDA), and governance impediments to development effectiveness the European Commission-had launched new strategies and poverty reduction; (ii) Bank-supported projects and on governance. These strategies shared some common programs increasingly addressing GAC impediments; and World Bank Engagement on Governance and Anticorrupti on: A Historical Summary 5 1. The GAC strategy had several objectives: • "to support poverty reduction ... :· • ".. . [by] developing capable and accountable states .... [undertaking] sound policies, improving service delivery, [establishing] rules for markets, combating corruption;• and • ".. .to ensure that its funds are used for their intended purposes:• 2. In addition, the"GAC guiding principles" were as follows: • Focus on "[a] capable and accountable state to create opportunities for poor people, provide better services, and improve development outcomes:· • Country ownership and leadership are key. Country government is the principal counterpart. • Remaining engaged so the poor do not pay twice. • Consistent approach across countries, even though one size does not fit all. • Engage broad set of stakeholders with focus on transparency, accountability, and participation. • Strengthen rather than bypass country systems. • Harmonization (the Bank will not act in isolation). Source: World Bank documents. (iii) countries and global partners valuing and respecting a more targeted effort involving 18 countries sought the Bank's capacity in this area (World Bank 2007a). It to enhance GAC responsiveness with the help of was envisaged that these objectives would eventually be considerable support provided under the Window reflected empirically in improvements in country gover- One facility of the Governance Partnership Facility nance performance. (GPF). 4 • GAC-in-Sectors and GAC-in-Projects. These efforts To this end, the IP proposed to deliver guidance aimed to strengthen incentive and accountability materials, tool, training, incremental resources, and frameworks in sector dialogue and project design, strategic staffing to help deepen Bank enegagement in as well as systematic risk assessment and man- the following areas: agement (for example, through the use of political economy analysis, actionable governance indicators, • GAC-in-Countries. These initiatives sought to and demand-side measures). Guidance notes and enhance Bank-country engagement on governance toolkits were designed to advise Bank teams on how and anticorruption issues. Initially, country-GAC to address GAC issues in the sectors and to support (CGAC) processes-comprising joint workshops, cross-cutting concerns, such as social accountability peer-to-peer learning events, clinics, and upstream (World Bank 2008b and 2009c). Also included were assessment activities-were launched in 27 coun- handbooks, tools, and training to support efforts to tries to help Bank teams systematically diagnose prevent fraud and corruption in projects. 5 A 2009 governance challenges and identify ways of address- Quality Assurance Group (QAG) survey of projects ing them through CAS design, sector strategies, approved in FY08 aimed to establish a baseline for and project preparation. The CGAC processes were incorporation of "generic" GAC elements in projects intended to deepen the Bank's understanding of (World Bank 2009f). what can be done to strengthen GAC in CASs and help identify governance entry points (for example, • Global GAC Efforts. The Bank proposed to increase core public management and accountability institu- its involvement in peer learning networks and col- tions, private sector engagement, and demand-side laborative governance initiatives. These included the capacities and frameworks) . Following the CGACs, Extractive Industries Transparency Initiative (EITI) 6 World Bank Country-Level Engagement on Governance and Anticorruption and global and regional legal conventions such as the Resourcing the Strategy. Significant incremental bud- Stolen Asset Recovery (StAR) Initiative, 6 Medicines getary and donor resources were deployed over the Transparency Alliance (MeTA), and Construction FYOS-12 period to support GAC implementation. This Sector Transparency (CoST) Initiative. In addition, comprised $54 million in incremental Bank budget as the Bank sought to harmonize GAC policies (for well as $61 million in donor funds allocated through the example, on cross-debarment) with other multilat- GPF. The GPF was supported by the United Kingdom, eral development banks, and to establish common the Netherlands, and Norway. response principles for high -risk countries under the auspices of Gov-Net. 7 Change Management. GAC implementation was viewed as a significant change management agenda. Institu- Internal Reforms. Other important internal reforms tional arrangements to support this Bank-wide initia- were carried out as complements to the GAC agenda, tive prominently featured a GAC Council, consisting of including implementation of Volcker Panel recommen- the Vice Presidents and chaired by the Managing Direc- dations on strengthening the Integrity Vice Presidency, 8 tors. The Council was supported by a GAC Secretariat, launch of a new WBI strategy emphasizing multistake- various other partnership secretariat units, and GAC focal holder engagements (World Bank 2009a), update of the points in Regional and network units. The Bank also peri- Bank's disclosure policy, launch of a new Operational odically sought the advice of a Group of External Advisers, Risk Assessment Framework (ORAF) for investment an Independent Advisory Board (that advises the Presi- lending, 9 the recruitment of a Chief Risk Officer, an dent and Audit Committee on Integrity Vice Presidency annual integrated risk monitoring report, and other performance), and an International Technical Advisory efforts to modernize investment lending (World Bank Group (that advises on the Use of Country Systems pilot). 20lla). World Bank Engagement on Governance and Anticorruption: A Histori cal Summa ry 7 2 Design of the Evaluation This chapter describes the design of the evaluation, including its logical framework, scope, main questions, and analytical methods. It reflects the Approach Paper, which was en- dorsed by the Committee on Development Effectiveness on June 8, 2010. In designing the evaluation of the 2007 GAC Strategy and Implementation Plan, lEG drew on the history of Bank engagement on governance issues and on previous evaluations. Framework ability) and "outcomes" (poverty reduction) . Each is described below. A logical framework, which linked GAC inputs to outcomes, formed the basis for this evaluation. Even Inputs-GAC Strategy and Implementation Activities. though the GAC strategy did not contain a formal logi- The strategy, the rolling implementation plan, and cal framework, the evaluation developed a results chain in order to make informed assessments of the Bank's implementation activities were the inputs designed to efforts (Figure 2.1). The results chain linked "inputs;' improve the Bank's ability to engage on GAC issues such as the GAC strategy and implementation activi- at the country, sector, and project levels as well as in ties, to "outputs;' that is, more GAC-responsive Bank international efforts. The GAC inputs include com- engagement in partner countries. More GAC-responsive munication of GAC strategic principles and supporting engagements in turn contribute to "intermediate out- operational policies and internal controls; guidance and comes" (enhanced state capacity and social account- support for risk review processes (for example, CGAC GAC Results Chain Inputs: GAC in Countries, Sectors, Projects & Global Strategy, IP, Policies, Guidance, Tools, Training, Communities of Practice, Funding and Staffing Outputs: GAC-Responsive Bank-Country Engagement Selectivity, System Strengthening, Signaling, Smart Design Intermediate Outcomes: Capable & Accountable States und Policies, Improved Service Delivery, Rules for Markets, Anticorruption Outcomes: Poverty Reduction Growth, Empowerment, Expanding Opportunity, Reduced Vulnerability Source: lEG approach paper. 9 processes, governance filters , red flags); development • Smarter design of projects by countries: The Bank of new product lines (for example, political economy and its clients should become better equipped to analysis); provision and rollout of operational tools, design innovative or "smarter" projects that are training, and advisory activities (for example, on action- cognizant of GAC issues. A "smarter" approach, able governance indicators); development of communi- according to the GAC strategy, ensures that project ties of practice (for example, GAC-in-projects, demand design and implementation arrangements are fitted for good governance, and political economy analysis), to the political economy. In addition, it would include strategic staffing (for example, GAC Advisers), and incre- appropriate measures to prevent fraud and corrup- mental resourcing (for example, budgetary resources and tion, to use country systems, and to employ transpar- donor funds); and change management arrangements. ency measures, such as third-party monitoring, so that citizen stakeholders are empowered to hold state Outputs-GAG Responsiveness of Bank Engagement. actors and service providers accountable. (Adapted The inputs are intended to enable the Bank and its from QAG. See World Bank 2009f.) 11 Smarter design country partners to better address GAC-related issues, therefore should contribute to intermediate gover- and to help relieve GAC-related constraints to poverty nance outcomes as well as higher-order development reduction. The main "outputs" are higher quality pro- outcomes. grams, portfolios, projects, and analytical and advisory activities that consistently and cost-effectively address • More effective strengthening of country institu- GAC concerns and risks. These "GAC-responsive" Bank- tions and systems: GAC-responsive Bank programs country engagements are characterized by (adapted from are characterized by borrower-led efforts to system- QAG; see World Bank 2009f): atically strengthen country systems. These include capacity building of cross-cutting and sectoral state • Enhanced selectivity of Bank country strategies and institutions (for instance, civil service and budget programs: In more selective Bank programs, candid management systems, revenue administration, local assessments of governance and political economy governments, ministries, and agencies) , account- risks would shape decisions regarding lending levels ability institutions (for instance, judiciaries, supreme and composition (or aid selectivity). 10 They would audit, and anticorruption bodies), and the private help identify viable governance entry points (such sector and civil society (World Bank 2005d). as core public management institutions, sectors, More GAC-responsive Bank portfolios theoretically demand-side and accountability institutions, or the are higher performing portfolios and therefore should investment climate), help clarify the rationale for the result in improvements in traditional measures of portfo- choice of financial and knowledge instruments (for lio quality. For instance, upstream diagnostic GAC efforts example, development policy versus investment lend- are expected to contribute to improved risk management, ing) in country portfolios, and strengthen the results design, and, therefore, project performance (World Bank frameworks (for example, through the use of action- 2009g). able governance indicators). • Improved signaling of GAC concerns and risks Intermediate Outcomes-Country Governance Per- through Bank portfolio processes: In GAC-respon- formance. Over time, GAC-responsive Bank support sive Bank portfolios, risks should be regularly and to countries should contribute to more capable and rigorously monitored over the course of implementa- accountable states that create opportunities for the tion (for example, through early warning of fraud poor. These governance improvements generally involve and corruption risks). They would also track the increasing the ability of states to correct market failures progress of governance reforms at the sector and proj- through the provision of public goods (for example, basic ect levels, for example, through the use of portfolio social and infrastructure services) and regulation of mar- management and actionable governance indicators. kets. In carrying out these public functions, governments Disclosure of portfolio reviews and efforts to engage are also responsible for mitigating government failures interested stakeholders in progress monitoring should or weaknesses in formulating and implementing sound be designed to signal progress on GAC issues and policies, providing public services, setting and enforc- promote proactive management of risks by borrowers ing rules governing markets, and combating corruption and the Bank. (World Bank 1997a, 2002, 2004, and 2005a). Improve- 10 World Ba nk Country-Level Engagement on Gove rna nce and Anticorruption ments in the quality of institutions have been captured undertook before-and-after comparisons of the GAC through an array of measurement techniques and gover- responsiveness of Bank operations during the four years nance indicators developed in the past decade. preceding (FY04-07) and the three years following (FYOS-10) the launch of the GAC strategy. Outcomes-Poverty Reduction. GAC outcomes are also poverty reduction outcomes, including: empowerment of citizens, expansion of opportunities through providing Main Questions access to markets and essential services, and provision of Pursuant to the objectives of the evaluation and based on security from vulnerability (including economic shocks, feedback from key stakeholders, lEG sought to address as well as crime, corruption, and violence) (World Bank the following questions: 2002). To what extent was the Bank's 2007 GAC strategy rel- evant? Scope ofthe Evaluation Given the multidimensional and wide-ranging na- i. Was the strategy based on diagnosis of the constraints ture of the GAC strategy, lEG sought to bound the to the Bank's poverty reduction goals and consistent scope of the evaluation by focusing on country-level with the priorities and needs of client countries? operational engagement. Covering other important ii. Was it aligned with the Bank's external authorizing and GAC-related interventions in this evaluation-for in- operating environment? stance, the organizational restructuring of the Integrity Vice Presidency and related reforms to strengthen the iii. Were the objectives, principles, scope, and priorities Bank's corporate investigations and sanctions regimes, of the 2007 GAC strategy coherent, internally consis- the organizational renewal ofWBI, or individual global tent, and realistic? partnership programs (such as the Stolen Asset Recovery iv. Did the strategy address trade-offs in pursuing GAC Program)-would have added considerable complexity to objectives and principles? an already ambitious evaluation. Therefore, it focused on country-level engagement on GAC issues. Was the design of the IP appropriate to GAC strategy objectives? In addition, lEG undertook in-depth analysis in specific sectors and thematic areas that were highlighted in the i. Were the IP's objectives, structure, priorities, as well first and second annual GAC Progress Reports. As such, as results and risk frameworks internally consistent the rollout of GAC activities was intended to expand into and appropriate to the strategy? sectors and thematic areas (for example, in health , educa- ii. Were policies, guidance materials, tools, and training tion, and infrastructure). While the evaluation assesses the activities rolled out under the IP appropriate to the relevance, efficiency, and effectiveness of GAC activities objectives and guiding principles of the strategy? across all sectors, it also includes in-depth analyses of GAC-in-sectors efforts in the roads and primary education iii. Was the distribution of inputs in line with principles that require consistency of treatment of countries, or sectors, and similar efforts in accountability institutions those that aim to make GAC "everybody's business"? (that is, non-executive institutions such as supreme audit, legislative oversight, other independent oversight bodies, iv. How appropriate was the guidance and operational as well as social accountability networks). 12 Criteria for se- support provided in priority areas, such as political lecting these sectors and thematic issues were priorities set economy analysis, and GAC-in-sectors? during early GAC-in-sectors efforts, areas with perceived To what extent was the GAC strategy implemented in higher risk exposure for the Bank, and areas with the line with plans? potential to add to existing lEG work. 1. Were resourcing and change management efforts sup- Finally, the evaluation defined FY04-10 as the period portive of the strategy? of review in order to facilitate pre- and post-GAC comparisons. The 2007 GAC strategy has been under ii. Were incremental Bank budgetary and trust fund implementation since FY08. To assess effectiveness, lEG resources transparently allocated, efficiently executed, Design of the Evaluation 11 and managed with view to ensuring sustainability? elements of both design and implementation processes. (World Bank 2007c) For source material, lEG relied on CASs, CAS Comple- tion Reports (CASCRs), Country Portfolio Performance iii. Was GAC support delivered in a user-friendly and Reviews, and related country program documents, as client-oriented manner in support of Bank teams in well as Project Appraisal Documents, Implementation operations, and ultimately, country partners? Status Reports, QAG analyses, and Implementation To what extent were implementation efforts effective in Completion Reports (ICRs). The analysis also drew on enhancing the GAC responsiveness of Bank-country lEG reports, including CASCR reviews, ICR reviews, engagement? Country Assistance Evaluations (CAEs), Country Program Evaluations (CPEs), and Project Performance i. To what extent has the Bank addressed GAC concerns Assessment Reports (PPARs), as well as on Integrity Vice more systematically in an increasing number of coun- Presidency Detailed Implementation Reviews (DIRs) and tries and sectors? other data, such as summary statistics on investigations. ii. To what extent did GAC implementation efforts have The sampling methodology was designed to facilitate an impact on the responsiveness of operations and three levels of analysis of the responsiveness of coun- Bank staff to address GAC concerns (World Bank try programs and projects to GAC issues (Appendix 2009c)? A). The first level analyzed whether the Bank has been To what extent did the Bank contribute to improved more systematic in addressing GAC issues at the country, governance in countries? lEG sought to identify coun- sector, and project level since the launch of the strategy. tries and sectors with evidence of tangible governance im- A second level assessed whether CGAC/GPF Window provements over the entire FY04-l 0 period. To the extent efforts contributed to improvements in the GAC respon- possible, the evaluation also tried to identify the contribu- siveness of Bank-country engagements. tion of Bank support to improvements in service delivery, financial management, investment climate, and so on. Country Case Studies. Building on the first and second In so doing, lEG was cognizant of two issues. First, the levels of analysis noted above, lEG undertook detailed country-level impact of GAC efforts over the FY08- l 0 pe- case studies of GAC in six country programs. Based riod is, in most cases, too recent to measure. Second, it is on in-depth field visits, this third level of analysis sought difficult to attribute improvements in country governance to identify what has worked and what has not in imple- to Bank support given the myriad other donor-supported menting GAC efforts.13 and indigenous efforts in client countries. Statistical Analyses. To assess the relevance, efficiency, In assessing results, the evaluation also sought to iden- and effectiveness of the GAC strategy, lEG conducted tify early lessons learned about what works to improve statistical analyses of the following issues: governance performance. It also determined whether linkages between GAC-responsive programs and inter- • Selectivity of aid flows from the Bank relative to mediate outcomes are stronger in certain areas compared those from other donors. To assess whether the 2007 with others. These form the basis for recommendations. GAC strategy helped maintain the Bank's policy of channeling scarce aid resources to their most effective use, the evaluation analyzed the degree to which aid Methods of Analysis flows from the Bank were more or less governance- To address these questions, lEG employed multiple ana- oriented relative to other channels (IDA 2009). The lytical methods, as described below: analysis drew on the existing literature as well as on data from the Development Assistance Committee Desk Reviews. To assess the effectiveness of the strategy of the Organization for Economic Cooperation and in improving the GAC responsiveness of Bank-country Development (OECD-DAC) and related data sources engagement, the evaluation conducted a desk review of to capture overall aid flows. Aid flows channeled 50 country programs and 200 lending and trust-fund- through the Bank included IDA, trust funds, and IBRD ed operations over the FY04-10 period. Using stan- resources. The analysis also used CPIA and other mea- dardized questionnaires, the desk reviews assessed GAC sures of governance. 12 World Bank Country-Level Engagement on Governance and Anticorruption • Selectivity of Bank support for capacity building evaluation also reviewed 32 analytical reports compris- relative to support from other donors. Drawing on ing formal economic and sector work (ESW) as well as OECD-DAC and related data sources, lEG analyzed freestanding PEA inputs to Bank strategies, projects, and the roles of various donors in directly supporting policy dialogue. system strengthening for core public sector, sectoral, and accountability institutions and civil society (for Process and Budget Reviews. GAC institutional ar- example, through technical cooperation and capac- rangements as well as resource and risk management ity building) in different country settings over the processes were analyzed. In particular, a detailed analy- FY04- 10 period. sis of the incremental Bank administrative resources and • Distribution of GAC inputs. lEG assessed the donor funds sought to review allocative efficiency, cost distribution of GAC inputs, including allocation of effectiveness, and sustainability. GPF and Bank budgetary resources across countries, sectors, and Bank units. Structured Interviews, Surveys, and Consultations. Structured interviews and surveys were conducted • Relative GAC Responsiveness of Operations. Based with key stakeholders involved in earlier and current on the desk reviews noted above, lEG undertook for- GAC efforts. In-depth interviews with current and for- mal statistical analyses to investigate the determinants mer senior Bank officials as well as key stakeholders were of various elements of GAC responsiveness and their used to ascertain the rationale, trade-offs, and emerg- possible impact on operational performance. ing implementation issues underpinning the 2007 GAC Sectoral and Thematic Analyses. These analyses re- strategy. Also, during the Cape Town GPF Workshop for viewed the evolution of Bank practice in the following Window One countries in September 2010 and during areas: GAC issues in roads, primary education, ac- country field visits, the team consulted extensively with countability institutions, and political economy analy- Bank Country Office staff, government officials, donors, sis. For each of these areas, lEG reviewed the literature, and civil society partners. In addition, lEG surveyed a and identified issues for more in-depth analysis in desk representative sample of operational staff to gauge their reviews and country cases. It subsequently drew on desk awareness of GAC issues and assess their experience with reviews and country case studies to identify portfolio GAC implementation. It also consulted with GPF donors trends in the sector, and identify lessons learned. For the and civil society organizations that provided their inputs thematic review of political economy analysis (PEA), the during the design of the 2007 GAC strategy. Design of the Evaluation 13 3 Relevance of 2007 GAC Strategy and Implementation Plan This chapter summarizes lEG's findings on the relevance of the 2007 GAC strategy and implementation plan. To assess relevance, the evaluation focused on consistency of the GAC strategy with the Bank's poverty reduction mandate as well as with regional, sectoral, and country priorities. It also assessed appropriateness of the IP design. Why and How Governance Matters- and state-building efforts. The limited success of these Recent Currents in the Literature efforts and the persistence of poor governance were often the result of deliberate counterstrategies ad- Literature reviews undertaken for this evaluation opted by elites and other political economy factors. For focused on the specific ways in which patterns of instance, political disincentives, rather than the weak governance affect poverty reduction. Given the recent capacity of providers, often accounted for the limited work by lEG on public sector reform issues, the evalu- access of the poor to high -quality services (World Bank ation did not conduct a comprehensive review of the 2003b). Similarly, influence-peddling offered firms spe- vast and well-established literature on institutions and cial access to subsidies or protection from competition, development. Rather it focused on specific topics (such but it also removed incentives to be dynamic and in- as the political economy of institutional change), as novative (World Bank 2004; Ramachandran, Shaw, and well as specific sectors (such as primary education and Tata 2007). Even when policies were intended to benefit transport) and thematic areas (such as accountability the poor, elites captured redistributive mechanisms to and the demand side). These reviews provided the basis narrowly and regressively target key constituencies. for evaluating the relevance of the GAC strategy to the Stark inequities persisted and perverse policy outcomes Bank's poverty reduction mandate. resulted. Governance and State Building. Over the past two Political economy factors also accounted for diverse decades, a considerable body of work emphasized the patterns of governance across countries and regions importance of governance in promoting broad-based and for distinct poverty reduction challenges. To be economic growth. Specifically, the ability of states to viable, reforms had to be fitted to local conditions. For correct market failures depends in part on their ability to instance: correct governance failures (World Bank 1997a, 2004, and 2005b). Common governance failures include the lack of • Persistent inequality across the Latin America and inclusive and transparent policymaking, allocations of Caribbean region resulted in regressive expendi- public resources based on loyalty rather than need, inef- ture patterns, political manipulation of antipoverty ficient program implementation, and arbitrary enforce- programs, and ongoing distributional conflicts (De ment of market rules. These, in turn, limit coverage and Ferranti 2004). While wholesale remedies were rare, responsiveness of service delivery, create inadequate and partial responses such as conditional cash transfers unpredictable resource flows, and exacerbate the risks showed promise (de Ia Briere and Rawlings 2006). and costs to economic agents of entry to markets. • In countries in East Asia and the Pacific, disenfran- To address governance failures, developing countries chisement of vulnerable groups and limited account- embarked on long-term institutional reengineering ability of government threatened the early gains 15 of public management reforms and anticorruption supreme audit institutions and legislation on the right to efforts. information), "space" for civic participation (for example, robust laws providing for nongovernmental organiza- • By the end of the first decade of post-socialist transi- tions and credible government-business forums), and tion, corruption was recognized as a central challenge provisions for citizen recourse and redress (for example, across Eastern Europe and Central Asia. Countries in ombudsman offices and other grievance mechanisms). Central Europe and the Baltics leveraged the Euro- Efforts to motivate demand-side responses have taken pean Union accession process to make progress, while various forms, including strengthening parliamentary others in the Commonwealth of Independent States committees involved in oversight functions, support for faced protracted political stalemates and the natural resource curse. independent media organs, and capacity building and financial support for civil society organizations (CSOs) • In South Asia, the poor bore the disproportionate and federated community organizations. burden of weak public administration, including po- liticization and limited accountability. Alternatives to Efforts to strengthen independent accountability improve service delivery included community-based systems have included direct support for specific and e-governance initiatives. institutions, as well as the use of country systems. More • In the Middle East and North Africa, pressures from knowledge on how to effectively sequence support for population growth and urbanization strained state domestic accountability systems is needed. For instance, institutions, heightened perceptions of corruption, as illustrated in hypotheses in Figure 3.1, when en- and in late 2010 led to unprecedented public protests. abling frameworks are weak but demand-side responses In addition to women, the youth and rural dwellers strong, development partners can proactively use CSOs lacked basic access to services, markets, and jobs. and other stakeholders in monitoring and oversight of Reforms targeting marginalized groups have met with their programs. If enabling frameworks are strong but limited success. demand-side response weak, donors can align their programs with country systems while supporting af- • In Sub-Saharan Africa, governance problems rooted firmative action efforts to build CSO capacity. A dilemma in colonial legacies, ethnic fragmentation , natural is how to strengthen domestic accountability in settings resource dependency, and low capacity were not with neither the space for social accountability nor the easily solved. Rather, populist efforts to indigenize forthcoming demand-side response. Efforts to strengthen inherited colonial states gave way to patronage and regulatory and legal frameworks may outpace the abil- unaffordable expansion of government (World Bank ity of social groups to engage constructively. Similarly, 2005c, p. 22). Structural adjustment and subsequent demand-side efforts may not be sustainable without capacity-building efforts also proved unsustainable. enabling frameworks. Yet, democratic transition in some countries still of- fers a chance to reshape state-society relations, even Governance and Sector Performance. Improved as conflict threatens further decline in others. performance in the social and infrastructure sec- Executive Restraint through Domestic Accountabil- tors typically involved strengthening both the sup- ity. The lack of restraint on executive discretion has ply side and demand side of governance. Supply-side undermined the legitimacy and responsiveness of the prerequisites of efficient and effective service provision state. The importance of accountability-the responsive- have been long established: dearly defined roles and ness of political executives to the needs and aspirations responsibilities, inclusive decision making, adequate and of citizens-gained prominence at the World Bank in predictable resources, and motivated staff. These require the 2000s (World Bank 2002; Mukhopadhyay and Meer cross-cutting public management systems (for example, 2004; Goetz and Hassim 2003; O'Neill, Foresti, and public financial management, pay and employment, and Hudson 2007; Schedler, Diamond, and Plattner 1999; intergovernmental relations). They also require sector- Jones and Stewart 2008). Well-functioning account- specific capabilities, which have evolved with advances ability systems require enabling frameworks that create in technology, awareness of vulnerabilities, and improve- space for citizen participation and demand-side responses ments in management practice (Campos and Pradhan from social actors. Enabling frameworks typically ensure 2008). For instance, performance management contracts government transparency and disclosure (for example, are commonly used for roads maintenance, as are build- 16 World Bank Country-Level Engagement on Governance and Anticorruption Strengthening Domestic Accountability Systems Demand-Side Responses Strong Weak Full Alignment with Inclusive Alignment with Country Systems Country Systems and Processes: with Affirmative Actions for CSOs: Budget support and Sectorwide .J Budget support and Sectorwide Approaches with CSOs in policy Approaches with range of dialogue, project design, project "' affirmat ive action for CSOs through implementation, and M&E a pooled funds ~~ lr:.. b c . . .. CSO Involvement through ... II'> Global knowledge sharing ... 0 u u Cll '0' ~ ..!!! Monitoring achievement of targets Cll VI ... a.. a: -~ .5 lii u < u < ... ..r:. 0 Operational Units ~ ~ Tailoring of GAC Pillars Application of Guidance Learning by Doing Client engagement Country knowledge Source: lEG. Relevance of 2007 GAC Strategy and Implementat ion Plan 23 4 Incentives and Institutional Arrangements This chapter revi ews the incentive and institutional arrangements designed to support GAC implementation, including funding, strategic staffing, accountability, and oversight arrangements. The chapter draws on a detailed review of financing provided under the Governance Partnership Facility. Financing GAC Implementation Incremental funding for GAC implementation rein- forced a historical pattern of increased Bank spending The 2007 strategy sought to make critical "change[s] on governance work-a trend that preceded the launch in the way the Bank did business"-an agenda that it of the 2007 strategy (Table 4.1). Bank budget (BB) suggested would entail significant costs. To meet these expenditures for governance work increased 21 percent, costs, the GAC IP made a.n explicit appeal for resourc- from $140 million in FY04 to $169 million in FY10, es. It argued that increased resources would provide an even as the institution continued to operate within a flat incentive for Bank teams to implement GAC activities. In real budget environment since FY06. In addition, exter- certain areas, additional financing would allow the Bank nal funding of governance work through Bank-executed to satisfy unmet demand (for example, in public finance trust funds (BETFs) increased by 168 percent-much management, fiduciary systems, governance diagnostics, more rapidly than Bank budget- over the same period. judicial reform, and social accountability). Beyond these By FY10, BETF financing equaled nearly 40 percent of immediate needs, the IP did not specify the future priori- BB spending on governance. ties or timeframe for resource allocation. However, it was careful to note that even more resources might be re- A total of$119 million was earmarked for GAC quired depending on "initial results, patterns of demand, implementation, in addition to governance work and lessons of initial experience" (World Bank 2007b ). already planned or ongoing. These funds included Sources Bank Budget (BB) 140 138 147 152 158 165 169 Reimbursables• 6 6 8 7 7 11 11 Sub-total 146 144 155 159 165 176 180 Bank-Executed Trust Funds• 25 34 35 42 43 52 67 Total Resources 171 178 190 201 208 228 247 Sources: Corporate Planning and Ana lysis Department. Notes: Governance work includes: (i) work in four sectors-Central Government Administration; Law and Justice; Sub-national Government Administration; and General Public Administration; and (ii) work not linked to specific operational products, expenditures for which are recorded in ' Internal Orders' in the Bank's budget management software.This definition of governance work is consistent with that used to arrive at the estimate of pre-FY07 BB spending on governance work as stated in the Implementation Plan. The Operations Policy and Country Services Vice Presidency (OPCS) provided a thematic aggregation of BB spending on governance work, which included; (i) work on five themes-Public Sector Governance; Rule of Law; Financial and Private Sector Development; Social Development, Gender and Inclusion; and Urban Development; and (ii) tasks recorded in Internal Orders. This data showed that BB spending increased from $147 million in FY07 to $156 million in FY10. a. Reimbursables comprise income from trust fund administration and trustee services, and income from operational services (for example, reim- bursable technical assistance and fee-based services). b. Includes GPF disbursements of $1.3 million for FY09 and $8.5 million for FY1 0. 25 $54 million in incremental BB allocated for FY08-ll Among Bank units, Regions were allocated 50 percent of and $65 million in GPF funds allocated for FY09-12. incremental BB for FY08-11. Among Regions, Sub-Sa- Although the incremental annual BB of $16 million was haran Africa, and Middle East and North Africa received mainstreamed from FY09 onwards, the Board papers on the lion's share of these BB resources. It is important to the FY09 and FYlO budgets stated five "key areas" that note, however, that BB increments were relatively small additional financing would support: GAC-in-countries, in relation to total BB spending on governance, which GAC-in-projects, GAC-in-sectors, governance diagnos- averaged $164 million per annum over the FY08- 10 tics, and governance indicators. The GPF funds remained period. Due to data limitations, the sectoral breakdown explicitly earmarked and were allocated in three rounds, of these expenditures was not available. the last in January 2010. The intended incentive effect of BB increments-to Deployment of Incremental Bank Budget. Incremental increase Regional spending on governance work-was BB expenditures were consistent with GAC IP priori- muted. The evaluation compared overall BB spending on ties. They also signaled strong preference for Regions governance work for FY08- 10 less the incremental BB re- and the newly restructured Integrity Vice Presidency ceived under GAC for those years.18 It found that Regions (Table 4.2 through Table 4.4) Y These GAC IP priorities received an estimated $20.6 million in incremental BB for included support for CGAC processes, recruitment of GAC implementation over FY08-1 0 but increased actual personnel with country and sector governance expertise, expenditures on governance by a lesser amount-an esti- and more active Bank participation in global initiatives mated $11 million-over the same period. Non-Regional such as StAR. In addition, nearly a third of incremen- VPUs received estimated incremental BB funding of tal BB in FY09 supported "other priorities;' such as $19.6 million over FY08-10 and increased their actual knowledge activities related to GAC-in-sectors, demand spending on governance by an estimated $25 million for good governance, and public sector management. over this period. Taken together, all Bank units received Incremental Bank Budget Funding of Operational Units by Purpose, FYOS-09 ($Million) Purpose of incremental funding FY08 FY09 CGACs 2.8 2.2 GAC in projects' ' 2.0 1.1 Staffing increments 3.8 StAR 1.6 1.5 Other initiativesb' 3.2 4.3 Total-Regions, Network Anchors and DEC/WBI 9.6 12.9 Sources: One-Year Progress Report and Corporate Planning and Analysis Department. a/Twenty-six countries were initially nominated by Regiona l vice presidencies; each country team was provided with incremental funds of $0. 1 mil - lion. In addition, the Poverty Reduction and Economic Managem ent Network (PREM) and OPCS received $0.1 million each. The Philippines was added later as a CGAC country. b/ The Year One Progress Report noted (see Appendix A): "The total of $3 million ea rmarked for GAC -in-projects between FY08 and FY09 was supple- mented by most Regional vice presidencies by core budget resources." However, details of these supplementary resources are not available. Incremental Bank Budget Funding for GAC-FYOS-11 ($Million) FYOS mid year FY09 FY10 FY11 Tot al Share of t otal Vice presidency actual actual planned planned FYOS-11 FYOS-11 Regions 6.0 7.8 6.8 6.2 26.8 SO% Network Anchors 3.1 3.8 3.0 3.0 12.9 24% DEC/WBI 0.5 1.3 1.3 1.3 4.4 8% INT 3.3 3.1 3.1 9.5 18% EXT 0.2 0.2 Total 9.8 16.2 14.2 13.6 53.8 100% Source: Corporate Planning and Analysis Department (CFRPA). EXT= External Affairs department, INT = Integrity Vice Presidency, DEC= Development Economics Department, WBI =World Bank Institute. 26 World Bank Country-Level Engagement on Governance and Anticorruption Incremental Bank Budget Funding by Region-FYOS-11 ($ Million) FYOS midyear FY09 FY10 FY11 Total Share of total Region actual actual planned plann ed FYOS-11 FYOS-11 i AFR 1.6 2.6 2.6 2.0 8.8 33% I I EAP 0.9 0.8 0.8 0.8 3.3 12% I ECA 0.9 0.9 0.9 0.9 3.6 13% I KR v 0.5 0.5 0.5 2.2 8% I MNA 0.9 2.0 2.0 2.0 6.9 26% I SAR 1.0 1.0 2.0 7% I Total 6.0 7.8 6.8 6.2 26.8 100% Source: Corporate Planning and Analysis Department. AFR =Sub-Saharan Africa, EAP = East Asia and the Pacific, ECA =Europe and Cen tral Asia, LCR = Latin America and the Caribbean, MNA = Middle East and North Africa, SAR = South Asia. an estimated $40 million in incremental BB funding for supported "(i) innovative, country level governance pro- GAC over FY08-10 but increased their total spending by grams; (ii) work on frontier areas of governance through an estimated $36 million. single or multi-country and global initiatives; and (iii) global GAC learning and knowledge platforms:'20 Among Given the fungibility of resources, Regions spent $9.6 major BETFs, the GPF was unique in two respects. First, million less on governance work than anticipated, tak- it was explicitly dedicated to the implementation of a ing into account the incremental BB funding over the Bank-wide strategy, more or less in line with the time- FYOS-10 period (Figure 4.1). Even after accounting for table for Phase I.2 1 Second, it involved donors directly a slow start in FY08, Regions fell short of the anticipated in the competitive selection of Bank-executed grants, FY08- 10 incremental GAC spending trajectory. With the through participation on a joint committee with Bank exception of South Asia, Regional spending on gover- staff. nance work in FY08-10 increased by less than the BB in- crements received.19 By implication, they spent more on The GPF channeled funds through multiple windows. other priorities. Since Regions accounted for 80 percent The complexity of design was a concern. Window One of BB spending on governance, this outcome appeared focused on the country level, Window Two on frontier inconsistent with the efforts of the GAC IP. Changes in Overall Regional Bank There are several plausible explanations for the deployment of part of the incremental Regional Budget Expenditures on Governance, budgets away from governance work. Several factors FY08-1 0 (US$ Million) could have caused the budget shifts, but these cannot be 25 determined on the basis of available information. First, in the eyes of operational units, governance work may 20 have already been adequately funded over the FY04- 07 15 period. Second, the zero real growth budget environment prevailing since FY06 created pressures in all Regions 10 to identify and act on redeployment opportunities. 5 Third, the availability of GPF funds in FY09 provided an .,. ~ alternative source for funding. The combination of these 0 factors likely induced Regions to move some incremental -5 resources out of governance work to other priorities. - 10 FY08 FY09 FY10 FY08-10 Use of Donor Funds-the Governance Partnership Fa- - 15 cility. In addition to incremental BB, donor funds were D Change in BB Spend D BB Incr. intended to jump-start changes in the way the Bank Net Change engaged GAC issues. The largely Bank-executed GPF Source: Corpora te Planning and Ana lysis Depart ment. Incentives and Institutiona l Arrangements 27 GAC areas at the country level, and Window Three percent). Regions as a whole received 76 percent of grant on global/regional knowledge and learning programs, funding-a much higher proportion than their 50 per- although eligible activities across windows were difficult cent share ofFY08-10 incremental BB.23 to distinguish from each other. GPF financed a portion of Bank staff and operating costs for implementation The bulk of GPF grants was managed by Bank units of CGAC plans in select countries, up to 100 percent of in the Poverty Reduction and Economic Management Bank staff and related costs of frontier work, and learn- Network and, in particular, public sector management ing and research costs that were shared with developing units (Figure 4.2). 24 PREM units in the Network Anchor countries and donor partners. 22 as well as in three Regional units (Sub-Saharan Africa, Europe and Central Asia, and South Asia) were the As of December 2010, the GPF had approved 94 largest beneficiaries of GPF grants both in volume and grants, totaling $65 million. The bulk of grant funding number of grants. Together, PREM received 47 out of94 was channeled through Window One and supported grants and $31 million out of the $65 million allocated. Sub-Saharan Africa. Forty-seven percent of the grant This finding explains concerns expressed by staff in other funding ($30.8 million) was channeled through Win- Networks (for instance, at the Bank's September 2010 dow One, 38 percent ($24.6 million) through Window Governance Partnership Facility Window One Workshop Two, and 15 percent ($9.9 million) through Window in Cape Town, South Africa) about the heavy PREM- Three. Sub-Saharan Africa received 33 percent of total orientation of GAC implementation. grant funding, followed by Europe and Central Asia (13 percent), East Asia and the Pacific (12 percent), Latin The distribution of GPF grants across countries did America and the Caribbean (7 percent), South Asia not follow any discernible pattern. Also, follow-on sup- (7 percent), and the Middle East and North Africa (4 port to CGAC countries was limited. For instance, GAC ~ Allocation of GPF Grants by World Bank Sector Units Public Sector $28.6 (N = 40) Country Services Panel $8.1 (N = 7) Social Development $6.9 (N = 11) Econ Policy & Public Sector $3.4 (N = 7) Social Protection $3.1 (N = 3) Others $2.4 (N =4) Operational Services $2.0 (N = 3) Water $1.8 (N = 3) Financial Management Private Sector Development $1.0 (N = 2) Economic Policy $0.9 (N = 3) Agriculture & Rural Dev. $0.9 (N =2) Procurement Poverty Reduction $0.6 (N = 2) Urban Development $0.5 (N = 1) Infrastructure $0.S (N = 1) Human Development $0.5 (N = 1) Education Communications Health, Nutrition & Population Econ Policy & Poverty Reduction $0.2 (N = 1) ,_-----,,------,------.------.------.------.------. 0 5 10 15 20 2S 30 35 [] Amount in $ million (no. of gra nts) Sources: Operations Portal; Governance Partnership Facility Secretariat, as of December 2010. 28 World Bank Country-Level Engagement on Governance and Anticorruption grants were not systematically awarded to countries with instance, through portfolio and transaction-level risk better or worse governance performance (as measured reviews or monitoring of actionable indicators), and even by the governance cluster of the CPIA). Nor did they fewer grants were focused on enhancing the Bank's capac- focus on scaling up initial funding provided for CGAC ity to exercise selectivity by identifying GAC entry points. processes in 27 countries. Only 10 of those countries received a grant from Window One, and eight of those The majority of grants were expected to achieve out- 10 also received a grant from Window Two. 25 Another comes relating to core public sector reform, and to a five CGAC countries received grants from Window Two. lesser extent, the demand side and accountability. Sev- Only 15 of the original CGAC countries received financ - enty percent of the grants in the sample identified state ing from one of the windows. capacity building at the federal, state, or local level as an expected outcome. 28 About a third identified at least one The majority of approved grants supported opera- outcome relating to the demand side of governance, such tional activities under the GAC-in-countries, and to a as support for CSOs. Fewer expected to achieve outcomes lesser extent, the GAC-in-sectors pillars (Figure 4.3).26 relating to formal accountability institutions, for instance, Sixty percent of grants reviewed explicitly linked their development of ombudsman offices and anticorruption development objectives to partner country strategies or commissions. It is also important to note that GPF grants country efforts, with less than a third linked specifically were far less focused on the investment climate than to the Bank's CAS. Approximately 40 percent of grants in were Bank operations. Investment climate issues were a the sample were aligned with the GAC-in-sector pillar. mainstay of Bank operational dialogue over the FY08-1 0 Only a few grants were linked to GAC-in-projects, even period, as they also were over the longer FY04-1 0 period. though some of these types of activities were supported However, these issues were absent from the objectives of through the country-focused grants mentioned above. GPF grants. Only one grant was explicitly linked to global initiatives. Financial management ofgrants was generally sus- Of the various elements of GAC responsiveness, GPF tainable. The Bank met its commitment to comple- grants were primarily focused on strengthening country ment grant-funded activities with its own budgetary institutions and supporting smarter project design. 27 resources. 29 Out of the 24 grants for which cost data Seventy percent of grants proposed outputs related to were available, half received BB funding lower than the country institutional strengthening, and about one-half grant amount, 13 percent received BB funding higher contained outputs relating to smarter project design. About than the grant amount, 26 percent received BB funding one third of the grants had at least one expected output equal to the grant amount, and a remaining 13 percent focused on improving the Bank's signaling of risks (for received no BB funding at all. While these contributions Number of GPF Grants by Expected Outcomes, Operational Activities, and GAC Elements Ill Smart Design 16 c cu Signaling 9 E cu Institutional Strengthening 23 w Selectivity 4 iii Ill GlobaiGAC c cu ·- ... 0 ·- ~ :~ GAC in Sectors and Themes 13 cu ... GAC in Projects 4 g< GAC in Countries 17 Internal Focus 6 "0 Ill cu cu Investment Climate ... E ... 0 cu ... Q..,. Accountability Institutions 6 >< :I wo Demand Side/Civil Society 11 Core Public Sector 21 0 5 10 15 20 25 Source: lEG desk review. Incentives and Institutiona l Arrangements 29 were varied-GPF grants covered multiple years and BB existing stock of the World Bank Group's GAC work nor is allocated on an annual basis-the level of complemen- all GAC-related innovations over the past year can be at- tary funding by the Bank was considerable. It was not tributed to implementation of the 2007 Strategy" (World clear that this level of BB funding would be maintained Bank 2009c ). Precisely for this reason, it was important through grant completion. for outputs financed by incremental resources to be tracked and reported. The progress reports laid out some The Bank used donor funds only sparingly to cover work program priorities for the second and third years personnel costs and so did not face immediate depen- of implementation. Yet, other than the 27 CGACs (see dency risks in this regard. Fixed- or staff-cost ratios on Table 4.2), the progress reports did not link the financing GPF grants varied widely by Region. By and large, these of GAC activities to any specific source, such as the pre- costs were below 20 percent of the grant amount, and in 2007 base spending on GAC work or incremental BB. only a few cases were grant funds used to hire new staff or a consultant. For 38 percent of the grants, fixed costs In addition, fragmented funding arrangements-one were 25 percent or below grant amounts, and in another for incremental BB allocated through the budget 38 percent there were no fixed costs at all. In other words, process, and another for GPF funds allocated through these costs were yet to be incurred or were all variable a separate competitive selection process-weakened (consultant fees, travel, and other non-personnel items) . overall incentives for prioritization. The evaluation found that Regions established different procedures The evaluation was unable to assess cost-effectiveness for preparation, review, and clearance procedures for of GPF grants. Unlike traditional BB-financed products, GPF applications. In some Regions, such as the Middle GPF outputs did not have historical unit cost data. Even East and North Africa and South Asia, staff expressed though Window One and some Window Two grants dissatisfaction with the lack of clarity on the rationale involved significant multiyear commitments, the evalua- underlying GPF selection decisions, particularly in the tion found no evidence that the Grant Funding Request first round; disruption caused by the three rounds of GPF (GFR) cost estimates had been adequately reviewed by allocations; and a perceived failure to consider regional resource management units in Regional units or cor- and country priorities. A Mid-Term Review attributed porate units. Requests for BB funding of comparable these concerns to a lack of convergence between GPF amounts (for example, $500,000 or more) were more design and distinctive regional governance strategies (for likely to have been vetted, before being considered by instance, in the Middle East and North Africa and Latin senior management (lEG 2011). America and the Caribbean) (GPF 2008b). These risks would have been mitigated through greater involvement Overall Approach to GAC Resourcing. Despite its pur- of Regional unit management in decision making on ported importance to the Bank's effectiveness, the GAC GPF grants, clearer linkages between planned activi- strategy was resourced at the margin, that is, through ties and total GAC funding (both GPF and incremental relatively small incremental BB funds. The Bank did BB), and more strategic and transparent prioritization of not review the degree to which its FY07 base BB spend- total funding through Bank budget and work planning ing on governance work was aligned with the priori- processes. 31 ties of the GAC strategy. As a result, there was no plan developed to redirect the base spending over the next few years, where needed. Instead, the increased reliance on Strategic Staffing donor funding posed dependency risks in implementing As part of the IP rollout, incremental BB helped the GAC strategy, especially in light of plans to maintain finance the recruitment of dedicated GAC staff. Based BB funding of governance work at current levels.30 While on an FY09 strategic staffing exercise and guidance pro- the Bank acknowledged these risks, it did not propose an vided by the GAC Council, Bank operational units pre- alternative. pared detailed strategic staffing plans involving of 64 new and redeployed positions. Actual recruitments across Even so, the Bank did not systematically link incre- four Regions (Sub-Saharan Africa, the Middle East and mental funding to incremental GAC activity (World North Africa, and South Asia) were aligned with planned Bank 2007b ). By the second year of implementation, levels, although the grade mix was top-heavy due to ad- the Bank itself had acknowledged that "neither the total ditional recruitments in the highest technical grades in 30 World Bank Cou ntry-Level Engagement on Governance and Anticorruption South Asia and Africa. Efforts served to offset reductions link sector interventions to broader public management in this high level complement within the Poverty Reduc- reforms. In still others, advisers conducted routine politi- tion and Economic Management Network public sector cal analyses to inform country strategy. family during the years preceding the launch of the GAC strategy. It is also important to note that Europe and Cen- Defining GAC competencies alone would not likely tral Asia adopted an alternative approach to mainstream- result in sustainable and cost-effective staffing; a ing GAC- one that emphasized training all staff rather realignment of some network and central units may than recruitment of dedicated GAC staff. have been required. Several networks have long been involved in supporting the Bank's governance work. In parallel, the Bank set out to develop competencies for For instance, the PREM Public Sector Management, distinct governance and public sector streams. However, OPCS Financial Management and Procurement, and these attempts proved challenging. Overlaps between the Social Development families have focused on strength- competencies were more pronounced than the differences. ening public sector capacity. The division of labor Areas of overlap included anticorruption, accountability, among these families has historically differed by region political economy, and related areas such as the justice sec- and country. Similarly, the newly established Integ- tor and decentralization. This exercise was complicated by rity Vice Presidency Preventive Services Unit, like the the diversity of job descriptions used for dedicated GAC Financial Management and Procurement families, has advisers. In some countries, Governance Advisers played sought to address fiduciary risks in projects. Frag- a role akin to portfolio managers and focused on fiduciary mentation and overlaps in unit mandates-and staff risk management issues. In other countries, they helped competencies-have been long recognized. Remedial Report of the 2005 World Bank Organizational Effectiveness Task Force: Diagnosis:"PREM (Public Sector Management) and OPCS (Procurement and Financial Management) both work on capacity building of country systems in financial management and procurement" and "Regarding local- level governance, decentralization, intergovernmental systems reform, and community-driven development have become high priorities across many countries. The Bank's ability to respond has been hampered by the fragmentation of our work ..." and, finally, "Given the strong (though fragmented) skill base already in place, within a fairly short period of time [a] new Network could become a preeminent global leader and innovator in addressing the challenge of how countries can improve governance, and how development partners can best provide support:' Recommendations: "The task force's judgment is that a Governance, Institutions, and Capacity Enhancement Network would sharpen focus and communicate globally the seriousness with which the Bank is engaging this frontier development issue:' And "the scope of work of this Network would need to include civil society institutions:' Report of the 2005 World Bank Task Force on Capacity Development in Africa: Diagnosis:"There is growing consensus within the Bank that its capacity development support remains fragmented and in some cases lacks coherence and client orientation. At the corporate level the operational and anchor units need to be better organized to ensure the coherence and effectiveness of the Bank's capacity development work. Bank management will need to exercise leadership, send consistent messages, and follow through ... especially on sensitive issues of governance, state effectiveness, and social inclusion ..." and "The inherently multisectoral nature of client capacity needs should always take precedence over the idiosyncrasies of organizational mandates for units in the Bank. It is imperative that Bank units coordinate their work more closely to meet partner needs:' Recommendations:"An lnstitutionwide focal point would spearhead analytical research, analytical and policy work to strengthen the Bank's knowledge base, and other knowledge creation and dissemination activities. The objective in the medium term should be to transform the Bank into a reputable center of excellence on the role of capacity, governance, and institutions in development:' Source: World Bank. Incentives and Institutional Arrangements 31 measures (for instance, by redesignating the Public of incremental BB and GPF grants (Sub-Saharan Africa, Sector family as the Public Sector Governance family) Europe and Central Asia, the Middle East and North have not satisfactorily addressed these issues. More Africa, and South Asia) put in place GAC steering groups fundamental reforms, such as those proposed in past to provide oversight. However, only Europe and Central Bank reviews, may be required (Box 4.1). Asia's steering group documented its work adequately. Outside of the Bank's management structure in Regions and networks, individual task team leaders ofGPF grants Coordination and Accountability were directly accountable to the GPF Steering Committee At the corporate level, the GAC Council served more through the GPF Secretariat. as an information-sharing forum than as a decision- making body with clear accountabilities. 32 Unlike Results Orientation. Reporting on GAC results suf- the Bank's Operations Committee, the Council did not fered due to lack of information on activities financed explicitly have responsibility for managing GAC-related through incremental BB. As noted above, three of the operational and corporate risks. Rotation of Council four Regions reviewed (Sub-Saharan Africa, the Middle chairs among Managing Directors every four months East and North Africa, and South Asia) lacked adequate further weakened accountability. The arrangement did documentation. Information was piecemeal and not not provide adequate time for any Managing Director sufficiently up-to-date to determine what was funded to steer the GAC agenda. Council meetings, which were and delivered through the use of the incremental BB regularly attended by a large number of nonmembers, resources. usually involved presentations by Bank units interested in showcasing their efforts. Council meetings would have Monitoring of GPF-financed activities was more benefited from critical review of what was working and systematic. The majority of grants reviewed used some what was not. combination of output and intermediate outcome indica- tors. A review of Grant Monitoring Reports from May Various units provided secretariat support to the 2009 through April2010 found that ratings for achieve- Council. Despite strong efforts by staff, these units ment of objectives were broadly evidenced-based. Most faced coordination challenges. A modestly staffed GAC grants were rated as at least moderately satisfactory in Secretariat in PREM supported agenda-setting for the the first year of implementation. Since most grants were Council, preparation of minutes and annual progress scheduled to close after FYll, these ratings would need reports, and outreach to internal and external stakehold- to be verified ex post. Also, nearly all monitoring reports ers. Front office staff of rotating Council chairs also sup- assessed risks and risk mitigation efforts. ported meetings and key communications. Separately, a GPF Secretariat within the PREM Public Sector Gover- Corporate reporting on GAC implementation focused nance Anchor facilitated grant submissions and selec- more on Bank inputs than on the quality of operations tions, grant monitoring, and overall reporting to the GPF and country governance performance. Year one and Steering Committee. It periodically provided reports to two progress reports to the Board primarily focused on the GAC Council. Other in-house secretariats (such as GAC inputs, such as analytical work, GAC-in-projects those for StAR and the Public Expenditure and Financial guidance, and CGACs. While they mentioned internal Accountability partnership) added complexity to GAC reviews of CGACs and GAC-in-projects, the progress reporting relationships. Lines of accountability upwards reports did not go far enough in presenting detailed find- through the PREM network to the Managing Directors ings or identifying lessons learned. Also, by their own ad- were fragmented and sometimes duplicative. mission, the reports were unable to establish the GAC IP activities added to the overall stock of Bank governance GAC oversight and accountability varied widely across work. In the final analysis, corporate progress reporting Regions. The four Regions that received the largest share did not adequately link resources to results. 32 World Bank Country-Level Engagement on Governance and Anti corrupti on 5 Implementation of the 2007 GAC Strategy This chapter reviews implementation of GAC IP activities or key operational"inputs"to the GAC results chain . These inputs included adherence to operational controls, provi- sion of guidance and tools, and delivery of support to Bank operational teams. They were intended to improve the GAC responsiveness of Bank operations-the focus of chapter 6. Adherence to Key Operational Controls Even so, the relationship between governance performance and IDA disbursements was affected- Selectivity ofAid Flows through the Bank. Over sometimes negatively-by the mix of financial instru- FYOS-10, the Bank continued to use governance ments in country portfolios, including the use of performance as a criterion for allocating conces- development policy loans (DPLs). Regression analysis sional resources across countries. The Performance undertaken as part of this evaluation found that use Based Allocation system gives considerable weight to offast-disbursing DPLs increased the likelihood that governance performance in IDA allocations and the countries-even when they had poorer governance- near doubling of available IDA resources since the would receive a flow of IDA funds (Appendix E, Tables Thirteenth Replenishment ensured that an increasing E.l-E.4). When only investment loans were used, share of commitments went to countries with higher however, countries required stronger institutional CPIA governance scores. Creditworthiness criteria frameworks to undertake transactions and draw down used to determine IBRD eligibility, by contrast, did not disbursements. At the same time, the use of DPLs was ensure that commitments to middle-income countries also associated with the achievement of various core were also governance-selective. It is important to note, public sector reform objectives (and more broadly, however, that the need to respond to the financial crisis economic reform objectives). These findings point to took precedence in determining lending volumes over the need for the Bank and its borrowers to make more the FYOS-10 period. informed choices about financial instruments based on their relative strengths and weaknesses vis-a-vis the Panel regression analysis confirmed that, before and GAC agenda. after the GAC strategy launch (and over the FY04-10 period), IDA flows were more selective than other Consistency of Risk Management in Lending Opera- types of aid flows. Improvements in CPIA governance tions. As noted in lEG's 2010 review of remedial scores were associated with increases in both IDA com- actions to strengthen IDA controls, GAC efforts have mitments and disbursements. Of the various compo- focused on measures to enhance the integrity of trans- nents of Bank flows, lending and trust fund commit- actions in Bank investment projects. Following the ments to IDA countries were more governance-selective Volcker Panel, the Integrity Vice Presidency scaled up its than lending or trust fund commitments to IBRD work on investigations, sanctions and debarments, im- countries. Also, among trust funds, recipient-executed plementation of the Voluntary Disclosure Program, and funds and financial intermediary funds were more prevention of fraud and corruption in Bank investment selective than Bank-executed funds (Appendix E, Tables operations. Furthermore, OPCS and other network units E.l-E.4). It should be noted, however, that individual developed GAC-in-projects tools that expanded the Bank-executed trust funds were allocated using criteria use of anticorruption measures in investment projects. that were sui generis and not necessarily linked to coun- Finally, the launch ofORAF in FYlO sought to standard- try governance performance. ize the Bank's approach to identifying and mitigating 33 risks on investment projects, even though it did not offer anticorruption plans and red flags) ; incorporating exter- guidance on the weighting of these risks. nal accountability (such as beneficiary participation and grievance mechanisms); and enhancing transparency and Early GAC efforts helped address material weaknesses disclosure. Supplemental supervision by the Bank was in operational controls on fraud and corruption in also indicated. Overall, the Bank's emphasis on internal Bank investment projects. But they focused primar- controls was balanced with some appreciation for the ily on the transaction level rather than on country benefits of transparency and independent oversight. systems, including those used in development policy lending. 33 In some instances, GAC-in-projects guid- Guidance was appropriate to managing fiduciary risks ance identified some entity-level controls (for example, on investment projects, but it did not make headway internal audit functions) that might affect transactions on the use of country systems (UCS). Hence, GAC in Bank investment projects. But, as of the end of FY11, Phase 1 guidance was focused narrowly on prevent- the Integrity Vice Presidency-a unit that was central to ing the misuse of the Bank's own funds rather than all the Bank's GAC efforts-had yet to develop its approach public funds in partner countries. It de facto maintained to ensuring the integrity of Bank resources channeled the restrictive "zero tolerance" stance on transactions in through either investment operations that supported investment projects rather than defining risk tolerances large scale incremental expenditures or DPLs. More for the UCS. The guidance also did not provide practical generally, GAC guidance was not adequately tailored to solutions for expanding UCS (except for suggestions on the specialized risks associated with DPLs, for instance, the use of supreme audit institutions). OPCS's efforts in those relating to budget support in settings characterized FY10 to reengage the UCS agenda came too late to affect by governance pathologies. GAC Phase 1 efforts. The layers and depth of risk review continued to Guidance on sectors and themes was primarily focused differ by the type of the lending instrument rather on applying generic project-level safeguards to specific than the risk profile of the operations in the post- sectors. The evaluation reviewed guidance materials in GAC period. The evaluation reviewed process maps for four areas: primary education, roads, accountability in- investment loans and DPLs prepared in the context of stitutions, and political economy analysis (Box 5.1). The the IDA controls review. By the appraisal stage, twice as findings are summarized below: many internal Bank units were involved in the review of investment loans compared to DPLs. Appraisals of GAC in Primary Education. Over GAC Phase 1, two investment projects took 1.6 times longer than those of background papers sought to broaden the Bank's focus DPLs. Anecdotal evidence from field visits indicated that from fiduciary risk management on education projects some GAC-in-projects efforts reinforced the relatively to incentives for sectoral performance. In line with heavier "footprint" of risk review units on the processing the current thinking on sector governance, the papers of investment loans compared to DPLs. Two-thirds of linked the supply side of public sector performance- operational staff, who responded to lEG's GAC survey, and to a lesser extent, the demand side of community reported that risk reviews in the post-GAC period were involvement-to education outcomes. Their treatment cumbersome. of key stakeholders and educational systems was consis- tent with the mainstream of thinking in the sector. Their discussion of demand-side factors included school-based Guidance and Tools management, community and parental involvement, and Cross-Cutting Issues. Cross-cutting guidance issued the role of public and private sector providers. over FYOS-10 concentrated on GAC-in-projects, and to a far lesser extent, on GAC-in-countries (Box 5.1). However, the papers were more concerned with im- Guidance relating to GAC-in-projects was primarily con- proving measurement and tracking of education sector cerned with mitigating the fiduciary risks to Bank opera- institutions than with providing operational guidance. tions. Recommendations included strengthening certain They emphasized precision in defining the elements of aspects of internal controls (for example, on financial sectoral governance. Complementary tools, available management and procurement); mainstreaming preven- on the GAC-in-human development Web site, offered tive measures against fraud and corruption (for example, suggestions on conducting Public Expenditure Tracking 34 World Bank Country-Level Engagement on Governance and Anticorruption Cross-Cutting GAC Guidance: Emerging Good Practices in GAC-in-Projects, 2009 (OPCS) Good Practice Note on GAC for Financial Management Specialists, 2009 (OPCS) Audit and Assurance Toolkit, 2008 (OPCS) Fraud and Corruption Awareness Handbook, 2009 (Integrity Vice Presidency) • Most Common "Red Flags" of Fraud and Corruption in Bank-Financed Projects, 2010 (Integrity Vice Presidency) Improving Development Outcomes, 2009 (OPCS) Sector-Specific and Thematic Guidance: GAC-in-Primary Education: Governance in Education: Raising Performance, 2009 (Human Development Network) Services Work: Indicators, Assessments, and Benchmarking of the Quality and Governance of Public Service Delivery in the Human Development Sectors, 2009 (Human Development Network) GAC-in-Roads: Deterring Corruption and Improving Governance in Road Construction and Maintenance, 2009 (Transport Network) Curbing Fraud, Corruption, and Collusion in the Roads Sector, 2011 (Integrity Vice Presidency) Accountability Institutions (in addition to cross-cutting guidance): • Guidance Note on Bank Multi stakeholder Engagement, 2009 (Legal department) Political Economy Analysis: • Political Economy of Policy Reform, 2008 (Social Development Network) • Problem-Driven Governance and Political Economy Analysis, 2009 (PREM) Source: World Bank documents. Surveys and other surveys in the education sector. The ing effective decision making, and improving relation- papers did not address relevant operational issues such as ships of "power and accountabilitY:' It also appropriately selectivity of entry points (for example, the supply-side sought to link Bank disbursements to sectoral gover- versus demand-side factors), the choice of instruments nance. To be effective, the final strategy would need to (for example, DPLs versus investment lending), and address long-standing issues such as the incentives for setting of risk tolerances (for example, how to respond education and public sector specialists to collaborate on to poor or deteriorating conditions of sectoral-level GAC-in-sector issues. governance). GAC in Roads. The Bank appropriately stressed pre- An early draft of the Bank's Education Sector Strategy ventive measures against fraud and corruption on cap- (Learning for All) included several measures relating ital-intensive roads projects. However, GAC guidance to GAC in education, and shows promise. The draft did not include an updated approach to UCS in the proposed to help strengthen the capacity of education roads sector. The guidance focused on reducing barriers systems to improve the "effectiveness of governance to market entry, fostering competition among eligible resources and aid financing" and to help "[reform] the bidders without government interference, and provid- governance, management, financing rules, and incen- ing a clear rationale for variation orders during imple- tive mechanisms in the system:' Equally important is the mentation. Preventive measures such as ring-fencing of draft's recognition of the role of information in promot- fiduciary controls-while effective in deterring fraud and Implementation ofthe 2007 GAC Strategy 35 corruption in specific Bank projects-often do not have supporting supreme audit institutions stressed the use of a wider impact on governance practices in the sector. partnerships. In countries where the Bank is not a major financier of roads sector expenditures, ring-fencing affects only a As part of the GAC strategy, some units also advocated nominal share of transactions both in number and in a more direct role for the Bank in channeling resources volume terms. Also, anecdotal evidence collected during directly to CSOs and other nonsovereign entities. Be- field visits suggested that the transaction costs and delays fore and after the launch of the GAC strategy, Bank-CSO in benefits resulting from ring-fenced procurement pro- relations largely consisted of facilitation of government- cesses can be significant. In this regard, recent Integrity CSO engagement within the context of Bank operations, Vice Presidency guidance that the Bank use more flexible bilateral Bank-CSO consultations "with the knowledge procurement processes was well-justified, but came too and support of member governments;' and other Bank late to affect Phase 1 GAC efforts. More generally, GAC partnerships with CSOs. Going beyond these efforts, guidance fell short of proposing a practical approach to direct Bank financial support to CSOs might attempt to using country systems, which would have helped prevent motivate social actors to apply demand-side pressures on corrupt or corruptible parties from simply shifting their executives. Advocates pointed to some early micro-level focus to non-Bank projects in the roads sector. examples that involved transfers through government to CSOs. While necessary, strengthening procurement on Bank roads projects is not sufficient to improve sector The direct financing proposal raised a number of governance. Guidance could have been better tailored operational issues, which were yet to be resolved. Bank to sector-specific issues such as pricing mechanisms proposals to scale up these micro-level efforts had yet to and land access benefits. In the roads sector, transfers clarify several concerns, some of which were shared by of substantial benefits and capture of windfall rents take GPF donors. Key operational issues included the Bank's place through mechanisms other than procurement. For comparative advantage relative to other agencies, impli- instance, the pricing mechanism can de facto transfer cations of direct support for the fiduciary risk profile of benefits from taxpayers to road users (in the case of Bank operations, the potential for capture by interested underpayment of roads costs by users), or roads user to parties, including party-affiliated CSOs, and potentially taxpayers (in the case of overpayment). Similarly, owners conflicting roles for the Bank (particularly in polarized of land adjacent to new roads may benefit from windfalls environments) as the financier of both sovereigns and unless access benefits are recovered through taxes or nonsovereigns. Moreover, the Bank's proposals should impact fees. The choice of projects that aims at equitable have referred to lessons learned from similar efforts over provision of access and mobility-and fair distribution of the 2000s to try to motivate private sector firms through benefits-through public participation and open infor- matching grants. mation are key GAC-in-roads issues, which were ignored by the guidance. Political Economy Analysis. The 2007 GAC strategy acknowledged links between governance, politics, and Accountability and the Demand Side. As a result of the poverty reduction prospects. It argued that governance GAC strategy's emphasis on accountability, the Bank was the result of the behaviors of different branches and provided timely guidance on how to engage non-state tiers of government, as well as the private sector, each of actors. A Note on Bank Multistakeholder Engagement which could function as its own "interest group:' Political- stressed the need to engage non-state actors and non- ly salient actors or elites shaped the incentives of individu- executive institutions in a manner consistent with the als within these institutions, as well as their willingness Bank's Articles.34 Accordingly, Bank teams were asked to support or obstruct poverty-reducing projects and to operate with experienced staff and sound knowledge programs. The quality of governance affected the condi- of political realities. Direct support to parliamentary tions under which elites benefitted from corruption, weak institutions was to be provided with the involvement enforcement oflaws, distortionary policies, and private of the executive. As such, the Bank's approach de facto diversion of public assets. Not surprisingly, efforts by the discouraged use of the kinds of standalone operations Bank to address poor quality governance including "cor- supported by other multilateral agencies, such as the ruption with deep political roots-such as state capture Inter-American Development Bank. Earlier guidance on and prqcurement corruption" had proven difficult. 35 36 World Bank Country-Level Engagement on Governance and Anticorruption The Bank argued that improved knowledge of political- rules. Second, the Bank should seek to assess rent-seeking economic interactions would enable it to support dynamics, as well as the likely impact of its operations on operations that were "better fitted" to country realities. the flow of rents and the stability of political compacts. It would be less likely to import or impose Western mod- Third, the Bank should be aware of the risks that technical- els in its operations. More realistic Bank support would ly-sound projects and programs might worsen the status enhance the likelihood that institutional reforms would quo. If necessary, the Bank should find "good enough succeed and help reduce poverty. By the same token, governance" alternatives that can produce measurable (if greater awareness of state fragility would not increase the limited) change. Finally, the conduct of PEA work should Bank's risk tolerance of various forms of corruption such involve more vigorous efforts to disseminate analyses in as patronage. Therefore, the Bank sought to strengthen its order to enrich public debate. own capabilities to support PEA in the context of country dialogue and project design. To promote the use of this type of PEA, the Bank mar- shaled considerable GAC resources. Earmarked financ- As part of GAC implementation, guidance on PEA was ing under the GPF was made available for carrying out developed in two primary reports: Political Economy PEA in the context of traditional Bank analytic and advi- of Policy Reform published by the Social Development sory activities, lending operations, and country strategy Department in 2008 and Problem-Driven Governance and development. The financial support was intended inter Political Economy Analysis published by PREM in 2009. alia to help diagnose the underlying political constraints While the former focused on the operational implications that client countries face and to push the "new frontiers" of political-economic analysis for sectoral reforms (with of governance work. Furthermore, a Political Economy an emphasis on agriculture and water), the latter iden- Community of Practice (PECoP) was established as tifies principles for incorporating "Governance and Politi- a focal point for supplying applied PEA in support of cal Economy" analyses in Bank programmatic activities. operations. PECoP aimed to lead Bank-applied PEA, to disseminate this knowledge, and to stimulate demand Guidance provided by these reports stressed four com- for PEA product lines, including "process support:' 36 The mon prescriptions for PEA. First, PEA should go beyond PECoP consisted of 25-30 core staff, 250 other interested technical assessments of formal institutions and should staff, and a roster of consultants. cover de facto and de jure mappings of institutional ar- rangements, structured feedback from local stakeholder The evaluation found that the Bank's goal of updat- representatives, analyses of distributional struggles, and as- ing its approach to PEA was appropriate. Since the late sessments of the divergence between formal and informal 1990s, lEG evaluations had consistently recommended USA/D's Democracy and Governance Assessment (2000): One of the earliest donor-based political assessments, the framework aims to identify comparable elements of liberal democratic government such as uliberty, open competition, the rule of law, and respect for pluralism and minority rights;' and entry points for promoting democracy. OF/D'S Drivers of Change (2004): This methodology assesses linkages between a country's political framework and its programs that have an impact on poverty reduction. Specifically, it focuses on structures, individual agents, and mediating institutions that affect political will. SIDA's Power Analysis (2005): In this approach, power asymmetries, access to resources, and influence over politics must be addressed if poverty is to be reduced. The analysis seeks to map the informal political landscape (including rules and structures) and to understand how development cooperation and donor activities are influenced by this landscape. The German Agency for Technical Cooperation's (GTZ) Governance Questionnaire (2004): This tool assumes that political reforms are strongly influenced by informal values, norms, customs and processes, rather than formal rules. It seeks to stimulate debate on how to support specific reforms. Sources: DFID (2004), Chakrabarti (2005), Sida (2005), Faust and Gutierrez (2004). Implementation of the 2007 GAC Strategy 37 that the Bank develop a stronger appreciation for po- some cases, country partners were not even aware litical economy realities in its operational work. Early that PEA work was being undertaken. Bank teams attempts to support political and institutional analysis in urgently required guidance and advice on how to in- the context of PRSs, however, lacked coherence and lost volve country partners in the initiation, preparation, momentum. Since then, other donors had advanced their and dissemination of PEA. For instance, a PECoP efforts to develop operationally-oriented PEA methods proposal to attribute Bank-financed PEA reports to (Box 5.2). GAC-era guidance provided a timely summa- individual authors (without the institution's "stamp tion of earlier Bank frameworks and perspectives. of approval") did not go far enough in considering the potential reputational risks to the institution of However, guidance and support for PEA could have gone disclosing highly sensitive analyses in an operational farther in addressing a number of analytical and opera- environment. tional issues: • The proposed development of free-standing PEA products could build on lessons learned from • A single definition of what constituted PEA was similar efforts in the past. The PECoP could usefully lacking. Despite efforts by the PECoP to integrate draw on lessons learned from the Bank's experience multiple perspectives and methodologies, the various with Institutional Governance Reviews, particularly toolkits, trainings, and outputs lacked consistency in responding to the needs of operational staff and and coherence. This problem was evident even in country partners. After GAC start-up financing the divergent views of PECoP members themselves. dries up, the current PEA effort may face challenges To some, PEA involved an analysis of stakeholders, in stimulating demand. Most respondents to lEG's while others understood it to mean an assessment of staff survey did not believe that PEA guidance and formal and informal institutions. Still others thought support had been packaged in a user-friendly way. In in terms of analyzing politics writ large, while others their written comments, survey respondents also did focused on formally simulating the incentives of bu- not express confidence in the ability of most political reaucrats and politicians involved in Bank operations. economy experts to advise on operational solutions. • Ex ante assessment of "good fit" was a major ra- Alternative ways of mainstreaming political economy tionale for undertaking PEA. While attractive in approaches could include more systematic peer theory, assessing "fit" was difficult in practice. It review of standard economic and sector work and development of project-level decision analysis tools. required analysts to estimate the degree to which in- stitutional design or "form" would deliver its desired "function" in a particular setting. What appeared to Delivery of Support be good fit ex ante might not be considered to be ex The Bank used various media to disseminate GAC post if it failed to produce the desired result. Similarly, guidance and promote the use of GAC tools. Network reforms that appeared to be "imported" (for instance, and regional units-to varying degrees-used training public financial management reforms) might, to the events, workshops, seminars, and print and Web-based surprise of analysts, become institutionalized and publications. In addition, internal communications from produce favorable results. senior management, the GAC Council, and GAC focal • Guidance could have more directly addressed the points were extensive, particularly in the first year. Also, the GAC Portal Web site served as a clearinghouse for political economy of aid including the Bank's own analytical and advocacy materials. role in partner countries. Unlike other donors (such as SIDA and GTZ) , the Bank did not explicitly ad- The emphasis on internal communications paid off. dress its own role in shaping incentives in sectors and Sixty-three percent of respondents to lEG's staff survey countries. Such an approach would have helped the were familiar with the 2007 GAC strategy and IP with Bank better assess reputational risks, particularly in no significant variation by regional unit or location (in aid-dependent countries. Country Office or Washington). Respondents with more • The internal marketing of PEA work within the than 10 years of Bank experience were significantly more Bank differed markedly from the largely muted familiar than those with less than 10 years of Bank expe- approach taken with country stakeholders. In rience (Appendix F). 38 World Bank Country-Level Engagement on Governance and Anticorruption Some media were more effective than others in dissem- still significantly higher than those of Washington-based inating information on GAC. A plurality (42 percent) staff. found training workshops a useful source of information. Fewer found other media-reading available Bank docu- Learning activities focused on GAC-in-projects and ments, attending events with senior managers, visiting country accountability institutions. GAC-in-sectors the GAC Portal, applying for GPF grants, or participat- was not given adequate attention. For instance, a total ing in the PECoP-as effective. Responses from Country of 56 learning events were supported by central units Office staff were significantly more favorable than those through September 2008, comprising seminars, train- from Washington-based staff. The degree of decentraliza- ing workshops, and partner events. Of these, 38 percent tion may account for more favorable responses from staff related to GAC-in-projects and 30 percent were on in certain Regions (for example, EAP). domestic accountability systems. Only 16 percent focused on GAC-in-sectors issues and even fewer on the demand Beyond access to information, relatively few staff report- side. During the first year ofGAC, 15 CGAC clinics and ed receiving tangible support to implement the GAC related country-oriented seminars provided a vehicle for strategy. To the extent they did, support focused on fraud more focused training. and corruption issues in projects (according to 45 percent of respondents). Far fewer reported receiving additional The GAC IP provided an opportunity to prioritize and resources, training and skilled personnel, or guidance on better coordinate learning activities, which could have institution building, or analytical support for PEA. been more fully exploited. Even before the GAC strategy, core courses were offered in public financial manage- The generally low staff ratings on the relevance of ment, civil service reform, justice sector reform, social specific guidance materials and tools translated in into accountability, and anticorruption, as well as project low utilization rates. Responses reflected the priority financial management and procurement. This menu was given to GAC-in-projects issues (Figure 5.1). Forty-five expanded to include courses around GAC pillars and, percent of respondents responded favorably to GAC-in- more recently, on the demand for good governance. Even projects guidance, in particular the OPCS Emerging Good though the PREM Anchor attempted to better organize Practice Notes and the Integrity Vice Presidency Fraud learning around program themes, the larger GAC training and Corruption Awareness Handbook. A third or fewer felt agenda-including its use of Web-related assets-lacked the same about materials relating to the demand for good adequate coordination. Given the feedback of Country governance and PEA. It is important to note that respons- Office staff, more could have been done to prioritize and es of Country Office staff on utilization-while low-were ensure the quality of decentralized offerings. F.IGURE 5.1 Staff Perceptions of Relevance and Use of Guidance Materials and Tools "to a Great Extent" Relevance of guida nce material and tools Utilization of gui dance mate rial and tools GAC in countries 43 GAC in countries 31 29 21 GAC in sectors 43 GAC in sectors 32 31 24 GAC -in-projects 52 GAC -in-projects 40 38 30 Demand for good Demand for good 31 37 governance 30 governance 25 31 25 Political Economy 27 Political Economy 21 0 10 20 30 40 so 60 0 20 40 60 Percent Percent D CO Staff D HQ Staff Source: 2010 lEG GAC Staff Survey. Implementation of the 2007 GAC Strategy 39 6 Effectiveness: Is the Bank More Responsive to GAC Issues? This chapter provides a review of the GAC responsiveness of Bank country programs and projects before and after the launch of the 2007 strategy. GAC responsiveness was defined in terms of the selectivity of appropriate entry points and instruments, country institu- tional strengthening, signaling and mitigation of risks, and smarter design of projects. The results were based on a desk review of 50 country programs and 200 projects over FY04- 07 (referred to as the "pre-GAC" period) and FYOS-1 0 (referred to as the "post-GAC" period). They also drew on six country case studies undertaken for this evaluation. Where relevant, econometric analysis was used to assess GAC elements in operations. To assess more dearly the changes that occurred in goal. While there have been signs of progress in certain the post-GAC period, this chapter presents data for areas, important opportunities have yet to be seized operations that were responsive "to a great extent?' As {Figure 6.1). noted in chapter 3, most Bank programs and projects were already at least somewhat GAC-responsive in the In country programs, the following trends were ob- pre-GAC period. There was little variation in somewhat- served: responsive programs and projects across time periods or other demographic parameters (for instance, region or • The Bank has continued to support good gover- levels of income or governance performance). Appen- nance objectives in virtually every country where dixes C and D present both sets of scores for all elements it has operations. Pre- and post-GAC CASs were of GAC responsiveness reviewed in this evaluation. similarly selective in identifying entry points for a GAC dialogue. In many countries, the Bank has sustained a medium-term dialogue on GAC issues Achievement of Implementation Plan and provided a program of support in areas such as Objectives: A Summary public financial management, sector service delivery, and the investment climate. Sustained engagement Over the FYOS-10 period, the Bank's response to GAC on these issues, even in challenging settings, remains issues in its country programs and projects demon- one of the Bank's strengths. strated continuity without systematic improvement as yet. Building on two decades of engagement on gover- • The Bank continued to be as committed to sup- nance issues, the 2007 GAC strategy acknowledged that porting country-led GAC efforts as it was before it "implied a change in the way the Bank does business:' the launch of the strategy. Virtually all CASs Strategic communications and engagement by Bank contained GAC-related pillars in the FY04-07 and senior management signaled this goal. The implemen- FYOS-1 0 periods with no variation across regions, tation plan and annual progress reports characterized levels of income, and governance performance. success as making systematic improvements in the GAC Pre- and post-GAC CASs were similarly selective in responsiveness of country operations, although they did · identifying entry points and choosing appropriate not set targets or a timeframe for the achievement of this instruments. 41 Country Programs and Projects Addressing GAC Issues "to a Great Extent;' Pre- and Post-GAC Periods Percent 0 10 20 30 40 so Portfoliowide fiduciary risk mitigation "' Gl ... ·;: c Selectivity :I 0 v c Signaling ~ < 1!1 Country institutional strengthening D PreGAC 0 10 20 30 40 so 0 Post GAC Quality of governance and PEA ~ ~ Quality of enhanced fiduciary aspects v ·a . Gl Q. Demand -side of governance .!: Use of country systems u < ~ 1!1 Quality of institutional strengthening Results orientation Source: lEG desk review. Note: Arrows indicate statistically sign ificant cha nges. • There was a statistically significant increase in the great extent" increased from 15 percent pre-GAC to number of countries where the Bank planned to 25 percent post-GAC. This type of analysis-and in support country institutional strengthening. The particular the assessment of informal institutions- share of CASs with plans to directly support country was associated with improved project "fit" to gover- institutions to a great extent nearly tripled, although nance realities. these were not explained by IP measures such as • The use of at least some country systems in projects CGACs and GPF window support (Box 6.1). increased significantly in countries with weaker • A minority of pre- and post-GAC country pro- institutional capacities and in Africa. In countries grams included explicit plans to signal risks. Even with CPIA governance scores lower than 3.5, 27 fewer sought to mitigate fiduciary risks. Less than a percent of pre-GAC projects used at least some coun- third of CASs in both periods aimed to use standard try systems-that is, public financial management, portfolio processes to identify, signal, and mitigate procurement, or personnel systems-compared to 41 GAC risks. Only 15 percent of pre-GAC CASs and percent of post-GAC projects. Also, in Africa, proj - 22 percent post-GAC explicitly focused on mitigating ects using at least some country systems increased fiduciary risks to a great extent. Also, the Bank did not from 11 percent pre-GAC to 40 percent post-GAC. In adopt a consistent risk-based approach to responding principle, the Bank's increased use of these systems in to governance crises or deteriorations. weaker settings allowed countries to more efficiently • Only a few CASs made commitments to adopt deploy limited capacities in meeting pressing domes- smarter project design approaches across the lend- tic priorities rather than sui generis donor project ing portfolio. The review did find that actual use of management and reporting requirements. smarter design in projects was more prevalent than • The share of projects rated highly for quality of these CAS commitments indicated. enhanced fiduciary aspects, GAC results measure- At the project level, the following trends were observed: ment, and for project-level institutional strengthen- ing components remained low and did not change • The number of projects that employed upstream significantly in the post-GAC period. For instance, governance and political economy analysis "to a about a quarter of pre-GAC projects rated highly in 42 World Bank Country-Level Engagement on Governance and Anti corrupti on these three areas; improvements in projects approved The GAC IP's goal of improving the Bank's reputa- post-GAC were only marginal with no variation tion on GAC issues was achieved partially. Donors, across regions or levels of income and institutional civil society organizations, and borrowing governments endowment. consulted during the evaluation were highly appreciative of the Bank's capacity to analyze and advise on gover- • Bank projects did not significantly expand their nance issues, and of the Bank's commitment to support overall use of demand-side measures in the post- longer-term institutional development. The GPF donors, GAC period. While the involvement of beneficiaries in particular, felt that the Bank's analytical strengths justi- in Bank projects was prevalent even before the launch fied a more proactive role in assessing political economy of the strategy, Jess than a third of pre- and post-GAC constraints and measuring governance performance. projects were found to be responsive to the demand During field visits, country stakeholders acknowledged side overall. Importantly, more post-GAC projects in the Bank's contributions in leading policy dialogue in post-conflict countries used these measures. areas such as public financial management, the invest- ment climate, and sector reform. Donors and some The quality and coverage of political economy issues government officials also noted that the Bank had higher in the Bank's economic sector work did not show the fiduciary standards than other partners. systematic improvements evident in projects. The operational benefits of free-standing PEA reports were Stakeholders inside and outside the Bank observed a often limited by an overly academic orientation, uneven potential conflict between its lending goals and its pur- methodological rigor, and a lack of consistency between suit of GAC objectives, particularly in poorly governed recommended actions and prevailing interpretations of settings. Within the first year of GAC implementation, the Bank's Articles of Agreement. the 2008 Gallup Global Survey revealed that opinion Who received CGAC and GPF window support? The Bank was already supporting "GAC as a pillar" in CASs in most countries that received CGAC and window support. For instance, 97 percent of pre-GAC and 84 percent of post-GAC CASs in CGAC countries were supporting GAC as a pillar, and 95 percent of pre-GAC and 88 percent of post-GAC CASs in window countries. Even though the Bank was not more likely to address GAC issues in CGAC and window countries relative to others, the evaluation found some patterns. For instance, countries with DPLs in their lending portfolios and lower CPIA scores were more likely to receive CGACs. Those in Africa were more likely to receive window support. Were CASs and projects in countries that received support more GAC-responsive? CASs and projects in CGAC and window countries were not systematically more GAC-responsive than CASsin other countries over the GAC implementation period. Country programs receiving CGAC and window support were not significantly different in their GAC- responsiveness than those that did not receive this type of support with two exceptions. Those that received support were 16 percent less likely to achieve domestic accountability objectives (relative to non-CGAC and window countries) in the post-GAC period. In addition, they were more likely to focus more on institutional strengt hening relative to non-CGAC and Window One countries in the post-GAC period (47 percent in all countries, compared to 55 percent in CGAC countries). Projects in CGAC and window countries were: • not significantly different over the pre- and post-GAC periods in terms of overall smart design. • thirty-two percent less likely to support rules-based decision-making and accountability"to a great extent" in the post-GAC period. • continued to have more risk management measures relative to projects in other countries in the post-GAC period. Source: lEG desk review. Effectiveness: Is the Ba nk More Responsive to GAC Issues? 43 leaders in both industrialized and developed countries ment climate. Among public sector reform entry points, were increasingly concerned about the Bank's approach the most prominent was public financial management, to lending in poorly governed countries. For instance, a and to a lesser extent, revenue management, decentraliza- third of respondents in lower-income countries felt that tion, and civil service reform. Private sector development the Bank should not lend to countries unless they took (including privatization and restructuring of state-owned serious actions to fight corruption. CSOs consulted in enterprises) and regulatory reform (including licens- the course of this evaluation agreed and recommended ing, permits, and customs) remained the most common that, in such settings, the Bank reduce lending or impose investment climate entry points before and after 2007. A stricter GAC-related conditions on loan disbursements. sizeable majority also identified domestic accountability Both Global Survey respondents and CSOs suggested institutions and the demand side in both the pre- and that the Bank should channel funds outside of govern- post-GAC periods. Domestic accountability institutions ment (for example, to civil society).37 Nearly half of Bank most frequently included judiciaries and supreme audit operational staff surveyed also believed that "the Bank's institutions, and to a lesser extent, legislative oversight, lending imperative conflicts with its ability to implement media, and procurement appeals bodies. Those that iden- the GAC strategY:' tified demand-side measures virtually all identified the need to strengthen CSOs, but only rarely identified the organized private sector and consultative mechanisms. GAC Elements in Country Strategies and Programs The number of country programs that justified their The evaluation assessed specific GAC elements of choice of instruments and identified GAC results country strategies and programs over the pre- and measures increased. Twenty-nine percent of pre-GAC post-GAC periods. These elements-selectivity, sig- CASs and 38 percent of post-GAC CASs were explicit naling, institutional strengthening, the use of country about their choice of instruments. Clearer rationale for systems, harmonization, and results frameworks-are instrument choice was particularly pronounced in CGAC discussed below. countries. Similarly, modest progress was made in the number of country programs that identified appropriate Selectivity of GAC Entry Points. As noted in chapter GAC results measures, even though results measurement 1, the strategy placed a premium on making "right" remained a weak area, particularly in Latin America choices about GAC objectives and entry points based and the Caribbean and Sub-Saharan Africa. Country on systematic diagnostic work. In both the pre- and programs that received financing from one of the GPF post-GAC periods, a majority of country programs re- windows identified GAC-related results measures more ceived high marks for incorporating explicit assessments frequently. of governance and political economy constraints to development objectives. To undertake these assessments, Yet, GAC selectivity was also associated with country the Bank most commonly proposed and used techni- programming decisions, in particular choice oflend- cal assistance (TA) activities as well as standard ESW ing instruments. Regression analysis found that the products such as Poverty and Social Impact Assessments GAC selectivity ratings for instrument choice and results (PSIA), Public Expenditure Reviews (PERs), and to a decreased significantly when a DPL was included in the lesser extent, Country Economic Memoranda (CEMs). country portfolio (Appendix E, Table E.S). Further, the Country Financial Accountability Assessments (CFAAs) probability that a CAS was GAC selective-that is, it and Country Procurement Assessment Reports (CPARs) incorporated governance assessments, justified the choice tended to be merged with PERs as integrated assessments of instruments, and identified GAC results-decreased in the post-GAC period, while new products, such as by 99 percent when a DPL was included in the portfo- freestanding PEA reports, became more prominent in lio. CASs were also 34 percent less likely to support the part due to availability of GAC funding. demand side if DPLs were used in the overall lending program (Appendix E, Table E.6) (lEG 2010}.38 These More than half of country programs drew on upstream findings suggest that the decision to use DPLs may be diagnostic work to justify their selection of GAC objec- associated with factors other than those related to GAC tives and entry points. Nearly all of them included sup- selectivity (for instance, economic stabilization, eco- port for core public sector institutions and the invest- nomic recovery, and short-term crisis response). The 44 World Bank Country-Level Engagement on Governance and Anticorruption use of ESW and TA appeared to work in the opposite the search for viable and durable reforms. Initiatives that direction. CASs with plans to undertake GAC analytics gained initial traction-for instance, public financial were 31 percent more likely to support non-executive management in Cambodia and GAC-in-roads in Moldo- accountability institutions. The finding points to the need va- became more established in subsequent CAS cycles. for CASs to be more explicit about trade-offs associated Nevertheless, the "venture capitalist" approach also weak- with using various instruments to pursue GAC and other ened incentives to make hard choices, for example, when development objectives.39 Bank support did not generate tangible, time-bound results (for example, public sector reforms in Liberia). Importantly, there were limits to how selective Bank CASs can be in pursuing GAC entry points. The Bank Strengthening Country Institutions. As noted earlier, typically has been opportunistic in responding to direct support for strengthening institutions in the changing country conditions. Only rarely was the Bank core public sector and institutions underpinning the compelled to organize its country programs around an investment climate was a mainstay of Bank assistance overarching national GAC priority (for instance, im- over the FY04-10 period. Sixty-five percent of pre- and proved management of petroleum revenues in Chad). post-GAC CASs aimed to support core public sector Furthermore, the evaluation also found no discernible reforms, with no statistically significant change. Another approach to strategic sequencing of different GAC re- prominent area of Bank support was the investment forms (for instance, public financial management before climate, with about a third ofCASs pre- and post-GAC decentralization, or regulatory reform before privatiza- supporting the development of market institutions. These tion) . Rather, Bank CASs typically adopted a "venture patterns were broadly consistent with trends in global aid capital" approach that supported multiple initiatives in commitments for capacity building (Box 6.2). Over the 2000s, the Bank, along with other bilateral and multilateral donors, increased support for institutional capacity building relative to the 1990s. Support for capacity building, driven by bilateral donors, reached its peak of a third of total official development assistance commitments in 2004, and this level of support was largely maintained over the Phase 1 GAC period. Consistently over the decade, Bank support to capacity building as a share of total Bank flows remained far higher (approximately 40 percent) than that of other donors. In countries with higher CPIA scores, the Bank and other donors focused on building technical and cross-cutting capacity, as well as private sector and civil society capacities. In countries with lower CPIA scores, the Bank and other donors focused on core public sector and domestic accountability systems. Share of Official Development Assistance Commitments to Capacity Building, 2000-09 70 60 50 20 10 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 D Total • • • • Bilateral - IBRD ••••••• IDA - Multilateral Sources: OECD Creditor Reporting System and World Bank. Effectiveness: Is the Bank More Responsive to GAC Issues? 45 The Bank's work on institutional strengthening in- lio reviews in countries in Europe and Central Asia, Latin creased in the service delivery sectors and domestic America and the Caribbean, and the Middle East and accountability institutions. The number of countries in North Africa was of concern. which the Bank supported sectoral institutions to a great extent increased from 35 percent pre-GAC to 43 percent An important test of the Bank's risk management post-GAC. The increase in support for sectoral institutions capabilities was how it responded to unanticipated in fragile states was significant. Similarly, the share of Bank governance events. The evaluation found the Bank's country programs that sought to strengthen accountability responses were highly varied, regardless of how much institutions increased from 23 percent to 35 percent in the it invested in ESW and TA. Over the FY04-l 0 period, post-GAC period. This expansion was explained by sig- more than half of unanticipated events were related to nificant increases in countries in Africa, in fragile states, GAC issues. These were categorized as stark downturns and countries that received GPF window support. (due to civil conflict, electoral and political instability, and breach of international agreements), deterioration (in- Addressing Country-Level Risks. In a majority of CASs, creased incidence of corruption, poor interministerial co- the Bank consistently identified GAC-related risks ordination, or legislative gridlock), and turnarounds (for and proposed mitigation measures. Political instability instance, peace agreements and the ascension to power of and weak state capacity were most frequently identified reformers) . Bank responses to these events varied from as risks in the pre-GAC and post-GAC periods. Fraud doing nothing to enhancing monitoring, conducting joint and corruption, as well as security were also identified. donor reviews, modulating lending levels, and disengage- In most cases, CASs that identified risks also proposed ment. The evaluation did not find a systematic relation- at least one mitigation measure. The most commonly ship between ESW ITA and Bank responses. In some proposed measures were lending scenarios, joint donor countries, however, ESW helped inform crisis response reviews, and anticorruption action plans. Of these, the (for example, in Poland and Bangladesh), but in others, most commonly proposed lending scenario was also the it did not (for instance, in Moldova). There was also no least likely used, in part because IDA discontinued their evidence that Bank responses were graduated. use. The most likely used were joint donor assessments. Using Country Systems as a Strategic Priority. Over the The Bank also used country portfolio review processes FY04-10 period, a third or fewer country programs to track governance issues in sectors and projects, al- identified the use of country financial management though it could have been more systematic about doing systems as a portfoliowide, strategic priority. The evalu- so. Nearly half of CASsin the pre- and post-GAC periods ation found that, in the pre-GAC period, the actual use of provided evidence that portfolio risks were routinely some country systems portfoliowide was more frequent monitored. These processes served to track and signal than statem ents in CASs would suggest. For instance, in GAC-related risks at the sector and project levels in 32 the pre-GAC period, the Bank used financial management percent of pre-GAC and 38 percent of post-GAC CASs. systems in 75 percent of countries, and intergovernmental, While modest, progress was evident in IDA countries audit, civil service and personnel systems, and national and in countries with lower CPIA scores. procurement systems in about half of sample countries. In the post-GAC period, the use of some country systems as Less than a quarter of Bank country programs regu- a portfoliowide measure declined in countries with lower larly disclosed or disseminated the results of their CPIA scores. Yet, in higher-CPIA countries, the evaluation portfolio reviews to implementing agencies-with found statistically significant improvements in CASs that important differences across countries and regions. In reportedly used intergovernmental fiscal rules and civil the post-GAC period, however, the evaluation observed service personnel rules. more frequent disclosure of portfolio reviews in IDA countries and fragile states. There was considerable Harmonizing GAC Efforts. The Bank generally main- variation in transparency efforts: for instance, country tained its support for harmonization of GAC efforts programs in East Asia and the Pacific set the standard in the FYOS-10 period, though its GAC-in-projects with 50 percent of pre-GAC and 75 percent of post-GAC approach led to backtracking in some cases. A sizeable CASs disclosing portfolio review results. Improvements majority of country programs in the pre- and post-GAC were also evident post-GAC in Sub-Saharan Africa and periods were committed to sharing information with oth- South Asia. Yet, lower frequency in disclosure of portfo- er donors and conducting joint analytical work on GAC 46 World Bank Country-Level Engagement on Governa nce and Anticorruption issues. A similar proportion of Bank CASs continued to the launch of the strategy. Over the period reviewed, support joint portfolio reviews, SWAps, and joint serial Bank-country engagement became increasingly budget support operations. In some countries, however, concerned with the management of natural resources the Bank sought to strengthen fiduciary controls on its (such as forestry and fisheries) and extractive indus- own projects by ring-fencing preventive measures rather tries (such oil, gas, minerals, and mining). Country than harmonizing them with other donor projects (for ex- programs in Africa, followed by those in Europe and ample, the required use of Good Governance Frameworks Central Asia and the Middle East and North Africa, and similar initiatives in Cambodia and Azerbaijan). were most likely to identify and support GAC efforts in these areas. Measuring GAC Results. In the pre-GAC period, the • The third most prominent GAC-in-sectors in CASs Bank's use of GAC-related process, actionable, and related to service delivery in the social sectors. Over institutional outcome indicators in CAS results frame- half of pre- and post-GAC CASs identified GAC in works was extensive. However, continued and consis- human development sectors as an entry point for tent identification and use of these indicators in the Bank-country engagement. These issues were most post-GAC period was a concern. Nearly all CAS results frequently identified in Europe and Central Asia and frameworks in the pre-GAC period included process and Sub-Saharan Africa pre-GAC, and in Latin America actionable indicators for core public sector and invest- and the Caribbean, and Sub-Saharan Africa post- ment climate entry points. A majority of CASs included GAC. process and actionable indicators for accountability institutions, although fewer did so for the demand side. • GAC-in-infrastructure was also a key entry point Virtually all of pre-GAC CASs collected data on process, in more than a third of CASs before and after the actionable, and to a lesser extent, institutional outcome launch of the strategy. The plurality of CASs that indicators relating to public sector reform and invest- identified infrastructure was in Sub-Saharan Africa, ment climate issues. Also, data on accountability and the followed by Europe and Central Asia and East Asia demand side were less frequently collected. and the Pacific. GAC-in-infrastructure was rarely identified as an entry point in Latin America and the Caribbean. GAC Elements in Sectors and Themes In its review of primary education, roads, and account- Efforts to address governance constraints in sectors ability institutions, the evaluation found that progress was a major thrust of Bank engagement before and on GAC-in-sectors was highly uneven. On the one after the launch of the strategy. Virtually all the sample hand, GAC elements in roads projects improved dramati- CASs-across regions, levels of income, and governance cally in quality of governance and PEA, fiduciary aspects, performance-identified strengthening of sector institu- and use of country systems. On the other hand, primary tions as a GAC entry point. The evaluation observed the education projects approved post-GAC weakened in the following patterns in the Bank's GAC-in-sectors work: quality of enhanced fiduciary aspects. Projects across these three areas scored worse for incorporation of the • Financial and private sector development were the demand side, identification of GAC results, and support most commonly identified GAC-in-sectors issues, for institutional strengthening. even though these were not formally part of the first two years of GAC implementation efforts. Over 85 percent of country programs in both the pre- and GAC Elements in Projects post-GAC periods identified financial and private Through its GAC-in-sectors work, the strategy sought sector development as a GAC entry point. Most of the to promote "smarter project design?' According to Bank CASs that identified this entry point were in Africa, guidance, smarter operations would systematically incor- followed by those in Europe and Central Asia, Latin porate political-institutional analysis; ensure good "fit" to America and the Caribbean, East Asia and the Pacific, governance realities; promote the use of country systems; the Middle East and North Africa, and South Asia. support institutional strengthening; include demand-side • Half of CASs before the launch of the strategy sup- measures; and effectively manage risks. Based on a desk ported efforts to improve governance in natural review of these elements across 200 operations, the evalu- resource management; these efforts expanded after ation observed the trends discussed below: Effectiveness: Is the Bank More Responsive to GAC Issues? 47 Institutional analysis (formal or informal) Formal institutions 83 90 Relevant historical legacies so 63 Cultural practices, norms, or other traditions 36 29 Informal relations among different levels of government 32 35 Social, regional, ethnic, religious, or linguistic relations 44 38 Electioneering and/or electoral cycles 15 13 Rent-seeking 26 38 Other 16 21 Any institutional analysis 92 92 ESW sources (used to inform analysis) Social Assessment 31 31 Poverty and Social Impact Assessment 26 40 Public Expenditure Review 23 28 Country Economic Memorandum 15 21 Institutional Governance Review 12 4 Technical Assistance 51 47 Political Economy Assessment Other 63 71 AnyESW 71 74 Source: lEG desk review. Note: Pre- and post-GAC columns show percentages of projects that use specific types of analysis. Incorporation of Political-Institutional Analysis. The pointed to some improvement in the following elements evaluation found that virtually all projects in the pre- of project fit: and post-GAC periods incorporated at least some analysis offormal or informal institutions (Table 6.1). • The proportion of projects for which design was Nearly 90 percent of projects included analyses of formal adapted to informal institutions to a great extent institutions, such as organizational structures, decision- increased slightly over pre- and post-GAC periods making rules, implementation arrangements, staff skills, {from 38 to 40 percent). Improvements were also ob- and disclosure mechanisms. They less frequently assessed served in IDA countries as well as those with higher informal institutions, such as history, culture, social re- CPIA governance scores. Variation across regions lationships, rent-seeking, and electoral processes. A high was found to be significant in the post-GAC period, proportion of projects drew their political and institu- with particularly well-adapted projects in Europe and tional analyses from formal Bank ESW The evaluation Central Asia and Latin America and the Caribbean assessed the intensity of political-institutional analysis on compared with Sub-Saharan Africa and South Asia. projects in the pre- and post-GAC periods and observed Regression analysis indicated that projects were better adapted when they included diagnoses of informal only marginal improvements. The average number of institutions and when they supported demand-side distinct political-institutional issues reflected in project objectives. Also, projects that used project implemen- documents increased slightly from 2.9 to 3.2. The average tation units were less likely to be "fitted" to informal number of ESW products cited increased from 2.2 to 2.5. realities (Appendix E, Table E.lO). "Fit" to Governance Realities. The evaluation found • Post-GAC projects more frequently supported modest-not statistically significant-improvements strengthening management of relevant public agen- in project "fit" to governance realities. The desk review cies (40 percent pre-GAC compared to 47 percent 48 World Bank Country-Level Engagement on Governance and Anticorruption post-GAC). Regression findings indicated that opera- operation, or its share of commitments in the country tions that used subprojects were 30 percent more likely portfolio. Presumably, such operations were more likely to strengthen management of relevant public agencies. to come under scrutiny for issues of inclusion. The Those that supported accountability objectives were 22 evaluation also confirmed that projects that were highly percent more likely, and those in the human develop- responsive or highly "fitted" to governance realities were ment sectors and in countries in Latin America and those that extensively used institutional analysis (Figure the Caribbean were 25 and 32 percent more likely to 6.2). For example, more than 50 percent of moderately support these agencies (Appendix E, Table E.ll). responsive and approximately 80 percent of highly responsive projects analyzed three or more types of • Post-GAC projects supported strengthening of institutional constraints. By contrast, less than 6 percent rules-based decision making and accountability of nonresponsive projects analyzed more than three types more frequently than pre-GAC ones. Statistically of constraints. significant improvements over the post-GAC period were observed in countries with higher CPIA scores, Planned and Actual Use of Country Systems by Bank as well as significant regional variation (for instance, Projects. Aside from statements of intent in CASs, 56 percent in East Asia and the Pacific compared to the Bank's commitment to use country systems is 27 percent in the Middle East and North Africa). best measured by evidence from projects. Projects Here too, regression analysis indicated that operations were reviewed along multiple dimensions: for instance, using subprojects and those supporting accountability designation of the project as on -budget, aligning with the objectives were 25 percent and 16 percent more likely fiscal calendar, relying on the range of budget preparation to support strengthening of rules-based decision and execution procedures, as well as intergovernmental making (Appendix E, Table E.ll). transfers, audit, and national competitive bidding (NCB) procedures for procurement:w At the outset, the evalu- • Projects approved in the pre- and post-GAC periods ation confirmed that the use of project implmentation employed proactive measures to include disadvan- units did not preclude the use of country systems. taged groups to more or less the same degree. The evaluation found a statistically significant increase A majority of pre- and post-GAC projects planned only for projects approved in CGAC countries. to use at least some country systems. The evaluation Overall, the evaluation identified three elements of observed more ambitious project-level commitments good project fit to informal realities-linkages to only in financial management. Pre-GAC projects most relevant ESW, incorporation of beneficiary feedback commonly proposed to use NCB procurement, civil mechanisms in project design, and the use of sub- service personnel, and budget execution procedures. The projects. Another element was the risk exposure of the evaluation observed a statistically significant increase in Institutional Analysis and Project "Fit" to Governance Realities 60 60 "' tl Not responsive .."' v ·e- Gl 40 / Gl -~ 40 -c. .. 0 c Moderately responsive .···.~ -c. .. 0 c .. .....·· ../.. Moderately responsive ~ . Gl 20 ; ... v 20 ~ Gl c. .. ·.. 0 0 0 2 3 4 5 6 7 8 0 2 3 4 5 6 7 8 Formal and informal institutional analysis Reliance on ESW Source: lEG desk review. Note: The horizontal axes reflect a cumulative score (0-8) of the types of institutions analyzed and the number of ESW used in projects Effectiveness: Is the Bank More Responsive to GAC Issues? 49 the share of post-GAC projects that committed to put the community organizations, monitoring and evaluation, project on-budget, align with fiscal calendars, and use and cross-cutting human resource systems. budget execution and auditing systems. Pre- and post- GAC projects in certain regions (South Asia and Latin Demand-Side Measures. The use of demand-side America and the Caribbean) were more ambitious than measures in projects was not new to the Bank. How- those in others (the Middle East and North Africa and ever, there was no systematic improvement in their use East Asia and the Pacific). In fragile states, projects that post-GAC except for greater provision for borrower planned to use some country systems dropped from 55 to disclosure (Figure 6.3). The desk review covered design 27 percent. aspects since implemention data were sparse. Nearly half of pre- and post-GAC projects employed mechanisms to The Bank largely followed through on commitments promote beneficiary participation in decision making and to use country systems in its projects approved during implementation, including special institutional arrange- the pre- and post-GAC periods. However, there was ments, stakeholder capacity building, and beneficiary no evidence of systematic improvement in UCS over involvement in subproject selection. Examples of institu- the post-GAC period. When they committed to do so, tional arrangements designed to promote participation approximately 80 percent of projects were on-budget, were parent-teacher, water user, or community forestry aligned with the fiscal calendar, and used budgeting pro- associations. Less frequently employed-in less than a cedures (through provision of large-scale operation and third of sample projects-were mechanisms by which maintenance), NCB procedures for procurement, civil beneficiary feedback could inform project redesign or service personnel rules, and audit procedures. Although other midcourse corrections during implementation. the quality of implementation data for projects approved Even fewer-11 percent in projects approved pre-GAC post-GAC was poor, the desk review found that use of and 4 percent approved post-GAC-provided for benefi- NCB rules declined. ciary or CSO verification before payments were made at the local level. Institutional Strengthening. A majority of pre- and post-GAC projects planned to support institutional Also relevant to the demand side of governance was strengthening, primarily for public sector organiza- disclosure and transparency ofproject information. In tions. Strengthening of cross-cutting financial manage- the post-GAC period, the Bank more systematically ment and procurement systems was proposed in half of strengthened provisions for borrower disclosure. Project all sample projects, but it grew significantly in projects in information covered by these provisions included pro- fragile states and IBRD countries, particularly in Eu- curement and overall audit reports, financing and budget rope and Central Asia. Far fewer operations-less than allocations, progress reports, and scorecards. Forty-eight a quarter in both pre- and post-GAC periods-focused percent of post-GAC projects included provisions for on strengthening human resource systems. Proposed borrower disclosure compared to 31 percent of pre-GAC strengthening of community organizations, particularly ones. Nearly a half of projects pre- and post-GAC support- in Africa (60 percent pre-GAC and 40 percent post- ed communication plans and result measures on transpar- GAC), and local governments, particularly in South ency mechanisms. Only three percent of sample projects-· Asia (67 percent pre-GAC and 20 percent post-GAC) approved in the pre-GAC period- required compliance dropped precipitously. with national right-to-information legislation. As with UCS, Bank projects in large measure followed The majority of Bank operations made provisions for through on plans to strengthen country institutions. beneficiary monitoring and oversight, while less than However, it is too soon to tell whether implementation a third included grievance and redress mechanisms. was more likely on projects approved post-GAC. When In general, the evaluation found no significant change they planned to do so, over three-quarters of Bank proj- post-GAC. Of projects that included beneficiary moni- ects over the entire period of review provided support toring, about a third each included third-party moni- for strengthening cross-cutting financial management toring, public service delivery surveys, and consumer and procurement systems, specific public sector organi- satisfaction surveys. Fewer adopted citizen charters or zations, and local governments. Slightly fewer (but still boards, citizen score cards, Continuous Social Impact over 70 percent) followed through on plans to strengthen Assessments, public expenditure tracking surveys, and 50 World Bank Cou ntry-Level Engagement on Governance and Anticorruption social audits. A third of projects incorporated grievance also proposed in pre- and post-GAC projects. While and redress mechanisms in their design. Of these, an the frequency of these measures increased, follow- overwhelming majority (over 90 percent in both pre- and through was mixed. Plans for anticorruption action plans post-GAC periods) relied on formal institutions such increased significantly in the post-GAC period. Other as appeals committees, ombudsman offices, and other frequently proposed measures included transparency complaints offices. About a third of pre- and post-GAC measures for fiduciary aspects, followed by monitor- projects also required records of complaints received and ing and evaluation of GAC measures, communications actions taken. Far fewer required action on all complaints plans, on-site field verification, random post-audits, and or written feedback. special arrangements for high-value, high -risk contracts. The evaluation observed that over 60 percent of pre- and Risk Management, Including Quality of Enhanced post-GAC projects followed through on their plans to use Fiduciary Aspects. A key priority for the GAC strategy transparency measures, on-site field verification, random was management of project-level fiduciary risks. The post-audits, and technical audits. Only half carried out evaluation confirmed that post-GAC Bank projects anticorruption action and communication plans, and more frequently cited the risk of the misuse offunds even fewer undertook reviews of high-value, high-risk (for instance, graft, fraud, corruption, or leakage). It is contracts. important to note, however, that more frequent identifi- cation of risks in project documents did not necessarily As part of its risk management efforts, the Bank also mean that these operations were inherently riskier. Only planned to supplement its own supervision efforts 7 percent of pre-GAC and none of the post -GAC projects mainly by leveraging the efforts of other development that were restructured cited fiduciary issues. partners. The frequency of enhanced supervision increased systematically post-GAC. About a quarter of Aside from standard fiduciary requirements for pre-GAC and a third of post-GAC projects sought to un- borrowers, specialized risk-mitigation measures were dertake joint supervision with other donors and decen- Demand-Side Measures in Pre- and Post-GAC Projects Percent 0 20 40 60 80 100 Institutional arrangements for beneficiary participation ~.2 ., - ... "' .. Q. Beneficiary involvement in decision making over sub-project selection or resource utilization ., ·u c: ·;; cu ... Beneficiary verification before payments made at the local level ~ "' 0.. Capacity building for stakeholder participation Provision for beneficiary feedback to impact program redesign or implementation Measures to comply with right-to-information rules ~ D Pre -GA c 0 c ..... cu.2 Provisions for public disclosure of project documents D Post GAC ., E - :I "' 0 ... ... 0 Communication plan with channel(s) for public disclosure .. -..: Indicators to monitor tranparency mechanisms Formal mechanisms for grievance redress .. . Grievance redress agency with authority c ~ to take or demand remedial action 01"0 cu .. a:: .. = "' a.c cu ... Formal grievance redress mechanisms obliged to take action on all complaints E~ Mechanisms to maintain records of 0 .!! u ~ complaints received and action taken Grievance redress agency obliged to provide written feedback on actions taken Source: lEG desk revi ew. Effective ness: Is the Ban k More Res po nsive t o GAC Iss ues? 51 tralized supervision. Post-GAC projects also more fre- post-GAC periods has remained uneven. Fewer projects quently planned to increase internal supervision budgets, post-GAC (10 percent) "fell through the cracks"-or owing in part to the more favorable internal resource received no risk review at all-relative to pre-GAC (16 environment for GAC efforts. The evaluation found that percent). Yet, the share of projects with no risk review the Bank followed through on its enhanced supervision during implementation increased significantly between plans in a majority of pre- and post-GAC projects. For pre-GAC (32 percent) and post-GAC periods {60 instance, two-thirds of sample projects followed through percent). More generally, the evaluation concluded that on plans for integrated procurement, financial manage- variations in "risk management intensity" were explained ment, and technical review, as well as decentralized in part by the choice of instruments, in particular DPLs supervision. The only exception was the very low number (Box 6.3). of projects (15 percent) in which joint donor supervision plans materialized. Results Orientation. Tracking implementation of generic GAC elements-such as demand-side, fidu- Despite the GAC strategy's emphasis on risk manage- ciary, and PEA aspects-through project-level results ment, application of controls during the pre- and frameworks was not new to the Bank. Indicators for Risk management intensity was defined as the total number of relevant upstream and downstream risk management measures used in a single project. Relevant measures included links to economic and sector work, governance and anticorruption plans, supplemental supervision, grievance mechanisms, and disclosure measures. Regression analysis found that risk management intensity fell significantly when the project was supported by a DPL and when it was mapped to either the Finance and Private Sector Development or PREM networks. At the same time, the intensity increased when the project was mapped to the Europe and Central Asia and South Asia Regions (Appendix E, Table E.l6a-b). As distinctions between fast-disbursing and investment lending become less pronounced, there is a need to ensure a consistent risk-management approach across financial products. In recent years, for instance, the use of large-scale financing of operations and maintenance expenditures, report-based disbursements, and subprojects have made traditional investment lending more flexible, and, in some cases, similar to targeted budget support. Yet, as noted in chapter 5, the control framework for investment loans and DPLs remains distinct both in design and application. Factors Associated with Risk Management Intensity of Bank Operations - 2.0 L. ,. _p_ 'vi{;.~...._q; diiS£ .L .,, Development Policy Loan - 1.6 ~ o e 0::"' z"' - 15 ~ 0 u 0 ~----~----~----~----~----~ - 20 L --,----------.----------.-----------.- - 15 - 10 -5 0 5 10 - 20 - 10 0 10 CPIA: Deviation from cluster average GNIPC: Deviation from cluster average Appendix A: Methods and Data Set Description 75 Based on the Level 1 desk review findings, and in line with evaluation questions, the evaluation identified those country programs with documentary evidence of some degree of GAG-responsiveness over the pre- and post-GAC periods (Table A.4) . Therefore, those country programs without relevant documentation were eliminated: Angola, Morocco, and Paki- stan, did not have CAS documents in the post-GAC period (FYOS-2010), and Guinea did not have a country strategy during the evaluation period. Subsequently, countries that posed logistical difficulties within lEG's tight timetable were also dropped (Iraq and Mozambique), even though they were found to be GAC responsive in the pre- and post-GAC periods. The resulting six case study countries were Azerbaijan, Bangladesh, Cambodia, Guatemala, Liberia, and Mol- dova. Field missions to these countries were undertaken between December 2010 and February 2011. Desk Review Implementation. Questionnaires were designed and used to evaluate country programs and projects in the Levell analysis. Each questionnaire consisted of questions on the content of governance and anti-corruption mea- sures in country program and project documents as well as ratings to evaluate the extent to which country programs and projects incorporated GAC measures in their design and implementation. The questionnaires were designed over a two-month period by two members of the lEG evaluation team. The project questionnaire was designed to build on the QAG 2009 GAC-in-Projects Benchmarking Survey as well as the logical framework used for this evaluation (see Figure 2.l).The questionnaires were peer-reviewed and pre-tested in May 2010. Out of existing Levell samples, five country program CASs and projects were randomly selected for pre-test. 47 Two team members evaluated a country program/project independently using the final draft of the questionnaires. Once completed, the results were matched and the team further elaborated on the questionnaire design based on findings of the pre-test. The following documents were used for the desk review: • For country questionnaires (used for country program evaluation): all CAS documents as defined by Operations Pol- icy and Country Services (OPCS), CASCRs for each CAS cycle, Country Portfolio Performance Reviews (CPPRs), Desk Review Ratings for Level 3 "Candidate" Countries GAC Responsiveness" Pre-GAC (2004-07) Post-GAC (2008-1 0) High/High + Improving 1. Bangladesh 13 13 2. Moldova 11 12 Somewhat+ Improving 3. Cambodia 9 10 4. Guatemala 7 10 S.lraq 9 10 Somewhat+ No Improve/Deteriorating 6. Liberia 10 10 1 7. Azerbaijan 12 10 8. Mozambique 10 10 Docs for only one period High but data in only one period 9. Pakistan 11 N/A 10. Morocco 11 N/A Somewhat but data in only one period 11 . Angola 9 N/A No docs for either period 12. Guinea j N/A l N/A ------------------------------------------------------------------------------------------------------------------------- ----------------------------------------------------- ----------------------------------- • GAC Responsiveness score out of a total of 15. 76 World Bank Country-Level Engagement on Governance and Anticorruption CAEs, programmatic (serial) Development Program Lending Implementation Completion Reports (DPL-ICR), and programmatic (serial} DPL-ICR Reviews by lEG. • For project questionnaires (used for project evaluation): PADs for all lending projects and the Trust Fund Agreement for all Trust Fund financed operations. For all closed projects, implementation was evaluated based on the ICR. For open documents, evaluators reviewed the last ISRs and where available the last Back-To-Office (Memorandum or Aide-Memoire for the last mission.48 The workload, which consisted of a review of 134 country programs in 44 countries 49 and 200 projects, was distributed equally among the eight lEG team members. Out of the 44 documented countries, Botswana, Chad, and Namibia only had one document. The rest of the countries had at least two documents (though not all of them supporting the post- GAC period). The maximum number of country documents in any one country was six (Bangladesh and Indonesia). Where documents only supported one of the periods, only the period's questionnaire, either pre- or post-GAC was filled out. For projects, there was an average of four documents per project, which yielded a catalogue of 1,118 project documents for all 200 projects. All projects had at least one document. The desk review of country programs was undertaken between July and September 2010 and the review of projects was carried out between September and November 2010. To ensure data quality, evaluators in the lEG team used Survey Monkey to fill out questionnaires. To ensure consistency and quality, each completed country CAS and project ques- tionnaires was peer reviewed by the task team leader and a senior team member. In addition to peer reviews, there was double data checking in data entry. Two team members reviewed data entry for each country and project questionnaire. Intercoder reliability, or the extent to which independent coders evaluate a characteristic of a message or artifact and reach the same conclusion, is a concern which is necessary to address for producing valid and reliable data results in content analysis. To ensure intercoder reliability there was a pre-test of the survey instrument and the authors tested for intercoder reliability across the eight coders to ensure individual coders would produce reliable and consistent results. Questionnaires that did not produce consistent results were discussed in an iterative review process with coders and re- vised where appropriate so coding would be consistent, reliable, and ultimately valid. Coders participated in a two-day training on how to complete the questionnaire before the testing period. Further, lEG was fully aware of potential reliability problems that might arise in sharing hypotheses with coders and made sure to provide coders with limited information about the details of the hypotheses before and during the evaluation pe- riod. The team involved two individuals in constructing the questionnaires used in the desk review who were not involved in the coding process. The peer review process of coders' findings further helped ensure reliability across coders. Thematic Reviews Four thematic analyses reviewed the evolution of Bank practice in the following areas: GAC issues in roads, primary education, accountability institutions, and political economy analysis. Each thematic review included a review of the literature on GAC in the state of development practice. The initial literature appraisal identified issues and questions to be covered in the desk reviews of projects and relevant ESW The subsequent analysis of each theme followed the meth- odologies described below. Political Economy Analysis Review. This review covered the main currents in the political-economic literature, Bank guidance provided in the context of the GAC strategy, as well as the actual application of political economy analysis (PEA) in Bank economic and sector reports and freestanding PEA "inputs" to operational work. For the review of ESW for political economic concerns, a review sample was selected. Based on the countries selected for Level-2 analysis in accordance with the approach paper for the larger GAC evaluation, all ESW for these countries from 2004 onwards was compiled. The final list consisted of 877 ESW reports covering 35 countries. This population of reports was then stratified according to pre-GAC or post-GAC timing (2004-07 versus 2008-10), Region (across the six World Bank Regions), and report type (core diagnostics consisting of Country Economic Memoranda, Public Expenditure Appendix A: Methods and Data Set Description 77 Political Economy Analysis and Sector Work (PEA-ESW) Sample Country Year I ReportType Project ID Azerbaijan FY09 Country Economic Memorandum (CEM) P107278 Bangladesh FY08 I Poverty Assessment (PA) P099963 Cambodia FY06 Poverty Assessment (PA) P085013 Haiti FY06 Country Economic Memorandum (CEM) P094865 Indonesia FY09 Development Po licy Review (DPR) P102317 Iraq FY08 Oil Sector Policy Note P096434 Kosovo FY08 Other Social Protection Study P107761 Liberia FY09 Public Expenditure Review (PER) P107304 Mali FY04 Other Infrastructure Study P078247 Morocco FY09 Other Public Sector Study P112606 Mozambique FY09 Food Prices Policy Note P113442 Nicaragua FY08 Institutional and Governance Review (IGR) P101317 Pakistan FYOS Country Financial Accountability Assessment (CFAA) P091615 Paraguay FY07 General Economy, Macroeconomics, and Growth Study P090165 Sri Lanka FY04 Development Policy Review (DPR) P084614 Syrian Arab Republic FYOS Investment Climate Assessment (ICA) P078466 Uzbekistan FY04 Public Expenditure Review (PER) P079105 Reviews, Development Policy Reviews). Taking into account these strata, a random sample of 17 documents was selected. The sample was balanced across Regions, between pre- and post-GAC periods, and between core and non-core reports. For the review of PEA inputs, a sample was drawn from the Political Economy Community of Practice's (PECoP) list of applied PEA completed since 2006 in 39 countries (another 4 are cross-national within regions, 2 are multi-regional or global. Of the 71 reports in that sample, 59-over 80 percent-were started after FY08. Forty one are country-level (or multi-country) analyses, another 38 are primarily sectoral-based PEA, and only 2 are project-specific risk assessments. The Africa and Latin America and Caribbean regions account for over 50 percent of completed PEA, while the rest are spread across four other regions. From these 71, 15 were selected for in-depth review, based on regional representation (the sample covers all six World Bank Regions), timing, and country/sector balance.5° Reviews of GAC-in-Sectors. Three additional thematic reviews examined GAC in the following sectors/themes: prima- ry education, roads and highways, and domestic accountability institutions. These reviews each had a similar framework. They reviewed the literature on GAC issues in the sector or thematic area, assessed the Bank's approach and guidance in the context of the GAC strategy, and summarized findings from lEG's desk review of GAC-responsiveness of projects in the relevant sector or thematic area. Findings and lessons were also drawn from the case studies. Resourcing Review Funding and Staffing. lEG's review of the resourcing of GAC included analysis of incremental Bank budgetary resourc- es as well as donor funds supporting GAC work via the GPF. Data on trends in staffing and training also were analyzed. For analysis, a sample of all approved GPF projects was selected, representative of the whole portfolio by regions and sectors, lending windows, grant size, and disbursement rate. Accordingly, the following filters were used: regions (by vice presidential unit), sectors (network vice presidential units), status (established trust funds only-77 of them as of June 22, 2010, when sampling was completed), amount (grants of above $500,000 should make roughly half of the sample), and have all three windows represented. 78 World Bank Country-Leve l Engagement on Governance and Anticorruption The main source for the population was a project database submitted by the GPF Secretariat. This was cross-checked with information available in the Bank's own Business Warehouse and on the GPF's official Web site. It should be noted that about one-third of commitments (amounting to around $20 million) in the GPF's official list were not considered as they were too recently approved and had low actual disbursements. This narrowed the population size to 56 projects with total commitments of $36 million, the population from which the sample was then drawn. To establish linkages with the GAC strategy, the sample needed to include all countries selected for the evaluation's Level 2 analysis. These include CGAC countries as defined by the GAC Council and all Window One countries. An overview of the 91 projects resulted in 13 projects to be channeled to Level2 countries, including 2 where at least one Level2 country was repre- sented. These were included in the sample. The rest of the sample was selected randomly out of the population of 56 projects less 13 CGACs and Window One, ensuring adequate representation of sectors and regions (as defined by central networks and regional vice presidencies), grant size (eliminating those grants under $0.5 million), disbursement rates and grant windows. Based on the principles laid out above, 30 actively disbursed projects were selected. Two tailed tests were conducted testing the difference in means between the sample and the population. The p-values produced showed no significant difference in CPIA Public Sector Module, income levels, and regional representation between the sample and population projects. 51 Process Reviews. An important thrust of the GAC strategy was strengthening Bank internal control,s. Therefore, as part of its evaluation of key Bank controls, lEG analyzed operational and GAC implementation progresses with a view to as- sessing the effectiveness of the Bank's risk management and internal change management efforts. This included reviews of Bank budgeting and GPF allocation processes, reviews of GAC implementation arrangements, and comparative analysis of existing process maps for IDA controls in development policy and investment lending. Surveys, Structured Interviews, and Consultations lEG GAC Staff Survey. A survey of staff was undertaken in October 2010 to determine attitudes and perceptions of the Bank's work on GAC issues. The survey was designed on the basis of the GAC strategy's First and Second Year Progress Reports, extensive consultations with operational and network staff across the Bank, feedback from attendants to a GPF Window One Workshop in Cape Town, South Africa, in October 2010, and minutes ofGAC Council and Board meet- ings over 2009 and 2010. Several iterations of this survey were tested with operational staff (including from South Asia, Sub-Saharan Africa, the Development Economics Vice Presidency, and East Asia and the Pacific). The team also undertook several review iterations to ensure consistency with the evaluation Approach Paper, as well as with early findings of desk reviews and thematic reviews (for example, of political economy analysis and GAC-in-sectors). The survey targeted a broad cross-section of staff at technical levels, GF, and above who have served as task team members for 892lending and trust-funded Bank operations approved between FY04 and FY10 from which the Level 1 sample of 200 project desk reviews was derived. Since these operations were undertaken in 50 countries, the target audience also includes staff (technical level and above) from Country Management Units in these 50 sample countries. The target audience covered staff mapped to headquarters and field offices, sector, and country management units, as well as families and networks. The survey was sent out to 1,942 staff members, 682 (35 percent) of whom responded. The survey was launched on October 20 and was closed on November 30, 2010. As an immediate step, the repre- sentativeness test was carried out. Overall, the network, duty station and regional affiliation of those who responded (682 staff) versus population (1942 staff) followed the same structure there was no evidence of over-representation of respondents versus population. Structured Interviews. Extensive interviews were conducted with current and former senior Bank officials, Bank teams at headquarters and in the field, development partners, as well as key country stakeholders to ascertain the rationale, Appendi x A: Methods and Data Set Description 79 trade-offs, and emerging implementation issues underpinning the 2007 GAC strategy. In addition, external surveys were reviewed to gauge perceptions of the Bank's work on GAC issues. Consultations. The evaluation team held separate group consultations-one with the GPF donors on AprilS, 2011 , and a second with civil society organizations (CSOs) on April19, 2011. The GPF donor consultation, held via videoconfer- ence, involved key representatives to the GPF Council from the Netherlands, Norway, and the United Kingdom. Through this consultation, lEG sought the donor views on the original intent of the GPF in the context of the GAC strategy, the strengths and weaknesses in the design of the GPF, and lessons learned to date. Similar, the consultation with CSOs focused on their perspectives on progress made over the course of GAC implementation, as well as the experience with Bank engagement on the demand side of governance. In conjunction with lEG's in-person session with CSOs, a Web- based discussion platform was launched to gather written feedback, including from institutions around the world. 80 World Bank Cou ntry-Leve l Engagement on Governance and Anticorruption APPENDIX B Resourcing of GAC Strategy-Funding and Staffing FIGURE 8.1 Regions' Spending on Governance Work-Change in FYOB-1 0 AFR EAP FY08 FY09 FY10 FYOS- 10 -6 ECA 2 LCR 4 1.5 2 :::f! GPF 14% Dedicated GAC Staffing Planned and Actual GAC Staffing (GF and Above)-AFR, ECA, MNA, and SAR AFRaJ ECA"1 SAR"' Number. of Staff by Location Planned Actual Planned I Actual Planned Actual Planned Actual Washington 1GH, 2GG, 2GH, 3GG, i 2GG, 1GF, 1GH, 1GG, 1GH,4GG 3GG 1GF 2GF i 1ETC 4ETC 3G ----~----~----G 3- ---l- ~- ~:-;- ~- /-+-- ~- __ 3G 1 Field-Local '' 4GF 1GH,2GG, --- i H~- ;-~- G-, ___, 3GF Field-International'' 1GF,6GG 1.5GG 1GH, 6GG Totals 15 13 5.5 9 20 15 Sources: GAC Secretariat and Regions. a/ Includes 1GG and 1GF who jointly manage the GAC program from Washington, and other staff who parti cipate substantively in GAC program delivery and advice. Grad e levels in "Actual" colum n assumed by lEGs, based on position titles: Lead PS Specialist-GH; Senior PS Specialist and Senior Econom ist-GG; PS Specia li st, Governance Special ist, Operations Officer, and YP-G F. b/ E CA specifically decided to make GAC work the responsibility of all its staff and, accordingly, did not create any dedicated positions for GAC. c/ Includes only staff in MNSPS. Information for other MNA units was not readily available. d/ SAR GAC staffing also includes 5 Program Assistants (G() and 1 Operations Analyst (GEl in the field. el Information provided by the Regions did not distinguish between Field-Local and Field- International Appointments. 82 World Bank Country-Level Engagement on Governance and Anticorruption Overall Staffing for PREMPS, OPCFM, and OPCPRC, FY04- 10 IBRD Total Staff Mapped to PREM-PS PREM-PS Staff by grade 80 __ ... ......... ... ... ------· GF GG GH Gl Total 70 60 2004 37 51 26 4 118 so ------ --- 2005 31 51 22 4 108 40 2006 23 54 18 4 99 30 ..... ~ ________• ......-..:..;-::::......__....<..::..;. 2007 18 59 18 4 99 20 2008 19 67 19 I 2 107 10 2009 27 67 20 3 117 0 •····························•····•··•····•·· 2004 200S 2006 2007 2008 2009 2010 2010 30 68 22 I 4 124 - - GF - - GG - GH ••••• Gl - - - ,- OPC-FM Staff by grade IBRDTotal Staff Mapped to OPC-FM 100 GF GG GH Gl Total 90 ....... ----- -------- 2004 49 68 18 2 137 80 70 60 ----- ______ .... ---- ------ so ...... .......... 2005 43 69 21 134 2006 40 74 20 135 40 30 2007 2008 41 46 80 84 20 21 142 152 20 10 0 +---" ·~· ...., . •i" -·~ •• · .._, ,·~ ._. • , ·• ··~ u •~ •• ·.._, u ••-r •~ •• •u• •"' ··~ , ·• • u ·r •~ •• ..,.,_, ••1· .., ,._, •• • •1 ,• ....,..... • .• .~ , ' ·~ . ,._•f.! " •._______, ·• "-' 2009 49 85 23 158 2004 200S 2006 2007 2008 2009 2010 2010 51 88 21 161 - - GF - - GG - GH ••••• Gl - - - ,- IBRDTotal Staff Mapped to OPC-PROC I 90 OPC-PROC Staff by grade ________ ... GF GG GH Gl Total 80 70 ____ ,...,. ...... ..... ............. ____ ,, ,,..----- ............. . 2004 47 62 23 1 133 ! 60 ~ ~: ~j ~ ~ ~~ so 40 30 20 I I 10 ··~·•••· o +-~~~~~AUoo~u·~••AU·•~·~••••~•··~~·•u•·•~•u•~ ~~~-, ;;:__ -~--L ;: ;; L :; j 2004 200S 2006 2007 - - GF - - GG - 2008 GH 2009 2010 ••••• Gl PREM-PS OPC-FM OPC-PROC by grade IBRD Total Staff Mapped to PREM-PS, OPC-FM & OPC-PROC 2SO - ...... ------ ...... GF GG GH I Gl Total 200 ----- ....... 2004 2005 133 117 181 181 67 65 I 7 6 388 369 1SO 100 .......... __ ................. ---- ........ ------ 2006 102 196 56 6 360 2007 111 212 56 6 385 so 2008 123 223 6o I 4 410 0 +-~·• •'-" "-' ••"-'•'-"•-"'••"-'·~ •r-c ·-"' •• "-' • '-" ·~ "-' ·· •r•~ ·· • "-' u•~ • y..t ••··~ • .,..,,.,,.u, ·~•••u•••t.O. .,.., ••.•u··~ , !.__--, · •••!.!• 2009 133 224 I 5 430 2004 200S 2006 2007 2008 2009 2010 - - GF - - GG - GH ••••• Gl 2010 130 235 :: I 6 439 ' - - - - - - ' - - - - - ' - - - - - ' - - - - - l_ _ _ _...L_.__ _ _ _, Source: World Bank human resources data. Appendix B: Resourcing of GAC Strateg y-Funding and Staffing I 83 APPENDIX C Summary Findings on GAC Responsiveness The ratings presented in the tables are based on desk review of Bank documents carried out by the lEG GAC Evaluation team. To a Great Extent Somewhat I #of CAS/CPS reviewed j To a Great Extent I Somewhat Please rate the extent to which the Bank's approach incorporates Explicit assessments of governance and political economy constraints l· . 59% 39% 41 59% 38% 37 Explicit choice of governance entry points 51 % 49% 41 49% 51 % 37 Mix of financial and knowledge inst ruments 29% 68% 41 38% 51 % 37 Ident ification of resu lts measures 18% 70% 40 24% 62% 37 Please rate the extent to which the CAS Includes country strengthening for ___ _ Core Public Sector Inst itutions l 66% 1 34% 41 65% 35% 37 Sectoral state inst itutions 35% 60% 40 43% 57% 37 I Domestic Accountability institutions 23% 52% 31 35% 35% 31 Civil Society and the Demand Side 14% 63% 35 9% 62% 34 The Investment Climate 34% 56% 41 35% 54% 37 Overall Rating of signaling of GAC concerns and risks through Bank portfolio pr ocesses--Please rate the extent to which Portfolio Risks are regularly monitored by the Bank ] 44% I 44% 41 51% 41 % 37 Portfolio processes track the progress of governance reforms at sector I I and project levels 32% I 56% ! 41 38% 46% 37 Portfolio reviews and results monitoring are regularly disclosed (proac- ti vely dissemi nat ed beyond im plementing agencies) _ J ·-- 17% __ _ 20% _j i 41 22% 16% 37 Overall Rating of GAC Responsiveness In the CAS Portfohowidefiduc1aryriskm1t1gation 15% 80% -~ 41 22% 72% 36 Enhanced selectivity of Bank country strategies and programs 46% 46% ! 41 44% 53% 36 I Improved signaling of GAC concerns and nsks t hrough Bank portfolio I processes I I 32% 51 % I l 41 25% 56% i 36 l~;;;~;~;:~;~ ~:::~:!~;;:~;;;!;~J~~~~::.~~- c~~::ries -------- _______ ;_~~-------- __ ~~: ________ ! ___ ___;_~_______________l______~~: ______U : _ _L________ ~! ___________ J co C/1 00 a- ~ Pre-GAC Post-GAC 0:: co lowCPIA' HighCPIA HighCPIA lowCPIA I :::J "' ! #of CAS/ n "" 0 c ~ To a Great Extent Some- what #ofCAS/ 1 Toa CPS reviewed Great Extent Some- ~ what #of CAS/ CPS reviewed Toa Great Extent I J Some- what #ofCAS/ CPS reviewed ! I I Toa Great Extent Some- j what CPS re- viewed -< ~ Please rate the extent to which the Bank's approach incorporates r ~ Explicit assessments of governance and political m :::J lD economy constraints 70% 30% 27 36% 57% 14 67% 33% 24 46% 46% 13 lD "' Ill Explicit choice of governance entry points 56% 44% 27 43% 57% 14 50% 50% 24 46% 54% 13 3 Ill Mix of financial and knowledge instruments 22% 78% 27 43% 50% 14 29% 63% 24 54% 31 % 13 ~ 0 Identification of results measures 23% 65% 26 7% 79% 14 25% 58% 24 23 % 69% 13 :::J Gl Please rate the extent to which the CAS includes country strengthening for ~ Core Public Sector Institutions 63% 37% 27 71 % 29% 14 63% 38% 24 69% 31 % 13 3 :::J "' Sectoral state institutions 31 % 65% 26 43 % 50% 14 42% 58% 24 46% 54% 13 () Ill Domestic Accountability institutions 30% 45% 20 9% 64% 11 40% 35% 20 27% 36% 11 :::J "' 0.. )> Civil Society and the Demand Side 18% 68% 22 8% 54% 13 14% 73% 22 0% 42% 12 ~ The Investment Climate 14 r;· 37% 52% 27 29% 64% 38% 50% 24 31 % 62% 13 Q 2 Overall Rating of signaling of GAC concerns and risks through Bank portfolio processes--Please rate the extent to which I ~ Portfolio Risks are regularly monitored by the Bank 41 % 48% 27 14 58% 42% 24 38% 38% 13 6" 50% I 36% :::J Portfolio processes track the progress of governance I reforms at sector and project levels 37% 56% 27 21 % i 57% 14 46% 46% 24 23% 46% 13 Portfolio reviews and results monitoring are regularly disclosed (proactively disseminated beyond imple- I menting agencies) 19% 19% 27 14% ! 21 % 14 25% 25% 24 15% 0% 13 l Overall Rating of GAC Responsiveness in the CAS Portfoliowide fiduciary risk mitigation 11 % 85% 27 21 % 71 % 14 17% 83% 24 33% 50% 12 Enhanced selectivity of Bank country strategies and programs 48% 44% 27 43% 50% 14 50% 46% 24 33% 67% 12 Improved signaling of GAC concerns and risks through Bank portfolio processes 33% 56% 27 29% 43% 14 29% 67% 24 17% 33% 12 Incorporated Smart design of programs and projects by countries 11 % 41 % 27 21 % 43% 14 22% 39% 23 17% 50% 12 Focused on country institutional strengthening I 11 % I 85% I 27 I 29% I 57% I 14 ! 42% I 58% I 24 i 58% I 25% ! 12 a. The evaluation team categorized countries with a CPIA governance cluster less than 3.5 as "Low CPIA" and those with a CPIA governance cluster of 3.5 and above as "High CPIA." IDA IBRD IDA IBRD To a # of CAS/ To a # of CAS/ # of CAS/ To a of CAS/ I I Toa Great Great Some- CPS Great I Some- CPS I Some- CPS Great I Some- I #CPS re- Extent what reviewed Extent what reviewed Extent what reviewed Extent what viewed Please rate the extent to which the Bank's approach incorporates Explicit assessments of governance and political economy constraints 65% 35% 23 50% 44% 18 62% 38% 21 56% 38% 16 Explicit choice of governance entry points 48% 52% 23 56% 44% 18 48% 52% 21 50% 50% 16 Mi x of financial and knowledge instruments 22% 78% 23 39% 56% 18 38% 57% 21 38% 44% 16 Identification of results measures 27% 68% 22 6% 72% 18 29% 62% 21 19% 63% 16 I Please rate the extent to which the CAS includes country strengthening for I Core Public Sector Institutions 70% 30% 23 61 % 39% 18 67% 33% 21 63% 38% 16 Sectoral state institutions 17% 83% 23 59% 29% 17 43% 57% 21 44% 56% ! 16 Domestic Accountability institutions 26% 53% 19 17% 50% 12 37% 47% 19 33% 17% 12 Civil Society and the Demand Side 11 % 63% 19 19% 63% 16 15% 65% 20 0% 57% 14 I I The Investment Climate 30% 65% 23 39% 44% 18 38% 52% 21 31 % 56% I 16 I I )> Overall Rating of signaling of GAC concerns and risks through Bank portfol io processes--Please rate the extent to wh ich ""0 ""0 Portfolio Risks are regula rly monitored by the Bank I 43% l 52% 23 I 44% r 33% l 18 l 62% I 38% 21 I 38% I 44% I 16 Ill ::J r Q_ ;;:· Portfolio processes track the progress of governance 0 reforms at sector and project levels I 30% I 70% I 23 I 33% I 39% i 18 I 48% ! 38% 21 I 25% I 56% I 16 l/) c Portfolio reviews and results monitoring are regularly 3 3 disclosed (proactively disseminated beyond imple- ; OJ -< ~ menting agencies} I 22% 22% 23 11 % 18 33% 19% 21 6% 13% ! 16 :::!:' I I I I 17% i I ! I J ::J '" Q_ Overall Rating of GAC Respons iveness in the CAS ::J D Portfoliowide fid ucia ry risk mitigation I 9% I 91 % I 23 I 22% I 67% I 18 I 14% ! 86% 21 I 33% I 53% i 15 "' 0 ::J Enhanced selectivity of Bank country strategies and Gl )> programs 39% 23 I 56% I 39% I 18 I 33% i 62% 21 I 60% I 40% i 15 n I I 52% I :A! Ill Improved signaling of GAC concerns and risks through ""0 "' 0 Bank portfolio processes I 30% I 65% I 23 I 33% I 33% i 18 I 33% I 57% 21 I 13% I 53% i 15 ::J ' ;::· "' Incorporated Smart design of programs and projects Ill ::J 18 47% 15 Ill by countries 13% 43% 23 17% 39% 20% 40% 20 20% "' "' ~ed on country institutional strengthening I 22% I 78% I 23 I 11 % I 72% 18 I 52% 48% 21 I 40% I 47% 15 ----------·-· - ():) '" CXl CXl ::E §:: ______ Pr~:G ~AC _ __ _ __ ___ _ Post-GAC l?i' CGAC/Window Others CGAC/Window Others ~ Toa #of Toa #ofCAS/ Toa #ofCAS/ Toa #of ~ Great Some- CAS/CPS Great Some- CPS Great Some- CPS Great Some- CAS/CPS § Extent I what reviewed Extent I what I reviewed Extent I what I reviewed Extent I what I reviewed Please rate the extent to which the Bank's approach ~ incorporates ~ ,,,,;c...re"'"'"" ,, .~'"·"~ '"' ,,m••, 1 1 I ~- r &ooomy""""'"" I 68% j 29% , 28 I 38% 62% 13 65% 35% 23 I SO% 43% 14 fil Explicit choice of governance entry points i 50% ! 50% .! 28 j 54% 46% 13 48% 52% 23 j 50% 50% 14 3 I i ! i ~ Mix of financial and knowledge instruments 1 25% i 75% 1 28 i 38% 54% 13 39% 52% 23 36% 50% 14 g Identification of results measures 19% !_ 70% I 27 I 15% 69% 13 30% 52% 23 • 14% 79% 14 g' Please rate the extent to which the CAS includes country strengthening for ~ Core Public Sector Institutions 71 % 29% T 28 54% 46% 13 74% 26% 23 50% 50% ! 14 ~ I ::J Sectoral state institutions 37% 59% 27 31 % 6 2% 13 48% 52% 23 36% 64% 14 ~ Domestic Accountability institutions 24% 52% 21 20% 50% 10 47% 42% 19 17% 25% 12 ~ CiviiSocietyandtheDemandSide 17% 57% 1 23 8% 75% 12 14% 71 % 21 0% 46% 1 13 ::J I g. ThelnvestmentCiimate 32% 54% 1 28 38% 62% 13 43% 48% 23 21 % 64% 14 0 2 Overall Rating of signaling of GAC concerns and risks through Bank portfolio processes~-Piease rate the extent to which Portfolio Risks are regularly monitored by the Bank l 39% 46% I 28 l 54% 38% 13 52% ~ 48% I 23 I 50% I 29% 14 Portfolio processes track the progress of governance I ! I i reforms at sector and project levels .~ 36% 50% ! 28 ! 23% 69% 13 35% 52% 23 43% 36% .~ 14 Portfolio reviews and results monitoring are regularly j disclosed (proactively disseminated beyond imple- I I I I l l '"'"""" '"'"' "' 1 18% 18% _I 28 1 15% 23% 13 22% j 22% 23 " " '" 14 Overall Rating of GAC Responsiveness in the CAS Portfoliowide fiduciary risk mitigation I 14% I 82% I 28 15% I 77% I 13 I 18% I 82% I 22 I 29% I 57% I 14 Enhanced selectivity of Bank country strategies and programs I 36% ! 54% [ 28 [ 69% I 31 % I 13 I 36% I 59% I 22 I 57% I 43% I 14 Improved signaling of GAC concerns and risks through Bank portfolio processes I 29% 57% I 28 ! 38% I 38% I 13 I 23% I 68% I 22 I 29% I 36% I 14 Incorporated Smart design of programs and projects by countries I 11 % 46% I 28 i 23% I 31 % I 13 I 23% I 50% l 22 I 15% I 31 % I 13 Focused on country institutional strengthening 18% 75% I 28 I 15% 77% 13 55% 45~~- -------~~---- 36% 50% 14 Pre-GAC Post-GAC Non Fragile Fragile Fragile' r Toa #ofCAS/ To a To a #ofCASl Toa #ofCAS/ Great Some- Extent I what CPS reviewed I Great Extent I : Some- I #ofCAS/ CPS I what reviewed Great Extent I Some- what CPS I reviewed Great Extent Some- what I CPS reviewed Please rate the extent to which the Bank's approach incorporates Explicit assessments of governance and political economy constraints 78% 22% 9 53% 44% 32 75% 25% 8 55% 41 % 29 Explicit choice of governance entry points 56% 44% 9 50% 50% 32 38% 63% 8 52% 48% 29 I Mix of financial and knowledge instruments 11 % 89% 9 34% 63% 32 13% 88% 8 45% 41 % 29 Identification of results measures 33% 56% I 9 13% 74% 31 63% 31% 62% 29 I 0% 8 Please rate the extent to which the CAS includes country strengthening for: Core Public Sector Institutions 67% 33% 9 66% 34% 32 50% 50% 8 69% 31% 29 i Sectoral state institutions 0% 100% 9 45% 48% 31 38% 63% 8 45% 55% 29 . Domestic Accountability institutions 14% 43% 7 25% 54% 24 33% 33% 6 36% 36% 25 1 Civil Society and the Demand Side 13% 63% 8 15% 27 13% 8 I 8% 62% 26 I 63% 63% 1 The Investment Climate 33 % 44% 9 !, 34% 59% l l 32 38% I 25% l 8 J 34% I 62% I 29 )> ~ Overall Rating of signaling of GAC concerns and risks through Ban.k portfolio processes--Please ra te the extent to which "0 "0 ! Portfolio Risks are regularly monitored by the Bank T 22% I 67% T 9 l 50% 38% 32 63% 38% 8 48% 41 % 29 !1) I ~ 0.. ;;:· I Portfolio processes track the progress of governance 1 0 reforms at sector and project levels 22% 67% 9 34% 53% 32 25% 50% 8 41 % 45% 29 (./) c Portfolio reviews and results monitoring are regularly 3 3 ! disclosed (proactively disseminated beyond imple- "' -< ~ menting agencies) 11 % 33% 9 19% 16% 32 25% 13% 8 21% 17% 29 -n I s· - 0.. ~ Overall Rating of GAC Responsiveness in the CAS __ ,i_ ~ I T lO 1 Portfoliowide fiduciary risk mitigation 0% 19% 32 0% 100% 8 29% 64% 28 1oo% 1 9 75% "' 0 ~ · Enhanced selectivity of Bank country strategies and Gl i )> programs 11 % 41 % 32 13% 75% 54% 46% n 9 56% 8 28 67% I ::l:l !1) Improved signaling of GAC concerns and risks 1 I "' '8 through Bank portfolio processes 33% 9 31 % 50% 32 25% 63% 8 25% 54% 28 ~ 56% I "' t6' Incorporated Smart design of programs and projects ~ !1) by countries 11% 22% 1 9 16% 47% 32 14% 29% 7 21% 46% 28 "' "' Focused on country institutional strengthening 0% 100% I 69% 32 38% 63% 8 50% 43% 28 9 22% a. Fragile states were those identified by IDA as fra gile and confl ict affected. http}/si teresou rces.worldbank.org/EXTLICUS/Resources/5 11 777-1269623894864/Fragile_Situations_List_FY11_ (Oct_19 _20 1O .pdf. 00 \0 \0 0 I I Pre-GAC I To a To a # of CAS/ To a To a # of CAS/ I I # of CAS/ Great Some- # of CPSCAS/ I Some- CPS Great Some- CPS Great I Some- I CPS I I Great ~ Extent what reviewed j Extent what j reviewed Extent what reviewed Extent I what I reviewed Q I I I a: Please rate the extent to which the Bank's approach incorporates Please rate the extent to which the Bank's approach o:J (lJ ::J Expl icit assessments of Expli cit cho ice of governance and political I governance entry "' n economy constra ints points 0 c ~ AFR 64% 36% 11 73% 27% 11 AFR 64% 36% 11 45% 55% 11 ~ EAP 75% 25% 4 50% 50% 4 EAP 50% I 50% 4 75% 25% 4 I 'rt> < I ~ ECA 63% 38% I 8 56% 44% 9 ECA 75% I 25% 8 56% 44% 9 m ::J LCR 50% 40% 10 38% 50% 8 LCR 40% 60% 10 38% 63% 8 lO (lJ lO MNA 60% 40% 5 100% 0% 3 MNA 40% 60% 5 0% 100% 3 rt> 3 rt> SAR 33% 67% i 3 50% 50% 2 SAR 0% 1 1oo% 3 100% 0% 2 ~ 0 Pre-GAC Post -GAC Pre-GAC Post-GAC ::J Gl To a # of CAS/ To a I # of CAS/ To a To a # of CA S/ 0 I I I # of CAS/ < Great CPS re- Great I Some- CPS re- Great Some- CPS re- Great Some- CPS re- rt> I Some- I Extent what viewed Extent what viewed Extent what viewed Extent I what I v iewed 3 (lJ ::J Please rate the extent to which the Bank's approach incorporates Please rate the extent to w hich the Bank's approach incorporates (") rt> (lJ ::J M ix of financial and knowl - I I Identification of I Q_ edge instruments I I I result s measures I )> I - - -1 I - ~ I ! i')' AF R 36% 64% 11 55% 45% I 11 AFR 27% 73% 11 18% 73% 11 I 0 ! I 2 EAP 50% I 50% 4 ! 50% 50% I 4 EAP 50% 50% 4 50% 50% 4 i I I ~ ' i I 6' ECA 38% i 63% 8 33% 56% i 9 ECA 13% 88% 8 22% 78% 9 I ::J l LCR 10% 80% 10 I 0% 63% I 8 LCR 0% 60% 10 0% 63% 8 I I MNA 20% 80% 5 l 33% 67% i 3 MNA I 0% 80% 5 33% 33% 3 I I 1 ' i I I 0% ! 2 SAR I 50% I 50% 2 100% 0% 2 Pre-GAC Post-GAC Pre-GAC Post-GAC To a To a To a To a # of CAS/ I I # of CAS/ I I # of CAS/ Great Some- CPS Great Some- CPS Great I Some- I # of CAS/ CPS Great Some- CPS Extent what reviewed Extent what reviewed Extent what reviewed Extent what reviewed Please rate the extent to which the CAS includes country strengthen ing for Please rat e the extent to which the CAS incl udes count ry strengthening for I I . - -- i Core publ ic sector lnstitu- Sectoral state in st i- i tions tutions I I AFR 55% 45% I 11 64% 36% 11 AFR 0% 100% 11 45% I 55% 11 i i i I EAP 100% 0% I 4 75% 25% 4 EAP 50% 50% 4 75% I 25% 4 ECA 63% 38% 8 56% 44% 9 ECA 50% 50% 8 44% 56% 9 LCR 80% 20% 10 88% 13% 8 LCR 33% 44% 9 25% I 75% 8 i MNA 20% 80% 5 33% 67% 3 MNA 40% 60% 5 33% I 67% 3 SAR 100% 0% 3 50% 50% 2 SAR 100% 0% 3 50% 50% 2 Pre-GAC Post-GAC i Pre-GAC Post-GAC I Toa ' #ofCAS/ Toa #of CAS/ # of CAS/ ! To a I I # of CAS/ To a I I CPS Great Some- CPS Great Some- I CPS ! Great Some- CPS Great Some- Extent what I reviewed I Extent what reviewed ! Extent what reviewed Extent what reviewed I I Please rate the extent to which the CAS includes country strengthening for ! Please rate the extent to which the CAS includes country strengthen ing for Domestic accountability I Civil society and the I institutions ! demand side I AFR 9% 64% I i 11 40% 20% 10 I i AFR 10% 50% I 10 9' 36% EAP 50% 50% i 4 50% 50% 4 I EAP 50% 50% I' 4 *' 25% 75% 4 ECA 40% 60% i 5 29% 43% 7 I ! ECA 20% 60% 5 Oo/o 71 % 7 : I LCR 0% 29% ' i 7 14% 43% 7 I ! LCR 11% 67% ! 9 Oo/o 71 % 7 M NA 0% 100% i 100% 0% 1 I i MNA 0% 75% I 4 0% 100% 3 SAR I 67% l 33% i 3 l 50% 50% 2 l SAR 0% 100% 1! 3 50% 50% 2 L----··········-··--------------------------------------------------------------------L---------------------------------- .................................... .: L ......:................................................................................................................................................................. Pre-GAC I Post-GAC To a To a #of CAS/ CAS/ I Great Great Some- CPS CPS I Some- I #of I Extent 1 what reviewed Extent what reviewed I Please rate the extent to which the CAS includes country strengthening for The Investment climate AF R 36% I ! 64% 11 I I 45% 45% 11 EAP 50% i 5o% I 4 i 5o% 50% 4 )> -o ECA 38% i 63% · 8 I 33% 67% 9 ~ LC R 10% 70% 10 o% 75% ::::l ! I 8 Cl. ;;:· MNA 40% ! 20% 5 67% 0% 3 0 VI c SAR - - - - - - - - - - ___________ t_ ________ - - - - - - - - l 67% i 33% 3 J 50% 50% 2 3 3 QJ 1 Pre-GAC Post-GAC Pre-GAC Post-GAC -< , I # of CAS/ I To a #of CAS/ To a # of CAS/ To a ! #of CAS/ To a Great 1 Some- CPS Great Some- CPS Great Some- CPS s· I Some- j CPS Great Cl. I Extent what j reviewed Extent what reviewed Extent what reviewed Extent what I reviewed ::::l I I I <0 V' Overall Rating of signaling of GAC concerns and risks through Bank portfolio processes Overall Rating of signaling of GAC concerns and risks through Bank portfolio processes 0 ::::l Portfolio processes ! I Gl )> track the progress of j I (} governance reforms , ! :r:J ro Portfolio risks are regularly at sector and project I j V' -o monitored by the Bank levels I 0 ::::l 64% 11 AFR 36% 64% 1 11 36% 36% 11 V' AFR 36% 11 55% 36% I I <" ro ::::l EAP 50% 50% 4 75% 25% 4 EAP 50% 50% 4 75% 25% ! 4 ro V' V' ECA 75% 13% 8 44% 56% 9 ECA 38% 50% 8 33% 67% ! 9 LCR 30% 40% 10 25% 50% 8 LCR 10% 60% 10 13% 50% ' i 8 M NA 40% 40% 5 100% Oo/o 3 MNA 40% 40% 5 I 67% 33% I 3 \0 SAR 33% 67% 3 50% 50% 2 SAR 33% 67% 3 50% 50% ! 2 \0 tv Pre-GAC Post-GAC # of CAS/CPS #of CAS/CPS To a Great Extent Somewhat reviewed Somewhat reviewed I I To a Great Extent ~ Overall Rating of signaling of GAC concerns and risks through Bank portfolio processes I 0:: OJ Portfolio reviews and OJ :l results monitoring are "' n regularly 0 disclosed (proactively c disseminated beyond § implementing agencies) "f AFR 9% 27% 11 27% I 18% 11 ~ & EAP 50% 50% 4 75% ! 0% 4 m :l lO ECA 25% 0% 8 11 % 33% 9 OJ lO LCR 0% 20% 10 (1) 0% 0% 8 3 (1) MNA 20% 0% 5 0% 0% 3 ~ 0 SAR 33% _l 33% 3 50% 50% 2 :l ---------------- ----· -------------- -------------------·----- Gl Pre-GAC Post-GAC Pre-GAC Post-GAC ~ i i I 3 OJ 1 #of CAS/ To a #of CAS/ To a #of CAS/ To a #of CAS/ :l To a Great I Some- CPS 1! Great Some- ~ CPS Great Some- CPS Great Some- CPS n I (1) Extent what , reviewed Extent what reviewed Extent what reviewed Extent what reviewed OJ I 1 :l Overall Rating of GAC Responsiveness in the CAS Overall Rating of GAC Responsiveness in the CAS a_ )> ~ I I I I Enhanced selectivity ·~ Portfoliowide fiduciary i of Bank country strat- r;· Q risk mitigation i I J 1 egies and programs ~ 2 AFR i 0% i 91 % 11 9% 82% i 11 AFR 27% i 64% 11 18% 73% 11 ~ a· EAP j 25% j 75% 4 50% 50% ! 4 EAP 50% ! 50% 4 50% 50% 4 :l I I t t ECA j 13% ! 88% 8 33% 67% ! 9 ECA 75% ! 25% 8 89% 11 % 9 LCR I 20% I 70% 10 13% 75% I 8 LCR 40% I so~. 10 38% 63% 8 MNA ! 20% j 80% 5 0% 100% ! 2 MNA 40% j 40% 5 0% 100% 2 • • t : I - ~~~-- - - - - - - - - - - - - - - ---__l----~~~---1---~~~-_j ______L _____L~o% ~-~~__j____~______ ~~------------------------ j____ ______} ______ - - - ~-~~-- - - - ~~~----J-------~-- - - - 1 - - - - ~?-~____ L__~}:n Pre-GAC Post-GAC Pre-GAC I Post-GAC Toa #ofCAS/ Toa #ofCAS/ #of CAS/ #ofCAS/ Toa I I I I I I I l! To a Great Some- CPS Great Some- CPS Great Some- CPS Great Some- CPS Extent what reviewed Extent what reviewed Extent what reviewed I Extent what reviewed Overall Rating of GAC Responsiveness in the CAS Overall Rating of GAC Responsiveness in the CAS I Improved signaling of Incorporated Smart I GAC concerns and risks design of programs through Bank portfolio and projects by I processes countries - I - J AFR 27% 64% 11 18% 55% 11 AFR 9% 55% 11 20% 50% 10 EAP 75% 25% 4 75% 25% 4 I EAP 50% 25% 4 SO% 25% 4 ECA 25% 63% 8 22% 67% 9 I ECA 0% 50% 8 22% 33% 9 I I LCR 20% 30% 10 13% 50% 8 LCR 10% 30% 10 0% I 50% 8 MNA 40% 60% 5 0% 100% I 2 MNA 20% 20% 5 0% 50% 2 SAR 33% 67% 3 50% 50% I 2 I SAR 33% 67% 3 50% 50% 2 ! I ' Pre-GAC Post-GAC I I # of CAS/CPS #of CAS/CPS To a Great Extent Somewhat reviewed To a Great Extent I Somewhat I reviewed Overall Rating of GAC Responsiveness in the CAS Focused on country insti- tutional strengthening _J AFR 18% i 82% 11 45% 45% 11 EAP - 1 I SO% ! ! 50% I 4 75% 25% 4 i ' ! I ECA i 0% l 88% 8 44% 56% 9 ! ' LCR 20% 60% 10 SO% 38% 8 MNA I 0% I 100% 5 0% 100% 2 I ' I l SAR 33% ! 67% 3 50% SO% 2 I Pre-GAC Post-GAC I # of Projects # of Projects To a Great Extent I Somewhat reviewed To a Great Extent I Somewhat I reviewed To what extent does the project design respond to Governance and PE constraints in the following ways? ~, )> I o Project design adapted to informal/actual reality 37% 55% ! 122 41 % SO% 78 il Strengthening management of relevant public agencies 40% -r 53% ' 122 47% 46% 78 ::::l i '' I I Q_ i ;;:· I Support for rule-based decision-making 0 l/) and accountability 31 % i SO% I I 121 37% 54% I 78 c I I 3 Proactive measures to include disadvantaged groups 34% 32% I 119 I 37% I 23% I 78 3 I .J.. Please rate the extent to which the Bank's approach incorporated smart design in the following areas: "' < , Quality of governance and political economy analysis ] 15% [ 61 % 115 25% 53% 76 s· I I 9: I I Quality of enhanced fiduciary aspects 22% I 62% 11 5 30% 62% 74 ::::l O i Demand-side of governance 31 % 44% 115 20% 50% 76 "' 0 I ::::l Use of country systems 23% 43% 111 32% 38% 69 C'l f) Quality of institutional strengthening 23% I 69% 116 23% 72% 75 ;J:J Ill Resu Its orientation 24% I 61 % 11 7 25% 60% 76 ::::l Overall Smart Design rating 40% i I 54% 122 45% 51 % 78 "' <'" Ill ::::l Ill "' "' '-0 w \() -1>- - ::::: Q Pre-GAC Post-GAC 0:: OJ LowCPIA HighCPIA LowCPIA HighCPIA ::J "' To a # of Proj- " n I i # of Toa I 0 To a Great I Some- I Projects # of Great Some - ! Proj ect s To a Great I Some- I # of Proj ects Great I Some- ects c Extent what reviewed Extent w hat i reviewed Extent w hat reviewed Extent what reviewed ~ ' -< To what extent does the project design respond to Governance and PE con stra ints in the following ways? r- Ill < Project design adapted to ~ m informal/ actual reality 40% 55% 82 30% 55% 40 I 39% I 50% 44 I 44% I 50% I 34 ::J I tO Strengthening manage- I I tO "' ment of relevant public Ill 3 agencies 44% 51 % 82 33% I 58% 40 I 48% I 43% 44 I 47% I 50% I 34 Ill ~ Support for rule-based 0 ::J decision-making and C\ accountability 38% 81 18% 40 30% 59% 44 47% 47% 34 0 I I 51 o/o I I I 50% ' I I i I I I < Ill 3 Proactive measures to ::J "' include disadvantaged (") Ill groups 38 I 18% 44 I 32% I 29% I 34 ::J l 36% 1 32% I 81 I 32% I 32% I 41 % "' Q_ Please rate the extent to which the Bank's approach incorporated smart design in the following areas: ; )> Quality of governance and I 19% 1 59% I 79 T 5% I 63% 38 I 19% I 57% ! 42 I 33% I 48% I 33 ~ r;· political economy analysis Q 2 Quality of enhanced fidu- ~ ciary aspects I 29% I 60% I 78 I 8% I 66% j 38 I 33% I 58% i 40 I 27% I 67% I 33 a· ::J Demand -side of gover- I nance 35% 48% 79 22% I 36% 36 17% 55% 42 24% 44% 34 ! Use of country systems I 27% I 44% I 78 I 12% i 42% 33 I 41 o/o I 30% 37 I 22% I 47% I 32 Quality of institutional strengtheni ng 28% 67% 79 14% 73% 37 27% 66% 41 18% 79% 34 Results orientation I 28% I 61 % I 79 I 16% I 61 % 38 I 24% I 64% 42 I 26% I 56% I 34 Pre-GAC Post-GAC IDA IBRD IDA IBRD # of # of To a Great Some- Projects # of To a Great I Some - I # of Projects To a Great Projects Toa Great Some- Projects I I I Some- I I Extent what reviewed Extent what reviewed Extent what reviewed Extent what reviewed To what extent does the project design respond to Governance and PE constraints in the following ways? Project design adapted to I I informal/ actual reality I 30% I 61 % 83 I 51 % I 41 % I 39 I 38% I 57% I 42 I 44% ! 42% I 36 Strengthening manage- ment of relevant public agencies I 39% i 58% 83 I 44% I 44% I 39 I 45% I 43% ! 42 I 50% I 50% I 36 Support for rule-based decision-making and accountability I 30% ! 50% I 82 I 33% I 51 % I 39 I 38% I 52% I 42 I 36% I 56% I 36 Proactive measures to include disadvantaged groups 31% i 40% 81 42% 16% 38 I 45% I 26% I 42 I 28% I 19% I 36 I ! t I I I Please rate the extent to wh ich the Bank's approach incorporated smart design in the follow- ing areas: Quality of governance and political economy analysis I 12% ! 62% ! 81 I 19% I 58% I 36 I 13% I 60% ! 40 I 40% I 46% I 35 )> ""0 Quality of enhanced fidu- ""0 !1> ciary aspects I 20% ! 62% i 79 I 27% I 62% I 37 I 33% I 54% I 39 I 26% I 71 % I 34 :J Q. ;;:· Demand -side of gover- n .. I VI nance 25% I 52% i 81 47% 26% 34 18% 60% 40 22% 39% 36 c 3 Use of country systems I 22% I 47% i 79 I 25 % I 34% I 32 I 30% I 41 % I 37 I 34% I 34% I 32 3 "' Quality of institutional -< ::I) 26% 71 % 35 :J strengthening ~ I 72% ! 81 1 29% 1 63% l 35 l L 20% 73% I 40 i 9-c 22% I 58% I 36 :J Resultsor~:~tion___ 17%__ j__!?~_ _j____~~----- ~~- 39% _ -------~----- ---~~- ~~--1---~~---- I <0 "' 0 :J C\ )> n ::0 !1> ""0 "' 0 :J :;::· "' !1> :J !1> "' "' \0 U1 \,() 0'> - ~ Q Pre-GAC Post-GAC 0:: I co OJ I CGAC/Window Others CGAC/Window Others ::J 7' ! #of n Toa I I #of Toa I I #of Proj- 0 I I To a Great I Some- I Projects Great Some- Projects To a Great I Some- I #of Projects Great Some- ects c 1 Extent what reviewed Extent what reviewed Extent what reviewed Extent what reviewed ~ -< To what extent does the project design respond to Governance and PE constraints in the following ways? r-- Ill T < Project design adapted to T Ill ! - informal/actual reality 40% 53% I I 83 31 % 39 I 41 % 48% ! 56 I 41 % i 55% I 22 m ' ::J lO Strengthening manage- OJ I I lO ment of relevant public Ill 3 Ill agencies I 37% I 55% I I 83 [ 46% 39 I 50% 46% I 56 I 41 % I 45% I 22 ::J ,... 0 ::J Support for rule-based decision-making and 1::I Cl ; 0 accountability 38% 44% I 82 18% I 64% I 39 I 32% 57% I 56 I SO% I 45% 22 < Ill 3 Proactive measures to OJ ::J include disadvantaged n ; Ill groups i 38% 31% 80 28% 33% 39 I 34% 27% I 56 I 45% i 14% I 22 OJ ,!, l I J l 1 ::J o._ Please rate the extent to which the Bank's approach incorporated smart design in the follow- )> ing areas: ~ ;=;· Quality of governance and Q political economy analysis i 18% I 59% i 80 I 8% I 65% I 37 I 23% j 58% I 53 I 32% I 41 % I 22 I : 2 ~ Quality of enhanced fidu- a· ciary aspects 31 % 58% 78 5% 71 % 38 37% 55% 51 14% 77% I 22 ::J I I I I I I Demand-side of gover- nance 34% I 43% I 79 25% 47% 36 20% 48% 54 18% 55% 22 '' Use of country systems 1 25% I 41 % I 75 I 17% I 47% I 36 I 35% 31 % 48 I 24% I 52% I 21 Quality of institutional 21 _ ;~~~~~~~!;_;~~~~~~-------------j_ _ ;_~_;_ _ _j_____;;.; ____ .l________ ; ; _______ _L___~_;_; ____l ___ ;;_ _ _ j_ _ _ _ _ ~;_ _ _ _ ;_ ;;;_j_ _ _ __ __l_ ____ __ ;;_j_ _ _ _ ~;_ _ __j_ _ _;;;_ ; __ ___l _j_ ____;;_ 22 Pre-GAC Post-GAC Fragile Non Fragile Fragile Non Fragile #of I 1 Toa #of #of To a #of Toa Great I Some- Projects ! Great Some- Projects Projects Great Some - Projects To a Great I Some- Extent what reviewed I Extent what reviewed Extent what reviewed Extent what reviewed I I I I I To what extent does the project design respond to Governance and PE constraints in the following ways? Project design adapted to informal/actual rea lit y 35% 52% I 23 37% 56% 99 50% 30% 10 40% 53% 68 Strengthening manage- ment of relevant public agencies 35% 61 % 23 41% 52% 99 50% 30% 10 47% 49% 68 I Support for rule-based i decision-making and accountability 14% SO% 22 35% 51 % 99 30% 40% 10 38% 56% 68 Proactive measures to include disadvantaged groups 27% 36% 1 22 36% 31 % 97 40% 40% 10 37% 21 % 68 Please rate the extent to which the Bank's approach incorporated smart design in the following areas: I T T 't T . Quality of governance and political economy analysis 14% 45% 22 15% 64% 95 33% 33% 9 24% 56% 66 > Quality of enhanced fidu - ciary aspects 18% 59% 22 23% 63% 94 14% 71 % 7 32% 61 % 66 5 :::l l. Demand-side of gover- · nance 13% 65% 23 36% 39% 92 22% 56% 9 19% 49% 67 1 Use of country systems 4% 43% 23 27% 43% 88 0% 50% 8 36% 36% 61 Quality of institutional ', strengthening 18% 68% 22 24% 69% 94 44% 44% 9 20% 76% 66 · Resu lts orientation 5% 82% 22 28% 56% 95 33% 44% 9 24% 63% 67 : :::l > :::l l ; J:) 'll > :::l > c:· 'll :::l 'll > > () ...., o.D CX> ~ Q 0:: Pre-GAC I Post-GAC co :::> "' Development Policy Opera- "' n Investment Lending tions Investment Lending Development Policy Operations 0 c To a Toa I # of Proj- ;::; I # of Toa # of I I # of I Great I Some- Projects Great I Some- I Projects To a Great i Some - Project s Great 1 Some- ects < Extent what reviewed Extent what reviewed Extent 1 what reviewed Extent , what reviewed r-- I I ro < To what extent does the project design respond to Governance and PE constraints in the following ways? ~ m :::> Project design ada pted to O informal/actual rea lity I 36% ' 56% I 104 I 47% I 47% I 17 I 41 % i 48% ! 58 I 42% 53% ! 19 O "' ro Strengthening manage- 3 ro ment of relevant public ;::; I 0 agencies I 38% ; 55% I 104 I 53% I 41% I 17 I 50% ! 43% i 58 I 42% i 58% i 19 :::> C\ Support for rule-based 0 < decision-making and ro 3 accountability I 26% i 52% ! 103 I 65% I 35% I 17 I 29% I 59% i 58 I 58% I 42% i 19 :::> "' (") Proactive measures to in- ro elude disadvantaged groups 1 35% 32% i 101 35% 35% l 17 34% 1 24% I 58 I 42% ! 21% I 19 :::> "' ! l I I J. Q_ )> Please rate the extent to which the Bank's approach incorporated smart design in the following areas: ;::; Quality of governance and ;=;· Q political economy analysis I 11 % i 61 % i 99 I 35% I 59% I 17 I 20% ! 52% I 56 I 44% i 56% ! 18 2 Quality of en hanced fid u- a· ciary aspects 19% 65% 99 44% 44% 16 28% 63% 54 39% 56% 18 :::> Demand-side of governance 32% 44% 98 25% 50% 16 25% 42% I 57 6% 72% 18 I I I Use of country systems 16% 45% 94 63% 31 % 16 22% 41% I 51 65% 29% 17 Quality of institutional strengthening 21 % 71 % 99 38% 56% 16 21 % 74% 57 29% 71 % 17 Results orientation I 20% 63% 99 I 47% I 47% I 17 I 23% 61 % 57 I 33% 61 % I 18 Pre-GAC Post-GAC Pre-GAC Post-GAC I I To a #of #of Strengthening man- To a #of To a Some- project Toa Great I Some- agement of relevant Great Some- Project design adapted to Great Projects project Great Some- #of project informal/actual reality Extent what reviewed Extent what reviewed public agencies Extent what reviewed Extent what reviewed AFR I 19% I 63% 27 28% 61 % 18 AFR 15% 78% 27 39% 44% 18 EAP 43% I 52% 23 43% 57% 7 EAP 43% 52% 23 71 % 29% 7 I I I ECA 53% I 47% 19 62% 31 % 13 ECA 53% I 42% 19 54% 46% 13 I LCR 45% 45% 22 53% I 42% 19 LCR 55% 32% 22 53% 42% 19 I I MNA 33% 58% 12 36% i 36% 11 MNA 33% 58% 12 45% 55% 11 I I I SAR ! 32% 63% 19 20% : 80% 10 SAR 47% ! 53% 19 30% 60% 10 Pre-GAC Post-GAC Pre-GAC Post-GAC Support for rule-based To a #of To a #~ Projects Proactive measures To a #of To a I decision-making and ac- Great Some- Projects Great Some- re- to include disadvan- Great Some- Projects Great Some- #of Projects countability Extent what reviewed Extent what viewed taged groups Extent what reviewed Extent what reviewed AFR 26% 41 % 27 39% SO% 18 AFR 38% 31 % 26 33% I 39% 18 EAP 36% 41 % I 22 57% 29% 7 EAP 32% SO% 22 57% I 14% i 7 ECA 26% 74% I 19 38% 62% 13 ECA 21 % 37% 19 46% 15% I 13 , I )> LCR 32% 55% 22 37% 63% 19 LCR 43% 14% 21 32% I 21 % I 19 "0 "0 Ill MNA 17% 42% 12 27% 36% 11 MNA 17% 25% 12 18% I 27% I 11 :::l a. SAR 47% 53% 19 30% 70% 10 SAR 47% 32% 19 SO% I 10% 10 x· 0 - - VI • Please rate the extent to which the Bank's approach incorporated smart design in the following areas: c I 3 I Pre-GAC Post-GAC Pre-GAC Post-GAC 3 "' -< Toa #of Toa Projects #~ Toa #of Toa I s· " Quality of governance and Great Some- Projects Great Some- re- Quality of enhanced Great Some- Projects Great Some- #of Projects a. political economy analysis Extent what reviewed Extent what viewed fiduciary aspects Extent what reviewed Extent 1 what reviewed :::l <0 V> AFR 4% 52% 27 19% 63% 16 AFR 11 % , 63% 27 38% ! 56% 16 0 :::l Gl EAP 9% 68% 22 14% 43% 7 EAP 10% I 62% 21 17% I 83% 6 )> n ECA 11 % 83% 18 31 % 69% 13 ECA 22% I 78% 18 46% ! 46% 13 :;lJ Ill LCR 30% 35% 20 I 42% 42% 19 I LCR 38% I 43% 21 21 % I 74% 19 :g I I ' :::l MNA 9% 73% 11 i 30% 20% 10 I MNA 10% I 90% 10 11 % ! 78% 9 V> <" Ill :::l SAR . 260jo 63% 19 -1 0% 80% 10 SAR 42% . 53% 19 40% l ~9%, ____ ___1_~ - Ill V> V> (Table continues on the following page.) "' "' 0 0 ~ro:t!I!IFJ '·'@fill¥ ~ 0:: co #of ::::l "' To a #of To a Projects To a #of To a n "' Demand-side of gover- Great Some- Projects Great Some- I re- Use of country Great Some- Projects Great Some- # of Projects 0 nance Extent what reviewed Extent what Iviewed systems Extent what reviewed Extent what reviewed c § AFR 30% 33% 27 19% 56% 16 AFR 11 % 41 % 27 40% 27% 15 "f fD EAP 27% 55% 22 14% 43% 7 EAP 29% 33% 21 0% 43% 7 & ECA 39% 39% 18 31 % 54% 13 ECA 27% 40% 15 45% 18% 11 m ::::l lO LCR 40% 35% 20 26% 37% 19 LCR 33% 33% 21 26% 58% 19 lO "' 11> MNA 9 9% 36% 11 MNA 0% 63% 8 25% 25% 8 3 11> SAR 19 10% 10 SAR 26% 63% 19 44% 44% 9 80% ~ 0 ::::l Pre-GAC Post-GAC Pre-GAC Post-GAC C'l To a #of To a #of To a #of To a #of ~ Quality of institutional Great Some- Projects Great Some- j Projects Great Some- Projects Great Some- Projects 3 strengthening Extent what reviewed Extent what reviewed Results orientation Extent what reviewed Extent what reviewed ::::l "' n 11> i AFR 8% 73% 26 13% 81 % 16 AFR 15% 62% l 26 I 31 % i 56% 16 ::::l "' Q_ I EAP 22% 74% 23 29% 57% 7 EAP 22% 65% I 23 57% I 43% 7 )> ~ i ECA 18% 82% 17 23% 77% ' 13 ECA 17% 72% i 18 I 15% ! 77% 13 () ' Q I LCR : 43% 48% 21 39% I 56% ; 18 : LCR 40% SO% I 20 I 32% I 47% : 19 I 2 ~ 6' ::::l 1 ; . . . . . .; . .1 l;'__ ~~·······1. : . . :: -:; l_; l_ ::_L~;·· · · · · · · · · · · · · · , ;J ' ; ... . .:;;_l. . . . :: . . .L J. :; 1. . . . ~--· · ·i To what extent does the project design respond to Governance and PE constraints in the ' Pre-GAC Post-GAC Pre-GAC ' Post-GAC To a #of To a #of Strengthening man- To a #of To a #of Project design adapted to Great Some- Projects Great Some- Projects agement of relevant Great Some- Projects Great Some- Projects informal/actual reality Extent what reviewed Extent what reviewed public agencies Extent what reviewed Extent what reviewed FPD 33% 58% 12 0% 70% 10 FPD 33% 58% 12 30% 60% 10 GOV/ EPOL 19% 76% I 21 40% 60% 20 GOV/EPOL 52% 38% 21 SO% SO% 20 HD 48% 48% I 27 41 % 47% 17 HD 41 % 59% 27 59% 41 % 17 NFRA 37% 44% I 27 64% 29% 14 NFRA 37% I 56% 27 64% 36% 14 SDV 37% 57% I 30 53% 42% 19 SDV 40% I 53% 30 32% 42% 19 ~ Pre-GAC Post-GAC Pre-GAC Post-GAC Support for rule-based To a #of To a #of Proactive measures To a #of To a #of decision-making and ac- Great Some- Projects Great Some- Projects to include disadvan- Great Some- Projects Great Some- Projects countability Extent what reviewed Extent what reviewed taged groups Extent what reviewed Extent what reviewed FPD 25% 67% 12 11 % 67% 9 FPD 8% 17% 12 11 % 22% 9 GOV/ EPOL 43% 48% 21 60% 40% 20 GOV/ EPOL 10% 38% 21 25% 10% 20 I I HD 26% 48% 27 29% 65% 17 HD 48% 48% 27 47% I 18% 17 I I NFRA 30% 52% 27 21 % 64% 14 NFRA 40% 28% 25 21 % 43% 14 )> I SDV 30% 50% 30 47% 42% 19 SDV SO% 23% 30 63% 1 21 % 19 "0 i • "0 - 11) - ::J Please rate the extent to which the Bank's approach incorporated smart design in the following areas: a_ ;;;:· ! Pre-GAC Post-GAC Pre-GAC Post-GAC 0 l l/) c To a #of To a #of To a #of To a #of '•· 3 Quality of governance and Great Some- Projects Great Some- Projects Quality of enhanced Great Some- Projects Great Some- Projects 3 political economy analysis Extent what reviewed Extent what reviewed fiduciary aspects Extent what reviewed Extent what reviewed OJ -< FPD 23% 54% 13 13% SO% 8 FPD 29% 57% 14 13% 75% 8 ::!1 ::J a_ GOV/ EPOL 5% 74% 19 38% 52% 21 GOV/EPOL 21 % 68% 19 38% 62% 21 5' <0 HD 24% 59% 29 8% 54% 13 HD 26% 67% 27 27% 64% 11 V> I I 0 ::J NFRA 9% 65% 23 i 27% 53% 15 NFRA 13% 57% 23 27% 67% 15 Gl )> SDV 13% 53% 30 _L 28% 56% 18 SDV 27% 57% 30 33% 50% 18 ~· n ::rJ 11) V> (Table continues on the following page.) "0 0 ::J V> <'' 11) ::J 11) V> V> s 0 rv ~ Pre-GAC 0:: OJ To a #of To a #of To a #of To a :::J "' Demand-side of gover- Great Some- Projects Great Some- Projects Use of country Great Some- Projects Great Some- #of project nance Extent what reviewed Extent what reviewed systems Extent w hat reviewed Extent what reviewed "' n - 0 c FPD 8% 46% 13 25% 38% 8 FPD 9% 45% 11 14% 57% 7 § GOV/EPOL 17% 61 % 18 '-;: 5% 67% 21 GOV/EPOL 32% 47% 19 50% 35% 20 fD HD 41 % 38% 29 I 14% 43% 14 HD 20% 48% 25 25% 50% 12 & I I m NFRA 27% 45% 22 27% 33% 15 NFRA 26% 43% 23 23% 23% 13 :::J lO SDV 43% 37% 30 33% 56% 18 SDV 23% 33% 30 29% I 35% 17 lO "' ro - 3 Pre-GAC • Post-GAC Pre-GAC ro Post-GAC ~ I # of To a # of To a To a #of 0 :::J Quality of institutional I To a Great Some- Projects Great Some- Projects Great Some- Great Some- Projects Gl strengthening Extent what reviewed Extent what reviewed Results orientation Extent what Extent what reviewed ~ FPD 8% 83% 12 25% 75% 8 FPD 7% 79% 14 13% 63 % 8 3 I :::J GOV/EPOL I 32% 58% 19 30% 65% 20 GOV/EPOL 16% 58% 19 29% 67% 21 "' (") ro ! HD 34% 59% 29 14% 86% 14 HD 24% 72% 29 29% 50% 14 :::J "' ! a. NFRA 22% 65% 23 13% 80% 15 NFRA 22% 57% 23 7% 80% 15 )> ~ SDV . I 17% 80% 30 28% 61 % 18 SDV 41 % 41 % 29 39% 44% 18 n· i Q 2 ~ 6' :::J APPENDIX D Detailed Desk Review Findings for Country Programs and Projects CAS Questionnaire- lEG desk review findings Pre-GAC Post-GAC # of # of # of # of CAS/ CAS/ CAS/ CAS/ CPS re - CPS re- CPS re- CPS re- viewed % Yes viewed % Yes viewed % Yes viewed I % Yes 1. Are the follow ing governance constraints d iag- 5. Which of the following specific ent r y points are nosed in t he CAS? identified? - Prope rty Rights and Rule-based Governance 41 76% 37 78% Pu blic Financial Management (PFM) 41 37 92% 100% I Quality of Budgeta ry and Financial Management 41 95% 37 95% Administration and Civi l Service Reform (CSR) 41 68% 36 50% Efficiency of Reve nue Mobilization 41 83% 37 81% , Revenue Management 41 73% 36 64% Qua lity of Public Administration 41 95% 37 92% I Decent ralization 41 56% 36 64% Transparency, Accountability, and Corrupt ion in t he 41 88% 37 86% 1 Anti-corru pt ion 41 56% 36 53% Pu blic Sector I Sectora~ity-service delivery 41 88% I 37 97% Gen der Equality 41 59% 37 65% 6. Which of the following instruments are used to support core public sector management as an entry Equity of Publ ic Resou rce Use 41 73% 37 73% point? Building Human Resources 41 83% 37 92% 24% 19% Other 41 36 Social Protection and Labor 41 71 % 37 73% DPL 41 39% 36 28% Policies and Institutions for Environmental Sustain- 41 71% 37 73% Programmatic DPL 41 41 % 36 44% ability I j iL 41 I 80% 37 68% I Business Reg ulatory Environment 41 85% 37 97% Programmatic IL 41 24% 37 24% Political Stability 39 33% 36 28% I Risk Management (e.g. Guarantees and Hedging insur- 40 I 15% 37 3% 2. Is governance one of the pillars of the CAS? - 41 I 95% 36 86% 1 ance pools) 3. Based on the CAS- Does the country have the I ! TA 41 93% 37 92% following mechanisms to I I I support the GAC objectives ? I I 41 90% 37 95% PRSP or Other Defi ned Strategy 41 90% 37 89% rESW Trust fund support I 41 51 o/o 37 65% Policy, Law and/ or Reg ulation 41 63% 37 65% 7.1s strengthening domestic accountability inst itu- I 41 68% 37 68% tions identified as an entry point in the CAS? Organizations res ponsible for implementi ng gover- 41 49% 37 62% 1 Legislative institutions 28 39% 25 44% nance reform I External Aud it function 27 56% 25 60% 4. Is core public sector management identified as an 41 37 100% I 1 100% I entry point in the CAS? ! Judiciary 27 74% 25 40% I Ombudsman I 27 4% 25 0% ' 0 w 0 ~ ~ I Pre-GAC Post-GAC Pre-GAC Post-GAC £< a: co #of #of #of #of ::J "' CAS/ CAS/ CAS/CPS CAS/ CPS re- CPS re- re- CPSre- n "' 0 viewed %Yes viewed %Yes viewed %Yes viewed I %Yes c ~ Global initiatives (e.g. AML/CTF, STAR) 27 11% 25 24% 11.Which of the following instruments are used to -< support promoting civil society and strengthening r-- (]) Public Procurement (Offices in charge regulations and/ 27 41% 25 36% the demand side as an entry point? < ~ or appeals review bodies) m Other 28 29% 26 27% ::J <.0 Media 27 15% 25 32% DPL 28 11% 26 8% <.0 "' (]) Right to Information 28 29% 25 28% Programmatic DPL 28 11% 26 15% 3 (]) 8. Which of the following instruments are used to IL 28 54% 26 50% ~ support strengthening domestic accountability 0 institutions as an entry point? ::J Programmatic IL 28 11% 26 8% Gl - 0 DPL 27 19% 25 28% Risk Management (e.g. Guarantees and Hedging insur- 27 4% 26 0% < (]) 3 Programmatic DPL 27 30% 25 44% ance pools) ::J "' IL 27 74% 25 64% TA 28 61% 26 73% (]) " Programmatic IL 26 8% 25 4% ESW 28 64% 26 58% ::J "' a. )> Risk Management (e.g. Guarantees and Hedging insur- 27 0% 25 0% Trust fund support 27 22% 26 58% ~ r;· ance pools) 12. Is strengthening the investment climate identi- 41 100% 37 97% £< fied as an entry point in the CAS? 2 TA 28 68% 25 84% ~ 13. Which of the following specific entry points are o· ESW 28 79% 25 84% identified? ::J Trust fund support 26 46% 25 56% Private sector development (privatization, SOE I 41 ! 85% I 36 I 86% Other I 26 23% 25 32% restructuring, private participation in infrastructure, 9. Are promoting civil society and strengthening the 41 68% 37 70% I enterprise support, financial linkages, etc.) demand side identified as entry points in the CAS? Regulatory reform (licensing & permits, customs, labor, I 41 ! 78% I 36 I 86% 10. Which ofthe following specific entry points are identified? corporate taxation) i I Promoting Civil Society and Demand side includes 26 88% 24 83% I Corporate governance 41 44% 36 33% promoting: Land and real estate markets 41 39% 36 31% I Civil Society Organizations 28 93% 26 77% Extractive industries 41 34% 36 33% I Organized Private Sector 28 25% 26 23% 14. Which of the following instruments are used to support strengthening the investment climate? Professional Associations 28 14% 26 I 0% I Other 40 15% 36 22% Consultative mechanisms (tripartite council, Business- 26 23% 26 I 31% I ! 1 DPL 41 29% 36 22% government dialogues) ------- - - - - -- - -~'_ _ _ _ __! Programmatic DPL 41 32% 36 36% #of #of # of CAS/ CAS/ CAS/ CAS/ CPS re- CPS re- CPS re- CPS re- viewed %Yes viewed %Yes viewed %Yes viewed %Yes 41 61% IL Programmatic IL 41 10% 36 36 69% 3% Self-standing political economy analysis - - - Country Procurement Assessment Review rm 9 28 22% 68% 2 9 0% 78% Risk Management {e.g. Guarantees and Hedging insur- 41 17% 36 19% Country Financial Accountability Assessment I 30 60% 9 56% ance pools) 18. Does the CAS propose to use any of the follow- TA 41 80% 36 86% ing country systems in programs and projects? ~ ESW 41 90% 36 83% Financial management {on-budget, fiscal calendar, - 41 I 32% I 37 I 35% Trust fund support 41 29% 36 33% provision for O&M/recurrent at appropriate levels, ESW 41 90% 36 83% budget execution) )> ""0 ""0 Trust fund support 41 29% 36 33% Audit 41 17% 37 19% (!) :::J 0.. 1 S. Was the justification for GAC entry points the 41 93% 37 I 86% Procurement {NCB-National Competitive Bidding) 41 22% 37 19% ;;:· result of political economy and governance analysis? Q Civil service and personnel rules {e.g. mainstreaming 41 15% 37 19% 0 16.1s any governance and political economy analysis ~ planned in the current CAS strategy? of project management units) ~- iD Social assessments 40 48% 36 47% Local government, e.g. Intergovernmental transfers 41 I 20% I 37 I 27% 0.. 0 Poverty and Social Impact Assessment 41 68% 37 54% 19. Were the following country systems actually (!) used in programs and projects during implementa- "A "' ::x:J Public Expenditure Review 40 68% 37 68% tion? (!) < Country Economic Memorandum 41 56% 37 49% -- Financial management {on-budget, fiscal calendar, - I 16 I 75% I 10 I 50% iii" :.E 41 I 10% 36 17% provision for O&M/recurrent at appropriate levels, ::!1 Institutional Governance Review :::J 0.. Technical Assistance Activity 41 73% 37 62% budget execution) s· \0 Audit 13 54% 8 25% Self-standing political economy analysis {PEA) 41 2% 36 14% 1 "' 0' ~ Procurement {NCB-National Competitive Bidding) 19 47% 8 25% Country Procurement Assessment Review {CPAR) 41 66% 37 27% n 0 I c Country Financial Accountability Assessment {CFAA) 41 I 68% 36 19% Civil service and personnel rules {e.g. mainstreaming 12 50% 7 29% ;:;. - - ... - of project management units) -< 17. Was governance and political economy analysis "U implemented using any of the following instru- 0 Local government, e.g. Intergovernmental transfers 14 I 57% I 8 I 38% \0 ments? -- -.-. ... OJ - 20. Were GAC issues addressed through the follow- Social assessments 24 67% 9 78% 3 ing coordination "' OJ Poverty and Social Impact Assessment 29 90% 10 60% mechanisms? ~ :::J 0.. "U Public Expenditure Review 33 88% 14 86% --- Harmonized/ multidonor serial budget support - 41 41 % 36 42% .2. (!) Country Economic Memorandum 26 65% 10 60% Sectorwide approaches 41 39% 36 44% ~ Institutional Governance Review 10 40% 4 50% Shared analytic work 41 85% 36 67% "' Technical Assistance Activity 28 93% 9 78% Multi-donor trust fund 41 39% 36 50% 0 V1 0 a- ~ o Pre-GAC Post·GAC Pre-GAC Post·GAC ~ i ro #of #of #of ! #of g) CAS/ CAS/ CAS/ i CAS/ "' CPS re- CPS re· CPS re- i CPS re· % ~ viewed %Yes viewed %Yes viewed i %Yes viewed Yes c ' ~ Joint portfolio reviews 41 32% 36 33% 2S. Were GAC entry points relating to PSM men- 41 98% ! 37 100% -< tioned in the CAS results framework (for instance, ~ Information sharing 41 93% 36 83% PFM, CSR, revenue administration, etc.)? < ~ 21. Are the following risks cited in the CAS? m -···· 26. If yes, which of the following types of indicators ~ Political stability 41 83% 37 70% were identified for PSM-related entry points? Q) ~ Security 41 46% 37 35% Process 41 76% 37 68% 3 ~ State institutions and capacity 41 83% 37 86% Actionable 41 78% 37 86% § Fraud and Corruption 41 46% 37 59% Institutional Outcome 41 34% 37 35% g' Fiduciary risk management 41 68% 37 70% 27. Of CASs that identified indicators, what share ~ 22. Does the CAS include any of the following Risk collected data? :::l g) Management Measures? Process indicators 31 94% 1 25 40% n ' ~ Early warning system . . 41 10~ 37 14% Actionable indicators 32 94% I 32 31 o/o a. Jomt finanoal management review With donors 41 32 *' 37 27% Institutional Outcome indicators 14 79% i 13 38% ~ ' 2. Risk reviews 41 15% 37 14% None (CASs which didn't identify any indicators for at 34 97% ! 34 41 o/o n Q Governance filters 41 5% 37 5% least one sub-entry point under Core PSM) 2 ~ Anti-corruption action plans 41 27% 37 27% 1 39 ! 51 % 1 36 1 67% o· . . 0 0 28. Were GAC entry points relating to Domestic ::J Partial/full disengagement plan 40 15 *' 37 5 *' Accountability institutions mentioned in the CAS Lending Scenarios 41 51 o/o 37 32% resu!ts frar_n~work (for ins!~nc.e, Legislative, External Audit, Jud1c1ary, Globalln1t1at1ves, etc.)? 23. Were any of the following Risk Mitigation Mea- sures implemented? ! 29. If yes, which of the following types of indica· Early warning system i 39 5% 15 7% tors were identified for Domestic Accountability . . . : institutions-related entry points? Joint financial management evaluation management : 39 26% 15 33% review with donors ! Process indicators 41 27% 37 32% Riskreviews ' i 39 8% 15 O% Actionableindicators 41 41 % 37 38% ! Governancefilters i 39 3% 15 0% lnstitutionaiOutcomeindicators 41 10% 37 14% I Anti-corruption action plans I 39 23% 15 13% None (that is, indicators were not identified for an 41 49% 37 14% Partial/ full disengagement plan II 39 8% 15 Oo/o ~~ point mentioned in the CAS) Adoption of lending scenarios 39 26% 15 7% 24. Did any unanticipated events affect the GAC -~ 41 39% 37 22% aspects of the CAS? #of #of CAS/ CAS/ #of CAS/ I I #of CAS/ CPS re- CPS re- % CPS re- CPS re- viewed % Yes v iewed Yes viewed % Yes viewed I % Yes 30. Of CASs that identified indicators, what share 36. If yes, which of the following types of indicators collected data? were identified for Investment Climate related entry points Process indicators 11 55% 12 33% Process indicators 41 56% 37 57% Actionable indicators 17 65% 14 14% Actionable indicators 41 73% 37 73% Institutiona l Outco me indicato rs 4 50% 5 0% Institutional Outcome indicators 41 24% 37 22% 31 . Were GAC entry points relating to Civil society 38 47% 36 47% and Demand side mentioned in the CAS results None (that is, indicators were not identified for an 41 85% 37 92% framework (for instance, CSOs, organized private sect or, professional associations etc.)? entry point mentioned in the CAS) )> 37. Of CASs that ident ified indicators, what share "0 32. If yes, which of the following types of indica- "0 tors were identified for promoting civil society and collected data? (l) :::J demand-side related entry points Process indicators 23 91 % 21 38% 0.. ;;:· Process indicators 41 32% 37 27% Q 1 Actionable indicators 30 83% 27 44% I 0 j Actionable indicators 41 22% 37 19% Institutional Outcome indicators 10 70% 8 13% ~ ! Institutional Outcome indicators 41 5% 37 8% "' iil I 0.. 41 44% 37 46% 0 I None (that is, indicators we re not identified for an (l) ! V> ! entry point mentioned in the CAS) :l:) "' (l) ~ 33. Of CASs that identified indicators, what share < collected data? ro· ~ I 1 Process indicators 13 69% 10 40% ::!1 :::J 0.. ! Actionable indicators 9 44% 7 29% :::J ! lO V> I Institutiona l Outcome indicators 2 0% 3 33% §' 34. Of CASs that identified indicators, w hat share n 0 c i collected data? ;::; i Process indicators 13 69% 10 40% -< I -o ! Actiona ble indicators 9 44% 7 29% 0 I I lO 2 Oo/o 3 33% Q) i Institutional Outcome indicators 3 V> 35. Were GAC entry points relati ng to Investment 40 90% 37 92% :::J Climate mentioned in the CAS Results Matrix (for "' 0.. instance PSD, Regulatory reform, corporate gover- -o nance, extractive industry etc.)? .Q. (l) ~ V> 0 ~ 0 (X) - ~ Q a: co #of #of #of #of Ill ::J Project Project Projects Projects "' re- re- re- re- n 0 viewed %Yes viewed %Yes viewed %Yes viewed I %Yes c ~ 1. Does the project address: 5. Does the project address the following Invest- -< r-- Domestic Accountability Institutions 122 28% 78 23% ment Climate entry points 7 !1> < l l Private sector development (privatization, SOE restruc- I 122 78 I 47% I 44% I ~ Primary Education 122 19% 78 12% m l turing, private participation in infrastructure, financial ::J \0 Roads and Highways 122 16% 78 13% Ill linkages, etc.) \0 !1> 2. Does the project address the following Core Public 3 Sector Management entry points? I Regulatory reform {licensing & permits, customs, labor, I 118 I 19% I 77 I 29% !1> ~ Public Financial Management {PFM) 122 ! 37% 78 46% corporate taxation) 0 i ::J Corporate governance 121 7% 78 17% Administration and Civil Service Reform (CSR) 121 ! 18% 78 27% C\ 0 < Revenue Management 122 ' 20% 78 32% Land and real estate markets 121 12% 78 14% !1> ! 3 Ill Decentralization 121 ! 39% 78 27% Extractive industries 122 6% 77 8% ::J i n i 6. Is the project based on an analysis of formal 121 !1> Sectoral capacity-service delivery 122 i 84% 78 77% 83% 78 90% Ill I institutions (e.g. organizational structures, decision- I ::J 0.. Anti-corruption 119 ! 17% 78 27% making rules, staff skills and capacity, and reporting > and accountability arrangements)? ~ 3. Does the project address the following Domestic l n· Accountability entry points? ! 7. Does the analysis mention any of the following Q ' informal institutions: 2 Legislative institutions 121 i 7% 77 8% -g Relevant historical legacies 120 50% 77 o· ExternaiAuditfunction 121 ! 12% 78 27% 1 64% ::J Cultural practices, norms, or other traditions influenc- I 121 I 36% 77 29% Judiciary 121 ' i 3% 77 12% ing project Ombudsman 121 ! 2% 77 3% Informal relations {conflict, cooperation) among differ- I 120 I 32% I 77 I 35% Global initiatives {e.g. AMUCTF, STAR) 121 ! 3% 77 3% ent levels of government Public Procurement {Offices in charge regulations and/ 122 ! 11% 78 17% i Social, regional, ethnic, religious, or linguistic relations I 119 I 44% I 74 I 38% or appeals review bodies) i i {including inequality, conflict, cooperation, etc.) Media 121 i 16% 76 8% Electioneering and/or electoral cycles 117 15% 75 1 13% Right to Information 122 I 11 % 77 1 9% Rent-seeking I 120 I 26% I 76 38% 4. Does the project address the following demand- i I side entry points? i 8. Do the project documents mention the following stakeholders: Civil Society Organizations 122 l 55% 77 40% Project beneficiaries 121 I 89% 78 91% Organized Private Sector 121 ' i 18% 78 21% I Adversely affected persons 122 I 48% 77 53% Professional Associations 121 ! 10% 76 12% Policy-makers/decision makers 121 86% 77 87% Consultative mechanisms {tripartite council, Business- 1OS ! 12% 77 19% I Donors 122 90% 78 83% government dialogues) ___________________________________ j_________________ _l _________ ···-- -··----------·····' Staff of implementing agencies 122 I 89% 78 81 % #of #of #of #of Projects Projects Projects Projects re- re- re- re- viewed %Yes viewed %Yes viewed %Yes viewed %Yes Civil society organizations 121 64% 78 65% Arrangements for random post-audits 122 25% 78 22% Private sector 121 62% 78 68% Arrangements for on-site field verification 122 34% 78 29% 9. Did governance and political economy analysis Transparency for fiduciary aspects 122 46% 78 45% draw on the following ESW? Governance and anti-corruption action plan 122 7% 77 26% Social assessments 121 31 % 78 31 % Special arrangements for high value/high risk con- 121 7% 77 13% Poverty and Social Impact Assessment 121 26% 78 40% tracts Public Expenditure Review 121 23% 78 28% Arrangements for a communication plan 122 36% 76 I 1 41 % Country Economic Memorandum 121 15% 78 21 % I )> M&E of GAC Measures I 122 I 34% I 77 44% D Institutional Governance Review 121 12% 78 4% D 14. Were the following measures to minimize risk of ro ::J Technical Assistance activity 121 51 % 78 47% misuse of funds implemented: 0.. x· Self-standing political economy analysis (PEA) 120 1% 78 Arrangements for technical/ program audit (in addition I 80 35% 33 21 % 1% I I I Q 0 10. Does the project include the following financial to FM audits) !:a OJ design features? Arrangements for random post-audits 86 27% 32 16% ro 0.. Subprojects (e.g. challenge grants) 122 42% 78 31 % I I Arrangements for on-site field verification 80 40% 33 27% 0 ro Results-based financing 122 16% 78 5% "' Transparency for fiduciary aspects 85 51 % 34 44% :x:J "' Incremental O&M financing 121 36% 78 32% ro Governance and anti-corruption action plan 78 10% 33 18% < iii' i Budget support 120 22% 78 26% :E I Special arrangements for high value/high risk con- 77 6% 32 6% ::!'! Complementary DPL and IL (implemented in parallel) 122 28% 77 22% tracts ::J 0.. 5' Other 121 89% 78 91 % Arrangements for a communication plan 86 36% 30 20% lO "' 11. Was the project restructured during implementa- 112 29% 60 7% M&E of GAC Measures 88 40% 32 28% §' tion? () 0 1S.ls there a self-standing PIU or not? 122 61 % 78 62% 12. Why was the project restructured during imple- c ;::; mentation? 16. Does the project propose to use country sys- 74 55% 48 SO% -< l Fiduciary risks 29 7% 3 0% tems? C) 0 ' Political factors 27 15% 3 33% 17. Does the project propose to use the following lO ! OJ country systems? 3 ! Institutional factors 31 29% 3 67% On-budget 74 55% 49 67% "' OJ l Technical reasons 27 59% 3 33% ::J 0.. Fiscal calendar 74 47% 49 65% C) ( 13. Were the following measures to minimize risk of .2. misuse of funds included: Provision for O&M/recurrent at appropriate levels 74 47% 49 45% ro ~ ! Arrangements for technical/program audit (in addition 122 32% 78 23% "' to FM audits) - ·-----·------- ~ 0 \0 0 - ~ 0:: OJ #of #of # of (lJ I #of ::::l Projects Projects Projects Projects "' re- re- re- , re- n % Yes 0 v iewed % Yes viewed I viewed % Yes I viewed I o/o Yes c: § Existing intergovernmental transfers 74 45% 48 44% 20. Did the project support institutional strengthen- "F Budget execution 74 62% 49 63% ing in the following areas: '11> < I Cross-cutting: Budget systems (FM and procurement) 76 57% 37 I 68% ~ Civil se rvice and personnel ru les i 74 65% 48 77% m i ::::l Aud it 75 59% 49 71 % Cross-cutti ng: HR systems 65 35% 26 i 31 % lO (lJ lO NCB Proc urement (Note: The issue is w hether procure- 74 Specific ministries, agencies, depart ments, or organi- I 95 92% 44 I 89% 11> 68% 49 61 % 3 I mentis allowed with minimal changes to normal zations 11> ! ::::l ' I ~ Local governments cou nt ry procedures, or else whether su bstantial I 75 64% 29 1 62% 0 ::::l I changes are required by t he Ban k, through a side let- I Comm unity organizations 78 64% 28 1 61 % C\ 0 i < J ter or so mething else). Monitoring and Evaluation 84 81 % 40 I 80% 11> I 3 (lJ 18. Did the project use the following country sys- 21 . Does the project identify the need for the fol- ::::l lowing supplemental supervision requirements for n tems? 11> I fiduciary aspects of the project: (lJ ::::l On-budget I 45 80% 26 88% a. Resources allocated to supervise GAC measures 108 6% 64 i 17% I I )> Fiscal calendar 43 74% 24 88% ! ~ I Decentralized or outsourced supervision including I I 113 26% I 70 I 23% () ' Provision for O&M/ recurrent at appropriate levels 45 73% 20 55% I 0 I I unannounced site visits 2 Existing inte rgovernmental transfers 41 73% 22 64% I ~ i Integrated Procurement, FM and Technical Reviews 109 20% 68 16% 6' 1 Budget execution 45 89% 22 86% I ::::l Higher percentage of ex-post reviews for high risk I 109 1 14% I 67 I 10% Civil service and perso nnel ru les I 54 83% 32 ! 78% I projects NCB Procurement ! 51 80% 24 83% j Plan for enhanced supervision by the Bank/Develop- I 113 25% I 71 I 37% . ! Audrt I 49 88% 24 79% ment partners 19. Does the project propose to support institutional I Other ! 111 28% 67 I 31 % strengthening in the following areas: I 22. Were any of the following supplemental supervi- Cross-cutting: Budget systems (FM and procurement) I 121 4 1% 78 51 % sion requirements for fiduciary aspects used during Cross-cutting: HR systems 121 23% 77 23% project implementation: i Resources allocated to supervise GAC measures I 82 7% 35 17% Specific ministries, agencies, departments, or organi- 121 84% 78 81 % 1 zations I i Decentralized or outsourced supervi sion including I I 85 I 25% I 42 I 19% I I I Loca l gove rnments 120 46% 77 unannounced site visits 36% I Integrated Procu rement, FM and Technical Reviews I Community organizations 122 50% 77 80 ! 24% 36 14% 39% i Monitoring and Evaluation 122 65% 77 Higher percentage of ex-post reviews for high risk I 82 l 12% I 34 I 6% 75% 1 - - - - - - - - - - -- - projects I Plan for enhanced supervision by the Bank/Develop- I 86 I 24% I 38 I 32% ment partners Other I 80 29% I 38 I 32% "' - Pre-GAC Post-GAC l Pre-GAC Po: #of #of #of ! #of ::E Projects Projects Projects Projects Q I I a: re- re- I re- re- co viewed %Yes viewed o/o Yes ! viewed I o/o Yes viewed I o/o Yes :J "' ..-----· --------·-------·---------------------· 31. Of projects that identified indicators, what share 40. Of projects that identified indicators, what share "' n collected data? collected data? 0 c Process indicators 101 60% 61 41 % ~ Process indicators 44 57% 31 35% -< Actionable indicators 101 70% 61 51 % r- Actionable indicators 44 59% 31 29% Ill Institutional Outcome indicators 101 31 % 61 20% < Institutional Outcome indicators 44 18% 31 26% ~ 32. Were GAC entry points relating to Domestic 118 26% 77 23% m Accountability institutions mentioned in the project :J 41. Based upon information and data provided to <0 results framework (for instance, Legislative, External Audit, Judiciary, Global Initiatives, Media etc.)? you-Please rate the extent to which risks of misuse <0 "' Ill of funds (graft, fraud, corruption, leakage) are iden- 3 33. If yes, which of the following types of indica- tified in the PAD? Ill tors were identified for Domestic Accountability :J ,... institutions-related entry points? .. To a great extent 26 21 % 28 36% 0 :J Process indicators 31 48% 18 61 % G'l Somewhat 65 54% I 38 49% I 0 Actionable indicators 31 48% 18 56% < Ill Not at all 30 25% I 11 14% Institutional Outcome indicators 31 32% 18 17% 3 42. What is the project's risk rating for Financial :J 34. Of projects that identified indicators, what share "' () collected data 7 Management? Ill :J "' 0. Process indicators 31 48% 18 33% High -- 20 20% 16% 1 21% )> Actionable indicators 31 39% 18 33% 17 17% 16% 23% Substantial I ~ Institutional Outcome indicators 31 29% 18 6% ;::;· Moderate 5 50% I 27 Q 35. Were GAC entry points relating to Civil society 118 56% 77 34% I 38% 2 and Demandside mentioned in the project results Low 15 15% I 4 6% ~ framework (for instance, CSOs, organized private i5" sector, professional associations etc.)? : 43. What is the project's risk rating for Procurement? :J tors were identified for promoting civil society and demand-side related entry points High 46 42% 28 I 39% Process indicators 65 60% 27 52% Substantial 16 15% 15 21 % Actionable indicators 65 55% 27 59% Moderate 33 30% 19 1 27% Institutional Outcome indicators I 65 25% I 27 26% Low 14 13% 9 13% 37. Of projects that identified indicators, what share 44. What is the project's risk rating for Overall Risks? collected data? High 17 18% 14 19% Process indicators 65 48% 27 30% Substantial 24 25% 27 37% Actionable indicators 65 43% 27 41 o/o I Moderate so 52% 27 37% Institutional Outcome indicators 65 20% 27 22% Low I 6 6% I 5 7% 38. Were GAC entry points relating to Investment l 116 36% 77 43% Climate mentioned in the project Results Matrix 39.1f yes, which of the following types of indicators were identified for Investment Climate related entry points Process indicators 44 64% 31 I 74% Actionable indicators I 44 70% 31 58% APPENDIX E Econometric Analysis This appendix uses data collected in the desk review (refer to Appendix A) and from World Bank databases to answer three overarching questions: • What factors were associated with the aid selectivity of World Bank lending commitments? • What factors were associated with the responsiveness of World Bank country programs and projects to various GAC elements? • What factors were associated with the number of risk review measures in projects? Two techniques were used to analyze the data. First, frequencies and cross-tabulations were used to assess correlations between ratings of GAC responsiveness and key variables. Pearson's Chi-2 significance tests were conducted for cross- tabulations and are reported in the main text. Frequencies for all ratings and data collected in the desk review are in appendixes C and D. Second, OLS, Logit, and appropriate multivariate regressions were used to analyze factors related to GAC responsiveness and aid effectiveness. All models report marginal effects of regression outcomes. Evaluating Aid Selectivity The basic aid-architecture estimation takes the following format: (1) ln (At =f30 +f3 1 (CPIA)i,t-l +f3 2 ln(GNI)i,t- J + f3 3 1n(Population)i,t- l +f3 4 (X)+ 111 + E;,r where A is the specific aid allocation, measured as the natural log of the annual commitments or disbursements to country i in year tin 2008 constant dollars, CPIA is the governance-duster CPIA score, GNI (Atlas method) in 2008 constant dollars, and Population is the total population, x is a vector of additional controls, fl represents cross-country invariant time-fixed effects, and e is a random disturbance. Among the controls used: indicators of human rights pro- tection and a measure of democracy in alternative estimations. In subsequent estimations, a binary variable coded 1 if the country received a development policy loan (DPL), 0 otherwise, is added. All dependent variables are lagged once to limit the risk that allocations are influencing any of these variables in the same year. The period under investigation is 2004-2010 for all estimations. All constant-dollar figures are converted from current dollars using the US CPl. Evaluating GAC Responsiveness GAC Responsiveness models assess how five sets of variables are associated with GAC Responsiveness. ¥; =a; +f3 1Postgac; +f3/'f; +f3 36; +f3 38; + 11; (2) In which: Postgac = A dummy variable for whether or not the project is in the Post-GAC period; y = Upstream design variables; 8 = Financial design variables; € = Project Entry points; and [) = Project-specific control variables 113 In addressing question one, tables E.S-, E.14 and E.17 explore what factors are associated with different measures of GAC re- sponsiveness (Ratings for selectivity, smart design of project, risk management intensity, use of country systems, institutional strengthening, and political economy factors in country programs and projects). All hypotheses associated with these vari- ables are two-tailed. Appendix E, table E.18 provides explanations for the operationalization of independent variables used in the multivariate analyses. One control variable of particular interest is the association between CGAC/Window countries and GAC responsiveness. Each table explores this relationship using a control variable and interaction term to see if CGAC/ Window countries are associated with more GAC responsiveness in the post-GAC period. Models that include the CGAC/ Window*Post-GAC interaction term report marginal effects of OLS regressions for ease of interpreting the interaction. Risk Management Intensity Risk management measures are assessed by looking at the count of risk management measures designed and imple- mented in projects-a variable called the risk management intensity. A list of risk management measures counted in operationalizing the risk management intensity are available in Table E.15. The risk management measures were selected in accordance with the QAG Benchmarking Survey and the DPL Readiness Framework. Counted measures included links to specified ESW, mitigation measures, supplemental supervision measures, disclosure measures, and grievance mechanisms. Tables E.l6a and E.l6b assess the risk management intensity of project design and implementation. Risk management intensity can further be seen as a measure of GAC responsiveness-the degree to which projects incorporate and imple- ment risk management measures. Table E.16a displays four models using Poisson and Negative binomial repressions to assess factors associated with risk management intensity. Because there is significant evidence of overdispersion in Model ( 1) assessing factors associated with risk management intensity in project design (= 38.16, p < 0.01 ), a negative binomial regression model (Model 2) is preferred to the Poisson regression model. Similarly, in assessing factors associated with the risk management intensity of project implementation (Model3), there is significant evidence of overdispersion (= 7.10, p < 0.01) in the Poisson model. Hence, a negative binomial regression model (Model4 below) is preferred. Further in assessing factors associated with risk management intensity of project design (Models 1 and 2 of Table E16a), the negative binomial regression model improves upon the under prediction of zeros in the Poisson regression model by in- creasing the conditional variance without changing the conditional mean. To further highlight the significance of variables such as the DPL, the zero- inflated count regression model is presented in table E.16b. A Vuong test of the Zero-inflated negative binomial regression versus the standard negative binomial regression is significant (z = 1.32 Pr>z = 0.0927). The significant positive value supports the use of the Zero-inflated negative binomial over the negative binomial regression. Table E.16b shows estimates of a Zero- Inflated Negative Binomial regression model responding to the failure of the Poisson regression model to account for dispersion and excess zeroes by changing the mean structure to allow zeros to be generated by two distinct processes. To make this clearer, the Poisson model and Negative binomial regression mod- els assume that every project has a positive probability of including at least one risk management measure. Substantive- ly, this may be unrealistic because some projects-such as DPLs that are based on prior actions-may not necessarily have the same kinds of risk management measures as ILs. The zero-inflated count model allows for this possibility, and in the process, it increases the conditional variance and the probability of zero counts. Let A = 1 if a project has a risk management measure, else A = 0. Whether or not a project includes at least one risk management measure is a binary outcome that can be modeled using a Logit or a Probit: (3 ) \{); =Pr(A;=Ii z;)= F(zti) whe re 3.5 1 1 - 0.242 0.100 0.194 - 0.130 (0. 191) (0.180) (0.135) (0.109) DPL 1 - 0.140 0.551 * - 0.629** - 0.995*** 1 (0.494) (0.320) (0.242) (0.268) IL - 0.196 - 0.302 - 0.0284 0.0800 I (0.241) (0.237) (0.158) (0.150) Trust Fund 0. 167 - 0.164 - 0.281 * - 0.272** I (0.210) (0.231) (0.153) (0.113) TA/ESW - 0.0189 - 0.0388 0.252** 0.216* (0.132) (0.142) (0.113) (0.113) Post-Conflict State - 0.00285 - 0.01 93 - 0.329** - 0.139 (0.192) (0.202) (0.125) (0.160) EAP - 0.147 0.0948 - 0.126 - 0.01 60 (0.272) (0.270) (0.2 14) (0.236) ECA - 0.138 - 0.0895 - 0.220 - 0.120 (0.201 ) (0.203) (0.156) (0.142) LCR I - 0.162 0.131 - 0.584*** - 0.285** (0.172) (0.176) (0.130) (0.111) MNA - 0.0245 0.283 - 0.130 - 0.0497 (0.225) (0.235) (0.180) (0.169) SAR - 0.4 12 0.2 10 - 0.0163 0.331 (0.313) (0.343) (0.311) (0.3 11 ) CGAC/Wind ow Country 0.231 - 0.0550 - 0.0321 0.0306 (0.194) (0.193) (0.152) (0.127) CGAC/Wind ow Country * Post - GAC - 0.159 0.03 15 0.250 0.129 (0.252) (0.249) (0.181) (0.165) Constant 0.937 0.241 1.120*** 1.1 53*** (0.6 18) (0.452) (0.288) (0.395) Observations 78 78 78 77 R' 0.155 0.116 0.378 0.365 Note: Models report OLSa with robus t standard errors in parentheses*** p<0.01, ** p 3.5 - 0.106 - 0.313* - 0.0252 (0.153) (0.160) (0.0295) CGAC/Window Country 0.0605 - 0.0735 -0.00674 (0.161) (0.205) (0.0170) Post-GAC - 0.159 - 0.0741 - 0.0759 (0.199) (0.203) (0.0771) Post-Conflict 0.0294 0.0187 0.0161 (0.120) (0.133) (0.0218) EAP 0.137 0.214 - 0.0232 (0.123) (0.130) (0.0266) ECA - 0.254 - 0.224 - 0.0120 I I (0.160) (0.171) (0.0209) LCR - 0.193 0.0647 - 0.0628 (0.150) (0.153) (0.0634) IMNA - 0.471** (0.232) - 0.0460 (0.204) - 0.00272 (0.00945) , SAR 0.0649 0.179 - 0.0361 I (0.139) (0.131) (0.0399) I CGAC/Window Country * Post-GAC 0.255 0.159 0.0758 (0.223) (0.229) (0.0774) Constant 0.714** 0.441 1.059*** (0.325) (0.300) (0.0656) Observations 78 78 78 R' 0.312 0.299 0.090 ------------- - Note: Models report OLS with robust standard errors in parentheses*** p 3.5 0.298* 0.279* 0.107 0.248* 0.000995 0.0271 (0.1S8) (0.149) (0.128) (0.133) (0.112) (0.130) I Post-GAC - 0.114 -0.117 -0.0430 - 0.159 - 0.0459 0.0659 (0.187) (0.184) (0.118) (0.135) (0.111) (0.141) DPL in Portfolio I - 0.00913 - 0.0602 0.317 0.552** 0.844*** - 0.280 (0.294) (0.424) (0.195) (0.259) (0.254) (0.232) IL in Portfolio 0.331 0.305 0.0298 0.124 0.0855 -0.00841 (0.257) (0.255) (0.148) (0.149) (0.155) (0.162) Trust Fund 0.0211 - 0.0256 - 0.0414 - 0 .191* - 0.0211 0.113 (0.199) (0.165) (0.137) (0.101) (0.123) (0.190) I I TA/ESW -0.109 - 0.0744 - 0.0355 - 0.0134 - 0. 119* -0.0887 (0.115) (0.119) (0.104) (0.0858) (0.0703) (0.103) Post-Conflict Country - 0.151 - 0.163 - 0.333*** -0.138 - 0 .304** - 0.390*** ~ (0.166) (0.151) (0.120) (0.146) (0.137) (0.119) EAP 0.365 -0.148 -0.124 0.171 0.426* 0.250 (0.250) (0.193) (0.182) (0.252) (0.237) (0.196) ECA - 0.197 - 0.266* - 0.329** - 0.0563 - 0.276** - 0.325** I I (0. 164) (0.157) (0.142) (0.156) (0.133) (0.133) LCR -0.0581 - 0.120 - 0 .329** - 0.330*** - 0.306** -0.158 (0.164) (0.153) (0.130) (0.11 9) (0.11 6) (0.150) MNA 0.139 0.153 -0.0146 0.0553 0.0621 0.0373 (0.250) (0.245) (0.173) (0.167) (0.180) (0.186) I SAR 0.479* 0.461 0.213 0.595** - 0.107 0.0504 (0.274) (0.280) (0.282) (0.253) (0.228) (0.278) CGAC/Window Country - 0.0310 - 0.0645 0.000826 -0.102 -0.123 - 0.00941 I (0.179) (0.169) (0.142) (0.118) (0.116) (0.134) CGAC/Window Country * 0.168 0.249 0.104 0.168 0.114 0.0351 Post-GAC (0.227) (0.221) (0.162) (0.173) (0.152) (0.183) Constant 0.0519 0.135 0.0298 - 0.368 - 0.479 0.642* i (0.437) (0.507) (0.282) (0.373) (0.360) (0.323) Observations ' 78 78 78 78 78 78 R' 0.282 0.273 0.296 0.371 0.354 0.324 Note: OLS with Robust standard errors in parentheses ••• p_n 69% 195 0 c ~ (1) ! (2) (3) (4) (5) (6) (7) (8} (9} (10} ! (11} (12} -< I Budget Budgetsys- HR sys- Specific agen- Specific agen- ! ~ systems tems tems HRsystems cles cles Local Govt I Local Govt CSOs CSOs M&E M&E I : ri5 Post-GAC 0.0379 0.0380 0.0706 0.148 0.0496 - 0-234** - 0.125 - 0.0465 - 0.0279 0.0944 0.161** 0.0749 m (0.0945} (0.134} (0.0621} (0.118) (0.0339} (0.110} (0.0833) (0.128) (0.0956} (0.117) (0.0676} (0.119) :::l 0.323*** 0 .230** lO ESW 0.0891 0.0681 0.0846 0.0723 0.0246 0.00551 - 0.167 - 0.110 0.220** 0.180** (0.112} (0.0895} (0.0712} (0.0696} (0.0542} (0.0662} (0.1 17) (0.0874) (0.125) (0.0742) (0.103} (0.0857} lO "' Analysis of Formallnstitu- 0.176 0.142 0.0845 0.0725 0.00459 0.110 0.103 0.0563 0.0741 - 0.00741 0.122 0.110 <1> tions 3 (0.203} (0.144) (0.105} (0.113} (0.0623) (0.118) (0.142} (0.108) (0.162) (0.118) (0.137} (0.125) <1> ~ Analysis of Informal - 0.109 - 0.0960 0.0382 - 0.00211 0.0491 0.0478 0.169 0.118 - 0.00104 0.000524 - 0.0680 - 0.0492 Institutions 0 :::l (0.153) (0.101 ) (0.101} (0.0855) (0.0634} (0.0793) (0.108) (0.0944} (0.152} (0.102) (0.0892) (0.0977} Gl Includes Beneficiary - 0.0411 - 0.0231 .0.295*** -0.182** - 0.0431 - 0.0897 0.202** 0.147* 0.468*** 0.361*** 0.285*** 0.256*** Feedback ~ <1> (0.102) (0.0698) (0.112} (0.0793} (0.0311} (0.0609) (0.102} (0.0868} (0.0832} (0.0751} (0.103) (0.0857} Includes Disclosure Provi- - 0.0994 - 0.0602 0.0370 0.0373 0.118** 0.161*** 0.167* 0.155* 0.153 0.119 0.0151 0.0135 3 sions :::l "' (0.101} (0.0681} (0.0690) (0.0662) (0.0483} (0.0610) (0.0911} (0.0824} (0.0999} (0.0733} (0.0813) (0.0742) n Has a PIU 0.127 0.0918 - 0.153** - 0.123* 0.0292 0.0232 <1> - 0.142 - 0.106 - 0.192* - 0.117 - 0.0190 - 0.0189 (0.0917) (0.0739} (0.0755} (0.0668} (0.0436} (0.0626) (0.103) (0.0787) (0.112) (0.0752} (0.0837) (0.0742) :::l "' Uses Country Systems 0.172* 0.132* 0.0364 0.0342 0.137* 0.168** 0.0554 0.0298 - 0.0712 - 0.0538 0.240** 0.202*** 0.. (0.0939} (0.0743} (0.0717) (0.0657} (0.0789} (0.0654) (0. 103} (0.0812} (0.0969) (0.0729} (0.0954} (0.0750} )> J DPL 0.169 0.106 0.0189 0.0416 - 0.199 - 0.176** - 0.0960 - 0.0774 -0.374*** -0.268** - 0.166 - 0.111 :::l rl ! (0.152} (0.116) (0.119} (0.115} (0.1 61) (0.0891} (0.124) (0.118} (0.0848} (0.107} (0.163} (0.117) n· ! Trust Fund - 0.124 - 0.0880 - 0.175** - 0.164** - 0.0458 -0.0825 0.00173 - 0.00338 0.157 0.0948 - 0.0506 - 0.0382 Q I (0.101 ) (0.0756) (0.0732} (0.0664} (0.0396) (0.0530) (0.0868} (0.0708} (0.0968) (0.0676) (0.0880} (0.0704) 2 ] Risk Exposure 0.0284 0.0322 - 0.0565 - 0.0525 - 0.0148 - 0.00769 0.0520 0.0443 0.0594 0.0471 0.0236 0.0186 I (0.0687) (0.0555) (0.0456) (0.0492} (0.0295} (0.0383) (0.0635} (0.0512} (0.0705) (0.0447) (0.0645} (0.0486) ~ (5" J CPIA (continuou s} - 0.101 - 0.0719 - 0.0734 - 0.0703 0.0550 0.115* 0.00809 - 0.0109 - 0.0620 - 0.0449 - 0.0914 - 0.0640 :::l (0.110) (0.0804} (0.0874) (0.0791} (0.0352) (0.0591} (0.103} (0.0790) (0.122) (0.0762) (0.0941) (0.0840) J FPD - 0.316*** - 0.231* 0 - 0.312*** 0.0706*** 0.175** - 0.344*** - 0.327*** - 0.387*** - 0.389*** - 0.0954 - 0.0869 (0.119) (0.122} (0} (0.0794} (0.0266} (0.0822} (0.0847} (0.120} (0.0744} (0.114} (0.159} (0.129} GovEPOL 0.498*** 0.397*** 0.176 0.179* 0.0525* 0.128 .0-230** - 0.211** - 0.338*** - 0.295*** - 0.0636 - 0.0540 (0.110} (0.104} (0.124} (0.103} (0.0288} (0.0850} (0.0941} (0.102} (0.0874} (0.0941} (0.125} (0.112} HD 0.0571 0.0331 0.0141 0.0152 0.0857*** 0.150** 0.0723 0.0407 - 0.0502 - 0.0518 0.181** 0.148* (0.125} (0.0957} (0.0905} (0.0885) (0.0295} (0.0709} (0.120} (0.0952} (0.113} (0.0916} (0.0735} (0.0783} INFRA 0.0339 0.00757 - 0.163** - 0.153* - 0.0139 - 0.0367 - 0.0666 - 0.0586 - 0.0174 - 0.0355 - 0.0401 - 0.0480 (0.134} (0.0989} (0.0773} (0.0888} (0.0392} (0.0757} (0.119} (0.102} (0. 125} (0.0924} (0.1 03} (0.0874} ECA 0.213 0.178 0.443*** 0-256** 0.0492 0.102 0.110 0.0811 - 0.208 - 0.0926 0.0908 0.0589 (0.142) (0.116) (0.166} (0.114} (0.0348) (0.0837} (0.151} (0.114} (0.141 } (0. 122) (0.101} (0.101} EAP -0.281** .0.199* - 0.00208 - 0.0139 0.0331 0.123 - 0.0166 - 0.0387 - 0.245** - 0.194 0.0918 0.0795 (0.125} (0.117} (0.131} (0.102) (0.0404) (0.103) (0.146} (0.124) (0.124} (0.128) (0.105) (0.109) MNA - 0.204 - 0.134 0.0570 0.0422 0.0405 0.0869 - 0.1 90 - 0.153 - 0.247** - 0.173 0.0594 0.0537 (0.136} (0.112) (0.123} (0.0984} (0.0283) (0.0948) (0.149) (0.113) (0.108} (0. 105) (0.0986) (0.109) j LCR 0.0139 0.00737 0.0399 0.00418 - 0.0325 - 0.0832 0.111 0.0546 - 0.233** -0.156* - 0.0291 - 0.0271 I (0.1 41 ) (0.103) (0.121) (0.0829) (0.0617} (0.0917) (0.139} (0.112) (0.113} (0.0928} (0.119} (0.108} l SAR -0.331*** -0-262** 0.0189 - 0.00147 0.0696** 0.121 - 0.0444 - 0.0540 - 0.00472 - 0.0200 - 0.181 - 0.114 (0.105} (0.121} (0.127) (0.117) (0.0314} (0.0794) (0.133} (0.119) (0. 177) (0.118) (0.178) (0.128) j Post-Conflict Country 0.179 0.126 - 0.000923 - 0.0118 - 0.0192 0.00118 .0-289** .0-280** - 0.0800 - 0.0556 - 0.142 - 0.0851 I (0.170} (0.129) (0.121) (0.115) (0.0656) (0.0993) (0.113} (0.120} (0.188) (0.146) (0.168) (0.115) j CGAC/Window Country 0.00178 - 0.0216 0.0718 - 0.00342 0.0949 0.0117 - 0.0453 - 0.0211 - 0.0499 - 0.000894 0.0381 - 0.00802 i (0.110) (0.105) (0.0821) (0.0977) (0.0724) (0.0775) (0.116) (0.109) (0.137) (0. 103) (0.0964) (0.0947) ! CGAC/Window Country • 0.0874 0.137 0.246* - 0.0503 - 0.153 0.0931 ~ Post-GAC - (0.156) (0.142) (0.131) (0.157) (0. 143) (0.145) I 0.389 0.556 ! Constant 0.0784 0.378 0.745** 0.409 I (0.3751 (0.352) (0.2961 (0.335) (0.3311 10.3381 ! Observations 195 195 194 194 195 195 193 193 195 195 195 195 l Pseudo/ R' 0.282 0.326 0.240 0.391 0.332 0.337 0.261 0.296 0.002 0.389 0.004 0.273 UQg Likelihood - 96.402 - 75.540 58.167 - 97.255 87.236 - 91.922 Note: Models 1, 3, 5, 7, 9, and 11 report ma rginal effects of a Log it with robust standard errors in parentheses; Models 2, 4, 6, 8, 10, and 12 report an OLS with robust standard errors in paren theses ·~· p 3.5 1=Governance-cluster Country CPIA is Greater than 3.5, 0 = Not FPD 1 = Project is an FPD project, 0 = Not GovEPOL 1 = Project is a Governance/ EPOL (PREM) project, 0 = Not HD 1 = Project is a Human Development project, 0 = Not INFRA 1 = Project is an FPD project, 0 = Not ECA 1= Project is located in ECA region, 0 = Not EAP 1= Project is located in EAPregion, 0 = Not MNA 1= Project is located in MNA region, 0 = Not LCR 1=Project is located in LCR region, 0 = Not 5AR 1= Project is located in SAR region, 0 = Not Post-Conflict Country 1 = Project is located in a Post-Conflict Country, 0 = Not CGAC/Window Country 1 = Project/ CAS is located in a CGAC/Window Country, 0 = Not CGAC/Window Country * Post-GAC Interaction term = CGAC/Window Country * Post-GAC Appendix E: Econometric Analysis I 133 Quality of PE Results I Analysis Project ICR/Latest ISR Rating OveraiiGAC 0.146* Responsiveness (0.069) Qualit y of Fiduciary 0.188* 0.496*** (0.020) (0.000) Demand side of Gov 0.121 0.483*** 0.119* (0.135) (0.000) (0.099) UCS Rating 0.155 0.535*** 0.344*** 0.108 (0.056) (0.000) (0.000) (0.135) Institutional 0.127 0.455*** 0.068 0.303*** 0.162** Strengthening (0.119) (0.000) (0.347) (0.000) (0.024) Resu lts 0.211 0.503*** 0.269*** 0.272*** 0.353*** 0.327*** (0.009) (0.000) (0.000) (0.000) (0.000) (0.000) Qual ity of PE Analysis 0.166 0.537*** 0.342*** 0.3 19*** 0.226** 0.345*** 0 .350*** (0.039) (0.000) (0.000) (0.000) (0.002) (0.000) (0.000) Note: R-squared reported with P values in parentheses,*** p<0.01, ** p <0.05, * p<0.1, Project Rating N = 155, GAC variables N = 200. iTABLE E.19o Project ICR/latest ISR Ratings for Development Objectives and GAC Measures Pairwise Correlations Project ICR/Latest I Overall GAC I Quality of I Demand side of I UCS ] Institutional Strengthen- Quality of PE ISR Rating Responsiveness Fiduciary Gov Rating ing I Results I Analysis Project ICR/ Latest ISR 1 Rating Overall GAC 0.136* Responsiveness (0.080) Qua lit y of Fiduciary 0.153 0.496*** (0.052)* (0.000) Demand side of Gov 0.085 0.483*** 0.119* (0.278) (0.000) (0.099) UCS Rating 0.1 51 * 0.535*** 0.344*** 0.108 (0.056) (0.000) (0.000) (0.135) Institutional 0.156** 0.455*** 0.068 0.303*** 0.162** Strengthening (0.048) (0.000) (0.347) (0.000) (0.024) Resu lts 0.165** 0.503*** 0.269*** 0.272*** 0.353*** 0.327*** (0.037) (0.000) (0.000) (0.000) (0.000) (0.000) Quality of PE Analysis 0.193** 0.537*** 0.342*** 0.319*** 0.226** 0.345*** 0.350*** (0.013) (0.000) (0.000) (0.000) (0.002) (0.000) (0.000) Note: R-squared reported with P values in parentheses,*** p <0.01, ** p